Teaching, Learning and Investigating Pragmatics : Principles, Methods and Practices [1 ed.] 9781443883320, 9781443877190

Citation preview

Teaching, Learning and Investigating Pragmatics

Teaching, Learning and Investigating Pragmatics: Principles, Methods and Practices Edited by

Sara Gesuato, Francesca Bianchi and Winnie Cheng

Teaching, Learning and Investigating Pragmatics: Principles, Methods and Practices Edited by Sara Gesuato, Francesca Bianchi and Winnie Cheng This book first published 2015 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2015 by Sara Gesuato, Francesca Bianchi, Winnie Cheng and contributors All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-4438-7719-0 ISBN (13): 978-1-4438-7719-0

TABLE OF CONTENTS

Preface ........................................................................................................ ix Sara Gesuato, Francesca Bianchi and Winnie Cheng Introduction ................................................................................................. 1 Francesca Bianchi Part I: Principles, Methods and Practices in Pragmatics and Pragmatics-Focused Pedagogy Section 1. Teaching and Learning Chapter One ............................................................................................... 13 Teachers’ Perceptions of Email Requests: Insights for Teaching Pragmatics in Study Abroad Contexts Eva Alcón-Soler Chapter Two .............................................................................................. 33 “Teacher! You Need to Give Me Back my Homework:” Assessing Students’ Needs for a Pragmatics Curriculum in an Academic ESL Program Patricia Frenz-Belkin Chapter Three ............................................................................................ 57 Pragmatic Competence, Length of Residence, Amount of Contact and Intensity of Interaction Zohreh R. Eslami and Soojin Ahn Chapter Four .............................................................................................. 85 Comparing Textbooks and TV Series as Sources of Pragmatic Input for Learners of Italian as a Second Language: The Case of Compliments and Invitations Elena Nuzzo

vi

Table of Contents

Chapter Five ............................................................................................ 109 Teaching L2 Pragmatics: From an Empirical Study to Recommendations for Pedagogical Practice Phyllisienne Gauci Section 2. Testing and Assessing Chapter Six .............................................................................................. 135 Designing Instructional Effect Studies for L2 Pragmatics: A Guide for Teachers and Researchers Kathleen Bardovi-Harlig Chapter Seven.......................................................................................... 165 “You Have No Choice:” Pragmatic Considerations in Current Language Test Instruments Richard Chapman Chapter Eight ........................................................................................... 185 DCTs versus Naturally Occurring Data in the Realization of Disagreement by Non-Native Speakers of English Carmen Maíz-Arévalo Chapter Nine............................................................................................ 207 The Use of Corpora to Identify the Pragmatic Knowledge Associated with Different Levels of Language Proficiency María Luisa Carrió-Pastor and Miguel Casas Gómez Part II: Principles, Methods and Practices in Specific Areas of Pragmatic Competence Section 3: Conversation Chapter Ten ............................................................................................. 231 An Interaction-Focused Pedagogy based on Conversation Analysis for Developing L2 Pragmatic Competence Anna Filipi and Anne-Marie Barraja-Rohan Chapter Eleven ........................................................................................ 253 Topic Management and Interactional Competence in Spanish L2 Conversation Marta García García

Teaching, Learning and Investigating Pragmatics

vii

Chapter Twelve ....................................................................................... 275 Peer and Teacher Feedback in Improving Pragmatic Conversation Skills William Collins Chapter Thirteen ...................................................................................... 297 Towards a Pedagogical Framework to Develop the Listenership of Japanese EFL/ESL Learners Pino Cutrone Section 4: Speech Acts Chapter Fourteen ..................................................................................... 323 Criteria for the Identification of Moves: The Case of Written Offers Sara Gesuato Chapter Fifteen ........................................................................................ 363 “Thanks a Bunch:” Cross-Cultural Comparison of the Speech Act of Thanking Winnie Cheng and Andy Seto Chapter Sixteen ....................................................................................... 387 The Variability of Compliment Responses: Italian and German Data Marina Castagneto and Miriam Ravetto Chapter Seventeen ................................................................................... 415 Universality and Relativity in Cross-Cultural Pragmatics: Requests and Apologies in English and Italian Loredana Pozzuoli Section 5: Aspects of Grammar Chapter Eighteen ..................................................................................... 441 The Acquisition of Emotional Competence in L2 Learners of Italian through Specific Instructional Training Anna De Marco and Emanuela Paone Chapter Nineteen ..................................................................................... 469 Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German Marion Weerning

viii

Table of Contents

Chapter Twenty ....................................................................................... 495 Instructional Advantages of a Pragmatic Account of Mood Distribution in Spanish Complements Patxi Laskurian-Ibarluzea Chapter Twenty One................................................................................ 519 Conversational Implicatures in the Croatian EFL Classroom Ana Werkmann Horvat and Ana Kevdeš Chapter Twenty Two ............................................................................... 547 Some – and Possibly All – Adults Compute Scalar Implicatures Yhara M. Formisano

PREFACE SARA GESUATO, FRANCESCA BIANCHI AND WINNIE CHENG

From 17 to 20 June, 2013, the Department of Linguistics and Literary Studies at the University of Padua, Italy hosted “Pragmatics on the go: teaching and learning about pragmatics – principles, methods and practices,” an international conference on applied pragmatics. The idea for such a conference emerged gradually in the minds of the two convenors, Sara Gesuato (University of Padua, Italy) and Francesca Bianchi (University of Salento, Italy) as a result of their experience as researchers and lecturers in EFL, as well as in discussions with colleagues – in Italy and around the world – involved in language teaching and crosslinguistic/cultural communication. It appeared that scholars and lecturers felt the urgent need to address scientific and pedagogic issues at the intersection between pragmatics and applied linguistics. It was equally clear, though, that such concerns were not adequately dealt with in teaching materials, scholarly publications, teacher training programs, workshops or refresher courses. The “Pragmatics on the go” conference provided a forum for focusing attention on these issues in a practical and systematic way. The conference brought together over 50 participants from 17 countries – linguists, teachers and other practitioners active in pragmatics and related disciplinary fields, such as sociolinguistics, discourse analysis, conversation analysis, language education, and communication studies. The large turnout of the academic event was due not only to the presence of three internationally renowned guest speakers – Eva Alcón-Soler (Jaume I University, Spain), Winnie Cheng (The Hong Kong Polytechnic University, China), and Kathleen Bardovi-Harlig (Indiana University, USA) – but also to the participants’ strong interest in discussing proposals for fostering the learning of linguistic pragmatics in second/foreign language education and for laying out clear practices in applied pragmatics.

x

Preface

The conference participants shared their experiences on how to implement the teaching of pragmatic phenomena in the language learning context with regard to several target languages. A variety of domains of investigation were covered (developmental pragmatics, computermediated communication, conversation analysis, intercultural competence, bilingualism, interpreting and emotional competence) and a number of topics were addressed (humour, verbal abuse, verb mood choice, discourse markers, academic discourse, deductive vs. inductive instruction, textbook writing, syllabus design and needs analysis). The conference convenors felt that all the commitment and dedication experienced at the conference and especially the rich and diverse contributions offered should be accessible to the wider scientificprofessional community. It was thought that the natural next step would thus be to compile a publication with the most significant conference contributions. Most of the conference participants enthusiastically answered the call to contribute an essay and one of the guest speakers, Winnie Cheng, generously agreed to take on the task of being a co-editor of the volume together with the conference convenors. Putting together this volume has required a lot of patient and meticulous work on behalf of all the people involved in their various capacities: everyone’s contribution is gratefully acknowledged. The papers in the volume testify to the commitment of practitioners in applied pragmatics to pursue such key goals as better understanding and accounting for linguistic behaviour; developing and field-testing teaching strategies; designing syllabi that make theoretical principles relevant and useful to language learners, and directly applicable to their interactional needs; and sharing materials, suggestions and research findings. We hope that this book may contribute to enabling the reader to achieve their own professional goals in linguistic pragmatics too.

INTRODUCTION FRANCESCA BIANCHI

The present volume aims to shed new light on pragmatic issues of relevance to applied linguistics, and in particular to language teaching and learning. To this aim, it brings together a number of academics, language specialists, and teachers working in a variety of contexts. We trust that the result can be of concrete use to applied linguists, PhD students in communication studies, language teachers, teacher trainers, examiners, materials developers, and experts in transcultural communication alike. The book is divided into two parts. Part 1 deals with issues of pragmatics and pragmatics-focused pedagogy, that is, the principles, methods and practices of its teaching and learning (Section 1), and the testing and assessing of pragmatic competence (Section 2). Part 2 focuses on specific areas of pragmatic competence, namely conversation (Section 3), speech acts (Section 4), and the functional use of aspects of grammar (Section 5). Section 1 investigates and discusses the factors, contexts, and inputs that facilitate or hamper the learning of pragmatic norms and habits. In Chapter 1, Eva Alcón Soler focuses on the influence of a studyabroad period on international teenage students’ performance of email requests. The author analyses international students’ requests before and after three months of stay in the UK, and compares the post-study-abroad data to data from British English students. She also considers teachers’ assessment of those e-mails, concluding that, despite some decrease in the use of direct strategies and some gains in accuracy at the end of the studyabroad period, the performance of international students does not approximate target language pragmatic norms, and that their e-mail requests are still generally inappropriate. This leads the author to recommend incorporating pragmatic instruction during study-abroad periods. Chapters 2 and 3 also focus on pragmatic acquisition in an L2 context, with specific attention to contextual factors such as length of residence and amount and type of language contact with native speakers. In Chapter 2,

2

Introduction

Patricia Frenz-Belkin analyses the performance of a group of (predominantly) Spanish-speaking immigrant students living in the United States, in a variety of scenarios. The author finds that schooling in the target language or time of residence in the target culture do not appear to have an impact on immigrants ability to produce appropriate utterances or to correctly evaluate the appropriateness of given utterances in face-threatening situations. This is especially true when the scenario is unfamiliar, such as an academic setting characterized by power imbalance between the interlocutors. In Chapter 3, Zohreh R. Eslami and Soojin Ahn analyse length of residence in the target community, too, alongside amount and type of language contact with native speakers. Their data on the performance of Korean advanced ESL learners when producing compliments and compliment responses show that learners with more frequent opportunities for interactive use of language develop their pragmatic competence significantly better than those with fewer opportunities for interaction. At the same time, the findings suggest that simple exposure to language is unlikely to be sufficient for the acquisition of L2 pragmatic knowledge and that diverse opportunities for interaction with other native and non-native English language users should be provided to learners. In Chapter 4, Elena Nuzzo compares and assesses two different sources of pragmatic input for foreign language learners, namely Italian-language textbooks and TV series. This author observers that the latter offer a wider and richer variety of examples of compliments and invitations, a finding in keeping with similar studies on other languages. Finally, in Chapter 5, Phyllisienne Gauci explores the potential of explicit and implicit teaching of Italian pragmatics in a foreign language context, focusing on requests and complaints from film extracts and other audiovisual material. The author compares the results of an explicit teaching experimental group, an implicit teaching experimental group, and a control group, and finds that both experimental groups performed better than the control group – merely exposed to the target language – in terms of production, as well as of awareness, and regardless of the teaching approach. Finally, the author notices that the positive effects of the instructional treatment are not always retained in the long-term and that results may be highly influenced by the testing instrument used. Section 2 addresses questions such as how to assess pragmatics acquisition, what is that tests assess, and which features should be assessed. This section opens with an essay by Kathleen Bardovi-Harlig, a practical guide for the development of research designs for studying the

Teaching, Learning and Investigating Pragmatics

3

effect of instruction on the development of L2 pragmatics in both host and foreign environments. This guide illustrates standard practices as they apply to pragmatics research and takes into account factors related uniquely to pragmatics. In Chapter 7, Richard Chapman analyses the testing material and practice of the Cambridge Proficiency: English examination and observes that this test, like many other similar ones, requires candidates to share the presuppositions of the testing organisation. Such presuppositions are so strongly imbued in the system that Cambridge exam writers and testers fail to notice ambiguous items, or regard as equivalent forms which are actually different in terms of illocutionary force or markedness. The author thus argues that this type of tests do not test real communicative competence, but rather the competence of the student to comply with previously acquired meta-rules as to how to approach the exam. In Chapter 8, Carmen Maíz-Arévalo compares naturally occurring online data produced on an asynchronous forum to DCT data produced by the same participants. This author shows that DCT data might mislead the researcher into rather simplistic views of non-native speakers’ performance. Finally, in Chapter 9, María Luisa Carrió-Pastor and Miguel Casas Gómez use learner corpus data to identify the use of hedges by learners of Spanish at different stages of second language learning. They also advocate the use of corpora for the creation of a list of the hedges associated with the different levels of second language learning, as a way to assist teachers and learners in the improvement of pragmatic competence in written discourse. Section 3 focuses on the analysis of conversational features and provides suggestions for developing conversation skills. In Chapter 10, Anna Filipi and Anne-Marie Barraja-Rohan show how non-native speakers have difficulties in initiating repair, producing appropriate responses, and closing conversations. From these observations they proceed to illustrate a pedagogical approach to teaching conversation skills which is based on four stages. The authors report on how this approach has been successfully used in teaching L2 students, and how it can serve as a valid model for training teachers of second languages. In Chapter 11, Marta García García examines how Spanish L2 learners confront the issue of topic management and, more specifically, how they solve conflicts in cooperative topic closures. In Chapter 12, William Collins discusses and compares the impact of teacher- and peer-feedback in helping learners to develop conversation skills, with particular attention to backchanneling, active listening, and the

4

Introduction

use of intonation to signal emotional involvement and empathy. The study, conducted with Japanese advanced learners of English in Japan, analyses data from conversation recordings, student transcriptions of the conversations, teacher feedback comments or student feedback comments, and student surveys concerning improvements in motivation and selfconfidence. The author also investigates the students’ ability to recognize gaps in listener participation in recorded conversations and to suggest suitable responses to speakers’ turns. The results suggest that both receiving and offering feedback has a positive impact on the students’ conversation skills. Finally, in Chapter 13, Pino Cutrone outlines a pedagogical framework for teaching and assessing backchannel skills. His framework develops from a detailed analysis of the literature on backchanneling, and considering variables such as: types and functions of backchannels; frequency; discourse contexts favouring backchannels; and conversational involvement. Section 4 offers insights into a rich variety of speech acts including offers, thanks, compliment responses, requests, and apologies. In Chapter 14, Sara Gesuato examines how interactants maximize their chances of interactional success when they realize offers in writing. After providing a definition of the act of offering and reviewing the literature on this type of speech act, the author contributes instantiations and a classification of the component moves of offers. She also proposes a move structure analysis of written offers and highlights the pros and cons of the research method applied to this study. In Chapter 15, Winnie Cheng and Andy Seto apply corpus linguistics analytical methods to highlight and compare the linguistic and pragmatic realizations of the speech act of thanking in four spoken English corpora representing respectively general English in the UK, Hong Kong English, and English in academic settings. The identification of the most frequent thanking expressions and their collocational and colligational patterns reveals a much greater variety of forms in the general usage of English by native speakers of English in a national corpus, compared to the other corpora. Chapter 16 by Marina Castagneto and Miriam Ravetto compares and contrasts Italian and German native speakers’ reactions to compliments and provide evidence to the variability of this type of speech act according to a range of variables, including geographical area, gender, topic, and illocutionary force. Specific attention is devoted to a discussion of how

Teaching, Learning and Investigating Pragmatics

5

Italian appears to be undergoing a phase of change in the pragmatics of compliment responses. Finally, Chapter 17 by Loredana Pozzuoli analyses apology and request strategies employed by Italian learners of English. The author compares them to those employed by English native speakers in the same situational contexts. This essay aims to establish the extent to which learners approximate the norms of native speakers and transfer Italian apology and request patterns into their L2. Section 5 opens with a contribution by Anna De Marco and Emanuela Paone on the decoding and encoding processes of emotional speech in L2 learners of Italian with a typologically distant mother tongue, Indonesian. This authors present the results of a pragmatic teaching approach aimed at encouraging awareness of the prosodic aspects of speech. The study shows that both the identification and the production of vocal emotions is affected by cultural features. Furthermore, it suggests that pragmatics plays an important role in social interactions involving emotional issues. Chapter 19 deals with the pragmatic value of three German particles with a very vague lexical meaning, and whose function consists in managing the interaction between speaker and listener. For a clearer understanding of the meanings of these particles, Marion Weerning provides exemplifications from German novels and their translations into Italian, a language which does not possess lexico-pragmatical equivalents of these words. Finally, the author critically examines how German textbooks used in Italian schools treat these three particles and proposes some new, more effective strategies for teaching their meanings and usages. Chapter 20 by Patxi Laskurian-Ibarluzea focuses on mood in Spanish and the way it is generally treated in teaching materials. The author suggests that “it is the meaning of a matrix [clause] in a given communicative context, rather than simply the meaning of a predicate, that licenses speakers to assert the propositional content of a complement” (p. 515). The author, thus, outlines a speaker-based pragmatic theory of mood in Spanish which can account for and predict mood choice by the speaker, depending on the discourse situation, and which represents a valid pedagogical tool for teaching mood selection in Spanish complements to students of Spanish as a second or foreign language. Finally, this section ends with two chapters on implicature. In Chapter 21, Ana Werkmann Horvat and Ana Kevdeš explore the use of active learning strategies for teaching conversational implicatures to Croatian learners of English. This essay reports the results of a student-centred instruction procedure focusing on conversational implicatures. It discusses

6

Introduction

them in the light of the students’ responses during a workshop and in a subsequent survey about the value of conversational implicatures instruction and about their overall awareness of the importance of acquiring pragmatic competencies. To conclude, in Chapter 22, Yhara M. Formisano investigates scalar implicature comprehension in English and Italian L1 and L2 students. The author reports two experiments aimed at testing whether adults in L1 and L2 interpret the quantifier some logically or pragmatically. Her results show that L1 as well as L2 speakers never give a logical interpretation of some. Non-target like answers are a consequence of the subject’s conjuring up of alternative realities or misinterpretation of an item. Moreover, the results show that pragmatic competence is not higher in L2 than in L1. Given the broad range of topics covered, the volume lends itself to multiple reading paths, synthetized into five lists below. Thus, depending on their interests, readers may want to read about specific speech acts (List 1), conversational features at large (List 2), specific learning or teaching contexts, or media (List 3), precise types of data or data collection methods (List 4), or various languages (List 5), as specified below. List 1: Speech acts Compliments (Chapter 3) Compliment responses (Chapters 16, and 20) Complaints (Chapter 4) Thanks (Chapter 15) Invitations (Chapter 3) Requests (Chapters 1, 4, and 17) Questions (Chapter 19) Offers (Chapter 14) Apologies (Chapter 17) Disagreements (Chapter 8) List 2: Conversational features Backchannelling (Chapters 12, and 13) Conversational strategies (Chapters 10, and 11) Mitigation and hedging (Chapters 1, 4, and 9) Implicature (Chapters 21, and 22) Mood (Chapter 20) Stance (Chapters 9, and 19) Emotions (Chapter 18)

Teaching, Learning and Investigating Pragmatics

7

List 3: Specific contexts Electronic interaction (Chapters 1, and 8) L2 contexts (Chapters 1, 2, 5, 10, and 18) FL contexts (Chapters 4, 9, 10, 11, 12, 13, 21, and 19) Native speaker’s performance (Chapters 4, 5, 9, 10, 11, 12, 13, 14, 15, 16, 17, and 22) Non-native speakers’ performance (Chapters 8, 17, 18, 19, 21, and 22) Explicit instructional treatment vs. implicit teaching or no teaching (Chapters 9, 18, and 21) List 4: Data and data collection methods DCTs (Chapters 4, 5, 8, 17) Role play (Chapter 4) (Un/solicited) free writing or conversations and naturally occurring data (Chapters 2, 8, 9, 10, 11, 12, 14, 15, and 16) Reference corpora (Chapter 4) Other types of data collection (Chapters 2, and 12) Textbooks and other set material (Chapters 3, 7, and 19) TV material (Chapter 3) List 5: Languages English (Chapters 1, 2, 7, 8, 9, 10, 12, 14, 15, 17, 21, and 22) Italian (Chapters 3, 4, 10, 16, 17, 18, and 22) Spanish (Chapters 11, and 20) German (Chapters 16, and 19) We are confident that any reader interested in applied pragmatics will find relevant topics covered and issues explored throughout the book.

PART I PRINCIPLES, METHODS AND PRACTICES IN PRAGMATICS AND PRAGMATICSFOCUSED PEDAGOGY

SECTION 1 TEACHING AND LEARNING

CHAPTER ONE TEACHERS’ PERCEPTIONS OF EMAIL REQUESTS: INSIGHTS FOR TEACHING PRAGMATICS IN STUDY ABROAD CONTEXTS EVA ALCÓN

1. Introduction 1.1 Background research Second language pragmatic research has frequently dealt with speech act performance. Among the different speech acts, requests have aroused a lot of interest in the field, and performance of requests has been examined by means of elicited data such as role-plays, discourse completion tests, or multimedia tasks. More recently, email communication has proved to be a context for collecting requests in a natural environment, providing us with information on how language is used in a widely accepted medium of interaction. In the context of virtual communication, the degree of directness and appropriateness of email requests has been addressed in a number of different studies. For instance, Chen (2001) examined differences between Taiwanese and American students in relation to requests for an appointment, requests for a recommendation letter and requests for special consideration. The author reported differences in the amount of lexical and syntactic modification employed by Taiwanese and American graduate students, while both groups favoured querypreparatory strategies. In a similar vein, Biesenbach-Lucas (2006, 2007) used emails that students sent to the researcher to examine the degrees of directness and indirectness in three types of requests (requests for an appointment, for feedback and for an extension of deadlines). Results of this study showed that both native and non-native speakers used direct requests for appointment and feedback, while there was a tendency to use

14

Chapter One

conventional indirect requests when asking for an extension of deadline, thereby suggesting that the level of imposition of the request may influence the degree of directness of the request. Moreover, studies by Hartford and Bardovi-Harlig (1996), Economidou-Kogetsidis (2011), and Pan (2012) have dealt with the performance and perception of email requests. Hartford and BardoviHarlig (1996) analysed email requests produced by native and non-native speakers of English and how they were evaluated by faculty members. Findings of their study revealed that learners did not employ mitigation devices, emphasised students’ needs and lacked status-congruent language. Economidou-Kogetsidis (2011) examined email requests sent by Greek Cypriot university students to faculty over a period of several semesters. The author analysed the degree of directness, mitigation and forms of address, reporting that students’ emails presented a high frequency of direct strategies, an absence of lexical mitigators, and inappropriate forms of address. The author also reported that such emails were perceived as impolite and were thus capable of causing pragmatic failure. Finally, Pan’s (2012) study examined internal and external modifiers in email requests produced by Chinese learners and compared them with those produced by American participants. Similar findings to previous studies on email requests were found, reporting that L2 users relied mainly on external modifiers and did not often use syntactic modifiers. In line with the above contrastive studies, Alcón-Soler (2013a) examined the use of request strategies and mitigation devices produced by International Baccalaureate students – British English speakers (BES) and International English speakers (IES) – during student-initiated email communication. Findings of the study show IES’s greater preference for direct strategies and external mitigators in comparison to BES, while no differences are found in the use of lexical and syntactic modifiers. Taking into account these findings, it was hypothesised that, besides lack of pragmalinguistic knowledge, participants’ perception of social distance from the interlocutor and perception of request imposition may explain students’ preference for direct requests. To further explore this tentative hypothesis, the author (2013b) analysed whether the use of internal and external request mitigators were influenced by participants’ judgements of the degree of request imposition and social distance with the email recipient. Results of the study show that participants do not frequently rely on mitigators, but whenever they perceive the need to mitigate the request, pragmatic variation is observed between BES and IES. Thus, when the request is perceived as demanding, BES activate their pragmalinguistic

Teachers’ Perceptions of Email Requests

15

knowledge and use a wider range of internal modifiers (both lexical and syntactic). In contrast, IES seem to lack the pragmalinguistic knowledge needed to soften the request. Finally, although teenagers do not perceive their relationship with their learning mentor as one of + social distance, both BES and IES show variation in their choice of form of address. Given that, as suggested by Crystal (1997), pragmatics deals with language use and its effects on participants in the act of communication, the present study, as a follow up to Alcón (2013a, 2013b), deals with how teachers perceive student-initiated email requests. More specifically, we will examine teachers’ perceptions of email requests in a context of insensitive exposure to the language, such as study abroad (SA). One of the factors that have been examined in relation to pragmatic learning is the environment. Barvodi-Harlig’s (2013) suggestion to look at learners’ interaction with the environment may explain the interest in exploring pragmatic learning in environments involving intensive exposure to language, such as SA. Previous cross-sectional studies have shown that SA is insufficient for pragmatic development (Bardovi-Harlig and Dörney 1998; Barron 2003; Matsumara 2003; Shauer 2009; Bella 2011; Taguchi 2011a, 2013; Vilar-Beltrán 2014; Félix-Brasdefer 2015). Besides, studies such as the one reported by Bardovi-Harlig and Bastos (2011) point out that variables such as level of proficiency or intensity of interaction make a difference when we look at pragmatic learning in SA contexts. These studies also report individual variation in pragmatic gains. Following the interest in pragmatic learning in SA contexts, and previous studies on performance of L2 email requests, the present study explores whether a short period of SA influences students’ performance and teachers’ perceptions of email requests. To date, the population of the studies conducted on learning requests during study abroad has been university students, with the exception of Achiba (2003), who studied her 7-year-old daughter, and Ellis’s (1992) study of two early adolescents. The present study deals with a different population: late adolescent learners in SA experiences. In adolescence there is an increasing use of both synchronous and asynchronous communication, and this affords the researcher the possibility to collect authentic language in an environment participants are familiar with. The following questions guided the study: RQ1: Do students differ in their performance of email requests after a short period of SA? RQ2: Does a short period of SA make a difference in teachers’ perceptions of email requests?

16

Chapter One

2. Methodology 2.1 Participants A total of 60 teenagers studying in three different state schools in the south of England were randomly selected from those who filled in the consent form to participate in the study and gave permission for the emails they sent to their mentors to be examined for research purposes. All participants were sixteen years old and were following the International Baccalaureate (IB) curriculum. Moreover, whether they were British English Speakers (BES) or International English Speakers (IES) was taken into consideration to form two participant groups: 30 BES and 30 IES. Since one of the BES moved to Australia four weeks after starting the IB programme, 29 BES and 30 IES finally took part in the study. The BES email requests were used as baseline data to examine differences between BES and IES (see Alcón 2013a; 2013b). IES’ performance of email requests was the focus of the present study. The IES were all Europeans who were studying in the UK for one academic year. Their level of English language proficiency was upper intermediate, as established by the standardised Quick Oxford Placement test (UCLES 2001), equivalent to Common European Framework level B2. In addition, six British teachers, who were mentors of the IB students, also participated in the study by forwarding to the researcher the email messages of those students who had previously agreed to have their email messages examined for research purposes.

2.2 Data collection and analysis For the present study we examined the emails that IES students sent to their learning mentors at two different times: September and December 2011. Thus, 150 email requests performed by IES were analysed in terms of level of directness and amount of mitigation at two different times: At Time 1 when participants arrived in the country (September, 2011) and at Time 2 after three months of SA (December, 2011). Considering the classification suggested by Blum-Kulka, House and Kasper (1989) and modified by Biesenbach-Lucas (2007) and Félix-Brasdefer (2012), the presence of direct strategies (imperative, performative, want statement, need statement, direct question, like/appreciate statement, expectation statement), conventionally indirect strategies (query preparatory, ability/willingness/ permission), and non-conventionally indirect strategies (hints) were used to examine frequency of request strategies (see Table 1).

Teachers’ Perceptions of Email Requests

17

TABLE 1. Request strategies in student-initiated email requests Directness Level

Request strategies Imperative Performative Want statements

Direct

Need statements Direct questions Like/appreciate statements Expectation statements

Conventionally indirect

Non-conventional indirectness

Query preparatory (ability/willingness)

Hints

Examples Send attachment again I am asking you information about… I want to confirm the day of the meeting I need to talk to you Where can I find the book? I'd like to have a meeting with you …I hope you can contact me no later than Friday Can I borrow your book at the end of this week? There seems to have been a problem with the web lately and I am not sure about your last deadline

In addition, following previous classifications on requests (BlumKulka, House and Kasper 1989; Trosborg 1995; Hassall 2012; Achiba 2003; Woodfield 2012; and Woodfield and Economidou-Kogetsidis 2010), the researcher examined the presence of internal and external modifiers (see Table 2 for examples of internal modifiers, and Table 3 for examples of external modifiers).

Chapter One

18

TABLE 2. Lexical and syntactic modification devices in studentinitiated email requests Type

Sub-type Please Openers (do you think..; would you mind…; is it all right…)

Lexical

Softeners (downtonerspossibly, perhaps, just, maybe…; understatersjust, a little, a minute…; hedges- kind of…) Intensifiers (really, I'm sure…) Subjectivisers Conditional structures

Tense

Syntactic Aspect Negation of preparatory condition Multiple syntactic modification

Examples Please, find attached a document with changes It is all right if I pop in Tuesday after lunch? Could you just let me know by the end of the term? I'd possibly need some feedback before the English class. I really need your help with that project. I suppose I could hand in the paper next week Could you please tell me when the deadline for the assignment is? Is it all right if I booked for the performance later in the week? I was wondering if what I sent for the geography paper was ok I don’t suppose there is any choice… I was wondering whether you could send doc III in word

Teachers’ Perceptions of Email Requests

19

TABLE 3. External modification devices in student-initiated email requests Type

Sub-type Preparators Grounders Disarmers

Expanders

External modification

Promises Imposition minimisers Apologies Preparators Grounders

Examples I really need to talk to you, Could we meet…? I have to go to the dentist tomorrow at 12.00. Could I write to you if…? I hate bothering you again, but could you confirm…? I would like to know about assignment 4 because I missed your class today (I was sick). Could you please explain what I have to do? Could I do it next week? I promise this will not happen again. I would like to see you before the Assembly, just for five minutes I'm very sorry, but I need to answer some questions… I really need to talk to you, Could we meet…? I have to go to the dentist tomorrow at 12.00. Could I write to you if…?

Finally, to examine teachers’ perception of students’ email requests, immediately after receiving the email, the British mentors were asked to evaluate their students’ emails as regards accuracy and appropriateness on a 5-point Likert scale. At the same time they were asked to explain their choice (see example 1).

Chapter One

20 (1)

Example of data collection on perception: Please read the following emails and choose the options that best represent your opinion. We are interested in your personal opinion, so your answers can never be wrong. Subject: Date: Thu, 22 Sep 2011 16: 52:20 +0200 Hi (name of the mentor) I would like to have a meeting with you next week. If you can before Wednesday, what do you think? Thank you (Name of the student) This email is grammatically correct 1. not at all 2. a little 3. so-so 4. quite correct 5. completely correct Please explain your choice This email is appropriate 1. not at all 2. a little 3. so-so 4. quite appropriate 5. completely appropriate Please explain your choice

A chi-square test of independence was used to establish statistically significant differences between frequency of request head acts and mitigators used by IES before and after the SA experience. A t-test was used to compare teachers’ perception of accuracy and appropriateness of IES before and after a short period of SA.

3. Results and discussions As illustrated in Table 4, IES resort to the use of direct strategies more frequently at the beginning of the SA than after three months of intensive exposure to email communication, the difference being statistically significant. In addition, analysis of the data shows a general tendency towards the use of query (Can we discuss my project for next term on Monday?) as a conventionally indirect strategy. Finally, the use of hints is practically absent in our data base. Findings related to strategies used by IES during a short period of SA indicate changes in the use of direct strategies, but not in the use of conventionally indirect strategies. Since the use of conventionally indirect strategies in L1 and L2 has also been reported in previous studies (House and Kasper 1987; Trosborg 1995;

Teachers’ Perceptions of Email Requests

21

Biesenbach-Lucas 2006, among others), it might be that L2 learners use this type of strategy as formulaic language in request performance. Finally, practically no use of non-conventional strategies is observed in our corpus. TABLE 4. Frequency of request strategies at the beginning and at the end of SA

Direct Conventionally indirect Non-conventionally indirect

TIME 1 TIME 2 Total TIME 1 TIME 2 Total TIME 1 TIME 2 Total

Participants 30 30 60 30 30 60 30 30 60

Total

%

54 114 168 144 145 289 1 2 3

32.14 67.86 100 49.44 50.56 100 49.15 50.85 100

Chi 2

df

Sig

25.02

7

0.00

14.70

4

0.01

0.66

1

0.42

In relation to the use of request mitigators, Table 5 shows that more lexical modifications are observed at Time 2 (53.78%) than at Time 1 (46.22%), although the difference is not statistically significant (F2=4.14; df=4; p=0.25). Neither is the difference statistically significant for the use of syntactic mitigators (F2=1.37; df=4; p=0.50). In the case of external modifications, the frequency of use between Time 1 and Time 2 is statistically significant (F2=11.53; df=4; p=0.03). TABLE 5. Frequency of request mitigators at the beginning and at the end of SA

Lexical

Syntactic

External

TIME 1 TIME 2 Total TIME 1 TIME 2 Total TIME 1 TIME 2 Total

Participants 30 30 60 30 30 60 30 30 60

Total

%

65 74 139 30 36 66 30 61 91

46.22 53.78 100 45.45 54.54 100 36.03 63.96 100

Chi 2

df

Sig

4.14

4

0.25

1.37

4

0.50

11.53

5

0.03

22

Chapter One

Changes in IES’ performance of email requests at Times 1 and 2 can be analysed in relation to BES’ performance in the same situational contexts. As reported by Alcón (2013a), BES and IES’ email requests entail directness, but IES show a greater preference for direct strategies and external mitigators in comparison to BES. If we compare the use of email requests by IES after SA and the BES email request performance, reported in Alcón (2013a), we could claim that IES preference for the use of external mitigators after the SA period is not in line BES performance of email requests. In addition, in high imposition requests, while BES show a preference for softeners, after a short period of SA, IES continue to rely on direct questions, the use of “could,” the overuse of “please” and the use of grounders. Thus, it seems that short periods of SA do not trigger changes towards target language pragmatic norms. In line with previous cross-linguistic research on email requests, pointing out differences between L2 and L1 in request emails (Economidou-Kogetsidis 2011; Félix-Brasdefer 2012), our findings suggest that these differences are also observed after a short period of SA. Findings from the present study support previous research conducted on pragmatic learning in SA that has provided evidence that a short period of SA is not enough for pragmatic development (Bardovi-Harlig and Dörney 1998; Barron 2003; Matsumara 2003; Shauer 2009; Taguchi 2011b, 2013; Vilar-Beltrán 2014; FelixBrasdefer 2015). To answer our research question 2, which deals with whether a short period of SA influences teachers’ perceptions of email requests, we examined teachers’ reports on perception of email requests both quantitatively and qualitatively. The quantitative analysis shows that after a short period of SA teachers’ perception of accuracy (M = 3.92; SD = .72) is higher than at the beginning of SA (M = 2.76; SD = .70), being the difference significant (t = 2.49 p < .001). A difference is also observed in perception of email appropriateness (M = 2.43; SD = .50, at the beginning of SA; M = 3.12; SD = .67, and at the end of SA). However, this difference is not significant (t = 5.88; p = .01). Taking into account these results, we can claim that a short period of SA makes a difference in the accuracy rather than in the appropriateness of email requests. To further understand why email requests are not perceived as appropriate at the end of the SA, we analysed whether general patterns of email requests influence teachers’ perception of students’ emails. It was found that some patterns of email request performance do influence teachers’ perception of email appropriateness. First, after three months of SA, and in line with their performance at the beginning of the SA, participants produce brief and direct email requests. In the case of low imposition requests, this

Teachers’ Perceptions of Email Requests

23

performance is perceived as appropriate, although sometimes emails are described as abrupt (see examples 2 and 3) (2)

(Name of the teacher), please could you confirm school postal address for me? Thanks. (Name of the student) This email is appropriate 1. not at all 2. a little 3. so-so 4. quite appropriate 5. completely appropriate Please, explain your choice “Appropriate, but sounds a bit abrupt as there is no greeting.”

(3) (Name of the teacher), please could I have the school postal address? I need to send the forms that you mentioned Thanks. (Name of the student) This email is appropriate 1. not at all 2. a little 3. so-so 4. quite appropriate 5. completely appropriate Please, explain your choice “This email is completely appropriate, the salutation and sign off are both polite, a clear reason is given for the request. It could have benefitted from a pre-close.”

Secondly, participants do not show any ability to mitigate high imposition requests, and they over-generalise the use of informal and direct language use in teacher-student email conversation, to which they have been exposed during their SA experience, in making both low and high imposition requests. In the case of high imposition requests, their linguistic behaviour has a negative impact on teachers’ perceptions of their emails. This is illustrated in example 4 when the teacher reports no consideration of the degree of imposition of the request.

Chapter One

24 (4)

(Name of the teacher), please where can I meet you before the lesson? Thanks. (Name of the mentor) This email is appropriate 1. not at all 2. a little 3. so-so 4. quite appropriate 5. completely appropriate Please explain your choice “This email makes absolutely no attempt to soften the request or provide any additional information, it is too direct and there is no attempt at politeness.”

Thirdly, learners do not seem to be aware of how the teacher-student relationship influences language choice. Teachers report a lack of conventions to show respect (example 5) and a lack of ability to open and sign off the email requests (example 6). (5)

Dear (Name of the teacher), Where can I find the tasks for next week? Not found in the folder Thanks. (Name of the mentor) This email is appropriate 1. not at all 2. a little 3. so-so 4. quite appropriate 5. completely appropriate Please explain your choice “I would say it is a little abrupt. It would be better if the email had some of the conventions which show respect to the social roles of the two parties. e.g., “Sorry to bother you. I hope you can find a minute to respond to my question. This convention may be a cultural one. I am English and perhaps this is not necessary in other cultures.”

(6)

Hi (name of the mentor), Here's the homework. Hope it is what you wanted.

Teachers’ Perceptions of Email Requests

25

This email is appropriate 1. not at all 2. a little 3. so-so 4. quite appropriate 5. completely appropriate Please explain your choice “The greeting is too familiar and there is no valediction or name at the end.”

In spite of how these patterns influence perception of email appropriateness, our data also show variation in teachers’ perception of the appropriateness of students’ emails. While three teachers consider brief and direct email requests as appropriate, considering the particular context of student-teacher email conversations (“Can’t see a problem here,” “I expect this level of informality,” “It is informal but polite for this context,” “The shortness of the emails reflects the relationships and the context”), the other three teachers consider them inappropriate and justify their assessment of email politeness on the use of opening and closings (“There is no name at the end,” “The greeting is too familiar and there is no valediction or name at the end,” “I would expect a more formal opening (Dear x) from a school pupil of his age,” “Greeting too informal”). In this case, we cannot explain this difference in terms of teachers’ cultural background, since all the teachers were British, but a tentative hypothesis could be that the age of the secondary school teachers also mediates in the perception of request imposition. This may explain why younger teachers, probably more used to the language employed in virtual communication, perceive brief and direct email requests as appropriate, while older teachers miss the linguistic devices expected in face-to-face interaction. Of course this is a tentative hypothesis to be explored in future studies. Finally, a difference is also observed in teachers’ pragmatic versus grammatical knowledge. If we compare individual teachers’ comments on the perception of accuracy and appropriateness of the emails, we can see that teachers’ knowledge of grammar outperforms their knowledge of pragmatics (see example 7), which will probably influence the attention paid to grammar versus pragmatics in the classroom. (7)

(Name of the teacher), please where can I find the book that we have to read for Wednesday? This email is grammatically correct 1. not at all 2. a little 3. so-so 4. quite correct 5. completely correct Please explain your choice

26

Chapter One “I think it is correct. How to explain why? It is a question comprising of a question word + modal verb + subject followed by a noun phrase which consists, in part, of a defining relative clause. The use of the definite article in the book is correct as the writer is referring to a specific or “Known” book. The defining clause is a dependent clause, in other words, it is not a complete sentence on its own.” This email is appropriate 1. not at all 2. a little 3. so-so 4. quite appropriate 5. completely appropriate Please explain your choice “The greeting is too familiar and there is no valediction or name at the end.”

It should be pointed out that we also found individual variations in email request performance during the SA experience. Thus, while some participants show changes towards directness in email request performance, others show a shift towards mitigated email requests. Further analysis of the data may provide insights on how those changes influence teachers’ perceptions of email requests. However, this issue is beyond the scope of the present study. Insights from this study show that SA is insufficient for pragmatic learning, suggesting the need to incorporate pragmatic instruction during the SA experience. It is possible that, as reported by Alcón (2013b), teenagers do not perceive their relationship with their teacher as one of + social distance, but the author also reported that when the request was perceived as demanding, BES activated their pragmalinguistic knowledge and used a wider range of internal modifiers (both lexical and syntactic). In contrast, data from this study show that IES seemed to lack the pragmalinguistic knowledge needed to soften the request after a short period of SA. Since teachers report a lack of ability to show respect, an open question is whether this is the result of a lack of pragmalinguistic or sociopragmatic knowledge. Besides, given that it seems that SA does not improve students’ ability to perform email requests, it would be interesting to explore the interaction effect of pragmatic instruction and SA. Training in email literacy during the SA period involves, first, awareness of how social norms are understood in a particular context or how individual perceptions of these norms influence the recipient’s perception of the email. Once learners become familiar with how the social context influences language choice, it will be the time to focus on how to improve their pragmalinguistic

Teachers’ Perceptions of Email Requests

27

knowledge, with an emphasis on choice of language according to sociocultural factors. From this perspective, teaching email requests can be approached as a process of discovery. In this process, cultural comparisons, explicit instruction, awareness-raising tasks, focused practice, and feedback may facilitate the process of learning how to perform email requests. Understanding the teaching of pragmatics as a process of discovery is also a challenge to encourage collaboration between teachers and researchers. We know that the acquisition of pragmatics does not necessarily deal with pedagogy, but it is also true that pedagogy can influence the learning of pragmatics in a second language. Nowadays different pedagogical proposals are available for teaching pragmatics (see Ishihara and Cohen 2010, for activities to teach pragmatics) and different studies have proved that instruction is beneficial for learning pragmatics (see Jeon and Kaya 2006; Rose 2005; Taguchi 2011a; Taguchi 2015 for a review of the effects of pragmatic instruction; see also Yates’ 2013 special issue for studies reporting on how pragmatic aspects of language use are treated in the area of language teaching and learning). However, few studies deal with the interaction effect of instruction and SA. To our knowledge the studies exploring the teachability of pragmatics during SA focus on performance rather than on perception of requests (Winkie and Teng 2010; Shively 2010, 2011; Winkie and Teng 2010; and Alcón 2015). Further studies are needed to examine the interaction effects of instruction and SA both on participants’ language use and interlocutors’ perception of pragmatic changes observed in instructed SA contexts. The present study is a first step in that direction, since it explores whether a short period of SA influences students’ performance and teachers’ perception of email requests. Exploring the impact of pragmatic instruction during SA would also provide us with information on how instruction and SA influence L2 learners’ pragmatic changes, and the reasons for and the consequences of these changes.

4. Conclusions For many years it has been believed that an SA experience is the best option to become fluent in a target language and acquire cultural knowledge in that language. Following this assumption, the present study examines whether a short period of SA influences performance and perception of email requests. Results of the study show that, after a short period of SA, a decrease in the use of direct strategies and some gains in accuracy are observed. Despite this general pattern in performance of email requests, teachers report no consideration of request imposition on

28

Chapter One

behalf of the student. In addition, high-imposition requests are perceived as abrupt and too direct, with a lack of conventions to show respect. Finally, individual variation is also present in teachers’ perception of email requests and in their knowledge of grammar and pragmatics. Given that a short period of SA does not seem to have an impact on perception of email appropriateness, the author suggests incorporating pragmatic instruction during SA. Further research is encouraged to examine the interaction effect of instruction and SA on performance and perception of email requests. Taking into account previous studies in the area of instructional pragmatics, it is suggested that the teaching of email requests be approached as a process of discovery, which is also a challenge to encourage collaboration between teachers and researchers.

References Achiba, M. 2003. Learning to Request in a Second Language. A Study of Child Interlanguage Pragmatics. Clevedon: Multilingual Matters. Alcón-Soler, E. 2013a. “Pragmatic variation in British and International English language users’ e-mail communication: A focus on requests.” Revista Española de Lingüísitca Aplicada 26: 25-44. —. 2013b. “Mitigating e-mail requests in teenagers’ first and second language academic cyber-consultation.” Multilingua 32 (6): 779-799. —. 2015. “Pragmatic Learning and Study Abroad: Effects of Instruction and Length of Stay.” In “Pragmatic learning across contexts,” edited by E. Alcón-Soler, and L. Yates, special issue, System 48C. Bardovi-Harlig, K. 2013. “Development L2 Pragmatics.” Language Learning 63: 68-86. Bardovi-Harlig, K., and M. T. Bastos. 2011. “Proficiency, length of stay, and intensity of interaction and the acquisition of conventional expressions in L2 pragmatics.” Intercultural Pragmatics 8 (3): 347384. Bardovi-Harlig, K., and Z. Dörnyei. 1998. “Do language learners recognize pragmatic violations? Pragmatic versus grammatical awareness in instructed L2 learning.” TESOL Quarterly 32: 233-259. Bardovi-Harlig, K., and B. S. Hartford. 1993. “Learning the Rules of Academic Talk: A Longitudinal Study of Pragmatic Change.” Studies in Second Language Acquisition 15: 279-304. Barron, A. 2003. Acquisition in interlanguage pragmatics. Learning how to do things with words in a study abroad context. Amsterdam: Benjamins.

Teachers’ Perceptions of Email Requests

29

Bella, S. 2011. “Mitigation and politeness in Greek invitation refusals: Effects of length of residence in the target community and intensity of interaction on non-native speakers’ performance.” Journal of Pragmatics 43: 1718-1740. Biesenbach-Lucas, S. 2006. “Making requests in E-mail. Do cyberconsultation entail directness? Towards convention in a new medium.” In Pragmatics Language Learning, vol. 11, edited by K. BardoviHarlig, C. Félix-Brasdefer, and A. S. Omar, 81-107. Honolulu: University of Hawai’i Press. —. 2007. “Student writing emails to faculty: An examination of epoliteness among native and non-native speakers of English.” Language Learning and Technology 11 (2): 59-81. Blum-Kulka, S., J. House, and G. Kasper. 1989. Cross-cultural Pragmatics: Requests and Apologies. Norwood: Ablex. Chen, E. 2001. “Making e-mail requests to professors: Taiwanese vs. American students” (presentation, American Association for Applied Linguistics Conference, St Louis, Missouri). Crystal, D., ed. 1997. The Cambridge encyclopedia of language (2nd ed.). Cambridge: Cambridge University Press. Economidou-Kogetsidis, M. 2011. “‘Please Answer Me as Soon as Possible:’ Pragmatic Failure in Non-native Speakers’ E-mail Requests to Faculty.” Journal of Pragmatics 43: 3193-3215. Ellis, R. 1992. “Learning to communicate in the classroom: A study of two language learners’ requests.” Studies in Second Language Acquisition 14: 1-23. Félix-Brasdefer, C., and M. Hasler-Barker. 2015. “Complimenting in Spanish in a Short-Term Study Abroad Context.” In “Pragmatic learning across contexts,” edited by E. Alcón-Soler, and L. Yates, special issue, System 48C. Félix-Brasdefer, J. C. 2012. “E-mail requests to faculty: E-politeness and internal modification.” In Interlanguage Request Modification, edited by M. Economidou-Kogetsidis, and H. Woodfield, 87-118. Amsterdam: Benjamins. Hassall, T. 2012. “Request modifiers by Australian learners of Indonesian.” In Interlanguage Request Modification, edited by M. Economidou-Kogetsidis, and H. Woodfield, (203-242). Amsterdam: Benjamins. Ishiara, H., and A. Cohen. 2010. Teaching and learning pragmatics. Where language and culture meet. London: Longman Applied Linguistics.

30

Chapter One

Jeon, E. H., and T. Kaya. 2006. “Effects of L2 instruction on interlanguage pragmatic development.” In Synthesizing research on language learning and teaching, edited by J. M. Norris, and L. Ortega, 165-211. Amsterdam: Benjamins. Matsumura, S. 2003. “Modelling the relationships among interlanguage pragmatic development, L2 proficiency and exposure to L2.” Applied Linguistics 24 (4): 465-491. Pan, P. C. 2012. “Interlanguage requests in institutional e-mail discourse: A study in Hong Kong.” In Interlanguage Request Modification, edited by M. Economidou-Kogetsidis, and H. Woodfield, 119-162 Amsterdam: Benjamins. Rose, K. R. 2005. “On the effects of instruction in second language pragmatics.” System 33: 385-399. Schauer, G. 2009. Interlanguage pragmatic development: The study abroad context. London: Continuum. Shively, R. L. 2010. “From the Virtual World to the Real World: A Model of Pragmatics Instruction for Study Abroad.” Foreign Language Annals 43 (1): 105-137. —. 2011. “L2 Pragmatic Development in Study Abroad: A Longitudinal Study of Spanish Service Encounters.” Journal of Pragmatics 43: 1818-1835. Taguchi, N. 2011a. “Teaching pragmatics: Trends and issues.” Annual Review of Applied Linguistics 31: 289-310. —. 2011b. “The effect of L2 proficiency and study-abroad experience in pragmatic comprehension.” Language learning 61: 904-939. —. 2013. “Production of routines in L2 English: Effects of proficiency and study-abroad experience.” System 41: 109-121. —. 2015. “Instructed pragmatics at a glance: where ILP studies are, were and should be going.” Language Teaching 48: 1-50. Trosborg, A. 1995. Interlanguage Pragmatics. Berlin: Mouton de Gruyter. Vilar-Beltrán, E. 2014. “Length of stay abroad: Effects of time on the speech act of requesting.” International Journal of English Studies 14: 79-96. Winkie, P., and C. Teng. 2010. “Using task-based pragmatics tutorials while studying abroad in China.” Intercultural Pragmatics 7 (2): 363399. Woodfield, H. 2012. “I think maybe I want to lend the notes from you: Development of request modification in graduate learners.” In Interlanguage request modification, edited by M. EconomidouKogetsidis, and H. Woodfield, 9-50. Amsterdam: Benjamins.

Teachers’ Perceptions of Email Requests

31

Woodfield, H., and M. Economidou-Kogetsidis. 2010. “I just need more time: A study of native and non-native requests to faculty for an extension.” Multilingua 29 (1): 77-118. Yates, L., ed. 2013. Teaching and learning pragmatics. TESL Canada Journal 30, Special issue 7.

Notes This study has been conducted within the framework of a research project funded by the Spanish Ministerio de Economía y Competitividad, co-funded by FEDER (FFI2012-38145). The data for the present study were collected thanks to the possibility that the author had to participate in the academic mobility program for senior researchers (PR2011-0486).

CHAPTER TWO “TEACHER! YOU NEED TO GIVE ME BACK MY HOMEWORK:” ASSESSING STUDENTS’ NEEDS FOR A PRAGMATICS CURRICULUM IN AN ACADEMIC ESL PROGRAM PATRICIA FRENZ-BELKIN

1. Introduction Instructors teaching ESL/EFL students often find their use of speech acts such as requests, apologies, and disagreements inappropriate, if not rude. Those with training in linguistics or TESOL may recognize in such utterances instances of pragmatic failure, whereas instructors of other disciplines may feel annoyed and form a negative impression of the students. Failure to produce pragmatically appropriate utterances in conversation is not limited to speech acts and occurs in a wide variety of speech contexts. Native speakers implicitly know how to manage a conversation, how turn-taking mechanisms work – for example, when it is appropriate to start a turn – and when backchanneling is called for to show that the listener is attending to what the speaker is saying. Language learners, on the other hand, do not automatically learn the pragmatic rules of the target language as they acquire linguistic competence in that language, and often transfer these rules from their first language. Therefore, they require instruction in the pragmatic norms that native speakers have internalized as part of the socialization into their culture. This is all the more important because pragmatic failure, unlike grammatical incorrectness, reflects negatively on the speakers’ character, or worse, on the ethnic group to which they belong. In turn, this may produce or re-produce ethnic stereotypes. One area of pragmatics in which cross-cultural miscommunication abound, is that of politeness. Cultures

34

Chapter Two

differ considerably in the ways in which politeness is expressed. Thus, requests may be issued directly and with a minimum of politeness markers (e.g., modal verbs or politeness formulae) in one culture, whereas in other cultures, speakers put forward more indirect requests couched in politeness markers. The focus of the present study is to assess in which ways ESL students at different levels of proficiency produce appropriate requests, refusals, and disagreements in different social encounters. Many of the ESL students who participated in this study were “1.5 generation students” (Rumbaut and Ima 1988) enrolled in U.S. colleges. They were predominantly Spanish-speaking immigrants, the majority of whom had come to the United States from the Dominican Republic with hopes for better educational and career opportunities than their home country could offer them. Like many immigrants to the United States, these too have minimum wage service jobs, typically working as cashiers in supermarkets, as home care providers, nannies or in other minimum wage service jobs. In other words, they come into daily contact with native speakers of English in situations where opportunities for pragmatically inappropriate language use and miscommunications abound. This study seeks to explore how these students, who have received all or at least some of their formal education in the United States, perform in terms of their pragmatic competence, especially within the context of speech acts that require a combination of mitigating politeness strategies.

2. Background The past three decades have witnessed researchers’ increased interest in the teaching of pragmatics to language learners, based on the recognition that exposure to the target language is mostly insufficient for learners to pick up how native speakers use language appropriately in different speech situations. Bardovi-Harlig and Mahan-Taylor (2003, 3) state that “left to their own devices with respect to contact with the target language in and out of the classroom, the majority of learners apparently do not acquire the pragmatics of the target language on their own.” A large number of earlier studies have investigated learners’ production and metapragmatic assessment of speech acts such as requests (Blum-Kulka and Olshtain 1986; Takahashi and DuFon, 1989; Rose 1998), refusals (Takahashi and Beebe 1987; Houck and Gass 1996) and apologies (Olshtain and Blum-Kulka 1985). Language learners’ pragmatic competence, which includes pragmalinguistic as well as sociopragmatic competence,1 has been examined in terms of first language (L1) transfer,

“Teacher! You Need to Give Me Back my Homework”

35

and learners’ overall proficiency in the target language and grammatical competence in particular. There is evidence that the production of speech acts may vary, at least partially, depending on the learner’s first language and culture; that is, failure to produce an appropriate speech act in the target language may be due to transfer of L1 norms. For example, Olshtain (1983), in her study of apology performance by English and Russian learners of Hebrew as a second language, found that the cultures under investigation differed in their judgement of the severity of an action or the circumstances related to certain behaviors which may give rise to the need to apologize. Different cultural norms may call for different types of apologies, which may also vary in the force of their illocution. Furthermore, Takahashi and Beebe (1987) not only found that Japanese learners of English transferred pragmatic norms from their L1 into their production of certain speech acts (e.g., apologies, refusals), they also noticed that pragmatic competence did not necessarily increase with linguistic competence, which may place more advanced learners at a disadvantage compared to less proficient speakers. Native speakers assume that speakers who exhibit linguistic fluency will also produce pragmatically felicitous utterances. If speakers fail to do so, it reflects negatively on their character rather than it being understood as a consequence of their lack of pragmatic competence in the target language. Other studies have found that learners at different proficiency levels produced speech acts that differed in terms of types and number of strategies in comparable contexts. Scarcella (1979) investigated the use of politeness strategies by beginning and advanced learners of English and found that the participants in her study seemed to acquire the pragmalinguistic rules before they acquired the rules for their appropriate use in social contexts. Based on her research, she concluded that learners’ pragmatic competence develops as their overall proficiency in the target language increases. In her study of three groups of Danish learners of English at different proficiency levels, Trosborg (1995) examined the acquisition of requests, complaints, and apologies. She found that learners with higher proficiency produced more native-like request strategies than those with lower proficiency. However, there were only slight differences among the three proficiency levels considered for apology and complaint strategies. This may suggest that some speech acts are acquired more easily than others. Bardovi-Harlig and Hartford (1993) conducted a longitudinal study that tracked the acquisition of suggestions and rejections by adult learners of English in academic counselling sessions; they found that, although

36

Chapter Two

learners’ overall proficiency increased, they were not able to mitigate suggestions and rejections appropriately. The researchers suggested that this finding may have reflected the learners’ opportunities (or lack thereof) for feedback or input in this particular institutional setting. The lack of opportunities for pragmatic learning in a formal classroom setting has been noted in other studies (e.g., Kasper 1989; Porter 1986) as well; for example, classroom discourse typically has been found to contain a narrower range of speech acts and a less frequent use of politeness markers than talk outside of the classroom. These findings are of particular interest to the present study, which focuses on the production of requests, disagreements and refusals by English language learners at different proficiency levels – especially their use of politeness markers – in a college setting. Since the use of politeness strategies by English language learners is central to this investigation, the literature reviewed below addresses this particular area of pragmatics. In terms of politeness in general, Brown and Levinson (1987) introduce the notion of face, which refers to the public self-image that all humans wish to maintain. They break down the concept of face into positive face and negative face; positive face includes people’s needs to be accepted, appreciated and included in a group, whereas negative face concerns individuals’ needs to act freely and without imposition. In order to soften potentially face-threatening acts, speakers use politeness strategies, including positive and negative politeness, to assure that hearers’ and speakers’ face needs are recognized. For example, when a hearer’s negative face is threatened, the speaker is expected to assure the hearer that his/her freedom of action and freedom from imposition are recognized and respected; strategies for negative politeness include apologizing, using hedges, and expressing deference. If an act threatens the hearer’s positive face, the speaker is assumed to treat the hearer as a friend, as a person whose character is liked, by using expressions of solidarity, informality, and familiarity (LoCastro 2012). Considering that speech acts may threaten the face of the hearer, the speaker or both, native speakers use politeness strategies to mitigate such threats; non-native speakers, on the other hand, often fail to produce the appropriate politeness markers, which can lead to serious misunderstandings or communication breakdown. Brown and Levinson (1987) claim that politeness is a universal phenomenon, implying that speakers from different linguistic and cultural backgrounds act based on the politeness model partially described by them. However, more recent research, especially by Asian researchers (e.g., Ide 2005, Matsumoto 1988, Cook 2006), has suggested that cultural

“Teacher! You Need to Give Me Back my Homework”

37

beliefs and practices as well as the communicative context may play an important role in the differential uses of politeness strategies. However, according to LoCastro (2012), “Brown and Levinson’s model of politeness has been and continues to be arguably the most influential approach to this form of human behavior,” particularly within a Western context. In addition to the linguistic forms and the socially appropriate usage of such forms, language learners need to be aware of the underlying preference structure of conversation. Research in conversation analysis (e.g., Pomerantz 1984) has recognized the importance of preference structure in the production of appropriate responses to certain utterances such as assessments, invitations, and requests. For example, in response to an assessment, agreement is the preferred or expected response, whereas disagreement is a dispreferred or unexpected move. In mundane conversation, speakers of American English try to avoid direct, unmitigated disagreement; instead, they use prefaces such as “well…,” token agreements (“yes, but…”); appreciation before disagreement (“What you are saying is really important…”); an account of reasons for their disagreement; and mitigated forms (LoCastro 2012). An expression of unmitigated disagreement would be considered impolite by speakers of American English; it would also be perceived as a face-threatening act.

3. Methodology 3.1 Setting and participants This research was conducted at a 2-year, bi-lingual (Spanish/English) community college in the South Bronx, New York, which is one of the poorest urban congressional districts in the United States. The college offers credit-bearing courses within an ESL program that offers a contentbased, academic ESL curriculum, which is comprised of four levels (beginners to high intermediate). The purpose of the program is to provide students with the linguistic, communicative, and academic skills as well as the core general education competencies that will enable them to succeed in college-level content courses related to their chosen majors. To that end, students must pass two high-stakes exit exams in English reading and writing, and therefore the curriculum focuses mainly on academic reading and writing skills. When the college was founded in the late 1960s, its mission was to provide high quality education to the growing Latino population in the South Bronx, New York. Today, the ESL student population has become more diverse, including students from West African countries, Bangladesh, Eastern Europe, and Southeast Asia.

38

Chapter Two

However, the majority of the English Language Learners (ELLs) are still immigrants from Latin America and the Spanish-speaking Caribbean, especially the Dominican Republic. For this study, one hundred students in the ESL Program completed a two-part questionnaire. These students were enrolled in different levels of the ESL program: 13 students attended the lowest level; 15 were enrolled in the intermediate level; 52 students were enrolled in the highest level. The students’ ages ranged from 18 to 40, with most students falling within the 20 to 25 year age range. They came from diverse linguistic backgrounds; however, the majority reported Spanish as their first language. Most of the participants had some formal English instruction either in the United States or in their home countries prior to enrolling at the college.

3.2. Data collection In order to identify troublesome areas in politeness strategies used by ESL students at the college and to assess whether students could identify inappropriate utterances before being able to produce appropriate ones, a two-part questionnaire was developed (for the full questionnaire, see the Appendix). In addition to providing personal information (age, country in which they were raised, languages spoken, time of residency in the United States, English instruction prior to enrolling at the college), in the first part of the questionnaire, which comprised eight short scenarios, the participants were asked to provide an opening move to requests and disagreements, as well as a refusal in situations of varying degrees of formality and increasing degrees of imposition. For example, one item asked: “You ask your professor to write a letter of recommendation for a scholarship. What do you say to her/him?” An appropriate request would require negative politeness strategies such as an apology, an expression of deference, and hedges because of the high degree of imposition of the request. An example of a less formal request with a low degree of imposition is: “You want the lunch special at the cafeteria. What do you say to the person behind the counter?” In the second part of the questionnaire, which was comprised of five dialogues, the students had to assess given utterances in terms of their appropriateness in speech situations that varied in terms of formality (responding to a friend, a classmate, strangers, a professor). For example, one item asked participants to rate the appropriateness of the following refusal:

“Teacher! You Need to Give Me Back my Homework”

39

Carla is talking to her classmate Maria. Carla: I’m glad you’re here. I wanted to invite you to my birthday party on Sunday. Maria: Oh, I can’t make it on Sunday.

This part of the questionnaire was meant to evaluate whether respondents noticed which of the utterances were inappropriate, thus displaying a certain degree of meta-pragmatic awareness. If the respondents identified a response as inappropriate, they were asked to provide a more appropriate response. All the speech events presented in the questionnaire reflected everyday situations that the students were likely to encounter on a regular basis, such as asking a professor about their homework or their grades, disagreeing with a classmate, or asking for directions. Choosing situations that, for the most part, reflected college life further ensured that the respondents were at least somewhat familiar with these speech events.

3.3. Data analysis In order to compensate for the imbalance in the numbers of respondents from different ESL levels determined by the fact that almost three quarters of the students were enrolled in the highest level of ESL, only twenty of the questionnaires from this group were considered in the primary data analysis, which reduced the total number of questionnaires to 48. This was done by eliminating one whole course. However, biographical data regarding years of residence in the U.S. and prior English instruction from the 52 questionnaires that were excluded in the primary analysis were considered in order to establish whether certain trends indicated by the findings were supported by the data. The data analysis comprised three steps. First, the students’ responses were evaluated in terms of their approximation to target language pragmatic norms by using three criteria, namely, appropriate forms of address with the interlocutor, mitigators, and politeness formulae. Responses were categorized as felicitous when all three criteria were present; they were classified as somewhat felicitous when they contained two of the criteria; they were labeled non-felicitous when they only included one or none of the criteria. For the second part of the questionnaire, the category noticed was added to label responses in which participants indicated that they considered a response inappropriate but could not provide an appropriate alternative form. After all responses had been categorized, a score for each participant (with the highest score of 12 for felicitous responses) in each ESL level was calculated, as well as the percentages of responses in each category,

40

Chapter Two

so as to establish which speech events appeared to be the most and the least troublesome for the respondents. Finally, the students’ scores were correlated with current ESL level, first language, residency in the U.S., and English instruction prior to enrolling at the college in order to identify possible salient relationships. From among the personal data provided by the respondents, students’ ESL level and time of residency in the U.S. were considered in the analysis to determine whether proficiency in and/or exposure to the target language had an impact on their responses. In addition, the first language(s) of the respondents were included in the analysis because the majority of the 1.5 generation students at the college are from Spanish-speaking countries, and it was expected that these students would exhibit the most difficulties due to their incomplete literacy development in their first language. Since the number of participants was too small to yield statistically significant results, the data from the questionnaires were listed in tables and color-coded in terms of felicitous, somewhat felicitous, non-felicitous answers as well as noticing.

4. Findings In this section, the findings of the data analysis will be presented starting with students’ responses to the items in the questionnaire (Tables 1 to 3) and then moving to the students’ scores in relationship to their ESL level and biographical data (Tables 4 to 6). Felicitous answers were entered as f, somewhat felicitous answers as sf, and non-felicitous answers as nf. In addition, for the second part of the questionnaire, a category n for noticing was used to indicate that a respondent had identified an utterance as inappropriate but was not able to provide an appropriate one. An invalid response was tagged as n/a (not applicable), which means that the respondent either did not provide an answer at all or submitted an answer that could not be considered in the analysis. Table 1 shows that students in the lower ESL level provided the most infelicitous responses to items e, and h in the first part of the questionnaire and to item D in the second part of the questionnaire. Item e in the first part asked respondents to provide a request asking their professor for a grade change, which is a severe face-threatening act because of the power relationship between interlocutors as well as the high level of imposition that the request poses. For item h the students were asked to place an order for the lunch special at the school cafeteria, which does not involve any serious face threat or imposition. Item D in the second part of the

“Teacher! You Need to Give Me Back my Homework”

41

questionnaire asked respondents to assess the appropriateness of the following request: Patricia doesn’t understand what the professor is saying about subject-verb agreement. Patricia: “Can you repeat what you were saying? I didn’t understand.”

In addition, if they deemed the utterance to be inappropriate, they were asked to provide an initial turn that would conform to the target language pragmatics norms in the given situation. TABLE 1. Assessment of students’ responses to individual questionnaire items (lower level) Item a b c d e f g h A B C D

f 0 0 0 4 (27%) 0 0 3 (20%) 0 1 (7%) 1 (7%) 1 (7%) 1 (7%)

sf 4 (27%) 0 5 (33%) 2 (13%) 1 (7%) 5 (33%) 5 (33%) 2 (13%) 1 (7%) 4 (27%) 1 (7%) 1 (7%)

nf 8 (53%) 13 (65%) 8 (53%) 7 (47%) 10 (67%) 6 (40%) 6 (40%) 10 (67%) 9 (60%) 9 (60%) 8 (53%) 13 (87%)

n

4 (27%) 1 (7%) 0 0

n/a 3 2 2 2 4 3 1 3 0 0 1 0

The items that seem to have been the least troublesome for the respondents were items d and g in the first part of the questionnaire. Item d asked: “You are discussing an important topic in class, and you disagree with what one of your classmates says. What do you say to her or him?” For item g the students were asked: “Your English-speaking employer asks you to work extra hours on the weekend. You cannot work on weekends because of family responsibilities. What do you say to your boss?” Table 2 shows that the group of intermediate students provided the most infelicitous answers to items b and e in the first part of the questionnaire and to item D in the second part.

Chapter Two

42

TABLE 2. Assessment of students’ responses to individual questionnaire items (intermediate level) Item a b c d e f g h A B C D

f 2 (15%) 2 (15%) 4 (31%) 4 (31%) 1 (8%) 3 (23%) 3 (20%) 3 (23%) 2 (15%) 2 (15%) 2 (15%) 0

sf 4 (31%) 2 (15%) 3 (23%) 3 (23%) 4 (31%) 4 (31%) 5 (33%) 3 (23%) 1 (8%) 3 (23%) 3 (23%) 4 (31%)

nf 6 (46%) 8 (61%) 6 (46%) 6 (46%) 7 (54%) 6 (40%) 5 (38%) 6 (46%) 4 (31%) 4 (31%) 3 (23%) 7 (54%)

n

6 (46%) 4 (31%) 5 (38%) 2 (15%)

n/a 1 1 0 0 1 3 1 1 0 0 0 0

The difference between the intermediate group and the lower-level students is that the intermediate students had difficulties with item b, which asked students to ask their professor about a homework assignment that the professor had not returned yet. Again, this is a face-threatening act that requires negative politeness strategies to mediate the possible implication that the professor has not done her job. The least troublesome items for this group were item g (refusing an employer’s request to work extra hours), item d (disagreeing with a classmate) and C. This last item asked respondents to assess Sarah’s response in terms of its appropriateness: Michael and Sarah are talking about a movie that they have both seen. Michael: I really loved that film. It was so moving. Sarah: I think it was rather stupid.

Table 3 shows that the students of the most advanced level provided the most infelicitous responses to items e, C, and D, which reveals that items e and D remained the most difficult across the three ESL levels. Item C, which posed difficulties for this group, is the movie scenario described above. The table also shows that the least troublesome items for the advanced students were items d and g, which is congruent with the findings of the other two groups.

“Teacher! You Need to Give Me Back my Homework”

43

TABLE 3. Assessment of students’ responses to individual questionnaire items (advanced level) Item a b c d e f g h A B C D

f 3 (15%) 2 (10%) 3 (15%) 9 (45%) 1 (5%) 7 (35%) 8 (40%) 6 (30%) 6 (30%) 6 (30%) 6 (30%) 4 (20%)

sf 7 (35%) 11 (55%) 11 (55%) 5 (25%) 7 (35%) 8 (40%) 8 (40%) 9 (45%) 7 (35%) 7 (35%) 2 (10%) 5 (25%)

nf 7 (35%) 7 (35%) 6 (30%) 4 (20%) 8 (40%) 2 (10%) 3 (15%) 4 (20%) 5 (25%) 3 (15%) 10 (50%) 11 (55%)

n

2 (10%) 4 (20%) 2 (10%) 0

n/a 3 0 0 1 4 3 1 1 0 0 0 0

The above findings suggest that items e and D posed the greatest difficulties irrespective of the respondents’ ESL level. Both items describe scenarios in which the respondents had to either assess (item D) or provide a request (item e) addressed to a professor; that is, they present situations in which a power distance exists between the interlocutors, and which involve a possible face threat. It should be noted that even though these items had the lowest scores in each level, the scores did improve, albeit only slightly, from one level to the next. For example, for item e, on the lowest ESL level, none of the respondents provided a felicitous request, and only one proffered a somewhat felicitous request. On the intermediate level, there was one felicitous request and four somewhat felicitous ones for the same item. In comparison, on the advanced level, there was again only one felicitous request in response to item e and seven somewhat felicitous requests. The findings regarding the least troublesome items also show consistency across levels; respondents across all levels did not seem to have problems with item g, which asked them to refuse their employer’s request to work on weekends or with item d, which asked students to disagree with a fellow student’s opinion in a class discussion. Tables 4 to 6 show students’ scores in relation to their age, first language, time of residency in the United States, and the kind of English instruction that they had received prior to enrolling at the college. In the tables, CLIP refers to the CUNY (The City University of New York) Language Immersion Program, a 1-year ESL program that students attend daily for 5 hours. GED (General Education Development) refers to a test in five subjects that students without a high school diploma can take to

Chapter Two

44

obtain a Certificate of High School Equivalency (the test is also available in Spanish). BCC (Bronx Community College) and KCC (Kingsborough Community College) are colleges within the CUNY system. HS is short for High School, and DR stands for Dominican Republic. TABLE 4. Students’ scores and biographical information for the lowest ESL level group Score

#of students

0

2

1

5

3

2

5

5

Age 47 22 23 24 19 21 22 n/a 21 32 21 29 18 20

L1 Spanish Spanish Spanish Spanish Spanish Spanish Spanish Spanish Spanish Spanish Spanish Spanish Spanish Spanish

Residency US 1.10 years 3 years 5 years 6 years 10years 2.6 years 2 years 5 years 3 years 5 years 3 years 2 years 2 years 3 years

English CLIP none Basic GED HS/US CLIP BCC CLIP none n/a CLIP Basic HS/US BCC

TABLE 5. Students’ scores and biographical information for the intermediate ESL level group Score

#of students

1

2

2

1

3

2

5

1

6

2

7

2

9 10 11

1 1 1

Age 18 21 21 20 32 20 20 30 31 22 21 19 27

L1 Spanish Spanish Spanish Spanish Spanish Spanish Spanish Korean Spanish Chinese Bengali Spanish French

Residency US 6 years 1.7 years 4 years 4 years 10 years 1.9 years 2 years 3 years 6 years 5 years 2.7 years 10 months 3 years

English HS/US HS/DR HS/US HS/US none Priv/DR ESL/DR ESL/US ESL GED HS/Bang HS/Ecuador KCC

“Teacher! You Need to Give Me Back my Homework”

45

TABLE 6. Students’ scores and biographical information for the advanced ESL level group Score 1

#of students 1

4

2

5

3

6

1

7

2

8

3

9

3

10

4

11

1

Age 35 23 25 28 18 20 23 21 26 21 23 21 23 28 35 25 30 35 40 40

L1 French/Ewe Spanish Spanish Spanish Spanish Spanish Spanish Spanish Spanish Twi Spanish Spanish Spanish Spanish French Spanish Bangla Spanish Italian Spanish

Residency US 5 years 8 years 10 years 14 years 4 years 1.5 years 7 years 5 years 6 years 5 years 4 years 3 years 3 years 10 years 2 years 4 years 13 years 18 years 19 years 10 years

English none HS/US HS/US HS/US HS/US HS/DR HS/US GED HS/DR GED ESL HS/US HS/DR HS/US n/a ESL/DR HS/Bang HS/DR n/a n/a

Tables 4, 5, and 6 show that the age of the respondents did not seem to have an impact on their scores, neither did the length of residency in the United States. However, there seems to be a relationship between respondents’ first language and their instruction prior to enrolling at the college and their scores, at least at the intermediate and higher levels. As for the intermediate level (Table 3), among the seven highest scoring students (scores between 6 and 11), four were not native speakers of Spanish, who otherwise constituted the majority of the students in this course (9 out of 13). These four students reported Korean, Bengali, Chinese, and French as their first languages. Moreover, one of the three high-scoring Spanish speakers came from Ecuador, whereas all the other Spanish speakers were immigrants from the Dominican Republic. In terms of the students’ English instruction prior to enrolling at the college, none of the seven students with the highest score in their level had attended U.S. high schools, whereas three students with the lowest scores (scores of 1, 2, and 3) had attended high schools in the United States. The student with the highest score (a score of 11) had attended another

46

Chapter Two

community college within The City University of New York, which was the most advanced level of prior instruction within the whole group. The students who attained a score of 9 and 10 both had learned English at high schools in their countries Bangladesh and Ecuador. As Table 4 shows, for the most advanced ESL level, a picture similar to that presented for the intermediate level emerges. Of the eight students with the highest scores (scores of 9, 10, and 11), three did not report Spanish as their first language; the reported first languages were French, Italian, and Bangladeshi. Except for two students, the remaining respondents on this level were native speakers of Spanish from the Dominican Republic. With regard to students’ prior English instruction, Table 4 shows that of the seven students with the lowest scores (scores of 1 to 6), five had attended high schools in the United States, with one student having attended the CUNY Language Immersion Program (CLIP) as well.

5. Discussion Regarding the questionnaire items that posed the most difficulty to students across the three ESL levels, one of these asked respondents to initiate an indirect request (a student wants to know if his/her professor has corrected their homework) and the other to produce a direct request (asking their professor for a better grade). Both speech situations constitute an imposition and a possible face-threatening act implying that the professor has not done her job properly, which requires that the speaker uses politeness markers to soften the threat. As Fraser (1980) stated, native speakers of English are expected to use mitigation markers when they wish to convey politeness toward the hearer. More specifically, the speech situations at hand require acts of negative politeness that assure hearers that the speakers recognize and respect their negative face wants; that is, the speakers are expected to convey to the hearers that they will not impose on the hearers’ freedom of action (Brown and Levinson 1987). This is especially important when there is an unequal power relationship between interlocutors, as was the case in the two scenarios at hand. Acts of negative politeness typically include some form of apology, expressions of deference, and hedges. Most of the respondents were unable to produce appropriate initiation moves to these requests. In the case of the missing homework, the majority of the students in the lower level chose a very direct, unmitigated approach; e.g., “Professor, why you don’t give back my assignment?,” which sounds more like an accusation than a request, and, thus, aggravates

“Teacher! You Need to Give Me Back my Homework”

47

the threat to the hearer’s negative face. The students in the highest ESL level typically used some kind of mitigator or politeness marker; however, most of their utterances still could not be labeled felicitous (e.g., “Professor, excuse me, do you remember my homework? Whenever is ready, thanks.”). With regard to the request for a grade change – a highly facethreatening act – most of the students across all levels failed to provide felicitous opening moves. There were some differences between the lower levels and the highest level in that the responses of the students in the highest level were more elaborate, which, in some cases, made their responses even more infelicitous. For example, whereas students from the lowest level often used an unmitigated approach, “Professor, I think that you should give me a better grade,” students from the highest level would typically say that they deserved a better grade and then give reasons as to why. In a few cases, the advanced students asked the professor to check her records again and explain the grade to them, which intensifies the imposition of the speech act. Based on these findings, it seems that greater fluency in the target language did not improve students’ pragmatic competence in certain areas such as face-threatening acts involving a power distance between speakers. This may put more fluent speakers at a disadvantage compared to their less fluent peers in communications with native speakers. Wolfson (1989) stated: “Since linguistic competence is an aspect of communicative competence, people who have one are expected to have the other and are therefore held responsible for sociolinguistic violations in a way in which those with less ability to communicate would not be.” (Wolfson 1989, 149) In other words, native speakers, whose knowledge of the pragmalinguistic and sociopragmatic rules of their language is implicit, may not even consider the infelicitous utterances of a non-native speaker who otherwise shows linguistic fluency to be the result of his or her not being aware of these pragmatic conventions. Interestingly, students across ESL levels seemed to have difficulties with the item that asked them to order the lunch special at the school cafeteria, a request which could be considered low-stakes and low on the face-threat spectrum. Students in the lower levels typically used want (“I want the lunch special”) or can (“Can I have the lunch special?”) with the occasional please tagged on to their requests. This could be explained by the fact that at this point in their language development, they have not yet mastered the appropriate use of modifiers such as could or may to convey politeness. Among the more advanced students, can was still used in some

48

Chapter Two

of the requests; however, students typically used a politeness marker such as please to mitigate their utterances. Very few of the requests on all levels included a polite summons such as excuse me. If a summons was expressed at all, the students tended to use informal addresses such as hey, hi, you guys. This could be an indication of L1 transfer in the case of the Spanish speakers because in many Spanish-speaking countries summonses in service encounters are more informal than they are in the United States, catering to the addressee’s positive face wants. Research on politeness strategies in Spanish-speaking countries in Latin America and the Caribbean (e.g., Chodorowska-Pilch 1999; Placencia and Garcia 2007) has suggested that speakers prefer positive face politeness strategies such as expressions of solidarity and affiliation over negative ones. The infelicitous responses of Spanish speakers in this study could, therefore, be attributed to transfer of politeness preferences from their culture, which are at odds with those of speakers of American English, who typically prefer negative politeness strategies in encounters in which interlocutors are not close. Lo Castro (2012) states that the ways in which service personnel such as waiters are addressed or summoned differ greatly from one culture to another. For example, in Mexico, waiters are called “joven” (young person), which applies to people of all ages. Such a summons would be perceived as overly familiar or rude in other countries. The data show that the respondents had little trouble in providing felicitous or somewhat felicitous refusals to their employer’s request to work extra hours on a weekend (item g). They seemed to have been aware that such a refusal needed to include an apology, an explanation and/or some form of redress (e.g., the offer to work extra hours on another day). The following are three refusals taken from the data provided by students from the three ESL levels: Lowest level: “I’m so sorry but I can’t work extra hours because I have responsibilities with my family.” Intermediate level: “I would like to work extra hours, Ms., but I have a lot of responsibilities.” Advanced level: “I’m sorry, I would like to help you but I can’t work on weekends.”

As these examples show, the differences in refusals among levels are not pronounced. As for reasons why this item proved to be less difficult than others discussed above, it is possible that this item presented a scenario with which the students, most of whom work while attending college, are very familiar. They may have had opportunities to hone their

“Teacher! You Need to Give Me Back my Homework”

49

pragmatics skills in work-related situations; for example, by observing native speaker interactions. On the other hand, the academic environment at the college is new to most of them; and, therefore they may be unfamiliar with the pragmatic norms that govern interactions between students and faculty. This would explain why the most difficult items presented them with situations in which they had to address requests to a professor. The following is another item (item d) that yielded many felicitous or somewhat felicitous responses: “You are discussing an important topic in class, and you disagree with what one of your classmates says. What do you say to him/her?”

According to the preference structure underlying conversations of American English speakers, disagreement is a dispreferred response which has to be mitigated in order not to offend the recipient. Most of the respondents across ESL levels showed an awareness of this conversational norm. The following is the response of a student from the intermediate level, which is representative of many of the responses from students across levels: “I understand your idea, but I am not agree with it.” The respondent shows an appreciation of the prior speaker’s statement or assessment before disagreeing. One reason why this item proved to be relatively easy for the respondents may be that at the college, instructors stress respect for others’ opinions in classroom discussions and model appropriate ways of responding to what students’ are saying. Here, again, it may be familiarity with a specific scenario that helped students to provide a felicitous or somewhat felicitous response. Interestingly, students in the lower and the advanced levels scored low on item C, which asked them to assess the appropriateness of a response to a friend’s evaluation of a movie. The given response was an unmitigated disagreement, which at least half of the respondents (53% in the lowest level and 50% in the highest level) deemed appropriate. These low scores may indicate that the students did not extend the norms that they have learned in the classroom, in this case that disagreement has to be mitigated, to conversations outside the classroom. These findings support what other researchers (e.g., Bardovi-Harlig and Hartford 1993, Kasper 1989; Porter 1986) have suggested with regard to the limited input that classroom discourse provides in terms of the range of speech acts and the lower frequency of use of politeness markers compared to conversations outside of the academic setting. Finally, the fact that of the twelve students who graduated from U.S. high schools, nine were among those students with the lowest scores in

50

Chapter Two

their levels, adds to what research findings have suggested for some time now; namely, that students who have moved to the United States before they became fully literate in their first language have difficulties writing essays and reading academic texts once they are in college (Oudenhoven 2006). The present study revealed that these students also seemed to have had difficulties in producing pragmatically felicitous utterances and in noticing utterances that were inappropriate.

6. Concluding remarks The findings of the present study confirm what previous research has suggested with regard to the relationship between formal instruction in academic reading and writing and the acquisition of pragmatic norms of the target language; namely, that formal instruction in these areas does not necessarily lead to the acquisition of pragmatics norms based on which native speakers of the language make linguistic choices. This research found that even students who received schooling in the target culture and language do not seem to have a significant advantage compared to students who studied the target language in a foreign language classroom in their home countries. Furthermore, the time that respondents have resided in the United States did not seem to have an impact on their ability to produce appropriate utterances or to correctly evaluate the appropriateness of given utterances. It appears, therefore, that students such as the participants in this study need explicit instruction in the pragma-linguistic and socio-pragmatic rules of American English in order to acquire a native-like communicative competence. The findings revealed that the particular group of students under investigation failed to provide felicitous responses in face-threatening situations within an academic setting, in which a power distance prevailed between interlocutors. Lack of familiarity with this type of encounter may be one explanation for this failure to produce responses that approximate the pragmatic norms of the target language; another may be that classroom discourse does not provide sufficient input for students to acquire these pragmatics norms without direct instruction. On the other hand, familiarity with a scenario seemed to have helped students to supply more felicitous responses, as was shown in the responses to the item that asked students to refuse an employer’s request to work on weekends. There was also evidence in the data that in some situations (e.g., the service encounter scenario), students transferred politeness preferences from their own cultures, which led to non-felicitous responses.

“Teacher! You Need to Give Me Back my Homework”

51

Based on the findings of this study, a course on inter-cultural communication is being developed within the college’s ESL department, which also includes a growing linguistics program. This course is intended to educate non-native speakers of English as well as students whose native language is English in the pragma-linguistic and socio-pragmatics norms to which native speakers adhere intuitively. The curriculum focuses on cultural values and beliefs underlying these norms, cross-cultural speech acts, and politeness among other topics. It is hoped that such a course will heighten the awareness of native and non-native speakers alike of the fact that pragmatic norms are not universal but differ from one culture/language to another, and that failure to produce linguistically appropriate expressions that fit the social context is not a reflection of the speaker’s character or culture.

References Bardovi-Harlig, K., and B. Hartford. 1993. “Learning the Rules of Academic Talk.” Studies in Second Language Acquisition 15: 279-304. Bardovi-Harlig, K., and R. Mahan-Taylor. 2003. Teaching Pragmatics. Washington, DC: U.S. Department of State. Blum-Kulka, S., and E. Olshtain. 1986. “Too Many Words: Length of Utterance and Pragmatic Failure.” Studies in Second Language Acquisition 8: 47-61. Brown, P., and S. Levinson. 1987. “Politeness: Some Universals in Language Usage.” Cambridge: Cambridge University Press. Chodorowska-Pilch, M. 1999. “On the Polite Use of Vamos in Peninsular Spanish.” Pragmatics 9 (3): 343-355. Cook, H. M. 2006. “Japanese Politeness as an Interactional Achievement: Academic Consultation Sessions in Japanese Universities.” Multilingua 25 (3): 269-291. Fraser, B. 1980. “Conversational Mitigation.” Journal of Pragmatics 4 (4): 341-350. Houck, N., and S. M. Gass. 1996. “Non-Native Refusal: A Methodological Perspective.” In Speech Acts across Cultures, edited by S. M. Gass, and J. Neu, 45-64. Berlin: Mouton de Gruyter. Ide, S. 2005. “How and Why Honorifics Can Dignity and Elegance: The Indexicality and Reflexivity in Linguistic Rituals.” In Broadening the Horizon of Linguistic Politeness, edited by R. T. Lakoff, and S. Ide, 45-64. Philadelphia: Benjamins.

52

Chapter Two

Kasper, G. 1989. “Interactive Procedures in Interlanguage Discourse.” In Contrastive Pragmatics, edited by W. Oleksy, 189-229. Amsterdam: Benjamins. LoCastro, V. 2012. Pragmatics for Language Educators: A Sociolinguistic Perspective. New York: Routledge. Olshtain, E. 1983. “Sociocultural Competence and Language Transfer: The Case of Apology.” In Language Transfer in Language Learning, edited by S. Gass, and L. Selinker, 232-249. Rowley, MA: Newbury House. Olshtain, E., and S. Blum-Kulka. 1985. “Degree of Approximation: NonNative Reactions to Native Speech Act Behavior.” In Input in Second Language Acquisition, edited by S. M. Gass, and C. Madden, 303-325. Rowley, MA: Newbury House. Oudenhoven, E. D. 2006. “Caught in the Middle: Generation 1.5 Latino Students and English Language Learning at a Community College.” PhD dissertation, Loyola University Chicago 2006. Digital Dissertations, AAT 3212980. Placencia, M. E., and C. Garcia. 2007. “Research on Politeness in the Spanish-Speaking World.” Mahwah, NJ: Lawrence Erlbaum. Pomerantz, A. 1984. “Agreeing and Disagreeing with Assessments: Some Features of Preferred/Dispreferred Turn Shapes.” In Structures of Social Action: Studies in Conversation Analysis, edited by J. Maxwell Atkinson, and J. Heritage, 57-101. Cambridge, UK: Cambridge University Press. Porter, P. A. 1986. “How Learners Talk to Each Other: Input and Interaction in Task-Centered Discussions.” In Talking to Learn: Conversation in Second Language Acquisition, edited by R. R. Day, 200-222. Rowley, MA: Newbury House. Rumbaut, R. G., and K. Ima. 1988. The Adaptation of Southeast Asian Refugee Youth: A Comparative Study. Final Report to the Office of Resettlement. Washington, DC: U.S. Office of Refugee Resettlement. Scarcella, P. 1979. “On Speaking Politely in a Second Language.” In On TESOL ’79: The Learner in Focus, edited by C. Yorio, K. Perkins, and J. Schachter, 275-287. Washington, DC: TESOL Takahashi, T., and M. A. DuFon. 1989. “Cross-Linguistic Influence in Indirectness: The Case of English Directives Performed by Japanese Speakers.” http://eric.ed.gov/?id=ED370439. Takahashi, T., and L. M. Beebe. 1987. “The Development of Pragmatic Competence by Japanese Learners of English.” JALT Journal 8: 131155.

“Teacher! You Need to Give Me Back my Homework”

53

Takimoto, M. 2006. “The Effects of Explicit Feedback and Form-Meaning Processing on the Development of Pragmatic Proficiency in Consciousness-Raising Tasks.” System 34 (4): 601-614. Trosborg, A. 1987. “Apology Strategies in Natives/Non-Natives.” Journal of Pragmatics 11: 147-167. Wolfson, N. 1989. Perspectives: Sociolinguistics and TESOL. Cambridge, MA: Newbury House.

Appendix Questoinnaire 1. Personal Information (this information will be kept confidential) 1. Male__ Female__ 2. Age __ 3. Country where you were raised ______________________________________ 4. Language(s) you speak (starting with your first language) _________________ 5. Time that you have been living in the U.S. _____________________________ 6. English instruction before you came to Hostos _________________ 2. Speech Situations. What would you say? In the following, you will read short descriptions of everyday situations. You will be asked what you would say in the given situation. Please write down how you would normally respond. a) You are in class and want to get your professor’s attention to ask a question about your essay. What do you say to her/him? ____________________________ __________________________________________________________________ b) Your professor has not returned your homework assignment. You want to know what happened to it. What do you say to him/her? __________________________ __________________________________________________________________ c) You ask your professor to write a letter of recommendation for a scholarship. What do you say to her/him? __________________________________________ __________________________________________________________________ d) You are discussing an important topic in class, and you disagree with what one of your classmates says. What do you say to him/her? ______________________ __________________________________________________________________ e) You think your professor should have given you a better grade. You go to her/him after class. What do you say? ___________________________________ __________________________________________________________________

54

Chapter Two

f) You are looking for the Financial Aid Office but you seem to be lost. You approach another student to get directions. What do you say? __________________________________________________________________ __________________________________________________________________ g) Your English-speaking employer asks you to work extra hours on the weekend. You cannot work on weekends because of family responsibilities. What do you say to your boss? _______________________________________________________ __________________________________________________________________ h) You want the lunch special at the cafeteria. What do you say to the person behind the counter? __________________________________________________ __________________________________________________________________ 3. Appropriate/Inappropriate Responses In the following, you will read responses to a situation or to what someone said. You have to decide whether what is written in bold is an appropriate response or not. If you think it is not, please write down what you think would be an appropriate response. A. Carla is talking to her classmate Maria. Carla: I’m glad you’re here. I wanted to invite you to my birthday party on Saturday. Maria: Oh, I can’t make it on Saturday. Appropriate Not Appropriate __________________________________________________________________ B. Paola is in an unfamiliar neighborhood looking for a post office. She sees two women talking in the street. Paola: Can you tell me where the nearest post office is? The women look at her. Appropriate Not Appropriate __________________________________________________________________ C. Michael and Sarah are talking about a movie that they have both watched. Michael: I really loved that film. It was so moving. Sarah: I think it was rather stupid. Appropriate Not Appropriate __________________________________________________________________ D. Patricia doesn’t understand what the professor is saying about subject-verb agreement. Patricia: Can you repeat what you were saying? I didn’t understand. Appropriate Not Appropriate __________________________________________________________________

“Teacher! You Need to Give Me Back my Homework”

55

E. At the library, a student is talking to a librarian Librarian: How can I help you? Student: I need to know where the reference books are. Appropriate Not Appropriate __________________________________________________________________

Notes 1 Pragmalinguistic competence refers to a learner’s knowledge of which linguistic forms are used to convey a particular pragmatic purpose whereas sociopragmatic competence describes a learner’s knowledge of which linguistic forms are appropriate in different social conditions.

CHAPTER THREE PRAGMATIC COMPETENCE, LENGTH OF RESIDENCE, AMOUNT OF CONTACT AND INTENSITY OF INTERACTION1 ZOHREH R. ESLAMI AND SOOJIN AHN

1. Introduction A movement in interlanguage pragmatics research from comparative studies to either cross-sectional or longitudinal research has led to more acquisitionally oriented ILP studies, linking interlanguage pragmatics research more directly to the scope of second language acquisition (SLA) research (Bardovi-Harlig 1999). To this end, some cross-sectional studies and a few longitudinal studies have traced the development of language learners’ pragmatic competence (Bouton 1999; Garcia 2004; Schmidt 1983; Siegal 1994; Taguchi 2008b, 2008c). Such studies revealed that the development of pragmatic competence is very complex and varies greatly from individual to individual depending on different factors such as attitude, proficiency, learning context, and length of residence in the target community (Bardovi-Harlig and Bastos 2011; Kasper and Schmidt 1996; Taguchi 2011). Research shows that even advanced learners with extended length of residence in the target community fail to approximate native speaker’s performance in terms of effective realization of various speech acts (e.g., Bardovi-Harlig 2001; Eslami-Rasekh 2005; Félix-Brasdefer 2004; Rose 2005). A critical issue in SLA is whether increased contact with users of the target language and length of stay in the target community is related to greater improvement in L2 proficiency (see, e.g., Bardovi-Harlig 2013; Bardovi-Harlig 2001; Taguchi 2011). Research shows that variables like length of stay in the target community and quality (type of contact) and quantity of input should be taken into account when assessing L2 learners’ pragmatic development (Taguchi 2011). Moreover, the inconsistency of research findings regarding the role of length of residence on learners’

58

Chapter Three

pragmatic development merits more inquiry (e.g., Churchill and DuFon 2006; Félix-Brasdefer 2004). As stated by Dietrich, Klein, and Noyau (1995), duration of stay in the target community is not as important as amount and intensity of interaction. Similarly, Kasper and Rose (2002) concluded that length of residence is not a reliable predictor for explaining pragmatic development in L2. Existing studies suggest that a number of relevant factors, including length of stay in the target-language community, may explain the rate and success of pragmatic development in L2 (Bardovi-Harlig 2013). Additionally, actual experience during residence in the target community, and the intensity of interaction is another important variable to consider in L2 pragmatic development (BardoviHarlig and Bastos 2011). Even though a number of studies have investigated the positive relationship between target language contact and second-language proficiency, conflicting evidence suggests that L2 contact does not necessarily result in greater improvement in L2 proficiency (Day 1984; DeKeyser 1986; Yager 1998; Segalowitz and Freed 2004). Similarly, in relation to pragmatic development, second language setting has been found to enhance pragmatic awareness and pragmatic competence (Bardovi-Harlig and Dörnyei 1998; Matsumura 2003; Schauer 2006). However, it has also been shown that, even learners with extended length of residence in the target community fail to achieve successful pragmatic performance when there is inadequate opportunities for social contact with native speakers (Bella 2012, 2011; Bouton 1994). Speech acts are one of the most studied components of pragmatic competence. Compliments, the focus of the present study, is one of the most important speech acts, which occur frequently in everyday encounters. Giving appropriate compliments and responding to compliments appropriately are among the most important speech acts needed by language learners (Rose and Kwai-fun 2001). Compliments serve several functions and can be used to open conversations, maintain conversations and end conversations. The objective of the current study is to examine whether length of residence in the target community, amount of language contact, and type of language contact (interaction intensity) can explain Korean ESL learners’ performance of complimenting behavior. We begin with a discussion of the important related theoretical concepts involved and then present the method of the study. Following the methodology, the results of the study are presented and discussed. The final section includes the conclusions and the pedagogical implications of the study.

Competence, Residence, Amount of Contact and Intensity of Interaction

59

2. Theoretical Background 2.1. Compliments Compliments and compliment responses have been extensively studied during the last there decades. They have been studied from various perspectives and in many different languages (e.g., Barnlund and Akari 1985; Chen 1993; Golato 2005; Herbert 1991, Holmes 1988; Liu 1995; Lorenzo-Dus 2001; Maíz-Arévalo 2012; Manes 1983; Wolfson and Manes 1981). Ethnographic work of Wolfson and Manes (1980) on compliments in American English is one of the earliest studies on this topic. Their research revealed the highly formulaic nature of compliments. They also identified different functions that compliments perform. They can be treated as a social strategy employed to start or maintain relationship in everyday interactions between colleagues, neighbors, or close friends. Holmes (1988, 486) considers compliments as “positively affective speech acts directed to the addressee which serve to increase or consolidate the solidarity between the speaker and addressee.” Wolfson and Manes (1981) found that the most frequent topics of compliments fall into two major categories: those having to do with appearance/possessions, and those addressing ability/performance. Compliments have been studied from various perspectives, in many different cultures and languages, in different modalities (face to face and online), and for cross-cultural comparison (see Chen 2010 for a recent overview). Some recent studies have extended the examination of complimenting behavior to online settings and films (e.g., Das 2010; Placencia and Lower 2013; Maíz-Arévalo 2013) As stated above, compliments can serve different functions. Under certain conditions, compliments replace speech acts such as apologies, thanking, and greetings (Billmyer 1990). They can also be used to soften the effects of criticism or other face-threatening acts such as requests (Billmyer 1990). Spencer-Oatey (2000) also noted that compliments may function as face-threatening acts since expressing admiration for something belonging to the addressee may be taken as an indirect request for the admired object. If the hearer believes that a compliment is too personal and is not comfortable with the level of intimacy implied, then the compliment can have a different effect than the one intended. Performing such multifunctional speech act seems to be a complicated and difficult task for L2 learners to pick up without intense interaction with other language users and instruction (Félix-Brasdefer and Hasler-Barker 2012).

Chapter Three

60

2.2. Compliment responses Compliment responses are one of the widely studied speech acts in English and in other languages as well (e.g., Holmes 1995; Pomerantz 1978; Lorenzo-Dus 2001; Maíz-Arévalo 2012). They are studied in different languages, cross-linguistically, and in different situations (see Chen 2010 for a review). One early study focusing specifically on compliment responses is Pomerantz’s (1978) descriptive analysis of compliment responses in American English. Based on her data, Pomerantz posited that agreement/acceptance and disagreement/rejection were the predominant compliment response types in American English. An important issue to consider when responding to complements is to be able to gracefully accept a compliment without seeming to praise oneself (Herbert 1986). As suggested by Pomerantz (1978), two politeness maxims are in conflict during response to compliments. One is Leech’s agreement maxim (Leech 1989) requiring one to “agree with your conversational partner”, and the other is to “avoid self-praise” (Leech’s modesty maxim). Thus, recipients have two basic options: to agree with the complimenter and violate the modesty maxim, or to disagree and thus violate the agreement maxim. Manes (1983) also addressed the dilemma posed to receivers of compliments and offered her own set of strategies which enable speakers to accept but not necessarily agree with the compliment. Some strategies can be seen in (1) – (3) below: (1) (2) (3)

A: Good shot. B: Not very solid though. A: You’re a good rower. B: These are very easy to row. Very light. A: You’re looking good. B: Great. So are you.

The strategy Manes identified in (1) above is for the receiver to play down the compliment. In (2), the recipient denies credit. In (3), Manes adds the dimension of accepting compliments outright or with a joke. Language learners are usually taught that the only correct way to respond to a compliment is to accept it with a simple “Thank you.” As Herbert (1991) pointed out, according to both etiquette books and native speakers’ awareness of prescriptive norms, “thank you” is considered the most appropriate response to a compliment in the United States. While this response may be appropriate, studies show that “an unadorned ‘thanks’ may unintentionally limit or even end an interaction between status equals,

Competence, Residence, Amount of Contact and Intensity of Interaction

61

and deflecting compliments may serve to extend the interaction between interlocutors, which may lead to interlanguage development” (Billmyer, Jakar, and Lee 1989, 17). Wolfson (1989) agreed with this view. According to Wolfson, a native speaker of English would consider the compliment a strategy to lengthen the conversation. In this case, a simple “thanks” may unintentionally bring about the opposite outcome by blocking opportunities to extend the interaction. When this occurs between native and nonnative speakers, the interlanguage development of the nonnative speakers may be hindered (Félix-Brasdefer and Hasler-Barker 2012; Wolfson 1989). By being aware of the rules and patterns that condition the behavior of native speakers of the target language, learners would be able to more satisfactorily develop relationships with native speakers and acquire native pragmatic knowledge and skill. Due to the complexity of performing this speech act and conflicting conversational maxims at play, responding to compliments is a challenging task for L2 language learners. Thus, in order to develop native like competence in using this speech act, interactional encounters with other language users are required.

2.3 Length of residence Length of residence is construed as one of the variables that affect the development of language learners’ pragmatic competence (Bella 2012). Many studies have used length of stay in a target speech community as an indicator of L2 pragmatic acquisition (Han 2005). Researchers argue that language learners living in a target speech community have many opportunities to interact in the L2, which leads to the learners’ successful acquisition of pragmatic competence (Cohen and Shively 2007; FélixBrasdefer 2004; Han 2005; Siegal 1994; Schauer 2007). Olshtain and Blum-Kulka’s study (1985) showed that the amount of external modification used by L2 learners approximated community pragmatic norms after five to seven years of stay in the target language environment, and that such convergence correlated positively with duration of stay. Similarly, Blum-Kulka and Olshtain (1986) reported that the length of residence in the target community accounted for the target-like perception of directness and politeness by non-native speakers of Hebrew. Bouton’s (1994) findings indicated that as ESL students’ length of residence on a U.S. university campus increased, they gradually acquired the ability to understand the conversational implicature. Additionally, it appears that even a short length of residence in the SL context affects pragmatic competence. Churchill (2001) recorded a decrease in direct want

62

Chapter Three

statements in the English request realizations of his Japanese Foreign Language learners over a month in the target language context. Overall, these studies suggest that longer stays abroad yield greater L2 pragmatic attainments Despite these findings, many questions still remain about the validity of the assumption that living abroad provides an ideal context for language learning. We know, for example, that not all individuals who live abroad for an extended time make the same linguistic gains. Although the studies mentioned above provide additional examples of the relation between pragmatic development and learners’ length of residence in the target language community, one might wonder to what extent pragmatic comprehension and pragmatic ability are generally influenced by the intensity of nonnative speakers’ exposure and social contacts in the target language, as opposed to the quantitative measure of length of residence in the target language community. Klein, Dietrich, and Noyau (1995), for example, suggested that what matters is intensity, not length of interaction. Likewise, Matsumura (2001) examined changes of Japanese students’ sociocultural perceptions with respect to the speech act of offering advice during an eight-month period of study-abroad in Vancouver and found no association between these students’ pragmatic development and length of residence in the target speech community, and asserted that acquisition of pragmatic competence is not associated with the length of stay, because learners vary individually in the amount of interaction in L2 as well as opportunities to interact in the target culture. 2.3.1. Amount of interaction vs. intensity of interaction A widespread assumption in SLA is that the amount of use of the target language is one of the crucial variables in the successful acquisition of the target language. In other words, simply because learners reside in the target language community does not necessarily mean that they have the desire to interact with the target speech community. Thus, amount of contact with the second language plays a more important role than length of residence in accounting for learners’ linguistic and pragmatic performance. Furthermore, intensity of interaction may account for more of the learning process than duration of stay in the L2 speech community or amount of contact with the target language. Research has shown that learners with more frequent and diverse opportunities for interaction with native speakers outperform learners with more extended length of residence but fewer opportunities for interaction, (e.g., Bardovi-Harlig and Bastos 2011; Bella 2012). Bella’s (2011) study on invitation refusals by

Competence, Residence, Amount of Contact and Intensity of Interaction

63

L2 learners of Greek, for example, revealed that opportunities for interaction are much more critical than length of residence in the target community for the development of learners’ sociopragmatic competence with regard to this particular speech act. Indeed, researchers suggest that L2 learners may acquire more target-like pragmatic norms through extended interaction in the target community rather than extended length of residence. In the absence of some amount of interaction in the target language, learners may not have an opportunity to improve their pragmatic knowledge of English. Therefore, in addition to investigating the impact of length of residence in the target community on pragmatic development, the impact of amount of interaction and intensity of interaction with other language users on pragmatic development is worth exploring further. It is claimed that what really matters is in language development is “intensity of interaction” (Klein, Dietrich, and Noyau 1995, 277), not length of residence. Therefore, reservations are expressed by some researchers as to whether pragmatic ability is influenced by the quality of the learners’ exposure and social contacts rather than the quantity and length of residence (Kasper and Rose 2002). The positive relationship between pragmatic gain and the level of the interactivity expected from learners can be explained by “Interaction Hypothesis” (Long 1983). The Interaction Hypothesis, formulated by Long (1983), contends that language acquisition is strongly facilitated by conversational interaction in a target language because the learner is afforded chances to access comprehensible input, opportunities for output, and implicit correction in the form of conversational feedback through the process of negotiated meaning. The interaction hypothesis attends to the interactional modifications; that is, L2 learners notice gaps between their L2 output and the response they receive from an interlocutor’s negative feedback and modify their own language use in response. This, in turn, affects their language learning and development. Swain (1985) asserted that it is not input itself that is important to L2 language acquisition, but rather the interaction in which learners become aware of a problem and are able to implement corrections. Based on Swain’s (1985) interaction hypothesis, Marriot (1995) outlined a framework for studying the acquisition of sociolinguistic competence by examining Australian secondary students who participated in exchange programs in Japan. She examined how learners benefit more from “selfand other-correction” procedures in interactive situations in a Japanese homestay context. Cooperative interactants who surrounded the learners, such as host family members, teachers, friends, and even members of their exchange organization, probably contributed significantly to the

64

Chapter Three

development of these learners’ L2 pragmatic awareness. These findings suggest that exchange students cannot acquire Japanese addressee honorifics unless they receive – and utilize – corrective feedback, either from their interlocutors or as a part of some form of instruction (Marriott 1995, 218-219). In light of the above discussion, the present paper intends to provide additional insights to the role that length of residence in the target community and amount and type of interaction with other target language users might play on the pragmatic performance of Korean ESL learners. The specific research questions addressed in this paper are: 1) Do differences in learners’ length of residence contribute to differences in learners’ pragmatic competence? 2) Do differences in amount of contact contribute to differences in learners’ pragmatic competence? 3) Do differences in type of contact contribute to differences in learners’ pragmatic competence? In what follows we will first provide a description of the participants. Second, the instruments used to gather data on the participants’ L2 pragmatic competence and information on learners’ language contact are discussed. Finally, the results are presented and discussions provided.

3. Method 3.1. Participants The participants of this study were 50 Korean graduate students majoring in different academic fields at Texas A&M University in the United States. There were twenty-seven male and twenty-three female participants whose age ranged from twenty-three to thirty-eight. All learners had passed the TOEFL test with 550 or higher to meet the university requirements. The length of time participants had spent in the United States ranged from two years to eight years.

3.2. Pragmatic competence measures A written discourse completion test (DCT) was used to determine the pragmatic competence of Korean ESL learners in the speech acts of compliment and compliment responses.

Competence, Residence, Amount of Contact and Intensity of Interaction

65

The DCT (Appendix A) included four situations requiring both compliment and compliment responses. The participants were asked to write down in English how they would make a compliment in the specific situation provided and also how they would respond to the compliments in each situation. To provide a suitable context for eliciting natural compliment sequences, the investigator chose a context based on the findings of Wolfson and Manes’ (1981) study, which found that the most frequent topics of compliments have to do with appearance/possessions and with ability/performance. Additionally, the investigator designed scenarios that the participants in the present study would be familiar with and would experience regularly in their daily living. The scenarios took into account social relationship between the two interlocutors (i.e., interlocutors are equal social status, or the addressee is either higher status or lower status). Each of the situations in the DCT specified mutually acquainted interlocutors, because much of the research has indicated that the great majority of compliments occur between interlocutors who are friends or acquaintances, rather than strangers (e.g., Manes 1983; Wolfson 1981, 1989). The following table shows the overview of the DCT: TABLE 1. DCT situations Situation 1 Situation 2 Situation 3 Situation 4

Distance -

Dominance/Power = + =

Compliment Type Ability Performance Appearance Possession

The DCT data were evaluated by two native speakers of English using Eisenstein and Bodman’s (1993) nativeness rating scale (e.g., 4 = Nativelike, 3 = Acceptable, 2 = Problematic, 1 = Not acceptable). Two ratings were assigned for each participant (one for compliments and the other for compliment responses). The two ratings were averaged and the average was considered as the final score for each participant. Pearson product moment Correlation was used to examine the degree of consistency between the two independent raters who scored the participants’ DCT scores. The result showed a high degree of correspondence between the ratings (Į = .82, p = .001).

3.3. Background questionnaire The questionnaire was a self-report instrument designed to elicit information on the total amount of time participants used English during

66

Chapter Three

the week both inside and outside the classroom (e.g., the time spent interacting in English, watching television or listening to the radio, reading books in English, and writing email), and the number of years spent in the United States. The data obtained from the background questionnaire was used to examine whether there was a positive relation among the variables of length of residence, amount of interaction, and type of interaction and the Korean ESL learners’ L2 pragmatic competence. The background information questionnaire (Appendix B) contained a total of 10 items. Six of these items were designed to elicit general information about the participants’ linguistic, educational, and personal background and were not considered in the assignment of contact scores for individual participants. The remaining four items were used to measure participants’ English language contact in daily life. To get the total number of contact hours, we totaled the number of hours given in answer to each question. Researchers (e.g., Al-Ansari 2000; Longcope 2003; Spada 1986) suggest that certain kinds of L2 contact activity might be weighted more than others. Based on their suggestion, the activities were categorized as more interactive (e.g., interaction in English, chatting via internet) and less interactive (e.g., watching television, listening to the radio). Also, quantitative values were differently assigned to each activity reflect the level of interactivity involved in using English. As Long (1982) maintains, interactional language use provides learners with more opportunities for getting comprehensible input. Through interaction speakers exchange information and provide feedback to each other. Modification process such as self-correction and clarification check occur during the negotiation process, which, in turn, should facilitate learners L2 development. Therefore, the subjects’ involvement in interaction in English whether it was with native or nonnative speakers of English (questionnaire item #7) was given 3 points (highest rank). Emailing or chatting on the internet (item #10) was given 2 points because this activity does not make the same communicative demands on the learner as engaging in spontaneous face- to- face conversational interaction. As Lluna-Mateu (2006) stated, online interactions may not imply the level of interactivity and negotiation of meaning that face to face interactions require. Reading books (#8) and watching television and listening to the radio (#9) were given 1 point, respectively. As asserted by Spada (1986), watching television does not make the same communicative demands on the learner as engaging in conversation. Therefore, this research focused

Competence, Residence, Amount of Contact and Intensity of Interaction

67

not only on the amount of contact but also on the type of contact based on the interactivity scale.

4. Results 4.1. Length of residence and learners’ pragmatic competence The third research question examined to what extent achievement of high pragmatic competence in the subjects’ L2 is related to the length of residence in the second language community. To examine the relationship between length of residence and pragmatic competence based on the DCT test, a Pearson product-moment correlation analysis was performed using SPSS with alpha set at .05. As seen in Table 10, the correlation coefficient for DCT with length of residence (r= -.141, p = .329) indicates that there is no significant effect of length of residence on the subjects’ pragmatic attainment. A number of studies supported the positive effect of length of residence on language acquisition (Ioup 1995; House 1996; Flege and Liu 2001). Contrary to what the investigator might expect, however, length of residence in the target country has not been identified as a predictor of learners’ L2 pragmatic achievement in this study. TABLE 2. Correlations between Pragmatic Competence and Length of Residence Variables Pragmatic Competence & Length of Residence

r -.141

p .329

As shown in Table 2, the relationship between the two variables was in the desired direction; that is, longer length of residence was more likely to lead to better outcomes in the L2 pragmatics, but it was relatively weak and non-significant. Many questions remain about the validity of the assumption that a lengthy residence in the target-language area provides an ideal context for language learning (Spenader 2005). We know, for example, that not all individuals who live abroad for an extended period make the same linguistic gains. This finding is congruent with other research in which there would be a negative relationship between length of residence in target community and language proficiency (Kondo 1997; Roever 2001; Rodriguez 2001). The results indicated that none of L2 performance benefited from length of residence in target language community in these studies.

Chapter Three

68

Overall, the results revealed that motivation variable examined demonstrated a positive and moderate relationship to the Korean ESL learners’ L2 pragmatic competence, which perhaps suggests that highly motivated learners may be superior in their pragmatic competence to those with lower motivation. However, overall amount of interaction and length of residence variables were not significant with these particular Korean ESL students even though it was expected that greater amount of interaction and a lengthy residence would lead to more increased levels of pragmatic competence. Nevertheless, the correlation between the reading category of interaction and the subjects’ pragmatic competence was moderately significant.

4.2. Amount of contact and learners’ pragmatic competence Correlational analysis was used to examine the data. For all the analyses alpha level was set at .05. Table 3 presents the means, standard deviations, skewness and kurtosis of the raw data for the items measuring amount of contact and participants’ pragmatic competence. TABLE 3. Amount of contact and pragmatic performance

DCT Amount of Contact Valid N (listwise)

N

Mean

Std.

50

3.58

.26

50

32.04

18.43

Skewness Statistic Std. Error .606 .337 .751

.337

Kurtosis Statistic Std. Error -.003 .662 -.633

.662

50

The mean level received on DCT representing the L2 pragmatic competence of the Korean ESL learners when performing complimenting behavior indicated that the Korean ESL participants attained a relatively high level of English pragmatic competence. A mean of 3.58 out of four (4 = Native-like, 3 = Acceptable, 2 = Problematic, 1 = Not acceptable) suggests that the Korean ESL learners’ DCT rating is between “acceptable” and “native-like.” The background information questionnaire examined various types of contact with English both in class and out of class. The items surveyed the total amount of time the participants had contact with English each week, including productive, more interactive use of language (e.g.,

Competence, Residence, Amount of Contact and Intensity of Interaction

69

conversational interaction in English, emailing or chatting on the internet) and receptive, less interactive use of language (e.g., reading books, watching television, and listening to the radio). The overall amount of English-language contact that the participants reported had a mean of 32.04 hours per week on average. The standard deviation for DCT was relatively small (lower than 1.0). However, the standard deviation for amount of contact was quite large, indicating that there was a great deal of variation in the total amount of time participants spent in contact with English each week. To test the normality of the raw data and to ensure further analyses, skewness (quantification of the asymmetry of the distribution) and kurtosis (quantification of the shape of the distribution) were examined. A review of the statistics shown in Table 1 revealed that the skewness and kurtosis values of the two variables all lie between ± 1.0, which means that the two variables fall within the “excellent” range as acceptable variables for further analyses (George and Mallery 2001). A Pearson product-moment correlation matrix was used to examine the relationship between the amount of contact with English as measured by the background questionnaire and Korean ESL students’ level of pragmatic ability in English as measured by the DCT. The correlation coefficient matrix is shown in Table 4. TABLE 4. Correlations between Amount of Contact and Pragmatic Competence Variables Pragmatic Competence & Amount of Contact

r .046

p .754

Contrary to what was expected, the correlation coefficient for amount of L2 contact was not statistically significant at p > .05. Given the nonsignificant relationship between the amount of contact with English and the students’ pragmatic ability, the hypothesis that greater amount of contact with English would lead to higher levels of pragmatic competence is not supported. The result was consistent with other research that shows that language contact does not necessarily result in L2 proficiency (Day 1984; DeKeyser 1986; Lapkin Hart, and Swain 1995; Yager 1998; Segalowitz and Freed 2004). With respect to lack of correlation between amount of L2 contact and pragmatic competence, previous research has shown that type of contact is more important than amount of contact in developing language proficiency. Therefore, we proceeded to analyze the correlation coefficient

Chapter Three

70

between the type of L2 contact and pragmatic competence in order to examine whether the type of contact, rather than the amount of contact, is more important in developing L2 pragmatics.

4.3. Type of contact and learners’ pragmatic competence In order to respond to this question, we calculated a Pearson productmoment correlation matrix between separate types of contact with and the learners’ pragmatic competence. To ensure a more reliable and valid analysis, first a review of summary statistics was conducted to determine if there is a normal distribution for the different types of contact variable. Descriptive statistics in Table 5 shows mean, standard deviation, skewness and kurtosis of the raw data for the four different types of contact variable. TABLE 5. Descriptive Statistics of the Four Types of Contact Variable

Conversational interaction Reading books Watching TV and listening to the radio Emailing or chatting via the internet Valid N (listwise)

Skewness Statistic Std. Error

Kurtosis Statistic Std. Error

N

Mean

Std.

50

6.69

1.85

-.803

.337

.925

.662

50

14.97

15.35

.958

.337

-.378

.662

50

6.99

5.47

.629

.337

-.374

.662

50

3.39

4.55

4.361

.337

24.038

.662

50

A review of the summary statistics showed an abnormal distribution for the variable “emailing or chatting via the internet.” For a normal distribution, values for skewness and kurtosis must be near zero (George and Mallery 2001). The measures of skewness and kurtosis were 4.36 and 24.04, respectively. Thus, a data transformation was executed on the variable “emailing or chatting via the internet.” Even after computing the square root of the variable and recalculating the summary statistics,

Competence, Residence, Amount of Contact and Intensity of Interaction

71

transformation failed to lower the skewness and kurtosis of the variable and therefore, this variable was dropped from further analysis. Table 6 presents the correlation between the participants’ DCT performance and the types of contact. We used the total number of hours per week reported in the background questionnaire across the different types of contact activities. TABLE 6. Correlations between Pragmatic Competence and the Different Types of Contact Variable Variables Pragmatic Competence & conversational interaction Pragmatic Competence & reading books Pragmatic Competence & watching TV and listening to the radio *. Correlation is significant at the 0.05 level (2-tailed)

r .377* .077 -.214

p .007 .593 .136

The results found only significant positive correlations between the variable “conversational interaction in English” and participants’ pragmatic competence at p < .05. The results show that interactions the learners had with native and non-native speakers of English were integral to their pragmatic competence. This finding lends support to the notion that more interactive language activities lead to greater language acquisition than less interactive language activities such as reading books, watching television, and listening to the radio.

5. Discussion Our findings indicated the relationship between length of residence and pragmatic competence to be non-significant. Furthermore the relationship between amount of contact and pragmatic competence was also insignificant. However, the relationship between type of contact and pragmatic competence was robust and statistically significant. Our findings indicate that the amount of interaction and length of residence in the community does not contribute significantly to learners’ pragmatic competence. As suggested by other researchers (e.g., Longcope 2003; Long 1982) interactional language use provides learners with more opportunities for noticing the linguistics forms and the gap between their performance and the native speakers’ performance. Specifically, the learners which had more substantial opportunities for social contact with other English speakers were found to exhibit a more native-like behavior with regard to the use of complimenting and compliment responses. The insignificant relationship between amount of contact and pragmatic

72

Chapter Three

competence indicates that it is not quantity but the intensity of language contact that is of primary importance in pragmatic development. In this sense, the findings of this study largely confirm views according to which pragmatic development depends greatly on the quality of the input available to learners (Bella 2012, 2011; Bardovi-Harlig and Bastos 2011; Klein, Dietrich, and Noyau 1995). Our findings provide further support to the claim that length of residence alone is not a sufficient condition for the development of pragmatic competence (Bella 2012). L2 research has shown that more interactive types of language contact can lead to greater L2 acquisition than less interactive types of language contact. Although it has been frequently claimed that length of residence can correlate positively with pragmatic development (e.g., Kasper and Rose 2002), the results of the present study seem to corroborate the view that the type of contact and the level of interactivity rather than simply the quantitative measure of length of residence and amount of contact is more instrumental in the learners’ pragmatic development. As Cohen and Shively (2007) contend, the types of experiences that learners have during their stay in an L2 speaking setting may present great variation depending on opportunities for social interaction with target language speakers, as well as their own individual characteristics. Therefore, claims about the positive effects of residence in the L2 setting should be considered with caution (Bella 2012; Félix-Brasdefer 2004), since the nature of contact and the level of interactivity of the situation are other important parameters to consider. The findings of the study lend support to the notion that productive, more interactive language activities lead to greater language acquisition than receptive, less interactive language activities such as reading books, watching television, and listening to the radio. When we compare our result to the results of studies that have found a significant role of length of stay for the development of pragmatic competence, we should keep in mind that in our study, the level of language proficiency was relatively controlled by including only advance learners. The participants were all graduate students with a TOEFL score of 550 or above. The results may be different with learners of different proficiency level. The extent to which length of stay is reported to be influential more likely reflects successful interaction which may also be enhanced by the learners’ level of language proficiency. One learner’s experience during residence in the target community might be different from another learner. Learners who seek and engage in highly interactive contacts have a better chance to improve their pragmatic competence.

Competence, Residence, Amount of Contact and Intensity of Interaction

73

It should also be noted that some aspects of pragmatic competence may take longer than others to develop. Due to the nature of pragmatic feature we investigated (complimenting) and its likely occurrence more in informal situations, academic context in which learners of this study had most of their daily contact with English, may not have provided sufficient opportunities for its use. Accordingly, length of residence may be defined by different purposes for the time in the host community (Bardovi-Harlig and Bastos 2011). The amount of daily interaction in the target language depends on the role of learner also. Students’ interaction patters and intensity for example, may be different and less intense than the amount of daily interaction in the target language and culture required by missionaries.

6. Conclusion This study aimed to investigate the pragmatic performance of Korean ESL learners when complimenting and responding to complements in relation to length of residence, amount of contact, and type of contact. Our findings show that learners with more frequent opportunities for interactive use of language with other language users show a significant increase in their pragmatic competence compare to the ones with less interactive activities. This study has certain limitations. The most important limitation concerns the use of the DCT, the disadvantages of which are extensively discussed in the literature (Eslami and Mirzaei 2014; Félix-Brasdefer 2010). As stated by these researchers, DCTs lack of authenticity of natural communication and elicited written answers may not reflect natural interactional features of oral conversations. Ethnographic research on speech act performance in natural settings is needed for a better understanding of the pragmatic development of learners in relation to amount and type of contact. Our study showed that students who spent more time in more interactive types of language contact demonstrate better pragmatic competence. More fine-tuned data is needed to better define the relationship found. Rather than simply saying which type of language contact is better, it may be more fruitful to inquire into the dynamics of the interactions between learners and target language users: for example, one can look at what the learner may be doing during L2 contact with target language users that may facilitate learning the L2, as well as what the learner’s interlocutor may be doing that may help the learner learn the L2. It is clear from the data presented in this study that the interactions the

74

Chapter Three

learners had with native speakers were integral to their pragmatic competence. In order to fully understand the role of interactions in L2 pragmatics, it is necessary to refer to the notion of noticeability: in other words, how much of the valuable input L2 learners will understand and notice. According to Schmidt (1993), the extent to which L2 learners notice and accommodate to the L2 pragmatic norms is dependent on learners’ motivation to master their L2. Learners who are more eager to establish “smooth relationships” with the target language community are likely to be more attentive to relevant features in the interactions and eager to sound “pragmatically correct” in their L2, which affects the actual L2 pragmatic competence of L2 learners. It should also be noted that the questionnaire data alone may not explain the week-by-week fluctuations in learners’ L2 contact. In addition to obtaining quantitative measures, it is also essential to use qualitative information about learners’ contact gathered longitudinally to provide vital complementary information as to the influences making for such fluctuation. To get a more complete picture of learners’ language contact, the researchers could use daily diary entries as a measure of contact, or learners could be requested to fill out a daily contact sheet which could specify not only the length and type of contact, but also, detailed information about the context in which the contact took place, the type of interlocutors learners interacted with as well as more information about the amount and type of contact that learners have outside the classroom. Despite these limitations, this research has shed some light on the largely under-examined relationships between the ESL learners’ pragmatic development and the amount and type of contact. The findings suggest that simple exposure to language is unlikely to be sufficient for acquisition of L2 pragmatic knowledge because the specific linguistic realizations are sometimes not salient enough for the learner to notice. The findings replicate the major conclusions of previous studies on the role of conversational interaction in L2 pragmatic acquisition. The findings have practical implications for promoting L2 pragmatic development. It is suggested that opportunities for interaction with other English language users be provided to learners. Also learners should be encouraged to seek diverse opportunities for interaction with native and nonnative speakers. It should also be pointed out that as Thomas (1983) claims, we cannot expect learners to acquire pragmatic norms without explicit teaching, regardless of the learning setting. Recent research on interlanguage pragmatics (Alcón-Soler and Martínez-Flor 2008; Eslami and Liu 2013; Taguchi 2009), highlights the importance of explicit pragmatic instruction for developing pragmatic competence in L2.

