(PDF) Questioning Linguistics

Questioning Linguistics Questioning Linguistics Edited by Ahmar Mahboob and Naomi Knight Cambridge Scholars Publishing Questioning Linguistics, Edited by Ahmar Mahboob and Naomi Knight This book first published 2008 by 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 © 2008 by Ahmar Mahboob and Naomi Knight 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): 1-84718-667-X, ISBN (13): 9781847186676 TABLE OF CONTENTS Acknowlegements ..................................................................................... vii Preface ...................................................................................................... viii Chapter One................................................................................................. 1 Questioning Linguistics Ahmar Mahboob & Naomi Knight Part I: Issues and Directions Chapter Two .............................................................................................. 18 Language-free Linguistics and Linguistics-free Languages Alastair Pennycook Chapter Three ............................................................................................ 32 Innocence: Realisation, Instantiation and Individuation in a Botswanan Town J R Martin Chapter Four.............................................................................................. 77 Reconciling the Co-articulation of Meaning between Words and Pictures: Exploring Instantiation and Commitment in Image Nuclear News Stories Helen Caple Chapter Five .............................................................................................. 95 “What the Hell is Wrong With You?” A Corpus Perspective on Evaluation and Emotion in Contemporary American Pop Culture Monika Bednarek Chapter Six .............................................................................................. 127 Free Word Order in Artificial Languages Alan Libert vi Table Of Contents Chapter Seven.......................................................................................... 143 Syntactic Encoding of Topic and Focus in Korean Hyeran Lee Chapter Eight........................................................................................... 164 Syndromes of Meaning: Exploring Patterned Coupling in a NSW Youth Justice Conference Michele Zappavigna, Paul Dwyer, J R Martin Part II: Applications and Variation Chapter Nine............................................................................................ 188 Demythologising CLT: Wanted – A Reorientation for Teachers in the 21st Century Anne Burns Chapter Ten ............................................................................................. 207 Fine-Tuning Discourse in Thai EFL Academic and Electronic Bulletin Board Writing Montri Tangpijaikul Chapter Eleven ........................................................................................ 225 The Semantics of Graduation: Examining ESL Learners’ Use of Graduation over Time Caroline Lipovsky & Ahmar Mahboob Chapter Twelve ....................................................................................... 242 Analysis of Japanese Spoken by Elderly Taiwanese: Word Usage, Particle Usages, and Predicate Forms Masumi Kai Chapter Thirteen...................................................................................... 258 What's the Use of Linguistics? Michael Walsh About the contributors ............................................................................. 272 ACKNOWLEGEMENTS We, the Co-Convenors of the First International Free Linguistics Conference (FLC), 2007, are grateful to the School of Letters Arts and Media (SLAM) and the Department of Linguistics at the University of Sydney for providing key support that made FLC possible. We would also like to thank Bridge Bookshop, Cafe Ottimo, Coop Bookshop, Gleebooks, and Starbucks Coffee for supporting FLC. Finally, we would like to thank our presenters, participants, contributors, reviewers, and the conference committeee members who made FLC a success and this volume possible. The editors, authors, and publishers are also grateful to those who have given permission to reproduce the following extracts and images of copyright material: Extract from Tears for the Giraffe by Alexander McCall Smith is reproduced by permission of Polygon, an imprint of Birlinn Ltd. www.birlinn.co.uk Extract from Morality for Beautiful Girls by Alexander McCall Smith is reproduced by permission of Polygon, an imprint of Birlinn Ltd. www.birlinn.co.uk Press photos from the Sydney Morning Herald published on: 1) February 22, 2005, p.11, 2) February 23, 2005, p.8, 3) March 22, 2005, p.3, & 4) March 25, 2005, p. 14, are reproduce by permission of The Sydney Morning Herald and The Sun-Herald, Level 4, One Darling Island Road, Pyrmont, NSW 2009. PREFACE This book originated from research presentations given at the 1st International Free Linguistics Conference that was held on October 6-7, 2007, at the University of Sydney, Australia. The conference was the first of its kind, and focused on the notions of freedom: from all linguistic subfield divisions, from rigid presentation themes, and from any fees. Scholars travelled to Sydney from around the world, and came together with a range of presentation topics from areas throughout the linguistic discipline to present them to an open and welcoming crowd of participants. This resulted in a combination of high quality papers that engaged with some of the most relevant and interesting issues in linguistics today, and these could all be related to a common undercurrent that flowed out of the Free Linguistics principles. That is, many of the presentations seem to orient towards the notion of freedom, and also delivered an important viewpoint towards the discipline of linguistics itself once seen from outside of its borders. We invited these presenters to submit their articles for a compiled volume centred around this resulting theme: questioning linguistics. The papers included in this volume reflect the nature of the conference and raise questions about our current understandings about languages and linguistics. All papers submitted for publication to this volume went through a rigorous review process and selection was based on the outcome of this review process: 21 papers were initially submitted for publication and 12 of them are included in this volume. All of the authors in this volume question language and linguistics in unique ways, and their range of scholarly backgrounds and theoretical perspectives enhance the chapters to offer a free and engaging pursuit of this theme. Questioning Linguistics then offers something cohesive but at the same time divergent- as it brings together authors who have broken, and here attempt to further remove, the disciplinary and sub-disciplinary borders that can be perceived in the field, and provide arguments for an introspective analysis of what we do as linguists and how this allows us to be ‘free’. Naomi Knight & Ahmar Mahboob Sydney, May 2008 CHAPTER ONE QUESTIONING LINGUISTICS AHMAR MAHBOOB & NAOMI KNIGHT 1. Questioning Linguistics Human curiosity in languages dates back to the earliest records of civilization – e.g., Śākaṭāyana, working on Sanskrit around 8th century BCE argued that all nouns derive from verbs. In fact, linguists, even at that early time, presented opposing points of view and arguments about the nature of language abounded (see Matilal, 1990, for an in-depth discussion of early Indian linguists). This is not surprising because language, in many ways, is like the proverbial group of blind people trying to describe an elephant – each person describes the elephant based on the part of the anatomy that they touch: everyone has a piece, but no one has the full picture. Linguists, like this group of blind people, describe only aspects of language that they focus on and no comprehensive theory exists to tell us all. The 20th century contributed greatly to our understanding of language and how it works within the human mind as well as in relation to the societies around us. As the last century unfolded, different schools of linguists emerged and positioned themselves as the ones that were best suited to describe language. The two major groups that have emerged from the twentieth century are the formal and functional schools (although other traditions also exist). These two schools are in themselves quite heterogeneous and linguists within them take a variety of positions. In addition to the linguists who work to describe language, there are associated fields and academic disciplines that use and apply this knowledge for diverse purposes and engage with them in different professional contexts. This book brings these different voices together into a single volume and allows readers to examine how linguists of diverse traditions study and use this expert knowledge of language. By doing so, the volume Questioning Linguistics invites us to reconsider the nature and 2 Chapter One focus of the field of study and questions a number of current thoughts about language theory, application, and use. The twelve original papers in this volume were selected from 37 papers presented at the First International Free Linguistics Conference (FreeLinguistics), 2007, held at the University of Sydney. FreeLinguistics is an initiative of the staff and students in the Department of Linguistics at the University of Sydney to create a space where linguists of all traditions and views can come together to present and engage with other perspectives on language – and to do this without any conference registration fees. The goal of FreeLinguistics is to provide a venue where linguists with different foci can share their descriptions of the language- elephant and thus help to draw a more comprehensive picture of the animal. As such, FreeLinguistics and the papers presented at the conference question linguistics. The selection of papers from FreeLinguistics included in this volume, representing diverse theoretical positions in linguistics and informed by a variety of research approaches, raise new questions about the nature of language and linguistics and their role in a globalized world. As such they represent the flavour of FreeLinguistics and paint a broader picture of language – and show us that we still need more studies to be able to fully comprehend the nature of the phenomenon. The volume, divided into two sections, first examines the goals of linguistic theory and the role of linguistics in our understanding of human society. The second section questions the current trends and practices in the application of linguistics in areas such as language teaching, language variation, and language attitudes. The following overview provides a richer description of each section and the papers that are included in them. 2. Issues and directions As we begin questioning linguistics, one of our first goals is to outline the language-assumptions that we take as facts and then highlight alternative ways of understanding linguistics. This is the goal of the collection of papers in the first part of the volume: relevant issues surrounding languages and linguistics are tackled by the authors, who question how we variously define and engage with concepts in and of languagee. Each chapter sheds light on areas that have been taken for granted, relatively ignored, or perceived unidimensionally, and the authors provide new suggestions about the directions we might take as linguists and researchers in thinking about and analysing language and beyond. Questioning Linguistics 3 The opening chapter of Part 1, “Language-free linguistics and linguistics-free languages” by Alastair Pennycook engages readers and questions the fundamental definitions of linguistics and of language itself with respect to the historical background of language description: The argument that linguistics might be better off if it were to get rid of the notion of languages as separate entities draws in part on Roy Harris’ (1990) remark that “linguistics does not need to postulate the existence of languages as part of its theoretical apparatus” (p.45). On the one hand, then, this chapter explores the ways in which languages are inventions of the discipline that makes them. It asks how we might go about exploring language diversity without positing the existence of languages. It looks at the historical and contemporary interests behind the long construction of things called languages and asks in whose interests we continue to divide language into these named entities. The other side of the question–whether languages, or language studies, would be better off without linguistics– explores the ways in which the narrow purview of linguistics limits what we can say about language-related issues. Here Blommaert’s (2005) observation that “linguists have no monopoly over theories of language, and as soon as one accepts that, far more candidates for critical potential offer themselves than SFL” (p.35) is a useful starting point. The point here is not to draw attention to the particular limitations of systemic functional linguistics (SFL) but to ask why it is that a certain form of linguistics has come to play such a dominant role in an enterprise such as critical discourse analysis (CDA), and how a wider vision of the operation of language might enable a more critical engagement with the social life of texts. (Pennycook) In arguing that the definitions and divisions of languages have been made in the interests of dominant ideologies, and that linguistics as a discipline is also limited by the parameters set by linguists, Pennycook brings us to the point that is underlined by many authors in this volume: language is a phenomenon that all can study and question as it is the mode through which most make meanings in their everyday lives. It is then important to broaden our understandings and interpretations by going back to basics, as it were, and seeing “local language understandings” (Pennycook, this volume, p.21). These, Pennycook finds, are prevented by the abstraction and quantifying of ‘languages’, and overshadowed by the privileging of scientific linguistic knowledge in systems and rules. This notion of ‘language’ is a socio-cultural concept that in its naming created a “language-object”, and instigated a quantification of languages into a hierarchy designed for colonial purposes. In calling into question these two constructions of language and linguistics, Pennycook makes explicit the underlying features of our systematisation of the phenomenon and 4 Chapter One turns our focus inward to problematize the concepts that we count on to do so. Those parameters and boundaries that have grown out of the linguistic discipline, creating oppositions rather than complementarities, are shown to obscure the way that linguists pursue their endeavours towards language. Pennycook not only takes a bold step in the breaking of boundaries between linguistics fields, a fundamental notion behind the principles of a ‘free linguistics’, but attempts to forge a new consideration of language if by any other name. Within linguistic theory as well, especially functional approaches, research has extensively explored the parameters of choice as a matter of oppositions. However, as J.R. Martin explains in “Innocence: realisation, instantiation and individuation in a Botswanan town”, one must shift focus to those neglected theoretical concepts involving complementarities in order to gain insight into the social meanings that have as yet not been exploited: In his No. 1 Ladies' Detective Agency series, Alexander McCall Smith presents four accounts of Mma Ramotswe's adopted daughter Motholeli's life - one biographical and three auto-biographical. In this chapter I explore some of the similarities and differences among these accounts from the perspective of systemic functional semiotic theory, focusing in particular on the complementary roles of three hierarchies: realisation, instantiation and individuation. I propose that our understanding of individuation needs to be elaborated to focus more clearly on identity and affiliation in relation to the rhetorical deployment of appraisal resources. In systemic functional linguistics (SFL), Martin shifts from the traditional focus on the theoretical concepts of realisation and rank to the development of the clines of instantiation and individuation in order to exemplify the individual strategies of speakers and their employment of expectations and variations in meaning. While discourse participants may have differentiated repertoires, their complementary deployment of resources display features of ideological constraints and categorizations, while instantiating different forms of consciousness in their identification. Microlinguistic questions specific to the theory are made by Martin, who pursues the notion that all of the fundamental features of the theory have a role to play in interpreting a text, and he presents the possibility that with a shift in theoretical orientation may come changes to the existing theory. This also underlines the perspective that linguistic theories are shifting and dynamic concepts themselves, which should firstly be able to sustain the development of each of their parameters, as well as being adaptable to the imminent flux that this development may have on the system. Language theories such as SFL are not impenetrable and stable, but are constantly Questioning Linguistics 5 changing, and as Pennycook has argued, we should not allow ourselves to be limited by our own parameters. This can be likened to the nature of language itself, a complex and shifting phenomenon that can be interpreted, described, and even created (as Alan Libert, this volume) shows, but cannot be contained within a set of scientifically determined concepts. The concept of linguistics has also been put into question by recent studies in multimodality, as modes of meaning-making such as gesture, facial expression, and image as well as their relation to verbiage are interpreted in terms of their communicative work in social interactions. In “Reconciling the co-articulation of meaning between words and pictures: Exploring instantiation and commitment in image nuclear news stories”, Helen Caple looks at the combination of text with images in image-nuclear news stories, illustrating that the authors play with this particular relation in order to engage with a specific ideologically constructed readership: Through the close analysis of a particular type of multimodal news story, this chapter investigates how intertextual references in newspaper headlines and press photographs contribute to the creation of solidarity between a newspaper and its readers. Using the concepts of instantiation and commitment from the systemic functional linguistic approach, I shall analyse how the “twoness” of meaning between headlines and photographs in image nuclear news stories combine to create an evaluative stance towards the news that is often playful. Such play on words and images also relies on the obliging reader’s (Kitis & Milapides, 1996) ability to activate other discourses that form his/her background knowledge in order to peel back the layers of meaning in the text. It is my suggestion that through this deliberate manipulation of the discourse the newspaper is able to express cultural and social solidarity with its readers, as the newspaper is assuming that readers share its understanding of the intertextual references being made in these texts. This can be labelled a kind of insiderism (Chang, 2004), including some readers and excluding others. As such, play of this nature may offer the kind of intellectual challenge that keeps readers interested in the news and still buying the newspaper. Through the use of two complementary modes of meaning, Caple finds that news authors are also able to manipulate expressions from within the linguistic framework by re-literalizing common idiomatic and intertextual expressions through what she describes as a “twoness of meaning” (p.57). While indicating that linguistics may involve more than verbiage in its full meaningful articulation, Caple shows that authors may use strategies of play with their various resources of communication in a culture to construe bonds with readers. Bringing other modalities besides speech into the 6 Chapter One study of linguistics opens up the discipline to a development of a new theoretical toolbox for explaining the role and potential of these modalities, and in their relation to surrounding text, it is clear that a great deal of meaning can be extrapolated in relations of co-articulation. This has led Caple to pursue the highly implicit relations of humorous play and cultural bonding that must be attended to once alternative modes are part of the description, and the strategic use of language through humour puts pressure on the theoretical frameworks of linguistics (in Caple’s study, that of SFL specifically) to incorporate tools for explaining and situating this phenomenon more fully. In particular, the implicit ‘dual’ meanings of humour cannot easily be reconciled to concepts such as the Instantiation cline, involving discrete choices instantiated from the system of meaning oppositions, and the position of images has yet to be established in a linguistic theory traditionally centred upon grammar. We must question how authors are able to so simply and strategically create invoked meanings that appeal specifically to the values that are shared with particular readers in the culture when these meanings are so difficult to interpret methodically in the current state of the theory. Perhaps we need to begin from the problematic texts in order to develop theories robust enough to handle these texts, or at least to shift our focus, as Martin (this volume) has argued, in order to attempt to capture these areas through the neglected features that are available. The approach that a linguist takes must then be determined based on the complexity of the phenomenon, and their own analytical purposes. In the next chapter, Monika Bednarek, for example, argues for a ‘three- pronged approach’ including a large-scale quantitative analysis, a small- scale corpus analysis, and a qualitative case study in order to fully capture the linguistic meanings and make any generalizable conclusions: In this chapter I take a corpus linguistic perspective on the language of evaluation and emotion working with a 1.5 million word corpus of American TV dialogue (from the TV series Gilmore Girls), focusing in particular on the expression of evaluation and emotion in American pop culture. The data for this study consist of a 1.5 million word corpus of transcripts of the popular American TV series Gilmore Girls which are analysed with the help of Scott’s (1998) Wordsmith tools. As will be shown, phrases like Oh my God and (what) the hell are used as “implicit cues” (Culpeper, 2001, p.172) to characterisation in TV dialogue, and work as conventionalised realisations of emotionality. The investigation of evaluation and emotion is embedded in the discussion of a three-pronged approach to the analysis of dialogue. More specifically, I argue for the necessity of combining a large-scale quantitative approach with a small- Questioning Linguistics 7 scale corpus analysis to be complemented by qualitative case studies (in this paper with the help of Martin and White’s (2005) Appraisal theory). In studying the influence of pop culture dialogue on our everyday lives, and how our emotions are shaped by culture, Bednarek also makes a case for acceptable data sources in the linguistic tradition. If television programmes such as the Gilmore Girls are so accessible and widely observed by members of the culture, why should it not be a source for examination as it is likely to have an important influence on our linguistic systems? As Bednarek argues, “the very popularity of television and TV series entails a huge influence of this dialogue in our daily lives” (p.63). Perhaps privileging sources of data from the academic pool over those of the common, everyday spheres limit our local understandings of language in the same way as the privileging of scientific knowledge does in linguistics (Pennycook, this volume). Thus, when Bednarek asks, “Why TV dialogue?”, she also problematizes the elitism of academic study in linguistic tradition, much as the conversation analysts (e.g. Sacks et al. 1974, Tannen 2005) have done in modern theory. From the analysis of everyday language to the creation of artificial languages, Alan Libert transacts explicitly with the notion of freedom, arguing for variation in “Free Word Order in Artificial Languages”: Language creators in theory have complete control over the form of their languages. Although most artificial languages are intentionally based on one or more natural languages, this is not true of all of them. In this chapter I shall examine the extent to which word order is free in artificial languages. We find that, although some such languages are freer than e.g. English or French, few are as free as the natural languages with the most freedom of order, such as Warlpiri. For example, although in Esperanto subjects, objects, and verbs can go in any order relative to one another, adpositions can only precede their complements, i.e. they are all prepositions. Indeed, there are some artificial languages which appear to be more rigid in word order than English. However, just as some artificial languages are not as free in word order as their designers or others suggest, at least some of those artificial languages that are said to have quite a fixed order appear to have exceptions. Aside from describing the range of freedom of order in artificial languages, I shall also attempt to explain restrictions that exist in artificial languages. An important point is the fact that to a large extent even those artificial languages which were not consciously based on natural languages still follow language universals. By challenging the relative freedom of rules of word order in languages in relation to one another, Libert shows the rigidity that is apparent despite the claims of language creators. In doing so, he exhibits the tendencies 8 Chapter One even of artificial languages, especially a priori and mixed languages, to follow certain universals. Clarity is also an issue once languages are brought into being, as they depend on designers with differing purposes, and are faced with some of the same constraints as natural languages of which they are unavoidably influenced. Thus, while claims have been made towards artificial languages in terms of their freedom of word order, Libert’s chapter makes clear that languages, even when designed, are influenced by the surrounding systems of language from which they are developed, and by their designers. They then involve more restrictions, rigidity, and universality than may be proposed. Hyeran Lee also questions the positioning of elements in descriptions of language systems, as she reformulates the interpretations of topicalisation, focalization and scrambling through an analysis of Complementizer Phrases in Korean in her chapter, “Syntactic Encoding of Topic and Focus in Korean”: This chapter aims to examine topicalisation, focalization and scrambling in Korean to show how such interpretations are syntactically obtained. It is claimed that discourse information such as topic and focus is encoded in the syntactic positions at the left periphery in this language. Topic is based-generated while an abstract element is moved to the left periphery. Focus directly moves to the left periphery for the interpretation to be licensed. Scrambling is analysed as an operation with no semantic import and thus the scrambled element does not move across the phase boundary to the left periphery. The operation that has been called long-distance scrambling is reanalysed as a focalisation process. Operations developed within languages are queried by Lee as once these elements are applied to different languages, their definitions will necessarily change. This is the case in Korean, in which the syntactic positioning of such elements as Topic and Focus need to be interpreted according to how they operate specifically for this language, and she argues that scrambling must also be reanalysed as a different process than traditionally defined. Lee suggests a new orientation to the structure of the Complementizer Phrase according to its use in Korean, and through this example, shows that while Libert has underlined the existence of language universals, there is still much variation that must be accounted for in every language. The operations that are put into use by language users are indicative in Lee’s chapter of this potential. Once a language is put in to use, we must then also consider how the users may use and adapt that language in various interactions. This consideration is made by Michele Zappavigna, Paul Dwyer and J.R. Martin in “Syndromes of meaning: Exploring patterned coupling in a Questioning Linguistics 9 NSW Youth Justice Conference”, as they theorize the delicate patterns of linguistic meanings used by mothers when in a situation that calls for strategic construal of emotion and responsibilities: This chapter explores patterning of evaluative meaning in a NSW Youth Justice Conference using Systemic Functional Linguistics (SFL) and appraisal theory (Martin & White, 2005). These conferences give young offenders the opportunity to meet with the victim of their crime and determine a punishment aimed at restoring the harm that they have done. We refer to patterns of meaning in these conferences as syndromes. Syndromes are formed through the coupling (Martin, 2000) of different meanings along what SFL terms the “cline of instantiation”. The text analysed in this paper is a transcribed conference held due to acts of vandalism by two teenage boys. We focus on the talk of their mothers and what appears a syndrome of ambivalence about the extent to which the mothers construe themselves as responsible for their sons’ behaviour. As the authors describe the use of associated meanings in an SFL framework, the authors not only show how mothers construe a “syndrome of talk” about responsibility for their sons’ actions, but they question how this type of meaning relation and patterning can be handled by the linguistic theory and what types of tools may be more or less suited to this strategic use of language. Zappavigna et al. deal with the problems of how to determine systemic probabilities once patterns are detected, and find that the model must be adapted in order to account for the multiple interactions of features that are co-instantiated, in SFL terms, through such elements as syndromes. While often resisted as a domain of linguistic study, visualisation is a concern in this chapter also, as once meanings are intricately and inextricably tied up in text, visualisations can help the researcher to properly explore this new complexity. The question of the nature and representation of linguistic evidence is pursued in this chapter, and linguistic theory challenged and adapted by the complexity of patterns becoming evident, it pushes linguists to endeavour to capture these meanings more systematically than has been done before. 3. Applications and Variation Part II of this book includes chapters based around the application of linguistic theory to language, and possibilities for variation within and between languages and dialects. The research concerns two broadly generalisable areas: that of language teaching and learning, and that of variation in and across languages. In relation to language teaching, the approach that has been overwhelmingly adopted in multicultural contexts 10 Chapter One is called into question and shown to be inefficient despite its widespread use. The learning of the English language by Thai and Japanese speakers is exhibited through an examination of modes and ontogenesis, while variation of features in Taiwan-Japanese are distinguished, and each of these studies exemplify the learning and use of languages through linguistics. Finally, the applicability of linguistics is shown to be widespread through its role in relation to work with Australian Aboriginal languages in the sociocultural context of Australia. Anne Burns investigates the suitability of a method of teaching that has been taken on without question by many language teachers and offers a more modern conceptualisation in “Demythologising CLT: Wanted – a reorientation for teachers in the 21st century”: Communicative language teaching (CLT) has been the “default” approach to teaching English for nearly four decades. Despite criticisms that it is based on westernised, imperialist philosophies unsuited to the numerous intercultural and multicultural contexts in which it is applied, its influence continues to permeate policies and practices worldwide. Over the last decade it has been taken up relentlessly in syllabus documents in countries whose ministries of education seek language teaching “reform”. The progressivist extremes advocated in strong forms of CLT, and frequently in “reformed” policy documents, have not manifestly resulted in widespread enhancement of English language teaching and consequently there continue to be calls for more teacher accountability, student testing, and measurable outcomes. A reorientation to effective language teaching is proposed. The hidden ideologies and imperialist philosophies that background the inception of this teaching approach mirrors the arguments set forth by Pennycook (this volume) towards languages themselves, and are similarly problematized by Burns who underlines the importance of placing more consideration into the approaches we subscribe to. She argues that the vague, unspecific definitions and lack of practical teaching strategies offered by the CLT approach make it a weak tool for effective language teaching, while in application it affects the students’ learning based on their own preparation and abilities rather than providing a structured scaffolding by the teacher. Burns questions the effectiveness of an approach that is so dependent on a consumerist learner-oriented functionality, and suggests a method by which teachers are able to work from a better understanding of the needs of the student to offer a comprehensive lesson structure based on the local dynamics of the classroom. This reorientation towards a teaching philosophy and methodological framework based on the interests of the student puts forth Questioning Linguistics 11 an important notion of the study and teaching of languages: that one must always pay attention to the shifting dynamics not only of the language itself but of the language user. In order to be effective teachers and effective language researchers, we need to see language in action within the contexts in which we interact with it. Montri Tangpijaikul also points out that the mode of linguistic exchange should be considered in language teaching, as is presented in his study on “Fine-tuning discourse in Thai EFL academic and electronic bulleting board writing”: This chapter examines the use of online bulletin board writing as a means of exercising learners’ use of expressions conveying epistemic modality and intensity. These linguistic features are important for the appropriate use of language, but ones which are underused by Japanese learners (Altman, 1982), Thai learners (Bhandhufalck, 1983), Chinese learners (Gibbon & Markwick-Smith, 1992), and lower level Finnish learners (Karkkainen, 1992). This is probably because learners lack the opportunity to express themselves in free interactions in English, a common problem for EFL learners. This study hypothesizes that Thai learners will exercise modality and intensity more in on-line bulletin board than in academic writing. Two parallel learners’ corpora are used, which comprise online bulletin board writing (BB corpus) and offline academic writing (ACAD corpus), a total of 115,980 words. They are drawn from the written outputs of 39 Thai EFL learners at Kasetsart University, Thailand. Data are analyzed using a computer concordancing program “ConcGram Corcordancer”. The analytical framework includes modal auxiliaries (e.g. will, would, may, might), epistemic stance adverbs (e.g. maybe, probably, actually, of course), adjectives (e.g. certain, possible), copular verbs other than “be” (e.g. tend, seem), intensifiers (e.g. rather, somewhat, quite, totally, absolutely), and comment clauses and phrases (e.g. I think, In my opinion). The findings show that Thai EFL learners are able to fine-tune their expressions with subtlety using these lexical and grammatical devices more in online bulletin board writing than in academic writing. As in Burns’ chapter, the success of language learners depends on their own possibilities for engaging with the English community outside of the classroom context, and this is something that creates difficulties for learners in relation to the socially relevant interpersonal use of such aspects as ‘fine-tuning devices’. Tangpijaikul shows that the forums available to learners in the classroom may not be entirely sufficient, but argues that the mode of online bulletin boards offer a potential open opportunity for learners to utilize and improve upon their linguistic potential in English, and for teachers to provide an impetus by offering it 12 Chapter One as an available mode of discourse. The interpersonal possibilities for learners can be improved within the classroom despite the difficulties of outside contexts for practice. By comparing the data of informal fine- tuning devices as used by Thai English learners in this online forum with their academic writing, the author exhibits how this modern and developing social tool can be used for educational purposes, while at the same time improving the students’ abilities to engage interpersonally with the English public. It is then significant to attend to the interpersonal meanings that are expressed by language learners as their adoption and development of these skills create greater possibilities for inclusion in the culture, implicating greatly on inter and intra-cultural relations. This development is tracked by Caroline Lipovsky and Ahmar Mahboob in their chapter, “The semantics of graduation: Examining ESL learners’ use of graduation over time”, as they examine the use of interpersonal meanings in a systemic functional framework: This paper examines ESL (English as a second language) learners’ use of graduation over time. Essays were collected at two points in time from 19 young adult learners from Japan enrolled in an intensive (6 hrs/day) ESL programme in the United States. These essays were coded using the Appraisal Framework, specifically the system of Graduation (Martin & White 2005). Analysis of these data from beginner learners shows that students initially relied on isolated lexemes and repetition, rather than semantic infusion, for intensifying and quantifying. In the later set of essays, learners started incorporating infusion. They also used more comparatives and enhancement, and quantified through listing. Language skills thus depend not only on the learning of grammar structures and vocabulary, but are importantly affected by the understanding and development of interpersonal evaluative meanings and their intensifications. While Tangpijaikul focuses upon modality and intensity in terms of the structures evident in students’ texts, Lipovsky and Mahboob consider the functionality of intensifying features of Japanese ESL learners’ discourse towards NESTs (Native English-Speaking Teachers) and NNESTs (Nonnative English-Speaking Teachers). By doing so, they exhibit the ontogenetic growth of the learners’ abilities to intensify their attitudinal meanings in the English language, indicating that the appropriation of a system of intensification in language is important for second language users to be able to more fully express their opinions and beliefs about the social world including the things and people that matter within it. Not only are language learners developing their knowledge of the language, but they are becoming more able to enact relationships and express degrees of feeling in different ways to do so more strategically. Questioning Linguistics 13 As an initial exploration into this area, the authors attend to the interests of the learners through their own individual development as evidence of learning not only the language, but language situated in the social context. Not only the social, but the historical context influences the development and use of a language by its users, and should also influence the researcher’s interpretation as it has in Masumi Kai’s “Analysis of Japanese Spoken by Elderly Taiwanese: word usage, particle usages, and predicate forms”: Taiwan was a Japanese colony for 50 years, up to 1945. Japanese language was taught in school and it was used as the official language during this period. It has been over 60 years since Chang Kai-shek’s Nationalist Party took over Taiwan and Mandarin Chinese became the official language of the island. However, there are still many elderly Taiwanese people who speak Japanese. This chapter analyzes the Japanese of four such people. The data shows that the Japanese spoken in Taiwan has deviations and features which are not seen in modern standard Japanese. The frequencies of deviations in the data vary depending on the subject’s educational