303 86 2MB
English Pages 280 [278] Year 2014
ELT: Harmony and Diversity
ELT: Harmony and Diversity
Edited by
Christoph Haase and Natalia Orlova
ELT: Harmony and Diversity, Edited by Christoph Haase and Natalia Orlova This book first published 2014 Cambridge Scholars Publishing 12 Back Chapman Street, Newcastle upon Tyne, NE6 2XX, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2014 by Christoph Haase, Natalia Orlova 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): 1-4438-5506-5, ISBN (13): 978-1-4438-5506-8
CONTENTS Preface ....................................................................................................... vii Acknowledgements .................................................................................... ix Section 1: Issues in Grammar Teaching and EAP Do Grammar Exercises Help? Assessing the Effectiveness of Grammar Pedagogy David Newby ............................................................................................... 3 Analysing the Development of Academic Writing Skills in English as a Lingua Franca Olga Dontcheva-Navratilova and Renata Povolná ................................... 17 Teaching Effective Academic Writing: IF-Conditionals in Economics Research Articles Silvia Cacchiani......................................................................................... 55 Communicating Time: L2 Students’ Mental Maps of Tenses Zuzana Procházková ................................................................................. 71 Issues in L2 Morphological Productivity Christoph Haase ........................................................................................ 97 Section 2: Teaching Expressivity and Culture How to Address Properly: Students’ Awareness of English and Czech Conventions Josef NevaĜil and Blanka Babická ........................................................... 113 Deceptive Similarities between French and English: A Challenge to Official Bilingualism and ELT in Cameroon Samuel Atechi .......................................................................................... 123 Teaching Speaking Across Cultures Hana Suchánková .................................................................................... 141
vi
Contents
Why Do We Need the Humanities and the Arts: Also in the English Classroom? OndĜej Skovajsa....................................................................................... 153 Section 3: Methodology, Technology and ELT Non-Native Language Teachers' Self-Perceptions: A Beginning vs Experienced Teachers’ Perpective Natalia Orlova ......................................................................................... 163 The Ambivalence of the Internet as a Means of ELT Cornelia Neubert ..................................................................................... 183 Social Networking: Using Facebook in Teaching English Idioms Darah Tafazoli, Masoumeh Kafshgarsouteh and Erfan Jalali ................ 197 On the Usefulness of the Pre-Listening Stage from the Adult Learner’s Perspective Hana Ždímalová ...................................................................................... 207 Typology of Senior Foreign Language Education Participants at U3A UJEP Iva Koutská .............................................................................................. 223 Section 4: Short Communications in Teaching Presenting the Present Tense: The Networking Approach Stanislava Kaiserová ............................................................................... 243 Blogs in ELT: Developing Self-Evaluation, Autonomy and Responsibility of the Learner Jana Pavlíková ........................................................................................ 251 ELT with Moodle at the Non-Philological Faculties Lada Klimová .......................................................................................... 257 The Vowel Length Difference before Final Voiceless Consonants: Practical Teaching Tips Dušan Melen............................................................................................ 263
PREFACE
The study of ELT constitutes a domain of research that is necessarily diverse and multi-facetted. The ongoing series of conferences hosted at the University of J.E. PurkynČ (UJEP) is committed to give a broad display of the approaches and perspectives on this domain. The volume in front of you represents the results of the latest conference, called CHALLENGES IV, which took place from 25 – 26 October, 2012 in Ústí nad Labem. As ususal, we invited scholars in order to share current views, new trends, innovative ideas and experience in foreign language teacher education with a special focus on the English language. We received submissions relevant to applied linguistics and education, ELT methodology, cultural studies and literary studies. Further, this event fell into an important period of academic activities at UJEP as it was linked to the EU project NEFLT, a project with a focus on creating and improving relationships between educators of foreign language teaching, on analyses of current foreign language teacher trainee programs and on mapping the qualification rates of foreign language teachers. CHALLENGES was sponsored by NEFLT (grant no. CZ.1.07/2.4.00/31.0074) The conference and the subsequent contributions to this volume reflect on the one hand the international spectrum of activities, on the other the more locally focused research which is displayed in the various articles in this volume. At the same time we believe to have created a comprehensive companion piece to the 2011 volume ELT – Converging Approaches and Challenges, edited by Christoph Haase and Natalia Orlova (CSP). The volume contains 18 chapters that are organized in four main sections dedicated to broad fields in ELT. The first, Issues in Grammar Teaching and EAP, starts with a paper by David Newby on his very individual take on a cognitive-communicative grammar. This important contribution sketches a hybrid grammar model with underpinnings in recent findings in cognitive linguistics. Three contributions look at academic writing at university level and beyond - Olga DontchevaNavratilova and Renata Povolná focus on academic writing skills in the lingua franca English; Silvia Cacchiani analyzes if-conditionals in economics research papers and Zuzana Procházková studies the mental mapping of student academic writing. The concluding segment by
viii
Preface
Christoph Haase looks at parameters of morphological productivity of second language learners in English. The second section, entitled Teaching Expressivity and Culture offers a diverse array of studies that include at first a systematic survey of English address forms used by non-native speakers by Josef NevaĜil and Blanka Babická and a paper on the heterogenous situation of English and French as competitors in Cameroon by Samuel Atechi. Hana Suchánková reflects on the teaching of speaking from a cross-cultural perspective and the section ends on a philosophical note with OndĜej Skovajsa’s flaming appeal to encourage the humanities in the English classroom. Section number three is the most technical with studies on Methodology, Technology and ELT. This section also spans across all levels of language teaching. In it, Natalia Orlova analyzes the selfperception of teachers and Cornelia Neubert discusses some surprising pros and cons of the internet as an ELT resource. In a similar vein, the use of social media in the Middle Eastern world has been an important issue in recent politics but Darah Tafazoli and his colleagues offer an empirical look at the teaching of English idioms via facebook in Iran. Hana Ždímalová identifies and analyzes the pre-listening stage in its relevance for adult learners and the paper by Iva Koutská takes this one step further up the age scale in her comprehensive study of senior FL education participants. The final section collects shorter contributions – reflections on a networked teaching of tenses by Stanislava Kaiserová, the use of blogs in ELT by Jana Pavlíková and trhe use of Moodle for non-philological learners at university level by Lada Klimová. The practical segment on vowel teaching by Dušan Melen concludes this volume. Thus, we see many different purposes in this collection and we hope that our readers may obtain new perspectives and inspiration for their own academic work – be it in teaching or in research.
ACKNOWLEDGEMENTS
We would like to express our hearfelt gratitude to the entire NEFLT team who helped to make this conference a success. Among the many, the contributions of Pavel Frajs should be highlighted. For the work on this volume our thanks go especially to Sarah Good and Joel Head for indispensable proofreading. Christoph Haase & Natalia Orlova, September 2013
SECTION 1: ISSUES IN GRAMMAR TEACHING AND EAP
DO GRAMMAR EXERCISES HELP? ASSESSING THE EFFECTIVENESS OF GRAMMAR PEDAGOGY DAVID NEWBY KARL-FRANZENS UNIVERSITY GRAZ
In this paper I shall present a framework for assessing the learning effectiveness of grammar exercises based on a Cognitive+ Communicative approach to grammar (Newby 2003). I shall consider basic premises of this approach and list two sets of criteria, pedagogical and communicative, which can be applied in the analysis of exercises. Finally, I shall give examples of how this analysis can be implemented.
1. Cognitive+Communicative Grammar Cognitive+Communicative Grammar (abbreviated to C+C grammar) is a model which I have developed to assist in the design and analysis of grammar pedagogy. Pedagogical tasks to be carried out by pedagogical grammarians, textbook writers and teachers include the setting of communicative grammar objectives, the formulation of rules and the creation of grammar exercises and activities. The ‘Cognitive’ part of the C+C axis derives to a large extent from the school of linguistics known as Cognitive Linguistics. The ‘Communicative’ part relates to the Communicative Approach to language teaching. It thus draws on insights both from linguistics and from language teaching methodology. (I shall capitalise these terms to denote these specific theoretical uses and to distinguish from the more general meanings they can express.) Both parts of the axis can be applied to two separate but complementary aspects of theoretical analysis: language description and language learning. At the core of C+C Grammar are a Communication Model of Language, which depicts language as a ‘speech event’ (see Hymes 1964, Langacker 2008), and a Cognitive Model of Learning, based on the learning stages and cognitive processes to be described in this paper (see Newby 2002, in
Do Grammar Exercises Help?
4
preparation). Incorporating both aspects is essential since if we are to teach grammar in a principled way we need both a knowledge of what grammar actually is and an understanding of how grammar is learned. A C+C approach to language seeks to analyse and describe grammar, on the one hand in terms of the mental processes that underlie the use of language (Cognitive), and on the other, as an act of communication - a dynamic process in which a speaker’s perceptions are encoded by linguistic means into messages (Communicative). A Cognitive analysis of language examines initially how the mind is structured and organised to perceive reality and how cognitive categorisation gives rise to grammar; a Communicative analysis sees language as a process of conveying messages and embeds the cognitive, semantic, and formal categories arising from a Cognitive analysis within other aspects of communication, such as contextual variables, pragmatic meaning, and register. A C+C approach to learning will focus on mental processes that are activated when grammar is learned (Cognitive) and explores how grammar can be developed as a skill (Communicative) by means of pedagogy. It should be added that Cognitive and Communicative are inseparable, overlapping, and complementary categories, which is why they are always linked by a ‘+’ sign. Elsewhere (Newby 2008, 2012a etc.), I have discussed how a C+C view of language impacts on areas such as the setting of communicative objectives and formulation of grammar rules; in this paper the focus will be on a C+C view of learning and how this impacts on grammar exercise design.
2. Hypotheses of a C+C view of learning Like any theory of learning, a C+C view is based on certain hypotheses. Some of the most important, together with their pedagogical implications are the following: a)
Language learning is concept learning; learning is meaning driven and goal directed. Implication: teaching objectives must be defined largely in terms of grammatical notions (Newby 1989, 1991 etc.); taskbased activities should be used. b) Language is embedded in a network of schematic constructs and contexts which facilitate both communication and language learning.
David Newby
5
Implication: Grammar activities should be embedded in contexts and learners should bring in their own ideas and knowledge. c) Knowledge of language emerges from language use (Croft and Cruse 2004: 1). Implication: grammar exercises should focus on developing competence and performance rather than just knowledge, as tends to be the case in traditional approaches (see Newby 2012). d) Learning is an active and dynamic process in which individuals make use of a variety of information and strategic modes of processing (O’Malley and Chamot 1990:217). Implication: Teachers must ensure that learners are as mentally active as possible and use their cognitive resources meaningfully.
3. Cognitive learning stages One of the premises of a Cognitive approach, identified by O’Malley and Chamot (ibid), is the following: ‘Learning a language entails a stagewise progression from initial awareness and active manipulation of information and learning processes to full automaticity in language use’. Most theories of learning see learning as a ‘stagewise progression’ and therefore use stage models; however, there are considerable differences in how they conceive of these stages. Figure 1 shows the C+C view of learning stages, which is loosely based on a stage-model of language processing used by certain cognitive psychologists such as Anderson (1990).
Do Grammar Exercises Help?
6
CONCEPTUALISATION
PROCEDURALISATION
PERFORMANCE AWARENESS
INPUT
OUTPUT
Fig. 1: A cognitive model of learning stages
Learning stages
The following chart explains what the terms in this learning model denote. Input – materials provided by the teacher/textbook + students’ existing knowledge and schemata Awareness students notice and focus on new grammar Conceptualisation students ‘understand’ a grammar rule; usually conscious knowledge Proceduralisation students are able to use grammar in ‘scaffolded’ exercises without a strong conscious focus on rules Performance students are able to use grammar in open contexts; focus on the overall message Output – what students say or write
David Newby
7
In some ways a Cognitive specification of learning stages can be compared to the teaching stages found in traditional grammar pedagogy: presentation – practice – production (PPP). However, whereas traditional grammar takes a teacher’s perspective, a cognitive view will see stages from the learner’s perspective and will focus on the tasks that need to be accomplished in the human mind at each stage in order for grammar to be internalised. It will be noted that this model sees grammar both in terms of competence and of performance. Here we can also see a link to the Communicative approach, which stresses the importance of seeing language in terms of both knowledge and skills. 4. Principles and criteria for assessing grammar activities Various researchers in the fields of cognitive psychology and neuroscience have identified factors that accelerate or optimise language learning. The C+C model provides specific parameters for analysing and assessing the effectiveness of grammar activities. I shall consider two categories: pedagogical principles, based on Cognitive views of learning, and communicative criteria, based on theories of language use. These categories have the following analytical tasks: Pedagogical Principles – to what extent does a grammar activity support learning by activating and optimising learning processes and thus contribute to the overall aims of learning grammar? Communicative Criteria – to what extent does a grammar activity support the development of both grammatical and communicative performance by simulating conditionsof real-life language use? Pedagogical Principles will determine whether an activity can be validated – i.e. plays a useful role in learning; Communicative Criteria will determine whether an activity can be authenticated – i.e. corresponds to communicative use. Validation and authentication can be seen from the learner’s perspective: if learners validate an exercise, they (subconsciously) accept that what they are doing makes a good contribution to their learning; if learners authenticate an activity, they recognise that the way they are using language corresponds to some extent to how they are likely to use it in real life. 5. Pedagogical principles It is the aim of grammar pedagogy to support learning processes. In this section I shall list principles which facilitate high quality learning and which therefore need to be incorporated into the design of grammar exercises and activities. They have been compiled from a variety of
8
Do Grammar Exercises Help?
theoretical sources in the disciplines of applied linguistics, cognitive psychology, and education neuroscience. a) Depth of processing The extent to which a new item of grammar becomes stored in the memory of the learner is partly dependent on how mentally active the learner is when doing grammar tasks. Depth of processing will be determined by the nature of the grammar task given to students. Traditional methodology tends to provide activities which are ‘cognitively shallow’. For example, at the awareness-conceptualisation stage, the mental activity of the learners may be restricted to understanding what they have been told by teachers, textbooks etc. As a result, they may hear or see the language input, but do not process it intensely enough - what might be called the ‘in-one-ear-and-out-of-the-other’ phenomenon. Discovery activities on the one hand lead to far deeper processing. Activities such as fill-in-the-gap tend not to require learners to strongly activate their mental energy. A cognitive view stresses the maximising of mental resources, and tasks are designed with this in mind. b) Commitment filter Students must be encouraged to ‘commit themselves’ to learning grammar. This may include affective aspects such as enjoyment or fun, but also what is sometimes termed ‘resultative motivation’: i.e. students experience feelings of achievement when doing grammar exercises and activities – an important justification for task-based activities. Also, cognitive needs such as curiosity, problem solving, drive for communication, acquiring knowledge, etc. Traditional, repetitive and boring grammar exercises are likely to impede good quality learning. c) Peer/social learning (oral activities) All learning is influenced by the learner’s social environment. The contribution to learning made by interaction between learners and their peers is an important factor. On some occasions, peers may be better at teaching than teachers! Group work activities can include peer monitoring as part of their design. d) Summative vs. formative exercises (testing vs. teaching/ supporting learning)
David Newby
e)
9
These are two terms commonly used in testing theory but are equally appropriate in analysing grammar exercises. A summative exercise has the main aim of testing whether an item of grammar has been learnt, whereas a formative activity has the specific aim of supporting the learning process. A shorthand way of distinguishing between them is testing vs. learning activities. Many exercises found in school textbooks do not help the learner to learn to any great extent but merely test declarative knowledge – whether a rule has been learnt. Summative testing exercises very often take the form of contrastive exercises (present perfect or past tense, etc.). A formative activity on the other hand has the specific learning aim of developing the learners’ ability to use a grammatical item and of building up their confidence with regard to this grammar. Do you know vs. can you use activities? Cognitive psychologists (for example, Anderson 1990) often distinguish between declarative and procedural knowledge. Many grammar exercises only focus on the former. However, students need exercises which require them not only to understand grammar but to use it if they are to go beyond the conceptualisation stage of learning.