Competence, Residence, Amount of Contact and Intensity of Interaction

75

References Adolphs, S., and V. Durow. 2004. “Social-cultural integration and the development of formulaic sequences.” In Formulaic sequences: Acquisition, processing and use, edited by N. Schmitt, 107-126. Amsterdam: Benjamins. Alcón-Soler, E., and A. Martínez-Flor. 2008. Investigating pragmatics in foreign language learning, teaching and testing. Bristol, UK: Multilingual Matters. Al-Ansari, S. 2000. “Sheltered curricular exposure and unsheltered extracurricular exposure as factors influencing the development of academic proficiency in ESL.” IRAL 38: 175-193. Barnlund, D. C., and S. Akari. 1985. “Intercultural encounters: the management of compliments by Japanese and Americans.” Journal of Cross-Cultural Psychology 16: 9-26. Bardovi-Harlig, K. 1999. “Exploring the interlanguage of interlanguage pragmatics: A research agenda for acquisitional pragmatics.” Language Learning 49: 677-713. —. 2001. “Empirical evidence of the need for instruction in pragmatics.” In Pragmatics in Language Teaching, edited by K. Rose, and G. Kasper, 13-32. Cambridge: Cambridge University Press. —. 2013. “Developing L2 Pragmatics.” Language Learning Supplement 1: 68-96. Bardovi-Harlig, K., and M.-T. Bastos. 2011. “Proficiency, length of stay, and intensity of interaction and the acquisition of conventional expressions in L2 pragmatics.” Intercultural Pragmatics 8: 347-384. Bardovi-Harlig K., and Z. Dörnyei. 1998. “Do language learners recognize pragmatic violations? Pragmatic versus grammatical awareness in instructed L2 learning.” TESOL Quarterly 32: 233-262. Bella, S. 2011. “Mitigation and politeness in Greek invitation refusals: Effects of length of residence in the target community and intensity of interaction on non-native speakers’ performance.” Journal of Pragmatics 43: 1718-1740. —. 2012. “Length of residence and intensity of interaction: modification in Greek L2 requests.” Pragmatics 22: 1-39. Billmyer, K. 1990. “‘I really like your lifestyle’: ESL learners learning how to compliment.” Penn Working Papers in Educational Linguistics 6: 31-48. Billmyer, K., V. Jakar, and M. P. Lee. 1989. “Developing pragmatic awareness through communicatively oriented materials” (presentantion,

76

Chapter Three

First Annual Pedagogy and Pragmatics Colloquium, International TESOL Convention, San Antonio, Texas). Blum-Kulka, S., and E. Olshtain. 1986. “Too many words: Length of utterance and pragmatic failure.” Studies in Second Language Acquisition 8: 165-180. Bouton, L. F. 1994. “Conversational implicature in the second language: Learned slowly when not deliberately taught.” Journal of Pragmatics 22: 157-167. —. 1999. “Developing nonnative speaker skills in interpreting conversational implicatures in English: Explicit teaching can ease the process.” In Culture in second language teaching and learning, edited by E. Hinkel, 47-70. Cambridge: Cambridge University Press. Chen, R. 1993. “Responding to compliments: A contrastive study of politeness strategies between American English and Chinese speakers.” Journal of Pragmatics 20: 49-75. Churchill, E. 2001. “The effect of a study abroad program on request realizations by Japanese learners of English.” Kanagawa University Studies in Language 24: 91-103. Churchill, E., and M. DuFon. 2006. “Evolving threads in study abroad research.” In Language learners in study abroad contexts, edited by M. DuFon, and E. Churchill, 1-27. Clevendon: Multilingual Matters. Cohen, A., and Shively, R. 2007. “Acquisition of requests and apologies in Spanish and French: Impact of study abroad and strategy-building intervention.” The Modern Language Journal 91: 189-212. Day, R. 1984. “Student participation in the ESL classroom or some imperfections in practice.” Language Learning 34: 69-102. DeKeyser, R. M. 1986. “From learning to acquisition? Foreign language development in a U.S. classroom and during a semester abroad.” PhD dissertation, Stanford University. Dietrich, R., K. Wolfgang, and N. Collette. 1995. The acquisition of temporality in a second language. Amsterdam: Benjamins. Eisenstein, M., and J. Bodman. 1993. “Expressing gratitude in American English.” In Interlanguage pragmatics, edited by G. Kasper, and S. Blum-Kulka, 64-81. Oxford: Oxford University Press. Eslami-Rasekh, Z. 2005. “Raising the pragmatic awareness of language learners.” ELT Journal 59: 199-208. Eslami, Z. R., and A. Mirzaei. 2014. “Speech act data collection in a nonWestern context: oral and written DCTs in the Persian language.” Iranian Journal of Language Testing 4: 137-154.

Competence, Residence, Amount of Contact and Intensity of Interaction

77

Eslami, Z. R., and C. Liu. 2013. “Learning Pragmatics through ComputerMediated Communication in Taiwan.” Iranian Journal of Society, Culture, & Language 1: 52-73. Flege, J.E., and S. Liu. 2001. “The effect of experience on adults’ acquisition of a second language.” Studies in Second Language Acquisition 23: 527-552. Félix-Brasdefer, C. 2004. “Interlanguage refusals: Linguistic politeness and length of residence in the target community.” Language Learning 54: 587-653. —. 2010. “Data collection methods in speech act performance: DCTs, role plays and verbal reports.” In Speech act performance: Theoretical, empirical and methodological issues, edited by A. Martínez-Flor, and E. Usó-Juan, 41-56. Amsterdam: Benjamins. Félix-Brasdefer, J.C., and M. Hasler-Barker. 2012. “Complimenting and responding to a compliment in the Spanish FL Classroom: From empirical evidence to pedagogical intervention.” In Speech Acts and Politeness across Languages and Cultures, edited by L. Ruiz de Zarobe, and Y. Ruiz de Zarobe, 241-273. New York: Peter Lang. George, D., and P. Mallery. 2001. SPSS for windows: Step by step. Boston: Allyn and Bacon. Golato, A. 2005. Compliments and compliment responses: Grammatical structure and sequential organization. Amsterdam: Benjamins. Han, S. 2005. “The interlanguage pragmatic development of the speech act of requests by Korean non-native speakers of English in an ESL setting.” PhD dissertation, University of Pensilvania. Herbert, R. K. 1986. “Say ‘thank you’ – or something.” American Speech 61: 76-88. —. 1991. “The sociology of compliment work: an ethnocentric study of Polish and English compliments.” Multilingua 10 (4): 381-402. Holmes, J. 1988. “Compliments and compliment responses in New Zealand English.” Anthropological Linguistics 28 (4): 485-508. —. 1995. Women, men and politeness. London: Longman. House, J. 1996. “Developing pragmatic fluency in English as a foreign language: Routines and metapragmatic awareness.” Studies in Second Language Acquisition 17: 225-252. Ioup, G. 1995. “Evaluating the need for input enhancement in post-critical period language acquisition.” In The age factor in second language acquisition, edited by D. Singleton, and Z. Lengyel, 95-123. Clevedon: Multilingual Matters. Kasper, G. 1992. “Pragmatic transfer.” Second Language Research 8: 203231.

78

Chapter Three

Kasper, G., and Rose, K. R. 2002. Pragmatic development in a second language. Oxford: Blackwell Publishing. Kasper, G., and R. Schmidt. 1996. “Developmental issues in interlanguage pragmatics.” Studies in Second Language Acquisition 18 (2): 149-169. Klein, W., R. Dietrich, and C. Noyau. 1995. “Conclusions.” In The acquisition of temporality in a second language, edited by R. Dietrich, W. Klein, and C. Noyau. 261-280. Philadelphia: Benjamins. Kondo, S. 1997. “Longitudinal study on the development of pragmatic competence in a natural learning context – Perception behind performance.” Proceedings of Sophia University Linguistic Society 12: 35-54. Lapkin, S., D. Hart, and M. Swain. 1995. “A Canadian interprovincial exchange: Evaluating the linguistic impact of a three-month stay in Quebec.” In Second language acquisition in a study abroad context, edited by B. F. Freed, 67-94. Amsterdam: Benjamins. Lee, J. S., and B .McChesney. 2000. “Discourse rating tasks: A teaching tool for developing sociocultural competence.” ELT Journal 64 (2): 161-168. Leech, G. N. 1983. Principles of pragmatics. London: Longman. Lewandowska-Tomaszczyk, B. 1989. “Praising and complimenting.” In Contrastive pragmatics, edited by W. Oleksy, 73-100. Amsterdam: Benjamins. Liu, D. 1995. “Sociocultural transfer and its effect on second language speakers’ communication.” International Journal of Intercultural Relations 19 (2): 253-265. Lluna-Mateu, F. R. 2006. “Development of Spanish L2 competence in a synchronous CMC (chat room) environment.” PhD dissertation, Louisiana State University. Long, M. H. 1982. “Native speaker/non-native speaker conversation in the second language classroom.” In On TESOL ’82, edited by M. Clarke, 207-25. Washington D.C: TESOL. —. 1983. “Does second language instruction make a difference? A review of the research.” TESOL Quarterly 17: 359-382. Longcope, P. D. 2003. “What is the impact of study abroad on L2 learning? A descriptive study of contexts, conditions, and outcomes.” PhD dissertation, University of Pennsylvania. Lorenzo-Dus, N. 2001. “Compliment responses among British and Spanish university students: a contrastive study.” Journal of Pragmatics 33: 107-127. Maíz-Arévalo, C. 2012. “At a loss for words or how to respond to compliments: A contrastive analysis of compliment response in

Competence, Residence, Amount of Contact and Intensity of Interaction

79

English and Spanish.” In New perspectives on (im)politeness and interpersonal communication, edited by L. Fernández Amaya, 157173. Newcastle upon Tyne: Cambridge Scholars. —. 2013. “Just click ‘Like’: Computer-mediated responses to Spanish compliments.” Journal of Pragmatics 51: 47-67. Manes, J. 1983. “Compliments: A mirror of cultural values.” In Sociolinguistics and Language Acquisition, edited by N. Wolfson, and E. Judd, 96-102. Rowley: Newbury House. Marriott, H. 1995. “The acquisition of politeness patterns by exchange students in Japan.” In Second language acquisition in a study abroad context, edited by B. F. Freed, 197-224. Amsterdam: Benjamins. Matsumura, S. 2001. “Learning the rules for offering advice: A quantitative approach to second language socialization.” Language Learning 51 (4): 635-679. —. 2003. “Modeling the relationships among interlanguage pragmatic development, L2 proficiency, and exposure to L2.” Applied Linguistics 24 (4): 465-491. Olshtain, E., and S. Blum-Kulka. 1985. “Degree of approximation: Nonnative reactions to native speech act behavior.” In Input in second language acquisition, edited by S. M. Gass, and C. G. Madden, 303329. Cambridge: Newbury House Publishers. Placencia, M. E., and A. Lower. 2013. “Your kids are so stinkin’ cute! :-): Complimenting behavior on Facebook among family and friends.” Intercultural Pragmatics 10 (4): 617-646. Pomerantz, A. 1978. “Compliment responses: Notes on the co-operation of multiple constraints.” In Studies in the organisation of conversational interaction, edited by J. Schenkein, 79-112. New York: Academic Press. Rodriguez, S. 2001. “The perception of requests in Spanish by instructed learners of Spanish in the second- and foreign-language contexts: A longitudinal study of acquisition patterns.” PhD dissertation, Indiana University, Bloomington. Roever, C. 2001. “A web-based test of interlanguage pragmalinguistic knowledge: Speech acts, routines, implicatures.” PhD dissertation, University of Hawaii. Rose, K. 2005. “On the effects of instruction in second language pragmatics.” System 33: 385-399. Rose, K., and C. N. Kwai-fun. 2001. “Inductive and deductive teaching of compliments and compliment responses.” In Pragmatics in language teaching, edited by K. Rose, and G. Kasper, 145-170. Cambridge: Cambridge University Press.

80

Chapter Three

Schauer, G. 2006. “Pragmatic awareness in ESL and EFL contexts: Contrast and development.” Language Learning 56: 269-318. —. 2007. “Finding the right words in a study abroad context: The development of German learners’ use of external modifiers in English.” Intercultural Pragmatics 4: 193-220. Schmidt, R. W. 1983. “Interaction, acculturation, and the acquisition of communicative competence: A case study of an adult.” In Sociolinguistics and language acquisition, edited by N. Wolfson, and E. Judd, 137-174. Massachusetts: Newbury House Publishers, Inc. —. 1993. “Consciousness, learning and interlanguage pragmatics.” In Interlanguage pragmatics, edited by G. Kasper, and S. Blum-Kulka, 21-42. Oxford: Oxford University Press. Segalowitz, N., and B. F. Freed. 2004. “Context, contact, and cognition in oral fluency acquisition: Learning Spanish in at home and study abroad contexts.” Studies in Second Language Acquisition 26: 173-199. Siegal, M. 1994. “Looking east: Learning Japanese as a second language in Japan and the interaction of race, gender and social context.” PhD dissertation, University of California, Berkeley. Spada, N. 1986. “The interaction between type of contact and type of instruction: Some effects on the L2 proficiency of adult learners.” Studies in Second Language Acquisition 8 (2): 181-199. Spencer-Oatey, H. 2000. Culturally speaking: Managing rapport through talk across cultures. London: Continuum. Spenader, A. J. 2005. “Cross-cultural adaptation and language acquisition in high school study abroad.” PhD dissertation, University of Minnesota. Swain, M. 1985. “Communicative competence: Some roles of comprehensible input and comprehensible output in its development.” In Input in second language acquisition, edited by S. Gass, and C. Madden, 235-253. Rowley, MA: Newbury House. Taguchi, N. 2008a. “Cognition, language contact, and the development of pragmatic comprehension in English as a second language.” Language Learning 58: 33-71. —. 2008b. “Pragmatic comprehension in Japanese as a foreign language.” Modern Language Journal 92: 558-576. —. 2008c. “The role of learning environment in the development of pragmatic comprehension: A comparison of gains between EFL and ESL learners.” Studies in Second Language Acquisition 30: 423-452. —. 2009. “Comprehension of indirect opinions and refusals in L2 Japanese.” In Pragmatic competence, edited by N. Taguchi, 249-274. Berlin: Mouton de Gruyter.

Competence, Residence, Amount of Contact and Intensity of Interaction

81

—. 2011. “The effect of L2 proficiency and study-abroad experience on pragmatic comprehension.” Language Learning 61: 1-36. Wolfson, N. 1989. Perspectives: Sociolinguistics and TESOL. New York: Newbury House. Wolfson, N., and J. Manes. 1981. “The compliment formula.” In Conversational routine, edited by F. Coulmas, 115-132. Berlin: Mouton de Gruyter. —. 1980. “The compliment as a social strategy.” Research on Language & Social Interaction 13 (3): 391-410. Yager, K. 1998. “Learning Spanish in Mexico: The effect of informal contact and student attitudes on language gain.” Hispania 81 (4): 898913.

Appendix A Discourse completion task Please read the description of each situation carefully and then write down in English what you would say and what the responses of the other person might be in that situation. Continue to write as much as you think is appropriate for each situation until the topic would change. Try to respond as you would naturally do in real-life language use. Consider that the friend is of the same gender as you are. Situation 1 You and your friend decide to co-write an academic paper. While working together, you notice that s/he is a very good writer. You: Your friend: You: Your friend: You: Your friend: Situation 2 You go to your professor’s house for an end of term potluck party and while leaving you would like to compliment his wife on the food. You: Your professor’s wife: You: Your professor’s wife: You: Your professor’s wife: Situation 3 You bump into an undergraduate student whom you go to the same church with and you notice that s/he is wearing a new pair of jeans today and s/he looks really good.

82

Chapter Three

You: An undergraduate student: You: An undergraduate student: You: An undergraduate student: Situation 4 Your friend comes to class one day, seems very excited, and sits next to you. S/he pulls out a picture – it’s a picture of her/his new car. You: Your friend: You: Your friend: You: Your friend:

Appendix B Background questionnaire Please report hours you spent on the activities listed in the questionnaire during the week just preceding administration of the questionnaire. 1. Age: _________ years old 2. Sex: Male/ Female 3. How old were you when you came to the U.S.: ________ years old 4. How long have you been in the U.S.? _________ years ____________ months 5. If you have ever taken TOEFL, what was your best score? 6. Before coming to the U.S. did you ever visit or live in an English speaking country such as Canada, Australia, Britain, etc? If your answer is yes, how long did you stay in the country? 7. How much time do you spend speaking English with English-speaking Americans or non-native speakers of English per week? (For example: teacher, friend, neighbor, etc.) _______________ hours per week 8. The average number of hours per week you read English books, newspapers, or magazines: _______________ hours per week 9. The average number of hours per week you watch TV and listen to the radio, tapes, or records in English: _______________ hours per week 10. The average number of hours per week you send email or chat in English via the Internet: _______________ hours per week

Competence, Residence, Amount of Contact and Intensity of Interaction

83

Notes 1

This paper is based on a PhD dissertation. The dissertation is available at http://repository.tamu.edu/bitstream/handle/1969.1/ETD-TAMU-2487/AHNDISSERTATION.pdf?sequence=1

CHAPTER FOUR COMPARING TEXTBOOKS AND TV SERIES AS SOURCES OF PRAGMATIC INPUT FOR LEARNERS OF ITALIAN AS A SECOND LANGUAGE: THE CASE OF COMPLIMENTS AND INVITATIONS ELENA NUZZO

1. Introduction Textbooks play a very important role in language classrooms, particularly in foreign language contexts, where they represent the main source of input learners are exposed to. However, their appropriateness for presenting learners with pragmatic features of the target language has been seriously questioned. As we will see in the next section, the existing literature on the topic, mostly devoted to the teaching of English as a second/foreign language, has shown that textbooks offer an “inaccurate and decontextualized presentation of the different pragmatic aspects examined, as well as a lack of natural conversational models representing the real use of language” (Martínez-Flor 2008, 246). A possible explanation for this inadequacy is that L2 textbooks rely heavily on nativespeakers’ intuitions about the pragmatic norms of their community rather than on empirical research (Boxer 2003; LoCastro 2003). Textbooks of L2 Italian have not been analysed from this perspective, at least not to my knowledge. Therefore, in order to observe whether they offer the same poor pragmatic input as their English counterparts, an exploratory study was carried out with a focus on the speech acts of complimenting and inviting. The current study proposes a comparison between textbooks and TV series, starting from the assumption that even though the language used in television is not the same as unscripted language, “the use of television dialogue as a surrogate for natural

86

Chapter Four

conversation for the analysis of certain linguistic features seems perfectly appropriate” (Quaglio 2009, 149). Given its exploratory nature, the present research is based on a small sample of data and does not intend to provide generalisable findings. Rather, it represents a first step towards a better understanding of L2 Italian teaching materials and a possible source of suggestions for textbook writers and practitioners.

2. Theoretical Background 2.1. Pragmatics in textbooks Textbooks have often been criticised by researchers for not representing good language models in the area of pragmatics. Bardovi-Harlig et al. (1991) surveyed conversational closings in twenty textbooks of American English, and reported that many of them failed to include complete examples of closings. Boxer and Pickering’s (1995) analysis of complaints in four American and three British English language textbooks revealed that the teaching materials focused mainly on direct complaining, whereas also indirect complaining is common in natural speech. Moreover, the authors noted that the main aim of the examples presented was to teach learners how to soften the face-threatening act of complaining through the use of polite expressions. In their investigation of suggesting and advising realisation strategies, Salazar-Campillo and Usó-Juan (2002) concluded that the conventionally indirect strategies were the most commonly ones presented in all the textbooks analysed, with almost no instances of indirect or direct strategies. Vellenga (2004) reported on a qualitative and quantitative study which focused specifically on the use of metalanguage, explicit treatment of speech acts, and metapragmatic information. Findings showed that the eight textbooks analysed included a paucity of explicit metapragmatic information, and teachers’ manuals rarely supplemented adequately. In her study on textbooks requests, Usó-Juan (2008) noted that coursebook writers showed a clear preference for using conventionally indirect request strategies in the activities presented to learners, and tended to opt for openers and the politeness marker please to modify the request moves. The activities focused almost exclusively on the acquisition of linguistic competence, and contextual information regarding interlocutors’ age, social status, degree of intimacy or degree of imposition of the request was not mentioned either explicitly or implicitly. The speech act of requesting was investigated also by Salazar-Campillo (2007), who concentrated on the use of mitigators. The transcript survey showed a

Comparing Textbooks and TV Series as Sources of Pragmatic Input

87

limited use and range of mitigating devices to soften the impact of the speech act. In order to support the claim that L2 textbooks are inappropriate for providing learners with accurate presentations of the pragmatic aspects of the target language, some scholars have compared naturally occurring data to textbook dialogues. Scotton and Bernsten (1988), for example, compared the two sources with regard to the communicative act of direction-giving, and found a mismatch between the two types of data, since the models provided in textbook dialogues did not include sequences that took place in natural exchanges. Williams (1988), who analysed the differences between the language of real meetings and the materials used for teaching them, argued that the language used in meetings was far more complex than that found in textbooks, and that the latter lacked typical features of spoken language, like unfinished sentences, false starts, interruptions, redundancy, repetitions and lengthy explanations. Mandala (1999) compared the structure of advice-giving exchanges in natural talk with such exchanges in textbook dialogues and found there was no correspondence between the two. It was observed that textbook dialogues were presented from an advice-giver’s point of view, therefore omitting conversational features that were common in authentic samples. In a comparison between textbook service encounters versus their authentic equivalent conversations, Gilmore (2004) found that textbooks dialogues were far from being authentic-like, differing considerably across a range of features such as turn-taking patterns, lexical density, number of false starts and repetitions, pausing, frequency of overlap or latching, and the use of hesitation devices and backchanneling. However, Gilmore (2004) observed that, when compared to older ones, recent publications were beginning to incorporate more of the discourse features found in authentic data. Kakiuki (2005) focused on the speech act of greeting and revealed that whereas one-turn greetings were presented accurately in some textbooks, other features such as number of turn-takings and particular greetings expressions were not reflected appropriately.

2.2. Pragmatics in TV material Whereas the findings summarized in the previous section suggest that textbooks are a poor source of pragmatic input for learners, films and video sequences are commonly regarded as valuable resources. Several studies report the benefits of employing this type of material for presenting learners with a contextualised view of language and real-life speech, as well as for illustrating different social realities and non-verbal aspects of

88

Chapter Four

communication. For example, Weyers (1999) measured the effects of exposure to a Spanish telenovela on the development of students’ communicative competence, intended as the confidence in generating output and the scope and breadth of their discourse. Washburn (2001, 22), analysing the use of sitcoms for the teaching of EFL pragmatics, concluded that: “sitcoms present many models of appropriate pragmatic language use among various characters of differing status, familiarity, gender, and in varied settings, such as at work, at home, in public places, and at formal gatherings.” Alcón-Soler’s (2005) study dealt with learners’ exposure to requests included in different excerpts from the series Stargate. After a period of instruction, the author provided positive evidence of the value of employing this type of material to make learners aware of the sociopragmatic and pragmalinguistic aspects involved in making requests. On the basis of her survey of modification devices in a corpus of ten films, Martínez-Flor (2008) claimed that the use of film was a good source of material for exposing learners to authentic samples of appropriate pragmatic input in a variety of contexts. Although the language used in films cannot be considered natural speech, several studies demonstrate that there are more similarities than divergences between TV series data and naturally occurring data as far as some pragmatic features are concerned. Rose (2001) compared the occurrence of compliments in forty American films with a corpus of naturally-occurring data (collected by Manes and Wolfson 1981) and found that syntactic formulae, compliment topic and compliment response strategy were similar in both sources, although some differences were identified regarding sociopragmatic features. Focusing on apologies, Kite and Tatsuki (2005) obtained similar results to those reported by Rose (2001), since the pragmalinguistic strategies employed to express apologies in both films and naturally occurring discourse were equivalent, whereas sociopragmatic factors, such as the gender of participants, appeared to differ in both sources. Tatsuki and Nishizawa (2005), after comparing the occurrence of compliments in 40 videotaped TV interviews, film data and naturally occurring data, observed that only some sociopragmatic features, such as gender, did not correspond with natural language use. In her comparison of requests in TV series and naturally occurring discourse Fernández-Guerra (2008) concluded that overall request head acts and their peripheral modification devices in video material corresponded fairly closely to the ones taking place in naturally occurring discourse. This implies that they can be regarded as a realistic representation of actual language use. As Quaglio (2009, 13) points out, “despite the natural restrictions imposed by the television medium,

Comparing Textbooks and TV Series as Sources of Pragmatic Input

89

television dialogue should sound natural; otherwise, viewer identification with the show characters can be negatively impacted, thus, potentially, affecting the success of the show” (original emphasis). This suggests that scripted language should resemble that of its audience.

3. This study 3.1. Goals and research questions From the literature review it emerges that, according to previous research, i) textbooks are a poor source of pragmatic input for learners, ii) audiovisual material is a valuable source that can present learners with samples of appropriate language use in a variety of contexts, and iii) scripted language can be an effective indicator of how natural conversation is perceived. As mentioned above, research in this field is essentially concerned with the teaching of English as a second/foreign language. As for L2 Italian, so far no studies have been conducted on the presentation of pragmatic features in textbooks. Therefore, this paper aims at contributing to the body of research on the topic and, in particular, addresses the following questions: 1) Do the dialogues presented in L2 Italian textbooks reflect the actual use of language, at least as it is represented in TV series, or are they as artificial and inappropriate as those found in their English counterparts? 2) In particular, how are compliments and invitations presented in L2 Italian textbooks as compared to TV series? Before presenting the data collection and data analysis procedures I will briefly introduce the two speech acts which this study focuses on, namely compliments and invitations.

3.2. The speech acts of complimenting and inviting Compliments and responses to compliments are among the most investigated speech acts across languages (see for example Manes 1983; Herbert 1990; Golato 2005; Ishihara 2010; on compliments in Italian: Alfonzetti 2009; De Marco 2011; Ravetto 2012). The reader is therefore invited to refer to the existing literature for a complete overview of the topic. It will suffice here to say that compliments express one’s admiration

90

Chapter Four

or approval of other’s work, appearance, tastes, abilities, and the like. They often function as a social lubricant, and contribute to establishing and maintaining solidarity. A distinction has been proposed between explicit and implicit compliments (Bruti 2006), but only the first type is considered here. Invitations and reactions to invitations are far less studied than compliments (but see Wolfson 1981; García 1999; Félix-Brasdefer 2003; Eslami 2005). Inviting is a directive act, whose point is to persuade the listener to do something. With an invitation the speaker asks the hearer to participate in or attend a certain occasion, mainly one hosted by the speaker. More specifically, for this study I included all acts referring to a future action to be performed jointly by the addresser and the addressee, to the advantage of the latter (or of both) and with a commitment of the inviter to sustain the costs.

3.3. Procedures of data collection and analysis Five textbooks were randomly chosen among the most commonly used ones in L2 Italian courses: Benvenuto! (Cittadini and Trotta 2006), Buona idea! (Bettinelli, Della Putta, and Visigalli 2011), Contatto 1 (Bozzone Costa, Ghezzi, and Piantoni 2005), Linea diretta 1 (Conforti and Cusimano 2001), Un giorno in Italia 1 (Chiappini and De Filippo 2002). All of them are elementary level books. As for TV material, five series were randomly chosen among the most popular ones on the Italian national broadcasting service (RAI): Il Commissario Montalbano, Ho sposato uno sbirro, Il Commissario Manara, Crimini, Medicina Generale. All of them, except the last one, which revolves around hospital life, belong to the genre of crime fiction. They are set in different Italian towns. I decided to randomly pick out ten compliment dialogues and ten invitation dialogues from both books and TV material. In order to do so, I first transcribed all the occurrences of the two acts I could find in some episodes of the TV series and in the books (including audio-recorded conversations), and then took two from each source (if available), possibly occurring in similar communicative situations. As only four compliment dialogues were found across the five textbooks, the database is not as balanced as it was expected (Table 1).

Comparing Textbooks and TV Series as Sources of Pragmatic Input

91

TABLE 1. Database Compliment dialogues Invitation dialogues Total

Textbooks 4 10 14

TV series 10 10 20

Each compliment and invitation dialogue was transcribed and then both the initiator’s act and the response were analysed with regard to strategies of realisation of the head act, internal and external modifiers. This kind of analysis has been widely used in the fields of cross-cultural and interlanguage pragmatics, as for example in Blum-Kulka, House and Kasper (1989), Trosborg (1995), Achiba (2003), Nuzzo (2007). The head act strategy is the semantic and grammatical form chosen by the speaker for the utterance(s) that convey(s) the illocutionary force of the main speech act. Internal modifiers are grammatical or lexical items whose presence is not necessary for conveying the propositional content; they contribute to modifying the illocutionary force of the speech act by intensifying or softening it. External modifiers are utterances that precede or follow the head act and contribute to modulating its illocutionary force, being not necessary nor sufficient to convey it. In order to classify the different types of strategies and modification devices used for the speech acts under investigation, a taxonomy was elaborated on the basis of previous literature in the filed (for example, Ishihara 2010 for compliments and responses to compliments; García 1999 for invitations; Félix-Brasdefer 2003 for responses to invitations). The categories are listed and exemplified in Tables 2-5.

Chapter Four

92

TABLE 2. Compliments Head act strategy

Internal modifier

Type Praise of performance/skills Praise of possession Praise of appearance

Example Questa cotoletta è eccezionale Proprio un bell’appartamento! Le sta benissimo

Superlative

… una bellissima casa!

Intensifier

Lei ha proprio una bella macchina! … ma che belle topoline

Exclamation Repetition

External modifier

Preparator

Veramente veramente un… una bellissima casa! A ma’ sai che te dico?

Translation This cutlet is outstanding It’s a beautiful flat, really! It suits perfectly on you … a beautiful house! Your car is really beautiful! … what cute little girls Really really a… a beautiful house! Mom, you know what?

TABLE 3. Compliment responses Head act strategy

Type Token of appreciation Downgrading

Example Grazie

Translation Thanks

È tutto compreso nello stipendio [complimenti alle bimbe dell’interlocutore] Beh non si lasci ingannare, sapesse quanto ci danno da fare

Avoiding the topic

[complimenti all’abilità dell’interlocutrice in cucina] Sai che mi ricordi quand’eri ragazzino…

Questioning or requesting reassurance Shifting the credit

Bello eh?

It’s all included in my salary [compliments on the addressee’s girls] Well don’t be deceived, if you only knew how much work they are [compliments on the addressee’s cooking skills] You know you remind me of when you were a child… Nice, isn’t it?

Veramente il vino è un’idea di Alberto!

The wine is Alberto’s idea, actually!

Disagreeing

Comparing Textbooks and TV Series as Sources of Pragmatic Input

93

TABLE 4. Invitations Head act strategy

Internal modifier

Type Locution derivable

Example … venite da me stasera?

Hypothesis

Suggestory formula

… potevamo andare a mmh a sederci da qualche parte eh… … perché non vieni a stare qui da noi?

Query preparatory

Vuoi venire?

Explicit performative Obligation statement Requesting permission Intensifier

… siete invitati a casa mia… … dovete assolutamente venire! … te lo fai offrire un caffè? … dovete assolutamente venire! … potevamo andare a mmh a sederci da qualche parte eh… boh non lo so a bere una cosa … mi farebbe molto piacere se lei venisse

Doubter

Distance from reality External modifier

Appealer Preparator Grounder

Anticipating refusal Exhorting

Senti… … pensavo una cosa… … visto che il destino c’ha fatto reincontrare… … vabè ma chissà lei quante cose avrà da fare Dai…!

Translation … will you come to my place this evening? … we could go mmh and sit somewhere eh… … why don’t you stay here at our place? Do you want to come over? … you’re invited to my place… … you must definitely come over! … will you let me offer you a coffee? … you must definitely come over! we could go mmh and sit somewhere eh… I don’t know have a drink … it would give me great pleasure if you would join Listen… … I was thinking… … since fate made us meet again… … well but who knows how busy you are Come on…!

Chapter Four

94

TABLE 5. Responses to invitations Head act strategy

Type Non-Performative “No” Negative willingness/ability Explicit acceptance Enthusiastic acceptance

Mitigated acceptance Wish Internal modifier

Distance from reality

Superlative

External modifier

Intensifier Statement of regret/apology Expressing gratitude/appreciation Alternative

Excuse/explanation Irony

Example no…

Translation no…

… non posso

… I can’t

Sì Va bene, veniamo a cena da te…

Yes All right, we’ll have dinner at your place… Ok, but I should be working I would really like to… It would be wonderful, great!

Sì però io avrei da fare Mi piacerebbe veramente… Sarebbe meraviglioso, bellissimo! Sarebbe meraviglioso, bellissimo! … grazie mille Mi dispiace ma… … grazie, con piacere … Perché invece non andiamo a vedere la mostra…? … solo che devo prendere un treno … e faremo un bel brindisi alla disciplina!

It would be wonderful, great! … thanks a lot I’m sorry but… … please, with pleasure … Why don’t we go and see the exposition…? … but I have a train to catch … and we will drink a toast to discipline!

Only those categories for which at least one occurrence was found in my database were included in the taxonomy.

Comparing Textbooks and TV Series as Sources of Pragmatic Input

95

3.4. Discussion of results 3.4.1. Compliment sequences Table 6 shows the occurrences of the different head act strategies and modifiers that were identified in the compliments analysed. The figures reveal that TV series compliments offer a wider variety of both strategies (3 versus 2) and modification devices (5 types versus 2), and also a much higher frequency of modification devices (an average of 1.6 per dialogue, versus 0.75 in book data). However, it is worth reminding that only four1 compliment dialogues were found across the five books; therefore the comparison between the two sources must be considered with caution when compliments are concerned. TABLE 6. Compliments: occurrences of head act strategies and modification devices in books and TV Type

Head act strategy

Internal modifier External modifier

Praise of performance/skills Praise of possession Praise of appearance Superlative Intensifier Exclamation Repetition Preparator

Occurrences in books (4 dialogues) 2

Occurrences in TV series (10 dialogues) 2

3 0 2 1 0 0 0

5 5 4 5 3 3 1

Following are some examples that illustrate the difference in the variety and frequency of modifiers. In (1) and (2), taken respectively from book data and from TV data, we find similar situations – complimenting someone else’s home – but different ways of modifying the compliment. Whereas in (1) the head act contains only one internal modifier (the Intensifier proprio), in (2) the complimenter uses an Intensifier (veramente), a Repetition (veramente veramente) and a Superlative (bellissima).

Chapter Four

96 (1)

(Conversation among colleagues; S1 is a new one, coming from another country) S1: (riferendosi all’appartamento di Alberto) Proprio un bell’appartamento! [S1: (referring to Alberto’s flat) It’s a beautiful flat, really!] S2: La stanza preferita da Alberto è sicuramente il salotto: in ogni salotto c’è sempre almeno una poltrona e la TV…2 [S2: Alberto’s favourite room must be the living room: in every living room there are always at least an armchair and a TV set…]3 (Benvenuto!, Unità 4)

(2)

(Conversation between Detective Montalbano and a woman whose husband is missing) S: Mh… Posso? (esce sul terrazzo, fa un giro e poi rientra) Complimenti eh? Veramente veramente un… una bellissima casa! [S: Mh… May I? (he goes out on the terrace, takes a look around and then goes back inside) Congratulations! Really really a… a beautiful house! A: Grazie. [A: Thanks.] (Il Commissario Montalbano, Il gioco delle tre carte)

The situations are similar also in (3) and (4), where the object of the compliment is a meal prepared by the complimentee. Again, in the textbook compliment we have only one internal modifier, namely the Superlative eccezionale used by the first speaker. On the other hand, the character of the son in the TV series episode modifies his compliment with several devices (Preparator, Repetition and Intensifier), thus sounding really enthusiastic about the meal. (3)

(Conversation among colleagues at dinner: A is a new one, coming from another country) S1: Questa cotoletta è eccezionale. [S1: This cutlet is outstanding.] S2: È vero! Complimenti anche per la scelta del vino! [S2: That’s true! Congratulations on the wine choice, too!] A: Veramente il vino è un’idea di Alberto! Comunque non penso di meritare tutti questi complimenti! [A: The wine is Alberto’s idea, actually! Anyway I don’t think I deserve all these compliments!] (Benvenuto!, Unità 6)

(4)

(Conversation between son and mother at dinner) S: A ma’ sai che te dico? Buono il pollo, buoni i peperoni, bona la cicoria, bona la frittata, bono tutto brava dieci e lode!

Comparing Textbooks and TV Series as Sources of Pragmatic Input

97

[S: Mom, you know what? The chicken’s good, the peppers are good, the chicory is good, the omelet is good, everything is good, top of the class for you!] A: Sai che mi ricordi quand’eri ragazzino no? Sai che ti facevo mangiare per diventare più alto, più grosso, più sano. [A: You know you remind me of when you were a child, don’t you? You know I used to feed you so you could became taller, bigger, healthier.] (Ho sposato uno sbirro, Per un figlio)

The results concerning the compliment response head acts are presented in Table 7. No modifiers were identified in the data. We can observe only slight differences between the two sources, namely a slightly wider variety of head act strategies in the TV series (6 types versus 4 in book data). TABLE 7. Compliment responses: occurrences of head act strategies and modification devices in books and TV Type

Head act strategy

Token of appreciation Downgrading Disagreeing Avoiding the topic Questioning or requesting reassurance Shifting the credit

Occurrences in books (4 dialogues) 1

Occurrences in TV series (10 dialogues) 2

0 1 0

1 1 2

1

1

1

1

In particular, the strategies of Avoiding the topic and that of Downgrading, respectively exemplified in (4) and (5), are never used in book dialogues. (5)

(Conversation between Detective Montalbano and one of his men, sergeant Fazio) S: Bravo Fazio, andiamo! [S: Good Fazio, let’s go!] A: È tutto compreso nello stipendio commissario. [A: It’s all included in my salary, Chief.] (Il Commissario Montalbano, La luna di carta)

Chapter Four

98

On the other hand, the remaining strategies are documented in both sources, as exemplified in (3) and (6) (Disagreeing), in (2) and (7) (Token of appreciation), in (3) and (8) (Shifting the credit), and in (9) and (10) (Questioning or requesting reassurance). (6)

(Conversation between the newly arrived babysitter and her employers; she is referring to the twin girls she is going to take care of) S: Buongiorno… ma che belle topoline, quanto son brave! [S: Good morning… what cute little girls4, and so good they are!] A: Beh non si lasci ingannare, sapesse quanto ci danno da fare… vero? [A: Well don’t be deceived, if you only knew how much work they are… aren’t they?] (Ho sposato uno sbirro, Missione tata)

(7)

(Conversation between two friends at A’s birthday party) S: Buon compleanno e complimenti perché la festa è bellissima! [S: Happy birthday and congratulations for the party, it’s great!] A: Grazie, mi fa piacere. [A: Thanks, I’m glad (you’re enjoying it).] (Contatto 1, Unità 1)

(8)

(Conversation between Detective Montalbano and a woman whose husband was killed) S: Lei ha proprio una bella macchina! [S: Your car is really beautiful!] A: Grazie, me l’ha regalata Angelo. [A: Thanks, it’s a present from Angelo.] (Il commissario Montalbano, La luna di carta)

(9)

(Conversation between a clerk and a customer who is paying a jacket he has just bought) A: Ecco qua. [A: Here you are.] S: Grazie. Le sta benissimo. [S: Thanks. It suits perfectly on you.] A: Bello eh? [A: Nice, isn’t it?] S: Molto. Lo tiene su? [S: Definitely. Are you keeping it on?] (Crimini, Mork e Mindy)

(10) (Conversation between two friends) S: Bella questa credenza… moderna, grande, ci starà un sacco di roba... con i cassetti rossi come il divano!

Comparing Textbooks and TV Series as Sources of Pragmatic Input

99

[S: This cupboard is beautiful… modern, big, you may put a lot of things inside… with red drawers like the sofa!] A: Ti piace? [A: Do you like it?] S: Sì molto. [S: Yes, very much.] (Contatto 1, Unità 9, esercizi)

3.4.2. Invitation dialogues The different types of head act strategies and modification devices that were found in the invitations analysed are reported in Table 8. The figures reveal that TV invitations offer a wider variety (8 types versus 3 in textbooks) and a higher frequency of modification devices (an average of 1.7 per dialogue versus 0.6 in books), as exemplified below. On the other hand, both sources show a similar range of head act strategies. TABLE 8. Invitations: occurrences of head act strategies and modification devices in books and TV Type

Head act strategy

Internal modifier External modifier

Locution derivable Hypothesis Suggestory formula Query preparatory Explicit performative Obligation statement Requesting permission Intensifier Doubter Distance from reality Appealer Preparator Grounder Anticipating refusal Exhorting

Occurrences in books (10 dialogues) 5 0 1 3 1 1 0 1 0 0 4 0 1 0 0

Occurrences in TV series (10 dialogues) 2 2 2 3 0 0 1 1 1 2 4 2 5 1 1

In (11) and in (12), taken respectively from book data and TV data, we have similar situations, namely an invitation for dinner among people who know each other. In (11) S produces a very simple invitation, modified only by the Appealer senti. On the other hand, in (12) S prepares carefully

100

Chapter Four

his invitation, introducing the head act with an Appealer and a Preparator, and providing support with two pieces of Grounding: his skill as cook and his willingness to meet his colleague’s girlfriend. (11) (Conversation between friends) S: […] Senti… venerdì faccio una cena per il mio compleanno. Vuoi venire? [S: […] Listen… on Friday I’ll give a dinner for my birthday. Do you want to come over?] A: Mi dispiace ma non posso, ho il corso d’inglese. [A: I’m sorry but I can’t, I have an English class.] (Buona idea!, Unità 6) (12) (Conversation between two colleagues who have just been reproached by their chief for not respecting the rules) S: Senti, pensavo una cosa, venite da me stasera? Eh? Cucino da dio… poi così conosco anche… come si chiama? [S: Listen, I was thinking, will you come to my place this evening? Eh? I’m an excellent cook… and so I can also meet… what’s her name?] A: Anna. [A: Anna.] S: Eh Anna. [S: Eh Anna.] A: Va bene, veniamo a cena da te e faremo un bel brindisi alla disciplina! [A: All right, we’ll have dinner at your place and we will drink a toast to discipline!] (Medicina Generale, Oltre le regole)

Following are two more instances of similar communicative events taken respectively from book data and from TV data. Again, whereas the textbook invitation in (13) is extremely simple and does not show any emotional involvement, the one from TV material (14) sounds friendly and appealing thanks to the humorous Preparator used by the inviter. (13) (Conversation between friends in a coffee shop) S: Bevi qualcosa? [S: Are you having anything to drink?] A: Sì grazie... un caffè. [A: Yes please... a coffee.] (Contatto 1, Unità 1, esercizi)

Comparing Textbooks and TV Series as Sources of Pragmatic Input

101

(14) (Conversation between Detective Montalbano and his vice Mimì Augello in front of a coffee shop; the two haven’t seen each other for some days) S: Eh Mimì, visto che il destino c’ha fatto reincontrare, te lo fai offrire un caffè? [S: Eh Mimì, since fate made us meet again, will you let me offer you a coffee?] A: Sì però io avrei da fare. [A: Ok, but I should be working.] (Il Commissario Montalbano, Il gioco delle tre carte)

Let us see one more example. The speaker in (15) modifies her invitation through several devices (Distance from reality, Doubter, Grounder) probably because she feels embarrassed to invite a person she has just met. In a similar situation – a conversation among colleagues who do not know each other well – the inviter in the textbook dialogue (16) produces a plain invitation, with no modifiers. (15) (Conversation among two strangers who have just met during a book presentation) S: No perché potevamo andare a mmh a sederci da qualche parte eh… boh non lo so a bere una cosa così ne… ne parlavamo con calma. [S: No because we could go mmh and sit somewhere eh… I don’t know have a drink so we could talk about… about it with some calm.] A: Mi piacerebbe veramente solo che devo prendere un treno. [A: I would really like to, but I have a train to catch.] S: No infatti. [S: No you're right.] A: Mi dispiace. [A: I’m sorry.] (Crimini, La doppia vita di Natalia Bloom) (16) (Conversation among colleagues) S: […] Comunque, siete invitati a casa mia a bere qualcosa, magari sabato prossimo. [S: […] Anyway, you’re invited to my place to drink something, maybe next Saturday.] A: Volentieri! [A: With pleasure!] (Benvenuto!, Unità 4)

In sum, we observe that whereas invitations in TV series are often formulated in a very cautious and tentative way, as exemplified in (12),

Chapter Four

102

(14) and (15), in books the same speech act is mostly realised in a direct and plain manner, as we can observe in (11), (13) and (16). As far as reactions to invitations are concerned, Table 9 shows a greater variety (7 types versus 4 in textbooks) and frequency (an average of 1.7 per dialogue versus 1 in books) of modifiers in the TV materials, although the difference is less evident than for invitations. Again, both sources show a similar range of head act strategies. TABLE 9. Responses to invitations: occurrences of head act strategies and modification devices in books and TV Type

Head act strategy

Internal modifier

External modifier

Non-Performative “No” Negative willingness/ability Explicit acceptance Enthusiastic acceptance Mitigated acceptance Wish Distance from reality Superlative Intensifier Statement of regret/apology Expressing gratitude/appreciation Alternative Excuse/explanation Irony

Occurrences in books (10 dialogues) 1 1 2 5 1 0 0 0 0 3 2

Occurrences in TV series (10 dialogues) 3 1 2 2 1 1 1 2 1 2 5

2 3 0

0 4 2

In both books and TV series, when the addressee declines an invitation the act is usually accompanied by modifiers that soften the threat of a refusal, as the Statement of regret/apology and the Excuse/explanation used in (11) and in (15). On the other hand, in case of acceptance some interesting differences emerge between books and TV series. Whereas in books positive reactions to invitations are always unmodified, the same speech acts in TV series show a variety of realisations, ranging from simple acceptance, like in (17), to elaborated response. In (12), for example, the addressee supports his acceptance with a humorous reference to something that had just happened to himself and the inviter. With this reaction A seems to make the acceptance sound more enthusiastic and friendly, but also more justified: inviter and addressee have something to drink to. Similarly, in (18) A intensifies her positive reaction with several

Comparing Textbooks and TV Series as Sources of Pragmatic Input

103

modifiers (two Superlatives, a Grounder) to her response. At the same time, she presents her acceptance as distant from reality through the use of the conditional mood (sarebbe). (17) (Conversation between Detective Montalbano and a woman whose husband has been killed) S: […] Ho fatto il caffè lo beve? [S: […] I made some coffee, do you want some?] A: Sì grazie, con piacere. [A: Yes please, with pleasure.] (Il Commissario Montalbano, La luna di carta) (18) (Conversation between acquaintances; A, sister of S3, is visiting his brother and sleeping in his hotel room) S: Senti no ma perché non vieni a stare qui da noi? Dai qui c’è posto! [S: Listen but why don’t you stay here at our place? Come on, we have room!] S2: Eh ma certo aggiungiamo un letto in camera mia. [S2: Eh sure, we can add a bed in my bedroom.] S3: Grazie grazie non è il caso. [S3: Thanks thanks, you don’t need to.] A: Sarebbe meraviglioso, bellissimo! Anche perché io odio le stanze d’albergo… [A: It would be wonderful, great! Even because I hate hotel rooms…] (Il Commissario Manara, Un morto di troppo)

To sum up, the results show that the use of head act strategies for compliments and invitations, and the respective responses, is similar in both textbooks and TV series. On the other hand, TV compliment and invitation dialogues differ from their textbook counterparts in the frequency and variety of modifiers.

4. Final remarks In this contribution five L2 Italian coursebooks and five Italian TV series were compared as sources of pragmatic input with reference to the use of compliments and invitations, and their responses. The findings revealed some differences concerning both the distribution and the linguistic realisation of the speech acts under investigation. Firstly, very few occurrences of compliment dialogues were found in books, while the same speech act is well represented in TV material. It is surprising that compliments receive little attention in textbooks: given the

104

Chapter Four

importance of this speech act as “social lubricant” in daily conversations (Golato 2005), learners of Italian as a second language could take advantage from being exposed to a large amount of examples of when and how to perform a compliment in everyday life. Secondly, for both compliments and invitations, and their responses, book dialogues present a more limited variety and a lower frequency of modification devices, when compared to the TV series data. These findings are in line with those of previous studies, as for example Grant and Starks (2001), who compare conversational closings in textbooks and soap operas, and Martínez-Flor and Fernández-Guerra (2002), who compare requests, suggestions and advice acts in coursebooks and films. In conclusion, the survey of Italian teaching materials thus provides support to the claim that L2 books are unlikely to help learners develop pragmatic competence, and that teachers and textbook writers should rely less on their native-speakers’ intuitions and more on naturalistic data, or at least on TV material. However, this is only an exploratory study based on a small sample of data. Future research should expand the investigation to a larger group of Italian textbooks, including intermediate and advanced levels, a larger corpus of dialogues, and naturally occurring data as well.

References Achiba, M. 2003. Learning to Request in a Second Language. A Study of Child Interlanguage Pragmatics. Clevedon: Multilingual Matters. Alcón-Soler, E. 2005. “Does Instruction Work for Learning Pragmatics in the EFL Context?” System 33: 417-435. Alfonzetti, G. 2009. I complimenti nella conversazione. Roma: Editori Riuniti University Press. Bardovi-Harlig, K., B. S. Hartford, R. Mahan-Taylor, M. J. Morgan, and D. W. Reynolds. 1991. “Developing pragmatic awareness: Closing the conversations.” ELT Journal 45: 4-15. Bettinelli, B., P. Della Putta, and M. Visigalli. 2011. Buona idea! Corso di lingua e cultura italiana. Livello elementare. Milano: Pearson Italia. Blum-Kulka, S., J. House, and G. Kasper. 1989. Cross-cultural Pragmatics: Requests and Apologies. Norwood: Ablex. Boxer, D. 2003. “Critical issues in developmental pragmatics.” In Pragmatic Competence in Foreign Language Teaching, edited by A. Martínez-Flor, E. Usó-Juan, and A. Fernández-Guerra, 45-67. Castelló: Servei de Publicacions Universitat Jaume I. Boxer, D., and L. Pickering. 1995. “Problems in the presentation of speech acts in ELT materials: the case of complaints.” ELT Journal 49: 44-58.

Comparing Textbooks and TV Series as Sources of Pragmatic Input

105

Bozzone Costa, R., C. Ghezzi, and M. Piantoni. 2005. Contatto 1. Corso di italiano per stranieri. Milano: Loescher. Bruti, S. 2006. “Cross-cultural Pragmatics: The Translation of Implicit Compliments in Subtitles.” The Journal of Specialised Translation 6: 185-197. Chiappini, L., and N. De Filippo. 2002. Un giorno in Italia 1. Roma: Bonacci Editore. Cittadini, R., and M. Trotta. 2006. Benvenuto! L’italiano per il lavoro e la vita quotidiana. Milano: Hoepli. Conforti, C., and L. Cusimano. 2001. Linea diretta 1. Corso di italiano per principianti. Perugia: Guerra Edizioni. De Marco, A. 2011. “Insegnare la pragmatica: complimentarsi in lingue e culture distanti dall’italiano.” In Apprendere l’italiano da lingue lontane: prospettiva linguistica, pragmatica, educativa, edited by R. Bozzone Costa, L. Fumagalli, and A. Valentini, 173-193. Perugia: Guerra. Eslami, Z. 2005. “Raising the pragmatic awareness of language learners.” ELT Journal 59 (2): 199-208. Félix-Brasdefer, J. C. 2003. “Declining an invitation: A cross-cultural study of pragmatic strategies in American English and Latin American Spanish.” Multilingua 22 (3): 225-55. Fernández-Guerra, A. 2008. “Requests in TV Series and in Naturally Occurring Discourse: A Comparison.” In Learning How to Request in an Instructed Language Learning Context, edited by E. Alcón, 111126. Bern: Peter Lang. García, C. 1999. “The three Stages of Venezuelan Invitations and Responses.” Multilingua 18: 391-433. Gilmore, A. 2004. “A Comparison of Textbooks and Authentic Interactions.” ELT Journal 58: 362-374. Golato, A. 2005. Compliments and Compliment Responses: Grammatical Structure and Sequential Organization. Amsterdam: Benjamins. Grant, L., and D. Starks. 2001. “Screening Appropriate Teaching Materials. Closings from Textbooks and Television Soap Operas.” International Review of Applied Linguistics 39: 39-50. Herbert, R. K. 1990. “Sex-based differences in compliment behavior.” Language in Society 19: 201-224. Ishihara, N. 2010. “Compliments and responses to compliments.” In Speech act performance. Theoretical, empirical and methodological issues, edited by A. Martínez-Flor, and E. Usó-Juan, 179-198. Amsterdam: Benjamins.

106

Chapter Four

Kakiuki, Y. 2005. “Greetings in English: Naturalistic Speech versus Textbook Speech.” In Pragmatics in Language Learning, Theory and Practice, edited by D. Tatsuki, 61-85. Tokyo: JALT – Japan Association for Language Teaching. Kite, Y., and D. Tatsuki. 2005. “Remedial Interactions in Film.” In Pragmatics in Language Learning, Theory and Practice, edited by D. Tatsuki, 99-118. Tokyo: JALT – Japan Association for Language Teaching. LoCastro, V. 2003. An introduction to Pragmatics: Social Action for Language Teachers. Michigan: Michigan Press. Mandala, S. 1999. “Exiting advice.” In Pragmatics and language learning, vol. 8., edited by L. F. Bouton, 89-111. Urbana-Champaign: University of Illinois. Manes, J. 1983. “Compliments: A Mirror of Cultural Values.” In Sociolinguistics and Language Acquisition, edited by N. Wolfson, and E. Judd, 96-102. Rowley: Newbury House. Manes, J., and N. Wolfson. 1981. “The Compliment Formula.” In Conversational Routine. Explorations in Standardized Communication: Situations and Pre-patterned Speech, edited by F. Columas, 259-271. The Hague: Mouton De Gruyter. Martínez-Flor, A. 2008. “Analysing Request Modification Devices in Films: Implications for Pragmatic Learning in Instructed Foreign Language Contexts.” In Intercultural language use and language learning, edited by E. Alcón-Soler, and M. P. Safont-Jordà, 245-280. The Netherlands: Springer. Martínez-Flor, A., and A. Fernández-Guerra. 2002. “Coursebooks and Films in Foreign Language Teaching: A Pragmatic Approach.” Studies in English Language and Linguistics 4: 181-206. Nuzzo, E. 2007. Imparare a fare cose con le parole. Richieste, proteste, scuse in italiano lingua seconda. Perugia: Guerra Edizioni. Quaglio, P. 2009. Television Dialogue. The sitcom Friends vs. natural conversation. Amsterdam: Benjamins. Ravetto, M. 2012. “Le risposte al complimento in italiano e in tedesco. Analisi di corpora di parlato.” Studi Italiani di Linguistica Teorica e Applicata XLI: 85-122. Rose, K. R. 2001. “Compliments and compliment responses in film: Implications for pragmatics research and language teaching” International Review of Applied Linguistics 39: 309-326. Salazar Campillo, P. 2007. “Examining Mitigation in Requests. A Focus on Transcripts in ELT Coursebooks.” In Intercultural language use

Comparing Textbooks and TV Series as Sources of Pragmatic Input

107

and language learning, edited by E. Alcón-Soler and M. P. SafontJordà, 207-222. The Netherlands: Springer. Salazar Campillo, P., and E. Usó-Juan. 2002. “The Presentation of Pragmatics in Current Tourism Texts: The Speech Acts of Suggesting and Advising.” In Proceedings of the 5th International Conference on Languages for Specific Purposes, edited by F. Luttikkhuizen, 311-318. Canet de Mar: Universitat de Barcelona. Scotton, C. M., and J. Bernsten. 1988. “Natural Conversations as a Model for Textbook Dialogue.” Applied Linguistics 9: 372-384. Tatsuki, D., and M. Nishizawa. 2005. “A Comparison of Compliments and Compliment Responses in Television Interviews, Film and Naturally Occurring Data.” In Pragmatics in Language Learning, Theory and Practice, edited by D. Tatsuki, 87-97. Tokyo: JALT – Japan Association for Language Teaching. Trosborg, A. 1995. Interlanguage Pragmatics. Requests, Complaints and Apologies. Berlin: Mouton de Gruyter. Usó-Juan, E. 2008. “The Presentation and Practice of the Communicative Act of Requesting in Textbooks: Focusing on Modifiers.” In Intercultural language use and language learning, edited by E. AlcónSoler, and M.P. Safont-Jordà, 223-243. The Netherlands: Springer. Vellenga, H. 2004. “Learning Pragmatics from ESL and EFL Textbooks: How likely?” TESL Electronic Journal 8 (2): 1-18. Washburn, G. N. 2001. “Using Situation Comedies for Pragmatic Language Teaching and Learning.” TESOL Journal 10: 21-26. Weyers, J. R. 1999. “The Effect of Authentic Video on Communicative Competence.” The Modern Language Journal 83: 339-349. Williams, M. 1998. “Language Taught for Meetings and Language Used in Meetings: Is there Anything in Common?” Applied Linguistics 9: 45-58. Wolfson, N. 1981. “Invitations, compliments and the competence of the native speakers.” International Journal of Psycholinguistics 25: 7-22.

Notes 1

One of these involves three interlocutors: two complimenters and one addressee. Therefore, in this dialogue we have two compliments, performed by two different speakers, and one response (see example 4). 2 In the examples taken from written sources the original punctuation has been maintained. In the transcriptions of spoken data punctuation has been added for an easier reading. 3 The English translations of the examples were kept as literal as possible. 4 Literally, “little mice.”

CHAPTER FIVE TEACHING L2 PRAGMATICS: FROM AN EMPIRICAL STUDY TO RECOMMENDATIONS FOR PEDAGOGICAL PRACTICE PHYLLISIENNE GAUCI

1. Introduction Various studies have been conducted on the effectiveness of teaching L2 pragmatics in the past 20 years. An extensive meta-analysis on instructed L2 pragmatics studies conducted by Jeon and Kaya (2006, 180) reveals that most of these studies deal with participants whose L1 is an Asian language, with a very few studies focusing on participants whose L1 is a European language besides English. The vast majority of studies employ university students (e.g., Liddicoat and Crozet 2001; Koike and Pearson 2005; Codina Espurz 2008; Félix-Brasdefer 2008b), who are probably easier to recruit. The same views are shared a couple of years later by Alcón-Soler and Martínez-Flor (2008, 7), who state that the most common characteristics of learners taking part in these studies are that they have English or Japanese as first language and that the university learning context is the main research setting. The range of targeted languages is also quite narrow, with a concentration of research interest in English and only more recently in other European languages (e.g., FélixBrasdefer 2008a, 2008b for Spanish). Research findings suggest that pragmatics is teachable and that the teaching of pragmatics outpaces the mere exposure to the target language. Apart from that, however, findings have been contrasting and many questions, especially those related to different teaching methods, the role of the teacher, the long-term effect of instruction and the choice of appropriate outcome measures seem to remain unanswered. As a

110

Chapter Five

researcher, but also as a language teacher and teacher trainer, I see this as a real shortcoming. If our ultimate aim is to include the teaching of pragmatics in schools or in the language curricula we really should attempt to find answers to these questions in order to be able to move from theory to practice. In Section 2 of this contribution, I will outline some important theoretical grounding and will attempt to summarise what we have learnt from a number of interventional studies on interlanguage pragmatics. In Section 3, I will then report results from an interventional study conducted in Malta which explored for the first time the teaching of Italian pragmatics in the foreign language (FL) context. This study aims to answer the question of which teaching approach is the most appropriate to enable learners to access and integrate pragmatic knowledge more quickly and efficiently. Section 4 will focus on the practical aspect of the issue under investigation, that is, the teaching of L2 pragmatics in the language classroom. Recommendations for pedagogical practice will also be put forward. Finally, Section 5 presents the main conclusions of the study.

2. Rationale of the study 2.1 Early interlanguage pragmatics research studies Early research in interlanguage pragmatics (ILP) conducted in the 1980’s and in the 1990’s focused primarily on pragmatics use (Gass and Selinker 2008, 292). These early studies have produced important empirical findings, primarily through the identification and comparison of speech act realisation patterns in various languages based on data from both native speakers (NSs) and non-native speakers (NNSs) (BardoviHarlig 2001). The studies were mostly of an observational nature and focused on the process of comprehension and production of L2 pragmatics. Later research was directed towards how pragmatic abilities are acquired in an L2, thus bringing ILP more directly into second language acquisition (SLA) research. L2 pragmatic development has been analysed from two different perspectives. Observational studies look at how pragmatic competence, usually in the form of specific learning targets such as performing speech acts or using pragmatic routines, develops over time when traditional input is provided in the classroom (see Ellis 1992) or in a natural setting (see Achiba 2003). Interventional studies, on the other hand, require the implementation of a specific treatment in the classroom followed by the measurement of whatever learning has taken place.

Teaching L2 Pragmatics

111

2.2 Interventional research studies While most studies agree on the effectiveness of pragmatic instruction, the fundamental question as to what is the most effective instructional approach remains unanswered. In this respect, most studies conducted in recent years use awareness-raising treatment conditions that reflect a continuum between the absolutely explicit and absolutely implicit extremes (Jeon and Kaya 2006). Implicit instruction employs consciousness raising activities during which learners are guided towards the self-discovery of target features from a given input, without any use of metalanguage. Explicit instruction, on the other hand, is less inductive in nature and is heavily dependent on teacher-fronted explanation of pragmalinguistic forms and sociopragmatic rules. A number of interventional studies comparing the effectiveness of the two different approaches have yielded contradictory findings. In an overview of the empirical findings gathered during the past 25 years with respect to the effects of instruction on the learning of pragmatic features, Takahashi (2010) rightly points out that the superiority of explicit instruction in some studies is questioned by the fact that other studies report either mixed or inconclusive results with respect to its effectiveness. In fact, while some studies report a clear advantage for explicit over implicit instruction (see Rose and Ng Kwai-Fun 2001; Takahashi 2001; Alcón-Soler 2005), others highlight the potential of implicit instruction (see Takimoto 2007). Still other studies report either inconclusive results (see Fukuya and Clark 2001) or positive results for both forms of instruction (e.g., Koike and Pearson 2005; Martínez-Flor and Fukuya 2005). Divergent results have also been observed on the issue of long-term effects of pragmatic instruction. This can usually be assessed exclusively if the research study includes a delayed post-test following the pre-test and the post-test stages. Kasper and Rose (2002, 272) highlighted the importance of delayed post-tests by stating that these: [...] should be a standard design feature in interventional research because without their use it is not possible to determine whether the gains that students made through instruction are durable.

The infrequent use of delayed post-tests in most interventional studies remains a prevalent methodological shortcoming of a series of instructed L2 pragmatic studies. Nonetheless, those which do employ delayed posttests do reveal interesting results. Some claim that the positive results are for the most part retained (Félix-Brasdefer 2008a) or that there is retention

112

Chapter Five

for some elements but not for others (Liddicoat and Crozet 2001). Other studies reveal that overall post-test results are not retained in the long term (Koike and Pearson 2005) and record different retention effects according to the type of task in which learners participate (Koike and Pearson 2005, Takimoto 2007). Others indicate a possible relationship between learner proficiency and the retention of the positive effects of instruction (Codina Espurz 2008). The issue of whether pragmatic development is maintained, changes, or is actually lost over time remains controversial.

3. A study on the teaching of Italian L2 pragmatics In the absence of L2 pragmatic studies targeting different European languages other than English, a study1 was conducted having for the first time the teaching of Italian pragmatics in the FL context as its main objective.2 The targeted pragmatic features are Italian internal modification devices used to modify the illocutionary force of requests and complaints. The study was conducted amongst 42 high-intermediate learners having Maltese as their first language and attending a high school in Malta. The idea of moving away from the traditional university setting and employing learners of a younger age (16-18) reflects the reality in which most instructed L2 learning takes place, that is at a pre-university stage. Learners were randomly distributed in three existing Italian L2 classes, two of which were treated as experimental groups having a differentiated treatment, while the third class acted as a control group.

3.1 Pilot Study, Grammar Proficiency Test, Testing Instruments First a pilot study was conducted with an existing Italian class – later on not involved in the study proper – in the same school where the study was carried out. The learners were assigned a written discourse completion task consisting of a number of situations aimed to elicit requests, complaints and apologies. A total of 176 elicited complaints and requests were analysed. This provided us with useful insights on the strategies, modifiers and supporting acts used by Maltese NNSs of Italian, and helped us identify which modification devices were not yet used and could therefore be taught during the instructional treatment. The pilot study also ensured that subjects would respond to instructions as planned and that they were actually able to use internal modification. The pilot study was followed by a grammar proficiency test administered to the three classes participating in the study. The aim was to test the

Teaching L2 Pragmatics

113

learners’ grammatical competence and to make sure that this would not be a variable that could hinder their pragmatic production (see BardoviHarlig 2001, 14; Bettoni 2006, 221; Ellis 2008, 196; Kasper 2001a, 504510; Rose 2000 on contrasting views regarding the relationship between pragmatic and linguistic development). The test gave us a clear indication that the learners were already familiar with some targeted structures such as the imperfect and the conditional and that they were able to use them at least in their primary (i.e., grammatical, not pragmatic) function. All learners participating in the study proved to be successful in the grammar proficiency test. Next, a pre-test was conducted through a battery of tests. These consisted of a written discourse completion task (DCT), a role-play (RP) and a multiple choice discourse completion task (MCDCT). The first two tasks were used to test our learners’ written and oral pragmatic production, while the third one was used to test their pragmatic awareness. The situations presented in the tasks aimed at eliciting requests, complaints and apologies –the last of which being used as distractors– and alternated between high-level and low-level of speaker difficulty3 (see Appendix 1 for an example of each task). The tests were administered to the two experimental classes as well as to the control class.

3.2 Instructional Treatment The pre-test was immediately followed by the instructional treatment. This was carried out in six weeks, and consisted of weekly lessons of approximately 55 minutes each. Half of the instruction time was devoted to the teaching of requests, while the other half to the teaching of complaints. Instruction was carried out by the same teacher in all the participating classes. In the two treatment classes, instruction involved the use of film segments and audiovisual material containing request and complaint exchanges drawn from a corpus of NS film speech acts. Consistency was maintained across both treatment classes by providing the same input, with the only independent variable being the type of instruction. In the explicit (EXPL) class, the teacher gave out metapragmatic information and instruction was based on teacher-fronted explanations of pragmalinguistic forms and sociopragmatic rules. In the implicit (IMPL) class learners were in a position to infer rules on their own since the treatment was inductive in nature and involved the application of inputenhancement techniques and awareness raising tasks with specific questions to guide the students’ discovery of pragmatic patterns. In

114

Chapter Five

contrast to what happened in the EXPL class, in the IMPL class the teacher did not give out any metapragmatic information. In order to operationalise explicit and implicit instruction, we adopted Housen and Pierrard’s (2005) definition of implicit and explicit form-focused instruction based on a comparison of a number of characteristics as outlined in Table 1. An attempt was made to be as faithful as possible to the definitions throughout the instructional treatment. Whenever this was not deemed possible, a strong attempt was made to be more or less faithful to them.4 The third class, the control (CONT) group, carried on with its usual language classes, which were mostly grammar-based, consisting mainly of written comprehension and written production tasks, as is the norm in the Maltese school attended by the learners. TABLE 1. Instructional treatment (adapted from Housen and Pierrard 2005, 10) Explicit Instruction Directs attention to target form + Is predetermined and planned (e.g., as the main focus and goal of a teaching activity) + Is obtrusive (interruption of communicative meaning) +/Presents target forms in isolation +/Uses metalinguistic terminology (e.g., rule explanation) + Involves controlled practice of target form +

Implicit Instruction Attracts attention to target form + Is delivered spontaneously (e.g., in an otherwise communication-oriented activity) +/Is unobtrusive (minimal interruption of communicative meaning) + Presents target forms in context + Makes no use of metalanguage + Encourages free use of the target form +/-

The targeted pragmatic features were request and complaint internal modification devices. These were selected from amongst the most frequently used request and complaint modifiers in a NS corpus collected by Nuzzo (2007). The taxonomy used (see Appendix 2) was elaborated by drawing on previous literature and from taxonomies used for languages other than Italian (e.g., Blum-Kulka, House, and Kasper 1989; Trosborg 1995; Sifianou 1999; Alcón-Soler, Safont-Jordà, and Martínez 2005) as well as on the taxonomy elaborated by Nuzzo (2007), which was modified after analysing the teaching practice of Italian as a FL. The six-week instructional treatment was immediately followed by a post-test during which the learners were tested with the same instruments utilised in the pre-test (a DCT, a RP and a MCDCT). A delayed post-test

Teaching L2 Pragmatics

115

was conducted four months later in order to verify whether the positive effects of the instructional treatment were retained in the long-term.

3.3 Learner and Native Speaker Corpora At the end of the three data collection sessions, a total of 504 written requests and complaints performed by the members of the three groups were collected thanks to the DCT. A total of 252 orally performed requests and complaints were collected from the RP task. The participants’ responses indicating their metapragmatic awareness throughout the three testing stages were also collected thanks to the MCDCT, in which participants were first presented with a situation and then asked to choose the most pragmatically appropriate answer. All oral production was audio recorded and transcribed. Additional data included the 12 audio recorded and transcribed treatment lessons, six lessons with the EXPL class and six with the IMPL class.5 The aim of the study was to investigate whether learners could improve their L2 pragmatic ability to use request and complaint modification devices and get closer to the behaviour of NSs of the target language, that is, the NS norm. Thus, along with the learner corpus, a NSs reference corpus was also set up. For the oral production task, the data previously collected by Nuzzo (2007) was used. This data consisted of 161 Italian NSs’ requests and complaints elicited by means of the same RP situations used in the present data-collection task. Since Nuzzo’s data is oral, a further cohort made up of 46 NSs of the same age as the learners were recruited to complete the DCT, and another 38 NSs completed the MCDCT. This was deemed necessary since later on all learners’ data was to be analysed in light of NSs’ data. In the next section, results from the pre-test and the post-test are presented and compared in order to examine the effects of the treatment as well as the differential effects of explicit and implicit instruction. Results on the pre-test (T1), post-test (T2) and delayed post-test (T3) are then compared in order to examine the differential retentive effects of the treatments on the target structures.

3.4 Findings and discussion The data collected through the two production tasks, the DCT and the RP was analysed through a counting of tokens of modifiers in both the learners and the NS datasets. Results indicated that at T1, the three learner groups taking part in the study used on average fewer modifiers compared

116

Chapter Five

to the NSs when performing their requests and complaints.6 In the written DCT, the NSs used on average 0.91 modifiers per head act while, in the same situations, the learners, on average, only used 0.49 modifiers per head act. In the RP, the NSs use on average 1.12 modifiers per head act against an average use of 0.41 modifiers per head act by the learners. This result is in line with previous research which reports that traditionally NSs use a higher number of modifiers than NNSs even at advanced stages of their interlanguage (Faerch and Kasper 1989; Rose 2000; Hassall 2001; Otcu and Zeyrek 2006; Bardovi-Harlig and Hartford 1993; Trosborg 1995; Nuzzo 2007, 2009, 2010). Results also showed that at T2 there was a statistically significant increase in the mean number of modifiers used per head act in the two treatment groups (IMPL and EXPL), while it remained practically the same for the CONT group.7 A positive effect of instruction could be observed in that the learners not only used a higher number of modification devices, but also expanded the variety of modification devices they used when passing from the pre-treatment to the posttreatment stage. The negative interrogative for example was practically unknown to the learners at T1, but was widely used at T2. Table 2 shows the frequency and type of modifiers used by learners at T1 and at T2. Results show marginal differences on the relative effectiveness of the teaching methods adopted, that is explicit vs. implicit instruction, with a slight advantage of implicit instruction. TABLE 2. Frequency and range of modifiers used by the learners at T1 and at T2 Frequency and Range of Modifiers EXPL IMPL Type Sub-type T1 T2 T1 T2 Lexical intensifiers 14 16 6 9 Modifiers softeners 4 7 1 8 politeness markers 22 21 6 12 doubters 0 0 0 0 subjectivisers 1 7 3 6 Syntactic negative interrogative 0 20 1 8 Modifiers conditional 2 4 2 14 imperfect 0 0 0 0 Total number of modifiers 43 75 19 57

CONT T1 T2 10 13 3 8 26 19 0 0 3 4 0 3 8 10 0 0 50 57

In an attempt to identify whether the increase in number of modifiers was mostly a result of progress in the learners’ written or oral production,

Teaching L2 Pragmatics P

117

the results w were analysedd further by lo ooking specifiically at the number n of modifiers ussed in the two separate tasks. Graph 1 indicates thee average num mber of modiifiers used by y the two treatment grroups exclusivvely in the DC CT task, whille Graph 2 thee average number of m modifiers usedd by the two treatment grooups exclusiveely in the RP task. In Graph 1, we see that in the DCT task thhere was a staatistically significant iincrease in thhe number of modifiers ussed from T1 to t T2 for both the EX XPL and the IM MPL groups. In Graph 2, hhowever, an important i difference eemerges. The increase in th he mean numbber of modifieers taught during the ttreatment sesssions between n T1 and T2 in the oral prroduction task was noot statistically significant fo or either grouup (although the t IMPL class appearrs to have obttained the bestt results). Thuus, the overalll increase in the number of modifierrs used can on nly be attributeed to an increase in the number of m modifiers in thhe written task k and not in thhe oral producttion task.

p < 0.05

p < 0.05

modifiers per act

1 0.8 T1

0.6

T2

0.4 0.2 0 EXPL L

IMPL

GRAPH 1. Increase in numbeer of modifiers used u - DCT resuults

Chapter Five

118

modifiers per act

p > 0.05

p > 0.05

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

T1 T2

EXPL L

IMPL

GRAPH 2. Increase in numbeer of modifiers used u - RP resultts

When loooking at the data qualitatively and com mparing it witth the NS data, some iinteresting finndings emerge. In Table 2 , we saw thatt learners showed a ppreference forr certain mod difiers over oothers. The politeness p marker, forr instance, was w very pop pular amongsst the learneers when mitigating their requestts and comp plaints both bbefore and after the instructionall treatment. This contrassts significanntly with thee use of politeness m markers by the t NSs. Furrthermore, thee learners sh howed an overall prefference for lexxical modifierrs as opposedd to what we observed among the IItalian NSs, who w generally y opted for syyntactic modiffiers over lexical ones. This comes as no surprisee. Results from m other studiees such as Faerch and K Kasper (1989), Barron (200 03) and Salazzar-Campillo (2008) ( all indicate the overuse of pooliteness mark kers such as “pplease” by NN NSs, who use them wiith the dual fuunction of ind dicator of the iillocutionary force and of mitigator. Additionnally, we notiice that whilee the use of m modifiers waas heavily situation-deppendent for the t NSs, who o appeared too be very sen nsitive to social variabbles, the same did not hap ppen for the N NNSs. In example (1), relevant to tthe situation Camicia da cambiare c (Chhange of shirt), we see that the learnners at T2 maade excessive use of modifiiers when mak king their request andd this resulteed in requessts which w would probablly sound unnatural too an Italian NS. N On the otther hand, thee NSs in this situation used very feew modifiers, as shown in example (2), indicating thaat the NS

Teaching L2 Pragmatics P

119

is aware thhat the salespperson has a duty to channge the shirtt and the customer haas the right to ask for this ch hange. (1) Mi sccusi, non è ch he mi potrebbee fare il piacerre di chambiarmi la caminncia perfavorre perchè ho o preso la misura sbaglliata?! (EXP PL/camicia/DCT T/T2) (Excuuse me, you coouldn’t change this shirt for m me could you please becauuse I bought it inn the wrong sizze?!)8 (2) Buonggiorno, mi cambbia la camicia? ? (NS/camicia/D DCT) (Goodd morning, willl you change the shirt for me?))

Thereforre, while an inncrease in the use of modifiiers might ind dicate that we are appproaching NS S behaviour (since it ressults in requ uests and complaints that are expressed less explicitly and lless aggressiv vely), this does not neccessarily meann that we are truly t approachhing the NS no orm. Graph 3 shows the average a numb ber of modifiiers used by the three learner grouups in both tasks t at T1, at T2 and att T3 and is useful to determine w whether the eff ffects of the in nstructional treeatment are reetained in the long-term m.

modifiers per act

p < 0.05

p < 0.05 5

p > 0.05

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

T1 T2 T3

EXPL

IMPL L

CO ONT

GRAPH 3. Aveerage number of o modifiers by the three learneer groups at T1, at T2 and at T3

The grapph indicates a sudden deecrease in thee use of mod dification devices eviddent in both thhe IMPL and the t EXPL classses at T3. Some of the positive effeects of instrucction which were w clearly viisible from T1 1 to T2 in

120

Chapter Five

both experimental groups were therefore retained after four months, but not all of them. In the CONT group, where no specific instruction on the use of modification devices took place, the use of modifiers per act registered at T3 approximates the amount of modifiers used at T3 in the two experimental classes. One, however, must keep in mind that the CONT class had a slight advantage on the other groups already at T1. On conducting a one-way ANOVA test we noticed that the increase in the number of modifiers from T1 to T3 was statistically significant for the two treatment classes, but not for the CONT group. The significance is probably to be attributed to the considerable increase in the use of modifiers from T1 to T2. Table 3 shows the MCDCT results for the NS group as well as for the three learner groups. Two observations can be made here. First, the learners’ responses reveal an increase in the number of correct responses from T1 to T3 in all three learner groups. A one-way ANOVA test, however, showed that after the instructional treatment there was a significant development in the learners’ ability to recognise pragmatically appropriate requests and complaints in the treatment classes, but not in the CONT class. This improvement was more evident for the learners in the IMPL group, who obtained higher scores than those of the EXPL group and who showed a constant progress from T1 to T3. The overall results obtained from the metapragmatic awareness data therefore seem comparable to those obtained from the production data. Some improvement was recorded in the learners’ ability to recognise pragmatically appropriate utterances in the treatment groups, but no significant improvement was detected among the learners in the CONT group. Furthermore, although in all situations the majority of the NSs chose the appropriate response, only in one situation, Valutazione errata (Wrong grading), did all the NSs appear to agree on the same response. This means that some of the “inappropriate” responses were in fact still attractive enough to be chosen even by some NSs. These results cast some doubt on the reliability of the MCDCT (see Fukuya and Clark 2001; Koike and Pearson 2005; Kasper and Roever 2005), especially due to the difficulty for the testers to develop responses that are unquestionably acceptable.

Teaching L2 Pragmatics

121

TABLE 3. MCDCT results

The results of this study have provided us with a number of interesting findings that contribute to a better understanding of pragmatic development amongst learners of Italian L2. We notice how in spite of an increase in the frequency and range of modifiers used by the learners after the instructional treatment, in most cases the use of modification devices

122

Chapter Five

was not situation dependent and the choice of modification devices used did not always coincide with the native-speaker norm. Lexical modifiers kept being chosen over syntactic modifiers and the learners commonly resorted to an excessive use of politeness markers such as per piacere and per favore (please). The learners found it hard to discriminate among different situations. They were less sensitive to social variables than the NSs, and tended to be polite to the same degree in all situations even after the instructional treatment. This may be due to the fact that the instructional treatment has failed to highlight the importance of the sociopragmatic aspect, which must be considered when using modification devices, or maybe the acquisition of sociopragmatic aspects requires more time than the acquisition of pragmalinguistic aspects.9 Kasper (2001b, 51-52) points out that “unlike inappropriate pragmalinguistic choices, which can often be identified within one utterance or turn, sociopragmatically infelicitous selections may be more difficult to locate.” The learner may choose a politeness style that sounds to the interlocutor or teacher as too strongly oriented towards negative or positive politeness. The study has shown that a significant increase in the use of modification devices was recorded mostly when the learners were tested through a written DCT, but that the same progress was not observable when the learners were tested through a RP. This result can be easily accounted for by the fact that the written mode leaves more time for reflecting on language production. The written DCT on its own may shed light on the learners’ acquired knowledge of pragmatic features, but might not reflect their online procedural skills.

4. Recommendations for pedagogical practice On the basis of the findings illustrated in the previous section, below are a few practical pedagogical recommendations for the teaching of pragmatics in the FL classroom. i. Maximising the use of authentic, audio-visual material The acquisitional context in SLA cannot be underestimated, since it determines the input which the language learners are exposed to, as well as the output and the feedback opportunities available to them. The study which has just been presented was conducted in a FL context, that is in a context with scarce opportunities for exposure to target pragmatic norms (input) and an impoverished environment for practice (output). In this context, teachers are often the only models of appropriate behaviour

Teaching L2 Pragmatics

123

(Martínez-Flor and Usó-Juan 2010, 10; Roever 2009) and learners are often not exposed to social situations and social roles. The findings of the study suggest that learners, especially those learning a language in a FL context, may be less sensitive to the types of politeness strategies or the types of modifiers available in the linguistic system and appropriate to specific contexts, and may not have the ability to choose among them the way NSs do. Authentic input is indispensable in this setting in order to raise learners’ awareness of the variety and contextual appropriateness of modifiers. Teachers should use authentic audio-visual material during their lessons as much as possible in order to expose learners to the real NS linguistic practices. ii. Highlighting the importance of social variables When presenting social interactions or when assigning communicative tasks to their learners it is crucial that teachers highlight the importance of social variables such as social distance, power difference and rank of imposition. The teaching of pragmatics must always take in account the participants and the context in which the interaction is taking place. Asking simple questions such as “Am I making the request to a sibling, to a friend or to a complete stranger? Am I complaining to the waiter/shop assistant or the boss? How valuable is the object that I am asking for or how grievous is the issue that I am complaining about?” helps language learners become more attentive to social variables and more confident when deciding on which politeness strategy to adopt. iii. Adopting different teaching methods Results showed marginal differences on the relative effectiveness of the teaching methods adopted, explicit and implicit instruction. In the DCT task there was a statistically significant increase in the number of modifiers used from T1 to T2 for both the EXPL and the IMPL group, whereas in the RP task the average increase in the use of modifiers taught during the treatment sessions between T1 and T2 was not statistically significant for either group. Results on the relative effectiveness of a teaching method over another therefore are not so clear-cut, although, on the whole, they seem to slightly favour implicit instruction. This could attest either to the minor role played by the teaching method or to the hardto-pin-down distinction between explicit and implicit instruction in SLA. Whichever the case, the study supports what Takahashi (2010, 137-138) has stated, namely that perhaps in pragmatics these two forms of intervention must be conceptualised on a continuum rather than being treated as binary and distinct concepts.

124

Chapter Five

iv. Presenting a variety of different situations and using discourse completion tasks and role-plays In the L2 classroom there needs to be time for students to express themselves and also for practice. Traditional language exercises are not necessarily the best way to give valid feedback regarding L2 pragmatics. An innovative approach is necessary, such that students are given the opportunity to use the language productively and creatively, otherwise they run the risk of not mastering the various pragmatic aspects of the language. Communicative tasks, in which learners are active agents in the communicative situation and during which they alternatively play the roles of speaker and listener may therefore be useful. This could also be followed by discussion on the sociopragmatic and pragmalinguistic appropriateness of utterances. The type of situations presented to the learners could be such as to elicit a large variety of speech acts including some face-threatening ones such as complaints, requests and apologies. Some acts are more difficult to handle than others. Performing an appropriate complaint, for instance, is a more arduous task than performing a request; in fact, the former speech act less frequently resorts to conventional formulae, is intrinsically more complex, and requires more attention in handling what are typically its two different components, the expressive component and the directive component (see Gauci 2012a). It is important that learners practice communicative tasks in both the written and the oral mode. Through a written DCT, the teacher gains access to information about the offline pragmatic knowledge of learners, whereas dialogues elicited through RP allow the teacher to observe learners’ online pragmatic knowledge.

5. Conclusion Each language has its set of resources or tools which we refer to as modification devices and which can be used for conveying particular communicative acts. This study has focused on the teaching of internal lexical and syntactic modifiers used to modify the illocutionary force of requests and complaints. There are, however, various other strategies which can be used in the realisation of speech acts, such as directness and indirectness, the use of pragmatic routines or optional supportive acts such as starters, which can also help to modify the effect of the performed speech act. The results of this study have confirmed previous research that NSs tend to use a higher number of modifiers than NNSs, and this often makes NNSs seem inappropriately over-assertive or domineering. Modification devices, however, can be taught and learnt.

Teaching L2 Pragmatics

125

The findings of the present study have shown that pragmatics is teachable and that the teaching of pragmatics outpaces the simple exposure to the target language. This is evidenced in an increase in frequency and variety of modification devices from the pre-test stage to the post-test stage in the two experimental classes. Such increase was not recorded in the control group. The positive results of instruction observed in production were also observed for pragmatic awareness. After the instructional treatment there was a significant development in the learners’ pragmatic ability to recognise pragmatically appropriate requests and complaints in the treatment classes, but again not in the control class. Pragmatic instruction is therefore beneficial, confirming previous studies such as Safont and Alcón 2000, Yoshimi 2001, Fukuya and Zhang 2002, Alcón 2005, Rose 2005, and Félix-Brasdefer 2008a, 2008b. The study has also shown that despite the increase in frequency, the learners’ choice of modifiers did not always conform to the NS norm. Thus, notwithstanding instruction, learners might not take into account situational and social factors present in the act of communication unless these are clearly pointed out. We saw that there is little difference between the teaching approach used (implicit vs. explicit) and that pragmatic instruction can benefit from using different methods along the implicitexplicit continuum. Finally, we also saw that results are highly influenced by the testing instrument used and that the positive effects of the instructional treatment are not always retained in the long-term, hence the importance of conducting a delayed post-test to check the long-term effectiveness of one’s approach/es. These findings have helped us understand that as classroom teachers we need to make changes in approaches to L2 teaching and to select a variety of teaching methods and pragmatically appropriate input if we really want our learners to become truly proficient in communicating in the target language. Unfortunately, pragmatics still plays a minor role in language classrooms, textbooks and course materials, and there is the need to probe further interventional studies in order to incorporate the teaching of L2 pragmatics as part of a set curriculum.

126

Chapter Five

References Achiba, M. 2003. Learning to request in a second language: A study of child interlanguage pragmatics. Clevedon: Multilingual Matters. Alcón-Soler, E. 2005. “Does instruction work for learning pragmatics in the EFL context?” System 33: 417-35. Alcón-Soler, E., M. Safont-Jordà, and A. Martínez. 2005. “Towards a typology of modifiers for the speech act of requesting: A sociopragmatic approach.” RæL, Revista Electrónica De Lingüística Aplicada 4: 1-35. Alcón-Soler, E., and A. Martínez-Flor. 2008. “Pragmatics in foreign language contexts.” In Investigating pragmatics in foreign language learning, teaching and testing, edited by E. Alcón-Soler, and A. Martínez-Flor, 3-21. Bristol: Multilingual Matters. Bardovi-Harlig, K. 2001. “Evaluating the empirical evidence: Grounds for instruction in pragmatics?” In Pragmatics in language teaching, edited by K. R. Rose, and G. Kasper, 13-32. Cambridge: Cambridge University Press. Bardovi-Harlig, K., and B. S. Hartford. 1993. “Learning the rules of academic talk. A longitudinal study of pragmatic change.” Studies in Second Language Acquisition 15: 279-304. Barron, A. 2003. Acquisition in interlanguage pragmatics: Learning how to do things with words in a study abroad context. Amsterdam: Benjamins. Bettoni, C. 2006. Usare un’altra lingua. Roma: Laterza. Blum-Kulka, S., J. House, and G. Kasper. 1989. Cross-cultural pragmatics: Requests and apologies. Norwood, NJ: Ablex. Brown, P., and S. Levinson. 1987. Politeness: Some universals in language use. Cambridge: Cambridge University Press. Codina Espurz, V. 2008. “The immediate vs. delayed effect of instruction on mitigators in relation to the learner’s language proficiency in English.” In Learning how to request in an instructed language learning context, edited by E. Alcón-Soler, 227-56. Bern: Peter Lang. De Marco, A. 2011. “Insegnare la pragmatica: complimentarsi in lingue e culture distanti dall’italiano.” In Apprendere l’italiano da lingue lontane: prospettiva linguistica, pragmatica, educativa, edited by R. Bozzone Costa, L. Fumagalli, and A. Valentini, 173-93. Perugia: Guerra. Ellis, R. 1992. “Learning to communicate in a classroom: A study of two learners’ requests.” Studies in Second Language Acquisition 14: 1-23.

Teaching L2 Pragmatics

127

—. 2008. The Study of Second Language Acquisition. Oxford: Oxford University Press. Faerch, C., and G. Kasper. 1989. “Internal and external modification in interlanguage request realization.” In Cross-cultural pragmatics: Requests and apologies, edited by S. Blum-Kulka, J. House, and G. Kasper, 221-47. Norwood, NJ: Ablex. Félix-Brasdefer, J. C. 2008a. “Pedagogical intervention and the development of pragmatic competence in learning Spanish as a foreign language.” Issues in Applied Linguistics 16 (1): 49-84. —. 2008b. “Teaching pragmatics in the classroom: Instruction of mitigation in Spanish as a foreign language.” Hispania 91 (2): 479-94. Fukuya, Y. J., and M. K. Clark. 2001. “A comparison of input enhancement and explicit instruction of mitigators.” In Pragmatics and language learning, edited by L. Bouton, 111-29. Urbana, IL: University of Illinois. Fukuya, Y. J., and Y. Zhang. 2002. “The effects of recasts on EFL learners’ acquisition of pragmalinguistic conventions of requests.” Second Language Studies 21 (1):1-47. Gass, S. M., and L. Selinker. 2008. Second language acquisition: An introductory course. New York: Routledge. Gauci, P. 2012a. “Insegnare a protestare in italiano L2.” In Competenze e formazione linguistiche. In memoria di Monica Berretta. Atti dell’XI Congresso Internazionale di Studi della Associazione Italiana di Linguistica Applicata (AItLA), edited by G. Bernini, C. Lavinio, A. Valentini, and M. Voghera, 383-96. Perugia: Guerra Edizioni. —. 2012b. “Teaching pragmatics in Italian L2: An empirical study in a foreign language context.” PhD dissertation, Università degli Studi di Verona, Italy. Hassall, T. 2001. “Modifying requests in a second language.” International Review of Applied Linguistics 39: 259-83. Housen, A., and M. Pierrard. 2005. “Investigating instructed second language acquisition.” In Investigations in instructed second language acquisition, edited by A. Housen, and M. Pierrard, 1-27. Berlin: Walter de Gruyter. Jeon, E., and T. Kaya. 2006. “Effects of L2 instruction on interlanguage pragmatic development: A meta-analysis.” In Synthesizing research on language learning and teaching, edited by J. Norris, and L. Ortega, 165-211. Amsterdam: Benjamins. Kasper, G. 2001a. “Four perspectives on L2 pragmatic development.” Applied Linguistics 22 (4): 502-530.

128

Chapter Five

—. 2001b. “Classroom research on interlanguage pragmatics.” In Pragmatics in language teaching, edited by K. R. Rose, and G. Kasper, 33-60. Cambridge: Cambridge University Press. Kasper, G., and C. Roever. 2005. “Pragmatics in second language learning.” In Handbook of research in second language teaching and learning, edited by E. Hinkel, 317-34. Mahwah, NJ: Lawrence Erlbaum. Kasper, G., and K. R. Rose. 2002. Pragmatic development in a second language. Malden, MA: Blackwell. Koike, D. A., and L. Pearson. 2005. “The effect of instruction and feedback in the development of pragmatic competence.” System 33 (3): 481-501. Liddicoat, A. J., and C. Crozet. 2001. “Acquiring French interactional norms through instruction.” In Pragmatics in language teaching, edited by K. R. Rose, and G. Kasper, 125-44. Cambridge: Cambridge University Press. Martínez-Flor, A., and J. J. Fukuya. 2005. “The effects of instruction on learners’ production of appropriate and accurate suggestions.” System 33: 463-80. Martínez-Flor, A., and E. Usó-Juan. 2010. “Pragmatics and speech act performance.” In Speech act performance: Theoretical, empirical and methodological issues, edited by A. Martínez-Flor, and E. Usó-Juan, 320. Amsterdam: Benjamins. Nuzzo, E. 2007. Imparare a fare cose con le parole. Richieste, proteste, scuse in italiano lingua seconda. Perugia: Guerra. —. 2009. “Ho bisogno dell’informazione per andare a Barcellona. Uno studio longitudinale sulle richieste di informazioni e suggerimenti in Italiano L2.” Linguistica e Filologia 28: 83-109. —. 2010.“Richiedere in italiano L1 e L2: Strategie di modulazione della forza illocutoria.” In La Comunicazione Parlata 3. Atti Del Congresso Internazionale, Napoli, 23-25 febbraio 2009 vol. I, edited by M. Pettorino, A. Giannini, and F. M. Dovetto, 513-532. Napoli: Università degli studi di Napoli L’Orientale. Nuzzo, E., and P. Gauci. 2012. Insegnare la pragmatica in italiano L2. Recenti ricerche nella prospettiva della teoria degli atti linguistici. Roma: Carocci. Otcu, B., and D. Zeyrek. 2006. Requesting in L2: Pragmatic development of Turkish learners of English. Duisburg: LAUD. Roever, C. 2009. “Teaching and testing pragmatics.” In The Handbook of Language Teaching, edited by M. H. Long, and C. J. Doughty, 560-75. Oxford: Wiley-Blackwell.

Teaching L2 Pragmatics

129

Rose, K. R. 2000. “An exploratory cross-sectional study of interlanguage pragmatic development.” Studies in Second Language Acquisition 22 (1): 27-67. —. 2005. “On the effect of instruction in second language pragmatics.” System 33: 383-99. Rose, K. R., and C. Ng Kwai-Fun. 2001. “Inductive and deductive teaching of compliments and compliment responses.” In Pragmatics in language teaching, edited by K. R. Rose, and G. Kasper, 145-70. Cambridge: Cambridge University Press. Safont, M. P., and E. Alcón. 2000. “Pragmatic competence and foreign language instruction. A study on the use of request strategies by English adult learners.” In Proceedings of the 24th International Conference of AEDEAN, edited by A. Mateos-Aparicio, and S. Molina, 61-65. Castilla la Mancha: Publicaciones de la Universidad de Castilla la Mancha. Salazar-Campillo, P. 2008. “Task analysis on mitigation in the speech act of requesting: Discourse completion task and role-play.” In Learning how to request in an instructed language learning context, edited by E. Alcón-Soler, 143-61. Bern: Peter Lang. Sifianou, M. 1999. Politeness phenomena in England and Greece. A cross-cultural perspective. Oxford: Oxford University Press. Takahashi, S. 2001. “The role of input enhancement in developing pragmatic competence.” In Pragmatics in language teaching, edited by K. R. Rose, and G. Kasper, 171-99. Cambridge: Cambridge University Press. —. 2010. “The effect of pragmatic instruction on speech act performance.” In Speech act performance: Theoretical, empirical and methodological issues, edited by A. Martínez-Flor, and E. Usó-Juan, 125-42. Amsterdam: Benjamins. Takimoto, M. 2007. “The effects of input-based tasks on the development of learners’ pragmatic proficiency.” Applied Linguistics 30 (1): 1-25. Thomas, J. 1983. “Cross-cultural pragmatic failure.” Applied Linguistics 4 (2): 91-112. Trosborg, A. 1995. Interlanguage pragmatics: Requests, complaints and apologies. Berlin: Mouton de Gruyter. Yoshimi, D. R. 2001. “Explicit instruction and the use of interactional discourse markers.” In Pragmatics in language teaching, edited by K. R. Rose, and G. Kasper, 223-44. Cambridge: Cambridge University Press.

130

Chapter Five

Appendix 1 Example of a DCT situation10 Camicia da cambiare (Change of shirt) Hai comprato una camicia per tuo padre, ma quando lui la indossa ti rendi conto che la misura è sbagliata. Torni al negozio dove l’hai comprata e cosa dici alla commessa? (You bought a shirt for your father, but you realise that it has the wrong size. You go back to the shop from where you bought it. What do you say to the salesgirl?) Tu (You): ....................................................................................................... Commessa (Salesgirl): Sì, va bene. Ha conservato lo scontrino? (Yes, ok. Have you kept the receipt?) Example of a RP situation ‘Furto’ di penna (‘Stolen’ pen) Ruolo A (Role A) All’inizio della lezione hai prestato una penna a un/a compagno/a di classe. Adesso la lezione è finita e tu vedi che il/la tuo/a compagno/a di classe sta mettendo la tua penna nella sua borsa. Cosa dici? (You lent a pen to your classmate at the beginning of the lesson. Now the lesson is over and you realise that your classmate has just put the pen in his/her bag. What do you say?) Ruolo B (Role B) All’inizio della lezione un/a compagno/a di classe ti ha prestato una penna. Adesso la lezione è finita e tu per errore stai mettendo la penna nella tua borsa. (Your classmate lent you a pen at the beginning of the lesson. Now the lesson is over and by mistake you are just putting the pen in your bag.) Example of an MCDCT situation Carrello sbagliato (Wrong trolley) Sei al supermercato. A un certo punto vedi che una persona sta mettendo le sue cose nel tuo carrello.Quale fra le due reazioni ti sembra la più delicata? (You are at the supermarket. At one point someone mistakenly places his own items into your trolley. Which of the two is the most gentle reaction?) A: ehm mi scusi, ehm sarebbe il mio carrello questo (A: excuse me, this would be my trolley) B: ehm mi scusi, ehm è il mio carrello questo (B: excuse me, this is my trolley)

Teaching L2 Pragmatics

131

Appendix 2 Request and complaint modifiers chosen for the instructional treatment Type Lexical Modifiers

Sub-type intensifiers softeners

politeness markers doubters subjectivisers Morpho-Syntactic Modifiers

negative interrogative conditional

imperfect

Examples davvero, proprio ecc. (really, definitely, etc.) un po’, un attimo, una monetina ecc. (a bit, a moment, small change, etc.) per favore, per piacere (please) forse, magari ecc. (maybe, etc.) credo, penso, immagino ecc. (I think, I imagine, etc.) non è che…? (can’t you...?, wouldn’t you...?) vorrei ..., potresti ...? ecc. (I would like..., could you...?, etc.) volevo ...ecc. (I wanted..., etc.)

Notes 1

The full version of the study can be found in Gauci (2012b). The issue of pragmatic development for Italian has to my knowledge only been addressed in two studies so far, both conducted in the second language (SL) context; an observational study conducted by Nuzzo (2007) and a study on pragmatic development resulting from pedagogical intervention in De Marco (2011). 3 Because pragmatic appropriateness is context and interlocutor dependent, situations in each task were chosen with reference to their categorisation within Brown and Levinson’s (1987) framework, which describes the three variables of social distance, power, and degree of imposition acting as universal constraints on linguistic action. The situations presented thus alternated in terms of degree of familiarity, participants’ position in a hierarchical structure and weight of imposition. 4 The + and +/- symbols have been added in the table and indicate, respectively, where the instructional treatment was completely faithful vs. partially faithful to the definition. 2

132

Chapter Five

5 For an extract of the transcribed lessons with the EXPL and with the IMPL classes, see Nuzzo and Gauci (2012, 100-102) and Gauci (2012a, 391-392). 6 Due to space constraints, results are being presented as the summing up of modifiers used in both speech acts of complaining and requesting. Results, however, can be viewed separately for the two speech acts in Gauci (2012b). 7 A number of examples of utterances produced by the learners in the pre-test and in the post-test stages as well as a thorough statistical analysis of the data can be viewed in Gauci (2012b). 8 In the examples taken from written sources, the original punctuation has been maintained. English translations were kept as literal as possible. 9 Thomas (1983) distinguished pragmalinguistic failure, which occurs when a learner tries to perform the right speech act but uses the wrong linguistic means (i.e., s/he deviates with regard to appropriateness of form) from sociopragmatic failure, which takes place when a learner fails to perform the illocutionary act required by the situation (i.e., s/he deviates with regard to the appropriateness of meaning). Appropriate communicative competence involves both appropriateness of meaning and appropriateness of form. 10 English translations of the situations are being provided for the benefit of nonItalian readers and are kept as literal as possible. These translations were not included in the original task.

SECTION 2 TESTING AND ASSESSING

CHAPTER SIX DESIGNING INSTRUCTIONAL EFFECT STUDIES FOR L2 PRAGMATICS: A GUIDE FOR TEACHERS AND RESEARCHERS KATHLEEN BARDOVI-HARLIG

1. Introduction This paper provides a practical guide for the development of research designs for studying the effect of instruction on the development of L2 pragmatics in both host and foreign environments. Building on accepted practices and standard recommendations in the study of effects of instruction in second language acquisition research more broadly, I will illustrate standard practices as they apply to pragmatics research and will take into account factors related uniquely to pragmatics. Pragmatic competence includes knowledge of how to interpret and perform speech acts, conversational management, discourse organization, turn taking, implicature, negotiation, pragmatic routines, and sociolinguistic aspects of language use such as choice of address terms. Research in pragmatics often distinguishes between pragmalinguistics – the language resources speakers use for pragmatic purposes – and sociopragmatics – the rules that guide use of language in context. I often tell my students that instructional effect studies are among the most challenging studies to design because they require both strong research design and good teaching. Researchers and teachers should know going into an instructional effect study that it is always possible that the teaching component could have been better. If the response to instruction is less robust than expected, longer instruction, better input, more practice, extended listening, or higher fidelity recordings (to name a few areas that have given us difficulty) are often suggested as modifications. This challenging design is made more challenging in pragmatics because the ways of teaching pragmatics are still being developed, although that also allows for flexibility and creativity.

136

Chapter Six

The number of instructional effect studies has grown since Rose (2005) reviewed 25 articles (from 1986 to 2005). More than 80 articles were published between 2000 and mid-2013 in journals and serial publications (Bardovi-Harlig 2015). The goals of instructional effect studies in pragmatics appear to be twofold: to determine what means of instruction promote acquisition of L2 pragmatics through classroom learning broadly defined, and to promote the teaching of pragmatics in second and foreign language classrooms worldwide. This paper attempts to provide a guide for researchers and teachers interested in conducting such studies and also includes a review and synthesis of instructional effect research in pragmatics. Given that somewhat immodest goal, a few caveats are in order. I consider the multiple components of the design of instructional effect research in a linear fashion and treat them as largely independent. I am fully aware that they may be neither. Features often cluster as do ideas; one choice may set off a chain of choices; choosing one instructional setting over another may determine variables as diverse as students and length of instruction. The core idea for a study may stem from watching learners struggle with pragmatics or from the availability of a new technology that lends itself to pragmatics instruction. The components that I present below are meant as a checklist, not a guide to ordered planning. Further, this paper is not a comprehensive review of the field; I have attempted to illustrate my observations and suggestions with previous studies so that the reader will have examples to follow, but have not exhaustively cited all the studies reviewed in Bardovi-Harlig (2015). (For a recent comprehensive review, see also Takahashi 2010).

2. The components of an instructional effect study The most important step in any research project is determining the research question and developing a research design to address that question. A variety of research questions can be asked by studies that investigate the effect of instruction on the development of secondlanguage pragmatics (Rose 2005). In what follows, I outline the components that would be found in most instructional effect studies and discuss standard practice in second language acquisition research for designing them. The major steps in designing an instructional effect study include determining the target of instruction, the participants (the students, the teachers, and the absentee students), the instruction (the type of pragmatic input, the method(s) of instruction, practice, duration, class format, control

Designing Instructional Effect Studies for L2 Pragmatics

137

groups, and documentation of teaching), evaluation (timing of evaluation [pretest, posttest, delayed posttest], means of evaluation, and evaluation of gains both linguistically and statistically), writing up the report, and reporting back to the teachers or program that hosted the study. Each stage will be illustrated with examples from published papers.

2.1. Determining the target of instruction Every study of the effect of instruction must identify a target for instruction. It may be something that is claimed to be difficult, something that has not been previously taught, or something that has not been taught well. In pragmatics, there are many options in each of these categories. Sources that identify instructional targets include acquisition research that identifies the areas in which learners have difficulty and the local language program curriculum. There may also be various forms of needs assessment, and these can include teacher observation of student pragmatic performance and student requests for help with pragmatic aspects of the target language. In our work in an English as a Second Language (ESL) program, we identified instructional targets from research and the curriculum. In one study, we identified the target through research. A study of recognition and use of conventional expressions in pragmatics demonstrated that learners either did not recognize the conventional expressions (such as No problem, That’d be great, and Do you have a minute) or recognized them, but did not know how to use them (Bardovi-Harlig 2009). This led us to develop teaching units for conventional expressions in social conversation (Bardovi-Harlig and Vellenga 2012). In a second study at the suggestion of a teacher, we worked with the low-advanced level curriculum of the program and targeted agreements and disagreements in academic discussion (Bardovi-Harlig, Mossman, and Vellenga 2015). Because clarifications often occur in the context of disagreements (Bardovi-Harlig and Salsbury 2004; Pomerantz 1984), we added clarifications to our instructional target. We worked with agreement expressions such as That’s true and You’re right, disagreements such as I agree…but, and related acts of clarification and rephrasing, such as What I mean and What you’re saying. Reviews of textbooks have shown that both ESL and English as a Foreign Language (EFL) textbooks lag well behind teachers’ interest in teaching pragmatics (Alcón-Soler 2008; Bardovi-Harlig et al. 1991; Boxer and Pickering 1995; Cheng and Cheng 2010, inter alia.) Many targets that are potentially useful to learners have yet to be explored pedagogically.

Chapter Six

138

2.2. Participants: students, teachers, and missing subjects Three roles are crucial to instructional effect studies: the students, the teachers, and the absentees. The design must take into account all three, especially the latter, whose lack of participation can quickly reduce the number of students who are ultimately included in the study. Although no researcher knows until the very end what the absentee rate will be, it is best to plan for it in the early stages to assure a sufficient number of students to evaluate the instruction. 2.2.1. The students Many teacher-researchers create experimental instructional units for the benefit of students who are enrolled in language programs in their own institutions. Other researchers may arrange to have their experimental lessons taught at another institution or program, especially when looking for more students than can be found at their home institution, or when looking for students who have particular characteristics such as specific first language, experience abroad, proficiency level, or age. Once learners are identified, it is important to establish their level of proficiency whether by institutional placement, standardized test (such as the TOEFL (Test of English as a Foreign Language) or the OPI (the Oral Proficiency Interview (OPI)), or by other means. A background questionnaire should be implemented to identify previous language experience and current language engagement that could augment or detract from the instruction and to facilitate comparison of learners. Researcher-teachers who implement their studies in their own institution should also include a background questionnaire, even if they feel familiar with the type of student the program attracts. Many background questionnaires exist (see for example the Language Contact Profile, Freed et al. 2004); additionally, questionnaires sensitive to pragmatics might include questions about study abroad and travel in areas where the target language is spoken, as well as extended contact with target-language speakers (Bardovi-Harlig and Dörnyei 1998). Engagement with target-language media (movies, TV, radio, podcasts, music, internet, and other computer-mediated forms of communication) as well as targetlanguage social media may also be relevant to pragmatics (Bardovi-Harlig and Bastos 2011). Reports of study abroad investigations are generally good sources of questionnaires, and Shively (2010) provides an example of a relevant set of questions used in a study that examined instructional effects in a study abroad setting. A thorough description of the students

Designing Instructional Effect Studies for L2 Pragmatics

139

allows other researchers and teachers to compare their students to the ones in the report. 2.2.2. The teacher(s) In instructional effect studies the researcher(s) and the teacher(s) are often different people. The teacher may be a co-author after the fact, but does not plan the study or instigate the research. There are two main reasons to divide the researcher and teacher roles. First, recruiting one’s own students may be seen as coercive. Students may be afraid to decline, but the option to decline to participate is a central tenet of informed consent. Whether intentionally or unintentionally, teachers may disfavor students who declined to participate or who did not participate wholeheartedly by not scoring well, not paying attention to instruction, or not attending all instructional sessions. Second, the researcher-as-teacher knows the design of the study and the intended outcome. Again, whether intentionally or unintentionally, the researcher-as-teacher may “teach to the test,” potentially leading the students to desired outcomes. Takimoto (2008a, footnote 1, 382) discussed this issue: teaching all the classes for four conditions himself, he reflected that “[i]n behavioral research, researcher expectancy can be a problem when the researcher teaches experimental groups.” As a result, he followed instructional guidelines to assure that the approaches were distinct and analyzed blinded data (i.e., the instructional group was not labeled) in order to prevent researcher bias toward the expected outcome. An alternative route is for researchers to recruit teachers. When using intact classes (classes that meet regularly in a language program), my colleagues and I recruit the regular teacher. Using the teacher who is already assigned to the class lends authenticity to the pragmatics instruction and better integrates it into classroom practices. If that is not possible because the researcher is the instructor of the class that will receive the experimental instruction, for example, then a guest teacher may give the lesson, or classes that meet at the same time can be combined. There are two approaches to assigning teachers when different instructional conditions are used. One is to arrange for different teachers to teach different instructional conditions so the teaching is distinct as possible. This avoids teachers “borrowing” between conditions, but may also introduce extraneous variables such as teacher personality. The second is to assign the same teacher to multiple conditions and to provide training and checklists for each condition and to follow up with

140

Chapter Six

observation, thus controlling extraneous variables while attempting to limit borrowing. Engaging other teachers to teach experimental classes means that the researcher and/or materials developers must meet with the teachers to explain both the materials and the instruction (but not the expected outcome). One of our materials developers met with the teachers to give them the materials and to explain the activities and materials for each of the four lessons they were asked to teach (Bardovi-Harlig, Mossman, and Vellenga 2015). Eslami and Liu (2013) developed the lessons and sent them to the teacher in Taiwan; they also gave the lessons to the graduate tutors to use for teaching through Computer-Mediated Communication (CMC). Koike and Pearson (2005) gave the teachers of the experimental groups explicit directions for how the experiment should be carried out and trained the teachers on the modeling and feedback procedures required for the study. Németh and Kormos (2001) developed opinion gap activities whose directions were worded by the research group, but were presented to the students by their regular teachers. Teachers are often interested in participating in experimental instruction. Researchers can enhance the experience by discussing the materials with the teachers, and possibly sharing the materials with the teachers after the study. 2.2.3. The absentees: planning for attrition Most instructional units in pragmatics do not take place on a single day (see section 2.3). That being said, however, the more meetings there are, the higher the chances of attrition. The students must at least attend the pretest and the posttest to be included in the analysis. Ideally, learners also complete the delayed posttest if there is one, although studies often report a reduced number of participants for the delayed posttest. In addition to attending both pretest and posttest, learners must also attend the majority – if not all – of the instructional sessions. Although researchers are often reluctant to exclude learners and thereby “lose” data, it is important to keep in mind that students do not learn from being assigned to a group, they learn from the instruction itself. It is true that deleting students who did not attend results in a smaller learner population, but it also yields the desired participant sample: students who have received the experimental treatment! One of the steps is to determine how many treatments or instructional class sessions a student must attend in order to be included in the study and thus be retained for analysis. Fordyce (2014) and S. Li (2011) required students to complete all the instructional sessions. In our

Designing Instructional Effect Studies for L2 Pragmatics

141

most recent study (Bardovi-Harlig, Mossman, and Vellenga 2015) which was conducted in regular meetings of communication classes in an intensive English program, we used a three-out-of-four lesson cutoff. Our student rosters went from an initial enrollment of 36 students in five ESL classes to 28 students who attended both the pretest and posttest and at least three of the four classes. Attendance by the control group, which took only the pretest and the posttest, went from 16 students at the pretest to 11 at the posttest. In an earlier study (Bardovi-Harlig and Vellenga 2012), we used a two-out-of-three cutoff for attendance in class, and 44 students of the original 66 satisfied the attendance requirements. Of the 44 eligible students, only 36 successfully recorded their oral responses. About 25% of the studies surveyed by Bardovi-Harlig (2015) reported attrition of students. Félix-Brasdefer (2008a, 2008b) reported that 16 of the 23 experimental students completed the study and similarly in the control group, 16 of 22 completed the tasks; students dropped the class or did not attend the posttest. Fordyce (2014) began with 143 students at a national university in Japan. However, in the final analysis, only the data from 81 students who took part in all parts of the study (i.e., they completed all three test batteries, attended all the classes, and completed all the tasks), and who had not spent more than 6 months living in an English-speaking country, were included. As the reader will note, not attending all the instructional sessions and not completing all the tests led to attrition, whereas not having lived in an English-speaking environment was a screening criterion. Either way, fewer students than anticipated were ultimately included in the study. More modest attrition was reported by Eslami and Liu (2013, 57), where only 118 of 130 available Taiwanese undergraduate students were included in the analysis because participants were absent for part of the treatment or for the pretest or posttest. Taking into account studies that reported attrition, planning for a 2025% rate of attrition would give a researcher a safe margin (cf. the studies discussed in this section and Fukuya and Clark 2001; Fukuya and Martínez-Flor 2008; Fukuya and Zhang 2002, 2006; S. Li 2011; Nyugen, Pham, and Cao 2013; Takahashi 2001; Taylor 2002).

2.3. Instruction Instruction here is understood broadly to encompass what both teachers and students do. Language instruction has many components, as all teachers know. For the purposes of planning an instructional effect study I have broken instruction down into six components which reflect separate choices on the part of the researcher: the type of input, method of

142

Chapter Six

instruction, duration of instruction, practice (if included), class format, and documentation of teaching. The discussion reflects the goal of improving conversation which is shared by most approaches to teaching pragmatics in instructional effect studies (Bardovi-Harlig 2015); the emphasis on conversation as a target mirrors research in interlanguage pragmatics more generally (Bardovi-Harlig 2010). Type of input and type of practice are of particular concern in pragmatics instruction because, along with the task used for evaluation, they require the operationalization of conversation (Bardovi-Harlig 2015). How conversation is represented in input, how it is practiced, and how it is evaluated will all likely impact the outcome of a study. 2.3.1. The type of pragmatic input Once the target is determined, the source of input must be identified. The type of pragmatic input for instruction is crucial because pragmatically relevant input is what is missing in most instructional contexts (Bardovi-Harlig 2001). A recent review of over 80 instructional effect studies in pragmatics (Bardovi-Harlig 2015) suggests that input to learners varies along two dimensions: modality and authenticity. Modality includes audio, audio-visual, and written input as well as combinations of audio-visual with written transcripts. Authenticity is a continuum, but points along the continuum that have been illustrated by recent studies include authentic, authentic-scripted, elicited, scripted, and pedagogical materials. Each of these can be aural, audio-visual, or written. With few exceptions, the pedagogical goal of most of the teaching units is to effect changes in the students’ pragmatics for spoken conversation. Going beyond the traditional pedagogical input written by textbook writers or researcher-teachers, instructional effect studies have used a variety of audio and audio-visual sources for input, illustrating not only what is possible, but what has been implemented by other teachers. Authentic-scripted input is found in movies, television, radio, or podcast shows that exist in the real world apart from instruction (thus authentic) but are scripted by the show’s writers. In written format, authentic-scripted examples are novels, short stories, magazines, and commercial websites. Studies that have used elicited language samples ask native speakers to perform tasks or role plays for the purposes of producing a model. Other studies have also used innovative pedagogical materials like Destinos (a 52-episode soap opera for students). Television shows used for instruction have included Friends, a situation comedy whose transcripts are readily available on the internet

Designing Instructional Effect Studies for L2 Pragmatics

143

with similarly easily accessible video segments (see for example, AlcónSoler 2012, and Silva 2003). Other TV shows have also been used and these include an adventure series, Stargate (Alcón-Soler 2005, 2007; Alcón-Soler and Guzmán-Pitarch 2010), a Spanish-language talk show, Cristina (Taylor 2002), and a reality show Super Nanny (Codina-Espurz 2008). Excerpts from films have also been used (Martínez-Flor 2008; Rose and Ng 2001). Some presentations of authentic conversation are made exclusively through written transcripts (Huth 2006; Liddicoat and Crozet 2001), and others have re-recorded selected transcripts from corpora to provide audio examples that are clearer for instructional purposes than the originals (Bardovi-Harlig, Mossman, and Vellenga 2015; Holmes and Riddiford 2010; Riddiford 2007; Riddiford and Joe 2010). Other studies have asked native speakers to perform the same task as the learners. These include mock job interviews (Louw, Derwing, and Abbott 2010), and a video of people asking Mr. Money for money (Fukuya and Clark 2001). Given the relative ease of working with digital audio-visual input and the importance of providing EFL learners with spoken models that they might not encounter outside the classroom – and ESL learners with models of language that they might not otherwise pay attention to – I would encourage teacher-researchers planning studies to look to audio-video input for conversational pragmatics. The greater principle is to match mode of input with intended mode of production. 2.3.2. The method(s) of instruction There is no single approach to teaching pragmatics, and even within the more common approaches, no single way of implementing an approach. The teaching approach clearly interacts with the type of input selected and whether or not practice is implemented, but these are still independent decisions. Instruction in pragmatics reflects interest in instructional approaches more generally, but also shows specific sensitivity to pragmatics. Several studies have compared explicit and implicit approaches to teaching with targets including suggestions (Fukuya and Martínez-Flor 2008; Martínez-Flor and Alcón-Soler 2007; Martínez-Flor and Fukuya 2005; Pearson 2006), requests (Alcón-Soler 2005, 2007), epistemic stance (Fordyce 2014), constructive criticism (Nguyen, Pham, and Pham 2012), and formulaic expressions (Tateyama 2001). Three- and four-way comparisons including implicit and explicit teaching have also been conducted. Q. Li (2012) compared the efficacy of

144

Chapter Six

three types of instruction on the teaching of requests: implicit Focus on Form (FonF) with visual enhancement, explicit FonF, and an input-output condition with no formal component which may allow incidental learning. Koike and Pearson (2005) investigated the effects of explicit and implicit pre-instruction and implicit and explicit feedback in four combinations to teach suggestions; these included a) Explicit pre-instruction + explicit feedback, b) Explicit pre-instruction + implicit feedback, c) Implicit preinstruction + explicit feedback, and d) Implicit pre-instruction + implicit feedback. Takahashi (2001) compared explicit, form comparison, form search, and meaning focus presentations of requests. Other comparisons include FonF and Focus of FormS in teaching requests (Fuyuka and Clark 2001), and inductive and deductive approaches to teaching compliments and compliment responses (Rose and Ng 2001), and requests (Martínez-Flor 2008). Not all comparisons involve “brand name” teaching approaches. Other studies have investigated a variety of approaches for their suitability in teaching pragmatics including comparisons of face-to-face teaching to computer-mediated instruction (Eslami and Liu 2013), form search to form comparison (Takahashi 2005b), consciousness raising to consciousness raising with metalinguistic feedback (Barekat 2013), a textbook-based presentation to the textbook-based presentation enhanced by consciousness raising (Tateyama 2007), and more input to less input (e.g., comparing the efficacy of presenting 16 dialogues to 8 dialogues; S. Li 2011). Other comparisons have included oral synchronous computer mediated communication (SCMC) to written SCMC for teaching refusals (Sykes 2005) and video with reconstruction to video with comprehension for teaching expressives and directives (Jernigan 2012). Many studies do not compare two approaches, but rather test the effect of instruction using a single approach. These studies implemented crosscultural comparison of refusals (Félix-Brasdefer 2004), metapragmatics, input, and consciousness-raising activities for requests (Codina-Espurz 2008), focused noticing of conventional expressions guided by metapragmatic questions (Bardovi-Harlig and Vellenga 2012), focused noticing of agreement and disagreement expressions in academic discussion guided by meta-pragmatic questions, metapragmatic information, and production practice (Bardovi-Harlig, Mossman, and Vellenga 2015), consciousness raising for hearsay evidential markers (Narita 2012), and noticing of biclausal requests (Takahashi 2005a). Several additional studies explore explicit approaches to teaching a variety of speech acts including requests (Halenko and Jones 2011; Holmes and Riddiford 2010; Safont-Jordà 2003), criticism (Nguyen 2013; Nguyen, Pham, and Cao 2013),

Designing Instructional Effect Studies for L2 Pragmatics

145

interactional discourse markers (Yoshimi 2001), job interviews and culture (Louw, Derwing, and Abbott 2010), suggestions (Fernández-Guerra and Martínez-Flor 2006), thanking (Ghobadi and Fahim 2009), requests and apologies (Eslami and Eslami-Rasekh 2008), and requests, apologies and complaints (Eslami-Rasekh, Eslami-Rasekh, and Fatahi 2004; Mirzaei and Esmaeili 2013). Cohen and Shively (2007) tested the effect of strategy training on requests and apologies in study abroad settings. As can be seen, the method of instruction reflects on-going pedagogical creativity. This is particularly true in pragmatics, where possibilities for instruction are as rich as the phenomena encompassed by the discipline. 2.3.3 Practice Some studies include a practice phase which follows the presentation of input, metapragmatic information, or other aspects of treatment. Practice is advocated by Shively (2010), following Cohen (2005), BarrajaRohan (2000), and authors of several chapters in Pragmatics: Teaching Speech Acts (Tatsuki and Houck 2010). If included, practice affords an opportunity for learners to engage in simulations of conversation. In a review of how conversation is operationalized in instructional effect studies in pragmatics at three points – input, practice, and evaluation tasks – I found that the most creative activities are often found during practice (Bardovi-Harlig 2015). Even non-production studies included production activities as practice (Alcón-Soler 2002; Eslami-Rasekh, Eslami-Rasekh, and Fatahi 2004; Martínez-Flor and Alcón-Soler 2007; Tateyama 2007). The level of creativity in practice sessions may be due to the fact that these sessions do not have to be graded or otherwise scored. From the perspective of research design, keep in mind that practice potentially adds additional time on task and so if one goal is to assess the effect of practice by one group, the same amount of time should be added to the activities of the non-practice group. 2.3.4. The duration of instruction Duration of instruction is always a challenge, given the desire to achieve multiple goals in a language class. Researchers whose instructional targets are aligned with the program curriculum or the course syllabus have an advantage in securing class time. It is possible to find instruction of all lengths (not including testing sessions), and more than any other feature of studies discussed here, length is a continuous variable.