background. The subject who received a higher education spoke Japanese in almost the same manner as a modern Japanese speaker. Only a few deviations were noticeable. However, the subjects who studied Japanese only until middle school showed several types of deviations. This chapter discusses two deviations among them: word usage and particle usage, as well as features of the predicate form. The most frequently occurring deviations were word usage. As for particle usage, which is said to be difficult for non-native Japanese speakers to acquire, the subjects did not show many deviations. One of the remarkable features was the lack of sentence ending morphemes. This caused the conversation to be fragmented and not well connected. Kai shows how the historical changes in Taiwan, as well as the relative educational backgrounds of the Japanese speakers has had a substantial impact upon their language use. While some deviations are related to the speaker’s concurrent use of the Taiwanese language of Min-nan, Kai argues that the degree of deviation depends on the subject’s educational background, and the features that set the uses apart from Japanese are brought together into a dialect named Taiwan-Japanese. In this chapter, the role of language in a changing political and social context, as well as its variation in a national population is presented, causing us to question our detection of those languages and dialects that have been uplifted. The final chapter in this section/volume raises one of the most common questions that is asked of linguists: “What’s the use of linguistics?” In his response to this question, Michael Walsh breaks down the borders 14 Chapter One between theory and application, and essentially softens the borders not only between Part I and Part II of this volume, but between those within the linguistic sphere: This is a personal response to a question often posed by students: “What's the use of linguistics?” Obviously my reaction to such a question will be a reflection of my own background and interests and should not be taken as purporting to be the definitive answer to a question that can be addressed in many other ways. As my background is mainly concerned with the study of Australian Aboriginal1 languages my responses will mainly relate to uses for linguistics within that arena. I will confine myself to three issues: saving language; linguistic disadvantage before the law, and, language and the land. While the issue itself is one that has been taken on through various examples by the preceding authors, Walsh attends to the usefulness of linguistics with three important and highly applicable subject areas of concern to many linguists today within Australia and abroad. Essentially he asks how linguistics can be useful in relation to Australian Aboriginal communities. Linguistics is shown to be applied in a revitalization of those languages that are on the brink of extinction in an effort not only to preserve the language itself as the main concern of the linguist, but more imperatively to improve the conditions of the Aboriginal communities that use it as a means of identification. It is shown that once a language is lost, so too is the identity of the language users, and this has a detrimental effect on the survival of the community. Linguistics then provides a means for aiding in the link between language and identity, and Walsh underlines the significance of this connection. He also shows how tradition and modern systems can clash without proper translation of language and of culture, as in the legal system, Aboriginal witnesses are often misinterpreted due to differences in traditional sociolinguistic practices towards the management of knowledge and reliance on particular modes and techniques of language interaction. Anglo-Australian and Aboriginal practices can be usefully bridged by linguists with a background knowledge of both relevant languages and cultures so that legality and fairness are upheld. Customs and traditions must also be acknowledged and respected when language is considered in relation to the land, as in Aboriginal languages, land and language are interdependent. Walsh makes suggestions about what can be done by linguists to create a more relationally equal and culturally sensitive context for these interacting communities, providing a strong argument for not only the usefulness of linguistics, but for a careful deployment of its resources through many languages. Questioning Linguistics 15 4. Beyond this volume The papers in this volume come together to lead us into reconsidering our understaning of what languages are and ways in which to study them. As such they represent a range of approaches within the field, although not all. They all contribute to our ability to paint a picture of the language- elephant, but the picture is far from complete. It is our belief, and the purpose of this book and FreeLinguistics, that a deeper understanding of languages is only possible if we look beyond the disciplinary boundaries and engage with different traditions, understandings, and approaches to linguistics. We hope that this volume helps us along that way. Notes 1. Walsh notes: I deliberately avoid the usage ‘Australian Indigenous Languages’, which is inclusive of the languages of the Torres Strait Islands, because I have virtually no expertise in those languages and my remarks in this paper are entirely directed towards Australian Aboriginal languages. References Altman, R. (1982). Interlanguage Modality. Paper presented at the 57th Annual Meeting of the Linguistic Society of America, San Diego, California. Bhandhufalck, F. (1983). Contrastive Analysis of Modals in English and Thai. Pasaa 13, 50-61. Blommaert, J. (2005). Discourse. Cambridge: Cambridge University Press. Chang, C. (2004). English Idioms and Interpersonal Meaning. Guangzhou: Sun Yak-sen University Press. Culpeper, J. (2001). Language and Characterisation. People in Plays and Other Texts. Harlow etc: Longman. Gibbons, J. & Markwick-Smith, V. (1992). Exploring the Use of a Systemic Semantic Description. International Journal of Applied Linguistics 2(1): 36-49. Harris, R. (1990). On redefining linguistics. In H. Davis & T. Taylor (Eds.) Redefining linguistics (18-52). London: Routledge. Kitis, E. & Milapides M. (1996). Read it and believe it: how metaphor constructs ideology in news discourse. A case study. Journal of Pragmatics 28: 557-590. 16 Chapter One Matilal, B. K. (1990). The word and the world: India's contribution to the study of language. Oxford: Oxford University Press. Martin, J.R. & White, P.R.R. (2005). The Language of Evaluation: appraisal in English. London: Palgrave. Sacks, H., Schegloff, E. and Jefferson, G. (1974). A simplest systematics for the organization of turn-taking for vcovnersation. Language 50: 696-735. Scott, M. (1998). Wordsmith Tools. Oxford: OUP. Tannen, D. (2005) Conversational Style. Oxford: Oxford University Press. PART I: ISSUES AND DIRECTIONS CHAPTER TWO LANGUAGE-FREE LINGUISTICS AND LINGUISTICS-FREE LANGUAGES ALASTAIR PENNYCOOK 1. The languages that linguistics produced In order to understand some of the background to the development of theories of languages, we need to turn first to an historical understanding of the interwoven projects of colonialism and language study. Linguists, as Errington (2008) explains “can be regarded as a small, rather special group of colonial agents who adapted European letters to alien ways of talking and, by that means, devised necessary conduits for communication across lines of colonial power” (p.4). As a result, the description of languages was intimately linked to the wider colonial emphasis on human hierarchies, so that “the intellectual work of writing speech was never entirely distinct from the “ideological” work of devising images of people in zones of colonial contact. It means also that language difference figured in the creation of human hierarchies, such that colonial subjects could be recognized as human, yet deficiently so” (p.5). Language descriptions cannot be abstracted from the colonial imperatives to control, subdue, and order. The description of languages, therefore, has to be seen not so much as a scientific division of a language spectrum along natural lines but rather a colonial project in the defining and dividing of colonized people. As Irvine and Gal (2000) describe the process of “linguistic description” of Senegalese languages by 19th century European linguists, “The ways these languages were identified, delimited, and mapped, the ways their relationships were interpreted, and even the ways they were described in grammars and dictionaries were all heavily influenced by an ideology of racial and national essences” (p.47). Grierson’s massive linguistic Survey of India, completed in 1928, provides interesting examples of how the Language-free Linguistics and Linguistics-free Languages 19 process of sorting languages and dialects occurred. Dialects tended to be considered spoken forms, while languages were accorded their special status according to other criteria such as regional similarities, family trees, or literary forms. One of the problems with this, however, was that while people had terms for their dialects - or at least terms for other people’s dialects (their own being considered the way one speaks) - they did not have terms for these larger constructions, “languages”. As Grierson (1907) explained: Few natives at the present day are able to comprehend the idea connoted by the words of a language. Dialects they know and understand. They separate them and distinguish them with a meticulous, hair-splitting subtlety, which to us seems unnecessary and absurd, but their minds are not trained to grasp the conception so familiar to us, of a general term embracing a number of interconnected dialects (p.350). Grierson makes several important moves here. He positions himself as able to perceive the reality of languages while local knowledge is dismissed as on the one hand an irrelevantly hair-splitting obsession with difference and on the other an inability to grasp the broader concept of languages. Having thus opened up a position in favour of a European understanding of superordinate languages, he is then able to explain why “nearly all the language-names have had to be invented by Europeans. Some of them, such as Bengali, Assamese, and the like, are founded on words which have received English citizenship, and are not real Indian words at all, while others, like “Hindostani”, “Bihari”, and so forth, are based on already existing Indian names of countries and nationalities” (p.350). While it is interesting at one level to observe simply that the names for these new entities were invented, the point of greater significance is that these were not just new names for extant objects (languages pre-existed the naming), but rather the invention and naming of new objects. The naming performatively called the languages into being. This invention of Indian languages has to be seen in the context of the larger colonial archive of knowledge. The British, as Lelyveld (1993) points out, “developed from their study of Indian languages not only practical advantage but an ideology of languages as separate, autonomous objects in the world, things that could be classified, arranged, and deployed as media of exchange” (p.194). This whole project was of course a cornerstone of the Orientalist construction of the colonial subject. Orientalism, suggests Ludden (1993), “began with the acquisition of the languages needed to gain reliable information about India. Indian languages became a foundation for scientific knowledge of Indian 20 Chapter Two tradition built from data transmitted to Europeans by native experts” (p.261). At the heart of the problem here is the underlying ideology of countability and singularity, reinforced by assumptions of a singular essentialized language-object situated and physically located in concepts of space founded on a notion of territorialization. The idea of linguistic enumerability and singularity is based on the dual notions of both languages and speakers of those languages being amenable to counting. It has been widely attested that there is a massive disparity between the number of languages that linguists believe exist and the number of languages people report themselves as speaking. Ethnologue, the Christian language preservation society, for example, notes that if we take an “approach to listing and counting languages as though they were discrete, countable units”, there are around 7000 languages in the world; yet there are some 40,000 or so names for different languages in use, and “the definition of language one chooses depends on the purpose one has in identifying a language” (Ethnologue, 2005). Remarkably, however, some linguists are prepared to overlook these problems and assert with confidence that “Once political considerations are firmly discarded, it is generally not a difficult matter to decide whether one is dealing with one language or more than one in a given situation” (Dixon, 1997, p.7). Other linguists, while on the one hand noting that political considerations cannot and should not be discarded, are nevertheless content to deal in terms of enumerative strategies which on the one hand reduce significant sociolinguistic concerns to the level of arithmetic, and on the other overlook both the problematic history of the construction of such languages and the contemporary interests behind their enumeration: “Over 95% of the world’s spoken languages have fewer than one million native users, some 5000 have less than 100000 speakers and more than 3000 languages have fewer than 10000 speakers. A quarter of the world’s spoken languages have fewer than 1000 users, and at least some 500 languages had in 1999 under a hundred speakers” (Skutnabb- Kangas, 2003, p.32). Mühlhäusler (2000) describes this as a continuation of the tradition of segregational linguistics, which insists that “languages can be distinguished and named” (p.358). To abstract languages, to count them as discrete objects, and to count the speakers of such languages, is to reproduce a very particular linguistic ideology. Language-free Linguistics and Linguistics-free Languages 21 2. The Linguistics that languages produced From the perspective of linguistic anthropology, with a particular interest in the notion of language ideologies, or regimes of language (Kroskrity, 2000), the question becomes one of asking how it is that languages are understood locally. As Woolard (2004) notes, such work has shown that “linguistic ideologies are never just about language, but rather also concern such fundamental social notions as community, nation, and humanity itself” (p.58). For linguistic anthropologists, the problem was that the “surgical removal of language from context produced an amputated “language” that was the preferred object of the language sciences for most of the twentieth century” (Kroskrity, 2000, p.5). By studying language ideologies as contextual sets of belief about languages, or as Irvine (1989) puts it, “the cultural system of ideas about social and linguistic relationships, together with their loading of moral and political interests,” (p.255) this line of work has shown the significance of local knowledge about language. At the very least, this sheds light on Mühlhäusler’s (2000) point that the notion of a “language” “is a recent culture-specific notion associated with the rise of European nation states and the Enlightenment. The notion of “a language” makes little sense in most traditional societies” (p.358). Discussing language use in Papua New Guinea, Romaine (1994) asks how we come to terms with the problem that speakers may claim to speak a different language when linguistically it may appear identical. She goes on to point out that the “very concept of discrete languages is probably a European cultural artifact fostered by procedures such as literacy and standardization. Any attempt to count distinct languages will be an artifact of classificatory procedures rather than a reflection of communicative practices” (p.12). Branson and Miller (2000) argue that we “must not only revel in linguistic difference but cope with that difference analytically. Let us recognize the culturally specific nature of our own schemes and search for new modes of analysis that do not fit other languages into a mould but celebrate and build on their epistemological differences” (p.32). The point here, then, is that while pluralist (socio)linguistics and applied linguistics focus on linguistic differences, they fail to address the metadiscursive concern of how we understand linguistic difference, failing thereby to engage with the ways in which languages and differences have been constructed. We have become blind to the linguistic metalanguages that have arisen as a result of language descriptions, and as a result we are unable to see that local language understandings matter. 22 Chapter Two As Heryanto (2007) suggests in his discussion of the imposition of Bahasa Indonesia: It took European colonialism to introduce the idea of ‘language’ before the old word bahasa came to articulate this newly-acquired concept. The adoption of a pre-existing word in East Asia to articulate a new concept from modern Western Europe helped make the concept appear universal. Language was –as it is today– believed to be a universal property of human species, in all its variations, existing in a separate sphere from, but universally referring to, more or less one and the same objective world (p.43). This introduced concept, Heryanto suggests, did not accord with local understandings of language since “at least in the two most widely spoken and influential languages in Indonesia, Malay and Javanese, there was no word for “language”. More importantly, there was neither a way nor a need to express its idea until the latter part of the 19th century” (p.43). This newly introduced concept of language entered “a world with no language”, in the process replacing vernacular views of language and how it worked. Samarin (1996) makes a similar point when he suggests that Africa was “a continent without languages.” This is not of course to suggest that Africans or Indonesians did not use language, but rather that languages as they came to be invented were not part of the linguascape: Africans used language in a linguistic sense to communicate with each other, and we have learned that these are beautifully complex and awesomely elegant means of verbal expression, not the primitive jabberings that they were first taken to be. But they were not languages in the socio-cultural sense. There is little in our knowledge of Africa to suggest ethnolinguistic self consciousness. Thus we can say before literacy there were no languages. (Samarin, 1996, p.390) In speaking of “language free communities” or a “continent without languages” the point, to be sure, is not that these contexts involved any less language use, but rather that these language users did not speak “languages”. This construction of language, either as an autonomous object or a linguistic system, has been challenged from several directions that suggest that linguistics has profoundly misconstrued language through its myths about autonomy, systematicity and the rule-bound nature of language, privileging supposedly expert, scientific linguistic knowledge over everyday understandings of language. Toolan (2003) rejects as a “powerful and misleading myth, any assumption that a language is Language-free Linguistics and Linguistics-free Languages 23 essentially an autonomous system which humans can harness to meet their communicational needs” (p. 123). Harris (1990) asks whether “The concept of a language, as defined by orthodox modern linguistics, corresponds to any determinate or determinable object of analysis at all, whether social or individual, whether institutional or psychological. If there is no such object, it would be difficult to evade the conclusion that modern linguistics has been based upon a myth” (p.45). From this perspective, then, it becomes clear that the European projects of colonial linguistics produced not only languages that did not fit local language use but also a body of knowledge about language that could not adapt to the locality of language. In order to construct itself as a scientific discipline, linguistics had to make an extensive series of exclusions, relegating people, history, society, culture and politics to a role external to languages. Nakata (2007) argues: If the history of a language and its users is not factored into the theory as a primary standpoint…then any knowledge generated about that language is flawed. This is not to reject entirely what linguists have done, or are currently doing. It is to make the point that the grammarians’ concentration on formal aspects of a language fundamentally separates the language from the people; it falsely separates the act of speaking from what is being spoken. Studies of this kind are content to describe and conclude with grammatical summations as if languages were floating in a vacuum, “ready-made” within a system of phonetic, grammatical and lexical forms and divorced from the social context in which the speech is being uttered (p.37). On the one hand, then, we have “the historical complicities between linguistics and colonialism (both “internal” and “external”) which still pervade its “neutral” systems of classification and nomenclature” and on the other hand, the problem of “the conceptual framework of linguistics as a science which still remains in place even with the subdisciplines of sociolinguistics and applied linguistics” (Parakrama, 1995, p.3). 3. Rethinking language and language education Why does this matter? It matters a great deal since the ways in which languages are described, legislated for and against, policed, and taught have major effects on many people. This is by no means a matter only for linguists to care about since those of us who work in areas such as applied or socio-linguistics often still employ precisely those terms and concepts that derive form this crooked history. When Heryanto (2007) speaks of 24 Chapter Two “language-free communities”, when Branson and Miller (2007) show how the move to constitute sign languages as “real languages” was also an act of epistemic violence, when, from an integrational linguistic perspective, Harris (1990) tells us that linguistics does not need to posit the existence of languages as separate and autonomous objects, and when linguistic anthropology draws our attention to the imperative of understanding local ideologies of language, we have clearly embarked on a different trajectory from much of applied and unapplied linguistics, with their belief in the existence and describability of discrete languages, their positing of languages as systems that exist outside and beyond communicative acts, their location of language within the heads of people, and their use of disembodied texts to represent language use. As Sinfree Makoni and I have argued (Makoni & Pennycook, 2007), current approaches to diversity, multilingualism and so forth, all too often start with the enumerative strategy of counting languages and romanticizing a plurality based on these putative language counts. While opening up questions of diversity with one hand, at the same time such strategies are also reproducing the tropes of colonial invention, overlooking the contested history of language inventions, and ignoring the collateral damage that their embedded notions of language may be perpetrating. By rendering diversity a quantitative question of language enumeration, such approaches continue to employ the census strategies of colonialism while missing the qualitative question of where diversity lies. In our view there is a disconcerting similarity between monolingualism and additive bilingualism in so far as both are founded on notions of language as “objects”. By talking of monolingualism, we are referring to a single entity, while in additive bilingualism and multilingualism the number of “language-things” has increased. Yet the underlying concept remains unchanged because additive bilingualism and multilingualism are at best a pluralisation of monolingualism (Makoni 2003). In the context of South African language policy Makoni (1998) argues that “emerging discourses about multilingualism derive their strength through a deliberate refusal to recollect that in the past multilingualism has always been used to facilitate the exploitation of Africans” (p.244-5).. Instead of the often static notions of language implied by concepts of multilingualism, we need to start to move towards concepts such as Jacquemet’s (2005) “transidiomatic practices”: “the communicative practices of transnational groups that interact using different languages and communicative codes simultaneously present in a range of communicative channels, both local and distant.” Transidiomatic practices, Jacquemet explains, “are the results of the co-presence of multilingual talk Language-free Linguistics and Linguistics-free Languages 25 (exercised by de/reterritorialized speakers) and electronic media, in contexts heavily structured by social indexicalities and semiotic codes. Anyone present in transnational environments, whose talk is mediated by deterritorialized technologies, and who interacts with both present and distant people, will find herself producing transidiomatic practices” (p.265). Such practices, however, are not only the product of contemporary linguistic contexts mediated by deterritorialized technologies, as they are the common ways in which languages have been and still are used throughout the world. These questions therefore go much further than challenging narrow linguistic and applied linguistic orthodoxies. The old issues of description versus prescription, linguistics applied versus applied linguistics simply fade from view as irrelevant. For some this might still imply little more than a turn towards sociolinguistics or pragmatics. Yet many of the assumptions of more socially oriented approaches to language study also need to come under critical scrutiny. The givens of sociolinguistics, such as bilingualism and multilingualism, notions such as language rights, or the idea of language pragmatics, are also questionable from this perspective since they are in a sense the by-products of the invented languages and metadiscursive regimes that linguistics has produced: If languages hadn’t been invented as isolated, enumerable objects separated from their environment in the first place, we wouldn’t need these add-on frameworks, and thus to talk of sociolinguistics or pragmatics is to uphold metalinguistic inventions. For one thing, this perspective suggests, following Halliday (2002), that we need to take semiodiversity as seriously as glossodiversity, the possibility of a multiplicity of meanings within a language as seriously as a multiplicity of languages (Pennycook, 2004). This argument is important for several reasons: It urges us to question the epistemologies or linguistic ideologies on which support for diversity may be based. Thus a rights- based approach to support for linguistic diversity and opposition to the English-Only movement in the US, as Sonntag (2003) points out, “has not fundamentally altered the American projection of its vision of global English… because a rights-based approach to promoting linguistic diversity reinforces the dominant liberal democratic project rather than dismantling it” (p.25). If oppositional strategies are conducted from within the same framework as that which they oppose, they run the danger of reproducing those same positions. A focus on semiodiversity rather than glossodiversity, then, can help us get beyond a view of diversity based only on language counts, and instead can engage with semiotic diversity. 26 Chapter Two This view has many implications for applied linguistic domains such as language testing (for a critical exploration, see Shohamy, 2001; 2006): Why is it, we might ask, that a language test such as the TOEFL (Test of English as a foreign language) remains so desperately monolingual? At first glance, this question may seem bizarre: It is a test of English, after all. Yet the linguistics of communicative activity developed by Thorne and Lantolf (2007), which opens up ways for us to see how languages may be mediational tools to develop each other, as well as the broader questioning of language inventions discussed above, suggests that a multilingual TOEFL may be a far more appropriate test (to the extent that testing can be appropriate) than a monolingual one. The point here is not of course that TOEFL should be offered in separate but discrete languages (Test of French, German, Japanese, Tsonga or Tagalog as foreign languages) but rather that to test language users in one narrow element of their linguistic repertoire while admitting of no leakage across the tight linguistic boundaries echoes a history of strange linguistic inventions. When we talk of “washback” in testing, it is more common to think of this in terms of the curricular effects of evaluation, but more broadly we might consider washback in terms of the “collateral damage” for language users, policy makers, citizens and educators of the strange notion that languages exist in separation from the world and each other and can be tested in isolation. Language education suffers similarly from such peculiar linguistic inventions. For a start, the enumerative strategies based on the notions of second language acquisition, or English as a second language become highly questionable. From the point of view outlined here, there is no good reason to separate and count languages in this way. The question to ask is what would language education look like if we no longer posited the existence of separate languages. As Busch and Schick (2007) show, it is quite possible to develop educational materials that challenge these language separations. While bilingual education has often successfully challenged its monolingual nemesis, it has also frequently operated with a vision of bilingualism that is little more than a pluralisation of monolingualisms, that is to say, it takes as a given the epistemology of the “bi”, the separability and duality of two languages. “What would language education look like,” however, asks García (2007), “if we no longer posited the existence of separate languages? How would we teach bilingually in ways that reflect people’s use of language and not simply people as language users?”(p.xiii). As she goes on to argue, “Translation of instructional material, offering the tests in the child’s language, bilingual teachers, bilingual pedagogy is not enough, for it is based on an invention and it rarely reflects the ways in which children communicate” Language-free Linguistics and Linguistics-free Languages 27 (p.xiv). Once again an answer might lie in starting to understand language and language education in terms of majority world local knowledge, in starting to relocate language learning from an additional to a transidiomatic practice. Further questions need to be addressed to other domains of linguistics and applied linguistics. What does translation start to look like if we start to rethink languages? The position I have been developing suggests that this boundary we set up between languages, making translation an issue when we speak “different languages” but not when we speak the “same language” is yet again a distinction that is hard to maintain. This does not dissolve translation into a meaningless activity; rather it suggests that all communication involves translation. The twin effects of metadiscursive regimes that divided languages into separable entities and pedagogical dictates that eschewed translation have had sadly detrimental effects on language education. As Kramsch (2006) suggests, language competence should be measured not as the capacity to perform in one language in a specific domain, but rather as “the ability to translate, transpose and critically reflect on social, cultural and historical meanings conveyed by the grammar and lexicon” (p.103). The role of the language teacher from this perspective, therefore, is “to diversify meanings, point to the meanings not chosen, and bring to light other possible meanings that have been forgotten by history or covered up by politics.” If English language teaching can escape its narrow vision of itself as a monolingual enterprise, it might finally be able to take up a more dynamic role in the world as a form of translingual activism (Pennycook, 2008). Language policy, meanwhile, becomes a very different project from its current orientation towards choosing between languages to be used in particular domains, or debating whether one language threatens another. If language policy could focus on translingual language practices rather than language entities, far more progress might be made in domains such as language education. An understanding of English as a global language not so much in terms of an entity that has spread but in terms of local language practices offers important insights into the ways in which English is locally mobilized (Pennycook, 2007a; 2007b). Areas such as critical discourse analysis, as Blommaert (2005) suggests, would be less reliant on particular versions of linguistics, and instead would have to incorporate ethnographic perspectives in order to account for texts in context. Applied linguistics more generally needs to address the question of what it might look like if we took seriously the implications of no longer positing the existence of separate languages, of acknowledging that if a science of language is an impossibility, so too is an applied science of language. But 28 Chapter Two as a domain of work more readily able to lead the way towards understanding the transidiomatic practices of speakers, applied linguistics may be able to help linguistics get over its unfortunate longterm obsession with the impossible study of languages. So, in trying to envision a language-free linguistics and linguistics-free languages, let us be aware of the (post)colonial legacies of what we do; let us consider more seriously the collateral damage and epistemic violence wrought by language descriptions. Let us not get trapped in linguistics- internal debates: Most schools of linguistics have far more in common than they admit. Let us acknowledge that language cannot be dealt with separately from speakers, histories, cultures, contexts, ideologies. And let us draw on linguistic anthropology, language ecology, emergent grammar, cultural studies, radical postcolonial creolistics, poststructuralism, whatever sheds light on language. Language questions are too important to be left to linguistics, and as applied linguists, we need to encourage movement towards a new era of language studies that can start to break with its colonial past. References Blommaert, J. (2005). Discourse. Cambridge: Cambridge University Press. Branson, J. & Miller D. (2000). Maintaining, developing and sharing the knowledge and potential embedded in all our languages and cultures: On linguists as agents of epistemic violence. In R. Phillipson (Ed.) Rights to language: Equity, power and education (28-32). Mahwah, NJ: Lawrence Erlbaum. Branson, J. & Miller D. (2007). Beyond ‘language’: Linguistic imperialism, sign languages and linguistic anthropology. In S. Makoni & A. Pennycook (Eds.) Disinventing and reconstituting languages (116-134). Clevedon: Multilingual Matters. Busch, B. & Schick J. (2007). Educational materials reflecting heteroglossia: Disinventing ethnolinguistic differences in Bosnia- Herzegovina. In S. Makoni & A. Pennycook (Eds.) Disinventing and reconstituting languages (216-232). Clevedon: Multilingual Matters. Dixon, R. (1997). The rise and fall of languages. Cambridge: Cambridge University Press. Errington, J. (2008). Linguistics in a colonial world: A story of language, meaning and power. Oxford: Blackwell. Ethnologue (2005). http://www.ethnologue.com/ethno_docs/introduction.asp#language_id Language-free Linguistics and Linguistics-free Languages 29 Accessed 19/11/05. García, O. (2007). Intervening discourses, representations and conceptualizations of language. In S. Makoni & A. Pennycook (eds.) Disinventing and reconstituting languages (xi-xv). Clevedon: Multilingual Matters. Grierson, G. (1907). Languages. In W. W. Hunter (Ed.) The Imperial Gazetteer of India, vol. 1: The Indian Empire—Descriptive (349-401). New edition. Oxford: Clarendon Press. Halliday, M.A.K. (2002). Applied linguistics as an evolving theme. Plenary address to the International Association of Applied Linguistics. Singapore, December, 2002. Harris, R. (1990). On redefining linguistics. In H. Davis & T. Taylor (Eds.) Redefining linguistics (18-52). London: Routledge. Heryanto, A. (2007). Then there were languages: Bahasa Indonesia was one among many. In S. Makoni & A. Pennycook (Eds.) Disinventing and reconstituting languages (42-61). Clevedon: Multilingual Matters. Irvine, J. (1989). When talk isn’t cheap: Language and political economy. American Ethnologist 16: 248-267. Irvine, J. & Gal S. (2000). Language Ideology and linguistic differentiation. In P.V. Kroskrity (Ed.) Regimes of language: Ideologies, Politics and identities (35-85). Santa Fe, NM: School of American Research Press. Jacquemet, M. (2005). Transidiomatic practices: Language and power in the age of globalisation. Language and Communication 25, 257-277. Kramsch, C. (2006). The traffic in meaning. Asia Pacific Journal of Education 26(1), 99-104. Kroskrity, C. (2000). Regimenting languages: Language ideological perspectives. In P.V. Kroskrity (Ed.) Regimes of language: Ideologies, Politics and identities (1-34). Santa Fe, NM: School of American Research Press. Lelyveld, D. (1993). The fate of Hindustani: Colonial knowledge and the project of a national language. In C. A. Breckenridge & P. van der Veer (Eds.) Orientalism and the Postcolonial Predicament: Perspectives on South Asia (189-214). Philadelphia: University of Pennsylvania Press. Ludden, D. (1993). Orientalist empiricism: Transformations of colonial knowledge. In C. A. Breckenridge & P. van der Veer (Eds.) Orientalism and the Postcolonial Predicament: Perspectives on South Asia (250-278). Philadelphia: University of Pennsylvania Press. Makoni, S. (1998). African languages as European scripts: the shaping of communal memory. In S. Nuttall & C. Coetzee (Eds.) Negotiating the 30 Chapter Two past: The making of memory in South Africa (242-248). Oxford: Oxford University Press. —. (2003). From misinvention to disinvention of language: Multilingualism and the South African Constitution. In Makoni et al. (Eds.) Black Linguistics: social and linguistic problems of languages in Africa and the Americas (132-151). London: Routledge. Makoni S. & Pennycook A. (2007). Disinventing and reconstituting languages. In S. Makoni & A. Pennycook (Eds.) Disinventing and reconstituting languages (1-41). Clevedon: Multilingual Matters. Mühlhäusler, P. (2000). Language planning and language ecology. Current Issues in Language Planning 1(3), 306-367. Nakata, M. (2007). Disciplining the Savages: Savaging the Disciplines. Canberra: Aboriginal Studies Press. Parakrama, A. (1995). De-hegemonizing language standards: Learning from (post)colonial Englishes about ‘English’. Basingstoke: MacMillan. Pennycook, A. (2004). Language policy and the ecological turn. Language Policy 3, 213-239. —. (2007a). Global Englishes and Transcultural Flows. London: Routledge —. (2007b). The myth of English as an international language. In S. Makoni & A. Pennycook (Eds.) Disinventing and reconstituting languages (90-115). Clevedon: Multilingual Matters. —. (2008). English as a language always in translation. European Journal of English Studies. Romaine, S. (1994). Language in Society: An Introduction to Sociolinguistics. Oxford: Oxford University Press Samarin, W. 1996. Review of Adegbija Efurosibina, Language Attitudes in Sub-Saharan Africa: A Sociolinguistic Overview. Anthropological Linguistics 38(2), 389-395. Shohamy, E. (2001). The power of tests: A critical perspective on the uses of language tests. London: Longman —. (2006). Language policy: Hidden agendas and new approaches. London: Routledge. Skutnabb-Kangas, T. (2003). Linguistic diversity and biodiversity: The threat from killer languages. In C. Mair (Ed.) The politics of English as a world language: New horizons in postcolonial cultural studies (31-52). Amsterdam: Rodopi. Sonntag, S. (2003). The local politics of global English: Case studies in linguistic globalization. Lanham, Maryland: Lexington Books. Language-free Linguistics and Linguistics-free Languages 31 Thorne, S. & J. Lantolf (2007). A linguistics of communicative activity. In S. Makoni & A. Pennycook (Eds.) Disinventing and reconstituting languages (170-195). Clevedon: Multilingual Matters. Toolan, M. (2003). An integrational linguistic view of coming into language. In J. Leather & J. van Dam (Eds.) Ecology of Language Acquisition (123-139). Dordrecht: Kluwer Academic Publishers. Woolard, K. (2004). Is the past a foreign country?: Time, language origins, and the nation in early modern Spain. Journal of Linguistic Anthropology 14(1), 57-80. CHAPTER THREE INNOCENCE: REALISATION, INSTANTIATION AND INDIVIDUATION IN A BOTSWANAN TOWN J R MARTIN 1. Hierarchies and complementarities Throughout its history the paradigmatic focus of systemic functional linguistics (hereafter SFL) has profoundly shaped its theoretical architecture. Over six decades most research has focused on the development of two hierarchies (realisation and rank) and two complementarities (metafunction and axis). Realisation organises language systems1 hierarchically into strata of abstraction (e.g. phonology, lexicogrammar, discourse semantics); and rank organises systems within strata hierarchically into orders of constitutency (e.g. phoneme, syllable, foot, tone group in English phonology). Turning to complementarity, metafunction organises systems on a given stratum at a given rank according to the kinds of meaning involved (ideational, interpersonal and textual) and the kinds of structure used to realise them (particulate, prosodic, periodic); and axis relates paradigmatic relations to syntagmatic ones by way of coordinating ranks and strata with one another. This results in effect in a highly saturated realisation hierarchy of the kind outlined by Matthiessen in many publications (e.g. Figure 1-13 in Halliday & Matthiessen, 2004, p.32). Less well developed in SFL is our understanding of two complementary hierarchies, instantiation and individuation. Instantiation explores the relationship between the reservoir of systems populating the realisation hierarchy and their actualisation in text. Individuation explores the relationship between this reservoir and the repertoire deployed by individuals. To date, work on register and genre informs our understanding of instantiation (e.g. Ghadessy, 1993, 1999); and work on coding orientation informs our understanding of individuation (e.g. Hasan, Innocence: Realisation, Instantiation and Individuation 33 1996, 2005). But in general terms this research is far less developed than that associated with realisation (especially with respect to the ongoing concentration on metafunction, rank and axis in lexicogrammar (e.g. Caffarel, Martin & Matthiessen, 2004; Halliday & Matthiessen, 2004). In this chapter I will suggest ways in which individuation research can be productively pursued; for complementary work on instantiation, focussing on the same data, see Martin (in press). 