6. Communicative criteria A Communicative approach to learning and teaching attempts to replicate the real-life conditions of language use and to apply them in the design of pedagogy. Communicative activities are essentially meaningdriven and goal-oriented. Whether grammar activities can be described as ‘Communicative’ can be determined according to certain criteria, listed below. It should be stressed that there is no simple binary distinction between ‘communicative’ and ‘non-communicative’ activities. It could be stated that the more of the communicative criteria which a grammar activity fulfils, the farther along the continuous incline towards ‘100% communicative’ it might be located. However, learning stages must also be taken into account. For example, discovery activities used at the awareness/conceptualisation stage of learning will fulfil few of the criteria, but in this case this does not matter since the aim is to raise awareness and support the conceptualisation of grammar, not to use it. On the other hand, activities at the performance stage should fulfil most of the communicative criteria.
Do Grammar Exercises Help?
10
a)
Contextualisation Language used in an explanation or exercise is embedded in a clear context, or the exercise facilitates contextualisation (imagining a context) by the student. b) Personalisation (compare pedagogical principle, 2) When speakers produce language they are representing information, ideas, knowledge, etc. from their own personal perspective. Grammar activities need to take into account this ‘personalisation’ aspect of language and give students the opportunity to apply their own schematic constructs, and express their own ideas, from their own perspective in order to produce utterances. c) Complex encoding Whenever human beings produce language, they are processing two general areas of cognition. On the one hand, they represent the world around them – what they see, think, remember, experience, etc.; on the other hand, they map their perceptions of the world onto language. If students are to get to the performance stage of the cognitive stage model, they must be given the opportunity as soon as possible to rehearse this complex encoding; that is to say, they must not only be required to add grammar to prefabricated sentences but to encode utterances. Grammar exercises which merely require students to fill in gaps leads to ‘simple’ encoding – i.e. just grammar. More complex encoding is required if students have to produce both grammar and vocabulary items. Most complex encoding takes place when students have to produce a complete utterance. d) Authenticity of process To produce language, learners apply processes that human beings use when encoding utterances. Fill-in-the-gap with words given in brackets of transforming direct into indirect speech are totally lacking in process authenticity, whereas paraphrasing an utterance is an authentic process. e) Interaction (oral activities) Learners use the grammar item to interact with other learners in ways which require a response – for example, in an oral group work activity. f) Task-based In addition to producing correct utterances, students fulfil a
David Newby
11
purposeful cognitive task which will have some kind of outcome or end product.
7. Analysing grammar exercises through C+C categories I shall consider three grammar activities to illustrate how the categories described above can be applied. These activities share common grammatical objectives: forming and asking questions and using tenses correctly. However, in other aspects they differ, as the analysis will show. Exercise a) fill-in-the-gap with the correct form. Where ____________________ (Mary - buy) her new dress? When ____________________ (you - go) to bed last night? What ____________________ (Jim - do) every Saturday? Exercise b) Write questions that would produce the answers. (Newby, 1992: Grammar for Communication, Exercises and Creative Activities, p.98, adapted) 1. A: Why are you going to bed already? B: Because I feel tired. 2. A. __________________________________________________? B: A hamburger, please. 3. A. __________________________________________________? B: Let's go to France. I've never been there. 4. A. __________________________________________________? B: English and Spanish. 5. A. __________________________________________________? B: For five years. She really loves it there. 6. A. __________________________________________________? B: Fantastic! We had seats in the front row!
12
Do Grammar Exercises Help?
Activity c) This is an oral version of activity b). Students work in groups of about four. Each group has a set of cards with questions on one side and answers on the other. Students take it in turns to read out the answers; the others group members have to guess the question. Front of card A hamburger, please.
Back of card What do you want for lunch?
Discussion: The chart below summarises the analysis which follows. Category
Activity b) proc.
Activity c) proc.
shallow
deep
deep
Commitment
X
9
99
Peer learning
X
X
9
Formative/summative
F
S
S
DYK
CYU
CYU
Contextualisation
X
(9)
(9)
Personalisation
X
9
9
Complex encoding
X
99
99
Authenticity of process
X
9
9
Interaction
X
X
9
Task-based
X
X
9
Learning stage
Activity a) conc.
Pedagogical principles Depth of processing
‘Do you know?/Can you use?’ Communicative criteria
David Newby
13
a) Learning stage: Like most grammar exercises found in school textbooks and pedagogical reference grammars, activity a) is restricted to the conceptualisation learning stage – students are being tested on their (declarative) knowledge of a grammatical rule, whereas exercises b) and c) operate at the procedural stage of learning – learners have to generate their own utterances; thus the process of automatisation is supported. b) Pedagogical principles: An examination of commercially available grammar exercises will show a preponderance of the exercise format ‘fill-in-the-gap with lexical prompts given in brackets’. As with many aspects of grammar teaching, such exercises have become part of pedagogical tradition; however, in terms of learning theory, there is very little to justify such an exercise type. Since learners are presented with pre-fabricated sentences, processing is limited to applying a rule deductively and is, as a result, very shallow. The fact that so much language is provided precludes the need for or the possibility of schematisation. Such exercises are, of course, fairly boring and satisfy neither cognitive nor affective needs. Commitment is therefore likely to be low. In some grammar practice books, such exercises are referred to as ‘revision’ exercises (for example, Swan and Walter 1997), but this is a misnomer: they are to be regarded as ‘re-testing’, rather than revision and are of a very largely summative nature. I have seen such exercises subsumed under the category of ‘English in use’, but this is even more of a misnomer. The only person who uses English in exercises of this format is the author of the book in which the exercise occurs: it is purely a ‘do you know?’ exercise without any element of ‘can you use?’ An analysis of activity b) will reveal a far greater activation of supportive learning processes. Since students have to generate their own utterances, processing will not only be deeper but will correspond more closely to real-life language processing. Commitment is likely to be higher than in exercise a) since it is more cognitively challenging. The fact that students are getting practice in using the grammar (‘can you use?’ rather than ‘do you know?’) gives this activity a strongly formative element. What has been said with regard to b) above also applies to c); however, in this case there are additional elements that activate learning processes. Firstly, since this activity has a game-like quality, both cognitive and affective commitment is likely to be high. Second, since it is an oral activity played in small groups, there is the opportunity of feedback from peers– correction, assistance, etc. Peer support is often more useful than teacher correction since not only is peer correction less threatening but
14
Do Grammar Exercises Help?
also peers are better able to identify with difficulties that may be experienced. c) Communicative criteria With the exercise type ‘fill-in-the-gap + prompted lexis’, it is usually the case that if the lexical items are changed into nonsense words, students are still able to get the answers right, which is clear proof of how little meaningful language processing is taking place (criterion: ‘authenticity of process’). Readers might like to attempt the ‘nonsense version’ of exercise a) themselves: Where ____________________ (Mary - blonk) her new blink? When ____________________ (you - zonk) to glog last night? What ____________________ (Jim - ving) every Saturday? An analysis based on the communicative criteria will show how and why this type of exercise can be regarded as ‘uncommunicative’: none of the criteria is fulfilled. As far as b) and c) are concerned, the picture is very different. Although each exercise consists of single sentences, rather than, say, a contextualised dialogue, students do have to imagine a possible context for each ‘mini-dialogue’ in order to come up with a meaningful utterance. Both exercises require a high degree of personalisation: students express their own ideas, thoughts, etc. Similarly, encoding has maximum complexity since students provide all the language required themselves – i.e. there are no language prompts to help them. (If teachers feel that students are not yet at the stage where they can do an unprompted exercise, prompts can be provided – for example, a bag of words at the top of the exercise.) In the analysis of pedagogical principles I referred to advantages of oral, as opposed to written, activities. With regard to exercise c) additional communicative criteria can be identified: first, the group work format provides for interaction between students. Second, since oral activities take place in real time, greater authenticity of process is achieved. Third, the activity becomes task-based due to the game-like nature of the activity. An added advantage is that contact time with the grammar to be learned is far greater than with written exercises: this activity generates a large number of questions.
David Newby
15
8. Conclusion My experience of conducting in-service workshops in many countries has revealed that many teachers tend to give little consideration to the question of how effective grammar exercises are in supporting learning. Often, the criterion of quantity, rather than quality, seems to be the overriding reason for selecting exercises. Yet it is important to reason that practice itself does not make perfect: as Van Lier points out (1996: 59), ‘we have to learn to distinguish between practice and malpractice’. The framework presented in this paper provides a theory-based mode of assessing grammar exercises which, as my students at the University of Graz will confirm, can not only be applied in the analysis of exercises but which acts as a springboard to activity design.
References Anderson, J.R. (1990). Cognitive psychology and its implications. (3rd ed.). New York: W.H. Freeman and Company. Croft, W. & Cruse, D.A. (2004). Cognitive linguistics. Cambridge: Cambridge University Press. Hymes, D. (1964). Toward ethnographies of communicative events. In P.P. Giglioli (Ed.), Language and social context. Harmondsworth: Penguin Books. Langacker, R.W. (2008). Cognitive grammar. A basic introduction. Oxford: Oxford University Press. Newby, D. (1989). Towards a notional grammar of English. In B. Kettemann et al (Eds.), Englisch als Zweitsprache. Tübingen: Gunter Narr. Newby, D. (1991). A notional grammar of tense and aspect. (Unpublished doctoral thesis). Karl-Franzens Universität, Graz. Newby, D. (1992). Grammar for communication: Exercises and creative activities. Vienna: Österreichischer Bundesverlag. Newby, D. (2003). A cognitive + communicative theory of pedagogical grammar. (Unpublished Habilitationsschrift). Karl-Franzens Universität, Graz. Newby, D. (2008). Pedagogical grammar: A cognitive + communicative approach. In W. Delanoy, & L. Volkmann (Eds.), Future perspectives for English language teaching. Heidelberg: Universitätsverlag Winter, 29-44. Newby, D. (2012a). Cognitive + communicative grammar in teacher education. In J. Huettner, B. Mehlmauer-Larcher, S. Reichl, & B.
16
Do Grammar Exercises Help?
Schiftner (Eds.), Bridging the gap: Theory and practice in EFL teacher education (pp. 101 – 123). Clevedon: Multilingual Matters. Newby, D. (2012b). Pedagogical Grammar. In M. Byram (Ed.), Routledge encyclopedia of language teaching and learning (revised 2nd ed.). London: Routledge. Newby, D. (in preparation). Cognitive + communicative grammar: From practice to theory. O’Malley, J.M. and Chamot, A. (1990). Learning strategies in second language acquisition. Cambridge: Cambridge University Press. Swan, M. & Walter, C. (1997). How English works. Oxford: Oxford University Press. van Lier, L. (1996). Interaction in the language curriculum. London: Longman.
ANALYSING THE DEVELOPMENT OF ACADEMIC WRITING SKILLS IN ENGLISH AS A LINGUA FRANCA OLGA DONTCHEVA-NAVRATILOVA AND RENATA POVOLNÁ MASARYK UNIVERSITY BRNO
The establishment of English as the lingua franca of all academic communication has resulted in the necessity to explore the academic writing skills of non-native university students since the ability to produce academic texts in English is absolutely essential to their future academic and professional careers. This chapter reports on the results of a survey into the academic writing skills of students of English at the Faculty of Education, Masaryk University, Brno, Czech Republic. The aim of the survey, which was based on questionnaires completed by students pursuing both Bachelor’s and Master’s degree study programmes, was to establish tendencies in the development of their academic writing skills. In addition, a corpus-based analysis of students’ diploma theses was carried out in order to map academic writing competencies they had attained by the end of their studies. The results of the investigation suggest implications not only for the (re)design of academic writing courses, but also for the development of students’ language competences.
1. Introduction In the ongoing process of increasing internationalization of all scholarship and with regard to the crucial role of English as the lingua franca of academia it has become indispensable for authors from different cultural backgrounds and intellectual traditions to achieve a certain level of academic literacy in English in order to become recognized members of their respective research fields within international academia. That is why
18
Development of Academic Writing Skills in English as a Lingua Franca
different academic discourses written in English have to be mastered by non-native writers, including university students, as part of their “secondary socialization” in educational and research institutions (Mauranen et al. 2010: 184). Modern language teaching emphasizes efficient and appropriate ways of communication and in particular students’ ability to express their ideas comprehensibly and effectively. In the globalized world this entails the acquiring of competence to communicate cross-culturally and across disciplines in a variety of genres using both spoken and written discourse. With the ever-increasing amount of information available and the necessity to present it to large audiences both in cross-cultural and crossdisciplinary communication, the written mode has become a substantial and indispensable means of communication in various contexts on both personal and professional levels. The ability to work with written texts, i.e. to read texts as well as to produce them, ranks among the most crucial and essential skills of educated modern people, whether in their native or a foreign language. Consequently, it is not surprising that, on account of the increasing importance of the written mode in academic communication, the teaching and learning of academic writing skills have recently become central issues of interest for numerous linguistically and pedagogically oriented studies dealing with the assessing of native and non-native writers’ performance in terms of genre, disciplinary and cross-cultural variation and concerned with the designing of courses aimed at developing students’ communicative competence in an academic setting such as university (e.g. Duszak 1997, Hyland 2002a, Hewings 2006, Paltridge and Starfield 2007). The present investigation of academic writing skills of Czech university students involved in Bachelor’s and Master’s degree study programmes in teacher training in the English language undertakes to diagnose the special needs and current writing skills of students as well as to explore the progress they make over the five years of their studies from general writing to specific academic writing skills, including the writing skills reflected in the academic discourse they produce when writing their final diploma theses. The aim of the investigation is to assess to what extent the current academic writing courses offered by the Department of English Language and Literature at the Faculty of Education, Masaryk University, Brno, Czech Republic correspond to the specific needs and expectations of the students, and also to provide the necessary data for a potential (re)designing of these courses so as to equip the students with the academic writing skills required for their professional careers in the globalized academic discourse community.