146

Chapter Six

Dividing the length variable, there seem to be “very short” instructional periods that run in total from one to two hours (Huth 2006; Koike and Pearson 2005; Narita 2012; Salazar 2003; Trosborg and Shaw 2008), “short” instructional periods which meet for complete class sessions for 24 weeks (Bardovi-Harlig, Mossman, and Vellenga 2015; Codina-Espurz 2008; Fordyce 2014; Takahashi 2005a, 2005b; Takimoto 2008a, 2008b), to “half-semester” distributed lessons which take place over 6-10 weeks (Alcón-Soler 2012; Belz and Kinginger 2003; Halenko and Jones 2011; Nguyen, Pham, and Pham 2012; Vyatkina and Belz 2006), and fullsemester delivery with sustained instruction at least once a week (AlcónSoler 2005; Nguyen, Pham, and Cao 2013; Riddiford and Joe 2010; Yoshimi 2001), and even full semester treatment in special communication courses devoted to pragmatic fluency and metalinguistic awareness (House 1996). Delivery of instruction can also be viewed as intensive (e.g., seven 50-minute sessions for seven consecutive days, Fukuya and Zhang 2006) and distributed (e.g., 50-minute sessions once a week for 10 weeks, Eslami and Liu 2013; or 30 minutes a week for 18 weeks, Ghobadi and Fahim 2009), or even a combination (e.g., six 2-hour sessions over 19 weeks, Fukuya and Martínez-Flor 2008). The most important consideration in determining the length of instruction is a realistic assessment of how long the lesson(s) would take under non-experimental conditions, and how they would best be distributed across class meetings. As guests in other teachers’ classrooms, researchers often reduce the estimated time of instruction in order to secure classroom time; in the case of lab studies or simulated classrooms, researchers attempt to make participation more appealing to students and thus shorter, but both potentially lead to retrospective evaluations that the instruction could have been longer. Reducing the instructional component could potentially reduce the impact of instruction for both the student (whom the study seeks to benefit) and the study itself. The best approach is an ecologically realistic one for the instructional setting. 2.3.5. Class format When we think of classes, we often think of established classes in a program or school. When we work with such classes, these are called intact classes. They show all the diversity and heterogeneity of classrooms everywhere. Working with intact classes has high ecological validity, as intact classes embody the multiple variables at play in institutional settings. Studies that used intact classes include Alcón-Soler (2012, inter

Designing Instructional Effect Studies for L2 Pragmatics

147

alia), Bardovi-Harlig, Mossman, and Vellenga (2015), Félix -Brasdefer (2008a, 2008b), and Nguyen (2013). Researchers may employ a lab setting or establish experimental groups that function like classes for a variety of reasons (Eslami and Liu 2013; S. Li 2011; Louw, Derwing, and Abbott 2010; Narita 2012; Németh and Kormos 2001; Sardegna and Molle 2010; Takimoto 2008a, 2008b; Taylor 2002; van Compernolle 2011). Occasionally, programs are not open to curricular innovations or do not allow modifications to a pre-planned syllabus (see, e.g., Takimoto 2008b, 8). In other cases, a researcher may wish to pre-select students for certain variables such as language proficiency, language background, or language experience more carefully than in non-experimental settings, or they may wish to distribute learner variables carefully balancing them across experimental and control groups. Takimoto (2008a, 2008b) offers a detailed description of how he established his class-like groups, including the recruitment process, payment of subjects, and securing a location. Web-based self-access instruction can be conducted anywhere and does not require a classroom, or other students, or a lab setting, and can monitor students’ completion of units or time at task. Teng and Fei (2013) created self-access web-based instruction to increase students’ awareness of the use of pragmatic formulas across speech acts in Chinese. They provided six units of 10-15 authentic pragmatic scenarios which learners completed over two months in sessions of 30 minutes to one hour. Cohen and Ishihara (2005) also developed self-access materials for teaching speech acts in Japanese, and Cohen and Sykes (2010) reported on the use of two websites, Dancing with Words and Croquelandia (a synthetic immersive environment) to teach Spanish requests, apologies, and service encounters. 2.3.6. The control group Instructional effect studies use control groups to determine whether instruction (alone) had an influence on learners and to rule out other possibilities. There are at least two types of control groups in the pragmatics literature: test only and input without treatment. Implementing a test-only control group addresses the question of whether taking the test twice improves a learner’s score. Because tasks can be learned, or at least become more familiar, this type of control group eliminates the possibility that a student’s score improves by just taking the test a second time. This type of control group is robustly represented in the pragmatics literature (e.g., Bardovi-Harlig, Mossman, and Vellenga 2015; Eslami and Liu 2013;

148

Chapter Six

Halenko and Jones 2011; Koike and Pearson 2005; Nguyen 2013; Nguyen, Pham, and Pham 2012; Nguyen, Pham, and Cao 2013; Pearson 2006; van Compernolle and Williams 2013). It works well when both the experimental group and the control group are enrolled in a language program and receive instruction for equal amounts of time (one group receives the specialized instruction, and the other group receives instruction according to the regular syllabus). A version of this is used in constructed classes by presenting a lesson on something unrelated in order to simulate alternative language instruction; for example, Takimoto (2008a, 2008b) offered TOEIC (Test of English for International Communication) practice to the control group. A second type of control group receives the same input as the group receiving the specialized instruction, but it does not get the treatment. In this approach, all groups get the same input, but only the experimental group(s) get the treatment that accompanies the input. Treatment includes enhancement, pre-listening questions, metapragmatic instruction, or roleplay practice. This simulates a comparison between a possible “natural condition” where learners encounter input but have no guidance and an instructed condition where learners are given input plus guidance. Studies that have used this type of control group include Eslami-Rasekh, EslamiRasekh, and Fatahi (2004), Eslami and Eslami-Rasekh (2008), FélixBrasdefer (2008a, 2008b), Fukuya and Zhang (2006), Ghobadi and Fahim (2009), Németh and Kormos (2001) and Sykes (2005). Another type of design compares two (or more) different instructional approaches to each other. In this type, there is often no control group (Rose 2005), although the use of a control group is not uncommon. For example, Alcón-Soler (2005) had two treatment groups, an explicit group that viewed video clips from a television series and received explicit training, and an implicit group that viewed the same video clips with input enhancement, and a control group that received the same input, but no additional treatment (thus constituting a control group of the second type). 2.3.7. Documentation of teaching Instructional effect studies often document the instruction. The idea is to make sure that the instructional components that are planned are those that are delivered or that the approach described by the teachers is the one that is used in class. We can do this in at least two ways: through direct observation and through checklists. Pearson (2006,) made audio recordings, and her account points out the importance of talking to the teachers who are conducting the classes:

Designing Instructional Effect Studies for L2 Pragmatics

149

All the course sections were taught by different teachers. For the class sections in Groups A and B, I provided the teachers with a teacher’s guide featuring instructions and a presentation script for the Spanish SA [speech act] lessons. Audiorecordings were made of the Spanish SA lessons to verify uniformity of the content presented to the learners. I consulted with the teachers teaching the Group C sections to ensure that they were strictly following the departmental syllabus and not duplicating material from the Spanish SA lessons. (Pearson 2006, 476).

Alcón-Soler (2002) asked one of the participating teachers to observe the lessons. Alcón-Soler (2005, 2007) and Alcón-Soler and GuzmánPitarch (2010) had teachers observe the lessons in order to document possible bias toward one of the groups. (Note that determining bias is particularly important if a single teacher teaches multiple instructional approaches, as discussed in “The teacher(s)” in Section 2.2.) Vellenga (2008) developed checklists to assure the fidelity of the lessons across teachers and sections. Because teachers completed a checklist to record each activity as it was accomplished, the checklists not only provided a record of what teachers did, but it also reminded them to include all the planned components. We adopted this approach and found it useful in addition because teachers included notes as well (BardoviHarlig and Vellenga 2012; Bardovi-Harlig, Mossman, and Vellenga 2015).

2.4. Evaluation 2.4.1 Pretests, posttests, and delayed posttests Designing an instructional effect study minimally requires a pretest and posttest. Rose (2005) reviews some early pragmatics studies that did not use pretests (e.g., Fukuya and Clark 2001; Tateyama et al. 1997; Wishnoff 2000), but it is now standard practice to do so. The main purpose of a pretest is to establish a baseline before instruction. Without one, neither researchers nor teachers can tell how much the learners improve after instruction. All studies have a posttest. In general, the posttest occurs shortly after the instruction is completed to capture what learners know as a result of instruction. Some studies offer the posttest immediately after instruction, sometimes even the same day, and others allow one to two days following the end of instruction. Delaying too long can allow learners to encounter additional input beyond instruction. Delayed posttests are used to determine whether learners retain what they learned during the instruction (see Takahashi 2010 for extended

150

Chapter Six

discussion). If learners show no or little retention over time, this might suggest that teaching needs to be cyclic, a concept promoted in different teaching approaches. However, a delayed posttest that shows that learners have retained their high scores (thus, presumably retaining the knowledge demonstrated during the posttest) may reflect not only effects of instruction, but also effects of intervening input from a variety of classroom and non-classroom sources (such as print media, TV, internet, and conversations). Generally, the longer the time between a posttest and the corresponding delayed posttest, the greater the opportunity for unplanned input. The range of time between the first posttest and a delayed posttest in pragmatics studies runs from a brief 2-3 weeks (AlcónSoler 2007; Félix -Brasdefer 2008a, 2008b; S. Li 2011), to a middle range of 5-8 weeks (Codina-Espurz 2007; Halenko and Jones 2011; Nguyen 2013; Nguyen, Pham, and Pham 2012) to the longer range of 6-12 months (Liddicoat and Crozet 2001; Pearson 2006). One additional consideration with a delayed posttest is convincing participants to return for a follow-up test. It is important to plan for the delayed posttest in the same semester or the same academic year, if possible. If an instructional study is planned and executed early in a standard semester, there may be time for a delayed posttest during the semester when students can still be located. Once the semester or session has ended it is much harder to gather the original learners. Difficulty gathering learners is amplified over summer vacation. Liddicoat and Crozet (2001) were able to test 6 of 10 learners after one year. In our seven-week sessions, we did not have time to run a delayed posttest before students left for summer, and we were able to retest only 3 of our 37 learners during the summer, thus effectively eliminating the delayed posttest from the design (Bardovi-Harlig, Mossman, and Vellenga 2015). 2.4.2. Tasks used for evaluation Much has been written about elicitation tasks in pragmatics research (see, e.g., Bardovi-Harlig 2010, 2015; Kasper and Rose 2002; Jeon and Kaya 2006). Although it is beyond the scope of this paper to review the full range of tasks here, this section reviews some of the factors to consider when selecting a task. The task used for evaluation should reflect the targeted mode of communication (this echoes the point made in “Instruction”). If the instruction addresses pragmatic features for conversation, then learners should be evaluated on speaking. If the instruction focuses on noticing pragmatic features in speech, then the task should be aural; and if the

Designing Instructional Effect Studies for L2 Pragmatics

151

instruction focuses on pragmatic features in written communication, the task should be written. Good matches between teaching and testing have greater face validity than mismatches. For example, Belz and Kinginger (2003) and Belz and Vyatkina (2005) used a written-for-written task for both input and evaluation. They arranged for their American students of German to engage in telecollaboration with native German speaking peers, and thus the input they received and their output for evaluation was produced in the same context and mode. The paring of video-clips from television shows as input with role plays for evaluation is one example of audio-visual input and spoken output (Alcón-Soler 2012; Alcón-Soler and Guzmán-Pitarch 2010; Silva 2003). Video-taped native speaker-native speaker (NS-NS) interactions as input with role plays as the task for evaluation is a second pairing (Martínez-Flor and Alcón-Soler 2007). Other examples of audio or audiovisual input paired with oral tasks for evaluation were used by Nguyen (2013), Nguyen, Pham, and Pham (2012), Holmes and Riddiford (2010), and Sardegna and Molle (2010). An additional consideration of selecting a task is to allow learners to demonstrate specific knowledge that they would not be able to demonstrate otherwise. Learners who have had opportunities to hear speech or to have oral practice may demonstrate an ease with production not found in learners who have not, and the task should be designed to reflect that; an oral task would capture spoken features, whereas a written task would not. Related to this, tasks should also allow individuals to perform. Consider discussion groups which represent an authentic context for academic language programs: the turns of one learner may prevent other learners from making a contribution if they think their point has already been made. Thus a simulated discussion where individual learners respond would better reveal changes for each learner (Bardovi-Harlig, Mossman, and Vellenga 2015). Thus, mode and opportunities to demonstrate knowledge are two key features in selecting tasks for instructional effect studies in pragmatics; general practices of task design also pertain.

152

Chapter Six

2.4.3. Evaluation of gains Measurement is closely related to task, but has not been discussed in pragmatics to the same degree as task has. Because studies of pragmatics investigate language use in context, almost any aspect of language, context, interaction, setting, or consequence is relevant for evaluation. Both quantitative and qualitative means of evaluation may be used. Nonproduction tasks often ask learners for interpretation, metapragmatic judgments, ranking, rating, comprehension, identification, and calculation of implicature. The measurement of performance on nonproduction tasks tends to straightforwardly follow from the design of the task, and the analyses are almost always quantitative. Because these tasks are more constrained, the analysis of the results is less controversial than the analysis of the production data, which by its nature can be viewed from many perspectives. Most studies analyze language production directly; the use of judges who provide ratings of performance on a preset scale is less often found in instructional effect studies than in interlanguage pragmatics more generally (Takimoto 2008a, 2008b). Although holistic scores can and do increase as a result of instruction, it is more difficult to pinpoint language change using such scores. Understanding specific developmental patterns, rather than having a general sense that students are better, may be more helpful in creating follow-up instruction. However, in cases such as mock job interviews where actual interviewers are very likely to use holistic scales, the use of scales by judges is an authentic choice (Louw, Derwing, and Abbott 2010). Instructional effect studies sit right on the border of instruction, which tends to value “right” or “wrong” answers, and acquisition, which values developmental sequences and interlanguage forms. This distinction brings a certain tension to scoring. To ascertain whether instruction brings about changes in learners’ interlanguage systems, it is important to track interlanguage development; namely, interim forms that show progress short of being completely target-like. For production data especially, without considering interim stages, instruction is likely to appear less effective than it is because getting to a target-like answer is metaphorically farther than getting to a better answer. Qualitative analyses may naturally avoid this issue because they do not convert observations about linguistic development into scores (see for example, Huth 2006; Liddicoat and Crozet 2001; Sardegna and Molle 2010).

Designing Instructional Effect Studies for L2 Pragmatics

153

Three fundamental principles of scoring are helpful when directly assessing the influence of instruction on the linguistic system through production data: 1) Do not score so generously at the pretest that there is no room for improvement at the posttest (in a warning, for example, be carefully is not be careful); 2) Do not be so strict at the end that the analysis does not reveal improvement. The result will be that student responses were not “right” before instruction and still not “right” after instruction. 3) Take development into account. Use an interlanguage analysis by documenting what learners do. An example from our study on the teaching of conventional expressions illustrates the value of separating the sociopragmatic from the pragmalinguistic aspects in evaluation (Bardovi-Harlig and Vellenga 2012). In the reciprocal thanking scenario (where the interlocutor says, “thanks for coming”) one student demonstrated a shift from You are welcome to Thanks for inviting me. Another student changed from You are welcome to Thanks for inviting Ø. After instruction, both learners identify the context as a thanking context (which is sociopragmatics). In addition, the first learner also used the full target-language expression, while the second learner is on his way to learning the expression. We might create a “used key words” category to capture the second learner’s progress. Using a “right/wrong” or “target-like/non-target-like” measure misses the benefit of instruction for the second learner. A hypothetical third learner may have stayed with You are welcome, which would have shown no change either in sociopragmatics or pragmalinguistics. Scoring the speech act and the expression separately records the progress by both learners. Scoring exact use of conventional expressions and use of key words separately from each other and separately from non-use captures what teachers instinctively know about teaching, namely, that it yields nuanced results and is not an all-or-nothing proposition. 2.4.4 Comparing groups statistically For quantitative studies, measuring gains linguistically is followed by a statistical analysis. Experimental groups are compared to each other and to a control group at both pretest and posttest, and at the delayed posttest, if there is one. Groups are compared at the pretest to make sure they are comparable. The assumption of comparability between or among groups is

154

Chapter Six

essential for instructional effect studies. If the groups are comparable at the pretest, an analysis is run using time and group as independent variables. Specifics will differ depending on the details of the study. If the groups are not comparable at the pretest, the researcher will need to control for the pretest scores statistically.

2.5. Writing up the report The need for a good write-up is not unique to instructional effect studies. However, because instructional effect studies have so many parts, it may be a particular challenge. (A checklist is provided in the Appendix). Nevertheless, documenting input, teaching approaches, practice activities, and tasks used for evaluation is particularly important in light of the two main goals of instructional effect studies in pragmatics: to allow for replication and to encourage the teaching of pragmatics in second and foreign classrooms more broadly. Although page limits and word counts have become increasingly more limited in journals, two journals stand out in their use of appendices to provide the necessary information. Applied Linguistics publishes extensive appendices on the journal’s website (e.g., Takahashi 2005a; Takimoto 2008b). Foreign Language Annals publishes relatively short reports, but provides extensive appendices (with text explanations) which enhance the reports (e.g., Fukuya and Martínez-Flor 2008; Shively 2010; Taylor 2002). Another option is to publish the pedagogical component of an instructional effect study separately. Although that is relatively unusual, it does address the goal of encouraging pragmatics teaching. Such sets of studies include Riddiford (2007), Holmes and Riddiford (2010), and Riddiford and Joe (2010) for research on teaching workplace requests and sociopragmatics and Riddiford and Newton (2013) for teaching, and Nguyen (2013), Nguyen, Pham, and Cao (2013), and Nguyen, Pham, and Pham (2012) for research on teaching constructive criticism, and Nguyen and Basturkmen (2010) for teaching. In contrast to journals and edited volumes, dissertations, theses, and inhouse reports are flexible in page length and format. Whatever the means of reporting the relevant information regarding the various aspects of the study, it is important to include as much information with as many examples as possible.

Designing Instructional Effect Studies for L2 Pragmatics

155

2.6. Report back to the teachers and program At the completion of the study, researchers should thank the teacher and the program for participating in or hosting the study. As Dörnyei (2003) observed, thanking participants and instructions is a basic courtesy which is often overlooked. In addition to writing thank-you notes, researchers should also report back to the people who supported the project. Sending a copy of the completed report is one way to report back, but offering to debrief the students after the instruction, telling the students the answers following the (delayed) posttest, or providing the teaching materials to the teachers for later use are all additional ways of maintaining interaction with the teachers, students, and programs that participated. Concordia University in Montreal has been very successful with their program of presenting “teacher-friendly research” at local and regional conferences where teachers can hear the results of studies in which they and their students participated. This also appears to encourage other teachers to participate. Establishing and maintaining good relationships with language programs makes subsequent studies possible and paves the way for other researchers as well.

3. Summary This paper presented the basic components of designing an instructional effect study for pragmatics. A checklist follows in the Appendix. In addition to addressing the goals of promoting the teaching of pragmatics and determining what approaches are effective in teaching pragmatics, instructional effect studies in pragmatics offer opportunities for innovative teaching in pragmatics. They form a natural bridge between research and teaching on the one hand, and researchers and teachers on the other. I hope this practical guide to instructional effect studies in pragmatics will encourage readers to undertake such investigations.

References Alcón-Soler, E. 2002. “Relationship between teacher-led versus learners’ interaction and the development of pragmatics in the EFL classroom.” International Journal of Educational Research 37: 359-377. —. 2005. “Does Instruction Work for Learning Pragmatics in the EFL Context?” System 33 (3): 417-35. —. 2007. “Fostering EFL Learners’ Awareness of Requesting through Explicit and Implicit Consciousness-Raising Tasks.” In Investigating

156

Chapter Six

Tasks in Formal Language Learning, edited by M. del Pilar Garcia Mayo, 221-41. Bristol: Multilingual Matters. —. ed. 2008. Learning How to Request in an Instructed Language Learning Context. Berlin: Peter Lang. —. 2012. “Teachability and Bilingualism Effects on Third Language Learners’ Pragmatic Knowledge.” Intercultural Pragmatics 9 (4): 51141. Alcón-Soler, E., and J. Guzmán-Pitarch. 2010. “The Effect of Instruction on Learners’ Pragmatic Awareness: A Focus on Refusals.” International Journal of English Studies 10 (1): 65-80, http://eric.ed.gov/?id=EJ912916. Bardovi-Harlig, K. 2001. “Evaluating the Empirical Evidence: Grounds for Instruction in Pragmatics?” In Pragmatics in Language Teaching, edited by G. Kasper, and K. Rose, 13-32. Cambridge: Cambridge University Press. —. 2009. “Conventional Expressions as Pragmalinguistic Resource: Recognition and Production of Conventional Expressions in L2 Pragmatics.” Language Learning 59 (4): 755-95. —. 2010. “Exploring the Pragmatics of Interlanguage Pragmatics: Definition by Design.” In Pragmatics across Languages and Cultures (Handbook of Pragmatics Vol. 7), edited by A. Trosborg, 219-59. Berlin: Mouton de Gruyter. —. 2015. “Operationalizing Conversation in Studies of Instructional Effect in L2 Pragmatics.” System 48: 21-34. http://dx.doi.org/10.1016/j.system.2014.09.002. Bardovi-Harlig, K., and M.-T. Bastos. 2011. “Proficiency, Length of Stay, and Intensity of Interaction and the Acquisition of Conventional Expressions in L2 Pragmatics.” Intercultural Pragmatics 8 (3): 34784. Bardovi-Harlig, K., and Z. Dörnyei. 1998. “Do Language Learners Recognize Pragmatic Violations? Pragmatic vs. Grammatical Awareness in Instructed L2 Learning.” TESOL Quarterly 32 (2): 233259. Bardovi-Harlig, K., B. Hartford, R. Mahan-Taylor, M. J. Morgan, and D. W. Reynolds. 1991. “Developing Pragmatic Awareness: Closing the Conversation.” ELT Journal 45 (1): 4-15. Bardovi-Harlig, K., S. Mossman, and H. E. Vellenga. 2015. “The Effect of Instruction on Pragmatic Routines in Academic Discussion.” Language Teaching Research 19 (3): 324-350. Bardovi-Harlig, K., and T. Salsbury. 2004. “The Organization of Turns in the Disagreements of L2 Learners: A Longitudinal Perspective.” In

Designing Instructional Effect Studies for L2 Pragmatics

157

Studying Speaking to Inform Second Language Learning, edited by D. Boxer, and A. D. Cohen, 199-227. Clevedon: Multilingual Matters. Bardovi-Harlig, K., and H. E. Vellenga. 2012. “The Effect of Instruction on Conventional Expressions in L2 Pragmatics.” System 40 (1): 77-89. Barekat, B. 2013. “Investigating Effects of Metalinguistic Feedback in L2 Pragmatic Instruction.” International Journal of Linguistics 5 (2): 197208. Barraja-Rohan, A.-M. 2000. “Teaching Conversation and Sociocultural Norms with Conversation Analysis.” In Teaching Languages, Teaching Cultures, edited by A. J. Liddicoat, and C. Crozet, 65-78. Melbourne: Applied Linguistics Association of Australia and Language Australia. Belz, J. A., and C. Kinginger. 2003. “Discourse Options and the Development of Pragmatic Competence by Classroom Learners of German: The Case of Address Forms.” Language Learning 53 (4): 591-647. Belz, J. A., and N. Vyatkina. 2005. “Learner Corpus Analysis and the Development of L2 Pragmatic Competence in Networked Intercultural Language Study: The Case of German Modal Particles.” The Canadian Modern Language Review 62 (1): 17-48. Boxer, D., and L. Pickering. 1995. “Problems in the Presentation of Speech Acts in ELT Materials: The Case of Complaints.” ELT Journal 49 (1): 44-58. Cheng, W., and P. Cheng. 2010. “Correcting others and self-correcting in business and professional discourse and textbooks.” In Pragmatics across Languages and Cultures (Vol. 7 of Handbooks of pragmatics), edited by A. Trosborg, 443-66. Berlin: Mouton de Gruyter. Codina-Espurz, V. 2008. “The Immediate vs. Delayed Effect of Instruction on Mitigators in Relation to the Learners’ Language Proficiency in English.” In Learning How to Request in an Instructed Language Learning Context, edited by E. Alcón-Soler, 227-256. Switzerland: Peter Lang. Cohen, A. D. 2005. “Strategies for Learning and Performing L2 Speech Acts.” Intercultural Pragmatics 2 (3): 275-301. Cohen, A. D., and N. Ishihara. 2005. “A Web-Based Approach to Strategic Learning of Speech Acts.” Center for Advanced Research on Language Acquisition, University of Minnesota. Accessed October 9, 2013, http://www.carla.umn.edu/speechacts/Japanese%20Speech%20Act%2 0Report%20Rev.%20June05.pdf. Cohen, A. D., and R. Shively. 2007. “Acquisition of Requests and Apologies in Spanish and French: Impact of Study Abroad and

158

Chapter Six

Strategy-Building Intervention.” The Modern Language Journal 91 (2): 189-212. Cohen, A. D., and J. M. Sykes. 2010. “Language Learner Strategies and Their Effect on Speech Act Performance.” Applied Linguistics Forum 30 (1). http://www.tesol.org/news-landing-page/2011/11/02/al-forumnews-volume-30-1-%28march-2010%29. Dörnyei, Z. 2003. Questionnaires in Second Language Research: Construction, Administration, and Processing. Mahwah, NJ: Erlbaum. Eslami, Z. R., and A. Eslami-Rasekh. 2008. “Enhancing the Pragmatic Competence of Non-Native English-Speaking Teacher Candidates (NNESTCS) in an EFL Context.” In Investigating Pragmatics in Foreign Language Learning, Teaching and Testing, edited by E. Alcón-Soler, and A. Martínez-Flor, 178-197. Bristol: Multilingual Matters. Eslami-Rasekh, Z., A. Eslami-Rasekh, and A. Fatahi. 2004. “The Effect of Explicit Metapragmatic Instruction on the Speech Act Awareness of Advanced EFL Students.” TESL-EJ 8 (2), http://tesl-ej.org/ej30/a2.html. Eslami, Z., and C. N. Liu. 2013. “Learning Pragmatics through ComputerMediated Communication in Taiwan.” Iranian Journal of Society, Culture & Language 1 (1): 52-73. Félix-Brasdefer, J. C. 2004. “Interlanguage Refusals: Linguistic Politeness and Length of Residence in the Target Community.” Language Learning 54 (4): 587-653. —. 2008a. “Pedagogical Intervention and the Development of Pragmatic Competence in Learning Spanish as a Foreign Language.” Issues in Applied Linguistics 16 (1): 49-84, http://escholarship.org/uc/item/03r9m3zb. —. 2008b. “Teaching Spanish Pragmatics in the Classroom: Explicit Instruction of Mitigation.” Hispania 91 (2): 477-492, http://www.jstor.org/stable/20063733. Fernández-Guerra, A., and A. Martínez-Flor. 2006. “Is Teaching How to Suggest a Good Suggestion? An Empirical Study Based on EFL Learners’ Accuracy and Appropriateness When Making Suggestions.” Porta Linguarum 5: 91-108, http://www.ugr.es/~portalin/articulos/PL_numero5/fernandez.pdf. Fordyce, K. 2014. “The Differential Effects of Explicit and Implicit Instruction on EFL Learners’ Use of Epistemic Stance.” Applied Linguistics 35 (1): 1-24.

Designing Instructional Effect Studies for L2 Pragmatics

159

Freed, B. F., D. P. Dewey, N. Segalowitz, and R. Halter. 2004. “The Language Contact Profile.” Studies in Second Language Acquisition 26 (2): 349-56. Fukuya, Y., and M. K. Clark. 2001. “A Comparison of Input Enhancement and Explicit Instruction of Mitigators.” In Pragmatics and Language Learning, Monograph Series 10, edited by L. F. Bouton, 111-30. Urbana, IL: Division of English as an International Language Intensive English Institute, University of Illinois at Urbana-Champaign. Fukuya, Y., and A. Martínez-Flor. 2008. “The Interactive Effects of Pragmatic-Eliciting Tasks and Pragmatic Instruction.” Foreign Language Annals 41 (3): 478-500. Fukuya, Y., and Y. Zhang. 2002. “Effects of Recasts on EFL Learners’ Acquisition of Pragmalinguistic Conventions of Request.” Second Language Studies 21 (1): 1-47, http://www.hawaii.edu/sls/wpcontent/uploads/2014/09/FukuyaZhang.pdf. —. 2006. “The Effects of Recasting on the Production of Pragmalinguistic Conventions of Request by Chinese Learners of English.” Issues in Applied Linguistics 15 (1): 59-91, http://escholarship.org/uc/item/77m4d8tj. Ghobadi, A., and M. Fahim. 2009. “The Effect of Explicit Teaching of English ‘Thanking Formulas’ on Iranian EFL Intermediate Level Students at English Language Institutes.” System 37 (3): 526-37. Halenko, N., and C. Jones. 2011. “Teaching Pragmatic Awareness of Spoken Requests to Chinese EAP Learners in the UK: Is Explicit Instruction Effective?” System 39 (2): 240-50. Holmes, J., and N. Riddiford. 2010. “From Classroom to Workplace: Tracking Socio-Pragmatic Development.” ELT Journal 65 (4): 376-86. House, J. 1996. “Developing Pragmatic Fluency in English as a Foreign Language: Routines and Pragmatic Awareness.” Studies in Second Language Acquisition 18 (2): 225-252. Huth, T. 2006. “Negotiating Structure and Culture: L2 Learners’ Realization of L2 Compliment-Response Sequences in Talk-inInteraction.” Journal of Pragmatics 38 (12): 2025-50. Jeon, E. H., and T. Kaya. 2006. “Effects of L2 Instruction on Interlanguage Pragmatic Development: A Meta-Analysis.” In Synthesizing Research on Language Learning and Teaching, edited by J. M. Norris, and L. Ortega, 165-211. Amsterdam: Benjamins. Jernigan, J. 2012. “Output and English as a Second Language Pragmatic Development: The Effectiveness of Output-Focused Video-Based Instruction.” English Language Teaching 5 (4): 2-14, http://www.ccsenet.org/journal/index.php/elt/article/view/15931.

160

Chapter Six

Kasper, G., and K. Rose. 2002. Pragmatic Development in a Second Language. Malden, MA: Wiley-Blackwell. Koike, D. A., and L. Pearson. 2005. “The Effect of Instruction and Feedback in the Development of Pragmatic Competence.” System 33 (3): 481-501. Li, Q. 2012. “Effects of Instruction on Adolescent Beginners’ Acquisition of Request Modification.” TESOL Quarterly 46 (1): 30-55. Li, S. 2011. “The Effects of Input-Based Practice on Pragmatic Development of Requests in L2 Chinese.” Language Learning 62 (2): 403-438. Liddicoat, A. J., and C. Crozet. 2001. “Acquiring French interactional norms through instruction.” In Pragmatics in Language Teaching, edited by G. Kasper, and K. Rose, 125-44. Cambridge: Cambridge University Press. Louw, K. J., T. M. Derwing, and M. L. Abbott. 2010. “Teaching Pragmatics to L2 Learners for the Workplace: The Job Interview.” The Canadian Modern Language Review 66 (5): 739-758. Martínez-Flor, A. 2008. “The Effect of Inductive-Deductive Teaching Approach to Develop Learners’ Use of Request Modifiers in the EFL Classroom.” In Learning How to Request in an Instructed Language Learning Context, edited by E. Alcón-Soler, 191-226. Switzerland: Peter Lang. Martínez-Flor, A., and E. Alcón-Soler. 2007. “Developing Pragmatic Awareness of Suggestions in the EFL Classroom: A Focus on Instructional Effects.” Canadian Journal of Applied Linguistics 10 (1): 47-76. Martínez-Flor, A., and Y. Fukuya. 2005. “The Effects of Instruction on Learners’ Production of Appropriate and Accurate Suggestions.” System 33 (3): 463-80. Mirzaei, A., and M. Esmaeili. 2013. “The Effects of Planned Instruction on Iranian L2 Learners’ Interlanguage Pragmatic Development.” Iranian Journal of Society, Culture & Language 1 (1): 89-100. Narita, R. 2012. “The Effects of Pragmatic Consciousness-Raising Activity on the Development of Pragmatic Awareness and Use of Hearsay Evidential Markers for Learners of Japanese as a Foreign Language.” Journal of Pragmatics 44 (1): 1-29. Németh, N., and J. Kormos. 2001. “Pragmatic Aspects of Task Performance: The Case of Argumentation.” Language Teaching Research 5 (3): 213-40.

Designing Instructional Effect Studies for L2 Pragmatics

161

Nguyen, T. T. M. 2013. “Instructional Effects on the Acquisition of Modifiers in Constructive Criticism by EFL Learners.” Language Awareness 22 (1): 76-94. Nguyen, T. T. M., and H. Basturkmen. 2010. “Teaching Constructive Critical Feedback.” In Pragmatics: Teaching Speech Acts, edited by D. H. Tatsuki, and N. R. Houck, 125-140. New York: TESOL. Nguyen, T. T. M., M. T. Pham, and T. H. Cao. 2013. “The Effects of Explicit Metapragmatic Instruction on EFL Learners’ Performance of Constructive Criticism in an Academic Setting.” In Pragmatics and Language Learning, Monograph Series 13, edited by T. Greer, D. H. Tatsuki, and C. Roever, 213-44. Honolulu: University of Hawai’i Press. Nguyen, T. T. M., T. H. Pham, and M. T. Pham. 2012. “The Relative Effects of Explicit and Implicit Form-Focused Instruction on the Development of L2 Pragmatic Competence.” Journal of Pragmatics 44 (4): 416-434. Pearson, L. 2006. “Teaching Pragmatics in Spanish L2 Courses: What Do Learners Think?” In Pragmatics and Language Learning, Monograph Series 11, edited by K. Bardovi-Harlig, C. Félix-Brasdefer, and A. S. Omar, 109-134. Honululu: University of Hawai’i Press. Pomerantz, A. 1984. “Agreeing and Disagreeing with Assessments: Some Features of Preferred/Dispreferred Turn Shapes. In Structures of Social Action, edited by J. M. Atkinson, and J. Heritage, 57-101. Cambridge: Cambridge University Press. Riddiford, N. 2007. “Making Requests Appropriately in a Second Language: Does Instruction Help to Develop Prgamatic Proficiency?” TESOLANZ Journal 15: 88-102. Riddiford, N., and A. Joe. 2010. “Tracking the Development of Sociopragmatic Skills.” TESOL Quarterly 44 (1): 195-205. Riddiford, N, and J. Newton. 2010. Workplace Talk in Action – An ESOL Resource. Wellington: School of Linguistics and Applied Language Studies, Victoria University of Wellington. Rose, K. 2005. “On the Effects of Instruction in Second Language Pragmatics.” System 33 (3): 385-99. Rose, K, and C. Ng. 2001. “Inductive and Deductive Teaching of Compliments and Compliment Responses.” In Pragmatics in Language Teaching, edited by G. Kasper, and K. Rose, 145-70. Cambridge: Cambridge University Press. Safont-Jordá, M. P. 2003. “Instructional Effects on the Use of Request Acts Modification Devices by EFL Learners.” In Pragmatic Competence and Foreign Language Teaching, edited by A. Martínez-

162

Chapter Six

Flor, E. Usó-Juan, and A. Fernández-Guerra, 211-32. Castelló De La Plana: Universitat Jaume I. Salazar Campillo, P. 2003. “Pragmatic Instruction in the EFL Context.” In Pragmatic Competence and Foreign Language Teaching, edited by A. Martínez-Flor, E. Usó-Juan, and A. Fernández-Guerra, 233-46. Castelló De La Plana: Universitat Jaume I. Sardegna, V. G., and D. Molle. 2010. “Videoconferencing with Strangers: Teaching Japanese EFL Students Verbal Backchannel Signals and Reactive Expressions.” Intercultural Pragmatics 7 (2): 279-310. Shively, R. L. 2010. “From the Virtual World to the Real World: A Modal of Pragmatics Instruction for Study Abroad.” Foreign Language Annals 43 (1): 105-37. Silva, A. J. B. D. 2003. “The Effects of Instruction on Pragmatic Development: Teaching Polite Refusals in English.” Second Language Studies 22 (1): 55-106, http://www.hawaii.edu/sls/wp-content/ uploads/2014/09/Silva.pdf. Sykes, J. M. 2005. “Synchronous CMC and Pragmatic Development: Effects of Oral and Written Chat.” CALICO Journal 22 (3): 399-432. Takahashi, S. 2001. “The Role of Input Enhancement in Developing Pragmatic Competence.” In Pragmatics in Language Teaching, edited by G. Kasper, and K. Rose, 171-99. Cambridge: Cambridge University Press. —. 2005a. “Noticing in Task Performance and Learning Outcomes: A Qualitative Analysis of Instructional Effects in Interlanguage Pragmatics.” System 33 (3): 437-61. —. 2005b. “Pragmalinguistic Awareness: Is It Related to Motivation and Proficiency?” Applied Linguistics 26 (1): 90-120. —. 2010. “Assessing Learnability in Second Language Pragmatics.” In Pragmatics Across Languages and Cultures (Handbook of Pragmatics Vol. 7), edited by A. Trosborg, 391-421. Berlin: Mouton De Gruyter. Takimoto, M. 2008a. “The Effects of Decuctive and Inductive Instruction on the Development of Language Learners’ Pragmatic Competence.” The Modern Language Journal 92 (3): 369-86. —. 2008b. “The Effects of Input-Based Tasks on the Development of Learners’ Pragmatic Proficiency.” Applied Linguistics 30 (1): 1-25. Tateyama, Y. 2001. “Explicit and Implicit Teaching of Pragmatic Routines.” In Pragmatics in Language Teaching, edited by G. Kasper, and K. Rose, 200-22. Cambridge: Cambridge University Press. —. 2007. “The Effects of Instruction on Pragmatic Awareness.” In JALT 2006 Conference Proceedings, edited by K. Bradford-Watts, 11891200. Tokyo: JALT.

Designing Instructional Effect Studies for L2 Pragmatics

163

Tateyama, Y., G. Kasper, L. Mui, H.-M. Tay, and O. O. Thananart. 1997. “Explicit and Implicit Teaching of Pragmatic Routines.” In Pragmatics and Language Learning (Vol. 8), edited by L. F. Bouton, 163-178. Urbana, IL: University of Illinois at Urbana-Champaign. Tatsuki, D. H., and N. R. Houck. ed. 2010. Pragmatics: Teaching Speech Acts. New York: TESOL. Taylor, G. 2002. “Teaching Gambits: The Effect of Instruction and Task Variation on the Use of Conversation Strategies by Intermediate Spanish Students.” Foreign Language Annals 35 (2): 171-89. Teng, C., and F. Fei. 2013. “A Consciousness-Raising Approach to Pragmatics Teaching: Web-Based Tasks for Training Study-Abroad Students.” Journal of Technology and Chinese Language Teaching 4 (1): 50-63, http://www.tclt.us/journal/2013v4n1/tengfei.pdf. Trosborg, A., and P. Shaw. 2008. “Deductive and Inductive Methods in the Teaching of Business Pragmatics: Not an Either/Or.” In Institutional Discourse In Cross-Cultural Contexts (LINCOM Studies In Pragmatics 14), edited by R. Geluykens, and B. Kraft, 193-220. Munchen: LINCOM. van Compernolle, R. A. 2011. “Developing Second Language Sociopragmatic Knowledge through Concept-Based Instruction: A Microgenetic Case Study.” Journal of Pragmatics 43 (13): 3267-83. van Compernolle, R. A., and L. Williams. 2013. “The Effect of Instruction on Language Learners’ Sociolinguistic Awareness: An Empirical Study with Pedagogical Implications.” System 41 (2): 298-306. Vellenga, H. E. 2008. “Instructional Effectiveness and Interlanguage Pragmatics.” PhD dissertation, Northern Arizona University. Vyatkina, N., and J. A. Belz. 2006. “A Learner Corpus-Driven Intervention for the Development of L2 Pragmatic Competence.” In Pragmatics and Language Learning, Monograph Series 11, edited by K. Bardovi-Harlig, C. Félix-Brasdefer, and A. S. Omar, 293-329. Honululu: University of Hawai’i Press. Wishnoff, J. R.. 2000. “Hedging Your Bets: L2 Learners’ Acquisition of Pragmatic Devices in Academic Writing and Computer-Mediated Discourse.” Second Language Studies 19 (1): 119-48. Yoshimi, D. R. 2001. “Explicit Instruction and JFL learner’s Use of Interactional Discoursemarkers.” In Pragmatics in Language Teaching, edited by G. Kasper, and K. Rose, 223-44. Cambridge: Cambridge University Press.

Chapter Six

164

Appendix Checklist for Planning and Writing

Target of instruction

Participants o o o

Instruction o o o o o o o

Type of pragmatic input Method of instruction Practice Duration Class format Control Group Documentation of teaching

Evaluation o o o o

Students Teachers Plan for absentee students

Scheduling of pretest, posttest, delayed posttest Tasks Measurement of gains (linguistic analysis) Statistical analysis

Report back to the teachers and/or programs involved

CHAPTER SEVEN “YOU HAVE NO CHOICE:” PRAGMATIC CONSIDERATIONS IN CURRENT LANGUAGE TEST INSTRUMENTS RICHARD CHAPMAN

1. Introduction Current international English language tests have their basis in a complex understanding of cognitive development, functional aspects of language use and interactive communicative skills.1 Recently, their justifications have become more explicitly practical and commercial, in keeping with the presumed needs of candidates, employers and institutions (the three main stakeholders, to use a term which is, significantly, in vogue in test descriptions). Cambridge English Language Assessment, for example, present their new C2 examination, invariably known as Proficiency, as a means to “show the world that you have the English to study or work at the highest levels.”2 But this practical bent to standardised, international tests, and the commercial or professional element that goes with it, risks perhaps distracting attention from the clear pragmatic elements present in any language assessment situation and which have historically been denied an explicit role in test development and evaluation. Because of their openly functional and even transactional nature, many current test instruments assume their pragmatic content without a theoretical analysis of the complete pragmatic picture, which would include more than just considerations of clarity of message, effect on the speaker and relevance. It is as if any pragmatic concerns had been absorbed into the functional and communicative approach to assessment of language skills. Instead, pragmatic aspects of language tests such as presupposition, the nature of performatives and the complexity of context do need to be granted their due importance.3 This paper aims to draw

166

Chapter Seven

attention to a lack of pragmatic understanding as regards test construction and design, and, at the same time, open some discussion about the actual pragmatic experience a language examination can produce. The lack of a role for pragmatics in language test design may be a logical consequence of the until recently held assumption that pragmatics has almost exclusively to do with spoken linguistic interchanges and that written language raises few pragmatic issues, as it is usually more formalised, lacks variables such as intonation and non-verbal markers, and exists in a very clearly defined contextual space; for instance, as contended by McEnery, Xiao, and Tono (2006, 104), “Written registers tend to be referentially explicit.” Cohen (2010, 4) also states, “pragmatics has conventionally focussed on the spoken medium and has paid little attention to writing.” While this may well overstate the case, and risks ignoring work on the pragmatic aspects of written texts, it does perhaps go some way towards explaining the lack of explicit pragmatic analysis of test items and the test experience. Research has already suggested that written language also has a significant pragmatic aspect (see Simpson 2004, 200) and we can see the content of emails and social network communication as indicative of the potential pragmatic load in new forms of written language use. Perhaps more important for our study will be the observation that the pragmatic aspects of language tests exist on various levels and pertain both to the linguistic content of the examination itself and the pragmatic experience of sitting the test and being assessed. As we shall see in our analysis, the complexity of the pragmatic aspect of language tests is partly due to the fact that there are at least two levels to each linguistic contribution, one on the level of the co-text and immediate context, and the other flowing from the reality of an examination situation in which candidates are being assessed (if not judged) and are aware of this assessment and the stakes involved. Short’s two-tier model of interaction in literature (Short 1989, in Simpson 2004) will be useful in our attempt to make more explicit what occurs in a test situation. Indeed, this two-tier pragmatic aspect is not confined merely to speaking sections of a language examination. As Fillmore (1981, mentioned in Shohamy 2001) suggests, types of texts and the way they are read and perceived change when they are utilised in a testing situation. Fillmore describes accommodation by the test taker to “truth” owned by the tester, and makes reference to the rubric of tests which he presents as “cold” and “most authoritative” (in Shohamy 2001, 124). Thus we will make reference to speaking and written parts of the test and each component may include different types of pragmatic issues.

“You Have No Choice”

167

2. Material and methods As a prototypical example of current international English language tests, I will analyse the Cambridge Proficiency: English examination (CPE) as a suitable source of data. The reasons for this choice were a combination of the recognition of the leading role Cambridge Assessment has played historically in the development of language tests, the significant position it holds in the marketplace today (an estimated 4 million candidates were predicted to take its examinations in 2013, for example; Weir 2013, 2), the fact that CPE is an examination at the C2 level of the Common European Framework of Reference for languages (CEFR) and so may be seen as a “final destination” for many language learners, and that it was substantially revised in 2013, with the first sitting of the new test taking place in March. It may be considered a genuine example of up-to-date language testing and at the same time a flagship of a popular and important participant in the field. I decided to utilise a paper from 2011 along with a sample produced in 2013 to provide a diachronic slant to the data, this being particularly important in an examination that has recently undergone changes. I also chose to analyse data in concert with descriptive material produced in reference to the test by the examination board itself (Cambridge English Research Notes and Brief Exam Guides). This would contextualise data and give balance to any interpretations based on the literature. The sample tests and descriptive material are readily available and were obtained from the Cambridge English Language Assessment website.4 Additional data, for purposes of comparison, were collected from other Cambridge Main Suite Examinations (KET, PET, FCE, and CAE, which correspond to the levels A2, B1, B2 and C1 of the CEFR respectively), and from preparation literature explicitly connected with the tests.5 It is worth noting here that there exists what we might call a complete “exam preparation subculture,” both in publications and on the internet, with its own materials, discourse and interpretation of what is meaningful linguistic behaviour and what is important in language learning.6 Cambridge English have long been aware of the washback effects7 of high-stakes examinations (indeed their literature claims to explicitly take this into account in the design and production of their tests),8 but from a pragmatic point of view it is interesting to recognise (and perhaps we should further investigate) the substantial nature of the influence of the examination sector on the development of approaches to study of the English language as a whole.

168

Chapter Seven

After data collection, the method of analysis consisted in familiarisation with the content and wording of the test through repeated readings. This led to the formulation of tentative descriptive criteria of the nature of the examination instruments used and their possible pragmatic content and effect. Findings and preliminary interpretations of them were then compared with the literature on testing and pragmatics. Conclusions are only intended to be the beginnings of insights into a largely unexplored area of language use (the pragmatic implications of language test items, which, as we shall see later, do not only consist in washback effects but also effects on the candidates themselves). A full description of the 2013 CPE would be out of place here,9 but a very brief outline will hopefully suffice. There are four papers: writing; listening; speaking; and a combined reading and “Use of English” test (an innovation in this revised edition of the examination). In the Reading and Use of English paper, texts for comprehension and language use are carefully chosen as to range of subject matter and relative length, and tasks are designed to provide the examiner with evidence of a candidate’s abilities in a way that is reliable and valid. Thus, there are test instruments that are repeatedly used: multiple-choice, gap-fill and multiple-matching. The listening paper uses similar instruments, while the productive skills are tested using carefully constructed stimuli to encourage responses that can be reliably assessed. However, the test methodology applied immediately brings us to our fundamental question: do the instruments offer a pragmatically valid experience of language and is the pragmatic aspect of it theorised and analysed at all?

3. Analyses 3.1. The Reading/Use of English paper The Reading/Use of English paper (CPE sample examination, 2013) perhaps provides the most striking exemplification of various (and perhaps unintended) pragmatic effects in language tests. Candidates are expected to make choices. In a total of 72 marks for the paper, 20 are attributed thanks to multiple choice answers (the candidate selects a, b, c, or d), and another 24 marks depend upon selections out of limited options (either repeatedly choosing one from the same eight possible paragraphs in Part 6, or opting for one of four paragraphs as the answer to each of ten questions in Part 7). Essentially, the test taker needs to be a good decision-maker, and to express her/his linguistic skills through minimal behaviour. Now, in testing reading skills, this has clear justifications in test design terms: with

“You Have No Choice”

169

minimal linguistic production a candidate can show a high level of reading comprehension and the examiner does not have this reading performance indicator filtered by issues of writing skills, explicit grammatical knowledge and the like. Marking can also be largely automated and, so, much more reliable (and fairer to all candidates). However, when we consider that this part of the examination also claims to assess candidate knowledge of semantic precision, idiom, grammatical structure and the understanding of detail, opinion, sequencing of information and more (CELA Brief exam guide 2013), we might pause and reflect on the heavy onus being placed on one-off decisions (multiple-choice and matching, 62% of the total marks) and one-word answers (22% of total). The practical and behavioural difficulties associated with these highstakes choices are perhaps aggravated by the advent of computer-based testing. The Cambridge English sample examination for 2013 was also available online and the experience of facing the test via the screen and keyboard clearly changes the examiner-candidate relationship and the nature of the choices being made. Especially in the paragraph selection task (Part 6) I can report distinct difficulty associated with scrolling the main text and scrolling possible answers, leaving only a partial view open to the examinee10 as each selection is made. While age may well be a significant factor here, it is important to acknowledge the substantial alteration to the examination experience that computer-delivered tests will bring about. Tests are to be administered many more times a year than was the case (even on demand in some systems: see the Oxford Test of English White Paper 2012), thus changing the ritualised, special-date element associated with formal examinations. These changes may be popular, and one hopes, beneficial, but they cry out for attention. The only section of the paper where more substantial production is required is the “key word transformations” task. Here candidates are asked to “complete the second sentence so that it has a similar meaning to the first” (CPE sample examination rubric, part 4, CPE Handbook, 2011). Strict parameters are given (answers between three and eight words, one word supplied that must in no way be changed), making the task as controlled as possible to ensure that candidates demonstrate working knowledge of certain structural and idiomatic forms in English. There are significant pragmatic problems with the task as it is conceived: the idea of re-writing or rephrasing is a perfectly “natural,” real-world task, but it is here reduced to a caricature. We may question the precise meaning of “a similar meaning” and debate what changes are acceptable or not, but, more importantly, when we look at individual items, we may see significant problems emerge. In the 2011 sample there were eight sentences, three of

Chapter Seven

170

which could only be “transformed” correctly by reversing the information sequence of the items (with, I suggest, substantial pragmatic effect) and altering the subject of the sentence. Thus, “I felt relaxed at Gita’s house because her parents greeted me so warmly”

was supposed to be rephrased as “Gita’s parents made me feel at ease with the warmth of their greeting.”11

It is almost superfluous to say that an affirmation in the first person about how “I felt” that is explained with an additional clause after because is pragmatically very different from a relation of what Gita’s parents did to me with a noun phrase added as explanation. Here we meet issues of fronting and emphasis, as well as transitivity. All of these differences may well come within the ambit of “complete the sentence so that it is similar to the first,” but its content seems unclear and open to interpretation, the student being required to identify the interpretation the examiner has in mind, and to make the correct linguistic choice – a choice which is not based precisely on her/his linguistic knowledge or even expertise, but rather on a subtle awareness (probably coming from practising similar tests) of what will be considered a variation that is similar to the first and so rephrases the sentence without violating the rules of test success. Here the metapragmatics has to do with understanding test rules rather than simply being a high-level speaker of English. Perhaps a more glaring example was the first in Part 4. Here the sentence to be rephrased was: “Serena really has no idea of the difficulty of finding a parking place.”

This was to be transformed as, “Little does Serena know how difficult it is to find a parking place.”

We might comment on the supposed naturalness (or lack of it) in the second sentence, although this is a highly subjective area for debate. What is beyond question, however, is how tenuous the claim to similarity is between the two phrases, “Serena really has no idea” and “little does Serena know.” Leaving aside the order of information issue mentioned above, it is worth underlining the rhetorical markedness of the little + inversion structure, which is different in pragmatic quality from the more usual negative structure reinforced with the adverb really. The original

“You Have No Choice”

171

sentence might be described as emphatic but expressed in a very common way; the second instead evokes a narrative and even ironic style. These briefly outlined examples already point to a characteristic of the examination: the carefully constructed instruments are as authenticlooking as possible, displaying an understanding of the need for variation in text content and the range of linguistic features over the whole paper, and are the product of expert test design with highly reliable and linguistically informed features. However, the need for reliability and costeffectiveness and the view of construct validity as being ultimately based on an idea of communicative competence risks producing tests that on occasion are pragmatically contradictory or meaningless. Perhaps it is no mistake that Part 4 of the Reading/Use of English paper is called Key word transformations, because that is exactly what its items can become: not a rephrasing of what was just said, but a substantial re-writing of a decontextualised item in an exam. The examiner has already decided “how far you can go” in creating your similar (but different) meaning; you have to guess it and thus show your knowledge of a set of English structures and idioms. It will be significantly different linguistically, but not so much so that it diverges from the permitted correct answer. But here we perhaps have an insight into how this part of the examination really works from a pragmatic point of view. Candidates (who often, it should be stated here, claim to enjoy, or at least prefer this part of the paper), are probably doing something quite different, in pragmatic terms, from the examiners’ assumptions of communicative competence or grammatical and lexical knowledge and use, when they answer these questions. Any “real” meaning that we might try to associate with an item such as the sentence about Gita’s parents (or how I felt?!) mentioned above is destined to be very subjective, depending, as it will, on an invented contextualisation. We imagine the scene, the participants, the message the speaker/writer wanted to convey (we do not know, indeed, if these eight sentences were supposed to have been written or spoken), and we construct our possible paraphrasing, or intralingual translation. Instead, the wily candidate in a CPE examination is probably doing something very different. Her/His objective is to pass the test, and the way to ensure this in Part 4 of the Reading/Use of English paper is to follow a series of previously acquired meta-rules as to how to approach this kind of question: try to maintain the same tense, look out for that list of typical structures and idioms that usually “come up” (“key” items?) and do not use more than eight words! Put another way, pragmatically, the candidates know what is expected and what has to be done. It is not really a question

Chapter Seven

172

of communicative ability, but much more a question of nous and showing you know what you are expected to know. A similar analysis might be presented regarding the multiple-choice modified cloze test.12 Here there are four options to choose from to fill each gap and, even in a well-constructed test, there will often be more than one answer that may be construed in such a way as to be correct. The skill for the student lies in being aware of what is the most likely answer (or the most “normal,” or the least rhetorical, or marked form?): Cambridge English asks for the answer which “best fills each gap.”13 In the sample from 2011 we see an instance of this: Keas – not just pretty parrots Few birds are as (1) ………. curious as keas.

The choices offered were: 1 A) insatiably

B) hungrily

C) thirstily

D) unmanageably

Now, it is interesting to note that we can quite easily recognise that the “correct answer” is A, but at the same time, we must recognise that D is a perfectly possible answer. It would be less common, but might represent a skilled use of language. Indeed, as there is reference later in the text to the keas’ destruction of “everything from rubbish bins to windscreen wipers,” the second choice finds pragmatic justification in the co-text. Yet, it is wrong (according to the official answer key, and probably to most native speakers and examination candidates alike). Collocation might provide a satisfactory rationale for the “correct” answer, but besides any quibbles we might have about the complex semantics concealed within collocation choices, we must also consider the greater creativity of the second answer, which does not instantiate the usual collocational pattern but expresses a semantically ideal concept in this (admittedly very limited) context. Pragmatically speaking, there are two messages here: firstly, it is tactically unwise to risk rhetorically emphatic choices; and secondly, preparation for the test really means developing a subtle awareness of the precise version of English that is required, the type of language expected, and entering the specific discourse of the Cambridge English CPE examination. In other words, candidates are asked to show skill at disambiguation. If we accept Lycan’s (2000, 170) assertion that “almost every sentence we encounter in life is technically ambiguous,” then we will understand why gap-fill tasks risk being pragmatically dubious if not downright unfair, if

“You Have No Choice”

173

we do not clarify what the true pragmatic parameters are in each instrument. In this task we are most of all asking our students to choose from a range of possible meanings effectively in order to interpret the text as the examiner has. We can accept this as a perfectly legitimate linguistic skill and indeed a life skill, but we should perhaps be more explicit about what is entailed in the multi-faceted choices we are forcing candidates to make. We should also ask ourselves if the context is really adequate. The open gap-fill task that made up Part 1 of the 2011 sample examination (moved to Part 2 of the new examination) shows a situation comparable to the one presented above, and which I will now briefly describe. Here ambiguity is not caused by a range of choices that includes more than one permissible answer, but instead a gap which, because it may be filled by any single word in English, allows for interpretative variation and again forces the candidate to imagine what the examiner considers linguistically more “normal” in each space: “the word which best fits each space” (part 1, CPE Sample Paper 2011, CPE Handbook 2011). It is revealing that the answer key provided in the handbook offers as many as four alternatives for one space and on two occasions our reading suggested at least another potentially legitimate answer. Again, the real skill the candidate is required to show is an understanding of a specific standard of English language and usage that “fits.” As befits a highly reliable examination, the reading instruments of the CPE are all designed to require minimal output, simply selecting a letter as proof of comprehension. Leaving aside any quibbles as to the potentially detached nature of the mere indication of a letter as a sign of understanding, the more complex issue arises of understanding as being in alignment with the expectations of the examination or the overall discourse community. Most answers will be quite clear, but some may depend upon a reading of reality that is essentially a part of the linguistic or even philosophical tradition the test itself comes from.14 Again we must ask ourselves what it is that candidates are being required to do with their language. And the reply might be a little different from the assumed yardsticks of grammatical and lexical knowledge, effective punctuation and organisation of discourse, performing well in the oral test and the like, that we would naturally come up with as an initial response.

3.2 The listening paper A similar range of testing instruments presents itself, not surprisingly, in the listening paper, which employs a similar approach to the reading tasks. Again, a substantial portion of the marks available are entrusted to

174

Chapter Seven

multiple-choice questions (11 out of a total of 30; 37%) and another 10 (33%) are assigned according to one-letter answers from a matching task. Only 9 marks depend on student-generated language (less than a third) and again the contribution is invariably limited to one or two words. Principles of reliability are clearly considered paramount, and this is understandable. The selection of texts for the listening paper shows significant pragmatic awareness on the part of the test designer: the monologues are varied in length and topic, and one of texts consists of a dialogue or discussion, which allows for contrasting verbal interactions and pragmatic behaviour. Despite the dependence on one-letter-answers, we must also give credit to the multiple-matching task for Part 4, which is a significant improvement on true-false, or true-false-not-in-text instruments that notoriously lack semantic credibility.15

3.3. The writing paper The writing paper is more overtly pragmatic in inception, as it requires candidates to produce two extended pieces of text, the first of which is obligatory and has the explicit aim of expressing opinions. Both compositions are graded according to four criteria (content, organisation, communicative achievement and language), and all of these can be said to have a clear pragmatic slant: content – the candidates are expected to satisfy the needs of the task set; organisation – the answer should appear logical and well-structured; language – explicitly includes the idea of range of language used (i.e. richness of vocabulary and variation of grammatical forms); and communicative achievement – which assesses issues of register and appropriacy. The examiner rubric speaks of irrelevance and “negative effect on the reader;” two highly pragmatic issues.16 Interestingly, I found that data from this paper were less important to our study, given that the instructions and tasks are so clearly defined and the marking criteria and grading scales so unequivocally laid down that there is little of significance for the pragmatic analysis of the test that is not clear from the outset. In other words, there is no mismatch between real and apparent pragmatic content.

3.4 The speaking paper The speaking paper is perhaps the section where we would expect to find the most data relevant to pragmatics. What interpretation of pragmatics is deemed as significant for communicative competence and so (overtly or covertly) favoured by the design and conduction of the

“You Have No Choice”

175

speaking component? Like the writing paper, the speaking paper is so precisely metered and structured as to reduce certain pragmatic variables in conversation (presumably this is intentional, as a way of guaranteeing reliability and thus fairness).17 There are three distinct parts to the test: firstly, an introductory conversation between the interlocutor and candidates; then a second part consisting of a decision-making task where candidates discuss a problem based on visual stimuli; and finally, a section in which each candidate has a “long turn” based on a prompt card, this followed by a group discussion. Immediately we notice a highly significant element which is indeed typical of all levels of Cambridge English Main Suite examinations: candidates have to interact in pairs (or possibly threesomes) when they show their oral skills. Speaking tests are not individual interviews, but something socially and pragmatically more complex. There has been much debate about the advantages and dangers of the paired oral-testing technique, but recent research probably points us in the direction of recognising the benefits as greater than the problems it causes.18 This format usually results in more interaction, more negotiation of meaning and richer candidate output. Here we can see pragmatics is taken into account, in that the very construct of the examination allows a representative sample of pragmatic ability to be produced by candidates: they can show turn-taking and linking skills, politeness techniques and rhetorical forms of speech (for the construction of convincing arguments). The scoring is also explicit in recognising the importance of pragmatic aspects of candidate performance: of the five criteria that inform rating of speaking, interactive communication clearly addresses pragmatic elements, and discourse management can also be seen to apportion considerable importance to pragmatic skills. Even criteria such as grammatical resource and lexical resource explicitly mention “flexibility of use” and “appropriate vocabulary” (CPE Handbook 2013). But these criteria, evaluating pragmatic skills as they do, can be seen as largely unsurprising, especially to anyone familiar with Cambridge English examinations or the recent history of English language testing. There are other, perhaps more interesting pragmatic aspects to the speaking test experience of the CPE examination, and these will be important for our overall understanding as to what actually happens, pragmatically, in a CPE examination. Although pragmatic awareness of aspects of language such as Grice’s Cooperative Principle (Lycan 2000, 187-195, gives a good summary of this) would clearly be of some benefit to a candidate, and awareness of turn-taking procedures and what we

176

Chapter Seven

might call Anglo-Saxon ideas of negotiation would presumably be of help, we can postulate another pragmatic component to the CPE speaking paper. In Part 2 (the decision-making section of the test) the candidate really needs to “make her/his contribution a little more than that which is required at the stage at which it occurs.” In other words, candidates are of course aware that every time they make an utterance it is an opportunity to “perform” and show what they know and can say. The decision-making task is, naturally, just a way of eliciting this verbal performance. It is implicit, but no less significant for that. The same happens in the “individual long turn” when a candidate speaks about a photograph in FCE or CAE and a prompt in CPE, but really has little pragmatic intention of talking about the stimulus, and every interest in making a good display of the language she/he knows. The extra comments from the other participant (invited by the interlocutor) exist, again, purely to elicit an utterance that has an illocutionary force that is almost entirely not to do with the topic or preceding monologue, and everything to do with the aim of “making a good impression” and so passing the examination. The three-way conversation as a whole is largely contrived from a pragmatic point of view, with an attempt at “authenticity” which is openly inadequate (indeed, a “real” conversation would not have scripted questions or strict timings).19 Here we can see that there are in fact two levels to the pragmatics of language tests: firstly the content of the linguistic situations created by the tests themselves (e.g. the supposed subject matter or meaning of a reading or listening text and the questions or tasks associated with it) and the metapragmatic content which is never absent and sometimes paramount (as in the speaking paper).20 The task is to “show yourself;” essentially to claim an identity as an accurate and skilled speaker of English by means of your performance in the various, fictitious linguistic mini-situations offered by the test. Thus every contribution is a performative21 in every sense of the word: it is part of the performance of language the candidate offers for assessment, and it is a pragmatic statement of ability and perhaps character. By expressing her/himself in a certain linguistic form, a candidate becomes a little better (or worse) in the judgement of the rater, within the parameters of the test itself. The use of can-do statements (an essential part of the way the CEFR is constructed and a significant part of Cambridge English descriptions of various examinations: see Cambridge Examination handbooks for KET, PET, FCE, CAE and CPE) only underlines how much of a performative the act of answering questions in language tests truly is: by constructing a certain contribution in the speaking test, the candidate is explicitly claiming and simultaneously

“You Have No Choice”

177

proving the ability to do things in the foreign language. The metapragmatic task is always to “show yourself.” This second level of pragmatics is hardly a new discovery. Short’s (1989, in Simpson 2004, 34-38) analysis of the two levels of interaction that are often to be found in literature is relevant here. Short postulated a two-tier pragmatics for the understanding of dialogue in drama, with one layer of discourse embedded within another. When characters interact in a play, film or even a novel, there is naturally a pragmatic element to the dialogues and actions that ensue. On a higher level, however, there is the ever-present, though possibly more hidden, interaction between the author and the audience or reader. The writer is presumably trying to appeal to her/his public, or convince them, or even leave them bemused. In the same way that characters in films or plays produce utterances that are simultaneously a part of a plot and a part of what the author tells the audience, so our candidates are obliged to make contributions that are relevant to the immediate linguistic mini-context of a particular examination task and, at the same time, to “play the game” and inform the assessor about their abilities. To put it slightly more theoretically, the real authenticity of examination pragmatics lies in this second tier of pragmatic action, where the illocutionary force of most contributions will be to reveal accurate grammatical and lexical knowledge or behave with visible linguistic skill or socio-linguistic dexterity at any given moment. Their aim must be to fit with the fiction of the language task, while remaining aware of the requirements of the pragmatic facts of a test situation. Naturally, there is a perlocutionary element to the state of affairs as well. The contributions of each candidate will have effects upon the assessor(s) and they will base their judgements on these effects in concert with the clear rating guidelines that are published by Cambridge English (e.g. in the Cambridge CPE handbook 2013 and more completely in examiner training material). Indeed, much of Cambridge English examination training is designed quite explicitly to modify and standardise the reactions of assessors to certain language items (e.g. “typical” grammar mistakes, divergences in pronunciation et cetera, all of which again is an attempt to reach a reliable means of assessment), through the use of video recordings of past candidates. Cambridge English also crosscompare the behaviour of assessors using complex statistical methods (e.g. Rasch analysis, see McNamara 1996, passim) with the aim of guaranteeing reliable rating.22 (A similar approach to rating is used with regard to the writing paper. Rating scales are furnished with clear descriptors and scorers are monitored for their own “performance” as consistent markers).

178

Chapter Seven

4. Concluding remarks Thus we have seen that an examination such as the CPE is full of pragmatic elements, both in its linguistic content and in the meta-context within which it must operate. But there is more to this second-level pragmatics, and again this is of vital importance for the candidate. The deficit we have just observed in awareness of the meta-pragmatics of standard international examinations has still further effects, and these might go some way towards explaining the difficulties we encountered earlier in the Reading and Use of English paper. One of the fundamental linguistic elements in interpretation is presupposition. How we interpret a sentence or text and especially disambiguate it (see section 3.1 for the importance of this for successful completion of tasks) largely depends upon the presuppositions we share. If we accept Stalnaker’s (2011) position that a speaker meaning is not a sentence meaning, and Seuren’s (2011) idea that presuppositions are essentially pragmatic and are effectively “default assumptions” which we apply to “make language work” in a given linguistic interchange, then we can begin to become aware of why it is that certain single-item tasks have only one permitted solution: discourse domains are “contextual constructions entertained by speakers and hearers” and they define what is allowable or not. In other words, it is not properties of language but rather “the common background beliefs and assumptions of speakers and addressees” (Archer and Grundy 2011, 17) that decide what a particular piece of language means at a certain moment, between certain users of that language.23 This is full of significance for us in our attempt to understand the pragmatic realities of international English language tests. When you as a candidate are faced with the dilemma of choosing between “insatiable” or “unmanageable,” it will not be linguistic difference that solves the problem, but instead you have no choice but to accept the presuppositions of the testing organisation (and perhaps the testing community or the whole Cambridge test discourse community), presuppositions that are “logical or neutral or uncontroversial in given contexts involving particular speakers and addressees” (Stalnaker, in Archer and Grundy 2011, 18). You need the pragmatic contextual elements in order to “spot” the only correct answer. This explains why it is that we can identify potentially imperfect items even in well-prepared tests such as the Cambridge English CPE and yet face a wall of certainty from the examination agency. It is probably not arrogance on their part but assurance based on confidence in their item-writing procedure and, more pertinently, on their pretesting: were an answer truly ambiguous, it would

“You Have No Choice”

179

presumably have been revealed during pretesting. If the inadequate item is not weeded out, then we must question the quality and thoroughness of the pretesting carried out (it is always a possibility that they have “cut corners”), or as is more likely, recognise the difference between the two solutions and why one is more acceptable, both to the experts and the student-guinea pigs who tried the test.24 Pragmatically, it is more than likely that Cambridge English and its trial testees are “on the same page” in terms of discourse. A mere glance at Cambridge English Language Assessment criteria for grading writing and speaking underlines the wealth of pragmatic content in a language test. Concepts such as coherence and cohesion, effect on the receiver, and even task completion (Cambridge CPE Handbook 2011) are all largely pragmatic in nature. Even pronunciation is assessed on supposed intelligibility for the listener. That this is not openly theorised and discussed is perhaps a “pragmatic deficit,” but as we have seen, the deficit goes much deeper. The whole nature of the examination and the way a student experiences it is shot through with pragmatic elements, from its original construction, to its character as a linguistic and personal experience “on the day” through finally to the results published. Not surprisingly, this pragmatic character can be seen as highly “western” in its inception: the evaluation of turn-taking is an example of this, as it is an ideological concept: “In societies with egalitarian values the right to a turn, that is, freedom of speech, in part defines one’s equality to one’s fellow citizens” (Pratt 1977, 101). When students are assessed on their respect for turn taking norms, they are in effect being assessed as “western citizens.” This is, perhaps, a direct reflection of the explicit politics of the Main Suite examinations as representing fairness, opportunity and equality.25 Choice fits neatly into this post-capitalist idea as well. It is not surprising, once we have taken the meta-pragmatic level of the examination into account, that candidates are faced with high-stakes choices as a way of representing their language proficiency. We could even say that they must be proficient in making choices. In our context we may well perceive “choice” as being a philosophical “good,” but all linguistic choices are replete with pragmatic significance and can risk reflecting what is not a choice at all, but rather an illusion of choice within a clearly predetermined linguistic and social context. The mismatch between the pragmatically “light” choice of hitting the a or b or c or d key on the computer and the pragmatically laden effect of that selection on the kind of individual a test-taker presents her/himself to be is of great importance here. Our tests play a part in the linguistic and social construction of identity and allow membership of groups, institutions and

180

Chapter Seven

companies (Shohamy 2001, especially chapter 18). The meta-pragmatic level of our examinations has always been present, but is ripe for better analysis. If examinations are to show the truth about candidates, if these candidates are to show their best side during them and if the results awarded are to be shown to employers and organisations, then we should feel obliged to show that we are aware of what English language tests are doing with the language they use and the language they elicit.

References Archer, D., and P. Grundy. 2011. “Linguistic Pragmatics,” in The Pragmatics Reader edited by D. Archer and P. Grundy, 11-18. London: Routledge. Brooks, L. 2009. “Interacting in pairs in a test of oral proficiency: Coconstructing a better performance.” Language Testing 26 (3): 341-366. Cambridge ESOL. 2005. FCE, CAE and CPE Handbooks for Teachers. Cambridge: Cambridge ESOL. —. 2011. CPE Handbook for Teachers. Cambridge: Cambridge ESOL. Cambridge English Language Assessment. 2013a. Cambridge English Proficiency: Brief exam guide. Cambridge: CELA. http://www.cambridgeenglish.org/images/21952-cpe-proficiencyleaflet.pdf —. 2013b. CPE Handbook for Teachers for exams from 2015. Cambridge: CELA. http://www.cambridgeenglish.org/images/168194-cambridge-englishproficiency-teachers-handbook.pdf. Cohen, A. D. 2010. “Coming to terms with Pragmatics,” In Teaching and Learning Pragmatics: where language and culture meet edited by N. Ishihara, and A. D. Cohen, 3-20. Harlow: Pearson Education Limited. Harrison, M. 2012. Cambridge English Proficiency Practice Tests (with key). Oxford: OUP. Hofstadter, D. 2007. I am a Strange Loop. New York: Basic Books. Hughes, A. 1989. Testing for Language Teachers. Cambridge: Cambridge University Press. Lycan, W. G. 2000. Philosophy of Language. London: Routledge. McEnery, T., R. Xiao, and Y. Tono. 2006. Corpus-Based Language Studies. Abingdon: Routledge. McNamara, T. 1996. Measuring Second Language Performance. Harlow: Longman. Milanovic, M. 2013. “A look into the future.” Cambridge English: Research Notes. Issue 51. February 2013, 31-33.

“You Have No Choice”

181

http://www.cambridgeenglish.org/images/130828-research-notes-51document.pdf. Oxford University Press. 2012. The Oxford Test of English White Paper. Oxford: Oxford University Press. —. 2012. The Oxford Test of English White Paper Summary. Oxford: Oxford University Press. Pennycook, A. 2010. Language as a Local Practice. Abingdon: Routledge. Phillipson, R. 1992. Linguistic Imperialism. Oxford: Oxford University Press. —. 2009. Linguistic Imperialism Continued. Hyderabad: Orient Black Swan/Routledge. Pratt, M. L. 1977. Toward a Speech Act Theory of Literary Discourse. Bloomington, IN: Indiana University Press. Searle, J. 1999. Mind, language and Society. London: Orion Books. Seuren, P. A. M. 2011. “Western linguistics: An historical introduction.” In The pragmatics reader edited by D. Archer, and P. Grundy, 55-67. London: Routledge. Reprint, Oxford: Blackwell, 1998. Shohamy, E. 2001. The Power of Tests. A Critical Perspective on the Uses of Language Tests. Harlow: Pearson Education Limited. Simpson, P. 2004. Stylistics. London: Routledge. Stalnaker, R. C. 2011. “Pragmatic Presuppositions,” reprint. In The pragmatics reader edited by D. Archer, and P. Grundy, 68-78. London: Routledge. Reprint, in “Semantics and Philosophy: Essays” edited by M. K. Munitz, and K. P. Unger, New York: New York University Press, 1974. Weir, C. J. 2013. “Measured constructs: A history of Cambridge English language examinations 1913–2012,” Cambridge English: Research Notes. Issue 51. February 2013, 2-10. http://www.cambridgeenglish.org/images/130828-research-notes-51document.pdf.

Notes 1

For an example of this thinking, see Milanovic (2013). Cambridge English Language Assessment (2013a). 3 For presupposition see Seuren, 1998 and Stalnaker, 1974; for performatives see Searle, 1999. It is perhaps interesting to note that neither Hughes (1989) nor Shohamy (2001) make any direct reference to pragmatics in their books on language testing. 4 The relevant Cambridge Esol material is to be found at: cambridgeenglish.org. 2

182 5

Chapter Seven

There are many publications that claim to mimic Cambridge examinations in order to facilitate candidates’ preparation for the examinations. Often pragmatic issues are revealed more obviously in these (admittedly peripheral) publications, allowing the analyst to enrich the data reading and evaluation process. An example used in this research is Harrison (2012). 6 Among the myriad websites, one of this kind that might be considered typical is examenglish.com (last access 15-06-2013), which offers brief examination descriptions, practice tests and practical advice for the would-be test taker. 7 Washback, or backwash, is a term used in testing to refer to the effects a particular type of examination may have on teaching in the classroom; a simple example might be the case of an entirely written end-of-year test which tempts teachers to ignore speaking skills and prioritise grammatical accuracy, orthography etc. A good, simple introduction to the concept is to be found in Hughes (1989, 1-2 and 44-47). 8 For example, see the Cambridge ESOL Key Features of Examinations as listed in their 2005 publications FCE, CAE and CPE Handbooks for Teachers, p. 2., where they claim to “relate the examinations to the teaching curriculum in such a way that they encourage positive learning experiences, and seek to achieve a positive impact wherever possible.” 9 See Cambridge English Language Assessment (2013a), for the main features of all four papers. 10 I sat the whole paper as a trial in March 2013, using the Cambridge English website version. 11 These examples are all taken from the sample CPE paper in Cambridge English Language Assessment (2013b). 12 This is a modified rather than a “pure” cloze test because instead of being a text subject to arbitrary, nth-word deletion, the gaps are selected, perhaps to examine knowledge of certain lexis, awareness of grammatical form or familiarity with collocation patterns. Again, being able to understand the rationale of selection (of gaps in the text) might be an advantage. 13 From the rubric to Part 1 of the CPE sample paper, Cambridge English Language Assessment (2013b). 14 Of relevance here is a reading of Phillipson (1992, and 2009, 111-112), and Pennycook (2010). 15 The question posed by Hofstadter (2007, 62) is a nice (if exaggerated) exemplification of the problem of true-false questions: “If the meanings of “true” and “false” were switched, this sentence wouldn’t be false.” Besides natural confusion that might be found in an individual true-false item that presupposes a simple bivalent interpretation (the more complex the text, the less likely it is that a simple true-false dichotomy can accurately reflect the meaning of a word, sentence or paragraph), we should also recognise the intrinsic difficulty in spotting something false in a text with little pragmatic sense due to its lack of context. In other words, in reading comprehension or listening comprehension texts, things tend only to be “false” in cohesive rather than coherent terms. In addition I must mention the problematic concept of “not-in-text” as an item answer which

“You Have No Choice”

183

discourages inference and implication and quickly becomes virtually unanswerable when it is not absolutely obvious. 16 These are available in the CPE Handbook for Teachers 2013 (Cambridge English Language Assessment 2013a; 2013b), or in preparation practice books such as Harrison (2012, 92-93). 17 Here I make reference to the highly structured way Cambridge English insist speaking examinations be performed. Besides the three-part superstructure of the test, the rubric lays down explicitly what the “interlocutor” may say and how many times a question may be repeated throughout the procedure. Again we can see this as an admirable attempt to ensure reliability (tests will not vary much in terms of questions asked or time given to the examination). 18 For a reasonably up-to-date treatment of this issue see Brooks (2009). 19 Cambridge English examiners are trained to respect the wording of the speaking tests very closely and to keep time almost religiously. This is yet another example of the respect for reliability considerations in the design and conduct of the examinations. 20 I suggest the term “metapragmatic” merely to emphasise the importance of the linguistic situation that exists beyond the immediate context created by a particular item or instrument. Candidates have to respond logically to a stimulus or question and at the same time satisfy another audience (the scorers). 21 I offer mild apologies for the rather contrived play on words, but the sense is precisely this: it is a performance which in many aspects conforms to the Austinian-Searlian conception of performatives. After you make your contribution, you are never quite the same again in the eyes of the assessor and, perhaps, ultimately in the published results of the examination board. See Searle (1999) for a good, accessible outline of the idea of performatives in language and speech acts in general. 22 This represents an admirable attempt by Cambridge to ensure credible assessment throughout the world. However, I would question the fact that tests are not all recorded (technologically a simple task and one which would modify rater behaviour significantly), and suggest that there is a real risk that other contextual factors have influence over test results (the non-verbal behaviour of candidates, how friendly or self-confident or even good-looking they are et cetera): these require systematic analysis. 23 For a brief but interesting discussion of what can become a philosophically very complex topic involving truth value and negation, semantics and pragmatics, see Archer and Grundy (2011, especially 16-17), and Seuren (2011). 24 For details on the method of item writing, banking and selection, and of pretesting, see the Cambridge ESOL (2011) and Cambridge English Language Assessment (2013b) and Cambridge English: Research Notes (various editions, especially Issue 51, February 2013). 25 Cambridge English are admirable in their efforts at inclusion, even, for example, of hearing impaired candidates. Their values may well be estimable, but that does not detract from their political and ideological content.

CHAPTER EIGHT DCTS VERSUS NATURALLY OCCURRING DATA IN THE REALIZATION OF DISAGREEMENT BY NON-NATIVE SPEAKERS OF ENGLISH CARMEN MAÍZ-ARÉVALO

1. Introduction In comparison with other speech acts (e.g., compliments, requests, complaints, etc.), the speech act of disagreement has received less scholarly attention despite its obvious complexity. Furthermore, most studies on disagreement focus on native varieties of English (Marra 2012; Gruber 2001; Holmes and Marra 2004; Muntigl and Turnbull 1998; ReesMiller 2000, inter alia) or contrast English with other languages such as Greek (Georgakopoulou 2001; Koutsantoni 2005; Sifianou 1992), Chinese (Bond, Zegarac, and Spencer-Oatey 2000; Cheng 2003; Liang and Han 2005), Japanese (LoCastro 1986; Nakajima 1997) or Spanish (García 1989; Santamaría-García 2003), to mention but a few. However, the realization of this speech act by non-native speakers of English has been scarcely studied, with some noticeable exceptions (Pearson 1986; Beebe and Takahashi 1989; Bardovi-Harlig and Salsbury 2004; Cheng and Warren 2005; Lawson 2009). Interestingly enough, the results of these studies are contradictory, which might be due to the use of different data collection methods – e.g., interviews, role plays or discourse completion tasks (DCTs henceforth). The debate about different methods to collect pragmatic data is not recent (see Yuan 2001; Golato 2003; Schauer and Adolphs 2006) but still far from having reached a unanimous agreement. The present paper aims to compare the realization of the speech act of disagreement produced by non-native speakers of English in naturally occurring online data (more concretely, in an asynchronous forum) and a

186

Chapter Eight

DCT filled in by the same participants. My ultimate aim is to observe whether or not results are actually influenced by the method of data collection. In other words, I intend to find out whether the participants express their disagreement in the same way when they interact naturally online or when they are asked to fill in a DCT. Should the method of data collection have an impact on the data, this paper would significantly contribute to reassess previous studies concerning the expression of disagreement by non-native speakers of English. In addition, this paper tackles naturally occurring data from a relatively under-explored approach since it focuses on the online production of students’ interaction rather than recording their oral interactions, which might also affect the way they speak. As for the approach adopted, it is both quantitative and qualitative, despite the limited size of the DCT sample (15 items) and the complexity of the naturally occurring data (where the combination of different strategies is pervasive). However, it is my belief that a combination of both approaches might help interpret the results more thoroughly. The rest of the paper is organised as follows: section two offers a brief overview of the most relevant studies concerning the expression of disagreement by non-native speakers of English, paying particular attention to the methodology employed and the results obtained. Section three describes the data and analytical methods employed in the present study. Section four presents and discusses the results. Finally, the last section of the paper offers some conclusions and points to future research.

2. Literature review Pearson (1986) can be considered the first seminal study concerning the expression of disagreement by non-native speakers of English. More specifically, she compared disagreements produced by Japanese learners of English as a foreign language (EFL henceforth) with those produced by native speakers of American English in recorded conversations. Her results showed that Japanese learners opt for explicit expressions of disagreement or for not expressing disagreement at all. Beebe and Takahashi (1989) also focused on the cross-cultural comparison between American and Japanese speakers in the realization of face-threatening acts such as disagreeing. Their data were gathered by means of field notes – a method whose drawbacks they acknowledged themselves (1989, 201) – and DCTs. They explained their methodological choices declaring that both methods give similar results since “subjects’ intuitions about what they would say correspond closely to what other subjects actually did say in the same situation.” Contrary to Pearson’s

DCTs versus Naturally Occurring Data

187

findings, Beebe and Takahashi found that Japanese EFL learners became extremely direct when expressing disagreement in English –even more so than American speakers –since they thought they had to abandon the Japanese formula of indirectness and implicitness when realizing this speech act in English. Slightly more recently, Bardovi-Harlig and Salsbury (2004) carried out a one-year longitudinal study of the disagreements produced by twelve learners of English as a second language (ESL henceforth) as they interacted with native speakers during conversational interviews. Instead of using DCTs, the authors recorded real exchanges between the native and non-native interlocutors. Interestingly, the longitudinal character of the study reflected progress in the students’ acquisition of pragmatic competence. Thus, they report that, at the very beginning, non-native speakers tended to express their disagreement in a direct way but ended up elaborating their disagreements as time passed. Cheng and Warren (2005) focused on the forms and pragmalinguistic realization of disagreement in a naturally occurring oral corpus of business discourse between native speakers of English and Chinese speakers of English as an L2. The authors compared the way disagreement is taught to Chinese learners in English textbooks with the real data, to find out, among other things, that strong disagreement is actually slightly more frequently used by native speakers than Chinese L2 speakers of English, even if it occupies a prominent place in textbooks. Furthermore, and as opposed to other studies, where strong or blunt disagreement seems to be common among non-native speakers, analysis of real data reveals that most of the disagreements are politely mitigated. Kreutel (2007) also compared the production of EFL learners with that of native speakers of American English. Using DCTs to gather her data, Kreutel’s findings confirm what has often been reported in the literature; namely, non-native speakers’ expressions of disagreement tend to lack complexity (Bell 1998; Nakajima 1997). EFL learners also show a tendency to express disagreement bluntly rather than employing mitigation strategies (Pomerantz 1984). Another common non-native strategy seems to be not to express any disagreement at all (see also Pearson 1986 and Horne 2004). In line with Bardovi-Harlig and Salsbury (2004), Lawson (2009) also recorded interviews between thirty Japanese speakers of English and thirty native speakers of American English discussing on polemic issues. Some of his findings replicate earlier studies (e.g., the use of more blunt expressions of disagreement in the case of non-native speakers such as “I don’t think so”). However, he also reports other findings that contradict previous research, such as the frequent use of the performative “I

188

Chapter Eight

disagree” among native speakers in contrast with non-native ones. The contradictory nature of the aforementioned results raises the question whether the different data collection methods employed in these previous studies may have played a significant role, rendering a comparison of two of these methods (DCTs and naturally occurring data) highly desirable.

3. Methodology 3.1. Discourse completion tests and online naturally occurring data Much has been written about discourse completion tests or tasks. It is not the intent of this section to repeat the arguments for or against that this frequently used data-gathering method has provoked (see Yuan 2001; Golato 2003). Suffice it to say that, despite its drawbacks, DCTs are undeniably useful when it comes to reflecting the speakers’ intuitions (Schauer and Adolphs 2006). Golato (2005, 14) puts it extremely clearly when stating that “[DCTs] are better suited to the study of ‘what people think they would say’ than to the study of ‘what people actually do say in a given speech setting.’” Following Golato (2003), the purpose of this study is precisely to compare the expressions of disagreement employed by a group of non-native speakers both in a natural environment (online forum) and in a DCT.

3.2. Participants The participants involved in this study were ten female students enrolled in the Master in English Linguistics of the Complutense University of Madrid. On the whole, the students’ level of English was fairly advanced. In fact, except for two of them who barely had a B1 level, the rest of the students in the group had a C1 or C2 level. Indeed, some of them are EFL teachers in their countries of origin. The participants, whose age ranged from mid-twenties to late thirties, came from different countries (e.g., Egypt, Iran, South Korea, Poland, Russia, Romania and Spain). English was used as a lingua franca both in the teacher-students and student-student interactions. With the purpose of avoiding biased exchanges, the subjects were not informed a priori of their participation in the experiment. However, they were informed once the whole experiment was over and were asked for their written consent, even if all the data were anonymous to protect their identities.

DCTs versus Naturally Occurring Data

189

3.3. Procedures of data collection and analysis The data used in the present study was gathered by means of two procedures: an online forum and a DCT. The following paragraphs describe both of them in detail. First, the ten students were asked to carry out a group assignment in two random groups of three members and one group of four. Their assignment consisted in analysing a series of multimodal texts. Students were required to carry out the analysis of the texts and upload a final paper with their results. As opposed to the traditional face-to-face group discussion, however, they were told to carry out their discussion online. For that purpose, a forum was created for each of the three groups and the students were given one week to do the activity. As their teacher, and to avoid biasing their negotiation process, I refrained from interfering, except when explicitly asked for help. Finally, the students were also told that they were free to organise their forum discussion as they considered best. The forum itself (as well as participation in it) was not to be assessed but just the final product (i.e., the paper each group had to upload). The decision not to assess the students’ participation or discussion in the forum was determined by the fact that, in previous studies, it had been shown that assessing the forums biased students’ spontaneity (Carretero, MaízArévalo, and Martínez 2014). As already mentioned, I tried to avoid biasing the students’ natural exchanges in the forums as much as possible. On the whole, this corpus comprises 15,598 words and fifty-nine cases of disagreement (n= 59). Even if limited, the activity students had to do boosted their discussion and negotiation and, as a result, the expression of disagreement. It is also necessary to mention that disagreement was never conflictive, and students reproduced in the forum the friendly atmosphere they also enjoyed in class, which proves that disagreement is not necessarily linked to conflict. In such a context, disagreement can even be regarded as a welcome collaborative effort to reach a satisfactory consensus. This does not mean, however, that disagreeing does not constitute a face threat. On the contrary, the participants are still extremely careful to “get one’s point across without seeming self-righteous or being injurious” (Locher 2004, 94). With regard to the DCT, it was presented in class as a practical example of the subject’s module devoted to data-gathering methods. Students were also provided with examples of other data-gathering procedures like questionnaires, interviews, field observation, etc. Thus, the students were requested to fill in a sample DCT so that they became familiar with how this data-gathering method worked. The DCT consisted of four situations, some of which they had actually encountered while

190

Chapter Eight

interacting in the forum. The students had to provide a written answer to the different scenarios (see Figure 1). Once again, and to avoid biasing results, they did not know a priori that their answers to the DCT would be under scrutiny or compared to their naturally occurring disagreements in the e-forum. Since the purpose of the present paper is to compare the expressions of disagreement produced by the students in L2 (English) in the DCTs and their actual expressions of disagreement in the naturally occurring data, only the students’ responses to the fourth situation (see Figure 1) have been taken into consideration. It is thus beyond the scope of this study to analyse the participants’ expression of agreement in L1 and L2 and their expression of disagreement in L1. Further research will focus on these aspects, with special interest in the comparison between expressions of disagreement in L1 and L2. The students provided 15 possible responses (since some of the students gave more than one possibility). Figure 1 shows what the students had to answer in the DCT: Situation 1: You are preparing an essay with two of your classmates (who are both the same nationality as you are) and one of them has made a comment you agree with. How would you express your agreement in your mother tongue? You would say: (please offer an English translation) ……………………………………………………………………………………… ……………………………………………………………………… Situation 2: You are preparing an essay with two of your classmates (who are both the same nationality as you are) and one of them has made a comment you do not agree with. How would you express your disagreement in your mother tongue? You would say: (please offer an English translation) ……………………………………………………………………………………… …………………….……………………………………………… Situation 3: You are preparing an essay with two of your classmates; the three of you belong to different nationalities and use English as lingua franca. One of your classmates has made a comment you agree with. How would you express your agreement in English? You would say: ……………………………………………………………………………………… ……………………………………………………………………… Situation 4: You are preparing an essay with two of your classmates; the three of you belong to different nationalities and use English as lingua franca. One of your classmates has made a comment you do not agree with. How would you express your disagreement in English? You would say: ……………………………………………………………………………………… ……………………………………………………………………… FIGURE 1. DCT employed in the current study

DCTs versus Naturally Occurring Data

191

Section 4 analyses the students’ expressions of disagreement during their online interaction in the forum and contrasts them to their answers in the DCT. Before the actual analysis is carried out, I shall briefly describe the taxonomy employed for that purpose.

3.4. A taxonomy of the expressions of disagreement Several taxonomies have been put forward in the study of disagreement. It is beyond the scope of this paper to review all of them in detail. All of them, however, share some common aspects, such as the distinction between “strong” and “weak” disagreement (Pomerantz 1984). Thus, strong disagreement consists only of disagreement components (e.g., “I don’t think so”) whereas weak disagreement includes other components like hesitation markers and fillers, token agreement, hedges, etc. Kakavá (1993) includes a third element in Pomerantz’ taxonomy: “mitigation.” Hence, she distinguishes between strong forms, strong yet mitigated, and mitigated forms of disagreement. The difference between the second and third types lies in the explicitness of disagreement in the second type as opposed to the third type, where disagreement may be accompanied by digressions or reformulations. Rees-Miller (2000) also considers three types of disagreement: softened disagreement, unmodified disagreement – i.e., Pomerantz’ strong disagreement – and aggravated disagreement. Finally, Cheng and Warren (2005) distinguish between strong disagreement, disagreement and polite disagreement. Disagreement, thus, can be mitigated or not, softened or uttered without any softeners. The speaker’s choice for one form or another is dependent on context. These taxonomies, however, have been proposed mostly with native speakers of English in mind. By contrast, Kreutel (2007) puts forth her classification after analysing the production of EFL students. Thus, she distinguishes between what she defines as “desirable features” and “undesirable features”, and establishes the following taxonomy (2007, 326): (a)

Desirable features: a. Token agreement b. Hedges c. Requests for clarifications d. Explanations e. Expressions of regret f. Positive remarks g. Suggestions

192

Chapter Eight

(b) Unddesirable featuures: a. M Message abanddonment b. T Total lack of mitigation m c. U Use of the perfformative “I disagree” d d. U Use of the perrformative neg gation “I don’tt agree” e. U Use of the baree exclamation n “no” f. B Blunt statemennt of the oppossite In my view, this distinction is hellpful but som mehow simplisstic, since disagreemennt is always contextually--bound and w what makes a certain expression ““desirable” inn one context might not bee so in a diffeerent one. Previous ressearch, howevver, seems to agree a that “m mitigated” disaagreement (what Kreuttel defines as “desirable”) “ teends to be “prreferred”, even if some researchers (see Lawson 2009) have proved p that nnative speakerrs do also produce “unndesirable disaagreements.” For the current analysis, I have reetained Kreuteel’s linguistic features. The reason for doing so is that I find them extremeely useful in the sense that a list off specific linguuistic realizations helps the researcher to carry out her analysis in a more sysstematic way, even if there are other equ ually valid and helpful taxonomies (e.g., ( Cheng and a Warren 20005). Howeveer, I have b “dessirable” and ““undesirable”” features omitted herr distinction between since I firmlly believe thatt desirability is i determinedd by context an nd not by the linguistiic form itselff. A more neutral distinctiion into stron ng versus mitigated diisagreements has been therrefore preferreed. Figure 2 illustrates i the taxonom my employed, together with some examplles:

FIGURE 2. Taxxonomy adopteed in the currentt study

DCTs versus Naturally Occurring Data

193

4. Analysis of results In this section, I present the findings, together with a comparison between the two data-gathering methods. For the sake of clarity, the general results will be summed up in Table 1 below, followed by a more detailed analysis of both strong and mitigated disagreement (sections 4.1. and 4.2. respectively). In the case of the DCT, students provided fifteen possible responses (some of them gave more than one) when asked how they would disagree in English. Focusing on their responses in L2, i.e., what they would say in English in this particular scenario, it can be initially observed that participants generally opted for expressing their disagreement in a mitigated way (see 4.2.). This already suggests that a high linguistic competence seems to have a decisive say in students’ pragmatic awareness. Figure 3 below reproduces students’ responses to the DCT in situation 4. Disagreement in L2 (English) I am not sure I am afraid I don’t quite agree Do you think so? I’m not really sure, maybe… I’m sorry but I don’t think so. Well… Maybe you’re right. I’m not 100% sure, but what about this? We can try both ways and then think about it more later [sic]. O.K.? I’m sorry, I don’t agree with you. Maybe not, who knows? Why? I don’t see your point… maybe it’s the other way round. I don’t think so. I don’t agree… I see your point here, but maybe we could also look at it from another perspective Aha ok, but maybe …. I am afraid I don’t agree with you. I’m afraid I can’t agree with you. It’s a good idea, but ….. FIGURE 3. Participants responses to the DCT

As for the naturally occurring data, the fifty-nine occurrences had to be manually identified by the researcher. Initially, a software programme like Wordsmith was tried but it only allowed identifying key words – e.g., “sorry”, “agree”, etc. and not discursive patterns like the ones actually used by the participants. This drawback has also been reported by other researchers on discourse (Carretero et al. 2014), who hence advocate for the need to use manual analysis. In fact, and as stated by Cheng and

194

Chapter Eight

Warren (2005, 248-249), “it is not possible to automatically search for disagreements.” Finally, it is important to mention that the whole naturally occurring corpus has not been included in the paper for the sake of space, although it will be exemplified thoroughly in the following sections. Table 1 summarises the general quantitative results, where “n” represents the number of occurrences in each of the corpora followed by its corresponding ratio. The following sections will compare the two sets of data in more detail. TABLE 1. Cases and ratio of disagreement in the DCTs and the naturally occurring data Type of disagreement

DCT (n= 15)

Strong disagreement Mitigated Token agreement disagreement Use of hedges Giving explanations Asking for clarification Expressions of regret Making positive remarks TOTAL

2 4 3 0 2 4 0 15

13% 27% 20% 0% 13% 27% 0% 100%

Naturally occurring data (n = 59) 0 0% 14 23.7% 17 29% 14 23.7% 9 15.2% 2 3.4% 3 5% 59 100%

4.1. Strong disagreement As already mentioned, strong disagreement is characterised by a total lack of mitigation (e.g., “no way”, “no, no, no”, “of course not”, etc.), the use of performatives or the blunt statement of the opposite. For Kreutel (2007), strong disagreement is undesirable and is more frequently employed by non-native speakers than by native ones. This might be due to her methodology or to the fact that the participants in her study did not have a strong linguistic competence. In their answers to the DCT, the participants of the present study demonstrated initial knowledge of what might be desirable ways to express disagreement in a collaborative context like the one in which they were involved. As their answers to the DCT show, participants were aware that the use of bare negative forms can be too face-threatening in a convivial context like this. Thus, their answers to the DCT showed a clear tendency to avoid strong disagreement, which only appeared on two occasions (13.3%), and both used by one of the Spanish speakers (who is also, significantly enough, one of the two students with a B1 level); namely, “I don’t think

DCTs versus Naturally Occurring Data

195

so” and “I don’t agree.” This may be due to an interference with Spanish, where the direct expression of disagreement is rather frequent in everyday conversation and far from considered impolite. This intuition is reflected in the naturally occurring data, where there is not a single case of blunt negations or the performative “I disagree.” On the three occasions where the negative performative “I don’t agree” appears, it is always carefully mitigated in quite complex combinations, which dismisses them as expressions of a “strong disagreement.” Example (1) illustrates this, where speaker (A) does not agree with (B) and asks for (C)’s opinion on the matter. (1)

There is one part in the text on which I still do not completely agree with [B]. The picture of the prince and Cinderella kissing, I think Cinderella is an actor in this case because it seems like she is kissing the prince back; however, [B] sees her as a passive participant; what do you think [C]?

Here, the negative performative (“I still do not completely agree with”) is preceded by a minimization of the disagreement (“there is one part”). This is followed by (A) giving reasons and explaining why she disagrees. All of it is accompanied by hedges (“I think…”, “it seems”). Finally, (A) asks for (C)’s opinion (“what do you think?”), which is also a mitigating strategy. In this case, it could be concluded that the students’ intuitions as to what one should do or not when expressing disagreement in such a context seem to reflect their naturally occurring answers closely. In the light of such results, it could be concluded that DCTs parallel naturally occurring data. However, as will be seen in the following sections, this is far from the whole picture.

4.2. Mitigated disagreement Mitigated disagreement is characterised by the use of elements that help minimise the directness of a blunt disagreement. To minimise being too direct, speakers can mitigate their disagreement by different means, namely: token agreement (i.e., “yes, but”), use of hedges, requests for clarification, expressions of regret, use of positive remarks, suggestions and explanations. The participants’ high linguistic competence shows that they are well aware of mitigating strategies, as reflected in their responses to the DCT (see figure 4).

Chapter Eight

196

4.2.1. Token agreement The DCT reveals that this strategy is already known to some of the students. Token agreement is employed on four occasions in the DCT (26.6% of the cases); for example: “I see your point here, but maybe we could also look at it from another perspective”, “Aha [sic] ok, but maybe…” or “It’s a good idea, but…”. When interacting with each other in the e-forum, students employed token agreement on fourteen occasions (23.7% of the cases), either alone or in combination with other mitigating strategies. Examples (2) to (4) below illustrate this: (2)

[Name], I totally agree with your first analysis fairy tales have not changed much in twenty years, as regards images in both Cinderella and Snow White. However, I would like to point something that strikes me in the images of Snow White in 2005 if you have a look at both the princes and the princess they are more or less at the same level, I mean in the picture. In the images of Cinderella in 2005 the princess is like in a superior position from the prince in a step, although she is the goal.

(3)

I agree with the last part. But, why do you think Cinderella is a phenomenon in 2005? I think the prince is not smiling at her since she’s looking out of the window.

In extract (3), token agreement (“I agree… but”) is combined with other mitigating mechanisms such asking for the addressee’s opinion (“why do you think…?”), hedges (“I think…”) and explanations (“since she’s…”). Example (4) is even more complex: (4)

Very interesting analysis, [Name]! Thank you for posting it! My ideas concerning Task 3 are basically the same, except for the third pair of images. I think the 1985 image rather than complementary/enhancing is contradictory...I don’t see any sign of the Prince and Snow White being happy in that image; it’s rather a gloomy image in my opinion. I agree with your reading of the Palace as a symbol of wealthiness and power.I agree with the last part. But, why do you think Cinderella is a phenomenon in 2005? I think the prince is not smiling at her since she’s looking out of the window.

In (4), the token agreement (“My ideas are basically the same, except for…”) follows other mitigating strategies – i.e., positive evaluation of the interlocutor (“very interesting analysis, [name]!”) and acknowledgement of her work (“Thank you for posting it! -”). The student closes her

DCTs versus Naturally Occurring Data

197

comments by adding a point for agreement. This combinatory pattern reflects what Kreutel (2007, 338) has defined as “sandwich pattern”: Most of the NSs [native speakers] use mitigation not only at the beginning of their utterances, but also at the end, creating a certain sandwich pattern that “wraps” the dispreferred speech act into preferred reactions. (Emphasis in the original)

Significantly enough, examples like (4) were produced by participants with an extremely high linguistic competence (they are practically bilingual speakers). In contrast to that, the weakest students in the group (with only a B1 level) did not produce any token agreement or combined responses like these. One of these two weaker students (she barely reaches a B1 level) even opted for message abandonment. In fact, she usually failed to respond to the questions posed by her group members, even if they explicitly asked for her opinion. 4.2.2. Hedges The DCT shows that hedges seem to be a frequent strategy employed by the participants (20% of the cases), which replicates other studies such as Lawson’s (2009) and Kreutel (2007) where non-native speakers also use hedges abundantly. Favourite hedges seem to be “I (don’t) think” and modal verbs (e.g., “could”, “may”) and adverbs (e.g., “maybe”). This is illustrated by the following examples from the DCT: “I am not sure”, “Maybe not, who knows?”, “Do you think so? I’m not really sure, maybe…”, “Well… Maybe you’re right. I’m not 100% sure, but what about this? We can try both ways and then think about it more later [sic].” Not surprisingly, hedges were also fairly frequent in the naturally occurring data (29% of the occurrences), as exemplified by extracts (5) and (6) below: (5)

It seems that we have a similar analysis more or less. However, I’m not sure whether they are represented as equals. I think that even though the gap between them is reduced in the second image, we still have the prince as the active actor. So, they don't have an equal representation. What do you think?

(6)

Why do you think the prince is depicted in this way?'..I'm not sure about this, but I think the fact that the prince is depicted with his whole body in the first image might enhance his strength..and therefore power over Snow White..the face, on the contrary, is not

Chapter Eight

198

clearly shown, since it is the part of our body that shows our emotions...so maybe in 1985 the prince's feelings towards Snow White were not considered so important. In the 2005 image, we can see the prince's face, his eyes closed while kissing Snow White, which might indicate a deep feeling of love towards her. Please let me know what you think about these issues and then maybe we could go on with the other images. -

Examples (5) and (6) also serve to illustrate how hedges were often employed together with other mitigating strategies like asking for the interlocutor’s opinion, requesting for clarifications and giving explanations. Hedges were used both by strong and weak students. In the case of the latter, however, expressions like “I (don’t) think” or “I’m not sure” were favoured over others like modal verbs. This might be due to the fact that the first type of expressions is learnt from the early stages whilst modal verbs are more difficult to master. Example (7) illustrates the use of hedges by the B1-level student: (7)

Hi girls, I am sorry to answer so late but, as you know [Name], I am still sick. Well, I think that when [Name] says there are two participants, she refers to the actor and the goal, but I am not sure if we have to consider Snow White as a participant because she is passive.

4.2.3. Giving explanations Giving an explanation is a natural mitigating strategy (Al Khatib 2006). However, their responses to the DCT seem to indicate that students were unaware of this strategy when expressing friendly disagreement. In fact, there is no instance of giving explanations in the DCT. This might mislead us into believing that non-native speakers do not employ this strategy when disagreeing. The analysis of the naturally occurring data, however, reveals that this is not the case. This again proves the drawbacks of using DCTs to collect data. Thus, whilst students never employed giving explanations in the DCTs, they used this strategy on fourteen occasions in the naturally occurring data (23.7% of the cases), as illustrated by example (8) below: (8)

I agree with your analysis of invited her to dance. I thought it was material because you could invite someone without the need of speech. His gesture would be an indication of this. However, I think we could consider it verbal.

DCTs versus Naturally Occurring Data

199

Example (8) illustrates Kreutel’s “sandwich pattern.” After the initial agreement (“I agree with your analysis…”), the speaker makes her point and explains why she thought so (“…because you could invite someone…”). Giving explanations seems to be more commonly used by stronger students although weaker students used it occasionally, too, as in example (9): (9)

In my opinion, she doesn’t kiss the prince because this is not possible. She is sleeping. But if you two think in other way, we can put that. But I don’t think she is an actor because she does nothing, because she is sleeping.

In (9), the student repeats her explanation twice, which proves that, for her, it is a critical factor for the analysis of the text (“because this is not possible. She is sleeping” and “because she is sleeping”). Although she uses hedges (“I don’t think”) and tries to please her classmate (“But if you think in other way, we can put that”), she ends up her comment by stating her disagreement once again. 4.2.4. Requesting for clarifications By requesting for a clarification, the speaker gives the addressee the opportunity to re-state their opinion. In the DCT, requesting for clarifications only appears on two occasions (13%) (“Do you think so? I’m not really sure, maybe…” and “Why? I don’t see your point… maybe it’s the other way round”). Inspection of the naturally occurring data shows this strategy is slightly more frequently used (15.2%). In fact, requests for clarification appeared on nine occasions, often as part of a combination with other strategies. Examples (10) and (11) serve to illustrate this (requests for clarification appear in bold for the sake of clarity): (10) I agree with the last part. But, why do you think Cinderella is a phenomenon in 2005? I think the prince is not smiling at her since she's looking out of the window. (11) Girls, I think we still need to work on the second part. There are some things which are not clear to me. [Name] mentions that there is the goal and then she says that there is no goal. [Name] could you clarify a bit? Maybe we should talk about it in here and then copy the final version to our file. We can leave it for tomorrow also, I have another assignment to work on so I think I will leave this one for today. Thanks [sic], [Name]

Chapter Eight

200

4.2.5. Expressions of regret Prior research agrees on the fact that expressions of regret tend to be over-exploited by non-native speakers (Kreutel 2007). A possible explanation might be that the phrase “(I’m) sorry” is one of the first functional expressions learnt by students (together with other formulaic language like “thank you”). This is also reflected in the participants’ response to the DCT, where expressions of regret account for more than one third of the total number of responses (27% of the cases), in different varieties: “I am afraid I don’t quite agree”, “I’m sorry but I don’t think so”, “I’m sorry, I don’t agree with you”, “I am afraid I don’t agree with you” or “I’m afraid I can’t agree with you.” Furthermore, expressions of regret in the DCT are not only reported by students with a lower linguistic competence but also by others whose linguistic competence is quasi-native. Once again, looking simply at the DCT might lead us to conclude that, in fact, non-native speakers overuse this mitigating strategy. However, the analysis of the naturally occurring data reflects there is a striking contrast between what students think they do (as reflected in their answers to the DCTs) and what they actually do. Thus, expressions of regret only occur twice (3.4%) in the naturally occurring corpus and, in both cases, they are produced by one of the students with a lower linguistic competence, as illustrated by example (12): (12) I am so sorry but I don’t see any Cinderella kissing the prince, do you mean Snow White? In this case, I think that the prince is the actor. She is not an actor because he is kissing her but not the other way around.

4.2.6. Making positive remarks about the speaker This strategy involves complimenting the interlocutor in an attempt to mitigate the impending disagreement. Complimenting the interlocutor is absent in the participants’ responses to the DCTs, which might reveal their unawareness that this strategy can also be used to minimise disagreement. However, inspection of the naturally occurring data reveals that students (especially those with a strong linguistic competence) do use this strategy, even if not particularly often (5% of the cases). Example (13) serves to illustrate this strategy: (13) Excellent analysis, [Name]! There’s just one point I don’t agree with: the fact that the two characters have the same role in the 1985

DCTs versus Naturally Occurring Data

201

image..It’s true that both the prince and Cinderella are Reacters and Phenomena (I think we can say this is a ‘reciprocal mental process’ or ‘bidirectional transactional reaction’...). However, whereas the prince only turns his head towards Cinderella, she turns her whole body towards him, so this is why I don’t think their roles are equal.

4. Conclusions This paper has aimed at contrasting the speech act of disagreement produced by non-native speakers of English in two different contexts: an artificial DCT and a naturally occurring online forum. This comparison intended to determine whether or not the method used for data collection plays a role in the way participants produce a particular speech act, in this case, disagreement among peers. DCTs show common patterns with the naturally occurring sample. Thus, strong unmitigated disagreement is generally avoided in both contexts. Likewise, hedges are favoured as mitigating strategies in both cases. However, differences outnumber similarities. On the whole, the sample DCTs offer extremely simple expressions of disagreement when compared to the naturally occurring data, where disagreement tends towards a high degree of elaboration and strategy integration in a quasinative manner. More specifically, there appear considerable differences between the disagreements collected by means of the DCTs and those of the naturally occurring data. Thus, whilst token agreement is used rather sparsely in the DCTs, it seems an extremely popular strategy in the naturally occurring data, especially among high-level students. Likewise, expressions of regret to express disagreement are extremely frequent in the DCT but virtually absent in the naturally occurring data. This proves that students are well aware of the double-edge nature of apologies in real interaction, which contradicts Kreutel’s results: I’m sorry is said to be generally overused by non-native speakers because it is acquired relatively early and used as a general means of avoiding confrontation by expressing humbleness and deference. Conversely, among native speakers, I’m sorry is usually associated with apologies, that is, the speaker acknowledges a mistake or failure on his or her part. […] This expression of reverence may be inappropriate when it comes to disagreement, indicating that a differing opinion is not necessarily a failure the speaker needs to apologize for. Accordingly, ESL learners should be sensitized that this feature reduces the authority and power behind a statement and may lead to the disagreement not being taken seriously by the listener. (Kreutel 2007, 331).

202

Chapter Eight

Hence, these students might be repeating early textbook learnt patterns in the DCTs, but avoid sounding too apologetic when interacting in a real context, especially if their linguistic competence is high. With respect to requests for clarifications, they are also virtually absent in the DCTs, which might mislead us into thinking they are a strategy unknown by learners. However, the naturally occurring data show that students do employ this mitigating strategy. A similar case is that of expressing positive remarks about the addressee or that of asking for clarification, which are practically absent in the DCTs but not in the naturally occurring corpus. Despite the limited size of the sample, it can however be concluded that these two different methods of data collection do have a distorting effect on the data. Results reflect that sticking just to DCTs might mislead the researcher into rather simplistic views of non-native speakers’ performance. Rather unfortunately, the size of this sample does not allow for generalisations. Future research will attempt to extend the collected data to corroborate present results. Likewise, other interesting aspects to be looked into in the future would be the contrast between different linguistic levels as well as the comparison between L1 and L2 expressions of (dis)agreement.

References Al Khatib, M. A. 2006. “The pragmatics of invitation making and acceptance in Jordanian society.” Journal of Language and Linguistics 5: 272-294. Bardovi-Harlig, K., and T. Salsbury. 2004. “The organization of turns in the disagreements of L2 learners: A longitudinal perspective.” In Studying speaking to inform second language learning, edited by D. Boxer, and A. D. Cohen, 199-227. Clevedon: Multilingual Matters. Beebe, L. M., and T. Takahashi. 1989. “Sociolinguistic variation in facethreatening speech acts. Chastisement and disagreement.” In The Dynamic Interlanguage: Empirical Studies in School Language Variation, edited by M. R. Eisenstein, 199-218. New York: Plenum. Bell, N. 1998. “Politeness in the speech of Korean ESL learners.” Working papers in educational linguistics 14 (1): 25-47. Bond, M. H., V. Zegarac, and H. Spencer-Oatey. 2000. “Culture as an explanatory variable: Problems and possibilities.” In Culturally speaking: Managing rapport through talk across cultures, edited by H. Spencer-Oatey, 293-315. London: Continuum.

DCTs versus Naturally Occurring Data

203

Burdine, S. 2001. “The lexical phrase as pedagogical tool: teaching disagreement strategies in ESL.” In Corpus linguistics in North America: Selections from the 1999 Symposium, edited by R. C. Simpson, and J. M. Swales, 195-210. Ann Arbor: University of Michigan Press. Carretero, M., C. Maíz-Arévalo, and M. A. Martínez. 2014. “‘Hope This Helps!’ An Analysis of Expressive Speech Acts in Online TaskOriented Interaction.” In Yearbook of Corpus Linguistics and Pragmatics 2014: New Empirical and Theoretical Paradigms, edited by J. Romero Trillo, 261-290. Switzerland: Springer. Cheng, W. 2003. Intercultural Conversation. Amsterdam: Benjamins. Cheng, W., and M. Warren. 2005. “//-> well I have a DIFferent // THINking you know //: A Corpus-driven Study of Disagreement in Hong Kong Business Discourse.” In Asian Business Discourse(s), edited by F. Bargiela-Chiappini, and M. Gotti, 241-270. Bern: Peter Lang. García, C. 1989. “Disagreeing and requesting by Americans and Venezuelans.” Linguistics and Education 1: 299-322. Georgakopoulou, A. 2001. “Arguing about the future: On indirect disagreements in conversations.” Journal of Pragmatics 33 (12): 18811900. Golato, A. 2003. “Studying compliment responses: A comparison of DCTs and recordings of naturally occurring talk.” Applied linguistics 24 (1): 90-121. —. 2005. Compliments and compliment responses: Grammatical structure and sequential organization, Vol. 15. Amsterdam: Benjamins. Gruber, H. 2001. “Questions and strategic orientation in verbal conflict sequences.” Journal of Pragmatics 33 (12): 1815-1857. Holmes, J., and M. Marra. 2004. “Leadership and managing conflict in meetings.” Pragmatics 14 (4): 439-462. Horne, B. 2004. “Culture and the language classroom. Challenging expectations of cultural behavior.” Bulleting of the Faculty of Education, Chiba University 52: 87-92. Kakavá, C. 1993. “Negotiation of disagreement by Greeks in conversations and classroom discourse.” PhD dissertation. Georgetown: Georgetown University. Kotthoff, H. 1993. “Disagreement and concession in disputes: on the context sensitivity of preference structures.” Language in Society 22: 193-216. Koutsantoni, D. 2005. “Greek cultural characteristics and academic writing.” Journal of Modern Greek Studies 23 (1): 97-138.

204

Chapter Eight

Kreutel, K. 2007. “‘I’m not agree with you.’ ESL learners’ expressions of disagreement.” TESL-EJ 11 (3): 1-35. Lawson, A. J. 2009. From the classroom to the bar-room: Expressions of disagreement by Japanese speakers of English. Master dissertation. Birmingham: University of Birmingham. Liang, G., and J. Han. 2005. “A contrastive study on disagreement strategies for politeness between American English & Mandarin Chinese.” Asian EFL Journal 7 (1): 1-12. LoCastro, V. 1986. “Yes, I agree with you. But…”: Agreement and disagreement in Japanese and American English. http://eric.ed.gov/?id=ED284425. Locher, M. A. 2004. Power and politeness in action: disagreements in oral communication. Berlin: Mouton de Gruyter. Marra, M. 2012. “Disagreeing without being disagreeable: Negotiating workplace communities as an outsider.” Journal of Pragmatics 44: 1580-1590. Muntigl, P., and W. Turnbull. 1998. “Conversational structure and facework in arguing.” Journal of Pragmatics 29 (3): 225-256. Nakajima, Y. 1997. “Politeness Strategies in the Workplace: Which Experiences Help Japanese Businessmen Acquire American English Native-like Strategies?” Working Papers in Educational Linguistics 13 (1): 49-69. Pearson, E. 1986. “Agreement / disagreement: An example of results of discourse analysis applied to the oral English classroom.” ITL Review of Applied Linguistics 74: 47-61. Pomerantz, A. 1984. “Agreeing and disagreeing with assessments: some features of preferred/dispreferred turn shapes.” In Structures of social action: Studies in Conversation analysis, edited by J. Atkinson, Maxwell, and J. Heritage, 57-103. Cambridge: Cambridge University Press. Rees-Miller, J. 2000. “Power, severity, and context in disagreement.” Journal of Pragmatics 32 (8): 1087-1111. Santamaría-García, C. 2003. “La negociación de acuerdo en la conversación coloquial: estudio contrastivo español-inglés.” PhD dissertation. Madrid: Universidad Complutense de Madrid. Schauer, G. A., and S. Adolphs. 2006. “Expressions of gratitude in corpus and DCT data: Vocabulary, formulaic sequences, and pedagogy.” System 34 (1): 119-134. Sifianou, M. 1992. Politeness phenomena in England and Greece: A cross-cultural perspective. Oxford: Clarendon Press.

DCTs versus Naturally Occurring Data

205

Verhelst, N., P. Van Avermaet, S. Takala, N. Figueras, and B. North. 2000. Common European Framework of Reference for Languages: learning, teaching, assessment. Cambridge: Cambridge University Press. Yuan, Y. 2001. “An inquiry into empirical pragmatics data-gathering methods: Written DCTs, oral DCTs, field notes, and natural conversations.” Journal of pragmatics 33 (2): 271-292.

CHAPTER NINE THE USE OF CORPORA TO IDENTIFY THE PRAGMATIC KNOWLEDGE ASSOCIATED WITH DIFFERENT LEVELS OF LANGUAGE PROFICIENCY MARÍA LUISA CARRIÓ-PASTOR AND MIGUEL CASAS GÓMEZ

1. Introduction In this chapter, we argue that identifying the types of pragmatic knowledge required at different levels of language proficiency is desirable. Our main objective is to identify which items related to pragmatic knowledge appear at different stages of language learning. In a pilot study, we compiled a corpus of 100 English essays written in English by Spanish learners of English, 50 being at a B1 level of proficiency, and the other 50 at a B2 level of proficiency. Our aim was to test whether the use of corpora of spontaneously produced speech/writing could help to identify what pragmatic knowledge is associated with different stages of second language learning, focusing specifically on language learners’ use of hedges. The essays were processed manually and five raters were involved in the tagging process to identify the use of the hedges associated with B1 and B2 levels of language proficiency. The findings suggest that the use of hedges did in fact differ according to the students’ level of proficiency. We concluded that the students’ communicative effectiveness was in part associated with their use of hedges. Consequently, we believe instruction should focus on activities which raise metadiscoursive awareness in order to support the acquisition of these discourse devices. We suggest that a list of the hedges associated with the different levels of second language learning be created, using

208

Chapter Nine

authentic corpora for this purpose, in order to assist teachers and learners in the improvement of pragmatic competence in written discourse.

2. Theoretical background As language teachers know all too well, knowledge of a language requires competence in several domains such as pragmatic competence. In addition, a language can be learnt both gradually and inadvertently, when speakers of one language come into direct and regular contact with those of another, or acquired purposely when it is learnt as a foreign or a second language. Traditionally, the interest of SLA researchers has focused on grammatical or lexical issues (James 1998; Ifantidou 2011), taking the view that these are the key issues in the mastering of a language. Also, teaching an L2 from the point of view of communication has become mainstream in the discipline of second language acquisition (SLA), as can be seen in the pioneering studies by Krashen (1982), Oxford (1989), Cook (1991), Ellis (1992), Rose and Kasper (2001), and Kasper and Rose (2002). Furthermore, some researchers have also focused on associating the different stages of language proficiency with different learning needs, motivation and levels of pragmatic awareness and competence (EslamiRasekh 2005; Takahashi 2005; Dastjerdi and Shirzad 2010; Carrió-Pastor and Mestre-Mestre 2013a). As Eslami-Rasekh (2005, 200) states: Pragmatic competence consists of illocutionary competence, that is, knowledge of speech acts and speech functions, and sociolinguistic competence. […]. An important question is whether learners need to be taught pragmatics. It can be argued that perhaps pragmatic knowledge simply develops alongside lexical and grammatical knowledge, without requiring any pedagogical intervention.

As this author explains, the difference between grammar, lexis and pragmatics may not have been resolved yet, as language learners may perceive them as implying the same competences. It may seem that language acquisition involves pragmatic aspects, but language interaction, reader engagement and the use of strategies to communicate may not be taught if pragmatic skills are not learnt. Furthermore, corpus linguistics can contribute to analysing how students use the resources of grammar, lexis and pragmatics and to find different ways to improve language acquisition. Corpus analysis provides second language teachers with information about the needs of learners and the most common strategies used in a language to communicate through

The Use of Corpora to Identify the Pragmatic Knowledge

209

learner corpora. The compilation of texts produced by second language students has been and continues to be a very fruitful way of identifying learners’ needs for language teachers, textbook designers and language researchers. Corpora have been used by researchers to identify stages in language learning, and learners’ needs and differences in the acquisition of languages have been studied by researchers such as Granger and Meunier (1994); Granger, Paquot, and Rayson (2006); Chen (2006); Granger and Vander (2007); Granger (2009); Granger and Paquot (2009, 2011) and Granger and Gilquin (2011), amongst others. Most of the above corpus studies have been concerned with determining the proficiency of students in different genres or at different stages of language acquisition from the perspective of grammatical or lexical needs (Granger and Meunier 1994; Granger and Paquot 2009, 2011), but little attention has been paid to the detection and classification of the errors produced by students’ pragmatic failure (Carrió-Pastor and Mestre-Mestre 2013b). A reason for this may be that pragmatic failure is not easy to detect. It could be argued that second language learners acquire pragmatic proficiency in their mother tongue and that, therefore, this is not of interest to second language teaching (Björkman 2011). Some researchers would say that pragmatic issues are implicit in language acquisition and so speakers of any language acquire pragmatic skills intuitively (Kasper and Rose 2002; Dahl 2004). In this study, we would like to argue that the above claim is not true. We think that most of the pragmatic knowledge associated with communication is acquired by speakers from their mother tongues, but also that there are several aspects to which due attention should be paid in second language instruction, as argued by Hyland and Milton (1997) and Alcón Soler (2005). As Alcón Soler (2005, 429) states: “[…] the acquisition of pragmatic aspects will require noticing, processing pragmatic input through practice, and storing that input in long-term memory.” In foreign language learning, the lack of naturally occurring input and feedback may result in speakers who are able to produce grammatically correct interaction, although they may be unable to communicate, as they do not have the adequate pragmatic competence. Pragmatic skills are as necessary as grammatical and lexical knowledge. In this vein, several researchers (Eslami-Rasekh 2005; Alcón Soler 2005; Carrió-Pastor and Mestre 2013a) have already shown that pragmatic competence requires knowledge in such domains as cohesion, coherence, the different types of text, etc. Furthermore, the incorrect use of the pragmatic conventions of language, and specifically metadiscourse, may lead to misinterpretations

210

Chapter Nine

when speakers with different mother tongues interact. Hyland (2004, 134) defined metadiscourse as “[…] the linguistic devices writers employ to shape their arguments to the needs and expectations of their target readers” or as “[…] the self-reflective expressions used to negotiate interactional meaning in a text, assisting the writer (or speaker) to express a viewpoint and engage with readers as members of a particular community” (Hyland 2005, 37). Dahl (2004, 1811) also succinctly described metadiscourse as “[…] overtly expressing the writer’s acknowledgement of the reader.” The effects of metadiscourse failure have been explained by several researchers such as Intaraprawat and Steffensen (1995), Hyland (1998, 2004), Dahl (2004), Gillaerts and Van de Velde (2010), Abdi, Rizi, and Tavakoli (2010), and Björkman (2011). Metadiscoursal features change the meaning of the message depending on the context in which they are used and the intention of the speaker, becoming a part of pragmatics that it is not easy to categorise or identify. This is the reason why in recent years there has been a growing interest in the individual use that writers make of metadiscourse devices to communicate in specific contexts. A common aspect in the considerations of all these authors may be that the acquisition of metadiscourse devices enables the writer to connect with the reader and facilitates the understanding of ideas. One example of the use of metadiscoursive devices in a language is hedging. The study of hedging used by learners (Hyland and Tse 2004) or native speakers of different languages (Dahl 2004; Hu and Cao 2011; MurDueñas 2011) has been of interest to researchers for the purpose of contrasting genres in different languages. Hedges are part of the interactional resources of metadiscourse. They are mainly used to involve the reader in the speaker/writer’ argumentation and, as Hyland and Tse (2004, 169) explain in their model of metadiscourse in academic texts, also to “[…] withhold [the] writer’s full commitment to [a] proposition.” The identification of what counts as a hedge varies from author to author. On the one hand, Lakoff (1972) pointed out a subcategory of hedges as performative hedges, “[…] which modify the illocutionary force of the speech act they accompany” (Crompton 1997, 272). This type of hedge expresses the factual interpretation of the writer that uses it to express his/her opinion. Later, Hyland (1998, 442) and Hyland and Tse (2004, 160) gave the following as examples of hedges: might, perhaps, it is possible and about, although about has been considered an example of vague language (Crompton 1997). On the other hand, Hu and Cao (2011, 2796) proposed the following list of English hedges: modal auxiliaries (e.g., might, could, would), epistemic lexical verbs (e.g., seem, assume, suggest), epistemic adjectives and adverbs (e.g., perhaps, likely, mainly)

The Use of Corpora to Identify the Pragmatic Knowledge

211

and miscellaneous expressions (e.g., in general, assumption (that)). Finally, Mur-Dueñas (2011, 3076) put forward the following list of words used as hedges in business English: 㻌 May, would, can, might, could, appear to, seem, suggest, indicate, expect, predict, propose, consider, tend to, believe, see, view, interpret, imply, argue, hypothesize, think, posit, point to, feel, anticipate, regard, theorize, assume, speculate, want, relatively, typically, likely, potentially, often, perhaps, not necessarily, about, somewhat, usually, commonly, quite, nearly, partially, just, almost, theoretically, probably, marginally, roughly, approximately, basically, maybe, normally, ostensibly, partly, presumably, virtually, tentatively, sometimes, slightly, cautiously, likely, potential, possible, hypothesized, common, unlikely, indicative, typical, apparent, feasible, presumed, probable, proposed, unclear, prone to, hypothesis, likelihood, argument, possibility, view, idea, probability, attempt, assumption, prediction, notion, conceptualization, perspective, tendency, implication, proposition, belief, expectation, feasibility, in general, in part, at least, to our knowledge, a priori, in theory, in broad terms, in our judgement, to some extent.

As can be observed, researchers have listed different items which can be used to indicate the level of commitment to a proposition, but little attention has been paid so far to the use of hedges by language learners – whether, how and to what extent they are learnt, and how they fit into the larger language learning process. This aspect has previously been pointed out by Intaraprawat and Steffensen (1995); Ädel (2006) and Carrió-Pastor and Mestre-Mestre (2013a), but these authors have studied metadiscourse used by learners from a general point of view. Some aspects such as the relationship between hedges and the different English proficiency levels have not been analysed in depth in previous studies. Hedges have traditionally been considered an aspect related to academic English, but this is a part of language that all students should master, not only students of academic discourse. We believe this is an important aspect that should be taken into account and that the teaching of hedges should be included in a second or foreign language teaching syllabus.