2. Realisation and instantiation SFL's best researched hierarchy is outlined in Figure 1, exemplified for written discourse (after Martin & Rose, 2007); in a model of this kind genre is a pattern of register patterns, register is a pattern of discourse semantic patterns, discourse semantics a pattern of lexicogrammatical patterns, which are in turn a pattern of graphological patterns - a relationship referred to as metaredundancy (Lemke, 1984). The co- tangential circles represent levels of abstraction from graphological or phonological substance, and if we follow Firth (1957), then all strata contribute a strand of meaning to discourse. Space precludes consideration of the theoretical and descriptive saturation of this hierarchy with work on rank, metafunction and axis. Landmarks of saturation include Halliday & Greaves, 2008 (phonology); Halliday & Matthiessen, 2004 and Caffarel et al., 2004 (lexicogrammar); Martin, 1992 and Martin & Rose, 2003 (discourse semantics); Ghadessy, 1993, 1999 (register) and Martin & Rose, 2007 (genre). Whereas realisation is a scale of abstraction, involving the recoding of one level of meaning as another, instantiation is a scale of generalisation, aggregating the meaning potential of a culture across instances of use. Halliday2 (e.g. Halliday & Matthiessen, 1999) uses the metaphor of climate and weather to characterise this hierarchy - since climate is a generalisation of trends in the weather we experience day by day. Analogising along these lines, instances of language use are related to the system as a whole in the same way that daily temperature readings are related to average temperatures for that time of year. Critically, every reading changes the average, albeit to a small degree. As lay persons, we notice changes from one day to the next, but not generally overall trends - unless momentous evolutions such as global warming are brought to our attention. An outline of the instantiation hierarchy, conceived in these terms, is presented as Figure 2, moving from system to its generic and registerial sub-potentials, through to texts and their generalisation as text types3, and culminating with readings of the meaning potential afforded by 34 Chapter Three individual texts. The basic idea of this scale is that the meaning potential of the culture, as consolidated along the hierarchy of realisation, is progressively narrowed until we arrive at the reading of a particular text. genre context field mode tenor discourse semantics lexicogrammar language graphology Figure 1: Realisation hierarchy-levels of the systemic abstraction Innocence: Realisation, Instantiation and Individuation 35 Figure 2: Instantiation hierarchy-relating system to instance 3. Realisation, instantiation and individuation Alongside realisation and instantiation we have what Matthiessen (2003 personal communication) has referred to as individuation. In Bernstein’s terms, this has to do with the relationship between the reservoir of meanings in a culture and the repertoire a given individual can mobilise. Bernstein (1996/2000, p.157f) discusses this relationship as part of his exploration of horizontal discourse (cf. Christie & Martin, 2007), which he defines as entailing "a set of strategies which are local, segmentally organised, context specific and independent, for maximising encounters with persons and habitats." I shall use the term repertoire to refer to the set of strategies and their analogic potential possessed by any one individual and the term reservoir to refer to the total of sets and its potential of the community as a whole. Thus the repertoire of each member of the community will have both a common nucleus but there will be differences between the repertoires. There will be differences between the repertoires because of the differences between members arising out of differences in members context and activities and their associated issues. ... From the idealisation constructed it is possible to see the inter-relations between Horizontal discourse and the structuring of social relations. The structuring of the social relationships generate the forms of discourse but the discourse in turn is structuring a form of consciousness, its contextual mode of orientation and realisation, and motivates forms of social solidarity. Horizontal discourse in its acquisition becomes the major cultural relay. (Bernstein, 1996/2000, p.157) The sets of strategies and their analogic potential referred to by Bernstein involve both recognition and realisation rules (Bernstein 1996/2000, 36 Chapter Three p.104f). Recognition rules allow speakers to identify the specificity or similarity of contexts, and thus to what is expected or legitimate in that context; realisation rules enable speakers to produce culturally specific texts and practices. As Bernstein comments "...one may be able to recognise that one is in a sociology class but not able to produce the texts and context-specific practices. In order to produce the legitimate text it is necessary to acquire the realisation rule" (p.105). To date in SFL, these ideas have mainly been explored by Hasan and her colleagues, focussing the coding orientations of mothers and pre-school children in relation to gender and social class (e.g. Hasan, 1996, 2005). This paper is oriented to development of this dialogue, keeping in mind that individuation complements and must be read in relation to the realisation and instantiation hierarchies noted above (as sketched in Figure 3; cf. Martin, 2006). 4. Four stories In order to reinforce the relation of individuation to realisation and instantiation this paper will focus on the same texts discussed in relation to realisation and instantiation in Martin (in press). The texts are from Alexander McCall Smith's The No. 1 Ladies detective Agency series; each is a version of the same story. Smith in fact tells the story four times - three times in second book in the series, Tears of the Giraffe, and the fourth time in the next book, Morality for Beautiful Girls. The full text of these stories appears in Appendix A. The relevant Botswanan characters from the series are as follows: Mma Ramotswe (lady detective) Mr J.L.B. Matekoni (her partner; mechanic) Motholeli (their adopted daughter; 13 years old) Puso (Motholeli’s younger brother) Rose (their maid) Mma Potokwane (orphan farm manager) Each of the four texts recounts the story of Motholeli and her brother Puso, from the moment their mother dies. Texts 2 and 4 are much fuller accounts, as told by Mma Potokwane (Tears of the Giraffe) and Motholeli (Morality for Beautiful Girls) respectively; texts 1 and 3 are sourced to Motholeli, are much briefer, and precede and follow Mma Potokwane's version in the same book. Generically speaking, text 1 is a very brief recount, and text 3 a short observation, while texts 2 and 4 are longer Innocence: Realisation, Instantiation and Individuation 37 recounts (Martin & Plum, 1997, Martin & Rose, 2007). Both texts 2 and 4 have typical recount structures. Text 2 takes readers to the point where Mma Potokwane is about to persuade Mr J L B Matekoni to adopt the children (without consulting with his fiancé Mma Ramotswe, the lady detective heroine of the series). The generic structure of text 2 is outlined below. instantiation system instance i n d i realisation v i d u a t i o n persona Figure 3: Three complementary hierarchies - individuation, realisation and instantiation [Abstract] [2] ...But while we are waiting, I'm going to tell you their story. You listen to this." [Orientation] ...There was a band of Masarwa who had come up from the Kalahari to hunt ostriches. ... [Record of events] - mother bitten - mother dies - mother buried 38 Chapter Three - brother rescued - recovery in hospital - nurse's charity - life with the nurse's family - Motholeli disabled - moving to the orphan farm [Re-orientation] That is their story, Mr J.L.B Matekoni. That is how they came to be with us here. Text 4, by Motholeli, extends the Record of Events to the point where Mma Ramotswe commits in her own mind to mothering the children in her home. [Abstract] [3] "Why don't you start off again?" she said. "I would like to hear your story." [Orientation] My name is Motholeli and I am thirteen years old, almost fourteen. ... [Record of events] - living in the bush - keepsake from mother - brother rescued - living with the nurse's family - disabled - offered a place at the orphan farm - arriving at the orphan farm - living at the orphan farm - finding parents - leaving the orphan farm -moving to Mma Ramotswe's house [Re-orientation] That is the story of my life. The main difference between the recounts, age notwithstanding, is that Mma Potokwane scaffolds the story as a single series of events leading from the children's mother's death to her orphan farm. Mma Potokwane smiled. "You are a good man, Mr J.L.B. Matekoni," she said. "I will send word for them to come. They will have to be fetched Innocence: Realisation, Instantiation and Individuation 39 from the fields. But while we are waiting, I'm going to tell you their story. You listen to this." [sequence of events] That is their story, Mr J.L.B Matekoni. That is how they came to be with us here. Mr J.L.B. Matekoni said nothing. He looked at Mma Potokwane, who met his gaze. She had worked at the orphan farm for almost twenty years-she had been there when it had started-and was inured to tragedy-or so she thought. But this story, which she had just told, had affected her profoundly when she had first heard it from the nurse in Francistown. Now it was having that effect on Mr J.L.B. Matekoni as well; she could see that. Motholeli on the other hand adopts a more “historical” approach (Coffin, 2006), dividing her life autobiographically into three phases-life with her family in the bush, life on the orphan farm and life with Mr J L B Matekoni and Mma Ramotswe. "So," said Mma Ramotswe. "You are reading a story to your little brother. Is it a good one?" Motholeli looked round and smiled. "It is not a story, Mma," she said. "Or rather, it is not a proper story from a book. It is a story I have written at school, and I am reading it to him." Mma Ramotswe joined them, perching on the arm of the sofa. "Why don't you start off again?" she said. "I would like to hear your story." I am a girl who has had three lives. My first life was when I lived with my mother and aunts and uncles, up in the Makadikadi, near Nata. ... That was the end of my first life. My second life started on the day I arrived at the orphan farm. ... I did not think it would be Mr J.LB. Matekoni who took us, but I was very pleased when he said that we could go to live in his house. That is how my third life began. ... I have had a fortunate life and I thank Mma Ramotswe and Mr J.L.B. Matekoni from my heart. 40 Chapter Three That is the story of my life. I am an ordinary girl from Botswana, but it is very lucky to have three lives. Most people only have one life. This story is true. I have not made any of it up. AFTER THE girl had finished, they were all silent. ... Rose, who had been listening from the corridor, looked down at her shoes and thought: What a strange way of putting it: three lives. This is a relatively mature way of mapping history for a 13 year-old girl (appreciated as strange by the maid Rose), and illustrates one of the many ways that Smith construes Motholeli's character as capable and wise beyond her years. Each text in addition includes some staging which is not predicted by extant descriptions of canonical recount structure (cf Martin & Rose, 2007). Potowkane begins with a pre-script, warning Mr J L B Matekoni (and readers) not to judge the Basarwa harshly for their customs (deflating a potential negative reaction to them burying Puso with his dead mother). You must understand, said Mme Potokwane, that although it is easy for us to criticise the ways of the Basarwa, we should think carefully before we do that. When you look at the life they lead, out there in the Kalahari, with no cattle of their own and no houses to live in; when you think about that and wonder how long you and I and other Batswana would be able to live like that, then you realise that these bushmen are remarkable people. And Motholeli ends with a post-script, proposing plans to work with Mr J L B Matekoni in his garage and care for her adopting parents at home (entailing further positive judgement of her character). I would like to be a mechanic when I grow up. I shall help Mr J.L.B. Matekoni in his garage and at night I will mend Mma Ramotswe's clothes and cook her meals. Then, when they are very old they will be able to be proud of me and say that I have been a good daughter for them and a good citizen of Botswana. These stages make texts 2 and 4 generically unique; the instances, in other words, conform to realisational predictions up to a point, and in addition diverge. This highlights the need for an instantiation perspective alongside realisation, with realisation taking responsibility for consolidating recurrent patterns of meaning generalised as system, complemented by instantiation which shows just how far specific texts both confirm expectations and uniquify meaning (as explored in Martin, 2006). Innocence: Realisation, Instantiation and Individuation 41 It also highlights the need for further work on individuation, since we need to be able to explain why it is that Mma Potokwane and Motholeli tell the story in different ways. Their complementary repertoires afford complementary instantiations of what went on. It is to this question of identity and affiliation that we now turn. 5. Identity The basic question we are asking here is how Mma Potokwane and Motholeli are construed by Smith as different. How do their different forms of consciousness, in Bernstein's terms, manifest themselves in different discourses? More specifically, how do they recognise this life- story telling context as similar and different to other contexts? And what realisation rules do they deploy to construct their texts? Smith is foregrounding ethnicity and generation as crucial variables here. But how does he realise them? As far as genre is concerned, as we have seen, Smith gives us a mixed message. Motholeli organises her life into three phases in text 4, a form of organization more strongly associated with the written discourse of schooled history than the sequential story telling of everyday life (Coffin, 2006, Martin & Wodak, 2003, Martin & Rose, 2007). This construes her in effect as more “mature” and “westernised” than Mma Potokwane. By the same token, Motholeli appears to lack some of the key writing resources she needs to realise these phases, since her metalanguage for referring to them is an everyday one (three lives). So instead of one life with three phases, we have three lives, something Rose the maid appreciates as very strange (but which some readers might experience as poetically insightful). The effect of this is to construe Motholeli as a precocious Basarwa child, rather than an educated Setswana adult (keep in mind she is 13 years old). Turning to appraisal (Martin & White, 2005), we can note another feature of the complementary identities. On the one hand we have Mma Potokwane, who relays other's emotions rather than her own, and on the other, Motholeli, who says how she herself feels (see Table 1). In the Tables, P "Basarwa stands for Mms Potokwane sourcing feelings to the Basarwa; and negative feelings (e.g. sad) have a minus sign in front of them (- sad). 42 Chapter Three Table 1: Vicarious and personal affect construed by Mma Potokwane and Motholeli source force inscribed affect trigger/target Potokwane comfortable Botswana P "Basarwa happy staying in camp P "mother comfortable sleeping position P "truckdriver [infused] astonished Motholeli at roadside P "nurse [infused] - concerned for the children P "people pity children Motholeli very - sad mother's death Motholeli happy staying alive nobody liked Motholeli's clan Motholeli love Jesus Motholeli - ashamed wet dress Motholeli proud cupboard Motholeli - cry missing family Motholeli very happy orphan farm Motholeli very pleased adoption by J.L.B. Matekoni We can also compare Mma Potokwane relaying positive and negative judgements made by other people with Motholeli inscribing her own, generally positive and often intensified opinions (Table 2). For Motholeli, in other words, attitudes are feelings she herself experiences; and while she does experience the emotional ups and downs of life, her judgements of herself are modest and of others overwhelmingly positive. In this respect Smith construes a young girl undaunted by her loss of her family and her disability, looking forward positively to life. Mms Potokwane inscribes less attitude (in fact warning Mr J.L.B Matekoni not to criticise the ways of the Basarwa), and concerns herself not with her own feelings but those of others. So as far as inscribing attitude is concerned she leaves it up to Mr J.L.B. Matekoni (and readers) to make their own judgements about the children, probably since in this instance, her ideational construal of what went on speaks for itself. Innocence: Realisation, Instantiation and Individuation 43 Table 2: Judgements sourced to Mma Potokwane and Motholeli source force inscribed trigger/target judgement P "people - criticise Basarwa P "people carefully think P "people [infused] remarkable Basarwa P "people - suspicious Basarwa Motholeli very clever Motholeli's clan Motholeli kind nurse's family Motholeli (neg) - better Motholeli Motholeli very (neg) - good Motholeli walking Motholeli kind housemother Motholeli how lucky Motholeli Motholeli very well other girls walking Motholeli right Mma Potokwane Motholeli fortunate life Motholeli good daughter Motholeli good citizen Motholeli ordinary girl Motholeli very lucky three lives Motholeli true story This takes us to what is perhaps the greatest difference between texts 2 and 4, the question of dialogism, or engagement as it is formalised in appraisal theory (Martin & White, 2005). Engagement theory developed out of Fuller's (1998) research into the ways in which Stephen Jay Gould bridged between the voices of humanities and science in his popular science writing. Martin and White's network of discourse semantic engagement systems is presented in Figure 4 below; lexicogrammatically speaking, realisations include modality and projection, concession and negation, and certain comment adjuncts, among others. 44 Chapter Three deny no, not, failed to… disclaim counter but, even, amazingly… contract confirm of course, naturally, rhet Q… proclaim pronounce I contend, indeed… commit endorse experts convincingly showed… entertain perhaps, might, possible, I reckon… expand acknowledge she said, I heard… attribute distance they claimed to have shown… Figure 4: Engagement systems (Martin & White, 2005) Taking projection as our way in (the [expand: attribute] options in Figure 4), we can observe that in text 2 Mma Potokwane focuses more on what people thought (11 instances) than what they said (4 instances), including ideas attributed to people in general and people other than herself (you must understand, you...wonder, you realise, you might or might not know, they thought, she decided, they found, as you have seen). Innocence: Realisation, Instantiation and Individuation 45 [YOU MUST understand], [said Mme Potokwane], that although it is easy for us to criticise the ways of the Basarwa, we should think carefully before we do that. when you think about that and [wonder] how long you and I and other Batswana would be able to live like that, then [you realise] that these bushmen are remarkable people. But, [as you might or might not know], Mr J.L.B. Matekoni, when a Mosarwa woman dies and she's still feeding a baby, they bury the baby too. [I suppose] [they thought] she would go back to her people, which they usually do. She saw the girl sitting at the hospital gate and [she decided] that she had nowhere to go. I spoke to the nurse about her, and [I understood] that although she was still a child herself, she was a good mother to the boy. and [I think] that sometimes people took pity on them and gave them money and [they found] that the tuberculosis had damaged the bones very badly. Now they are with us, [as you have seen]. [YOU MUST understand], [said Mme Potokwane], that although it is easy for us to criticise the ways of the Basarwa, we should think carefully before we do that. The people up there are sometimes suspicious of Basarwa, as [they say] that they steal their goats and will milk their cows at night if they are not watched closely. [According to the daughter], her mother died before the baby even woke up. [I said] we could take them, and I went up to Francistown to collect them just a few months ago. Now they are with us, [as you have seen]. Conversely, Motholeli focuses mainly on what other people say to her (16 instances) in comparison to what she herself thinks (7 instances): so I scraped it off his face and [saw] that he was still breathing. [I remember] picking him up and running through the bush until I found a road. [I do not remember] what happened there, Sometimes at night, I would wake up and [think] how lucky I was. But [I wondered] if they remembered me, because I was just a girl and I had run away from them. [I did not think] that this was possible [I did not think] it would be Mr J.LB. Matekoni who took us, but I was given to a woman [who said] that I could live in her yard. 46 Chapter Three But [the woman said] that she could not have a wheelchair girl living in her yard, as that would not look good and [people would say]: What are you doing having a girl in a wheelchair in your yard? That is very cruel. Then a man came by [who said] that he was looking for orphans to take to his orphan farm. There was a lady from the Government with him [who told me] that I was very lucky to get a place on such a fine orphan farm. But [I must always remember] to love Jesus, [this woman said]. [I replied] that I was ready to love Jesus and that I would make my little brother love him too. [One of the ladies there told the children] to go off and play, and not to stare at us, I worked hard, and [Mma Potokwane, who was the matron, said] that one day, if I was lucky, she would find somebody who would be new parents for us. but I was very pleased when [he said] that we could go to live in his house. [She said] that she always felt very sad when one of the orphans went, as it was like a member of the family was leaving. and [she told me] he was one of the best men in Botswana. I would be very happy in his house, [she said]. [She said] that she would be my new mother, and she brought us to her house, which is better for children than Mr J.L.B. Matekoni's house. Then, when they are very old they will be able to be proud of me and [say] that I have been a good daughter for them and a good citizen of Botswana. This pattern of attribution is summarised in Table 3 below. From this we can observe that whereas Motholeli sticks to what she knows (what she herself thinks or recalls people saying to her), Mma Potokwane moves beyond her own directly observable experience to generalise beliefs (you must understand, you...wonder, you realise) or in infer ideas (you might or might not know, they thought, she decided, they found, as you have seen). Innocence: Realisation, Instantiation and Individuation 47 Table 3: Sources of mental and verbal projection in texts 2 and 4 mental projection verbal projection Text 2 Mma Potokwane 3 2 source Other source 8 1 Text 4 Motholeli source 7 1 Other source - 16 Alongside projection the other main resource for expanding the number of voices in a text is modality (the [expand: entertain] options in Figure 4 above), which relativises the voice of the speaker/writer in relation to ability, obligation, inclination, probability and usuality. Two of Mma Potokwane's mental projections, the first person present tense I suppose and I think can in fact be re-read here as subjectively explicit modalisations of probability (pace Halliday & Matthiessen, 2004). [I suppose] they thought she would go back to her people, which they usually do. and [I think] that sometimes people took pity on them and gave them money Both Mma Potokwane and Motholeli foreground modulations of ability (13 each), reflecting their story's concern with what people were and weren't able to do. In addition Mma Potokwane has 3 modulations of obligation (must, should, had to); Mothtoleli's text has 1 of these, but it is actually sourced, via projection, to the lady from the government who offers Motholeli and Puso a place on Mma Potokwane's orphan farm (But I must always remember to love Jesus, this woman said). Motholeli is not construed as being in a position to tell people what to do. Turning to modalisation, the texts both draw on usuality to generalise about people's behaviour - in Mma Potokwane's case other people's behaviour and in Motholeli's case her own: The people up there are sometimes suspicious of Basarwa, but were sleeping under the bushes, as they often do. They're always digging up roots and stripping bark off trees, The girl herself had tuberculosis, which is not at all unusual, I suppose they thought she would go back to her people, which they usually do. 48 Chapter Three Sometimes she would go and beg outside the railway station, and I think that sometimes people took pity on them and gave them money, I had never had a cupboard before and I was very proud of it. I had never had such beautiful things, and I would keep them under my pillow, where they were safe. Sometimes at night, I would wake up and [think] how lucky I was. But I would also cry sometimes, because I was thinking of my first life and I would be thinking about my uncles and aunts and [wondering] where they were now. Beyond this, Mms Potokwane relativises several propositions through with respect to probability, including certainty (must). These high valued realisations might be better read as contracting the heteroglossic dimensionality of text 2 (the [contract: proclaim: commit: pronounce] options in Figure 4), since they arguably proclaim Mma Potokwane's opinions. They must have found water in the salt pans and then wandered on towards one of the villages along the road to Maun. but nothing like that can deal with a lebolobolo bite, which is what this must have been. He must have been astonished to see a young Mosarwa child standing there with a baby in her arms. But, as you might or might not know, Mr J.L.B. Matekoni, when a Mosarwa woman dies and she's still feeding a baby, they bury the baby too. I suppose they thought she would go back to her people, and I think that sometimes people took pity on them and gave them money, The results of this analysis are summarised in Table 4 below, which includes 2 alternative realisations of the relevant engagement option here (i.e. [expand: entertain]) for each text (if, seemed)). Taken together with the attitude and projection analyses reviewed earlier, a clearer picture of Mma Potokwane's authoritative identity and Motholeli's unassuming one is beginning to emerge. Innocence: Realisation, Instantiation and Individuation 49 Table 4: Types of modality in texts 2 and 4 modulation: modalisation: ability obligation usuality probability Mma 13 [could] 3 [should] 9 [would] - (+2) [if, Potokwane seemed] (text 2) Motholeli 13 [could] 1 [must]* 7 [sometimes] 7 (+2) [if] (text 4) Turning from options which expand the play of voices in discourse to those which contract, Mma Potokwane and Motholeli use negation and concession in comparable ways to deny what they consider to be mistaken beliefs and to expectations: There are some of these people who wandered around on the edge of the Makadikadi Salt Pans, up on the road over to the Okavango. I don't know that part of the country very well, but I have been up there once or twice. My mother and father are late. I am very sad about this, but I am happy that I am not late too and that I have my brother. What distinguishes them is the range of proclaiming options Mma Potokwane uses to confirm and pronounce upon what she says: YOU MUST understand, said Mme Potokwane, that although it is easy for us to criticise the ways of the Basarwa, we should think carefully before we do that. When you look at the life they lead, out there in the Kalahari… when you think about that and wonder how long you and I and other Batswana would be able to live like that, then you realise that these bushmen are remarkable people. This band had made a camp about two or three miles outside the village. They hadn't built anything, of course, but were sleeping under the bushes, as they often do. Unfortunately there was a snake at her feet, and she rested her heel on its head. The snake bit her. That's how most snakebites occur. People are asleep on their sleeping mats and snakes come in for the warmth. Then they roll over onto the snake and the snake defends itself. According to the daughter, her mother died before the baby even woke up. Of course, they didn’t lose any time and they prepared to bury the mother that morning. But, as you might or might not know, Mr J.L.B. 50 Chapter Three Matekoni, when a Mosarwa woman dies and she's still feeding a baby, they bury the baby too. There just isn't the food to support a baby without a mother. That's the way it is. The driver slowed, and then stopped. He must have been astonished to see a young Mosarwa child standing there with a baby in her arms. Of course he could hardly leave her, even though he could not make out what she was trying to tell him. He was going back to Francistown and he dropped her off at the Nyangabwe Hospital, handing her over to an orderly at the gate. They looked at the baby, who was thin, and suffering badly from a fungal disease. The girl herself had tuberculosis, which is not at all unusual, and so they took her in and kept her in the TB ward for a couple of months while they gave her drugs. The nurse had heard of the orphan farm, and so she wrote to me. I said we could take them, and I went up to Francistown to collect them just a few months ago. Now they are with us, as you have seen. Motholeli is not assertive in this way; readers are positioned to accept her propositions at face value. It's not that she is uncertain, as she was construed by Smith (2002a) in text 3: [3] There were some people who had mothers to look after them. She was not one of those people, she knew. But why had her mother died? She remembered her only vaguely now. She remembered her death, and the wailing from the other women. She remembered the baby being taken from her arms and put in the ground. She had dug him out, she believed, but was not sure. Perhaps somebody else had done that and had passed the boy on to her. And then she remembered going away and finding herself in the strange place. (p.123) Recover from this she does in text 4 in the next book in the series. But she does not attempt to coerce her audience, smothering them with her authority. Although engagement theory has not explicitly to date dealt with internal conjunction in this regard (Halliday & Hasan, 1976, Martin, 2002, Martin & Rose, 2003), Mma Potokwane's authority is further enhanced by the number of justifications she gives for what she is saying. For Motholeli, causal relations are external, relating the events she has experienced to one another - for example: Innocence: Realisation, Instantiation and Individuation 51 After she died, I rescued my little brother, who had been buried in the sand with her. He was just under the sand, so I scraped it off his face and saw that he was still breathing. I used to do things for those kind people. I did their washing and hung it out on the line. I cleaned some pots for them too, as they did not have a servant. But I would also cry sometimes, because I was thinking of my first life and I would be thinking about my uncles and aunts and wondering where they were now. Mma Potokwane makes several of these event-based causal connections as well; but beyond this she relates propositions rhetorically by explaining why she takes up the position she does. The contrast we are focusing on here relates to the ambiguity of a proposition such as They arrived because I saw them - on the external reading, this means something like “I saw them and told them to come so they're here” whereas on the internal reading my seeing them is the reason why I'm justified in telling you they're here. YOU MUST understand, said Mme Potokwane, that although it is easy for us to criticise the ways of the Basarwa, we should think carefully before we do that. [I'm advising this because] When you look at the life they lead, out there in the Kalahari, with no cattle of their own and no houses to live in; when you think about that and wonder how long you and I and other Batswana would be able to live like that, then you realise that these bushmen are remarkable people. There was a band of Masarwa who had come up from the Kalahari to hunt ostriches. [which justifies me inferring that] They must have found water in the salt pans and then wandered on towards one of the villages along the road to Maun. Unfortunately there was a snake at her feet, and she rested her heel on its head. The snake bit her. [which allows me to generalise that] That's how most snakebites occur. People are asleep on their sleeping mats and snakes come in for the warmth. Then they roll over onto the snake and the snake defends itself. 