Olga Dontcheva-Navratilova and Renata Povolná
19
2. Academic writing in English as a lingua franca The lingua franca status that English has acquired in the global academic world at the beginning of the 21st century has made the acquisition of academic English indispensable for anyone contemplating university studies and/or a professional career in science, research or academia. Despite the variety of academic literacies and epistemologies that have evolved over the ages, Anglo-American academic writing conventions have for a long time been regarded as the preferred norm for academic communication by educators involved in academic English courses and by the mostly Anglophone ‘gatekeepers’ reviewing publications for established academic publishers and periodicals (cf. Clyne 1991, Swales 1997, Tardy 2004). However, with the increasing number of non-native speakers forced to communicate using the modern academic lingua franca, in the last two decades numerous studies scrutinizing academic English from a discourse analytic and pedagogigal perspective have problematized the educated native speaker as the model of good and fluent language performance and questioned the Anglo-American tradition of academic writing as the prevailing discourse convention (e.g. Flowerdew 2008, Mauranen et al. 2010). While considering the success of the numerous programmes of English for academic purposes offered by various universities, strong voices within the international academic discourse community have pointed out that “there are no native speakers of academic English” (Mauranen et al. 2010: 184) and warned against the danger of a ‘Tyrannosaurus Rex’ impact of English on the numerous ‘nonEnglish’ rhetorical, register and genre conventions (e.g. Swales 1997, Tardy 2004). The resulting ideological challenges to Anglo-American academic writing conventions have called for the use of a more flexible approach to academic writing in English which respects heterogeneity and takes into consideration existing variations in meaning and organization of academic texts across different disciplines, languages and cultures (cf. Duszak 1997). The raising of awareness of the existing cross-cultural variation in academic literacy conventions has been fostered by an important change in the understanding of academic writing, shifting “from a formal text-based perspective to a functional perspective that concentrates on the writer and the writing process and, even more, on the reader and the cognitive construction of discourse in a community” (Schmied 2011: 1, cf. Hyland 2010). Within this functional approach, academic writing conventions are seen as being affected by the dynamics of the national and international academic discourse communities and by their preferred ways of
20
Development of Academic Writing Skills in English as a Lingua Franca
expressing interpersonal meanings. The potential of this approach to explain reasons for cross-cultural variation in academic writing can be illustrated by a comparison of the Anglo-American and the Czech (Central European) writing conventions, which are scrutinized in the present investigation. Previous research into the Anglo-American and Central European academic literacies has shown that they differ considerably in many respects and especially in the way they approach writer-reader interaction (e.g. Clyne 1987, Mauranen 1993, ýmejrková and Daneš 1997, Duszak 1997, Kreutz and Harres 1997, Chamonikolasová 2005, Stašková 2005, Povolná 2010b, Dontcheva-Navratilova 2012a). This divergence seems to be motivated to a large extent by the size and the level of heterogeneity of the respective discourse communities (Mauranen 1993), which affect the relations of power and solidarity among their members (ýmejrková and Daneš 1997). The large Anglo-American academic discourse community is linguistically and culturally heterogeneous. Consequently, when writing for a highly varied and depersonalized readership, its members have to compete for research space in a territory densely packed with occupied ‘niches’ (Duszak 1997). This forces authors to invest a greater persuasive effort and adopt a more reader-friendly attitude and urges them, especially those working in the soft sciences, to deviate from the traditional scientific paradigm advising objectivity and the avoidance of human agency (Hyland 2001, Bennett 2009). As evidenced by recent research (e.g. Gosden 1993, Kuo 1999, Tang and John 1999, Hyland 2001, 2002a, Harwood 2005), Anglo-American authors typically opt for a marked authorial presence and strive to negotiate their claims and debate their views with the implied audience, while guiding the reader through the text by using signals such as discourse markers indicating intratextual connections and logical relations holding between segments of discourse. It thus seems reasonable to assume in agreement with ýmejrková and Daneš (1997: 57) that the potential to cope with a lower degree of shared knowledge by opting for explicit marking of discourse organization and a higher level of dialogicality makes the present-day Anglo-American norm of academic writing particularly suitable for purposes of cross-cultural communication within the global academic discourse community. The Central European academic literacy subsumes several considerably smaller and therefore much more homogeneous discourse communities. Due to the considerable amount of common knowledge and methodological principles which members of these discourse communities share, their preferred patterns of interaction tend to be marked by symbiosis and avoidance of tension. This motivates the use of rather monologic, more
Olga Dontcheva-Navratilova and Renata Povolná
21
implicit and less structured texts, whose authors seem to favour conceptual and terminological clarity rather than persuasion and discourse organization. Thus, while Anglo-American writers take responsibility for making their academic texts intelligible, Czech and Central-European authors have traditionally left the responsibility for understanding the text to the reader. However, since English has become widespread as a lingua franca, the Central European academic literacies have started to change under pressure to adapt to the requirements and conventions of the global discourse community. Thus, as ýmejrková and Daneš (1997: 42) point out, Czech academic writing has recently been profoundly affected by the spread of English academic norms. This tendency is likely to be most obvious in the works of Czech authors writing in English as an academic lingua franca, since when producing their academic discourse they are likely to draw on both Czech and Anglo-American writing conventions. Interference between the Czech and the Anglo-American academic literacies is also likely to affect the teaching of academic English at Czech univeristies, since the educational system is supposed to transmit a particular epistemological and literacy tradition to new members of the academic discourse community. It is therefore relevant to study what kind of instruction on academic writing conventions students are provided with in university academic writing courses as well as to analyse students’ academic discourse reflecting the level of acquisition of academic writing skills and conventions. This can shed light on whether the use of English as an academic lingua franca by Czech students shows some degree of adaptation to Anglo-American writing conventions and whether the Czech academic writing tradition has a significant impact on students’ English academic discourse.
3. Methodology and data The aim of this research is to discover current tendencies in the development of academic writing competences acquired by university students in Bachelor’s and Master’s degree programmes for teacher training in English and, in addition, to compare students’ expectations concerning academic style and academic writing skills with the actual development of their language competences as reflected in the academic discourse they produce when writing their diploma theses. Accordingly, this chapter first presents and discusses the results of a needs analysis carried out in two study groups, one of the Bachelor’s and one of the Master’s degree programme at the Department of English Language and Literature, Faculty of Education, Masaryk University, Brno,
22
Development of Academic Writing Skills in English as a Lingua Franca
Czech Republic. Secondly, the results, which are drawn from a small-scale corpus-based analysis of students’ diploma theses, are exemplified and discussed in order to map academic writing competences achieved by students at the end of their studies with regard to the use of selected language issues, namely author-reference devices and discourse markers. The present needs analysis starts with questionnaires as a method of data collection suitable in particular to the present situation analysis (cf. Jordan 1997), since this type of analysis “concerns starting where the students are and refers to information about learners’ current proficiencies and ambitions” (Hyland 2006: 74, Dudley-Evans and St John 1998) and it is thus viewed as a good starting point for the needs analysis. The questionnaires also include questions aimed at the students’ knowledge of basic academic style issues, such as working with sources, citing, paraphrasing, plagiarism, i.e. issues which broaden the scope of information usually gathered by means of questionnaires. The questionnaires were administered to Czech students in the Bachelor’s and Master’s degree programmes in teacher training in English. The total number of 78 students, who voluntarily completed the anonymous questionnaires, comprises 46 students involved in the second year of the Bachelor’s degree programme and 32 in the second year of the Master’s degree programme. Since more than 90 per cent of all respondents were female students, it was decided that it was not relevant to consider gender differences concerning the students’ academic writing skills. While the second-year Bachelor’s degree students have been exposed to instructions on academic writing skills only exceptionally, the Master’s degree students included in the investigation received some formal tuition in academic writing during their previous university studies. Therefore, the respondents in the latter group were given a more detailed questionnaire. The questionnaire for second-year Bachelor’s degree students consists of 15 items, which are divided into two sections, titled Writing in General and Academic Writing. The questions in Section One concern writing issues in general, while those in Section Two search for instructions received on and knowledge of some basic academic style issues. Most entries in both sections of the questionnaire are multiple-choice questions. Only a few entries are yes/no questions, sometimes supplemented by openended questions asking respondents to specify reasons for their affirmative answers. Section One was designed to collect the kind of data that enables the mapping of students’ previous experience and writing strategies, for example what kinds of writing assignments they have done previously,
Olga Dontcheva-Navratilova and Renata Povolná
23
what instructions on the process of writing they have received during their studies and what resources they have used when collecting data on a certain topic. Section Two focuses on the students’ goals and expectations concerning their future tuition in academic writing skills before writing their Bachelor’s theses. This section also includes questions which reveal the Bachelor’s degree students’ awareness of basic concepts in academic writing, such as citation styles, academic writing style and its typical features, paraphrasing and plagiarism. The questionnaire for Master’s degree students is aimed at more specific needs and, in addition to gaining data about students’ writing experience and previous academic writing tuition, it attempts to map their current academic skills related to such discipline- and culture-specific aspects of academic writing as discourse organization through signposting, citation practices and authorial presence. The choice of these aspects is motivated by the fact that they are related to the expression of interpersonal meanings the role of which is particularly important at a more advanced level of academic writing, as they help writers to persuade readers of the credibility of their claims and the validity of their research (cf. e.g. Duzsak 1997, Hyland 2002a, Swales 2004, Flowerdew and Peacock 2010). The questionnaire comprises 18 items, which are divided into three sections. Section One includes questions concerning previous experience, Section Two consists of questions concerning instructions received on general and discipline-specific academic writing, while Section Three comprises text-based tasks intended to assess the students’ academic writing skills. Most of the entries in the first two sections are yes/no questions, usually complemented by an open-ended question asking respondents to specify, exemplify and/or provide reasons for their answers. In the text-based tasks in the last section of the questionnaire students are asked to identify target structures, assess their stylistic appropriateness and comment on their functions. Concerning the results of the needs analysis, it is necessary to emphasize that the statistical data yielded by this survey are regarded as a starting point for an essentially qualitative analysis of the needs and writing skills of the students involved in the Bachelor’s and Master’s degree teacher training programmes in English. The corpus-based investigation of students’ diploma theses focuses on a relatively small sample amounting to about 176,000 words and comprising ten diploma theses (5 in the field of linguistics and 5 in the field of methodology) written by students of English in their final year of study at Masaryk University, Brno, Czech Republic. In agreement with
24
Development of Academic Writing Skills in English as a Lingua Franca
Flowerdew (2004: 18), specialized corpora are assumed to be more appropriate than large general corpora for an analysis of particular language features in one particular genre, which in the case of the present chapter entails author-reference devices and discourse markers, notably in students’ diploma theses. Since diploma theses represent university students’ final written achievements before they embark on their professional careers, they are often conceived as “the most sustained and complex piece of academic writing” (Swales 2004: 99) most students ever undertake. This is the main reason why they are worthy of researchers’ attention, in particular now that international academic communication in the majority of fields university graduates enter is conducted in English and “internationalization of scholarship opens up new challenges for our understanding of academic communication phenomena” (Duszak 1997: 19). As for the methods applied in the investigation of students’ diploma theses, all the texts were first computer-processed using the AntConc concordancer and then examined manually for both quantitative and qualitative results with regard to the features selected for analysis, i.e. author-reference devices and discourse markers. The most important results are given in tables, exemplified and discussed below.
4. Undergraduate students’ academic discourse: results and discussion The Bachelor’s degree students who participated in the needs analysis were 46 second-year students who were studying English and another subject, such as Czech, German, French, Russian, history, mathematics, civics or special pedagogy. These respondents were involved in what is usually labelled a double-subject Bachelor’s study programme. At the time when the survey was carried out, most of them did not have any teaching experience since they had started their university education immediately after completing their secondary education. The results concerning Section One of the questionnaire, titled Writing in General, show that the students had a relatively considerable amount of experience in writing argumentative (59% of respondents) and literary analysis (83%) essays. This finding is very important, since “the presentation of information and [above all] argumentation is usually something that language users are not very much aware of, but which is extremely important in academic writing” (Schmied 2011: 42). By contrast, they had little experience in producing research papers (22%), which is hardly surprising because all were at the beginning of their
Olga Dontcheva-Navratilova and Renata Povolná
25
university studies. As regards the type/token ratio, most students produced different types of academic genres only in a few cases (mostly 1-5). In accordance with what is usually recommended in academic style manuals (cf. e.g. Bennett 2009), most respondents had received instruction on the overall organization of text (85%) and its language and style (67%) before writing an essay. Only a few (26%) had been instructed on the process of argumentation, i.e. they had received instruction on how to form their reasoning, justify their beliefs, and draw conclusions; this result is probably caused by the fact that most respondents had not had a chance to attend a course in general writing, in which they would have been instructed on how to formulate arguments. Three students had not been given any instruction at all and only seven (9%) had been advised on grammar and citing, which is rather unexpected since both grammatical issues and referencing are traditionally mentioned among features included in manuals on academic writing (Hamp-Lyons and Heasley 2006, Bennett 2009). As regards courses in academic writing, only four students had had a chance to attend such a course, but sixteen students were attending a course in written fluency at the time of the needs analysis. Those who had received advice and/or encouragement from their teachers mostly remembered they had been advised to avoid plagiarism (93%), to understand and consider the format of the text they were supposed to produce (57%) and to do brainstorming before writing an academic text (50%). Fewer than half of the respondents (48%) had been encouraged to revise and edit their writing and cite sources carefully, while only seventeen (37%) wrote drafts of their texts and nine (20%) did some research prior to writing an essay. This seems to be caused by the fact that these students of two different subjects, such as English plus Czech or German, tended to be distracted by assignments from subjects other than English. In order to find out whether the students participating in the survey were capable of acquiring a level of academic literacy necessary for success in their studies and future teaching profession, the questionnaire also included items asking about research they had done prior to writing. The results testify that for all students who did some research prior to writing an essay (63% of all respondents) the internet was their main source of information. Other sources of reference frequently resorted to were books, magazines and newspaper articles; professional journals and personal interviews were used only exceptionally. Section Two of the questionnaire, labelled Academic Writing, comprises questions concerning students’ experience in academic writing and their
26
Development of Academic Writing Skills in English as a Lingua Franca
knowledge of some relevant academic style issues, such as citing, paraphrasing and plagiarism. Similarly to diploma theses written at the end of Master’s degree programmes (cf. Swales 2004: 99), Bachelor’s theses represent the most complex and sophisticated piece of academic writing students are expected to produce at the end of the Bachelor’s stage of their studies and, consequently, these are expected to be of good quality and worthy of researchers’ attention. Since it is believed that, apart from instructions on and practice in features typical of academic writing style in general, students need information concerning the format of the Bachelor’s thesis, its content and possible ways of presenting arguments, as well as instructions on fieldspecific academic writing conventions, it was considered necessary to relate the results of the needs analysis to the field of interest on which the students intended to concentrate in their future Bachelor’s theses. Before discussing the results, it should be noted that all the respondents from the second year of the Bachelor’s degree programme were double-subject students, which means that for their Bachelor’s theses they could choose topics from linguistics, literature or culture, provided they wanted to write their Bachelor’s theses in English, or, alternatively, they could opt for topics from their other subject of study or even from common core subjects, such as psychology and pedagogy. Topics related to ELT methodology can be taken only by students in the Master’s degree programme. As the results of the survey prove, the respondents showed a clear preference for cultural studies and literature (19 of 46 students), common core subjects (11), and linguistics (8). Only a few did not provide an answer when the questionnaire was administered, which indicates that these respondents were not yet sure which topics to select. As regards the writing of the future Bachelor’s thesis, the respondents were highly motivated to use sources, in particular to master an appropriate style of academic writing (23 students), while only a few (7) wanted to achieve a better overall quality in their work (cf. ýmejrková et al. 1999). Among the sources these respondents were planning to use, books and/or textbooks (18) and the internet (15) were mentioned most typically. Four students also listed knowledge items they expected to acquire in an academic writing course. In order to find out what the students had already learnt about academic writing style, they were asked questions about their knowledge of important academic writing issues before they started their university studies and at the moment when the present needs analysis was carried out. The highest number of students claimed they had known what plagiarism was (41), while a considerably lower number believed the same
Olga Dontcheva-Navratilova and Renata Povolná
27
about paraphrasing (23). Sixteen had known what academic writing style was. These findings prove it is crucial for students to become “aware of the conventions involved … in a BA thesis in their specialization” (Schmied 2011: 5) and thus it is of great importance to introduce courses in academic writing skills as early as possible (for typical writing skills, cf. ýmejrková et al. 1999), preferably at the very beginning of the Bachelor’s degree programme. The lowest number of students (11) knew about possible citation and referencing styles, such as APA, MLA and Chicago style. During their university studies a relatively high number of respondents believed they had learnt more about academic writing style (27), possible citation and referencing styles (25) and paraphrasing (21); these results prove that the students included in the investigation had received tuition in some of the most important academic style issues (Bennett 2009) in the first half of their Bachelor’s studies. As for the items of the questionnaire concerning the students’ awareness of what academic writing style was, nine respondents felt they knew what it was and also what its typical features were. Only a few provided an adequate (7) or partly correct (7) definition, such as one of the following: x a way of writing at a high formal level, with regard to the form and relevant content and specific rules of argumentation x a formal way of writing on some topic with quotations (use of several sources) x it’s a style used for academic purposes and has certain rules that have to be followed x a formal style for academic purposes, its aim is to inform, to express ideas clearly without repetition, formal words are used, plain expressions are avoided, passive voice is preferable x academic writing usually has topics that are of interest to the academic community. It has its rules and is supposed to inform or argue about something x a very formal writing style x standard written form of the language – it is formal, objective, responsible – correct use of language and style The above definitions selected from the students’ responses provide evidence that many Bachelor’s students knew the most important characteristics of academic writing style, such as complexity, formality, precision, objectivity, explicitness, accuracy, hedging and responsibility (e.g. Hamp-Lyons and Heasley 2006, Bennett 2009). Many respondents
28
Development of Academic Writing Skills in English as a Lingua Franca
(33) admitted that they had heard the term ‘academic writing style’ but were not totally familiar with its typical features, and four admitted that they were not familiar with the concept of academic writing style at all. The students who intended to attend a course in academic writing in the future expected to receive instruction above all on vocabulary and collocation and how to cite sources and avoid plagiarism. They also mentioned that they expected advice from experienced people and feedback on mistakes in their academic writing. Here are some student responses from the needs analysis: x it would be good to write essays and then talk about mistakes which students usually make x I expect to know how to write in a very formal and polite way x I expect to learn the typical way of writing an academic text x the most important thing is to understand how to cite sources, how to quote, just all the things necessary to avoid plagiarism x actually I take part in an academic writing course and I especially appreciate vocabulary and collocation x what is acceptable and what is not; what is important to know before writing; I would like to hear some advice from experienced people As for the citing of sources referred to in academic texts, many respondents stated they had been given instructions by their teachers (22) and/or had tried to find out for themselves (11). Not a single student considered it possible but not necessary to cite sources, which is a positive finding, since citing sources is an important aspect of every academic text (Hamp-Lyons and Heasley 2004, Wallace 2004). Only three students admitted they did not know how to cite sources at all. When asked to provide a definition of paraphrasing, ten respondents provided an adequate definition, while 24 gave only a partly correct, inaccurate definition. Here are some of the definitions of paraphrasing given in the questionnaires: x using different/my own words for saying what someone else said and it is necessary to mention his name (an example of a right definition) x I use the idea of what somebody else said and use my own words to explain it to someone else, for example (an example of a partly right definition)
Olga Dontcheva-Navratilova and Renata Povolná
29
x I say what someone else has said in my own words (an example of a wrong definition) Since plagiarism is often difficult to recognize for novice writers, the respondents were offered four possible definitions in the questionnaire. The highest number (40 students; 40%) recognized copying material/someone else’s ideas without acknowledging the source as plagiarism. Next comes the use of material from the internet without mentioning the source (34; 74%), followed by the paraphrasing of someone else’s ideas without mentioning the source (32; 70%). The greatest difficulty for the respondents was to recognize as plagiarism the use of material where the author has been identified without quotation marks to indicate his/her original words; this was identified correctly as plagiarism only by fewer than half of all respondents (22; 48%). Summing up, it can be postulated that the second-year double-subject Bachelor’s students had practice in writing argumentative, literary and advantages/disadvantages essays, which seems to result from their sufficient exposure to a relatively wide range of writing assignments. These respondents also seem to have been given sufficient instruction in the overall organization of an essay. However, they were only rarely encouraged to do research and write drafts before writing an essay. The needs analysis has proved that most respondents were able to recognize definitions of paraphrasing and plagiarism and need more instruction and/or practice in referencing.