3. This study 3.1. Goals and research questions In this paper, we argue that the hedges used in different genres should be identified through the use of learner corpora, and that the different stages of language learning should be taken into account to be able to tag

212

Chapter Nine

hedges and associate them to the different stages of language acquisition. We believe that second language learners should be made aware of the usefulness of hedges by means of the collection of examples that illustrate their meaning and connotations. Consequently, in this study our intention is to test whether the use of learner corpora can be helpful in identifying the hedges associated with different stages of second language learning. The objectives of this study are: 1) To identify the hedges produced in written texts by Spanish learners of English as a foreign language at two levels of language proficiency, B1 and B2 (according to the competences described in the Common European Framework of Reference for Languages, Council of Europe 2001). 2) To classify the hedges that may be associated with the B1 and B2 levels of language proficiency. 3) To find out whether the use of hedges differs depending on language proficiency and see whether the use of learner corpora helps in identifying pragmatic knowledge, specifically with regard to the acquisition of hedges.

3.2. Method 3.2.1. Subjects A pilot study was carried out at Universitat Politècnica de València, involving one hundred learners of English. They were selected from one hundred and twenty Spanish students who were enrolled on different degree programmes at the university, had taken a test to assess their language proficiency and had been classified as being at B1 (preintermediate) or B2 (intermediate) level. The tests for both levels included sections assessing different areas of their English language proficiency: reading, writing, listening, speaking and use of English. The students selected were enrolled in the 2012-13 academic year in one of the following English courses: Technical English (B1 level) and English B2 level, depending on the test results. We selected fifty students from those who had passed the B1 level test, disregarding those that had obtained marks above or below this level. We also selected fifty students from those who had passed the B2 level test, similarly disregarding those students who had demonstrated a higher or lower level of language proficiency. The written assignments given to both groups of learners consisted of responses to emails received from friends. In the B1 level test, learners had to write an assignment of around 300 words on whether they thought that teaching how to use social media was a good idea from the point of view

The Use of Corpora to Identify the Pragmatic Knowledge

213

of legal aspects. In the B2 level test, learners had also to write an assignment of about 300 words, giving advice to a friend who was going to organise a concert at the college and suggesting things the organiser should do. In this way, we were able to compile a corpus of fifty essays written in English by Spanish students of English with a B1 level of proficiency in English and fifty by Spanish students of English with a B2 level of proficiency. The students’ written texts were entered into our database. They comprised a total of 28,302 words (13,459 in the B1 essays, and 18,853 in the B2 essays). 3.2.2. Data analysis We produced a list of hedges from all those proposed by other researchers (Hyland 1998; Hyland and Tse 2004; Hu and Cao 2011; MurDueñas 2011). The initial list of hedges searched for in the corpus was considered to be the most complete list, but we realised that the list needed to be checked, as hedges can only be identified in context. Most of the hedges identified by the above authors as being used in academic English were not found in the corpus we had compiled, given the specific nature of our corpus, so a manual analysis and tagging of hedges was carried out. The one hundred essays were processed manually and four raters undertook the tagging process to identify the hedges associated with B1 and B2 levels of language proficiency. Given that hedges are a type of interactional metadiscoursive strategy and are frequently used in academic writing, in this study hedges were taken to be those devices used to qualify the writer’s commitment to a proposition, to show (un)certainty about the truth of an assertion and to establish alternative viewpoints (Hu and Cao 2011). There were several meetings with the raters involved in the identification of hedges to discuss the identification of some hedges. The hedges thus identified in the corpus of this study and used by learners with B1 and B2 levels, which coincided with those identified by previous studies, were: may, would, can, might, could, seem, consider, think, often, perhaps, usually, quite, maybe, possible, probable, in general, believe, want. After this selection of the hedges used by the learners, we also decided that a further set of items should be included, on the grounds that these phrases also show the writer’s degree of commitment to a proposition, their stance on a subject or their claims about a topic: recommend, persuade, honest opinion, absolutely sure, my suggestion, so sorry, my opinion, doubt, sorry, refuse. They were included as they are all features that indicate, as Intaraprawat and Steffensen (1995, 258) explain, “[…] the writers’ sincerity and intention to produce an ethical text by accurately

214

Chapter Nine

indicating what they believe, what they know, and what they assume;” (that is the case of recommend, persuade, honest opinion, my suggestion, my opinion) or to show uncertainty or doubts about the truth of an assertion and to establish alternative viewpoints (that is the case of absolutely sure, so sorry, doubt, sorry, refuse). Some examples of the context of the use of these words as hedges are shown in the Results section of this paper.

3.3. Results and discussion The occurrences of hedges found in the corpus were extracted following a corpus-driven methodology. Their count was compared to the total number of words in the corpus and the frequency of occurrence of each hedge was compared to the total number of hedges. Given that hedges should be detected in context, as their function is to reduce commitment and negotiate meanings between the reader and the writer, the results obtained in this analysis are presented in different stages. In the first place, some examples found in the corpus of hedges listed in the literature (Hyland 1998; Hyland and Tse 2004; Hu and Cao 2011; MurDueñas 2011) are shown and the implication of hedges commented on, following a qualitative methodology. Then, the quantitative results of the study are shown in Table 1. As the results obtained in the texts written by students with B1 and B2 levels of language proficiency are quite different, the examples included in this section are a sample of the most frequent hedges that show the procedure followed by raters to detect and tag them. Subsequently, some examples of the hedges tagged in the corpus that have not been mentioned in previous studies are shown and commented on, and they are listed in Table 2 to show the different frequencies of hedges. Examples 1, 2, 3, 4, 5 and 6 are examples of the hedges listed in the previous literature and found in the corpus: (1) “You told me you want to celebrate the party in the dinning-room. If I have to be honest, I would do it outside at night. And I would rent some light beams.” (B2-08).

In (1), the writer of the text uses the hedge would to express a tentative opinion or persuade the reader, modifying the initial proposal and establishing an alternative viewpoint. (2) “But I prefer to celebrate it in the college garden. We are in summer and I think that people love seeing a concert under the brightly sun.” (B2-15).

The Use of Corpora to Identify the Pragmatic Knowledge

215

In (2), the writer is making a tentative suggestion: that the concert should be held in the college garden, and the use of the hedge think makes the statement not too authoritarian or bossy. (3) “In reference with the best place to hold it, I think it could be the college garden because it’s bigger than the college hall, so more people will be able to attend.” (B2-29). “I read your letter and I think I can help you with your friend’s issue. I think she is embarrassed.” (B1-07).

In (3), we see two learners of different levels using think, with the higher-level student also employing could. These hedges qualify the writers’ commitment to propositions, conveying tentativeness. (4) “I am sure she will apologise and maybe she will repair your car. If she doesn’t apologise, she is not a good friend.” (B1-28). “There are several groups composed by young people that might not be very expensive. But maybe we can make a survey in order to know the opinion of the rest of the people.” (B2-12).

In (4), in the first example, the writer uses am sure to sound confident to the reader, but the writer also express uncertainty about certain ideas in this same sentence by using the hedge maybe. The modal verb maybe is also used in the second example to express uncertainty, which is reiterated in the second example with the use of might and maybe. The reader considers it a good idea to organize a concert and the writer expresses uncertainty about the issues that may be taken for granted by the reader. (5) “In my opinion, I quite prefer a pop-rock concert. I believe it’s the best option for this kind of concert. Try to find a young group near here!” (B2-29).

In (5), the writer uses several hedges (my opinion; quite; believe) to qualify his/her commitment to the proposition of the holding of a pop-rock concert and to help the reader make decisions about his/her stance on the concert. He/She also finishes the paragraph with an exclamation mark in order to sound very sure about his/her proposal. (6) “It might rain and the equipment can’t be wet. Moreover, the sound will be better than outside.” (B2-43).

In (6), the writer does not want to be negative and he/she uses the hedge might to be tentative in the statement concerning the possibility of

Chapter Nine

216

rain. The writer is trying to convince the reader that holding the event outside is not a good idea by suggesting the possibility of rain and introducing a negative element into the discourse, and it is this negative possibility that leads to the following statement. Table 1 shows the frequency of occurrences in the corpus of the hedges listed in the literature referred to above. The first column shows the hedges found in the corpus, the second and third columns the relative frequencies of hedges produced by the B1 learners rendered as the percentage values and as the number of occurrences represented in relation to the relative frequency per 1,000 words. The fourth and fifth column show the relative frequencies of hedges produced by the B2 learners detailed in percentage values and in number of occurrences represented in relation to the number of words in this sub-corpus. TABLE 1. Occurrences of hedges found in the corpus and listed in Hyland (1998), Hyland and Tse (2004), Hu and Cao (2011), and Mur Dueñas (2011) Hedges

may would can might could seem consider think often perhaps usually quite maybe possible probable in general believe want average values

B1 learners Percentage 10.6 9.7 7.8 2.9 6.7 1.9 4.8 9.7 7.8 5.7 3.8 3.8 4.8 5.7 5.7 3.8 1.9 2.9 5.5

B1 learners Per 1,000 words 0.8 0.7 0.6 0.2 0.5 0.1 0.4 0.7 0.6 0.4 0.3 0.3 0.4 0.4 0.4 0.3 0.1 0.2 0.4

B2 learners Percentage 8.6 14.1 19.5 5.3 10.1 3.8 4.4 5.1 3.8 2.8 2.2 2.8 5.3 2.2 2.5 2.2 2.5 2.8 5.5

B2 learners Per 1,000 words 1.8 3.1 4.8 1.2 2.6 0.8 0.9 1.1 0.8 0.6 0.4 0.6 1.2 0.4 0.5 0.4 0.5 0.6 1.2

The Use of Corpora to Identify the Pragmatic Knowledge

217

Examples 7, 8, 9, 10, 11 and 12 are examples of the hedges tagged in the corpus and not mentioned in previous studies: (7) “You finally have managed to persuade the principal to do the concert, you are the best! I think this is a good opportunity to show people our songs and how well we play.” (B2-15).

In (7), the writer uses the hedge persuade to emphasise the fact that he/she appreciates what has been explained, and this is reinforced by the use of an exclamation mark. The verb persuade is a hedge revealing the writer’s opinion about what is appropriate, which is meant not only to inform, but also to affect the addressee. (8) “You said that you are trying to decide the place. In my opinion, it could be better outside in the college garden. We have wonderful views in the background as well as we have a lot of space in the garden.” (B2-07).

In (8), the writer wishes to reduce the certainty of the proposition conveyed in the first sentence with the use of in my opinion and could. These hedges qualify, by reducing it, the degree of certainty of the opinion expressed. (9) “I am absolutely sure she will buy a new one. I think it is very important that you are friends.” (B1-19).

The writer uses the hedge absolutely sure to transmit the certainty his/her opinion and then qualifies his/her commitment to a particular idea with I think. Absolutely sure is a performative hedge (Lakoff 1972) modifying the illocutionary force of the speech – it express the writer’s factual interpretation of his/her opinion. (10) “This is the most difficult question. The best choice will be to have audience, otherwise, it will be a waste of money for the college and you will not be able to organise another one in the future. So I recommend you a mix of rock and roll and pop music.” (B2- 01).

The writer gives his/her point of view with the hedge recommend, revealing his/her strong decision-making stance about the music that should be played in the party. (11) “I am sorry but I am not interested in introducing the musicians when they come on stage because I get very nervous when I have to talk to a lot of people.” (B2- 18).

Chapter Nine

218

The writer explains a situation that is uncomfortable for him/her and uses the attitudinal hedge sorry to mitigate his/her refusal to introduce the musicians. (12) “I think if your friend is a good person, he will not refuse to apologise. I think you have to talk to him seriously.” (B1-32).

The writer uses the hedge refuse to reinforce the if-clause and to make the statement less forceful and assertive. This hedge is also toned down by the use of think twice. The occurrences of the second set of hedges, those not included in the list shown in Table 1, can be seen in Table 2. The first column shows the hedges found in the corpus, the second and third columns the relative frequencies of hedges produced by the B1 learners rendered as percentage values and as the number of occurrences represented in relation to the relative frequency per 1,000 words. The fourth and fifth column show also the relative frequencies of hedges produced by B2 learners detailed as percentage values and as the number of occurrences represented in relation to the relative frequency per 1,000 words. TABLE 2. Hedges found in the corpus and not listed in previous studies Hedges

recommend persuade honest opinion absolutely sure my suggestion so sorry my opinion doubt sorry refuse average values

B1 learners Percentage 5.0 10.0 0.0 25.0 10.0 10.0 15.0 10.0 5.0 10.0 10.0

B1 learners Per 1,000 words 0.07 0.14 0.00 0.37 0.14 0.14 0.22 0.14 0.07 0.14 0.09

B2 learners Percentage 7.3 15.4 4.1 7.3 9.8 11.3 13.1 14.7 12.2 4.8 10.0

B2 learners Per 1,000 words 0.6 1.3 0.3 0.6 0.8 0.9 1.1 1.2 1.0 0.4 0.8

It should be observed that Table 1 shows a higher frequency of hedges than Table 2. The difference between the B1 and the B2 learners may mean that these hedges are acquired during the transition from a B1 to a B2 level of language. In this sense, the results extracted from this study may support the claim that learners learn how to use hedges at different stages of language acquisition.

T The Use of Corrpora to Identify y the Pragmaticc Knowledge

219

The heddges identifieed in this stu udy can be classified in different grammaticall classes to be b incorporatted in the Ennglish curricu ula as: a) lexical verbbs with evaluaative implicattions (recomm mend, persuad de, doubt and refuse)) and b) adjectival, adverbial and noominal modall phrases (honest opinnion, absoluteely sure, my su uggestion, so sorry, my opinion and sorry). Thesse hedges maay be incorporrated into currricula of Eng glish as a foreign langguage. Finally, Figure 1 shoows a compaarison of the hedges foun nd in the corpus analyysed:

FIGURE 1. Comparison of thee occurrences found f in the heddges of the corp pus

As can bbe observed, most m of the heedges found inn the corpus were w used more frequeently by the B22 learners and d this finding m may suggest that t when learners movve from a B11 to a B2 leveel of English pproficiency, pragmatic p knowledge oof the target laanguage is acq quired.

220

Chapter Nine

4. Discussion It was observed in Tables 1 and 2 that the learners at B2 level of proficiency used a higher number of hedges than the English learners at B1 level. This finding might mean that, although the lower-proficiency language learners under study may have been aware of the need to express their stance on a particular subject, or to strengthen their claims, their language proficiency prevented them from using hedges in the same way as the higher-proficiency students. For example, the hedge honest opinion was not used by the B1-level learners, as can be seen in Table 2 and Figure 1. This nominal modal phrase was used by the B2-level learners to reduce the risk of opposition from the reader. The results showed that Spanish learners of English at the B1 level of language proficiency more frequently used the hedges may, would, think, absolutely sure and my opinion. With these hedges, the writers explained their commitment to a proposition, showed uncertainty about one assertion and established alternative viewpoints with hedges. It was shown in Table 1 and 2 that B1-level learners did not use a wide variety of hedges, but they still expressed their commitment to the propositions expressed in the texts. On the other hand, the Spanish learners of English at the B2 level of proficiency preferred to use the hedges may, would, can, could, persuade and doubt. These hedges expressed the speakers’ commitment to the prepositions, as shown in examples (1) and (7). As can be observed in Table 1, both groups of learners more frequently used the hedges may and would to show uncertainty about an assertion, but both groups preferred the use of other hedges (i.e., consider, think, perhaps, probable) to convey imprecision. Hyland (2004, 148) points out that: “[…] the way that writers present themselves, negotiate an argument, and engage with their readers is closely linked to the norms and expectations of particular cultural and professional communities.” To this idea, it has been shown in this research that the way writers present themselves may be also closely linked to the language proficiency of writers, as writers can engage readers only if they are proficient in the language they use. It can be observed in Figure 1 that the more proficient the learners were in English, the more hedges they used. This might entail that the pragmatic knowledge of learners may be associated to the acquisition of language skills.

The Use of Corpora to Identify the Pragmatic Knowledge

221

5. Final remarks This study was designed to identify and classify hedges produced at two levels of English language proficiency, to find out if the use of hedges differs depending on the language proficiency level of the speakers, and if the use of corpora can help in identifying pragmatic knowledge, specifically with regard to the acquisition of hedges. In particular, the hedges identified in this study and used by English learners at B1 and B2 levels of language proficiency, qualified the learners’ commitment to a proposition, showed uncertainty about the truth of an assertion and established alternative viewpoints (see Hu and Cao 2011). In this sense, the learners of English included in this research were conscious of the need to withhold the opinion of the writer and to take into consideration the presence of the reader. The learners were able to use different hedges to be polite and mitigate the effect of their assertion, but it was interesting to discover that learners with lower level of language proficiency did not use hedges as frequently as learners at a higher language proficiency level. In this sense, it might be stated that the higher the level of language proficiency, the higher the use of hedges. The lack of language proficiency may be one of the causes that might prevent the imitation of pragmatic strategies when writing in a second language. Furthermore, we believe that the use of authentic corpora collected by students should be used to raise their awareness of the pragmatic competence acquired at the different levels of language proficiency. Samples extracted from authentic texts may show, for example, the hedges associated to the different professional needs, as for instance, the hedges used in a job interview. In this way, learners may be able to observe the authentic use of language and this may be used as accessible models of hedges for second language learners. We believe the compilation of corpora illustrating the actual context-dependent use of pragmatic devices, at the different levels of language proficiency may be helpful to students when acquiring the knowledge associated to each profession. Indeed, this study has provided evidence that the use of a corpus of real texts is very useful in detecting the use of hedges. Context is important in the identification of hedges and other self-reflective or addressee-oriented linguistic expressions, which express the writer’s degree of commitment to, and attitude towards, a given proposition (Crompton 1997, 2012; Hyland 1994, 2005). Hyland (1994, 244) indicates that: “Control over this feature [hedging] of academic discourse is therefore an important communicative resource for L2 writers at any proficiency level.”

222

Chapter Nine

Finally, in this study we have shown that the use of a corpus of real texts to identify the hedges used by learners made it possible to determine which ones learners use at different stages of the learning process. We concluded that second language learners might require pragmatic training in order to achieve effective communication in different contexts. Consequently, we believe instruction should include activities which raise pragmatic awareness, specifically with regard to the use of hedges. As Hyland (2004, 148) states: “[…] metadiscourse offers teachers a useful way of assisting students towards control over disciplinary-sensitive writing practises. Because it shows how writers engage with their topic and their readers, exploration by students of metadiscourse […] can offer useful assistance for learning about appropriate ways to convey attitude, mark structure, and engage with readers.” We are conscious of the limitations of this study. This study only included a set of students and focused on two levels of language proficiency. Future studies should focus on the identification of metadiscoursive features associated with different levels of proficiency in language learning, in order to gain a better understanding of the process of the acquisition of pragmatic knowledge in written communication in different contexts.

References Abdi, R., M. T. Rizi, and M. Tavakoli. 2010. “The cooperative principle in discourse communities and genres: A framework for the use of metadiscourse.” Journal of Pragmatics 42: 1669-1679. Ädel, A. 2006. Metadiscourse in L1 and L2 English. Amsterdam: John Benjamins. Alcón Soler, E. 2005. “Does instruction work for learning pragmatics in the EFL context?” System, 33: 417-435. Björkman, B. 2011. “Pragmatic strategies in English as an academic lingua franca: Ways of achieving communicative effectiveness?” Journal of Pragmatics 43 (4): 950-964. Carrió-Pastor, M. L., and E. Mestre-Mestre. 2013a. “Increasing pragmatic awareness in the L2 classroom.” In Studies and global perspectives of second language teaching and learning, edited by J. Schwieter. New York: Information Age Publishing. —. 2013b. “A proposal for the tagging of grammatical and pragmatic errors.” Research in Corpus Linguistics 1: 7-18.

The Use of Corpora to Identify the Pragmatic Knowledge

223

Chen, C. W. 2006. “The use of conjunctive adverbials in the academic papers of advanced Taiwanese EFL learners.” International Journal of Corpus Linguistics 11 (1): 113-130. Cook, V. 1991. Second language learning and language teaching. London: Arnold. Council of Europe. 2001. Common European Framework of Reference for Languages: learning, teaching, assessment. Cambridge: Cambridge University Press. Crompton, P. 1997. “Hedging in academic writing: some theoretical problems.” English for Specific Purposes 16 (4): 271-287. —. 2012. “Characterising hedging in undergraduate essays by Middle Eastern students.” The Asian ESP Journal 8 (2): 55-78. Dahl, T. 2004. “Textual metadiscourse in research articles: a marker of national culture or of academic discipline?” Journal of Pragmatics 36: 1807-1825. Dastjerdi, H. V., and M. Shirzad. 2010. “The impact of explicit instruction of metadiscourse markers on EFL learners’ writing performance.” The Journal of Teaching Language Skills 2 (2): 155-174. Ellis, R. 1992. Second language acquisition and language pedagogy. Clevendon: Multilingual Matters. Eslami-Rasekh, Z. 2005. “Raising the pragmatic awareness of language learners.” ELT Journal 59 (3): 199-208. Gillaerts, P., and F. van de Velde. 2010. “Interactional metadiscourse in research article abstracts.” Journal of English for Academic Purposes 9: 128-139. Granger, S. 2009. “The contribution of learner corpora to second language acquisition and foreign language teaching: a critical evaluation.” In Corpora and language teaching, edited by K. Aijmer, 13-32. Amsterdam: John Benjamins. Granger, S. and G. Gilquin. 2011. “From EFL to ESL: evidence from the International Corpus of Learner English.” In Exploring secondlanguage varieties of English and learner Englishes: bridging a paradigm gap, edited by J. Mukherjee, and M. Hundt, 55-78. Amsterdam: John Benjamins. Granger, S., and F. Meunier. 1994. “Towards a grammar checker for learners of English.” In Creating and using English language corpora, edited by U. Fries, and G. Tottie, 79-91. Amsterdam: Rodopi. Granger, S., and M. Paquot. 2009. “Lexical verbs in academic discourse: a corpus-driven study of learner use.” In Academic writing: at the interface of corpus and discourse, edited by M. Charles, D. Pecorari, and S. Hunston, 193-214. London: Continuum.

224

Chapter Nine

—. 2011. “Language for specific purposes learner corpora.” In Language for specific purposes. The Encyclopaedia of applied linguistics, edited by T. Upton, and C. Chapelle. Oxford: Blackwell-Wiley. Granger, S., M. Paquot, and P. Rayson. 2006. “Extraction of multi-word units from EFL and native English corpora. The phraseology of the verb “make.”” In Phraseology in Motion I: Methoden und Kritik. Akten der Internationeln Tagung zur Phraseologie, edited by A. Häcki Buhofer, and H. Burger, 57-68. Basel: Baltmannsweiler Schneider Verlag Hohengehren. Granger, S., and V. Vander. 2007. “Corpus linguistics, language learning and ELT.” Mindbite 1: 11-14. Hu, G., and F. Cao. 2011. “Hedging and boosting in abstracts of applied linguistics articles: A comparative study of English and Chinese medium journals.” Journal of Pragmatics 43: 2795-2809. Hyland, K. 1994. “Hedging in academic writing and EAP textbooks.” English for Specific Purposes 13 (3): 239-256. —. 1998. “Persuasion and context: The pragmatics of academic metadiscourse.” Journal of Pragmatics 30: 437-455. —. 2004. “Disciplinary interactions: metadiscourse in L2 postgraduate writing.” Journal of Second Language Writing 13: 133-151. —. 2005. Metadiscourse. London: Continuum. Hyland, K., and J. Milton. 1997. “Qualification and certainty in L1 and L2 students’ writing.” Journal of Second Language Writing 6 (2): 183205. Hyland K., and P. Tse. 2004. “Metadiscourse in Academic Writing: A Reappraisal.” Applied Linguistics 25 (2): 156-177. Ifantidou, E. 2011. “Genres and pragmatic competence.” Journal of Pragmatics 43: 327-346. Intaraprawat, P., and M. S. Steffensen. 1995. “The use of metadiscourse in good and poor ESL essays.” Journal of Second Language Writing 4 (3): 253-272. James, C. 1998. Errors in Language Learning and Use. London: Longman. Kasper, G., and K. R. Rose. 2002. Pragmatic development in a second language. Oxford: Blackwell. Krashen, S. 1982. Principles and practice in second language acquisition. Oxford: Pergamon. Mur-Dueñas, P. 2011. “An intercultural analysis of metadiscourse features in research articles written in English and in Spanish.” Journal of Pragmatics 43: 3068-3079.

The Use of Corpora to Identify the Pragmatic Knowledge

225

Oxford, R. 1989. Language learning strategies: What every teacher should know. New York: Newbury House. Rose, K. R., and G. Kasper. 2001. Pragmatics in language teaching. Cambridge: Cambridge University Press. Takahashi, S. 2005. “Pragmalinguistic awareness: Is it related to motivation and proficiency?” Applied Linguistics 26 (1): 90-120.

PART II PRINCIPLES, METHODS AND PRACTICES IN SPECIFIC AREAS OF PRAGMATIC COMPETENCE

SECTION 3 CONVERSATION

CHAPTER TEN AN INTERACTION-FOCUSED PEDAGOGY BASED ON CONVERSATION ANALYSIS FOR DEVELOPING L2 PRAGMATIC COMPETENCE ANNA FILIPI AND ANNE-MARIE BARRAJA-ROHAN

1. Introduction Over the last twenty years, studies in the application of findings of Conversation Analysis in a number of fields and professions have been noticeably increasing. This shift from description to application has not been without its opponents. Indeed, questions have arisen about the ethical considerations of such applications in the medical field (see for example Antaki et al. 2013). However, professions in education can benefit greatly from such applications because improvements in practice have the potential to enhance students’ success in school as well as their social interaction skills, which may lead to more successful participation in society. Such is the case for immigrant learners and international students. In language testing the study of the interactions that occur in standardised speaking tests and their applications also have an important role to play in addressing issues of validity and responsible test design, fairness, and the impact of tests on both the individual and society at large. It is against the backdrop of studies in the second language (L2) classroom and standardised testing that our present work is to be understood. We draw on our research and work as practitioners in both fields to suggest a possible pedagogy for the development of skills in pragmatics and interaction more generally. We start by providing an overview of the research that has been conducted in describing the classroom and its social organisation, and how this is manifested in the

232

Chapter Ten

second language classroom with respect to ESL and other second languages, and in language testing. Next we provide examples focusing on specific problematic areas in students’ L2 skills in interaction, by drawing on data from an adult ESL classroom and from published data of an L2 Italian final high school examination and an IELTs speaking test, and finally we suggest a pedagogy that provides instruction in and for interaction.

1.1. Pragmatics, interaction and Conversation Analysis Traditionally, pragmatics in interaction has been concerned with the structure of conversation. Studies in the field have examined implicature, speech acts, the information structure and presuppositions in conversation (Grice 1981, Levinson 1983, Searle 1979). More recently, attention has been drawn to the organisation and mechanics of interaction arising from the increasing number of studies in Conversation Analysis, which is grounded in observation and analysis of naturally occurring interactions. This is in opposition to the native speaker’s intuition, which has been demonstrated to be inaccurate (Golato 2003). With its focus on the microdetails of talk as it unfolds moment by moment and its attention to the collaborative accomplishment of the interactants, Conversation Analysis provides a set of methods for exposing the organisation of pragmatic competence. However, it does so in ways that give prominence to what is visible as speakers work together to establish what is referred to as intersubjective (i.e., commonly shared) understanding. From this perspective, pragmatic competence is not seen as residing in what a speaker says, thinks or expects intuitively, but rather in her/his actions in interaction (Sidnell 2013). Since the pioneering early work of Sacks, Schegloff and Jefferson in the 1960s and early 1970s; for example, their seminal work published in 1974 on the organisation of turn-taking in conversation (Sacks, Schegloff, and Jefferson 1974), an impressive number of investigations have uncovered the workings of interaction in a range of languages (see for example, the review in Mori 2007 for language classrooms, and Pallotti and Wagner 2011 for L2 contexts more generally), in a range of contexts for adults (see for example Drew and Heritage 1992, and more recently, Sidnell and Stivers 2013) and children (see for example, Filipi 2009; Gardner and Forrester 2010). In studies of the classroom, the last decade has seen an increasing body of work that has built on the early work of McHoul (1978, 1990) and Mehan (1979) on the organisation of classroom interaction from the

An Interaction-Focused Pedagogy based on Conversation Analysis

233

perspectives of turn-taking and sequence organisation in both teacherfronted classrooms and more recently in student-centred classrooms (Seedhouse 2004), and in the transitions in lesson phases (Markee 2004). (See Gardner 2013 for a review of this work.) Gardner (2013) notes that the majority of this work has been on the second language classroom. While there are still controversies in the field with respect to what Conversation Analysis can contribute to second language learning and its relationship to second language acquisition (Gardner 2013, Mori 2007), there is a growing acknowledgment that it is well placed to provide both a set of findings and an ideal methodological tool to track learning over time (Barraja-Rohan 2011; Brouwer and Wagner 2004; Filipi 2013; Firth and Wagner 2007; Hellermann 2006; Markee 2008; Nguyen 2011, 2012). Within the context of learning, second language learners do not necessarily acquire pragmatic competence ‘in the wild’. However, they have been shown to use their language resources “creatively” to achieve participation in interaction (Gardner 2013). In the foreign language classroom, opportunities to develop this competence are generally insufficient for two reasons: there is no specific focus on pragmatics and interactional skills (Rose 2005), and course books generally do not reflect authentic interactions (Huth and Taleghani-Nikazm 2006; Wong 2002, 2011). With new approaches to teaching languages that centre on authenticity and learning through interaction, the need to do this has become apparent (Huth and Taleghani-Nikazm 2006). The set of methods in Conversation Analysis for uncovering the interactional resources displayed by the interactants together with the amassed research findings in the field are available for application by educators (Barraja-Rohan 1997, 2011; Huth and Taleghani-Nikazm 2006). Findings in second language acquisition from a Conversation Analysis perspective give insights into how L2 learners interact in both the classroom and outside the classroom. These findings provide valuable information to educators so that they can gain a deeper understanding of learners’ behaviours and use of language in order to better target specific areas of pragmatics. For instance, the studies conducted by Golato (2002) and Taleghani-Nikazm (2002) showed how L2 speakers transfer first language (L1) cultural norms of interaction when interacting in L2. Golato’s (2002) study focussed on how transfers caused communications problems in compliments proffered by American English speakers while using German as L2, and Taleghani-Nikazm (2002) explored how Iranian L2 speakers of German, engaged in a how are you? sequence, reproduced their L1 pattern.

234

Chapter Ten

1.2. Conversation Analysis and Second Language testing Conversation Analysis methods have proven to be very useful in investigating the interactions in a second language examination context. Studies over the past 20 years have examined the interactional behaviour of both examiners and test takers. Investigations of the examiners’ utterances have been shown to have a bearing on the performance of the test taker because ultimately the interactions in the tests are the achievement of both the candidate and the assessor (Filipi 1994, 1998; He 1998; Lazaraton 1996; Seedhouse and Egbert 2006). Using the microanalytic approach of Conversation Analysis, Filipi (1998) showed that examiners in a final L2 Italian oral state exam displayed styles that were either more interaction-like or more interrogation-like leading to different displays of ability from the students. With respect to test takers, research has shown that some have poor skills in initiating repair in the second language. They will, for example, laugh, be silent or produce a minimal response token such as ‘yeah’ in response to an examiner question they do not understand (reported for ESL by He 1998, and L2 Italian by Filipi 1994). Some test takers have also been shown to be unable to produce extended turns (Filipi 1994). Despite the investigations and attempts to address gaps through teacher (and student) training, students still show gaps in their interactional competence in exam conditions (Seedhouse and Egbert 2006). The above findings highlight the need for a targeted pedagogy in the second language classroom that focuses on the students’ ability to answer questions in appropriately fitted ways, to initiate repair and self-correct, to use minimal response tokens in ways that are expected in that language, and to extend turns. Test developers of speaking tests also need to be aware of the limitations that particular designs impose. Seedhouse and Egbert (2006) for example maintain that given the design of the IELTS test where the examiner is the interviewer and the test taker the interviewee, there is minimal scope for students to engage in a variety of sequence types and turn-taking behaviours, including asking questions (Seedhouse and Egbert 2006), managing topics (Seedhouse and Harris 2011) and opening and closing a conversation (Filipi 1997; Seedhouse and Egbert 2006). There can sometimes be a tension between what the aims of a second language speaking test should be; for instance, whether indeed it is possible (Shamoosi and Tavakoli 2010) or even desirable to provide information about a student’s capacity to communicate in authentic ways. However, since the introduction of communicative methodologies (Nunan 1991), one of the purposes of teaching a second language has been to

An Interaction-Focused Pedagogy based on Conversation Analysis

235

provide students with the opportunity to develop skills to enable them to interact in authentic ways, whether this be with texts or with people. This is potentially where a problem emerges because there is an inherent assumption that L2 students have a solid understanding of their L1 pragmatic competence. However, we know that such understandings operate at a subconscious level (Wong and Waring 2010), much like knowledge of grammar does (O’Grady et al. 2001). Therefore there is a need to develop a pedagogy that allows students to compare and contrast their L2 with their L1 in order for them to develop pragmatic awareness in both L1 and L2, and pragmatic competence in L2. The rest of this paper will provide examples from examinations and the ESL classroom to illustrate the issues just discussed, and then describe an approach that focuses on the development of skills in interaction.

2. Method Some of the data for this paper (extracts 2, 3 and 6) is drawn from past research. This is indicated in the header of the relevant extracts. Other data comes from unpublished sources in the form of classroom interactions of an ESL adult setting in Australia. The published data are derived from a study of a classroom of low intermediate ESL adult students (BarrajaRohan 2011), a final high-stakes Italian as a second language oral examination in Victoria, Australia (Filipi 1994, 1998), and the high-stakes IELTS speaking test (Seedhouse and Egbert 2006). Two features will be analysed and discussed. The first, which is common to the three interactional contexts of extracts 1, 2 and 3, is the students’ inability to initiate repair or produce an appropriate second pair part response to a question. The second which applies to an ESL adult context consists in the absence of assessments and terminal components, evident, for example, when students move straight into leave-taking without closing down the conversation. In order to bring consistency to the transcripts, a uniform set of notations based on Jefferson (1984), Gardner (2001) and Schegloff (2007) have been applied. (See Appendix).

Chapter Ten

236

3. Analyses 3.1. Common interactional problem 1: Failure to initiate a repair or produce an appropriate second pair part response to a question Repair is an important mechanism in conversation. It enables conversationalists to deal with interactional troubles related to (mis)hearing and misunderstanding as well as to check their own understanding. This can be accomplished through actions such as correction, clarification requests, confirmation checks, understanding checks (I mean X not Z) and questions like pardon?. Repair helps to maintain shared understanding or intersubjectivity (Schegloff 1991). Showing one’s understanding of a question usually involves a response in the next turn; thus the question and answer form a relational pair called an adjacency pair made up of a first pair part (the question) and a second pair part (the answer) (Schegloff 2007). Adjacency pairs create a sequential environment whereby a response constitutes a natural, logical, relevant action in response to the previous turn, which brings closure to the communicative exchange initiated by the previous turn. Its absence or incongruity is thus noticed. This may require the recipient to undertake remedial action. Each of the first three samples below illustrates problems that arise in the adjacency pair sequences leading to a need for repair. Extract 1 is taken from an ESL class during a “free conversation” session during which students could choose their conversational partner and their own topic. As we will see below, none of the speakers use repair strategies. Moreover, Minh, the first speaker, fails to produce any listener response while Zhu, the second speaker, is in engaged in an extended turn, as might be expected. Extract 1 ((A Chinese student at CEFR B2 level (Zhu) and a Vietnamese student (Minh) at B1 level talking about their weekend.)) 1 2 3 4 5 6 7 8 9

MIN: er what did you do er in actually er- whiZHU: weekend? (...) ((Zhu goes onto explain what he did at the week-end.)) ZHU: how about you. MIN: ˆ mm hm ((Uttered with a smiling voice.)) ˆ (1.0) ZHU: yeah, I got a few australian friends but er- (1.0) but it’s hard to: you can talk to them: it’s more: in different of topics but if you just talk to them

An Interaction-Focused Pedagogy based on Conversation Analysis 10 11 12 13 14 15

ˆ ˆ

237

there’s some possilibity, or anything else. (…) you don’t have any- any language an you problem can talk any talk. yunno if it’s not about your idea. ha? an then they can talk your idea. so if you talk to aus-australian people. so it’s hard for- for you, cos to talk is …

The conversation starts with Minh asking the ritual question how’s your weekend in line 1; however, it is Zhu who completes the question in line 2. Zhu answers the question by producing an extended telling to Minh’s question (data not shown) mentioning that he likes visiting friends. Then Zhu returns the question to Minh in 3, but Minh produces an unusual next action to Zhu’s question by giving a minimal response (with a smiling voice) instead of a second pair part or an other-repair initiation; the latter would have been appropriate as a next action if Minh did not understand Zhu’s question. The minimal response that Minh utters does not involve rising intonation thus we can assume that it is not designed to initiate repair. The question that Zhu asks “what about you” was a routine question that students in this class were familiar with, therefore what Minh has missed is the action intended here. This shows a lack of pragmatic understanding on his part. However, Zhu does not initiate a third turn repair (Schegloff 1997) after Minh has produced a sequentially inappropriate response. Instead, Zhu continues his extended telling launching into the topic of practising English with Australians. During this long extended telling (which goes on beyond line 14), Minh does not produce any verbal listener responses. Because it is an audio recording, there is no access to body language, and we cannot ascertain whether Minh is nodding or not. Nonetheless, Minh’s lack of verbal feedback is noticeable, as he does not even produce a comment, particularly when Zhu says in lines 13 and 14 it’s hard for you. This is an assessment showing empathy towards Minh. Minh is not a beginner and has an intermediate language level so it is even more striking that he does not contribute at all during Zhu’s extended turn. In the next extract, taken from an L2 Italian oral examination, the two examiners are engaged with a test taker during the first part of the examination. This is a general conversation about everyday topics. Extract 2 ((E1: examiner 1, E2 examiner 2, C: candidate)) 1 2 3

E1: C: ˆ E1:

cosa coltivano? ((what do they cultivate?)) sì sì. ((yes yes.)) cosa COSA coltivano? ((what WHAT do they cultivate?))

Chapter Ten

238 4

C:

5 6 7

E2: ˆ C:

um (0.4) cosa coltivano. um:: ((um (0.4) what do they cultivate. um)) (0.8) frutteti o:: ((orchards or::)) sì. ((yes.)) (from Filipi 1994, 125)

The candidate’s response sì sì ‘yes yes’ is an inappropriate fit to the examiner’s open question which is asking for very explicit information in answer to his question. This is a strong indication that the candidate has not understood the question or what kind of question it is. Indeed, the examiner orients to her response as a lack of understanding by initiating a self-repair. However, he has misinterpreted the source of the problem. By repeating cosa loudly, the examiner is targeting a problem of hearing, which has not, in fact, arisen. Rather, as the candidate’s turn in line 4 shows, it is a problem of understanding the word coltivano ‘cultivate’. At this point we note the second examiner accommodates through the beginnings of an ‘or’ question, but again the candidate replies with sì ‘yes’. We cannot say whether the candidate does not want to (perhaps as a strategy to mask her lack of knowledge of vocabulary) or does not know how to adequately initiate a direct repair. However, we can conclude that the candidate has not understood either word (coltivano ‘cultivate’ or frutteti ‘orchards’), and produces sì ‘yes’ inappropriately. The third extract is taken from the first part of an IELTS speaking test. Extract 3 ((NB modifications to the original have been made to numbering and the placement of pauses.)) (E: examiner, C: candidate) 1 2 3 4 5 6 7 8 9 11 12

E:

E:

ˆ ˆ ˆ ˆ ˆ ˆ

C:

ˆ

C: E: C:

do people (0.6) (0.2) yeah (0.5)

yeah (2.7) so (1.2) do people generally prefer watching films (.) at home 10 (0.3) mm hm (0.6) (from Seedhouse and Egbert 2006, 10)

As was the case in extract 2, here the examiner is also accommodating. In his first turn he slows down his speech for the key part of his question.

An Interaction-Focused Pedagogy based on Conversation Analysis

239

The candidate is quick to respond with a yes. The examiner then produces an “add on question” (Lazarton 2002) in line 6, uttered slowly and after a gap. This provides a second choice after possible completion. The student again produces yeah as a response. A decidedly long gap ensues before s/he adds so which itself is followed by a long gap (line 7). The examiner has thus created space for a more adequate response. Indeed, the candidate orients to the inadequacy of his/her response lag through so. The examiner’s recycled turn that follows is an exact repetition of his first question, but this time there is prosodic emphasis on generally. This action suggests that the examiner’s repair is targeted as a possible mishearing. However, the candidate responds with the even more minimal mm hm. Through these actions it is clear that the examiner is providing the candidate with an opportunity to extend his/her turn, or to initiate repair, but neither of these actions is forthcoming. In interaction, participants strive for intersubjectivity or mutual understanding. The locus for uncovering this achievement is in the next turn at talk and what participants do next. The analytical question in Conversation Analysis here is “Why that now?” (Schegloff and Sacks 1973). Seedhouse and Egbert (2006) maintain that intersubjectivity is not a goal of the IELTS speaking test. However, the phenomenon of repair is itself a display of intersubjectivity, and therefore a crucial skill that language learners need to acquire and demonstrate as part of the interactional skills in their L2 in both the classroom and in a testing situation. If it is true that intersubjectivity is not a goal of speaking tests, then to what extent can a test inspire confidence in being able to generalise about test takers’ ability to function and apply their skills in the real world? A test taker’s and student’s ability to repair successfully (rather than give an incorrectly “fitted” answer or provide a minimal response as shown in the above samples) ought to be judged as an ability to deal with trouble in talk as (native) speakers of English would in such situations.

3.2. Common interactional problem 2: Lack of assessment/terminal components and moving straight into leave-taking without closing down the conversation Assessments are statements of evaluation and when they are found in conversation closings, they signal the end of the conversation before the interactants engage in a terminal exchange such as leave taking (bye-bye). Assessments of the likes that’s good and response tokens like okay are usually uttered with a falling intonation and are referred to as terminal components because of their placement in a pre-closing sequence.

Chapter Ten

240

Extract 4 ((HUY, HOA, both students at level A2)) 1 2 3 4 5 6 7

HUY: HOA: HUY: HOA: HUY: HOA: HUY:

ˆ

last-last week? last wek. yeah. mm hm e::r (0.6) yeah, I- I see you later er see you uh bye bye. bye. uh

Extract 5 ((ANH, THU, both students at level A2)) 1 2 3 4 5 6

ANH: ˆ THU: ANH: THU:

ehr (0.4) I- I would: uhn like uhn (0.5) say something but I'm- I'm not (good). er see you later. (0.4) see you later? yes see you. bye, bye.

In extracts 4 and 5, the students move straight into leave-taking without closing down their conversation. This results in an abrupt ending. It is customary in English to move the conversation into closing before leavetaking with terminal components such as yeah or okay uttered with falling intonation (Button 1987). In these extracts, it is noticeable that before moving into leave-taking the speakers produce some hesitation, which seems to indicate that they are unsure as to what their next move should be or how to properly verbalise their next action. In extract 4, Huy utters mm hm followed by e::r and a silence (0.6) and another response token yeah pronounced with rising intonation before uttering the formulaic leavetaking. In extract 5, Anh indicates that she would like to say something else before closing the conversation; however she is unable to express herself and abandons her utterance (lines 1 and 2: I- I would: uhn like uhn (0.5) say something). Moreover, she adds but I’m- I’m not (good). er, by which she implies that her English level is not adequate for her to fully express herself. This utterance seemingly denotes her frustration, hence the abandonment of her previous attempt. These two extracts show a lack of pragmatic competence in properly closing a conversation. Even though this does not cause any miscommunication, it still results in an abrupt ending which might be consequential to the interaction if it were to occur with more “expert” speakers.

An Interaction-Focused Pedagogy based on Conversation Analysis

241

All the excerpts above show that students produced a sequentially inappropriate next action. In excerpts 1-3, the students did not engage in a repair strategy with an open class initiator like “mm?” and instead uttered a response token. In excerpts 4-5, the students did not produce pre-closing components to take leave of each other. All of these point to a lack of pragmatic competence rather than a gap in vocabulary per se.

4. Developing an interaction-focused pedagogy The above analyses have illustrated the issues that students have with key aspects of interaction. These include their failure to produce a repair or an appropriate second pair part, a finding reported elsewhere (for example, Cafarella 1997; He 1998), and their inappropriate skills in closing a conversation. While these problems should not be seen as belonging to one participant only in each of the dyadic or triadic contexts, it is nonetheless evident that explicit teaching, and in the context of examinations, training of examiners1, is needed to address issues of interactional competence. Barraja-Rohan (1997) advocated a focused, direct and staged teaching methodology based on concepts in Conversation Analysis. Earlier Scarino et al. (1988) proposed a model for developing L2 communication. We have combined these concepts and developed a model by borrowing and adapting the terminology in Scarino et al. (1988) to focus specifically on spoken interaction, rather than communication in general, and have added metacognition, an important and essential feature of sound pedagogy, which is present at each phase of the model itself. When linked to actual interaction, metacognitive strategies may involve returning to any phase of the model, if particular features of conversation still pose problems. The processes in the pedagogy are thus iterative.

242

Chapterr Ten

FIGURE 1. Peddagogy Wheel

As illustrateed in Figure 1, the first stag ge is to raise sstudents’ awaareness of the feature tthat is to be taaught. This caan be done in a number of ways, w but will entail ssome kind off presentation n of naturallyy occurring in nteraction accompanied by exercises that involve students in nnoticing the feeature; for example, inddividual note--taking follow wed by group aand/or class sharing s as in the think, pair, share teechnique. The nextt stage will enntail exercisess that providee students with h focused practice of tthe feature thaat is being shaaped. Studentss will be encouraged to engage in diiscussion withh others, inclu uding studentss as well as thee teacher, and, they wiill be encouraaged to take notes. n In additiion, students are given opportunitiees for self-obsservation thro ough filming and recording g as they interact withh other studennts. In the thhird phase, preeparing for in nteraction, stuudents will be asked to have a connversation in front of thee class. The emphasis heere is on spontaneity,, not prepared speech. The class c will provvide feedback k. The finaal stage involvves students in n seeking out oopportunities for actual interaction ooutside of the classroom. In n the case of an ESL class, this t is not as challenginng as for studdents in the foreign languagge classroom, who may have fewer opportunitiess to hear and d engage witth the languaage being learned outsside school orr the immediaate classroom. For schools that have ready accesss to commuunities of sp peakers, this is less frau ught with difficulties even thoughh building relationships taakes time an nd effort. However, scchools that havve little directt contact with native/expertt speakers of the L2 lannguage have an a extra challeenge. One posssible solution n is to use

An Interaction-Focused Pedagogy based on Conversation Analysis

243

Web 2.0 tools that make interaction possible – Skyping between sister schools, for example. Here too time is needed to develop sound relationships (Colmar Brunton, DEECD, 2012). Two important features of this pedagogy are the need for reflection essential to metacognition, which comprises both metacognitive knowledge and metacognitive regulation (Baird and Posner 2000; Flavell, Miller, and Miller 2002), and the interactions that support learning. Strategies for developing reflective practices in students are essential. One way of doing this is the inclusion of self- and peer-assessment, which makes explicit what students are looking for in behaviour, and which builds awareness leading to improvement. Another is to ask students to write down what they notice about their own and others’ interactions in both languages through journaling, note-taking or blogging. While the focus is on developing the skills of interaction, students will also be interacting with each other in the learning process through discussion about the particular interactional feature being noticed and/or worked on. This provides an interaction-centred pedagogy. In the foreign language classroom, an important consideration will be on which language these conversations will take place in, and this meta-talk can provide further scope for students to use their linguistic resources to develop a wide range of interactional skills in both their languages. Reflecting on their use of L1 and L2 will further enable students to build metacognitive skills. In this important respect, Conversation Analysis provides not just a structure and insights for teaching interactional competence but also a grammar of interaction.

4.1. Application of the pedagogy The methodology outlined above, which includes the application of concepts from Conversation Analysis, led to successful interactions as exemplified in the extract below. This extract shows a complete conversation between two Vietnamese students at beginner level, who had displayed similar problems to those outlined in the analysis above. The students produced this conversation after six lessons that focussed on the following aspects of interactional competence: x conversational openings (including greetings with appropriate socio-cultural norms of interaction such as an how are you sequence) x adjacency pairs (paired utterances such as greeting/return greeting, question/answer)

Chapter Ten

244

x initiating and closing a topic including using assessments such as that’s good) x keeping a conversation going through questions like what about you? x response tokens to show active listener behaviour, such as mm (hm), yeah, okay x closing a conversation with terminal components such as okay with falling intonation x appropriate leave-taking. Extract 6 ((Hung and Truc are both male Vietnamese students at level A2.) 1 2 3 4 5 6 7 8 9 10

TRUC: HUNG: TRUC: HUNG: TRUC: HUNG:

11 12 13 14

HUNG: TRUC: HUNG: TRUC:

TRUC: HUNG: TRUC:

hello Hung. how are you today.= =good thanks. are you are er you going back er the next course? yes of course. what did you plan to do on the holiday. o::h, my holiday? em I-I will looking er for a job, if I don't get a job I come back this school. oh yes that good. how about you? mm I:: stay home, listen the music an’ I review my lesson. oh:: I'm sorry, I-I’m busy I must go now. okay. s[ee you], bye bye. [see you] (from Barraja-Rohan 2011, 494)2

In this conversation, the students display a fundamental understanding of how a basic conversation works in English. Truc starts the conversation with a greeting followed by the routinised how are you question. Hung responds briefly in a culturally appropriate manner in line 2. It is Truc who introduces the first topic through a question (line 3) which is responded to in the next turn by Hung (line 4). Truc keeps the conversation going by asking another question (line 6), which engenders a longer response involving a compound turn constructional unit3 of the type I will … if I … (Lerner, 1996) in lines 6 and 7. This compound turn constructional unit is a complex clause which had not been taught in class. This indicates that developing interactional competence may possibly involve producing complex grammar (Barraja-Rohan 2013). In the next turn, Truc produces a response token oh yes followed by an assessment, which shows his pragmatic understanding that good. Then Hung returns the question to

An Interaction-Focused Pedagogy based on Conversation Analysis

245

Truc, who until this point had initiated all the questions. Hung initiates another action in line 12 by moving into closing, which Truc orients to in the next turn (line 13) by producing the terminal okay with falling intonation. In line 14 Hung appropriately initiates leave-taking which is simultaneously reciprocated by Truc in an overlap (line 15), thus indicating Truc’s pragmatic understanding of where the conversation is heading. This conversation which on the surface appears simple, reveals some complexity, as indicated above. It shows how both interactants orient to the structure of conversation in an orderly fashion through the opening, centring and closing produced with socio-culturally appropriate norms of interaction (Barraja-Rohan 1997) as they alternate between speaker and listener roles. Ultimately, it is successful because the interactants display intersubjectivity and use an adequate bank of interactional resources to reach their social goal.

5. Concluding remarks Students continue to display problems in speaking tests. This is evident for example in the absence of interactional features in the students’ utterances as analysed in the IELTS speaking test and Italian oral examination samples discussed above. Further examples of these can be found in the original studies of Filipi (1994, 1998), Seedhouse and Egbert (2006), Seedhouse and Harris (2011). Some of these problems may be attributed to the design of the test that does not permit students to have control over initiating, as reported by Ducasse and Brown (2011), Filipi (1997), and Seedhouse and Harris (2011). However, they may also be attributed to an L2 classroom pedagogy that neglects to focus on explicit awareness and the teaching of culturally appropriate interaction as it happens in the wild. The pedagogy outlined above proceeds from the identification of issues in students’ interactional behaviour and knowledge of both students’ L1 and their L2. It then proposes a series of pedagogic interventions. These include exposing students to unscripted/naturally occurring conversations and discussion of concepts of Conversation Analysis as part of their awareness building, and explicit teaching of these concepts for authentic participation in authentic contexts. In this process, students engage in cross-cultural discussions on L1 and L2 so that they develop an awareness of the socio-pragmatic differences and similarities between them. However, this process also demands that teachers themselves become aware of the organisation of social interaction and how

246

Chapter Ten

pragmatics impacts on communication by placing value and importance on peer to peer interaction in the classroom (Barraja-Rohan 2011, Hellermann 2008). This requires that teachers build opportunities into their teaching for identifying students’ L2 pragmatics issues in order to be able to help them remedy them through explicit teaching. Students also need to become aware of their own learning in this context and build metacognitive skills. For this to occur, teachers and students need explicit Conversation Analysis training through tasks that use corpora of actual and naturally occurring data (Barraja-Rohan and Pritchard 1997), and include a description of their relevant contexts. In this way, they can attend to authentic interaction, make explicit their implicit understanding of the rules of conversation in their language(s), and appreciate that pragmatics is learnable. Finally, in a world that is becoming increasingly multilingual, they can understand and feel confident in making distinctions between what is universal about interaction and what is culturally variant.

References Antaki, C., Finlay, W. M. L., Williams, V. et al. 2013. “Conversation Analysis and Interventions for Change” (presentation, 13th International Pragmatics Conference, New Delhi, India, September 813). Baird, A. J., and I. M. Posner. 2000. “Executive Attention and Metacognitive Regulation.” Consciousness and Cognition 9: 288-307. Barraja-Rohan, A.-M. 1997. “Teaching Conversation and Sociocultural Norms with Conversation Analysis.” In Teaching Language, Teaching Culture. Australian Review of Applied Linguistics, Series S, Special Issue, edited by A. J. Liddicoat, and C. Crozet, 14: 71-88. —. 2011. “Using Conversation Analysis in the Second Language Classroom to Teach Interactional Competence.” Language Teaching Research 15 (4): 479-507. —. 2013. “Orienting Towards L2 Interactional Competence: Learning to Engage in Multi-unit Storytelling” (presentation, Thinking, doing, learning: Usage based perspectives on second language learning Conference, Odense, Denmark). Barraja-Rohan, A.-M., and C. R. Pritchard. 1997. Beyond Talk: A course in communication and conversation skills for intermediate adult learners of English. Melbourne: Western Melbourne Institute of TAFE.

An Interaction-Focused Pedagogy based on Conversation Analysis

247

Brouwer, C. E., and J. Wagner. 2004. “Developmental Issues in Second Language Conversation.” Journal of Applied Linguistics 1 (1): 29-47. Button, G. 1987. “Moving Out of Closings.” In Talk and Social Organisation, edited by G. Button, and J. R. E. Lee, 101-151. Clevedon: Mutilingual Matters. Cafarella, C. 1997. “Assessor Accommodation in the V.C.E. Italian Oral Test.” Australian Review of Applied Linguistics 20: 21-41. Colmar Brunton. 2012. “Summary of Key findings: The Benefits, Impacts and Success Factors for Successful Sister School Relationships – Key Findings from the Research.” DEECD Sister School Relationships Final Report. http://www.education.vic.gov.au/Documents/school/principals/manage ment/factsisschrels.pdf Drew, P., and J. Heritage, eds. 1992. Talk at Work: Interaction in Institutional Settings. Cambridge: Cambridge University Press. Ducasse, A., and A. Brown. 2011. “The Role of Interactive Communication in IELTS Speaking and its Relationship to Candidates’ Preparedness for Study or Training Contexts.” IELTS Research Reports 12: 125-150. Filipi, A. 1994. “Interaction in an Italian oral test: The role of some expansion sequences.” In “Spoken Interaction Studies in Australia. Australian Review of Applied Linguistics,” edited by R. Gardner, special issue, Australian Review of Applied Linguistics Series S 11: 119-136. —. 1997. “Managing Roles in the Openings and Closings of an Italian Oral Test.” Monash University Linguistics Papers 1 (2): 3-10. —. 1998. “Interaction or Interrogation? A Study of Talk Occurring in a Sample of the 1992 VCE Italian Oral Common Assessment Task (CAT 2).” Australian Review of Applied Linguistics 21 (2): 123-137. —. 2009. Toddler and Parent Interaction: The Organisation of Gaze, Pointing and Vocalisation. Amsterdam: Benjamins. —. 2015. “The Development of Recipient Design in Bilingual ChildParent Interaction.” Research on Language & Social Interaction 48 (1): 100-119. Firth, A., and J. Wagner. 2007. “Second/Foreign Language Learning as a Social Accomplishment: Elaborations on a Reconceptualized SLA.” The Modern Language Journal 91 (Focus Issue): 800-819. Flavell, H. J., H. P. Miller, and A. S. Miller. 2002. Cognitive Development (4th Edition). Upper Saddle River, NJ: Pearson Education.

248

Chapter Ten

Gardner, H., and M. Forrester, eds. 2010. Analysing Interactions in Childhood: Insights from Conversation Analysis. Chichester, UK: Wiley-Blackwell. Gardner, R. 2001. When Listeners Talk: Response Tokens and Listener Stance. Amsterdam: Benjamins. —. 2013. “Conversation Analysis in the Classroom.” In The Handbook of Conversation Analysis, edited by J. Sidnell, and T. Stivers, 593-611. Malden, MA: Wiley-Blackwell. Golato, A. 2002. “German Compliment Responses.” Journal of Pragmatics 34: 47-571. —. 2003. “Studying Compliment Responses: A Comparison of DCTs and Recording of Naturally Occurring Talk.” Applied Linguistics 24 (1): 90-121. Grice, H. P. 1981. “Presupposition and Conversational Implicature.” In Radical Pragmatics, edited by P. Cole, 167-181. New York: Academic Press. He, A. W. 1998. “Answering Questions in LPIs: A Case Study.” In Talking and Testing: Discourse Approaches to the Assessment of Oral Proficiency, edited by R. Young, and A. W. He, 101-116. Amsterdam: Benjamins. Hellermann, J. 2006. “Classroom Interactive Practices for Developing L2 literacy: A Microethnographic Study of Two Beginning Adult Learners of English.” Applied Linguistics 27 (3): 377-404. —. 2008. Social Actions for Classroom Language Learning. Clevedon, UK: Multilingual Matters. Huth, T., and C. Taleghani-Nikazm. 2006. “How Can Insights From Conversation Analysis be Directly Applied to Teaching L2 Pragmatics?” Language Teaching Research 10 (1): 53-79. Jefferson, G. 1984. “Notes on a Systematic Deployment of the Acknowledgement Tokens ‘yeah’ and ‘mm hm’.” Papers in Linguistics 17: 197-216. Lazarton, A. 1996. “Interlocutor Support in Oral Proficiency Interviews: The Case of CASE.” Language Testing 13 (2): 151-172. Lerner, G. H. 1996. “On the "Semi-permeable" Character of Grammatical Units in Conversation: Conditional Entry into the Turn Space of Another Speaker.” In Interaction and Grammar, edited by E. A. Schegloff, E. Ochs, and S. A. Thompson, 238-276. Cambridge: Cambridge University Press. Levinson, S. C. 1983. Pragmatics. Cambridge: Cambridge University Press.

An Interaction-Focused Pedagogy based on Conversation Analysis

249

Markee, N. P. 2004. “Zones of Interactional Transition in ESL Classes.” The Modern Language Journal 88: 583-596. —. 2008. “Toward a Learning Behavior Tracking Methodology for CAfor-SLA.” Applied Linguistics 29 (3): 404-427. McHoul, A. 1978. “The Organization of Turns at Formal Talk in the Classroom.” Language in Society 7 (2): 183-213. —. 1990. “The Organization of Repair in Classroom Talk.” Language in Society 19: 349-377. Mehan, H. 1979. Learning Lessons: Social Organization in the Classroom. Cambridge, Mass.: Harvard University Press. Mori, J. 2007. “Border Crossings? Exploring the Intersection of Second Language Acquisition, Conversation Analysis, and Foreign Language Pedagogy.” The Modern Language Journal 91: 840-862. Nguyen, H. T. 2011. “A Longitudinal Microanalysis of a Second Language Learner's Participation.” In L2 Learning as Social Practice Conversation-Analytic Perspectives, edited by G. Pallotti, and J. Wagner, 17-44. Honolulu, HI: University of Hawai'i, National Foreign Language Resource Center. —. 2012. “Social Interaction and Competence Development: Learning the Structural Organization of a Communicative Practice.” Learning, Culture and Social Interaction 1 (2): 127-142. Nunan, D. 1991. Language Teaching Methodology: A Textbook for Teachers. London: Prentice Hall. O'Grady, W., J. Archibald, M. Aronoff, and J. Rees-Miller. 2001. Contemporary Linguistics. Boston: Bedford/St. Martin's. Pallotti, G., and J. Wagner, eds. 2011. L2 Learning as Social Practice: Conversation-analytic Perspectives. University of Hawai’i at MƗnoa: National Foreign Language Resource Center. Rose, K. R. 2005. “On the Effects of Instruction in Second Language Pragmatics.” System 33 (3): 385-399. Sacks, H., E. A Schegloff, and G. Jefferson. 1974. “A Simplest Systematics for the Organisation of Turn-taking in Conversation.” Language 50 (4): 696-735. Scarino, A., D. Vale, P. McKay, and J. Clark. 1988. Australian Language Levels (ALL) Guidelines, Book 2: Syllabus Development and Programming. Canberra: Curriculum Development Centre. Schegloff, E. A. 1991. “Conversation Analysis and Socially Shared Cognition.” In Perspectives on Socially Shared Cognition, edited by L. Resnick, J. Levine, and S. Teasley, 150-171. Washington, DC: American Psychological Association.

250

Chapter Ten

—. 1997. “Third Turn Repair.” In Towards a Social Science of Language: Papers in Honor of William Labov, edited by G. R. Guy, C. Feagin, D. Schriffin, and J. Baugh, volume 2, 31-40. Amsterdam: Benjamins. —. 2007. Sequence Organization in Interaction: A Primer in Conversation Analysis, volume 1. Cambridge: Cambridge University Press. Schegloff, E. A., and H. Sacks. 1973. “Opening Up Closings.” Semiotica 8 (4): 289-327. Searle, J. R. 1979. Expression and Meaning. Studies in the Theory of Speech Acts. Cambridge: Cambridge University Press. Seedhouse, P. 2004. The Interactional Architecture of the Language Classroom: A Conversation Analysis Perspective. Oxford: Blackwell. Seedhouse, P., and M. Egbert. 2006. “The Interactional Organisation of the IELTS Speaking Test.” IELTS Research Reports 6: 161-205. Seedhouse, P., and P. Harris. 2011. “Topic Development in the IELTS Speaking Test.” IELTS Research Reports 12: 55-110. Shomoossi, N., and M. Tavakoli. 2010. “Authenticity and Authentication in Language Testing: An Operational Perspective.” Modern Journal of Languages 2 (1): 1-26. Sidnell, J. 2013. “Basic Conversation Analytical Methods.” In The Handbook of Conversation Analysis, edited by J. Sidnell, and T. Stivers, 77-99. Malden, MA: Wiley-Blackwell. Sidnell, J., and Stivers T., eds. 2013. The Handbook of Conversation Analysisi. Malden, MA: Wiley-Blackwell. Taleghani-Nikazm, C. 2002. “A Conversation Analytical Study of Telephone Conversation Openings Between Native and Nonnative Speakers.” Journal of Pragmatics 34: 1807-1832. Wong, J. 2002. “‘Applying’ Conversation Analysis in Applied Linguistics: Evaluating Dialogue in English as a Second Language Textbooks.” International Review of Applied Linguistics 40: 37-60. —. 2011. “Pragmatic Competency in Telephone Conversational Openings.” In Pragmatics: Teaching Natural Conversation, edited by N. Houck, and D. Tatsuki, 135-152. Alexandria, VA: TESOL. Wong, J., and H. Z. Waring. 2010. Conversation Analysis and Second Language Pedagogy: A Guide for ESL/EFL Teachers. New York: Routledge.

An Interaction-Focused Pedagogy based on Conversation Analysis

251

Appendix Transcription Notations The conventions used in the transcriptions are the following: Intonation contours: . full fall , continuing contour or slight rise ? full rise Other conventions used are as follows: ˆ the arrow indicates a point of interest today.= latching indicates continuous stretch of talk [bye] the square brackets indicate simultaneous talk o::h the colon indicates lengthening of sound but I’m the underline indicates sentence stress last-last the hyphen indicates abrupt cut off or glottal stop (.) indicates a very short pause or a micropause (0.5) indicates the length of the silence in relation to the surrounding talk the signs >< indicate talk that is slower than the surrounding talk ((Nods.)) the double brackets indicate co-activity relevant to the interaction ( ) indicates talk that is not clearly audible cosa COSA capitalisation indicates loud talk

Notes 1 The results of this study (reported in Filipi 1994, 1998) were in fact used by the first author in the institution responsible for examination development (the then Victorian Board of Studies) to train examiners across all languages as well as teachers and students of Italian throughout the 1990s. 2 See Barraja-Rohan (2011) for a full description of pre-instruction and postinstruction conversations by the same students. 3 A turn constructional unit is a basic segment of speech as defined by Sacks, Schegloff and Jefferson (1974).

CHAPTER ELEVEN TOPIC MANAGEMENT AND INTERACTIONAL COMPETENCE IN SPANISH L2 CONVERSATION MARTA GARCÍA GARCÍA

1. Conversation and interactional competence Conversation is “the primordial site of social life” (Drew 2005, 74), in terms of both quantity, “accounting for more than 90% of all spoken language” (Cheng 2003, 12), and quality: it is through conversation that we can share opinions, experiences and feelings. It is also through conversation that we become acquainted with people and become friends, but we also criticize, disagree, argue, and resolve those disagreements through conversation. In sum, conversation enables us to establish, maintain and terminate social relationships. As the central element of socialization, conversation is also the “matrix for language acquisition” (Levinson 1983, 284). In the same vein, Hatch (1978, 404) suggested that “language learning evolves out of learning how to carry on conversations,” which applies to both first (L1) and second or foreign (L2) languages. For L2 students, the reason for mastering conversation is thus twofold: it is both a medium for learning the foreign language and a means of becoming full participants in the target community (Barraja-Rohan 2000, 65). What, then, is required for people to successfully engage in conversation? We can find an early conceptualization of the ability to take part in a conversation in studies of discourse pragmatics (e.g., House 1996, 2002, and Riggenbach 1991). Pragmatic fluency is here defined – or operationalized – in terms of the ability to make substantial contributions at an appropriate speech rate and through the appropriate use of routinized pragmatic phenomena and formulae (for signaling alignment, for topic initiating and change, and for repair), and the following argument is made:

Chapter Eleven

254

[I]t is [this] linguistic and pragmatic knowledge for performance which is of prime importance and should thus be given primary attention in attempts to improve learners’ interactional competence in ELF [English as Lingua Franca]. (House 2002, 263)

Whereas discourse pragmatics is primarily concerned with the appropriate linguistic forms and routines that language learners use, the construct of interactional competence (IC) that House’s quote mentions has been conceptualized slightly differently in the socially oriented realm of L2 research. For Hall and Pekarek Doehler (2011, 2), IC implies the following: [a)] knowledge of social-context-specific communicative events or activity types […] [b)] the ability to deploy and to recognize context-specific patterns by which turns are taken, actions are organized and practices are ordered […] [c)] the prosodic, linguistic, sequential and nonverbal resources conventionally used for producing and interpreting turns and actions, to construct them so that they are recognizable for others, and to repair problems in maintaining shared understanding of the interactional work we and our interlocutors are accomplishing together.

Although similarities are initially evident, in particular, in interlocutors’ “competence in participating,” as evidenced by Conversation Analysis (CA), the conceptualization of IC outlined above involves the following properties: -

-

-

it is not the property of an individual; rather, it is “co-constructed by all participants” (He and Young 1998, 7); it is not general and transferrable to every situation; rather, it is local, that is, specific for a determined interactive (Hall 1995) or discursive practice (He and Young 1998); it is neither abstract nor discrete knowledge that is somehow ready for use, that is, not a “competence for action” but rather a “competence-in-action:” it unfolds within the action and is “a constitutive and inseparable dimension of the action itself” (Pekarek Doehler 2006, 28);1 it is both a resource and an object for learning insofar as the L2 learner’s existing competencies enable her to participate in new practices “while helping develop novel competencies” (Kasper 2006, 87).

Because the specific practice of conversation “is, in the awareness of participants, talking about topics” (Orletti 1989, 76), the main objective of

Topic Management and Interactional Competence in Spanish

255

this paper is to seek to understand how a particular aspect of IC – topic management – is accomplished among Spanish L2 students.

2. Topic management in conversation: the perspective of CA The above quotation by Orletti highlights the point that topic in conversation is a highly intuitive, commonsense notion that is always accessible to participants, whose awareness of “what is being talked about” forms the basis for their contributions. For researchers, however, an uncontested definition of topic remains elusive. It is difficult not only to say what a topic is but also to determine the segment of speech (e.g., an utterance, a turn, or a chain of speech) to which the topic applies (Schegloff 1990, 52). Rather than questioning what a topic is or what participants talk about and focusing on content or semantic issues, CA is interested in the structural function of topic management in conversation and in the formal procedures utilized by conversationalists. Topical coherence is therefore viewed primarily as an interactional achievement and not as a matter of content. First, topical talk is a “collaborative phenomenon” that involves one speaker doing topic-developmental utterances and the other(s) supporting him or her with attention signals, questions, reactions and so forth “to keep the line of talk going” (Maynard 1980, 266). Second, as Levinson (1983, 315) states: [T]opical coherence is something constructed across turns by the collaboration of participants. What needs to be studied is how potential topics are introduced and collaboratively ratified, how they are marked as ‘new’, ‘touched-off’, ‘misplaced’ and so on, how are they avoided or competed over and how they are collaboratively closed down.

In a nutshell, “all manner of conversational procedures are implicated in the management of topic” (Heritage 1989, 29), an issue that makes topic analysis an extraordinarily complex endeavor. The CA literature has treated topical coherence as the tension between maintenance and progression: on the one hand, conversationalists have to fit their turn with the prior one; on the other, they are expected to say something new and move the conversation forward. Therefore, [E]ach segment of discourse [...] represents some kind of resolution of the tension between the two opposing needs, that of staying on topic and that of renewing topics. (Linell and Korolija 1997, 167).

256

Chapter Eleven

The issue of renewing topics is what speakers usually understand as topic change, and although CA terminology has not been very consistent in this sense, it is possible to distinguish between topic initiation and topic shift as two aspects of topic change. Topic initiation refers to the introduction of a new topic at the beginning of a conversation or after a close or a series of silences, and topic shift is the issue of “shifting emphasis within a topic or moving towards a new topic” (Wong and Waring 2010, 104). If this move goes unnoticed and topics flow into another, the interlocutors use a stepwise topic change, what is generally seen as the “best way to move from topic to topic” (Sacks 1992, 566). However, more often than not, conversationalists may want to introduce a new topic in a marked or disjunctive way, that is, by means of expressions such as actually or by the way that signal such turns “as not tightly fitted to the ongoing talk” (Wong and Waring 2010, 115). Furthermore, the on-going topic may come to an end when the participants run out of things to talk about. This happens in very structured and collaborative ways, as Howe (1991, 9) observed: there is a “regular sequence of ending indicators,” such as summary assessments,2 acknowledgment tokens and pauses. Thus, topic closings are conversational environments marked by a reduced number of primary speakers (i.e., participants who make topical contributions), by shorter utterances and by a growing frequency of minimal responses and longer between-turn gaps (Bublitz 1988, 73). The issue of closing a topic represents a potential problem for conversationalists, who have to collaboratively come to an end and, at the same time, have to find a way to restore continuous talk. Topic closings are therefore a highly interesting “actional microcosm” (Fazel Lauzon, Pekarek Doehler, and Pochon Berger 2009) to observe the IC of participants. Most likely owing to its aforementioned complexity, the early work within the CA tradition on topical management “has fallen away almost completely” (Seedhouse and Harris 2011, 8). The literature on L2 topic management is also scarce. Studies have traditionally focused on interactions between native and non-native speakers in the context of language assessment (Young and Halleck 1998; Franco Cordón 2011; Gan, Davison, and Hamp-Lyons 2009; see also Seedhouse and Harris 2011 for an overview) or in informal settings, such as conversations among colleagues or friends (Cheng 2003). More learning-oriented environments are addressed in the studies by Nguyen (2011) – casual chats with the teacher during office hours, – and König (2013) – conversations between the au-pair student and the host family. Thus, there is a gap not only in the literature on L2 conversational topic management in general,

Topic Management and Interactional Competence in Spanish

257

but also in the research specifically examining the interactions between two or more learners without the participation of (a) more proficient speaker(s). The current study aims to contribute to filling this gap in the research and to expand the results of a previous study (García García 2013). Adopting a CA approach, it examines how Spanish L2 learners confront the issue of topic management and, more specifically, how they solve conflicts in cooperative topic closures and who in particular (and with what resources) takes the initiative to solve them.

3. Data and method The examples that I will present in the following section were obtained from a larger corpus (García García 2014) of 10 audiotaped conversations between learners of Spanish at two German universities. The participants were 30 students (monolinguals and bilinguals in German) whose proficiency levels in Spanish ranged from A2 to B2 according to the Common European Framework of Reference. Three students volunteered to participate in each conversation; all were familiar with one another, as they were enrolled in the same language course. The following recording setting was used: outside of lesson hours, the three volunteers met with the teacher-researcher3 in the same room as the regular course and were prompted to speak for approximately half an hour. As a starting point, the students could draw on a text4 that they had received the previous week, but they were explicitly told that the use of such a text was optional and that they could talk about any topics they wished. In CA, analysis begins with a process of repeatedly listening to the data and meticulous transcription: “[t]his process is to be seen not as a means to capture the data for later analysis, but as a tool to become as closely familiar with the object of inquiry as possible, thereby turning the act of transcription into an act of analysis” (Gardner 2004a, 269)5. As I examined the data, I adopted the analytical stance of “unmotivated looking” (Psathas 1990), that is, I did not seek to find certain instances of pre-determined linguistic phenomena so as to prove a hypothesis. Rather, I allowed the data to drive the questions (Lazaraton 2002, 77) and to speak for themselves. In doing so, I found many sequences in which the participants showed (verbally and nonverbally) that they were having problems maintaining the flow of conversation: substantial contributions became fewer, and minimal responses and rather long silences were common. The sequential environment of topic closures appeared to be

Chapter Eleven

258

especially problematic for the participants. These sequences of interest constitute the collection of instances that I worked on.

4. Analysis In the following sections, I will comment on four examples that exemplify which resources the participants drew on and which ones they failed to display during their performance of topic management actions.

4.1. Non-collaborative closings Excerpt (1) is taken from a conversation between three German girls (Bettina, Bianca and Sylvie) at a B1 level of competence in Spanish. Whereas both Bianca and Bettina speak very fluently, Sylvie speaks more hesitantly and at a slower speech rate. In this fragment, the students are talking about their experiences abroad and the possibility of living in a foreign country. (1) (BF: Bettina, BL: Bianca and SW: Sylvie) 1

BF

2 3

BL

4

BF

5

SW

6

BF

7 8

BL BF

9 10

BL BF

11 12

BL BF

13

sí yo pienso que es también (.) el mismo– yes I think that it is also the(*masc.) same also me: encantaría (.) eh vivir en Espa[ña: well(=Ger) I’d love to live in Spain [sí yes o en Sudamérica (.) un poco (.) eh uno o [dos años or in South America a short time one or two years [-soloonly sí: porque eh eh siempre no puedes sentarte como una española yes because always you can’t *feel like a Spanish person [hm [o como un una persona de: Sudaméri[ca or like a(masc.) a person from South America [hm y: mh sí (.) eh no te sientas como a ca– eh en casa and yes you don’t *feel like in ho– at home [he–he–he [°en Alemania° in Germany (1.0)

Topic Management and Interactional Competence in Spanish 14

BL

15

BF

16

BL

17

BF

18

SW

19

BL

20

SW

21

BL

22

SW

23 24 25

BF SW

26 27

BL

28 29

BF

30 31 32 33

SW BF

34 35

BL BF

36 37 38

BL BF

39

SW

posiblement[e es probably it is [sí: Yes más fácil eh con un– un novio de España y de: easier eh with a boyfriend from Spain and from sí:! yes sí:! sí claro yes yes of course América del Sur South America porque en ese caso es una cosa familiar .hh because in that case it is a familiar thing sí yes y: and hh ºsíº (.) y: mi patria es– (.) es en Alemania y mi– mi familia es (.) en and my home is is in Germany and my my family is in Ale[mania Germany [sí (1.5) para mí también he–he–he yes for me too (.) hm .hhh pero es muy interesante eh conocer a estos costumbres hm but it is very interesting to know (*to) these customs o cómo or how (.5) hm la gente son porque son diferente the people are(*pl.) because they are(*pl.) different hm y: puedes (.5) ehm: (.8) ehm (1.0) eh (0.8) also (1.0) and you can well(=Ger) aprender (.2) mucho eh para tu vida learn a lot for your life [hm [cómo vivir (.2) eh la how to live the [hm

259

Chapter Eleven

260 40 41

BF

[vida porque (.5) ºes (.2) un poco diferenteº (.2) he–he life because it is a little different (41.)

We can observe that this fragment is divided into two parts with two different patterns of interaction. The first part (lines 1-27) is characterized by a high level of agreement and collaboration among the three participants. It is an example of topical talk as a collaborative phenomenon, as mentioned in section 2. The two opinions stated (that it would be nice to live in Spain or South America despite the difficulty posed by not feeling at home and that it would be easier with a boyfriend from that country) are accompanied by alignment (line 3), receipt (lines 5 and 23), acknowledgement (lines 7 and 9) and agreement (lines 17, 18, 21, and 27) tokens. In line 29, Bettina introduces a new aspect (“but it is very interesting to know these customs”) as an attempt to relativize the previous negatively loaded arguments about a long stay abroad. From that moment forward, she is the only person who makes substantial contributions. The others show agreement with the minimal-acknowledgement token hm (Sylvie in line 32 and 39; Bianca in lines 34 and 37), but it appears that they had nothing to say. Bettina’s last comment (line 40) ends with laughter, which is a typical resource (Howe 1991; García García 2014; Holt 2010) used to signal that one has finished and – as long as no one has any further statements – also functions as self-alignment. The laughter is followed by a long silence, and the topic and the sequence close. This abrupt manner of closing a topic is particularly noticeable. There is no summary assessment, final comment, collaborative disengaging or interactional work. The topic is simply terminated or maybe abandoned because none of the participants makes any further contributions. Excerpt (2) illustrates a similar case. The participants in this conversation are Nora, Evelyn and Sebastian, all of whom are attending a B1 Spanish course: Nora speaks slightly more fluently than Evelyn, but both can get their meanings across. Sebastian has fewer linguistic resources at his disposal: he shows a very low speech rate and also takes the fewest turns. In this excerpt, they are talking about the role of physical appearance when living abroad. (2) (EG: Evelyn, NH: Nora, SB: Sebastian) 1 EG es que pasó al chico- al Paco (.) que: el chico habla *that what happened to the guy to *the Paco that the guy speaks 2 en alemán y no en español (.) a mí también eso pasó aj! German and not Spanish to me also that happened

Topic Management and Interactional Competence in Spanish 3 4

NH

5

EG

6

NH

7

EG

8 9

NH

10 EG 11 SB 12 EG 13 NH 14 EG 15 NH 16 17 EG 18 NH 19 20 21 SB 22 NH 23 24 EG 25 26 NH 27 28 NH

261

muchísimas veces en Barcelona porque mi cara lots of times in Barcelona because my face [sí yes [típicamente no es española tampoco typical it isn’t Spanish either y tu pelo and your hair y (.) si yo empezó- empecé a hablar español la gente and if I began(*3rdsg.) began to speak Spanish the people eh contesta en inglés answer in English sí yes GRR he-he me volvía loco! it was driving me crazy(*masc.) sí yes he-he sí y .h por ejemplo mi tía en Egip- Egipto eh puede hablar alemán yes and for example my aunt in Egypt can speak German y cuando estamos eh eh jóvenes? and when we *are young hm siempre hablamos en alemán y luego? cuando aprendió always we spoke German and then when I learned(*3rd) el árabe (.3) y yo hable con ella en árabe ella contesta (.) Arabic and I *talked with her in Arabic she answers en (.) alemán (.) porque no puede hablar conmigo en árabe in German because she can’t speak with me in Arabic .hhh es it is (.) hm (.5) muy raro very weird (3.0) tú est- tienes experiencias como esta en (.4) Columbia? you do you have experiences like this in Colombia

262

Chapter Eleven

The excerpt above is also divided into two parts (lines 1-14 and 15-28); both are reports of personal experiences. In the first, Evelyn speaks about her problems trying to speak Spanish in Barcelona. Because she does not look like a Spanish woman (and perhaps also because she speaks Spanish with a foreign accent), during her stay in Barcelona she frequently received answers in English, which markedly confused her (line 12, me volvía loco, “It was driving me crazy”). Evelyn’s report is met by Nora’s support, which is signaled by the acknowledgement tokens sí (lines 4, 9 and 13) and her collaborative turn in line 6 (y tu pelo, “and your hair”). Following Evelyn’s demonstration of annoyance (line 10), Sebastian laughs (line 11), and so does she (line 14) to bring this phase of the interaction to an end. Immediately after that, Nora self-selects. The acknowledgement token sí (“yes”), with which the turn begins, functions as a “pivot” (Jefferson 1993), a link between previous and actual talk, and shows engagement with Evelyn’s talk. Furthermore, Nora prefaces her turn with the words “for example” (por ejemplo, line 15); in doing so, it is clear to the co-participants that the subsequent talk is characterized as “analogous” (Linell and Korolija 1997, 180) to the previous anecdote. In contrast to the first segment, in this second segment, involvement signals are very scarce: solely in line 17 does Evelyn respond with a confirmation (hm) to Nora’s word-search (jóvenes?, “young?”).6 After the long turn (lines 18-21), the only reaction is an inbreath from Sebastian. Nora produces a summary assessment (es muy raro, “it is very weird”), but no comment follows. After a 3-second pause (line 27), Nora initiates a new topic with an invitation to Sebastian – the only one who has not yet told a personal anecdote – to produce topical talk. The interesting feature of this excerpt is that Nora takes all the responsibility for the interactional work: she self-selects by introducing the anecdote, she comments on her own story, and because nobody further contributes to the conversation, she produces a topic shift to restore topical talk and keep the conversation going. Finally, she accomplishes all of these tasks with a high degree of other-attentiveness.

4.2. Restoring topical talk It is apparent that not only closing a topic but also restoring the turntaking mechanism can be difficult on some occasions, an issue that will be discussed with greater detail in this section. The third example is taken from a conversation between Eileen, Markus and Arthur. Both Eileen and Markus were born in Germany, but to Turkish and Croatian parents, respectively. Arthur has lived in Germany

Topic Management and Interactional Competence in Spanish

263

with his Russian parents since he was 12 years old. All of them are attending an A2+ Spanish course. Eileen has reached a high level of fluency, Arthur is an average student, and Markus’s level of proficiency is similar to Sebastian’s: he speaks relatively slowly and participates very little. Immediately prior to this fragment of the conversation, Eileen had expressed her wish to live abroad in a southern country, possibly Spain, and she explains her reason in the following excerpt. (3) (EE: Eileen, AS: Arthur, MS: Markus) 1 EE sí es diferente la gente son diferente yes it is different the people are(*3rdpl.) different 2 y: (.5) ºme gusta esoº and (.5) I like that 3 (2.5) 4 EE [pero no sé but I don’t know 5 AS [pero por trabajar? but *for working 6 (1.0) 7 EE [tal vez– tal vez maybe maybe 8 AS [(en España) in Spain 9 EE también por also for 10 AS hm 11 EE trabajar to work 12 AS tam– también? also 13 EE sí: si hay un trabajo para mí: ahí yes if there is a job for me there 14 (.) 15 AS un bueno [trabajo a *good job 16 EE [por qué no? sí: pero no sé why not yes but I don’t know 17 (.5) 18 AS es difícil de: it is difficult to 19 EE hm y [mi: novio hm and my boyfriend

264 20 AS 21 EE 22 23 EE 24 AS 25 26 EE 27 28 MS 29 30 EE 31 32 EE

Chapter Eleven [(de hacer) to do es de Alemania entonces es un poco difícil is from Germany then it is a little difficult (1.5) no sé I don’t know es alemán? is he German (.5) hm (5.0) un poco difícil no? a little difficult isn’t it (.5) hm (.5) porque él no quiere: vivir en un otro país y: he–he because he doesn’t want to live in one another country and

In lines 1-2, the closure of the previous sequence occurs; Eileen has just explained why she would like to live in Spain. This explanation is followed by a silence of 2.5 seconds, and because no one else takes the floor, Eileen self-selects and, in line 4, expresses insecurity (pero no sé, “I don’t know”), which overlaps with Arthur’s question (pero por trabajar?, “but *for working?”). The “but” in this formulation indicates that Arthur objects to this possibility. Thus, both Eileen and Arthur have doubts about the viability of this plan, but for different reasons, as I will show below. Because the two utterances overlap, Eileen needs a short period of time to process Arthur’s question, and a gap of 1 second follows (line 6). This silence moves Arthur to produce an “increment” (Gardner 2004b) to the question with the adverbial phrase en España (“in Spain”), which overlaps with Eileen’s delayed answer (tal vez tal vez, “maybe maybe”) in line 7. This overlap is most likely the reason for Eileen’s subsequent confirmation (también por trabajar, “also for to work”). Arthur’s question (tamtambién?, “also?”) suggests surprise, and Eileen insists that finding a job would be a problem but that living in Spain would not (sí: si hay un trabajo para mí ahí, “yes if there is a job for me there”). Arthur adds (un*bueno trabajo, “a good job”), an observation that indicates, again, that he is questioning that possibility. This turn partially overlaps with Eileen’s repeated remark (por qué no?, “why not?”), and Eileen does not align with Arthur’s comment, instead repeating, in line 16 (as in line 4), no sé (“I don’t know”). Arthur acknowledges the difficulty (es difícil de, “it is

Topic Management and Interactional Competence in Spanish

265

difficult to”), and his turn is interrupted by Eileen, who finally discloses the reason for her insecurity, namely, the fact that her boyfriend comes from Germany and “therefore, it is a little difficult” (entonces es un poco difícil, line 21), something that actually does not concern the job issue. After this comment, Arthur has no more objections, and a silence follows (line 22). Eileen repeats, for the third time, no sé, which can be understood as a topic-closing statement and her final contribution to this “discussion.” Arthur’s question in line 24 (es alemán?) does not actually produce more topical talk, as Eileen has just revealed this piece of information. Eileen simply answers with a minimal response (hm), followed by a long silence of 5 seconds. We then find a typical closing environment (final comment + silence), which is similar to the previous excerpt: we arrive at an abrupt topic ending with only one comment by only one participant, while the others remain silent. Thus, line 28 would be an adequate sequential environment in which to produce a topic introduction. Instead, Markus produces an on-topic statement (un poco difícil), a repetition of Eileen’s final comment in line 21. This statement is, in a sense, an unsuccessful attempt to restore turnby-turn talk because Eileen can merely provide a minimal response (hm), exactly as in the question-answer pair (lines 24-26) before. Thus, at this point, after a topic closure (line 23) and two failed attempts to restore the turn-by-turn mechanism by Arthur and Markus, only Eileen takes the initiative (and the responsibility) to maintain the flow of the conversation. She does not change the topic – an action that would be inappropriate after Markus’s comment – and instead provides further explanation of the difficulty mentioned (porque él no quiere vivir en un otro país, “because he doesn’t want to live in one another country”). Two observations are key to this excerpt. The first is that both Markus and Arthur display a preference for maintaining the ongoing topic, even beyond topic boundaries. In so doing, they reveal the difficulties they experience when introducing new referents. The second observation concerns the context sensitivity of two of the speakers: Markus lacks not only the ability to introduce or restore topical talk but also the context sensitivity needed to treat the previous topic as closed. By contrast, Eileen displays both context sensitivity and the ability to reinitiate continuous talk.

4.3. Dealing with sensitive issues Sometimes, conversations come to an uncomfortable point, and exiting the topic may be the best strategy in order to re-establish mutual understanding. Excerpt (4) is part of the same conversation as excerpt (3).

266

Chapter Eleven

Immediately prior to this sequence, Markus has said that he prefers to live in Europe rather than in the United States. Arthur subsequently asks him for his opinion about people with “intercultural experience.” (4) (EE: Eileen, AS: Arthur, MS: Markus) 1 AS y creo que (.) la gente? and I think that the people 2 (.) 3 EE hm 4 AS que viví eh vivía en extranjero? who lived(*1stsg) lived in foreign country 5 (.2) 6 EE hm 7 MS hm 8 (.) 9 AS son muy interesantes son muy: eh are (*3rdpl) very interesting(pl.) are(*3rdpl) very 10 (0.5) 11 MS multicultura *multiculture 12 AS multicultura! *multiculture 13 EE hm 14 AS qué piensas? qué piensas de:? what do you think what do you think of 15 (0.5) 16 MS ((carraspeo)) ((clears throat)) 17 (1.0) 18 AS tú cre– eh (1.5) º(tú crees?)º do you th– do you think 19 (0.5) 20 MS creo que es (1.0) igual eh (1.5) en Alemania o en (1.5) en I think that it is the same in Germany or in in 21 Francia (2.0) a mí es no importante France to me is not important 22 (2.5) 23 AS mí eh tus– eh tienes amigos alemanos? me you(+s) do you have friends German 24 (2.5) 25 MS sí: (1.5) ein paar he–he–he yes a couple of them(=Ger) 26 EE he–he–he 27 AS he–he–he

Topic Management and Interactional Competence in Spanish 28 29 MS 30 EE 31 MS 32 AS 33 MS 34 AS 35 EE 36 AS 37 EE 38 39 MS 40 41 AS 42 MS 43 AS 44 45 MS 46 47 EE 48 49 AS

267

(0.5) sí yes -pocosa few y: croatas y:– and Croatians and y no [tienen– and don’t they have [italianos Italians una experiencia extranjeros? (1.0) los ale[manes an experience abroad(*pl.) the Germans [en otras países in other(*fem.) countries en otras países in other(*fem.) countries hm (.) no no (0.5) solo (1.5) sólo– sólo alemanes (1.0) qui no eh vivi– vivían only (1.5) only only Germans who didn’t live didn’t live (°no sé°) I don’t know que no vivían (.) en un otra país who didn’t live in one another(*fem.) country (2.0) no sé: °sicher° I don’t know sure(= Ger) (4.0) cómo es para tu padres por ejemplo? porque para how is it for your(*sg.) parents for example because for mi padres es muy difícil vivir a[quí my(*sg.) parents it is very difficult to live here [también también also

Arthur begins this fragment with an opinion (y creo que (.) la gente? que viví eh vivía en extranjero? son muy interesantes, “and I think people who have lived abroad are very interesting”) and finishes his contribution with a request for Markus’s opinion (line 14, qué piensas?, “what do you think?”). Markus exhibits some linguistic problems (hesitations and lapses) in delivering the answer as well as some problems understanding

268

Chapter Eleven

the sense or relevance of Arthur’s question because, for Markus, es igual (“it is the same”) and es no importante (“it is not important”). A silence of 2.5 seconds follows, and Arthur is clearly not satisfied with this response, insisting (¿tienes amigos alemanes?, “do you have German friends?”). Markus code-switches to German to give an ironic answer (line 25, ein paar, “a couple of them”), and this irony is understood by Eileen, who laughs and adds pocos (“a few”), feeling quite amused. Markus’s subsequent turn (lines 31-33, y croatas y italianos, “and Croatians and Italians”) can be understood as a type of self-assertion because he does not understand why Arthur is questioning whether he has German friends. However, Arthur is not interested in the “nationalities” of Markus’s friends; rather, he is interested in how much “experience abroad” – if any – his German friends have. Thus, Arthur continues in an almost inquisitorial tone: y no tenían una experiencia en el extranjero? (“and didn’t they have experience abroad?”). Markus answers twice with no sé (“I don’t know”), in lines 42 and 45, with markedly decreased volume, and he code-switches again to German with a nearly inaudible answer (line 45, sicher, “sure”), which is placed between two long pauses7 (lines 44 and 46). At this critical point in the conversation, with two co-participants involved in an obvious misunderstanding, Eileen self-selects and shifts the topic with a question to Arthur (cómo es para tus padres por ejemplo? “how is it for your parents for example?”). With this strategy, she achieves two goals: (1) she moves away from the problematic talk and relieves Markus of the pressure to respond, as he obviously feels uncomfortable being interrogated, and (2) she also displays “interactional cohesiveness” (Jefferson 1984), that is, the topic shift that Eileen proposes is otherattentive with Arthur. Moreover, she also reveals some information about her family (“for my parents, it is very difficult to live here”), and in doing so, she provides Arthur with a model for an answer, while steering him away from the questioning mode that he had used with Markus before.

5. Discussion and implications In the excerpts examined, the participants appear to have problems actively maintaining turn-by-turn talk in the sequential environment of topic closings. In excerpts (1) and (2), the students opt to allow the topic to fade (rather than collaboratively and smoothly terminating it or shifting topics); in excerpt (3), they maintain the topic (rather than initiating a new topic) even after exceeding topic boundaries. In excerpt (4) – after a misunderstanding – they simply remain silent. Therefore, this analysis reveals the rather passive attitude of (some of) the participants. Thus,

Topic Management and Interactional Competence in Spanish

269

taking the initiative in topic management and changing the course of conversation appear to be complex skills for language learners. The results also allow us to detect different levels of ability in topic management among the participants. Operating at the lowest level are Sebastian and Markus, who only make minimal contributions and do not initiate any topic movement. At a next level of ability are Arthur, Bettina, Bianca and Sylvie, all of whom have no difficulties producing questions or comments, but they dare not attempt more complicated topical actions. Finally, speakers such as Nora and Eileen show an ability that enables them to take the initiative and steer the conversation as they like. Whether these stages represent a fixed order in the acquisition of topic management skills is a question that merit further research. Interestingly, these differences in topic management ability do not match directly with the proficiency level of the participants, as is demonstrated in the cases of Markus, Arthur and Eileen, all of whom were attending the same Spanish A2-level course at the time of the recording. A further finding regarding the relationship between ability in topic management and overall language proficiency is that the linguistic resources needed for successful topic management are not particularly complex. For example, the information that Eileen provides in line 32 in excerpt (3) and her question in line 47 in excerpt (4) are enclosed in linguistic structures that are feasible for all participants to produce; the same occurs in excerpt (1) with Nora’s summary assessment in line 26 and her topic shift in line 28. These findings are consistent with those by König (2013, 245), who reported changes in the ability to introduce topics that affect “mainly the turn architecture,” rather than the linguistic items or structures. What distinguish Eileen and Nora from their co-participants is not, then, the possession of more resources but rather the ability to mobilize them at a particular moment in the interaction, which, in turn, reinforces the conceptualization of IC as competence-in-action as presented in the first section of this paper: a knowledge of resources that cannot be separated from the “ability to use” them (Hymes 1972). Moreover, the special cases of Sebastian and Markus highlight the conceptualization of IC as both a resource and an object for learning (Kasper 2006): those students who lack interactional resources fail to substantively participate, which further prevents them from gaining new learning opportunities. The current study has meant to contribute to the research agenda of understanding how L2 speakers organize their participation in L2 talk (Kasper 2009) and highlights the difficulties in topic management encountered by speakers at different levels of conversational proficiency.

270

Chapter Eleven

These issues are relevant both to second and foreign language instruction and to teacher education.

References Barraja-Rohan, A.-M. 2000. “Teaching conversation and sociocultural norms with conversation analysis.” In Teaching languages, teaching cultures, edited by A. J. Liddicoat, and C. Crozet, 65-78. Melbourne: Language Australia NLLIA. Bublitz, W. 1988. Supportive fellow-speakers and cooperative conversations. Amsterdam: Benjamins. Cheng, W. 2003. Intercultural conversation. Amsterdam: Benjamins. Drew, P. 2005. “Conversation analysis.” In Handbook of language and social interaction, edited by K. L. Fitch, and R. E. Sanders, 71-102. Mahwah, N.J.: Erlbaum. Fazel Lauzon, V., S. Pekarek Doehler, and E. Pochon Berger. 2009. “Identification et observabilité de la compétence d'interaction: le désaccord comme microcosme actionnel.” Bulletin suisse de linguistique appliquée 89: 121-142. Franco Cordón, A. I. 2012. “Control y colaboración en la gestión del tema en una prueba de evaluación oral.” Marcoele 14: 1-22. Gan, Z., C. Davison, and L. Hamp-Lyons. 2009. “Topic negotiation in Peer Group Oral Assessment Situations: A Conversation Analytic Approach.” Applied Linguistics 30 (3): 315-34. García García, M. 2013. “Iniciativa y manejo de los temas en la conversación en ELE.” Revista Nebrija de Lingüística Aplicada 13, número especial. http://www.nebrija.com/revista-linguistica/revistalinguistica-nebrija13/htm/GarciaMarta.htm. —. 2014. La competencia conversacional en español como lengua extranjera: Análisis y enfoque didáctico. Alcalá de Henares: Universidad de Alcalá. Gardner, R. 2004a. “Conversation analysis.” In The handbook of applied linguistics, edited by A. Davies, and C. Elder, 262-84. Malden, Mass.: Blackwell. —. 2004b. “On delaying the answer: Question sequences extended after the question.” In Second language conversations, edited by R. Gardner, and J. Wagner, 246-66. London: Continuum. Hall, J. K. 1995. “(Re)creating our worlds with words: A sociohistorical perspective of face-to-face interaction.” Applied Linguistics 16 (2): 206-32.

Topic Management and Interactional Competence in Spanish

271

Hall, J. K., and S. Pekarek Doehler. 2011. “L2 Interactional Competence and Development.” In L2 interactional competence and development, edited by J. K. Hall, J. Hellermann, and S. Pekarek Doehler, 1-15. Bristol: Multilingual Matters. Hatch, E. 1978. “Discourse analysis and second language acquisition.” In Second language acquisition: A book of readings, edited by E. Hatch, 401-35. Rowley, Mass.: Newbury House. He, A.W., and R. Young. 1998. “Language proficiency interviews: A discourse approach.” In Talking and testing: Discourse approaches to the assessment of oral proficiency, edited by R.Young, and A. W. He, 1-24. Amsterdam: Benjamins. Heritage, J. 1989. “Current developments in conversation analysis.” In Conversation: an Interdisciplinary Perspective, edited by D. Roger, and P. Bull, 21-47. Clevedon: Multilingual Matters. Holt, E. 2010. “The last laugh: Shared laughter and topic termination.” Journal of Pragmatics 42: 1513-1525. House, J. 1996. “Developing pragmatic fluency in English as a foreign language: Routines and metapragmatic awareness.” Studies in Second Language Acquisition (18): 225-52. —. 2002. “Developing pragmatic competence in English as lingua franca.” In Lingua franca communication, edited by K. Knapp, and C. Meierkord, 245-67. Frankfurt: Peter Lang. Howe, M. 1991. “Collaboration on topic change in conversation.” Kansas Working Papers in Linguistics 16: 1-14. Hymes, D. H. 1972. “On communicative competence.” In Sociolinguistics: selected readings, edited by J.B. Pride, and J. Holmes, 269-93. London: Penguin Books. Jefferson, G. 1984. “On stepwise transition from talk about a trouble to inappropriately next-positioned matters.” In Structures of Social Action: Studies in Conversation Anaylisis, edited by A. J. Maxwell, and J. Heritage, 191-222. Cambridge: Cambridge University Press. —. 1993. “Caveat speaker: Preliminary notes on recipient topic-shift implicature.” Research on Language and Social Interaction 26 (1): 130. Kasper, G. 2006. “Beyond Repair: Conversation Analysis as an Approach to SLA.” AILA Review 19: 83-99. —. 2009. “Categories, Context, and Comparison in Conversation Analysis.” In Talk-in-interaction: Multilingual perspectives, edited by Nguyen, H. t. and G. Kasper, 1-28. Honolulu, HI: National Foreign Language Resource Center, University of Hawai’I at Manoa.

272

Chapter Eleven

König, C. 2013. “Topic management in French L2: A longitudinal conversation analytic study.” In Eurosla Yearbook: Volume 13 (2013), edited by L. Roberts, A. Ewert, M.Pawlak, and M. Wrembel, 226-50. Amsterdam: Benjamins. Lazaraton, A. 2002. A qualitative approach to the validation of oral language tests. Cambridge: Cambridge University Press. Levinson, S. C. 1983. Pragmatics. Cambridge: Cambridge University Press. Linell, P., and N. Korolija. 1997. “Coherence in multi-party conversation: Episodes and contexts in interaction.” In Conversation. Cognitive, Communicative and Social Perspectives, edited by T. Givón, 167-205. Amsterdam: Benjamins. Maynard, D. W. 1980. “Placement of topic changes in conversation.” Semiotica 30 (3/4): 263-90. Nguyen, H. t. 2011. “A Longitudinal Microanalysis of a Second Language Learner’s Participation.” In L2 learning as social practice: Conversation-analytic perspectives, edited by G. Pallotti, and J. Wagner, 17-44. Honolulu, HI: National Foreign Language Resource Center, University of Hawai’i at Manoa. Orletti, F. 1989. “Topic organization in conversation.” International Journal of the Sociology of Language 76: 75-85. Pekarek Doehler, S. 2006. “Compétence et langage en action.” Bulletin suisse de linguistique appliquée 84: 9-45. Psathas, G. 1990. “Introduction: Methodological issues and recent developments in the study of naturally occurring interaction.” In Interaction Competence, edited by G. Psathas, 1–29. Washington, D.C. University Press of America. Riggenbach, H. 1991. “Toward an understanding of fluency: A microanalysis of nonnative speaker conversations.” Discourse Processes 14: 423-41. Sacks, H. 1992. Lectures on conversation. Oxford: Blackwell. Schegloff, E. A. 1990. “On the organization of sequences as a source of “Coherence” in talk-in-interaction.” In Conversational organization and its development, edited by B. Dorval, 51-77. Norwood, NJ: Albex. Schiffrin, D. 1988. “Conversation Analysis.” In Linguistics: The Cambridge Survey: Vol. IV: The Socio-Cultural Context, edited by F. J. Newmeyer, 251-76. Cambridge: Cambridge University Press. Seedhouse, P., and A. Harris. 2011. “Topic development in the IELTS Speaking Test.” IELTS Research Reports 12: 55-110. Young, R., and G. B. Halleck. 1998. “‘Let Them Eat Cake!’ or How to avoid losing your head in cross-cultural conversations.” In Talking and

Topic Management and Interactional Competence in Spanish

273

testing: Discourse approaches to the assessment of oral proficiency, edited by R. Young, and A. W. He, 355-82. Amsterdam: Benjamins. Wong, J., and H. Z. Waring. 2010. Conversation Analysis and Second Language Pedagogy. A Guide for ESL/ESF Teachers. New York: Routledge.

Appendix Transcription conventions .hh ? ! CAPITALS : ºno sé° ((clears throat)) (.) [

Inbreath Rising intonation Animated talk Stretch of talk that is louder than the surrounding talk Lengthened sound or syllable Abrupt ending or cutoff Stretch of talk that is quieter than the surrounding talk Nonverbal action Short pauses of less than 0.2 seconds Start of overlapping talk

() -solohe-he *feel

Uncertain transcription Smiley voice Laugh particles Non-standard Spanish word or expression in the original Words in German Non-standard grammatical feature in the original

(=Ger) (*fem.)

Notes 1

In this manner, IC differs radically from the “knowledge for performance” conceptualization posited by House (2002, 263). 2 A summary assessment is “a comment on topic 1 which seems to close off the topic for further discussion. It may also function as a formulation, […] clarifying the central point of the topic […] or stating the consequence of what has been talked about (the upshot). More often, however, it functions as a coda, a concluding remark, after which further conversation relevant to topic 1 is unnecessary.” (Howe 1991, 3) 3 The influence of personalities on individual contributions to conversation cannot be completely excluded in this study; nevertheless, since I was personally acquainted with all of the participants, I can testify that they were talkative and extroverted students who were always willing to speak and participate in the classroom. 4 The text pertained to Paco, a Spanish lorry driver who drives all over Europe.

274 5

Chapter Eleven

Transcription conventions are provided in the Appendix and are based on the guidelines developed by Lazaraton (2002). 6 At the time of the recording, Nora is a 22-year-old young woman. The word Nora is most likely looking for is pequeñas, “children.” 7 “It may be that […] silences are utilized as a resource to focus off the line of talk such that a topic change is invoked to restore continuous speaker transitions. In a sense, […] the production of silences [is the way] to ‘wind down’ a topic” (Maynard 1980, 280).

CHAPTER TWELVE PEER AND TEACHER FEEDBACK IN IMPROVING PRAGMATIC CONVERSATION SKILLS WILLIAM COLLINS

1. Introduction Large university English Communication classes focusing on improving speaking and listening skills present a challenge to both learner and teacher. A key element in improving students’ skill and confidence in their speaking ability is regular feedback. But the large size of the classes makes it difficult for the teacher to provide detailed feedback to students’ on their speaking. Learners’ in my Japanese university EFL setting often report feeling a lack of confidence due to an inability to measure their improvement. The current study reports the results of a one-year study concerning the impact of regular teacher and peer feedback on improving listener participation in pair-recorded story conversations in three university English Communication classes. The research questions in the study were (1) Given explicit instruction in using clarifying and comment strategies, how well would students be able to give peer feedback to other students? (2) Would regular feedback on students’ use of the comment and clarifying strategies in recorded conversations lead to improvements in their use by listeners? If so, which kind of feedback would show the greatest improvement at the end of the semester? (3) What impact would each component of the study have on improving students’ enjoyment of English study and self-confidence in their speaking and listening abilities?

2. Theoretical background The key components in the current study were active listening, interactional feedback and student noticing, intonation and storytelling. Studies concerning each of these are discussed below. The feedback in the

276

Chapter Twelve

current study was given in two ways, real-time pair conversation recording and feedback suggestions inserted on conversation transcripts. While extensive research has been done to substantiate the efficacy of real-time feedback during collaborative listening in promoting L2 development, very little research has been done on giving and receiving feedback on written transcripts. Storytelling was the focus of the recorded conversations in which one partner recounted a personal anecdote and the other listened and responded. These will be called story conversations in the current study. Mackey (2006) examined the role of interactional feedback and student noticing in L2 development. The interactional feedback was provided by the instructor as part of an in-class game show activity and took the form of either recasts or negotiation of meaning. The study found that a significant percentage of the students who reported noticing forms actually went on to develop the ability to produce the target form. Foster and Ohta (2005) investigated peer interaction in recorded conversations and looked at the difference between peer negotiation moves as signals of communication problems and pragmatic moves as signals of interest and encouragement. The study found that when students collaborate to create discourse through scaffolding, helping continue sentences and suggesting ideas to continue conversational flow, “language development might occur as the gap between individual and joint performance is filled and learners develop increased independence” (Ohta 2005, 414). Rost (2002, 143) argued that “collaborative listening, in which learners interact with each other, is established as a vital means of language development.” Rost identified comment strategies including (a) responding – providing a personal, relevant response to information or ideas presented, and (b) inferencing – drawing inferences about incomplete information based on incomplete information. Washburn and Christianson (1996) showed that pair-taping helped students achieve higher fluency and listening comprehension. Their study also stressed the importance of the listener signalling emotional involvement in the conversation by using backchannel cues, asking for and giving clarification, making comments and asking follow-up questions. Wajnryb (2003) argued that storytelling is a valuable tool for languagelearning because it gives students exposure to accessible language, allows them the chance to use the language and, when communicating with other students, a motivation to speak and listen. She also stressed that storytelling gives a natural context for students to receive feedback from both peers and teacher. She proposed recording of story conversations in

Peer and Teacher Feedback in Improving Pragmatic Conversation Skills 277

which students share personal experiences on a given topic and respond as useful for learning in two stages: first, recording in real-time which allows for a focus on fluency, and later transcription and written feedback as a way to address things like grammar, word-choice and intonation. Brazil, Coulthard, and Johns (1980) proposed what they termed an interactional explanation of the significance of intonation. They argued that it is not linguistic considerations that determine tone choice, but “the speaker’s assessment of the relationship between the message and the audience” (Brazil, Coulthard, and Johns 1980, 18). As the speaker continually makes this assessment in the course of a speech, conversation, etc., she bases tone choice on whether to “refer to sections of his message as part of the existing common ground or to proclaim them as an addition to it” (Brazil, Coulthard, and Johns 1980, 18). Wells (2006) discussed the significance of intonation in signalling degree of emotional involvement on the part of the listener. He argues that there are two special tone meanings associated with rise-fall intonation. One signals that the speaker is impressed, the other is a challenge. Bachorowski (1999) explored the acoustic properties of emotional speech and finds that “listeners are able to accurately judge emotions from speech at rates far greater than chance” (Bachorowski 1999, 53). Collins and Ruhl (2008) explored the impact of pair-recording and active listening on students’ enjoyment of and confidence in their English conversations. Students in the study reported that pair-recording and active-listening helped them enjoy English more and improved their conversations. Collins (2013) investigated the impact of two reflective listening exercises on improving students’ use of the active-listening strategies under different degrees of pre-planning in a recorded speakingexam. Sinclair and Coulthard (1975) first proposed the Birmingham Model for analyzing classroom interaction. In this framework, classroom discourse was divided into 5 ranks: lesson, transaction, exchange, move and act. The three types of exchanges were eliciting, informing and directing and there were three parts or moves to an exchange: initiation, response and follow-up. Finally, moves were further subdivided into acts, the smallest unit of spoken discourse. In a classroom exchange from Brazil (1995, 17) dealing with pyramids, The teacher’s opening move was: “They were pharaohs./ Erm do you know anything about them? / They were great for building something you make in math.” The student answered “pyramids” and the teacher acknowledged the correct answer and pursed the topic further with another student. Each teacher and student turn is called a “move,” the units

278

Chapter Twelve

within slashes are called “acts,” each three-turn unit (question /answer/ follow-up) is called an “exchange” and the continuation of this sequence about pharaohs and pyramids until the teacher moves on to another segment of the lesson is called a “transaction.” Francis and Hunston (1992) modified this framework so that it was “flexible and adaptable enough to cope with a wide variety of discourse situations (including) casual conversations between friends” (Francis and Hunston 1992, 123). The ranks in Francis and Hunston’s (1992) framework were: interaction, transaction, exchange, move and act. Collectively the studies suggest that collaborative listening and interactive feedback, particularly during real-time conversations, contribute to learner’s L2 development. The Wells and Bachorowsky studies bolster the idea that intonation is important in signalling emotional involvement and empathy, and the Washburn and Christianson and Collins and Ruhl studies suggest pair-recording and active listening increase students’ enjoyment and motivation and can improve fluency and listening comprehension.

3. Participants, materials and methods The study was conducted with three classes of pharmacy majors over one year in a first-year English Communication class. Forty students participated in the teacher-feedback group, forty-two in the peer-feedback group, and forty-three in the no-feedback group. All of the students were Japanese EFL learners. The students’ proficiency level ranged from intermediate to high-intermediate to advanced. The students were given the Penguin Readers Placement Tests and were placed at level 5 and 6, with level 6 being the highest. The three student populations were of comparable age, ethnic and linguistic background. Data collected in the study was taken from the following sources: (a) four story-conversation recordings, conducted at three-week intervals over the semester; (b) student-transcriptions of the conversations; (c) teacher feedback comments and (d) student feedback comments; and (d) student surveys concerning improvements in motivation and self-confidence. Students were given explicit instruction in using a set of active-listening comment and intonation strategies to signal emotional involvement and empathy. Three classes participated in the study and in all three classes, students regularly recorded their conversations. The three classes were given, respectively, peer-feedback, teacher-feedback and no feedback concerning their use of the active-listening and intonation strategies.

Peer and Teacher Feedback in Improving Pragmatic Conversation Skills 279

The study employs the Francis and Hunston (1992) model of spoken discourse analysis, which modified the 5 ranks of the Birmingham Model to include non-classroom conversational interaction, and Brazil, Coulthard, and Johns’s (1980) intonation model, which posited an interactional explanation of the significance of intonation, to measure improvements in listener participation in recorded story conversations.

3.1 Active-listening strategies and feedback criteria The evaluation of improvement over the course of the four recordings and the content of teacher and peer feedback focused on the students’ ability to use the active-listening strategies and signal empathy with intonation. The following paragraphs introduce the target active-listening strategies that were the focus of instruction, and the guidelines provided for peer-feedback. Samples of both peer and teacher feedback are also presented. 3.1.1 Active-listening strategies The strategies were taught and practiced in class, and consisted of “basic” and “advanced” strategies (Collins and Ruhl 2008). The basic strategies included backchanneling (“Oh yeah?”/“Oh really?”/“Uh-huh.”), comments/rejoinders (“That’s + adjective”/“Wow!”/“No Way!”/“Oh no!”), and clarifying cues (repeating an unfamiliar word or phrase). The advanced strategies consisted of personalizing, speculating and generalizing, examples of which are shown in Example 1 below. The distinction between strategies as basic or advanced was based on degree of cognitive processing required of the listener to form and produce the response in the real-time flow of the conversation. The role of listener here refers to the second, responding participant in the conversation. With the basic strategies, it was sufficient for the listener to use either emotionally neutral keep-the-conversation-going acts (such as “uh-huh,” “yeah,” “right” and so on), or to recognize the personal experience being recounted by the speaker as a happy, sad, or surprising one and respond accordingly (such as “That’s great” or “Oh no” or “No way!” and so on.). With the advanced-strategies, greater cognitive-processing was required as the listener had to draw a comparison between the speaker’s remark and his own experience, or make a general observation about the kind of experience referenced by the speaker.

280

Chapter Twelve

EXAMPLE 1: Advanced active-listening strategies Interlocutor’s Comment: “I recently took my driving test.” Personalizing [Oh, I-----------too.] Oh, I took my test too./Oh, I’m taking my test soon too. [Oh, I---------but---------] Oh, I took my test, but I haven’t passed yet. [Oh really? (In my case)---------] Oh really? I passed my test last month! Generalizing [(doing----)] is –(adjective), isn’t it? Taking the driving test is stressful, isn’t it? [(Noun)] is –(adjective), isn’t it? The driving test is difficult, isn’t it? It’s –(adjective) [(to do---/doing----)] isn’t it? It’s hard to pass the driving test, isn’t it? [I think a lot of people----] I think a lot people take the test their freshman year. [I’ve heard that----] I’ve heard that the driving test is more difficult than the written test. Speculating [I guess ---- ] I guess it was difficult. / I guess you were nervous. [I bet ---- ] I bet that was stressful. / I bet you were nervous. [----must have been----/ [done]---)] That must have been stressful. / You must have been nervous.

3.1.2 Story Guidelines At the beginning of the semester, students were introduced to a selection of 50 story themes. Themes were chosen that dealt with change, character-building, taking risks, and dealing with challenges or disappointment. Themes also dealt with emotional experiences, such as funny, scary, sad or moving experiences. Finally, to ensure that the speaker’s story would be long enough to promote fluency building and encourage regular responses and involvement of the listener, students were instructed to make their stories at least 2 minutes. In practice, the length of the 40 students’ story conversations over the four recordings was around 2 minutes. Examples of story topics are shown in Appendix 1.

Peer and Teacher Feedback in Improving Pragmatic Conversation Skills 281

3.1.3 Evaluating improvement: regular story-conversation pair recording .

Four conversation recordings were made respectively week 3, 6, 9 and 12 of the 16 week semester. After each recording, the students who took part in the recorded conversation transcribed it. In the peer-feedback and teacher-feedback groups these transcriptions were used to give feedback to the students who recorded the conversations. 3.1.3.1 Student feedback In the peer-feedback group, the students regularly exchanged conversation transcriptions and examined listener participation in the conversation. The students when giving peer-feedback were instructed to consider three aspects of listener-participation in their feedback: (a) did the intonation of the listener’s active-listening responses match the emotional content of the speaker’s story? (For example, if the speaker’s move expressed something exciting, surprising, happy, scary or sad, did the listener’s intonation signal empathy?); (b) Did the listener give activelistening responses suitable to the emotional content of the story? (c) Did the listener give enough comment and advanced active-listening responses to show their understanding and signal emotional involvement? Students were instructed to make these judgments by imagining they were listening and responding to the speaker’s remarks themselves. Students wrote or typed on the transcript suggested responses for the listener. Students were not instructed where or how often to give feedback. Example 2 shows one instance of student feedback. The bracketed words are peer-feedback suggestions, while bracketed bold words are praise. In every case of peersuggested feedback, there had been no listener response in the original recorded conversation. In this example, A is the speaker who recounts the story, and B is the listener who responds with active listening. The peer feedback was given later by a student who had not participated in the conversation and was thus hearing the speaker’s story and the listener’s responses for the first time.

282

Chapter Twelve

EXAMPLE 2: Student Feedback Sample A: Hi Yuu! B: Hi Kouta! A: I would like to tell you about a great memory I had. B: Oh, please tell me. A: It was when I was a young child. When I was young, I like all sports, but especially baseball. [B: Oh yeah? I liked baseball my favorite too. Or My favorite sport was---] A: I liked watching baseball game so much. B: Yeah. [intonation up for question] A: When I was a first grade student of elementary school, my father took me to my first game. I was so excited. [B: Yes, going to games is really exciting] A: I often went watching high school baseball games and cheered some teams. B: Great. A: I had a favorite pitcher in other high school. [B: Really? Or You did?] A: He was so tall and cool on the mound. And his ball was so great. He was famous for his arm in Nagasaki. [B: Wow. I guess he was a great player] A: I sometimes saw him in stadiums when I went watching baseball games, but there was great distance between the stands and the mound. B: I see. A: So I had never seen him close up. B: Yeah. [That’s too bad.(intonation down)] A: One day, when I got into a train, I found that someone tall was standing behind me. [B: Someone tall?] A: It was he! B: Really? [Good intonation So, how did you feel? ] A: I was so excited and felt my heart beating fast. B: Wow! A: Then, a friend of mine who was with me spoke to him, and told him that I was his fan. B: He did? Your friend must be kind. A: Yeah. Hearing that, he smiled and he shook hands with me. B: Sounds great! A: I was so happy and almost moved to tears. It was a very good memory.

3.1.3.2 Teacher feedback Teacher feedback was given to the second student-group. Teacherfeedback differed from peer-feedback in that suggestions on specific wording (such as “That’s good” or “How did it go?”) were not given, but only suggestions of feedback type (such as “rejoinder” or “question”).

Peer and Teacher Feedback in Improving Pragmatic Conversation Skills 283

Teacher feedback suggestions for advanced active listening was given in the form of sentence heads (“That must---” or “I guess---”) rather than the entire wording. In cases where two suggestions are separated by a slash, this indicates a choice of two feedback suggestions. Example 3 shows a sample of teacher feedback: EXAMPLE 3: Teacher Feedback Sample A: I tell you about my scary experience. B: Oh yeah. [IntonationВ] A: I used to go to cram-school and I walked home. B: Walked home? A: The school was far so I had to walk a big distance. B: Uh-huh. [Question/That sounds----] A: It was at night time. B: Oh I see. A: So one night I left my cram-school to walk home. B: Uh-huh. A: First part was many stores and cars and people— B: Cars and people? A: Yes, so I felt comfortable with many people. B: Uh-huh. A: But last part was in dark area with no car or people. B: Ohh. [Sounds--/That must---] A: Then I saw strange man coming from on the street. B: Strange man? Ohh. It’s scary. A: Yes. So I walked a faster. B: Yeah? [Good intonation] A: Then I heard talking. B: Talking? A: He was talking by himself. B: That’s strange. [Good intonation] A: Yes. I thought he is crazy. [Pause] B: Oh yeah. [You must have been---] A: I decided to run home. B: Run home? A: I heard the man was running too! B: No way! [Good intonation] A: Then I was at my house. B: That’s good! A: I went inside and closed the door. B: That was very scary for you.[Good] A: Yes. Thank you for listening. B: Thank you.

284

Chapter Twelve

3.2. Measuring listener participation To measure improvements in students’ awareness of and ability to use the active-listening and intonation strategies, the students’ peer-feedback, conversation-transcriptions and audio recordings were analyzed. The parameters measuring use of active-listening strategies included all of the following: ratio of listener backchanneling acts to comment acts, ratio of basic comment to advanced comment acts and mean word count of comment acts. The subdivision of basic active-listening strategies into backchanneling (“Yeah,” “Uh-huh,” “Right,” etc.) and comment acts (“That’s great,” “that’s difficult,” “that sounds scary,” etc.) was based on the idea that comments required greater listener-engagement with the content of the speaker’s talk than backchanneling. This is because appropriate use of the comment strategies required a certain degree of listener comprehension of the speaker’s story, so that an appropriate use of, for example, “That’s scary,” or “That’s too bad” signalled the speaker that the sad or scary content of her story had been successfully conveyed to the listener; backchannelling cues such as “Yeah,” or “uh-huh,” on the other hand, could be used without the listener actually understanding what they speaker was saying. Advanced comments (such as “I guess your friends went to a lot of trouble to prepare for the surprise party” or “It’s difficult when you have trouble understanding your teacher”) were more detailed and specific than any of the basic comments and thus required greater listener comprehension and signalled more active involvement to the speaker. In addition, as a way to measure effectiveness of listeners’ use of intonation to signal empathy, ratio of emotional speaker-utterances to suitable intonation response acts were measured. Emotional utterances were those that met two criteria: (1) they referenced personal experiences that were strongly emotional, i.e., happy, sad, surprising or admirable events, (see Example 5), and (2) the speaker signalled the emotional nature of the event by giving heightened stress and rising (for happy/surprise) or falling (sad/disappointing) intonation. Suitable intonation response acts were those responses whose tonal-contours signalled listener acknowledgement of the emotional tenor of the events just referenced by the speaker. Example 4 shows examples taken from the student data. Words with no intonation marker over them were spoken with a level tone.

Peer and Teacher Feedback in Improving Pragmatic Conversation Skills 285 EXAMPLE 4: Examples of Suitable Intonation Responses Speaker Event I had to practice my piano two hours every day. I read the paper and it said I passed (the entrance exam)!

Then the teacher hit me.

Listener Response ВГ В ГВГ Wow. That’s a lot! ВГ ВГ Great! You must have been very ВГ excited. ВВ ВГ He hit you? That’s terrible!

3.3 Measuring students’ improved awareness of passive listener participation Another factor examined in the data was how well students identified listener-omissions in peers’ conversations and how suitable their suggestions were. This notion of listener omission was introduced to raise students’ awareness of the greater need to signal listener empathy or interest with certain kinds of story content than with others. Given that these judgments are often more subjective than those about grammar, a more appropriate term than omission is perhaps “passive listener involvement,” This was determined by teacher review of the student conversation transcriptions, identifying instances of passive listener involvement in the conversation, and comparing teacher judgments with peer-feedback protocols. Two kinds of passive listener participation were noted: (1) the listener said nothing when a response was clearly called for or (2) listener response suggested a lack of listener empathy or interest (i.e., listener responses were limited to tonally flat backchanneling such as a tonally flat “yeah,” “right” or “uh-huh,” despite the speaker’s previous remark referencing an emotional turn of events in the story and signalling with a marked rise in intonation. Instances of each from the student conversations are shown in Example 5. The emotional tenor of the turn of events is shown underlined in brackets.

286

Chapter Twelve

EXAMPLE 5: Passive Listener Involvement in Student Data Speaker’s Emotional Utterance “But then when I found my dog on the street. She was hit by a car.” [sad/shock] “By the end of the day I had lost 50000 yen!” (at Pachinko)!” (rising intonation, strong accent on “50000 yen”)….[pause] [upset/disappointment] When I went into my room many friends said “Surprise!” happily! [happy/surprise] “Finally I won the first prize for the speech contest.” [happy/satisfying]

Passive listener response (no response) Right (tonally flat)

(no response) Uh-huh (tonally flat)

This teacher-review process yielded data on the ratio of listener feedback-omissions to suitable peer-feedback suggestions (see Table 1). For example, if there were five instances of listener-omission in the conversation and the peer-feedback identified and suggested suitable feedback responses for all 5, the ratio for that conversation would be 5:5. If the teacher identified five listener-omissions but the student only recognized and gave suitable feedback for four, the ratio would be 5:4. If a student correctly identified all five omissions but one feedback suggestions were unsuitable, this would be calculated as 5:4.5. Feedback suggestions were deemed unsuitable for a variety of reasons. Sample instances of student feedback-suggestions deemed unsuitable are shown in Example 6: EXAMPLE 6: Samples of Unsuitable Feedback Suggestions A:I was a member of the table-tennis club in high-school B: Many people do so. [vague, need to specify “do”] A: I took the math test, but I failed the test. B: I guess you are bad student. [non-empathic, judgmental] A: Last Saturday I went to the shopping mall B: No way! [strong surprised response not suitable] A: I went to a movie last weekend. B: That’s great! [strong positive response not suitable].