52 Chapter Three They gave her some of their herbs. They're always digging up roots and stripping bark off trees, but nothing like that can deal with a lebolobolo bite, [I'm inferring this because] which is what this must have been. But, as you might or might not know, Mr J.L.B. Matekoni, when a Mosarwa woman dies and she's still feeding a baby, they bury the baby too. [I'm justifying this because] There just isn't the food to support a baby without a mother. That's the way it is. The driver slowed, and then stopped. [which justifies me inferring that] He must have been astonished to see a young Mosarwa child standing there with a baby in her arms. [which justifies me inferring that] Of course he could hardly leave her, even though he could not make out what she was trying to tell him. Then, they let them go. Beds on the TB ward were needed for other sick people and it was not the hospital's job to look after a Mosarwa girl with a baby. [which justifies me inferring that] I suppose they thought she would go back to her people, which they usually do. This nurse and her husband fed the children, but they couldn't take them into the family properly, [I'm saying this because] as they had two children of their own and they did not have a great deal of money. I spoke to the nurse about her, [which justifies me inferring that] and I understood that although she was still a child herself, she was a good mother to the boy. Sometimes she would go and beg outside the railway station, [which justifies me inferring that] and I think that sometimes people took pity on them and gave them money, but she preferred to earn it if she could. The nurse and her husband had to move. [I'm saying this because] Innocence: Realisation, Instantiation and Individuation 53 The husband worked for the meat-packing firm and they wanted him down in Lobatse. These internal cause/effect relations accumulate rationale by way of confirming the authority of Mma Potokwane's propositions; she is not just one voice among others, but a voice eclipsing others in the heteroglossic sphere she construes and controls. The ways in which Mma Potowkane deploys projection and modalisation mean that she claims to know more than she can have ever directly observed. Sometimes she explicitly acknowledges this, as with the must modalisations of certainty just reviewed. Other times she simply generalises behaviour beyond the instance she occasionally might have observed, modalised as sometimes, often, usually or always - upping the ante occasionally with proclamations about how life is: That's how most snakebites occur. People are asleep on their sleeping mats and snakes come in for the warmth. Then they roll over onto the snake and the snake defends itself. But, as you might or might not know, Mr J.L.B. Matekoni, when a Mosarwa woman dies and she's still feeding a baby, they bury the baby too. There just isn't the food to support a baby without a mother. That's the way it is. And then there is all the knowledge she has learned through the verbal and mental processes noted earlier. Mma Potokwane's conjecture, generalisations and subsumed knowledge are summarised in Appendix C; the way they result in a story that moves back and forth between observable experience and this extended knowledge is outlined in Table 5 below for the first few phases of her recount (for the complete analysis see Appendix D). What I am suggesting here is that reading down the left- hand “Observable experience” column gives us a complete story, of the kind Motholeli tells, which in Mma Potokwane's case is considerably elaborated through her extended knowledge on the right-hand side. 54 Chapter Three Table 5: Mma Potolwane alternating between observable experience and extended knowledge in text 2 Observable experience Extended knowledge YOU MUST understand, said Mme Potokwane, that although it is easy for us to criticise the ways of the Basarwa, we should think carefully before we do that. When you look at the life they lead, out there in the Kalahari, with no cattle of their own and no houses to live in; when you think about that and wonder how long you and I and other Batswana would be able to live like that, then you realise that these bushmen are remarkable people. There are some of these people who wandered around on the edge of the Makadikadi Salt Pans, up on the road over to the Okavango. I don't know that part of the country very well, but I have been up there once or twice. I remember the first time I saw it: a wide, white plain under a white sky, with a few tall palm trees and grass that seemed to grow out of nothing. It was such a strange landscape that I thought I had wandered out of Botswana into some foreign land. But just a little bit farther on it changes back into Botswana and you feel comfortable again. There was a band of Masarwa who had come up from the Kalahari to hunt ostriches. They must have found water in the salt pans and then wandered on towards one of the villages along the road to Maun. The people up there are sometimes suspicious of Basarwa, as they say that they steal their goats and will milk their cows at night if they are not watched closely. Innocence: Realisation, Instantiation and Individuation 55 This band had made a camp about two or three miles outside the village. They hadn't built anything, of course, but were sleeping under the bushes, as they often do They had plenty of meat - having just killed several ostriches - and were happy to stay there until the urge came upon them to move. There were a number of children and one of the women had just given birth to a baby, a boy. She was sleeping with him at her side, a little bit away from the others. She had a daughter too, who was sleeping on the other side of her mother. The mother woke up, we assume, and moved her legs about to be more comfortable. Unfortunately there was a snake at her feet, and she rested her heel on its head. The snake bit her. That's how most snakebites occur. People are asleep on their sleeping mats and snakes come in for the warmth. Then they roll over onto the snake and the snake defends itself. They gave her some of their herbs. They're always digging up roots and stripping bark off trees, but nothing like that can deal with a lebolobolo bite, which is what this must have been. According to the daughter, her mother died before the baby even woke up. Of course, they didn’t lose any time and they prepared to bury the mother that morning. But, as you might or might not know, Mr J.L.B. Matekoni, when a Mosarwa woman dies and she's still feeding a baby, they bury the baby too. There just isn't the food to support a baby without a mother. That's the way it is. 56 Chapter Three Given the conjectural status of much of this extended knowledge (about B events in Labov & Fanshel's 1977 terms), we can perhaps read Mma Potokwane's proclamations and rationalising internal conjunction as in a sense over-determined - her strategy for asserting her authority over propositions she cannot have directly experienced on her own. Motholeli on the other hand sticks to what's she done and what people say to her, and it is this more than anything which distinguishes her consciousness of lived experience from Mma Potokwane's learned wisdom. The result is a discourse that Mr J.L.B. Matekoni finds impossible to resist, and he adopts Motholeli and Puso without even consulting with his fiancé Mma Ramotswe: He sat down heavily on an upturned drum. These children were Mma Ramotswe's responsibility now, and he had not even asked her opinion. He had allowed himself to be bamboozled into taking them by that persuasive Mma Potokwane, and he had hardly thought out all the implications... (Smith, 2002a, p.105) Both rhetorics are of course designed to align readers, as well as the Botswanan characters with whom Mma Potowkane and Motholeli directly engage. Mma Potokwane aligns us persuasively, humanitarians, rallying around her judgement of Mr J.L.B. Matekoni's decision to care for the children: "What have I done?" muttered Mr J.L.B. Matekoni, under his breath. Mma Potokwane gave him his answer. "A very good thing," she said. (Smith, 2002a, p.89) Motholeli aligns us innocently, as proud relations, admiring the undaunted spirit and precocious courage of the child: AFTER THE girl had finished, they were all silent. The boy looked up at this sister and smiled. He though: I am a lucky boy to have such a clever sister. I hope that God will give her back her legs one day. Mma Ramotswe looked at the girl and laid a hand gently on her shoulder. She thought: I will look after this child. I am now her mother. Rose, who had been listening from the corridor, looked down at her shoes and thought: What a strange way of putting it: three lives. (Smith, 2002b, p.86) For my part, there is only one point in Motholeli's story where her bonding rhetoric fails, and that is where she agrees to love Jesus, and make Puso love him too, as a condition for being accepted at the orphan farm. This kind of manipulative conditional love is abhorrent to a confirmed Innocence: Realisation, Instantiation and Individuation 57 atheist such as myself. As a humanist, however, and parent, I cannot turn against Motholeli. As outlined by Knight (in press) in her work on humour and bonding, I thus had no choice but to take this part of her story as humorous and laugh off the coupling of love and Jesus to which Motholeli so readily subscribes. Given the rest of her tale, a smile is all it takes to keep me onside. Then a man came by who said that he was looking for orphans to take to his orphan farm. There was a lady from the Government with him who told me that I was very lucky to get a place on such a fine orphan farm. I could take my brother, and we would be very happy living there. But I must always remember to love Jesus, this woman said. I replied that I was ready to love Jesus and that I would make my little brother love him too. (Smith, 2002b, p.83-84) 6. Exploring individuation In this chapter I have made some provisional suggestions as to ways in which individuation can be explored, focusing on appraisal systems and the distinctive ways they can be used to express feelings and align readers. There is of course no way to construe identities other than by instantiating them in texts; and there is no way to form texts other than by drawing on the realisational resources members of a culture share. We must, in other words, keep all three hierarchies, realisation, instantiation and individuation in mind when exploring semantic variation, since all systems proposed for a given language and culture along the realisation hierarchy instantiate, and all individuate as well. This point is outlined in Figure 5 below. Changing levels of abstraction brings us no closer to or further away from instances of language use, nor closer or further away from individual language users. In this respect our exploration of identity and affiliation in this paper has been an extremely limited one. We've explored just two identities, as manifested in just 3 texts. Smith has a lot more to say in his The No. 1 Ladies' Detective Agency series about Mma Potokwane and Motholeli than is revealed here; and as these texts unfold, Motholeli's identity develops, ontogenetically speaking. She's the same girl but she matures, a process Smith choses not to pursue, backgrounding her character in more recent books. For data on the development of teenage identities, researchers will have to look elsewhere. As far as relevant realisation resources are concerned, we are barely scratching the surface. Alongside appraisal, the entire pandora's box of involvement systems passed over in Martin & White's (2005) appraisal 58 Chapter Three system are of extreme relevance (naming, technicality, abstraction, swearing, anti-language, etc.). And as outlined in Martin (in press), the way in which these resources couple with ideational meaning, within and across modalities of communication, is critical. We don't after all simply affiliate with feelings; we affiliate with feelings about people, places and things, and the activities they participate in, however abstract or concrete. We align through the bonds we make as we share these couplings. And we share identity, since identity is always already the ways that we belong. Figure 5: All strata instantiate; all strata individuate Notes 1. A system is a set of choices or interrelated options that are organized paradigmatically in a ‘system network’ in which the options are represented at increasing levels of delicacy, and from which speakers choose in order to make Innocence: Realisation, Instantiation and Individuation 59 meaning. SFL considers meaning as choice between semiotic alternatives in the meaning potential of language, and it is through systems of paradigmatic choices at each stratum that meaning is made in interaction. 2. Halliday (2005) provides a clear account of the realisation and instantiation complementarity. 3. Halliday and Matthiessen (1999) treat register and text type as alternative terms for the same level on this hierarchy, depending on whether we are looking from the system (register) or the text (text type) end; the terminology here is potentially confusing since Martin (e.g. 1992) uses register and genre as the names of contextual strata on his realisation hierarchy (and Halliday and Matthiessen do not distinguish register from genre). References Bernstein, B. (1996). Pedagogy, Symbolic Control and Identity: theory, research, critique. London: Taylor & Francis. [Revised Edition 2000] Caffarel, A. Martin, J.R. & Matthiessen, C.M.I.M. (Eds.) (2004). Language Typology: a functional perspective. Amsterdam: Benjamins Christie, F & Martin, J.R. (Eds.) (2007). Knowledge Structure: functional linguistic and sociological perspectives. London: Continuum. Coffin, C. (2006). Historical Discourse: the language of time, cause and evaluation. London: Continuum. Firth, J.R. (1957) A Synopsis of Linguistic Theory, 1930-1955. Studies in Linguistic Analysis (Special volume of the Philological Society) (1- 31). London: Blackwell. [reprinted in Palmer, F.R. (1968). (Ed.) Selected Papers of J R Firth, 1952-1959 (168-205). London: Longman.] Fuller, G. (1998). Cultivating science: negotiating discourse in the popular texts of Stephen Jay Gould. In J.R. Martin & R. Veel (Eds.) Reading Science: critical and functional perspectives on discourses of science (35-62). London: Routledge. Ghadessy, M. (Ed.) (1993). Register Analysis: theory and practice. London: Pinter. Ghadessy, M. (Ed.) (1999). Text and Context in Functional Linguistics. Amsterdam: Benjamins (CILT Series IV). Halliday, M.A.K. (2005). Computing meanings: some reflections on part experience and present prospects. [Computational and Quantitative Studies]. In J. Webster (Ed.) Collected Works of M A K Halliday, Volume 6 (239-267). London: Continuum. Halliday, M.A.K & Greaves, W.S. (2008). Intonation in the Grammar of English. London: Equinox. 60 Chapter Three Halliday, M.A.K & Matthiessen, C.M.I.M. (1999). Construing Experience through Language: a language-based approach to cognition. London: Cassell. Halliday, M.A.K & Matthiessen, C.M.I.M. (2004). An Introduction to Functional Grammar, 3rd ed. London: Arnold. Halliday, M A K & Hasan, R. (1976). Cohesion in English. London: Longman. Hasan, R. (1996). Ways of Saying: ways of meaning. In C. Cloran, D. Butt & G. Williams (Eds.) Selected papers of Ruqaiya Hasan. London: Cassell (Open Linguistics Series). —. (2005). Language, Society and Consciousness. In J. Webster (Ed.) The Collected Works of Ruqaiya Hasan (106-129). London: Equinox. Knight, N. (in press) “Still cool…and american too!”: an SFL analysis of deferred bonds in internet messaging humour. In N. Nørgaard (Ed.) Odense Working Papers in Language and Communication (Special Issue of Papers from 34th International Systemic Functional Congress), Odense. Labov, W & Fanshel, D. (1977). Therapeutic Discourse: psychotherapy as conversation. New York: Academic press. Martin, J.R. (1992). English Text: system and structure. Amsterdam: Benjamins. —. (2006). Genre, ideology and intertextuality: a systemic functional perspective. Linguistics and the Human Sciences 2.2. (Special Issue on Genre edited by J. Bateman), p.275-298. —. (in press) Tenderness: realisation and individuation in a Botswanan town. In N. Nørgaard (Ed.) Odense Working Papers in Language and Communication (Special Issue of Papers from 34th International Systemic Functional Congress), Odense. —. (to appear). Boomer dreaming: the texture of recolonisation in a lifestyle magazine. In G. Forey & G. Thompson (Eds.) Text-type and Texture. London: Equinox. Martin, J.R. & Plum, G. (1997). Construing experience: some story genres. Journal of Narrative and Life History 7(1-4), 299-308. (Special Issue: Oral Versions of Personal Experience: three decades of narrative analysis. M.Bamberg Guest Editor). Martin, J.R. & Rose, D. (2003). Working with Discourse: meaning beyond the clause. London: Continuum [2nd Revised Edition 2007]. Martin, J.R. & Rose, D. (2007). Genre Relations: mapping culture. London: Equinox. Innocence: Realisation, Instantiation and Individuation 61 Martin, J.R. & Wodak, R. (Eds.) (2003). Re/reading the past: critical and functional perspectives on discourses of history. Amsterdam: Benjamins. Martin, J.R. & White, P.R.R. (2005). The Language of Evaluation: appraisal in English. London: Palgrave. Smith, A.M. (2002a). Tears of the Giraffe. New York: Anchor Books. [Time Warner Audio Books 2003]. —. (2002b). Morality for Beautiful Girls. New York: Anchor Books. [Time Warner Audio Books 2003]. 62 Chapter Three APPENDIX A: THE STORIES [1] "Do you live here?" he asked. "On the farm?" She nodded. "I have been here nearly one year," she answered. "I am here with my young brother. He is only five." "Where did you come from?" She lowered her gaze. "We came from up near Francistown. My mother is late. She died three years ago, when I was nine. We lived with a woman, in her yard. Then she told us we had to go." Mr J.L.B. Matekoni said nothing. Mma Potokwane had told him the stories of some of the orphans, and each time he had found that it made his heart smart with pain... (Smith, 2002a, p.49) [2] YOU MUST understand, said Mme Potokwane, that although it is easy for us to criticise the ways of the Basarwa, we should think carefully before we do that. When you look at the life they lead, out there in the Kalahari, with no cattle of their own and no houses to live in; when you think about that and wonder how long you and I and other Batswana would be able to live like that, then you realise that these bushmen are remarkable people. There are some of these people who wandered around on the edge of the Makadikadi Salt Pans, up on the road over to the Okavango. I don't know that part of the country very well, but I have been up there onceor twice. I remember the first time I saw it: a wide, white plain under a white sky, with a few tall palm trees and grass that seemed to grow out of nothing. It was such a strange landscape that I thought I had wandered out of Botawana into some foreign land. But just a little but farther on it changes back into Botswana and you feel comfortable again. There was a band of Masarwa who had come up from Kalahari to hunt ostriches. They must have found water in the salt pans and then wandered on towards one of the villages along the road to Maun. The people up there are sometimes suspicious of Basarwa, as they say that they steal their goats and will mile their cows at night if they are not watched closely. This band had made a camp about two or three miles outside the village. They hadn't built anything, of course, but were sleeping under the bushes, as they often do. They had plenty of meat-having just killed several ostriches-and were happy to stay there until the urge came upon them to move. There were a number of children and one of the women had just given birth to a baby, a boy. She was sleeping with him at her side, a little bit away from the others. She had a daughter too, who was sleeping on the Innocence: Realisation, Instantiation and Individuation 63 other side of her mother. The mother woke up, we assume, and moved her legs about to be more comfortable. Unfortunately there was a snake at her feet, and she rested her heel on its head. The snake bit her. That's how most snakebites occur. People are asleep on their sleeping mats and snakes come in for the warmth. Then they roll over onto the snake and the snake defends itself. They gave her some of their herbs. They're always digging up roots and striping bark off trees, but nothing like that can deal with a lebolobolo bite, which is what this must have been. According to the daughter, her mother died before the baby even woke up. Of course, they didn’t lose any time and they prepared to bury the mother that morning. But, as you might or might not know, Mr J.L.B. Matekoni, when a Mosarwa woman dies and she's still feeding a baby, they bury the baby too. There just isn't the food to support a baby without a mother. That's the way it is. The girl hid in the bush and watched them take her mother and her baby brother. It was sandy there, and all they could manage was a shallow grave, in which they laid her mother, while the other women wailed and the men sang something. The girl watched as they put her tiny brother into the grave too, wrapped in an animal skin. Then they pushed the sand over them both and went back to the camp. The moment they had gone, the child crept out and scrabbled quickly at the sand. It did not take her long and soon she had her brother in her arms. The was sand in the child's nostrils, but he was still breathing. She turned on her heels and ran through the bush in the direction of the road, which she knew was not too far away. A truck came past a short time later, a Government truck from the Roads Department. The driver slowed, and then stopped. He must have been astonished to see a young Mosarwa child standing there with a baby in her arms. Of course he could hardly leave her, even though he could not make out what she was trying to tell him. He was going back to Francistown and he dropped her off at the Nyangabwe Hospital, handing her over to an orderly at the gate. They looked at the baby, who was thin, and suffering badly from a fungal disease. The girl herself had tuberculosis, which is not at all unusual, and so they took her in and kept her in the TB ward for a couple of months while they gave her drugs. The baby stayed in the maternity nursery until the girl was better. Then, they let them go. Beds on the TB ward were needed for other sick people and it was not the hospital's job to look after a Mosarwa girl with a baby. I suppose they thought she would go back to her people, which they usually do. One of the sisters at the hospital was concerned. She saw the girl sitting at the hospital gate and she decided that she had nowhere to go. So she took her home and let her stay in her backyard, in a lean-to shack that they had used for storage but which could be cleared out to provide a room of sorts. This nurse and her husband fed the children, but they couldn't take them into the family properly, as they had two children of their own and they did not have a great deal of money. 64 Chapter Three The girl picked up Setswana quite quickly. She found ways of making a few pula by collecting empty bottles from the edge of the road and taking them back to the bottle store for the deposit. She carried the baby on her back, tied in a sling, and never let him leave her sight. I spoke to the nurse about her, and I understood that although she was still a child herself, she was a good mother to the boy. She made his clothes out of scraps that she found here and there, and she kept him clean by washing him under the tap in the nurse's backyard. Sometimes she would go and beg outside the railway station, and I think that sometimes people took pity on them and gave them money, but she preferred to earn it if she could. This went on for four years. Then, quite without warning, the girl became ill. They took her back to the hospital and they found that the tuberculosis had damaged the bones very badly. Some of them had crumbled and this was making it difficult for her to walk. They did what they could, but they were unable to prevent her from ending up unable to walk. The nurse scrounged around for a wheelchair, which she was eventually given by one of the Roman Catholic priests. So now she looked after the boy from the wheelchair, and he, for his part, did the little chores for his sister. The nurse and her husband had to move. The husband worked for the meat-packing firm and they wanted him down in Lobatse. The nurse had heard of the orphan farm, and so she wrote to me. I said we could take them, and I went up to Francistown to collect them just a few months ago. Now they are with us, as you have seen. That is their story, Mr J.L.B Matekoni. That is how they came to be with us here. (Smith, 2002a, p.84-88) [3] There were some people who had mothers to look after them. She was not one of those people, she knew. But why had her mother died? She remembered her only vaguely now. She remembered her death, and the wailing from the other women. She remembered the baby being taken from her arms and put in the ground. She had dug him out, she believed, but was not sure. Perhaps somebody else had done that and had passed the boy on to her. And then she remembered going away and finding herself in the strange place. (Smith, 2002a, p.123) [4] ...Mma Ramotswe joined them, perching on the arm of the sofa. "Why don't you start off again?" she said. "I would like too hear your story." MY NAME is Motholeli and I am thirteen years old, almost fourteen. I have a brother, who is seven. My mother and father are late. I am very sad about this, but I am happy that I am not late too and that I have my brother. Innocence: Realisation, Instantiation and Individuation 65 I am a girl who has had three lives. My first life was when I lived with my mother and aunts and uncles, up in the Makadikadi, near Nata. That was a long time ago, and I was very small. They were bush people and they moved from place to place. They knew how to find food in the bush by digging for roots. They were very clever people, but nobody liked them. My mother gave me a bracelet made out of ostrich skin, with pieces of ostrich eggshell stitched into it. I still have that. It is the only thing I have from my mother, now that she is late. After she died, I rescued my little brother, who had been buried in the sand with her. He was just under the sand, so I scraped it off his face and saw that he was still breathing. I remember picking him up and running through the bush until I found a road. A man came down the road in a truck and when he saw me he stopped and took me to Francistown. I do not remember what happened there, but I was given to a woman who said that I could live in her yard. They had a small shed, which was very hot when the sun was on it, but which was cool at night. I slept there with my baby brother. I fed him with the food I was given from that house. I used to do things for those kind people. I did their washing and hung it out on the line. I cleaned some pots for them too, as they did not have a servant. There was a dog who lived in the yard too, and it bit me one day, sharply, in my foot. The woman's husband was very cross with the dog after that and beat it with a wooden pole. That dog is late now, after all that beating for being wicked. I became very sick, and the woman took me to the hospital. They put needles into me and they took out some of my blood. But they could not make me better, and after a while I could not walk anymore. They gave me crutches, but I was not very good at walking with them. Then they found a wheelchair; which meant that I could go home again. But the woman said that she could not have a wheelchair girl living in her yard, as that would not look good and people would say: What are you doing having a girl in a wheelchair in your yard? That is very cruel. Then a man came by who said that he was looking for orphans to take to his orphan farm. There was a lady from the Government with hi who told me that I was very lucky to get a place on such a fine orphan farm. I could take my brother, and we would be very happy living there. But I must always remember to love Jesus, this woman said. I replied that I was ready to love Jesus and that I would make my little brother love him too. That was the end of my first life. My second life started on the day I arrived at the orphan farm. We had come down from Francistown in a truck, and I was very hot and uncomfortable in the back. I could not get out, as the truck driver did not know what to do with a girl in a wheelchair. So when I arrived at the orphan farm, my dress was wet and I was very ashamed, especially since all the other orphans were standing there watching us come to their place. One of the ladies there told the children 66 Chapter Three to go off and play, and not to stare at us, but they only went a little way and they watched me from behind the trees. All the orphans lived in houses. Each house had about ten orphans in it and had a mother who looked after them. My housemother was a kind lady. She gave me new clothes and a cupboard to keep my things in. I had never had a cupboard before and I was very proud of it. I was also given some special clips which I could put in my hair. I had never had such beautiful things, and I would keep them under my pillow, where they were safe. Sometimes at night, I would wake up and think how lucky I was. But I would also cry sometimes, because I was thinking of my first life and I would be thinking about my uncles and aunts and wondering where they were now. I could see the stars from my bed, through a gap in the curtain, and I thought: if they looked up, they would see the same stars, and we would be looking at them at the same time. But I wondered if they remembered me, because I was just a girl and I had run away from them. I was very happy at the orphan farm. I worked hard, and Mma Potokwane, who was the matron, said that one day, if I was lucky, she would find somebody who would be new parents for us. I did not think that this was possible, as nobody would want to take a girl in a wheelchair when there were plenty of first-class orphan girls who could walk very well and who would be looking for a home too. But she was right. I did not think it would be Mr J.LB. Maketoni who took us, but I was very pleased when he said that we could go to live in his house. That is how my third life began. They made us a special cake when we left the orphan farm, and we ate it with the housemother. She said that she always felt very sad when one of the orphans went, as it was like a member of the family was leaving. But she knew Mr J.L.B. Maketoni very well, and she told me he was one of the best men in Botswana. I would be very happy in his house, she said. So I went to his house, with my small brother, and we soon met his friend, Mma Ramotswe, who is going to be married to Mr J.L.B. Maketoni. She said that she would be my new mother, and she brought us to her house, which is better for children than Mr J.L.B. Maketoni's house. I have a very good bedroom there, and I have been given many clothes. I am very happy that there are people like this in Botswana. I have had a fortunate life and I thank Mma Ramotswe and Mr J.L.B. Maketoni from my heart. I would like to be a mechanic when I grow up. I shall help Mr J.L.B. Maketoni in his garage and at night I will mend Mma Ramotswe's clothes and cook her meals. Then, when they are very old they will be able to be proud of me and say that I have been a good daughter for them and a good citizen of Boswana. That is the story of my life. I am an ordinary girl from Botswana, but it is very lucky to have three lives. Most people only have one life. This story is true. I have not made any of it up. Innocence: Realisation, Instantiation and Individuation 67 AFTER THE girl had finished, they were all silent. The boy looked up at this sister and smiled. He though: I am a lucky boy to have such a clever sister. I hope that God will give her back her legs one day. Mma Ramotswe looked at the girl and laid a hand gently on her shoulder. She thought: I will look after this child. I am now her mother. Rose, who had been listening from the corridor, looked down at her shoes and thought: What a strange way of putting it: three lives. (Smith, 2002b, p.82-86) 68 Chapter Three APPENDIX B: MODALITY ANALYSIS how long you and I and other Batswana would be able to live like that, but nothing like that can deal with a lebolobolo bite, and all they could manage was a shallow grave, in which they laid her mother, while the other women wailed and the men sang something Of course he could hardly leave her, even though he could not make out what she was trying to tell him. but which could be cleared out to provide a room of sorts. This nurse and her husband fed the children, but they couldn't take them into the family properly, as they had two children of their own and they did not have a great deal of money. but she preferred to earn it if she could. They did what they could, but they were unable to prevent her from ending up unable to walk. I said we could take them, and I went up to Francistown to collect them just a few months ago. YOU MUST understand, said Mme Potokwane, we should think carefully before we do that. The nurse and her husband had to move. The people up there are sometimes suspicious of Basarwa, but were sleeping under the bushes, as they often do. They're always digging up roots and stripping bark off trees, The girl herself had tuberculosis, which is not at all unusual, I suppose they thought she would go back to her people, which they usually do. Sometimes she would go and beg outside the railway station, and I think that sometimes people took pity on them and gave them money, They must have found water in the salt pans and then wandered on towards one of the villages along the road to Maun. which is what this must have been. But, as you might or might not know, Mr J.L.B. Matekoni, when a Mosarwa woman dies and she's still feeding a baby, they bury the baby too. He must have been astonished to see a young Mosarwa child standing there with a baby in her arms. I suppose they thought she would go back to her people, and I think that sometimes people took pity on them and gave them money, Innocence: Realisation, Instantiation and Individuation 69 if they are not watched closely. with a few tall palm trees and grass that seemed to grow out of nothing. who said that I could live in her yard. But they could not make me better, and after a while I could not walk anymore. which meant that I could go home again. But the woman said that she could not have a wheelchair girl living in her yard, I could take my brother, I could not get out, as [the truck driver did not know] what to do with a girl in a wheelchair. I was also given some special clips which I could put in my hair. I could see the stars from my bed, through a gap in the curtain, and [I thought]: I did not think that this was possible, when there were plenty of first-class orphan girls who could walk very well but I was very pleased when [he said] that we could go to live in his house. Then, when they are very old they will be able to be proud of me and say that I have been a good daughter for them and a good citizen of Botswana. But I must always remember to love Jesus, this woman said I had never had a cupboard before and I was very proud of it. I had never had such beautiful things, and I would keep them under my pillow, where they were safe. Sometimes at night, I would wake up and [think] how lucky I was. But I would also cry sometimes, because I was thinking of my first life and I would be thinking about my uncles and aunts and [wondering] where they were now. if they looked up, if I was lucky, 70 Chapter Three APPENDIX C: MMA POTOKWANE'S INFERENCES, GENERALISATIONS AND PROJECTED KNOWLEDGE conjecture They must have found water in the salt pans and then wandered on towards one of the villages along the road to Maun. but nothing like that can deal with a lebolobolo bite, which is what this must have been. He must have been astonished to see a young Mosarwa child standing there with a baby in her arms. generalisation The people up there are sometimes suspicious of Basarwa, as they say that they steal their goats and will milk their cows at night if they are not watched closely. I think that sometimes people took pity on them and gave them money, but she preferred to earn it if she could. but were sleeping under the bushes, as they often do I suppose they thought she would go back to her people, which they usually do. They're always digging up roots and stripping bark off trees, but nothing like that can deal with a lebolobolo bite, which is what this must have been. That's how most snakebites occur. People are asleep on their sleeping mats and snakes come in for the warmth. Then they roll over onto the snake and the snake defends itself. But, as you might or might not know, Mr J.L.B. Matekoni, when a Mosarwa woman dies and she's still feeding a baby, they bury the baby too. There just isn't the food to support a baby without a mother. That's the way it is. Innocence: Realisation, Instantiation and Individuation 71 vicariously acquired knowledge [hearsay] The people up there are sometimes suspicious of Basarwa, as they say that they steal their goats and will milk their cows at night if they are not watched closely. According to the daughter, her mother died before the baby even woke up. I spoke to the nurse about her, and I understood that although she was still a child herself, she was a good mother to the boy. [assumption] YOU MUST understand, said Mme Potokwane, that although it is easy for us to criticise the ways of the Basarwa, we should think carefully before we do that. When you look at the life they lead, out there in the Kalahari, with no cattle of their own and no houses to live in; when you think about that and wonder how long you and I and other Batswana would be able to live like that, then you realise that these bushmen are remarkable people. The mother woke up, we assume, and moved her legs about to be more comfortable. But, as you might or might not know, Mr J.L.B. Matekoni, when a Mosarwa woman dies and she's still feeding a baby, they bury the baby too. Beds on the TB ward were needed for other sick people and it was not the hospital's job to look after a Mosarwa girl with a baby. I suppose they thought she would go back to her people, which they usually do. and she decided that she had nowhere to go. I spoke to the nurse about her, and I understood that although she was still a child herself, she was a good mother to the boy. I think that sometimes people took pity on them and gave them money, but she preferred to earn it if she could. and they found that the tuberculosis had damaged the bones very badly. Some of them had crumbled and this was making it difficult for her to walk. 72 Chapter Three APPENDIX D: OBSERVABLE EXPERIENCE AND EXTENDED KNOWLEDGE (TEXT 2) Observable experience Extended knowledge YOU MUST understand, said Mme Potokwane, that although it is easy for us to criticise the ways of the Basarwa, we should think carefully before we do that. When you look at the life they lead, out there in the Kalahari, with no cattle of their own and no houses to live in; when you think about that and wonder how long you and I and other Batswana would be able to live like that, then you realise that these bushmen are remarkable people. There are some of these people who wandered around on the edge of the Makadikadi Salt Pans, up on the road over to the Okavango. I don't know that part of the country very well, but I have been up there once or twice. I remember the first time I saw it: a wide, white plain under a white sky, with a few tall palm trees and grass that seemed to grow out of nothing. It was such a strange landscape that I thought I had wandered out of Botswana into some foreign land. But just a little bit farther on it changes back into Botswana and you feel comfortable again. There was a band of Masarwa who had come up from the Kalahari to hunt ostriches. They must have found water in the salt pans and then wandered on towards one of the villages along the road to Maun. The people up there are sometimes suspicious of Basarwa, as they say that they steal their goats and will milk their Innocence: Realisation, Instantiation and Individuation 73 cows at night if they are not watched closely. This band had made a camp about two or three miles outside the village. They hadn't built anything, of course, but were sleeping under the bushes, as they often do They had plenty of meat - having just killed several ostriches - and were happy to stay there until the urge came upon them to move. There were a number of children and one of the women had just given birth to a baby, a boy. She was sleeping with him at her side, a little bit away from the others. She had a daughter too, who was sleeping on the other side of her mother. The mother woke up, we assume, and moved her legs about to be more comfortable. Unfortunately there was a snake at her feet, and she rested her heel on its head. The snake bit her. That's how most snakebites occur. People are asleep on their sleeping mats and snakes come in for the warmth. Then they roll over onto the snake and the snake defends itself. They gave her some of their herbs. They're always digging up roots and stripping bark off trees, but nothing like that can deal with a lebolobolo bite, which is what this must have been. According to the daughter, her mother died before the baby even woke up. Of course, they didn’t lose any time and they prepared to bury the mother that morning. But, as you might or might not know, Mr J.L.B. Matekoni, when a Mosarwa woman dies and she's still feeding a baby, they bury the baby too. There just isn't the food to support a baby without a mother. That's the way it is. 74 Chapter Three The girl hid in the bush and watched them take her mother and her baby brother. It was sandy there, and all they could manage was a shallow grave, in which they laid her mother, while the other women wailed and the men sang something. The girl watched as they put her tiny brother into the grave too, wrapped in an animal skin. Then they pushed the sand over them both and went back to the camp. The moment they had gone, the child crept out and scrabbled quickly at the sand. It did not take her long and soon she had her brother in her arms. There was sand in the child's nostrils, but he was still breathing. She turned on her heels and ran through the bush in the direction of the road, which she knew was not too far away. A truck came past a short time later, a Government truck from the Roads Department. The driver slowed, and then stopped. He must have been astonished to see a young Mosarwa child standing there with a baby in her arms. Of course he could hardly leave her, even though he could not make out what she was trying to tell him. He was going back to Francistown and he dropped her off at the Nyangabwe Hospital, handing her over to an orderly at the gate. They looked at the baby, who was thin, and suffering badly from a fungal disease. The girl herself had tuberculosis, which is not at all unusual, and so they took her in and kept her in the TB ward for a couple of months while they gave her drugs. The baby stayed in the maternity nursery until the girl was better. Then, they let them go. Beds on the TB ward were needed for other sick people and it was not the hospital's job to look after a Mosarwa Innocence: Realisation, Instantiation and Individuation 75 girl with a baby. I suppose they thought she would go back to her people, which they usually do. One of the sisters at the hospital was concerned. She saw the girl sitting at the hospital gate and she decided that she had nowhere to go. So she took her home and let her stay in her backyard, in a lean-to shack that they had used for storage but which could be cleared out to provide a room of sorts. This nurse and her husband fed the children, but they couldn't take them into the family properly, as they had two children of their own and they did not have a great deal of money. The girl picked up Setswana quite quickly. She found ways of making a few pula by collecting empty bottles from the edge of the road and taking them back to the bottle store for the deposit. She carried the baby on her back, tied in a sling, and never let him leave her sight. I spoke to the nurse about her, and I understood that although she was still a child herself, she was a good mother to the boy. She made his clothes out of scraps that she found here and there, and she kept him clean by washing him under the tap in the nurse's backyard. Sometimes she would go and beg outside the railway station, and I think that sometimes people took pity on them and gave them money, but she preferred to earn it if she could. This went on for four years. Then, quite without warning, the girl became ill. They took her back to the hospital and they found that the tuberculosis had damaged the bones very badly. Some of them had crumbled and this was making it difficult for her to walk. 76 Chapter Three They did what they could, but they were unable to prevent her from ending up unable to walk. The nurse scrounged around for a wheelchair, which she was eventually given by one of the Roman Catholic priests. So now she looked after the boy from the wheelchair, and he, for his part, did the little chores for his sister. The nurse and her husband had to move. The husband worked for the meat-packing firm and they wanted him down in Lobatse. The nurse had heard of the orphan farm, and so she wrote to me. I said we could take them, and I went up to Francistown to collect them just a few months ago. Now they are with us, as you have seen. That is their story, Mr J.L.B Matekoni. That is how they came to be with us here. [88] CHAPTER FOUR RECONCILING THE CO-ARTICULATION OF MEANING BETWEEN WORDS AND PICTURES: EXPLORING INSTANTIATION AND COMMITMENT IN IMAGE NUCLEAR NEWS STORIES HELEN CAPLE 1. Introduction In The Sydney Morning Herald, a daily broadsheet newspaper in Sydney, NSW, news stories are often presented as short image nuclear news stories (Caple, forthcoming a), where a news photograph combines with a heading and a short caption in the telling of a recent newsworthy event (see Figure 1). There is no extended text with such stories, and one of the distinguishing features of this particular type of multimodal news story genre is the fact that the heading and image invariably combine to create an initial reading of the story that is based in a play on the words and image. This play is realised in a number of different ways and forms the focus of the analyses presented in this chapter. 