5. Graduate students’ academic discourse: results and discussion The investigation into graduate students’ academic discourse in English aimed at comparing the results of a needs analysis based on the subjective perceptions of the students to the findings of a corpus-based analysis of their academic writing perfomance as illustrated by their diploma theses. The respondents to the questionnaire were Master’s degree single-subject full-time and distance students, who were writing their diploma theses on English literature, linguistics or English-language teaching methodology, and Master’s degree double-subject full-time students, who could choose to write their theses with the Department of English Language and Literature or a topic related to their second subject, which similarly to the Bachelor’s degree programme is typically Czech, German, French, Russian, civics, history, mathematics or special pedagogy. All the students taking part in the survey were in the last year of
30
Development of Academic Writing Skills in English as a Lingua Franca
their Master’s degree programme; they had taken an academic writing course during their studies and, at the time when the investigation was carried out, were writing or had just completed writing their diploma theses. The analysis of students’ academic discourse was based exclusively on diploma theses written in English in the fields of linguistics and English-language teaching methodology.
5.1 Questionnaire results As mentioned above, the questionnaire administered to graduate students aimed at finding out the scope and focus of the schooling in academic writing they had received during their university studies. The first section of the questionnaire aimed at exploring the academic writing experience of the respondents. The results of the survey revealed that apart from having attended general academic writing courses during their Bachelor’s or Master’s degree studies, half of the graduate students were involved in an academic writing course focusing specifically on the writing of a diploma thesis. Eighty-two per cent of the students indicated that advice and instruction received prior to writing their theses encouraged them to refer to online resources on general academic writing style (typically OWL Purdue and MLA), printed style guides and research articles on academic writing in English. This suggests that the students were acquainted with the so-called scientific paradigm commonly recommended by Anglo-American academic style manuals, which advises adherence to “clarity, economy, rational argument supported by evidence, caution and restraint” (Bennett 2009: 52) and the avoidance of explicit reference to human agency (Hyland 2001b). However, since students also reported that they had received advice from their, mostly Czech, teachers and supervisors (reported by 64% of respondents) or from their colleagues (18% of respondents), it is likely that they were also influenced by Czech academic writing conventions, which they had also experienced in the previous course of their studies and/or through the other subjects that they were studying. Since apart from ‘generic’ skills transferable across disciplines novice writers need to be initiated in literacy skills specific to the purposes of particular disciplinary academic communities (Hyland 2002: 385), the survey explored the main subject areas of the academic writing experience of the students. While the respondents stated that they had submitted written assignments in various subject areas, their experience in writing an extensive academic text was confined to their Bachelor’s thesis. The most frequently reported subject areas of Bachelor’s theses were literary
Olga Dontcheva-Navratilova and Renata Povolná
31
(32.6%) and cultural studies (24.4%), while theses on linguistics (18.6%) and ELT English-language teaching methodology (15.1%) were less frequent (9.3 per cent of the respondents had not written their theses in English). Since the majority of the graduate students involved in teacher training programmes write their Master’s degree thesis in Englishlanguage teaching methodology, it is likely that without additional instruction in discipline-specific academic writing conventions the academic discourse of the students would show some variation concerning, for instance, authorial presence choices and citation practices. As to the scope of the academic writing skills of the students, the results of the survey indicated that similarly to undergraduate students, the majority of the graduates had received instruction in citation practices and referencing styles (MLA, APA, etc.), text organization, paragraphing and layout, preferred level of formality (mainly related to lexical choices), and the use of complex structures (including passive voice, nominalization, non-finite clauses and extraposition). It is obvious that the range of features listed by graduate students is considerably broader and includes several interpersonal devices, such as hedging, impersonality vs. personality choices, cohesion and argumentation strategies. However, as the findings of the survey show, the students had not received explicit instruction in generic structure, as only one of the respondents reports to have received instruction in abstract writing; this may result in a relatively low degree of genre awareness on the part of the students. Graduate students’ academic discourse is strongly influenced by instructions on general and discipline-specific academic writing provided by academic writing courses and supervisors of diploma theses, as these are supposed to set the requirements that the submitted theses should meet. Therefore the second section of the questionnaire focuses on advice received on specific aspects of academic writing, such as authorial presence, citation, cohesion and coherence. One of the aspects of of academic writing which shows considerable variation across different cultural traditions and disciplinary conventions is the construal of authorial presence (e.g. Mauranen 1993, ýmejrková and Daneš 1997, Duszak 1997, Chamonikolasová 2005, Mur DueĖas 2007, Dontcheva-Navratilova 2012) typically related to the choice between personal and impersonal forms. As evidenced by recent research, soft sciences, and especially applied linguistics, are characterized by prominent occurrence of personal structures (e.g. Hyland 2002a, Tang and John 1999, Hewings and Hewings 2002, Harwood 2005); it is therefore relevant to consider how students were instructed in the building-up of their authorial presence prior to writing their theses. The results of the survey show that
32
Development of Academic Writing Skills in English as a Lingua Franca
in agreement with the generally established scientific paradigm the majority of respondents (78.1%) were advised to use impersonal structures, such as passive voice and formal subjects, to achieve formality and objectivity. In addition, half of the students (16) were instructed on the possible use of first person pronouns for indicating authorial presence. When listing the main functions of author reference singular pronouns (I/me/my), namely stating goals and purposes in the introduction, summarizing results in the conclusion, expressing personal views, elaborating an argument and describing procedure, the respondents showed awareness not only of the interpersonal potential of authorreference pronouns (e.g. Tang and John 1999, Hyland 2002a) but also of their occurrence in specific components of the generic structure of the diploma thesis (cf. Paltridge and Starfield 2007). Mentioned by only 31 per cent of the respondents, instruction provided on the functions of the plural author-reference pronouns (we/us/our) concerns the use of the inclusive we for reader involvement and to describe disciplinary knowledge and practices, and the use of the exclusive we for authorreference in co-authored texts. In addition, the students were advised to use the exclusive we for author-reference in single authored texts and instructed that this was a more appropriate choice than the use of I. While the inclusive/exclusive divide may be used to enhance persuasion (Harwood 2005), the advice to use the exclusive we as a default authorreference device may be interpreted as interference from the Czech academic literacy (Chamonikolasová 2005), which is thus transmitted through the educational system. Another important aspect of the interpersonal dimension of academic writing is citation practices, as they allow writers to negotiate views on accepted theory and previous research findings in their argumentation (Hewings et al. 2010). First of all, the respondents were asked to comment on the issue of plagiarism. As the results of the questionnaire suggest, graduate students are fully aware (90% of respondents) that copying material from printed or internet sources as well as paraphrasing or using someone else’s ideas without acknowledging the source is not acceptable and therefore qualifies as plagiarism. However, the survey showed that similarly to the undergraduate students, more than half of the graduates (57%) did not indicate that using someone’s original words without quotation marks is also regarded as plagiarism, even if reference to the source is indicated. As to the available referencing styles, most respondents (78%) reported that they were instructed to use the MLA and the APA styles of referencing, which they successfully identified as the author-title and the
Olga Dontcheva-Navratilova and Renata Povolná
33
author-date referencing styles respectively. Since the focus of most academic writing courses in the teacher training study programme is on the author-title style, it is not surprising that MLA was indicated as the preferred referencing style of the majority of the students. It should be mentioned, however, that the results of the survey did not indicate an awareness of existing disciplinary variation in citation practices on the part of the respondents. To explore in greater detail the graduate’s knowledge of citation practices, the questionnaire comprised a task in which the respondents were asked to correct inconsistencies in referencing in an extract from a research article (Hyland 2008: 5) which originally uses the APA referencing style: The study of formulaic patterns has a long and distinguished history in applied linguistics, dating back to Jespersen 1924 and to Firth, who popularised the term ‘collocation’ along with the famous slogan that ‘you shall judge a word by the company it keeps’. More recently, Nattinger and DeCarrico (1992 - 32) have emphasised the importance of frequent multiword combinations as a way of assisting communication by making language more predictable to the hearer. Wray and Perkins, 2000, for instance, argue that such sequences function as processing short-cuts by being stored and retrieved whole from memory at the time of use rather than generated anew on each occasion.
The results of the survey indicate that while the majority of the respondents noticed the inconsistencies in the references to the date of publication and suggested a consistent use of the year in brackets, most students were unsystematic in their corrections of punctuation. The fact that several respondents suggested the inclusion of page numbers, which in the text under consideration is actually not necessary, may reflect the preference of novice writers for the use of direct quotations, with which the indication of page numbers is required. The last aspect of academic writing targeted by the questionnaire is related to discourse organization and linking. The issues of cohesion and coherence are traditionally treated carefully in academic writing style manuals (Bennett 2009) and given considerable space in academic writing courses. In order to find out whether the students had an advanced awareness of the role of linking signals in academic discourse, the questionnaire focused on the use of structuring signals organizing larger stretches of discourse (e.g. This sections deals with, In this part of the thesis I will address) and cross-referencing markers directing the reader elsewhere in the text (e.g. Drawing in the classification suggested in section 3.2, As shown in Figure 5). While 63 per cent of respondents reported that they had been advised to use such devices at boundaries of
34
Development of Academic Writing Skills in English as a Lingua Franca
larger sections of discourse, to improve readability, connect ideas, and make explicit the logical organization of the text, the rest of the graduate students did not indicate having received instruction on these issues. As to linking at the micro-discourse level, the majority of the students (30) was aware of the need to use formal discourse markers to enhance the coherence and cohesion of academic discourse and mentioned the following functions of discourse markers listed here with decreasing frequency: listing, contrast, addition, exemplification, cause-effect, summation/conclusion, focus and topic change, reformulation. The absence of temporal markers on this list is significant, as it suggests a lack of awareness of their potential to perform discourse deictic functions. The text-based task (reproduced below) asking the respondents to identify the discourse markers used and, if necessary, to replace the suggested markers by more suitable ones, focused on the most frequent functions of discourse markers in academic discourse, such as listing, contrast and cause-effect. Why is it so difficult to solve the unemployment problems of the developing world? There are three main reasons. One, there is the constant pressure of a rapidly rising population. However, this problem is made worse in cities by the drift of people from country areas to escape the poverty of rural life. Then there are problems of bad manpower planning. Therefore, a feature of unemployment in the developing world is educated unemployment – the lawyers or arts graduates who have been trained at great expense for jobs which do not exist. The point here is that the manpower plan has not been matched to the production plan. But the major reason for many countries failing to solve the unemployment problem has been their government preference for large-scale capitalintensive projects which use up scarce resources and have little impact on unemployment. In fact, by destroying local craft-based industry, some projects may even create further unemployment.