4. Findings Table 1 shows the ratio of passive listener participation to suitable peer-feedback suggestions for the four sets of peer-feedback. The aim of collecting this data, which was done only for the peer-feedback group, was

Peer and Teacher Feedback in Improving Pragmatic Conversation Skills 287

two-fold: (1) to determine how well students understood when and how to use the active-listening and intonation strategies being learned in class at each stage in the study, and (2) to ascertain the quality of feedback students were receiving from peers. All figures are group means, calculated as the average count of each measurement (instances of passive listener participation, suitable feedback suggestions, unsuitable listener intonation and suitable intonation suggestions) for each week’s set of recordings, and the standard deviations for each mean are shown under each ratio. The standard deviations for the left number in each ratio are shown immediately below the ratio and the standard deviation for the right number in the ratio is shown two places below the ratio. TABLE 1: Suitability of Peer-Feedback on Pragmatic Skills

Active-Listening Suggestions (Ratio of passive listener participation to suitable feedback suggestions) Intonation Suggestions (Ratio of unsuitable listener intonation to suitable intonation suggestions)

1st (Week 3) 9:3 1.231: 0.960

2nd (Wk 6) 7:4 1.109: 0.7162

3rd (Wk 9) 6:5 0.5749: 0.6991

4th (Wk 12) 3:3 0.5991 1.0989

5:1 0.952: 0.534

4:2 0.752: 0.655

3:2 0.611 0.574

2:2 0.602: 0.563

4.1 Four conversation recordings The week three recording was done immediately after the basic activelistening strategies had been introduced, but before the two feedback groups had received any feedback. The week six recording was after students had received peer and teacher feedback concerning use of the basic strategies. The ninth week recordings were done immediately after the advanced active-listening strategies had been introduced. The final recordings were done after students had received peer and teacher feedback concerning use of the advanced strategies. Table 2 shows the increase in listener participation in four recordings for the teacherfeedback group. Each ratio was calculated by taking the mean for all listener ratios in that group for that recording. The standard deviation for each mean is shown below it. “BC” denotes backchanneling. (See appendix 2 for an example of a coded transcription.)

288

Chapter Twelve

TABLE 2: Increase in Listener-Participation (Teacher-Feedback Group)

Listener BC to comment acts ratio Basic to advanced comment acts ratio Comment Act Mean Word-count Ratio of emotional speakerutterances to accented intonation response acts

1st (Week 3) 7:2 0.812: 0.766 2:0 0.693: 0 2.1 0.648 1:1 0.485 0.479

2nd (Wk 6) 10:2 0.823: 0.634 2:0 0.693: 0 2.6 0.775 2:1 0.656: 0.547

3rd (Wk 9) 11:6 0.647: 0.553 4:2 0.747: 0.663 3.6 0.897 3:2 0.664: 0.622

4th (Wk 12) 12:8 1.084 0.787 5:3 0.671: 0.621 3.9 0.844 3:3 0.685: 0.583

Table 3 shows the increase in listener participation in four recordings for the peer-feedback group. TABLE 3: Increase in Listener-Participation (Peer-Feedback Group)

Listener BC to comment acts ratio Basic to advanced comment acts ratio Comment Act Mean Word-count Ratio of emotional speakerutterances to accented intonation response acts

1st (Week 3) 6:2 0.764: 0.537 2:0 0.468: 0 2.2 0.563: 1:1 0.446: 0.437

2nd (Wk 6) 9:2 0.655: 0.578 2:0 0.641: 0 2.8 0.779: 2:2 0.521: 0.497

3rd (Wk 9) 12:7 0.622: 0.724 5:2 0.741: 0.522 3.8 1.06 3:3 0.615: 0.563

4th (Wk 12) 13:9 0.684: 0.733 5:4 0.826: 0.646 4.2 0.873 3:3 0.764: 0.621

Table 4 shows the increase in listener participation in four recordings for the no-feedback group.

Peer and Teacher Feedback in Improving Pragmatic Conversation Skills 289

TABLE 4: Increase in Listener-Participation (No Feedback Group)

Listener BC to comment acts ratio Basic to advanced comment acts ratio Comment Act Mean Word-count Ratio of emotional speakerutterances to accented intonation response acts

1st (Week 3) 7:1 0.647: 0.432 1:0 0.563: 0 2.0 0.655 1:1 0.521: 0.507

2nd (Wk 6) 8:1 0.943: 0.511 1:0 0.554: 0 2.1 0.638 2:1 0.644: 0.592

3rd (Wk 9) 9:2 0.788: 0.621 1:1 0.633: 0.589 2.7 0.955 3:1 0.744: 0.604

4th (Wk 12) 10:3 1.272: 0.644 2:1 0.755: 0.624 2.6 0.844 3:2 0.763: 0.727

Analysis of student-recordings and transcriptions showed differing degrees of improvement in the four measurements of listener-participation across the three groups of students. The findings for the first recording, conducted before any feedback had been given for the two feedbackgroups, were quite similar for the three groups: between eight and nine listener acts, and between six and seven backchanneling acts; between one and two basic comment acts, and no advanced comment acts; between 2.0 and 2.2 mean words for comment acts; and 1 emotional speaker-utterances to 1 accented intonation response act. The subsequent three recordings showed much larger growth in listener-participation in the two feedback groups than in the largely static results for the no-feedback group, and the growth was largest in the peer-feedback group. The ratio of listener backchanneling to comment acts grew from 7:2 to 12:7 in the teacherfeedback group, 6:2 to 13:9 in the peer-feedback group, and 7:1 to 10:3 in the no-feedback group. The growth in listener use of comment acts was particularly static in the no-feedback group. The mean word-length of comment acts grew from 2.1 to 3.9 in the teacher group, 2.2 to 4.2 in the peer group, and 2.0 to 2.6 in the no-feedback group. The ratio of listener basic to advanced comment acts grew from 2:0 to 5:3 in the teacherfeedback group, 2:0 to 5:4 in the peer-feedback group, and 1:0 to 2:1 in the no-feedback group. The ratio of emotional speaker-utterances to accented intonation response acts was the least variable indicator between groups, with 1:1 to 3:3 in the teacher and peer group, and 1:1 to 3:2 in the nofeedback group.

Chapter Twelve

290

4.2 Student survey At the end of the semester, the author conducted a student survey to ascertain the impact the key components of the study had on students' enjoyment of English and confidence in their speaking and listening ability. The four components were (1) recording their conversations; (2) Using active listening to respond in real-time recorded conversations; (3) receiving written feedback suggestions (Teacher and peer-feedback groups only); (4) giving written feedback suggestions (peer-feedback group only). These are indicated in Table 5 by the bracketed number in the “study component” row. Likert-scale questions, the most widely used scale in survey research (Brown, 2001), were used. In the survey, respondents were asked to indicate the degree of their agreement with a set of statements concerning each of the four key areas. Students circled 1, 2, 3, or 4 with the descriptors strongly agree, moderately agree, moderately disagree, and strongly disagree. An even number of options was given to prevent students from choosing a “safe,” neutral option. Convenience sampling procedures were adopted (see Dornyei, 2003). Each statement was written as “Recording conversations with a partner helped me enjoy English more” and “Recording conversations with a partner improved my confidence in my speaking and listening ability” and so on. It was explained to students that “enjoy English more” means more than their experience of learning English prior to the study. The survey results for the teacher-feedback group are shown in Table 5. N/A indicates that this component was not part of the study in that group. TABLE 5: Student Survey Results Concerning Study Components (Teacher-Feedback) Survey Statement Study Component Strongly Agree Moderately Agree Moderately Disagree Strongly Disagree MEAN Standard Deviation

Helped Enjoy English More (1)

(2)

(3)

(4)

Improved Speaking and Listening Confidence (1) (2) (3) (4)

55%

40%

42%

N/A

60%

55%

38%

N/A

43%

58%

48%

N/A

40%

45%

51%

N/A

2%

2%

10%

N/A

0%

0%

11%

N/A

0%

0%

0%

N/A

0%

0.4%

0%

N/A

3.55 0.504

3.38 0.54

3.3 0.648

N/A N/A

3.60 0.496

3.55 0.503

3.28 0.64

N/A N/A

Peer and Teacher Feedback in Improving Pragmatic Conversation Skills 291

Table 6 shows the survey results for the peer-feedback group. TABLE 6: Student Survey Results Concerning Study Components (Peer-Feedback) Survey Statement Study Component Strongly Agree Moderately Agree Moderately Disagree Strongly Disagree MEAN Standard Deviation

Helped Enjoy English More (1)

(2)

(3)

(4)

Improved Speaking and Listening Confidence (1) (2) (3) (4)

58%

50%

46%

45%

64%

57%

48%

51%

42%

48%

50%

51%

36%

43%

50%

49%

0%

2%

4%

4%

0%

0%

2%

0%

0%

0%

0%

0%

0%

0.4%

0%

0%

3.6 0.496

3.48 0.554

3.45 0.597

3.5 0.555

3.65 0.483

3.58 0.501

3.48 0.554

3.52 0.506

Table 7 shows the survey results for the no-feedback group. TABLE 7: Student Survey Results Concerning Study Components (NoFeedback) Survey Statement Study Component Strongly Agree Moderately Agree Moderately Disagree Strongly Disagree MEAN Standard Deviation

Helped Enjoy English More (1)

(2)

(3)

(4)

Improved Speaking and Listening Confidence (1) (2) (3) (4)

38%

36%

N/A

N/A

40%

37%

N/A

N/A

52%

51%

N/A

N/A

53%

55%

N/A

N/A

10%

13%

N/A

N/A

7%

8%

N/A

N/A

0%

0%

N/A

N/A

0%

0.4%

N/A

N/A

3.2 0.603

3.12 0.634

N/A N/A

N/A N/A

3.28 0.572

3.23 0.554

N/A N/A

N/A N/A

All three groups reported a positive impact for the first two components, recording their conversations and using the active-listening

292

Chapter Twelve

strategies, on both their enjoyment of English study and their confidence in their speaking and listening ability. For the statement “recording conversations helped me enjoy English study more,” the three groups reported, respectively, 98%, 100% and 90% (combining percentage who chose either “strongly agree” or “moderately agree”). The groups also reported that story recording had a positive impact on their confidence (100%, 100% and 93% respectively). For the active-listening strategies, the students reported positive impact on both enjoyment (98%, 98% and 87%) and confidence (100%, 100% and 92%). For receiving feedback, the teacher-feedback group reported a positive impact on both enjoyment (90%) and confidence (89%). For the peer-feedback group the combined strongly agree and agree results were 96% and 98% respectively. Finally, for giving feedback, the peer-feedback group reported a positive impact on both enjoyment (96%) and confidence (100%).

5. Discussion At the outset of the paper three research questions were posed. The first question concerned how well students given instruction in the activelistening strategies would be able to give suitable feedback to classmates regarding ways to improve listener participation in the conversation. Analysis of the peer feedback written on conversation transcriptions for each of the four recordings revealed the incidence of passive listener participation decreased over the four recordings, and the rate of suitable feedback increased. This finding suggests that over time, peers were better able to recognize listener-omissions and offer suitable suggestions, and that listeners were becoming more active participants in the conversations. These results echo Mackey’s (2006) findings on the impact of noticing on subsequent ability to use target forms as well as Foster and Ohta’s (2005) findings on the benefits of pair collaboration on lifting subsequent individual performance. The second research question was whether and which kind of feedback would lead to improvements in listeners’ use of the strategies. The two classes that received regular feedback showed much greater improvement over the course of the four recordings in the use of the strategies, with greater use of comment acts relative to backchanneling and advanced comment acts relative to basic. The mean word count of comment acts also increased. Of the two types of feedback, the class that received and gave regular peer-feedback showed a slightly greater use of advanced comment, but the differences between the two groups was not statistically significant. There was also improvement over the course of the four recordings in students’ appropriate use of intonation

Peer and Teacher Feedback in Improving Pragmatic Conversation Skills 293

in all three groups, with no difference between the two feedback groups and the no feedback group. Student survey reported positive impact on their enjoyment of English and their confidence in their speaking and listening in all three groups, with somewhat lower levels in the no feedback groups. Students also reported very positive impact of receiving and giving feedback on their enjoyment and confidence.

6. Conclusion The findings in the current study suggest that the two feedback-groups’ ability to use the active-listening strategies in recorded conversations improved after receiving regular feedback. The marked improvement in students’ use of the strategies between the first pre-feedback recorded conversation and the other three conversations in the two feedback-groups was not seen in the no-feedback group’s conversations. In the three recordings post-feedback recordings, listener-participation as measured by listener BC to comment acts ratio, basic to advanced comment acts ratio, and comment act mean word-length, all increased markedly in the two feedback-groups, but remained relatively static in the no-feedback group. The one skill where this disparity was not seen was intonation which, measured by ratio of emotional speaker-utterances to accented intonation response acts, was similar in all three student-groups. Student surveys suggested that giving peer-feedback was just as beneficial as receiving it in enabling the students in the peer-feedback group to improve their proficiency with the strategies in successive recorded conversations. There was no very little difference in the degree of improvement for the peer and teacher feedback groups. There are a number of limitations to the current study. The criteria for determining what constituted a gap in listener-response was imprecise as it was based on a subjective intuition of where in the conversation a signal of listener empathy was called for. To more firmly establish this criterion, future research should examine inter-rater reliability of this construct. The relatively small student population and their high proficiency-level make it difficult to generalize the positive impact of peer and teacher-feedback for improving listener-participation to lower-proficiency students as well. Future research should test the findings with a larger student-group. Further, research into cultural differences in use of intonation to signal emotional involvement and empathy in English and the learner’s first language would be useful in establishing whether students from certain language-backgrounds might have more difficulty in learning to use intonation in English than those from others. Also, students’ perceptions

294

Chapter Twelve

of how to identify where a listener response was called for need to be examined to establish a greater empirical basis for what constitutes a listener-omission. These findings bolster Washburn and Christianson’s (1996) argument concerning the importance of the listener signalling emotional involvement. Finally, larger samples of transcribed student conversations might prove fruitful in developing a comprehensive corpus of EFL learner language so that patterns in speaker and listener interaction could be studied in greater depth.

References Bachorowski, J-A. 1999. “Vocal expression and perception of emotion.” Current Directions in Psychological Science 8: 53-57. Brazil, D., M. Coulthard, and C. Johns. 1980. Discourse Intonation and Language Teaching. London: Longman. Brazil, D. 1995. Classroom and Spoken Discourse. Birmingham: University of Birmingham Press. Brown, J. D. 2001. Using surveys in language programs. Cambridge: Cambridge University Press. Collins, W., and D. M. Ruhl. 2008. “Speaking and listening skills through storytelling, talking journals, and active listening.” In JALT 2007 Conference Proceedings, edited by K. Bradford Watts, T. Muller, and M. Swanson, 598-612. Tokyo: JALT. Collins, W. 2013. “Testing interactional English conversation skills in a university speaking exam.” In JALT 2012 Conference Proceedings, edited by N. Sonda, and A. Krause, 670-683. Tokyo: JALT. Dörnyei, Z. 2003. Questionnaires in second language research. London: Lawrence Erlbaum Associates. Foster, P., and A. S. Ohta. 2005. "Negotiation for meaning and peer assistance in second language classrooms." Applied Linguistics 26 (3): 402-430. Francis, G., and S. Hunston. 1992. “Analyzing everyday conversation.” In Advances in Spoken Discourse Analysis, edited by M. Coulthard, 123161. London: Routledge. Mackey, A. 2006. "Feedback, noticing and instructed second language learning." Applied Linguistics 27 (3): 405-430. Rost, M. 2002. Teaching and researching listening. Harlow, UK: Pearson Education. Sinclair, J. McH., and M. Coulthard. 1975. Towards an Analysis of Discourse. Oxford: Oxford University Press. Wajnryb, R. 2003. Stories. Cambridge: Cambridge University Press.

Peer and Teacher Feedback in Improving Pragmatic Conversation Skills 295

Washburn, N., and K. Christianson. 1996. “Teaching conversation strategies through pair-taping.” The Internet TESL Journal, iteslj.org/Techniques/Christianson-PairTaping.html. Wells, J. C. 2006. English Intonation. Cambridge: Cambridge University Press.

Appendix 1 1. 2. 3. 4. 5.

Story Conversation Topics Sampling Tell me about a time you had a big change in your life A memory or experience that meant a lot to you A scary experience you had A time you pushed yourself to do something you didn’t think you could do A time you thought “I know I shouldn’t do this but…”

Appendix 2 Coded Student Transcription* A Today I’d like to talk about a club when I was joined at high school. B Uh-huh. 1 [1] (BC act) (BC = backchanneling) A I was joined orchestra club at high school. B Wow. 2 [1] (BC) Orchestra? 3 [1] (BC) That sounds cool. 4 [3] (comment act) A Yes. I was playing violin there, I was concertmaster who conduct the orchestra instead of conductor. B Wow. 5 [1] (BC) Being conductor is difficult, isn’t it? 6 [6] (comment act) A A little bit difficult. B I guess you are the main person of the club. 7 [10] (comment act) A Thank you. We often volunteered at hospitals and kindergardens B Sounds like a nice activity. 8 [5] (comment act) A Thank you. Also we were give much energy from children. B I enjoy being with children. 9 [5] (comment act) A Yes, me too. And we were told many interesting topics from old people. B It must be very stimulating. 10 [5] (comment act) A Yes, it’s good. One thing I learned from that experience is that volunteer is very profitable for each other. B I think it’s good to volunteer. 11 [6] (comment act) A Thank you. We should do more volunteer. B I think so too. 12 [4] (comment act) A I want to volunteer at the university too. B Oh, it’s very nice. 13 [4] (BC) A Thank you. Thank you for listening. B Thank you. *Number of act [number of words] (type of act)

CHAPTER THIRTEEN TOWARDS A PEDAGOGICAL FRAMEWORK TO DEVELOP THE LISTENERSHIP OF JAPANESE EFL/ESL LEARNERS PINO CUTRONE

1. Introduction In the modern age of globalization, more people around the world are studying EFL/ESL to communicate across borders in such fields as international politics, academia, business and science. Perhaps no nation has expended greater resources encouraging its citizens to study English than Japan. Despite the great emphasis put on the Japanese to learn English, the results to date have been largely disappointing (Nikolova 2008), particularly concerning oral skills (Roger 2008). A key aspect of effective oral communication involes what McCarthy (2002, 2003) calls listenership, namely, being able to give effective listener feedback to one’s interlocutor (Farr 2003; O’Keeffe, McCarthy, and Carter 2007), and this is a specific area in which Japanese EFL/ESL English speakers have experienced problems (Cutrone 2005, 2010, 2013). It is becoming increasingly clear that what constitutes effective feedback seems open to interpretation, and there is potential for intercultural pragmatic failure and misunderstanding when listening styles differ. In an attempt to inform language pedagogy in Japanese EFL/ESL contexts, the main aim of this article is twofold: to demonstrate why this feature of language warrants more attention in the language classroom, and to offer a pedagogical framework for the analysis of this aspect of Japanese EFL/ESL speakers’ English.

298

Chapter Thirteen

2. What is listenership? There exist many terms to describe what is meant by listenership (Fujimoto 2007 lists 24 terms); however, the one most commonly associated with listening behaviour in the research literature remains the term coined by Yngve (1970), backchannel, which he describes as follows: When two people are engaged in conversation, they generally take turns... In fact, both the person who has the turn and his partner is [sic] simultaneously engaged in both speaking and listening. This is because of the existence of what I call the backchannel, over which the person who has the turn receives short messages such as yes and un-huh without relinquishing the turn. (Yngve 1970, 568).

While some researchers such as Oreström (1983) have followed Yngve’s (1970) definition of backchannels in their studies, others have broadened it. Most notably, Duncan (1974) and Duncan and Fiske (1977) extend backchannels to include sentence completions, requests for clarification, brief statements, and non-verbal responses such as head nods and headshakes. In various studies involving Japanese participants specifically, Cutrone (2005), Maynard (1997) and White (1989) differed slightly in their identification of backchannels. White (1989), focusing solely on non-word vocalisations such as mmhm, yeah, uh-huh, oh, and hmm, limited her analysis to audio recordings and thus did not include non-verbal behaviour. Cutrone (2005) and Maynard (1986, 1987, 1990, 1997), on the other hand, used a broader identification of backchannels as proposed by Duncan and Fiske (1977), in that they too included sentence completions, requests for clarification, brief statements, and non-verbal items such as head nods and laughing. Hence, in this paper, the term backchannel refers to the following: minimal non-word vocalisations, brief utterances and non-verbal behaviour by the listener, in that they serve as messages to the primary speaker. One of the most difficult aspects in identifying a backchannel is determining whether a particular behaviour constitutes a backchannel or a separate turn. As Maynard (1986, 1084) points out, much of the confusion stems from distinguishing between having a turn and having the floor, and can be attributed to self-contradictions in Yngve’s (1970) definition. Although Yngve’s earlier definition of a backchannel is given in terms of not relinquishing a turn, he cites the following example as backchannel behaviour:

Towards a Pedagogical Framework to Develop the Listenership

299

In one case, what looked like backchannel activity consisted of filling in needed personal background so that the person having the floor could continue. This went on for about thirty seconds and involved a number of sentences. It is interesting to note that this extensive backchannel activity was in turn provided with back-back channel activity of the ‘uh-huh’ variety. (Yngve 1970, 568).

In this example, Yngve appears to be identifying backchannel behaviour on the basis of holding the floor, rather than having the turn. Similarly, longer utterances such as You’ve finished it then in response to the primary speaker’s talk can cause confusion because this utterance may allow the primary speaker to continue holding the floor; yet it appears to be a speaking turn in itself. Thus, what starts as a backchannel can actually end up as a turn, if the primary speaker shows no willingness to continue speaking. To differentiate between backchannels and turns, this study identifies backchannels in the context of Markel’s (1975) turn-taking system, which he describes as follows: A speaking turn begins when one interlocutor starts solo talking. For every speaking turn there is a concurrent listening turn, which is the behaviour of one or more nontalking interlocutors present. (Markel 1975, 190).

Thus, in line with the work of Cutrone (2005) and Maynard (1997), this paper identifies a brief statement as a backchannel and not a primary turn when it serves only to react to what the primary speaker is saying (listening function) and not to add any new information to the conversation (speaking function). Accordingly, brief questions such as Is that right? or Oh really?, which are formed as requests for clarification, are regarded as backchannels. However, a question such as Why did he do that? is considered a full speaking turn because it serves a speaking function in terms of driving the conversation in a new direction. Moreover, responses to questions are not considered backchannels. This follows one of the tenets of Ward and Tsukahara’s (2000) practical definition of backchannels, according to which, unlike responses to questions, backchannels are optional and not required. Additionally, answers to questions, which are sometimes quite brief and include ellipsis, would also seem to provide new information that helps drive the conversation forward, thus constituting a change of primary speakership. Finally, analysts have to make decisions regarding how to deal with pauses and utterances found between turns at talk. In Cutrone’s (2005) intercultural analysis, utterances are identified as backchannels only when they occur immediately after the primary speaker stops talking (within one

300

Chapter Thirteen

second) and are followed by a substantial pause before the next turn at talk starts (exceeding one second). This decision was made because it was felt that these backchannels are produced in response to the primary speaker’s speech, and because they occur before a substantial turn transitional period starts. This paper follows the approach used by Cutrone’s (2005) earlier study.

3. Analytical/practical issues concerning backchannels There are several issues involving the analysis of backchannel behaviour in the research, as well as the subsequent application of this skill-set in the EFL classroom. First, while researchers have suggested a range of backchannel functions, there has been little consensus on them to date, as there appear to be reliability issues in measuring this aspect of conversation. With this mind, Maynard (1997) has identified the following functional categories of backchannels: continuer, understanding, agreement, support and empathy, emotive, minor additions, and clarification. Examples of backchannels in each of these categories are presented in Cutrone’s (2005, 2010, 2013) previous analyses. The one thing these functions have in common is that they all seem to serve a paramount function, namely, to acknowledge and support the primary speaker, so he/she can continue driving the conversation forward; however, as Section 4.4 discusses, there may exist important functional differences across cultures where Japanese EFL/ESL are concerned, which is one of the main reasons I believe that backchannel behaviour warrants more attention in EFL/ESL classes involving Japanese learners. To date, the teaching of backchannel behaviour has been largely neglected in the EFL classroom. Only recently has there been any mention of backchannels in learner textbooks and teacher education materials, and these references appear to be limited in number and do not recognize listener responses in any great depth (Thonus 2007). Similarly, the research on the learnability of L2 backchannel behaviour has been scant, and the few studies that have been conducted seem to contain methodological issues that would preclude them from adequately informing classroom application. For instance, in addition to the problems associated with comparing experimental groups with completely different sociolinguistic characteristics (i.e., age and socio-economic status), Brozyna (2007) appears to have based her results solely on two-minute extracts pre- and post-instruction (with no delayed post-test). Besides the validity concerns involving the brevity of her analysis, Brozyna (2007) seems to equate success solely with the ability to produce more

Towards a Pedagogical Framework to Develop the Listenership

301

backchannels. However, the relationship between backchannels and discourse is not as straightforward as simply counting the number of backckhannels a person or group has sent. Rather, the study of backchannel behaviour is complex and multifaceted, with frequency comprising only one of the many interrelated and overlapping components in this skill-set. Moreover, even where frequency is the focus, what determines success will likely differ depending on the issues of a particular cultural group, as Japanese EFL/ESL learners, who are generally thought to send backchannels too frequently (see Sections 4.1 and 5.1), should be instructed to provide backchannels less frequently. In a more recent study, which involved Japanese EFL learners, the criteria that Sardegna and Molle (2010) used for assessing backchannel behaviour also appears to be oversimplified, as the lone determinant of success was the observation that the Japanese EFL learners provided fewer backchannel forms common in Japanese and more that are common in English. Further, it seems that Sardegna and Molle’s (2010) study contained brevity issues in both treatment and analysis, as the treatment consisted of only a two-hour lesson (conducted by video conference), and their conclusions appear to be based solely on the limited amount of data presented in a short post-treatment conversational excerpt involving 5 JEFLs and the teacher. Finally, the fact that this study excluded non-verbal backchannels such as head nods, which are used quite frequently by Japanese EFL/ESL speakers, seems to bring the results further into question.

4. Why is listenership important for JEFL/ESL learners? The main reason listenership is becoming such an important topic in language learning is due to researchers becoming increasingly aware of its great impact on intercultural communication (IC). As O’Keeffe, Clancy and Adolphs (2011, 100) caution, attempts to move between L1 and L2 pragmatic norms can feel like “a minefield for learners of a language.” Concerning Japanese EFL/ESL pedagogical contexts, various studies in this area have shown that Japanese EFL/ESL speakers’ listening behaviour differs from that of native English speakers (NESs) in many respects (Clancy et al. 1996; Cutrone 2005; Maynard 1986, 1987, 1990, 1997; White 1989) and these differences sometimes lead to miscommunication and negative perceptions across cultures (Blanche 1987; Boxer 1993; Cutrone 2005; LoCastro 1987). This section, thus, outlines some of these differences in listening behaviour and discusses their potential impact on

302

Chapter Thirteen

IC involving Japanese EFL/ESL speakers. To this end, this section also serves to identify some of the sub-categories of listenership and emphasise why each should be taken up in the language classroom.

4.1 Frequency of backchannels Some initial clarification is needed to explain what the term Frequency means in the category described in this section. Frequency, as an independent category, refers to the overall frequency of backchannels and should not be confused with the inclusion of the word frequency used in subsequent subsections where the frequency of backchannels within categories, such as variability and discourse contexts favouring backchannels, are also measured. It is also important to note that the frequency of one person’s backchannels is directly related to how much their interlocutor speaks (see Section 5.5). So, for instance, if one person in a dyadic conversation were to produce a large number of backchannels, it would also demonstrate that that person spent much of the conversation in a listening role. Accordingly, it would be expected that their interlocutor spoke much more and likely provided far fewer backchannels (i.e., since they were mostly in a primary speaking role and not a listening role, the interlocutor would have had far fewer opportunities to produce backchannels). With this in mind, the results from a number of studies have consistently shown Japanese producing more backchannels than NESs in both intracultural (using their L1s) and intercultural conversations (using L2 English). For instance, Maynard (1986, 1990, 1997), comparing Japanese and Americans’ backchannel behaviour in both intracultural and intercultural conversations, observed that the Americans and Japanese participants’ backchannel behaviour in the intercultural conversations were similar to that within their own cultural context. On average, the Japanese participants sent backchannels every 4.5 seconds of their interlocutor’s primary speaking turn, while the American participants sent backchannels every 19.25 seconds. While most other studies used a different analytical technique to count the frequency of backchannels (i.e., analysts counted backchannels per interlocutor word rather than the timing of backchannel responses, due to potentially different rates of speech thought to affect the results), the results were similar nonetheless, and showed Japanese participants sending backchannels more frequently than NESs. For instance, White (1989) compared Japanese and Americans in intracultural and intercultural conversations and also found that the Japanese participants provided more

Towards a Pedagogical Framework to Develop the Listenership

303

backchannels in both contexts. In the intracultural dyads, the ratio was 3:1, and in the intercultural dyads, the ratio was 1.5:1. Clancy et al. (1996) found that, in intracultural conversations, Japanese participants used slightly more backchannels than the Americans (1.06:1) in her study. Cutrone (2005) examined Japanese and British intercultural conversations in English and found the Japanese participants employed more backchannels than their British interlocutors at a rate of 1.14:1. Lastly, Ike (2010) investigated backchannel behaviour in Japanese English vis-à-vis Australian English and confirmed that backchannels occurred more frequently in Japanese English (1.95:1). Besides the fact that such frequent interjections could be (mis)taken as a sign of impatience and demand for a quick completion of the statement (Cutrone 2005; Lebra 1976; Mizutani 1982), some people might perceive the Japanese EFL/ESL speakers’ lack of primary speakership as an unwillingness to communicate (Cutrone 2005; Sato 2008).

4.2 Variability of backchannel types It is necessary to point out that some of the intercultural studies mentioned above were carried out differently. For instance, regarding the use of participants, Cutrone’s study (2005) involved Japanese and British adults and took place in Okayama, Japan (in an EFL context), whereas both Maynard (1997) and White (1997) used Japanese and American university students in their intercultural pairings, which took place in New Jersey (Rutgers University) and Hawaii (the University of Hawaii), respectively (in ESL contexts). In all three studies however, the Japanese EFL/ESL speakers were highly proficient in English. Furthermore, another important difference in these studies is that Cutrone (2005) examined the variability of the backchannels used by the participants of each cultural group in great depth, while Maynard (1997) dealt with this aspect of backchannel behaviour only briefly, and White (1989) did not include it in her analysis at all. In their studies, Cutrone (2005) and Maynard (1997) found that the Japanese EFL/ESL speakers’ backchannels consisted mainly of non-word vocalisations and head nods, while the NESs (American and British) displayed greater variability in the types of backchannels they sent. In his most recent intercultural analysis, Cutrone (2013), using Japanese and American university students in an EFL setting, expanded his list of categories to include the following verbal and non-verbal backchannel forms:

304

Chapter Thirteen

Types of Verbal Backchannels x Simple (isolated) x Simple with head nod(s) x Compound (isolated) x Compound with head nod(s) x Complex (isolated) x Complex with head nod(s) x Japanese x Minimal Response x Extended Response Types of Non-verbal Backchannels x Single head nod (isolated) x Multiple head nods (isolated) x Smile (isolated) x Smile with verbal backchannel x Laughter (isolated) x Laughter with verbal backchannel x Raised Eyebrows (isolated) x Raised Eyebrows with verbal backchannel x • 2 non-verbal backchannels occurring simultaneously (isolated) x • 2 non-verbal backchannels occurring simultaneously with verbal backchannel While many of the differences in the subcategories above were negligible, a few of the notable ones are worth pointing out here. That is, similar to the results of the previous studies, the NESs were able to produce a greater range of complex backchannels (i.e., brief sentences containing several different words), as well as content words (such as “brilliant”) in their backchannel responses. Furthermore, the Japanese participants in this study relied mostly on simple backchannels (i.e., nonword vocalisations such as un and head nods) and repetition. In this regard, it is worth observing that lack of variability, as well as repetition, in backchannel form may be (mis)interpreted as boredom and inattention (Cutrone 2005, 2013; McCarthy 2002, 2003).

4.3. Discourse contexts favouring backchannels The term discourse contexts favouring backchannels, used by Maynard (1986, 1990, 1997), refers to the places or points identified in the primary

Towards a Pedagogical Framework to Develop the Listenership

305

speaker’s speech where the non-primary speaker frequently sends backchannels. Although various studies have provided different degrees of detail in their analyses, the following discourse contexts have been examined: Discourse Contexts x At or directly after a pause Clausal Boundaries x At or near any clausal boundary x At or near internal clause boundary x At or near final clause boundary At or directly after points in which both clausal boundaries and pauses occur together x At or directly after any clausal boundary, accompanied by a pause x At or directly after internal clause boundary, accompanied by a pause x At or directly after final clause boundary, accompanied by a pause Gesticulation x After primary speaker’s non-verbal gesture x After primary speaker’s head nod(s) Simulataneous Speech x During the primary speaker’s speech A common finding has been that Japanese discourse contexts favouring backchannels vary considerably (in both the L1 and L2 English), while clausal boundaries, especially coinciding with a pause, are the single most important discourse contexts for NESs (Cutrone 2005; Maynard 1986, 1990, 1997; White 1989). It is not known whether such differences affect perceptions across cultures. Moreover, related to the discussion on overall frequency above, a great portion of the backchannels provided by Japanese EFL/ESL speakers are sent during the primary speakers’ speech, creating simultaneous speech (Cutrone 2005; Hayashi 1988; Maynard 1997). As various researchers such as Lebra (1976) and Mizutani (1982) have hypothesised that some may take these frequent interjections as interruptions, a sign of the listener’s impatience and demand for a quick completion of the statement, White (1989) and Cutrone (2005) conducted correlation analyses to find out what effect Japanese people’s frequent backchannels might have on their intercultural interlocutors’ perceptions of them. Interestingly, the ten

306

Chapter Thirteen

Americans in White’s (1989) study perceived more frequent backchannels sent by their Japanese interlocutors as a positive trait (i.e., showing signs of comprehension, encouragement, and interest and concern), while the eight Britons in Cutrone’s (2005) intercultural analysis perceived more frequent backchannels as interruptions and signs of impatience.

4.4 Functions of backchannels While this has been an under-researched area overall, there may be some key functional differences across cultures, as Blanche (1987) and Cutrone (2005) have provided anecdotal evidence of Japanese people providing unconventional backchannels in English, such as by employing continuer-, understanding-, agreement-, and/or empathy-/support- type backchannels in situations when they did not understand or disagreed with what their interlocutor was saying at the time. In the common classroom scenario described by Blanche (1987) and Cutrone (2005), NES teachers sometimes misinterpret students’ nods coupled with vocalisations of yes and mhm at seemingly appropriate times as displays of understanding, rather than simply polite expressions of attending. When teachers discover much later on that students have not understood them, they may sometimes feel perplexed and/or even slightly annoyed by what they perceive to be mixed signals, or in extreme cases, deceptive messages, resulting in the squandering of valuable class time. Sometimes these misunderstandings can have more dire consequences, as was the case in the Hitachi-Mitsubishi trial (The Japan Times 1983, 2). One of the defendants in the case, Mr Takaya Ishida of Mitsubishi, claimed that he had not agreed with the FBI undercover agents when they told him he had to steal some information/documents. His defense counsellor argued that Mr Ishida’s responses of yeah and uhuh were not to show agreement, but rather to indicate he was listening and to allow the other person to continue, which is a common way to respond in Japanese culture.

4.5 Conversational involvement As Section 4.1 alludes to, several studies involving the intercultural analyses of communication styles have shown that the Japanese EFL/ESL speakers in these studies spoke less than NESs, did not elaborate as much, and were less likely to engage in small talk (Cutrone 2005, 2013; Hill 1990; Sato 2008). This is potentially a source of misunderstanding in an English conversation, as the importance of making small talk, taking the

Towards a Pedagogical Framework to Develop the Listenership

307

initiative to speak, and elaboration towards making a positive impression have been documented by several sources (Cutrone 2005; McCarthy 2002, 2003; McCroskey 1992; Sato 2008; Stubbe 1998).

5. Establishing targets for teaching The fact that backchannel behaviour is an especially difficult skill-set to measure may be one of the main reasons why it has been rarely taught and studied (Thonus 2007). In an earlier paper (Cutrone 2010), I attempted to survey the field of research in this area in order to piece together a descriptive account of how NESs generally employ backchannels. In the current study however, I chose not to use rumerical targets because I felt that they would be largely impractical to the teachers that such targets are designed to help. Thus, rather than providing strict (and potentially unrealistic) numerical goals, this paper attempts to offer general directional targets for teachers to guide their students towards. Thus, based on the descriptions of NESs’ backchannel behaviour, as well as on the documented problems some Japanese EFL/ESL speakers have experienced in this area, this section proposes the following instructional goals for Japanese L2 English speakers where listenership is concerned.

5.1 Frequency of backchannels As discussed in Section 4.1, Japanese EFL/ESL speakers employ backchannels considerably more than NESs and this may, at times, negatively affect IC. Therefore, if we choose as one of our instructional goals that of approximating the listenership of native (and/or fully proficient) speakers of English in the international community, then Japanese EFL/ESL speakers should be taught how to send backchannels less frequently. A target pace for sending backchannels according to American English (which, as discussed in Section 7 is used as a model only because other varieties of English have not yet been wholly described in the research literature) has been shown to be approximately every 30-40 words (Cutrone 2005; Maynard 1997; White 1989). However, due to the spontaneous nature of how conversations develop, teachers would be well advised to not focus on strict numerical targets such as this. Rather, teachers should first aim to raise their learners’ awareness in this area, that is, showing the potential negative impact of frequent minimal backchannels. Subsequently, as the sections below will discuss, teachers need to make concerted efforts in the classroom to have their students focus on quality over quantity where backchannels are concerned.

308

Chapter Thirteen

5.2 Variability of backchannel types As Section 4.2 reports, the variability of Japanese EFL/ESL speakers’ backchannels tends to be more limited than that of NESs, and this lack of variability tends to negatively affect their intercultural interlocutors’ perceptions of them. Consequently, as a pedagogical objective, Japanese EFL/ESL speakers would benefit from trying to use a more diverse repertoire of listener responses in their intercultural encounters in English. Accordingly, the distinction in the research arena between what is a backchannel vis-à-vis what is a primary speaker turn (as described in Section 2) becomes blurred and less important in the classroom. That is, whether listener responses are backchannels or full speaking turns, the use of longer utterances and content words over minimal responses (i.e., nonword vocalisations) should be emphasized in the language classroom. However, more variability in itself should not be the only goal here; it is perhaps more important that the listener responses that learners produce actually fit the situation they are in. In this way, instruction regarding the functional use of backchannels will go a long way towards increasing the variability that Japanese EFL/ESL speakers employ in the backchannels they use. Thus, for instance, if one friend told another friend that they had just won the national lottery, most people would expect an emphatic response such as That’s great! or Congratulations! instead of a simple head nod or conversational grunt such as un or mhm. To this end, one activity that a teacher can administer in the classroom is to have learners simply match statements with appropriate responses (see Appendix A).

5.3 Discourse contexts favouring backchannels As Section 4.3 outlines, clausal boundaries and pauses (especially occurring simultaneously) have been identified as primary discourse contexts favouring backchannels. Still, it is important to keep in mind that backchannels that are sent in other discourse contexts are not necessarily considered incorrect, and the placement of backchannels largely depends on the context of the conversation and the function that the non-primary speaker wants to express at that particular time. Hence, while it may be helpful to guide students towards sending their backchannels at clausal boundaries, teachers are advised to only alter learners’ behaviour in this area when it is seen to negatively impact IC, as mentioned in the following sub-section. Concerning simultaneous speech backchannels, Japanese L2 English speakers, unlike NESs, have a tendency to send backchannels while their

Towards a Pedagogical Framework to Develop the Listenership

309

interlocutor is speaking, which can negatively influence IC in English (as stated in Section 4.3). That is, these frequent interjections may sometimes be (mis)taken as rude and interruptive. Hence, learners should be instructed to avoid sending backchannels whilst their interlocutor is in the midst of their speaking turn. Ideally, learners should aim to send backchannels when their interlocutor completes a clause and/or when a pause in the primary speaker’s speech occurs.

5.4 Form and function of backchannel responses Instruction on form and function is crucial to helping students master this overall skill-set. However, since backchannel behaviour is largely idiosyncratic and contingent upon contextual variables, success in this area is extremely difficult to measure. Adding to this complexity, there is considerable overlap between forms and functions. Hence, it seems that assessing how well a backchannel form corresponds to a backchannel function is something that needs to be done on a case-by-case basis. To assist teachers in this area, Cutrone (2010) has surveyed the research in this area and has provided a list of forms for each of the specific backchannel functions mentioned in this paper. Still, as Section 5.2 briefly discussed, there exist moments in a conversation where taking a full-speaking turn might be a better option than producing a backchannel. For instance, when Japanese EFL/ESL speakers produce backchannels to feign understanding and/or agreement is a case in point. In these particular cases, learners need to be taught explicitly what expressions they can use when they do not understand and/or disagee with what their interlocutor is saying. For instance, creating situations in the classroom (e.g., roleplays) where students can practice expressions such as I’m not really sure what you mean or I’m afraid I disagree would be helpful in this regard. Beyond these specific situations, instruction designed to provide learners with options to navigate a number of other difficult conversational situations would be useful (see Appendix B for a list of phrases associated with various conversational repair strategies and management techniques). To this end, Cutrone (2013) advocates the teaching of specific conversational micro-skills as they relate to backchannel behaviour. These micro-skills include conversational repair strategies such as the ability to self-repair, the ability to ensure comprehension on the part of the listener (e.g., comprehension checks), the ability to initiate repair when there is a potential breakdown (e.g., clarification requests), and the ability to employ compensatory strategies (e.g., paraphrasing, circumlocution, and the use of

310

Chapter Thirteen

fillers and hesitation devices). Such repair strategies are often realised in conversations by means of a specific set of typical conversational phrases and routines such as I beg your pardon or what does that mean. Further examples can be seen in two language teachers’ resource books that have been entirely based on these structures: Function in English (Blundell, Higgens, and Middlemiss 1982) and Conversation and Dialogues in Action (Dörnyei and Thurrell 1992). Section 6, which deals specifically with the teaching of listenership, offers general suggestions for classroom intergration.

5.5 Conversational involvement As Section 4.1 alludes to, it is important to understand that teaching backchannel behaviour and teaching learners how to speak more seem to be directly related pedagogically. That is, the inextricable link that exists between producing a backchannel and producing a speaking turn was pointed out by Schegloff (1982), who claimed that when non-primary speakers provide backchannels, they are essentially forsaking the opportunity to take a primary speaking turn. This is supported by the fact that Transition-Relevant Places (i.e., moments in the conversation at which an exchange of turn is appropriate) in Sacks, Schegloff and Jefferson’s (1974) turn-taking system, are often found in similar discourse contexts as backchannels. This was evident in Cutrone’s (2005) intercultural analysis, as several of the Japanese EFL speaking participants acknowledged that they often employed backchannels as a way to avoid taking primary speakership, yet still contribute to the conversation. Therefore, as noted above, it seems necessary to examine backchannel behaviour in tandem with other aspects of listener behaviour such as the ability to take over primary speakership of the conversation and/or ask follow-up questions, etc. Thus, in addition to the conversational repair strategies shown in Appendix B, learners would benefit from instruction in the following areas: return questions, follow-up questions, new topic initiation, expansion techniques, etc. These features are consistent with what some interlocutors might hope to encounter in an English conversation, as the importance of making small talk, taking the initiative to speak, and elaboration have been documented by several sources (Cutrone 2005; McCarthy 2002, 2003; McCroskey 1992; Ross 1998; Sato 2008; Stubbe 1998; Yashima 2002). According to Widdowson (1989), a large part of communicative competence is simply a matter of knowing how to employ such conventionalised expressions, or as he terms them, “partially preassembled patterns” and “formulaic frameworks” (Widdowson

Towards a Pedagogical Framework to Develop the Listenership

311

1989, 135). These expressions and formulae comprising conversational micro-skills would seem to lend themselves ideally to explicit teaching, and can serve as useful language input for conversation classes (Dörnyei and Thurrell 1992, 1994). Although many of the variables discussed above are interrelated to learners’ listening behaviour in one way or another, an obvious and recurring theme in the research literature has been the descriptions of Japanese EFL/ESL speakers’ reticent behaviour and perceived unwillingness to communicate, which has negatively affected interlocutors’ perceptions of them in intercultural encounters. In addition to providing learners with the ability to use a diverse repertoire of English backchannels, it is imperative that instruction helps Japanese EFL/ESL speakers demonstrate more of an overall willingness to communicate (WTC). Increasing WTC would not only improve learners’ confidence and reduce their anxiety, it would also facilitate their future language development and positively influence the perceptions their interlocutors have of them and their communicative abilities.

6. Approaches to the teaching of listener responses Due to the vital role backchannel behaviour has in intercultural communication, and the fact that it is largely neglected in EFL/ESL pedagogy, listenership warrants a higher priority in ELT, particularly concerning Japanese EFL/ESL speakers. Accordingly, this paper has described how language practitioners can best integrate this important feature of conversation into the language classroom. Although Schmidt’s (1983) longitudinal study demonstrated that adequate L2 backchannel behaviour can be acquired implicitly (i.e., in an immersion setting) over an extended period of time (i.e., seven years), various studies have advocated an explicit (and perhaps more expedient) approach towards the teaching of English backchannels in the language classroom (Brozyna 2007; Cutrone 2013; Thonus 2007). Since backchannel behaviour tends to differ significantly across cultures, and many learners may not even be conscious of this skill, Thonus (2007) believes that the first step, as in many language-learning tasks, is raising awareness of this area. Gilewicz and Thonus (2003) suggest that close vertical transcriptions (i.e., transcriptions that attempt to replicate not only words, but also silences and interruptions, signs and sounds, in order to provide full and accurate accounts of the talk represented) of NS-NS (native speaker) and NNSNNS (non-native speaker) conversations are an excellent means of raising awareness, as analysis of transcripts permits focused study of backchannel

312

Chapter Thirteen

variations and functions. To this end, Thonus (2007, 145) recommends the following four pedagogical strategies: x Contrasting listener responses across languages x Creating opportunities for students to become conversational researchers x Providing multiple examples of listener responses in varying contexts x Coaching listener response use in fluency-focused exercises Thonus (2007), recognising that the acquisition of listener responses is a process that takes time, points out some of the problems that learners may need to overcome. For example, learners may progress through stages of underuse to overuse (i.e., flooding), and may also fixate on one backchannel form to the exclusion of others but broadening their repertoire over time Also, the incorrect use of learners’ repetition-as-backchannel is also common (Thonus 2002, 2007). Cutrone (2013) believes the best way forward may be a combination of methods in which implicit methods are used to supplement an explicit approach. That is, the norms of backchannel behaviour in English, as well as how different listening behaviours can be misconstrued across cultures, should be systematically taught, and this should be supported by teacher feedback, peer discussions and self-analysis. With the above-mentioned framework in mind, I would like to suggest two phases of instruction in this area. The first phase of instruction involves having learners take part in consciousness-raising activities, which are designed to raise awareness of how conversational behaviours might be interpreted (and sometimes misunderstood) across cultures. One of the ways which EFL/ESL teachers can raise their learners’ awareness in this area is by, first, having students watch video clips and/or listen to audio excerpts which are specifically chosen (or created) to draw attention to particular features of conversation and, subsequently, engage in deconstruction/discussion activities to become more cognizant of how different communicate styles might affect IC. In the second phase of instruction, learners focus more on developing strategies for application. This second phase involves the following two steps: skill building and practice. In the skill-building step, students are given tasks that enable them to explore the target language and develop strategies for output. For instance, students may be presented with an incomplete conversational transcript (i.e., with all the listener responses removed) and asked to write in appropriate backchannel forms at

Towards a Pedagogical Framework to Develop the Listenership

313

appropriate places in the primary speaker’s speech. Subsequently, in the practice phase, learners take part in real-time role-plays (or conversations with prompts) and focus on applying what they learned in the previous step. It would be a good idea to have the teacher, as well as other students, observe the conversations and offer constructive feedback. Ideally, the conversations would be recorded and played back for reference, as the teacher and students provide feedback.

7. Conclusion Having presented a general framework for the inclusion of listenership in the language classroom above, I acknowledge that various obstacles still remain. First, it seems unrealistic to expect teachers to go from little or no instruction in this area to suddenly dedicating entire classes and/or syllabi to it. A reasonable approach would be to start with modest goals and gradually progress to larger ones, as teachers become more familiar and comfortable teaching this feature of language. A good place to introduce listener responses would be in EFL/ESL textbooks designed for oral communication classes. As researchers such as Thonus (2007) and Wannaruk (1997) have advocated, these materials need to start incorporating sections explaining how backchannels are used and activities to practise using them. By laying out some clear goals and targets for the instruction of backchannels in the language classroom in this paper, it is the writer’s hope that English language teachers will find it easier to integrate this elusive aspect of pragmatic competence into their lessons. Finally, while the use of NES norms as a model for backchannel behaviour is presently justified in this paper (as there are no other concrete and comprehensive options currently available where backchannels are concerned), it would be a positive development to see researchers one day piece together a more thorough and representative description of English backchannel behaviour as it exists in the international community. By doing this, a broader set of goals and norms could be presented in EFL/ESL contexts around the world. Nonetheless, it is important to point out that the instructional goals for EFL/ESL backchannel behaviour will probably always have much more to do with the recognised tendencies and idiosyncrasies of a particular group of learners than following the prescriptive norms of any one NES variety and/or international standard. For instance, irrespective of which English model was used, it is clear from the research presented above that many of the general goals would remain the same for the Japanese EFL/ESL speakers, such as sending minimal backchannels less frequently (especially while one’s interlocutor

314

Chapter Thirteen

is speaking), with greater variability (but at context-appropriate moments), while asking questions and taking the primary speakership in the conversation more often, and initiating conversational repair strategies when they do not understand and/or disagree rather than feigning understanding and agreement. In intercultural encounters with Japanese EFL/ESL speakers, issues involving backchannel behaviour often stem from, and result in, intercultural misunderstandings. I hope to have shed some light on this complex phenomenon, which EFL/ESL teachers can only deal with successfully if they are properly informed. The research to date has contributed to our understanding of backchannel behavior across cultures and provides useful insights to teachers as they consider classroom methods and practices. Nonetheless, there is much we do not yet know about backchannel behaviour and future research in this area would, thus, do well to examine diverse groups of EFL/ESL learners, and investigate how backchannel behaviour is affected by other factors such as larger group dynamics, varying conversational registers, interlocutor familiarity, and the topic of the conversation.

References Blanche, P. 1987. “The Case for a Pedagogy of Pragmatics in Foreign or Second Language Teaching.” RELC Journal 18, 46-71. Blundell, J., J. Higgens, and N. Middlemiss. 1982. Function in English. Oxford: Oxford University Press. Boxer, D. 1993. “Complaints as Positive Strategies: What the Learner Needs to Know.” TESOL Quarterly 27: 277-299. Brozyna, B. 2007. The Teachability of Interactional Discourse Signals in the Classroom Environment. Master’s thesis, University of Reading. Clancy, P., S. Thompson, R. Suzuki, and H. Tao. 1996. “The Conversational Use of Reactive Tokens in English, Japanese, and Mandarin.” Journal of Pragmatics 26: 355-387. Cutrone, P. 2005. “A Case Study Examining Backchannels in Conversations between Japanese-British Dyads. Multilingua - Journal of Cross-Cultural and Interlanguage Communication 24: 237-274. —. 2010. “The Backchannel Norms of Native English Speakers: A Target for Japanese L2 English Learners.” University of Reading Language Studies Working Papers 2: 28-37. —. 2013. Assessing Pragmatic Competence in the Japanese EFL Context: Towards the Learning of Listener Responses. Newcastle upon Tyne: Cambridge Scholars Publishing.

Towards a Pedagogical Framework to Develop the Listenership

315

Dörnyei, Z., and S. Thurrell. 1992. Conversation and Dialogues in Action. Hemel Hempstead: Prentice Hall. —. 1994. “Teaching Conversation Skills Intensively: Course Content and Rationale.” ELT Journal 48: 40-49. Duncan, S. Jr. 1974. “On the Structure of Speaker-Auditor Interaction during Speaking Turns.” Language in Society 2: 161-180. Duncan, S. Jr., and D. Fiske. 1977. Face to Face Interaction: Research, Methods, and Theory. Hillsdale, NJ: Erlbaum. Farr, F. 2003. “Engaged Listenership in Spoken Academic Discourse: The Case for Student-Tutor Meetings.” Journal of English for Academic Purposes 2: 67-85. Fujimoto, D. 2007. “Listener Responses in Interaction: A Case for Abandoning the Term Backchannel.” Journal of Osaka Jogakuin College 37: 35-54. Gardner, R. 1997. “The Conversation Object mm: A Weak and Variable Acknowledging Token.” Research on Language and Social Interaction 30: 131-156. —. 1998. “Between Speaking and Listening: The Vocalisation of Understandings.” Applied Linguistics 19: 204-224. Gilewicz, M., and T. Thonus. 2003. “Close Vertical Transcription in Writing Center Training and Research.” Writing Center Journal 24: 40-55. Goodwin, C. 1986. “Between and Within: Alternative Sequential Treatment of Continuers and Assessments.” Human Studies 9: 295-27. Hayashi, R. 1988. “Simultaneous Talk – From the Perspective of Floor Management of English and Japanese Speakers.” World Englishes 7 (3): 269-288. Hill, T. 1990. “Sociolinguistic Aspects of Communicative Competence and the Japanese Learner.” Dokkyo University Studies in English 36: 69-104. Ike, S. 2010. “Backchannel: A feature of Japanese English.” In JALT 2009 Conference Proceedings, edited by A. Stoke, 205-215. Tokyo: JALT. Ito, A. 2007. Functions of Back Channels in Japanese Casual Conversations: Comparing Single Back Channels and Repeated Back Channels. Master’s thesis, Kobe College Graduate Division of English. Lebra, T. 1976. Japanese Patterns of Behavior. Honolulu: The University of Hawaii Press. LoCastro, V. 1987. “Aizuchi: A Japanese Conversational Routine.” In Discourse across Cultures: Strategies in World Englishes, edited by L. Smith, 101-113. London: Prentice Hall. Markel, N. 1975. “Coverbal Behavior Associated with Conversation

316

Chapter Thirteen

Turns. In Organization of Behavior in Face-to-Face Interaction, edited by A. Kendon, R. M. Harris, and M. R. Key, 189-197. The Hague: Mouton. Maynard, S. K. 1986. “On Back-Channel Behavior in Japanese and English Casual Conversation.” Linguistics 24 (6): 1079-1108. —. 1987. “Interactional Functions of a Nonverbal Sign: Head Movement in Japanese Dyadic Conversations.” Journal of Pragmatics 11: 589606. —. 1990. “Conversation Management in Contrast: Listener Responses in Japanese and American English. Journal of Pragmatics 14: 397-412. —. 1997. “Analyzing Interactional Management in Native/Non-Native English Conversation: A Case of Listener Response.” IRAL 35: 37-60. McCarthy, M. 2002. “Good Listenership Made Plain: British and American Non-Minimal Response Tokens in Everyday Conversation.” In Using Corpora to Explore Linguistic Variation, edited by R. Reppen, S. M. Fitzmaurice, and D. Biber, 49-72. Philadelphia: Benjamins. —. 2003. “Talking Back: ‘Small’ Interactional Response Tokens in Everyday Conversation.” Research on Language and Social Interaction 36: 33-63. McCroskey, J. C. 1992. “Reliability and Validity of Willingness to Communicate Scale.” Communication Quarterly 40: 16-25. Mizutani, N. 1982. “The Listener’s Responses in Japanese Conversation.” Sociolinguistics Newsletter 13: 33-38. Nikolova, D. 2008. “English Teaching in Elementary Schools in Japan: A Review of a Current Government Survey.” Asian EFL Journal 10 (1). http://asian-efl-journal.com/quarterly-journal/2008/03/25/englishteaching-in-elementary-schools-in-japan-a-review-of-a-currentgovernment-survey/. O’Keeffe, A., B. Clancy, and S. Adolphs. 2011. Introducing Pragmatics in Use. London: Routledge. O’Keeffe, A., M. McCarthy, and R. Carter. 2007. From Corpus to Classroom: Language Use and Language Teaching. Cambridge: Cambridge University Press. Oreström, B. 1983. Turn-Taking in English Conversation. Lund: Lund University Press. Roger, A. 2008. “Teaching the Speaking Skill to Japanese Students Part 1: Construct & Practice.” The Journal of Kanda University of International Studies 20: 1-26. Ross, S. 1998. “Divergent Frame Interpretations in Oral Proficiency Interview Interaction.” In Talking and Testing: Discourse Approaches

Towards a Pedagogical Framework to Develop the Listenership

317

to the Assessment of Oral Proficiency, edited by R. Young, and A. W. He, 333-353. Philadelphia: Benjamins. Sacks, H. E., E. Schegloff, and G. Jefferson. 1974. “A Simple Systematics in the Organization of Turn-Taking for Conversation.” Language 50: 696-735. Sardegna, V., and D. Molle. 2010. “Videoconferencing with Strangers: Teaching Japanese EFL Students Verbal Backchannel Signals and Reactive Expressions.” Intercultural Pragmatics 7 (2): 279-310. Sato, Y. 2008. “Oral Communication Problems and Strategies of Japanese University EFL learners.” PhD dissertation, University of Reading. Schegloff, E. 1982. “Discourse as an Interactional Achievement: Some Uses of ‘UH-HUH’ and Other Things that Come Between Sentences.” In Georgetown University Roundtable on Language and Linguistics, Analyzing Discourse: Text and Talk, edited by D. Tannen, 71-93. Washington: Georgetown University Press. Schmidt, R. 1983. “Interaction, Acculturation, and the Acquisition of Communicative Competence: A Case Study of an Adult.” In Sociolinguistics and Language Acquisition, edited by N. Wolfson, and E. Judd, 137-174. Rowley: Newbury House. Selting, M. 1994. “Emphatic Speech Style – with Special Focus on the Prosodic Signalling of Emotive Involvement in Conversation.” Journal of Pragmatics 22: 375-408. Stubbe, M. 1998. “Are you Listening? Cultural Influences on the Use of Supportive Verbal Feedback in Conversation.” Journal of Pragmatics 29 (3): 257-289. The Japan Times. “Japanese Mannerism is a Key Point in IBM case.” The Japan Times, January 23, 1983. Thonus, T. 2002. “Tutor and Student Assessments of Academic Writing Tutorials: What is ‘Success’?” Assessing Writing 8 (2): 110-134. —. 2007. “Listener Responses as a Pragmatic Source for Learners of English.” CATESOL Journal 19: 132-145. Uematsu, S. 2000 “The Use of Back Channels between Native and NonNative Speakers in English and Japanese. Intercultural Communication Studies 10: 85-98. Wannaruk, A. 1997. “Back-Channel Behaviour in Thai and American Casual Telephone Conversation.” Suranaree Journal of Science and Technology 4 (3): 168-174. Ward, N. 2004. “Pragmatic Functions of Prosodic Features in Non-Lexical Utterances.” Speech Prosody 4: 325-328. Ward, N., and W. Tsukahara. 2000. “Prosodic Features which Cue BackChannel Responses in English and Japanese.” Journal of Pragmatics

318

Chapter Thirteen

32 (8): 1177-1207. White, S. 1989. “Backchannels across Cultures: A Study of Americans and Japanese.” Language in Society 18: 59-76. Widdowson, H. G. 1989. “Knowledge of Language and Ability for Use.” Applied Linguistics 10 (2): 128-137. Yashima, T. 2002. “Willingness to Communicate in a Second Language: The Japanese EFL context.” The Modern Language Journal 86: 54-66. Yngve, V. 1970. “On Getting a Word in Edgewise.” Chicago Linguistic Society 6: 567-578.

Appendix A Match the utterance on the left with an appropriate response on the right. An example is given (shown in italics below). 1. I’ve lost my wallet.͒_f_ 2. I passed my exam. ͒____ 3. My grandmother passed away.͒____ 4. I couldn’t get to sleep last night.͒____ 5. My grandmother feels much better now.____ 6. Thank you very much for the present.͒____ 7. Do you know the way to the train station?͒___ 8. I think Japan is the safest places to live.____ 9. Justin Beiber is coming to Japan.͒____ 10. My father is a carpenter.͒____ 11. This small town seems quite boring.͒____

Appendix B Saying you don’t know I’m sorry I don’t know. I’m not sure. I have no idea. Saying you don’t understand I can’t understand. (Could you give me an example?) I’m not sure what you mean. I’m not quite with you.

a. Really? b. Not really. c. Congratulations! d. You’re welcome. e. I don’t think so. f. Oh no! g. What’s that mean? h. I agree. i. That’s too bad. j. That’s good to hear k. I’m so sorry to hear that.

Towards a Pedagogical Framework to Develop the Listenership Disagreeing with someone I’m afraid I disagree. Sorry, I don’t really think so. Not me, I thought … Oh really? I didn’t think so. Finding out about meaning What does _____ mean? What’s another word for _______? What’s the opposite of ________? What’s the difference between _____ and _____? Does ______ mean _________? Finding out about appropriateness What should I say when …? When can I say …? In what situation could I …? Asking someone to repeat something Could you repeat that more slowly please? I beg your pardon. Pardon me. Asking about pronunciation How do you say ______ (in English)? How do you pronounce ______? Asking for the written form Could you write _____ down please? How do you spell _____ please? Finding about the correctness Is it correct to say ______? Can I say ______? Is this sentence correct? … Which sentence is correct? _____ or ______

319

320

Chapter Thirteen

Giving yourself time to think Just a moment please. May I think about it for a moment?

SECTION 4 SPEECH ACTS

CHAPTER FOURTEEN CRITERIA FOR THE IDENTIFICATION OF MOVES: THE CASE OF WRITTEN OFFERS SARA GESUATO

1. Introduction Speech acts have been the focus of attention of both theoretical pragmaticians and applied linguists for decades now. Theoretical pragmaticians are interested in the “actional power” of speech acts, which, by definition, enable language users to do things with words. Their goal is therefore to accurately describe what speech acts consist in, that is to identify, define and account for their communicative functions, pre-conditions and possible outcome. Particularly worthy of investigation are those speech acts that do not fall squarely into one class (e.g., expressives), since they call for a thorough and in-depth illustration of the verbal and non-verbal circumstances in which their performance makes sense to the interactants involved. Applied linguists are interested in how communicative practices affect the outcome of interaction from both a transactional and an interpersonal point of view. They point out what makes communicative events pragmalinguistically efficient and socio-pragmatically acceptable in specific linguistic and cultural contexts, and their goal is to enable language users (especially second/foreign language users) to accurately and appropriately produce, interpret and react to speech acts. Worthy of special attention are therefore those speech acts that are potentially dangerous to perform and respond to because they involve, or affect, interpersonal rights and duties (i.e., interactants’ needs, goals, options, constraints) and personal feelings (i.e., their preferences, likes and desires). Some speech acts combine both sets of characteristics, and offers are a case in point. On the one hand, they have a dual pragmatic nature, partly

324

Chapter Fourteen

directive and partly commissive (see section 2). That is, they are meant to influence the addressee’s behaviour – by having them accept a good/service – while announcing the addresser’s intention to act – that is to provide that good/service. In addition, the realization of their commissive force requires not only the offerer’s willingness to act, but also the interlocutor’s consent: the provision of the offered item is conditional on the addressee’s willingness to accept it. On the other hand, an offer is a potential minefield from the interpersonal point of view. Although intended as an act of liberality towards the addressee, an offer may backfire, for instance if the offered item does not match the addressee’s taste, or if the circumstances of its proposed provision (e.g., amount, time, place) place a burden on the addressee. That is, the acceptability, likeability and success of an offer partly depend on it being meant to be good for the addressee, and partly on its interpretation and conegotiation with the addressee (Kerbrat-Orecchioni 2004, 421-422). In this paper, I examine how interactants maximize their chances of interactional success when they realize offers in writing. My research questions are: A) what makes up an offer when this is the focus of the interaction? B) How can its component moves be defined? That is, what do they contribute to the overall strategy of the interaction? and C) How can its component moves be identified? That is, how do they signal their communicative functions? To address the above issues, I first provide a definition of the act of offering and present an overview on the literature on offers. Then I outline the data used as the object of analysis. There follows a description of the method of analysis I adopted. The findings include instantiations and a classification of the component moves of offers. In the discussion and conclusion, I propose a move structure analysis of written offers and highlight the pros and cons of the research method applied to this study.

2. Definition An offer is the verbalization of the addresser’s willingness to make a good/service available to the addressee, or a third party, intended as a benefit to the addressee. Since it expresses willingness to act, an offer is often an initiating, non-solicited act, but – depending on the context in which it is formulated – it can also be solicited by the interlocutor, and motivated by the larger interactional-transactional circumstances of the communicative event (see Chodorowska-Pilch 2002, 21). The intended provision of the good/service is presented as merely possible, although hoped-for, and thus announced rather than realized:

Criteria for the Identification of Moves: The Case of Written Offers

325

indeed, the addresser has to obtain the addressee’s consent before the offered item can be provided. For this reason, the offer can be defined as a conditional promise (Rabinowitz 1993, 24;1 see Wunderlich 1977, 42-43; Leech 1983, 217, 219), that is an act of commitment to do something on condition that the addressee’s approval of the future act is ensured.2 At the same time, if the addressee’s consent is obtained, the act of offering is binding; that is, not proceeding with the announced provision of the good/service would be considered a “breach of contract,” even if the “contract” arises out of self-imposed obligation.3 The offered item is regarded by the addresser as desirable and agreeable to the addressee, although not necessarily in their best interest (e.g., the provision of a cigarette): the goal is to benefit the addressee by pleasing them.4 Because its illocutionary goal coincides with the social goal of sustaining social harmony, an offer counts as a convivial act, which abides by the Generosity Maxim (i.e., minimizing benefit to oneself, while maximizing benefit to others; Leech 1983).5 First, an offer enhances the addressee’s positive face – it reveals the addresser’s awareness of and intention to match the addressee’s preferences besides projecting an image of the offeree as one worth making the offer to (Ruiz de Zarobe 2012, 177). Second, an offer also enhances the addresser’s positive face. That is, proposing to act generously towards the addressee – thus showing concern for them (Ruiz de Zarobe 2012, 177; Kerbrat-Orecchioni 2004, 422) – and revealing one’s ability to guess their tastes (i.e., to accurately judge on the appropriacy of the offered item) lead the addresser to earn social credits with the addressee, especially if the offer is accepted (Ruiz de Zarobe 2012, 178).6 However, an offer is also a competitive act (Leech 1983), that is one whose illocutionary goal competes with the social goal, since it fails to fully abide by the Tact Maxim of minimizing cost, while maximizing benefit to others (Leech 1983). Indeed, an offer threatens the addressee’s negative face because it imposes on them: A) the addresser manifests the desire to influence the addressee’s behaviour according to the addresser’s choices, even if these match the addressee’s preferences; B) the addressee is required to make up his/her mind about and respond to the offer; and C) if the offered item is provided, the addressee will be socially indebted to the addresser.7 At the same time, an offer threatens the addresser’s negative face: A) in offering a good/service, the addresser commits to a course of action that will set limits to his/her freedom, and B) if the offer is accepted, the addresser will have to invest resources (e.g., time, energy, money) on something not directly beneficial to him/her (see Havertake

326

Chapter Fourteen

1994, 108-115; Barron 2003, 125-126; Chodorowska-Pilch 2002, 23, 2003, 313; Kerbrat-Orecchioni 2004, 422).8 An offer, therefore, is a tentative commissive act (Yotsukura 1997, 6566; Barron 2003, 126), which comes with directive strings attached (see Hancher 1979, 6; Chen, Lei and Zhang 1995, 151; Kerbrat-Orecchioni 2004, 421),9 and its “fuzzy nature” (Aijmer 1996, 189) accounts for the large number of strategies used to encode it.10 As Schneider (1980), quoted in Barron (2003, 125), observes, an offer can be realized as: a preference question (i.e., “Would you like…?”), which highlights its conditional nature; an execution question (i.e., “Can I do X for you?”), which highlights its commissive nature; or an imperative (i.e., “Have X”), which signals its directive nature. Indeed, its successful performance requires a consideration of its dual illocutionary nature. With regard to its propositional content, an offer predicates a future, albeit tentative, act of the addresser (i.e., the provision of a good/service), as is the case with commissives, but this act also requires the addressee’s transactional cooperation (i.e., the acceptance of the good/service), as in the case of directives.11 Similarly, the preparatory circumstances applying to the performance of an offer highlight the commissive-directive nature of the act. Relevant to commissives are the following: A) the addresser’s ability and willingness to do the proposed act; B) the addresser’s belief that the performance of this act might correspond to the addressee’s wants or wishes; C) the addressee’s ignorance about the addresser’s ability and willingness to act. Additional circumstances, more relevant to directives, include: D) the non-obvious nature of the proposed act, which requires the addressee’s approval (Yotsukura 1997, 67-70); E) the addressee’s ability to perform the required co-act; F) the addresser’s knowledge about it; G) and the non-obvious nature of the performance of that co-act, which indeed requires the act of offering itself as a prompt. The sincerity of an offer involves both the addresser and the addressee: on the one hand, A) the addresser does intend to perform the proposed act (as is the case with commissives in general), although B) it is not clear whether this corresponds to the addressee’s preferences (which marks this kind of commissive as tentative); at the same time, C) the addresser hopes the addressee will perform the co-act of accepting the offered item (as is the case with directives).12 Overall, the offer consists in the undertaking of a tentative but binding commitment to do the proposed act, as well as an attempt to get the addressee to co-operate with the addresser towards the complete fulfilment of the proposed act.

Criteria for the Identification of Moves: The Case of Written Offers

327

The felicity conditions of offers are summarized below (cf. Barron 2003, 126):13 • •

•

•

A) Propositional: - i) a future act of the addresser - ii) a future co-act of the addressee B) Preparatory: - i) the addresser is able and willing to do the proposed act - ii) the addresser believes that the addressee might want him/her to do the proposed act - iii) the addressee does not necessarily believe that the addresser is able/willing to do the proposed act - iv) it is not obvious to both the addresser and the addressee that the addresser would do the proposed act without the addressee’s prompt, permission or consent - v) the addressee is able and willing to perform the expected co-act - vi) the addresser knows that the addressee is able to do the co-act - vii) it is not obvious to the addresser and the addressee that the addressee would do the co-act unless prompted by the addresser C) Sincerity: - i) the addresser wants to do the proposed act - ii) it is not clear that the addressee wants the addresser to do the proposed act - iii) the addresser wants the addressee to perform the coact D) Essential: - i) the utterance counts as the undertaking of a tentative commitment to do the proposed act, which is binding, should the addressee want the addresser to act that way - ii) the utterance counts as an attempt by the addresser to get the addressee to perform the co-act.

Offers are complex illocutionary acts because of their twofold pragmatic nature (see Zilles Pohle 2007, 202). They are both faceflattering and face-threatening acts (Kerbrat-Orecchioni 2004, 422). They address and affect both parties’ preferences (i.e., what they want for themselves and from each other) and future courses of action (what they may choose or have to do next). As a result, the managing of offers (i.e.,

328

Chapter Fourteen

performing and reacting to them) calls for the tactful handling of the interactants’ interpersonal needs and goals.

3. Literature review Previous studies have analysed several aspects of the speech act of offering, namely A) the content, lexico-syntactic formulation and positioning of the head acts of offers in conversations, in private and in business contexts, B) the content and formulation of offers of courses by private and public institutions on the Internet, C) the role of offers in balancing social debts and credits, D) cross-linguistic manifestations of offers, and E) the learnability of offers in an L2 context.

3.1. The realization of offers in conversation Rabinowitz (1993) explored the illocutionary and perlocutionary nature of the speech act of offering, provided a motivated definition of it, outlined the possible realization of offer sequences in oral interaction, and identified and described the social and linguistic parameters typical of offers in US middle class interactions. She noticed that in that context the offered item was mostly food; that most offers (i.e., 91%) were produced spontaneously; that most offerers were female; and that offers tended to be characterized by such linguistic features as rising intonation and the use of the verbs want, like or need. The author also observed that the interpretation of an utterance as an offer was mainly affected by the context of situation, which could override linguistic cues (i.e., its overt verbal encoding). Roulet (1977)14 investigated the syntactic-semantic realizations of offers in French, which include declarative, interrogative and imperative structures. He observed that the frequent realization of offers as indirect speech acts involves stating or questioning one of their felicity conditions, except the essential one, and more specifically, stating what the offerer already knows for sure (i.e., his/her ability to act, his/her willingness to act, the reference to a future act; e.g., Je peus vous accompagner; Si cela peut vous arranger, je vous accompagne) or questioning something the offerer is not yet sure of (i.e., whether the offeree wants the offered item, whether he/she will let the offerer provide it, and whether he/she considers it useful; e.g., Voulez-vous que je vous accompagne?; Me permettez-vous de vous accompagner?; Ça te serait utile que je te passé ma calculatrice?; p. 537). Finally, the author observed that the great number of indirect realizations of offers signals that offering is a non-solicited act, which

Criteria for the Identification of Moves: The Case of Written Offers

329

intrudes into the discourse, such that the offerer may feel apologetic about it (see previous examples). Conein (1986) examined offer sequences in French telephone dialogues, the term offer comprising such acts as solicitations, suggestions, invitations and indirect requests (p. 115). He observed how a so-called offer cannot, in fact, be defined as such unless the response to it is also taken into consideration. The author also noted that the frequent encoding of the first interlocutor’s contribution to discourse as a question, especially with a verb of volition in the second person, highlights the uncertainty of the future course of event being mentioned, which thus appears negotiable and subject to change. This way, the first interlocutor can inquire about the felicity conditions of the offer, and the second interlocutor can report on its relevant circumstances; this allows them to avoid an explicit, dispreferred, refusal. In particular, Conein also pointed out that the first interlocutor’s contribution to discourse calls for a response which opens up a range of options for the second interlocutor. These options include, quite rarely, acceptance or rejection, and, much more frequently, mitigated responses such as: reference to a problem, feigning acceptance and then changing one’s mind, expressing a condition for accepting the offer, inquiring about the contextual options/constraints of the offer, making a counter-offer. These options realize a form of delayed or conditional acceptance which requires the first interlocutor to further motivate the offer. Conein also explained the preference for mitigated, qualified responses on behalf of the second interlocutor as being due not only to rejection being a dispreferred response requiring qualification, but also to immediate acceptance being interpretable as socially stigmatized eagerness to receive a benefit. Overall, he defined offers as contextually informed by the local interactional activity carried out by the participants. Adopting a CA perspective, Davidson (1984, 1990) examined the types and sequencing of turns produced by two interlocutors engaged in the production of, and reactions to, offers.15 The analysis of a series of conversational exchanges showed how interlocutors constantly monitored each other’s utterances and negotiated their interactional roles; for instance, A) an offer set up a relevant next turn that could be an acceptance, a rejection, a weak agreement or a silence; B) a rejection could be accepted and finalized by the offerer, or give rise to a (series of) modified versions of the previous offers; C) these modified re-offers, in turn, called for the same types of relevant responses (acceptance, rejection, weak agreement, silence); D) a silence or a weak agreement was interpreted as a pre-rejection (i.e., the manifestation of some problem), which gave rise to a modified version of the previous offer or, in the case

330

Chapter Fourteen

of silence, to its extension beyond its first possible completion point; E) a rejection that met with silence or a rejection could give rise, respectively, to its extension beyond its first possible completion point or to modified version(s) of the original rejection. The interactants’ turns thus appeared to cyclically re-instantiate the same broad types of interactional moves; this was mainly due to the addresser’s noticing the display of, or deriving the implication of, an actual or possible dispreferred response from the addressee. This often led to a reformulation of the addresser’s original utterance, and the consequent provision of a place for another response from the addressee. Fukushima (1990a) examined the structure (i.e., number, type and sequence of functional components) and syntactic formulation of offers elicited in writing from native speakers of English by means of an openended questionnaire (i.e., a discourse completion task) illustrating interactional scenarios in which the degree of social distance between the interlocutors varied (from low through medium to high). The author observed that the most frequent realization consisted of the head act only, but that the degree of indirectness (i.e., syntactic elaboration) increased as social distance increased. Aijmer (1996, 189-195) overviewed the large number of strategies that can be employed in expressing offers in oral interaction. Her corpus-based analysis showed that these included: a commissive realized as a firstperson statement in the future (i.e., “I/we WILL VP”; e.g., I’ll buy a cup of tea);16 a request for permission realized as a first-person modalized interrogative (e.g., Can I give you X?); an inquiry about the addressee’s preferences realized as a first-person modalized interrogative (e.g., Shall we VP?); an inquiry about the addressee’s wants realized as a secondperson want interrogative (e.g., Do you want/like/need X?); a strong suggestion encoded in a let imperative (i.e., Let’s V or Let me V); a strong suggestion realized as a declarative sentence with a (semi-)modal of obligation like have to or must (e.g., You will have to come over). Accompanying moves were found to justify the offer (e.g., It might be interesting), inquire about its acceptability (e.g., if you can), express the offerer’s willingness to be of help (e.g., You’re welcome; I’m at your service), negotiate aspects of the offered item (i.e., the time) possibly leading to a joint social commitment, be preceded by a pre-sequence setting the stage for and signalling the upcoming illocution (e.g., I’ve just boiled some water. Would you like some tea?). Finally, Aijmer also listed typical responses to offers of minor favours, namely: requesting the hearer to act (e.g., Please); accepting the offer (e.g., Ok); expressing appreciation (e.g., Thanks), expressing intention (e.g., I will).

Criteria for the Identification of Moves: The Case of Written Offers

331

Curl (2006) described the phrasing of head acts of offers in phone conversations. She observed that different formulations of offers (e.g., “Can I X?”; “Do you want me to X?”; “If you want X, then I will X”) showed two correlations: one was with the goals and constraints of the interactional situation (e.g., whether an offer of assistance was the reason for calling or was generated within the interaction; whether the targeted problem was overtly or covertly expressed); the other was with the placement of the offer in the interaction (i.e., its sequencing with respect to other turns) and the content of the preceding turns. Chodorowska-Pilch (2002) examined the linguistic mechanisms encoding politeness in Spanish offers produced in travel agencies, and the contexts in which this polite attitude is conveyed. She found that mitigating strategies in offers encoded deferential politeness, and involved the indirect encoding of the offers themselves, or their combination with other illocutions; typical realizations included indirect questions, suggestions, explanations; the use of the verbs querer and poder; conditional and impersonal constructions; future, present subjunctive and imperfect verb forms, and politeness markers (e.g., ¿me entiendes? in the case of problematic circumstances not (totally) under the offerer’s control). Chodorowska-Pilch (2003) analysed the linguistic mechanisms employed to codify polite offers in Spanish. The author observed that, although some offers were encoded directly, that is by means of the performative offrecer (i.e., “to offer”), most were expressed indirectly through such strategies as: indirect questions (usually in association with the verb poder “can”); expressions of hope (i.e., a ver sì “let’s see whether”);17 conditionals of the form (e.g., Si quieres, te lo averiguo “If you want me to, I’ll look into it”); the politeness marker si quieres together with imperatives or declaratives (e.g., Siéntate, si quieres. “Have a seat, if you like;” Te hago copias, si quieres “I’ll make you the copies, if you want”). The author also identified additional forms of mitigation in offers, which showed deference to the addressee – through the avoidance of imposition or the giving of options – such as the use of the imperfect tense, future tense, conditional mood, subjunctive mood as well as conversational implicatures. Yotsukura (1997) adopted a CA approach to the study of Japanese business phone conversations in which the caller reports a customer service-related problem and the callee offers assistance in response to that. The author identified a recurrent five-part structure to such transactional conversations: an opening section (e.g., self-identification, greeting); a transition section expressing a general reason for calling; the report of a

332

Chapter Fourteen

problem; a summary of the matter with an offer of assistance; and a closing section. The study therefore highlighted the reacting nature of offers produced in customer service phone interactions. Zilles Pohle (2007) examined offers in Irish English and German business negotiations, in terms of their transactional function, linguistic formulation and topics addressed. The author noticed that in business transactions, offers were proposals of the provision of goods/services, which also involved the expectation of reciprocation, so that both parties could benefit from the exchange. They were therefore interested offers. Their encoding options included: the frequent use of conventionally indirect formulas, the use of performative verbs, reference to their felicity conditions, or the expression of joint (i.e., both parties’) commitment to action. Their topics addressed business details such as change in price, procedural actions smoothing the negotiation process, but also the longterm relationship with the business partner. Chen, Lei and Zhang (1995) explored the content and formulations of the responding act of refusal, elicited through a questionnaire, produced after various types of initiating acts, including offers. The authors noticed, first of all, that all refusals in Chinese were meant not only to be effective, but also to preserve the face of the refuser and the refusee, and secondly that refusals could be substantive or merely ritual. Next, the researchers identified and classified the following refusal strategies from the semantic point of view: direct refusal (i.e., denial of compliance), expressing regret, giving reasons, suggesting an alternative course of action, stating a principle conflicting with compliance, using folk wisdom, attempting to dissuade the interlocutor, verbalizing limited acceptance, and avoiding a direct response. The scholars observed that the refusal strategies most often used after offers were, in this order: dissuading the interlocutor (which expressed the notion that the offer was perceived as appropriate although not needed), expressing a direct refusal (which showed the refuser’s concern about reducing the refusee’s cost), and giving reasons (which could hurt the refusee’s face precisely because it justified, and tried to legitimize, the refusal). The authors also noticed that in the case of ritual refusal (i.e., pretended non-compliance, which showed modesty and gauged the interlocutor’s sincerity), the refusers gave reasons that considered the cost of the offer to the refusees. Finally, they pointed out that an offer-response interaction involved a structured, conventionalized multi-turn sequence where both interlocutors negotiated their communicative/transactional goals, in the case of both substantive and ritual refusals.

Criteria for the Identification of Moves: The Case of Written Offers

333

3.2. The realization of offers by institutions on the Internet Ruiz de Zarobe (2012) examined informative-commercial offers advertising courses, posted by Spanish and French organisations on the Internet. The author analysed the offer strategies of the head acts in terms of their lexico-syntactic formulation, the types and combinations of the information notions expressed, and the semantic roles assigned to the entities and phenomena mentioned therein. Five strategies were identified. The first strategy, the performative verb in the third person (e.g., te ofrece Z; propose Z), is unthreatening and explicit in its offer of a benefit, and is the only commissive one. The second strategy, the possibility statement (e.g., usted podrá X; vous pouvez X), requests the addressee’s action (i.e., looking at the courses on offer); it thus resembles a directive, although it gives ample margin of options to the addressee on how to respond. The third strategy, a future state of the addressee (e.g., Encontrará aqui W; Vous trouverez ci-dessous W), presents the offered item as being of obvious interest and already made available to the addressee. The fourth strategy, the imperative (e.g., Consulta Y; Découvrez Y), clearly directs the addressee toward the realization of actions (e.g., looking for information) and the experience of events (e.g., discovering benefits/information). The fifth strategy, a pre-offer that announces the actual offer (e.g., Vous souhaitez H?; Busca H?), inquires about, or occasionally states, the addressee’s desires or interests; since it makes the addressee’s need explicit, the addressee can feel induced to recognize it and then accept the offer. The first three strategies were more frequent in the Spanish data, while the fourth and fifth were more frequent in the French data. The author noticed the influence of asynchronous communication on the formulation of these offers: A) the compulsory nature of the offerer’s action was taken for granted and shown to be already fulfilled when the reader accepted the offer, rather than conditional on the offeree’s acceptance of the offer. B) The offerers appeared to be sure of the addressees’ desire. C) The involvement of the addressee in the joint action of the offer (i.e., searching for information on web pages, but especially taking the initiative in looking for relevant web pages) was an effect of the medium of communication. In general, the offerers assumed that the offers were of interest and beneficial to the addressees, and so their offers contained no mitigation and are rich in positive politeness strategies.

334

Chapter Fourteen

3.3. The role of offers in balancing social debts and credits Koutlaki (2002) examined the role of offers in the context of Persian ritual politeness (called tæ’arof), which pivots on two notions of face: “pride” and “honour.” She explained how in Persian, an offer reflects and enhances both interlocutors’ positive face: on the one hand, it enables the offerer to appear to be generous, and on the other, the offer and the reaction to it allow interlocutors to express their recognition and adherence to the socially sanctioned rule of interdependence. Taleghani-Nikazm (1998) analysed the preference format of offers in Persian. The author observed how rejections, which are considered polite, have the marks of preferred seconds (i.e., they are delivered immediately upon completion or in terminal overlap with the offer turn), while acceptances, which are considered impolite, have the marks of dispreferred seconds (i.e., they incorporate features of delay). Astruc, Vanrell, and Prieto (forthcoming) examined the contribution of intonation to the encoding of politeness in offers and requests in Catalan. Discourse completion tasks were used to elicit speech acts in scenarios controlled for level of social distance, power, and cost of the action). The findings regarding offers showed that: A) in high-cost offers, speakers preferred questions with a rising pattern (that is, high-cost situations triggered more rising pitch patterns); B) a falling pattern applied to the majority of the offers (61%); C) most offers were encoded as interrogative clauses (67%); D) in high-distance offers, that is offers to strangers, speakers were more likely to use more frequently deferential and indirect language (e.g., the V form (vostè “you, polite”) combined with a conditional or imperfect verbal form); more generally, E) high-power situations (i.e., situations involving a higher social and work-related risk such as addressing the company’s director at a work lunch) also elicited more occurrences of deferential and indirect language than low-power situations (e.g., addressing a colleague at a work lunch); finally, F) highcost situations elicited fewer instances of deferential and indirect language. Edmondson and House (1981, 136-142) examined the illocutionary and interactional role of offers in oral communication. They observed that offers are illocutions that initiate exchanges, which thus require a responding move. The authors also noticed that offers can be realized tentatively (e.g., I can lend you mine if you want,18 p. 136), directly (e.g., Help yourself, p. 137) or even authoritatively (e.g., I’ll take you in my car, p. 137). The authors further noticed that offers can be produced A) after a complaint by the interlocutor, as a way to compensate them for the trouble caused to them; B) after a request as an alternative to, and compensation for, the inability or failure to undertake the action required of the current

Criteria for the Identification of Moves: The Case of Written Offers

335

speaker (e.g., A: Have you got the OED there / B: Sorry it’s at home – I can bring it for you tomorrow if you like, p. 138); C) after the interlocutor has made an offer, as a form of reciprocation (e.g., I’ll buy you one next time, p. 141); D) as a supporting move of a request; and E) after the interlocutor has granted permission to do something as a way of showing that that permission will not be abused (e.g., A: Can I borrow your black dress for this evening / B: Sure / A: Thanks – I’ll look after it – but if it gets dirty I’ll have it cleaned for you, p. 142). Kerbrat-Orecchioni (2004) examined the formulation of offers and reactions to offers within the context of face work. She observed how offers can be formulated “bluntly” through imperatives, possibly accompanied by some form of intensification, if they are focused on their face-flattering aspect, or in a softened way (e.g., through a volition question), if they are focused on their face-threatening aspect. The author made several observations. A) both acceptance and refusal, the envisaged reactions to an offer, are often accompanied by an act of thanking, which focuses on the cost of the offer to the offerer. B) The preferred response to an offer, namely acceptance, can be realized directly, but is more often characterized by the markers of a dispreferred response (e.g., hesitation, delayed response, softeners, reference to a conditional constraint). This property is due to the double bind experienced by the offeree: their need, on the one hand, not to offend the offerer, and on the other, not to take advantage of their generosity. C) Since a refusal is a face-threatening act, it is often accompanied by some softener such as an act of temporary acceptance. This makes it possible for the offerer to reassert the offer, stressing its sincerity, which gives rise to a step-by-step negotiation with the addressee, a negotiation leading to a final acceptance or at least a mutually acceptable rejection. The author concluded that the variety of formulations of offers and the reactions to them is due to “la complexité de leur statut par rapport au système des faces” (Kerbrat-Orecchioni 2004, 428).

3.4. Cross-linguistic realizations of offers Kasper (1981, 140-151) compared the realization of offers/invitations – treated together – in English by English native speakers and German learners. She noticed that the German learners performed these speech acts less often than the native speakers, and also that about 47% of the time the offers/invitations produced by the German learners were characterized by errors such as an inappropriate choice of register, modality and/or level of

336

Chapter Fourteen

directness, and/or by too explicit a formulation of the propositional content. Fukushima and Iwata (1987) studied offers in Japanese and English. The authors found that in both languages, the offerer tried to maximize the benefit to the offeree, tried to shorten the perceived distance between the interlocutors (i.e., by making the offeree feel good), and tended to use set phrases. They also found that the Japanese offerer made more distinctions in strategies and expressions depending on the interlocutor’s status, and more often exploited the Modesty Maxim than the English offerer did. Matoba (1996) investigated the encoding of offers in German and Japanese. The strategies identified were first divided in terms of whether they referred to the speaker’s or hearer’s action or state, and next in terms of whether they explicitly or implicitly expressed their commissive goals. Explicit commissives included: declaratives (e.g., I want to offer X; I will do X), interrogatives (e.g., Can I give X?; Shall I do X?), and wish interrogatives (e.g., Do you wish X?). Implicit commissives included: nonimpositive (e.g., beneficial to the hearer) imperatives (e.g., Do X), permission declaratives (e.g., You can do X), intention interrogatives (e.g., Do you want X?) and hints (e.g., I hope you like X). The author also investigated correlations between the formulation of offers and the perceived distance from (i.e., degree of inhibition/carefulness felt toward) the interlocutor in discourse completion tasks. In German, the higher the perceived distance, the more frequent the choice of a speaker-oriented perspective was. In Japanese, the hearer-oriented perspective was the most frequent choice independently of the perceived distance. The choice of the referential perspective was said to reflect socially and culturally learned values. Tsuzuki, Takahashi, Patschke and Zhang (2005) examined offers in English and Chinese. They observed how in both languages, positive politeness, as encoded through an imperative, was preferred when the addressee was close and equal, while negative politeness, as encoded through an interrogative, was preferred when the addressee was not close and/or in a higher position. Wierzbicka (1985) explored the different cultural assumptions underlying the conceptualization and formulation of speech acts, including offers, in English and Polish. She observed how the preference for encoding offers through interrogative forms in English can be related to the Anglo-Saxon implicit social norms placing a high value on the rights and autonomy of the individual. On the other hand, the author pointed out that the preference for the imperatives and diminutives in Polish signals that the offerer insists on doing what they think is best for the addressee,

Criteria for the Identification of Moves: The Case of Written Offers

337

regardless of the latter’s (expressed) desires, while downplaying the magnitude of the offered item. Astruc and Vanrell (forthcoming) examined the pragmalinguistic realizations and intonational cues of politeness in offers/invitations19 and requests used by native speakers and by learners of Spanish. They found that offers were overwhelmingly realized as interrogative utterances (83%) and associated with volitive verbs (querer, gustar, 67%), but also that high-cost offers were most often realized as declaratives and that epistemic verbs (poder, 15%) and possession verbs (pasar, dar, 9%) were also instantiated. The offers were frequently accompanied by the low-rise tone (61%), which positively correlated with low cost offers, and less frequently by the high-rise tone (36%), which positively correlated with high cost offers. Overall, the authors noticed that the learners transferred the intonational patterns of their first language, in particular with more frequent use of high rises than the native speakers and also with the use of falling intonation patterns.

3.5. The learnability of offers in an L2 context Barron (2003) examined Irish learners of German’s development of L2 pragmatic competence over a 10-month abroad period. This involved, among other things, a gradual but partial approach to the L2 offer-refusal exchange structure occurring in oral interaction. This type of exchange comprised an initiation (the formulation of a first offer) by one speaker, a refusal by the interlocutor, a single re-offer by the speaker, and the eventual reaching of an agreement with satisfaction of the provision and acceptance of the offered item. Fukushima (1990b) examined Japanese learners of English’s performance of offers. The author found that pragmatic failure occurred in the formulation of the head acts, which were not often appropriate to the illocution, the situation, or the degree of closeness with the addressee, and thus sounded like commands or too informal.

3.6. Rationale of the current research Of the above studies, one that carried out an in-depth move structure analysis of instances of offers is Barron (2003). By following the approach introduced by Edmondson and House (1981) for describing the structure of oral interactions, the author illustrated and accounted for sequencing and functions of turns in so-called offer sequences (i.e., acts of offering, and reactions to them) in terms of a few moves: Initiate, which initiates an

338

Chapter Fourteen

exchange; Satisfy, which produces an outcome; Contra, which is an attempt to get the interlocutor to withdraw the offer; Reoffer,20 which is an attempt to reiterate an offer;21 and Counter, which is an attempt to get the interlocutor to modify or qualify the preceding offer. However, her goal was not to describe how offers are carried out, but rather how they are organized in terms of an exchange structure involving two interlocutors (Edmondson and House 1981, 130-131).22 Also, while she investigated issues of internal modification, she did not examine the external modification of offers. Similarly, Rabinowitz (1993, 120-129) analysed the responses to offers that may occur in interaction, and identified these types: voiced acceptance, implied acceptance, voiced refusal, implied refusal, noncommitment, mentioning an indirectly related topic, switching to an unrelated topic and ignoring the offer. She then outlined the internal turntaking structure of longer offer-response interactional sequences, identifying these types: A) offering, accepting, giving, thanking; B) offering, refusing, switching to a new topic; C) offering, weak agreement (or refusal or silence), repeated offering, accepting, giving; D) offering, refusing, elaborating on the offer, repeated refusal, repetition of the elaboration; E) offering, requesting clarification, repeating the offer, refusing; and F) offering, failing to acknowledge the offer. In addition, Aijmer (1996) examined the conversational routines taking place in oral interactions realizing and responding to offers, and identified moves accompanying the offering head acts (e.g., signalling the upcoming illocution, justifying the illocution, inquiring about its suitability, expressing one’s willingness to be of help, negotiating the terms of the offered item). Finally, to my knowledge, the only study that has examined written offers has only focused on the head acts of offers produced by educationalacademic institutions on the Internet, without considering their larger cotext (Ruiz de Zarobe 2012). The research so far, therefore, has described in part the behaviour of communicative participants who produce offers and react to them during larger oral interactional exchanges. However, previous studies have not examined other manifestations of offers, namely: A) offers realized as extended stretches of discourse; B) offers expressed in writing in a private context; C) the principles for defining and describing the component moves of offers. The present study aims to contribute to answering the above questions: A) it illustrates how offers are instantiated when their illocution is the only or main focus of the communicative event in which they occur how; B) it describes how offers are encoded in writing by

Criteria for the Identification of Moves: The Case of Written Offers

339

individuals in a private context; C) it presents a taxonomy of the moves making up offers, whose applicability and validity can be tested; in particular, it innovatively offers both identification and definition criteria of the offering head act and its external modifications so that they may be liable to verification and improvement in future studies.

4. Data The data considered for the study comprises 37 texts (3,000 words), elicited under experimental conditions. The elicitation material consisted of written descriptions of interactional scenarios based on real-life events (experienced by me as a participant or witness). These scenario descriptions called, but did not explicitly prompt, for the realization of offers, and were slightly adapted so as to be suitable to the written medium. The scenarios described situations involving interactants in different role-relationships in terms of social distance (close vs. distant) and degree of power (equal vs. subordinate vs. superior). A total of six sets of scenario descriptions was thus assembled. These were meant to elicit texts addressed to: intimate and equal interlocutors (set A: six scenarios); distant and equal interlocutors (set B: two scenarios); intimate and subordinate interlocutors (set C: six scenarios); distant and subordinate interlocutors (set D: two scenarios); intimate and superior interlocutors (set E: three scenarios); distant and superior interlocutors (set F: one scenario).23 Sample scenario descriptions are reproduced in the Appendix. The study participants were native speakers of English attending a US university, and were rewarded with a small amount of money for their participation. During each elicitation session, I asked each participant to choose three scenario descriptions (one each from sets A, B and C, or from sets D, E and F),24 and to react in writing to them by imagining to be directly involved in the situations therein described. The participants were not given any time limit to complete the task or any indication as to the expected length of their texts. They were also told that they could revise and edit their texts until they felt satisfied, or delete them, or more simply opt out of any or all parts of the task, with no questions asked. Sample texts are provided in the Appendix. Of all the texts elicited, four had to be discarded. Although they were in line with the scenario descriptions, that is, they instantiated plausible verbal reactions to the situations depicted therein, they realized other kinds of speech acts than those expected (e.g., requests, thanks, compliments).

340

Chapter Fourteen

5. Method The data was analysed through a combined qualitative and quantitative approach. The qualitative approach involved, first of all, familiarizing myself with the texts through repeated readings. This led me to a preliminary, inductive detection of the moves that could be plausibly said to be instantiated in 10 texts, on the basis of the content and lexico-grammatical wording of text segments (i.e., recurrent encoding of given notions through similar phrases). Next, I provided tentative definitions of the communicative functions of these text segments. I then applied the inductively developed coding scheme to the remaining data so as to test its validity and, if need be, modify the description of the content and/or wording of the moves so as to better account for the whole data. Next, I refined the move coding scheme by developing identification criteria of moves, in addition to the definitions already outlined. While definitions specify the primary function of the moves, the identification criteria highlight clues for their recognition “on the surface of the text.” They comprise four parameters: A) content: a description of the specific content (i.e., steps) making up a move (semantic parameter); B) wording: an indication of the likely or possible lexico-syntactic encoding of the move (grammatical parameter); C) sequencing: an indication of the positioning of the move in the overall text and with respect to other moves (textual parameter); and D) heuristics: an interrogative – or, less frequently, declarative – prompt for detecting the types of functions and the types of information characterising text segments (functional parameter).25 The goal of the qualitative analysis was to characterize the communicative nature of the offers examined, by specifying the micro-functions realized in their component text segments. In the next phase of my analysis, I counted move tokens so as to determine their frequency of occurrence in the whole corpus, their distribution across the texts, and their position in each text with respect to other moves. The goal of this quantitative analysis was to outline the most recurrent strategies for realizing offers and to describe the typical textual structure of this speech act.

6. Findings The qualitative analysis of the texts led to the identification of a set of micro-level strategies collectively contributing to the realization of offers. These moves qualify offers as such, outlining their communicative profile.

Criteria for the Identification of Moves: The Case of Written Offers

341

The quantitative analysis led to a description of the degree of similarity among tokens of the act of offering. This describes offers as members of the same speech act type, but not necessarily to the same degree.

6.1. Move types Eleven types of component moves are instantiated in the texts examined. They are presented in the order in which they tend to appear in the texts (see section 6.2). ALERTER is the move that establishes contact with the addressee. It may include an attention-getter, a greeting and/or the addressee’s name. It can therefore be realized as a proper name; a general noun phrase, comprising a title or adjective used as an address term (including an endearment term); an interjection, possibly in combination with a noun phrase; or finally a prepositional phrase, when the addressee is a group (i.e., in the case of notices meant to be posted on bulletin boards). Examples are “Hey”26 or “Hi” or “Hello” (+ Name); “Hey Mom”; “Dear Dr. [...].” The text segment realizing an ALERTER can be identified if it can provide relevant answers to such questions as How does the offerer call for the offeree’s attention? How does he/she (try to) establish contact with the offeree? Who is the offeree? What is the interlocutors’ likely social relationship (e.g., formal, intimate)? OPENER creates a textual and/or interactional bond with the offeree. Content-wise, it may involve the offerer’s introducing oneself, inquiring about the offeree’s health or somehow engaging in ice-breaking small talk; alternatively, it may consist in reacting to previous interactional input, that is, making reference to previous communicative exchanges. This move can be realized as a declarative or interrogative clause or group of clauses. Instances of it are “How’s it going?”; “I hope you have been doing well.”; “Thank you for your email this morning”; “I’d be happy to give you some feedback on your paper”; “My name is Dr. […].” Given its ice-breaking function, this move can be expected to occur at the beginning of the text. Relevant heuristic prompts for this move are How does the interaction get underway? or How does the offerer get it started? INTRODUCER introduces the topic the offer will be relevant to, but does not hint at the upcoming illocution. It mentions an event or situation involving or relevant to the offeree, mostly encoded as a declarative accompanied by an evidential. Examples are “I am moving out of town because I just accepted a new job”; “I’ve heard about the interview”; “I am glad to hear that the home repair is going so well.” As the move that paves the way for the illocution, it is likely to occur at the beginning of the text.

342

Chapter Fourteen

Relevant interrogative prompts are What is the offerer writing about? and What is the offerer bringing up for the offeree to consider? Or How does the offerer set the stage for the illocution? PRE-CONDITION is the move through which the offerer mentions or inquires about the pre-requisites that make the offer possible, feasible and sensible. Through this move, therefore, the offerer refers to his/her possession of a good, or his/her willingness or ability to provide a good/service; alternatively, he/she may inquire about the offeree’s possible interest in, preference for, or willingness to accept a good/service. In the latter case, the offerer sounds out the addressee about the suitability either of the offering illocution or of the realization of the act mentioned in the proposition by showing consideration for the addressee’s preferences. PRECONDITION can thus be realized as an if-clause or interrogative clause (i.e., a volition, preference or interest query), a declarative clause (a statement of fact or feasibility), or a combination thereof. The corpus includes the following examples: “Also, if you need someone to baby-sit Maggie that day”; “I have the time available”; “And knowing that my brother works as an interior designer”; “Interested in purchasing sustainable and local fruits & vegetables?” As it refers to the conditions that must be satisfied for the provision of the offered item, or at least for the offer to sound feasible and sensible, it can be expected to be realized before the head act. The text segment realizing this move can be identified by means of such prompts as A condition exists such that it makes sense to express the offer and Does a condition exist such that it makes sense to express the offer? HEAD-ACT realizes the offer by expressing the illocution and/or its object, or possibly by stating a pre-condition or perlocution of it.27 By default, therefore, this move manifests the offerer’s readiness, intention or ability to provide a good/service, or informs about the actual provision of it; alternatively, it may simply refer to a good/service to be provided to the offeree or even prompt the offeree to produce a request for a good/service. Because of the possible variety of its content, the move can be realized as a declarative (i.e., statement of the offerer’s attitude, volition or possibility/ability to act, possibly with a performative; or a statement about the provision of a good/service); an interrogative (i.e., a volition query); an imperative (i.e., a request for action), or a combination of options. Relevant examples are “and would be more than happy to look after her;” “Im28 free to watch Lisa,” “Im giving away a chest, a bookshelf and a filing cabinet,” “Would you like me babysit the munchkin?;” “Take them;” “I have a chest of drawers, a bookshelf, and a filing cabinet I’d like to give away.” Given its crucial role in the text, the HEAD-ACT may occur at the beginning of the text itself, to maximize communicative

Criteria for the Identification of Moves: The Case of Written Offers

343

effectiveness (i.e., to immediately let the reader know what the topic brought up for consideration is all about). This move provides relevant answers to the following queries: What does the offerer aim to make available to the offeree? What does he/she want to do, or what can he/she do, for the offeree? MOTIVATOR justifies the offer by making explicit its appropriacy, reasonableness or feasibility. It may mention a problem faced or a need to be met by the offeree, a goal the offerer or the offeree may want to pursue, or the offeree’s previous generous, commendable behavior. Alternatively, it can refer to the likeability of the offered good/service to the offeree or a negative consequence to avoid through the offered item. Finite and nonfinite declaratives, prepositional phrases and combinations thereof realize this move. Instances of it are “i know you will have a lot on your hands during this time;” “and as a result I have a bunch of old furniture that I can’t take with me;” “to help you out with shopping.” Because its communicative role is to rationally sustain the illocution, it may be expected to occur after the HEAD-ACT, that is, once its logical contribution to the discourse becomes more easily interpretable. The move directly addresses the point of the text by answering such questions as What makes the offer relevant and beneficial to the offeree?, What need or problem does it address? and What benefit does it bring? BACKGROUND mentions or inquires about the characteristics of the offered good/service, or the logistics of its provision. It illustrates aspects characterizing (the provision of) the offered item or refers to the contextual options and constraints applying to the realization of and reaction to the offer. BACKGROUND can be encoded through a full or ellipted declarative, a phrase or a combination thereof. Relevant instances are (“detergent, toothpaste, rice, dry pasta”; “next week”; “My shift at the theatre doesn’t begin until 6”; “There isn’t much I can do from email”; “while you go on your interview”; “sorry, no delivery – you must pickup)!”. Because it specifies details of the object of the illocution or its giving, it can reasonably be expected to occur after the HEAD-ACT has been realized. Several interrogative prompts are suitable to this move such as Where/ when/ how is the offer realized?, Where/ when/ how is the offered item (to be) used? and What is the offered item like? ACCEPTANCE-MAXIMIZER is a strategy for increasing the chances of the offeree’s accepting the offer. It may involve offering incentives likely to ensure the offeree’s cooperation, reassuring the offeree about the feasibility of the offer or urging the offeree to accept the offered item. More specifically, A) it can point out the negative consequences of failure to accept the offered item or highlight its beneficial value; B) it can pre-

344

Chapter Fourteen

empt the addressee’s objections by reassuring about the envisaged positive outcome or effect of the offer, the negligible cost – or the cost-free nature – of the offered item, the sincerity of the offer, the offerer’s ability to provide the offered good/service, the pleasure experienced in committing to the offer or the offeree’s positive attitude toward the offered item; and C) it can re-instantiate the offer by repeating it, by requesting the offeree to accept it or by checking whether the offeree’s reluctance to accept is sincere. Phrases, declaratives, and combinations of declaratives plus imperatives are attested instantiations of this move, exemplified by such formulas as “All in good condition!”; “itll be my pleasure”; “Hurry—these items will go quickly!”. Because of its communicative function, this move is much more likely to occur after the HEAD-ACT. ACCEPTANCEMAXIMIZERS can be identified by interrogative prompts like Why is it important, useful, beneficial to have the offer accepted? What are the negative consequences of the failure to have the offer accepted? How beneficial, valuable, agreeable is the offer to the parties involved? How pleasant and agreeable is the offered item and how negligible its cost to the offerer? and How sincere are the offerer and the offeree in making vs. rejecting, respectively, the offer? FACE-ENHANCER protects or sustains the offeree’s positive face by building or stressing solidarity and closeness with them. It expresses affection, liking, concern, gratitude, compliments or good wishes. It can be rendered as a phrase, a declarative or an ellipted declarative, any of which can be expressed emphatically, that is, with the addition of an exclamation mark at the end (e.g., “Congratulations! Such wonderful news”; “We’re so proud of you!”). This move reveals the writer’s attitude towards the addressee, and supports the occasion-specific or long-term relationship between them. It can therefore occur in close proximity of the HEAD-ACT. The relevant text segment appears to be the appropriate answer to such questions as How does the offerer try to make the offeree feel good, understood, accepted, appreciated? and How does the offerer try to prevent the offeree from feeling bad, excluded, disapproved of or criticized? NEXT-STEPS announces, proposes, inquires about or otherwise acknowledges the (verbal) actions enabling or favouring the realization and acceptance of the offer. This move may involve soliciting a response, offering or requesting information, making arrangements or seeking commitment from the addressee. This content can be conveyed by means of imperative or declarative structures, possibly combined, as in “Let me know if you want me to”; “and I look forward to hearing back from you.” Because of its content and function, this move is likely to occur at the end

Criteria for the Identification of Moves: The Case of Written Offers

345

of the text, as a way to channel future interaction. A NEXT-STEPS token can be identified when it appears to aptly answer such questions as What can/shall I/we do next? OR Shall/Can I/ we X next? or when it appears to more specifically paraphrase the general statements or suggestions I/We’ll do X next and Let’s do X next. CLOSING is the move that brings the interaction to a close, and which is therefore likely to occur at the end of the text. It may express pleasantries, small talk, or a closing formula such as a salutation, a signature and/or a post-scriptum. Relevant encoding options are phrases, imperatives, adverbials and combinations thereof. Possible instantiations are “Much love”; “Thanks”; “Take care”; “Sincerely,” + Name. This move answers the question How does the offerer take his/her leave? The criteria outlined above are meant to guide the researcher in the identification of move tokens. However, there are also text segments that cannot be unambiguously classified, since they simultaneously fit the identification criteria for two move types. Such text segments express two goals, either both relevant to the transaction at hand (e.g., “I’m going to start the dish right now so it will be ready when you get home. BACKGROUND+NEXT-STEPS) or one relevant to the specific transaction and the other to the management of the interaction (e.g., “I am very sorry to hear about your headache. I imagine it must be very difficult to concentrate, as you said in your message” INTRODUCER+FACE-ENHANCER; “Good luck on the interview!” CLOSING+FACE-ENHANCER). That is, they realize two moves simultaneously. In addition, text segments realizing distinct moves may partially overlap. That is, a move token may occupy part of a text segment that realizes an additional move. For instance, the whole of “Let me know if you’re up for it, we’d love to come by and pay you a visit” realizes the move NEXT-STEPS, since it proposes the future stages in the interaction, namely A) soliciting a response or commitment from the addressee (“Let me know if you’re up for it”) and B) tentatively suggesting arrangements involving a shared experience with the reader (“we’d love to come by and pay you a visit”); however, at the same time, “we’d love to come by and pay you a visit” expresses the positive emotional impact on the addresser of the tentative solidarity-oriented experience with the addressee, thus realizing the FACE-ENHANCER move. The occurrence of partially and completely overlapping text segments calls for co-coding, that is the simultaneous coding of a given stretch of text for two or more communicative functions. In this paper, I will therefore classify text segments realizing two moves as instances of the generic CO-CODING move.

346

Chapter Fourteen

A final consideration is in order here. Most of the above-outlined move types can be grouped under two broad categories: illocution-oriented and interaction-oriented. The former sustain the communicative purpose of the text. The latter allow the writer to take care of interactional needs and goals, which are relevant to any instance of (written) communication. However, two move types, namely FACE-ENHANCER and CO-CODING, fall somewhat in between the two above-mentioned move type categories, namely illocution-oriented and interaction-oriented. FACE-ENHANCER addresses interpersonal needs and goals which sustain social harmony in general, as well as contribute to motivating the sincerity of the offering speech act (see the preparatory felicity conditions in section 2. above). (That is, it is certainly compatible with the illocution, but not necessarily strictly relevant to it.) CO-CODING is a generic label that may encompass reference to interactional, illocutionary as well as interpersonal aspects of the communication. Thus, the communicative nature of the former move type is a bit ambiguous, and that of the latter a bit vague. Keeping the above clarifications in mind, I now proceed to describe the patterns of occurrence of the above move types in the corpus.

6.2. Frequency, distribution and sequencing of move tokens The moves identified in the corpus are not all exemplified with the same frequency of occurrence or similarly distributed across the texts, and do not necessaritly appear in all of them, as shown in Table 1. First of all, no move is particularly prominent with respect to the others – the most frequent one accounts for 13% of the data, and the least frequent one for 3%. This is probably due to the relatively high variety of move types attested in the corpus (i.e., 12, including CO-CODING). The most frequently instantiated move is HEAD-ACT (13%), the gist of the offer. Of the illocution-oriented moves, those that most often sustain it are BACKGROUND, MOTIVATOR, FACE-ENHANCER, and ACCEPTANCEMAXIMIZER, together accounting for almost 40% of the data. The other four moves relevant to the offering illocution, namely PRE-CONDITION, NEXT-STEPS, CO-CODING and INTRODUCER, instead, realize about 26% of all move tokens. CO-CODING applies to 6% of the data, and its most common instantiation combines the INTRODUCER and MOTIVATOR moves (8 instances). Other attested combinations include CLOSING+FACEENHANCER, INTRODUCER+FACE-ENHANCER, OPENER+FACE-ENHANCER, INTRODUCER+FACE-ENHANCER+MOTIVATOR and BACKGROUND+NEXTSTEPS.

Criteria for the Identification of Moves: The Case of Written Offers

347

Of the interaction-oriented moves, the most frequent is ALERTER (11%), followed by CLOSING (8%), while OPENER only realizes 3% of all move tokens. Overall, the frequency hierarchy of the moves in the corpus is this: HEAD-ACT > BACKGROUND > ALERTER > MOTIVATOR, FACE-ENHANCER, ACCEPTANCE-MAXIMIZER > CLOSING, PRE-CONDITION > NEXT-STEPS > CO-CODING > INTRODUCER > OPENER. TABLE 1: Frequency of occurrence of move tokens, and distribution of move types across texts. Move types HEAD-ACT BACKGROUND ALERTER MOTIVATOR FACE-ENHANCER ACCEPTANCE-MAXIMIZER CLOSING PRE-CONDITION NEXT-STEPS CO-CODING INTRODUCER OPENER Total

Tokens 41 (13%) 40 (12%) 36 (11%) 30 (9%) 29 (9%) 28 (9%) 27 (8%) 25 (8%) 23 (7%) 20 (6%) 14 (5%) 9 (3%) 322 (100%)

Texts 37 (100%) 26 (70%) 36 (97%) 23 (62%) 18 (49%) 20 (54%) 27 (72%) 22 (59%) 21 (57%) 15 (41%) 14 (38%) 9 (24%) 37 (100%)

The distribution29 of the moves in the corpus contributes to outlining the profile of the genre, by highlighting the functions that most often characterize it. Table 1 shows, first of all, that there is a general correspondence between frequency and distribution, that is, the most frequently instantiated moves are usually the ones that are also most widely distributed. Partial exceptions are BACKGROUND, which is the second most frequent, but the fourth most widely distributed, and ACCEPTANCE-MAXIMIZER, which is the sixth most frequent, but the eighth most widely distributed. Second, only one move appears to be crucial to the genre, namely HEAD-ACT, which is found in all the texts. All the other moves are optional to the speech act, although to varying degrees. Third, illocution-oriented moves (i.e., BACKGROUND, MOTIVATOR, PRECONDITION, NEXT-STEPS, ACCEPTANCE-MAXIMIZER) characterize a minimum of 38% and a maximum of 70% of the texts, while interactionoriented moves (i.e., ALERTER, CLOSING, OPENER), apply to a minimum of 24% up to a maximum of 97% of the texts. Fourth, the move types that fall

348

Chapter Fourteen

in between the illocution- vs. interaction-oriented move types, namely FACE-ENHANCER and CO-CODING, characterize slightly less than half of the texts. Overall, the distribution hierarchy of the moves in the corpus is the following: HEAD-ACT > ALERTER > CLOSING > BACKGROUND > MOTIVATOR > PRE-CONDITION > NEXT-STEPS > ACCEPTANCEMAXIMIZER > FACE-ENHANCER > CO-CODING > INTRODUCER > OPENER. A deeper understanding of the written offers examined comes from a consideration of the sequencing patterns of their component moves. ALERTER, which has an interaction-initiating phatic function, indeed often occurs in text-initial position (94% of the time). OPENER, the “icebreaker”, when instantiated, always occurs in second position in the text after ALERTER. INTRODUCER, which, if present, is the first move directly relevant to the communicative purpose of the text, comes right after OPENER 62% of the time. PRE-CONDITION, which inquires about the feasibility of the offer, is often (i.e., 88% of the time) found in the vicinity of HEAD-ACT, namely either right next to it or separated by an intervening move. It can occur either before or after it (56% vs. 64% of the time, respectively). HEAD-ACT can be found one or two moves away from PRE-CONDITION (88% of the time), MOTIVATOR (67% of the time) or BACKGROUND (55% of the time). MOTIVATOR, unlike what I expected, can occur either before or after the head act (i.e., 52% vs. 48% of the time). I can think of one reason for this: while motivation for one’s communicative purpose may indeed be more easily interpretable once that communicative purpose has been clarified, which would justify its occurrence after HEAD-ACT, it may also serve the strategic interpersonal goal of preventing possible objections from the addressee, if the communicative purpose in question is known to require the addressee’s consent. Therefore, giving one’s motivations in advance may lead the addressee to be favourably disposed toward the upcoming expression of one’s communicative purpose. On the other hand, as expected, BACKGROUND and ACCEPTANCEMAXIMIZER are very often found after HEAD-ACT (i.e., 92% and 85% of the time, respectively). Indeed, details about an offer, and the likely act of giving to follow, can more sensibly be discussed once the intention to provide a good/service has been conveyed. Similarly, incentives to accept can logically be formulated once the need for a relevant reply to an offer has been conveyed. NEXT-STEPS is found after HEAD-ACT 92% of the time, and more generally in the last three positions in the text 83% of the time. The very

Criteria for the Identification of Moves: The Case of Written Offers

349

nature of its content (i.e., reference to a course of action following the current main one) motivates its typical position in the texts. Similarly, CLOSING is found after NEXT-STEPS 64% of the time, in the last two positions in the text 100% of the time, and in text-final position 96% of the time, in line with its communicative role of bringing the communicative event to a close. FACE-ENHANCER, on the other hand, has no preferred position in the texts. It is found separated from HEAD-ACT by at least two intervening moves; it occurs more frequently before than after the HEAD-ACT (i.e., in 57% vs. 43% of the texts). In sum, the typical sequencing patterns of the moves is as follows: ALERTER > OPENER > INTRODUCER > FACE-ENHANCER, PRE-CONDITION, MOTIVATOR > HEAD-ACT > PRE-CONDITION, MOTIVATOR > BACKGROUND > ACCEPTANCE-MAXIMIZER, FACE-ENHANCER > NEXT STEPS > CLOSING

7. Discussion The above small-scale corpus-based study shows that an extended written offer consists of: A) an obligatory core move realizing the illocution (i.e., HEAD-ACT); B) possible additional moves sustaining the illocution (i.e., INTRODUCER, PRE-CONDITION, MOTIVATOR, BACKGROUND, ACCEPTANCE-MAXIMIZER, NEGOTIATIONS, NEXT-STEPS); and C) further possible moves sustaining the interaction as i) a self-contained event (i.e., ALERTER, CLOSING), ii) the prelude to a following course of action (i.e., NEXT-STEPS) and iii) a social event (i.e., OPENER, FACE-ENHANCER). The fact that only HEAD-ACT is the obligatory move suggests that the speech act of offering can be described in terms of prototypical and marginal, rather than obligatory, component moves. Also, the fact that both the illocution- and the interaction-oriented moves are variously distributed across the texts – neither group being more prominent than the other – shows the flexible realization of the speech act, whose transactional effectiveness appears to correlate with the instantiation of some of the moves that (can) rationally sustain it, and whose interactional adequacy depends on the satisfaction of at least one of the requirements of phatic communication. The two most frequently distributed moves are those that provide the interactional boundaries of the communicative event (i.e., ALERTER and CLOSING), but which therefore are not tied to the communicative purpose

350

Chapter Fourteen

of the texts. The next most frequently distributed moves are those that most closely sustain the illocution, namely BACKGROUND and MOTIVATOR. Relevant to more than half of the texts are three more illocution-sustaining moves, namely PRE-CONDITION, NEXT-STEPS and ACCEPTANCE-MAXIMIZER. The analysis also shows that moves are not combined together in a random fashion. Thus, the more directly they support the illocution, the closer they occur to HEAD-ACT (i.e., MOTIVATOR, PRE-CONDITION and BACKGROUND; see section 6.2). Also, their sequencing with respect to HEAD-ACT can mostly be accounted for in terms of the specific supporting function they play, with some minor exceptions. Therefore, although only one move appears obligatory in the corpus, the others cannot merely be classified as optional to the speech act in question. Rather, a typical instantiation of a written offer is likely to contain at least one, or possibly, more of these supporting moves and to conform to the above sequencing patterns. The micro-strategies instantiated in the corpus are in line with those characterizing dialogic offer sequences (see section 3). Indeed, studies on oral interaction have shown that, as an initiating speech act produced “in pursuit of the desired response” (Davidson 1984, 126-127, endnote 5) of acceptance, but which may be met with a dispreffered response (e.g., refusal, non-commitment, switch to another topic, deliberatly ignoring the offer; Rabinowitz 1993, 120-123), an offer is likely to trigger a subsequent version of itself; this may involve supplying “a new perspective” on the offer (see Rabinowitz 1993, 126-128), “adding more components, providing inducements, giving reasons for acceptance” (Davidson 1984, 107), and “‘urging’ to accept” (Davidson 1984, 127, endnote 9). In asynchronous written interaction, which only makes provision for a delayed response from the addressee, no co-constructed negotiation of the proposed transaction is possible, but preventive steps need to be taken so that the addressee’s preferences and freedom of (re)action are safeguarded. The data analysed shows that writers sustain their offering head acts before these can be challenged, and address possible objections from the addressees before any can be expressed by means of moves appealing to the addressee’s reason and emotions, moves that make the offer more palatable, convincing and thus likely to prevent a dispreferred response. It therefore comes as no surprise that the moves identified in the corpus are functionally similar to those exemplified in Aijmer (1996; see section 3.), which included setting the stage for and signalling the upcoming illocution, justifying the offer, inquiring about its acceptability, negotiating aspects of the offered item in view of a joint social commitment. The

Criteria for the Identification of Moves: The Case of Written Offers

351

partial similarity of the present findings to those of previous studies suggests two considerations. First, the strategies used to realize an offer are not strictly tied to the specific circumstances of its performance. Rather, in all contexts, the offering illocution necessarily calls for the same kinds of means to the end of overcoming potential obstacles to acceptance. Second, in both synchronous and asynchronous communication, the addresser is aware of and takes care of the addressee’s interactional needs: in the former case, by reacting in a relevant way to the addressee’s contributions to discourse (see Chen, Lei, and Zhang 1995, in section 3 about refusal strategies after offers), and in the latter, by building information units into his/her text such that they may satisfy the addressee’s envisaged needs. The texts analysed all instantiate initiating speech acts, which expect and even pursue a reaction from the addressees without having first been prompted by any interlocutors (but see Edmondson and House 1981 for counterexamples). However, in line with the prompts that elicited them, these written offers are verbal reactions to contextual circumstances that make the offerers aware of problems or opportunities that the offerees may have to address or may want to seize, respectively (see Edmondson and House 1981), as is evidenced in the MOTIVATORS. The written offers instantiated, therefore, have the marks of acts of generosity which, if accepted, will be of benefit to the offerees, will inconvenience the offerers, but will also tip the balance of the interlocutors’ relationship in favour of the offerers, who will gain social credits. Both the PRE-CONDITIONS and, especially, the ACCEPTANCE-MAXIMIZERS are meant to partly redress the envisaged social imbalance by convincing the offeree that the specific benefit to be gained in the short term will not put the offerer to too much trouble – and thus will not really leave the offeree indebted to the offerer – and by leaving it up to the addressee to decide whether to accept the offer or not, and thus to incur the inevitable social debt. The texts examined, that is, reveal the offerers’ awareness of the moral strings attached to the preferred second being sought after, namely acceptance, but also duly take into consideration the need to balance the interlocutors’ current and future social debts and credits. The study also explicitly addressed the question of the recognisability, and therefore the identification, of moves, an issue often neglected in speech act, as well as genre analysis.30 The move types identified in the corpus were defined by means of detailed formulations of their shared content and communicative goals. This served to make explicit their contribution to the overall strategy of the communicative event and to keep this distinct from their verbal realization. More specifically, I

352

Chapter Fourteen

proposed a set of criteria collectively contributing to signalling the communicative function of text segments: content, wording, positioning of text segments, and their relevance to specific heuristic prompts. Content can be likened to the steps usually specified as micro-components of moves in genre analysis. Unlike the definition of the move type, which outlines the goal and some general information it conveys, content specifies the detailed notions of a text segment and their micro-functions, which may characterize specific instances of it. Wording refers to the surface, textual-linguistic (lexico-semantic and morpho-syntactic) realization of the move. These may vary from one instance to the next, but should conform to common lexico-grammatical patterns, thus highlighting the common traits of tokens of the same move type. Positioning contributes to making sense of micro-textual strategies because it highlights the relevance of a given text segment to another, thus favouring its interpretability. Finally, the heuristic prompts enable the researcher to check the suitability of the definitions of moves – possibly provided on the basis of intuitive descriptions of textual patterns – in an explicit test-like fashion. Taken together, the identification criteria highlight differences between moves with a partly similar focus (e.g., PRE-CONDITIONS vs. MOTIVATOR vs. HEAD-ACT). More importantly, they introduce a more objective instrument for analysing texts. The taxonomy of the component moves of offers that is presented here, therefore, makes its own descriptive criteria explicit, uses the same and all descriptive criteria for all the components it comprises, and can be subjected to verification. The study highlights the importance of devising specific and testable functional, verbal and content-related criteria for both defining and identifying moves in extended speech acts.