78 Chapter Four Figure 1: An Image Nuclear News Story (SMH, 22/03/2005, p.3). (Full page in Appendix A) In a corpus of 900 such stories, I have been able to track the emergence of this play between the heading and image, and the many different ways in which it is manifested in the stories themselves. Given that 95% of these stories engage in this initial play between the heading and the image, this co-construal of such complex meanings in image nuclear news stories is an area that has proved itself worthy of further investigation. The aim of this chapter then is to examine the ways in which the words and images combine to make meaning and the analytical tools employed for the task will include the concepts of instantiation and commitment from within the linguistic framework of Systemic Functional Linguistics (hereafter SFL). Thus, this chapter will address the issue of the co-construal of meanings between words and pictures in image nuclear news stories, by examining the multistratal and bimodal implications of the play between the photographs and the words accompanying them. 2. Theoretical underpinnings According to SFL theory, language, as a complex semiotic system, can be analysed both syntagmatically and paradigmatically. What is meant by this is that structural operations, such as the use of this lexeme over that deployed in this order rather than that, are seen as “realizing systemic choices” (Halliday & Matthiessen, 2004, p.24). As Halliday and Matthiessen (2004) state “when we analyse a text, we show the functional organization of its structure; and we show what meaningful choices have Reconciling the Co-articulation of Meaning between Words and Pictures 79 been made, each one seen in the context of what might have been meant but was not” (p.24). This is the realisation hierarchy of stratification. In SFL, language is stratified into three levels: meanings (discourse semantics) realised as wordings (lexicogrammar), which are in turn realised as sounds (phonology) or letters (graphology). Likewise, for the analysis of photographic images, the systemic functional framework has been usefully adapted by taking a metafunctional approach (see Kress & van Leeuwen, 2006; O’Toole, 1994) to analyse the systemic choices that have been made in the composition of such images. Another global semiotic dimension from which this new news story genre can be viewed is that of instantiation. Like stratification, the cline of instantiation also looks at how language is organised or functions within a particular context, however, it is different to stratification in that it allows us to maintain two perspectives at once: that of language as system and language as text (Halliday & Matthiessen, 2004). Thus, instantiation is described as a cline, where we may observe a text from the point of view of the system at one end of the cline or from the standpoint of the instance at the other. By system we mean the overall generalised meaning potential of the system of language, and by instance, we mean the affording instance, the actual text as it is manifested in a particular context at a particular time. In between these two poles we have (in Halliday’s (1978) terms) the functional varieties of language or registers (viewed from the system end of the cline), and text types or the patterns of meanings in a sample group (as viewed from the instance end of the cline). Martin and Rose (2003) describe this shifting perspective as apparent flux (at the instance end of the cline), as inertia (at the system end of the cline), or as something in between. Figure 2 shows the cline of instantiation. Different to Halliday and Matthiessen’s version of the cline, Martin and Rose (2003) introduce reading as a final step on this model. This extra stage will be dealt with more fully later in this chapter. Figure 2: The Cline of Instantiation (Martin & Rose, 2003). 80 Chapter Four I have collected and analysed 900 of these image nuclear news stories, and as a corpus the relative frequencies of language choices being made at the instance end of the cline can be distilled as systemic probabilities at the potential pole of the cline. Thus, as a corpus, the patterns of meanings across these texts can be usefully examined using the cline of instantiation. 3. Instantiation and image nuclear news stories One of the most important considerations to be remembered as far as instantiation is concerned is that every stratum on the realisation hierarchy instantiates. Thus it may be useful to look at how stratification and instantiation intersect. Halliday (2000, cited in Matthiessen, 2005) modelled the intersection of the cline of instantiation with the hierarchy of stratification in the “Instantiation-stratification matrix” (see Table 1) as an alternative way of modelling language in context. Table 1: Instantiation-stratification matrix (Halliday, 2000, cited in Matthiessen, 2005) INSTANTIATION system subsystem/instance type instance STRATIFICATION institution context culture situation situation type register semantics semantic system [text as] meaning text type register lexicogrammar grammatical system [text as] wording text type Since it is difficult to talk of stratification in the photographs in my corpus, as multimodal theorists have yet to commit to any definitive systematic way of describing the levels of abstraction in images, we can, of course, describe stratification and the verbiage in these stories. The cline of instantiation, however, may prove itself very useful in examining text-image relations, in particular where they co-construe meaning. Thus, Reconciling the Co-articulation of Meaning between Words and Pictures 81 while in the following the analysis of the image itself may remain necessarily scant, given the restrictions of this chapter, it is hoped that a fruitful analysis of text-image relations will be offered. By examining how these two global dimensions intersect in relation to the choices being made in creating image nuclear news stories, we can demonstrate the cline of instantiation at work. The entire meaning potential of the English language is made available in the creation of image nuclear news stories. However, those choices are immediately constrained by the context in which we find such stories, that is, in Western print news media, in particular in the broadsheet, The Sydney Morning Herald. At the level of context then, the instance of the image nuclear news story can be viewed as the selection expression of the variables of field, tenor and mode: field examines what the language is being used to talk about, tenor gives us the role relationships between the interactants and mode looks at the role language is playing in the interaction (Eggins, 1994). In terms of field choices, this news story genre retells recent newsworthy events, be they destabilising, restabilising or established events taking place in both the local and international news arena. Modally, we see the combination of a salient image, a heading and a caption as illustration of this news event that is usually separated from other stories on the page using a single black line rule (refer to Appendices for full page view). This is, then, a multimodal news story genre. Tenor relations in this news story genre are between the creators of this text and the readers, that is, between certain sub-editors who are represented externally via the newspaper as an institution and as an ideological construct to the audience – an educated, middle-class reading public. Image nuclear news stories are different from other multimodal news stories, in that there is a direct and meaningful relationship between the heading and the image. Other multimodal news stories, on the other hand, tend to display a stronger relationship between the headline and the verbiage, usually the lead paragraph, with the photograph and its caption residing somewhere within the general layout of the story. In such stories, the represented participants in the photograph may or may not be referred to at any stage of the retelling of the news event in the main body of the text or they may only be referred to in the caption underneath the photograph. Contrary to this, in image nuclear news stories, at the level of discourse semantics, meaning is created through the combination of the words in the heading and the image itself. Often, several layers of meaning are built up in these stories and one may be required to draw on other intertextual references, or to draw on both interpersonal and ideational metafunctional resources, to begin to unpack these layers of meaning. The 82 Chapter Four caption in image nuclear news stories usually reveals the more congruent meanings, which are more closely tied to the retelling of the news event, and which may be viewed as the original driving force behind the inclusion of such a story in the newspaper in the first instance. In creating these layers of meaning in image nuclear news stories, most of which are ultimately humorous, manipulation of both the lexis and grammar takes place at the level of lexicogrammar. Sometimes famous quotations are taken from their original context and redeployed in image nuclear news stories. For example, the saying “Beam me up, Scotty”, an infamous line (that was never actually phrased as such) from the popular science fiction television series Star Trek, has been slightly modified and redeployed in the heading of an image nuclear news story (see Figure 1) as “Beam me up, Scottie: library's treasures go online”. The change in spelling of “Scotty” to “Scottie” further enhances the fact that this quote has been totally recontextualised, and that it now refers not to the Starship Enterprise’s transporter chief, Montgomery "Scotty" Scott, but rather to Scott Wajon (pictured in Figure 1), from the State Library of NSW's imaging services, and to the efforts being undertaken to make available, via the library website, electronically scanned images of the precious artefacts lodged with the State Library. A similar treatment of the words and phrases associated with literary or filmic titles, or with proper names can also be observed in other image nuclear news stories. Certain compositional choices are also in evidence in image nuclear news stories, as the images chosen for this news story genre are often those that have been carefully composed to achieve a sense of balance in the image (see Caple, fortcoming b). The overall effect of such images is often very powerful visually and, to a degree, aesthetically motivated. Particular typographical choices are also in evidence in the heading, separating this news story from others on the same page, as these stories use a modern, sans serif font that until recently was used exclusively in this news story genre (see full pages in the Appendices). One of the defining features of image nuclear news stories, however, is the fact that the heading and image invariably combine in a playful relationship. This is one that relies on the reader having extensive linguistic, cultural, and general knowledge that he or she can draw on in unpacking the layers of play built up in these stories. This coupling of image and heading and the meanings that such coupling affords will be explored further in the following section. Reconciling the Co-articulation of Meaning between Words and Pictures 83 4. Coupling and commitment in image nuclear news stories The co-construal of meaning in image nuclear news stories occurs across two semiotic modes, that of verbal language in the heading and that of the visual representation of an event in the photographic image. Each, of course, can be read or viewed separately. However, given the fact that the heading always appears above the image, with white space to the right of the heading and sometimes also along the side of the image (as in Figure 4), and that the ideational content of the heading is usually reflected in the ideational content of the image, the reader should also attempt to read the two together. It is to this coupling of the heading and image to form a playful relationship that we now turn. Martin (2000) first introduced the notion of “coupling” in his seminal work on the system of APPRAISAL within the SFL framework. He refers to coupling as the binding of two meanings, be it across genres, systems, or contexts, at any point along the cline of instantiation (Martin, in press). He has also used the term to explain humour and sarcasm “as involving discordant couplings – either between appraisal selections and what is being appraised, or among the appraisal variables themselves” (Martin 2000: 164). If we now turn our attention to the play in image nuclear news stories, we can see that there are two principle ways in which the coupling between words and images in this genre are manifested. These are the recontextualisation of established phrases such as famous quotations from movies or speeches or the use of proper names and titles, again removed from their original context, and the manipulation of common idiomatic expressions. Famous quotations are taken from movies, novels, or famous speeches. These tend to iconise the character or actor (e.g. Captain Kirk, Humphrey Bogart) and are seen as membershipping devices, creating a fandom, or bondicons (Stenglin, 2004; Martin 2007) for people who love, for example, Shakespeare, Crocodile Dundee, or Dirty Harry. According to Stenglin (2004), bondicons are rallying devices, realised not only through language but also through other semiotic modes, for example flags, iconic buildings and people, which are used to crystallise strong interpersonal attitudes to ideational meanings. Thus, as far as famous quotations are concerned, the readers must essentially know where the original quote has come from if they are to appreciate the play, and as a bondicon, the quote has become interpersonal, charged up as a rallying resource (Stenglin, 2004). Also, a contextual shift has occurred, where these references are taken out of their 84 Chapter Four original context (as in a movie or a poem for example) and placed in a new one, in a multimodal print news story. A similar phenomenon occurs in the use of proper names and titles. It is also important at this stage in our analysis to consider the division of labour between the different modalities in image nuclear news stories, whether the work being done by the verbiage in the heading is also reflected in the image or not. The stories in Figures 4 and 6 (and full pages in Appendices 2 & 3) will be taken as examples of this. Figure 4: SMH, 22/02/2005, p.11. The heading in Figure 4 reads “Gold ahead, make my day”. The intertextual reference here is to the film Sudden Impact, starring Clint Eastwood, and to one of the most famous lines in the movie where Clint Eastwood, aka Dirty Harry, says, while pointing his famous magnum revolver at a youth, “Go ahead, make my day”. Being able to appreciate where this quotation comes from is further enhanced by the coupling of the heading with the image. The way that the represented participant is standing and holding a weapon is the exact mirror image of the stance of Dirty Harry in that scene in the movie, but also looks very similar to the way that Dirty Harry was portrayed in the posters advertising the movie (see Figure 5). Reconciling the Co-articulation of Meaning between Words and Pictures 85 Figure 5: Advertising material for the Dirty Harry movies. Thus, this coupling of image and heading works to enable the reader to draw on the intertextual reference to the Dirty Harry movies, and this occurs before requiring the reader to attend to the actual news story itself. Once the reader has attended to this meaning, and perhaps congratulated him/herself on knowing how this intertextual reference could be viewed as playful, he or she can also appreciate that the heading has also been slightly manipulated in that ‘Gold’ is used in place of ‘Go’. This enables the readers to shift their focus onto the news event at hand and by engaging with the caption to the side of the picture (which reveals that the person in the image is taking part in an “Olympic” event), they can then understand why the word ‘Gold’ has been used in this context. This may also further enhance the playful nature of the story. 86 Chapter Four Figure 6: SMH, 25/05/2005, p.14. In the text in Figure 6, however, the division of labour between the heading, image and caption works quite differently, albeit a similar kind of playful stance on the story is also ultimately achieved. Here, the heading “Cry freedom” may also evoke intertextual references to the movie of the same title, or to the works of Donald Woods, a South African journalist who wrote extensively on apartheid and in particular about Steve Biko, about whom the film Cry Freedom was written. The photograph, however, depicts unidentifiable (certainly in an Australian context that is) Caucasian represented participants running towards each other with highly emotional expressions on their faces. Unlike in the previous example, there is no intertextual reference in this photograph to the movie, Woods, or Biko, as we would have experienced such references again in the movie posters (see Figure 7) or in the movie itself. Instead, we are compelled to read the heading on a more literal level, by taking the emotion shown by the represented participants as equating to the ‘Cry’ in the heading, and their running action and gestures as leading them away from what appears to be official-looking persons in the background towards what we may surmise as being “freedom” and into the arms of their loved ones. It is only once we have engaged with the caption and realised that this news event is describing the release of Romanian journalists who had been held hostage in war-torn Iraq that we may then be able to fully appreciate and activate Reconciling the Co-articulation of Meaning between Words and Pictures 87 the meanings we may have originally associated with the movie ‘Cry Freedom’. Another method, and the one which is most commonly employed in creating a playful stance on the events in image nuclear news stories, is that which involves the use and manipulation of common idiomatic expressions in the heading, coupled with the image. Such expressions are often referred to as narrow idioms or as dead metaphors (Lakoff & Johnson, 1980; Chang, 2004) because they have become so conventionalised and specialised in their meaning that it is often difficult to establish what their origins may have been. The decoding of the play using such metaphorical expressions, then, remains very much within the linguistic knowledge of the reader. Figure 7: Advertising material for the movie ‘Cry Freedom’. I will now present an example of this in an image nuclear news story. In the story in Figure 8 (full page in Appendix D), the heading reads “Seeing red”. The idiom ‘see red’ means to “get very angry” (OERD 2003) and since this is the meaning that has been naturalised over time in our language, this is the meaning that we initially invoke when we encounter this attitudinally charged phrase. However, the words ‘see’ and ‘red’ can also be read as individual lexical items, or in combination as in ‘to see the colour red’. If we then combine this heading with the photograph of the Forth Bridge in the UK, lit up in red lights, it is on this latter meaning that we now focus our attention. In this photograph we can literally see something that is red in colour and the point at which this combination of heading and image becomes playful is when we are able to 88 Chapter Four hold both the figurative and literal meanings of ‘seeing red’ in our gaze at the same time. Thus, readers have to decide which of these meanings they attend to more. Initially, they would attend to the figurative meaning as this is the meaning that they have been apprenticed into within their culture. Then, in viewing the image and heading together, they can shift their gaze to the literal rendition of this phrase. Thus, it is this shifting back and forth between the literal and figurative that creates the play. In fact, this is how puns in English become humorous, by holding both the literal and figurative meaning in our gaze at the same time. Figure 8: An image nuclear news story, SMH, 23/02/2005, p.8. Thus far two distinct methods for creating a playful relationship between the headings and images in image nuclear news stories have been introduced: (1) employing intertextual references drawing on our cultural/popular/general knowledge, and (2) employing and/or manipulating common idiomatic expressions from within our linguistic framework. However, in every instance, the success of the play between the heading and image relies on our ability to hold two meanings in our gaze at the same time: that of the original context or sense of the wording and that of its redeployment in combination with a photograph in this particular news story genre. This is what I term twoness of meaning. Raskin (1985) describes certain jokes in a similar way as having an Reconciling the Co-articulation of Meaning between Words and Pictures 89 element that triggers a switch from one semantic script to another, with the switch making up the joke. If we are unable to hold both semantic scripts in our view at the same time then we are unlikely to find the joke humorous. We can reason about this twoness of meaning in another way, by now shifting our gaze back to SFL theory and re-examining instantiation and commitment theory. Each of the three example stories examined in this chapter require a certain amount of meaning potential to be recovered from the system in order to make meaning. This notion of recovering meaning potential from the system in terms of instantiation has been described by Martin (in press) as commitment, and the degree of commitment of meaning potential differs from one instance to the next. In terms of commitment, then, in image nuclear news stories we first commit to the more expected meaning in the wording of the heading, be it as a quote from a famous movie, or as an idiomatic expression. Then, there has to be a certain degree of de-commitment of meaning from the original source text, which has then to be re-committed in the news story text, and it is in the tension created between these different contexts and the meanings that they generate that the play between the heading and image is being created. Again, it is in our ability to hold both meanings/contexts in our gaze at the same time that the play is created. 5. Syndromes of meaning in image nuclear news stories Zappavigna (this volume) describes syndromes as clusters of associated signs where the relationship between the signs is the way in which they co-construe more complex meanings. In terms of instantiation, she further describes syndromes as “the recurrent co-instantiation of patterns of linguistic potential” (Zappavigna, this volume). In a particular text, or even across a large corpus then, there may be certain patterns that reveal themselves, patterns that have a tendency to occur more frequently than others. This may also be true of image nuclear news stories. In a corpus of 900 of these stories, the heading and image combine in a playful stance on the news event in approximately 95% of the instances (the remaining 5% do not engage in any form of play and deal with more serious hard news events), and this play is predominantly manifested in the manipulation of common idiomatic expressions (68%), in the recontextualisation of famous quotations (11%) and in the use of proper names or titles (10%). That such regular patterns are visible in these news stories has major implications for the reader of these texts and the extent to which they are expected to take up the challenge of decoding the play. 90 Chapter Four 6. Reading position in image nuclear news stories The extra stage added by Martin and Rose (2003) to the cline of instantiation is that of reading or subjectified meaning (refer to Figure 2 above). They suggest that as readers we are continually engaging with the readings that the texts themselves afford, and in so doing we tend to shuffle back and forth between the text and the reading, thus creating a kind of recursive loop at the end of the cline, until we have recovered the negotiation of meanings between the two. What those readings are can be looked at in terms of compliant or tactical responses to the text. In image nuclear news stories, it is the play between the heading and the image that we, as readers, are ultimately expected to resolve. The headline and image together make one meaning which one attends to several times over until one has unpacked all the play. Since 95% of these stories do attempt to create a pun, or play on words, at least between the heading and photograph, one may conclude that this genre naturalises the play, setting up a reading position from which the compliant, or obliging (Kitis & Milapides, 1996) reader is expected to figure out this play. A tactical reading would be anything that does not fall into what has been naturalised by the genre. So a tactical reader may only partially engage with the story, may go straight to the caption to find the news, ignore the play and thus not engage with this as an image nuclear news story, but rather just as a news story. However, one additional reading position may be necessary for describing this particular news story genre, and that is one where the reader does not “get” the play, and is thus excluded from the group that can “get” the play. Some readers may lack the general knowledge needed to get at certain intertextual references, or may not be a native English speaker and may thus be unable to recognise certain idiomatic expressions. Such a reader may also get some of the play only some of the time and may be frustrated by such stories at other times. Thus, we may need to introduce a choice between excluding or including the reader to a system for reading position before deciding whether one engages in a compliant or tactical reading of the text. One’s ability to engage with the play in image nuclear news stories has important implications for the newspaper’s ability to create a bond (Stenglin, 2004) with the reader, that is, a community of readers that share in the newspaper’s ideological stance on world events. It is my suggestion that through this deliberate manipulation of the discourse, the newspaper is able to express cultural and social solidarity with its readers, as it is assuming that readers share its understanding of the intertextual references being made in these texts. This has been labelled a kind of insiderism or Reconciling the Co-articulation of Meaning between Words and Pictures 91 membershipping device by theorists like Chang (2004), including some readers and excluding others. Furthermore, as far as the use of idiomatic expressions is concerned, McCarthy (1998) states that “idioms are communal tokens that enable speakers to express cultural and social solidarity” (p.145, cited in Chang, 2004: 76). In this way then, punning or wordplay serves to membership the readers into “belonging to a community with shared linguistic and cultural values” (Grauberg, 1989, cited in Chang, 2004: 105). Of course, an important effect of being able to form such a community of readers is that The Sydney Morning Herald can then articulate quite accurately the demographic that is its readership to advertisers. 7. Conclusion One of the effects of this playful relationship between the heading and image in image nuclear news stories is the intellectual challenge that such play affords the readers of The Sydney Morning Herald. Newspapers today face tough competition for reader loyalty. Other media platforms compete for the interests of media audiences and if the claims of the “attention economy” (Goldhaber, 1997, cited in Gauntlett, 2000: 9) are to be believed then they are also competing for less and less of our time. It is interesting to note that going very much against the trend of other Australian newspapers, the circulation of The Sydney Morning Herald has increased in recent years (Fairfax Media, 2007). Thus, by engaging the readers in this “linguistic gymnastics” that so clearly defines image nuclear news stories, The Sydney Morning Herald may just be doing enough to keep readers interested in the news and still buying the newspaper. References Caple, H. (forthcoming a). Intermodal relations in image nuclear news stories. In L. Unsworth (Ed.) Multimodal Semiotics: Functional Analysis in Contexts of Education. London: Continuum. —. (forthcoming b). What You See and What You Get: The Evolving Role of News Photographs in an Australian Broadsheet. In V. Rupar (Ed.) Journalism and Sense-making: Reading the Newspaper. Hampton Press. —. (to appear). Composition and the photographic image. Chang, C. (2004). English Idioms and Interpersonal Meaning. Guangzhou: Sun Yak-sen University Press. 92 Chapter Four Eggins, S. (1994). An Introduction to Systemic Functional Linguistics. London: Pinter. Fairfax Media (2007). Fairfax Media Publications Circulation And Readership Results For Newspapers, Inserted Magazines And Websites – To June 2007. Retrieved 21st October, 2007, from www.fxj.com.au. Gauntlett, D. (2000). Webstudies: rewiring media studies for the digital age. London: Arnold. Halliday, M.A.K. (1978). Language as Social Semiotic. London: Arnold. Halliday, M.A.K. & Matthiessen C.M.I.M. (2004). An Introduction to Functional Grammar, 3rd ed. London: Arnold. Kitis, E. & Milapides M. (1996). Read it and believe it: how metaphor constructs ideology in news discourse. A case study. Journal of Pragmatics 28: 557-590. Kress, G. & T. van Leeuwen (2006). Reading Images The Grammar of Visual Design, 2nd ed. London: Routledge. Lakoff, G. & M. Johnson (1980). Metaphors we live by. Chicago: University of Chicago Press. Martin, J.R. (2000). Beyond exchange: APPRAISAL systems in English. In S. Hunston & G. Thompson (eds.) Evaluation in Text: Authorial Stance and the Construction of Discourse (142-175). Oxford: OUP. —. (in press) Tenderness: realisation and individuation in a Botswanan town. In N. Nørgaard (Ed.) Odense Working Papers in Language and Communication (Special Issue of Papers from 34th International Systemic Functional Congress), Odense. Martin, J.R. & Rose D. (2003). Working with Discourse: meaning beyond the clause. London: Continuum. Martin, J.R. (2007). Intermodal reconciliation: mates in arms. Plenary address at the ASFLA Conference, 27-29 September, University of New England, Armidale. Matthiessen, C.M.I.M. (2005). The “architecture” of language according to systemic functional theory: developments since the 1970s. Department of Linguistics, Macquarie University. O’Toole, M. (1994). The Language of Displayed Art. London: Leicester University Press. Oxford English Reference Dictionary (OERD) (2nd ed.) (2003). Oxford: OUP. Raskin, V. (1985). Semantic mechanisms of humor. Dordrecht: D. Reidel. Stenglin, M. (2004). Packaging curiosities: Towards a grammar of three- dimensional space. Unpublished PhD Thesis, Department of Linguistics, University of Sydney. Wray, A. (2002). Formulaic language and the lexicon. Cambridge: CUP. Appendix A: Figure 1: SMH, 22/03/2005, p.3. Appendix B: Figure 4: SMH, 22/02/2005, p.11. APPENDICES Reconciling the Co-articulation of Meaning between Words and Pictures 93 94 Appendix C: Figure 6: SMH, 25/05/2005, p.14. Appendix D: Figure 8: SMH, 23/02/2005, p.8. Chapter Four CHAPTER FIVE “WHAT THE HELL IS WRONG WITH YOU?” A CORPUS PERSPECTIVE ON EVALUATION AND EMOTION IN CONTEMPORARY AMERICAN POP CULTURE MONIKA BEDNAREK 1. Introduction In this chapter I take a corpus linguistic perspective on the language of evaluation and emotion working with a 1.5 million word corpus of American TV dialogue (from the TV series Gilmore Girls). I am interested in occurrences of evaluative and emotional phrases such as the following: 1 y? LORELAI: Really. RORY: I don't believe this! Oh my God, I'm going to Chilton! 2 ce things were dirty . RORY: It's 7:16. LO RELAI: Oh my God, I wa s gonna wear my 3 ok this morning. Goo d morning, Jackson. S OOKIE: Oh my God, toda y was Rory's 4 s working. The phone is ringing.) LORELAI : Hey. Oh my God, you have good 5 s in my office if yo u'd like to see them. RORY: Oh my God, I to tally would. 6 is that? LANE: 12 calories. RORY: Here LANE: Oh my God bles s you! LANE: Man, 7 sit down. RORY: No I can't sit down. I'm too -- Oh my God, He k issed me! (Mrs. 8 Rich has this amazin g hair. RORY: Really? LANE: Oh my God, it's so perfect. It's 9 . (Emily takes a bit e and makes a face.) EMILY: Oh my God, it's horrible! What 10 ought we were meeti ng at Luke's. RORY: W e were? Oh my God, I'm so sorry. I 1 >LUKE: What the hell is going on? 2 >LUKE: What the hell is he terrified of? 3 CHRIS: What the hell are you doing here w ith my daughter? 4 LOGAN: What the hell is wrong with you? 5 >LUKE: What the hell is wrong with them? Why can't they just le t the bozo in 6 isbee! What the hell has gotten into you? LUKE: I told Rory we were e 7 RELAI: What the hell is that? 8 YPSY : What the hell are you doing here? 9 >LUKE: What the hell are we doing in a st eam room? As will be shown, phrases like Oh my God and (what) the hell are used as “implicit cues” (Culpeper, 2001, p.172) to characterisation in TV dialogue, and work as conventionalised realisations of emotionality. The investigation of evaluation and emotion is embedded in the discussion of a three-pronged approach to the analysis of dialogue. More specifically, I 96 Chapter Five argue for the necessity of combining a large-scale quantitative approach with a small-scale corpus analysis to be complemented by qualitative case studies (in this paper with the help of Martin and White’s (2005) Appraisal theory). This approach extends previous studies in linguistics that involve both corpus and discourse analysis, for instance corpus-based discourse analyses or Matthiessen’s (2006) “two-pronged approach” (Matthiessen 2006: 110). It allows us to see quantitative intertextual patterns (large- scale corpus analysis), semantic-pragmatic patterning (small-scale corpus analysis) as well as textual development (qualitative case study). Thus, a three-pronged approach to discourse analysis aims at providing a kaleidoscopic lens which fractures perception, and simultaneously shows us different aspects of emotionality in television dialogue. 2. Analysing TV dialogue 2.1 Why TV dialogue? Given that television is a popular medium and commonly looked down on in terms of its quality, why choose TV dialogue as data to analyse? On the one hand, the very popularity of television and TV series entails a huge influence of this dialogue in our daily lives. As Roman (2005) points out, “[t]elevision programs and characters have a unique ability to become an intimate part of a household and family” (p.130). This has been called the phenomenon of “parasocial interaction”–a “one-sided friendship television viewers have with a mass communication ‘persona’ or character” (Lewis, 1994, p.4). There are also reasons why it might be particularly interesting to look at realisations of emotionality in TV dialogue. No matter how emotions are defined, there is a debate whether they are biologically given or culturally construed. In a strong view, the sociologist Meštrovic has argued that “emotions have been McDonaldized, petrified, routinized, and otherwise made artificial” (Meštrovic, 1997, p.146). However, it has been suggested that emotions are not necessarily either wholly universal or wholly culturally determined, and that these two views might not be completely incompatible (Parrot & Harré 1996: 2, Planalp 1999: 195). Thus, while some aspects of emotion do appear biological/universal, emotional experience and our attitudes towards and beliefs about it are clearly influenced by culture. For instance, the cultural psychologists Kitayama and Markus (1994) state that “emotions are socially and culturally shaped and maintained, especially by means of collective "What the Hell is Wrong With You?" 97 knowledge that is represented in linguistic convention, everyday practice, and social structure” (p.10). It is in this respect that the analysis of linguistic emotionality in TV dialogue may well be worthwhile.1 2.2 The corpus: Why the Gilmore Girls? The corpus for this paper consists of 1.5 million words of fan transcripts of the American TV series Gilmore Girls, which ran for seven seasons from 2000 to 2007 (GiGi corpus). Note that the transcripts cover all episodes, consist of word-to-word dialogue as well as some descriptions of setting, action scenes and/or camera movements, and include names of speakers (e.g. Lorelai, Rory, Luke). The transcripts are available online and are characterised by multiple authorship, as is common in TV and film production. In order to allow the reader to follow the discussion of characterisation in Section 3.1.3 below more easily, here is a short description of what the Gilmore Girls is about: Set in a storybook Connecticut town (Stars Hollow) populated with an eclectic mix of everyday folks and lovable lunatics, GILMORE GIRLS is a humorous multigenerational series about friendship, family and the ties that bind. […] [T]he series revolves around thirtysomething Lorelai Gilmore and her college-age daughter, and best friend in the world, Rory. Lorelai has made her share of mistakes in life, but she has been doing her best to see that Rory doesn’t follow in her footsteps. That may be easier said than done, considering that the two share the same interests, the same intellect, the same coffee addiction and the same eyes. From the beginning, this unique mother-daughter team has been growing up together. Lorelai was just Rory’s age when she became pregnant and made the tough decision to raise her baby alone. This defiant move, along with Lorelai’s fiercely independent nature, caused a rift between her and her extremely proper, patrician, old-money parents, Emily and Richard. However, Lorelai was forced to reconcile with them when she found herself in desperate need of money for Rory’s tuition. Continuing to add to the unmistakable style of Stars Hollow is a colorful roster of town characters, including Lorelai’s best friend and business partner Sookie St. James, Miss Patty, the local dance teacher and social commentator, Michel Gerard, the haughty concierge of the Dragonfly Inn, and Kirk, the town’s jack-of-all-trades and master of none. Rory’s two best friends are her intense classmate Paris Geller and childhood pal Lane Kim, who just married fellow musician, Zack, in a typically quirky Stars Hollow celebration. (http://www.cwtv.com/shows/gilmore-girls/about; official website) 98 Chapter Five As becomes apparent, the Gilmore Girls is a drama/comedy hybrid in terms of TV genres as recognised by Roman (2005). There are several reasons for choosing dialogue from this series as data. Firstly, it is very contemporary; all episodes are available on DVD, and transcripts are available online for all seasons. Secondly, the series has been very successful in a variety of ways: Commercially, it has attracted as many as 5.2 million viewers in the US (season 2) and by season 5, it was the WB channel’s second most watched primetime show. It also showed a 120% increase in the cost of an average commercial from 2000 to 2003 (www.nielsenmedia.com). Culturally, the series has received many awards and award nominations, indicating that the series is recognised in the film and television industry. Internationally, the Gilmore Girls was/is broadcast in at least 52 different countries, and academically, papers have been presented on the series in research on popular culture, with a recent edited volume dedicated solely to the Gilmore Girls (Calvin, 2008). 3. A three-pronged analysis In order to approach the analysis of dialogue I propose a three-pronged analysis consisting of three different types of research: • Large-scale quantitative corpus analysis • Small-scale corpus analysis • Qualitative discourse analysis (case study) Each of these approaches has certain foci and advantages as outlined in Table 1 below (for more detail see Bednarek, 2008, Chapter 6): "What the Hell is Wrong With You?" 99 Table 1: A three-pronged approach (from Bednarek, 2008) Large-scale corpus Small-scale corpus Qualitative discourse analysis analysis analysis frequency, distribution, key, stance, functions logogenesis, socio- patterns, hypothesised of emotion terms, cultural analysis, co- general functions patterns with respect to articulated meanings co-articulated meanings most representative less representative least representative less detailed more detailed most detailed description, more description, more description, most descriptive interpretive interpretive Thus, a large-scale corpus analysis uses a large (more than 1 million words) corpus and computer software to investigate the frequency and distribution of lexico-grammatical items. If the corpus design is reasonable, it can provide a relatively representative sample of language that allows us to hypothesize about general functions of the respective lexico-grammatical items. On account of the large size of the corpus, however, the description is less detailed than would otherwise be possible. In contrast, a small-scale corpus analysis uses a corpus of a size that is amenable to manual analysis, e.g. of semantic or pragmatic meanings, and that still shows us intertextual patterning of such meanings. On account of its smaller size, such a corpus will be less representative, but its analysis will be more detailed and more interpretive. Finally, undertaking one or several case studies allows for qualitative discourse analysis, which can take into account textual development (logogenesis) and the socio-cultural context (e.g. of production and reception) of the discourses at hand. It will be least representative, in that researchers will be least able to generalise from findings but its analysis will be most detailed and most interpretive. The focus in this paper is on large-scale quantitative corpus analysis (Section 3.1) and on qualitative discourse analysis (Section 3.3), although some comments shall also be made on small-scale corpus analysis (Section 3.2). 