As the results of the survey show, the majority of the respondents managed to identify all the discourse markers in the text. Practically all graduate students chose an adequate replacement for the listing/addition devices: typically firstly or first for one, and the more formal secondly, in addition or furthermore for then. The replacements proposed for the rather informal contrastive marker but, i.e. however and nevertheless, were also adequately chosen. It is significant that none of the students suggested dropping any of the discourse markers, although nearly all sentences in the text contained one. This is in conformity with the findings of previous research (Vogel 2008, Povolná 2010a and 2010b), which showed that the rate of sentence linkers in texts by Czech novice writers often exceeds the standard rate in native-speaker discourse. This suggests that when trying to
Olga Dontcheva-Navratilova and Renata Povolná
35
comply with the requirement of explicit signposting and linking stated in most academic style manuals (Bennett 2009), students prefer to rely on the use of discourse markers learned through exposure to overt teaching of target structures. To sum up the results of the survey, it is obvious that thanks to the extensive instruction received on academic writing the respondents have successfully acquired general academic writing skills. The experience gained in university academic writing courses has undoubtedly helped the students improve the coherence and cohesion of their texts and raise their awareness of some of the interpersonal dimensions of academic writing, such as citation practices and authorial presence. However, due to scarce and unsystematic instruction on genre and disciplinary conventions, the ability of the students to cope with disciplinary and cultural variation in academic writing is likely to be insufficient. In order to explore to what extent the graduates managed to build persuasive argumentation and a coherent authorial voice, the diploma theses written by the students were analysed with a focus on author-reference devices and selected discourse markers.
5.2 Corpus-based analysis of diploma theses: findings and discussion The investigation of the graduate students’ discourse scrutinized ten diploma theses (5 in the field of linguistics and 5 in the field of ELT methodology) totalling 176,000 words. Since the sample included in this specialized corpus is rather small, quantitative analysis was used only as a starting point of an investigation aimed at identifying the major tendencies in the use of author-reference devices and discourse markers by the students in their diploma theses. These were then studied qualitatively to explore the functional specialization of the language devices used and to discuss the reasons for the choices made by the novice writers. a) Author-reference devices The construal of authorial presence by the graduates is likely to be influenced by advice provided by style guides, academic writing instructors and supervisors, which as the questionnaire results indicate recommend the use of the traditional scientific paradigm. This preference towards the backgrounding of authorial presence and the use of impersonal structures is in conformity with the established conventions in Central European academic discourse. It is therefore not surprising that the novice authors’ discourse is marked by a prominent use of passive voice, it-
36
Development of Academic Writing Skills in English as a Lingua Franca
clauses, and discourse entity (thesis, chapter, section) and research process (analysis, comparison) subjects. However, as a quantitative analysis of the frequency of occurrence of author-reference devices – personal pronouns (I/me and we/us) and possessive determiners (my, our) – in the corpus shows (Table 1), Czech graduate students also use expressions indicating high author visibility to construe their authorial voice. Author reference devices I/me/my we/us/our Total
Linguistics Raw Norm. No. rate 478 5.25 75 0.82 553 6.08
Methodology Raw Norm. No. rate 182 2.14 139 1.64 321 3.78
Raw No. 660 214 874
Total Norm. rate 3.75 1.22 4.97
Tab. 1: Rate of personal pronouns in corpus (normalized rate per 1,000 words) The findings of the analysis show that the frequency of exclusive pronouns (singular I/me and plural editorial we/us) is higher than the rate of inclusive ones (we/us). There is also evidence of disciplinary variation which seems to be associated with the rate of author-reference forms, rather than with their functional specialization. The rather descriptive character of methodology theses, in which authors present their observations on the teaching and learning processes and create communality on the basis of shared experience of language learning, seems to favour a higher rate of inclusive and a lower rate of exclusive pronouns. The more interpretative character of linguistics theses, on the other hand, is more readily associated with the expression of personal commitment to claims, which results in a more prominent occurrence of exclusive pronouns. When analyzing the discourse of novice non-native speakers, it is relevant to consider both the extent to which novice writers use the potential of personal pronouns to create a coherent authorial voice and what motivates their choices. When compared to the rate of different forms of author-reference devices in published research articles (Table 2), the findings show that graduate students underuse all author-reference forms, which confirms the results of previous research into novice academic discourse (e.g. Hyland 2002b; Samraj 2008). Nevertheless, these findings clearly indicate that graduate students are aware of the importance of interpersonal meanings in academic discourse, as all theses included in the corpus use inclusive and exclusive author-reference devices. While
Olga Dontcheva-Navratilova and Renata Povolná
37
singular first person forms commonly occur in the Introduction and Conclusion sections and often in the Materials and Methods, Discussion and Conclusions sections, plural pronouns are more frequent in the Introduction, Literature Review and Methodology sections. The most common pronominal form for projecting the authorial voice into the discourse is the subjective I, while we – the second in frequency in published research – is distinctly underused, especially in the linguistics theses. The possessive forms – which have the potential “to promote the writer’s contribution by associating them closely with their work” (Hyland 2001: 223) – also show a significant rate; for instance, the frequency of my in the linguistics theses is even higher than in published research. Authorreference devices I me my we us our
Linguistics theses
Methodology theses
Research articles
2.96 0.48 1.81 0.47 0.18 0.18
1.19 0.12 0.84 1.42 0.04 0.18
3.61 0.3 0.97 2.54 0.28 1.45
Tab. 2: Comparison of the frequency (per 1,000 words) of pronouns in the corpus and their rate in applied linguistics research articles as reported by Hyland (2001) An analysis of the functional specialization of the author-reference pronouns used by the students may shed light on reasons for the variation in the rate of the author-reference forms. In agreement with Hyland’s (2002a: 1099) findings, the results of the analysis indicate that the most frequent functions of author-reference devices are the stating of discoursal goal/purpose or the explaining of procedure. In the Introduction, the exclusive personal pronoun I is commonly used for making centrality claims, stating the goal and previewing the organization of the thesis, as in (1). Writer involvement may be further enhanced by the use of the possessive my with text and research nouns (e.g. thesis, work, study, analysis). (1) In my diploma thesis, I have decided to survey the methodological field and focus my attention to one of the four language skills, namely speaking. My concern with this topic is not accidental for I consider
38
Development of Academic Writing Skills in English as a Lingua Franca
speaking one of the major concepts in language teaching. (Methodology) Explicit author reference is also used to facilitate coherent transition between thematic segments (2) and structural units, thus fostering the dialogicality of the discourse. (2) Let me return to the analysed news items and discuss the overuse of references to women’s appearance. (Linguistics) The prominence of the discourse-organizing role of author-reference pronouns indicates a growing awareness on the part of graduate students of the academic conventions of explicit discourse organization. At the same time, however, it shows a preference for low-risk writer roles related to signalling intentions, outlining argument structure and guiding the reader through the discourse, which, although foregrounding writer visibility, do not involve commitment to claims and expressing opinions (cf. Hyland 2002a). The use of the agentive I for explaining procedures and collecting data in the Materials and Methods section also seems to involve a low degree of personal exposure (3). However, when the writer commits him/herself to subjective choices concerning data processing and research methods, author-reference involves claiming the authority of a competent member of the discourse community to take such decisions (Hyland 2002a: 1102). This use of I is not very frequent and is more common in linguistics theses. (3) For my analysis I have used texts from London-Lund Corpus (for text classification, see Svartvik 1990), namely two face-to-face conversations (S.1.4 and S.1.13) and two telephone conversations (S.8.1 and S.8.3). (Linguistics) In the methodology theses, however, some authors use the exclusive editorial we for self-reference, especially when guiding the reader through the text; this might be interpreted as interference from the Czech academic tradition, which tends to use this strategy to reduce personal attributions (Chamonikolasová 2005): (4) We devote the initial section to the class atmosphere set-up as it plays an important role in creating a positive emotional climate in class. (Methodology)
Olga Dontcheva-Navratilova and Renata Povolná
39
It should be mentioned, however, that this use of we may also be seen as exploiting the exclusive/inclusive ambiguity of the first person plural pronoun (cf. Harwood 2005, Hyland 2001) to achieve reader involvement and assume the agreement of the reader with the claims put forward by the author. An example of this is provided by (5), where the first occurrence of we may be interpreted as referring exclusively to the author; nevertheless, due to the inherent ambiguity of we it may also be seen as granting the writer the authority to assume the agreement of the reader with his/her views. The second occurrence of we in (5) is clearly inclusive and by achieving reader involvement and the perception of commonality it contributes to the persuasiveness of the argument. (5) We can state that the wide range of the listening activities, applied to the mainstream students, is used for the development of the listening skills of the learners with SLD as well. If they suffer from the specific difficulties described above, we need to use some extra methods to strengthen their listening skills. (Methodology) The use of author-reference pronouns with cognitive verbs (e.g. think, believe, assume, explain, suggest, claim, conclude, consider, doubt) for expressing opinions, elaborating arguments and stating findings can be found primarily in conclusions, where while restating the purpose and summarizing the main findings and claims in the theses, the authors explicitly take responsibility for them: (6) I have put forward the hypothesis that newspapers, mostly tabloids, adopt an ideology that views women as ‘the weaker sex’. Having analysed in detail news items of the tabloid, I have concluded that a prejudicial categorization is manifested in a widespread manner. (Linguistics) It is hardly surprising that author-reference structures indicating powerful authorial presence are not very frequent in the graduate students’ discourse. This is in consonance with the findings of previous research which shows that non-native students tend to “either underuse writer pronouns or use them unadventurously, referring to their texts rather than their ideas” (Hyland 2002b: 353). The restricted use of authoritative author-reference devices by graduate students may be interpreted as a sign of their reluctance to assume the responsibility of a clear commitment to opinions and claims, which may be enhanced by the influence of the L1
40
Development of Academic Writing Skills in English as a Lingua Franca
academic writing tradition and an insufficient awareness of academic genre conventions. A comparison of the results of the survey with the analysis of the graduate students’ discourse shows that while the students were instructed on how to use the scientific paradigm, they project their authorial voice into the text by the use of inclusive and exclusive first person pronouns and possessive demonstratives. This seems to be the result of instruction received in academic writing courses and advice provided by supervisors of diploma theses. In addition, the students might have realized the importance of the interpersonal dimension in academic discourse when consulting secondary sources during the writing of their theses. The fact that some of the students use the exclusive we for self-reference is obviously motivated by the Czech academic discourse tradition and advice provided by instructors. Although this may be regarded as a drawback from the point of view of Anglo-American literacy, especially in consideration of the fact that the students are future teachers of English, it is hardly perceived as such by the examiners, who themselves are formed by Czech academic conventions. It can therefore be concluded that the graduate students’ discourse may be seen as an instance of English as a lingua franca in which Anglo-American literacy is influenced by Czech academic norms. b) Discourse markers Since “the process of creating coherent texts involves an indication of relationships between the things one is ‘on about’ ” (Halliday and Hasan 1989: 94), it is not surprising that an appropriate use of signals indicating relationships between adjacent or more distant segments of discourse by the writer is often listed among language phenomena mentioned by authors of academic style manuals (Bennett 2009) and practised in courses of academic writing skills offered to students in university settings. In addition, the frequent use of linking signals reflects “the characteristic choice of [academic written discourse] to mark the links between ideas overtly” (Biber et al. 1999: 880). An appropriate use of linking devices, including discourse markers (DMs), can enhance a deductive process of interpretation and understanding of the text by the prospective reader(s), which is especially important in academic written discourse, where authors need to present and support their arguments to an academic audience in a straightforward and comprehensive way (for differences between academic texts written in English and Czech, cf. e.g. Chamonikolasová 2005). The easier and faster processing of the text by the reader(s) can be achieved by signposting
Olga Dontcheva-Navratilova and Renata Povolná
41
through text organizers such as DMs (cf. Haberlandt 1982). Their explicit use is closely related to a dialogic, more interactive and reader-oriented character regarded as typical of academic texts written in English. Such text characteristics are in contrast to rather monologic, less interactive and more writer-oriented texts usually connected with Teutonic intellectual traditions (Galtung 1985) and attributed to academic texts written in German, Polish and Czech (Clyne 1987, Duszak 1997, Chamonikolasová 2005). Owing to the expected differences between academic texts produced by native speakers of English and those written by non-native speakers, in this case Czech university students, it was considered important to discover to what extent novice writers included in the investigation follow the tendency to produce more reader-oriented and thus reader-friendly texts with overall better organization of discourse, which is in agreement with Anglo-American academic style conventions for building coherent and at the same time cohesive texts. As stated above, the results discussed and exemplified below have been drawn from a small-scale corpus-based analysis of ten students’ diploma theses. In this chapter the analysis is only concerned with causal and contrastive DMs, i.e. the two semantic categories of DMs which rank among the most informative and at the same time most complex semantic relations of all that can hold between segments of discourse (Kortmann 1991: 160-164). The analysis deals with causal and contrastive relations obtaining at clausal and higher levels of discourse, since at these levels the markers relate two separate messages, hence functioning as DMs (Fraser 1999: 939-940). Moreover, the expression of semantic relations at these higher levels of discourse is of crucial importance when building coherent texts of greater length such as students’ diploma theses. From the morphological point of view, the DMs under scrutiny are drawn primarily from conjunctions (e.g. because, although, but), adverbs (e.g. therefore, however, nevertheless) and prepositional phrases (e.g. on the other hand), and their meaning is procedural rather than conceptual (cf. Fraser 1999). The author uses DMs intentionally in order to enable his/her reader(s) to derive meaningful discourse from the text. If a marker is absent, the propositional content of the respective discourse segments remains the same, but without any explicit guiding signal it could be more difficult for the reader(s) to arrive at the interpretation intended by the author. Therefore it is assumed that students are aware of the necessity of using linking devices including DMs in an appropriate way and be instructed accordingly by their diploma thesis supervisors and/or academic writing skills teachers (cf. Bennett 2009).
42
Development of Academic Writing Skills in English as a Lingua Franca
As for the syntactic viewpoint, DMs can be divided into those expressing hypotactic and those involved in paratactic relations. The reason for this division is the expected difference between the two syntactic groups in the frequency rates within both semantic categories of DMs selected for the analysis. The hypotactic relation is usually expressed overtly by certain markers, such as because and although, while the paratactic relation, apart from being indicated by certain markers, such as thus, however and nevertheless, can often remain unexpressed explicitly. Consequently, a high number of overtly expressed DMs, in particular those occurring in hypotactic relations, was expected, because these are mostly signalled overtly and, as stated in Taboada, the semantic relations of cause and concession are “typically expressed through subordination” (2006: 576). Table 3 testifies that in students’ diploma theses causal relations expressed by explicit hypotactic DMs, although realized by three different markers only (because, as and since), tend to be relatively frequent (1.92 tokens per 1,000 words) when compared with those expressed by explicit paratactic markers (2.28). This result confirms that hypotactic relations are mostly marked overtly in academic discourse. In contrast to the hypotactic DMs, all of which are relatively frequent in the data, the paratactic ones are rather unevenly distributed, some of them having a frequency rate of less than 0.2 (40 tokens) in the data; owing to their rather limited use, the DMs accordingly, as a consequence, as a result, consequently, hence, for, in consequence, now, of course, so that, then and somehow have been excluded from the table, although they are included in the last two lines. (See e.g. 2.36/2.93 in the first column, where 2.36 is the frequency rate of paratactic DMs per 1,000 words listed in the table, while 2.93 equals the total frequency rate of all paratactic DMs found in the linguistics subcorpus.) Of the 15 different types of paratactic causal DMs searched for during the analysis only three, namely so, therefore and thus occur with noteworthy frequency (40 or more occurrences); these are all listed in the following table.