8. Future perspectives The study reported in this paper, and the method of analysis adopted to carry it out, can benefit from some corrective actions. First and most importantly, the taxonomy of offer moves presented here needs to be field-tested for validation. Thus, its application to the same set of texts by more than one coder and/or to new sets of texts by the same or another coder would at least partly verify its suitability for text analysis. Secondly, the analysis has been far from thorough. For instance, the semantic-syntactic strategies for encoding head acts have not been fully described. Since the enconding options of head acts of oral offers have, instead, been investigated in a few previous studies (see Astruc, Vanrell,

Criteria for the Identification of Moves: The Case of Written Offers

353

and Prieto, forthcoming; Roulet 1977; Aijmer 1996; Curl 2005; Chodorowska-Pilch 2002, 2003; Zilles Pohle 2007), a comparison of the means of encoding offering illocutions in spoken vs. written communication might shed further light on the nature of this speech act. In particular, wording can be more thoroughly and systematically specified along several dimensions: A) syntactic function (e.g., declarative), B) communicative function (e.g., statement), C) syntactic strategy (e.g., ifclause; passive voice; subordination), D) semantic strategy (e.g., ability query; volition statement), and E) lexical-phraseological preferences (e.g., evaluative terms; expression of opinion). This will allow for a more accurate description of each move. Thirdly, the variable degree of (in)directness with which offering head acts can be realized (see section 6.1) suggests the need to establish even more stringent criteria for their identification than have been outlined above. This would make it possible to determine with a higher degree of certainty whether, and in which specific instances, acts of offering can occur with no overt head act – that is, only with supporting moves31 – or whether they must be described as always having some head act, however indirect – which was the option adopted in this study (see section 6.1.). Along similar lines, the possible correlation between the variable degree of directness in the encoding of an offer and contextual variables such as the cost of the offer and the power and distance between the interlocutors is worth exploring, as has been done for oral offers (see Astruc, Vanrell, Prieto, forthcoming). This study only describes the realization of offers in one English variety. The data collection procedure adopted could be used to gather and examine comparable texts in another variety or language so as to identify, by comparison, possible linguistically and or culturally specific ways of encoding offers in writing. For now, the data collected suggests that, as in oral offers, English speakers do place a high value on the rights and autonomy of the individual (see Wierzbicka 1985), given that the head acts are often verbalized tentatively (e.g., with the conditional and/or a modal of possibility) and accompanied by a PRE-CONDITION that inquires about the addressee’s willingness to accept the offered item. At the same time, the data also shows that with close and subordinate addressees (i.e., family members), the offerer may be more direct and insistent (see Tsuzuki, Takahashi, Patschke, and Zhang 2005 about close and equal addressees). Finally, given its descriptive, rather than pedagogic orientation (unlike, e.g., Barron 2003; Fukushima 1990b), this study does not address the issue of the learnability of offers by non-native speakers of English. However, the pragmatic analysis of the strategic components of offers that is

354

Chapter Fourteen

presented here could be used to raise English learners’ awareness of the transactional-interactional aspects of communication to be addressed when performing an offering illocution and of their possible wording. The potential usefulness of the taxonomy could be explored in an explicit second or foreign language instruction classroom context.

References Aijmer, K. 1996. Conversational routines in English: conventions and creativity. Studies in Language and Linguistics. London: Longman. Astruc, M. L., and M. M. Vanrell. Forthcoming. “Intonational phonology and politeness in L1 and L2 Spanish.” Probus. Astruc, M. L., M. M. Vanrell, and P. Prieto. Forthcoming. “Cost of the action and social distance affect the selection of question intonation in Catalan.” In Interdisciplinary approaches to intonational grammar in Ibero-Romance intonation, edited by M. Armstrong, N. Henriksen, and M. M. Vanrell. Amsterdam: Benjamins. Barron, A. 2003. Acquisition in interlanguage pragmatics. Learning how to do things with words in a study abroad context. Amsterdam: Benjamins. Barros García, M. J. 2010. “Actos de habla y cortesía valorizadora: las invitaciones.” Tonos. Revista Electrónica de Estudios filológicos 19: 113. Chen, X., Y. Lei, and Y. Zhang. 1995. “Refusing in Chinese.” In Pragmatics of Chinese as a native and target language, Technical Report 5, edited by G. Kasper, 119-163. Honolulu, HI: Second Language Teaching and Curriculum Center, University of Hawai’i at Manoa. Chodorowska-Pilch, M. 2002. “Las ofertas y la cortesía en español peninsular.” In Actos de habla y cortesía en español, edited by M. E. Placencia, and D. Bravo, 21-36. Munich: Lincom Europa. Chodorowska-Pilch, M. 2003. “Las ofertas corteses en el español peninsular.” In Actas del Primer Coloquio del Programa EDICE. La perspectiva no etnocentrista de la cortesía: identidad sociocultural de las comunidades hispanohablantes, edited by D. Bravo, 306-314. Estocolmo: Universidad de Estocolmo. Conein, B. 1986. “Conversation et interaction sociale. Analyse de sequences d’offre et d’invitation.” Langages 81: 111-120. Curl, T. S. 2006. “Offers of assistance: constraints on syntactic design.” Journal of Pragmatics 38 (8): 1257-1280.

Criteria for the Identification of Moves: The Case of Written Offers

355

Davidson, J. A. 1984. “Subsequent versions of invitations, offers, requests and proposals dealing with potential or actual rejections.” In Structure of social action: studies in conversation analysis, edited by J. M. Atkinson, and J. Heritage, 102-128. Cambridge: Cambridge University Press. Davidson, J. A. 1990. “Modifications of invitations, offers and rejections.” In Interaction competence, edited by G. Psathas, 149-179. Washington, DC: International Institute for Ethnomethodology and Conversation Analysis. Edmondson, W., and J. House. 1981. Let’s talk and talk about it: a pedagogic interactional grammar of English. München: Urban & Schwanzenberg. Fukushima, S. 1990a. “A descriptive analysis of English speakers’ performance in offering and requesting.” In Eibei Bungaku to Gengo: In search for the new horizon of research, edited by Bibulos Research Group, 536-549. Tokyo: Homerosusha. —. 1990b. “Offers and requests: performance by Japanese learners of English.” World Englishes 9: 317-325. —. 1996. “Request strategies in British English and Japanese.” Language Sciences 18 (3-4): 671-688. Fukushima, S., and Y. Iwata. 1987. “Politeness strategies in requesting and offering.” JACET Bulletin 18: 31-48. Hancher, M. 1979. “The classification of cooperative speech acts.” Language in Society 8: 1-14. Havertake, H. 1994. La cortesía verbal. Estudio pragmalingüístico. Madrid: Gredos. Kasper, G. 1981. Pragmatische Aspekte in der Interimpsprache. Eine Untersuchung des Englischen fortgeschrittener deutscher Lerner. Tübingen: Gunger Narr Verlag. Kerbrat-Orecchioni, C. 2004. “Il fait vraiment chaud aujourd’hui! Vous voulez boire quelche chose? Le ‘Travail de Faces’ dans l’Echange Initié par un Offre.” In Structures et Discours. Mélanges Offerts à Eddy Roulet, edited by A. Auchlin, M. Burger, L. Fillietaz, A. Grobet, J. Moechler, L. Perrin, C. Rossari, and de Saussure L., 417-432. Québec: Nota Bene. Koutlaki, S. A. 2002. “Offers and expressions of thanks as face enhancing sacts: tæ’arof in Persian.” Journal of Pragmatics 34: 1733-1756. Leech, G. 1983. Principles of pragmatics. London: Longman. Matoba, K. 1996. “Referential perspective in speech acts: a comparison between German and Japanese.” In Contrastive sociolinguistics, edited by M. Hellinger, and U. Ammon, 411-445. Berlin: Mouton de Gruyter.

356

Chapter Fourteen

Pérez Hernández, L. 2001. Illocution and cognition, a constructional approach. Logroño: Universidad de la Rioja. Rabinowitz, J. F. 1993. “A Descriptive Study of the Offer as a Speech Behavior in American English.” PhD dissertation, University of Pennsylvania, Philadelphia. Roulet, E. 1977. “Etude des realisations directes et indirectes de l’act d’offre en français parlé.” Studi italiani di linguistica teorica e applicata VI (3): 525-539. Ruiz de Zarobe, L. 2000-2001. “Estrategias de invitación en español e imagen social de los hablantes: Un estudio empírico.” Pragmalingüística 8-9: 261-278. —. 2012. “Learning-offers on the Internet in Spanish and French.” In Speech acts and politeness across languages and cultures, edited by L. Ruiz de Zarobe, and Y. Ruiz de Zarobe, 173-196. Bern: Peter Lang. Schneider, K. P. 2002. “No problem, you’re welcome, anytime: responding to thanks in Ireland, England and the USA.” In The pragmatics of Irish English, edited by A. Barron, and K. P. Schneider, 101-139. Berlin: Mouton de Gruyter. Taleghani-Nikazm, C. 1998. “Politeness in Persian interaction: the preference format of offers in Persian.” Crossroads of Language, Interaction, and Culture 1: 3-11. Tsuzuki, M., K. Takahashi, C. Patschke, and Q. Zhang. 2005. “Selection of linguistic forms for requests and offers: comparison between English and Chinese.” In Broadening the Horizon of Linguistic Politeness, edited by R. T. Lakoff, and I. Sachiko, 283-298. Amsterdam: Benjamins. Wierzbicka, A. 1985. “Different cultures, different languages, different speech acts: Polish vs. English.” Journal of Pragmatics 9 (2-3): 145178. Wierzbicka, A. 1987. English speech act verbs: a semantic dictionary. Sydney: Academic Press. Wunderlich, D. 1977. “Assertions, conditional speech acts, and practical inferences.” Journal of Pragmatics 1 (1): 13-46. Yotsukura, L. A. 1997. “Reporting problems and offering assistance in Japanese business transactional telephone conversations: toward an understanding of a spoken genre.” PhD dissertation, Ohio State University, Columbus. Zilles Pohle, S. 2007. “Offers in Irish English and German business negotiations: a cross-cultural pragmatic analysis.” In The use of English in institutional and business settings. An intercultural perspective, edited by G. Garzone, and C. Ilie, 199-228. Bern: Peter Lang.

Criteria for the Identification of Moves: The Case of Written Offers

357

Appendix Sample scenario descriptions and texts Set A, Scenario OF.W2.A1 You [Susan/Stephen] are a part-time student and work as a cashier at a cinema at weekends. A dear friend of yours, Silvia/Simon, has sent you an email to inform you that he/she has a job interview early next week. Since you know he/she has a four-month old boy/girl, and you are free on the day of the interview, you volunteer to baby-sit her child. What do you write to him/her? Set A, Text OF.W2.A1.01 Hey Simon, How’s it going? I heard about the interview and I just wanted to say congratulations! Also, if you need someone to baby-sit Maggie that day I am free all day and would be more than happy to look after her. Let me know if you want me to. Good luck! [Abbreviated first name] Set B, Scenario OF.W2.B.9. One of your friends has moved to the countryside and now lives and works on a farm where he/he grows wonderful and inexpensive produce. You post a notice on the bulletin board of your block of flats (OR your apartment block) informing them that you can arrange to have fresh produce delivered weekly to your address for whoever is interested. What do you write? Set B, Text OF.W2.B9.01 Attention Residents of Bayview, If you like fresh produce like I do then sign up to receive fresh produce sent to your house, for free! That’s right, for free! My friend has started his own business and is trying to get potential customers. His phone number is listed on the bottom of this announcement. Or you can come see me if you would like to try the merchandise first. I’m in room 524 if you didn’t know already. Thanks! Your friendly neighbour, [given-name surname]

358

Chapter Fourteen Set C, Scenario OF.W2.C1.

You run a small company (e.g., a small shoe factory) that employs half a dozen people. You have noticed that one of your employees, Nancy/Rick, has been suffering from backache for some time. You are on familiar terms with a famous local orthopaedist you went to school with. Since you know that you can easily talk your doctor friend into finding time in his/her busy schedule to see your employee, you think you can arrange an appointment for Nancy/Rick with that specialist next week. You email Nancy/Rick. What do you write to him/her? Set C, Text OF.W2.C1.01 Hello Nancy, I have noticed you have been suffering with back pain for some time now and I would like to offer something that may help. I have a friend who is a very well known orthopaedist. If you would like, I can arrange an appointment for you next week with him. (Normally the wait time for an appointment with him is quite lengthy.) Please let me know and if you are interested, I will make the call and set something up. [first-name] Set D, Scenario OF.WE.D3. You are a university professor supervising an undergraduate student. The student has emailed you a chapter of her/his thesis, as expected. After going over the text for over an hour, you realize that he/she needs in-depth help re-organizing and editing her/his work. You reply to his/her email offering to give him/her extra help at your home over the weekend. What do you write? Set D, Text OF.W2.D3.01 Student, I have read the chapter of your thesis that you emailed to me, after looking over it I think that you require assistance in re-organizing and editing your work to allow your thesis to flow more smoothly. There isn’t much I can do from email, but if you would like to come to my house over the weekend I will be more than happy to help you because with some work it could be a great thesis.

Criteria for the Identification of Moves: The Case of Written Offers

359

Set E, Scenario OF.W2.E2 You [Eve/Edward] work as a secretary to Ms/Mr Clive, the manager of a large firm. Today she/he has to prepare her/his annual financial report for tomorrow’s meeting. But she/he’s got a bad headache and she/he can’t concentrate, and has told you about it in a short email to you. You think you can prepare her/him a draft of the report that she/he can later revise. What do you write in your reply? Set E, Text OF.W2.E2.02 Dear Ms. Clive, Thank you for your email this morning; I am very sorry to hear about your headache. I imagine it must be very difficult to concentrate, as you said in your message. In light of this, I would like to offer my assistance in preparing the annual financial report for tomorrow’s meeting. If you would prefer I could prepare a draft and send it to you promptly for revision when you are feeling less ill. Furthermore, if there is anything else I might be able to offer my assistance with, please do not hesitate to contact me at your convenience. Yours sincerely, [given-name surname] SET F, Scenario OF.W2.F1. You are a part-time English lecturer at University. You’ve heard from a colleague of a possible short-term appointment at a local international school, where there is at least one student interested in taking translation classes on an independent study basis. You write to the head of the school to offer your professional services to them. What do you write to him/her? Set F, Text OF.W2.F1.02 To Whom It May Concern: My name is Professor Jumabhoy and I am a very qualified and experience [sic] English lecturer at a university. I have heard that there is interest in translation classes at your prestigious school and would like to offer my profession [sic] services to you. Attached here is my resume and I look forward to hearing back from you. Thank you.

360

Chapter Fourteen

Notes 1 For an in-depth discussion of the communicative and transactional nature of the speech act of offering, see Rabinowitz (1993, 15-64). 2 Edmondson and House (1981, 136) explain that the difference between an offer and a promise – both classified as Willing illocutions – depends on their interactional status: the former is an initiating act, while the latter is a response to a request (i.e., an undertaking). 3 See Rabinowitz (1993, 216-218) about communication breakdowns arising from failure to follow through with the giving, once an offer has been accepted. 4 Most of the time, offers extend the provision of assistance, hospitality or gifts (Barron 2003, 123). 5 However, Chodorowska-Pilch (2002, 23; 2003, 307) observes that offers cannot be intrinsically polite utterances, otherwise there would be no instances of mitigated offers, as are found in her data. 6 Barros García (2010, 6) makes a similar comment about invitations, a manifestation of solidarity- and affiliation-oriented politeness (cortesia valorizadora), which presents the addresser as an agreeable and supportive person, and the addressee as an in-group member who is the object of one’s affection. 7 However, Ruiz de Zarobe (2012, 177) also observes that as a result of the offer, the offeree’s territory is also increased and thus their negative face enhanced. 8 Ruiz de Zarobe (2000-2001, 263-265) similarly observes that invitations in Spanish A) both enhance and threaten the positive and negative face of both interlocutors, and B) imply or manifest both positive and negative politeness, signalling the inviter’s awareness of the addressee’s needs for involvement and independence. 9 However, Wierzbicka (1987, 192) points out that the directive nature of offers is optional rather than necessary, as when someone offers to bring someone else books from the library. 10 From a complementary perspective, Kerbrat-Orecchioni (2004, 419) observes how the offer “ne dispose pas de formulation qui lui appartiene en propre.” 11 Following Hancher (1979), Rabinowitz (1993, 21) clarifies that an offer is a precooperative speech act, projecting and calling for a combined effort involving both parties: obligation to act for the offerer, and obligation to respond with an acceptance or refusal for the offeree. 12 For Pérez Hernández (2001: 307), quoted in Ruiz de Zarobe (2012), an offer is more of a commissive than a directive act, because it presents the offerer as the agent of the action specified in the message, and the offeree as the participant who only has to accept the offer. 13 Roulet (1977, 533-534) reduces them to six: A) the speaker is able to perform a given act; B) the speaker is willing to perform that act; C) the hearer wants the speaker to perform the act; D) the hearer allows the speaker to perform the act; E) the act is useful to the hearer; and F) the speaker mentions a future act. Similarly, Matoba (1996, 418) includes the following: A) preparatory: “S is able to perform A. H is able to accept A from S.”; B) sincerity: “S intends to do A. H wants S to perform A.”; C) propositional: “S predicates a future A of S”; D) essential: “Counts as the understanding by S of an obligation to do A” (p. 418; original emphasis).]

Criteria for the Identification of Moves: The Case of Written Offers 14

361

The examples in this paragraph are from Roulet (1977). In both publications the author also examined other types of speech acts and the reactions to them. 16 The examples in this paragraph are adapted from Aijmer (1996, 189-195). 17 The examples in this paragraph are from Chodorowska-Pilch (2003). 18 In this paragraph, the examples are from Edmondson and House (1981), reproduced without the original diacritics, and with slashes signaling new lines. 19 The authors use the terms invitation and offer somewhat interchangeably in their work. 20 Reoffers can be substantive, when they are meant literally, or ritual (Barron 2003, 127), when they are meant to satisfy the sincerity condition, that is, after an Initiate offer which expresses illocutionary intent, and a ritual refusal reacting to it. 21 A Reoffer can be a specific type of the more general Contra. 22 The author also provided coding categories for lexico-semantic-syntactic offer strategies, semantic refusal strategies, lexical and phrasal downgraders of offers, and syntactic downgraders of offers. 23 The different number of scenarios per set is due to the personal source of inspiration for them (see above). 24 I chose sets A, B and C for the first two study participants, then D, E and F, for the third and fourth participants, then again A, B and C for the fifth and sixth participants, and so on. The choice depended on what the participants had done immediately before, during an oral elicitation session, in which they had been engaged, in pairs, in role-plays elicited through scenario descriptions calling for the realization of, and reaction to, oral offers. These scenarios, too, were grouped into six sets (from A to F), in which the variables of social distance and degree of power differential were similarly controlled for. Therefore, if a pair of participants had been assigned sets A, B and C during the oral elicitation session, they were assigned sets C, D and F during the written one, and vice versa. 25 This is the second functional parameter, after the definition of the move. 26 Here and elsewhere, all the examples are faithful quotations from the corpus. 27 This occurs when no other more explicit head act occurs in the text. 28 Here and elsewhere, the participants’ typing has been faithfully reproduced. 29 The term is not used in the general, rather than statistical, sense of ‘presence across a number of texts’. 30 In a study on gratitude acknowledgements, Schneider (2002) observes that studies in pragmatics do not always make it clear how the classification of illocutionary strategies is arrived at, how they are kept distinct, why they are glossed the way they are, “how many categories should be distinguished, and how individual realizations relate to the postulated strategies.” The problem is said to be aggravated in those expressive speech acts performed by routine formulas, which, therefore, are ambiguous or opaque. Also, most studies on realization strategies “often neglect the systematic description of the “conventions of form”, i.e., of the structural patterns and lexico-semantic devices conventionally employed, and especially their frequency and distribution in language use” (Schneider 2002, 107). 31 This option has been adopted for the act of requesting, for instance (Fukushima 1996, 673). 15

CHAPTER FIFTEEN “THANKS A BUNCH:” CROSS-CULTURAL COMPARISON OF THE SPEECH ACT OF THANKING WINNIE CHENG AND ANDY SETO

1. Introduction Speech act theory is a branch of philosophy which attempts to classify spoken language in terms of what is done rather than what is said. It was first introduced by Austin (1962) and then developed by others, most notably by Searle (1965, 1969, 1975). Speech act theorists are concerned with the functional value of utterances rather than the form of utterances. Austin (1962) identified a type of utterance or sentence, called a “performative,” in which the words constitute the performing of an action. Examples of performatives are I thank you, I apologise, I pronounce you man and wife, and I say that you are wrong. While the work of Austin in identifying the ways in which performatives perform actions helped to shape Pragmatics, there are some fundamental problems with the idea that only performatives perform actions (Thomas 1995, 44): 1) there is no formal (grammatical) way of distinguishing performative verbs from other sorts of verbs; 2) the presence of a performative verb does not guarantee that the specified action is performed; and 3) there are ways of “doing things with words” which do not involve using performatives. When confronted with the shortcomings of his performative hypothesis, Austin (1962) set about trying to distinguish between the meaning of words and their “illocutionary force.” Austin (1962) claims that when someone speaks, “acts” are performed, and that within an utterance a

364

Chapter Fifteen

speaker usually performs three acts. These acts are separated out for the purposes of analysis and classification by speech act theorists: (1) locutionary act, i.e., the formal, literal meaning of the words uttered; (2) illocutionary act, i.e., the act performed in saying the utterance; and (3) perlocutionary act, i.e., the effect the utterance might have. Others have attempted to classify the functional value of utterances with a variety of speech act typologies which are invariably criticized for failing to account for all the possible functions of language in the real world. The most well-known example of such a typology is Searle’s (1976), which has five basic kinds of actions. Representatives commit the speaker to the truth of the expressed proposition; directives are attempts by the speaker to get the addressee to do something; commissives commit the speaker to some future course of action; expressives express a psychological state; declarations effect immediate changes in the institutional state of affairs and tend to rely on elaborate extra-linguistic institutions.

2. Theoretical background Thanking is perceived as an expression of gratitude or appreciation. As Searle (1969, 65) puts it, “[w]hen I thank someone, I imply that the thing I am thanking him for has benefited me (or was at least intended to benefit me).” Previous research on thanking has studied diverse forms and functions of thanking, compared speakers of different varieties of English, and examined a range of speech events in different contexts of communication (see, for example, Coulmas 1981; Wolfson 1981; Eisenstein and Bodman 1986; Herbert 1986; Mey 1993; Jung 1994; Stenström 1994; Aston 1995; Jacobsson 2002; Wichmann 2004; Cheng 2009; Grainger 2013; Jautz 2013; Liao 2013). In Jung’s (1994) research of thanking and responses to thanking in American English, findings showed that, in general, thanking was used to express appreciation of benefits and to enhance rapport between interlocutors. These basic functions were extended to conversational opening, conversational closing, leave-taking, and offering positive reinforcement. Another function of thanking was expressing dissatisfaction or discomfort indirectly, which was often accompanied by sarcasm or a different intonation pattern. Regarding responses to thanking, Jung (1994) identified six types, namely acceptance, denial, reciprocity, comment, nonverbal gesture, and no response. Comparing the use of thanks in Italian and English bookshop encounters, Aston (1995) found that the use and

Cross-Cultural Comparison of the Speech Act of Thanking

365

realizations of thanking in conversation closings were influenced by cultural differences concerning perceptions of the overall situation in the talk and preferred procedures of conversational management. Stenström (1994) was one of the first researchers who examined speech acts in corpus data. She categorized thanking as an initiating act used to express gratitude, or as a politeness device. As a politeness device, thanking could be used to terminate a conversation. Stenström found that the most frequent realizations of in the London-Lund Corpus of Spoken English (0.5 million words) were thank you, thanks, and thanks very much, in a descending order of frequency. Other linguistic realisations included thanks a lot, thanks awfully, thanks very much indeed, thank you so much, thank you very much, and thank you very much indeed. Adopting a corpus-based approach, Cheng (2009) investigated the genre of conversation between in intercultural communicative contexts in the Hong Kong Corpus of Spoken English (HKCSE). In the sub-corpus of conversation (258,882 words), the study identified 110 instances of thanking realized in thirteen thanking expressions and performing nine situated thanking functions. The most frequent three thanking expressions were thank you (55.5%), thank you very much (10.9%), and thanks (10%), in line with the findings of Stenström (1994). Thanking functions, in a descending order of importance, comprised expressing gratitude, and complimenting on a service or a favour received (47.3%); expressing gratitude, and agreeing to and accepting an offer or a compliment (29.1%); expressing gratitude and shifting to a related topic (10%); expressing gratitude in response to a greeting (2.7%); expressing gratitude when making a request (2.7%); thanking after rejecting to maintain a polite and friendly social atmosphere (2.7%); stating appreciation, and establishing and maintaining a polite and friendly social atmosphere (1.8%); expressing gratitude, and closing the current topic or terminating the current discourse (1.8%); and expressing gratitude in response to a request being granted (1.8%). In another study of Hong Kong English, Wong (2010) examined the Hong Kong component of the International Corpus of English (ICEHK) and found mainly brief expressions of gratitude thanks and thank you, used as closing signals and constituting a complete turn, suggesting that “Chinese may be too reserved to express their gratitude openly and explicitly” (Wong 2010, 1243). Investigating the linguistic ritual for thanking in Japanese (o-rei) in telephone conversations in the Japanese end-of-the year gift-giving season, Ohashi (2013) found “many means of indicating o-rei that are not predicted by most politeness and speech act theorists” and that thanking

366

Chapter Fifteen

functioned more than “expressions of gratitude and appreciation” but involved the “mutual and reciprocal aspects” of interaction (Ohashi 2013, 2150). In Turkish, Zeyrek (2012) identified nine thanking formulae and nine thanking strategies – three explicit thanking routines, three implicit thanking formulae, and three strategies with other illocutionary forces and functions. The analysis of the use of modifiers in thanking expressions revealed that thanking in the context of Turkish is an intrinsically courteous and polite act. Zeyrek (2012, 83) concluded that politeness in Turkish is “not a categorical construct with negative and positive poles, but rather a gradient one with highly modified and innovative means on one end, and direct, unmodified expressions on the other.” Based on Brown and Levinson’s (1987) theory of politeness, Siebold (2012) conducted a contrastive analysis of Spanish and German expressions of thanking in terms of their frequency, discourse organization, internal structure and modification, using data collected from role-plays of eight everyday situations. Spanish expressions were found to be characterized by a strong tendency towards positive politeness, while German expressions tended to be associated with negative politeness. In their resource book on pragmatics for students, Archer, Aijmer, and Wichmann (2012) remarked that thanking can be considered an important part of other speech acts. The combination of thanks and a compliment is said to be used when an offer is accepted (Thank you they look very nice) and also when it is rejected (No thanks – it looks very appetizing) (Archer, Aijmer, and Wichmann 2012, 41). The function of speech acts, including thanking, is found to be dependent on the global context and the type of communicative situation. The review has shown that the speech act of thanking, as an important part of interpersonal communication in different languages and cultures, has been widely researched in pragmatics and intercultural communication, with the use of methodologies such as conversation analysis and corpus linguistics.

3. This study 3.1. Research aims The present study is about the speech act of thanking, which is a paradigm case of an expressive in Searle’s (1976) classification of illocutionary acts. It aims to identify and compare the variety of forms that perform the thanking speech function in different domains and contexts of situation and by speakers of different national cultural backgrounds. The

Cross-Cultural Comparison of the Speech Act of Thanking

367

study also aims to identify core thanking expressions, if any, across different corpora, followed by an investigation into collocational and colligational patterns of these core forms of thanking. The corpora examined were the Hong Kong Corpus of Spoken English (HKCSE), the British Academic Spoken English Corpus (BASE corpus), the spoken part of the British National Corpus (BNC), and the Michigan Corpus of Academic Spoken English (MICASE). Implications of the findings will be discussed.

3.2. The four spoken corpora The HKCSE, compiled by the English Department of The Hong Kong Polytechnic University between 1996 and 2000, consists of four subcorpora that represent the main overarching spoken domains found in the Hong Kong context, namely academic, business, conversation, and public discourse. Each sub-corpus consists of a variety of discourse types and participants. The one-million-word HKCSE was further enriched as a research, learning and teaching resource by adding a prosodic transcription (i.e., the indication of speakers’ intonation in the transcript) (Cheng, Greaves, and Warren 2008). The BASE corpus, developed by the Universities of Warwick and Reading, is a collection of transcripts of lectures and seminars recorded at the two universities in the UK during 1998-2005. It consists of 160 lectures and 39 seminars recorded in a variety of university departments, containing 1,644,942 tokens. Spoken data were obtained from four broad disciplinary groups, each comprising 40 lectures and 10 seminars, namely Arts and Humanities, Social Studies and Sciences, Physical Sciences, and Life and Medical Sciences. The BNC is a 100-million-word collection of samples of written and spoken language from a wide range of sources, designed to represent a wide cross-section of British English from the later part of the 20th century, both spoken and written. The spoken part (10%) consists of orthographic transcriptions of unscripted informal conversations (recorded by volunteers of different ages, from different regions and belonging to different social classes in a demographically balanced way) and other types of oral interaction collected in different contexts, ranging from formal business or government meetings to radio shows and phone-ins. The MICASE is a spoken language corpus of approximately 1.8 million words (nearly 200 hours) focusing on contemporary university speech within the microcosm of the University of Michigan, in Ann Arbor, Michigan. In the MICASE, “academic speech” is defined as that speech

368

Chapter Fifteen

which occurs in academic settings. As it is not pre-defined as “scholarly discussion,” such genres as jokes, confessions, and personal anecdotes, as well as definitions, explanations, and intellectual justifications are included.

3.3. Data analysis method and procedure The approach to the present study is corpus-based (Tognini-Bonelli 2001), with the corpora used as inventories of data from which to extract appropriate material in order to support intuitive knowledge, to extract illustrative examples, to quantify linguistic phenomena, to verify expectations, or to find proof for existing language theories (TogniniBonelli 2001; Biber 2009). In the study, the linguistic realizations of thanking defined in previous linguistic research were assumed and corpus research was conducted “to describe the patterns of variation and use associated with those grammatical features” (Biber 2009, 278). On the basis of the literature review on thanking (see, for example, Eisenstein and Bodman 1986; Herbert 1986; Stenström 1994; Aston 1995; Cheng 2009; see section 2), nineteen forms of thanking were identified, and they are: thank you very much, thank you, thanks very much, thanks, I can’t thank you enough, I don’t know how to thank you, thank you for (inviting me), thank you so much, thanks so much, thank you so much for (such a lovely time), thank you very much, thank you very much indeed, thank you very much for (the dinner), thanks a bunch, thanks a lot, thanks a lot for (the meal), thanks for (the wonderful meal), thanks for everything, and thanks awfully. These forms of thanking vary in length from one word (thanks) to seven words (I don’t know how to thank you). Each of the four corpora was searched for instances of each of the nineteen forms of thanking to find out whether these are used in the four spoken corpora, their relative frequencies of occurrence, and their proportional frequency counts. Frequencies are vital as they show which words and structures are central in a language, and proportional frequencies show the prevalence of specific types of linguistic realizations of the speech act concerned in the four corpora. This was followed by examining the corpora to identify forms of thanking other than those reported in the literature that has been reviewed for this study. This was followed by identifying the core thanking expressions, if any, based on frequencies of occurrence. The core forms thus identified were then examined to find out collocational and colligational patterns. The program used to search the BASE corpus and HKCSE was ConcApp developed by Chris Greaves, The Hong Kong Polytechnic

Cross-Cultural Comparison of the Speech Act of Thanking

369

University. The BNC was interrogated through the BYU-BNC website developed by Mark Davies, Brigham Young University. The MICASE was interrogated through a customized search engine developed by the Humanities Text Initiative section of the University of Michigan Library and by Alan Pagliere.

4. Results and discussion 4.1. Linguistic realizations of thanking A total of 9,221 instances of thanking were found. Table 1 (in the Appendix) shows the corpus-specific frequencies of the nineteen forms of thanking, as reported in the literature, in a descending order of frequency. Two of the nineteen forms of thanking, I can’t thank you enough and I don’t know how to thank you, are not found in any of the corpora. Figure 1 shows the five most frequent forms of thanking in the corpora, accounting for 94.14% of all the seventeen forms. They are thank you (5,246; 56.9%), thanks (1,440; 15.6%), thank you very much (1,292; 14%), thanks very much (357; 3.9%), and thank you for (346; 3.8%). 169

thank you

4245

459 373 52

thanks

80

998 310

BASE

51

thank you very much

1058

136 47 35

thanks very much

24 3 19

thank you for

47 25

0

BNC HKCSE MICASE

295

255

1000

2000

3000

4000

5000

FIGURE 1. Top five most frequent forms of thanking in the four corpora that are reported in the literature reviewed

370

Chapter Fifteen

Some examples of these five thanking forms are in the following: (1) Thank you mm0341: that would be terrific thank you nf0340: that’s what we’re up to at the moment … (BASE corpus: Life and Medical Sciences) So erm, that who we are, what we do and how we get our work and I hope that been interesting for you. Thank you. Very nice. Any questions? (BYU-BNC: D94 S_meeting) A: b:

oh I didn’t know if it was an allergy or a cold okay well good luck thank you (HKCSE: A039)

SU-f: Teresa do you have a Post-it by chance? S1: yeah actually i think i do, right here SU-f: thank you. what's going on with the thing? (MICASE: SVC999MX148) (2) Thanks sm0156: the other person would be like oh well you know he wouldn't have up to that point had he not be so over the top in his thanks nm0146: yeah yeah (BASE corpus: Arts and Humanities) I'm just giving you a spread sheet and this blank page and Oh bless you, thanks! give you a did you give him a blank page? (BYU-BNC: J8D S_meeting) A:

b: A: b:

just do an exit page thank you well we worked on we worked on organization a little bit on er grammar but mostly er I think the problem here was organization and not really er grammar problem okay this part is okay yea thanks yea mhmm (HKCSE: A039)

right right right it could be anything from chapter, five six, eight nine. how's that? okay, thanks, you've cleared that up are there gonna be any review sessions? (MICASE: OFC280SU109)

Cross-Cultural Comparison of the Speech Act of Thanking

371

(3) Thank you very much can you leave them say on this desk and i will come back and collect them at three okay all right thank you very much (BASE corpus: Arts and Humanities) Right, am I let off the hook now? Yeah, your off the hook. Thank you very much, thank you Chris. (BYU-BNC: D96 S_meeting) A1: and we will have er thank you very much that’s the end of our briefing section okay any questions and comments (HKCSE: A014) i hope you do i think i think you can. yeah. well thank you very much. okay you're welcome. (MICASE: ADV700JU047) (4) Thanks very much nm5225: thank you nm5219: thanks very much could I bring Connie back to the front as well (BASE corpus: Life and Medical Sciences) You'll have to come up sometime I'll try and do that I'll give you a ring, a ring for you to come up Right, thanks very much Lovely to see you Goodbye (BYU-BNC: KB0 S_conv) a: ** anyway enjoy your flight tonight B: thanks very much a: bye bye (HKCSE: B030) could you give me a hand sorry. sure. thanks very much. appreciate it. (MICASE: TOU999JU030) (5) Thank you for nm5033: well thank you for coming along this morning (BASE corpus: Physical Sciences) Is this a note in your bag? Yes. Well thank you for telling me then Nat. (BYU-BNC: KC5 S_conv) a1: good morning everybody thank you for for coming and er as you can see um my topic er of the (HKCSE: A017)

372

Chapter Fifteen

back next week so okay, no problem. no problem. thank you for meeting with me. sure. (MICASE: OFC320SU153)

The four spoken corpora were then examined for forms of thanking not reported in the literature that has been reviewed. Findings show a total of 91 tokens of thanking in 74 forms (Table 2 in the Appendix). Topping the list are thanks to (88; 24.9%), followed by thanks very much for (43; 12.2%), thank you to (27; 7.6%), thank you (subject) for (15; 4.2%), and thank you both very much (13; 3.7%). The five most frequent forms constitute 52.6% of all the forms. Another finding is that 43 of the 74 (12.15%) forms of thanking occur only once across the corpora. The relatively low percentage of many forms hence shows the rich variety of forms used to perform the speech act of thanking in real-life interactive situations. Findings also show that the specialized corpora (BASE and MICASE) and the intercultural corpus (HKCSE) are highly restricted in the variety of forms used to express thanking: the former displays four types of forms while the latter instantiates eleven types of forms. The native-speaker and general English corpus (BNC) is, however, very different compared to the other three, with 69 types of forms that express thanking. Another finding is that certain forms are corpus-specific, for instance, thank you guys for (n=2) is found only in American English MICASE. Figure 2 shows the frequencies of the forms of thanking across the corpora. Out of the 74 forms, only two of them are found in all the four corpora: thanks to (88; 24.7%) and thank you to (27; 7.6%); two are found in three corpora: Thanks very much for (43; 12.2%) and Thank you all for (7; 2.8%); and four in two corpora. The remaining 66 of the 74 forms (89.2%) occur in only one spoken corpus.

Cross-Cultural Comparison of the Speech Act of Thanking

thank you (subject) for

0 0 0

thank you both very much

0

15

2

11

0

BASE

6

thank you to

BNC

16

2 1

HKCSE

6

thanks to

5

thanks very much for

0 0

373

MICASE

69

8

3 3

37 20

40

60

80

FIGURE 2. Top five most frequent forms of thanking in the corpora that are not reported in the literature reviewed

Examples of the five most frequent forms of thanking are given in (6) to (10) below: (6) Thank you (subject) for Thank you Betty for helping me to win this award. (BYU-BNC: HDN S_speech_scripted) (7) Thank you both very much Thank you very much, for this very nice time. Thank you both very much. (BYU-BNC: KS0 S_meeting) b1: thank you both very much (.) that’s all we have for you this week (.) join us again next week for another edition of Newsline (.) goodnight (HKCSE: P129) (8) Thank you to We thank you to everybody who's contributed to to our speaking this morning. (BYU-BNC: KB0 S-conv) I’m here really to say thank you a huge thank you to those frontline medical staff and also the er those in who are in the administration whose given themselves selflessly to Hong Kong (HKCSE: P041)

374

Chapter Fifteen

Beth um, first of all thank you to everybody who came to Knitwits on Sunday (MICASE: MTG999SU043) (9) Thanks to nm5082: simply by force of of volume so these things thanks to the Laurium mines were big business they were thrown out in their millions (BASE corpus: Arts and Humanities) Give my respect and thanks to the author, whose novel is the first English one and the French are only romances now, that I have been able to read for many a day. (BYU-BNC: K60 S_classroom) we are investing significant resources into our infrastructure manpower and services as well as worldwide marketing and promotion of Hong Kong as the preferred destination thanks to your efforts and support in the past we have built an enviable position as the preferred city destination in Asia (HKCSE: P010) and Bill Rathert who, showed me how to use the air scribe and without it i would never have nice clean fossils to study. so thanks to them. (MICASE: DEF305MX131) (10) Thanks very much for nm0296: thanks very much for coming to our symposium on educational research (BASE corpus: Life and Medical Sciences) Right. Thanks very much for your help, okay, bye (BYU-BNC: KB7 S_conv) A: well Emily Lau thanks very much for being with us a: thank you (HKCSE: P162)

4.2. Distribution of forms of thanking The proportion of the five most frequent forms of thanking differs across the corpora. The uneven distribution raises the question of the prevalence of these forms. Figure 3, showing the distribution of the five most frequent forms of thanking that have been reported in the literature reviewed, reveals that 70% of each of the forms can be found in the BNC.

Cross-Cultural Comparison of the Speech Act of Thanking

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

thank you

thank you for

thank you very much

375

thanks

thanks very much

BASE

3%

5%

4%

4%

10%

BNC

81%

74%

82%

69%

83%

HKCSE

9%

14%

11%

6%

6%

MICASE

7%

7%

3%

21%

1%

FIGURE 3. Relative frequencies of the five most frequent forms of thanking across the corpora

Figure 4, showing the five most frequent forms outside of what has been reported in the literature reviewed for this chapter, shows that two forms, thank you (subject) for and thank you both very much, occur exclusively in the BNC. The major difference in the frequency of occurrence of the forms of thanking across the four corpora is most likely due to the differences in corpus size and the spoken communicative contexts and domains across the corpora. As described above, out of the 74 forms of thanking formulae, only two occur in all four corpora (thank you to and thanks to); two occur in three corpora (thank you all for and thanks very much for); four occur in two corpora (thank you again, thank you all, thank you all very much, and thank you both very much). The remaining 66 forms are found in one corpus only, 61 of which in the BNC (92.4%). In addition, among the 74 forms, 43 occur once in the corpora (58.1%). These findings suggest that across the four spoken corpora, the forms of thanking do not only vary considerably, but also are not evenly distributed.

Chapter Fifteen

376

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

thank you (subject) for

thank you both very much

thank you to

thanks to

thanks very much for

BASE

0%

0%

30%

7%

7%

BNC

100%

100%

59%

78%

86%

HKCSE

0%

0%

7%

9%

7%

MICASE

0%

0%

4%

6%

0%

FIGURE 4. Relative frequencies of forms of thanking not found in the literature reviewed

4.3. Collocational and colligational patterns Many forms of thanking in the corpora examined are formulaic; for example, thank you and thanks. These forms often co-occur with items of particular word-classes such as an adverb phrase (e.g., very much) and/or a preposition (e.g., for). Analyzing the collocational patterns of thanking formulaic expressions in different corpora suggests that certain expressions and certain grammatical classes are preferred in a thanking context (Berglund 2005). Thank you and thanks are the common base words or “nodes” coselected with other words. Out of the 91 forms identified in the corpus data, only 16 (17.2%) are preceded by words from particular word-classes, including articles (the), adjectives (many, enormous, particular and special), auxiliary verbs (cannot), and lexical verbs (cries out, give and offer) in L1 and L2 positions. The most frequent collocate that precedes thanks is the adjective many. The remaining 77 forms are followed by items of various word-classes in positions R1 up to R10, including adjectives (innermost), adverbs (a lot, again, also, awfully, beyond, deeply, enough, entirely, ever, here, indeed, largely, lot, mainly, obviously, so much, today and very much), articles (the, a), verbs (must), prepositions

Cross-Cultural Comparison of the Speech Act of Thanking

377

(for, from, of, over, through, to and unto), conjunctions (that), determiners (every, all), nouns (bunch, everybody, everything, guys, heart and part), phrases (on behalf of and ought to), prepositional phrase (due to), pronouns (both, my, and you), quantifiers (all [of]), and verbs (are, be, extended and go). The most frequent collocates following thank you and thanks in positions R1, R2 and R3 are for (preposition) and very much (adverb phrase) with 26 instances (33.8%).

5. Conclusion In this chapter, the forms of thanking identified both in previous empirical studies of thanking and in four spoken English corpora are discussed. The twofold focus of the study is the identification of the most frequent thanking expressions and their collocational and colligational patterns. This study concludes that corpus evidence shows a great variety of formulations of thanking, the majority of which are, however, realized as thank you or thanks (Stenström 1994; Cheng 2009). Comparisons across the four spoken corpora indicate a much greater variety of forms in the general usage of English by native speakers of English in a national corpus (BNC), whereas the forms in the two specialized corpora (BASE and MICASE) and in the intercultural corpus (HKCSE) are found to be rather limited. This corpus-based study has also highlighted the collocational and colligational patterns which characterize the use of thank you and thanks. Another conclusion of the study is that the linguistic findings related to the speech act of thanking in different spoken corpora can inform communicative activities in, for example, professional communication and learning and teaching.

6. Implications for language learning and teaching An important recommendation to be made for English language teachers and teaching materials writers, especially those involved in teaching English as a Lingua Franca, is that teaching materials need to incorporate a more accurate and wider range of linguistic realizations of speech acts, in order to better reflect the realities of actual language use, and thus enhance learners’ language awareness of the use of English, both written and spoken (Cheng 2009, 2012; Römer 2009; Seto 2009). With the aid of computational and analytical skills, innovative pedagogical practices aim at identifying and depicting how linguistic realizations of different speech acts can be implemented at classroom level.

378

Chapter Fifteen

Cheng (2012), for instance, describes student projects that adopt corpus-based or corpus-driven and data-driven approaches undertaken by language learners at an advanced level of study so that they can learn to become language researchers. These approaches could be used in pragmatic studies to investigate thanking and other speech acts. For instance, concordance outputs can be used for learners to deduce the functions of the forms of different speech acts (Tribble and Jones 1990). Alternatively, concordance outputs can be turned into a gap-filling exercise by deleting the keywords which are particular linguistic realizations of speech acts. Learners can be asked to supply the keywords themselves or can be given a list of linguistic realizations to select from (Tribble and Jones 1990; Meunier 2002). Learners can be divided into groups and asked to record their language output illustrating the use of different speech acts. The data can then be compared to the data retrievable in naturally occurring spoken discourse. In the process, learners can recognize the possible errors in their usage and know what is correct and valid (Meunier 2002). The present study has produced corpus evidence that can contribute to English language learning and teaching in a number of important and productive areas, ranging from curriculum development to pedagogical practices. Moreover, corpus data, being an important source of authentic English language usage in different real-life situations, is particularly valuable as language is ever changing. Pedagogical corpus applications could focus on both indirect and direct approaches to using corpora in teaching and learning. Indirect applications of corpora can help researchers and materials writers not only with the design of the language teaching syllabus and curriculum, but also with the content in reference works and teaching materials. Corpora of specialized texts and research findings based on them can also help improve pedagogical practice, curriculum development and material design. Direct applications of corpora can help language learners and teachers to understand the language patterns in an autonomous way with the datadriven learning method. Data-driven learning activities can also be applied to specialized corpora such as learner corpora and parallel (or translation) corpora. In English language education, it is problematic to rely on the introspections of teachers and textbook writers, and greater attention needs to be given to real-world language use. To provide effective English learning education and promote communicative competence among learners, teachers and textbook writers need to incorporate more accurate linguistic realizations into our teaching materials to better reflect the

Cross-Cultural Comparison of the Speech Act of Thanking

379

authentic use of English in naturally occurring spoken discourse. Corpus data is useful in offering authentic learning materials to help enhance the language awareness of not only teachers and textbook writers but also learners. Corpus-driven approach and data-driven learning are hence essential for the purpose of enhancing language awareness. Given the availability of corpora and searching tools, both teachers and learners can use corpus data in language teaching and learning contexts without specific training in or knowledge of information technology. Linguistic realizations of selected speech acts can be used in specific pedagogical and learning contexts (see for example Tsui 2005; Cheng 2006; Cheng and Cheng 2010; Mahlberg 2006). Corpus evidence can be used in many ways. At the regional level, it can serve as a basis for curriculum development. At the school level, it can be tailored to specific pedagogical goals and learning requirements. At the classroom level, it can be integrated into teaching and learning materials. The overall focus could be on selecting those lexico-grammatical items and other linguistic features with relatively high frequencies of occurrence in corpora and applying them to areas of curriculum development, pedagogical practices, and language awareness. On the basis of corpus evidence, more attention needs to be given to those items and features that are overlooked in existing curriculum and textbooks. Given the increasing availability of corpus data in the 21st century, both language teachers and learners can be actively engaged in exploring the use of authentic English in our daily life.

References Archer, D., K. Aijmer, and A. Wichmann. 2012. Pragmatics: An Advanced Resource Book for Students. Abingdon: Routledge. Aston, G. 1995. “Say ‘thank you’: Some pragmatic constraints in conversational closings.” Applied Linguistics 16 (1): 57-86. Austin, J. 1962. How to Do Things with Words. Oxford: Clarendon Press. Berglund, Y. 2005. Expressions of Future in Present-day English: A Corpus-based Approach. Uppsala: Acta Universitatis Upsaliensis. Biber, D. 2009. “A corpus-driven approach to formulaic language in English: Multi-word patterns in speech and writing.” International Journal of Corpus Linguistics 14 (3): 275-311. Brown, P., and S. C. Levinson. 1987. Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press.

380

Chapter Fifteen

Cheng, W. 2006. “Describing the extended meanings of lexical cohesion in a corpus of SARS spoken discourse.” International Journal of Corpus Linguistics 11 (3): 325-344. —. 2009. “‘Thank you’: How do conversationalists in Hong Kong express gratitude?” In Language in Life, and a Life in Language: Jacob Mey – A Festschrift, edited by B. Fraser, and K. Turner, 39-48. Bingley, UK: Emerald. —. 2012. Exploring Corpus Linguistics: Language in Action. Abingdon: Routledge. Cheng, W., and P. Cheng. 2010. “Correcting others and self-correction in business and professional discourse and textbooks.” In Pragmatics across Languages and Cultures, edited by A. Trosborg, and K. K. Luke, 443-466. Berlin: Mouton de Gruyter. Cheng, W., C. Greaves, C., and M. Warren. 2008. A Corpus-driven Study of Discourse Intonation. Amsterdam: Benjamins. Coulmas, F. 1981. “‘Poison to your soul’: Thanks and apologies contrastively viewed.” In Conversational Routine: Explorations in Standardized Communication Situations and Prepatterned Speech, edited by F. Coulmas, 69-91. The Hague: Mouton Publishers. Davies, M. 2004. BYU-BNC: The British National Corpus. http://corpus.byu.edu/bnc. Eisenstein, M., and J. W. Bodman. 1986. “‘I very appreciate’: Expressions of gratitude by native and non-native speakers of American English.” Applied Linguistics 7 (2): 167-185. Grainger, K. 2013. “Of babies and bath water: Is there any place for Austin and Grice in interpersonal pragmatics?” Journal of Pragmatics 58: 27-38. Herbert, R. K. 1986. “Say ‘thank you’ – or something.” American Speech 61 (1): 76-88. Jacobsson, M. 2002. “Thank you and thanks in early modern English.” ICAME Journal 26: 63-80. Jautz, S. 2013. Thanking Formulae in English: Explorations across Varieties and Genres. Amsterdam: Benjamins. Jung, W.H. 1994. Speech acts of “Thank You” and responses to it in American English. Paper presented at the 16th Annual Meeting of the American Association for Applied Linguistics. Baltimore, MD. Liao, B. 2013. “On appropriacy of thanking: Dynamic compensation and adaptation.” English Language Teaching 6 (5): 71-80. Mahlberg, M. 2006. “Lexical cohesion: Corpus linguistic theory and its application in English language teaching.” International Journal of Corpus Linguistics 11 (3): 363-383.

Cross-Cultural Comparison of the Speech Act of Thanking

381

Meunier, F. 2002. “The pedagogical value of native and learner in EFL grammar teaching.” In Computer Learner Corpora, Second Language Acquisition and Foreign Language Teaching, edited by S. Granger, J. Hung, and S. Petch-Tyson, 119-141. Amsterdam: Benjamins. Mey, J. 1993. Pragmatics: An Introduction. Oxford: Blackwell. Ohashi, J. 2013. Thanking and Politeness in Japanese: Balancing Acts in Interaction. Basingstoke, UK: Palgrave Macmillan. Römer, U. 2009. “Corpus research and practice: What help do teachers need and what can we offer?” In Corpora and Language Teaching, edited by K. Aijmer, 83-98. Amsterdam: Benjamins. Searle, J. R. 1965. “What is a Speech Act?” In Philosophy in America, edited by M. Black. New York: Allen & Unwin. —. 1969. Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. —. 1975. “Indirect Speech Acts.” In Syntax and Semantics, vol. 3, edited by P. Cole and J. L. Morgan. New York: Academic Press. —. 1976. Speech Acts. London: Syndics of the Cambridge University Press. Seto, A. 2009. “‘I agree with you’: A corpus-based study of agreement.” 3L: The Southeast Asian Journal of English Language Studies 15: 4167. Siebold, K. 2012. “Implicit and Explicit Thanking in Spanish and German.” In Speech Acts and Politeness across Languages and Cultures, edited by L. Ruiz de Zarobe, and Y. Ruiz de Zarobe, 155172. Bern: Peter Lang. Stenström, A.-B. 1994. An Introduction to Spoken Interaction. London: Longman. Thomas, J. 1995. Meaning in Interaction: An Introduction to Pragmatics. London: Longman. Tognini-Bonelli, E. 2001. Corpus Linguistics at Work. Amsterdam: Benjamins. Tribble, C., and G. Jones. 1990. Concordances in the Classroom: A Resource Book for Teachers. London: Longman. Tsui, A. 2005. “ESL teachers’ questions and corpus evidence.” International Journal of Corpus Linguistic 10 (3): 335-356. Wichmann, A. 2004. “The intonation of Please-requests: A corpus-based study.” Journal of Pragmatics 36 (9): 1521-1549. Wolfson, N. 1981. “Compliments in cross-cultural perspective.” TESOL Quarterly 15 (2): 117-124.

382

Chapter Fifteen

Wong, M. L.-Y. 2010. “Expressions of gratitude by Hong Kong speakers of English: Research from International Corpus of English in Hong Kong (ICE-HK).” Journal of Pragmatics 42: 1243-1257. Zeyrek, D. 2012. “Thanking in Turkish: A Corpus-based Study.” In Speech Acts and Politeness across Languages and Cultures, edited by L. Ruiz de Zarobe, and Y. Ruiz de Zarobe, 53-88. Bern: Peter Lang.

Cross-Cultural Comparison of the Speech Act of Thanking

383

Appendix TABLE 1. Frequencies of occurrence of the nineteen forms of thanking across the corpora Linguistic realizations of thanking Thank you

BASE

BNC

HKCSE

MICASE

Total

169

4,245

459

373

Thanks

52

998

80

310

Thank you very much Thanks very much

51

1,058

136

47

35

295

24

3

Thank you for

19

255

47

25

Thanks for

9

139

13

21

Thank you very much indeed Thank you very much for Thanks very much indeed Thanks a lot Thank you so much Thank you so much for Thanks so much Thanks awfully Thanks a lot for Thanks a bunch Thanks for everything I can’t thank you enough I don’t know how to thank you TOTAL

2

123

4

0

9

79

14

2

3

63

0

0

5,246 (56.89%) 1,440 (15.62%) 1,292 (14.01%) 357 (3.87%) 346 (3.75%) 182 (1.97%) 129 (1.40%) 104 (1.13%) 66 (0.72%)

7 0 0

0 9 3

1 3 2

16 5 1

24 (0.26%) 17 (0.18%) 6 (0.07%)

0 0 0 0 0

2 3 0 1 1

2 0 0 0 0

1 0 2 0 0

5 (0.05%) 3 (0.03%) 2 (0.02%) 1 (0.01%) 1 (0.01%)

0

0

0

0

0

0

0

0

0

0

356 (3.86%)

7,274 (78.89%)

785 (8.5%)

806 (8.74%)

9,221 (100%)

Chapter Fifteen

384

TABLE 2. Frequency of occurrence of forms of thanking in the corpora that are not reported in the literature Linguistic realizations of thanking Thanks to

BASE

BNC 69

HKCSE 8

MICASE 5

6

Thanks very much for Thank you to Thank you (subject) for Thank you both very much Thanks very much indeed for Thanks ever so much Many thanks to Thanks to (subject) for Many thanks Thank you all Thank you all for Thank you all very much Thank you every so much for Thank you all very much indeed Thank you all very much for Thank you ever so much Thank you to (subject) for Thanks also to Give thanks Many thanks indeed Special thanks to Thank you again Thank you again for Thank you guys for Thank you that

Total

3

37

3

0

8 0

16 15

2 0

1 0

88 (24.86%) 43 (12.15%) 27 (7.63%) 15 (4.24%)

0

2

11

0

13 (3.67%)

0

12

0

0

12 (3.39%)

0 0 0

12 11 10

0 0 0

0 0 0

12 (3.39%) 11 (3.11%) 10 (2.8%)

0 0 0 0

7 6 5 6

0 1 1 1

0 0 1 0

7 (2.82%) 7 (2.82%) 7 (2.82%) 7 (2.82%)

0

5

0

0

5 (1.41%)

0

5

0

0

5 (1.41%)

0

4

0

0

4 (1.12%)

0

4

0

0

4 (1.12%)

0

4

0

0

4 (1.12%)

0 0 0 0 0 0 0 0

4 3 3 2 1 0 0 2

0 0 0 0 1 2 0 0

0 0 0 0 0 0 2 0

4 (1.12%) 3 (0.85%) 3 (0.85%) 2 (0.56%) 2 (0.56%) 2 (0.56%) 2 (0.56%) 2 (0.56%)

Cross-Cultural Comparison of the Speech Act of Thanking Thank you very much indeed to Thank you very much, (subject), for Thank you very, very much Thanks are due to Thanks be to Cannot thank you enough for Cries out thanks beyond to Enormous thanks go to Many thanks for Many thanks indeed for Offer thanks on behalf of Particular thanks must go to Special thanks also to Thank all very much for Thank ever so much (subject) for Thank you (subject) also for Thank you all very, very much indeed for Thank you also for Thank you both Thank you both for Thank you both very much for Thank you both very much indeed Thank you both very much indeed for Thank you from Thank you over Thank you today for

385

0

2

0

0

2 (0.56%)

0

2

0

0

2 (0.56%)

0

2

0

0

2 (0.56%)

0 0 0

2 2 1

0 0 0

0 0 0

2 (0.56%) 2 (0.56%) 1 (0.28%)

0

1

0

0

1 (0.28%)

0

1

0

0

1

0 0

1 1

0 0

0 0

1 1

0

1

0

0

1

0

1

0

0

1

0 0

1 1

0 0

0 0

1 1

0

1

0

0

1

0

1

0

0

1

0

1

0

0

1

0 0 0 0

1 1 1 1

0 0 0 0

0 0 0 0

1 1 1 1

0

1

0

0

1

0

1

0

0

1

0 0 0

0 1 1

1 0 0

0 0 0

1 1 1

Chapter Fifteen

386 Thank you very deeply from the innermost parts of my heart for Thank you very indeed for Thank you very much indeed for Thank you very much to Thank you very, very, very, very much Thanks be unto Thanks entirely to Thanks ever so Thanks ever so much for Thanks every so much all of you Thanks every so much for Thanks everybody here for Thanks from Thanks go to Thanks largely to Thanks mainly to Thanks obviously through Thanks ought to be extended Thanks that Thanks to you lot Thanks very much to Thanks you for TOTAL

0

1

0

0

1

0

1

0

0

1

0

0

1

0

1

1

0

0

0

1

0

1

0

0

1

0 0 0 0

1 1 1 1

0 0 0 0

0 0 0 0

1 1 1 1

0

1

0

0

1

0

1

0

0

1

0

1

0

0

1

0 0 0 0 0

1 1 1 1 1

0 0 0 0 0

0 0 0 0 0

1 1 1 1 1

0

1

0

0

1

0 0 0 0 18 (5.08%)

1 1 1 1 295 (83.33%)

0 0 0 0 32 (9.04%)

0 0 0 0 9 (2.54%)

1 1 1 1 354 (100%)

CHAPTER SIXTEEN THE VARIABILITY OF COMPLIMENT RESPONSES: ITALIAN AND GERMAN DATA MARINA CASTAGNETO AND MIRIAM RAVETTO

1. Introduction Most studies about compliments have tried to define them, to identify their form and main functions. Others have focused their attention on specific constitutive aspects – i.e., topics, complimented items and syntactic construction – of the speech act of complimenting in a wide range of languages. There are also many studies comparing compliments and compliment responses (CRs) in very different varieties. In these contrastive analyses, one of the frequently studied languages is English (e.g., English-Japanese: Daikuhara 1986; American English-Chinese: Chen 1993; Australian English-Iranian Persian: Razi 2013; EnglishBulgarian: Slavinova 2011). Very few studies compare compliments and CRs among German and Italian speakers (Ravetto 2012; Payne 2013). In our paper1 we propose a corpus-based contrastive analysis of CRs in Italian and German. The two languages nicely lend themselves to a pragmatic comparison, as both are pluricentric languages, characterized by the co-presence of more than one geographic variety. In fact, because of the late unification of both these nations, their linguistic varieties did not conform, until recently, to the model of a single one, such as the prestigious variety spoken in the capital city. Furthermore, as far as we know, the literature on compliments has very seldom dealt with diatopic differences in the same linguistic area, or, at least, in the same country. Yet, linguistic varieties are astonishingly different in pragmatic conventions and practices, to the point that it is sometimes very difficult for people coming from different areas of the same country to feel like a linguistic community. According to Labov (1972, 120), a linguistic

388

Chapter Sixteen

community is determined by the common sharing of norms and abstract schemes of evaluation; now, the pragmatic management of speech acts (especially those acts focused on politeness such as compliments) is sometimes so different as to lead to the formation of reciprocal stereotypes among people living in the same country. So in our study we decided to pay attention also to the diatopic variation of CRs in Italy. The paper consists of four main parts. In 1.1 we define compliments and CRs, describing their constitutive aspects, their structure and the dia-levels across which they may show variation. As a second step, we present our corpus (1,880 CRs), the method of data collection and our CR classification framework set up to typologize our corpus of CRs. In 3. we compare Italian and German CRs, and we analyse the Italian data according to diatopic variation. In the last section we discuss the findings of our study.

1.1 Compliments: definition and constitutive aspects Compliments have been considered “verbal presents” (KerbratOrecchioni 1987, 10) consisting in the expression of personal admiration or the positive evaluation of a specific item. In this way, the act of complimenting performs its main function, the building and reinforcement of solidarity between the interlocutors (Pomerantz 1978; Wolfson 1983, 86). Beside this main function, compliments involve other secondary functions, such as reinforcing desired behaviors (Manes 1983) or softening face-threatening acts such as criticism or requests. The highly routinized formulation of compliments makes them easily identifiable in the stream of speech and minimizes the chance that they might be misinterpreted by the addressee. Wierzbicka (1987, 201) identifies the semantic components of the speech act of complimenting as follows: I perceive something good about your Y I want to say something good about you because of that I say: (something good about X and X’s Y) I feel something good about thinking about it I say this because I mean to cause you to know that I am thinking something good about you I assume that you will feel something good because of that.

The definition highlights as one of the requirements of an act of complimenting the presence of an addressee whose characteristics are complimented on. This aspect is also pointed out by Holmes (1988, 446), who defines the compliment as:

The Variability of Compliment Responses: Italian and German Data

389

a speech act which explicitly or implicitly attributes credit to someone other than the speaker, usually the person addressed, for some “good” (possession, characteristics, skill, etc.) which is positively valued by the speaker and the hearer.

Compliments are “remarkably formulaic speech acts” (Holmes and Brown 1987, 529), since they are realized by means of a very narrow range of syntactic and semantic patterns and of a restricted set of lexical items (Knapp, Hopper, and Bell 1984). Manes and Wolfson (1981) showed for American English that the following three most commonly occurring syntactic formulae could account for 85% of the compliments examined: (i) (a) NP BE (INT) ADJ (Your hair is really great) (b) NP BE looking INT ADJ (You are looking very pretty) (ii) I (INT) like/love NP (I simply love that skirt) (iii) PRO BE (INT) ADJ NP (That’s really a great juice)

With reference to German, Golato (2005) argues that complimenting utterances are characterized by a high-frequency of the formula (e.g., die ist schön, die Tasche, “that is nice, the bag”) and of positive adjectives, such as schön (“nice”) or lecker (“tasty”). In parallel, the choice of complimented topics is also characterized by repetitiveness and regularity. Moreover, the overwhelming majority of compliments place positive value upon objects or traits which are new, temporary or which are the results of one’s effort (e.g., make-up). In contrast, intimate characteristics or taboo situations are rarely selected as topics of compliments.

1.2 Compliments and linguistic variability Unlike Chen (1993, 52), who claims that, in a contrastive study like hers, “differences resulting from socio-ethnopragmatic factors are not crucial and may be disregarded,” we are firmly convinced that what is really interesting in compliments is their variability along the axes of the so-called dia-levels. Compliments, and in particular CRs, may show dramatic diatopic variation,2 also in the very same country (see 3.3). Variability applies, for one thing, to the objects of the complimenting illocution. Indeed, it appears that acceptable topics of compliments are closely related to the specific cultural norms of given societies. For instance, Chinese speakers give more compliments on abilities (Cheng 2003), while the most popular

390

Chapter Sixteen

topics in Egyptian interactions are natural appearance attributes, such as the color of the eyes (Nelson, El Bakary, and Al Batal 1996, 118). Furthermore, there are a number of functions which vary cross-culturally. For instance, complimenting is often used as an information-seeking means in Poland (Jaworski 1995, 79ff.): by giving a positive evaluation of an event or phenomenon, Polish speakers try to have information about the complimented topic (e.g., the price of the object). In other countries and cultural groups, compliments may be interpreted as implicit requests: the speaker’s expression of admiration for an object imposes, in general, an obligation on the complimentee to offer that object to the complimenter (Herbert and Straight 1989, 38; Holmes and Brown 1987, 526). Compliments also vary on the diastratic level. Many studies have taken into account the weight of gender in performing and receiving a compliment,3 pointing out especially that women are more likely than men to give, receive and reciprocate a compliment, but other sociolinguistic factors, for instance the interactants’ age, have not attracted the same amount of attention. Moreover, compliments vary on the diaphasic axis. Unfortunately, hardly any study has focused on some particular linguistic register.4 However, many scholars have often considered the role-relationship between complimenter and complimentee,5 mostly to conclude that compliments are exchanged between people of equal status. The diamesic variation, instead, was examined in depth by using three different methods (semi-spontaneous interactions, DCTs and role-plays; for details, see Golato 2005, 12ff.) to elicit compliments and CRs.

1.3. CRs In contrast to the formulation of compliments showing a mostly standardized nature (see section 1.1.), CRs vary both in their strategies and in their semantic structure. Pomerantz (1978) pointed out the state of inbetweenness of complimentees, who have to deal with two diametrically conflicting conversational principles: Principle I: Agree with the compliment Principle II: Avoid self-praise

The preferred reaction is an agreement with the complimenter, which is associated with the acceptance of the compliment. However, this norm results in praise of self, which minimizes sympathy and solidarity. Consequently, if complimentees agree with their interlocutors, they

The Variability of Compliment Responses: Italian and German Data

391

disregard the principle of avoiding self-praise, and if they reject the compliment, the Agreement Maxim is violated. Studies on CRs reveal that speakers use different answer types, some of them as a form of compromise between the conflicting maxims of agreement and avoidance of self-praise. That is why many studies categorized CRs by ordering them on a scale from a higher degree of acceptance or agreement through various forms of conflict between the two maxims to the refusal of the compliment (Pomerantz 1978; Frescura 1996; Ruhi 2007; Alfonzetti 2009; see section 2.2). In the next section, we introduce a new fine-grained typologization of CRs, trying to manage the conflict among the two maxims, as different balancing in politeness have been used by our informants. We add new tags to the previous typologizations of CRs in the literature on this field in order to have a more adequate and precise classification.

2. Method In the following sections we describe our corpus and the method of data collection. Then we illustrate the categorization framework of CRs used in our study; our classification integrates and modifies the previous taxonomies available in the literature, which we used as starting point for our analysis.

2.1. Corpus Our corpus consists of 1,880 CRs: 1,273 produced by Italian speakers and 607 produced by German ones. It was collected by using three different methods. The main method involved the recording of semispontaneous interactions. Compliments were deliberately elicited by Italian students talking to Italian native speakers in Italy and to German native speakers in Germany (the Italian students were studying as Erasmus students at German universities). CRs and the following turns were, on the contrary, fully spontaneous. Interactions took place during activities that the speakers would normally engage in with each other, such as lunches in the cafeteria, studying together, dinners, parties and so on. The interactions were recorded and transcribed, including contextual information such as the age and gender of the informants, the towns where the data were collected and the geographic origin of the complimentees. This made it possible to collect 1,060 CRs from Italian informants and 409 CRs from German speakers. Furthermore, we used a DCT (see an example in Appendix B2) to collect CRs from Italian and German speakers (156 and

392

Chapter Sixteen

131, respectively). The third method of data collection was a role-play thanks to which 57 CRs produced by Italian speakers and 67 by German ones were collected (see Appendix B3). The complimenters and the complimentees were all university undergraduate/graduate students who ranged in age from 18 to 32. The Italian data were collected in North-Western Piedmont (i.e., Vercelli, Novara) and in two towns in the South (Cancello Arnone, province of Caserta and Grottaglie, province of Taranto), and the German data in central-Western Germany (i.e., Marburg, Mannheim). Compliments concerned four different items: physical appearance, personal belongings, character and abilities (see Appendix A). In case of semi-spontaneous interactions and role-plays, in order to identify CRs we considered the conversational turns from the elicitation of the semi-spontaneous compliment to the point of the interaction in which the compliment topic was abandoned. The length of the interaction after the compliment varied from zero conversational turns (no CR) up to twenty-six turns. For the data collected by using the DCT, we considered a CR the whole text after each prompt. Following in other scholars’ wake (Pomerantz 1978; Knapp, Hopper, and Bell 1984; Holmes 1986; Herbert 1989; Chen 1993; Frescura 1996; Saito and Beecken 1997; Golato 2005; Tran 2007), we did not identify the CRs in our corpus on the basis of lexico-grammatical wording of given text segment, but rather on the basis of its main pragmatic function inside the whole act of the compliment.6 Furthermore, the Italian data were collected and organized on the basis of different contextual factors, for example the geographic area of data collection, the communicative style of the interlocutors (see 3.3), the gender of the complimentees (206 CRs), the morpho-syntactic form of the compliment formula (164 compliments formulated through a plain adjective or its superlative form). Up to now, we have not been in a position to do the same for the German data (i.e., to classify it) on the basis of the gender of the complimentees and the nature of the complimented item. The collection of contextual information regarding the German subcorpus is one of the main aims of our future research.

2.2 Classification of CRs The CR classification adopted in our study modifies and integrates the previous categorization frameworks, taking Frescura (1996) as a starting point. It consists of four categories and fourteen micro-categories.7

The Variability of Compliment Responses: Italian and German Data

393

TABLE 1. Classification of CR types and examples (I = Complimenter; C = Complimentee) I. Direct Acceptance

CR types 1. Thanking

2. Pleased Acceptance 3. Acceptance 4. Non-verbal Acceptance 5. Return II. Limited Acceptance

6. Ironic Acceptance

7. Minimization

Examples I: What a nice hat! C: Thanks8 I: Your jacket is very nice! C: I hope so. It cost me a fortune. I: Your shoes are really nice! C: Yes I: You are very kind C: (smiling) I: You are a very good cook! C: You too I: What a nice jacket. You look great! C: Hey, I make everything look good! I: Your bag is nice C: Yeah, quite nice

8. Deflection 8a. Lateral Deflection of the Merit 8b. Lateral Deflection of the Quality 8c. Lateral Deflection of the Topic (Explanation / Justification) 9. Reassurance Request III. NonAcceptance

10. Reductive Deflection 11. Rejection 12. Discredit of the Complimentee

13. Discredit of the Complimenter IV. Ignoring

14. Ignoring

I: Good job! C: Clara helped me I: Your pullover is really nice C: It is warm I: You have nice shoes C: I bought them in Germany

I: You always have good ideas C: Really? I: This roast beef is tasty! C: It is insipid! I: Good, you can explain very well! C: I don’t think so I: You look very nice C: I look like the witch from Hansel and Gretel! I: What a nice hairstyle! C: You are blind! I: Your bag is really nice C: Can you give me my mobile phone?

394

Chapter Sixteen

As in the model proposed by Tran (2007), the CR types were ordered, within each of the four categories, from the highest to the lowest degree of acceptance. The new micro-categories we introduced in our taxonomy were: Acceptance (No. 3 in the Table 1), Nonverbal Acceptance (No. 4), Lateral Deflection of the Topic (No. 8c), Discredit of the Complimentee (No. 12). We introduced the tag Acceptance (No. 3), not found in Frescura’s taxonomy, for the cases in which the complimentee chose to answer by means of a very short agreement token (e.g., yes).9 But compliments can also be accepted only by nonverbal means, such as a smile (Lee 2009), or kisses and hugs (Ruhi 2007). In this way the complimentees show their agreement with the compliment but, at the same time, they mitigate face threats otherwise evoked by an overt verbal response. Among the Limited Acceptances, at a first step in a hypothetical continuum from acceptance to denial, we inserted the micro-category Ironic Acceptance. Considering just the realization of CRs, it is very difficult to discriminate between Ironic acceptance and Pleased acceptance. This is a form of acceptance through which the complimentee agrees with the previous utterance and increases its complimentary force, as in the instance offered by Tran (2007): A: nice car! B: Thanks. Brand new.10 Yet, in our opinion, Ironic Acceptance, so similar on the formal level to Pleased Acceptance, has to be considered a kind of Limited Acceptance, because it displays a joking or playful connotation. In compliment sequences like the following: A: Your eyes are really beautiful B: Didn’t you ever notice it before? I’m famous for that, if the key11 were not ironical, the CR would be a real case of boasting and vainglory, in obvious violation of the Maxim of Modesty. So, by means of Ironic Acceptance the complimentee is actually cooperating with the complimenter, because he/she is trying to trigger the non-natural meaning of a joke in her/his interlocutor’s mind, at the same time mitigating the complimentary force of the speaker’s positive evaluation. Basically, in Ironic Acceptance the complimentee resorts to Gricean conversational implicature, and, in order to be understood by the complimenter, she/he resorts to hints such as a marked intonational profile of her/his utterance and non-verbal or prosodic signals such as smiles. Moreover, the content of the assessment contained in this kind of CR is often hyperbolic in the context – very few people can declare in truth to have wonderful eyes. So, the interlocutor is led to understand that he/she is facing an ironic response, and that the complimentee is trying to evade from the conversational duty to respond to the compliment received: if you say the contrary of what you mean, it is difficult to assign a truth value to the

The Variability of Compliment Responses: Italian and German Data

395

statement. Maybe it is not by chance if Ironic Acceptance was not so frequent in the corpus, and it was limited to dialogic interactions with a strong intimacy between the complimenter and the complimentee, when the risk of misunderstandings was reduced.12 Here is an instance of Ironic Acceptance from our Italian sub-corpus: (1)

I001: sai fare pure queste cose, che bravo C002: Eppure non trovo moglie, ma si può I003: C004: le donne (English translation) I001: you know well how to do such things, bravo! [sowing potatoes] C002: yet, I don’t find a wife, how is it possible? I003: C004: women

In the category Limited Acceptance, we added a new kind of Lateral Deflection side by side to Lateral Deflection of the Merit and of the Quality: Lateral Deflection of the Topic. This tag was used for both the cases in which the complimentee gave further unrequested information about the object of the compliment (as in the instance in Table 1) and the cases in which the complimentee tried to save her/his positive face by explaining why she/he had the particular object or ability mentioned in the compliment. We did not choose a tag like Comment History (as in Herbert 1989; Golato 2005) or Informative Comment (Holmes 1986), preferring the tag Explanation13 because we wanted to highlight the difference between Explanation and Justification: by Explanation the complimentee depersonalized the compliment force by giving impersonal details,14 as the following example from our German sub-corpus shows: (2)

I001: du hast eine schöne Tasche C002: bei bei engelhorn gibt’s viele von diesen taschen (English translation) I001: you have a nice bag C002: in in engelhorn there are many of these bags

Instead, through Justification she/he tried to avoid looking vain (Example 3). This last kind of CR has never before been identified in the literature on compliments.

Chapter Sixteen

396 (3)

I001: wie schön du hast schöne Schuhe C002: ich war gestern so ärgerlich ja und wenn ich schlecht gelaunt bin, dann kauf ich mir was also, hab mir ein neues paar gekauft (English translation) I001: how nice you have nice shoes C002: yesterday I was so angry ja and when I am in a bad mood, I buy me something so I have bought a new pair

A very problematic micro-category is Reassurance Request (No. 9). It was placed in very different positions in previous CR taxonomies. For instance, according to Knapp, Hopper, and Bell (1984) it was a sort of acceptance with amendment. Herbert (1989) grouped it under the “NonAgreement” category, asserting that, in this case, it is hard to understand if the complimentee wants to accept or reject the compliment. Chen (1993) considered it a deflection, while, according to Frescura (1996), it was a form of non-acceptance. The Reassurance Request is difficult to categorize because it is a recycling move (Alfonzetti 2009) extending the compliment over more than two conversational turns (see 3.3), that is, it solicits a repetition of the compliment after the turn with the Reassurance Request. This response type can either be a form of disagreement or work as a politeness strategy of pre-acceptance. Furthermore, it can be a way to fish for more compliments (Tran 2007). Sometimes the Reassurance Request questions the sincerity of the complimenter and so the condition of appropriateness of the speech act. Here is an instance from our Italian sub-corpus, in which the Reassurance Request buono? (“Is it?” turn C002) was immediately followed by the complimentee’s sudden withdrawal from the conversation. She evaded the duty to respond to the compliment by physically removing herself from the communicative context: (4)

I001: che buon caffettino, Rituccia C002: buono? Vabbuò, io vado a fare la doccia I003:

(English translation) I001: what a good coffee, Rituccia C002: is it? Ok, I go to shower I003:

The Variability of Compliment Responses: Italian and German Data

397

We placed Reassurance Request in the last position within the category Limited Acceptance, because we thought that through this CR speakers neither agreed with the compliment nor rejected it. They preferred to stay in a sort of in-betweeness, as they did not show at once their position about the praising aspect of the compliment; rather, they waited for a second elicitation by complimenters, who probably repeated or reinforced their first evaluation. Indeed, complimentees normally expressed their opinion (i.e., agreement or disagreement) about the compliment after it had been repeated (they agreed or disagreed with the compliment). In the end, unlike other studies, we considered Ignoring (No. IV/14; Example 5) an independent category. In Golato’s (2005) taxonomy it was placed among the solutions between acceptance and non-acceptance. Holmes (1986) defined it as Deflect/Evade. However, ignoring a given compliment is not always intentional, so we preferred to classify Ignoring neither as a form of acceptance nor as a non-acceptance CR. (4)

I001: bella questa borsa, Francy I002: figa [No answer] (English translation) I001: nice bag, Francy I002: cute [No answer]

3. Analysis of the data 3.1 Italian and German CRs in comparison Table 2 shows the distribution of CRs among the four categories in the Italian and the German data:15 TABLE 2. Distribution of CRs among the four categories for Italian and German Categories Direct Acceptance Limited Acceptance Non-Acceptance Ignoring Total

Italian 154 (50.16%) 81 (26.38%) 61 (19.86%) 11 (3.58%) 307 (100%)

German 96 (30.37%) 159 (50.31%) 56 (17.72%) 5 (1.58%) 316 (100%)

398

Chapter Sixteen

With regard to the global distribution of the four categories, we noticed that all of them were exemplified in the two sub-corpora. The most frequent one was Direct Acceptance (250 total occurrences). Limited Acceptance was somewhat less frequent than Direct Acceptance (240 total occurrences), followed by Non-Acceptance with 117 occurrences, while Ignoring was the rarest category in our data (only 17 total occurrences). The comparison between the Italian and the German sub-corpora led to further findings. Unlike the German data, which showed a preference for Limited Acceptance, the Italian CRs had a preference for the Direct Acceptance strategy, mostly through the Thanking micro-category (Table 2; see Ravetto 2012, 2013). This was especially true when the informants were young people living in the North-Western area of Italy. The Ignoring category gave us precious information on the pragmatic management of dangerous linguistic acts in the two cultures as well: the German informants tended to give an answer to compliments in any case (only 1.58% of Ignoring), while the Italian speakers chose not to reply more than twice as often (3.58% of occurrences). This could potentially be considered a pragmatic clue of the German preference for respecting the Gricean Cooperative Principle and the Manner Maxim when it enters into conflict with the Politeness Principles. As a matter of fact, Ignoring is a way to escape from the conversational duty to respond to a compliment: as such, silence can be ambiguous about the degree of agreement and disagreement of the complimentee, and the German informants tended to avoid any kind of ambiguity. The comparison between German and Italian at the micro-category level is provided in Table 3. The German informants basically avoided Nonverbal Acceptance (just 0.31%), while among the Italian speakers this micro-category accounted for 9.44% of the data: again, the Germans preferred an overt formulation of information. Through an open statement of accepting or refusing, the speaker took on herself/himself the responsibility of the speech act she/he was performing, thus avoiding again the violation of the fourth Gricean maxim “Be clear and unambiguous.” Moreover, Pleased acceptance and Ironic Acceptance were strongly dispreferred in our German sub-corpus: Pleased acceptance was almost half as frequently instantiated in the German as in the Italian data (5.69% vs. 10.09%), and we had just a single occurrence of Ironic Acceptance (0.31%) in the German data, compared to 14 occurrences (4.56%) in the Italian sub-corpus. Again, joking was probably considered too ambiguous by the German informants, who tended to be more oriented towards clarity and plain speaking.

The Variability of Compliment Responses: Italian and German Data

399

TABLE 3. Distribution of CR types in Italian and German CR types Direct Acceptance Thanking Pleased Acceptance Acceptance Nonverbal Acceptance Return Limited Acceptance Ironic Acceptance Minimization (a) Lateral Deflection of the Merit (b) Lateral Deflection of the Quality (c) Lateral Deflection of the Topic Reassurance Request Non-Acceptance Reductive Deflection Rejection Discredit of the Complimentee Discredit of the Complimenter Ignoring Ignoring

Italian 154 (50.16%) 48 (15.63%) 31 (10.09%) 36 (11.72%) 29 (9.44%) 10 (3.25%) 81 (26.38%) 14 (4.56%) 8 (2.60%) 4 (1.30%) 9 (2.93%) 41 (13.35%) 5 (1.62%) 61 (19.86%) 22 (7.16%) 24 (7.81%) 9 (2.93%) 6 (1.95%) 11 (3.58%) 11 (3.58%)

German 96 (30.37%) 42 (13.29%) 18 (5.69%) 14 (4.43%) 1 (0.31%) 21 (6.64%) 157 (50.31%) 1 (0.31%) 14 (5.06%) 16 (5.06%) 11 (3.48%) 50 (15.82%) 65(20.56%) 56 (17.72%) 23 (7.27%) 3 (0.94%) 29 (9.17%) 1 (0.31%) 5 (1.58%) 5 (1.58%)

Going deeper into the contrastive analysis, we noticed considerable differences between the Italian and the German data in responding to compliments on different topics (see Tables 4 and 5). As a comparison of Table 4 and Table 5 reveals, for the German speakers the management of compliments on physical appearance and personal belongings was harder to deal with than for the Italian speakers; on the contrary, the Italian informants were more cautious in accepting compliments on character and abilities. In fact, among the tokens of Direct Acceptance in the German data (96), 48 cases concerned character, 22 occurrences were on abilities. In particular, 21 cases of Thanking were produced after compliments on character (twice as many cases as on compliments for abilities, four times as many cases compared to Thanking for compliments on other topics). Moreover, with its 14 occurrences, the choice of returning compliments on character was by far the preferred choice as opposed to returning compliments on other topics.

Chapter Sixteen

400

TABLE 4. Distribution of CRs in relation to the object of the compliment – Italian sub-corpus

Direct Acceptance

Limited Acceptance

NonAcceptance

Ignoring Total

Thanking Pleased Accept. Acceptance Nonverbal Accept. Return Ironic Accept. Minimization (a) Lateral Deflection of the Merit (b) Lateral Deflection of the Quality (c) Lateral Deflection of the Topic Reassurance Request Reductive Deflection Rejection Discredit of the Complimentee Discredit of the Complimenter Ignoring

Physical appearance 26 (32.5%) 5 (6.2%)

Personal belongings 11 (14.2%) 16 (20.7%)

7 (8.7%) 15 (18.7%)

18 (23.3%)

2 (2.5%) 7 (8.7%) 2 (2.5%)

Character

Abilities

7 (9.2%) 5 (6.5%)

4 (5.4%) 5 (6.7%)

4 (5.2%) 11 (14.4%)

7 (9.4%) 3 (4%)

2 (2.6%) 6 (7.8%)

6 (8.1%) 1 (1.3%)

6 (7.7%) 2 (2.5%)

2 (2.7%)

3 (3.7%)

6 (7.7%)

7 (8.7%)

9 (11.6%)

5 (6.5%)

3 (3.7%)

1 (1.2%)

1 (1.3%)

1 (1.2%)

4 (5.1%)

7 (9.2%)

10 (13.5%)

2 (2.5%)

2 (2.5%)

12 (15.7%) 5 (6.5%)

8 (10.8%) 4 (5.4%)

2 (2.5%)

4 (5.2%)

77 (100%)

7 (9.2%) 76 (100%)

80 (100%)

20 (27%)

4 (5.4%) 74 (100%)

The Variability of Compliment Responses: Italian and German Data

401

TABLE 5. Distribution of CRs in relation to the object of the compliment – German sub-corpus

Direct Acceptance

Limited Acceptance

NonAcceptance

Ignoring Total

Thanking Pleased Accept. Acceptance Nonverbal Accept. Return Ironic Accept. Minimization (a) Lateral Deflection of the Merit (b) Lateral Deflection of the Quality (c) Lateral Deflection of the Topic Reassurance Request Reductive Deflection Rejection Discredit of the Complimentee Discredit of the Complimenter Ignoring

Physical appearance 6 (7.5%) 4 (5%)

2 (2.5%)

Personal belongings 5 (6.1%) 6 (7.4%)

Character

Abilities

21 (28%) 8 (10.6%)

10 (12.5%)

2 (2.4%)

4 (5.3%) 1 (1.3%)

7 (8.7%)

1 (1.2%)

14 (18.6%)

5 (6.2%) 1 (1.2%) 8 (10%) 14 (17.5%)

3 (3.7%)

5 (6.6%) 2 (2.4%)

2 (2.5%)

9 (11.1%)

11 (13.7%)

29 (35.8%)

6 (8%)

4 (5%)

37 (46.2%)

14 (17.2%)

5 (6.6%)

9 (11.2%)

3 (3.7%)

9 (11.1%)

4 (5.3%)

7 (8.7%)

4 (5.3%)

1 (1.2%) 14 (17.5%)

3 (4%) 75 (100%)

80 (100%)

2 (2.4%) 11 (13.7%)

1 (1.2%)

80 (100%)

2 (2.4%) 81 (100%)

On the other hand, Lateral Deflection of the Topic, by which it is possible for the complimentee to reply to the previous turn without addressing the compliment, and Reassurance Request were far more frequent in compliments on personal belongings and physical appearance (91 occurrences on 159 cases of Limited Acceptance). This suggests that these two topics were considered harder for the complimetee to handle.

402

Chapter Sixteen

Vice versa, the Italian data showed that, for the Italian speakers, compliments on character and abilities were more difficult to accept. These two topics, then, were interpreted to be more dangerous in terms of face. So Lateral Deflection of the Topic was more frequent for compliments on character and abilities (25 occurrences, including 20 occurrences of topic shift when complimenting on abilities), and so was Reductive Deflection. Compliments on character and abilities were rejected in 20 cases (compliments on physical appearance only in 4 cases) and, as they were so risky, they could be ignored as well (there was no case at all of ignoring compliments on different topics). Physical appearance were easier for the Italian speakers to accept: on 80 compliments on this topic, 55 were directly accepted (and almost half of them by Thanking), and only in two cases were they rejected. The Italian and German pragmatic behaviour was not so similar, then, in compliment management.

3.2 An ongoing pragmatic change in Italy? We have to consider that this sub-corpus of 307 compliments concerns all young informants coming from Eastern Piedmont. The trend towards Direct Acceptance was an unexpected result, considering that the few existing studies on Italian compliments underlined how Italian speakers tended to resort to Limited Acceptance (Frescura 1996; Alfonzetti 2007). In our opinion, the Italian way of handling compliments is becoming more similar to that in use in the USA and New Zealand. In these countries a Direct Acceptance by Thanking is the norm (Holmes 1986; Chen 1993; Frescura 1996). We did not carry out a longitudinal study and we did not set up a research protocol to collect compliments in order to be compared at specific time intervals. Nevertheless, we claim that Italy may be going through a pragmatic change if we consider the percentage of Direct Acceptances in our sub-corpus side by side with the data shown in Frescura (1996), and interpreting them in an apparent diachronic perspective. In Frescura’s (1996) corpus (979 CRs elicited in Italy) only 10,82% of CRs were ascribable to Direct Acceptance, while 82% were cases of Limited Acceptance. According to Frescura, the Italian speakers were therefore a perfect example of what Chen (1993) called conflict culture, a culture in conflict between Leech’s Agreement and Modesty Maxims. Let us compare Frescura’s data to the CRs in our sub-corpus of 164 compliments (Aina 2013). This sub-corpus was built on a balanced sample, considering the grammatical form of the adjective (plain or

The Variability of Compliment Responses: Italian and German Data

403

superlative), and the gender and the age of the informants (see 2.1). Complimenters and complimentees were unfamiliar to each other. The CRs, thus divided into eight groups, were elicited about three topics: personal belongings, natural physical attributes (smile, eyes) and temporary physical attributes, due to the care and effort of the complimentee (haircut, make-up). As Table 6 shows, the percentage of Direct Acceptance in our study, ranging between 57% and 75%, exceeded that of Frescura (1996) by 10.82%. TABLE 6. Direct Acceptance in Aina’s sub-corpus

Women Men Total

Young informants (18-30) 79% 70% 75%

Adults (>30) 59% 56% 57%

In Aina’s sub-corpus, Direct Acceptance was much higher among young informants than among adults. Direct Acceptance was used 57% of the time among the adult informants, while the young informants chose this category in 75% of the time. Moreover, we should not forget that Frescura’s work dated back to 1996, 18 years ago, and her corpus of CRs had been collected in the preceding two years. Summing up these data, the hypothesis of an ongoing pragmatic change in Italy does not seem to be unwarranted. The trend to boldly accept compliments on physical appearance, felt as potential FTAs in most cultures, is another side of this pragmatic change. In many Western countries, such as the USA, compliments on appearance are considered embarrassing if they are about natural physical attributes, but they are absolutely welcome if they concern clothes or hair (Wolfson 1983, 90). Manes (1983, 99) agreed with Wolfson’s statement, adding that the topics particularly likely to attract compliments were those aspects of appearance “which are the result of deliberate effort.” Natural physical attributes, on the other hand, were an appropriate topic for a compliment among Egyptian speakers (Nelson, El Bakary, and Al Batal 1996). Now, in our sub-corpus the young Italian speakers from the north-western area gladly accepted compliments on every aspect of physical appearance. Maybe beauty has become too much a matter of business, by now, so it has quite always been produced by the addressee’s skill or effort, as once was a good meal or a job well done. In particular, it was really surprising that, when the topic of the compliment was a natural physical attribute, Direct Acceptance was the

404

Chapter Sixteen

only possible choice for young women (100%; all through Thanking), even if they are exactly the social group that might have found it the hardest to accept the compliment in order for them to be able to save their positive face. Young men accepted directly the compliments on this kind of topics “just” in 72.72% of the compliment sequences. Similar results were found in the Aina sub-corpus when analyzing Direct Acceptance of compliments on personal belongings: young women accepted through Thanking in all cases but one, and young men were only a little bit more cautious (Thanking in 62% of the occurrences). Considering the form of the adjective in the compliment (superlative vs. plain form), we noticed that young women were more sensitive than adult women to the form of the adjective in the compliment; when the adjective displayed its superlative form the percentage of direct acceptances by Thanking was significantly lower (83,33%). Adult informants did not appear to be so sensitive to this pragmalinguistic factor. If we are really facing a pragmatic change, Hudson’s (1980) assumption that linguistic changes are started by young women would be confirmed on the pragmatic level too.

3.3 The importance of diatopic variation In order to check whether the different varieties of Italian display different complimenting patterns, two sub-corpora were collected (see 2.1). The first one, consisting of 105 compliments elicited in Eastern Piedmont and in Grottaglie (province of Taranto), raised ethnolinguistic questions about the possible power of the complimenter to cast an evil eye on the complimentee (see Sidraschi 2011). Sidraschi’s informants from Grottaglie were young men and women, aged from 18 to 30 years. All of them hold a secondary-school diploma, but they are actually jobless or they are doing odd jobs (waiters, factory workers). In Grottaglie 52 semi-spontaneous CRs were collected. The fear for the evil eye potentially cast by a compliment emerged in interviews and in some spontaneous conversations collected through the ethnographic method. Only a few informants declared that compliments have been perceived as dangerous just by old people; on the other hand, spontaneous conversations showed that this belief has been lively also among the young. In particular, an informant (30 years old) attributed the whining behaviour of his sister to the evil eye cast by compliments to her. The other sub-corpus consisted of 125 compliments from Eastern Piedmont and 125 from Cancello Arnone (province of Caserta). These compliments were about five topics: three of them were “classic” topics in

The Variability of Compliment Responses: Italian and German Data

405

the literature on CRs (personal belongings, natural physical attributes like eyes or hair, personal qualities or abilities), while the remaining two topics were thought to be embarrassing and potentially difficult to accept (to be sexy, to be a swottier; see Di Francesco 2010). A contrastive analysis of the data showed a difference between the North and the South of Italy (see 3.2). Table 7 reveals a prevalence of Direct Acceptance in the Northern data, while on the Southern side we see a preference for Limited Acceptance, matching the findings presented by Frescura (1996) and Alfonzetti (2009) about the whole Italian situation. TABLE 7. Distribution of CRs on the four categories in Cancello Arnone (Campania) Categories Direct Acceptance Limited Acceptance Non-Acceptance Ignoring

Piedmont 40.0% 32.8% 20.8% 6.4%

Campania 35.2% 44.0% 16.8% 4.0%

Considering the micro-categories, we saw how the Northern data were consistent with the observations drawn from the contrastive analysis between the Italian and the German data (see 3.1) and from the analysis of the Aina sub-corpus aimed to check the different illocutionary force of the superlative form of the adjective (see 3.2). Direct Acceptance prevailed on Limited Acceptance, and among the former the most frequent answer was Thanking (32% of the occurrences of Direct Acceptance). The plain acceptance, when the complimentee just answered yes, showed twice as many occurrences than in the Southern data. With regard to Non-Acceptance, the Northern Italian informants more frequently chose the Reductive Deflection, an open rejection (16 occurrences vs. 12 in the Southern sub-corpus), or ignored the compliment (8 cases vs. 5 in the South). Moreover, the selection of the micro-categories Thanking, Reductive Deflection, Rejection and Ignoring was definitely more frequent when the informant was a woman (15 occurrences of Thanking vs. just one by a man; 28 occurrences of Reductive Deflection, Rejection and Ignoring altogether but 4 in men’s CRs). Looking at the Southern Italy sub-corpus, we saw how Limited Acceptance prevailed on Direct Acceptance. Among Limited Acceptance, two micro-categories were definitely preferred, and both were not so frequent in the Northern data: Ironic Acceptance (20% of Limited Acceptance) and Reassurance Request (25%). In the case of our Southern

406

Chapter Sixteen

Italian sub-corpus, Reassurance Request was a strategy for avoiding an open answer, thus avoiding the violation of Leech’s maxim of Modesty (in case of acceptance) or to refuse the verbal gift of the complimenter (in case of rejection). Sometimes the need to protect one’s positive face was so strong that the complimenter chose to answer the compliment by discrediting him/herself (I001: you are so sexy C002: I look like Benevento’s witch), or by denigrating the complimenter. Discredit of the Complimenter and Discredit of the Complimentee are both quite rare in the sub-corpus, because they are proper FTAs. Yet, we had four occurrences in Cancello Arnone and just two in Eastern Piedmont. Three occurrences out of four concerned the topic “being sexy.” The Discredit of the Complimenter was extremely rare (just one occurrence in Eastern Piedmont and one in Cancello Arnone, according to the Di Francesco sub-corpus) because it attacks the complimenter’s positive face. The greater attention paid in Campania to politeness strategies in general was confirmed by the following linguistic clue: compliments tended to be performed as long compliment sequences, and the compliment interactions were in some cases very long, including up to 26 conversational turns. Other people (relatives, boyfriends) contributed to supporting the conversation too. Often at a first glance they could look aggressive, but, as a matter of fact, their intervention freed the complimentee from the need to give a response to the compliment, which allowed him/her to protect his/her own face. Sometimes pragmatics is very tricky and a supportive conversational behaviour can be masked as aggressive: southern Italian informants disguised themselves as rude in order to be extremely polite.

4. Conclusion As far as we know, there are only a few contrastive studies about compliments among Italian and German speakers, and they are all devoted to the analysis of some particular sociolinguistic factor of the compliment (as is the case of Payne 2013). Our study dealt with the realization of CRs by Italians and Germans. On the basis of a wide corpus, we showed similarities and differences between the two groups of informants. The Italian speakers had a preference for Direct Acceptance strategies, while the German informants mainly used Limited Acceptance CRs. Moreover, Nonverbal Acceptance and Ignoring were sometimes attested in the Italian data; on the contrary, the German speakers clearly preferred an overt and a verbal response to compliments. If we consider the topics of compliments,

The Variability of Compliment Responses: Italian and German Data

407

the two sub-corpora revealed further differences. In reacting to compliments on physical appearance and personal belongings, the German informants mainly used Limited and Non-Acceptance CRs. In the Italian data, these kinds of CRs were far more frequent with compliments on character and abilities. Furthermore, compliments and CRs varied not only crosslinguistically but also within the same language. In fact, our sub-corpora of Northern and Southern Italian varieties displayed clear differences as regards the CRs types. Direct Acceptance CRs were frequently attested in the North of Italy, which showed the tendency of an open acceptance or refusal of compliments. The Southern Italian informants more frequently chose Limited Acceptance CR types. Yet, in order to obtain an extensive analysis of such a complex linguistic act, much work is still to be carried out. A further step of our study will be a contrastive analysis of the data looking specifically at the gender of both the complimenters and the complimentees in the interaction (to be divided in four groups: woman-woman; woman-man; man-woman, man-man). Moreover, we will collect additional, balanced data to study the diatopic pragmatic variation of compliment behaviour in the same country: the Italian and German ways of replying to compliments will be investigated in different geographic (and cultural) regions of the two nations.

References Aina, A. 2013. “‘Mi scusi…sa che ha degli occhi bellissimi?’ La pragmatica del complimento rivolto ad estranei.” PhD dissertation, Università del Piemonte Orientale, Vercelli. Albrecht, J. 1986. “‘Substandard’ und ‘Subnorm’. Die nichtexemplarischen Ausprägungen der ‘historischen Sprache’ aus varietätenlinguistischer Sicht.” In Sprachlicher Substandard, edited by G. Holtus, E. Radtke, vol. I, 65-88. Tübingen: Niemeyer. Alfonzetti, G. 2009. I complimenti nella conversazione. Roma: Editori Riuniti. Berruto, G. 1987. Sociolinguistica dell’Italiano contemporaneo. Roma: NIS. Chen, R. 1993. “Responding to compliments. A contrastive study of politeness strategies between American English and Chinese speakers.” Journal of Pragmatics 20: 49-75. Cheng, W. 2003. Intercultural Conversation. Amsterdam: Benjamins. Coseriu, E. 1973. Lezioni di linguistica generale. Torino: Boringhieri.

408

Chapter Sixteen

Daikuhara, M. 1986. “A study of compliments from a cross-cultural perspective: Japanese and English.” Working Papers in Educational Linguistics 2 (2): 103-135. Di Francesco, L. 2010. “Stessi complimenti, reazioni diverse. Stili comunicativi a confronto nel Piemonte Orientale ed in Campania.” PhD dissertation, Università del Piemonte Orientale, Vercelli. Farghal, M., and M. A. Al Khatib. 2001. “Jordanian college students’ responses to compliments. A pilot study.” Journal of Pragmatics 33: 1485-1502. Frescura, M. 1996. “The conflictual behavior of Italian speakers in responding to compliments.” Rassegna Italiana di Linguistica Applicata 28: 89-110. Furkó, B. P., and É. Dudás. 2012. “Gender differences in complimenting strategies with special reference to the compliment response patterns of Hungarian.” Argumentum 8: 136-157. Golato, A. 2005. Compliments and compliment responses: grammatical structure and sequential organization. Amsterdam: Benjamins. Herbert, R. K. 1989. “The ethnography of English compliments and compliment responses: a contrastive sketch.” In Contrastive Pragmatics, edited by W. Oleksy, 3-35. Amsterdam: Benjamins. —. 1990. “Sex-based differences in compliment behaviour.” Language in Society 19: 201-224. Herbert, R. K., and H. St. Straight. 1989. “Compliment-rejection versus compliment-avoidance: listener-based versus speaker-based pragmatic strategies.” Language and Communication 9: 35–47. Holmes, J. 1986. “Compliment and Compliment Responses in New Zealand English.” Anthropological Linguistics 28 (4): 485-508. —. 1988. “Paying Compliments: A Sex-Preferential Politeness Strategy.” Journal of Pragmatics 12: 445-465. Holmes, J., and D. F. Brown. 1987. “Teachers and students learning about compliments.” TESOL Quarterly 21: 523-546. Hudson, R. 1980. Sociolinguistics. Cambridge: Cambridge University Press. Hymes, D. 1974. Foundations of sociolinguistics: An ethnographic approach. Philadelphia: University of Pennsylvania Press. Jaworski, A. 1995. “This is not an empty compliment! Polish compliments and the expression of solidarity.” International Journal of Applied Linguistics 5: 63–94. Kerbrat-Orecchioni, C. 1987. “La description des échanges en analyse conversationelle: l’example du compliment.” DRLAV-Revue de Linguistique 36/37: 1-53.

The Variability of Compliment Responses: Italian and German Data

409

Knapp, M., R. Hopper, and R. Bell. 1984. “Compliments: A Descriptive Taxonomy.” Journal of Communication 34: 19-31. Labov, W. 1972. Sociolinguistic patterns. Oxford: Blackwell. Lee, C.-L. 2009. “Compliments and responses during Chinese New Year celebrations in Singapore.” Pragmatics 19: 519-541. Manes, J. 1983. “Compliments: a mirror of cultural values.” In Sociolinguistics and language acquisition, edited by N. Wolfson, and E. Judd, 96-102. Rowley, MA: Newbury House. Manes, J., and N. Wolfson. 1981. “The Compliment Formula.” In Conversational Routine: Explorations in Standardized Communication Situations and Prepatterned Speech, edited by F. Coulmas, 116-132. The Hague: Mouton Publishers. Nelson, G. L., W. El Bakary, and M. Al Batal. 1996. “Egyptian and American compliments: Focus on second language learners.” In Speech across cultures. Challenges to communication in a second language, edited by S. M. Gass, and J. Neu, 109-128. Berlin: Mouton de Gruyter. Parisi, C., and P. Wogan. 2006. “Compliment Topics and Gender.” Women and Language 29: 21-28. Payne, S. 2013. “Compliment Responses of Female German and Italian University Students: A Contrastive Study.” Language Studies Working Papers 5: 22-31. Pomerantz, A. 1978. “Compliment responses. Notes on the co-operation of multiple constraints.” In Studies in the organization of conversational interaction, edited by J. Schenkein, 79-112. New York: Academic Press. Ravetto, M. 2012. “Le risposte al complimento in italiano e in tedesco. Analisi di corpora di parlato.” SILTA 1: 85-122. Ravetto, M. 2013. “Das Komplimentieren in deutsch-italienischen Kontaktsituationen.” In Fremdes wahrnehmen, aufnehmen, annehmen – Studien zur deutschen Sprache und Kultur in Kontaktsituationen, edited by B. Hans-Bianchi et al., 247-260. Frankfurt a.M.: Peter Lang Verlag. Razi, N. 2013. “A Contrastive Studies of Compliment Responses among Australian English and Iranian Persian Speakers.” Procedia. Social and Behavioral Sciences 70: 61-66. Rees-Miller, J. 2011. “Compliments revisited: Contemporary compliments and gender.” Journal of Pragmatics 43: 2673-2688. Ruhi, ù. 2007. “Higher-order Intentions and Self-Politeness in Evaluations of (Im)politeness: The Relevance of Compliment Responses.” Australian Journal of Linguistics 27: 107-145.

410

Chapter Sixteen

Saito, H., and M. Beecken. 1997. “An Approach to instruction of pragmatic aspects: implications of pragmatic transfer by American learners of Japanese.” The Modern Language Journal 81 (3): 363-377. Sidraschi, D. 2014. “Pragmatics and Ethnolinguistics of Compliments: Compliment Responses in Novara and Grottaglie.” Lingue e Linguaggi 11: 225-239. Slavianova, L. 2011. “Contrastive Analysis of Compliments in Contemporary Bulgarian and English Language”, http://conf.uniruse.bg/bg/docs/cp11/6.3/6.3-23.pdf. Tran, G. Q. 2007. “Compliment Response Continuum Hypothesis.” The International Journal of Language Society and Culture 21. http://www.aaref.com.au/en/publications/journal/journal-articles/issue21-2007/. Wierzbicka, A. 1987. English speech act verbs: A semantic dictionary. Sydney: Academic Press. Wolfson, N. 1983. “An empirical based analysis of compliments in American English.” In Sociolinguistics and language acquisition, edited by N. Wolfson, and E. Judd, 82-95. Rowley, MA: Newbury House. Yuan, Y. 2002. “Compliments and compliment responses in Kunming Chinese.” Pragmatics 12 (2): 183-226.

Appendix (A) Examples of stimuli (German), with English translation Physical appearance: Deine neue Frisur finde ich toll! / Du hast schöne Augen // I find your hairstyle wonderful! / You have nice eyes Personal belongings: Deine Schuhe, wie schön! / Du hast aber eine schöne Tasche! // Your shoes, how nice! / You really have a nice bag! Character: Du bist sehr charismatisch, ich mag deinen Charakter / Wie kooperativ bist du! // You are very charismatic, I like your character / You are so cooperative! Abilities: Du kannst sehr gut Italienisch sprechen / Du bist ein toller Koch! // You can speak Italian very well / You are a fantastic cook! (B) Examples of stimuli (Italian), with English translation (B1) SEMI-SPONTANEOUS COMPLIMENTS: Che bella borsa /begli occhi/bel taglio di capelli che hai // What a beautiful bag/eyes/haircut you have! (B2) DCT

The Variability of Compliment Responses: Italian and German Data

411

Completa scrivendo come ti comporteresti o la risposta che credi appropriata, e, se lo ritieni, anche più risposte. Cerca di essere il più naturale possibile. Grazie per la collaborazione! Directions: please respond to the following situations. For each situation, you might find more than one response socially appropriate. In that case, please write all of them in the lines provided. Thanks for you cooperation! 1. Incontri per strada un amico/a che ti dice: “Lo sai che hai davvero dei begli occhi? Non li avevo mai notati!”. Tu rispondi: // Walking in the street, you meet a friend who says: “You have really beautiful eyes. I never noticed them before!”. You say: A:________________________________________________________________ B:________________________________________________________________ C:________________________________________________________________ D:________________________________________________________________ 2. Hai superato con successo un esame. Un compagno/a di Università ti dice: “30? Che bravo!”. Tu rispondi: // You successfully passed a University exam. A mate says: “You got A? You are so clever!”. You say: A:________________________________________________________________ B:________________________________________________________________ C:________________________________________________________________ D:________________________________________________________________

3. Hai appuntamento con un amico. Quando vi incontrate, indossi un paio di scarpe nuove. Ti dice: “Che belle scarpe hai!”. Tu rispondi: // You arranged to meet a friend. When you meet, you are wearing new shoes. He says: “You are wearing such beautiful shoes!”. You say: A:________________________________________________________________ B:________________________________________________________________ C:________________________________________________________________ D:________________________________________________________________ (B3) ROLE PLAYS: 1. A incontra il suo amico/la sua amica B per strada e nota per la prima volta che ha davvero dei begli occhi // A meets her/his friend B, noticing for the first time that she has really beautiful eyes. 2. A incontra un/a compagno/a di Università (B), che ha appena fatto un esame (B ha preso 30 e lode, ma A non lo sa) // A meets the University mate B, who has just passed an exam (B got A+, but A does not know the result) 3. A e B hanno appuntamento: quando B arriva, A nota le sue scarpe nuove // A and B arranged to meet: when B comes, A notices her/his new shoes.

412

Chapter Sixteen

Notes 1

Both Marina Castagneto and Miriam Ravetto collected the data for the analysis; both transcribed, classified, compared and analysed them. In the present paper Ravetto wrote sections 1, 1.1, 1.3, 2, 2.1, 3.1 and 4, while Castagneto sections 1.2, 2.2, 3.2 and 3.3. We want to thank the anonymous reviewers for their suggestions and the students who helped us to collect the corpora, above all Lina Di Francesco and Alessia Aina. 2 In order to designate the sociolinguistic dimensions of variation inside a language, we adopt the terms diatopic, diastratic, diaphasic and diamesic variations. Diatopic variations are related to geographic varieties, diastratic variations are due to sociolinguistic factors (such as social classes), diaphasic variations have to do with the context in which the linguistic act is performed, and diamesic variations show the role of the oral or written channel on linguistic production. This meta-linguistic terminology has been extensively used in the Italian tradition of General Linguistics and of Italian Linguistics since the publication of Berruto’s Sociolinguistica dell’Italiano contemporaneo (1987). Yet, these terms were introduced by Coseriu (1973), who had drawn on the Norwegian linguist Flydal. For an overall account of the widespread use of these metalinguistic terms, see Albrecht (1986). 3 See, among others, Knapp, Hopper and Bell (1984); Holmes and Brown (1987); Wolfson (1983); Herbert (1990); Farghal and Al Khatib (2001); Parisi and Wogan (2006); Rees-Miller (2011); Furkó and Dudás (2012). 4 But see Lee (2009), who demonstrated that compliments in the Chinese new Year period are also gender-sensitive and generation-sensitive. By linguistic register we mean a specific linguistic variety whose peculiar structure and selected items depend on the reciprocal status of the speakers, the goal and the context of the linguistic interaction. 5 Among the most famous studies are: Wolfson (1983); Knapp, Hopper and Bell (1984); Holmes and Brown (1987); Holmes (1988); Chen (1993). 6 A partial exception to our semantic-pragmatic tagging concerns the category Thanking. As there is no other way of thanking than expressing the wish to thank through the word thanks itself, we could say that this tag was assigned to CRs in our corpus on the basis of lexico-grammatical wording, even in consideration of the high frequency of this response. 7 This taxonomy was applied both to the Italian and to the German sub-corpus. Marina Castagneto transcribed or supervised the annotation of the Italian data, while the German sub-corpus was in charge of Miriam Ravetto. Ambiguous cases were discussed between the authors until an agreement was reached on how to label the problematic turns. 8 The English examples in the table are made-up. 9 In this case we separated agreement tokens from acceptance tokens to be sure that we were dealing with a case of acceptance and not a response aimed just to reinforce the comity with the complimenter, as part of politeness management. And we must not forget, however, that the need to distinguish agreement tokens from appreciation tokens was already present in the pionieristic work by

The Variability of Compliment Responses: Italian and German Data

413

Pomerantz (1978), and that it is also evidenced by the choice of different tags for these two functions in more than one taxonomy of CRs. Tran (2007), for instance, considered agreement tokens a higher step in his acceptance-denial continuum compared to appreciation tokens. 10 The name of this tag in Tran (2007) is Compliment Upgrade, but, leaving aside labelling issues, this kind of CR fully corresponds to our Pleased Acceptance. 11 Here the notion of ‘key’ refers to Hymes’s (1974) “SPEAKING” model. 12 Yet, we are aware that the real communicative intention of the speaker is not always easy to recognize, so we understand why some scholars decided to keep together the two kinds of CRs, as in the case of Herbert (1989), who included also CRs with a joking connotation in the Praise Upgrade tag. Other scholars created a tag Joke or Joking for all kinds of joking CRs (e.g., Chen 1993, 53) to tag compliments such as the following: A: Wow, what a watch! I wish I had one like that. B: Get yourself a girlfriend that will buy one for you, but some CRs offered in the literature look really like our Ironic Acceptance responses. Lee (2009, 528) for instance, referred the case of a compliment on a smart dress responded to by the complimentee like this: A: dangran, wo de shencai hen mei (‘of course, I have a beautiful figure’). 13 The tag Explanation was already in Tran (2007), as a synonym of Comment History, and in Chen’s (1993) Explaining, to mark a slight different pragmatic function from the one identified in our taxonomy. 14 As in Herbert’s Comment History tag (Herbert 1989, 13), the function of the tag Explanation was to categorize the CRs by which the speaker “although agreeing with the complimentary force of the previous utterance, does not accept the praise personally; rather, speakers impersonalizes the complimentary force by giving (frequently irrelevant) impersonal details”, as in his instance: F1: I love that outfit; F2: I got it for the trip to Arizona. 15 The German and the Italian sub-corpora in their grand total were not fully balanced in the sample, given that the German sub-corpus consisted of almost half of the compliments in the Italian sub-corpus and given that the Italian sub-corpus consisted of more sub-corpora elicited with the explicit goal of investigating the weight of a particular factor. So, to achieve the goal of this work, we decided to present only a section of both the Italian and the German sub-corpora, consisting of almost 300 CRs each. These CRs were very similar regarding their contextual factors and the method of data collecting (see the description of the corpus in 2.1). By selecting just a representative (but large) sample of the German and the Italian sub-corpora, it was easier to recognize the general trends in responding to compliments in the two respective countries, improving the degree of control over the research design and the generalizability of the findings.

CHAPTER SEVENTEEN UNIVERSALITY AND RELATIVITY IN CROSS-CULTURAL PRAGMATICS: REQUESTS AND APOLOGIES IN ENGLISH AND ITALIAN LOREDANA POZZUOLI

1. Introduction Verbal communication is a social and cultural phenomenon (SpencerOatey 2000). A wide range of aspects of verbal communication such as intonation, cohesion, pausing and indirectness can vary from culture to culture and each of these aspects contributes in different ways to the communication of meaning through speech (Tannen 1984). Moreover, speakers of different languages may give different interpretations to the contextual factors related to a specific communicative situation based on the socio-cultural conventions they are familiar with. For instance, they may have different expectations concerning the rights and obligations or the degree of power or distance separating two interlocutors in verbal interactions. They may also have different conventions for selecting conversational strategies and interpreting their use in a specific communicative context. Therefore, communication is influenced by one’s culture and, consequently, to speak one language rather than another means to convey different cultural and social values. In light of the above considerations, a cross-cultural pragmatic research project on the use of request and apology strategies was carried out in order to investigate (1) how Italian and English speakers perform these two speech acts in their native languages and (2) how Italian learners of English deal with requests and apologies in their second language (L2). Thus, in keeping with the main goals of the study, this paper is intended to

Chapter Seventeen

416

contribute to the field of cross-cultural pragmatics and derive implications for inter-language pragmatic research as well.

1.1. Cultural variation in speech acts performance and contrastive studies of speech acts 1.1.1. Overview A cross-linguistic comparison of speech act realizations has proven to be an essential instrument for shedding light on the general principles which govern the production of speech acts and on the degree to which rules of usage can vary from language to language. As suggested in several cross-cultural pragmatic studies (Clyne 2006; Lucy 1992; Matsumoto 1988), there is a universal as well as a language-specific component in the realization of speech acts. Universal pragmatic knowledge is shared across languages and it explains, for example, that the same basic strategies are used in the realization of requests, apologies, compliments and many other speech acts (Blum-Kulka and Olsthain 1984). However, in addition to this basic shared knowledge, there are also different types of interactional behaviors and remarkable cross-linguistic as well as cross-cultural differences in the selection, distribution and realization of speech acts from one language to another. For instance, socio-cultural variables such as authority, social distance, and situational setting are supposed to influence the appropriateness and effectiveness of the communicative strategies that are used to realize particular speech acts (Wierzbicka 1985). Contrastive analysis of speech acts can also contribute to the development of inter-language pragmatic research. The comparative investigation of the realization of speech acts may lead to the identification of remarkable differences in the employment of the strategies used by second language learners and those used by native speakers in the production of the same speech actions. In order to correctly interpret what is being said in a certain L2, learners need to have a good understanding of the cultural and pragmatic norms of the specific language and context. On the contrary, inappropriate transfer or “negative transfer” of sociolinguistic norms from L1 into L2 within situations of intercultural communication may lead to miscommunication and pragmatic failure (Thomas 1983). More specifically, “negative pragmatic transfer” is generally referred to as a wrong generalization of pragmatic knowledge from L1 to a communicative situation in L2 which is likely to entail adverse effects on communicative success (Kasper 1992). In this context, learners of a certain

Universality and Relativity in Cross-Cultural Pragmatics

417

L2 should be aware of the fundamental distinction between the propositional content of a linguistic expression and its illocutionary force potential (Searle 1969) in order to understand not only which expression can be used for the performance of a particular illocutionary act, but also the specific context in which such an act can be appropriately carried out. Therefore, a good command of the socio-pragmatic formulas used in a target language is indispensable to the acquisition of communicative competence in a L2. These formulas, indeed, can improve the learner’s productive and receptive performance and develop an understanding of the target culture (House 1996, 226). In this regard, communicative competence can be defined as the second language learners’ ability to properly use language in social interactions by interpreting and conveying suitable illocutionary forces according to the socio-cultural context in which utterances are used (Kasper 1983). From the above considerations, it follows that contrastive studies of speech acts profitably contribute to the development of second language acquisition research by shedding light on how L2 learners understand and perform actions in a target language. Requests and apologies are the most frequently examined speech acts and previous research has developed specific taxonomies of request and apology realization strategies that have become the most preferred models in later studies. In this regard, following the request coding scheme introduced by Blum-Kulka, House, and Kasper (1989), Trosborg (1995) and Schauer (2009), requestive strategies differ with regard to level of directness with which their illocution is conveyed and can be categorized into three different groups: direct requests, conventionally indirect requests and non-conventionally indirect requests or hints. As shown in Table 2 (section 2.3.1), direct strategies are “procedures that realize the act by reference to the contextual preconditions needed for its performance, as conventionalized in a given language” (Blum-Kulka and Olshtain 1984, 201). The illocutionary force conveyed is directly derivable from linguistic indicators or from the semantic content of the utterances. In conventionally indirect strategies, the interpretation of the utterance as a request relies on conventional usage (Blum-Kulka, House, and Kasper 1989). Finally, in non-conventionally indirect strategies, the interpretation of the utterance as a request is determined by the specific communicative context, since the locution is vague or ambiguous (Schauer 2009). Independently of the strategies encoding them, requests can also be accompanied by external modifications that qualify and support the illocution. More specifically, external modifiers or supportive moves (Blum-Kulka, House, and Kasper 1989)1 are additional statements made

418

Chapter Seventeen

by the speakers to support the head act, that is, the core of the request utterance containing the requestive verb which determines the level of directness of the request. According to the apology coding scheme developed by Blum-Kulka, House, and G. Kasper (1989) and Trosborg (1995), people usually apologize by means of semantically different types of expressions. More specifically, as described in Table 4 (section 2.3.2), a person apologizing may choose to express his/her apology directly or indirectly. The most direct realizations of an apology is by means of an explicit illocutionary force indicating device (IFID), which selects a direct routinized expression of regret. The IFIDs are performed through a performative verb expressing apology, such as be sorry, apologize, regret, etc. Another way to perform an apology is by resorting to indirect ways of apologizing through the use of indirect strategies. IFIDs and the indirect strategies are not mutually exclusive, which means that they may be used either alone or in combination with one another (Blum-Kulka, House, and Kasper 1989). Finally, as Trosborg (1995) and Blum-Kulka, House, and Kasper (1989) show, in addition to selecting apology strategies, the apologizer may also modify them by using external modifications which are meant to mitigate the magnitude of the offence committed. The classification system used for the present investigation was based on the above-mentioned studies and will be discussed in detail in subsections 2.3.1 and 2.3.2. The following two sub-sections, instead, contain a review of the most cited and influential studies on requests and apologies carried out within the research areas of inter-language pragmatics and cross-cultural pragmatics in the last decades. 1.1.2. Contrastive studies of requests In this part of the study, a brief review of prominent contrastive studies on requests is introduced. This section begins with the introduction of the Cross-Cultural Speech Act Realization Project led by Blum-Kulka, House and Kasper in 1989 whose categorization model has been frequently employed in cross-cultural pragmatic and Inter-language pragmatic research. This will be followed by a discussion on the results yielded by House and Kasper (1987) and those presented in Trosborg (1995) and in Yu (1999). Finally, a review of what Schauer (2009) reported on her developmental study on requests will be provided. One of the most important accomplishments in the analysis of requests and apologies from a cross-cultural perspective is the Cross-Cultural Speech Act Realization Project (CCSARP) carried out by an international

Universality and Relativity in Cross-Cultural Pragmatics

419

group of researchers (Blum- Kulka, House, and Kasper 1989). Results yielded in this project showed, for example, that both German and Danish learners of British English deviated from the British norm and followed their L1 norms in the performance of requests by choosing direct strategies to a greater extent than native speakers. House and Kasper (1987) focused on the comparison of requests performed by German and Danish learners of English. They pointed out that non-native speakers used longer requests and a higher level of directness than native speakers. They also noted that the external modification “grounder” was the one most frequently employed by all the participants, suggesting that the reason for this language behavior may be the fact that “it is psychologically most plausible to make the addressee understand the reason(s) behind a request” (House and Kasper 1987, 1281). The learner data showed a marked overuse of external modifications in comparison to the native speaker data, which in the case of the German participants resulted in lengthier request utterances. This finding was in line with the results of the study carried out by Blum-Kulka and Olshtain (1986), who found that the learner group over-employed external modifiers. Trosborg (1995) studied English requests made by Danish learners at three proficiency levels. In contrast to most studies, the results of this study showed that direct requests increased with proficiency level, whereas the use of hints, that is the most indirect way to perform requests, was the most frequent in less advanced learners. Trosborg argued that direct requests were uncommon among low-proficiency learners, indicating that perhaps they thought that they would be impolite if they had used them. Using a DCT, Yu (1999) investigated the ways in which Chinese learners of American English performed requests in their L2. He found that the learners’ performance did not approximate the English native speakers’ behavior, especially regarding the employment of mitigating devices (past tense forms or modal verbs), which were used to a greater extent by the group of native speakers. Schauer (2009) examined the inter-language pragmatic development of German learners of English at a British University over a period of one academic year. To obtain insights into the effect of the L2 context in the performance of the speech act of request, the study-abroad learners’ data was compared with those of British native speakers and German learners of English in their home country. Data were collected with the Multimedia Elicitation Task and the results revealed that the conventionally indirect strategy “ability” was the most frequent strategy in all of the three groups

420

Chapter Seventeen

of participants. This finding was in line with previous request studies (House and Kasper 1987; Woodfield 2008), according to which this strategy is a standardized way of formulating a request. Overall, Schauer’s study showed that the learners who studied abroad and the native speakers used a much broader range of request strategies than the learners who studied in their own country. The results also suggested that members of the study abroad group and the native speaker group tended to have a much broader repertoire of external modifiers than the other group of learners (Schauer 2009, 187). The results of the studies reviewed in this section suggest that the overuse of external modifications when formulating requests was a typical feature in English learners who, as a consequence, tended to be more verbose than English native speakers. The ability strategy and conventionally indirect requests in general appeared to be frequently used and often favored by both learners of English and English native speakers. However, learners of English often tended to prefer direct over indirect strategies, performing direct strategies to a greater extent than native speakers. On the contrary, Trosborg’s study reported contrasting results, according to which direct requests increased with higher proficiency learners. 1.1.3. Contrastive studies of apologies This section focuses on a review of remarkable investigations on apologies carried out in cross-cultural pragmatics and inter-language pragmatics. Firstly, the results derived from Trosborg (1995) will be discussed in comparison with those yielded in the CCSARP (1989). Secondly, a contrastive study on apologies performed in Italian and American English will be presented (Lipson 1994). Trosborg (1995) examined the apology speech act from an interlanguage pragmatic point of view. She found that the expression ‘‘I’m sorry” or “sorry” was the most common realization strategy, as confirmed also by Olshtain and Cohen (1983) and Rintell and Mitchell (1989). The “request for forgiveness” strategy, on the other hand, was only associated with situations with higher severity of offence. This strategy was less frequently used by both the native speaker group and the non-native speaker group than the other IFID strategies. The results yielded in the CCSARP project demonstrated no significant difference in the strategy selection, and they further revealed strong similarities in the employment of IFIDs, in that they were employed to a similar extent by native speakers and foreign language learners alike.