3.1 Large-scale quantitative corpus analysis For the quantitative corpus analysis Scott’s (1998) Wordsmith software package was used. This software is widely used in corpus linguistics, and is a useful tool for the analysis of large-scale data that cannot be analysed 100 Chapter Five manually. It allows the user to produce frequency lists (lists of words and word clusters in the corpus – the WordList tool), concordances (occurrences for words or phrases – the Concord tool) and key words (words/phrases that are statistically unusual when comparing two corpora – the KeyWords tool). 3.1.1 Frequency lists Let us look first at frequency lists. Excluding names, the 20 most frequent words in GiGi are listed in Table 2: Table 2: 20 most frequent words in Gigi excluding names N Word Freq. 1 I 44,176 2 THE 41,839 3 YOU 41,266 4 TO 33,188 5 A 27,725 6 AND 26,317 7 IT 19,276 8 THAT 14,415 9 OF 13,990 10 IN 12,632 11 IS 12,414 12 WHAT 10,261 13 S 9,177 14 ME 9,090 15 ON 9,049 16 HAVE 8,717 17 THIS 8,633 18 FOR 8,582 19 NOT 8,463 20 SO 8,428 Although this frequency list tells us something about the data in itself, e.g. that it might be interesting to look in more detail at questions with what (No 12 in the list) or negation with not (No 19), we do not yet know which of these frequencies are particular to the Gilmore Girls and which reflect general language tendencies. In other words, if we analyse the corpus that we are interested in (what Scott and Tribble, 2006, p.58 call the “node- "What the Hell is Wrong With You?" 101 text”) we need another corpus that works as a standard of comparison or baseline to establish what is special about the node corpus (what is called the “reference” corpus by Scott & Tribble 2006, p.58). This reference corpus “should be an appropriate sample of the language which the text we are studying […] is written in” (Scott & Tribble, 2006, p.58). In the following I use wordlists from two reference corpora: the Santa Barbara Corpus (transcripts of words spoken American English) and the Longman Spoken American Corpus (ca 4,8 million words of spoken American English). It must be noted that the wordlists were compared manually rather than automatically, because the Santa Barbara Corpus (SB) is not big enough to act as a reference corpus in Wordsmith (according to research mentioned by Scott and Tribble (2006, p.65), a reference corpus should be around five times bigger than the node corpus), and because I had no direct access to the Longman Spoken American Corpus (LSAC) but had to work with wordlists that are available in Mittmann (2004).2 3.1.2 N-grams in the Gilmore Girls Because Mittmann (2004) does not provide word lists for individual word forms in LSAC, WordList was used to produce a list of the most frequent words in SB. Comparing the lists from GiGi and SB manually, it is interesting to observe that number one to ten in both lists are very similar in both corpora with more striking differences further below: • what: No12 GiGi vs. No 19 SB • me: No 14 GiGi vs. No 51 SB • not: No 19 GiGi vs. No 55 SB • on: No 15 GiGi vs. No 26 SB • for: No 18 GiGi vs. No 34 SB This confirms the conclusion drawn from looking at the frequency list of GiGi above, namely that it would be interesting to investigate what and not further. For example, when we look at 3-word clusters (or 3-grams) for what, we can see that the first evaluative/emotional cluster is What the hell at No 16 (55 occurrences). We will look at this phrase in more detail in Section 3.1.3. In order to compare clusters or n-grams (on clusters/n-grams/lexical bundles see e.g. Biber et al., 1999, Stubbs & Barth, 2003, Mittmann, 2004, Römer, 2008) wordlists from GiGi were compared with wordlists from the LSAC provided in Mittmann (2004). Starting with 2-grams, it is striking that there is still a great deal of overlap between GiGi and LSAC – eight of ten GiGi 2-grams are also among the ten most frequent 2-grams in LSAC, 102 Chapter Five with only are you and to the over-represented in GiGi. This seems to indicate that the spoken discourse in GiGi has been well designed to mirror “natural” spoken American English. The high frequency of to the in GiGi is perhaps the result of the descriptive passages, whereas the key phrase are you might point to the fact that GiGi contains more questions to interlocutor(s) than ordinary conversation. Using Concord for looking at clusters with are you gives us more information on the kind of questions that might be interesting to analyse further (Table 3): Table 3: Are you clusters N cluster Freq. 1 what are you 761 2 are you doing 471 3 are you talking 175 4 you talking about 169 5 why are you 167 6 are you going 165 7 are you sure 159 8 how are you 159 9 where are you 156 10 you doing here 143 Some of these shall be investigated further in connection with 3- grams. Comparing 3-grams in GiGi and in LSAC still shows some degree of overlap: ten of the 15 most frequent 3-grams in GiGi are also among the 15 most frequent 3-grams in the LSAC. Significantly, I don’t know is the most frequent 3-gram in both corpora (and, incidentally, also in the SB). What are you, do you think, out of the, are you doing and oh my god are not among the 15 most frequent 3-grams in the LSAC. However, what are you is number 30 in the LSAC, do you think is number 32 and out of the occurs 60 times (out of 270 total occurrences) in descriptive passages. This leaves us with the more interesting remaining 3-grams are you doing and oh my god. Of the occurrences of are you doing in GiGi the majority (203 occurrences) are in the form what … are you doing …? These can be sub- divided into three major groups: • locative (167): o What are you doing here? (138) o What are you doing Location (29) "What the Hell is Wrong With You?" 103 • evaluative (17): o What are you doing X-ing (7) o What are you doing with X (6) o What are you doing, Vocative (4) • temporal (16): o What are you doing TIME As becomes apparent, the great majority of these questions are locative, especially in the form What are you doing here? indicating character surprise at the presence of other characters. Interestingly, it achieves this without using any explicitly evaluative or emotional language. The evaluative instances can probably be classified as “demands for explanation” (Spitz, 2005, p.316)–instances of a speech act that belongs to speakers’ “argumentative resources” (Spitz, 2005, p.245). The evaluative phrase What are you doing X-ing, which is grammatically “unusual”, is particularly interesting in as far as the whole phrase what … are you doing works to evaluate the following non-finite clause negatively, as shown by some examples: and what are you doing agreeing to come over here all the time? what the hell are you doing calling Christopher what are you doing talking to him about stuff like that what are you doing yelling ‘Fire’ what are you doing telling my daughter to lie As for oh my god, as we will see later, this is also frequently associated with surprise, confirming the overall significance of this emotion in Gilmore Girls. This may be the result of the putative intent of a TV series’ creators to capture the interest of viewers by having unexpected things happen to characters in the series. Skipping 4- to 6-grams, and moving on to 7-grams in Gilmore Girls, the most important ones in GiGi are:3 • can I talk to you for a (13) – followed by either sec/second/minute or moment • I don’t want to talk about it (8) • I don’t want to talk about this (8) • I wanted to talk to you about (8) • What do you want me to do (8) 104 Chapter Five Two points can be made with respect to this. Firstly, talking emerges as a central action in GiGi, where the normal expectation is that issues and problems are talked about and “talked through”, though characters may refuse to obey this expectation. This might be related to the character of Gilmore Girls as a “female” drama – reflecting social action that is (perhaps stereotypically) associated with women. Additional 7-grams with talk which occur further below in the wordlist confirm this: I talk to you for a sec (6); don’t want to talk about this anymore (5); have to talk to you about something (5); I have to talk to you about (5); I need to talk to you about (5); I want to talk to you about (5); I don’t know what you’re talking about (5)). Secondly, these phrases point to confrontational or at least problematic issues, in particular: I don’t want to talk about it (8) I don’t want to talk about this (8) I don’t know what you’re talking about (5) Further, I wanted to talk to you about; I have to talk to you about; I need to talk to you about often seem to be used to introduce confrontational or problematic issues, such as in the following examples: I wanted to talk to you about Rory and this ridiculous accusation of her being a loner and how that’s somehow something bad I have to talk to you about how it’s all feeling wrong I need to talk to you about something serious Again, this seems to be the result of what we might call a “dramatic” element in Gilmore Girls that is used to capture the viewers’ interest. Other 7-grams reflect the emotionality in the series, which fulfils similar functions as well as construing character relations (I just want you to be happy (7); If it makes you feel any better (7); What the hell is wrong with you (6)). In order to investigate n-grams in Gilmore Girls in more detail, further analysis is necessary, which takes into account context and usage. However, the kind of analysis that has been presented so far is important in as far as it points the researcher to where s/he might need to look more closely. "What the Hell is Wrong With You?" 105 3.1.3 Evaluation and emotion in the Gilmore Girls Before looking more closely at linguistic realisations of evaluation and emotion in Gilmore Girls, it is important to point out that my interest is not in emotion or evaluation as mental processes as such. Rather, the focus is on the linguistics of emotion and evaluation, and, more specifically, evaluative and emotional phrases. Further, the occurrence of such phrases is not necessarily a reflection of some inner state of mind; rather, these phrases work as conventionalised signs. The following analysis is limited to a) expletive interjections and b) the evaluative phrase What the hell is wrong with you. Expletive interjections Following the finding of the importance of oh my god in the analysis of 3-grams (Section 3.1.2 above), GiGi was searched for occurrences of a selection of common expletives (compare e.g. Ivarsson Ahlin, 2006) used as interjections. These are: Jesus Christ geez (including oh geez etc) hell damn damn it shit fuck shoot yuck god oh god for the love of god dear god good god (oh) thank god for X’s sake (for heaven’s sake, for Pete’s sake, for God’s sake, for goodness sake) In Ameka’s (1992) terms, these include both primary interjections (yuck), secondary interjections (damn) and interjectional phrases (oh thank god; for the love of god), all of which are emotive rather than cognitive. Expletives have been linked to affect or emotion in a variety of approaches in linguistics (Bednarek, 2008, Chapter 1). For instance, Martin and White (2005) talk about swearing as construing “emotional outbursts” (p.68). And interjections have been defined as “relatively conventionalised vocal gestures […] which express a speaker’s mental state, action or attitude or reaction to a situation (Ameka 1992, p.106). Importantly, “they are all produced in reaction to a linguistic or extra-linguistic context, and can 106 Chapter Five only be interpreted relative to the context in which they are produced” (Ameka, 1992, p.108). For example, looking at occurrences of oh my god in its context in GiGi (not reproduced here for copyright reasons), this can be associated with emotions of • positive or negative surprise; • annoyance/exasperation, panic, disgust; • pleasure/happiness, admiration. There are also other, more bleached meanings where oh my god indicates involvement or emotionality, or marks a sudden realisation. This means that the meaning of interjections becomes clear only by looking at the context and gesture/facial expression/tone of voice etc. Counting expletive interjections will thus only tell us about emotionality/ involvement but not about the particular emotions involved. Nevertheless, a corpus analysis does show a number of interesting things, which I will only briefly summarise here: 1) There are no occurrences for (Jesus) Christ, shit or fuck, although there is one humorous occurrence of Jesus, Mary, Joseph and a camel. This reflects the “family-friendly” character of the show which was funded by the Family Friendly Programming Forum – a clear influence of external factors. In this sense, no unbiased reflection or construal of reality is possible in GiGi. 2) There are 1009 occurrences of expletive interjections uttered by female characters and 239 occurrences uttered by male characters. Even taking into account the number of turns of the six main female and the ten main male characters,4 this means that female characters use more exclamatory expletives than male characters. This is interesting because, as Precht has shown using a corpus of 1 million words of spoken American conversation, “men’s expletive frequencies were significantly higher than women’s” (Precht, 2006, p.24). Since such exclamatory expletives signal emotionality, the portrayal of men in Gilmore Girls mirrors the stereotypical conception of men as less “emotional” than women at least with regards to expletive interjections. 3) There are differences in terms of which expletives are preferred by women and which are preferred by men. Table 4 below lists the most frequent expletives, respectively (% > 5.5): "What the Hell is Wrong With You?" 107 Table 4: Expletive interjections Female Raw % Male Raw % oh my god 462 45.8 oh my god 42 17.6 god 178 17.6 geez 40 16.7 oh god 88 8.7 god 35 14.6 (oh) thank god 76 7.5 for X's sake 23 9.6 geez 59 5.8 (oh) thank god 20 8.4 my god 19 7.9 damn 15 6.3 damn it 14 5.9 oh god 14 5.9 Other 146 14.5 Other 17 7.1 Total 1009 100 Total 239 100 Table 4 shows that male characters in GiGi are more varied than female characters in their usage of the most frequent expletives (% > 5.5) as shown by the presence of the expletives for X’s sake, my god, damn, damn it in addition to oh my god, god, oh god, (oh) thank god and geez. Further, if we compare speakers of individual expletives, some interjections seem more “male”, others more “female” (Figure 1). Figure 1 suggests that oh my god/god is perhaps more female, whereas geez, damn and damn it (perhaps also for X’s sake/my god) are more male. This partially reflects Precht’s (2006) findings that damn is significantly higher in men and god significantly higher in women in American conversation (p.25). It also confirms Ivarsson Ahlin’s (2006) analysis of movie dialogue, which also found that oh my god was the most frequent expletive used by female characters, though on the basis of only very low frequencies. 108 Chapter Five Figure 1: “Male” and “Female” interjections 4) Conversely, we can say that those female characters who do use the “strong” or “male” expletives geez, damn/damn it and hell stand out or that the situation where they use these expletives is marked as particularly emotional. Compare the explicit description of a situation as markedly emotional in the following extract: LORELAI [stunned]: Rory called. LUKE: I know! She called and yelled at me. LORELAI: No, she called and yelled at me! LUKE: Yeah, but I’m the one who had to hear it, and she was loud! And she said – ‘hell’. I never heard her say ‘hell’. I didn’t even know she knew how to say ‘hell’. [Flustered.] She was mad and she yelled and she said ‘hell’. (Gilmore Girls, extract from series 6, episode 7, “Twenty-one is the loneliest number”) Paris is one of the female characters who does use “male” expletives; she has both damn it and hell among her three most preferred expletives, and it is part of her character that she is harsh, insensitive, undiplomatical and blunt. Her usage of expletives clearly contributes to this construal. Both Lorelai and Rory also use geez, which aligns them with both Luke and "What the Hell is Wrong With You?" 109 Jess who are the most prolific geez users among the male characters. 5) This points to the fact that similar characters are related by their usage of expletives. In fact, for both Lorelai and Rory (mother-daughter) oh my god and god are preferred expletive interjections; both Luke and Jess (uncle-nephew) use geez most frequently; both Richard and Emily (husband-wife) like to use for X’s sake. 6) It is also interesting to look more closely at characters’ non-use of exclamatory expletives, which also seems very much part of their character construal. For example, Richard and Emily hardly ever use expletives such as geez or yuck, which are too “casual” (and perhaps “young”) for them. And Rory tends not to use stronger expletives, in line with her construal as slightly innocent and very studious (compare the ‘hell’ example above). In fact, a few characters have “signature” expletives; i.e. they use certain expletives most frequently of all characters (taking into account turns): • Emily: for X’s sake • Jess: geez • Paris: hell, damn it • Kirk: damn, damn it • Michel: (oh) thank god • Sookie: oh my god • Chris: my god • Logan: hell • Jackson: shoot • Doyle: god Summing up these findings, interjections are clearly part of surge features that work as implicit cues to characterisation (Culpeper (2001, p.190 and research cited there). These, he says, “are frequently used by authors […] as a conventional way of signalling that a character has a particular emotion or attitude” (Culpeper, 2001, p.192-193). They are part of the construal of individual personae in the series.5 What the hell is wrong with you? Moving on to the evaluative phrase What the hell is wrong with you, the decision to look more closely at this phrase results from the importance 110 Chapter Five of what and the occurrence of what the hell as cluster found in Section 3.1.2 above. Regarding occurrences for What the hell…? itself we find the following important clusters: • am I/are they/we/you/X doing (17) • is/was that/this (14) • is going on/happening (11) • is wrong with you (them) (8) I shall discuss only the last phrase, as this is the only one including both emotional and evaluative components: What the hell and wrong with you. This phrase also occurred in the list of 7-grams described in Section 3.1.2. In terms of the two components, it seems that what the hell works both to intensify the negative evaluation that is already present in What is wrong with you (increasing the force of the utterance), and to generally signal the speaker’s emotionality or involvement. The phrase what is wrong with you is interesting semantically. For instance, it is not listed in the Oxford Advanced Learner’s Dictionary (OALD), even though it has no equivalents in languages such as German and French. Wierzbicka (2006) in fact dedicates a whole chapter on analysing the cultural embeddedness of right and wrong in English, though she focuses mainly on right rather than wrong. Admittedly, the OALD does list examples such as What’s wrong with eating meat (defined as “not morally right”), and There’s something wrong with the printer (defined as “causing problems”) but these arguably do not have the exact same meaning. Firstly, it seems that it is the whole phrase What is wrong with you that carries meaning, and it is, for instance, significant that the contracted form is not used. Secondly, the phrase is very complex in providing both an evaluation of someone’s action/behaviour as inappropriate and of the person who acted that way as not “functioning properly”, as it were. Not only are speakers saying that they disapprove of some (unspecified) action that has to be retrieved from the context; they are also saying that the action is the result of “something being wrong” (causing problems) with the person. This particular phrase hence seems to span the meanings of “causing problems” and “not morally right”, in that some “fault” in a person is said to cause them to act in a certain (bad) way. This is interesting to discuss with respect to Martin and White’s Appraisal theory, which will also be relevant in the qualitative analysis in Section 3.3 below. Starting from the assumption that texts position and re-position readers evaluatively in certain ways, Appraisal researchers focus on what they call appraisal resources (the linguistic devices of expressing appraisal). One Appraisal system is Attitude, involving “resources for […] "What the Hell is Wrong With You?" 111 reacting emotionally (affect), judging morally (judgement) and evaluating aesthetically (appreciation)” (Martin, 1995, p.28). Attitude thus has three sub-systems: Affect, Appreciation and Judgement. Affect systems characterise phenomena by reference to emotion: Affect (emotion): I’m happy, She’s proud of her achievements, he's frightened of spiders Appreciation systems include resources used to evaluate the (aesthetic) quality of processes and products (and human beings when they are seen as entities, and when they are not evaluated in terms of their emotions or behaviour): Appreciation (aesthetics: evaluating text/process/phenomenon): It’s a fantastic book (cf. Martin, 2000, p.145f) Both Affect and Appreciation have further sub-categories (Martin & White, 2005), which, however, will not be applied in this paper. Judgement systems consist of resources for morally evaluating human behaviour, by reference to a set of norms: Judgement (ethics: evaluating behaviour): a brutal tyrant, a skilful performer, don’t be cruel Judgement is subdivided in two broad categories: judgements of social esteem (normality, capacity, tenacity) and judgements of social sanction (veracity, propriety). Thus, in terms of Judgement subcategories, What … is wrong with you is an evaluation of Capacity that works to evaluate retrospectively the Propriety of someone’s behaviour, and it would need to be doubly classified. This adds further weights to arguments made in Bednarek (forthcoming) that we need a topological perspective on interpersonal meaning to deal with phenomena such as appraisal blends within categories like Judgement. 112 Chapter Five 3.2 Small-scale corpus linguistics The second perspective on the analysis of discourse is that gained from small-scale corpus linguistics. For reasons of scope this will not be demonstrated in this paper, but the methodology has been applied in both Bednarek (2006a) and Bednarek (2008). It refers to the analysis of small- scale corpora – ideally between 70,000 and 100,000 words – corpora that are small enough for manual, context-sensitive analysis but large enough to show at least some patterns and allow some generalisibility. In analogy to large-scale corpus research, a distinction can be made between text- based and text-driven studies (Bednarek, 2006b). 3.3 Qualitative Analysis The third perspective on the analysis of discourse is that provided by qualitative research–the close-reading type of analysis favoured by discourse analysts. This allows researchers to take into account textual development, or logogenesis (Halliday & Matthiessen, 2004, p.43) and social context. Consider this extract of conflict talk from Gilmore Girls: 1. EMILY: Lorelai. 2. LORELAI: Mom, hello. 3. EMILY: I wanted to talk to you about the Christmas dinner this Friday. 4. LORELAI: Ah, Christmas dinner 5. EMILY: You forgot. 6. LORELAI: Well mom, there’s been a lot going on around here lately, your Christmas shindig’s not exactly high on my list of things to obsess about. 7. EMILY: Well I’m sorry if the timing is bad, Lorelai, but the world doesn’t always revolve around you. 8. LORELAI: Well thanks for the tip 9. EMILY: Cocktails are at 6, dinner’s at 8. 10. LORELAI: I probably won’t be there for cocktails. 11. EMILY: Why not? 12. LORELAI: Because I have to work. 13. EMILY: You can’t leave work early? 14. LORELAI: No I can’t. 15. EMILY: Why not? 16. LORELAI: Because it’s not in my job description. 17. EMILY: Well then don’t come. 18. LORELAI: What? 19. EMILY: Don’t come. It’s obviously an enormous burden for you. 20. LORELAI: Yeah but - 21. EMILY: Just send Rory. 22. LORELAI: You’re telling me not to come to the Christmas party? "What the Hell is Wrong With You?" 113 23. EMILY: Well you’re obviously too busy. 24. EMILY: I had the German measles in the 5th grade, I still had to show up to the Christmas party. 25. LORELAI: Lorelai let’s be honest here, I’m not too happy with you right now and I assume you’re not too happy with me. 26. LORELAI: My polka dot dress matched my face and still I had to sit through 12 courses. 27. EMILY: I am tired of forcing you to do all those terrible things that infringe upon your life and I do not have the energy to pretend that the way you treated my the other day was in any way acceptable. 28. LORELAI: So you’re uninviting me to Christmas dinner? 29. EMILY: Yes I am. 30. LORELAI: Fine. 31. EMILY: Fine. 32. LORELAI: Ok, anything else? 33. EMILY: I believe that’s all. 34. LORELAI: Ok well, great mom, it’s been swell talking to you. 35. EMILY: Bye Lorelai. 36. LORELAI: Bye. (Gilmore Girls, extract from series 1, episode 10, “Forgiveness and Stuff”) While this text does not exhibit the prototypical argument format as identified in previous linguistic research (e.g. Spitz 2005), and while this does not really represent an argument about a specific topic, the presence of other features signals that this is indeed conflict talk. For instance, we can find the use of linguistic features that research has shown to occur in arguments. There is an interruption, which has been “related to the accomplishment of aggravated opposition” (Spitz, 2005, p.162), there are dissent markers (well) that “emphasise the oppositional character of […] turns” (Spitz, 2005, p.160), there is “contrastive mirroring” (Spitz, 2005, p.201) (fine – fine), and there is not much mitigation. However, in the following sections the focus is on typical speech acts and Attitude (the complete analysis for both is provided in the Appendix). The following speech acts occur in this extract: Greeting and statement (especially at the beginning and end) Accusation (5, 19, 23, 24, 26, 27) Disqualification (6, 7, 24, 26, 27) Explanation (12, 16, 19, 23) Demand for explanation (11, 15, 18) Demand for clarification (22, 28) Directive (17, 19, 21) Justification (6) 114 Chapter Five Insincere apology (7) Insincere thanks (8) Clarification (29) Contradiction (14) possible Counter-claim (20) (compare Spitz 2005 for definitions of some of these) Of these, Accusation, Disqualification, Demand for explanation, Directive, Contradiction and Counter-claim have been classified as “argumentative speech act(ion)s” (Spitz, 2005, p.245). Accusations and Disqualifications are particularly frequent in the extract. Interestingly, we can find hearer speech acts that seem a preferred response to speaker speech acts, but the resulting adjacency pairs (Accusation – Justification or Demand for Explanation – Explanation) are nevertheless confronting, since the second pair part involves a “negative” speech act. Thus, a Justification may be done via a Disqualification or an Explanation may be done via an Accusation: 5. E You forgot. Accusation 6. L Well mom, there’s been a lot Justification / Disqualification going on around here lately, your Christmas shindig’s not exactly high on my list of things to obsess about. 18. L What? Demand for explanation 19. E Don’t come. Directive It’s obviously an enormous Explanation / Accusation burden for you. 22. L You’re telling me not to come to Demand for clarification the Christmas party? 23. E Well you’re obviously too busy. Explanation /Accusation An Appraisal analysis also shows that this is conflict talk. Looking at all Attitude values in terms of Martin and White’s (2005) classification, a majority involves negative Judgement followed by negative Appreciation. Judgement is used by Emily and Lorelai to evaluate each other’s behaviour negatively, whereas Appreciation concerns evaluations of the Christmas dinner. In other words, negative Judgement and Appreciation "What the Hell is Wrong With You?" 115 are associated with the speech acts of Accusation and Disqualification introduced above. There is only some Affect (wanted to, not too happy, tired of; conventionalised: I’m sorry, thanks) and the only instance of positive Judgement (let’s be honest here) arguably intensifies the force of the following proposition rather than evaluating Emily’s behaviour positively. Interestingly, there are many embedded attitudinal appraisals, for instance those presented in Table 5 below: Table 5: Embedded appraisals Speaker Text Appraisal Appraiser Appraised Category Emily I am I am tired E Lorelai neg (E) tired of of forcing Judgement forcing you to do you to all those do all terrible those things that terrible infringe things upon your that life infringe I am E forcing … neg Affect upon tired of your life your life forcing E action neg Judgement terrible Lorelai things neg (attributed Appreciation infringe by E) things neg Appreciation This example is notable in that the complete clause (I am tired of … life) works to express a negative Judgement of Lorelai on the part of Emily. Looking at the embedded negative Judgement and Appreciations (forcing, terrible, infringe) it becomes clear that Emily is being ironic and implicitly attributes some of these evaluations to Lorelai (ie, that only Lorelai thinks these things are terrible and infringe upon her life). This implicit attribution in itself contributes to the negative evaluation of Lorelai by accusing her of having negative attitudes towards what should not be evaluated negatively according to Emily. The fact that Emily has to force Lorelai (with forcing an arguably negative action) reflects on Lorelai rather than Emily. This is reinforced by negative Affect, with forcing as a tiring action for Emily (tired of is negative Affect: displeasure). All in all, 116 Chapter Five the complex evaluations suggest that it is Lorelai who is to blame, not Emily. Such implicit attributions and irony are difficult to capture in anything other than a qualitative discourse analysis. Other aspects that are of note in this extract include the use of deontic modality, or modulation (obligation and readiness: have to, had to, can’t) and the use of irony in Lorelai’s utterances which turns positive evaluations into negative ones (thanks for the tip, fine; great mum, it’s been swell talking to you). In conjunction with other argumentative resources (as mentioned above) and paralinguistic features not investigated (volume increase, tone, stress, emphatic intonation), speech acts and appraisal thus clearly contextualise this interaction as conflict talk. 4. Concluding remarks This study showed how linguistic features can work as implicit cues to characterisation and also demonstrated the importance of conflict, drama and surprise in an American TV series. It also suggested that a corpus perspective on evaluation and emotion in popular culture can provide a useful input to studying TV dialogue, showing a way into the more detailed analysis of selected features. For example, the importance of the 3-gram are you doing resulted in an analysis of the larger patterns in which this 3-gram was embedded and in the identification of a longer phrase (What are you doing here?) that indicates character emotion (surprise) without any explicitly evaluative or emotional language. Further, the frequency of the 3-gram oh my god took us into a more detailed investigation of expletive interjections in general in terms of character and relationship construal. Finally, the importance of what and, more specifically, of the cluster what the hell resulted in a decision to look more closely at the evaluative phrase What the hell is wrong with you in terms of its evaluative and emotional meanings. The investigation of (argumentative) speech acts and Appraisal in a selected case study completed the study of evaluation and emotion in the Gilmore Girls and showed that more intricate features (e.g. irony) become apparent only in a qualitative analysis. Such analysis cannot yet be automated reliably (how can we identify all relevant speech acts and Appraisal instances in a large corpus?) and can only be undertaken for selected corpus extracts. However, a semi-automated study of semantic-pragmatic features would be feasible in a small-scale corpus, allowing us to see intertextual patterning of such features and providing a third perspective on the construal of emotionality and evaluation in discourse. The different perspectives that become apparent through these three methodologies "What the Hell is Wrong With You?" 117 extend previous discourse studies, and are the reason why I argue for a three-pronged approach to the analysis of TV dialogue and to the analysis of discourse in general. Notes 1. It is also worth noting that TV and film dialogue can have a significant influence on learners of English in non-English speaking countries, who may buy the DVDs and watch the original versions. For such learners, “films [and TV series, M.B.] may be one of the best opportunities to hear a foreign language spoken. This also means that film [and TV, M.B.] language becomes an influential model for advanced learners of English” (Mittmann, 2006). 2. For a comparison of dialogue in Friends, Golden Girls and Dawson’s Creek and everyday spoken English see Mittmann (2006). 3. 7-grams from GiGi could not be compared to lists of 7-grams from either the LSAC or the SB because they are not provided by Mittmann (2004), and because the SB is too small to yield many occurrences of 7-grams. Frequency is inverse to cluster size: as cluster size goes up, the raw frequency of occurrence goes down, but differences between corpora appear to become more pronounced. 4. Female: Lorelai, Rory, Emily, Sookie, Lane and Paris; male: Luke, Richard, Logan, Dean, Jess, Christopher, Kirk, Michel, Zach and Jackson (the result of a Concord search for the name in capitals followed by a colon, as this is how speakers are marked in the corpus). 5. It is interesting to compare the findings for expletives in Gilmore Girls with findings by Mittmann (2006) for the American TV series Friends, Golden Girls and Dawson’s Creek. She notes that god and hell occur frequently in these TV series (in contrast to the f-word ), with god “[taking] over the functions of many other swear words. It is interesting to note in this context that in the films the word hell occurs predominantly as part of the sequence WH- + the hell (mostly what/who/how the hell).” The latter finding is also confirmed by the analysis of What the hell below, which may point to these features being characteristic of TV dialogue in general (not just dialogue in the Gilmore Girls). References Ameka, F. (1992). Interjections: the universal yet neglected part of speech. Journal of Pragmatics 18: 101-118. Bahktin, M. (1981). Discourse in the novel. In M. Holquist (Ed). The Dialogic Imagination (259-422). Austin: University of Texas Press. Bednarek, M. (2006a). Evaluation in Media Discourse. Analysis of a Newspaper Corpus. London/New York: Continuum. —. (2006b). Epistemological positioning and evidentiality in English news discourse – a text-driven approach. Text and talk 26(6): 635-660. 118 Chapter Five —. (forthcoming). Polyphony in Appraisal: typological and topological perspectives. Linguistics and The Human Sciences. —. (2008). Emotion Talk across Corpora. Basingstoke/New York: Palgrave Macmillan. Biber, D., Johansson, S., Leech, G., Conrad, S. & Finegan, E. 1999. Longman Grammar of Spoken and Written English. London: Longman. Bubel, C. (2006). The Linguistic Construction of Character Relations in TV Drama: Doing Friendship in Sex and the City. Unpublished PhD dissertation, Universität des Saarlandes, Saarbrücken, Germany. Calvin, R. (2008) (Ed). Gilmore Girls and the Politics of Identity: Essays on Family and Feminism in the Television Series. Jefferson: McFarland and Company. Culpeper, J. (2001). Language and Characterisation. People in Plays and Other Texts. Harlow etc: Longman. Goffman, E. (1974). Frame Analysis. An Essay on the Organization of Experience. London: Harper and Row. Halliday, M.A.K. & Matthiessen, C. M.I.M. (2004). An Introduction to Functional Grammar, 3rd ed. London: Arnold. Ivarsson Ahlin, M. (2006). Female voices in horror: a linguistic study of female stereotyping in two slasher movies. Unpublished thesis, Växjö University (Sweden). Available online from http://www.diva- portal.org/vxu/abstract.xsql?dbid=621 Kitayama, S. (1994). (Ed). Emotion and Culture. Empirical Studies of Mutual Influences. Washington, DC: American Psychological Association. Kitayama, S. & Markus, H. R. (1994). Introduction to cultural psychology and emotion research. In S. Kitayama (Ed.) Emotion and Culture. Empirical Studies of Mutual Influences (1-19). Washington, DC: American Psychological Association. Lewis, F. (1994). Getting By: Race and Parasocial Interaction in a Television Situation Comedy. Unpublished PhD dissertation, University of Kentucky. Available at http://www.dissertation.com/book.php?method=ISBN&book=1581120 59. Martin, J.R. (1995). Reading positions/positioning readers: judgement in English. Prospect: a Journal of Australian TESOL 10: 27-37. —. (2000). Beyond exchange: appraisal systems in English. In S. Hunston & G. Thompson (Eds.) Evaluation in Text: Authorial Stance and the Construction of Discourse (142-175). Oxford: OUP. "What the Hell is Wrong With You?" 119 Martin, J.R. & P.R.R. White (2005). The Language of Evaluation. Appraisal in English. Basingstoke/New York: Palgrave Macmillan. Matthiessen, C.M.I.M. (2006). Frequency profiles of some basic grammatical systems: an interim report. In S. Hunston & G. Thompson (Eds.) System and Corpus: Exploring Connections (103-142). London: Equinox. Meštrović, S. (1997). Postemotional Society. London: Sage. Mittmann, B. (2004). Mehrwort-Cluster in der englischen Alltagskonversation. Unterschiede zwischen britischem und amerikanischem gesprochenen Englisch als Indikatoren für den präfabrizierten Charakter der Sprache. Tübingen: Gunter Narr. —. (2006). With a little help from Friends (and others): Lexico-pragmatic characteristics of original and dubbed film dialogue. In C. Houswitschka, G. Knappe & A. Müller (Eds.) Anglistentag 2005, Bamberg–Proceedings (573-585). Trier: WVT. Parrott, W. G. & Harré, R. (1996). Introduction: some complexities in the study of emotions. In R. Harré & W. Parrott (Eds.) The Emotions. Social, Cultural and Biological Dimensions (1-20). London etc: Sage. Planalp, S. (1999). Communicating Emotion. Social, Moral and Cultural Processes. Cambridge: CUP. Precht, K. (2006). Gender differences and similarities in stance in informal American conversation. Unpublished manuscript, Kent State University. Available at www.kprecht.net/MyWork/Precht%20Gender%202006.pdf Römer, U. (2008). Identification impossible? A corpus approach to realisations of evaluative meaning in academic writing. Functions of Language [special issue on evaluation and text types] 15(1), p.115- 130. Roman, J. (2005). From Daytime to PrimeTime. The History of American Television Programs. Westport, Connecticut/London: Greenwood Press. Scott, M. (1998). Wordsmith Tools. Oxford: OUP. Scott, M. & Tribble, C. (2006). Textual Patterns. Key Words and Corpus Analysis in Language Education. Amsterdam/Philadelphia: Benjamins. Spitz, A. (2005). Power Plays: the Representation of Mother-Daughter Disputes in Contemporary Plays by Women. A Study in Discourse Analysis. Unpublished PhD dissertation, Universität des Saarlandes. Available to download at http://scidok.sulb.uni-saarland.de/volltexte/2006/595/. 120 Chapter Five Stubbs, M., & Barth, I. (2003). Using recurrent phrases as text-type discriminators: A quantitative method and some findings. Functions of Language 10(1): 61-104. Thompson, G. & Hunston, S. (2006). Introduction: system and corpus: two traditions with a common ground. In S. Hunston & G. Thompson (Eds). System and Corpus. Exploring Connections (1-14). London: Equinox. Tognini-Bonelli, E. (2001). Corpus Linguistics at Work. Amsterdam: John Benjamins. White, P.R.R. (2000). Dialogue and inter-subjectivity: reinterpreting the semantics of modality and hedging. In M. Coulthard, J. Cotterill & F. Rock (Eds.) Dialogue Analysis VII: Working with Dialogue (67-80). Tübingen: Niemeyer. —. (2001). Appraisal outline. Manuscript (word processor version). www.grammatics.com/appraisal. Accessed July 1, 2004. —. (2002). Appraisal. In J. Verschueren, J. Östman, J. Blommaert & C. Bulcaen (Eds.) 1995-2002. Handbook of Pragmatics (1-27). Amsterdam/Philadelphia: John Benjamins. Wierzbicka, A. (2006). English. Meaning and Culture. Oxford: OUP. www.nielsenmedia.com (Nielsen Media Research – reports on advertising and ratings) "What the Hell is Wrong With You?" 121 APPENDIX Speech acts 1 E Lorelai. Greeting 2 L Mom, hello. Greeting 3 E I wanted to talk to you about the Statement Christmas dinner this Friday. 