Olga Dontcheva-Navratilova and Renata Povolná
No. of words in texts Hypotacti c DMs as because since All hypot. DMs Paratactic DMs (and) so therefore thus All parat. DMs TOTAL DMs
Linguistics theses (87,636) Raw No. 54 82 30 166
Norm. rate 0.62 0.94 0.34 1.89
Raw No. 36 84 87 207/25 7 423
Norm. rate 0.41 0.96 0.99 2.36/2.9 3 4.83
Methodology theses (88,628) Raw Norm. No. rate 88 0.99 68 0.77 16 0.18 172 1.94 Raw No. 20 36 42 98/14 5 317
Norm. rate 0.23 0.41 0.47 1.11/1.6 4 3.58
43
Total (176,264) Raw No. 142 150 46 338
Norm. rate 0.81 0.85 0.26 1.92
Raw No. 56 120 129 305/40 2 740
Norm. rate 0.32 0.68 0.73 1.73/2.2 8 4.20
Tab. 3: Most frequent DMs for hypotactic and paratactic causal relations in students’ theses The novice writers included in the investigation use certain causal markers very frequently, such as because (150), as (142), thus (129) and therefore (120), or relatively frequently, such as so (56) and since (46). It is worth noting that the most commonly applied markers of all, i.e. because, as, thus and therefore, represent almost three quarters of all causal markers found in the corpus. It can thus be concluded that the students prefer to use a relatively limited repertoire of the more common causal DMs than resorting to a wider choice of the DMs which are at their disposal (18 different types), notably when expressing paratactic relations, which enable a ‘natural’ way of sequencing discourse segments from known to new information (cf. e.g. consequently and thus in (7) below). There are, of course, differences between the two fields of study in terms of both types and tokens of DMs the students apply. These can be caused in particular by students’ comprehension and individual preferences in writing habits resulting from overt field-specific instructions provided by theses supervisors, which, for example, in the case of linguistics theses has led to the overuse of certain more frequent DMs, as illustrated in (7), which comprises both causal and contrastive DMs:
44
Development of Academic Writing Skills in English as a Lingua Franca
(7) The questionable character of Sample 19 is twofold. The first pair liable and responsible may at first seem to fall in the category of transition, but in the author’s opinion it is not, because the structure of the binomial is not auxiliary + main verb (or its past participle) as in Sample 15, but auxiliary + main verb + adjective. The second pair may not appear to be in transition because it is preceded by a clause and consequently found in the thematic part. However, in the author’s opinion it is a transition element because the two clauses are connected by the link and, and thus making the relation of the two predicates coordinate. (Linguistics) Table 4, provided below, provides evidence that, contrary to expectations, contrastive relations expressed by hypotactic DMs are considerably less frequent (1.08 tokens per 1,000 words) in the corpus than those expressed by paratactic DMs (4.36), although hypotactic relations are usually marked overtly in discourse (cf. above). A preference for the use of paratactic DMs can result from the tendency favoured by the students to order segments of discourse in a ‘natural’ way, which entails the placement of a segment with known information first, as with however in (8), i.e. before new and/or unexpected information (for the ‘basic distribution of CD’, cf. Firbas 1992). (8) It is evident from the previous statement that in order to communicate both types of skill have to be employed. However, there is still another step to be taken in the long and effortful process of learning to speak a second language. (Methodology) The most interesting finding in Table 4 is the uneven distribution of the contrastive DMs (altogether 959) in terms of both types and tokens. Of the 38 different types included in the analysis only six occur with noteworthy frequency, i.e. having a normalized frequency rate higher than 0.2 (40 occurrences); this concerns only one hypotactic marker (although) out of the nine included in the analysis, and five paratactic ones out of the 29 searched for in the data
Olga Dontcheva-Navratilova and Renata Povolná
. No. of words in texts Hypotactic DMs although All hypot. DMs Paratactic DMs but however neverthele ss on the other hand yet All parat. DMs TOTAL DMs
Linguistics theses (87,636) Raw Norm. No. rate 32 0.37 32/118 0.37/1.3 5 Raw Norm. No. rate 180 2.05 58 0.66 34 0.39
Methodology theses (88,628) Raw Norm. No. rate 25 0.28 25/73 0.28/0.8 2 Raw Norm. No. rate 162 1.83 85 0.96 10 0.11
45
Total (176,264) Raw No. 57 57/191 Raw No. 342 143 44
Norm. rate 0.32 0.32/1.0 8 Norm. rate 1.94 0.81 0.25
36
0.41
22
0.25
58
0.33
56 364/45 0 568
0.64 4.15/5.1 3 6.48
1 280/31 8 391
0.01 3.16/3.5 9 4.41
57 644/76 8 959
0.32 3.65/4.3 6 5.44
Tab. 4: Most frequent DMs for hypotactic and paratactic contrastive relations in students’ theses Similarly to causal DMs, some contrastive DMs are very frequent, such as but (342) and however (143), or relatively frequent, such as on the other hand (58; frequently used probably under the influence of the Czech phrase na druhé stranČ), although (57), yet (57) and nevertheless (44). Of these, the informal paratactic marker but (342) represents more than one third of all tokens found in the data, while being almost six times more frequent than the most common hypotactic DM although (57). Example (9) illustrates a mistake some of the students occasionally make, i.e. the use of a comma after a hypotactic marker such as although and though. This type of mistake, caused in our opinion by a student’s inability to distinguish between conjunctions and adverbs, occurs above all in the methodology theses. (9) Though the performances were not filmed again, the final discussion was very interesting and lively, as the students fully
46
Development of Academic Writing Skills in English as a Lingua Franca
identified with their new identities and situations. Although, the organization of the project in this mixed ability class was rather complicated and difficulties were also encountered with making the students start the activities, finally, they got fully involved in and enjoyed the project. (Methodology) No. of words in texts Hypotactic DMs although All hypot. DMs Paratactic DMs but however neverthele ss on the oth. hand yet All parat. DMs TOTAL DMs
Linguistics theses (87,636) Raw Norm. No. rate 32 0.37 32/118 0.37/1.3 5 Raw Norm. No. rate 180 2.05 58 0.66 34 0.39
Methodology theses (88,628) Raw Norm. No. rate 25 0.28 25/73 0.28/0.8 2 Raw Norm. No. rate 162 1.83 85 0.96 10 0.11
Total (176,264) Raw No. 57 57/191 Raw No. 342 143 44
Norm. rate 0.32 0.32/1.0 8 Norm. rate 1.94 0.81 0.25
36
0.41
22
0.25
58
0.33
56 364/45 0 568
0.64 4.15/5.1 3 6.48
1 280/31 8 391
0.01 3.16/3.5 9 4.41
57 644/76 8 959
0.32 3.65/4.3 6 5.44
Tab. 5: Most frequent DMs for hypotactic and paratactic contrastive relations in students’ theses Table 5 does not include the following contrastive DMs, although, similarly to Table 3 above, they are listed in the lines which give total numbers of both hypotactic and paratactic DMs. The exclusion concerns the hypotactic albeit, at any rate, by comparison, despite the fact (that), even if, even though, except (that), in comparison, in spite of the fact (that), or else, though, while and whereas, and the paratactic actually, after all, all the same, alternatively, anyhow, anyway, at the same time, besides, by contrast, conversely, in any case, in contrast, in spite of that, instead, nonetheless, notwithstanding, on the other side, oppositely, still and though. As stated in Altenberg (1986), with DMs such as albeit,
Olga Dontcheva-Navratilova and Renata Povolná
47
notwithstanding and oppositely, zero occurrence is not at all surprising, since these markers are not likely to appear in any corpus of contemporary English. As with the choice of causal DMs, it can now be postulated that when writing diploma theses the students do not frequently resort to the whole repertoire of DMs that are at their disposal. Even though there is a much wider availability of contrastive markers (38 types) than causal markers (18 types), the spectrum of DMs the students really often choose to use when expressing cause and contrast is rather narrow for both semantic categories (only 6 different types). These findings can be further evidenced by the findings drawn from the comparison of the use of causal and contrastive DMs by Czech university students as writers of diploma theses and native speakers of English as authors of research articles published between the years 2000 and 2008 and selected from the journal Applied Linguistics (cf. Povolná 2012b) .Most common causal and contrastive DMs as because since (and) so therefore thus although but however nevertheless on the other hand yet
Linguistics theses (Czech students) 0.62 0.94 0.34 0.41 0.96 0.99 0.37 2.05 0.66 0.39 0.41 0.64
Methodology theses(Czech s)
Research articles (native E speakers)
0.99 0.77 0.18 0.23 0.41 0.47 0.28 1.83 0.96 0.11 0.25 0.01
0.29 0.66 0.40 0.26 0.79 0.66 0.74 2.03 1.62 0.16 0 0.11
Tab. 6: Comparison of the frequency rates (per 1,000 words) of the most common causal and contrastive DMs in students’ theses and research articles by native speakers of English (as presented in Povolná 2012b) Table 6, in which only the normalized frequency rates per 1,000 words of the most common causal and contrastive DMs are given, further testifies to the tendency favoured by the diploma thesis students to resort
48
Development of Academic Writing Skills in English as a Lingua Franca
to a narrow range of DMs they know well and thus prefer to use, sometimes rather repeatedly (cf. (7) above). This tendency is evidenced by much higher frequency rates of some of the DMs listed above, notably in linguistics theses (e.g. therefore, thus, nevertheless), in which the students tend to overuse certain DMs they like to apply while introducing every other segment of discourse with a marker, as illustrated in (7) and mentioned above in the discussion on the results drawn from the graduate students’ questionnaires. Table 5 also confirms Biber et al.’s finding (1999) that however, therefore and thus rank among the most common DMs of all in academic texts written by native speakers of English. Finally, it can be postulated that in their diploma theses the graduate students attempt to apply explicit links between discourse segments as far as they are able and thus contribute to the dialogic and reader-oriented character of their theses, notably in agreement with Anglo-American conventions of building coherent reader-friendly texts, although they tend to apply only a limited repertoire of linking devices which they sometimes overuse, probably under the influence of overt instructions by teachers of academic writing and diploma thesis supervisors.
6. Conclusions The results of this investigation have confirmed that academic writing as a highly complex skill requires systematic tuition throughout the whole course of university studies. It is therefore important to raise students’ awareness of the specificities of academic writing and gradually create opportunities for them to acquire the established academic writing conventions in order to become competent writers (and readers) in a foreign language. The results of the needs analysis survey indicate that Czech students of English entering university have previously worked on and received tuition mainly in general writing skills and for many of them academic writing and its specific features are first encountered at university. Thus, when writing academic texts, students rely on advice provided by printed and on-line style manuals, instructions received in academic writing courses and advice provided by thesis supervisors. The survey shows that Bachelor’s students have sufficient practice in writing different kinds of essays resulting from sufficient exposure to a relatively wide range of writing assignments. However, they need more encouragement to do research and write drafts when producing academic texts. Graduate students should also be encouraged to study and edit their own academic texts, thus adopting an approach akin to the ‘on-line’ genre analysis
Olga Dontcheva-Navratilova and Renata Povolná
49
procedure advocated by Flowerdew (1993) for highlighting discipline-, genre- and culture-specific academic conventions. They would benefit from further tuition focusing on argumentation strategies, patterns of textual organization and lexicogrammatical choices associated with particular academic genres and disciplines (Paltridge 2002: 127). Even though both Bachelor’s and Master’s degree students are aware of the serious problems related to plagiarism, not many of them are able to identify its various kinds, which further supports the view that all theoretical tuition has to be accompanied by meaningful activities and assignments. The comparison of the results of the survey with the analysis of graduate students’ discourse has revealed that although the students generally follow the instructions received in academic writing courses, their diploma theses show variation in terms of the use of discourse markers and devices indicating authorial presence. Despite being instructed to use the scientific paradigm, graduate students seem to have realized the importance of the interpersonal dimension in academic discourse in the field of humanities, as they use various self-reference devices to project their authorial voice into the text and discourse markers indicating relationships between segments of discourse when attempting to build coherent texts. The choice of devices reflects their personal preferences, advice received in academic writing courses or provided by diploma thesis supervisors, and in some cases shows signs of interference from the Czech academic discourse tradition. This divergence from AngloAmerican academic writing standards suggests that the future teachers of English are using English as a lingua franca in which Anglo-American academic writing conventions are affected by Czech academic literacy. As to the implications for teaching academic writing at the university level, the results of this investigation suggest that the acquisition of academic writing skills should not be dealt with separately or in isolation from other courses. By contrast, it should be naturally incorporated into various university courses, whether of linguistics, literature or cultural studies, thus enabling students to gain valuable experience concerning argumentation strategies and stylistic appropriateness of language use from expert sources. A syllabus for an efficient academic writing course should reflect the move from teaching general academic writing skills in the Bachelor’s degree programme to discipline- and genre-specific skills required in the Master’s degree programme and include the analysis of authentic academic texts as well as the analysis and editing of students’ works (cf. an approach akin to the ‘on-line’ genre analysis procedure).
50
Development of Academic Writing Skills in English as a Lingua Franca
Based on the above discussion and the results drawn from the questionnaires and corpus-based analysis, we can now postulate our aims for future research, which, in our opinion, should include further comparisons of students’ assessment of their needs and views on courses and teachers of academic writing and supervisors of Bachelor’s/diploma theses, as well as views on students’ needs, interviews with students and teachers to clarify their views, and, last but not least, corpus-based analyses of students’ works exploring the use of argumentation strategies, patterns of textual organization and lexicogrammatical choices.
References Altenberg, B. (1986). Contrastive linking in spoken and written English. In Tottie, G. and Bäcklund, I. (Eds.), English in speech and writing: A symposium (pp.13-40). Uppsala: Almqvist & Wiksell. 13-40. Bennett, K. (2009). English academic style manuals: A survey. Journal of English for Academic Purposes 8, 43-54. Biber, D., Johansson, S., Leech, G., Conrad, S., & Finegan, E. (1999). Longman grammar of spoken and written English. London: Longman. Brown, J. D. (1995). The elements of language curriculum: A systematic approach to program development. Boston: Heinle & Heinle. Chamonikolasová, J. (2005). Comparing the structures of academic texts written in English and Czech. In M. Huttová (Ed.), Slovak Studies in English 1 (pp. 77-84). Bratislava: Univerzita Komenského. Clyne, M. (1987). Cultural differences in the organisation of academic texts. Journal of Pragmatics 11, 211-247. ýmejrková, S. & Daneš, F. (1997). Academic writing and cultural identity: The Case of Czech academic writing. In A. Duzsak (Ed.), Culture and Styles of Academic Discourse (pp. 40-62). Berlin: Mouton de Gruyter. ýmejrková, S., Daneš, F. & SvČtlá, J. (1999). Jak napsat odborný. Prague: Leda. Dontcheva-Navratilova, O. (2008). Reporting verbs as indicators of stance in academic discourse. Porta Lingea-2008. Szakember, Szaktudás, Szakuyelv. Debrecen, 2008/1, 97-107. —. (2009). Evaluation in non-native writer’s academic discourse: Stance devices. In S. Hanušová et al. (Eds.), Research in English Language Teacher Education. Spisy Pedagogické fakulty Masarykovy univerzity, vol. 131, Brno: Masaryk University. —. (2012a). Cross-cultural differences in the construal of authorial voice in the genre of diploma theses. In C. Berkenkotter, V. Bhatia, & M.