Universality and Relativity in Cross-Cultural Pragmatics

421

Lipson (1994) compared apology strategies in American English and in Italian by using an American television sitcom as an instrument for data collection. This sitcom was shown in the original language to ten Italian students who had to rewrite the apology episodes for an appropriate Italian context. The differences between the original script and the students’ versions were then analyzed and compared. Lipson found that, in stark contrast to the strategies preferred in English, which were “expressions of regret” and “minimizing of the offence.” her Italian data contained a predominant use of “requests for forgiveness.” Lipson suggests that in an Italian context the expression of regret is favored in situations when the offence is perceived as very severe by the apologizer and when a greater degree of reprimand is expected. The above-mentioned contrastive studies of the performance of apologies reveal cross-cultural similarities in the employment of the IFID strategies, which were extensively used by both learners of English and native speakers. Conversely, differences between English and Italian speakers were detected in the use of the strategies “expressions of regret" and “minimizing of the offence,” which were mostly preferred by English native speakers. 1.1.4. Conclusion The studies discussed in the previous sub-sections have significantly contributed to shedding light on L2 learners’ productive pragmatic competence when the performance of apologies and requests is involved. They have given rise to an effective analytic approach for the crosslinguistic investigation of requests and apologies and have shown that speech acts have different realizations across languages and cultures. However, cross-cultural analysis of speech acts involving Italian is an area that needs to be further investigated. Accordingly, this study aims to shed some light on the inter-language variation in the realization patterns of requests and apologies among Italian learners of English.

2. Method 2.1. Participants and data collection procedure Two groups of subjects participated in this study. The first group consisted of 10 English native speakers. They were all British students studying at Lancaster University and enrolled in different postgraduate academic programs. The group consisted of 7 females and 3 males and

422

Chapter Seventeen

their average age was 24. The second group of participants consisted of 10 Italian learners of English. All but two were females, ranging in age from 26 to 27. They were university students at the University of Perugia in Italy, all enrolled in a master program “Modern Languages and Literatures.” They all had been studying English for at least eight years, and nine of them had spent at least three months in an English-speaking country.

2.2. Instrument and procedure The data for the present research were collected through a modified version of the Discourse Completion Test (DCT) used in the CCSARP (Blum-Kulka, House, and Kasper 1989). In this study, participants were asked to write what their response would be in particular scenarios by designing situations which a university student would normally encounter in everyday contexts. The DCT designed for the present study consisted of sixteen different situations: eight of the scenarios called for the performance of requests and the other eight scenarios called for the realization of apologies. An Italian version of the DCT was created by translating and adapting the same situations as the English version. The Italian participants were asked to fill in the DCT in English as well as in Italian. For this reason, three different types of performances were collected in this study: performances obtained by English native speakers, performances obtained in Italian and in English by the same group of Italian native speakers. In Figure 1, examples of actual scenarios included in the English and the Italian version of the DCT are provided. Participants were encouraged to complete the DCT as spontaneously as possible without worrying too much about grammatical and syntactic mistakes which they might make in the English DCT. The reason for this was that the actual aim of the present study was not to assess linguistic competence, but rather to analyze language use cross-culturally.

Universality and Relativity in Cross-Cultural Pragmatics

423

FIGURE 1. Examples of actual scenarios included in the English and the Italian version of the DCT SITUATION 10 (Request) A student in the library is making too much noise and disturbing you and other students. You decide to ask the student to be quiet. What would you say? SITUATION 6 (Apology) You have borrowed your friend’s notes and because of the rain the day before, some of the notes have got wet and damaged. In the next lecture your friend sits down next to you. What would you say? SITUAZIONE 10 (Richiesta) Uno studente sta disturbando la quiete della biblioteca dove ti trovi a studiare con altri studenti. Decidi di chiedere allo studente di smettere di fare rumore. Che cosa diresti allo studente? SITUAZIONE 6 (Scusa) Hai preso in prestito degli appunti da una tua amica. A causa della pioggia del giorno prima alcune pagine del suo quaderno si sono bagnate e rovinate. Cosa diresti alla tua amica seduta accanto a te a lezione l’indomani mattina?

For each situation included in the DCT, the contextual setting in which the interactional encounter takes place was specified, including also dominance (Dom. in Table 1), which is the power relationship between the participants as determined by their social status, and social distance (Dist. in Table 1), which is their level of intimacy (see Trosborg 1995). Dominance was marked with the “+” value if the requestee had authority over the other, or with the “-” value if the participants were of equal status. Social distance instead was marked with “+” value if the participants had never met, or with “-” value if they knew each other well. The sequence of the sixteen scenarios was determined randomly and was the same in all the DCTs given to the participants.

Chapter Seventeen

424

TABLE 1. Categorization of the 16 scenarios according to the dominance and social distance variables (ST = student; P = professor; SC = secretary) Situation 1

Speech act Request

2

Request

3 4 5 6

Request Request Request Request

7

Request

8 9

Request Apology

10 11

Apology Apology

12

Apology

13

Apology

14

Apology

15 16

Apology Apology

Contextual setting

Dom.

Dist.

Role

“Ask to close the window” “Ask for an extension” “Ask for a meeting” “Ask to be quiet” “Ask for a book” “Ask for help with the photocopier” “Ask to clean up the kitchen” “Ask for a lift” “Being late with your supervisor” “Spilling tea” “Ruining your friend’s notes” “Stepping on a student’s foot” “Being late with your friend” “Forgetting to return a form” “Missing the class” “Forgetting your friend’s book”

-

+

ST-ST

+

-

ST-P

+ + +

+ +

ST-P ST-ST ST-P ST-SC

-

-

ST-ST

+

-

ST-ST ST-P

+ -

-

ST-SC ST-ST

-

+

ST-ST

-

-

ST-ST

+

-

ST-SC

+ -

-

ST-P ST-ST

2.3. Method of analysis The categories used for the analysis of the data in this paper were based on the request and the apology classification schemes used by Blum-Kulka, House, and Kasper (1989) in the CCSARP, Trosborg (1995) and Schauer (2009), which have been combined and modified to better clarify the data in the present study. In particular, with regard to requests, the different types of performatives and hints (hedged and unhedged performative as well as mild and strong hints) have been combined into two categories called “performatives” and “hints. With regard to

Universality and Relativity in Cross-Cultural Pragmatics

425

apologies, Blum-Kulka, House, and Kasper (1989) classified “concern for the hearer” as intensification. Following Trosborg (1995), in the present study this concept has been treated as a separate strategy. 2.3.1. The request coding scheme As explained in Section 1.1.1, a request can be performed by means of different strategies and linguistic devices that modify the illocutionary force conveyed by the interlocutor in the speech act. Table 2 exemplifies the types of request strategies employed by the informants in doing their DCTs.2 TABLE 2. Overview of request strategies (1) (2)

Strategy Imperative Performative verb

(3)

Obligation statement

(4)

Want statement

(5) (6)

Locution derivable statement or question Availability question

(7) (8)

Willingness question Ability question

(9) (10)

Permission question Prediction question

(11)

Hint

Example “Sara, clean up the kitchen!” “I would like to ask you to give me your pen” “Oh, Mark, you really have to clean up that mess” “I would like you to lend me this book” “Do you know how to use this photocopier?” “Have you got a minute to help me with this?” “Will you clean the kitchen?” “Could you please give me an extension?” “Can I borrow this book?” “Is there any possibility to arrange a meeting?” “You have left the kitchen in a total mess”

In Table 3, the list of the request external modifications used in this study is presented. The modifications are signaled by italics in order to distinguish them from the main head act.

Chapter Seventeen

426

TABLE 3. Overview of request external modifications Request Modification Alerter

Preparator

Disarmer

Grounder

Cost-minimizing

Promise of reward

Moralizing

Appreciator

Considerator

Example It is employed at the beginning of a request to get the hearer’s attention. e.g., “Excuse-me, could you tell me where the museum is?”; “Mark, can you answer the phone?” It can precede or replace the alerter and prepares the hearer for the request. e.g., “Are you busy next Monday? I would like to ask you to come with me to the hospital” It is used to try to avoid any potential objections the hearer might make about the request. e.g., “I know I am disturbing you but would you mind helping me?” It provides explanations, reasons as well as justifications for the request. e.g., “I have missed the bus. Could you please give me a lift?” It is used to minimize any possible cost to the hearer in performing the requested action. e.g., “Can you lend me your book? I’ll return it as soon as possible” It presents an offer of compensation/reciprocation and is used to increase the possibility of compliance. e.g., “If you come to the wedding with me, I’ll get you the ticket for your favorite band’s concert” The speaker may invoke general moral maxims in order to gain more credibility. e.g., “ Can I borrow your car? We are good friends and we should help each other as much as possible” It is used to show appreciation towards the hearer for the envisaged performance of the requested action. e.g., “Can you close the window? Thank you”; “Could you lend me your notes? I’d really appreciate that” It is used to show consideration towards the interlocutor for the cost in carrying out the requested action3) e.g., “Only if it does not cause you any trouble, of course”

2.3.2 The apology coding scheme As explained in the Introduction, interlocutors can apologize in a variety of ways. In Tables 4 and 5, it is possible to observe the apology strategies and modifications used by the participants in this study.

Universality and Relativity in Cross-Cultural Pragmatics

427

TABLE 4. Overview of apology strategies 1) 2) 3) 4) 5) 6) 7) 8)

Strategy Expression of regret Offer of apology Request for forgiveness Acknowledgment of responsibility Explanation Offer of repair Promise of forbearance Concern for the hearer

Example “I’m sorry”; “I regret that…” “I apologize”; “Accept my apology” “Forgive me”; “Excuse-me” “It’s my fault” “The bus was late” “You can borrow my notes instead” “It won’t happen again” “Are you ok?”

TABLE 5. Overview of apology external modifications Apology modification Alerter

Distracting from the offence

Preparator

Appreciator

Checking on reaction

Example It is employed at the beginning of an apology as an attention getter. e.g., “Jane, I’ve ruined your notes. I’m sorry” It is used to pacify the hearer by diverting the hearer’s attention from her/his responsibility for the offence. e.g., “Sorry, this driver is indeed reckless” (Reference to the external causes of the damage); e.g., “Sorry I’m late but If you hadn’t been waiting for me so long, you would not have met that nice guy! You owe me a favor!” (Reference to the positive side-effects of the damage through the use of humor or irony) It is used to prepare the hearer for the content of the apology that the speaker is about to make. e.g., “Sarah, I’ve something to tell you. I’ve ruined your notes, sorry” Presenting the damage caused to the addressee as though it were a benefit received by them. It can be used by the speaker to mitigate the impact of the offence on the hearer. e.g., “Sorry, I was stuck in traffic, Thank you so much for waiting” It is used to find out whether the infraction committed by the speaker has somehow damaged the relationship between the interlocutors. e.g., “Sorry, I forgot your book. Are you mad at me?”

Chapter Seventeen

428

3. Results and discussion 3.1. Italians requesting in English The overall distribution of the request strategies for each group of participants according to the three main levels of the indirectness scale is summarized in the table below. The following abbreviations are used to refer to the three groups of informants involved in the project: ES (English speaker), EL (English Learner), IS (Italian Speaker). TABLE 6. Use of request strategies in the three groups of participants Participants

Direct

Conventionally Indirect

%

7.5

%

87.5

NonConventionally Indirect % 5

Raw %

6 15

Raw %

70 83

Raw %

2

Raw %

12 24

Raw %

67 72

Raw %

1 4

Raw

19

Raw

58

Raw

3

English speakers 4

English learners

Italian speakers

The most frequent requestive strategies used by all of the participants were conventionally indirect strategies. Non-conventionally indirect requests, that is Hints, were used by a higher number of English speakers (ESs) (5%) than English learners (ELs) (2%). It was also observed that the Italian speakers (ISs) employed almost the same number of Hints (4%) as the ESs. The ELs employed a higher number of the most direct category of request strategies (15%) than the ESs (7.5%). A reason for this quite remarkable difference might be that, as shown in previous studies (Schauer 2009), given the more explicit and clearer expression of their illocutionary force, utterances of this type are more likely to be correctly inferred by the hearer. From Table 6, it is possible to observe that the ISs tended to perform requests in a more direct way than the ESs, who, instead, focused more on the respect of the other’s freedom of action by avoiding particularly impositional strategies.

Universality and Relativity in Cross-Cultural Pragmatics

429

GRAPH 1. Comparison of the three subject groups’ request strategies in percentage values

Graph 1 shows the comparison of the individual groups’ preferences with regard to individual request strategy types. The Ability strategy was one of the most frequent strategies in all three groups. However, the ESs employed it far less frequently that the ELs preferring, instead, other conventionally indirect strategies, such as Permission, Willingness, and Prediction, which is the most frequent strategy employed by the ESs in this study. The ELs also employed a very wide range of the Ability strategy in the Italian version of the questionnaires, which may be the main reason why they also used it in their English requests. The use of direct strategies was less frequent in the ESs’ responses than in the ELs’ responses, who employed a higher number of Imperatives, Performatives, Locutions derivable, and Want statements. The only direct strategy which was used more than once by the ESs was Locution derivable. The ESs and the ISs used almost the same number of Hints. On the other hand, this kind of request was employed only once by an EL. Nonconventionally indirect requests are the least direct request strategy type and demand a high degree of inferential ability on the part of the hearer as well as good linguistic skills on the part of the speaker. Accordingly, we would like to suggest that learners tried to avoid this strategy type because it was more likely to be misunderstood. The ELs tended to employ a greater number of direct strategies than the ESs both in their L1 and L2

430

Chapter Seventeen

performances, especially in situations characterized by: “-social distance” and“-dominance.” On the contrary, although the ESs tended to be much less direct than the ELs, they preferred to employ direct strategies, such as Locution Derivable and Performatives, in situations in which they had to deal with unknown, higher-status interlocutors (+ social distance, + dominance), as in scenario 6 (see Table 1). The graph below shows the frequency of use of external modifications in the three groups of participants and reveals a remarkable difference between the ESs and the ELs.

GRAPH 2. Comparison of the three subject groups’ request external modifications in percentage values

As can be seen in Graph 2, the ELs provided more supportive moves than the ESs. This is unlike the results in Trosborg’s (1995) study, who observed that her learners-subjects tended to employ a lower number of external modifications than her native speakers. However, the results of the present research are in line with findings of other previous studies (Blum-Kulka and Olshtain 1986; Schauer 2009). Edmondson and House (1991) coined the term waffle phenomenon to refer to an overuse of external modifications by non-native speakers. A plausible explanation for the very wide use of external modifications in the learners’ responses might be the influence of the learners’ L1, since the Italian native speakers employed the highest number of external modifiers. The data also showed that the Alerter and Grounder external modifications were the most widely employed supportive moves by the learner and

Universality and Relativity in Cross-Cultural Pragmatics

431

native speaker participants. This finding seems to indicate that these external modifiers constitute core elements of request utterances (Schauer 2009, 186). Moreover, the ELs employed a remarkably greater number of Preparators than ISs in their Italian data. This may suggest that learners tended to establish rapport before presenting the request. On the whole, the ELs as well as the ESs did not often use Disarmers, Cost-minimizers or Promises of reward as modifications. The Appreciator modifier (“thank you” in English, “grazie” in Italian), instead, was widely employed by all of the three groups of participants, almost always, at the end of the request.

4.2. Italians apologizing in English Graph 3 displays the employment of the apology strategies in the three groups of informants revealing slight differences between the ESs and the ELs. In Graph 3 it is possible to observe the common tendency to use the IFID across the three groups of subjects. More specifically, the ESs employed the IFID in 95% of the scenarios. The ELs showed a similar communicative behavior using the IFID overall in 89% of the situations. As shown by Sbisà (1999), expressions like “scusa” are the most conventional Italian IFIDs and, similarly, they were observed 76% of the time in this study as well. In all scenarios, except for scenario 12, all the informants accompanied the IFID with additional apology strategies. This is in line with Trosborg (1995, 390), whose results showed a lower number of apologies performed by means of a single IFID. In this context, from Graph 3, a general conclusion may be drawn: the ELs employed an overall greater number of apology strategies than the ESs. As shown, the ESs employed Explanations in 21% of the situations, while the ELs used it in 26% of the cases. This result is in contrast with Trosborg (1995), who observed that learners provided fewer explanations than native speakers. However, this finding is in line with the overall findings of this study, which revealed the learners’ as well as the ISs’ tendency to employ more strategies in the apology speech act than the ESs. Moreover, most of the time, learners provided lengthy utterances resorting to the use of several external modifications. Once again, the overuse of supportive moves might be related to the Italian language’s influence on the learners’ performance in their L2. In fact, the ELs generally used almost the same number of apology strategies in their Italian responses as well.

432

Chapter Seventeen

GRAPH 3. Comparison of the three subject groups’ apology strategies in percentage values

Another very common strategy that was widely employed by the subjects was Repair, which was used by the ESs in 13% of the time and by the ELs in 19 % of the time. The ESs as well as the ELs employed it almost always in the same situations, suggesting that they gave the same assessment to the contextual variables related to offering repair situations. Acknowledgment of responsibility was employed to the same extent by both the ESs and the ELs, that is 5% of the time. Just as in Trosborg (1995) and in Blum- Kulka, House, and Kasper (1989), the Promise of forbearance strategy was not frequent in any of the groups (2% of the time in each). All the groups involved employed it in the communicative situation in which participants had to apologize to a friend for ruining her/his notes. Therefore, significant cross-cultural agreement in the evaluation of the use of the strategy under analysis was observed. Concern for the hearer was employed by the ESs and the ELs to the same extent (both groups 5%). From these findings, it seems that Concern for the hearer was a strategy used both in Italian and in English to deal with unintentional physical offence, as shown in scenario 12 of this study.

Universality and Relativity in Cross-Cultural Pragmatics

433

GRAPH 4. Comparison of the three subject groups’ apology external modifications in percentage values

In Graph 4, it is possible to observe a remarkable difference in the frequency of the supportive moves yielded by the three groups of participants. The ELs tended to use supportive moves to a much larger extent than the ESs. Indeed, learners employed almost twice as often the moves employed by the ESs. The ELs also provided the same number of supportive moves in their Italian responses, and these matching results might be due to the Italian language’s influence on the learners’ L2 performance. The ELs did not differ from the ESs in their use of the Preparator move, which was the most frequent external modification in both groups of informants. Accordingly, the ELs as well as the ESs provided the preparatory move in 11% of their responses. The Distracting from the offence move was used by 6% of the ISs as well as the ELs and it was never realized by the ESs. Some of the learners employing this move also resorted to ironic expressions to divert the hearer’s attention from the offence occurred and to reduce the seriousness of the violation. The ESs did not use either the Distracting from the offence move or the Checking on reaction move. This might suggest that we are dealing with a specific Italian cultural convention which, overall, did not appear to address the concerns of the ESs.

434

Chapter Seventeen

4. Conclusion Socio-cultural factors possibly affecting the selection and the distribution of linguistic elements as well as the speech act realization in any particular language should be taken into consideration, since verbal behavior cannot be truly understood without reference to cultural values and attitudes (Wierzbicka 1985). To this end, the aim of the present research was to examine the requests and apologies uttered in English by Italian ELs and by ESs in order to identify similarities and differences in the realization patterns of these two speech acts under the same circumstances. Furthermore, this study aimed at investigating any possible L1 pragmatic knowledge affecting learners’ L2 performance when realizing requests and apologies. As for the request speech act, the results of this study are in line with the those yielded by the CCSARP project, which showed the crosslinguistic preference for conventionally indirect strategies. However, the analysis of requests has also shown that the ELs often resorted to more direct strategies than the ESs in their L1 and L2 speech act performance. The Italian tendency to use more explicit linguistic means to carry out the request speech act in specific situational settings might have influenced the ELs in the performance of the same speech act in their L2 as well. Non-conventionally indirect requests, that is Hints were used by a higher number of ESs as well ISs than ELs. This might suggest that native speakers of a given language, who are confident in their linguistic and pragmatic competence, feel comfortable and safe in using more ambiguous linguistic means, whereas learners of a certain target language, on the other hand, would mostly rely on strategies which can ensure that their communicative intent is properly understood by the hearer. The Italian preference for semantically rich requesting speech acts is another remarkable difference between the English and the Italian data. In fact, results also reveal that ISs employed a greater number of external modifications thus giving support to earlier findings that show that in some languages speakers may compensate for the level of directness in their requestive behavior by means of external modifications. The same pragma-linguistic behavior was observed in the ISs’ responses in which, the directness of the head act was softened by “verbose” supportive moves. As regards the apology speech act, an overall cross-cultural agreement between the ESs and the ELs in the employment the IFIDs can be observed. In fact, IFIDs were found in the majority of the responses in the data obtained. Most of the participants also tended to accompany the direct

Universality and Relativity in Cross-Cultural Pragmatics

435

apology with other indirect strategies included in the apology speech act set, such as Explanation and Repair which, however, were used to a greater extent by the ELs and the ISs than the ESs. Data obtained from this study show an orientation on the part of the majority of the respondents to protect their positive face by avoiding some apology strategies such as Concern for the hearer and Promise of forbearance, which are most damaging to the speaker’s face. Results also suggest that the Italian tendency to provide more supportive moves was highly manifest in the apology performances as well. More specifically, the most frequent moves in ELs’ performances were Preparator and Distracting from the offence. Albeit to a lesser degree, the former was also employed by the ESs. The latter, instead, was never found in the ESs’ responses. In conclusion, from the data obtained in this study we might venture to suggest that the learners’ inclination towards a larger employment of direct strategies for the performance of the request speech act was felt to be a safer linguistic option when their primary goal was to make themselves understood. However, culturally speaking, this also seems to indicate that the ISs, in situational contexts in which dominance and social distance were not involved, tended to perform requests in a more direct way than the ESs who, instead, appeared to focus more on the respect of others’ freedom of action by avoiding explicit impositive acts as much as possible. Furthermore, the ELs’ tendency to employ a larger number of external moves producing more elaborate requests as well as apologies can be regarded as a specific deviation from native English speakers’ general pragmatic behavior, due to the effect of certain pragmatic conventions in Italian according to which speech acts are generally performed in a more verbose way. The conclusions drawn in this study need to be validated with further research. In addition to a higher number of subjects with more varied backgrounds (linguistic, cultural, geographical, social, etc.), different variables, such as age and gender as well as the employment of collection data instruments that allow the analysis of naturally occurring data should be considered in further studies. This would be needed in order to reach more accurate conclusions regarding the comparison between English and Italian speakers’ realization of requests and apologies.

436

Chapter Seventeen

References Blum-Kulka, S., J. House, and G. Kasper. 1989. Cross-cultural Pragmatics: Requests and Apologies. Norwood: Ablex. Blum-Kulka, S., and E. Olshtain. 1984. “Requests and apologies: A crosscultural study of speech act realization patterns (CCSARP).” Applied Linguistics 5 (3), 196-213. —. 1986. “Too many words: Length of utterances and pragmatic failure.” Journal of pragmatics 8: 47-61. Clyne, M. 2006. “Some thoughts on pragmatics, sociolinguistic variation, and intercultural communication.” Intercultural Pragmatics 3 (1): 95105. Edmondson, W. J., and J. House. 1991. “Do learners talk too much? The waffle phenomenon in interlanguage pragmatics.” In Foreign/second language pedagogy research, edited by R. Philipson, E. Kellerman, L. Selinker, M. Sharwood-Smith, and M. Swain, 273-287. Clevedon: Multilingual Matters. House, J. 1996. “Developing pragmatic fluency in English as a foreign language. Routines and metapragmatic awareness.” Studies in Second Language Acquisition 18 (2): 225-252. House, J., and G. Kasper. 1987. “Interlanguage pragmatics: requesting in a foreign language.” In Perspectives on language performances, edited by W. Loerscher, and R. Schulze, 1250-1288. Tuebinger: Narr. Kasper, G. 1983. Strategies in interlanguage communication. Longman: London. —. 1992. Pragmatic transfer. Second Language Research 8(3): 203-231. Lipson, M. 1994. “Apologizing in Italian and English.” International Review of Applied Linguistics in Language Teaching 32: 19-39. Lucy, J. 1992. Language diversity and thought: a reformulation of the linguistic relativity hypothesis. Cambridge: Cambridge University Press. Matsumoto, Y. 1988. “Reexamination of the universality of face: Politeness phenomena in Japanese.” Journal of Pragmatics 12 (4): 403-426. Olshtain, E., and A. D. Cohen. 1983. “Apology: A speech act set.” In Sociolinguistics and Language Acquisition, edited by N. Wolfson, and E. Judd, 18-35. Rowley, MA: Newbury House. Rintell, E., and C. Mitchell. 1989. “Studies of requests and apologies: An inquiry into method.” In Cross-cultural pragmatics: Requests and apologies, edited by S. Blum-Kulka, J. House, and G. Kasper, 248272. Norwood, NJ: Ablex.

Universality and Relativity in Cross-Cultural Pragmatics

437

Sbisà, M. 1999. The room for negotiation in apologizing: evidence from the Italian speech act of scusarsi. Paper read at the International Conference Pragma99 Pragmatics and Negotiation, Tel Aviv and Jerusalem, 13-16 June 1999. Retrieved from: http://www2.units.it/dipfilo/sbisa/scuspap.html on 10/01/2011. Schauer, G. A. 2009. Interlanguage pragmatic development: The study abroad context. London: Continuum. Searle, J. R., 1969. Speech acts: an essay in the philosophy of language. Cambridge: Cambridge University Press. Spencer-Oatey, H. 2000. Culturally Speaking: Managing Rapport through Talk across Cultures. London: Continuum. Tannen, D. 1984. The pragmatics of cross-cultural communication. Applied linguistics 5: 189-195. Thomas, J. 1983. “Cross-cultural Pragmatic failure.” Applied linguistics 4: 91-112. Trosborg, A. 1995. Interlanguage Pragmatics: Requests, complaints and apologies. Berlin: Mouton de Gruyter. Wierzbicka, A. 1985. “Different cultures, different languages, different speech acts.” Journal of Pragmatics 9: 145-178. Woodfield, H. 2008. “Interlanguage requests: A contrastive study.” In Developing contrastive pragmatics: Interlanguage and cross-cultural perspectives, edited by M. Puetz, and J. Neff van Aertselaer, 231-264. Berlin: Mouton de Gruyter. Yu, M. 1999. “Universalistic and culture-specific perspectives on variation in the acquisition of pragmatic competence in a second language.” Journal of Pragmatics 9 (2): 281-321.

Notes 1

It should be noted that the terms of external modifications and supportive moves are used interchangeably in the literature and in this study. 2 In the current table, as well as in the following ones, the reported examples are drawn from the data collected in this study. 3 It is worth noting that the last two external modifications, namely, the Considerator and the Appreciator, were introduced by Schauer (2009) in her research. As they have been considered adequate to the particular expressions used by the participants involved in this research, they have been employed for analyzing the external modifications in this study as well.

SECTION 5 ASPECTS OF GRAMMAR

CHAPTER EIGHTEEN THE ACQUISITION OF EMOTIONAL COMPETENCE IN L2 LEARNERS OF ITALIAN THROUGH SPECIFIC INSTRUCTIONAL TRAINING ANNA DE MARCO AND EMANUELA PAONE1

1. Introduction The expression of emotions involves psychological and physiological processes which constitute a primary response of adaptation to external stimuli. Such processes are influenced by cultural factors (Levenson, Soto, and Pole 2007). Cultural aspects of emotion comprise belief and practices about what types of emotions can/should, or cannot/should not, be manifested, when, where and how. Emotions are culture-specific, that is, they “vary across cultures” (Mesquita and Walker 2003, 777). Culture affects the expression and the perception of emotions at different levels, as Levenson, Soto, and Pole (2007, 788) explained: Cultural influences may vary for aspects of emotion (e.g., subjective experience, language, expressive behavior, peripheral physiology), type of emotion (e.g., negative emotion, positive emotion, self-conscious emotion), context (e.g., presence of culturally salient cues), and the individual (e.g., extent of identification with cultural traditions).

The vocal expression of emotions involves prosodic cues, as “the voice has emerged as a basic system for the signalling of the emotional states of individuals, due to variations in pitch, intensity, rhythm, pauses, and vocal qualities” (Anolli, Wang, Mantovani, and De Toni 2008, 566). Research on the vocal expression of emotions has investigated two aspects of emotional speech: 1) The encoding process, which aims at exploring vocal cues involved in the expression of emotions, from a speaker-centered perspective, whether these are specially elicited – through simulation or mood induction

442

Chapter Eighteen

procedures (MIPs; see Gerrards-Hesse, Spies, and Hesse 1994) involving the use of films, images, hypnosis – or whether they are spontaneously produced (see Campbell 2000; Toshinori Ishi and Campbell 2012); 2) Relevant study methods include submitting vocal stimuli – which may be conveyed through sentences, words or non-linguistic vocalisations like laughter, yawns, sighs and so on – and collecting listeners’ responses to them through forced-choice procedures or ratings (i.e., Scherer, Banse, Wallbott, and Goldbeck 1991). Only a few studies have investigated both the encoding and the decoding processes of emotions due to the complexity of the phenomenon subject. In addition, the different methods employed have given rise to a wealth of heterogeneous data. However, emotional speech should be examined from both perspectives, since it is these combined perspectives that provide information2 about speakers (gender, personal features, etc.) and listeners’ ability to decode and infer vocal cues from speech. These two processes are even more important in an L2 learning environment. As the next section on cross-linguistic research illustrates, L2 learners (especially those from distant cultures) experience great difficulties in producing and perceiving emotional speech. In addition, learners are not always aware of the fact that many of the misunderstandings with native speakers are the result of a failure to identify and produce prosodic cues of speech. In light of the above, the analysis carried out in this study focused on both encoding and decoding processes in the emotional speech of L2 learners of Italian. The goal was to test the effectiveness of pragmatic training aiming at encouraging L2 learners’ awareness of the prosodic aspects of speech in order to improve their perception and production of emotional stimuli in Italian.

2. Cross-linguistic research on vocal emotions In the last years, studies on vocal emotions have been carried out from various disciplines, paving the way for new cultural perspectives on an understanding of emotions (Scherer and Bänziger 2004; Scherer 2009; Grandjean, Bazinger, and Scherer 2006; Hammerschmidt and Jürgens 2007; Johnstone and Scherer 2000, Pavlenko 2006). In particular, recent research conducted from the speaker’s perspective has examined emotional speech cross-linguistically, confirming the existence of universal acoustic patterns shared by different languages and cultures (Banse and Scherer 1996; Banse and Scherer 2009).3 Other studies (addressing emotions from the listener’s perspective) have demonstrated

The Acquisition of Emotional Competence in L2 Learners of Italian

443

the ability of humans to recognize emotional states expressed in a language other than their own, with above chance accuracy (Scherer, Banse, and Wallbott 2001; Galatà and Romito 2010; Pell, Monetta, Paulmann, and Kotz 2009; Bryant and Barrett 2008). Scherer, Banse and Walbott (2001) carried out a cross-linguistic research involving speakers from nine different countries. These were asked to recognize vocal emotions expressed in German. The sample included subjects from European, U.S. and Asian countries. The results showed an overall accuracy of 50%. However, the accuracy percentage decreased as the typological/cultural difference from German increased. In fact, the Indonesian group (speakers of non Indo-European language) scored the lowest value. The researchers concluded that: the most important goal of the present study was an empirical test of whether the recognition of vocal emotion portrayals is universal or culture dependent. The considerable main effect of the country factor […] provides clear evidence that there is a substantial country effect. (Scherer, Banse, and Wallbott 2001, 87).

Furthermore, Scherer discussed these results in 2003, claiming that: the data also suggest the existence of culture- and/or language specific paralinguistic patterns in vocal emotion expression, since, in spite of using language-free speech samples, the recognition accuracy decreased with increasing dissimilarity of the language spoken by the listeners from German. (Scherer 2003, 235).

These results have been confirmed and expanded by more recent studies (Bryant and Barrett 2008; Chen, Gussenhoven, and Rietveld 2004; Thompson and Balkwill 2006; Galatà and Romito 2010). Chen, Gussenhoven and Rietveld (2004) carried out a research on the perception of paralinguistic intonational meanings in British English and Dutch. Native speakers of both languages were asked to listen to a series of stimuli in their mother tongue and to judge each stimulus on semantic scales (i.e., self-confident vs. not self-confident; friendly vs. not friendly; surprised vs. not surprised, etc.). Stimuli “differed in pitch contour, pitch register and pitch span in Experiment 1, and in pitch register, peak height, peak alignment and end pitch in Experiment 2” (Chen, Gussenhoven, and Rietveld 2004, 311). The results revealed that native speakers of both languages perceived stimuli differently, supporting “the theory of paralinguistic meaning based on the universality of biological codes which, however, acknowledges a language-specific component in the

444

Chapter Eighteen

implementation of these codes” (Chen, Gussenhoven, and Rietveld 2004, 311). Thompson and Balkwill (2006) supported the idea that a combination of universal and cultural-specific cues encodes emotional prosody. This is one of the few studies involving western and non-western cultures. Their study involved native speakers of English, German, Japanese, Chinese and Tagalog. These were asked to produce neutral sentences conveying four different emotional cues (joy, anger, sadness and fear). A sample of English native speakers judged the emotional content conveyed by utterances produced by the subjects. The results revealed that accuracy was above chance in all four languages; nevertheless, the listeners scored the highest identification accuracy for English utterances and the lowest identification accuracy for Japanese and Chinese utterances. Galatà and Romito (2010) carried out a research on four European languages. A sample of actors and native speakers from different European countries were invited to reproduce a standard sentence in Italian, French, English and German. Italian listeners were asked to identify vocal emotions. The results showed that even those who had a lower language proficiency level were able to infer the emotional content from the sentences. Indeed, listeners who had claimed to have high English proficiency achieved a lower accuracy index. On the contrary, they reported higher accuracy in German, a language in which they were not proficient. Therefore, the above study excluded a straight correlation between language proficiency and the ability to identify vocal emotions. Some studies provide evidence that L2 learners experience increasing difficulties in identifying and reacting to emotional speech, the more distant they are from the culture of their L2. Being able to judge the interlocutor’s emotional state is crucial for successful communication; […] SLA research shows that learners from “distant” cultures experience significantly greater difficulties in identifying emotion in the L2 and in judging the intensity of that emotion than do fellow learners from “closer” cultures with similar levels of proficiency […] and language teachers need to be aware that cultural/typological distance between the learners’ L1(s) and their L2 is an important obstacle in mastery of emotional speech. (Dewaele 2005, 376).

In addition, previous research has focused on learners’ recognition process of vocal emotions (i.e., Chua and Schirmer 2011), but only a few studies (i.e., Armour 2013) have specifically addressed the process of encoding emotions by L2 users. The acquisition of emotional competence4 in L2 also implies the control and management of para-verbal components

The Acquisition of Emotional Competence in L2 Learners of Italian

445

of communication (intonation, rhythm, etc.). Recent studies have confirmed that such supra-segmental features represent a huge obstacle in second language acquisition, especially in adulthood (Marotta and Boula de Mareüil 2010), as the influence of L1 prosody is unconsciously transferred to the L2. Thus, the vocal expression of emotions in L2 should be taken into account in L2 teaching practices.

3. Objectives and hypothesis The present study started from the assumption that the perception and reproduction of vocal emotions in an L2 is an essential component of learners’ communicative and especially pragmatic competence. The aim of the research project was to explore whether training focused on pragmatic awareness encourages L2 learners from distant cultures to develop the ability to perceive, identify, and produce emotional stimuli in Italian. Therefore, we considered it extremely important to take into account both perspectives of emotional speech, i.e., speaker-centered (encoding) and listener-centered (decoding) perspectives. As regards the encoding process, we hypothesized that learners’ emotional speech would be affected by cultural features, as reported in a recent study on Chinese vocal emotions (Anolli, Wang, Mantovani, and De Toni 2008).5 Such cultural features might produce variations in fundamental frequency (F0) and pitch range values, intensity and speech rate with respect to native speakers of Italian. These differences might also represent an obstacle to the correct identification of emotions; thus, as regards the decoding process, we expected a) learners to possibly experience difficulty in identifying vocal emotions expressed in Italian by a native speaker and b) Italian listeners to possibly experience the same difficulty in identifying vocal emotions expressed in Italian by L2 learners. We therefore also expected pragmatic training to help Italian L2 learners to identify vocal emotions expressed by Italian speakers as well as to reproduce vocal emotions as similar as possible to those of the target language. Although we expected pragmatic training to help L2 learners to approximate native speakers’ vocal productive and interpretive behaviour, we did not expect the training to affect learners’ emotional dimension – that is the experiment did not aim at teaching learners to feel emotions as natives do, but rather to be functional in social interactions with Italian speakers. Thus, after the training, we also expected Italian native listeners to be able to infer emotional cues from learners’ emotional speech.

446

Chapter Eighteen

4. Method: participants and material In a class of Italian L2 learners we selected two groups of 5 female Indonesian students (aged 22-27) at a B1 level of proficiency, so as to study their vocal expression of emotions. The other learners did not take part in the experiment, some of them because they were from European countries (Spain, France, etc.) and others because they did not agree to participate. Whereas the first group of Indonesian students was involved in the pragmatic training, the second one acted as a control group and participated in the first stage of the training only (see section 4.1). We also involved an Italian speaker, a 26-years-old woman,6 in order to compare her vocal emotions with those of the Indonesian learners. We selected three pairs of “basic emotions”7 as the object of our investigation: anger/disgust, fear/sadness, joy/surprise. These are called “basic” because human beings are believed to share them with their primate ancestors (see Darwin 1982; Ekman 1982; Galati 1993; Balconi, 2004; Matsumoto 2009). The terms in each pair belong to the same semantic area and entail similar emotional implications (Anolli and Ciceri 1992).

4.1. The standard sentence and the recording procedure Our work dealt with simulated emotional states and adopted the elicitation procedure employed in previous research on vocal emotions in Italian (Anolli and Ciceri 1992; Anolli, Wang, Mantovani, and De Toni 2008). This procedure consisted in reading out a set of different stories, each containing a specific emotional stimulus conveyed by a standard sentence. The one employed in these studies was Non è possibile, non ora (“It’s not possible, not now”), which we reduced to Non è possibile. In addition, the original stories were rearranged and simplified, so as to facilitate the learners’ comprehension.8 The learners were supposed to read the six different stories on their own and to pronounce the sentence six times, each with a different emotional coloring, corresponding to anger, disgust, joy, surprise, fear and sadness, while trying to be as natural and spontaneous as possible. At the same time, they were supposed to express the same emotional states through their facial expressions as well. This additional task was expected to facilitate the participants’ identification with the emotional context proposed, as reported by Laird and Strout (2007, 56):

The Acquisition of Emotional Competence in L2 Learners of Italian

447

People can be induced to smile, and they report feeling happy […], to sit in a slumped posture to produce a feeling of sadness […] and to speak in a harsh tone to produce anger […].

These recordings took place in the first phase of the training and at the end of it (see section 4.4). We adopted the same elicitation procedure described above with the Italian native speaker. Data obtained was also employed in the second part of the research dealing with the decoding process. For each recording session, we employed a directional microphone (Dynamic supercardioid vocal microphone N/D767A) and a professional audiotape recorder. Recordings were collected in a classroom at the University of Calabria; although the room was silent, it was not soundproof.

4.2. The acoustic analysis We carried out a series of acoustic analyses of the emotional speech of the experimental group (recorded in the first and in the last phase of the training) and of the control group (collected in the first phase). In addition, we analyzed the native speaker’s vocal emotions, in order to make a comparison with those of the learners. We took into account the following parameters: fundamental frequency (frequency of vibration of the vocal folds, measured in Hertz); pitch range (the distance between the highest and the lowest F0 values); intensity (the amplitude of the sound waves, measured in decibels); duration of the sentences (measured in milliseconds); and speech rate (the number of syllables per second, including pauses). We focused on F0 and pitch range values in order to determine whether the productions were expressed by a dynamic or flat intonation. With regard to pitch range values, we normalized the data by converting hertz to semitones. We adopted ‘t Hart’s approach9 (‘t Hart, Collier, and Cohen 1990). The Acoustic analysis was carried out by using PRAAT,10 a speech analysis software (version 5.2) developed by Paul Boersma and David Weenik, from the Institute of Phonetic Sciences of the University of Amsterdam.

448

Chapter Eighteen

4.3. The perception test 4.3.1. Indonesian learners’ perception test The second part of the research focused on the perceptual aspect of emotional speech. In the first place, we wanted to check the possible improvements of the learners’ decoding proficiency. A perception test exemplifying the above six vocal emotions expressed by the female native speaker (see section 4.1) was administered to both groups of learners in the first phase of the training (T1), and to the experimental group in the last phase (T2) and a month after the training (T3). Students were asked to identify the emotions conveyed through the use of voice in those recorded sentences by choosing from six options (joy, anger, disgust, sadness, fear and surprise). 4.3.2. Italian listeners’ perception test After checking the learners’ decoding proficiency, we wanted to check the communicative effectiveness of the experimental group over the control group. Therefore, two perception tests were administered to a sample of 50 Italian listeners (25 women and 25 men), aged 18-54. The first one dealt with (a) the experimental group’s emotional speech produced in the first phase of the training and (b) that produced by the control group in the same phase; the second one contained the experimental group’s emotional speech collected in the last phase of the training. The listeners, mostly from Calabria and Southern Italy, were briefly informed about the goal of the experiment and invited to pay attention to the intonation patterns of each sentence and to choose from six different emotional states (anger, disgust, joy, fear, surprise and sadness). Emotional utterances were played randomly; no time limit was imposed. Both in the learners’ perception test and in the Italian listeners’ one, we opted for a multiple choice test in order to give the subjects a set of possible solutions and to reduce the possibility of collecting extremely varied responses, a risk associated with the use of open-question tests.

4.4. Pragmatic training The training of the experimental group lasted 4 weeks and was organized in four phases:

The Acquisition of Emotional Competence in L2 Learners of Italian

449

1) In the first stage, both groups participated in a general discussion aimed at exchanging ideas about emotional expressions in Italian and Indonesian cultures. The goal was therefore to highlight the main difficulties involved in communication. For example, the Indonesian speakers perceived the increasing loudness of voice of Italians as irritation or anger. In this phase we administered to both groups the above-mentioned perception test which dealt with the identification of six emotions expressed in Italian. The goal was to spot vocal emotions which might confuse L2 learners. In this phase, we also recorded both groups’ vocal emotions, by resorting to the elicitation procedure described in section 4.1. 2) The second stage involved only the experimental group and lasted one week. In this phase the training focused on audio-visual tools in order to give concrete examples of emotional states expressed in Italian. We selected three Italian TV series (Boris, Inspector Montalbano and Love Bugs), and three cartoons (Bambi, The little mermaid and Toy story). For each emotion we also provided a description of its prosodic features (intonation, speech rate, etc.) according to previous studies (see Anolli and Ciceri 1992, 322324). The goal of these descriptions was to make the learners aware of physiological changes (i.e., fast heartbeat, rapid rate of breathing) and the corresponding effects on the vocal expression of emotions (fast/slow speech rate, broad/narrow pitch range, high/low intensity and so on). The learners were interested in this kind of information, even though some aspects (such as intonation patterns) were more difficult to comprehend. 3) The third phase of the training focused on para-verbal communication through a series of specific exercises. We divided the experimental group into pairs and arranged the members of each pair back to back, to avoid visual contact between them. Since the group consisted of 5 participants, one of them did the exercise twice. In each pair, a learner was asked to express her emotions by pronouncing numbers from 1 to 4 (so as to prevent any semantic interference effect), giving her utterances specific coloring. The other learner was supposed to identify the emotional state expressed by her colleague by focusing on her intonation patterns only. The learners enjoyed this task and experienced no particular difficulties in doing it (although some emotions, like disgust or fear, were trickier to express than others). 4) In the last stage, the L2 learners were engaged in role-play activities dealing with the expression of emotions (i.e., the

450

Chapter Eighteen

performance of dialogues containing specific emotional contexts). They were also involved in a shadowing training activity, where they had to repeat sentences produced by native speakers of Italian while conveying a specific emotional state. Thus, the experimental group put into practice the prosodic suggestions given to them in phase 3. Finally, we recorded their vocal emotions once again, adopting the above-mentioned procedure. The data collected were compared with those recorded in phase 1 of the training.

5. Results: the encoding process Table 1 summarizes all the parameters taken into account in the acoustic analysis. Each of them will be discussed separately, but initially it is useful to observe the overall vocal cues of each emotion produced by the native speaker, the control group and the experimental group (in the last phase of the training). Briefly, the native speaker’s productions were characterized by more dynamic intonation, faster speech rate and higher intensity than those of the learners of both groups. On the other hand, the experimental group’s parameters were more similar to those of the native speaker and varied for each emotion (unlike the parameters of the control group, which were quite homogeneous for each production).

The Acquisition of Emotional Competence in L2 Learners of Italian

451

TABLE 1. Overview of the vocal cues for the six emotions in the native speaker by Italian and the Indonesian learners of the experimental and control groups Vocal cues F0 Mean

Pitch Range

Intensity

Duration

Speech Rate

Speakers

Anger

Sadness

Fear

Joy

Disgust

Surprise

Native speaker Experim. Group Control Group Native speaker Experim. Group Control Group Native speaker Experim. Group Control Group Native speaker Experim. Group Control Group Native speaker Experim. Group Control Group

290

218

330

319

286

203

281

212

327

236

276

198

201

197

259

198

224

183

28.4

22.4

31.3

10.6

27

14.8

16.2

16.6

10.3

11.4

17.6

5.5

8.5

6.2

7.8

6.2

7.6

4.2

70

60

69

70

62

56

68

62

68

64

60

55

63

61

55

55

59

51

1,449

1,594

1,546

1,029

1,053

1,331

1,070

1,710

1,460

1,840

1,380

2,060

1,210

1,350

1,290

1,190

1,450

1,360

4.1

3.7

3.8

5.8

5.6

4.5

5.6

3.5

4.1

3.2

4.3

2.9

4.9

4.4

4.6

5

4.1

4.4

5.1. F0 and pitch range values We compared the F0 and pitch range values of each vocal emotion produced by the native speaker, the control group and the experimental group (in the last phase of the training), as shown in Graphs 1 and 2. The native speaker’s values were higher than those of both groups (in

452

Chapter Eighteen

particular, the pitch range was much wider). On the other hand, the experimental group’s values were higher than those of the control group. We also compared the experimental group’s emotional speech produced in the first and in the last phase of the training: the acoustic analysis confirmed our hypothesis, showing that in the first phase, vocal emotions were characterized by low F0 values and a narrow pitch range, which gave rise to a flat tone. Thus, the learners employed one intonation contour for all emotions. In the last phase of the training, the experimental group’s values were extremely heterogeneous. F0 and pitch range increased, producing a more dynamic intonation in emotions like anger, fear, joy, disgust and surprise, as shown in Graphs 1 and 2. Only sadness values did not vary and remained low, in accordance with previous studies on vocal emotions11 (Anolli and Ciceri 1992; Laukka 2004). Tables 2 compares the control group’s vocal emotions with those produced by the experimental group in the last phase of the training. The control group’s emotional speech was characterized by static values. That is, F0 and pitch range were lower than those of the experimental group and did not vary with regard to the emotion expressed (as shown in Graphs 3 and 4).

GRAPH 1. F0 mean values for each vocal emotion produced by the native speaker and the learners of the experimental and control groups

The Acquisition of Emotional Competence in L2 Learners of Italian

453

GRAPH 2. Pitch range values for each vocal emotion produced by the native speaker and the learners of the experimental and control groups

TABLE 2. Experimental group’s F0 means and pitch range values (in the first and in the last phase of the training) Emotions

F0 means

F0 means

Anger Sadness Fear Joy Disgust Surprise

(First phase) 199.650 180.520 200.550 258.230 187.200 214.460

(Last phase) 281.910 198.560 236.630 327.240 212.914 276.210

Pitch range (First phase) 8.6 4.3 7.2 7.1 7.0 7.2

Pitch range (Last phase) 16.2 5.5 11.4 10.3 16.6 17.6

GRAPH 3. Experimental group’s F0 mean values (in the first and in the last phase of the training)

454

Chapter Eighteen

GRAPH 4. Pitch range

5.2. Intensity As shown in Graph 5, the experimental group’s intensity values were quite homogeneous for each emotion, except for anger and joy (whose values were high) and sadness (whose intensity was low). The control group’s values were lower; in particular, intensity was low in sadness and fear and joy; on the contrary, it was high in anger and, to a lesser degree, in disgust and surprise. Graph 6 compares the intensity values of the six vocal emotions expressed, respectively, by the native speaker, the experimental group and the control group. The native speaker’s values were higher than those of the learners of both groups.

GRAPH 5. Intensity values of vocal emotions expressed by the experimental and control groups

The Acquisition of Emotional Competence in L2 Learners of Italian

455

GRAPH 6. Intensity values of vocal emotions expressed by the native speaker and the learners of the experimental and control groups

5.3. Duration and speech rate In the native speaker’s productions, duration was extremely variable. It was short in emotions like anger, fear and surprise, whereas it was longer in the case of disgust, joy and sadness. As far as the control group is concerned, duration was almost the same for each emotion. By contrast, the experimental group’s vocal emotions presented variable durations. In particular, duration was short in emotions like anger and surprise, whereas it was longer in the case of sadness, disgust and fear (as shown in Graph 7). Concerning speech rate, the native speaker’s values were more variable than those of the Indonesian learners. Speech rate was fast in fear and surprise, whereas it was slower in sadness, anger, disgust and joy (as shown in Graph 8). The experimental group’s speech rate was fast in anger, joy and surprise; it was slower in sadness and a little less slow in fear and disgust. The control group’s speech rate was almost the same for each emotion, except for anger and fear, whose speech rate was higher than that of surprise, joy, sadness and disgust (as shown in Graph 9).

456

Chapter Eighteen

GRAPH 7. Duration values of vocal emotions expressed by the native speaker and the learners of the experimental and control groups

GRAPH 8. Native speaker’s speech rate values

GRAPH 9. Experimental and control groups’ speech rate values

The Acquisition of Emotional Competence in L2 Learners of Italian

457

5.4. Pitch contours Graphs 10, 11 and 12 show the intonation contours of anger produced, respectively, by the native speaker, one learner of the control group and one of the experimental group.

GRAPH 10. Native speaker’s pitch contour

GRAPH 11. Control subject’s pitch contour

GRAPH 12. Experimental subject’s pitch contour

The intonation contours of the Italian speaker and that of the experimental subject were characterized by a series of F0 peaks, which produced a more dynamic intonation contour. On the contrary, the control

458

Chapter Eighteen

subject’s contour was more regular and static, with only a few significant peaks.

5.5. Synopsis of the acoustic profiles of the six emotions The analysis revealed the following acoustic profiles with regard to the emotions expressed. Anger - High F0 values and broad pitch range characterized the native speaker’s speech. Duration was short and speech rate was medium. Intensity values were high. - The experimental group’s vocal profile was characterized by high speech rate, short duration and medium intensity. The F0 values were high and the pitch range was wide. - The control group’s vocal profile was characterized by medium speech rate and duration. Intensity was high, even though its values were lower than those of the experimental group, as were F0 and pitch range. Disgust - The native speaker’s vocal profile was characterized by medium F0 and wide pitch range. Intensity values and speech rate were medium. Duration was long. - Medium duration and speech rate characterized the experimental group’s speech. F0 values were medium and pitch range was wide. Intensity values were medium. - The control group’s vocal profile was characterized by medium speech rate, duration and intensity, low F0 and narrow pitch range. Surprise - High F0 and wide pitch range characterized the native speaker’s speech. Intensity, duration and speech rate values were medium. - The experimental group’s vocal profile was characterized by high speech rate and medium duration. F0 and pitch range were high, whereas intensity values were medium. - The control group’s vocal profile revealed medium F0 and pitch range, low intensity, fast speech rate and medium duration. Joy - High F0 and broad pitch range characterized the native speaker’s speech. Intensity was high, while duration and speech rate values were medium. - The experimental group’s speech was characterized by high F0,

The Acquisition of Emotional Competence in L2 Learners of Italian

459

broad pitch range, and medium intensity, duration and speech rate. The control group’s vocal profile revealed medium F0, pitch range, duration and speech rate. Intensity values were high. Sadness - Low F0 and narrow pitch range characterized the native speaker’s speech. Intensity, duration and speech rate values were medium. - The experimental group’s speech was characterized by low F0, narrow pitch range, long duration, low speech rate and intensity values. - The control group’s vocal profile revealed similar variations, although F0 and pitch range and intensity values were lower than those of the experimental group. Duration was long. Fear - High F0 and narrow pitch range characterized the native speaker’s speech. Intensity was high, while duration was medium and speech rate was fast. - The experimental group’s speech was characterized by medium F0 and narrow pitch range, high intensity, long duration and medium speech rate. - The control group’s vocal profile was characterized by low F0, narrow pitch range, high intensity, and medium duration. Speech rate values were high. -

To conclude, the native speaker’s vocal cues were more variable than those of the learners. In particular, F0 values were higher and pitch range was broader. Duration and speech rate varied depending on the emotion expressed. By contrast, the Indonesian learners’ values were lower and more static; nevertheless, the experimental group’s values were more variable than the control group’s and increased through the training, thus reducing the flatness of the learners’ productions.

6. Results: the decoding process The second part of the research focused on the decoding process of vocal emotions. Analysis of the perception test took into account the percentage of correct identifications.

6.1. The learners’ perception test As described in section 4.3, in the first phase of the training (T1) we administered the perception test to both groups of learners, and to the

Chapter Eighteen

460

experimental group also at the end of the training (T2) and a month after the training (T3). In the first stage (T1), we aimed to identify which emotions caused confusion among the learners. Results revealed that anger scored the highest percentage of correct identifications, followed by surprise. The other emotions were not correctly decoded. The mean of correct identifications per learner was 1.6 out of 6. In T2 and T3, the mean increased to 4 correct identifications per learner. Joy and anger obtained the highest percentage of correct identifications and the other emotions were correctly identified, including those which had raised confusion (surprise/joy, sadness/fear, etc.) Table 3 displays the percentages of correct identifications in relation with the emotion expressed. The mean values refer to the average number of correct identifications per learner at T1, T2 and T3 (also shown in Graph 14). Graph 13 shows the learners’ increasing accuracy as time progressed. T3 results confirmed that the experimental group enhanced their emotional competence even in the long run. TABLE 3. Experimental group’s correct identifications of vocal emotions Emotion Anger Surprise Sadness Disgust Joy Fear Mean

T1 100% 67% 0% 0% 0% 0% 1.6

T2 100% 100% 67% 33% 100% 33% 4

T3 100% 67% 100% 67% 67% 67% 4

GRAPH 13. Experimental group’s correct identifications of vocal emotions

The Acquisition of Emotional Competence in L2 Learners of Italian

461

GRAPH 14. Mean of correct identification per learners (T1, T2, T3)

GRAPH 15. Percentage of correct identifications: Control group vs. experimental group

Finally, Graph 15 shows the percentage of correct identifications scored by the control group (in the first phase of the training) and the experimental group (in the last phase). Results confirmed that the learners involved in all the phases of the pragmatic training were able to identify vocal emotions with higher accuracy than the control group.

6.2. Italian listeners’ perception test Table 4 summarizes the results. As Graph 16 shows, the emotional speech produced by the experimental group in the first stage of the training obtained 48% of correct identifications. In the productions collected in the last phase, accuracy percentage increased (60%). The pairs of emotions which had caused confused identifications, anger/disgust, as well as sadness/fear, showed improvement. On the other hand, the percentage of correct identifications for joy/surprise decreased.

462

Chapter Eighteen

TABLE 4. Italian listeners’ degree of accuracy in recognizing vocal emotions produced by the learners of the experimental group in the first and in the last phase of the training Emotion Disgust Anger Joy Surprise Fear Sadness Mean

First phase (%) 36 45 82 55 9 64 48.5

Last phase (%) 78 94 57 34 27 73 60

GRAPH 16. Italian listeners’ degree of accuracy in recognizing vocal emotions produced in the first and in the last phase of the training

GRAPH 17. Italian listeners’ degree of accuracy in recognizing vocal emotions produced by the control group and the experimental group

The Acquisition of Emotional Competence in L2 Learners of Italian

463

A comparison of the results of the two groups shows that, on the one hand, the experimental group’s emotional speech (produced in the last phase of the training) was mostly recognized by the Italian listeners. Anger, disgust and sadness obtained the highest percentage of accurate identifications, whereas surprise and fear scored lower accuracy levels. On the other hand, the listeners experienced greater difficulty in identifying emotions produced by the control group, as shown in Graph 17.

7. Discussion and conclusion The data obtained in this study support the three hypotheses we wanted to test: (a) the learners’ emotional speech in Italian was affected by cultural features, (b) the Italian learners experienced difficulties in identifying vocal emotions expressed in Italian by a native speaker and (c) the Italian listeners experienced the same difficulties in identifying vocal emotions expressed in Italian by Italian L2 learners. The training focused on the “pragmatic” gap identified between the Italian native speakers and the Italian L2 learners, and was adapted to a group of learners whose L1 (i.e., Indonesian) was typologically distant from Italian. The linguistic distance between L1 and L2 affected the emotional dimensions of the learners’ interaction with the native speakers, giving rise to misunderstandings. The instructional training helped the Indonesian learners to understand a series of pragmatic and prosodic aspects involved in the vocal expression of emotions in the Italian culture, and encouraged them to “play” with intonation and to observe differences at a perceptual level. The findings appear to be encouraging: on the one hand, the learners became aware of some of the cultural differences between their own and the native speakers’ vocal expression of emotions; this increased awareness helped them to improve their ability to decode vocal emotions. The degree of accuracy recorded in the perception tests administered at the end of the training and after a month confirmed this improvement. On the other hand, the Italian L2 learners improved their emotional speech as well, reducing the flatness of their intonation. In particular, the acoustic analysis registered an increase of F0 and pitch range values. In line with this finding, the Italian listeners were able to infer emotional cues from the standard sentence produced by the L2 learners. The findings suggest that pragmatics plays an important role in social interactions involving emotional issues. An awareness of the huge obstacles which learners face when interacting with native speakers should

464

Chapter Eighteen

persuade L2 teachers to adopt a specific didactic method to counterbalance such difficulties. Nevertheless, acquiring emotional competence in L2 is a long and hard process, which involves many variables. Teaching this skill is an arduous task. The training proposed here was employed in a pilot study; thus we expect that it might be improved in future implementations. It could represent the starting point of future research involving other typologically distant languages and the use of more specific instruments. Furthermore, the experiment could be implemented with a more heterogeneous group of students and over a wider training period. Although our research did not include the analysis of vocal emotions expressed in learners’ L1, we believe that its analysis is important in order to determine how L1 influences L2 productions in emotional contexts, and this will certainly be investigated in our future research.

References Anolli, L., and R. Ciceri. 1992. La voce delle emozioni. Verso una semiosi della comunicazione vocale non-verbale delle emozioni. Milano: Franco Angeli. Anolli, L., L. Wang, F. Mantovani, and A. De Toni. 2008. “The voice of emotion in Chinese and Italian young adults.” Journal of CrossCultural Psychology 39 (5): 565-598. Armour, W. 2013. “Voice to text: Assessing the expression of emotions by L2 learners of Japanese reading manga aloud.” In Disciplines: the lenses of learning, edited by K. Coleman, and A. Flood, 188-201. Champaign, IL: Common Ground. Balconi, M. 2004. Neuropsicologia delle emozioni. Roma: Carocci. Banse, R., and K. Scherer. 1996. “Acoustic profiles in vocal emotion expression.” Journal of personality and social psychology 70 (3): 614636. —. 2009. “The dynamic architecture of emotion: evidence for the component process model.” Cognition and emotion 23 (7): 1307-1351. Boersma, P., and D. Weenink. 2014. Praat: doing phonetics by computer, version 5.3.76. Computer program, http://www.praat.org/. Bryant, G. A., and C. H. Barrett. 2008. “Vocal emotion recognition across disparate cultures.” Journal of Cognition and Culture 8 (1-2): 135-148. Campbell, N. 2000. “Databases of emotional speech.” Speechemotion2000 2 (11): 34-38.

The Acquisition of Emotional Competence in L2 Learners of Italian

465

Chen, A., C. Gussenhoven, and T. Rietveld. 2004. “Language-specificity in the perception of paralinguistic intonational meaning.” Language and Speech 47: 311-349. Chua, S. M., and A. Schirmer. 2011. “Perceiving verbal and vocal emotions in a second language.” Cognition and Emotion 25 (8): 13761392. Darwin, C. 1872. The expression of emotions in man and animals. London: John Murray. Translated by G. A. Ferrari as L'espressione delle emozioni negli uomini e negli animali (Torino: Boringhieri, 1982). Dewaele, J. 2005. “Investigating the psychological and emotional dimensions in instructed language teaching: obstacles and possibilities.” The modern language Journal 89: 367-380. D’Urso, V., F. Cavicchio, and E. Magno Caldognetto. 2008. “Le etichette lessicali nelle ricerche sperimentali sulle emozioni: problemi teorici e metodologici.” In Comunicazione parlata e manifestazione delle emozioni, edited by E. Magno Caldognetto, F. Cavicchio, and P. Cosi, 201-222. Napoli: Liguori. Ekman, P. 1982. Emotion in the human face. Cambridge: Cambridge University Press. Galatà, V., and L. Romito. 2010. “Gli italiani e il decoding di emozioni vocali in quattro lingue europee.” In Atti del convegno AISV, Parlare Con Le Persone, Parlare Con Le Macchine, edited by F. Cutugno, P. Maturi, R. Savy, G. Abete, and J. Alfano, 197-231. Torriana: EDK. Galati, D. 1993. Le emozioni primarie. Torino: Bollati Boringhieri. Gerrards-Hesse, A., K. Spies, and F.W. Hesse. 1994. “Experimental inductions of emotional states and their effectiveness: a review.” British Journal of Psychology 85 (1): 55-78. Grandjean, D., T. Bazinger, and K. Scherer. 2006. “Intonation as an interface between language and affect.” Progress in Brain Research 156: 235–247. Hammerschmidt, K., and U. Jürgens. 2007. “Acoustical correlates of affective prosody.” Journal of Voice 21 (5): 10. ‘t Hart, J., R. Collier, and A. Cohen. 1990. A perceptual study of intonation. An experimental-phonetic approach to speech melody. Cambridge: Cambridge University Press. Johnstone, T., and K. Scherer. 2000. “Vocal Communication in Emotion.” In Handbook of Emotion, edited by M. Lewis, and J. M. Haviland, 220-235. New York: Guilford.

466

Chapter Eighteen

Laird J. D., and S. Strout. 2007. “Emotional Behaviours as Emotional Stimuli.” In Handbook of emotion elicitation and assessment, edited by J. A. Coan, and J. B. Allen, 54-64. Oxford: Oxford University Press. Laukka, P. 2004. “Vocal expression of emotion: discrete-emotions and dimensional accounts.” PhD dissertation, Uppsala University. Levenson, R. W., J. Soto, and N. Pole. 2007. “Emotion, biology and culture.” In Handbook of cultural psychology, edited by S. Kitayama, and D. Cohen, 780-796. New York: Guilford. Marotta, G., and P. Boula de Mareuil. 2010. “Persistenza dell'accento straniero. Uno studio percettivo dell'italiano L2.” In La dimensione temporale del parlato. Atti della V conferenza dell’associazione italiana di scienze della voce, edited by S. Schmid, M. Schwarzenbach, and D. Studer, 475-494. Rimini: EDK. Matsumoto, D. 2009. “The origin of universal human emotions.” http://davidmatsumoto.com/content/NG%20Spain%20Article_2_.pdf. Mayer J. D., and P. Salovey. 1997. “What is emotional intelligence?” In Emotional development and emotional intelligence, edited by P. Salovey, and D. J. Sluyter, 3-31. New York: Basic books. Mesquita, B., and R. Walker. 2003. “Cultural differences in emotions: a context for interpreting emotional experiences.” Behaviour Research and Therapy 41 (7): 777-793. Mozziconacci, S. 2002. “Prosody and emotions.” In Proceedings of the 1st International Conference on Speech Prosody, edited by B. Bel, and I. Marlien, 1-9. Aix-en-Provence: Laboratoire Parole et Langage. Pavlenko, A. 2006. Emotions and multilingualism. Cambridge: Cambridge University Press. Pell, M. D., L. Monetta, S. Paulmann, and S. A. Kotz. 2009. “Recognizing emotions in a foreign language.” Journal of Nonverbal Behavior 33: 107-120. Saarni, C. 1999. The development of emotional competence. New York: Guilford. Scherer, K.2003. “Vocal communication of emotion: A review of research paradigms.” Speech Communication 40: 227-256. —. 2009. “The dynamic architecture of emotion: Evidence for the component process model.” Cognition and Emotion (23) 7: 13071351. Scherer, K., R. Banse, and H. G. Wallbott. 2001. “Emotion inferences from vocal expression correlate across languages and cultures.” Journal of Cross-Cultural Psychology 32 (1): 76-92.

The Acquisition of Emotional Competence in L2 Learners of Italian

467

Scherer, K., R. Banse, H. G. Wallbott, and T. Goldbeck. 1991. “Vocal cues in emotion encoding and decoding.” Motivation and emotion 15: 123-148. Scherer, K., and T. Bänziger. 2004. “Emotional expression in prosody: a review and an agenda for future research.” In Speech prosody, edited by B. Bel, and I. Marlien, 359-366. Nara, Japan: SProSIG. Sorianello, P. 2006. Prosodia. Roma: Carocci. Thompson, W. F., and L. Balkwill. 2006. “Decoding speech prosody in five languages.” Semiotica 158 (1): 407-424. Toshinori Ishi, C., and N. Campbell. 2012. “Analysis of acoustic-prosodic features of spontaneous expressive speech.” Revista de estudos da linguagem 12 (2): 38-49.

Notes 1

The paper is the result of the joint work of the two authors. In particular, De Marco is responsible for sections 1, 2, 4.3, 4.3.1, 4.3.2, 4.4, 5.5, 6, 6.1, 6.2, 7 and Paone for sections 3, 4, 4.1, 4.2, 5, 5.1, 5.2, 5.3, 5.4. 2 “Speech communication conveys more than the syntactic and semantic content of sentences. Various prosodic cues are available to speaker and listener in order to encode and decode the full spoken language. […] prosodic cues provide information such as the speaker’s gender, age and physical condition, and the speaker’s view, emotion and attitude towards the topic, the dialogue partner, or the situation” (Mozziconacci 2002, 1). 3 To further clarify this point, we report a series of acoustic profiles proceeding from Banse and Scherer’s study (1996, 616): “Anger: Anger generally seems to be characterized by an increase in mean F0 and mean energy […] the rate of articulation usually increases. Fear: […] increases in mean F0, in F0 range, and high-frequency energy. Rate of articulation is reported to be speeded up. Sadness: […] A decrease in mean F0, F0 range, and mean energy is usually found […] There is evidence that high-frequency energy and rate of articulation decrease. Joy: […] we find a strong convergence of findings on increases in mean F0, F0 range, F0 variability, and mean energy. There is some evidence for an increase in high-frequency energy and rate of articulation. Disgust: […] the results for disgust tend to be inconsistent across studies. The few that have included this emotion vary in their encoding procedures from measuring disgust […] at unpleasant films to actor simulation of the emotion. The studies that used the former found an increase in mean F0, whereas those that used the latter found […] a lowering of mean F0.” (Banse and Scherer 1996, 616). 4 For a more in-depth discussion of emotional competence we refer to three publications: a) Mayer and Salovey (1997, 5) wrote that: “emotional intelligence involves the ability to perceive accurately, appraise, and express emotion; the ability to access and/or generate feelings when they facilitate thought; the ability to

468

Chapter Eighteen

understand emotions and emotional knowledge; and the ability to regulate emotions to promote emotional and intellectual growth”; b) D’Urso, Cavicchio and Magno Caldognetto (2008) claimed that native speakers of any linguistic community develop the ability to identify and name emotional states. Such [an] ability, called emotional competence, varies across cultures. It helps to communicate emotional states to oneself and to the others; c) Saarni (1999, 2) stated that: “Emotional competence entails resilience and self-efficacy (and selfefficacy includes acting in accord with one’s sense of moral character). When one is emotionally competent, one is demonstrating one’s self-efficacy in emotionallyeliciting [sic] transactions, which are invariably social in nature.” 5 “Chinese participants had a generally slower speech rate and longer pauses in the vocal expression of emotions than Italian participants. With regard to vocal pitch cues in expressing emotions […], Chinese participants varied much less from their neutral vocal cues […] than did their Italian counterparts; […] furthermore, as regards voice intensity, Chinese participants displayed much less variation in VoInt (M) […] and VoInt (Max) across different emotions than did the Italians” (Anolli, Want, Mantovani and De Toni 2008, 579). 6 The experiment involved a female Italian speaker in order to avoid gender differences in the encoding process, and consequently to obtain comparable data. 7 “Research has demonstrated that a class of emotions known as basic emotions has a unique set of characteristics that distinguish them from all other emotions. These characteristics include unique physiological signatures, distinctive changes in mental activities and attention, subjective experience, and reliable nonverbal signals. Moreover, these characteristics are universal to all people of all cultures. For now, the basic emotions include anger, contempt, disgust, fear, enjoyment, sadness, and surprise.” (Matsumoto 2009, 3). 8 In the case of disgust, we created a new story, in which the female protagonist went to a restaurant and ordered a vegetables pizza without basil (as she really hated it). When she took a bite off the pizza, she realized it contained a lot of basil and exclaimed: “It’s not possible! It’s disgusting!”. 9 We applied the following formula: D = 12*log2(f1/f2) = 12/log102*log10(f1/f2). See also Sorianello (2006). 10 For further information we refer to http://www.praat.org/. 11 “[I]t is well established that mean, range, and variability of F0 rises for “active emotions” (e.g., anger, fear, happiness) and decreases for “passive” emotions (e.g., sadness)” (Laukka 2004, 17).

CHAPTER NINETEEN DENN, EIGENTLICH, ÜBERHAUPT: THREE “PRAGMATIC PARTICLES” IN GERMAN MARION WEERNING

1. Introduction The German language is rich in particles with a very vague or shadowlike lexical meaning whose function lies in managing the interaction between speaker and listener: in a German dialogue, every 8th to 12th word (Helbig and Kötz 1981, 43) is such a pragmatic particle. Many languages, such as Italian, do not have this kind of words (or only a few of them), so that they can cause difficulties for learners of German as a foreign language, who are used to setting a lexical equivalent from their mother tongue beside each new word they acquire in their target language. Even if a considerable amount of literature has been published on these particles, they remain one of the most controversial issues in German grammar, as is reflected in German grammarians and textbook writers’ inability to agree on what to call them. After briefly describing their features and differentiating them from other particles, this study focusses on three typical representatives: denn, eigentlich and überhaupt. In order to illustrate their meaning, their use and the difficulty of rendering them into Italian, which does not possess lexical equivalents of them, it then furnishes examples taken from the Italian translation of a German novel (Heidenreich 2002, 2004). Once the complexity of these three words has been worked out, the study goes on to critically examine how German textbooks for beginners in Italian schools (Catani et al. 2013a, 2013b; Curci, Bente Pieper, and Roth 2005a, 2005b, 2005c, 2005d; Motta 2004a, 2004b; Vanni and Delor 2010a, 2010b; Weerning 2007a, 2007b) treat these three frequent German words. Finding that they pay little or no attention to pragmatic particles in general, this article proposes new strategies for teaching them effectively.㻌

470

Chapter Nineteen

2. Abtönungspartikeln 2.1. General features Modality can be expressed in a variety of ways. Speakers (or writers) can show, for example, their interest in obtaining information from listeners (or readers) by explicit verbal means (Ich würde gern / muss wissen: [...]? “I would like to / I must know: [...]?”).1 Every language has its own lexical and/or grammatical means. In spoken language, speakers can also modulate their interest by prosodic and non-verbal means. Whereas Italian speakers underline attitudinal aspects through facial expressions accompanied by gestures, German L1 speakers normally avoid this kind of physical interaction: they express modal aspects within the speaker-listener interaction in a less direct and less explicit way than Italians. German offers a means of expressing stance that many other languages do not possess:2 so-called Abtönungspartikeln,3 “modulating particles”, a set of words with a similar function but with a very subtle difference between each other and which have practically lost their lexical meaning. For example, even a German mother tongue speaker cannot easily explain the difference between Wie spät ist es?, Wie spät ist es denn?, Wie spät ist es eigentlich? and Wie spät ist es überhaupt? “(Tell me,) what time is it (, then)?” When conducting a conversation, German speakers are not usually aware of the contribution that Abtönungspartikeln make to the orientation of listeners (Hentschel 1986, 32), in much the same way that Italian speakers are not aware that they are raising their eyebrows when communicating. The speaker uses particles to modulate the illocution like a painter uses shades to modulate a colour. Indeed, in German, abtönen, “tone down”, is a verb used in painting. The addressee perceives not only a general characteristic, for example [+GREY], if we are considering this colour inside an image, or [+INTEREST], if we are considering the manner in which the illocution of asking something is performed by the speaker inside a dialogue; but in the same way that (s)he can see the gradations between light grey and dark grey when looking at an image, (s)he can perceive greater or lesser degrees of interest (or other modal features, such as certainty, surprise or anger) expressed by the speaker in a dialogue. However, Abtönungspartikel (AP) is not the only term modern German grammar and textbook writers use when talking about these words, as Table 1 shows.

Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German

471

TABLE 1. Terms for APs in German reference grammars Reference Grammar Eisenberg (2006), Zifonun et al. (1997), Hentschel and Weydt (2002), Hoffmann (2013) Duden (2006) Engel (2004) Weinrich (1993) Boettcher (2009) Helbig and Buscha (1980)

Term Abtönungspartikel

Gloss modulating particle

Abtönungspartikel (Modalpartikel) Abtönungsangabe Modalpartikel Einstellungspartikel Partikel

modulating particle (modal particle) modulating adjunct modal particle attitude particle particle

Sometimes they are also described simply as Füllwörter, “filling words”, or Würzwörter, “spice words.” Indeed, APs can be omitted without changing the propositional content of the sentence and its validity. They are not informative, but meta-informative. For a long time, German grammarians did not deal with APs at all, only considering them parasites within utterances.4 Only when language started to be considered more than simply a system of rules, that is, a means of interaction, in general, (Köller 2004, 528) and a means of realising (speech) acts, in particular, did linguists and grammar writers begin to pay attention to these pragmatic particles. Only by analysing conversational German could it become clear that it is nearly impossible to structure linguistic interaction processes without them (Köller 2004, 528). Indeed, “they act as a kind of lubrication in dialogue” (Durrell 1992, 134). This implies that they have to be examined not by looking for a function they may serve in verbalising objects and facts, but by considering their function in the interaction process (Köller 2004, 528). Studies of language use in interaction have shown that APs are interpretable only in relation to the context and situation, being the most important intermediaries of attitudes, expectations and different connotations (Weinrich 1993, 843). The function of these illocution indicators (“illokutive Indikatoren”, Helbig and Kötz 1981, 16) “is broadly to clarify or emphasize to the listener the speaker’s attitude to what he or she is saying” (Durrell 1992, 134), even if in many grammars the listener is often not mentioned (as in Weinrich 1993). Others, such as Burkhardt (1986, 146), argue that APs do not actually make attitudes to the proposition clear, but rather evaluate the listener’s assessments, and Kegel (2006, 431-32) arrives at the conclusion that APs enable the speaker to have her/his intentions understood with the highest degree of probability,

472

Chapter Nineteen

while the recipient receives additional linguistic hints leading to an almost unequivocal understanding of the speaker’s utterance; seen this way, APs serve to optimise communication.

2.2. APs vs. other particles APs, however, are only one sub-class of the larger class of particles5 that characterise German, and their peculiarity can only be fully grasped when they are compared to the other particles available in German. These include a quite heterogeneous group of particles that can be found as independent clause constituents (i.e., not as a part of a clause constituent) in the Mittelfeld6 of the German sentence, often merely called adverbs. By applying a narrow definition, adverbs are only those invariable words which can answer wh-questions (Wann? - Heute. Wo? Hier. Wie? - Gern. “When? - Today. Where? - Here. How? - With pleasure.”). Adverb-like words (i.e., not true adverbs) include – beside APs – words such as trotzdem, “nevertheless”, deshalb, “therefore”, also, “so”, which are often called Konnektivpartikeln, “connecting particles”, because they connect the event or situation represented in the clause (to which they belong) to another event or situation which has just been mentioned (Anna ist krank, sie bleibt deshalb zu Hause. “Anna is ill, therefore she’s staying at home.”); other adverb-like words are wahrscheinlich, “probably”, vielleicht, “perhaps”, and leider, “unfortunately”, which are often called Modalpartikeln,7 “modal particles”, or Kommentarpartikeln (KPs), “commenting particles”, because, through them, the speaker expresses her/his subjective view or perception of the degree of validity of the utterance (Burkhardt 1985, 266). Validity-qualifier KPs and illocution-modifier APs differ in subtle ways. For example, the KP wahrscheinlich and the AP wohl are both usually translated as “probably.” However, with wahrscheinlich, the speaker undercuts the validity of an utterance by expressing a subjective degree of validity (in Anna kommt wahrscheinlich, “Anna is probably coming”, the degree of validity of Anna’s coming is quite high); instead, with wohl the speaker conveys her/his attitude to what (s)he is saying (in Anna kommt wohl, “Anna is coming, probably”, the speaker’s attitude is one of relatively high confidence). For a non-linguist, it may be quite difficult to catch the subtle difference between both words in this way, though syntactic criteria allow her/him to separate APs clearly and easily from similar words belonging to different sub-classes of particles.

Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German

473

The place of an AP is in the centre of the Mittelfeld of the German sentence, together with commenting particles, text-connecting particles and adverbs, but APs can never occupy the Vorfeld8 (e.g., Anna kommt wohl, but *Wohl kommt Anna); this makes them completely different from adverbs and other adverb-like particles (Anna kommt wahrscheinlich / deshalb / heute, as well as Wahrscheinlich / Deshalb / Heute kommt Anna, “Probably / For this reason / Today Anna is coming”). As their place is in the Mittelfeld of the German sentence, we can find them in all kinds of clauses, but the prevailing pragmatic function of each AP permits them to be sorted according to functionally classified clause types (Dürscheid 2000, 64-65) or speech acts; for example, mal with its banalising sense is typical of imperative clauses and requests, and denn with its function of signalling the speaker’s interest belongs to interrogative clauses or the speech act of asking something.9 Further syntactic criteria that distinguish APs from other particle classes are the following: x Contrary to discourse markers such as hallo, “hello“, or hm, APs are integrated in the sentence and cannot stand alone; x Contrary to prepositions, APs do not determine the morphological behaviour of other elements in the sentence; x Contrary to coordinating and subordinating conjunctions, APs cannot introduce a sentence, and they do not determine the syntactic behaviour of other elements in the sentence; x Contrary to response particles, APs cannot answer a polar question; x Contrary to adverbs (in a narrow definition), APs cannot answer a non-polar question; furthermore, they cannot be modified by intensifying particles such as sehr, “very.” Contrary to other independent clause constituents, APs cannot be negated by nicht, “not”, and they cannot be coordinated by und, “and”, or oder, “or”, but they can be combined with each other as far as their syntactic distribution (Zifonun et al. 1997, 1211) allows (Was wollt ihr denn eigentlich überhaupt machen? “Tell me, but …, what do you want to do, then?”). Furthermore, it is controversial whether they can be stressed; a few of them can, either increasing the pragmatic effect or markedly changing the pragmatic meaning of the utterance, as in Was [']macht ihr denn? (Lernt ihr?), “What are you doing? (Are you studying?)” as opposed to Was macht ihr [']denn (wenn ihr nicht lernt)? “What are you doing, then (if you aren’t studying)?”10

474

Chapter Nineteen

Even if it is possible, as we has seen, to distinguish APs from other particles by means of syntactic criteria, it is controversial whether APs should be treated as a word class of their own or as a set of abnormal, nonprototypical adverbs.

2.3. APs and homonymy When we find an AP in the Mittelfeld of a sentence, its classification is not always clear at first sight because nearly all APs have homonyms (Hentschel and Weydt 2002, 647) in other word classes. Whereas the words considered APs in modern German have acquired a new function (which is merely pragmatic)11 and lost nearly all of their original meaning (which has been reduced to a shadow, Boettcher 2009, 165),12 their homonyms have maintained a full lexical meaning. So we have, for example, the AP ruhig, which lacks a lexical meaning, as well as the adjective ruhig, which means “quiet” (e.g., in Schlaf ruhig! ruhig can be interpreted as an AP “Don’t worry, just sleep!” or an adjectival adverb “Sleep quietly / peacefully!”). Nonetheless, for Hoffmann (2013, 403) the fundamental meaning of APs plays a central role in their usage, in agreement with Weydt (1977, 222), who suggests tracing the meaning of the different homonyms back to one global meaning in order to understand them better (in the case of ruhig, the global meaning of both AP and adjective is the lack of tension). The next section will examine denn, eigentlich and überhaupt as typical representatives of the AP class and illustrates the above-mentioned aspects of APs from the point of view of Italian, a language which does not possess this particle class.

3. Denn, eigentlich and überhaupt as typical representatives of the AP class Let us consider a textbook dialogue in which two classmates – at the beginning of the school year – are talking about their new timetable and their teachers, and at a certain point one of them asks the other if he knows who their teacher will be that year (Catani et al. 2013a, 40): Sag mal, wer ist eigentlich unser Klassenlehrer?

However, the same question could also be formulated by omitting the AP or using other APs:

Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German

475

Sag mal, wer ist unser Klassenlehrer? Sag mal, wer ist überhaupt unser Klassenlehrer? Sag mal, wer ist denn unser Klassenlehrer?

The differences among these four questions are very subtle. Without an AP, the question seems quite brusque; it seems unnatural, as if something were missing. But what is the difference between denn, eigentlich and überhaupt? The DWDS, “Digital Vocabulary of German Language” (BBAW 2013), a monolingual dictionary edited by the Berlin-Brandenburgische Akademie der Wissenschaften (BBAW), furnishes überhaupt and denn as synonyms for eigentlich, and eigentlich and denn as synonyms for überhaupt. But, as we can see below, denn, eigentlich and überhaupt are not synonymous; they are only similar. Moreover, bilingual dictionaries cannot provide further assistance in revealing the differences between these words because, like other APs, they have undergone a semantic bleaching; the most frequent translation offered by German-Italian dictionaries is ma, “but”,13 for all three. Let us try, in the following sub-sections, to go deeper and discover the differences between these three APs, taking a look at their fundamental meanings, their homonyms and their translatability into Italian.

3.1. Denn 3.1.2. Fundamental meaning㻌 㻌 The etymology of the German denn is the same as that of dann: both denn and dann have the same common root as the English then and than. In the Hamburg area, denn is used instead of dann (Bis denn! instead of Bis dann! “Till then!”), whereas in standard German denn can be replaced by dann in conditional sentences in which denn is unstressed, in exclamations where denn is followed by stressed doch, and in questions where stressed denn implies wenn nicht das gerade Ausgeschlossene, “if not that, what was just excluded”: […] ich klingelte einfach [...], und wenn es denn sein soll, dann bist du auch hier. (Heidenreich 2002, 202). = […], und wenn es dann sein soll, dann bist du auch hier. [[…] I rang the doorbell […], and, God willing, you are here, too.] Solche unerwarteten Dimensionen beunruhigten die Tölzer denn [']doch […]! (Heidenreich 2002, 61).

Chapter Nineteen

476

= Solche unerwarteten Dimensionen beunruhigten die Tölzer dann [']doch […]! [Such unexpected dimensions worried the inhabitants of Tölz […]!] Ihr macht keine Ferien? Was macht ihr [']denn? = Was macht ihr [']dann? [You aren’t making a journey? But what are you doing, then (if you aren’t making a journey)?]

In addition, in declarative and imperative clauses containing a connecting so … denn, which belongs to an informal register, denn can be replaced by dann: Viele hatten Fieber. So passierte es denn, dass auch Anna krank wurde. = So passierte es dann, dass [...]. [Many people had fever. So it happened that Anna fell ill, too.] So nimm denn meine Hände (German hymn) = So nimm dann meine Hände [Take my hands]

In light of this, perhaps we could say that, even in modern German, both denn and dann have maintained their demonstrative sense relating to the situational context: denn to the causal context and dann to the temporal or conditional context. 3.1.2. Homonymy Denn, as a coordinating conjunction, introduces a main clause in the same way as und, “and” – not a secondary clause like the German weil, English because or Italian perché, which are subordinating conjunctions. The denn-clause furnishes a reason for what was said previously: Ich war traurig, denn ich liebte [...] Gertrud. (Heidenreich 2002, 227). [I was sad because I loved [...] Gertrud.] 㻌

The adjunction14 denn, “than”, occurs in a few comparative expressions, such as (mehr) denn je, “(more) than ever”, although it is usually replaced in modern German by als after a comparative. In contemporary German, the AP denn is particularly frequent (Zifonun et al. 1997, 1230). It is typical of interrogative clauses, especially yes-noquestions and questions regarding information or motivation (Dahl 1988, 64). The speaker wants to know more, and denn makes her/his question seem more natural or friendly (E-Großwörterbuch Deutsch als Fremdsprache 2003): it intensifies the whole question. It expresses interior

Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German

477

expectation, active interest, impatience, surprise (BBAW 2013) and doubts (E-Großwörterbuch Deutsch als Fremdsprache 2003). Denn always creates a relation to what preceded it in the context: the denn-question results directly from the situation (Zifonun at al. 1997, 1230). We can find several examples of this use of the AP denn in Gisela Heidenreich’s autobiographical novel Das endlose Jahr, for example, when Gisela’s mother, looking around at the landscape where she had lived many years before, asks: Wo ist denn der Wasserfall? (Heidenreich 2002, 11). [(But ...) where is the waterfall?]

In this example, the question about the waterfall just arises from the situational context (i.e., Gisela’s mother is looking around): denn anchors what is asked for inside the context. 3.1.3. Translatability into Italian 㻌 As regards the translatability of the AP denn into Italian, the GermanItalian dictionary DIT (2002, 186) suggests dunque, allora, poi, mai, “so, then, ever”, as possible translations for denn in interrogative clauses; Bianco et al. (2008) suggest e, “and”, ma, “but”, and even ma si può sapere, “but can we know.” To Burkhardt (1985, 265), poi, “then”, seems to be the best Italian correspondent, in the same way as Durrell (1992, 136) sees a “possible English equivalent” in then at the end of the sentence (Did you …, then?). This interpretation coincides with the etymology of the word and its closeness to the temporal adverb dann, while Helling (1983, 30) argues that denn is placed in polar and non-polar questions expressing the search for a reason, which puts it nearer to the causal conjunction denn. Analysing Belli’s translation of Heidenreich’s novel, we can see that, very often, denn is not translated at all: Wie heißt du denn? (Heidenreich 2002, 131). Come ti chiami? (Heidenreich 2004, 137). [What’s your name?]

When it is translated, we find ma, “but”, or e, “and”, for example: Wo ist denn der Wasserfall? (Heidenreich 2002, 11). Ma dov’è la cascata? (Heidenreich 2004, 13). [But where is the waterfall?]

478

Chapter Nineteen Warum hast du denn so einen alten Mann geheiratet? (Heidenreich 2002, 41). E perché hai sposato un uomo così vecchio? (Heidenreich 2004, 44). [And why did you marry such an old man?]

The examples analysed confirm the vague lexical meaning of denn and the difficulty of rendering its full pragmatic function into Italian.㻌

3.2. Eigentlich 3.2.1. Fundamental meaning㻌 㻌 The global meaning of eigentlich has to do with the real essence of something: etymologically, eigentlich is composed of eigen, “own”, + the suffix -lich, “-ly.” This fundamental meaning can be traced in all uses of eigentlich.㻌 㻌 3.2.2. Homonymy㻌 㻌 As an attributive (and therefore declined) adjective, its place in the noun phrase is between article (if there is one) and noun (der eigentliche Wert, “the real / true value”); it is not possible to use it as a predicative adjective (*Der Wert ist eigentlich㻌“The value is true.”). Contrary to APs, the KP eigentlich can also occupy the first place in a sentence. Trying to paraphrase the KP eigentlich in German, the DWDS (BBAW 2013) suggests im Grunde genommen, “basically”, but Belli, rendering Heidenreich’s novel into Italian (Heidenreich 2004), does not translate it at all: Eigentlich war ich ein sehr schüchternes Kind [...]. (Heidenreich 2002, 74). All’epoca ero una bambina molto timida […]. (Heidenreich 2004, 79). [At that time I was a very shy child [...].]

There are only a few cases in which it is translated: Es ging eigentlich nur darum, wer die meisten und schönsten Käfer gefangen hatte. (Heidenreich 2002, 234). Gli unici due criteri che contavano erano in realtà il numero e la bellezza degli esemplari. (Heidenreich 2004, 242). [The only two criteria which really counted were the number and the beauty of the beetles.]㻌

Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German

479

The classification of eigentlich as an AP is controversial and difficult: for Zifonun et al. (1997, 1209), it is a peripheral AP, like überhaupt, and contrary to denn, which belongs to the core group of APs. The AP eigentlich – which like überhaupt is used much less frequently than the AP denn, also in Heidenreich (2002) – is typical for interrogative clauses and is normally unstressed. It “tones down questions and makes them sound casual” (Durrell 1992, 137) or clear and confidential (BBAW 2013). For Albrecht (1977, 651), this is the case only if the partners in the dialogue know each other. This confidentiality sometimes appears brusque to Bianco et al. (2008). Boettcher (2009, 168) goes further, perceiving a confrontational connotation, one which we can also locate in Heidenreich (2002, 175), when Gisela sits down to drink a cup of coffee with her mother and […] fragte für sie völlig unvermittelt: “Weißt du eigentlich, wie der Bruder von meinem Vater heißt?” [[…] asked completely abruptly: “(Tell me,) do you know the name of my father’s brother?”]

Métrich and Faucher (2009, 289-90) admit a large spectrum of speaker emotions in eigentlich-questions, including annoyance and surprise. For Helbig and Kötz (1981, 34), eigentlich introduces a new, or more important aspect compared to what was said before, as in Heidenreich (2002, 36), when Gisela’s mother talks about her journey to Oslo and Gisela asks: Wieso eigentlich mit dem Zug? [(And …) why by train?]

For Métrich and Faucher (2009, 289-90), using eigentlich can also make the question stand out against its background situation, putting the question in the foreground and emphasising the thematic relation between the question and the situational background which can be real or feigned (Métrich and Faucher 2009, 289-90). 3.2.3. Translatability into Italian Possible translations of the AP eigentlich into Italian are offered, for example, by the DIT (2002, 221) which suggests in realtà, (ma) dunque, (ma) insomma, “really, (but) so, well”, and by Bianco et al. (2008), who propose dimmi un po’, a proposito, allora, e, ma, insomma, “tell me, by the way, then, and, but, well.” Analysing Belli’s translation of

480

Chapter Nineteen

Heidenreich’s novel (Heidenreich 2004), we find that, contrary to denn, eigentlich is nearly always translated: Hast du eigentlich den Bücherschrank ausgeräumt? (Heidenreich 2002, 100). Hai per caso svuotato l’armadio dove la nonna tiene i libri? (Heidenreich 2004, 106). [Did you, by chance, empty the bookcase where grandmother keeps the books?] [...] wie hieß der Bruder eigentlich? (Heidenreich 2002, 169). [...] ma come si chiamava il fratello? (Heidenreich 2004, 175). [[…] but what’s his brother’s name?] Wieso eigentlich mit dem Zug? (Heidenreich 2002, 36). Ma perché mai in treno? (Heidenreich 2004, 39). [But why were you going by train, then?] Warum bist du eigentlich ausgerechnet nach Oslo versetzt worden […]? (Heidenreich 2002, 35). Ma perché con tutti i posti che c’erano ti hanno mandato proprio a Oslo […]? (Heidenreich 2004, 38). [But why did they send you to Oslo out of all the places that were available?

In the last example, the translator explicitly adds facts contained only implicitly in the original text; in this way, he can give more importance to what is asked for. The examples examined above show that the presence of eigentlich in the utterance weighs more heavily on the listener (or reader) than the presence of denn: Belli (Heidenreich 2004) nearly always tries to translate it in some way, even if he has no fixed expression at his disposal (rendered twice with per caso “by chance”, once with a simple ma “but”, and twice with ma + intensifier), whereas denn often remains untranslated. This more incisive presence of eigentlich in the sentence can be attributed to its syllable structure (it is a heavy, i.e., long, word compared to the monosyllabic denn, which can be reduced in informal spoken language even to a clitic, as in Wo is’n der Wasserfall? instead of Wo ist denn der Wasserfall?) as well as to its stronger pragmatic function. Whereas denn signals the thematic relation between what is asked and the context, eigentlich does more than that: it connects what is asked for to the context, putting it into the foreground and moving the context to the background.

Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German

481

3.3. Überhaupt 3.3.1. Fundamental meaning㻌 㻌 In the same way as we traced the functions of denn and eigentlich to their fundamental meaning in the previous sections, we can analyse the etymology of überhaupt in order to understand its function. Überhaupt derives from Middle High German über + houbet, “over the head(s)”; this compound has its origins in ranching (“without counting the single heads”, Pfeifer 2013), from which its global meaning, “looking at everything as a whole” is derived.㻌 㻌 3.3.2. Homonymy㻌 㻌 As a KP, überhaupt can also be placed in the Vorfeld: Und überhaupt war das Essen sehr gut. (Heidenreich 2002, 29). [And anyway, the food was excellent.]

This use is not very frequent; normally we only find überhaupt together with a negation (überhaupt nicht / nichts, “not / nothing at all”, etc.). As an AP, it is typical for questions. It “casts doubts on a basic assumption” (Durrell 1992, 142) made before; the überhaupt-question furnishes a kind of pre-condition (Métrich and Faucher 2009, 867) for what represents the topic, as in Heidenreich (2002, 293), where Gisela, thinking about her father and her mother, asks herself: Aber - gäbe es mich überhaupt ohne diese Vergangenheit? [But - would I exist without this past?]

In wh-questions, it increases the vividness of the question (BBAW 2013); for Paul (2002, 1041), the speaker signals to the hearer that (s)he really does not know anything about what (s)he is asking, as in Heidenreich (2002, 123): Was macht Michael bei deiner Mutter, was wird hier überhaupt gespielt? [What’s Michael doing at your mother’s, what’s going on here at all?]

Sometimes, especially in rhetorical questions, it signals prejudice; indeed, for Bianco et al. (2008), it may underline a problem and, at times, may sound aggressive, as in Heidenreich (2002, 91):

482

Chapter Nineteen Hören Sie, meine Mutter ist achtzig Jahre alt. […] Wie kommen Sie überhaupt auf den Namen meiner Mutter […]? [Listen, my mother is eighty years old. […] (But …) how did my mother’s name occur to you […]?]

㻌 It is controversial to consider überhaupt an AP: some authors include it in their AP list, others do not. For example, Helbig and Kötz (1981, 16) list it as an illocution indicator, but it is absent from the rest of the book in which they describe individual APs. For Zifonun et al. (1997, 1209), it is a peripheral AP like eigentlich.㻌 3.3.3. Translatability into Italian㻌 㻌 The DIT (2002, 995) suggests ma, poi, in sostanza, “but, then, in essence”, as possible translations of the AP überhaupt into Italian; Bianco et al. (2008) propose insomma, “well, really.” Trying to paraphrase überhaupt in German, the DWDS (BBAW 2013) suggests aufs Ganze gesehen, insgesamt (gesehen), “all in all”, for the KP überhaupt, and denn and eigentlich for the AP überhaupt. Whereas Belli (Heidenreich 2004) introduces his Italian version of the eigentlich-question often by an adversative ma, “but” (according to its function of expressing the opposition between what is asked for as the foreground of the question and the context as the background of the question), he connects his version of überhaupt-questions (as far as he translates the AP) to what is said before by an additive e, “and” (once by a simple e and three times by e + intensifier): Warum gingen sie überhaupt zu Fuß? Der Weg zur Kaserne war weit. (Heidenreich 2002, 89). E perché andavano a piedi? La caserma era lontana. (Heidenreich 2004, 95). [And why did they go on foot? It was a long way to the barracks.] Wie kommen Sie überhaupt auf den Namen meiner Mutter […]? (Heidenreich 2002, 91). E poi chi le ha fatto il nome di madre? (Heidenreich 2004, 97). [And who told you my mother’s name […], then?] Was macht Michael bei deiner Mutter, was wird hier überhaupt gespielt? (Heidenreich 2002, 123). Cosa fa Michael da tua madre e cosa diavolo sta succedendo? (Heidenreich 2004, 128). [What’s Michael doing at your mother’s and what the Hell is happening here?]

Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German

483

Conveying überhaupt into Italian in this way is consistent with the function of überhaupt insofar as it adds what is asked for as a new element to the context. Indeed, contrary to denn and eigentlich, überhaupt does not anchor what is asked for in the context. Nevertheless, Belli (Heidenreich 2004) does not translate it at all in half of the cases in which it occurs in the original version, acknowledging the difficulty of rendering its full pragmatic function into this Romance language. In one passage of Heidenreich’s novel (Heidenreich 2002), the translator is forced to express the modulation of the speaker’s interest when two questions – the first containing eigentlich and the second überhaupt – follow one right after the other: 㻌 Warum bist du eigentlich zum “Lebensborn”, warum bist du überhaupt aus Tölz wieder weggegangen? (Heidenreich 2002, 117). Perché sei andato a lavorare per il Lebensborn, e perché mai te ne sei andata di nuovo via da Tölz? (Heidenreich 2004, 123). [Why did you go to work for the Lebensborn, and why did you ever leave Tölz again?]

The modulation of the speaker’s interest in getting an answer expressed by eigentlich (high interest) and überhaupt (even higher as expressed by eigentlich) is realised in Belli’s translation into Italian by intensifying the interrogative element from a simple perché, “why”, to e perché mai, “and why ever.”㻌

3.4. Denn vs. eigentlich vs. überhaupt and denn + eigentlich + überhaupt We have closely examined denn, eigentlich and überhaupt in order to better understand the subtle differences between: Sag mal, wer ist denn unser Klassenlehrer? Sag mal, wer ist eigentlich unser Klassenlehrer? Sag mal, wer ist überhaupt unser Klassenlehrer?

Now we can say that both denn and eigentlich anchor the questions in the situational context (“verankern”, Métrich and Faucher 2009, 292). Denn signals this anchoring and eigentlich focuses the question as the speaker’s only relevant interest that arises in that moment, whereas überhaupt increases the vividness of this interest even more than eigentlich.

484

Chapter Nineteen

As we have seen, these functions can be traced to the fundamental meaning of each AP: what is asked for is tagged by denn (cf. “then”) as connected to the temporal-causal-conditional-situational context, by eigentlich (cf. “own”) as the essential part of this context and by überhaupt (cf. “over the heads”) as important in general, independently of the context. If we put them in a hierarchy, we can summarise that denn is less specific than eigentlich and überhaupt, and the latter is the most specific one with which the speaker can attach the greatest importance to what (s)he asks for. According to Zifonun et al. (1997, 1212), when APs are combined, the less specific or important one precedes the more specific or important one; thus, we can find denn eigentlich, denn überhaupt, eigentlich überhaupt and even denn eigentlich überhaupt. Their combination further increases the pragmatic effect of underlining the speaker’s interest in obtaining the answer: Sag mal, wer ist denn eigentlich überhaupt unser Klassenlehrer?

The comparison between denn, eigentlich and überhaupt in section 3 illustrated how their lack of lexical meaning poses problems of interpretation and translation. This is a challenge even for professional translators, as we have seen. Whereas, on the one hand, it is difficult to replace APs when translating from German into languages which do not contain this category of words, on the other hand, inexperienced translators translating into German often forget to use them in German: they do not consider that using them makes oral language simply more natural and fluent.

4. Teaching APs If translators have problems using APs, learners at school will avoid them all the more. If we consider that learners generally use only a few modifiers (see Gauci, this volume), using modifiers without a lexical meaning is even more difficult for them. Their coursebooks could, thus, be a potentially valuable source of assistance with this difficult area of language learning. Therefore, it is important to examine just how textbooks used in schools treat APs.

Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German

485

4.1. APs in textbooks Normally, the words encountered in the first textbook with which learners begin to study a foreign language are the most frequent and important words. As denn, eigentlich and überhaupt are among these,15 we can already find them in textbooks for beginners, although unaccompanied by relevant explanations: one of the first APs learners encounter is denn. The textbook by Catani et al. (2013a), for example, introduces it in the very first lesson (pp. 25-26), together with the AP mal (Sag mal, was schenkst du denn deinem Vater? “Tell me, what are you giving your father as a present?” and Was macht ihr denn? “What are you doing?”). Further down (p. 40), we can find another dialogue containing not only mal, but also eigentlich and the APs etwa, doch, schon – all together in one dialogue. What happens, then, when students come home and they cannot remember what denn, eigentlich and the other particles mean? If they look them up in the alphabetical glossary at the end of the book, they can find the lexical equivalents indicated in Table 2. However, these are the equivalents of denn, eigentlich, etc., when they do not function as APs, but rather as members of other particle classes (denn as conjunction, eigentlich as commenting particle, and so on). With these lexical indications, it is impossible to make sense of the textbook dialogues we have just seen (“Tell times [sic], what do you give because [sic] to your father as a present?” and “What are you doing because [sic]?”). 㻌 㻌 TABLE 2. Lexical equivalents indicated in the glossary of Catani et al. (2013a, W1-W8), accompanied by an English gloss

denn eigentlich etwa doch schon mal

Italian equivalent (Catani et al.) perché in realtà circa eppure; certo, sicuro già (in matematica) per

Gloss of the Italian term because really about, approximately (and) yet; certainly already (in mathematics) times

Furthermore, checking all grammar chapters that deal with interrogative clauses and imperative clauses in Catani et al. (2013a, 2013b) as well as in the textbooks written by Motta (2004a, 2004b) and Vanni and Delor (2010a, 2010b), we cannot find any mention of denn, mal, eigentlich or other APs.

486

Chapter Nineteen

Fortunately, not all textbooks pretend that APs do not exist at all. Curci, Bente Pieper, and Roth (2005a, 2005b, 2005c, 2005d) and Weerning (2007a, 2007b) correlate APs to grammar topics, giving a short, explicit overview of their use in certain speech acts, thus preferring a deductive approach. Both authors offer a practice session, too: besides traditional exercise types such as “choose the best AP to complete the sentence” or “express the meaning of the sentence in another way using the appropriate AP” in reference to given sentences or passages, they also include awareness-raising exercises (such as one in Curci, Bente Pieper, and Roth 2007c, L24 in which students have to listen to utterances and indicate whether they perceive them as intrusive questions, requests, expressions of surprise or other illocution types). In sum, we find denn, eigentlich, überhaupt and other APs in all textbooks. We see them in dialogues already in the first lessons, further on in authentic texts and even in many exercise texts: but their presence is not emphasised. Moreover, those textbooks which do treat APs in an explicit way do not provide a systematic approach to them. So, for example, we can locate them in Weerning’s phrase banks dedicated to individual speech acts at the end of every lesson (Weerning 2007a, 2007b): already in lesson 4 (2007a, AB29), we find mal in sag mal, which functions as a signal that you are going to ask something (“[s]egnalare che stai per chiedere qualcosa”), but there are no exercises and no hints regarding mal in either the word list (AB28) or the grammar explanation (AB29-30) at the end of the lesson, even though one of the grammar topics discussed is Adverbs (“L'avverbio”); only in lesson 10 (AB61), pragmatic particles (“Le particelle communicative”), is there an explicit grammar topic which explains the pragmatic nature of mal, denn and other APs.㻌

4.2. Pedagogical suggestions How a systematic approach to APs in textbooks should be realised is controversial. APs can be introduced in the exposure phase not only by means of more or less constructed textbook dialogues (such as the one mentioned in Catani et al. 2013a, 40), but also through authentic texts of scripted dialogic language, such as comics, TV movies or spontaneously produced informal Internet communications (Heggelund 2001), like chatting, blogging or emailing. All the textbooks analysed use these kinds of text sources, but without pointing out the presence of APs and their natural integration in dialogic language. Thus, they forgo the chance to spur students’ curiosity towards a word category which their own language does not possess.

Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German

487

Opting for an inductive method, textbooks could help students discover the phenomena on their own, thus improving their achievements: discovering something by oneself increases awareness and effectiveness. The textbooks considered in this study use partly inductive approaches, too, but not in reference to pragmatic issues. Cognitive explanations  such as those given by Curci, Bente Pieper, and Roth (2005a, 2005b, 2005c, 2005d) and Weerning (2007a, 2007b)  should come after the phase of habituation, sensitisation and raising students’ curiosity (“Phase der Gewöhnung, Sensibilisierung und NeugierWeckung”, Barkowski and Krumm 2010, 241) toward the presence of APs in everyday German. These detailed explications concerning the use of particular APs and general AP features, such as their syntactic behaviour and lack of lexical meaning, should be graded. For a learner at A1-level, it is not important to understand the subtle differences between denn, eigentlich and überhaupt, but, as bumping into the AP denn is unavoidable even for a beginner of German as a FL, there is no plausible reason why (s)he should not be aware of its pragmatic function in questions even at A1-level.16 On the other hand, eigentlich and überhaupt, owing to their lower frequency, can wait to be introduced at A2- or even B1-level.17 Students, above all beginners, can also acquire an awareness of the pragmatic nature of APs through non-verbal means (Barkowski and Krumm 2010, 241). It is especially easy in Italy to make the most of miming and gestures. When, further on, students encounter the conjunction denn, they should understand that they are facing a homonym. Finally, learners have to practise APs. They can do so more or less implicitly, when they simulate dialogues and role-plays under the supervision of their teacher who can help them to add some AP to their dialogues,18 also exploiting this moment to take care of intonation, to which teachers often pay too little attention in the classroom: both APs and intonation are linked to the whole utterance. Naturally, learners can practise APs with explicit exercises as well: in order to fix a particular AP (e.g., inserting denn in a set of questions embedded in mini dialogues) or to distinguish different APs (e.g., inserting denn and mal in mini dialogues or underlining them in a given text) and homonyms belonging to the AP class and to other particle classes (e.g., reading a given text which contains both the AP denn and the conjunction denn in order to determine which category each denn belongs to). Unfortunately, the textbook exercises analysed above do not cover the whole range outlined here. As indicated by Muhr (1989, 651), exercises on APs should give a clear, pragmatic frame: the kind of communicative situation being evoked

488

Chapter Nineteen

(formal, informal, etc.) must be clear at first sight, as well as the relation between the communication partners (familiar, unfamiliar, etc.), the intended effect of the speech act (descriptive, evaluative, etc.) and the use of forms of personal address such as personal forms (du, Sie, “you”, without and with distance, etc.). Muhr’s criteria are only partly satisfied in the textbook exercises analysed: for example, in Weerning (2007a, AB73, exercise 16) and in Curci, Bente Pieper, and Roth (2005c, L24-25, exercises 1 and 4) they are, whereas in Motta (2004b, 188, exercise 6)19 the context frame is missing. These criteria also imply that isolated sentences, such as those in Curci, Bente Pieper, and Roth (2005c, L24-25, exercises 3 and 6; L31, exercise 1), should be avoided. Ultimately, it would be a desideratum that the alphabetical glossaries at the end of a lesson or the entire textbook clearly separate APs from their homonyms (as in Box 1). There is no plausible reason why learners should not know that, for example, denn in the Mittelfeld of a question is an AP, which is difficult to translate by verbal means, whereas denn introducing a main clause is a conjunction we can easily translate (“because, for”). 㻌 BOX 1. How denn could appear in an A1- or A2-textbook glossary denn 1 (modulating particle) ĺ (in interrogative clauses it indicates the speaker’s interest; often it is not necessary to translate it; it can be translated with then at the end of the sentence: Did you …, then?) denn 2 (coordinating conjunction) ĺ because

5. Conclusion Given their pragmatic importance in the German language, APs should no longer be treated as parasites, but rather as useful animals (“Nutztiere”, Dalmas 2006, 417): they optimise communication, and German native speakers use them all the time. Learners of German as a FL should know that an adequate use of APs identifies them as both competent and socially more acceptable speakers (Barkowski and Krumm 2010, 241). At the same time, teachers should be aware that failure to expose pupils to APs, intonation and other pragmatic features denies them the chance of achieving and demonstrating such competence. Without a doubt, APs are a complex domain which is difficult to describe (as we have seen in sections 1 to 3). But this does not justify hushing them up or hiding them away, like an ugly stepchild (as section 4.1. illustrated with examples). In their textbooks and classrooms, teachers can find many opportunities to highlight APs: while reading or listening to a text, simulating a dialogue, focussing on a feature of a particular text-

Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German

489

type, of a particular speech act, of a particular grammar topic or intonational cluster, or comparing two languages in a contrastive way. This brief analysis of some current Italian textbooks shows that pragmatic aspects only enter into Italian classrooms with difficulty.20 In this respect, the treatment of APs is similar not only to that of intonation in FL teaching in general, but also to that of gestures in Italian as a FL – always used by native speakers, but still without a fixed place in textbooks. Textbook authors and teachers need to change their attitude towards pragmatic aspects of interaction and integrate them systematically alongside vocabulary, grammar and phonology teaching, if they want to prepare their students to communicate effectively in a FL. Too often, they forget that the classroom probably represents the only available environment where learners can try out what using a living foreign language feels like.㻌

References Albrecht, J. 1977. “Wie übersetzt man eigentlich eigentlich?” In Aspekte der Modalpartikeln. Studien zur deutschen Abtönung, edited by H. Weydt, 19-37. Tübingen: Niemeyer. BBAW. 2013. Digitales Wörterbuch der deutschen Sprache. http://www.dwds.de. Bianco, M. T., E. Lima, S. Palermo, and L. Tonelli. 2008. “Anmerkungen zu den Abtönungspartikeln aus italienischer Sicht.” In ProGr@mm. http://hypermedia.ids-mannheim.de/call/public/gruwi.ansicht?v_typ=u&v_id=3372. —. 2010. “Die Abtönungspartikeln und ihre italienischen Entsprechungen.” In ProGr@mm. http://hypermedia.ids-mannheim.de/call/public/gruwi.ans icht?v_typ=o&v_id=3840. Boettcher, W. 2009. Grammatik verstehen I – Wort. Tübingen: Niemeyer. Burkhardt, A. 1985. “Der Gebrauch der Partikeln im gesprochenen Deutsch und im gesprochenen Italienisch.” In Gesprochenes Italienisch in Geschichte und Gegenwart, edited by G. Holtus, and E. Radtke, 236-75. Tübingen: Narr. —. 1986. Soziale Akte, Sprechakte und Textillokutionen. Tübingen: Niemeyer. Cardinaletti, A. 2007. “Für eine syntaktische Analyse der Modalpartikel.” In Gesprochene Sprache – Partikeln, edited by E. M. Thüne, and F. Ortu, 89-102. Frankfurt a/M: Lang.

490

Chapter Nineteen

Catani, C., H. Greiner, E. Pedrelli, and C. Wolffhardt. 2013a. Kurz und gut! Ein Lehrwerk für deutsche Sprache und Kultur. 1. Bologna: Zanichelli. —. 2013b. Kurz und gut! Ein Lehrwerk für deutsche Sprache und Kultur. 2. Bologna: Zanichelli. Curci, A. M., S. Bente Pieper, and S. M. Roth. 2005a. Vitamin D. Ein modulares Lehrwerk für Deutsch als Fremdsprache. 1. Ismaning: Hueber and Le Monnier. —. 2005b. Vitamin D. Arbeitsbuch. 1, Ismaning: Hueber and Le Monnier. —. 2005c. Vitamin D. Ein modulares Lehrwerk für Deutsch als Fremdsprache. 2. Ismaning: Hueber and Le Monnier. —. 2005d. Vitamin D. Arbeitsbuch. 2. Ismaning: Hueber and Le Monnier. Dahl, J. 1988. Die Abtönungspartikeln im Deutschen. Heidelberg: Groos. Dalmas, M. 2006. “Modalfunktion als Mittel zur Textgestaltung.” In Grammatische Untersuchungen. Analysen und Reflexionen. Gisela Zifonun zum 60. Geburtstag, edited by E. Breindl, L. Gunkel, and B. Strecker, 417-30. Tübingen: Narr. Diewald, G. 2007. “Abtönungspartikel.” In Deutsche Wortarten, edited by L. Hoffmann, 117-41. Berlin: Mouton de Gruyter. DIT. Il dizionario Tedesco Italiano – Italiano Tedesco. 2002. Turin: Paravia. Duden. Die Grammatik. 2006. Mannheim, Leipzig, Lucern: DudenVerlag. Durrell, M. 1992. Using German. A Guide to Contemporary Usage. Cambridge: Cambridge University Press. Dürscheid, C. 2000. Syntax. Grundlagen und Theorien. Göttingen: Vandenhoek and Ruprecht. E-Großwörterbuch Deutsch als Fremdsprache. 2003. Berlin: Langenscheidt (CD-ROM). Eisenberg, P. 2006. Grundriss der deutschen Grammatik. Das Wort. Stuttgart: Metzler. Engel, U. 2004. Deutsche Grammatik. Neubearbeitung. Munich: Iudicium. Gauci, P. This volume. Glaboniat, M., M. Müller, P. Rusch, H. Schmitz, and L. Wertenschlag. 2005. Profile deutsch. Lernbestimmmungen. Kannbeschreibungen. Kommunikative Mittel. Niveau A1-A2. B1-B2. C1-C2. Berlin: Langenscheidt (CD-ROM). Heggelund, K. T. 2001. “Zur Bedeutung der deutschen Modalpartikeln in Gesprächen unter besonderer Berücksichtigung der Sprechakttheorie und der DaF-Perspektive.” Linguistik Online 9, 2/01. http://www.linguistik-online.de/9_01/Heggelund.html.

Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German

491

Heidenreich, G. 2002. Das endlose Jahr. Die langsame Entdeckung der eigenen Biographie – ein Lebensborn-Schicksal. Frankfurt a/M: Fischer Taschenbuch. —. 2004. In nome della razza ariana. Il viaggio di una donna alla ricerca della propria identità. Translated by Marco Belli. Milan: Baldini Castoldi Dalai. Helbig, G., and J. Buscha. 1980. Deutsche Grammatik. Ein Handbuch für den Deutschunterricht. Leipzig: Enzyklopädie. Helbig, G., and W. Kötz. 1981. Die Partikeln. Leipzig: Enzyklopädie. Helling, C. 1983. Die deutsche Modalpartikel im Übersetzungsvergleich. Udine: Del Bianco. Hentschel, E. 1986. Funktion und Geschichte deutscher Partikeln. Ja, doch, halt und eben. Tübingen: Niemeyer. Hentschel, E., and H. Weydt. 2002. Handbuch der deutschen Grammatik. Berlin: Mouton de Gruyter. Hoffmann, L. 2013. Deutsche Grammatik. Grundlagen für Lehrerausbildung, Schule, Deutsch als Zweitsprache und Deutsch als Fremdsprache. Berlin: Schmidt. Kegel, J. 2006. “Zu verwendeten Partikeln als pragmatisches Mittel der Rezeptionslenkung.” In "Wollt ihr den totalen Krieg?" Eine semiotische und linguistische Gesamtanalyse der Rede Goebbels’ im Berliner Sportpalast am 18. Februar 1943, edited by J. Kegel, 430-44. Tübingen: Niemeyer. Köller, W. 2004. Perspektivität und Sprache: zur Struktur von Objektivierungsformen in Bildern, im Denken und in der Sprache. Berlin: Mouton de Gruyter. Leech, G. N. 1983. Principles of Pragmatics. London: Longman. Linke, A., M. Nussbaumer, and P. R. Portmann. 1996. Studienbuch Linguistik. Tübingen: Niemeyer. Métrich, R., and E. Faucher. 2009. Wörterbuch deutscher Partikeln. Unter Berücksichtigung ihrer französischen Äquivalente. Berlin: Mouton de Gruyter. Motta, G. 2004a. Direkt. Ein Lehrwerk für Deutsch als Fremdsprache. Kursbuch + Arbeitsbuch. 1. Turin: Loescher. —. 2004b. Direkt. Ein Lehrwerk für Deutsch als Fremdsprache. Kursbuch + Arbeitsbuch. 2. Turin: Loescher. Muhr, R. 1989. “Zur Didaktik der Modalpartikel im Unterricht Deutsch als Fremdsprache.” In Sprechen mit Partikeln, edited by H. Weydt, 64560. Berlin: Mouton de Gruyter.

492

Chapter Nineteen

Ortu, F. 2007. “Einleitung in die Beiträge der Arbeitsgruppe Partikeln.” In Gesprochene Sprache – Partikeln, edited by E. M. Thüne, and F. Ortu, 83-88. Frankfurt a/M: Lang. Paul, H. 2002. Deutsches Wörterbuch. Tübingen: Niemeyer. Pfeifer, W. 2013. Etymologisches Wörterbuch des Deutschen. s.v. “Überhaupt.” http://www.dwds.de/?view=1&qu=%C3%BCberhaupt. (Digital and regularly updated version as a component of BBAW 2013 of: Pfeifer, W. 1995. Etymologisches Wörterbuch des Deutschen. Munich: DTV). Reiners, L. 1943. Stilkunst. Ein Lehrbuch deutscher Prosa. Munich: Beck. Universität Leipzig. 2001. “Rangliste der deutschen Wörter.” WortschatzPortal. http://wortschatz.uni-leipzig.de/Papers/top1000de.txt. Vanni, A., and R. Delor. 2010a. Stimmt! Deutsche Sprache und Kultur für junge Leute. Kursbuch + Arbeitsbuch + Libro Attivo. 1. Milan: Pearson. —. 2010b. Stimmt! Deutsche Sprache und Kultur für junge Leute. Kursbuch + Arbeitsbuch + Libro Attivo. 2. Milan: Pearson. Weerning, M. 2007a. Basis Deutsch A1+. Lehrwerk für Deutsch als Fremdsprache. Kursbuch und Arbeitsbuch. Florence: Le Monnier and Hueber. —. 2007b. Basis Deutsch A2+. Lehrwerk für Deutsch als Fremdsprache. Kursbuch und Arbeitsbuch. Florence: Le Monnier and Hueber. Weinrich, H. 1993. Textgrammatik der deutschen Sprache. Mannheim: Dudenverlag. Weydt, H. 1969. Abtönungspartikel. Die deutschen Modalwörter und ihre französischen Entsprechungen. Bad Homburg: Gehlen. —. 1977. “Ungelöst und strittig.” In Aspekte der Modalpartikeln, edited by H. Weydt, 217-25. Tübingen: Niemeyer. Zifonun, G., L. Hoffmann, and B. Strecker, et al. 1997. Grammatik der deutschen Sprache, Band 2. Berlin: Mouton de Gruyter.

Notes 1

All English glosses and translation are provided by the author. For Burkhardt (1985, 266), Italian has at most two Abtönungspartikeln: poi and mai. He classifies other words, such as proprio, davvero, etc. in a different way (see his schema on page 267). For him, only ma, be’, eh?, vero? and sometimes magari substitute for German Abtönungspartikeln in Italian. For Bianco et al. (2010), only mica assumes the probable function of an Abtönungspartikel. 3 The term Abtönungspartikel was coined by Weydt (1969), who began to publish his ground-breaking studies of pragmatic particles in the 1960s. Ortu (2007, 83) calls this moment the U-turn after Weydt (“Weydt-Wende”). 2

Denn, eigentlich, überhaupt: Three “Pragmatic Particles” in German

493

4 “Alle diese Flickwörter wimmeln wie Läuse im Pelz unserer Sprache herum” [All these patch words are teeming in our language like lice in our fur] (Reiners 1943, 340). 5 The definition of the term particle is controversial; in a broad sense, all invariable, i.e., uninflected words are particles. 6 We can compare the German sentence to a bridge with two pillars, one on the left (i.e., the beginning of the sentence) and the other on the right (i.e., the end of the sentence). The left pillar is made up of the subordinating conjunction or, if there is none, the finite verb; the right one consists of other verbal forms such as the infinitive, past participle, separable prefixes (if any) and – if the sentence is introduced by a subordinating conjunction – the finite verb. In most types of sentences, we can find at most one clause constituent in front of the left pillar; this area is called the Vorfeld. Most of the elements of the sentence are placed between the two pillars, in the area called the Mittelfeld. 7 Unfortunately, some authors use the term Modalpartikeln for APs, which produces considerable terminological confusion (see Table 1). 8 That is one of the reasons why some linguists call them nichtsatzgliedfähig, “incapable of being a clause constituent”, besides the fact that they cannot be lexical heads like nouns or syntactic heads like prepositions. 9 What counts, however, is the illocution, not the formal clause type. A request expressed by an interrogative clause can contain mal (Kannst / Würdest du mal kommen? “Can / would you come?” or Komm mal! “Come (here)!”), in the same way that we find denn in a question expressed as an exclamation (Ist das Wetter denn nicht herrlich! “Isn’t the weather gorgeous, then!” or Ist das Wetter denn nicht herrlich? “Isn’t the weather gorgeous, then?”) or in an indirect question (Er fragte mich, wie ich denn heiße. “He asked me what my name is.”). 10 ['] precedes the stressed syllable. 11 Let us keep in mind that a pragmatic function is also a genuine grammatical function: “General pragmatics is a set of strategies and principles for achieving success in communication by the use of grammar. Grammar is functionally adapted to the extent that it possesses properties which facilitate the operation of pragmatic principles” (Leech 1983, 76). 12 It is controversial whether APs have lost their original (lexical) meaning or not. For Cardinaletti (2007, 100), they have no lexical meaning (“tragen keine lexikalische Bedeutung”). Indeed, Diewald (2007, 127) calls them synsemantic words and Dalmas (2006, 417) function words. For Linke, Nussbaumer and Portmann (1996, 273), semantic dimensions are nearly absent. Heggelund (2001) insists that APs are not completely lacking in meaning: they are only poor in meaning, or they have a vague meaning. 13 Used in this way, ma in Italian and but in English are also no longer syntactic connectors, but pragmatic connectors. 14 German grammar sometimes distinguishes adjunctions from prepositions: both introduce a noun clause but, contrary to prepositions, adjunctions do not determine the case of the noun clause. 15 According to a study ranking the frequency of German words (Universität Leipzig, 2001), denn occupies position 185 (near etwas, “something”), überhaupt

494

Chapter Nineteen

position 452 (near Oktober, “October”) and eigentlich position 458 (near vielleicht, “perhaps”). This study, however, does not make distinctions between homonyms in different word classes. 16 This claim is controversial. For Glaboniat et al. (2005), denn should be part of the receptive vocabulary of A2-students and of the productive vocabulary of B1students. 17 For Glaboniat et al. (2005), eigentlich should be part of the receptive vocabulary of A2-students and of the productive one of B2-students, and überhaupt part of the receptive vocabulary of B1-students and of the productive one of B2-students. Helbig and Kötz (1981, 47) also suggest distinguishing highly frequent APs, which students should use productively and receptively very early, from less frequent APs that they should know only receptively. Contrary to this point of view, Muhr (1989, 656) is convinced that APs should not be taught before the intermediate level (i.e., B-level). 18 Muhr (1989, 656) excludes an imitative acquisition of APs by beginners whose use of them could soon get out of control. 19 In this question-answer exercise, one student must formulate questions which are all structured in the same way (Sag mal, was macht eigentlich ein [Berufsbezeichnung]? “Tell me, what does a(n) [job title] do?”), expecting the answer of the second student. The context frame is only clear if the exercise is seen in relation to a previous text about professions. 20 Looking at the summaries of modern textbooks in use in Italy, speech acts have their place beside vocabulary and grammatical topics. At first sight, we could interpret that as a tangible sign that pragmatics has entered textbooks. But what is effectively dedicated to pragmatics inside the individual lessons is quite limited: teaching pragmatics should be more than a page with a phrase bank or an illustrative dialogue.

CHAPTER TWENTY INSTRUCTIONAL ADVANTAGES OF A PRAGMATIC ACCOUNT OF MOOD DISTRIBUTION IN SPANISH COMPLEMENTS PATXI LASKURAIN-IBARLUZEA

1. Introduction In Spanish, the indicative mood (a verb form which makes a statement or asks a question) can be used in independent clauses. The subjunctive mood is found primarily in dependent clauses, but of course the indicative can occur there as well. We can distinguish two basic types of subordinate or dependent clauses, by examining the context where the indicative and subjunctive alternation occurs. On the one hand, we have dependent clauses that complement a predicate (with an argument in a specific semantic role), as in examples in (1). On the other hand, we have dependent clauses that modify a predicate. In this group we distinguish clauses that modify a verb (adverbial clauses), as in examples in (2), from those that modify a noun (adjectival clauses), as in examples in (3): (1)

a. No pienso que él viva aquí. not think+1SG that-COMP he live+3SG+SUBJ here [I don’t think he lives here.] b. Pienso que él vive aquí. think+1SG that-COMP he live+3SG+IND here [I don’t think he lives here.]