4 L Ah, Christmas dinner. Statement (repetition) 5 E You forgot. Accusation 6 L Well mom, there’s been a lot Justification/Disqualification going on around here lately, your Christmas shindig’s not exactly high on my list of things to obsess about. 7 E Well I’m sorry if the timing is Insincere bad, Lorelai, but the world doesn’t Apology/Disqualification always revolve around you. 8 L Well thanks for the tip. Insincere thanks 9 E Cocktails are at 6, dinner’s at 8. Statement 10 L I probably won’t be there for Statement cocktails. 11 E Why not? Demand for explanation 12 L Because I have to work. Explanation 13 E You can’t leave work early? Question (but very likely rhetorical: Statement) 14 L No I can’t. Answer (very likely: Contradiction) 15 E Why not? Demand for explanation 16 L Because it’s not in my job Explanation description. 17 E Well then don’t come. Directive 18 L What? Demand for explanation 19 E Don’t come. Directive It’s obviously an enormous Explanation/Accusation burden for you. 20 L Yeah but - Introducing possible Counter-claim 122 Chapter Five 21 E Just send Rory. Directive 22 L You’re telling me not to come to Demand for clarification the Christmas party? 23 E Well you’re obviously too busy. Explanation / Accusation 24 L I had the German measles in the Disqualification / Accusation 5th grade, I still had to show up to the Christmas party. 25 E Lorelai let’s be honest here, I’m Statement (possibly not too happy with you right now Disqualification) and I assume you’re not too happy with me. 26 L My polka dot dress matched my Disqualification / Accusation face and still I had to sit through 12 courses. 27 E I am tired of forcing you to do all Disqualification / Accusation those terrible things that infringe upon your life and I do not have the energy to pretend that the way you treated my the other day was in any way acceptable. 28 L So you’re uninviting me to Demand for clarification Christmas dinner? 29 E Yes I am. Clarification 30 L Fine. Statement 31 E Fine. Statement 32 L Ok, anything else? Question 33 E I believe that’s all. Answer 34 L Ok well, great mom, it’s been (Closing) Statement swell talking to you. 35 E Bye Lorelai. Greeting 36 L Bye. Greeting "What the Hell is Wrong With You?" 123 Attitude S Text Appraisal Appraiser Appraised Appraisal category E Lorelai. L Mom, hello. E I wanted to I wanted to Emily talk to Affect (desire) talk to you Lorelai about the about Christmas Christmas dinner this dinner Friday. L Ah, Christmas dinner. E You forgot. You forgot Emily Lorelai’s neg Judgement behaviour L Well mom, a lot going on Lorelai happenings (neg) Appreciation there’s been in Lorelai’s a lot going life on around here lately, not exactly high Emily’s neg Appreciation your on my list of Christmas Christmas things to obsess party shindig’s about not exactly obsess action neg Judgement high on my list of things to obsess about. E Well I’m I’m sorry Emily the timing is Affect (conventionalised) sorry if the bad timing is bad timing of neg Appreciation bad, Emily’s Lorelai, but party the world the world Lorelai neg Judgement doesn’t doesn’t always always revolve around revolve you around you. L Well thanks thanks for Lorelai the tip pos Affect for the tip. (conventionalised) NEGATIVE IRONY E Cocktails are at 6, dinner’s at 8. 124 Chapter Five L I probably won’t be there for cocktails. E Why not? Why? Emily that Lorelai neg Appreciation won’t be there L Because I have to Lorelai work modulation: obligation have to work. E You can’t can’t Emily leave work modulation: readiness leave work early (ability) early? L No I can’t. can’t Lorelai leave work modulation: readiness early (ability) E Why not? Why? Emily that Lorelai neg Appreciation can’t leave work early L Because it’s not in my job description. E Well then don’t come. L What? What? Lorelai that Lorelai neg Appreciation shouldn’t come E Don’t it’s an enormous Lorelai coming to neg Appreciation come. It’s burden (attributed the party obviously by Emily!) an enormous burden for you. L Yeah but - E Just send Rory. L You’re You’re telling Lorelai Emily’s neg Judgement telling me me not to come behaviour not to come to the party to the Christmas party? E Well you’re you’re too busy Emily Lorelai neg Judgement obviously too busy. "What the Hell is Wrong With You?" 125 L I had the I had the Lorelai Emily’s neg Judgement German German measles behaviour measles in … party. or situation the 5th still had to show up to Counter plus obligation grade, I still the had to show Christmas up to the party Christmas party. E Lorelai let’s let’s be honest Emily content of pos Judgement but works be honest here following as intensifier here, I’m utterance not too I’m not too Lorelai neg Affect happy with happy with you you right you’re not too Lorelai Emily neg Affect now and I happy with me (attributed assume by Emily!) you’re not too happy with me. L My polka My polka dot Lorelai Emily’s neg dot dress dress … courses behaviour Judgement/Appreciation matched my or situation face and still had to sit through Counter plus obligation still I had to 12 courses sit through sit through 12 courses neg Appreciation 12 courses. E I am tired I am tired of Emily Lorelai neg Judgement of forcing forcing you to do you to do all those terrible all those things that terrible infringe upon things that your life infringe I am Emily forcing … neg Affect upon your tired of your life life and I do forcing Lorelai action neg Judgement not have the terrible (attributed things neg Appreciation energy to infringe by Emily!) action neg Appreciation pretend that I do not have the Emily Lorelai neg Judgement the way you energy to treated my pretend that the the other way you treated day was in me the other day any way was in any way acceptable. acceptable I do not Emily (I) Judgement: capacity have the to pretend Appreciation: complexity energy that … [not] the way neg Judgement acceptable Lorelai treated E. 126 Chapter Five L So you’re you’re Lorelai Emily’s neg Judgement uninviting uninviting me to behaviour me to Christmas dinner Christmas dinner? E Yes I am. L Fine. Fine Lorelai pos Appreciation NEGATIVE IRONY E Fine. Fine Emily pos Appreciation NEGATIVE IRONY L Ok, ok Lorelai anything else? E I believe that’s all. L Ok well, ok Lorelai great mom, great pos Appreciation it’s been NEGATIVE IRONY swell swell talking to pos Appreciation talking to Emily NEGATIVE IRONY you. E Bye Lorelai. L Bye. CHAPTER SIX FREE WORD ORDER IN ARTIFICIAL LANGUAGES ALAN REED LIBERT 1. Introduction In this paper I explore the degree to which word order is free or fixed in artificial languages (ALs). Some creators of ALs assert that their languages have a quite free word order, but often it may not be that free. One the other hand, some AL designers say that their languages are rigid in word order, but this sometimes is not completely accurate. Aside from investigating these points, I shall briefly attempt to account for the restrictions on word order which exist in ALs. The problem in many cases is one of description: the AL designer (or other author writing about a language) has made unclear, contradictory, or confusing statements about the language; to some extent this chapter is about what authors say about ALs rather than about the ALs themselves. Creators of ALs would advance their causes by providing good descriptions of their languages, but they often do not do this. An example of lack of clarity in discussing word order comes from Igbinẹ´wẹ´ká (1987), writing about Guosa: In a complete sentence in Guosa Language, subject are [sic] placed before the verb. This, however depends on the mode of sentence and thought. e.g. Mr. Abasiemeka killed a snake: “Mazi Abasiemeka gbù máchìjí”. To reverse a sentence as natural as this is an incomplete sentence and wrong grammatical expression in Guosa language (p.6). It can therefore be difficult to find out the facts about an AL. There may be those who are dubious about the theoretical importance of artificial languages. I would contend that they are of considerable interest–for one thing, languages which have been consciously created 128 Chapter Six may be able to tell us something about the language faculty: since language designers can give their languages any form they wish, if their creations still follow some linguistic universals (particularly the more abstract ones that they might not be consciously aware of), it may mean that even the activity of language creation is constrained in some ways by the language faculty. For the purposes of this chapter I am only interested in those artificial languages which have been designed with some practical purpose in mind (usually that of facilitating international communication), such as Esperanto, i.e. I shall not deal with languages created as part of or in association with works of fiction or for playful purposes. I am also not concerned here with computer languages or systems of logic. Even with these restrictions, we have more than 1000 languages which can be examined, though I shall only be looking at a small fraction of these. Many researchers have used a classification of ALs based on the extent to which they are derived from natural languages: ALs whose elements (mainly) come from one or more natural languages are called a posteriori languages, while those whose designers do not (consciously) draw on natural languages as sources are known as a priori languages; ALs which have substantial a posteriori and a priori parts are mixed languages. In previous research (Libert, 2000; 2003) I have found some indications that as exotic as some a priori and mixed languages may appear on the surface, they do appear to follow many linguistic universals, i.e. in some respects they are not all that strange in underlying structure. Thus, even when a language creator does not intend to base an AL on natural languages, some parts of the AL are influenced by natural languages and/or by our language faculty. What is more, ALs do not even seem to have the same range of freedom of word order that is found in natural language; thus the word order possibilities of ALs are a subset of those of natural languages. There is a possible exception to this: although most of those ALs which are said to have a rigid word order are not as fixed as their designers indicate, there is at least one AL which is rather rigid, perhaps more so than any natural language, as we shall see. I shall focus on ALs which are said (by their designers, by proponents of them, or by others) to be either very free or very rigid in their word order, i.e. on the extremes of word order freedom, and not on languages which have a basic order but which allow variation. For example, Weisbart (1912, p.9) gives “Subjekt, Prädikat, Objekt” (“subject, predicate [presumably = ‘verb’], object”) as the “Gewöhnliche Folge der Satzglieder” (“Usual sequence of sentence members”) of his language Free Word Order in Artificial Languages 129 Europal. Since he calls this the “usual” order, one might assume that it is possible to have other orders. 2. ALs Asserted to Have Free Word Order Let us start with two quotations about Esperanto, by far the most successful and most used AL. In my view, at least the first of these gives the incorrect impression that there are few limitations on word order in this language.1 But why do we need case endings anyway, no matter how simple they are? The reason is that there are no rules in the Esperanto grammar that prescribe the word order in a sentence. […] But you cannot give rules for word order, even for an artificial language, in just a few lines, and it is probably for this reason that Zamenhof decided to allow the word order in Esperanto to be as free as possible (within the boundaries dictated by common sense of course). (anon., no date, <http://download.travlang.com/Ergane/es_06.htm>) Another point to notice in Esperanto sentence structure is the extremely free word order. There are just a few word types that do require strict word order in Esperanto, as in English. You can say al la lernejo [to the school], but you can't say lernejo la al [school the to]. Generally speaking, the words in Esperanto that require a fixed word order are the ones that don't require endings (Fantom, 2003). I would contend that if a language does not allow adpositions to precede or follow their objects it does not have an “extremely free word order”. While Esperanto is certainly more free than English, it does not approach the freedom of some natural languages. To see this we shall now look at some examples of considerable freedom of word order of natural languages. When discussing this, some authors may focus on the relative order of subjects, objects, and verbs, and we find some natural languages in which these can occur in any order relative to one another, though some orders may be more frequent than others. The relative positions of attributive adjectives and their nouns may also come up. However, languages with freedom in the placement of all of these items may not all necessarily be that free, because there may still be rigidity about the placement of other items (such as adpositions and articles), and there might be some restrictions on separating parts of constituents. However, there are natural languages which allow discontinuous constituents to at least a fair degree. Consider the following remark by Dixon (2002): 130 Chapter Six In some [Australian] languages it is not just the order of phrases in a clause that is free, but the order of words in a clause. Dyirbal is an extreme example – if there is one NP in A function, another in O, and a further NP in dative case, each consisting of several words, then the words in each phrase may be freely scattered throughout the clause (p.143). Discontinuous constituents can occur in some Indo-European languages, including Latin, as shown in Example (1):2 (1) grandia per multos tenuantur great-NOM/ACC-N-PL through many-ACC-M-PL are-tapered flumina rivos rivers-NOM/ACC-N-PL brooks-ACC-M-PL ‘Great rivers are channelled into many brooks’ (Ovid Remedia amoris 445, cited in Kessler, 1995, p.4) English is basically a prepositional language, in that the vast majority of its adpositions precede their objects, and there are languages such as Turkish, which are postpositional, with most or all adpositions following their objects. There are languages with a great degree of freedom in this regard, at least for some adpositions, i.e. they have what have been called ambipositions (a term used by Schaeder (2000, p.38) and others), adpositions which can occur in either position relative to their object. Sometimes there are restrictions on this in terms of case assignment and/or meaning (e.g. with one meaning an adposition must precede and with another meaning follow its object), but this is not always true, as shown in Example (2) from Hungarian. English arguably has a small number of such items; an example of one is in (3). (2) a. a híd-on át the bridge-SUPERESS over ‘over the bridge’ b. át a híd-on (same (truth-conditional) meaning, Marácz 1986: 236) (3) Notwithstanding these objections/these objections notwithstanding, they pressed ahead with their proposal. Free Word Order in Artificial Languages 131 In spite of the great freedom of word order in some natural languages, there do seem to be limits on what is possible. For example, I know of no languages in which a definite article can be in any position relative to its noun. Also, languages in which many elements of main and subordinate clauses can be mixed up with one another are very rare or non-existent. However, in at least one language (Latin) one can move something out of a subordinate clause, creating a discontinuous subordinate clause. An example of this is in Example (4) (with the parts of the subordinate clause in italics): (4) Tametsi tu scio quam Although thou-NOM-M/F/N-SG know-INDIC-PRES-1SG how sis curiosus art-SUBJ-PRES-2SG interested-NOM-M-SG ‘Although I know how interested you are’ (Caelius to Cicero, Epistulae ad Familiares 8, 1, 1, cited by Kessler, 1995, p.8) Having had a brief view of the degree of freedom of word order which exists in natural languages, let us now see how free ALs are, or are claimed to be. Consider first the following remark by Couturat and Leau (1979 [1903]) about Myrana, an AL based on Latin, designed by Joseph Stempfl: L’auteur laisse la construction entièrement libre: il considère comme impossible d’astreindre tous les peuples à une construction fixe et rigide; c’est, dit-il, créer et chercher des difficultés. Il faut que les flexions grammaticales indiquent suffisamment le rôle de chaque mot, quelle que soit sa place. C’est le meilleur moyen d’éviter tous les idiotismes de syntaxe. (“The author leaves (sentence) construction completely free: he considers it impossible to compel all peoples to (use) a fixed and rigid construction; it is, he says, to create and look for difficulties. It is necessary that the grammatical inflections indicate sufficiently the role of each word, whatever its position. It is the best means of avoiding all the idioms of syntax.”) (p.406). However, the word order is apparently not “completely free” since a small number of words do have fixed positions. For example, the negative word no is “mis devant le mot à nier” (‘placed before the word to be denied’) (Couturat & Leau, 1979 [1903], p.405). Sotos Ochando’s language (an a priori AL) was also said to have a large degree of freedom of word order. The first of the following quotations on it is from Couturat and Leau (1979 [1903], p.64), and the 132 Chapter Six second is from Gisbert and Lorrio (1862, p.94), a pedagogical work on the language: Pour la construction, il ne donne aucune règle, parce que la syntaxe permet toutes les inversions, comme en latin. Il recommande seulement de ne pas abuser de cette faculté, et de suivre autant que possible l’ordre logique. (“For sentence construction, he [= the designer, Sotos Ochando] does not give any rules, because the syntax permits all inversions, as in Latin. He recommends only to not abuse this ability, and to follow as much as possible the logical order.”) La Lengua Universal, que tambien tiene estos casos, puede por consiguiente tener cuanta latitud se apetezca en el uso del hipérbaton, sin mas limitacion que la que naturalmente impone la razon …. Esto no obstante en el uso ordinario seguirá la Lengua Universa la construccion natural que es la que indica el órden lógico de las ideas que se emitan en el discurso. (“The Universal Language, which also has these cases [nominative, genitive, etc.], can consequently have as much latitude as is desired in the use of hyperbaton, without any more limitation than reason naturally imposes … Notwithstanding this in ordinary use the Universal Language will follow the natural construction which is that which is indicated by the logical order of the ideas which are expressed in the discourse.”) In the absence of rules forbidding a certain word order and of native speakers, it is impossible to prove that a language such as Sotos Ochando’s language is not as free as e.g. Dyribal; texts in the language could only give us evidence for the existence of certain orders, not evidence against it (though they could provide an indication of frequency of different orders). Further, all of the texts that I have in this language are from Gisbert and Lorrio (1862), in which, as a pedagogical work, one might not expect unusual orders to occur. However, one might note that in all of the direct wh-questions in the texts in this work there is wh- movement. This could make one wonder whether it was meant to be compulsory, indicating that there could be at least one limitation on word order. There is one respect in which Sotos Ochando’s language is freer than most natural and artificial languages: the articles (of which there are four different ones) can occur before or after their noun. However, if they occur after it, they are not separate words but affixes, and their form is different. This is shown in Example (5) with il, which is equivalent to the English definite article in its most common use; as a suffix it is -i: Free Word Order in Artificial Languages 133 (5) ‘the man(NOM): il ibacala / ibacalai (Couturat & Leau, 1979 [1903], p.60) The same sort of thing happens in Danish, although the two possibilities are not in free variation, as they are here. One might see such cases as less interesting than if the article as a separate word could either precede or follow its noun, but I do not know of any instances of this happening in an AL. Kosmos, another AL based on Latin, has a very free word order, according to Couturat and Leau (1979 [1903]): La seule indication relative à la syntaxe est celle-ci: la place normale de l’adjectif est après le substantif. D’ailleurs, l’auteur n’est nullement partisan d’une construction rigide, et laisse toute liberté sur ce point, grâce à la déclinaison. (“The only indication relative to the syntax is this: the normal place of the adjective is after the noun. Moreover, the author is in no way a supporter of a rigid (set of rules for) construction, and allows complete freedom on this point, thanks to [the existence of] the declension.”) (p.378) I suspect that the creator of this language would not really allow “complete freedom” of word placement and there is evidence that he would not. Couturat and Leau (1979 [1903], p.375) state that one way of forming comparatives and superlatives is with the words magis ‘more’ and maxime ‘most’ respectively “placés devant l’adjectif” (“placed before the adjective”), from which we can assume that this is their only possible position when they have this function. Similarly, substantivized adjectives have the article before them, i.e. it cannot follow them. Esperantido (a modification of Esperanto) would seem to be very free with respect to word order given the following quotation: La ordo de la vorton en la frazo estas tute libera, se nur la signifo de la frazo restas klara. Le pley bona vortordo estas tiu, kiu estas la pley klara, le pley logika k. la pley agrabla slo la stila vidpunkto. (“The word order in the sentence is completely free, as long as the meaning of the sentence remains clear. The best word order is that which is at the same time the most clear, the most logical, and the most agreeable according to the point of view of style.”) (de Saussure, 1919, p.9) It might be difficult to judge “when the meaning…remains clear” but it would be surprising if discontinuous constituents occurred in this language. It would also be surprising to see adpositions following their complements as all of the adpositional phrases that I have seen in texts in 134 Chapter Six de Saussure (1919) contain prepositions, not postpositions. Similarly all of the definite articles that I have seen in this work are before their nouns. Weferling (1974) says of INTAL, “Le pozition del vortes in le satse es generalim tre liber. Ma le regulari ordin es: subjekt – verbe – objekt” (“The position of the words in the sentence is generally very free. But the regular order is subject – verb – object”) (p.15). Again there is apparently not complete freedom: I assume that adpositions always precede their object without any intervening words permitted and there is a statement from the language designer to back this up: Weferling (1974) states, “Prepozitiones star nonmediat avan ler komplement” (“Prepositions stand immediately before their complement”) (p.14). To summarize in regards to ALs which are stated to be very free in their word order, it is unclear whether any ALs have total freedom, and some of them may even lack the possibility to have any prepositions before or after their objects, a possibility which exists to some extent in many natural languages. I have never seen a sentence in an AL of the sort illustrated by (4), i.e. where a subordinate clause is discontinuous. 3. ALs Asserted to Have Rigid/Invariable Word Order We shall now look at ALs which are claimed to have a fixed word order in one or more than one respect, and we will see that sometimes these assertions are inaccurate. We can distinguish two situations, and hence two types of inaccurate statements. On the one hand an assertion about an order can be inaccurate because in certain situations another order is used, and perhaps must be used, as when statements and some kinds of questions require different orders. Thus a language designer (or someone else writing about an AL) may simply be careless and make a blanket statement about word order, forgetting that for certain situations he prescribes a different order. On the other hand, a designer may state a rule and then not always follow it, possibly through negligence in his use of his own language (or another author may give a rule, based on what a language creator states, and then have examples by the creator which violate it). We shall look at instances of both situations. Couturat and Leau (1979 [1903]) state that in Lingua Komun “le sujet est toujours avant le verbe” (“the subject is always before the verb”; emphasis their own) (p.482). However, in the sample passage of this language (the “Lord’s Prayer”) that they give there are two places where this does not hold: Free Word Order in Artificial Languages 135 (6) sante esa tue nómine; vena imperio tue holy be thy name come kingdom thy ‘hallowed be thy name; thy kingdom come’ (ibid.) The designer of this language (F. Kürschner) was probably influenced by the word order of the Latin and/or German versions of this (Lingua Komun is at least largely derived from Latin), so one might say that this is not a fair example; however, a strong statement is made (presumably Couturat and Leau were accurately reporting on what was said in the original source on this language, which I do not have), and is violated. Idiom Neutral has a rigid order in some respects: attributive adjectives must follow their nouns, and all adpositions are prepositions. Couturat and Leau (1979 [1903]) state, “Le verb est toujours précédé du sujet et suivi de ses compléments.” (“The verb is always preceded by the subject and followed by its complements”) (p. 491). However, this is contradicted in the following sentence, “Dans les phrases interrogatives, le verb est précédé du sujet, et du mot interrogatif (qui peuvent être identiques)” (“In interrogative sentences the verb is preceded by the subject and by the interrogative word (which can be the same)”). Couturat and Leau (1979 [1903]) also say , “Le place de l’adverbe est après le verbe (sauf la négation no) mais avant le mot modifé, s’il n’est pas un verbe: multe grand, très grand …” (“The place of the adverb is after the verb (except for the negation no), but before the word modified, if it is not a verb: multe grand ‘very big’ …”) (p.492). Aside from the fact that this does not account for adverbs modifying attributive adjectives of subjects (which would presumably be before verbs, since subjects are usually before them), in Couturat and Leau (1979 [1903], p.496) there is a text in Idiom Neutral, presumably from an authoritative source on the language, in which an adverb (plu ‘more’) precedes a verb: (7) traduksion no plu es nesesar translation no more is necessary ‘translation is no longer necessary’ One might say that plu precedes the verb here because it is part of a negating phrase, along with no (which is stated to be before verbs), but this is not explicitly allowed for by the rule given above. Let us now look at aUI, an a priori language. Weilgart (1979, p.50D), the designer of the language, gives a strict rule: “WORD-ORDER is always: “SUBJECT—VERB—OBJECT—INDIRECT OBJECT.”3 However, several pages earlier (Weilgart, 1979, p.45) is a weaker version of the rule: 136 Chapter Six INVERSION is to be avoided. Best is always SUBJECT—VERB— OBJECT order, even in questions and relative clauses: Instead of Whom did he see? rather he saw whom? ‘vu iOpAv hu?’ He saw the man whom we know = he saw the man, we know whom = vu iOpAv pI u, fnu gUv xu.’ ( = we know which). In any case, the SV order is not followed in the sentence in (8) (8) hE c´ fE? what be this ‘What is this’ (Weilgart, 1979, p.28) This is probably due to the author’s confusion of subject and subject complement. One of the causes of poor descriptions of languages by language designers may be their lack of linguistic knowledge. There is also more than one instance in Weilgart (1979) of an indirect object preceding a direct object, so he does not always follow the strict rule that I have quoted, or his statement that the indirect object “always comes last” (Wielgart, 1979, p.46). Weilgart (1979) says of questions, “No inversion! ‘What does he eat?’ = He eats what?” (p.49). It is unclear whether he means that in wh- questions in general (i.e. not only when the wh-phrase is an object) there is no wh-movement, but in fact in his texts he puts wh-adverbs in several places: in situ, e.g. (9a), in clause-initial position, e.g. (9b), and also in the second position of a clause, e.g. (9c). One might think that this last (rather usual) placement is an error, but (9c) is not the only instance of it. (9) a. bu c´ ha? you be where ‘Where are you?’ b. ha c´ bu? where be you ‘Where are you?’ c. vu ha tev-yEc at bEna? he where come-CTF to land ‘Where would he land (come-to land)?’ (Weilgart, 1979, p.56, 95) We thus see some apparent freedom in one aspect of word order of this language. Free Word Order in Artificial Languages 137 Given the incorrect statements about ALs that we have seen, it must be asked whether any ALs really are invariable or rigid in word order. There is an AL, Interglossa, which keeps the same word order in statements, questions, and relative clauses, as Hogben (1943, p.34) explicitly states. He is aware that this may not be all that easy to follow this, as he says, “The beginner has to get accustomed to the trick of preserving the word-order of an equivalent simple sentence in a relative clause. This will offer no difficulty to anyone who is familiar with colloquial Anglo-American.” There is a relative pronoun in this language, but it is only used as the subject of a relative clause (10a) and for possessors of subjects; there is no overt relative pronoun for noun phrases in other functions and hence the statement word order is maintained, since there is nothing to move to the beginning of the clause (10b): (10) a. Un anthropi; su pre dicte re; non habe bio SG man who before say it not have life ‘The man who said so is dead’ b. Un anthropi; mi pre vise; non habe bio SG man I before see not have life ‘The man I saw is dead’ (Hogben, 1943, p.35) Yes-no questions have the particle que as their first constituent (unless there is a vocative phrase, which will precede it) and otherwise are identical to the corresponding statements (11a). Wh-questions have their wh-phrase (which will be introduced by quo ‘which’) in situ (11b): (11) a. Que re habe thermo Q it have heat ‘It is hot?’ b. Tu esthe volo quo re you feel desire which it ‘What do you want?’ (Hogben, 1943, p.91-92) Interglossa is thus one of the few ALs which really has a fixed (or as Hogben (1943, p.34) puts it, “invariable”) word order under all circumstances for S, V, and O. The “general formula for all types of sentence or clause” (Hogben, 1943, p.35) is given in (12): 138 Chapter Six (12) (1) Vocative cluster (if present) followed by a colon … (2) Interrogative particle or imperative particle or linkword (if present). (3) Subject cluster. (4) Verboid cluster. (5) Direct and Indirect Object clusters with accompanying qualifying clusters. Hogben is then more specific about the relative order of the “clusters” mentioned in (5) in this list, giving, as he puts it, “The rule of precedence with reference to the Direct and Indirect or Instrumental Object clusters”: “the shorter of the two comes first” (Hogben, 1943, p.35). He also states the order of modifiers of nouns and their nouns: “phrases” (by which he means prepositional phrases and “clusters equivalent to a participial phrase”) are placed after the noun, while “single words” are placed before it. One of his examples is in (13): (13) U palaeo gyna in horta SG old woman in garden ‘The old woman in the garden’ (Hogben, 1943, p.35) His reference to “single words” as opposed to phrases is misleading (even if we do not follow the standard generative practice of considering all adjectives to be heads of adjective phrases, whether or not there is anything else in these phrases), since if an attributive adjective is modified by what would be an adverb in English, apparently it will still precede the head noun: (14) mega longo historo much long story ‘a very long story’ (Hogben, 1943, p.35) Hogben is thus not immune to lack of clarity or errors in descriptions of ALs. However, the template that he gives for “substantive clusters” covers (14): (15) (1) Directive (i.e. preposition equivalent). (2) One of the following: (a) pronoun-equivalent; (b) general article; (c) numeral. (3) A qualifier of (4), i.e. an adverb-equivalent. (4) A qualifier of (5), i.e. an adjective-equivalent. (5) A noun-equivalent … (Hogben, 1943, p.36) Free Word Order in Artificial Languages 139 This template does not obviously cover cases when two or more adjectives modify the same noun, although Hogben brings up such situations. Here is thus another (relatively minor) problem with his description of his language, and there may be others. Even with Hogben’s (1943) rigid and specific set of word order rules for Interglossa, there seems to be some room for variation, since one of the “rules for clause-order” (these concern the orders of different clauses in a sentence) is that “An adverbial clause preferably precedes the principal [i.e. main clause]” (1943: 36). Since he uses the word “preferably” I assume that this is not an absolute rule. His other “rules for clause order” are: “A noun clause follows the principal without a conjunction equivalent to that” and “A relative clause immediately follows the substantive which it qualifies” (p.36). Hogben (1943) explicitly allows variation in one aspect of word order: The general rule that any single qualifying word must immediately precede the word it qualifies admits of one exception to allow for afterthought. Words or expressions which qualify a sentence or clause as a whole may come at the beginning of it or at the end, as do surely and a long while in the English sentences: (a) surely you don’t mean that; (b) he has been staying there a long while (p.36). Interglossa appears quite rigid in most aspects of word order, and perhaps more rigid than most, if not all, natural languages. However, I know of no AL (or any natural language) which is completely unvarying and rigid in all aspects of word order, and this perhaps tells us something about the nature of language. This is in line with Universal 297 of the Universals Archive: “In all languages, inversion of the word order can be used as a logical or emotional emphasis.” 4. Conclusion To conclude, I shall touch on the causes of word order restrictions in ALs. Not surprisingly, ALs which are largely based on one or more natural languages often “inherit” the word order rules of those languages. Situations of greater interest would be those in which ALs deviate from the word order of the languages on which they are based, or those of a priori languages, or those which hold of ALs in general. The facts about articles and adpositions brought up in this chapter are likely due to the fact that the native languages of the creators of the vast majority of ALs are one of the major Indo-European languages (English, French, German, etc.), in which articles occur at the beginnings of noun phrases and 140 Chapter Six adpositions are usually prepositions. As noted above, a priori and mixed languages often behave in accord with language universals, which is interesting since a language designer could do anything he wanted with his language. For example, to my knowledge no artificial language violates the following: “Question-word Movement Universal: All languages that move [a] WH-question word to clause-peripheral position move it to the left, not the right” (Universal 1379 of the Universals Archive). It should be pointed out that the Universals Archive is a collection of posited universals, not universals which have been proved to hold; many of them have counter-examples, and indeed this one does. However, even if a universal is not without exception, if it states something which holds in the overwhelming majority of languages, this may still tell use something about the language faculty, which could play a role in constraining the design of ALs. Interglossa may come close to violating the universal cited in the previous section (Universal 297 of the Universals Archive), since it does not allow much “inversion of the word order” (although a precise definition of such inversion would be needed to fully judge this). Aside from the effects of natural languages and of language universals (i.e. of the language faculty), the design of grammars of ALs is often influenced by considerations of ease of learning and use; after all, most major ALs were designed to be international auxiliary languages and one of their frequently asserted advantages was that they were less difficult to learn than natural languages. Some authors may believe that freedom of word order, at least in some domains, is a liability; consider the comment of Couturat and Leau (1979 [1903]) on the fact that in Idiom Neutral adpositions must precede their complement: Cette règle, qui parait toute simple aux Francais, est très utile pour les Allemands, chez qui le substantif est tantôt précédé, tantôt suivi de la préposition …, ou, ce qui est pis encore, encadré entre deux prépositions (“This rule, which appears quite simple to the French, is very useful for the Germans, among whom the noun is sometimes preceded, sometimes followed by the preposition …, or, what is worse, enclosed between two prepositions”) (p. 492). However, Clark (1907) has a different opinion, although he is speaking more generally, and presumably not about adpositions, since Esperanto does not have ambipositions. In the following passage he is arguing for the value of the accusative marker of that language: The flexibility of the language is enormously increased; the words can be put in any order without obscuring or changing the sense. Ex.: Free Word Order in Artificial Languages 141 La patro amas sian filon = the father loves his son. Sian filon amas la patro (in English "his son loves the father" has a different sense). Amas la patro sian filon (= the father loves his son, but...). La patro sian filon amas. Sian filon la patro amas (= it is his son that the father loves). In every case the Esperanto sentence is perfectly clear, the meaning is the same, but great scope is afforded for emphasis and shades of gradation. Further, every nation is enabled to arrange the words as suits it best, without becoming less intelligible to other nations. Readers of Greek and Latin know the enormous advantage of free word order. For purposes of rendering the spirit and swing of national works of literature in Esperanto, and for facilitating the writing of verse, the accusative is a priceless boon. Is the price too high? (p. 43-4) Probably in many cases a designer’s views about the advantages of having a free or fixed word order has influenced the rules that he gave to his language. Notes 1. Admittedly these are not academic sources. However, they do give an indication of how some people who (presumably) know Esperanto think of this language, and the first of them is from one of a series of lessons on it. 2. This is an example from poetry, and one might argue that such examples should not be cited, since poetic language is more free than that of prose; however, as poetic language is a part of language, its data are relevant, and even in poetry there may indeed be word orders which are not well-formed. 3. Indirect objects are in prepositional phrases headed by at ‘to’. References Anonymous (no date). Lesson 6–Cases. Retrieved 30th September, 2007, from http://download.travlang.com/ Ergane/es_06.htm. Clark, W.J. (1907). International Language. Past, present & future. Retrieved 11th October, 2007, from http://www.gutenberg.org/files/16737/ 16737-h/16737-h.htm. Couturat, L. & Leau, L. (1979 [1903]). Histoire de la langue universelle. Hildesheim: Georg Olms. de Saussure, R. (1919). Fundamento de la Internacia Lingvo Esperantida. Bern Dixon, R.M.W. (2002). Australian Languages. Cambridge: Cambridge University Press. 142 Chapter Six Fantom, I. (2003). Achieving the impossible with Esperanto. Retrieved 2nd October, 2007, from http://www. languageadvantage.com/features/esperanto2003.htm. Gisbert, D.L. & Lorrio, D.P. (1862). Manual de lengua universal. Madrid: no publisher given. Hogben, L. (1943). Interglossa. Harmondsworth: Penguin. Kessler, B.