Olga Dontcheva-Navratilova and Renata Povolná
51
Gotti (Eds.), Insights into Academic Genres (pp. 301 – 328). Bern: Peter Lang. —. (2012b). Lexical bundles in academic texts by non-native speakers. Brno Studies in English 38/2, 25-46. Dudley-Evans, T. & St John, M. J. (1998). Developments in English for specific purposes. Cambridge: Cambridge University Press. Duszak, A. (Ed.). (1997). Culture and styles of academic discourse. Berlin: Mouton de Gruyter. Firbas, J. (1992). Functional sentence perspective in written and spoken communication. Cambridge: Cambridge University Press. Flowerdew, J. (1993). An educational, or process approach, to the teaching of professional genres. English Language Teaching Journal 47, 305313. —. (2000). Discourse community, legitimate peripheral participation, and the non-native-English-speaking scholar. TESOL Quarterly, Vol. 34/1, 127-150. Retrieved from http://www.jstor.org/stable/3588099. —. (2004). The argument for using English specialized corpora to understand academic and professional language. In U. Connor, & T. Upton (Eds.). Discourse in the professions (pp. 11-33). Amsterdam and Philadelphia: John Benjamins. —. (2008). Scholarly writers who use English as an additional language: What can Goffman’s ‘Stigma’ tell us? Journal of English for Academic Purposes 7, 77-86. Flowerdew, J. and Peacock, M. (2010). Research perspectives on English for academic purposes. Cambridge: Cambridge University Press. Fraser, B. (1999). What are discourse markers? Journal of Pragmatics 31, 931-952. Galtung, J. (1985). Struktur, Kultur und intellectueller Stil. In A. Wierlacher (Ed.). Das Fremde und das Eigene (pp. 151-193). Munchen: Iudicum Verlag. Gosden, H. (1993). Discourse functions of subject in scientific research articles. Applied Linguistics 14, 56-75. Haberlandt, K. (1982). Reader expectations in text comprehension. In J.F. Le Ny & W. Kintsch (Eds.). Language and Comprehension (pp. 239249). Amsterdam: North-Holland. Halliday, M. A. K. and Hasan, R. (1989). Language, context, and text: Aspects of language in a social-semiotic perspective. (2nd ed.). Oxford: Oxford University Press. Hamp-Lyons, L. (2001). English for academic purposes. In R. Carter, & D. Nunan (Eds.). The Cambridge guide to teaching English to speakers of other languages. Cambridge: Cambridge University Press.
52
Development of Academic Writing Skills in English as a Lingua Franca
Hamp-Lyons, L. & Heasley, B. (2006). Study writing. A course in writing skills for academic purposes. (2nd ed.). Cambridge: Cambridge University Press. Harwood, N. (2005). We do not seem to have a theory… The theory I present here attempts to fill this gap: Inclusive and exclusive pronouns in academic writing. Applied Linguistics 26, 343-375. Hedgcock, J. S. (2005). Taking stock of research and pedagogy in L2 writing. In E. Hinkel (Ed.). Handbook of Research in Second Language Teaching and Learning (pp. 597-613). Mahwah and London: Lawrence Erlbaum. Hewings, M. (Ed.). (2006). Academic writing in context. London: Continuum. Hewing, A., Lillis, T. & Vladimirou, D. (2010). Who’s citing whose writings? A corpus based study of citations as interpersonal resource in English medium national and English medium international journals. Journal of English for Academic Purposes 9, 102-115. Hinkel, E. (2004). Teaching academic ESL writing. Mahwah, NJ: Lawrence Erlbaum. Hyland, K. (1996). Second language writing. Cambridge: Cambridge University Press. —. (2001). Humble servants of the discipline? Self-mention in research articles. English for Specific Purposes 18, 207-226. —. (2002a). Options on identity in academic writing. ELT Journal 56/4, 351-358. —. (2002b). Specificity revisited: How far should we go now? English for Specific Purposes 21, 385-395. —. (2006). English for academic purposes: An advanced resource book. Abingdon and New York: Routledge. —. (2008). As can be seen: Lexical bundles and disciplinary variation. English for Specific Purposes 27, 4-21. —. (2010). Constructing proximity: Relating to readers in popular and professional science. Journal of English for Academic Purposes 9, 116-127. Jordan, R. R. (1997). English for academic purposes: A guide and resource book for teachers. Cambridge: Cambridge University Press. Kortmann, B. (1991). Free adjuncts and absolutes in English. London: Routledge. Kreutz, H. & Harres, A. (1997). Some observations on the distribution and function of hedging in German and English academic writing. In A. Duzsak (Ed.). Culture and styles of academic discourse (pp. 181-201). Berlin: Mouton de Gruyter.
Olga Dontcheva-Navratilova and Renata Povolná
53
Kuo, C.-H. (1999). The use of personal pronouns: Role relationships in scientific journal articles. English for Specific Purposes 18/2, 121-138. Mauranen, A. (1993). Cultural differences in academic discourse – Problems of a linguistic and cultural minority. In L. Löfman, L. KurkiSuonio, S. Pellinen, & J. Lehtonen (Eds.). The Competent Intercultural Communicator (pp. 157-174). AFinLA Yearbook 1993. Mauranen, A., Hynninen, N. & Ranta, E. (2010). English as an academic lingua franca: The ELFA project. English for Specific Purposes 29, 183-190. Mur-DueĖas, P. (2007). I/we focus on…: A cross-cultural analysis of selfmentions in business management research articles. Journal of English for Academic Purposes 6, 143–162. Paltridge, B. (2002). Thesis and dissertation writing: An examination of published advice and actual practice. English for Specific Purposes 21, 125-143. Paltridge, B. & Starfield, S. (2007). Thesis and dissertation writing in a second language. Oxon/New York: Routledge. Povolná, R. (2012a). Causal and contrastive discourse markers in novice academic writing. Brno Studies in English 38/2, 105-122. —. (2012b). Enhancing coherent interpretation in academic written discourse: Cross-cultural variation in the use of discourse markers. In O. Dontcheva-Navratilova, & R. Povolná (Eds.). Discourse interpretation: Approaches and applications (pp. 179-208). Newcastle upon Tyne: Cambridge Scholars Publishing. —. (2010a). Can non-native speakers of English use contrastive discourse markers correctly when writing academic texts? In M. Adam et al. (Eds.). For thy speech bewrayeth thee. A Festschrift for Libuše Dušková (pp. 209-231). Praha: Univerzita Karlova v Praze. —. (2010b). Exploring sequential relations in learner discourse. In A. Ciuk, & K. Molek-Kozakowska (Eds.). Exploring space: Spatial notions in cultural, literary and linguistic studies. Vol. 2: Space in language studies (pp. 177-188). Newcastle upon Tyne, UK: Cambridge Scholars Publishing. Samraj, B. (2008). A discourse analysis of Master’s theses across disciplines with a focus on introductions. English for Academic Purposes 7, 55-67. Schmied, J. (2011). Academic writing in Europe: A survey of approaches and problems. In J. Schmied (Ed.). Academic writing in Europe: Empirical perspectives. REAL Studies 5 (pp. 1-22). Göttingen: Cuvillier Verlag.
54
Development of Academic Writing Skills in English as a Lingua Franca
Stašková, J. (2005). Options of identity: Authorial presence in research articles abstracts. Slovak studies in English 1 (pp. 201-207). Bratislava: Univerzita Komenského. Swales, J. (1997). English as Tyrannnosaurus Rex’, World Englishes, 16 (3). 373-382. —. (2004). Research genres: Explorations and applications. Cambridge: Cambridge University Press. Taboada, M. (2006). Discourse markers as signals (or not) of rhetorical relations. Journal of Pragmatics 38/4, 567-592. Tardy, C. (2004). The role of English in scientific communication: Lingua franca or Tyrannosaurus Rex? Journal of English for Academic Purposes 3. 247-269. Tang, R. & John, S. (1999). The ‘I’ in identity: Exploring writer identity in student academic writing through the first person pronoun.’ English for Specific Purposes 18, 23-39. Vogel, R. (2008). Sentence linkers in essays and papers by native vs. nonnative writers. Discourse and Interaction 1/2 (pp. 119-126). Brno: Masaryk University. Wallace, M. J. (2004). Study skills in English. (2nd ed.). Cambridge: Cambridge University Press.
TEACHING EFFECTIVE ACADEMIC WRITING: IF-CONDITIONALS IN ECONOMICS RESEARCH ARTICLES SILVIA CACCHIANI UNIVERSITY OF MODENA AND REGGIO EMILIA
This contribution looks into knowledge construction in research articles (RAs) in economics. Based on insights from insiders (for one, Brandis 1968) and discourse analytical investigation carried out manually by Merlini Barbaresi (1983), we proceed on the assumption that knowledge construction results from the complex interplay of the macro-speech acts (in the sense of Searle and Vandervecken 1985) of hypothesis, analysis / interpretation / generalization and prediction, where hypothesis and analysis / interpretation / generalization are functional to generalization. Using the tools of corpus linguistics (Scott 2010 [1997]), we provide grounded insights into this central claim (Cacchiani 2011). We then concentrate on the grammar of if-conditionals as a key cohesive device in the text, that is briefly touching upon issues of grammatical form and meaning (in the sense of Purpura 1996).
1. Introduction: Effective writing in the disciplines In a recent paper on writing in the academy, Ken Hyland (2013) makes a case for seeing writing and specialist forms of academic literacy within individual disciplinary practices and beliefs as key to knowledge construction and to students’ education. At the same time, writing is also central to making and negotiating a professional academic career. Depending on their participation or engagement, members of a community of practice can be seen as marginal, peripheral and core (Wenger 1998). Broadly speaking, core participants in a community of practice can be seen also seen as parent members of what Swales (1990) labels a discourse community. Discourse communities share common goal, distinctive
56
Teaching Effective Academic Writing
participatory mechanisms and communicative behaviours and routines, while also owning one or more genres and specific lexis. Within academic communities and specific disciplinary culture, core participants comprise expert members, who shape and use genre features and conventions effectively. Novices are expected to align with these conventions. The empowering potential of writing, Hyland’s (2013) goes on to argue, cannot be overestimated: it is through the full command of language and genre features that peripheral members such as doctoral students can achieve full participation in the specific disciplinary community. And, given the central role of English in academic publishing, catering for the pressing needs of non-native academics to develop academic literacy, and thus connect with the discipline and progress to core participants and credible insiders by aligning with the international community, turns into a most compelling requirement for scholars in non-English-medium institutions. In this context, banking on Cacchiani (2012), the relevant patterns and functions are discussed along the lines of Declerck and Reed’s (2001) comprehensive description of English conditional utterances. The discussion will be centred on discourse-signaling devices in the HEM corpus of research articles, a collection of 2,700,000 tokens equally distributed among the HEM-Economics (HEM-E), HEM-History (HEM-H) and HEM-Marketing (HEM-M).1 Understanding the writing and knowledge construction practices of RAs in economics, we shall see, calls for a radical rethinking of the traditional distinction into three types of canonical conditionals (realis, potentialis, irrealis, andtense patterns 1, 2, 3). The paper concludes with some preliminary remarks on the contribution of corpus-informed genre analysis to L2 EAP writing in the context of a small, elective EAP course intended for PhD students and scholars in economics.
2. Keywords and key lexical bundles in RAs in economics Drawing on findings from Cacchiani (2011), in this section we engage in an interdisciplinary dialogue within the humanities by briefly considering the discourse signaling devices that are unusually frequent, or
1
The HEM corpus was built and is currently hosted at the University of Modena and Reggio Emilia. It comprises research articles from a panorama of academic journals and a broad range of disciplines within economics (HEM-Economics: HEM-E), history (HEM-History: HEM-H) and business / marketing /management (HEM-Marketing: HEM-M) in the years 1999-2000. (See http://www.cla.unimore.cofin/ for more on this point.)
Silvia Cacchiani
57
key (Scott 2010 [1997]) in RAs in economics (HEM-E) in comparison with the HEM-H and the HEM-M. As seen in Table 1, the most salient keywords clearly point to prediction and related macro-speech acts. The typical configuration of RAs in economics, the data shows, is a combination within an expository-argumentative text (Werlich 1976) of the macro-speech acts of hypothesis, data analysis / interpretation / generalization and, though less represented, prediction. RAs in economics proceedthrough the interaction of empirical hypothesis and conditional prediction with data analysis, bearing on models, factors and exogenous and endogenous variables specified by the analyst. This explains recourse tokeywords that realize or name the underlying (macro-) speech act (i), which may come with different degrees of certainty and probability, expressed by modals or other verbs (ii). With content words that describe and present data and statistics (iii), self-reflexive markers pointing to the disciplinary-specific reliance on visuals are also key (iv). Lastly, related to interpretation and generalization are markers of hypothesis or cause and reason relations (cf. Siepmann 2005) between elements within the clause or sentence (v). (Ranking and keyness are given in square brackets.) a) Hypothesis: assume[135; 863,3 / 150; 526,7], assumption[263; 503 / 247; 346,4], assumed [360; 362 / 220; 381,5], suppose[384; 335,7 / 136; 578,2](HEM-E; HEM-H / HEM-HEM-M); b) Analysis/interpretation/generalization:estimated[92; 1.115,10 / 236; 361,4] (HEM-E; HEM-H / HEM-E; HEM-M); estimates[103; 1.069,5], analysis[246; 542,1], analyse [460; 273,3], find[454; 282,7] (HEM-E; HEM-H); c) Prediction: expectations[423; 304,1] (HEM-E; HEM-H); expected [233; 361,6] (HEM-E; HEM-M); d) Modals: will[1.613,4], can[1.414], may[816,2], likely[268,8] (HEM-E; HEM-H); e) Other verbs:denote[336,3 / 314,7], implies [318,7 / 488,8] (HEM-E; HEM-H / HEM-E; HEM-M), shows [318,7] (HEM-E; HEM-H); f) Data: results [15; 3.663,7], data [48; 1.765,3], findings [377; 339,1] (HEM-E; HEM-H); g) Statistics: probability [73; 1.307,3 / 172; 472,7] (HEM-E; HEM-H; HEM-E; HEM-M); model [111; 5.443,3], variables 26; [2.800,1], variable [46; 1.839,7], sample [53; 1.718,6], models [11; 1.013,7], hypothesis [141; 835,7], mean [383; 335,5], conditional [437; 294,7] (HEM-E; HEM-H);
19.918
6.575
4.405
4.242
11.833
2.640
1.821
7.728
2.749
1.997
5.794
6.663
1.467
1.227
7 Model
11 Table
15 Results
19 We
26 Variables
46 Variable
47 If
48 Data
53 Sample
59 Will
65 Can
73 Probability
92 Estimated
1 Is
WORD
4 Are
N
0,05
0,06
0,26
0,22
0,08
0,11
0,3
0,07
0,1
0,45
0,16
0,17
0,25
0,76
77
107
2.756
2.051
162
417
3.037
83
94
4.293
341
329
594
6.697
1.115,1 423 expectations
1.307,3 409 then
0,11 1.414,0 402 shows
0,08 1.613,4 384 mean
1.718,6 384 suppose
0,02 1.765,3 383 denote
0,12 1.806,3 377 findings
1.839,7 360 assumed
2.800,1 353 therefore
0,18 3.173,8 330 due
0,01 3.663,7 263 assumption
0,01 3.900,2 252 hence
0,02 5.443,3 246 analysis
0,27 6.015,5 218 follows
HEMH. KEY N WORD LST NESS % 0,76 9.526,1 203 case
Teaching Effective Academic Writing
HEME. FREQ. FREQ. LST % 44.683 1,71 18.537
Keywords (keyness) [HEM-E; HEM-H]*
58
593
3.252
1.091
879
492
379
503
905
1.854
1.014
993
1.180
2.109
1.012
0,02
0,12
0,04
0,03
0,02
0,01
0,02
0,03
0,07
0,04
0,04
0,05
0,08
0,04
123
1.828
374
246
67
28
70
243
793
277
209
283
784
171 0,03
0,07
0,02
0,01
0,03
0,01
0,01
304,1
312,4
318,7
335,5
335,7
336,3
339,1
362
367,4
397,9
503
530,8
542,1
605,8
HEMHEMM. KEY E. FREQ. FREQ. LST NESS LST % % 3.341 0,13 1.453 0,06 639,7
1.225
1.146
4.838
16.541
135 Assume
141 Hypothesis
146 May
171 Be
7.728
4.405
3.341
2.369
3.252
32 If
37 Table
75 Case
95 Since
131 Then
0,12
0,09
0,13
0,17
0,3
3.836
2.190
3.364
4.114
8.888
42
0,07
0,04
0,06
584,5 349 there
772,3 272 denote
910,3 247 assumption
0,07 1.407,9 236 estimated
0,16 1.515,5 233 expected
3.980,7 220 assumed
1,07 5.574,1 201 follows
0,07
59.130
1.912
1,71
44.683
719,5
740,4 460 analyse
816,2 460 analyse
5 Is
0,46
0,09
835,7 454 find
868,3 449 likely
1.013,7 439 denotes
1.069,5 437 conditional
11 Fig
105
5.243
2.228
138
156
219
75
HEMKEY M. NES N WORD LST S % 6,04 6.723,9 172 probability
0,04
0,63
0,19
0,04
0,05
0,06
0,04
Keywords (keyness) [HEM-E; HEM-M]* HEME. FREQ. N WORD FREQ. LST % 3 The 198.074 7,58 334.918
948
1.527
111 Models
179 Implies
1.157
103 Estimates
Silvia Cacchiani
0,01
0,01
0,05
0,05
0,01
33
33
543
494
6
19
0,02
0,02
273,3
273,3
282,7
268,8
292,9
294,7
4.710
379
993
1,227
2.340
905
1.012
0,18
0,01
0,04
0,05
0,09
0,03
0,04
7.432
170
886
1.250
2.904
724
411
0,13
0,02
0,02
0,05
0,01
0,01
246,7
314,7
346,4
361,8
361,6
381,5
298,7
HEMHEMM. KEY E. FREQ. LST NESS LST % % 1.467 0,06 1364 0,02 472,7 FREQ.