(2)

a. Juan vendrá cuando tú llames. Juan come+3SG+FUT when you call+2SG+SUBJ [Juan will come when you call.]

Chapter Twenty

496

b. Juan vino cuando tú llamaste. Juan come+3SG+PRET when you call+2SG+PRET+ IND [Juan came when you called.] (3)

a. Busco un secretario que hable inglés. look for+1SG a secretary that-RP speak+3SG+SUBJ English [I’m looking for a secretary that speaks English.] b. Busco un secretario que habla inglés. look for+1SG a secretary that-RP speak+3SG+IND English [I’m looking for a secretary that does speak English.]

The account of mood selection in adverbial and adjectival clauses has been straightforward. The verb in an adverbial clause will be in the subjunctive if the action/state being represented is posterior with respect to the action/state represented by the governing verb, as in example (2a) above. The verb will be in the indicative if the action/state being represented is viewed as habitual or having been completed with respect to the action/state represented by the governing verb, as in example (2b). The verb in an adjectival clause, on the other hand, will be in the subjunctive if the noun or pronoun modified is negated, nonexistent or evokes a nonspecific referent, as in example (3a). The verb will be in the indicative if the noun modified evokes a specific discourse referent, as in (3b). The account and class instruction of mood selection in complement clauses, on the contrary, has been a highly controversial debate. There is an overwhelming tendency among university-level textbooks of beginner and intermediate Spanish to introduce the indicative mood in dependent clauses as the ‘default’ one, thus focusing the instruction on mood selection on explaining when to use the subjunctive. However, the vast majority of explanations are predominantly semantic in nature, in the sense that they tend to advocate that the semantic nature of the governing predicate determines mood in the verb of the dependent clause (Sandstedt and Kite 2014; Guzmán et al. 2013; Pellettieri et al. 2011; Blanco and Colbert 2009). Moreover, these explanations do not tend to focus on what the subjunctive means but just point out the fact that it is frequently associated (but not always) to expressions of doubt, non-reality, emotions or desires, as illustrated in the following examples (4)-(7) from Sandstedt and Kite (2014, 155-156): (4)

Dudo que Paco ahorre su dinero. doubt+1SG that-COMP Paco save+3SG+SUBJ his money [I doubt hat Paco would save his money.] (‘doubt’ subjunctive)

Advantages of a Pragmatic Account of Mood Distribution (5)

Quieren que recemos want+3PL that-COMP pray+1PL+SUBJ [They want us to pray for him.]

497

por él. for he (‘desire’ subjunctive)

(6)

Manuel niega que Don Paco sea Manuel deny+3SG that-COMP Mr. Paco be+3SG+SUBJ un hombre generoso. a man generous [Manuel denies that Don Mario is a generous man.] (‘unreality’ subjunctive)

(7)

Lamento mucho que perdamos. regret+1SG much that-COMP lose+1PL+SUBJ [I very much regret that we would lose.] (‘emotion’ subjunctive)

As a result, university-level Spanish textbooks present students with long lists of verbs and expressions (such as certain impersonal expressions) that will trigger the use of the subjunctive in the complement. As a consequence, students are often even encouraged to memorize the various cases covered by the use of the subjunctive in Spanish complements with the help of acronyms such as the popular and widely used WEIRDO. WEIRDO is a mnemonic device used by instructors that stands for ‘wish, emotion, influence, request/recommendation, doubt/denial’ and ‘ojalá’ (‘I wish’). If the content of the governing predicate expresses or conveys one of those meanings, then the subjunctive mood is chosen in the verb in the complement; if the meaning of the governing predicate does not express any of the WEIRDO meanings, then the indicative mood should be chosen. However, not only is this type of instruction merely a description instead of an explanation, but it is also obvious that the sole memorization of lists limited to formal categories is not the procedure by which people learn to use a language effectively. In any case, and more importantly, the notions of doubt, desire, non-reality, and feeling are not provided by the subjunctive mood in the examples above, but by the meaning of the governing predicate itself. This chapter argues that the selection of one mood over the other is determined by what the speaker intentions are, as reflected by the meaning of the main ‘matrix’ in a given discourse context. The use of the term ‘matrix’ instead of predicate has to do with the different meanings that governing predicates can possess in different communicative contexts, and it has some clear pedagogical advantages, since it forces students to go

498

Chapter Twenty

from trying to identify simple lexical items as the members of a list, to the interpretation of the meaning of those elements in the concrete communicative context in which they are produced. The main claim is that the indicative mood ‘pragmatically asserts’ the content of the proposition where it is contained. The subjunctive, on the other hand, is used when the content of the proposition where it is contained is not asserted (MejíasBikandi 1994, 1998). This generalization provides an elegant account for mood variability, since it recognizes the speaker’s ability to qualify the canonical communicative use of the governing matrix by means of a very specific communicative intent. In instructional terms, the assertion vs. non-assertion meanings of the indicative and the subjunctive help students of Spanish language understand any use of the indicative and subjunctive forms, enabling them to become more aware of the uses of mood in Spanish by using their own L1 language (whatever it may be) as a model. Furthermore, the conceptual distinction between pragmatic assertion and non-assertion can also be extended to account for mood selection in adjectival and adverbial subordinate clauses. A significant number of the examples used in this chapter to illustrate the points under discussion were taken from leading newspapers and news outlets in Spain (Diario Vasco, El Correo, El Periódico de Catalunya, Cinco Días, El País, El Mundo, and ANSA). The examples were extracted from all sections of these newspapers and news outlets, and all of them were published in between 2008 and 2012. These examples are intended to be interpreted as illustrative of the points being discussed, and were chosen because they reflect contextualized instances of different mood selections in Spanish complements. The rest of this chapter is organized as follows: in section 3, the leading accounts of mood distribution in Spanish complements are presented. Drawing on Mejías-Bikandi’s (1994, 1998) account, section 4 introduces a simplified pragmatic explanation of mood distribution in Spanish complements and deals with its application in accounting for cases of mood variability in complements. Section 5 extends the analysis of mood distribution in complements to mood selection in adjectival and adverbial subordinate clauses. And finally, section 6 deals with the primary objective of this paper: the instructional advantages of the proposed pragmatic account of mood distribution in Spanish complements, and serves as a summary of this chapter.

Advantages of a Pragmatic Account of Mood Distribution

499

2. Theoretical background: Mood distribution in Spanish complements 2.1. Terrell and Hooper’s semantic account of mood distribution in Spanish complements In their seminal work, Terrell and Hooper (1974) provide a semantic generalization of the role of presupposition and assertion in the selection of mood in Spanish complements, proposing a nearly perfect correlation between the semantic notion of assertion and the indicative mood, and non-assertion and the subjunctive mood. This account draws the distinction between presupposition and assertion in terms of the content or truth-conditions of the proposition expressed. Terrell and Hooper (1974) establish six categories of sentences based on the semantic notions of assertion and presupposition. The six predicate categories (see Table 1 below) are divided into three groups. The first group is composed of predicates that take assertions as complements (assertive predicates and report predicates). The second group is composed of factive predicates that take presupposed complements (mental act predicates and comment predicates). And the third group includes those predicates that take complements that are neither asserted nor presupposed (doubt predicates and volition and command predicates). TABLE 1–Taxonomy of predicates according to Terrell and Hooper (1974) Semantic notion a) Assertion

b) Presupposition

c) Neither

Class of predicate (i) Assertion (asegurar ‘assert’, pensar ‘think’, creer ‘believe’…) (ii) Report (decir ‘say’, contar ‘tell’, explicar ‘explain’…) (iii) Mental Act (saber ‘know’, ver ‘see’, darse cuenta ‘realize’…) (iv) Comment (lamentar ‘regret’, alegrarse ‘be happy’,…) (v) Doubt (dudar ‘doubt’, negar ‘deny’, ser probable ‘be probable’…) (vi) Volition (desear ‘wish’, sugerir ‘suggest’, ordenar ‘to command’…)

According to Terrell and Hooper’s (1974) generalization, complements that are assertions, that is, complements of assertive and report predicates,

Chapter Twenty

500

such as the examples in (8), carry the indicative mood in the complement. Nonassertive complements, that is, presupposed complements as in (9a), and complements that are neither presupposed nor asserted, i.e., complements to nonassertive predicates and complements to volitional predicates, such as the examples in (9b) and (9c), respectively, all carry the subjunctive mood.1 (8)

a. Creo que Juan viene. believe+1SG that-COMP Juan come+3SG+IND [I think that Juan is coming.] b. Digo que Juan viene. say+1SG that-COMP Juan come+3SG+IND [I say that Juan is coming.]

(9)

a. Lamento que Juan venga. regret+1SG that-COMP Juan come+3SG+SUBJ [I regret that Juan is coming.] b. Dudo que Juan venga. doubt+1SG that-COMP Juan come+3SG+SUBJ [I doubt that Juan is coming.] c. Ordeno que Juan order+1SG that-COMP Juan [I order that Juan come.]

venga. come+3SG+SUBJ

However, the correlation established in Terrell and Hooper’s (1974) generalization between the semantic notion of presupposition and the subjunctive mood cannot predict mood distribution in the complement of some factive predicates. Note, for instance, that although the complement to the factive matrices of the mental act is semantically presupposed, it nevertheless carries an indicative verb form. Consider examples (10) and (11): (10) Ella sabe que tenemos un examen hoy. she know+3SG that-COMP have+1PL+IND a exam today [She knows that we had an exam today.] (11) Ella se dio cuenta de que tenemos un examen she realize+PA+3SG of that-COMP have+1PL+IND a exam hoy. today [She realized that we had an exam today.]

Advantages of a Pragmatic Account of Mood Distribution

501

Unable to provide a satisfactory answer, Terrell and Hooper consider this type of matrix to be an exception to their generalization that asserted complements are in the indicative while non-asserted complements are in the subjunctive (Terrell and Hooper 1974, 488). Similarly, the complement of the factive predicate el hecho de is semantically presupposed, and it should therefore take a complement carrying the subjunctive mood. Yet, el hecho de clauses are often found carrying complements in indicative.

2.2. (Non)veridicality and model shift accounts of mood distribution In a series of works, Giannakidou (1994, 1995, 1998, 2011) proposes an account of mood selection in complements based on the notion of (non)veridicality. The author posits a divide within intensional verbs (verbs that take complement clauses as arguments) based on whether at least one epistemic agent (the speaker or the subject of the main verb) is committed to the truth of the complement. If a propositional attitude verb expresses a commitment to the truth of the complement, it will be veridical and select the indicative; if it does not, it will be nonveridical and select the subjunctive. In this analysis, the divide between veridical and nonveridical propositional attitude verbs corresponds to the distinction between weak and strong intensional predicates. Giannakidou (2011, 1675) defines (non)veridicality as follows: Veridicality is a property of sentence embedding functions: such a function F is veridical if Fp entails or presupposes the truth of p. If inference to the truth of p under F is not possible, F is nonveridical. More specifically, veridical operators express certainty and an individual’s commitment to the truth of a proposition, but nonveridical expressions express uncertainty and lack of commitment.

Following a similar model, Quer (2001, 2010) proposes that embedding predicates of propositional attitude introduce specific types of models into the context. According to the author, “Mood overtly marks information about the models where the clauses are to be interpreted: mood shift signals a change of type of model for the evaluation of the proposition or property at hand” (2001, 84). Quer explains that the use of the indicative in the complement of weak intensional predicates does not entail a shift in the type of model, since these complements “take us from the epistemic model of the speaker to some individual’s epistemic model” (2001, 87). On the contrary, when using the subjunctive in the complement, “the buletic models associated to strong intensional

502

Chapter Twenty

predicates (MBul(x)) do induce model shift, because they take us from the epistemic model of the speaker to the buletic model of an individual”2 (2001, 88). Quer’s account is not substantially different from Giannakidou’s approach to mood selection in complement clauses. In the case of these clauses, which are the main focus of the present study, both Quer and Giannakidou follow a selectional notion of subjunctive. They consider the indicative as the default mood in Spanish, and both endorse a ‘selectional theory’ of mood distribution, according to which the subjunctive is ‘triggered’ by certain semantic properties of the embedding context.3 That is, the subjunctive is possible only in complement clauses of predicates that share the particular semantic characteristic of being ‘strong intensional’ predicates. In any case, Quer’s (2001, 2010) mood distribution account cannot be considered to be exclusively semantic in nature, since it acknowledges the importance of different aspects of discourse interpretation and context change to distinguish core cases of subjunctive selection from more peripheral ones. However, the main issue with Giannakidou’s and Quer’s account of mood distribution is very similar to the issue with Terrell and Hooper’s account: namely, neither Quer’s nor Giannakidous’s account can predict mood distribution in the complement of certain factive predicates. Considering that the complement of factive verbs is presupposed to be true – Giannadikou says that “all factive verbs are veridical” (2011, 1675) – factive predicates are considered weak predicates and hence they should take complements carrying the indicative. But factive comment predicates such as lamentar ‘to regret’, which presuppose the truth of the proposition in the complement, invariably requiring the subjunctive, and not the indicative. I consider that mood distribution in the complement of such factive predicates – and by extension in the complement of all predicates – can be more adequately accounted for by following a speaker-based pragmatic notion of presupposition and assertion rather than the notion of (non)veridicality4 or the notion of semantic assertion and presupposition, as in Terrell and Hooper (1974).

2.3. Pragmatic presupposition and assertion, and mood selection in Spanish complements There are numerous studies that emphasize the role of pragmatic notions such as ‘relevance’ and ‘new/old information’ in the selection of mood in Spanish complements (Lunn 1989; Lavandera 1983; Klein 1975; Guitart 1991; Mejías-Bikandi 1994, 1998). Many investigators claim that

Advantages of a Pragmatic Account of Mood Distribution

503

it is necessary in the study of presupposition and assertion to distinguish between the meaning of a sentence (semantic notion), which remains constant regardless of contextual conditions, and the information conveyed by an utterance (pragmatic notion), which depends on the mental states of the interlocutors. Along those lines, Mejías-Bikandi (1998) proposes a speaker-based pragmatic definition of presupposition and assertion based on the notions of ‘new/old information’. This allows him to maintain Terrell and Hooper’s (1974) original generalization about mood distribution in terms of assertion (indicative) vs. non-assertion (subjunctive), without having to resort to the construction of two types of presupposition and assertion. Addressing Terrell and Hooper’s inconsistency, Mejías-Bikandi accounts for mood distribution in factive predicates like saber (Mental Act) and lamentar (Comment). Comment and mental act matrices are problematic in Terrell and Hooper’s account, since both take semantically presupposed complements, but the distribution of mood in both types of complement is different; the indicative is used with mental act predicates, and the subjunctive with comment predicates. Mejías-Bikandi (1994, 1998) adopts a pragmatic generalization to account for mood distribution in Spanish complements. According to this analysis, the indicative marks pragmatic assertion, and the subjunctive marks non-assertion (presupposition; and neither presupposition nor assertion). Mejías-Bikandi argues that a proposition is pragmatically presupposed when the speaker assumes “that the information has been entertained or discussed previously by the hearer” (Mejías-Bikandi 1998, 943). His notion of pragmatic assertion, on the other hand, follows Faucounnier’s (1985) “Mental Spaces” theory, and it is initially used to explain the use of the indicative in the complement of mental act matrices. Mejías-Bikandi’s definition of assertion is pragmatic because it is based on the notion of speaker intention and defined independently of the notion of semantic presupposition: a speaker asserts a proposition P when the speaker intends to indicate that P is contained in some space R, that is, when the speaker intends to indicate that P provides information about some individual's view of reality. (Mejías-Bikandi 1994, 895).

Mejías-Bikandi explains the use of indicative in sentences with assertive matrices by arguing that the speaker denoted by the grammatical subject of the sentence is making an assertion about her representation of her view of the world. By contrast, in comment predicates the speaker has no intention of indicating that the proposition expressed by the

504

Chapter Twenty

complement is contained in any space R and thus the subjunctive is employed. According to Mejías-Bikandi, in order to predict the mood of complement clauses in Spanish, one must take into consideration what the speaker intentions are, as reflected by the meaning of the matrix clause. Thus, if the speaker intends to present a proposition P as part of some individual's view of reality, P will be asserted and the indicative mood will be used. On the contrary, if the speaker does not intend to present P as part of some individual's view of reality, P is not asserted and the subjunctive mood will be used. This generalization provides an elegant account for mood distribution in the factive comment and mental act matrices, which was problematic in Terrell and Hooper’s (1974), Giannakidou’s (1994, 1995, 1998, 2011) and Quer’s (2001, 2010) accounts of mood distribution in complements. Consider the following examples in (12) and (13): (12) Pedro se alegra de que Pedro REFL please+3SG of that-COMP [Pedro is happy that you are coming.]

vengas. come+2SG+SUBJ

(13) José se dio cuenta de que Pedro estaba José REFL realize+3SG of that-COMP Pedro be+IMP+2SG+IND Allí. there [José realized that Pedro was there.]

By using a comment matrix in (12), the intention of the speaker is not to indicate that P is true for the speaker or for any other person. The complement is assumed to be pragmatically presupposed; that is, it is assumed to belong in the ongoing discourse, and the speaker is simply commenting on the proposition expressed by the complement, which is signaled by the use of the subjunctive mood. By using a mental act matrix in (13), on the other hand, the meaning of the matrix phrase is such that the speaker’s intention when uttering it is precisely to indicate that the proposition expressed by the complement clause is true for José. The intention of the speaker is to indicate that P is contained in R(p). So the speaker is thus asserting P, in spite of the fact that P is semantically presupposed. This assertion in the complement is signaled by the use of the indicative mood. Therefore we see that the relevant presupposition in this analysis (the presupposition that triggers the use of subjunctive mood) is the pragmatic presupposition rather than the semantic one.

Advantages of a Pragmatic Account of Mood Distribution

505

3. A simplified account of mood distribution in Spanish complements In previous sections Terrell and Hooper’s (1974), Giannakidou’s (1994, 1995, 1998), Quer’s (2001, 2010) and Mejías-Bikandi’s (1994, 1998) models of mood selection in complements have been reviewed. It has been shown that Terrell and Hooper’s model is unsuccessful in accounting for mood selection in the complement to knowledge and knowledge acquisition factive predicates. Giannakidou’s and Quer’s models, in turn, have been proven to be unsuccessful in accounting for mood selection in the complements to comment factive predicates. Only Mejías-Bikandi’s (1994, 1998) model and his speaker-based notion of pragmatic assertion and presupposition has proven to adequately explain mood selection in all factive predicates. Therefore, and taking into consideration language instruction, we can account for mood distribution in Spanish complements by simplifying Mejias-Bikandi’s generalization and providing students with a binary operational value for the indicative and subjunctive moods. This operational value will be as follows: i. The speaker uses the indicative mood to represent the pragamatic assertion of the proposition where it occurs. ii. The speaker uses the subjunctive mood to represent the non-assertion of the proposition where it occurs. The generalization proposed here provides an elegant account for mood selection and use, and, being speaker-intent based, it nicely accounts for mood variability in Spanish complements.

3.1. Accounting for mood variability There are two basic types of mood variability in Spanish complements. On the one hand, we find matrices such as el hecho de ‘the fact’ that may carry both the indicative and the subjunctive in their complements. On the other hand, we find cases of the indicative mood being used by speakers in complements where the subjunctive is traditionally expected, and the subjunctive mood being used by speakers in complements where the indicative is traditionally expected.

506

Chapter Twenty

3.1.1. Mood distribution in el hecho de clauses There has been great interest in the distribution of mood in el hecho de clauses. The complement of this predicate takes both the indicative and the subjunctive mood, and numerous authors have studied this mood variability (Woehr 1975; Lipski 1978; Krakusin 1992; Jary 2009). El hecho de clauses are found playing any semantic argument role in the sentence; that is, they can be found as the complement (direct, indirect, or prepositional) of certain governing predicates and, in some cases, as the subject of a governing predicate or a copula. Consider the following examples: (14) a. Me refiero al hecho de que es REFL refer+1SG to+the fact of that-COMP be+3SG+IND importante important que estemos representados todos los partidos. [I am referring to the fact that it is important that we all the parties are represented.] (“Esperamos aumentar la representación, por lo menos, con dos concejales.” Diario Vasco, May 17, 2011: on-line edition) b. Me refiero al hecho de que la Diputación REFL refer+1SG to+the fact of that-COMP the Diputación de Bizkaia haya aprobado of Bizkaia have+3SG+SUBJ approve+PPRT el pago anual al Guggenheim Bilbao. [I am referring to the fact that the Diputación de Bizkaia has approved the payment to the Guggenheim-Bilbao.] (“Remedios vendo, que para mí no tengo.” El Correo, July 29, 2012: on-line edition) (15) a. […] las tiendas de discos […] lamentan el hecho de que the stores of records regret+3PL the fact of that-COMP cada vez más artistas venden los discos each time more artists sell+3PL+IND the records que promocionan a la salida de sus bolos. [The record stores regret the fact that more and more artists are selling the discs that they promote at the end of their performances.] (“La venta 'en caliente' de discos en directo se implanta en España.” El Periodico de Catalunya, October 4, 2010: on-line edition)

Advantages of a Pragmatic Account of Mood Distribution b. Merkel lamentó el hecho de que el Reino Merkel regret+PA+3SG the fact of that-COMP the Kingdom Unido haya vetado United have+3SG+SUBJ veto+PPRT una reforma normal del Tratado de la UE. [Merkel regretted the fact that the United Kingdom has vetoed a normal reform of the Treaty of the EU.] (“Merkel subraya que se ha dado ‘un paso muy importante’ en un largo camino.” El Periodico Mediterraneo, December 10, 2011: on-line edition) (16) a. Desde el punto de vista gastronómico sorprende el hecho de que (Maastrich) surprise+3SG the fact of that-COMP (Maastrich) es be+3SG+IND la ciudad holandesa que más estrellas Michelin atesora. [From the gastronomic point of view it surprises the fact that (Maastrich) is the Dutch city that has the most Michelin stars.] (“Arte y compras a orillas del Mosa.” Cinco Dias, August 8, 2011: p. 31) b. […] me sorprende el hecho de que una entidad que CL surprise+3SG the fact of that-COMP a entity that-RL presume de su vocación de servicio boast+3SG of its vocation of service dispense give+3SG+SUBJ ese trato discriminatorio y absurdo a un consumidor […] [I am surprised by the fact that an entity that boasts of its vocation of service gives that absurd and discriminatory treatment to a consumer.] (“Pocas facilidades para pagar recibos.” El Periodico de Catalunya, November 27, 2012: on-line edition) (17) a. Pese a ello, Defensa considera […] que está llamada a jugar un papel cada vez más importante a medio plazo, como demuestra el hecho de que tanto la OTAN como demonstrate+3SG the fact of that-COMP both the NATO as la UE animan the EU encourage+3PL+IND a sus socios a compartir capacidades militares. [Despite this, Defense considers […] that it is called to play an increasingly important role in the medium term, as evidenced by the fact that both NATO and the EU encourage their partners to share military capabilities.]

507

508

Chapter Twenty (“Un general español aspira a ‘número dos’ de la AED.” El País, Sept 20, 2012: p. 15. Comunidad Valenciana edition) b. […] el ya famoso 'Ecce Homo' de Nerja […] tiene sin embargo cierta gracia, como demuestra el hecho de que la noticia demonstrate+3SG the fact of that-COMP the news esté dando la vuelta al mundo. be+3SG+SUBJ give+PRT the tour to the world [The already famous 'Ecce Homo' of Nerja is nevertheless somewhat funny, as evidenced by the fact that the news is going around the world.] (“¡Viva el arte!” (Letters to the Editor). El Correo, September 21, 2012: on-line edition)

Examples in (14) and (15) contain sentences where el hecho de has an object semantic argument role. The governing predicate in (14a) and (14b), referirse is considered an assertive predicate, that is, a predicate that takes sentential complements in the indicative mood; the governing predicate in (15a) and (15b) lamentar, on the contrary, is considered an non-assertive predicate, that is, a predicate that takes sentential complements in the subjunctive mood. However, the examples show that both predicates will take el hecho de complements carrying the indicative in (14a) and (15a) and the subjunctive in (14b) and (15b). Examples in (16) and (17), on the other hand, contain sentences where el hecho de clauses have a subject semantic argument role: the governing predicate in (16a) and (16b), sorprender is considered a non-assertive predicate, that is, a predicate that takes sentential complements in the subjunctive mood; the governing predicate in (17a) and (17b) demostrar, on the contrary, is considered an assertive predicate, that is, a predicate that takes sentential complements in the indicative mood. But, again, the examples show that both predicates will take el hecho de subjects carrying the indicative in (16a) and (17a) and the subjunctive in (16b) and (17b). Moreover, the same copulative expression can take a propositional subject in the indicative and subjunctive. Consider examples in (18) and (19): (18) a. […] escribió hoy La Gazzetta, según la cual no sería un obstáculo el hecho de que el not be+3SG+COND an obstacle the fact of that-COMP the argentino no podrá jugar Argentinean not be able+3SG+FUT play+INF en la Liga de Campeones ‘[[…] wrote today La Gazzetta, according to which it wouldn’t be

Advantages of a Pragmatic Account of Mood Distribution

509

an obstacle the fact that the Argentinian won’t be able to play the Champions League.] (“Tévez: DT Mancini lo ve afuera del Manchester City.” ANSA Noticiero en español. November 18, 2011) b. El nuevo auto judicial señala que ha de llevarse a cabo la clausura de esas plazas de rotatorio, "para lo que no es obstáculo el hecho de que la obra not be+3SG obstacle the fact of that-COMP the work haya sido ejecutada totalmente." have+3SG+SUBJ be+PPRT execute+PPRT totally [The new court order notes that the closure of these parking spaces has to be carried out, “given that it is not an obstacle the fact that the work has been executed completely.] (“La Justicia ordena al alcalde Valladolid la clausura del aparcamiento de Portugalete.” El Mundo, September 26, 2008: p. 5. Valladolid edtion) (19) a. Para mí es importante el hecho de que for me be+3SG important the fact of that-COMP soy capaz de sentir las flores be+1SG+IND able of feel+INF the flowers [For me, it is important to me the fact that I am able to smell the flowers.] (“Ver que llega el frío, que vienen las primeras violetas...” El Periodico de Catalunya, July 6, 2012: on-line edition) b. También es importante el hecho de que tanto Berta also be+3SG important the fact of that-COMP both Berta como yo hayamos trabajado en televisión. as I have+3PL+SUBJ work+PPRT in television [The fact that both Berta and I have worked in Television is also important.] ("Los mayores pueden ayudar a superar la crisis." Cinco Dias, February 19, 2011: p. 45)

As evidenced by the examples above, it is not possible to claim that mood selection in el hecho de clauses is determined by the nature of the governing predicate or copula. However, it is possible to account for this mood variability in the complement to el hecho de clauses by following the binary operational value for the indicative and subjunctive moods proposed at the beginning of this section: 1. The speaker uses the indicative in the complement of el hecho de clauses when she intends to assert the propositional content of the

510

Chapter Twenty

complement; i.e., when her communicative intent is to indicate that the propositional content of the complement is true for the speaker or for any other person. 2. The speaker uses the subjunctive, i.e., does not assert the propositional content of the complement proposition when she intends to present the proposition as a pragmatically presupposed discourse referent; i.e., when her main communicative intent is not to indicate that the propositional content of the complement is true for the speaker or for any other person. To illustrate this generalization for mood selection in el hecho de clauses, consider the following example in (20): (20) El presidente de la Fed mostró su preocupación por las amenazas inflacionistas que sufre la economía estadounidense, así como por el estancamiento del consumo interno y por el hecho de que el sector de la vivienda “continúe debilitándose.” El hecho de que Bernanke se mostrara the fact of that-COMP Bernake REFL show+IMP+3SG+SUBJ preocupado concern+PPRT tanto por el bajo crecimiento económico, como por el comportamiento de la inflación, […]. Bernanke se mostró especialmente pesimista a la hora de evaluar el comportamiento futuro de los precios de los combustibles, cuyo crecimiento revela el hecho de que el incremento en la demanda de los the fact of that-COMP the increase in the demand of the últimos años last years no se ha visto acompañado not REFL have+3SG+IND see+PPRT accompanied por una mayor producción. [The president of the Fed showed his concern for the inflationary threats that the American economy suffers, as well as for the stagnation of the internal consumption and for the fact that the sector of the housing keeps on weakening. The fact that Bernake showed concern both for the low economic growth as well as for the behaviour of inflation, […]. Bernanke was especially pessimistic when evaluating the future behaviour of the fuel prices, which growth reveals the fact that the increase in demand in recent years has not been accompanied by an increase in production.] (“Bernanke pinta un escenario pesimista.” El Mundo, July 16, 2008: p. 35. Andalucía edition)

Advantages of a Pragmatic Account of Mood Distribution

511

The subjunctive mood is used in the sentential complement in El hecho de que Bernake se mostrara preocupado ‘the fact that Bernake showed concern’ because the propositional referent denoted by the complement, ‘Bernake is concerned’ is assumed by the speaker to be pragmatically presupposed. This is because it has been explicitly activated in the previous sentence El presidente de la Fed mostró su preocupación ‘the president of the Fed showed his concern’, and therefore the main communicative intent of the speaker is not to indicate that the propositional content of the complement is true for the speaker or for any other person. On the contrary, by using the indicative in the el hecho de que el incremento en la demanda de los últimos años no se ha visto acompañado por una mayor producción ‘the fact that the increase in demand in recent years has not been accompanied by an increase in production’, the speaker is signaling that she assumes that the referent denoted by the propositional content of the complement ‘the increase in demand in recent years has not been accompanied by an increase in production’ is not of present concern in the discourse; that is, the speaker does not assume that it is pragmatically presupposed. In this case, the speaker is actually pragmatically asserting the content of the propositions in the complement, since her main communicative intent with this utterance is to indicate that the propositional content of the complement is true for the speaker or for any other person. 3.1.2. Non-canonical cases of mood distribution in Spanish complements Regarding the second type of mood variability in Spanish complements by which speakers produce non-canonical uses of indicative and subjunctive, the claim is that speakers, with very specific communicative intents, have the ability to force the canonical communicative schema of the governing matrix and use the indicative mood in complements where the subjunctive is traditionally expected and vice-versa. Consider the following examples in (21) and (22): (21) Lamento que has estado enfermo. regret+1SG that-COMP have+2SG be+PPRT sick [I very much regret that you have been sick.] (22) El prisionero admite que viniera. the prisoner admit+3SG that-COMP come+IMP+3SG+SUBJ el inspector the inspector [The prisoner admits that the inspector came.]

512

Chapter Twenty

In (21) lamentar ‘to regret’ is a governing matrix that speakers customarily use to introduce pragmatically presupposed complements, that is, non-asserted complements, and therefore it should take complements carrying the subjunctive, yet in this instance the indicative is being used. The speaker in this instance has chosen to assert the proposition in the complement and comment on it in the same utterance, in some sort of biclausal utterance, where both clauses are asserted ‘you’ve been sick and I regret it’. In (22), on the other hand, admitir ‘to admit’ is a governing matrix that customarily takes complements in the indicative, yet in this instance, the subjunctive is being used by the speaker, whose communicative intent is not to indicate that the propositional content of the complement is true for the speaker or for any other person. The complement is assumed to be pragmatically presupposed; that is, it is assumed to belong in the ongoing discourse, and the speaker is simply commenting on the proposition expressed by the complement, which is signaled by the use of the subjunctive mood.

4. Extending the analysis to adjectival and adverbial subordinate sentences The analysis proposed for mood selection in Spanish complements can also be extended to adjectival and adverbial clauses. That is, the conceptual distinction between pragmatic assertion and non-assertion applies to the mood selection of both adjectival and adverbial subordinate sentences in Spanish. Adjectives modify a noun (a discourse referent) but do not themselves evoke discourse referents. The traditional semantic account of mood distribution in adjectival clauses is based on the notion of specificity and non-specificity of the referent being modified (Rivero 1977; Cressey 1971). The difference between indicative and specificity, on the one hand, and subjunctive mood and non-specificity, on the other, is related to notions of assertion and non-assertion. The subjunctive in the adjectival clause is motivated by the fact that it is not possible to assign a property to something that may not exist, therefore the speaker will not assert the propositional content of the subordinate as being true for the speaker or any other person. The indicative, on the contrary, is motivated when the speaker identifies the referent and assigns this property with certainty, asserting the propositional content of the subordinate as being true for the speaker or any other person. Consider the examples in (23):

Advantages of a Pragmatic Account of Mood Distribution

513

(23) a. La chica que vive en esta casa tiene que ser the girl that-RP live+3SG in this house have+3SG to be+INF very rich muy rica. [The girl that lives in this house has to be very rich.] b. La chica que viva en esta casa tiene que the girl that-RP live+3SG+SUBJ in this house have+3SG to ser muy rica be+INF very rich [The girl that may live in this house has to be very rich.]

In (23a), by using the indicative, the speaker identifies the referent denoted by la chica ‘the girl’ as specific and identifiable. In (23b), on the other hand, the speaker does not identify the referent denoted by la chica ‘the girl’ as specific and identifiable, therefore using the subjunctive in the subordinate clause. Adverbial subordinate clauses, on the other hand, modify the main clause and express circumstances of the main verb. As in the case of adjectival clauses, mood distribution in adverbial subordinates can be related to the notions of assertion and non-assertion. Similarly to adjectival subordinates, the subjunctive is used in adverbials when the situation expressed by the subordinate is an ‘unrealized’ event/state and cannot therefore be asserted as being true for the speaker or for any other person. The indicative, on the other hand, will be used in adverbials when the situation expressed by the subordinate is not an ‘unrealized’ event/state and their propositional content can therefore be asserted as being true for the speaker or any other person.

5. The relevance of the present study This paper subscribes to the idea that the adequate generalization in mood distribution in Spanish complements is assertion vs. non-assertion. However, contrary to the previous analyses in Terrell and Hooper (1974), Giannakidou (1994, 1995, 1998), and Quer (2001, 2010), this paper claims that it is the meaning of a matrix in a given communicative context, rather than simply the meaning of a predicate, that licenses speakers to assert the propositional content of a complement. The account and/or description of mood selection in Spanish complements in terms of the pragmatic notion of assertion and nonassertion is actually quite simple but certainly not overly so. It might be thought that the account offered here is just a little more than a restatement

514

Chapter Twenty

of the assertion/non-assertion account proposed many years ago by Terrell and Hooper (1974). However, I argue that the current proposal does much more than that because it disposes of the semantic notion of assertion and non-assertion as theoretical tools for determining mood in noun complements, and employs a speaker-based pragmatic notion, which is fundamental for a well-developed theory of utterance interpretation. I propose that, due to its lexical semantics, matrices impose a specific construal on the situation expressed in their complements, and hence motivate a specific construction. Some matrices are more flexible, and potentially compatible with several construals. In those cases the selection is made according to the pragmatics of the speech situation. The semantic value of an expression involves the way the speaker chooses to think about it and represent it, as well as the properties inherent to the situation being expressed in the complement. This is a pivotal claim of the Cognitive Grammar framework, and is expressed by Langacker (1987, 7) as follows: In choosing a particular expression or construction, a speaker construes the conceived situation in a certain way, that is, she selects a particular image (from a range of alternatives) to structure its conceptual content for expressive purposes.

In other words, the meaning of an expression includes both the knowledge system it evokes when the expression is activated, as well as the particular ‘construal’ the speaker imposes on the situation. As a consequence of the perspectival nature of meaning proposed in this chapter, meanings change. New experiences and changes in our environment require that we adapt our semantic categories to transformations of the circumstances, and that we leave room for nuances and slightly deviant cases. This dynamic nature explains the flexibility of the ‘meaning’ of matrices and resulting mood variation. The theory dealing with the analysis of mood selection in Spanish presented in this chapter not only outlines the communicative intricacies that lead to a particular mood choice by speakers, but it is also able, to some extent, to predict mood choice by speakers depending on the discourse situation. This feature is in turn what makes the current analysis a valid pedagogical tool for teaching mood selection in Spanish complements to students of Spanish as a second or foreign language.

Advantages of a Pragmatic Account of Mood Distribution

515

References Blanco, J. A., and C. María. 2009. Ventanas. Curso intermedio de lengua española. 2nd ed. Boston, MA: Vista Higher Learning. Cressey, W. 1971. “The subjunctive in Spanish: a transformational approach.” Hispania 5 (1): 895 896. Fauconnier, G. 1985. Mental Spaces. Cambridge, Mass.: MIT Press. Giannakidou, A. 1994. “The semantic licensing of NPIs and the Modern Greek Subjunctive.” In Language and Cognition 4, Yearbook of the research Group for Theoretical and Experimental Linguistics, edited by A. de Boer, H. de Hoop, and H. de Swart, 55-68. Groningen: University of Groningen. —. 1995. “Subjunctive, habituality and negative polarity.” In Semantics and Linguistic Theory (SALT) V, edited by M. Simons, and T. Galloway, 132-150. Ithaca, NY: University, CLC Publications. —. 1998. Polarity Sensitivity as (Non)veridical Dependency. Amsterdam: Benjamins. —. 2011. “Positive polarity items and negative polarity items: variation, licensing, and compositionality.” In Semantics: An International Handbook of Natural Language Meaning, 2nd ed., edited by C. Maienborn, K. von Heusinger, and P. Portner. 1660-1712. Berlin: Mouton de Gruyter. Guitart, J. M. 1991. “The pragmatics of Spanish mood in complements of knowledge and acquisition of knowledge predicates.” In Discourse Pragmatics and the Verb. The Evidence from Romance, edited by S. Fleischman, and L. R. Waugh, 179-193. New York: Routledge. Guzmán, E., P. Lapuerta, J. E. Liskin-Gasparro, and M. Olivella de Castells. 2013. Identidades. Exploraciones e interconexiones, 3rd ed. Upper Saddle River, NJ: Pearson Education. Jary, M. 2009. “Relevance, assertion and possible worlds: A cognitive approach to the Spanish subjunctive.” In Utterance interpretation and cognitive models, edited by P. De Brabanter, and M. Kissine, 235-277. Bingley: Emerald Group Publishing. Klein, F. 1975. “Pragmatic constraints on distribution: the Spanish Subjunctive.” In Papers from the Eleventh Regional Meeting. Chicago: Chicago Linguistic Society. April 18-20, 1975, edited by R. E. Grossman, L. J. San, and T. J. Vance, 353-365. Chicago: Chicago Linguistic Society. Krakusin, M. 1992. “Selección del Modo después de el hecho de que.” Hispania 75 (5): 1289-1293.

516

Chapter Twenty

Langacker, R. W. 1987. Foundations of cognitive grammar (vol. 1): theoretical prerequisites. Stanford: Stanford University Press. Lavandera, B. 1983. “Shifting moods in Spanish discourse.” In Discourse Perspectives on Syntax, edited by F. Klein-Andreu, 209-236. New York: Academic Press. Lipski, J. M. 1978. “Subjunctive as fact?” Hispania 61: 931-934. Lunn, P. 1989. “The Spanish subjunctive and relevance.” In Studies in Roman Linguistics, edited by C. Kirschner, and J. De Cesaris, 250-260. Amsterdam: Benjamins. Mejías-Bikandi, E. 1994. “Assertion and Speaker's Intention: A Pragmatically Based Account of Mood in Spanish.” Hispania 77: 529539. —. 1998. “Pragmatic Presupposition and Old-Information in the Explanation of the Use of the Subjunctive Mood in Spanish.” Hispania 81: 941947. Pellettieri, J., N. Lopez-Burton, R. Hershberger, R. Gómez, and S. NancyDavis. 2011. Rumbos. Curso intermedio de español, 2nd ed. Boston, MA: Heinle Cengage Learning. Quer, J. 2001. “Interpreting Mood.” Probus 13: 81-111 —. 2010. “On the (Un)stability of Mood Distribution in Romance.” In Modality and Mood in Romance: Modal Interpretation, Mood Selection, and Mood Alternation, edited by M. Becker, and E.-M. Remberger, 163-179. Berlin: Walter de Gruyter. Rivero, M. L. 1977. Estudios de gramática generativa del español. Madrid: Cátedra. Sandstedt, L. A., and R. Kite. 2014. Espacios. Boston, MA: Heinle Cengage Learning. Terrell, T., and J. Hooper. 1974. “A Semantically Based Analysis of Mood in Spanish.” Hispania 57: 484-494. Woehr, R. 1975. “Grammar of the factive nominal in Spanish.” Language Science 36:13-19.

Notes 1

Assertive predicates are assertive not because they are themselves assertive but because their complements are assertions (weak and strong). Non-assertive predicates are non-assertive because their complements are non-assertions. 2 The buletic model, contrary to the epistemic one, is the semantic model that deals with somebody’s wishes or desires. 3 Giannakidou (1994, 1995) claims that the subjunctive is ‘triggered’ by certain semantic properties of the embedding context pretty much the way polarity items

Advantages of a Pragmatic Account of Mood Distribution

517

(PIs) are triggered by their licensers; the subjunctive can thus be viewed as a PI of some kind. 4 Despite its shortcomings, it is important to mention that Quer’s (2001, 2010) account of mood shift as model shift aims to provide a comprehensive understanding of the use of mood in all embedded sentences, allowing for a unified analysis of mood distribution in argument, adjectival and adverbial clauses. The present study accounts for mood distribution in embedded clauses in terms of discourse referent activation, and since the only clauses with referential properties are complement clauses, the focus of the present study is solely mood distribution in complement clauses.

CHAPTER TWENTY ONE CONVERSATIONAL IMPLICATURES IN THE CROATIAN EFL CLASSROOM ANA WERKMANN HORVAT AND ANA KEDVEŠ

1. Introduction In a traditional foreign language classroom teachers usually set the focus of their lessons on widely recognizable and easily testable language skills (Richards 2011) like reading, speaking, listening and writing. Other aspects of language, in particular pragmatic competencies, have been considered less important and are thus the most difficult to both teach and acquire, mostly due to their “subtlety and complexity” (Grossi 2009, 53). Pragmatic competencies seem to take the backseat in a traditional foreign language classroom. However, it is important to note that sociocommunicative and pragmatic competencies are as important as other competencies to achieve full communicative competence (Grossi 2009). Nowadays, it is becoming widely accepted that teaching and learning languages includes far more than barely targeting grammatical and lexical systems. The rationale for this change of direction in teaching lies in the fact that what has recently been considered the main goal in the process of learning a foreign language is, in fact, developing communicative skills (Bardovi-Harlig 1996; Rose and Ng 2001; Bardovi-Harlig and Griffin 2005). Thus it should not come as a surprise that more emphasis is recently being put on acquiring pragmatic competencies. This is especially due to recent studies supporting the idea that students have problems at the interpersonal level in conversations with native speakers, “in spite of having a perfect dominion of the foreign language grammar rules” (Fernández Amaya 2008, 11). In line with the above considerations, our paper explores the classroom procedure of applying an explicit instructional approach, active learning strategies in particular, in teaching conversational implicatures to Croatian

520

Chapter Twenty One

EFL learners. In addition to that, we question the learners’ attitudes towards the applicable value of pragmatic competence in second language acquisition. Our paper opens with an outline of previous research in the field. While we briefly explore issues like pragmatic competence and incompetence, we will present in depth some of the main research issues in the acquisition of pragmatic compentencies, especially the teachability of these compentencies, and the explicit vs. implicit approach to their teaching. After reviewing studies dealing with the acquisition and teaching of conversational implicatures in EFL learning, we will introduce the research itself, which will include our research objectives, the methods and instruments applied, and the main characteristics of our participants. The next section presents the findings obtained, both from our one-session workshop and from the survey that followed it. In our discussion section, we will discuss the possible implications of our results and identify some of the challenges instructors and learners encounter in an EFL classroom.

2. Literature review 2.1. Teaching pragmatics Kasper (1997) calls pragmatic compentencies “the knowledge of communicative action and how to carry it out, and the ability to use language appropriately according to context”, while Thomas (1983) defines them, along the same line, as the ability to communicate effectively that includes knowledge beyond the level of grammar. Pragmatic incompetence, on the other hand, is the lack of unawareness of, and inability to use, the communicative rules of a language and culture. It may lead to pragmatic failure (Thomas 1983), that is, the negative, unintentional outcome of the transfer of L1 pragmatic rules into a second/foreign language context, and to misunderstandings and frustrations in communication (Murray 2011; Brock and Nagasaka 2005). One of the basic issues tackled in interlanguage pragmatics research is the teachability of pragmatic competencies (Rose 2005). According to Yates (2004), there is increasing evidence that different aspects of pragmatics are teachable. Studies show that pragmalinguistic routines can be taught successfully (House 1996) even at the beginner level. For example, an empirical study on German-English interlanguage conducted by Wildner-Basset (1994) supports the idea that learners face difficulties when acquiring pragmatic declarative and procedural knowledge. However, it was found that at an elementary level of language instruction these competencies can, in fact, be successfully acquired. Also, Bouton

Conversational Implicatures in the Croatian EFL Classroom

521

(1992, 1994) explores how efficient are the learners in acquiring pragmatic competencies (e.g., in interpretation of implicatures) if they are not taught. His studies show that the learners’ efficiency could be improved by focused instruction. On the other hand, Kasper (1997) claims that knowledge of pragmatic competence cannot be definitively taught, but rather possessed, developed, acquired, used or lost. However, the author notes that there are ways in which we can arrange “learning opportunities” so students can develop their pragmatic competencies in their L2. According to the scholar, each teaching goal or, to put it differently, each aspect of pragmatic competence requires different teaching methods, procedures and strategies in order to be fully efficient. Some suggestions of these strategies can be found in Smith (2009). Based on empirical research, the author suggests implementing strategies that mimic real-life speech acts such as compliment, condolence and apology responses. Similarly, Takimoto (2008) explored the effects of deductive and inductive teaching approaches to the acquisition of pragmatic competence on Japanese learners of English as a foreign language. The study showed that “inductive instruction is effective when combined with problem-solving tasks or structured input tasks for which the emphasis is on pragmalinguistic and sociopragmatic resources” (Takimoto 2008, 381). This suggests that the teacher should keep in mind that form-only activities are of limited applicability since activities that involve more pragmalinguistic and sociopragmatic processing yield better results. These observations gave rise to the idea that different activities and strategies bring different results. With regard to language teaching, Knutson (2006, 592) notes that “introducing students to a broad range of cultural material, all the while keeping L1 use in classroom interaction to a minimum – has become the unspoken imperative”, and proposes the introduction of curricular modules on cross-cultural awareness before the language courses on the intermediate level, so as to incite the language learners to reflect on various aspects of culturally conditioned behavior. The findings of Rafieyan, Majid and Eng’s (2013) survey on the relation between attitude toward the target language culture instruction and pragmatic comprehension development in Malaysian English language learners suggest that positive attitudes towards the culture in the affective, behavioural and cognitive aspect enhance the development of pragmatic competencies. Furthermore, such results are in line with the previous studies by Albirini (2009) and Sarçoban and Çalúkan (2011) which indicated learners’ positive attitudes toward learning about the cultural

522

Chapter Twenty One

perspectives of the target language. However, Ishihara and Cohen (2010, 15) warn that apart from acquiring a socio-cultural background, “it is important to know the language forms to use in a host of sometimes delicate cross-cultural situations, depending on who is being spoken to, the relative positions of authority of the conversational partners, and the context of the communication.” The teachability and the different ways of teaching pragmatic competencies, or in other words, whether pragmatic competence should be taught explicitly by instruction or whether exposure is sufficient represent an extremely influential part of work done on the acquisition of pragmatic competencies, especially among second language learners. Traditionally, the explicit approach involves input and practice followed by description, explanation and discussion of pragmatic features, while the implicit approach involves bare input and practice without a metapragmatic component (Kasper 1997). These two approaches have been extremely important in experimental studies on the acquisition of pragmatic competencies by L2 learners. For example, Tateyama et al. (1997) explore how beginning L2 learners of Japanese acquire and develop pragmatic competencies when receiving explicit as opposed to implicit language instructions. The results of the study suggest the idea that explicit instruction is more efficient than implicit instruction. Similarly, Alcón Soler’s (2005) study shows that learners’ awareness of requests is raised from both explicit and implicit instruction. However, both her study and the one by Mirzaei and Esmaeili (2013) carried out on Iranian L2 learners of English show that participants exposed to explicit, rather than implicit, instruction achieve better results in pragmatically appropriate uses of language. Halenko and Jones (2011) showed that not only does the explicit instruction facilitate learning pragmatic competencies, but it is also found valuable by learners themselves. In particular, the participants reported that the instruction helped them to communicate more effectively, as well as to focus their attention on different sociopragmatic aspects of communication Overall, studies in interlanguage pragmatics suggest that explicit instruction is particularly beneficial to raising pragmatic awareness and fostering the development of pragmatic competence in students.

2.3. Teaching and acquiring implicatures In the previous section we have seen that scholars have explored various facets and components of pragmatic competence. One of these is conversational implicature. This is of relevance to interaction in general,

Conversational Implicatures in the Croatian EFL Classroom

523

and to language learning in particular, since their accurate pragmatic interpretation requires the ability to understand the intrinsic, literal meaning of a message, but also the speaker’s intention and the context (Grice 1975; Kordiü 1991). Grice introduces the term conversational implicature in his seminal 1975 work, Logic and Conversation. The term refers to what is not directly said in an utterance, or even not, strictly speaking, implied, but rather entailed by it. According to Grice, the existence of conversational implicatures reveals that the cooperative principle is obeyed. The speakers intend to convey meaning that is not literally conveyed in their utterance, thus creating a distinction between what is said and what is intended. In order to understand the meaning that is not pronounced, members of a speech community rely on shared background, common knowledge and experience (Gumperz 1982, 1996; Ochs 1993). However, if a foreign language student is not familiar with the interactional maxims or the cultural background of a native speaker, this can result in cross-cultural pragmatic failure. Early studies dealing with this aspect of pragmatic competence (Bouton, 1992; Takahashi and Roitblat 1994) suggest that L2 learners experience difficulties when processing implied meaning in their L2. Later studies such as Blight (2002) and Taguchi (2002, 2005), however, did not provide conclusive evidence in this regard. For instance, Blight (2002) examined different types of explicit instruction in teaching conversational implicatures to Japanese learners. His findings showed that difficulties could emerge even when explicit instructions are used, especially with non-advanced learners who have problems with grasping the complexity of the Cooperative Principle he used as framework in teaching, as well as the interplay of social elements which stipulate the realization of the Theory. Taguchi (2002) explored comprehension of implied meaning in general and indirect replies among Japanese learners of different levels of proficiency. However, her results showed that, regardless of proficiency levels, learners were able to understand implied meaning on the basis of contextual information. In addition to that, Taguchi (2005) investigated whether L2 proficiency has any affect on comprehension of implied meaning in spoken discourse. The results showed that more proficient speakers are, in fact, more accurate, but not faster. The research so far, therefore, does not make it clear whether an effective method can be implemented for fostering L2 learners’ understanding of conversational implicatures.

524

Chapter Twenty One

3. Research design and methodology 3.1. Research objectives This study aimed at analysing the use of active learning strategies in introducing conversational implicatures, in particular to EFL students who are not in command of metapragmatic linguistic resources. We focused on the following objectives: 1) to design and implement a student-centred instruction procedure for teaching conversational implicatures 2) to analyse the effect of the teaching procedure through students’ responses during the workshop 3) to survey the students’ attitudes towards the applicable value of conversational implicatures instruction in SLA and their overall awareness of the importance of acquiring pragmatic competencies.

3.2. Participants The central phase of our research was conducted in May 2013. The participants were first-year students at the Faculty of Teacher Education Osijek, Department in Slavonski Brod. Their participation was elicited within their regular EFL course; they were invited to participate, and twenty five of them, all female, agreed to join the research. We did not test the students prior to our research, but in order to enrol at a Croatian university, in accordance with the Law on education in primary and secondary schools, they had all been required to pass the English state graduation exam, which implies they had all attained a B2 level of proficiency in English. Their educational background did not include any previous explicit instruction in conversational implicatures or pragmatics in general, either in their native language or the foreign language they were studying, and they had little or no access to the metalinguistic resources that were relevant to our research topic.

3.3. Research method and instruments The research was organised in two main parts. The first part was the implementation of a one-session workshop with the students, while the other one consisted in the administration of a questionnaire so as to explore the students’ attitude towards the applicability of conversational implicatures in foreign language teaching, as well as their general awareness of the role pragmatic skills play in the communication process.

Conversational Implicatures in the Croatian EFL Classroom

525

In line with the majority of previous studies (e.g., Bouton 1994; Rose and Ng 2001; Nguyen, Pham, and Pham 2012) which claim that explicit methods are more effective, we applied the explicit approach, designing student-centred strategies followed by elaborate metalinguistic instructions, description and discussion. It is important to note that the students participating in the research had previously had very limited access to metalinguistic resources. 3.3.1. Design of the workshop The workshop format was chosen for the numerous advantages it offers in comparison to other approaches, especially in contrast with traditional teacher-fronted approach. The workshop method is considered to be intellectually inciting and helpful in developing participants’ organisational, social and communication skills (Kyriacou 1988, 66), as well as in increasing and maintaining the participants’ motivation through carefully structured interactive activities (Martinko 2012). Some recent studies favour content-based cultural workshops, since these “encourage fulfilling, productive, and rewarding language learning experiences” (Robinson 2011, 15). Furthermore, there are indications that participating in workshops is especially beneficial for students who opt for teaching professions (such as the participants of our study), because it “significantly increases their self-efficacy and effective teaching behaviours and decreases their public speaking apprehension” (Boman 2013, 100). Due to the aforementioned considerations, the workshop method was chosen and organised so as to raise the students’ interest, motivate them for experiential learning and provide the optimal environment for our research purposes. The workshop was divided into an introductory, central and final part, encompassing five different activities. The aim of the initial activity was to announce the topic of the workshop and to make the students aware of the existence of this type of linguistic structures, since one of the chief goals in the classroom was raising pragmatic awareness (Bardovi-Harlig 1996). The students were given handouts containing ten short dialogues such as: (4)

Q: Would you like some coffee? A: I think I would die without it.

and: (5)

Q: Does she know how to sing? A: You’d better bring some ear plugs.

Chapter Twenty One

526

Their task was to state whether the respondents’ answer to the question is positive or negative (YES or NO). This was followed by a group discussion on how they had reached their answers, how the hearers were able to interpret them, which knowledge and information was used in the exchange, etc. Based on the input from this activity, each student was instructed to derive a rather simple meta-linguistic definition of conversational implicatures by filling in the following template: (6)

When speaking to other people, we often _____________ ideas that we do not express____________, but rather___________. These expressions are called ____________. What helps the hearers to understand them is their knowledge about the _____________, _____________ and shared common ___________.

with the key-words imply, directly, indirectly, conversational implicatures, context, culture, and knowledge. We opted for a simpler metalinguistic task, since previous studies had shown that an elaborate presentation of Grice’s theory is often difficult to understand even for advanced learners (Blight 2002; Murray 2011). However, we wanted to include a metalinguistic activity involving a definition-oriented task, since these are the activities that students most frequently face in their L1 and L2 classes. The aim of this activity was to help students pin down their intuitions about conversational implicatures, their function and meaning, which were stimulated through the first activity. The next activity was aimed at exploring the preconditions for the derivation of conversational implicatures and the process of their interpretation. This analysis was done on the examples from the scripts of the popular television series That ‘70s Show (1998 – 2006), Scrubs (2001 – 2010), House, M.D. (2004 – 2012), The Office: An American Workplace (2005 – 2013), Big Bang Theory (2007 – ) and the film Yes Man (2008). (7)

MR. SUTTON: I’m a garbage man. How many times do you think I get thanked? TURK: Six! MR. SUTTON: You’re off by about six! TURK: Twelve! (Scrubs, 2001 – 2010)

(8)

MIDGE: Hmm, I'm just loving the time we spend together Donna. In my new book 'Our Mothers, Our Selves' it says we're supposed to be friends. We should talk to each other, and listen. Your stupid father never listens.

Conversational Implicatures in the Croatian EFL Classroom

527

DONNA: Is there anything in your book about not insulting my father? MIDGE: Well I don't know, I've only read the first chapter. (That ‘70s Show, 1998 – 2006)

The participants were divided into groups and assigned the task of examining the excerpts and discussing whether and how the implicatures were interpreted by the characters. The aim of this activity was to expose the students to conversational implicatures in naturally occurring contexts and to provide them with an opportunity to apply the metalinguistic knowledge from the previous activity.1 Furthermore, this activity was designed to motivate the students by using examples from popular culture, rather than the usual textbook examples which tend to be contrived and lacking in originality (Bardovi-Harlig 1996). In the next activity the participants received handouts with a text describing part of an American woman’s everyday routine. The text contained a number of conversational implicatures, the majority of them being culture-bound. The participants’ task was to identify the conversational implicatures, indicate which of these were culture-bound and explain why. Furthermore, they had to discuss how these implicatures could be interpreted by members of different speech communities and then list similar examples present in the Croatian language. Also, they had to consider the influence of non-verbal elements as well as the level of formality of a particular conversation on the successful use of implicatures. The aim of this activity was to raise awareness of the cultural relevance of conversational implicatures.2 The final activity provided a synthesis of the workshop and offered the participants an opportunity to practice their productive skills and express their creativity. We were reluctant to use a DCT or similar data elicitation instruments because they present non-authentic communication contexts and originate non-spontaneous interactions which do not carry real consequences for the interlocutors. We therefore feared that such elicitation tasks would affect the participants’ reactions and responses, which would be different from those produced in spontaneous interaction. (Beebe and Cummings 1996). Role play exercises, on the other hand, would enable us to use made-up situations but to elicit more realistic responses, since the participants would be pretending, acting “as if” and thus relieved of possible communicative inhibitions. Maley and Duff (2002) note that such classroom activities help to place the language in a cultural, social and personal context and serve to enhance the students’ motivation.

Chapter Twenty One

528

The students were divided into groups and each received a handout with a short dialogue between two characters (two mobsters, two lovers, student and teacher, patient and doctor, two business partners, guest and waiter, hair-dresser and customer), which included an occurrence of a misinterpreted conversational implicature, as in the following examples: (9)

(The teacher is orally questioning the student. However, he has not prepared for the exam.) TEACHER: Can you explain the basic principles of the theory of relativity? STUDENT: Can’t you ask me something a little less complicated? TEACHER: Sure, would you like to recite the alphabet? STUDENT: A, B, C, D .... TEACHER: .......

(10) (The patient has been experiencing massive headaches. After having done some scans, the patient comes to see his doctor with his results.) DOCTOR (looking at the scans): OK, let’s see what we have here... All right... Mhm. I see. So, I believe we won’t be seeing much of each other anymore... PATIENT: ......

The participants were instructed to guess the possible implications of the misinterpretation and try to imagine what the worst conceivable consequences would be. Based on this, they were to write the scenarios for these stories and prepare to act them out in front of the entire group. The workshop ended after the participants performed their sketches. The aim of this activity was to wrap up the workshop and examine the students’ overall comprehension of conversational implicatures. A production task was regarded as best suited, as it would require the student’s overall involvement. 3.3.2. The survey The second part of the research was conducted through the administration of a brief questionnaire which explored the participants’ attitudes toward and opinions on the applicability of conversational implicatures in foreign language classes, along with their general familiarity with the pragmatic aspect of communication and its significance in intercultural interaction. The value of the survey lies in the fact that it addresses the question of students’ choices and attitudes and

Conversational Implicatures in the Croatian EFL Classroom

529

how these largely affect their motivation in classroom, which was raised by several scholars (Bardovi-Harlig 1996; Ishihara 2010). The questionnaire was constructed in English and consisted of eight closed-ended questions, six of them presenting Likert scales of attitudes and two requiring ranking of answer options. The questions investigated if participants had prior knowledge of conversational implicature, the level of agreement with claims regarding their view on the importance of implicature for the understanding of the speaker’s intended meaning, the role of culture and context, the effectiveness of the workshop in terms of implicature explanation, the introduction of conversational implicatures into foreign language teaching and raising the awareness of culture and context-dependency of conversational implicatures. The participants were also asked to rank the listed teaching methods with respect to their effectiveness. Finally, the participants were given the chance to freely comment on the workshop (see Appendix 1).

4. Findings In this part we will first present the results which we obtained from a structured observation of the implementation of the workshop, participants’ responses and their classroom activity. After that, we will provide a quantitative analysis of the responses to the questionnaire which we administered afterwards. In the opening activity the students were to discover the implied meaning of the answers from the short dialogues. They solved the task very successfully and were able to detect the meaning conveyed. This might be due to the fact that the dialogues were simple, short and that there were no cultural peculiarities. However, there was one dialogue that they were not quite certain about: (11) Q: Would you like some pizza? A: Do birds fly?

When discussing the answers and how they had reached them, the participants noted that they were not sure whether the interlocutor wanted pizza or not, since it was not clear how precise and specific he/she was. With other dialogues like the following: (12) Q: Has Mark arrived yet? A: His car is here.

530

Chapter Twenty One

they were able to distinguish immediately whether the answer was positive or negative. Together we examined the information they relied on and the clues which yielded the correct answers. The participants pointed out that they had made use of their knowledge of the world and human behaviour in particular situations in order to construct the meaning which was not directly expressed. These conclusions were helpful in solving the following task where the participants derived a meta-linguistic definition of conversational implicature by using the template given to them. All of the participants accomplished the task quite easily: (13) When speaking to other people, we often imply ideas that we do not express directly, but rather indirectly. These expressions are called conversational implicatures. What helps the hearers to understand them is their knowledge about the context, culture and shared common knowledge.

This suggests that students feel more comfortable when doing explicitly formulated tasks as opposed to those that are more abstract and implicit, maybe because this is what they usually encounter in contemporary teaching materials. The next activity dealt with the context in which conversational implicatures are employed and interpreted. This proved to be the most difficult task for our students; they had trouble detecting which implicatures were successfully used and explaining how the interlocutors interpreted these implicatures in the given conversations. With the excerpts where the interlocutors were both pragmatically competent and, more important, willing to cooperate; e.g.: (14) CARLA: Hey, Dr. Cox, you wanna put in for some lottery tickets? DR. COX: O gosh, Carla! I would, I really would! But you see, I already set fire to a big pile of money just this morning. CARLA: Hey, we have the same chance of winning as anyone else. (Scrubs, 2001 – 2010)

they were eventually able to identify and interpret the conversational implicatures. However, in cases where the communication background was more complex and the interlocutors repeatedly failed to comply with the Cooperative Principle, the majority were not able to reach the solution: (15) CUDDY: I was expecting you in my office 20 minutes ago. HOUSE: Really? Well, that’s odd, because I had no intention of being in your office 20 minutes ago.

Conversational Implicatures in the Croatian EFL Classroom

531

CUDDY: You think we have nothing to talk about? HOUSE: No, just that I can’t think of anything that I’d be interested in. (House M.D., 2004 – 2012)

The reason for students’ difficulties might be that complex conversational implicatures including multiple violations of the Maxims are often the most difficult to process and understand (Blight 2002) as well as the fact that students are rarely exposed to real communicative contexts in their learning materials. We facilitated their decoding process by providing additional input on the communicative environment, the characters’ traits and relationships, as well as on the plot and genre of the television series. Also, we encouraged more intensive group work and peer support. The activity which followed seemed to be much easier for our participants. When working on the text, they were perfectly capable of identifying the conversational implicatures. However, they required more detailed clarification for some of the expressions, like Xerox and similar ones. They were able to explain how and why the respective conversational implicatures are culture-bound probably because they are constantly exposed to elements of popular culture in the US through TV and the Internet. Furthermore, in the discussion that ensued, they thought of many examples of conversational implicatures in Croatian which foreigners would not be able to interpret without additional explanations being provided. The participants also remarked that the non-verbal aspect of communication might alter the interpretation of conversational implicatures, as well as that the use of implicatures is dependent on the interpersonal relationships of the interlocutors. The final activity drew on all the previously addressed issues and gave the participants a chance to express themselves creatively. In pairs, they assumed the roles of the characters assigned to them and wrote and later performed the possible endings of the dialogues. They were able to anticipate the probable misinterpretations and possible consequences (wrong haircuts or food orders, heated arguments, fights, etc.), but did not always think of the worst imaginable scenarios. When writing the dialogue sequences, they also produced new conversational implicatures. Most of the sketches were humorous, so it was evidently enjoyable for the participants to both perform and view them. This supports the idea that activities such as role-plays help personalize the language and have a great impact on increasing students’ motivation, which might be the reason why students performed successfully in such a complicated task. Furthermore, such activities “can potentially train a new L2 speaker to perform better in

532

Chapter Twenty One

L2 in difficult conditions, for example, when under pressure or when being observed” (Babayants 2011, 96). Due to space limitations, in what follows we will present only the crucial and most interesting findings of the survey as relevant to the following discussion. The other part of the research findings comprises the quantitative analysis of the answers elicited through the questionnaire. In one of the questions, the participants were asked to express the level of agreement with the statement Being able to interpret conversational implicatures is important for understanding the speaker’s intended meaning. As presented in Figure 1, the vast majority of participants agreed and strongly agreed with the statement, while 20% them neither disagreed nor agreed. Strongly agree

20%

Agree

60%

Neither disagree nor agree

20%

Disagree

0%

Strongly disagree

0%

FIGURE 1. Responses to the prompt “Being able to interpret conversational implicatures is important for understanding the speaker’s intended meaning.”

In one of the next questions, the participants ranked the skills that improve the most as a result of being familiar with conversational implicatures. Overall language proficiency and Maintaining satisfying social relations with others were the highest ranked answers and Learning about other cultures the lowest ranked answer (Figure 2).

Conversational Implicatures in the Croatian EFL Classroom

Learning about other cultures

533

2.5

Translation

2.7

Maintaining satisfying social relations with others

2.75

Overall language proficiency

2.75

FIGURE 2. Responses to the prompt “Being familiar with conversational implicatures improves one’s skills in:”

The following questions explored the students’ impressions of the workshop. First of all, we wanted to know if the workshop was helpful in fostering understanding the phenomenon of conversational implicature. As evident from Figure 3, most of the students agreed (55%) or strongly agreed (25%) on this point, 10% of them neither disagreed nor agreed, while 5% of them disagreed and the 5% strongly disagreed. Strongly agree

25%

Agree Neither disagree nor agree

55% 10%

Disagree

5%

Strongly disagree

5%

FIGURE 3. Responses to the prompt “The workshop activities have helped me to understand the phenomenon of conversational implicature.”

Moreover, 20% of the participants strongly agreed and 70% of them agreed with the statement The assignments raised my awareness of culture- and context-dependent use of conversational implicatures. The last two questions were related to the introduction and applicability of conversational implicatures in foreign language classes. Firstly, the great majority expressed a positive attitude (80% agreed and 10% strongly

534

Chapter Twenty One

agreed; 10% neither disagreed nor agreed) towards the claim Conversational implicatures should be introduced into foreign language teaching. Strongly agree

10%

Agree

80%

Neither disagree nor agree

10%

Disagree

0%

Strongly disagree

0%

FIGURE 4. Responses to the prompt “Conversational implicatures should be introduced into foreign language teaching.”

Secondly, they ranked the teaching approaches listed with respect to their effectiveness as presented in Figure 5. None of the participants provided their own suggestions for other teaching approaches.

Context based approach (literature, films...)

2.95

Non-instructional, communication based approach

2.9

2.45

Theory based approach

Other

0

FIGURE 5. Replies to the query “Which strategies and methods would be the most effective in teaching conversational implicatures?”

5. Discussion In what follows we will discuss the general implications of the findings with respect to previous studies (see Section 2). The observation of students’ responses and the input they provided when solving the tasks

Conversational Implicatures in the Croatian EFL Classroom

535

suggests the effectiveness of explicit instruction, which is consistent with the findings of previous studies (e.g., Bouton 1994; Rose and Ng 2001; Ngyuen, Pham and Pham 2012). Also, it was observed that students responded well to explicit guidance and elaborate discussions before and after the activities. It is interesting to note that following only one teaching session, the participants were able to identify, explain and produce simple forms of conversation implicatures in L2. Moreover, they were able to relate this pragmatic target to examples from their L1. Finding similarities and differences facilitated the correct interpretation of conversational implicatures. It is evident that the group benefited from receiving metapragmatic explanation which was helpful for the understanding of the sociopragmatic features (Ngyuen, Pham and Pham 2012) of the use of conversational implicatures. Paying heed to the participants’ low metalinguistic knowledge, we avoided introducing Grice’s Cooperative Principle, but the simple working definition which was provided nevertheless enabled them to disclose the reasoning process underlying the use of conversational implicature. A few of the activities assigned included materials with authentic language written by native speakers, which served the participants not only as an optimal input of the forms taught (Bardovi-Harling 1996), but also as a valuable source of sociopragmatic information and a vehicle for developing cross-cultural language competencies (Yates 2004). Furthermore, we used the excerpts from popular televisions series, which the majority of participants were familiar with, in order to underline that conversational implicature, and pragmalinguistic phenomena in general, do not only appear in classroom-based environment, but are an integral part of everyday communication in every speech community. Also, the contents of the teaching materials, which are usually associated with leisure time and entertainment, rather than with formal education and commonly used textbook, may have encouraged communicative practices beneficial for pragmatic development (Kasper 1997). However, some difficulties in the implementation arose, primarily in the third activity, which dealt with misused conversational implicatures. This suggests that the recognition and correct interpretation of these was especially challenging for foreign language students in cases of severe and/or multiple violations of pragmatic conventions, as in excerpt (12), where Dr House intently discards Cuddy’s implicatures about missing their appointment and the necessity to discuss problematic issues, but rather focuses on the literal meaning of her words. We also noticed that on some occasions the students intuitively understood that the interlocutors

536

Chapter Twenty One

were misbehaving in the pragmatic sense but could not explain why and how exactly: (16) DR BOB KELSO: Who the hell ate my scone? DR COX: That would be me, Bobbo, and it was delicious - my compliments to the lady. DR BOB KELSO: I made those! DR COX: I know! (Scrubs, 2001 – 2010)

These uncertainties were expected, since studies on classroom procedures for providing instruction in conversational implicature warn the reader that the ambiguity of implicatures and variety of possible interpretations might pose difficulties for learners (Blight 2002). In such cases, recasts or explicit corrections are deemed successful in helping students to notice the wrong forms (Ngyuen, Pham and Pham 2012). However, in the majority of cases we made an effort to contextualise dialogues in detail, providing extra information on the communication environment in which the dialogues took place. This enabled the students to be more successful at recognising the motivation behind the characters’ pragmatic choices. Also, we encouraged all forms of group work, since it not only provided peer support which facilitated the assignment solving but also enabled the students to negotiate the possible meaning and, under conditions of increased naturalness of communication (Bardovi-Harlig and Griffin 2005), to practise and develop their interactional competencies (Young 2011). Throughout the entire workshop, emphasis was placed on the participants’ active participation and engagement; activities were taskoriented and the discussions which ensued heuristically moderated. The workshop method and student-centred activities served to overcome the shortcomings of a teacher-fronted classroom (see Kasper 1997), which is characterised as an impoverished learning environment in terms of acquiring pragmatic competencies. A shift in the power dynamics in teacher-student relationship created the space for a more symmetrical discourse, allowing students to practise conversational management and employ a larger diversity of communicative acts. We aimed to create what Kasper (1997) calls learning opportunities, where the students were able to develop their pragmatic competence in general, that is become aware of the pragmatic dimension of the language use, in their mother tongue, as well as in the foreign languages. The survey in the final phase was a very important part of our research. In the majority of previous empirical studies in the field of pragmatics, the

Conversational Implicatures in the Croatian EFL Classroom

537

researchers used the input from participants to draw conclusions on their abilities, to analyse the effectiveness of the particular pedagogical approaches which were adopted and give practice-informed recommendations for classroom implementation. However, the learners are rarely directly included in the process of making those recommendations or consulted on the matters of learning pragmatics, although some scholars raise the question of students’ motivation to learn L2 pragmatics and value their own pragmatic development (e.g., Ishihara 2010; Bardovi-Harlig 1996). Therefore, in addition to obtaining feedback on our workshop, we were interested in learning what the participants’ impressions were after their first encounter with explicit instruction in pragmatics. The results of the questionnaire analysis certainly indicated an increased level of pragmatic awareness in the foreign language acquisition and use. With regard to attitudes toward being taught pragmatics, the majority of the participants showed that they recognised the importance of being familiar with conversational implicatures and understanding their context- and culture-dependent use. Also, they regarded it as efficient in improving one’s overall language proficiency and establishing agreeable relations with others. Overall, most of the participants evaluated the workshop activities as helpful in understanding the phenomenon of conversational implicatures, which provides evidence in support of the effectiveness of the active learning strategies and the explicit approach we applied. Furthermore, it is encouraging that there was a positive attitude towards the introduction of conversational implicatures in foreign language teaching, even if based on a single workshop session. Among the preferred teaching approaches, the participants rated as the highest the context-based approach, relying on literature, films and pop-culture as a learning environment/medium, and the communication-based approach. Such answers could be related to Young’s (2011) description of the pragmatics of interaction.

6. Conclusion The theoretical and practical aspects of second language acquisition have been of interest to many scholars worldwide; Croatian linguists, however, have not yet given much attention to the pragmatic segment of this field. For this reason, some of the findings and conclusions in our paper are largely speculative and leave a number of research questions open. However, we successfully addressed a few research issues. Firstly, we implemented a student-centred instruction procedure for introducing conversational implicatures. The student-centred approach, as

538

Chapter Twenty One

evident from the participants’ activity and motivation provided a favourable context for the successful implementation of the workshop. The students’ reactions drew attention to the more challenging aspects of this topic. Secondly, the analysis of the students’ responsiveness and feedback during the workshop offered a valuable insight into the choices teachers make and ought to make when designing and using different teaching materials and activities. Furthermore, it confirmed the effectiveness of explicit methods in teaching pragmatics. Thirdly, the answers collected through the administration of the evaluation questionnaire indicated the students’ positive attitudes towards the introduction of pragmatics-related assignments into foreign language classes and the use of an active, communication-based approach. The students are rarely consulted when creating new curricula, but such findings, although preliminary, definitely suggest the necessity to explore the possibilities of a more extensive introduction of pragmalinguistic contents into foreign language teaching, even if only for experimental purposes. Some contextual factors posed limitations to the study. This was a small-scale, one-session study, conducted with a rather homogeneous group of participants, that is, not a broadly representative sample, which would limit the possibilities for sweeping generalisations of our findings. Also there were some external factors (e.g., time of the day, fatigue) which may have influenced the students’ performance in the workshop. In designing further studies, we would suggest research on larger, more heterogeneous samples of participants or different homogenous groups for comparison purposes (pre- and post-tests). Some further research issues include the application of various teaching methods and their comparison and analysis, an investigation of the factors that influence students’ motivation for engaging in pragmalinguistic activities, and comparison of the explicit and implicit approach in a workshop environment.

References Albirini, A. 2009. “Using technology, literature and guest speakers to raise the cultural awareness of Arabic language learners.” The International Journal of Language Society and Culture 28: 1-15. Alcón Soler, E. 2005. “Does instruction work for learning pragmatics in the EFL context?” System 33 (3): 417-435.

Conversational Implicatures in the Croatian EFL Classroom

539

Babayants, A. 2011. Acting and Second Language Pragmatics: Pedagogical Intersections. Master’s thesis, University of Toronto, Ontario Institute for Studies in Education, Toronto. Bardovi-Harlig, K. 1996. “Pragmatics and language teaching: Bringing pragmatics and pedagogy together.” Pragmatics and Language Learning 7: 21-39. Bardovi-Harlig, K., and R. Griffin. 2005. “L2 pragmatic awareness: Evidence from the ESL classroom.” System 33: 401-415. Beebe, L., and M. Cummings. 1996. “Natural speech act data versus written questionnaire data: How data collection method affects speech act performance.” In Speech acts across cultures: Challenges to communication in a second language, edited by S. Gass, and J. Neu, 65-86. Berlin: Mouton de Gruyter. Bennett, M. J. 1997. “How not to be a fluent fool: Understanding the cultural dimension of language.” In New ways in teaching culture, edited by Fantini A. E., 16-21. Alexandria, VA: TESOL International Association. Blight, R. 2002. “Classroom procedure for explicit instruction in conversational implicature.” JALT conference proceedings. http://jaltpublications.org/archive/proceedings/2002/142.pdf. Boman, J. S. 2013. “Graduate student teaching development: Evaluating the effectiveness of training in relation to graduate student characteristics.” Canadian Journal of Higher Education 43 (1): 100114. Bouton, L. 1992. “The interpretation of implicature in English by NNS: Does it come automatically without being explicitly taught?” In Pragmatics and Language Learning Monograph Series, vol. 3, edited by L. Bouton, 64-77. Urbana-Champaign, IL: Division of English as an International Language, University of Illinois. —. 1994. “Conversational implicature in the second language: Learned slowly when not deliberately taught.” Journal of Pragmatics 22: 15767. Brock, M. N., and Y. Nagasaka. 2005. “Teaching Pragmatics in the EFL Classroom? SURE You Can!” TESL Reporter 38 (1): 17-26. Fantini, A. E., ed. 1997. New ways in teaching culture. Alexandria, VA: TESOL International Association. Fernández Amaya, L. 2008. “Teaching culture: is it possible to avoid pragmatic failure?” Revista Alicantina de Estudios Ingleses 21: 11-24. Grice, H. P. 1975. “Logic and conversation.” In Syntax and Semantics volume 3: Speech Acts, edited by P. Cole, and J. L. Morgan, 41-58. New York: Academic Press.

540

Chapter Twenty One

Grossi, V. 2009. “Teaching pragmatic competence: Compliments and compliment responses in the ESL classroom.” Prospect Journal 24 (2): 53-62. Gumperz, J. J. 1982. Discourse strategies. Cambridge: Cambridge University Press. —. 1996. “The linguistic and cultural relativity of conversational inference.” In Rethinking linguistic relativity, edited by J. Gumperz, and S. Levinson, 374-400. Cambridge: Cambridge University Press. Halenko, N., and C. Jones. 2011. “Teaching pragmatic awareness of spoken requests to Chinese EAP learners in the UK: Is explicit instruction effective?” System 39 (2): 240-250. House, J. 1996. “Developing pragmatic fluency in English as a foreign language.” Studies in Second Language Acquisition 18: 225-252. Ishihara, N. 2010. “Instructional pragmatics: Bridging teaching, research, and teacher education.” Language and Linguistics Compass 4 (10): 938-953. Ishihara, N., and A. D. Cohen. 2010. Teaching and Learning Pragmatics: Where language and culture meet. Harlow, UK: Longman/Pearson Education. Kasper, G. 1997. “Can pragmatic competence be taught?” NetWork 6: 105-119. Knutson, E. M. 2006. “Cross-cultural awareness for second/foreign language learners.” Canadian Modern Language Review/La Revue canadienne des langues vivantes 62 (4): 591-610. Kordiü, S. 1991. “Konverzacijske implikature.” Suvremena lingvistika 17: 87-96. Kyriacou, C. 1988. Temeljna nastavna umijeüa: metodiþki priruþnik za uspješno pouþavanje i uþenje. Zagreb: Educa. Maley, A., and A. Duff. 2002. Drama techniques in language learning. Cambridge: Cambridge University Press. Martinko, J. 2012. “Radionica – metoda interaktivnog uþenja i pouþavanja odraslih.” Andragoški glasnik Glasilo Hrvatskog andragoškog društva 16 (2): 165-174. Mirzaei, A., and M. Esmaeili. 2013. “The effects of planned instruction on Iranian L2 learners’ interlanguage pragmatic development.” International Journal of Society, Culture & Language 1 (1): 89-100. Murray, J. C. 2011. “Do Bears Fly? Revisiting Conversational Implicature in Instructional Pragmatics.” TESL-EJ 15 (2). http://www.teslej.org/wordpress/issues/volume15/ej58/ej58a4/. Nguyen, T. T. M., T. H. Pham, and M. T. Pham. 2012. “The relative effects of explicit and implicit form-focused instruction on the

Conversational Implicatures in the Croatian EFL Classroom

541

development of L2 pragmatic competence.” Journal of pragmatics 44 (4): 416-434. Ochs, E. 1993. “Constructing social identity: A language socialization perspective.” Research on Language and Social Interaction 26 (3): 287-306. O’Sullivan, K. 1994. Understanding ways. Sydney: Hale & Iremonger. Rafieyan, V., N. B. A. Majid, and L. S. Eng. 2013. “Relationship between Attitude toward Target Language Culture Instruction and Pragmatic Comprehension Development.” English Language Teaching 6 (8): 125-132. Richards, J. C. 2011. Competence and performance in language teaching. Cambridge: Cambridge University Press. Robinson, T. 2011. “Rock and Roll English Teaching: Content-Based Cultural Workshops.” English Teaching Forum 49 (4): 14-23. Washington, DC: US Department of State, Bureau of Educational and Cultural Affairs. Rose, K. R. 2005. “On the effects of instruction in second language pragmatics.” System 33 (3): 385-399. Rose, K. R., and C. Ng. 2001. “Inductive and deductive teaching of compliments and compliment responses.” In Pragmatics in language teaching, edited by K. R. Rose, and G. Kasper, 145-170. Cambridge: Cambridge University Press. Sarçoban, A., and G. Çalúkan. 2011. “The influence of target culture on language learners.” Journal of Language and Linguistic Studies 7 (1): 7-17. Smith, B. 2009. “Learner strategies in L2 pragmatics: The case of Spanish compliment responses.” In Pragmatics Applied to Language Teaching and Learning, edited by R. Gómez Morón, L. Fernández Amaya, M. Padilla Cruz, and M. d. l. O. Hernàndez Lòpe. Newcastle upon Tyne: Cambridge Scholars Publishing. Taguchi, N. 2002. “An application of relevance theory to the analysis of L2 interpretation processes: The comprehension of indirect replies.” IRAL 40 (2): 151-176. Taguchi, N. 2005. “Comprehending implied meaning in English as a foreign language.” The Modern language journal 89 (4): 543-562. Takahashi, S., H. Roitblat. 1994. “Comprehension process of second language indirect requests.” Applied Psycholinguistics 15 (4): 475-506. Takimoto, M. 2008. “The Effects of Deductive and Inductive Instruction on the Development of Language Learners’ Pragmatic Competence.” The Modern Language Journal 92 (3): 369-386.

542

Chapter Twenty One

Tateyama, Y., Kasper, G., Mui, L. P., Tay, H. M., and Thananart, O. O. 1997. Explicit and implicit teaching of pragmatic routines. Pragmatics and language learning 8: 163-178. Thomas, J. 1983. “Cross-cultural pragmatic failure.” Applied Linguistics 4: 91-112. Tomalin, B., and S. Stempelski. 1993. Cultural awareness. Oxford: Oxford University Press. Wildner-Bassett, M. E. 1994. “Intercultural pragmatics and proficiency: ‘Polite’ noises for cultural appropriateness.” IRAL 32 (1): 3-17. Yates, L. 2004. “The secret rules of language: tackling pragmatics in the classroom.” Prospect: An Australian Journal of TESOL 19 (1): 3-21. Young, R. F. 2011. “Interactional competence in language learning, teaching, and testing.” Handbook of research in second language teaching and learning 2: 426-443. Young, R. 1997. “A Critical-Pragmatic Theory of Classroom Talk.” In Encyclopedia of Language and Education, vol. 3, Oral Discourse and Education, edited by B. Davies and D. Corson, 11-20. Dordrecht: Kluwek Academic Publishers.

DVDs Braff, Zach and John C. McGinley, Donald Faison, Ken Jenkins, Sarah Chalke. Scrubs. DVD. Created by Bill Lawrence. Burbank: Walt Disney Studios Home Entertainment, 2011. Carell, Steve andf Ed Helms, Rainn Wilson, John Krasinski, BJ Novak. The Office: An American Workplace. DVD. New York: NBC, 2013. Carrey, Jim and Zooey Deschanel. Yes Man. DVD. Directed by Peyton Reed. Burbank: Warner Home Video, 2009. Galecki, Johnny and Jim Parsons, Kaley Cuoco, Simon Helberg. The Big Bang Theory. DVD. Created by Chuck Lorre and Bill Prady. Burbank: Warner Home Video, 2013. Kutcher, Ashton and Topher Grace, Mila Kunis, Laura Prepon. That 70’s show. DVD. Created by Mark Brazill. Los Angeles: 20th Century Fox Home Entertainment, 2013. Laurie, Hugh and Omar Epps, Robert Sean Leonard, Jesse Spencer, Lisa Edelstein. House MD. DVD. Created by David Shore. Los Angeles: Universal Pictures, 2012.

Conversational Implicatures in the Croatian EFL Classroom

543

Appendix 1 Questionnaire on the use of conversational implicatures By completing this questionnaire you are taking part in a research conducted for the conference “Pragmatics on the go – Teaching and learning about pragmatics: principles, methods and practices.” Please, answer truthfully, the survey is anonymous and your responses will be used for research purposes only. Thank you for your time! Please, indicate the level of agreement with the following statements. 1.

I have already heard/learned before about conversational implicatures.

1. strongly disagree 2. disagree 3. neither disagree nor agree 4. agree 5. strongly agree 2.

Being able to interpret conversational implicatures is important for understanding the speaker’s intended meaning.

1. strongly disagree 2. disagree 3. neither disagree nor agree 4. agree 5. strongly agree 3.

Cultural context is important for the correct interpretation of conversational implicatures.

1. strongly disagree 2. disagree 3. neither disagree nor agree 4. agree 5. strongly agree 4.

Being familiar with conversational implicatures improves one’s skills in (write numbers 1-4 next to the answers, so that (1) is the least and (4) the most important answer):

a) overall language proficiency____ b) maintaining satisfying social relations with others____ c) translation____ d) learning about other cultures_____

Chapter Twenty One

544 5.

The workshop activities helped me to understand the phenomenon of conversational implicature.

1.strongly disagree 2. disagree 3. neither disagree nor agree 4. agree 5. strongly agree 6.

The assignments raised my awareness of culture- and context-dependent use of conversational implicatures.

1.strongly disagree 2. disagree 3. neither disagree nor agree 4. agree 5. strongly agree 7.

Conversational implicatures should be introduced into foreign language teaching.

1. strongly disagree 2. disagree 3.neither disagree nor agree 4. agree 5. strongly agree Which strategies and methods would be the most effective in teaching conversational implicatures? (Answer only in case you circled 4 or 5 in the previous question, write numbers 1 4 next to the answers, so that (1) is the least and (4) the most effective approach) 8.

a) theory based approach____ b) non-instructional, communication based approach____ c) context based approach (literature, films…)____ d) other: __________________________ _____

Comments: ____________________________________________________________ ____________________________________________________________ ____________________________________________________________

Conversational Implicatures in the Croatian EFL Classroom

545

Notes 1

The discussion partially encompassed the inferring process as well, but it was not overtly introduced in order to maintain the focus on implicatures. 2 The reason for this was that several authors (O’Sullivan 1994, Fantini 1997, Tomalin and Stempelski 1993, Bennett 1997) highlight the importance of activities that raise cultural awareness by exploring social, political, historical and institutional circumstances, while at the same time boosting learners’ interest for both language and culture (Young 1997).

CHAPTER TWENTY TWO SOME – AND POSSIBLY ALL – ADULTS COMPUTE SCALAR IMPLICATURES YHARA M. FORMISANO

1. Introduction Recent research suggests that the ratio of scalar implicatures (hereafter SI) drawn by adults in their L1 is not 100%. In other words, there is a percentage of native speakers that interpret scalar terms such as some in their logical sense (“some, possibly all” in this case). Moreover, as Slabakova (2010) points out, L2 learners seem to perform better than native speakers on SI understanding. The aim of the present research is twofold. First, the investigation aims at verifying whether the percentage of logical interpretation of the scalar term some in L1 is due to a real preference for some understood as “possibly all,” or whether this is an effect of the testing method. Second, data from L2 learners are also analyzed to check whether they have a higher pragmatic competence compared to native speakers, something that seems highly unlikely. Results show that some is never given a logical interpretation, either in L1 or in L2, and non-target like answers are due to the conjuring up of alternative realities or are explainable in terms of misinterpretation of the items tested. Moreover, as expected, results show that pragmatic competence is not higher in L2 than in L1. This paper explores the interpretation of scalar implicatures in L1 and L2. The research is based on the notion of implicit meaning –in particular, as conveyed through scalar implicatures– put forward by Grice (1957; 1975) and more recent studies on their experimental validation. Scalar implicature processing has been widely investigated from different perspectives (see for example Bott and Noveck 2004; Chierchia 2004; Katsos 2008; Katsos and Cummins 2012). Some questions, though, remain open. It is not clear, in fact, why scalar terms are sometimes interpreted in

Chapter Twenty Two

548

their logical sense. This research aims at understanding why SIs are sometimes interpreted in their logical sense by analyzing the motivations participants provide to explain their answers. The paper is organized as follows: section 2 discusses the notion of scalar implicatures and previous research findings that are relevant to the present analysis. Section 3 is devoted to the description of the experiments conducted, in both L1 and L2, aimed at determining the interpretation of the scalar term some and the results of these experiments. Section 4 compares and discusses the results. Section 5 presents the conclusions.

2. Scalar implicatures The notion of implicature was first put forward by Paul Grice (1957; 1975). Grice recognizes the central role of inferential processing during utterance interpretation, and analyzes the different types of implicit meanings conveyed through an utterance. In his words, a speaker means something by uttering x if he “intended the utterance of x to produce some effect in an audience by means of the recognition of this intention” (Grice 1957, 385). He proposes the concept of implicature to provide an explanation of the reason why a logical term like and, for example, is interpreted in natural languages as “and then,” and to define the difference between a logical inference and a pragmatic inference. The fundamental difference that Grice underlines is between “what is said,” namely the truth-conditional content of an utterance, and “what is implicated,” the meaning that the speaker wants to convey, which is processed by the hearer by means of reasoning. As Levinson (2000, 171) puts it, “what is said is the input to the pragmatic reasoning responsible for output of implicatures: what is implicated is calculated on the basis of what is said.” Implicatures are a part of what is communicated without being part of what is “said” by the speaker. They are pragmatic inferences derived via the tacit assumption, shared by the speaker and his/her audience, that what is communicated has to be true, informative, relevant, and clear.1 Grice (1957, 1975) divides implicatures into two types, conventional and conversational implicatures. Conventional implicatures are non-truthconditional inferences that do not derive from superordinate pragmatic principles, but are conventionally attached to certain lexical items and, therefore, do not change according to the communicative context. An example is provided in (1). (1)

She studied very hard for her exam, therefore I think she’ll pass it. [therefore engenders the implicature that studying is linked in some way to passing an exam]

Some – and Possibly All – Adults Compute Scalar Implicatures

549

Conversational implicatures, on the other hand, have to be inferred on the basis of the communicative context and, depending on the degree of their context dependence, they are further divided in generalized conversational implicatures (GCIs) and particularized conversational implicatures (PCIs) conversational implicatures. GCIs are derived assuming that the context of the utterance is an ordinary conversation (see example (2)), while PCIs are highly context-dependent and can only be interpreted with reference to the specific context in which they are generated (example (3)). (2)

She went home and had dinner [+> she went home and then had dinner]

(3)

A: Would you like some coffee? B: Coffee would keep me awake [+> No – only in a context in which A knows that B would like to sleep]

Scalar implicatures belong to the GCI category. They are generated by items belonging to an ordered scale (Horn 1972), the use of which implicates that the use of a stronger item on that scale is not possible in terms of relevance of what is being communicated. Scales can be generated for example by quantifiers (all vs. some), adverbs (always vs. often), modals (may vs. must), and predicates (freezing vs. cold). Consider: (4)

Some elephants have trunks

This utterance is true from a logical point of view given that, as a logical operator, some is interpreted as “some, possibly all,” and thus it is logically true that some elephants have trunks because all of them do. Under the Maxim of Quantity,2 however, this interpretation is not possible. Following this maxim, the contribution has to be as informative as required; thus, the choice of a weak term like some pragmatically means that the interpretation of “possibly all” is ruled out. The pragmatic interpretation of some is “some, not all” and therefore the sentence in (4) is underinformative and, thus, pragmatically infelicitous given that all elephants have trunks. The processing of scalar implicatures in L1 varies depending on the age of the individuals interpreting the utterance. Noveck (2001) conducted an experiment with adults and found that 59% of the time the utterance in (4) was considered to be false. Bezuidenhout and Morris (2004) investigated scalar implicatures in adult L1 with an eye-tracking

550

Chapter Twenty Two

experiment. The authors monitored individuals’ gazing time of sentence pairs like the ones in (5): (5a) Some books had colour pictures. In fact all of them did, which is why the teachers liked them. (5b) The books had colour pictures. In fact all of them did, which is why the teachers liked them.

They noticed that the gazing time on all was longer for the sentences containing some compared to the pair without it. This provided evidence in favour of the hypothesis of a pragmatic interpretation of the quantifier some as “not all,” which had to be re-analyzed after all was read. The reverse pattern emerges when investigating the rate to which children draw scalar implicature. Noveck (2001) and Chierchia et al. (2004) both report the preference for a logical over pragmatic interpretation of quantifiers and logical operators in children, pointing at what the authors call Pragmatic Delay Hypothesis (Chierchia et al. 2004, 297). This hypothesis accounts for children’s logical interpretation of quantifiers in terms of failure to apply the Gricean Maxim of Quantity related to a developmental factor. Nevertheless, as discussed in Guasti et al. (2005), children behave in an adult-like fashion and show a high percentage of scalar implicature computation if provided with explicit contextual support for sentence interpretation. This has also been confirmed by Feeney et al. (2004), who examined the interpretation of some in 7-8 year-old children, and found that they were able to associate the pragmatic meaning of “not all” in pragmatically rich contexts. The authors concluded that “children’s sensitivity to the scalar implicature associated with some is largely dependent on context” (Feeney et al. 2004, 130). Moreover, as Papafragou and Musolino (2003) point out, children do not treat all scalar terms alike and their processing of SIs is connected to the awareness of the goal of the task at hand. These results show that testing methods affect the way in which linguistic phenomena in general, and implicatures in particular, are investigated, leading to possibly divergent, or at least complementary, findings. Implicature processing has also been investigated in second language acquisition studies (see for example Bouton 1994; Ifantidou 2013; Taguchi, Shuai, and Yan 2013). In this field, though, attention devoted to SI processing has been scant. Slabakova (2010) analyzed scalar implicature processing in a group of Korean learners of English both in their L1 and in their L2 to assess the rate of pragmatic interpretation of the scalar term some. Not only did her participants perform better in their L2 than in their L1, but they also showed a higher rate of pragmatic

Some – and Possibly All – Adults Compute Scalar Implicatures

551

interpretation of some in English compared to a group of English native speakers that judged underinformative sentences (i.e., pragmatically infelicitous) true 54% of the time, hence attributing some the meaning of “some, possibly all.” Even when provided with adequate contextual support, L2 learners in this experiment showed higher pragmatic accuracy, interpreting some as “not all” 90% of the time compared to native speakers who did so 62.5% of the time. The author concluded that “learners behave more pragmatically in their second language than in the native language, no matter if the judgements are supported by context or not” (Slabakova 2010, 2456). Slabakova puts forward the processing limitation hypothesis, according to which L2 learners lack some processing resources and so they “may not be able to come up with alternative contexts as often as they do in their native language, hence they derive L2 implicatures significantly more often than native adults” (Slabakova 2010, 2457). The findings and the interpretation proposed seem surprising, both for the relatively high percentages of responses showing a logical interpretation of the scalar term in English by native speakers (55.4% in the case of sentences without a context, and 37.5% in the case of sentences in pragmatically enriched contexts), and for the higher pragmatic accuracy displayed by L2 learners. Thus, the need emerged to investigate further the nature of this residual “logical” interpretation of scalar some. In addition to this, I argue that the suggested data interpretation according to which coming up with alternative contexts corresponds to a logical interpretation of the scalar item is likely to be flawed. This is because, if a participant agrees with an underinformative sentence as “Some elephants have trunks” explaining that an elephant without the trunk might actually exist, (s)he is still interpreting some as “some, not all.” The only motivation that could justify the proposal of a logical meaning attribution to some would be something like i. “I agree because they all do” or ii. “I agree because if all elephants have trunks, then some of them do too.” This paper examines the motivations that participants provide for their interpretation of sentences containing the quantifier some. The goal is to ascertain whether evidence can indeed be found of logical meaning attribution when the interpretation selected is “some, possibly all.” In other words, motivations are analyzed to check for the presence, or absence, of answers like those in i. and ii. that would be considered as evidence of a logical meaning attribution.

552

Chapter Twenty Two

3. The present research The present study has a twofold aim and thus reports on two experiments. In the first experiment, I tested adult native speakers of Italian on their interpretation of the scalar term some to verify how often they would draw the scalar implicature connected to it and to analyze the motivations provided to ascertain the presence of responses that seem to conform to the logical interpretation of the term. In the second experiment, I tested adult Italian learners of English to determine the extent to which they would provide a pragmatic interpretation of some in English and to check whether their pragmatic competence in L2 would be higher than that of the group of Italians tested in their native language. The results of both studies are then compared and discussed.

3.1. Experiment 1 This experiment was aimed at verifying the frequency with which SIs are drawn by adult L1 speakers. Moreover, it aimed at analyzing the answers provided by participants so as to examine the motivations they gave for judging as true pragmatically underinformative sentences such as (4). In keeping with previous research, it was hypothesized that the acceptance of pragmatically infelicitous sentences could be explained in terms of a logical interpretation of the scalar term, or the conjuring up of alternative realities. Participants could thus agree with the sentence “some elephants have trunks,” even though it is underinformative, by explaining that some could mean “some, possibly all,” conforming to logical meaning attribution. Conversely, they could attribute the pragmatic meaning to some and disagree with the underinformative sentence, or alternatively, they could agree with it, providing an alternative interpretation by conjuring up a different context, for example the existence of sick elephants that have lost their trunks. In some cases, pictures accompanied the sentences to provide a context for interpretation and they served as a means of verifying the influence of the context on the rate of scalar implicature drawing. 3.1.1. Participants Thirty-four Italian native speakers whose age ranged from 18 to 56 (mean age of 27) participated in the study. They were recruited through acquaintances and in a high school in Perugia, Italy,3 and volunteered to take part in the research.

Some – and Possibly All – Adults Compute Scalar Implicatures

553

3.1.2. Method I administered a test made up of 43 PowerPointTM slides. The test was structured as follows: 1 opening and 1 closing slide; 1 slide with the instructions for the test;4 and 40 slides testing participants on their interpretation of sentences. Each of the 40 slides contained a sentence in Italian that tested subjects’ interpretation of different items. The sentences were divided as follows:5 -

8 with a felicitous (i.e., pragmatically adequate) some (e.g., Some bears are white). 8 with an infelicitous (i.e., pragmatically inadequate) some (e.g., Some spiders have eight legs). 8 with a true all (e.g., All rabbits have ears). 8 with a false all (e.g., All houses have a fireplace). 8 absurd sentences with either some or all (e.g., All computers have noses; Some lions study linguistics).

Half of the slides (except for those with the absurd sentences) included pictures that provided a context for interpretation. This was meant to ascertain the possible influence of the context on the drawing of scalar implicatures. Thus, of the 8 slides with a felicitous some, 4 slides had a picture (e.g., “Some cookies are round” was accompanied by a picture of differently shaped cookies, some of which were round) and 4 slides only contained the sentences; the same applied to the 8 slides with infelicitous some, true all, and false all. The slides were presented in a random order so as to avoid any priming effect. Participants were tested one at a time, in a quiet environment, and were asked to sit in front of the computer screen. Before starting the test, they were asked to read the instructions carefully and the researcher made sure that they had thoroughly read and understood them before they actually took the test. Instructions required the participants to read the sentences on the slides and say whether they agreed or not with the content of the sentence and, also, to then explain why or why not. In order to relieve pressure, they were also told that a “don’t know” answer was possible, if they were not sure what to respond. They were instructed on the fact that in the slides with a picture and a sentence, the sentence referred to the pictures, and so that they had to refer to that for sentence interpretation, while for text-only slides they were asked to refer to their common world

554

Chapter Twenty Two

knowledge. The researcher was with the participants during the test and took note of their answers and explanations on an answer sheet. 3.1.3. Results Pragmatic interpretation of the scalar term some was marked as “target answer;” vice versa, judging underinformative sentences as true was marked as “non target.” Finally, “I don’t know” responses were marked as such. Table 1 presents the percentages of the three type of responses on the four items tested. Table 1 shows that the adult Italian speakers tested in their native language mostly gave target responses. It also shows that all the items tested had a percentage of non-target answers (i.e., failure to draw the scalar implicature generated by some), which account for a minority of the data, but are not equally distributed across the four items tested. In particular, as found in previous research (Chierchia et al. 2004; Noveck 2001; Slabakova 2010), sentences that tested participants’ processing of the scalar implicature deriving from some proved to be the most difficult and resulted in the highest percentage of non-target answers (18%). Finally, no “I don’t know” responses were recorded. Thus, at first glance, results of the first experiment seemed to be in keeping with data from previous research. TABLE 1. Percentages of target, non-target, and “I don’t know” responses in Italian L1 Tested item Felicitous some Infelicitous some True all False all

Target answers 98.5% 82.0% 92.6% 96.9%

Non-target answers 1.5% 18.0% 7.4% 3.1%

Don’t know 0% 0% 0% 0%

Previous researchers justified similar findings by stating that scalar terms are sometimes given a logical interpretation and thus that participants accept underinformative sentences such as “Some elephants have trunks” because they assign some the “some, and possibly all” meaning. To check whether this interpretation also applied to my findings, I examined the explanations that the participants had provided for their non-target responses to infelicitous some. Various reasons were offered, but no participant provided a logical interpretation of the scalar item, that is none of them offered a logical meaning attribution to it. Non-target answers were examined also to check for context dependence effects. Providing a picture for sentence interpretation was thought to foster the

Som me – and Possibbly All – Adultss Compute Scallar Implicaturess

555

derivation oof scalar impllicature. Insteead, the particcipants appearred to be misled by thhe pictures in some cases (half ( of the noon-target answ wers were on sentencess accompanied by pictures)), either becauuse they did not n follow the instructions and ignnored the picture, or becaause they tho ought the picture was not clear enouugh. Text-onlly underinform mative sentencces that somee participants judged j as true are the ffollowing:6 -

-

-

“Som me cars have foour wheels:” th he participantss who said thaat they agreed with the conntent of the sen ntence explaineed that there arre cars oken cars mayy have less than four that hhave three wheeels or that bro wheells. “Som me mothers havee at least one child:” the particcipants who saaid that they aagreed with the content of th he sentence expplained, for exaample, that a mother is alw ways a motherr even after thee loss of a child, so motheers may have noo children and still s be motherss. “Som me books have words w in them:”” the participannts who said thaat they agreed with the conttent of the senttence explainedd that there are books that oonly have picturres in them.

As already stated, thee presence of a picture did nnot always co ontributed to the drawiing of the scallar implicaturee. Half of the non-target an nswers on infelicitous ssome were in sentences acccompanied byy a picture, as shown in Figure 1 andd Figure 2.

FIGURE 1. Som me – Example of o sentence accompanied by piictures

The slide in Figure 1 showed four pictures of oopen suitcasess with the sentence “some suitcases are open.” Participants who agreed with the content of thhe sentence gave g diverse explanations e ffor their interp pretation. Some had problems wiith the pictu ures themselvves. In one case, c the

Chapter Tweenty Two

556

participant w was not sure whether the suitcase wass actually opeen, while others did nnot reference to the picturees and said “ssometimes you are not able to closee them” or “suuitcases are meant to be clossed.”

FIGURE 2. Som me – Another example e of senteence accompannied by pictures

The slidee in Figure 2 showed three pictures of giirls wearing sk kirts with the sentence “some girlls are wearin ng a skirt.” IIn this case, too, the participants gave two kinds of explanattions as to whhy they agreed d with the sentence prrovided. Som me said that it i was not cclear from the picture whether onlly some or all girls had sk kirts, and otheers said “they y can also wear trouserrs,” without making m referen nce to the pictuures. It appearrs from these explanations that the quanntifier some was w never given its loggical interprettation. All the non-target annswers were due d either to conjuringg up differennt realities orr to not folloowing the insstructions thoroughly. Even in thhe 18% of non-target n annswers, the pragmatic p interpretatioon of some held. h Thereforre, with regaards to the lo ogical vs. pragmatic aattribution of meaning to the quantifierr some, the pragmatic p interpretatioon was presentt 100% of the time. The sam me is true of the t non-targett answers to tthe sentences with the felicitous soome. In this case, c the samee participant ddid not agree with the following teext-only sentennces: -

“Som me books are innteresting:” the participant expplained that he thinks all boooks are interestting. “Som me dogs have loong ears:” the participant expplained that “alll dogs have llong ears, it is people p who cut them.”

Again, som me was attributed its prag gmatic meaniing here, eveen if the expected answer would have been “agrree.”

Some – and Possibly All – Adults Compute Scalar Implicatures

557

The text-only sentences with true and false all that gave rise to nontarget answers were the following: -

-

-

“All trees have roots:” the participants who disagreed with the content of the sentence said that they did not know whether all trees have roots or only some of them do. “All birds have wings:” the participants who disagreed with the content of the sentence said that they did not know every kind of bird, so they could not be sure of the fact that all birds have wings. “All men smoke:” only one participant agreed with the content of the sentence saying “they don’t smoke only cigarettes.” “All students speak Spanish:” the participants who agreed with the content of the sentence said that they could not know if actually all students speak Spanish.

The slides accompanying true and false all sentences, which were considered “not clear” or were ignored, were the following. -

-

-

-

Slide: visual: picture of three buildings with balconies; text: “all the buildings have a balcony.” The participants who disagreed with the content of the sentence said that some buildings do not have balconies. Slide: visual: picture of a bunch of roses; text: “all the flowers are roses.” The participants who disagreed with the content of the sentence said that there are many kinds of flowers. Slide: visual: picture of three bedrooms, of which only two had a double bed and one had a single bed; text: “all the bedrooms have a double bed.” The participants who agreed with the content of the sentence said that all the bedrooms showed in the pictures had double beds. Slide: visual: picture of three cups, of which only two had a teaspoon with them and one did not; text: “all the cups have a teaspoon.” The participants who agreed with the content of the sentence said that all the cups shown in the pictures had a teaspoon.

The above findings showed that the pragmatic interpretation of some held in all cases. All, too, was always interpreted as contrasting with some. None of the participants tested in their native language provided a logical interpretation of the quantifier under analysis. The next section describes the data of the second experiment that tested the interpretation of some in L2.

558

Chapter Twenty Two

3.2. Experiment 2 The second experiment was a replica of the first one. The only variable that changed was the language in which participants were tested. The aim of this experiment was to verify whether, as shown in Slabakova (2010), L2 learners are more pragmatically accurate than native speakers in the processing of scalar items. Such an outcome seemed highly unlikely, considering that linguistic competence in L2 is never higher than in L1. Moreover, as discussed in section 3.1.3, participants tested in their L1 drew the scalar implicature in some in 100% of the cases under analysis. 3.2.1. Participants Thirty-four Italian native speakers learning English as L2 participated in the study. Their age ranged from 18 to 56 (mean age 27), and their English level was B1. They were recruited through acquaintances and in a high school in Perugia, Italy, and volunteered to take part in the research. 3.2.2. Method The method was the same as that used in Experiment 1 (see section 3.1.2). The original test was translated into English and the researcher made sure participants understood both the instructions given to them and the sentences. Considering that the test was not about L2 production, and to avoid putting the participants under too much pressure, they were told that they could provide explanations of their answers in the language they preferred (English or Italian). 3.2.3. Results The results of the experiment conducted with the L2 learners showed a higher percentage of non-target answers on items that tested the processing of the scalar implicature, namely those about infelicitous some, compared to the other test items. Table 2 presents the general results of the second experiment. Table 2 shows that the item that proved to be the most difficult was infelicitous some (18%), followed by true all (6.9%). Felicitous some and false all were the least problematic, the two items in fact showed a very low percentage of non-target answers. No “I don’t know” responses were recorded.

Some – and Possibly All – Adults Compute Scalar Implicatures

559

TABLE 2. Percentages of target, non-target, and “I don’t know” responses in English L2 Tested item Felicitous some Infelicitous some True all False all

Target answers 99.5% 82.0% 93.1% 99.5%

Non-target answers 0.5% 18.0% 6.9% 0.5%

Don’t know 0% 0% 0% 0%

As it had been done in Experiment 1, the comments to non-target answers were analyzed to check for possible logical explanations motivating the agreement with the underinformative sentences, and to examine the influence of pictures in the process of scalar implicature processing. Again, the motivations provided were diversified, and they had nothing to do with a logical interpretation of the scalar term. Text only underinformative sentences with whose content some participants agreed are the following: -

-

-

-

“Some cars have four wheels:” the explanations provided for expressing agreement with the content of this sentence were, for example “some do and some don’t,” or “after an accident a car may have less than four wheels.” “Some mothers have at least one child:” the participants who agreed with the content of the sentence explained, for example, that “a lot of mothers have only one child” or “they can decide to have more.” “Some books have words in them:” the participants who agreed with the content of the sentence explained, for example, “some have words and some have pictures. “Some turtles have shells:” the explanations provided made reference to the existence of turtles without shells.

In the same way as in Experiment 1, the presence of pictures did not always affect the interpretation of the sentences. Slightly more than one third of the non-target answers were provided on sentences accompanied by a picture. They occurred in the following cases: -

-

Slide: visual: picture of four open suitcases; text: “some suitcases are open.” The participants who agreed with the content of the sentence explained that one of the suitcases was closed, or that it was not clear from the picture whether they were all open. Slide: visual: picture of three girls wearing skirts; text: “some girls are wearing a skirt.” The participants did not make reference to the picture, providing explanations like “skirts are typical for girls.”

Chapter Tweenty Two

560

These tw wo items also proved difficcult to interprret in Experim ment 1. It could be thhus hypothesized that theere was someething uncleaar in the pictures theemselves (seee 3.1.3), ratheer than somee linguistic processing problem. Coonsidering thee motivations given in fact,, it appeared clear that the scalar teerm some waas given a praagmatic interppretation 100 0% of the time by parrticipants testeed in their seccond languagge as well. Th his is true also for the oother non-targget answers. In the caase of true alll, non-target answers a on texxt-only senten nces were due to the cconjuring up of o different realities, for exxample, wheree birds do not have w wings (for thee sentence “A All birds havee wings”). Non-target N answers on sentences acccompanied by b a picture were again motivated m either by paarticipants’ ignnoring the piccture and makking reference to their general worrld knowledgee (a few of them did), or bby their stating g that the picture was not clear, as in Figure 3:7

s accom mpanied by a piccture. FIGURE 3. Alll – Example of sentence

In the caase of Figure 3, 3 for examplee, a participannt said “they have h their eyes closed, but I cannot be b sure that alll of them are actually sleep ping.” Infelicitoous some and false all had very low perrcentages of non-target n responses. T The relevant teest items are th he following: -

-

Slide:: visual: picturee of four children eating, onlyy some of which h were eatingg ice cream; teext: “Some chilldren are eatingg an ice cream.” The explan anation providedd was that all th he children weree eating ice cream. Slide:: text: “All studdents speak Sp panish.” The paarticipant agreed with the coontent of the sentence s explaining that it is true if they stu udy in Spainn.

In the seecond experim ment too, non ne of the item ms tested wass given a logical interrpretation conntrary to whaat was foundd in previous research

Some – and Possibly All – Adults Compute Scalar Implicatures

561

(Slabakova 2010). The analysis of the explanations proved to be a fundamental tool to interpret the data and, through that analysis, I was able to conclude that the rate to which scalar implicature was drawn was 100% in participants tested in their L2 as in those tested in their L1.

4. Discussion Results of the two groups tested in this study would seem to corroborate the hypothesis proposed in Slabakova (2010). L2 learners showed an overall better performance compared to participants tested in their native language. Looking at the results on infelicitous some, though, there appeared to be no difference between the two groups (18% of nontarget answers for both groups). Table 3 compares the non-target results of the L1 and the L2 participants. TABLE 3. Percentages of non-target answers in L1 and L2. Tested item Felicitous some Infelicitous some True all False all

L1 1.5% 18.0% 7.4% 3.1%

L2 0.5% 18.0% 6.9% 0.5%

As shown in Table 3, the two groups performed in the same way on the processing of the scalar implicature. This result seemed to go against the processing limitation hypothesis put forward in Slabakova (2010). Not only is the rate of computation of scalar implicatures the same for both groups but, also, by analyzing the participants’ motivations for their answers, it was possible to understand that participants, both in their L1 and in their L2, never interpreted the scalar item in its logical sense of “some, possibly all.” These results are opposite to those discussed in Chierchia et al. (2004) and Noveck (2001), who found a percentage of logical meaning attribution to some. In addition to the results on some, the percentages of non-target answers on the other items tested were never motivated by a logical interpretation. In the case of false all, for example, participants who disagreed with the statement motivated their answers by pointing out the difference between some and all and emphasizing that the sentence could be applied to “only some.” Moreover, from personal communication with some of the participants after the administration of the test, it was clear that they completely ignored the logical meaning of some and kept on insisting that “if you use some it means that that thing is not true for all of

562

Chapter Twenty Two

them” (C. 27; 3 – L2), “the sentence ‘some elephants have trunks’ could only be true if you imagine that there are some sick elephants that don’t” (G. 49; 8 – L1), or “only educated people know or understand the logical meaning of some” (P. 51; 2 – L1). Instructing participants to interpret the sentences according to their basic world knowledge and everyday language use provided a basis for attributing to some its scalar meaning. The presence of pictures providing a context for sentence interpretation did apparently favour the pragmatic interpretation of the scalar item. This reflects the findings of Feeney et al. (2004) and Guasti et al. (2005), according to whom, providing participants with a pragmatically adequate context fosters the drawing of implicatures. Nevertheless, pictures did not always foster the pragmatic interpretation in the way that was expected. Half of the non-target answers in L1 and a third in L2 were on sentences accompanied by a picture. However, the interpretation problems could not be ascribed to the attribution of a logical meaning to the sentences. The pictures were either ignored, and participants referred to general world knowledge, or they were considered not clear. Moreover, the same test items turned out to be problematic in L1 and L2, which could be due to an intrinsic difficulty or ambiguity with the sentences or the pictures themselves, rather than their linguistic interpretation. It could also be hypothesized that the lack of determiners or demonstratives in the sentences accompanied by pictures (see note 7), negatively influenced sentence interpretation and hindered comprehension. In both cases, though, the analysis of the motivations showed that the pragmatic interpretation of the scalar “some” always held. The general (slightly) better performance of participants tested in their L2, also found in Slabakova (2010), could be explained in terms of motivation and familiarity with being administered language tests. L2 learners are used to being tested on their knowledge of the L2, thus being administered a language test is not a new experience for them. They know that they have to show their proficiency in the language they are learning, and so they pay more attention to the task in general, and to instructions and testing items in particular. On the other hand, native speakers of a language are not as used to being tested on their mother language so, when they are administered a test in L1, they might think that they are being tested on their reasoning abilities rather than on their proficiency in L1 and, thus, they could be more prone to conjure up alternative explanations, or provide unusual interpretation of sentences. Moreover, it has to be pointed out that L2 learners who participated in the study were never given specific instructions about how to interpret

Some – and Possibly All – Adults Compute Scalar Implicatures

563

some in English. They were only taught its meaning and its use as opposed to any. Thus, their understanding and correct use of its pragmatic meaning of “not all” could either derive from a universal mechanism for the computation of scalar items or from a transfer from their L1.8

5. Conclusions This paper reports the results of two experiments conducted to investigate the interpretation of the scalar term some in L1 and L2. This quantifier has both the logical meaning of “some, possibly all” and the pragmatic meaning of “some, not all.” In everyday language use, people are expected to interpret some in its pragmatic sense, considering that communicative exchanges conform to the Cooperative Principle and the Conversational Maxims (Grice 1957, 1975, 1989). Nevertheless, previous research findings (Chierchia et al. 2004; Noveck 2001; Slabakova 2010, among others) concur with the difficulty on the part of native speakers to interpret this kind of implicit meaning and put forward the idea that in some cases both adults and children do not compute the scalar implicature and, thus, attribute the logical meaning to the quantifier. Children have been shown to accept the logical interpretation of some more often than adults, who, however, do not draw the scalar implicature 100% of the time in their L1. L2 learners, on the other hand, have been shown to have a higher pragmatic accuracy in the interpretation of this scalar item in their L2 compared to their L1, and to a group of native speakers (Slabakova 2010). The two experiments reported in this paper tested adults in L1 and L2 on their interpretation of the quantifier some to verify how it is interpreted (logically vs. pragmatically). The test not only asked participants whether they agreed or not with the content of the sentences presented to them but, also, to motivate their answers. Overall results agree with previous research (Slabakova 2010) in that participants tested in their L2 generally performed better than participants tested in their L1. Nonetheless, the detailed analysis of the results, which involved examination of the motivations provided, showed that neither in L1 nor in L2 did participants ever attribute the logical meaning to the quantifier some. The percentage of non-target answers on the relevant items tested was due to the conjuring up of alternative realities or to the ambiguity of the items being tested. In both cases, the pragmatic interpretation of the scalar term held. The analysis of the explanations provided as motivation of the answers proved to be a fundamental tool to interpret the data. By considering these explanations, it was possible to understand that the slightly worse

564

Chapter Twenty Two

performance of the L1 participants was due to external factors such as motivation or attention, and not to a real preference for a logical meaning attribution. Analysis of participants’ explanations could also be a useful tool in research about children’s computation of scalar implicatures, even though children might not have the kind of meta-linguistic awareness needed to justify the answers they provide. Further research could also aim at analyzing production data to check for the use of quantifiers and the meaning attributed to them. Moreover, research is needed to investigate the nature of scalar implicature processing in L2 to verify whether it derives from a universal mechanism of implicature calculation, or if it depends on a transfer of interpretative practices from L1 to L2.

References Bezuidenhout, A. L., and R. K. Morris. 2004. “Implicature, Relevance and Default Pragmatic Inference.” In Experimental Pragmatics, edited by I. A. Noveck, and W. Sperber, 257-282. London: Palgrave. Bott, L., and I. A. Noveck. 2004. “Some utterances are underinformative: the onset and time course of scalar inferences.” Journal of memory and language 51: 437-457. Bouton, L. F. 1994. “Conversational implicature in a second language: learned slowly when not deliberately.” Journal of Pragmatics 22 (2): 157-167. Breheny, R., N. Katsos, and J. Williams. 2006. “Are generalised scalar implicatures generated by default? An on-line investigation into the role of context in generating pragmatic inferences.” Cognition 100: 434-463. Carston, R. 1998. “Informativeness, relevance, and scalar implicatures.” In Relevance Theory: Applications and Implications, edited by R. Carston, and S. Uchida, 179-236. Amsterdam: Benjamins. Chierchia, G. 2004. “Scalar implicatures, polarity phenomena, and the syntax-pragmatics interface.” In Structures and Beyond, edited by A. Belletti, 39-103. Oxford: Oxford University Press. Chierchia, G., M. T. Guasti, A. Gualmini, L. Meroni, S. Crain, and F. Foppolo. 2004. “Semantic and Pragmatic Competence in Children’s and Adults’ Comprehension of Or.” In Experimental Pragmatics, edited by I. A. Noveck, and D. Sperber, 283-300. New York: Palgrave Macmillan. Feeney, A., S. Scrafton, A. Duckworth, and S. J. Handley. 2004. “The Story of Some: Everyday Pragmatic Inference by Children and

Some – and Possibly All – Adults Compute Scalar Implicatures

565

Adults.” Canadian Journal of Experimental Psychology 58 (2): 121132. Grice, H. P. 1957. “Meaning.” The Philosophical Review 66 (3): 377-388. —. 1975. “Logic and Conversation.” In Syntax and Semantics, edited by P. Cole, and J. Morgan, 41-58. New York: Academic Press. —. 1989. Studies in the way of words. Cambridge, MA: Harvard University Press. Guasti, M. T., G. Chierchia, S. Crain, F. Foppolo, A. Gualmini, and L. Meroni. 2005. “Why children and adults sometimes (but not always) compute implicatures.” Language and cognitive processes 20 (5): 667696. Horn, L. 1972. On the semantic properties of logical operators in English. Bloomington: Indiana University Linguistics Club. Ifantidou, E. 2013. “Pragmatic competence and explicit instruction.” Journal of Pragmatics 59: 93-116. Katsos, N. 2008. “The Semantics/Pragmatics Interface from an Experimental Perspective: the Case of Scalar Implicature.” Synthese 165: 385-401. Katsos, N., and C. Cummins. 2012. “Scalar implicature: theory, processing and acquisition.” Nouveaux Cahiers de Linguistique Française 30: 39-52. Kook-Hee, G., H. Marsden, and M. Whong. 2011. “L2 Acquisition of any: Negative Evidence, Negative Implicature and Negative L1 Transfer.” In Selected Proceedings of the 2010 Second Language Research Forum, edited by G. Granena, J. Koeth, S. Lee-Ellis, A. Lukyanchenko, G. Prieto Botana, and E. Rhoades, 29-39. Somerville, MA: Cascadilla Proceedings Project. http://www.lingref.com/cpp/slrf/2010/paper2613.pdf. Levinson, S. C. 2000. Presumptive Meanings. The Theory of Generalized Conversational Implicature. Cambridge, MA: the MIT Press. Noveck, I. A. 2001. “When children are more logical than adults: Experimental investigation of Scalar Implicatures.” Cognition 78 (2): 165-188. Papafragou, A., and J. Musolino. 2003. “Scalar implicatures: Experiments at the semantics/pragmatics interface.” Cognition 86: 253-282. Sbisà, M. 2007. Detto non detto. Le forme della comunicazione implicita. Bari: Laterza. Slabakova, R. 2010. “Scalar Implicatures in second language acquisition.” Lingua 120: 2444-2462. Sperber, D., and D. Wilson. 1986. Relevance: communication and cognition. Oxford: Blackwell.

566

Chapter Twenty Two

—. 1995. Relevance: communication and cognition, 2nd ed. Oxford: Blackwell. —. 2004. “Relevance Theory.” In Handbook of Pragmatics, edited by L. R. Horn, and G. Ward, 607-632. Oxford: Blackwell. Taguchi, N., L. Shuai, and L. Yan. 2013. “Comprehension of conversational implicature in L2 Chinese.” Pragmatics and Cognition 21 (1): 139-157. White, L. 2011. “Second language acquisition at the interfaces.” Lingua 121: 577-590.

Appendix 1. Instructions for the test in Italian Ci sono 40 frasi in questo test. A volte ci sono delle figure insieme alle frasi, in questi casi la frase si riferisce alla figura. Per ogni frase bisogna dire se si è d’accordo oppure no, e spiegare per quale motivo.

2. Instructions for the test in English There are 40 sentences in this test. Sometimes there are pictures together with the sentences. In those cases, the sentence refers to the picture. For each sentence, you are asked to say whether you agree or not with what is stated in it. You should also motivate your answer.

Notes 1

Namely, the assumption is that the speaker is conforming to the Cooperative Principle and to the Conversational Maxims of Quality, Quantity, Relevance, and Manner (see Grice 1957; 1975; 1989). 2 The traditional Gricean paradigm will be used throughout this study. I am aware of the theoretical debate on the nature of implicatures, namely whether they derive from the Maxims or from the principle of Optimal Relevance (Sperber and Wilson 1986,; 1995; 2004). I agree with Sbisà (2007) that the principle of relevance is an overly bare substitute of the Cooperative Principle and its Maxims, though recognizing that Relevance Theory is undoubtedly one of the most prominent theories in pragmatics. No position will be taken here instead on the Default view (Levinson 2000; Chierchia 2004) vs. the Context-driven Approach (Carston 1998;

Some – and Possibly All – Adults Compute Scalar Implicatures

567

Sperber and Wilson 1986; Breheny, Katsos, and Williams 2006) considering that this point is not relevant with regard to the data under analysis. 3 I am grateful to Laura Alberti for the recruitment of the latter group of participants. 4 See Appendix. 5 The number and type of items tested replicate the ones found in Slabakova (2010). 6 The original items, here translated to be consistent with the language of the article, were: ‘Alcune automobili hanno quattro ruote,’ ‘Alcune madri hanno almeno un figlio,’ ‘Alcuni libri contengono parole’. 7 Being aware of the grammatical non-appropriateness of the sentence accompanying the picture, for which ‘All the dogs are sleeping’ would have been the expected description, the absence of the determiner was a strategic choice aimed at keeping the same syntactic form in sentences with and without pictures. The same is true for the sentences in Italian, in which a determiner or a demonstrative would have clarified the relation between the sentence and the picture. 8 See White (2011) for interfaces and second language acquisition and Kook-Hee, Marsden, and Whong (2011) for the role of transfer on implicature processing.