340
340
1.320
1.245
259
313
59
1.225
1.180
948
150 Assume
167 Hence
168 Implies
0,04
0,05
0,05
0,02
662
948
966
132
0,01
0,02
0,02 488,8
494,1 495 denotes
526,7 493 shall
578,2 379 are
Teaching Effective Academic Writing 0,76 142
88
36.996 178, 1
178,1
0,67 227,03
Tab. 1: Keywords (keyness) [HEM-E; HEM-H]; [HEM-E; HEM-M]. (adapted from Cacchiani 2011, Tables 1 and 2)
259
206
19.918
*: Italics is used for words that are key in both [HEM-E; HEM-H] and [HEM-E; HEM-M].
492
136 Suppose
60
Silvia Cacchiani
61
a) Self-reflexive markers for visuals:table [11; 3.900,2 / 37; 1407,9] (HEM-E; HEM-H / HEM-E; HEM-M); fig [11; 3.980,70] (HEM-E; HEM-M); b) Markers of hypothesis or cause and reason relations:if [47; 1.806,3 / 32; 1.515,5], case [203; 639,7 / 75; 910,3], hence [252; 530,8 / 167; 494,1], then [409; 312,4 / 131; 584,5] (HEM-E; HEM-H; HEM-E; HEM-M);due [330; 397,9], therefore [353; 367,4] (HEME; HEM-H); since [95; 772], follows [201; 298,7] (HEM-E; HEMM). Analysis of 3-, 4-, and 5-token key lexical bundles1 along the lines of recent research on disciplinary variation (Hyland 2008) corroborates and lends support to these findings.2If we now set aside structuring signals (Hyland 2008), or text-oriented bundles which are concerned with organization and purpose of the paper, it is clear that the remaining bundles help construct knowledge around the complex interplay and partial overlap of the macro-speech acts of hypothesis (a), analysis / interpretation / generalization, and prediction, where hypothesis is empirical and prediction conditional. More particularly, definitions and parameter settings (b) are functional to data analysis and discussion. Also, given the heavy reliance of economics on statistics and figures, graphs and tables, as well as quantification bundles comprising description and location signals, text-oriented bundles of the structuring type are also frequent and highly salient (c). When applying a model, individual variables are tested in turn, and framing signals are used to delimit an argument and specify cases as well as conditions and circumstances of application of models and variables within the model. As such, they score high among the top 25 key bundles (d). Reliable inferencing is behind the choice of resultative signals, which allow the analyst to discuss research processes and outcomes in data interpretation and generalizations (e) with different degrees of certainty, e.g. via recurrence to stance and engagement bundles, also in conditional predictions. Some examples (Cacchiani 2011, modified):
1
Lexical bundles (Scott 2010 [1997]: clusters) are word strings that appear in a genre more frequently than expected by chance, and (ii) occur in multiple texts in that genre (Biber et al. 1999). Considering corpus size and research purpose, frequency cut-off point at 2 hits per million word. 2 See Cacchiani (2011) for extensive discussion of the functional categorization of lexical bundles in RAs in economics into non-mutually exclusive sets, and for the terms adopted.
62
Teaching Effective Academic Writing
a) Hypothesis: it is assumed that the [17; 52,7 / 14; 111.6] (HEM-E; HEM-H / HEM-E; HEM-M);we assume that the [10; 175,5], is assumed to be [11; 171,4] (HEM-E; HEM-H); b) Procedural signals for definitions and parameter setting: is a function of the [59; 36,9 / 37; 84,2] (HEM-E; HEM-H / HEM-EHEM-M); can be thought of as [25; 97,9], is the ratio of [91; 52,1] (HEM-E; HEM-M); c) Quantification signals: an increase in the [1; 354,4 / 3; 251,4],significantlydifferent from zero [26; 113,4 / 14; 97], the # level of significance[30; 46,1 / 50; 79,7] (HEM-E; HEM-H / HEME; HEM-M); significant at the # level [3; 134,5] (HEM-E; HEMH); to an increase in the [3; 47],the size of the [9; 127,6], (HEM-E; HEM-M); d) Location signals:in table # and table[7; 80,3 / 11; 132](HEM-E; HEM-H / HEM-E; HEM-M), panel B of table #[21; 51,4], the lefthand side of [50; 38,2] (HEM-E; HEM-H); at the end of the[4; 193,5], at the beginning of the [6; 141,2] (HEM-E; HEM-M); e) Description signals:the standard deviation of the[72; 33], the present value of the[73; 55], the dependent variable is the[79; 31,6] (HEM-E; HEM-H); percentage change relative to the[74; 68,3], the probability of success of [81; 66] (HEM-E; HEM-M); f) Structuring bundles:table # and table #[1; 242,6 / 1; 418,9], are presented in table #[5; 101,5 / 5; 177,6], are reported in table [8; 76,5 / 12; 127,5] (HEM-E; HEM-H / HEM-E; HEM-M). g) Framing signals: in the presence of the[23; 124,2], with respect to the[48; 90,6], if and only if the[67; 34,3], if andonly if[5; 197,7], the case in which [155; 28,7], other things being equal[478; 31,6] (HEM-E; HEM-H); in the context of the[19; 102,5], in the case of a[20; 102,5], as in the case of [25; 100,2], if and only if[6; 151,3] (HEM-E; HEM-M). h) Resultative signals:these results are consistent with[44; 39,5 / 70; 68,3] (HEM-E; HEM-H / HEM-E; HEM-M);table # reports the results [26; 47,5], this is due to the[84; 31,6], leads to an increase in[129; 26,4], is determined by the[87; 71,9], the reason is that[110; 65,9], these results suggest that[157; 55,4], this implies that the[166; 54] (HEM-E; HEM-H); as a result of the [10; 134,3], due to the fact that [23; 100,2], on the basis of the [48; 79,5], it follows that the[400; 28,5] (HEM-E; HEM-M) i) Stance and engagement bundles:it can be shown that [31; 46,1 / 16; 109,3], can be thought of as [12; 59,3 / 25; 97,9] (HEM-E; HEMH; HEM-E; HEM-M), it is easy to see [66; 79,7] (HEM-E; HEM-
Silvia Cacchiani
63
M), also marking prediction, as in may not be [186; 123,1] or in are more likely to be [20; 50,1] (HEM-E; HEM-H), ranking among the 25 key bundles.
3. If-conditionals The analysis to this point strongly suggests that the centrality of empirical hypothesis and conditional prediction is perhaps the most distinctive feature of RAs in economics, which sits in well with the keyness of if, assum*, therefore, thus, suggest*, denote,impl*, hypothes*, estimate*, result*, find*, likely or of lexical bundles such as if and only if, these results suggest that, this implies that. Turning to if as the most key of all cohesive devices, itsvery keynessalso suggests that the relevance of the type of coherence relations instantiated in conditional sentences is not minimal. We therefore focus on if-conditionals in an attempt to describe the full range of meanings that they express in economics RAs.3
3.1. Possible-worlds In line with Declerck and Reed (2001), we use P / P-clause for the protasis, or conditional clause, introduced here by the most basic conjunction if, and Q /Q-clause for the apodosis, or main clause. The major reason why a closer look at if-conditionals (If P, Q structures) is needed is that the categorization into canonical (tense) pattern 1, 2 and 3 conditionals offered in pedagogic grammars and textbooks on the basis of form (i.e., tense) may be problematic for RAs in economics. Examples (1a)to (1c) illustrate the three patterns in predictive conditionals (term from Dancygier 1998): (1a) If we reduce environmental standards, the production in the pollution-intensive sector will become more efficient. (constructed) (1b) If we reduced environmental standards, the production in the pollution-intensive sector would become more efficient. (constructed) (1c) If we had reduced environmental standards, the production in the pollution-intensive sector would have become more efficient. (constructed)
3
This section elaborates and revises a previously published paper by Cacchiani (2012).
64
Teaching Effective Academic Writing
In (1a), a case of potentialis,the validity of P is represented as possibly true (Declerck and Reed 2001: open). P-clause is in the present tense and Q-clause is in the future tense (canonical pattern 1). Thirdly, if can be replaced by assuming that, supposing that, supposing, assuming, or exhortative let, suppose, assume.If the conditional is assumed to be fulfilled in the actual world, we would have a closed-P conditional, as in (1d):if can be replaced by given that, andan inferential relation links the Pclause and the Q-clause, which may refer to different times. (1d) If, as you say / as I believe, they don’t reduce environmental standards / hadn’t reduced environmental standards yet, the production in the pollution-intensive sector will not become more efficient. (constructed) In (1b), a case of irrealis, the P-proposition is represented as unlikely to be true (Declerck and Reed 2001: tentative); P-clause is in the past tense and Q-clause is in the conditional tense (canonical pattern 2). In (1c), a case of irrealis, the validity of P is represented as contrary to fact and, thus, falsein the actual world (Declerck and Reed 2001: counterfactual);Qclause is in the past perfect and Q-clause is in the conditional perfect tense. (2a) and (2b) would be non-canonical pattern 1 realisconditionals. They are unbounded set-identifying-P conditionals (Declerck and Reed 2001; Dancygier 1998: non-predictive): the P-clause is interpreted generically, the Q-clause is atemporal (a state), and the if-clause can be replaced by a restrictive clause or by anatemporal or restrictive when / whenever clause. The P-clause is neutral. That is, actualization of the Pclause is not presupposed at the time of speaking (Declerck and Reed 2001). (2a) If debtors are unable to pay their own debts, debtors are insolvent. (constructed) (2b) We cannot speak of insolvent debtors if they are able to pay their own debts. (constructed) Gnomic(universally true) conditionals are also neutral: (3) You cannot marry a man if you are a man. You cannot marry your sister if you are a boy. (HEM-E~1\eer\432) (12) In terms of a possible-world typology of conditionals, neutral-P, closed-P,open-P, tentative-P and counterfactual-P conditionals are nonfactual or theoretical (i.e., the if-clause creates a supposed world). Another option are realisand factual-P conditional, in which the P-
Silvia Cacchiani
65
situation is part of the actual world (i.e., known to be true in the actual world), would be (4). What looks like a non-canonical pattern 2 conditional makes reference to past habits based on repetitive actualizations of the same situation.In case could be used here with an actualization-conditioning meaning (that is, the situation in the in caseclause triggers the situation in the Q-clause. (4) Manufacturers of category A items could apply for prices increases if they had incurred increases in allowable costs. (HEM-E~1\joeb\511) There is an important point to make about examples (1) to (4): the realis-potentialis-irrealis distinction does not correlate exactly with pattern 1-3 conditionals. Secondly, form- (i.e. tense-) based distinctions relating to (canonical) patterns 1 to 3 do not provide a useful basis for describing conditionals. To complicate the picture further, data from the HEM-E shows that, with very few exceptions (5) (a tentative-P conditional), canonical patterns are not a feature of economics RAs. Examples (6) and (7) illustrate non-canonical pattern 1 and 2 conditionals. Importantly, pattern 3 counterfactuals are essentially absent, which is in keeping with the epistemology of the discipline, and empirical analysis and conditional prediction in particular. (5) If the limiting pivot magnitude were different when the Official is effective than when she is ineffective, then for a large n the voters would infer that the event of being pivotal was overwhelming evidence in favour of the Official’s type that yields the higher pivot magnitude […]. (HEM-E~1\eer\) (6) If environmental standards are reduced, production in the pollution-intensive sector becomes more efficient. (HEM-E~ejope\4) (7) […], and it is possible that we would prefer A over B if we took these correlations into account. (HEM-E~1\iref\93)
3.2. Interpreting conditionals In order to provide a list of the standard meanings and functions of conditionals in economics RAs, our approach has involved a qualitative investigation into the extended concordances of the if node in the HEM-E. Adapting Declerck and Reed’s (2001) fine-grained categorization of the functions served by P- and Q-clauses in discourse, conditionals are divided into case-specifying, where Q applies if P obtains, and rhetorical
66
Teaching Effective Academic Writing
conditionals, whose P-clause is not case-specifying. Based on the link between P-clause and Q-clause, the categories identified were as follows. Case-specifying-P conditionals, where P specifies a case or the cases in which Q obtains, comprise: (i) actualization conditionals and (ii) direct inferentials, which are implicative (with causal, resultative, licensing, preclusive or non-preclusive link between the P-clause and the Q-clause, also signaled by then), and (iii) purely case-specifying-P conditionals. In keeping with the epistemology of the discipline, actualization conditionals and direct inferentials- where the Q-clause asserts the conclusion drawn from the premise in P -, account for the vast majority of examples in the corpus. In (6) above, an actualization-conditionalreducing environmental standards is a (cause-effect or enabling) condition for more efficient production in the pollution intensive sector. Consider also (8), in which differentiation results into or licenses profitable technology transfer in a closed conditional. Actualization-conditioning also seems to be at play in (9), where if we solve […] and denote signal specific step in the analysis. In direct inferentials, the P-clause is premise-expressing, as in (10), with a reduced P-clause (if small enough), and the inferential premise-conclusion link particular to mathematical-statistical analysis signaled by then. Lastly, though less represented, purely-case-specifying-P conditionals, which are non-implicative (and, therefore, cannot take then) are also present (3). (8) If products are sufficiently differentiated, technology transfer is always profitable. (HEM-E~1\iref\) (9) If we solve Eq. 2.16 for (w)r and denote the solution by wh(r), then it represents the wage rate that industry h can bear under zeroprofit condition. (HEM-E~1\eer\432) (10) Clearly, *-0>2-* if 2+/(+)=1+/, i.e., if 22=2. However, if small enough then *-0