(1995). Discontinuous Constituents in Latin. Retrieved 11th October, 2007, from artsci.wustl.edu/~bkessler/latin-discontinuity/discontinuity.ps. Igbinẹ´wẹ´ká, A. (1987). The Dictionary of Guosa Language Vocabularies, Vol. 1. Ikeja, Lagos: Guosa Publications. Libert, A. (2000). A Priori Artificial Languages. Munich: Lincom Europa. Libert, A. (2003). Mixed Artificial Languages. Munich: Lincom Europa. Marácz, L.K. (1986). Dressed or Naked: The Case of the PP in Hungarian. In W. Abraham & S. de Meij (Eds.) Topic, Focus, and Configurationality (227-252). Amsterdam: John Benjamins. Schaeder, B. (2000). Ambiposition. In H. Glück (Ed.) Metzler Lexikon Sprache, 2nd ed. (38). J. B. Metzler: Stuttgart. Universals Archive. Retrieved 11th October, 2007, from http://typo.uni- konstanz.de/archive/intro/. Weferling, E. (1974). Standard-Gramatika de International Auksiliari Lingue, 6th ed. Braunschweig: self-published. Weilgart, J.W. (1979). aUI: The Language of Space, international 4th ed. Decorah, IA: Cosmic Communication Co. Weisbart, J. (1912). Europal. Hamburg: self-published. CHAPTER SEVEN SYNTACTIC ENCODING OF TOPIC AND FOCUS IN KOREAN HYERAN LEE 1. Introduction In this chapter, I attempt to examine topicalization, focalisation and scrambling in Korean to elucidate the articulated structure of Complementizer Phrases (CP) 1. I claim that discourse information such as Topic and Focus is syntactically realized at each position in the CP area driven by the features. Topic is old information while Focus is new information. Topic is what the sentence is about while the rest of the sentence is what is said about the topic. Focus is elements that are focused as new information. Contrastive Topic and Contrastive Focus are formed when the presupposed entities exist in context, having the contrastive reading or the exhaustive list reading. I show that the syntactic position determines the particular reading between Topic (Contrastive Topic) and Focus (Contrastive Focus). The question is then what syntactic positions are held in the clausal periphery to encode discourse information such as Topic and Focus. The abbreviations used in this chapter are introduced as below. Complementizer Phrase (CP) introduces a clausal complement by complementizers such as that, whether, and if. Force Phrase (ForceP) encodes the force of the clause, the sentence type. Finite Phrase (FinP) determines the finiteness of the clause, whether it is tensed or non-tensed. Inflectional Phrase (IP) determines inflections such as gender, number or person agreement. Tense Phrase (TP) determines the tense, whether a clause is tensed or non-tensed. Topic Phrase (TopP/TopicP) is introduced for languages like Korean/Japanese that hold topics roughly in the sentence initial position. Focus Phrase (FocP/FocusP) is a phrase that has a focused element. Verb Phrase (VP) is a lexical category that is the locus of 144 Chapter Seven theta roles. The little vP is projected right above VP as a shell structure, having theta-relation with its specifier. Specifiers (Spec) are linearized to the left of their head in English and Korean as the outcome of the second merge while complements are the outcome of the first merge. Noun Phrase (NP) contains a noun optionally preceded by determiner and followed by a prepositional phrase. In the sentence, The woman gave the girl the gift, the woman assumes Nominative Case (NOM) while the girl assumes Dative Case (DAT) and the gift, Accusative Case (ACC). Present (PRES) represents the present tense while Past (PAST), the past tense. Declarative (DEC) indicates the sentence type, the declarative. Contrastive Focus (CF) is contrastively focused with respect to other entities in discourse. Contrastive Topic (CT) is also contrastively topicalized with respect to other entities. 2. Theoretical Background Rizzi (1997, 2004) argues that specific discourse functions such as Topic and Focus are syntactically expressed with different functional categories within the CP area. Rizzi proposed the articulated CP structure that is analogous to Pollock’s (1989) structure and Larson (1988)’s VP structure. He suggests the following structure of CP. (1) [ForceP [TopicP [FocusP [TopicP [FinP [IP… ForceP encodes the sentence type, the force of the clause, while the FinP determines the finiteness of the clause. TopicP and FocusP appear between the upper and lower bound, ForceP and FinP respectively in the left periphery. The discourse information is syntactically realized in the CP area with distinctive heads in his system. Korean may make use of the articulated CP structure to better express all types of preposing such as topicalization, focalization, and scrambling that have not been very clear in their definition and distinction. The elaborated CP structure in Korean doesn’t have to be the same with (1) above. Rizzi (1997) says that the topic-focus system is present in the structure only if it is needed. Based on his claim I posit the following structure in Korean: (2) [CP [TopP [FocP [TP [vP [VP]]]]]]] FocP could have multiple Specifiers (Specs) since the multiple focusing is possible in Korean (Choe 1995). Unlike FocP, TopP does not have multiple Specs since there is only one topic in a sentence in Korean Syntactic Encoding of Topic and Focus in Korean 145 (Choe 1995, Han 1998). See the following examples (the bold faced letters indicate that the NP is focused): (3) a. John-i Mary-eykey senmwul-ul café-eyse cwu-ess-ta John-NOM Mary-DAT gift-ACC café-in give-PAST-DEC ‘John gave Mary a gift in a café’ b. John-i Mary-eykey senmwul-ul café-eyse cwu-ess-ta c. Mary-eykey John-i senmwul-ul café-eyse cwu-ess-ta d. Senmwul-ul John-i Mary-eykey café-eyse cwu-ess-ta e. Café-eyse John-i Mary-eykey senmwul-ul cwu-ess-ta f. Mary-eykey senmwul-ul café-eyse John-i cwu-ess-ta g. Senmwul-ul Mary-eykey café-eyse John-i cwu-ess-ta h. Café-eyse Mary-eykey senmwul-ul John-i cwu-ess-ta The above sentences are different from the scrambling sentences in spite of the same word order changes, which will be accounted for in the discussion of “Scrambling” below. Sentences (3b-e) show that an element is preposed for focusing effects, while sentences (3f-h) show that the multiple focus phrases are possible. The notions such as topic (TOP), contrastive topic (CT), and contrastive focus (CF) have been used in a confusing way in terminology, since the same –nun marker is used for different readings. For this analysis, I adopt Han’s (1998) definition as below. Han’s examples are slightly adapted for my own purposes. (4) John-un Mary-lul coaha-n-ta John-TOP/CT Mary-ACC like-PRES-DEC ‘As for John, (he) likes Mary’ ‘As for John, (he) likes Mary (Tom likes Jane, and David likes Laura)’ (5) John-i Mary-nun coaha-n-ta John-NOM Mary-CF like-PRES-DEC ‘John likes Mary (but not others)’ (6) Mary-nun John-i coaha-n-ta Mary-CT/TOP John-NOM like-PRES-DEC ‘As for Mary, John likes her’ ‘As for Mary (not for others), John likes (her)’ The topic and the contrastive topic readings are available for John in (4) and (6). In (5), the contrastive focus reading is the only available interpretation when the object NP is –nun marked. The same –nun marker for Mary in (5) and (6) thus has different readings, as in the following examples: 146 Chapter Seven (7) a. John-un Mary-eykey senmwul-ul café-eyse cwu-ess-ta John-TOP Mary-DAT gift-ACC café-in give-PAST-DEC ‘As for John, (he) gave Mary a gift’ b. John-un Mary-eykey-nun senmwul-ul café-eyse cwu-ess-ta John-TOP Mary-DAT-CF gift-ACC café-in give-PAST-DEC ‘As for John, (he) gave a gift to Mary in a café, not to others’ Sentence (7a) shows that there is a topic in the sentence initial position. Sentence (7b) shows that there is a single topic in the sentence initial position, though the two –nun markers appear. The second –nun marker in (7b) is not interpreted as topic but as contrastive focus, meaning that John gave a gift to Mary in a café, not to others. In this section I have briefly reviewed Rizzi’s (1997, 2004) CP structure, from which it has been proposed that the Korean CP structure employs topic and focus phrases. Terminology related with topic and focus has been checked, adopting Han’s (1998) definition. A detailed discussion of topic and focus will be made in the following sections. 3. Topic and Topicalization Korean is generally known as a discourse (topic) prominent language. Li and Thompson (1976) classified Korean as both subject prominent and topic prominent language. In a topic prominent language, a variety of topics are often present while in the subject prominent language the subject is obligatorily present (EPP). Located in the sentence initial position, the topic may have the –nun marker in Korean, expressing what the sentence is about with respect to the rest of the sentence. The rest of the sentence is called the comment clause, which follows the topic and says something about the topic. The unmarked topic is definite and generic and has familiarity to the speaker and the hearer. (8) John-un ku nonmwun-ul ilk-ess-ta John-TOP/CT the paper-ACC read-PAST-DEC ‘As for John, (he) read the paper’ ‘As for John, (he) read the paper, (Tom read X, and David read Y)’ (9) Ku nonmwun-un John-i ilk-ess-ta the paper-TOP/CT John-NOM read-PAST-DEC ‘As for the paper, John read (it)’ ‘As for the paper, John read (it), (but not other papers)’ John-un and ku nonmwun-un have the topic/contrastive topic reading with all the characteristics of the topic as mentioned above. Syntactic Encoding of Topic and Focus in Korean 147 The topic marker –nun is not indispensable to express topicality in Korean. Without the topic marker -nun, the thematic subject or object in the sentence initial position can be interpreted as topic as below. (10) John-i ku nonmwun-ul ilk-ess-ta John-NOM the paper-ACC read-PAST-DEC ‘As for John, (he) read the paper’ (11) Ku nonmwun-ul John-i ilk-ess-ta the paper-ACC John-NOM read-PAST-DEC ‘As for the paper, John read (it)’ In (10), the thematic subject John is understood as topic, what the sentence is about, and the rest of sentence is what is said about John. In (11), the object ku nonmwun-ul is what the sentence is about and the rest of sentence is what is said about ku nonmwun-ul. The –nun marker could have the contrastive focus reading as below. (12) John-i ku nonmwun-un ilk-ess-ta John-NOM the paper-CF read-PAST-DEC ‘John read the paper, (but not others)’ Though the same –nun marker is used in (12) and (9), the NP with –nun in (12) is CF while the NP with -nun in (9) is TOP/CT. The same morphological form –nun is interpreted in a different way by the syntactic position that the element occupies. Chomsky (1977) assumes that topics are base-generated in their surface position and topicalization involves movement of an abstract element from an argument position to a position immediately adjacent to topic. Such movement is similar to the operator movement of the relative clause. Following Chomsky (1977, 1986, 1995), Choe (1995) argues that topics in Korean are base-generated in the sentence initial position, while the abstract element, the null operator, moves in LF in order to form the topic-comment relation by strong binding (Chomsky 1986, 1995). Strong binding states that every variable must have its range fixed by a restricted quantifier, or have its value determined by an antecedent (Chomsky 1995: 56). Choe’s proposed structure is presented as below. (13) [CP [τ topici [TopP 0i [FP focusj [TP [ ti… tj… ]]]]]] According to Choe (1995, p.270), the null pronominal operator (0) moves to the Spec of TopP (0i, ti) and it is bound to topici (topici, 0i) by 148 Chapter Seven strong binding, constituting the topic-comment relation in LF. τ is a category representing the topic-comment relation. Focusj may be information focus or contrastive focus or wh-focus. Foci undergo preposing either in Syntax (Focus movement) or in PF (PF scrambling). The null operator is not involved in focalization. Using this structure, Choe (1995, p.269) claims that foci and topics in Korean should move to certain positions either at S-structure or in LF to be properly licensed and interpreted in LF. Under the recent minimalism where LF and PF exist with the elimination of S-structure, we have to reformulate her proposal with respect to the topic and focus movement as follows: (14) Topics (the null operator) and foci in Korean should move to certain positions in the CP area before Spell-Out to be properly licensed and interpreted. Based on Chomsky (1977, 1986, 1995) and Choe (1995), the proposed structure in (2) is used for our analysis. Sentences (8-10) are schematized as below. (15) John-un ku nonmwun-ul ilk-ess-ta John-TOP/CT the paper-ACC read-PAST-DEC ‘As for John, (he) read the paper’ ‘As for John, (he) read the paper, (Tom read X, and David read Y)’ [CP [TopP John-un 0i [TP ti [vP ti [VP ku nonmwun-ul ] ilk]-ess]]-ta] (16) Ku nonmwun-un John-i ilk-ess-ta the paper-TOP/CT John-NOM read-PAST-DEC ‘As for the paper, John read (it)’ ‘As for the paper, John read (it), (but not others)’ [CP [TopP ku nonmwun-un 0i [TP [vP ti [vP John-i [VP ti ] ilk]]-ess]]-ta] (17) John-i ku nonmwun-ul ilk-ess-ta John-NOM the paper-ACC read-PAST-DEC ‘As for John, (he) read the paper’ [CP [TopP John-i [TP (John-i) [vP (John-i) [VP ku nonmwun-ul ] ilk]- ess]]-ta] (18) Ku nonmwun-ul John-i ilk-ess-ta the paper-ACC John-NOM read-PAST-DEC ‘As for the paper, John read (it)’ [CP [TopP ku nonmwun-ul [TP [vP (ku nonmwun-ul) [vP John-i [VP (ku nonmwun-ul) ]ilk]]-ess]]-ta] In (15) and (16), the topic is base-generated with the null operator. In (15), the object is assigned accusative Case by Agree (Chomsky 2001) in situ. Syntactic Encoding of Topic and Focus in Korean 149 The subject in (16) has nominative Case as default in situ following Schütze (2001) and Lee (1992). In case the topic is not marked with –nun, assuming the nominative Case marker in (17) or the accusative Case marker in (18), the NP is topicalized directly from the Case position without the null operator movement. In (19) below, the topic is base-generated with null operator movement, providing the topic reading. The subject NP with FOCUS is raised to the Spec of FocP, providing the contrastive focus reading. (19) ku nonmwun-un John-un ilk-ess-ta the paper-TOP/CT John-CF read-PAST-DEC ‘The paper, John read’ (‘As for the paper, it is John (not others) who read (it)’) (‘As for the paper (not for other papers), it is John (not others) who read (it)’ [CP [TopP ku nonmwun-un 0i [FocP John-un [TP (John) [vP ti [vP (John) [VP ti ] ilk]]]-ess]]]-ta] The null pronominal operator moves to the position immediately adjacent to the topic, forming the topic-comment relation by strong binding with the base-generated topic. John-un with the contrastive focus reading moves to the Spec of FocP, which is lower than TopP. The second -(n)un marker cannot be interpreted as the topic marker due to the syntactic position. The above data have shown that the sentence initial element located in the Spec of TopP is always interpreted as topic regardless of the existence of –nun. The –nun marked element right below TopP is interpreted as contrastive focus. The topic reading and the –nun marker do not necessarily go together, though the sentence initial –nun marker always indicates topic/CT. The topic reading is obtained by the syntactic position, the Spec of TopP. Thus elements are supposed to move to the Spec of TopP position by topicalization to provide the topic interpretation. 150 Chapter Seven 4. Focus and Focalization É. Kiss (1998) classifies foci as informational focus and identificational focus. (20) Informational foci: What is not presupposed in a topic-focus (theme- rheme) structure. (21) Identificational focus: Identificational focus expresses exhaustive Identification. Cinque (1993) proposes the Nuclear Stress Rule (NRS) for informational focus, claiming that sentences with informational focus have the nuclear stress on the syntactically lowest element in the structure. É. Kiss (1998) says that identificational focus is different from informational focus in that the stress is put on the element that is identificationally focused. He also says that identificational focus expresses exhaustive identification like the cleft construction in English. Based on Han’s (1998) definition, contrastive focus in Korean has the –nun marker and it has the same exhaustivity as identificational focus. The difference is that contrastive focus presupposes other entities in the context. For our analysis we would like to adopt all three foci in Korean. See the following (Example (12) is repeated in (24)). Informational focus (22) John-i ku nonmwun-ul ilk-ess-ta John-NOM the paper-ACC/FOC read-PAST-DEC ‘John read the paper’ Identificational focus (23) John-i ku nonmwun-ul ilk-ess-ta John-NOM/FOC the paper-ACC read-PAST-DEC ‘John read the paper’ (‘It is John who read the paper’) Contrastive focus (24) John-i ku nonmwun-un ilk-ess-ta John-NOM the paper-CF read-PAST-DEC ‘John read the paper’ (‘John read the paper, (but not others)’) In Korean, focus is phonetically manifested with stress in situ or by preposing. The NP, ku nonmwun-ul in (22) is the lowest element in the sentence so that it is informational focus by the NSR. On the other hand, John in (23) is identificational focus. Sentence (24) shows contrastive focus with the nun-marker. I propose that foci, informational focus, identificational focus, and contrastive focus, are licensed and interpreted Syntactic Encoding of Topic and Focus in Korean 151 by preposing (overt focus movement) or by covert movement in LF interface. First, we would like to examine the canonical order. The sentence below shows the covert movement of informational focus. (22) is repeated below. (25) John-i ku nonmwun-ul ilk-ess-ta John-NOM the paper-ACC read-PAST-DEC ‘John read the paper’ [CP [TP [vP John-i [VP ku nonmwun-ul ] ilk]-ess]-ta] LF: [CP [FocP ku nonmwun-ul [TP [vP John-i [VP (ku nonmwun-ul)] ilk] ess]]-ta] In (25), nominative Case is assigned as default in Korean following Schütze (1997, 2001) so that the subject does not move. Accusative Case is assigned in situ by Agree following Chomsky (2001, 2005). Before Spell-Out, the NP ku nonmwun-ul is moved to the Spec of FocP to be properly licensed and interpreted. The subject NP with focus below is raised to the Spec of FocP, while the object NP stays in situ. Sentence (23) is repeated below. (26) John-i ku nonmwun-ul ilk-ess-ta John-NOM/FOC the paper-ACC read-PAST-DEC ‘John read the paper’ (‘It is John who read the paper’) [CP [FocP John-i [TP (John-i) [vP (John-i) [VP ku nonmwun-ul] ilk] ess]-ta] Nominative and accusative Case markers, -i /-ka and -ul, overlap with the focus markers in morphological form in Korean (Schütze 1997, Jung 2001). In (26), John assumes the proper morphological form that indicates both nominative Case and focus. The subject NP with FOCUS and the object NP with FOCUS raise to the multiple Specs of FocP. Note that the multiple focus phrases are possible in Korean (Choe 1995). (27) John-i ku nonmwun-ul ilk-ess-ta John-NOM/FOC the paper-ACC/FOC read-PAST-DEC ‘John read the paper’ (‘It is John who read the PAPER’) [CP [FocP John-i [FocP ku nonmwun-ul [TP (John-i) [vP (ku nonmwun- ul) [vP (John-i) [VP (ku nonmwun-ul)] ilk]-ess]-ta] 152 Chapter Seven The focus phrase with the –nun marker located below the higher TopP or FocP is contrastive focus in Korean. The contrastive focus reading is provided by the syntactic position. (28) John-i ku nonmwun-un ilk-ess-ta John-NOM/FOC the paper-CF (Contrastive Focus) read-PAST-DEC ‘John read the paper’ (It is John who read the paper, but not other papers.) [CP [FocP John-i [FocP ku nonmwun-un [TP (John-i) [vP (ku nonmwun) [vP (John-i) [VP (ku nonmwun) ] ilk]]-ess]]]-ta] In (28), ku nonmwun-un cannot be a topic, since it is located in the Spec of FocP. In case that ku nonmwun-un is located in the sentence initial position, the NP has the topic or contrastive topic reading as in (16). The particular interpretation is obtained by the position that the element occupies. Next, I would like to examine the cases where the object is overtly preposed. When the object NP with FOCUS is raised to the Spec of FocP, the focus reading is provided. The subject NP with nominative Case may be either neutral or focused. For the former reading the subject NP stays in the vP area (i), while for the latter reading it raises to the Spec of FocP (ii). (29) ku nonmwun-ul John-i ilk-ess-ta the paper-ACC/FOC John-NOM(i) (or FOC(ii)) read-PAST-DEC (i) ‘The paper, John read’ (‘It is the paper that John read’) [CP [[FocP ku nonmwun-ul [TP [vP (ku nonmwun-ul) [vP John-i [VP (ku nonmwun-ul)] ilk]]-ess]]-ta] (ii) ‘The paper, John read’ (It is the paper that JOHN read’) [CP [FocP ku nonmwun-ul [FocP John-i [TP (John-i) [vP (ku nonmwun-ul) [vP (John-i) [VP (ku nonmwun-ul)] ilk]]-ess]]]-ta] Sentence (29) is different from (11) since it has the focus reading. The same word order has different readings, the focus reading in (29) and the topic reading in (11), depending on the structural differences. With respect to the subject NP, (29i) shows the neutral reading, while (29ii) shows the focus reading. Accordingly, the same word order provides all three possibilities, (11), (29i), and (29ii), in interpretation depending on the syntactic position. The subject NP raises to the Spec of FocP and the object NP also raises to the Spec of FocP, providing the contrastive focus reading and the focus reading. Syntactic Encoding of Topic and Focus in Korean 153 (30) ku nonmwun-ul John-un ilk-ess-ta the paper-ACC/FOC John-CF read-PAST-DEC ‘The paper, John read’ (‘It is the paper that John, not others, read’) [CP [TopP [FocP ku nonmwun-ul [FocP John-un [TP (John) [vP (ku nonmwun-ul) [vP (John) [VP (ku nonmwun-ul)] ilk]]-ess]]]]-ta] The same word order has different interpretations depending on the syntactic position. Let us compare (30) with (31a,b). Sentence (19) is repeated in (31b). (31) a. ku nonmwun-ul John-un ilk-ess-ta the paper-ACC John-CF read-PAST-DEC ‘The paper, John read’ (‘As for the paper, It is John who read (it)’) [CP [TopP ku nonmwun-ul [FocP John-un [TP (John) [vP (ku nonmwun-ul) [vP (John) [VP (ku nonmwun-ul) ] ilk]]-ess]]]-ta] b. ku nonmwun-un John-un ilk-ess-ta the paper-TOP/CT John-CF read-PAST-DEC ‘The paper, John read’ (‘As for the paper, it is John who read (it)’) (‘As for the paper (not for other papers)), it is John (not others) who read (it)’) [CP [TopP ku nonmwun-un [FocP John-un [TP (John) [vP (ku nonmwun) [vP (John) [VP (ku nonmwun)] ilk]]-ess]]]-ta] In (30) and (31), John has the contrastive focus reading, located below the topic or the focus. In (30), the two NPs are focused, undergoing the multiple focus movement. In (31), the sentence initial NPs are positioned in the Spec of TopP, providing the same topic reading. In spite of the different morphological markers for the sentence initial NPs, (31a,b) could have the same interpretation due to the same syntactic position taken. The slight difference in reading comes from the possibility of the contrastive topic reading with ku nonmwun-un in (31b). Sentence (30) has a different interpretation from (31a) despite the same morphological marker –ul attached to the sentence initial NP: the former is characterized with the focus reading and the latter, with the topic reading. To summarize, interpretations differ based on the syntactic position that the element occupies. Anything that occupies the Spec of FocP has the focus reading. 5. Long-Distance Preposing as Topicalization and Focalization Long-distance preposing in Korean/Japanese has been analyzed as long-distance scrambling. Saito (1992) and Bošković and Takahashi 154 Chapter Seven (1998) argue that long-distance scrambling in Japanese is semantically vacuous, completely undone in LF. (32) Daremo-ni dareka-ga [Mary-ga t atta to] omotteiru everyone-DAT someone Mary-NOM met that thinks ‘Everyone, someone thinks that Mary met’ ∃ > ∀; *∀ > ∃ With the quantifier preposing, movement apparently seems to be undone in LF. The same applies to the Korean counterpart. However, the undoing effects do not seem to be always true with non-quantifier preposing. (33) I chayk-ul Mary-ka [John-i t this book-ACC (or FOC) Mary-NOM John-NOM sa-ss-ta-ko ] sayngkakha-n-ta buy-PAST-DEC-COMP think-PRES-DEC ‘Mary thinks that John bought this book’ i) ‘As for this book, Mary thinks that John bought (it)’ ii) ‘This book, Mary thinks that John bought (it)’ In addition to the focus reading (33ii), the topic reading (33i) is also added to be more precise. As we noted earlier, the element without the topic marker can function as a topic. The above sentence can thus be interpreted as (33i) and (33ii). Such long-distance topicalization and focalization cannot be undone in LF. Thus I propose that the long-distance preposing is not long-distance scrambling that brings the undoing effects in Korean. As evidence, the possible questions for the preposed sentence are different from those of the non-preposed one. The non-preposed sentence can serve as an answer for the following questions (English counterparts of the Korean sentences are provided): (34) a. Answer: Mary thinks that John bought this book b. What does Mary think that John bought? c. What does Mary think/do? d. Who does Mary think bought this book? Sentence (34a) is the non-preposed English counterpart of the Korean sentence (33). To have the proper answer in (34a), questions such as (34b,c, and d) are possible. Note that focus can be assigned in situ with stress in PF in Korean without preposing though it undergoes the LF movement to be properly licensed and interpreted as our proposal. Concerning the preposed sentence with the topic interpretation in (33i), the questions for this answer are as follows. Syntactic Encoding of Topic and Focus in Korean 155 (35) a. Answer: As for this book, Mary thinks that John bought it. b. As for this book, who does Mary think bought it? c. As for this book, what does Mary think (do)? d. *What does Mary think that John bought? Sentence (35a) is the preposed English counterpart of (33i) with the topic interpretation. (35b,c) can be possible questions for the answer (35a). Sentence (35d) cannot be a possible question for the same answer since what is asked is focus rather than topic. The preposed sentence with the focus interpretation in (33ii) could have the following questions: (36) a. Answer: This book, Mary thinks that John bought it. b. What does Mary think that John bought? c. *Who does Mary think bought it? d. *What does Mary think/do? In focalization, (36b) is the only possible question for the answer (36a), but (36c,d) are not. We thus know that topicalization and focalization are different from the so-called long-distance scrambling that can be completely undone. Choe (1995) also states that the long-distance preposing always triggers the focus reading. In our analysis, we have seen that not only the focus reading but also the topic reading is obtained by long-distance preposing. The term, (long-distance) scrambling, is thus called into question to define the precise nature of preposing in Korean. Evidence that long-distance preposing in Korean is not long-distance scrambling comes from anaphor binding. It is well known that clause internal scrambling remedies the violation of Condition A in Korean/Japanese. Lee (1990, 1993) and Yoon (1991) argue that long-distance scrambling also enhances the violation of Condition A, creating a new binder. (37) a.*na-nun [seloi-ka [Chelswu-wa Youngswu]i -lul I-TOP each other-NOM Chelswu-and Youngswu-ACC Miwueha-n-ta-ko] sayngkakha-n-ta hate-PRES-DEC-COMP think-PRES-DEC 'I think that each other hates Chelswu and Youngswu' b. ? [Chelswu-wa Youngswu]i-lul na-nun [seloi-ka ti Chelswu-and Youngswu-ACC I-TOP each other-NOM Miwueha-n-ta-ko] sayngkakha-n-ta hate-PRES-DEC-COMP think-PRES-DEC Mahajan (1990) and Saito (1992, 2004) judge the Japanese counterpart of (37b) ungrammatical. Unlike Japanese data, (37b) should not be undone in Korean, being quite natural to most Korean native speakers. Unlike 156 Chapter Seven Japanese, the violation of Condition A is much improved with the movement across the clause boundary. This means that the undoing effects do not apply in Korean. The long-distance preposing is not a semantically vacuous movement in Korean. Then the term, scrambling, should not be used for long-distance preposing in Korean. In sum, we have seen that the overt object preposing requires different questions from the non-preposed sentence in order to be the proper answer, and the violation of Condition A is improved with the preposed object antecedent across the clause boundary. In this respect, long-distance scrambling in Korean is not scrambling in the real sense, but topicalization or focalization that brings semantic import and thus cannot be undone. 6. Scope Phenomenon In the previous section, I have shown that the apparent long-distance scrambling is not scrambling but topicalization or focalization. However, it cannot be denied that there are undoing effects in the long-distance preposing sentences with regard to the scope phenomenon. (38) Motun salam-ul nwukwunka-ka [Mary-ka t mana-ss-ta-ko] everyone-ACC someone Mary-NOM meet-PAST-DEC-COMP sayngkakha-n-ta think-PRES-DEC ‘Everyone, someone thinks that Mary met’ ∃ > ∀; *∀ > ∃ The scope phenomenon apparently supports the undoing effects in (38). The sentence exhibits that the movement seems to be undone to the original position since the universal quantifier always takes narrow scope in spite of preposing. I suggest that such undoing effects with respect to scope change are due to the scope reconstruction effects, not due to the whole NP reconstruction that leads to the semantic vacuity. The copy left in the original place based on the copy theory (Chomsky 1995) is involved in the scope effects. The preposed NP is responsible for the focusing effects. Thus I argue that the long-distance preposing of a quantifier brings the scopal vacuity, not the total semantic vacuity. Apart from the scope phenomenon, I show that the long-distance preposing with quantifiers has some semantic effects as focusing (see below). (39) a. Everyone, someone thinks that Mary met b. Who does someone think Mary met? c. *What does someone think/do? Syntactic Encoding of Topic and Focus in Korean 157 d. *Who thinks that Mary met everyone? For the answer (39a), (39c,d) cannot be the questions, but (39b) can. That is why (39a) does not have the same interpretation as the non-preposed sentence, the sentence with undoing effects. I have discussed long-distance preposing in Korean and argued that it is not semantically vacuous and thus must not be undone unlike Japanese. It has been shown with preposing of non-quantified NPs in the previous section. In this section I have shown that the apparent undoing effects with long-distance preposing of quantifiers are due to the scope reconstruction effects and the copy left, affecting part of the interpretation only. The scopal interpretation is part of interpretation, not the whole NP interpretation. Though the long-distance preposed sentence is scopally vacuous, it does not mean that it is semantically vacuous: scopal vacuity does not lead to the total semantic vacuity. The semantic effects by focalization are still present in the sentence with long-distance preposing of quantifiers when sentences are tested by appropriate questions. This way the same analysis applies to the long-distance preposing of the non- quantified NPs and the quantified NPs. The same focusing effects are held for both, but the scopal vacuity is unique to the quantified NPs only. Due to the focusing effects, they are not semantically vacuous, contra Saito (1992) and Bošković and Takahashi (1998). 7. Scrambling So far we have seen that the apparent long-distance scrambling in Korean must be analyzed as topicalization or focalization. The apparent undoing effects are subject to the scope reconstruction effects that contribute to part of the interpretation. Then the question is how scrambling in the real sense is obtained. First of all, scrambling without semantic import occurs in PF, not in LF. The place for scrambling is thus the PF component, not in syntax. Next, the clause internal preposing with informational focus could be scrambling (PF-scrambling). The following is a case of scrambling with no overt focus or topic movement in spite of the word order change. The subject NP and the object NP are scrambled within the vP area. (40) a. John-i ku nonmwun-ul ilk-ess-ta John-NOM the paper-ACC read-PAST-DEC ‘John read the paper’ [CP [TP [vP John-i [VP ku nonmwun-ul ] ilk]-ess]-ta] 158 Chapter Seven b. Ku nonmwun-ul John-i ilk-ess-ta the paper-ACC John-NOM read-PAST-DEC ‘The paper, John read’ [CP [TP [vP ku nonmwun-ul [vP John-i [VP (ku nonmwun-ul)] ilk]]-ess]-ta] In (40), there is no overt focus movement, though informational focus lies on the lowest element by NSR. I suggest that this is a true case of PF- scrambling in Korean. See more examples below. (41) a. John-i caki-lul salangha-n-ta John-NOM self-ACC love-PRES-DEC ‘John loves (him)self’ b. Caki-lul John-i t salangha-n-ta Self-ACC John-NOM love-PRES-DEC ‘(him)self, John loves’ Sentence (41b) shows the undoing effects in terms of the binding theory, providing the correct binding in the original place. It is a case of scrambling when the word order change occurs within vP at PF. However, when caki-lul occupies the Spec of FocP or the Spec of TopP, it induces the focus reading and the topic reading respectively. (42) a. *Selo-uy sensayngnim-i kutul-ul salangha-n-ta each other-GEN teacher-NOM them-ACC love-PRES-DEC ‘each other’s teacher loves them’ b. Kutul-ul selo-uy sensayngnim-i t salangha-n-ta them-ACC each other-GEN teacher-NOM love-PRES-DEC Sentence (42b) shows the improvement of Condition A by preposing. The preposing of them must occur in syntax as an A-movement providing the binder, and thus cannot be undone. The preposed NP has the neutral reading, the non-focused/ non-topicalized reading as a case of scrambling. The focus reading and the topic reading are also possible when the NP kutul-ul occupies the appropriate syntactic positions. Next, the clause internal quantifier preposing could belong to scrambling. (43) Motun salam-ul nwukwunka-ka t salangha-n-ta everyone-ACC someone-NOM love-PRES-DEC ‘Everyone, someone loves’ ∃ > ∀; ∀ > ∃ Syntactic Encoding of Topic and Focus in Korean 159 Sentence (43) shows the wide and narrow scope of existential quantifiers. With the existential quantifier taking wide scope, the undoing effects are obtained, which is a case of scrambling. With the universal quantifier taking wide scope, the undoing effects are not obtained, which is a case of syntactic A-movement. The non-focused/ non-topicalized neutral reading is obtained when scrambling occurs within vP. Focalization or topicalization effects are manifested when the universal quantifier moves to the Spec of FocP or to the Spec of TopP. Data without focalization or topicalization are a case of real scrambling. Either the subject or the object can be raised to the sentence initial position. Scrambling in the real sense is actually PF-scrambling with no semantic import. The preposing with focalization and topicalization in the previous sections, however, has the semantic import with focality and topicality, which cannot be characterized as scrambling. The different types of preposing that are found in Korean can be supported by Russian data in Bošković (2007). He suggests that Slavic languages could have scrambling, topicalization and focalization. Unlike Korean, he shows that preposing across the clause boundary could involve scrambling in addition to topicalization and focalization. The following sentence, quoted from his work, provides three way ambiguous readings by fronting the embedded object. (44) Ivanai tvrdis da ona voli ei Ivan-ACC you-claim that she loves ‘You claim that she loves Ivan’ ‘Ivan, you claim that she loves’ (Bošković 2007: 9) In (44), Ivan could be undone as scrambling, or it is interpreted as a topic or a focus. Concerning Korean, we can test possible readings with clause internal preposing, not with long-distance preposing. Long-distance preposing in Korean always triggers focusing or topicalization effects so that it cannot have scrambling based on our analysis. Sentence (40b) is repeated below to test possible interpretations caused by scrambling, topicalization, and focalization. (45) ku nonmwun-ul John-i ilk-ess-ta the paper-ACC John-NOM read-PAST-DEC i) ‘The paper, John read (it)’ [CP [TopP [FocP [TP [vP ku nonmwun-ul [vP John-i (ku nonmwun- ul)] ilk]]-ess]]]-ta] ii) ‘As for the paper, John read (it)’. 160 Chapter Seven [CP [TopP ku nonmwun-ul [TP [vP(ku nonmwun-ul) [vP John-i (ku nonmwun-ul)] ilk]]-ess]]-ta] iii) ‘The paper, John read (it)’ (‘ It is the paper that John read’) [CP [FocP ku nonmwun-ul [TP [vP (ku nonmwun-ul) [vP John-i (ku nonmwun-ul)] ilk]]-ess]]-ta] iv) ‘The paper, John read (it)’ (‘It is the paper that John, not others, read’) [CP [FocP ku nonmwun-ul [FocP John-i [TP (John) [vP (ku nonmwun-ul) [vP (John) (ku nonmwun-ul)] ilk]]-ess]]]-ta] The preposed element ku nonmwun-ul could be completely undone in (45i) which is a case of scrambling. The preposed object is interpreted as a topic in (45ii). It could be interpreted as a focus when focus falls on the preposed element as in (45iii). Two focus phrases are possible as in (45iv) since the multiple foci are allowed in Korean. Depending on the syntactic positions, all four interpretations are licensed. Like the Russian sentence above, preposing in Korean, clause internal preposing in particular, has ambiguous readings and the ambiguity results from the syntactic positions that the elements occupy. 8. Conclusion First, this analysis demonstrates the syntactic differences between topicalization, focalization, and scrambling. Second, the topic reading is always obtained in the Spec of TopP with the topic marker –nun. The topic reading is also obtained without the topic marker –nun as long as the NP (NP-ka /NP-lul) is raised to the Spec of TopP. Third, the focus reading is obtained by raising to the Spec of FocP. Differently from Rizzi’s (1997, 2004) structure, the multiple focus phrases are possible in Korean. Fourth, when the NP with –nun marker is located below TopP or FocP (below the most prominent NP), it has the contrast focus reading. From this we can see that discourse information is encoded in the syntactic positions at the left periphery in Korean. Finally, this analysis introduces a new type of analysis for scrambling. On one hand, the long-distance preposing always brings some semantic effects such as focalization and topicalization in Korean: scopal vacuity does not lead to the total semantic vacuity. On the other hand, the clause internal preposing provides ambiguous interpretations including scrambling: PF scrambling with no semantic import, focalization, and topicalization. Syntactic Encoding of Topic and Focus in Korean 161 Notes 1. A different version of this paper has also been published in Studies in Generative Grammar vol.18, No.1 (p.137-171) in Korea in February 2008 with the title Left Periphery in Korean: Topicalization, Focalization, and Scrambling. References Aoun, J. (1985). A grammar of anaphora. Cambridge, MA: MIT Press. Bošković, Z. (2007). Scrambling. To appear in T. Berger, K. Gutschmidt, S. Kempgen, & P. Kosta. (eds.) The Slavic Languages. Berlin: Mouton de Gruyter Bošković, Ž.& Takahashi, D. (1998). Scrambling and last resort, Linguistic Inquiry, 29: 347-366. Choe, H.-S. (1995). “Focus and Topic movement in Korean and licensing” in K. É. Kiss (ed.) Discourse configurational languages. Oxford: Oxford University Press: 269-334 Choi, H.-W. (1997). “Topic and focus in Korean: The information partition by phrase structure and morphology” in H.-M. Sohn & J. Haig (eds.) Japanese/Korean Linguistics 6. Stanford: CSLI: 545-562. Chomsky, N. (1977). “On wh movement” in P. Culicover, T. Wasow, & A. Akmajian (eds.) Formal syntax. New York: Academic Press: 71- 132. —. (1986). Knowledge of language: Its nature, origin and use. New York: Praeger. —. (1995). The minimalist program. Cambridge, MA: MIT Press. —. (2000). “Minimalist inquiries: The framework” in R. Martin, D. Michaels, & J. Uriagereka (eds.) Step by step: Essays on minimalist syntax in honour of Howard Lasnik. Cambridge, MA: MIT Press.: 89- 155. —. (2001). “Derivation by phase” in M. Kenstowicz (ed.) Ken Hale: A Life in Language. Cambridge, MA: MIT Press: 1-52. —. (2005). “On phases”. To appear in R. Freidin, C.P. Otero & M.L. Zubizaretta (eds.), Foundational Issues in linguistic theory. Cambridge, MA: MIT Press. —. (2006). Approaching UG from below. Ms. MIT. Cinque, G. (1993). “A null theory of phrase and compound stress”. Linguistic Inquiry 24: 239-298. É. Kiss, K. (1998). “Identificational focus versus informational focus”. Language 74: 245-273. 162 Chapter Seven Han, C.-H. (1998). “Asymmetry in the interpretation of –(n)un in Korean” in N. Akatsuka, H. Hoji, S. Iwasaki, S.-O. Sohn, & S. Strause (eds.) Japanese Korean Linguistics, Vol. 7. Stanford: CSLI Publications. Hornstein, N. & Weinberg, A. (1990). “The necessity of LF”. The Linguistic Review 7(2): 129-168. Jung, Y.-J. (2001). “Multiple Case marking and its PF interpretation”. Korean Journal of Linguistics 26: 239-268. Larson, R. (1988). “On the double object construction”. Linguistic Inquiry 19, 335-392. Lee, E.-J. (1992). On the extended projection principle. Doctoral dissertation, University of Connecticut, Storr. Lee, Y.-S. (1990). “Investigation of scrambling in Korean”. Proceedings of the International Conference in Korean Linguistics. Hanshin, Seoul. —. (1993). Scrambling as Case driven obligatory movement. Doctoral dissertation. University of Pennsylvania. Li, C. & Thompson, S. (1976). “Subject and topic: A new typology of language” in C. Li (ed.) Subject and topic. New York: Academic Press: 457-489. Mahajan, A. (1990). The A-A' distinction and movement theory. MIT Dissertation. Cambridge, MA: MITWPL. May, R. (1985). Logical form. Cambridge, MA: MIT Press. Pollock, J.-Y. (1989). “Verb movement, universal grammar and the structure of IP”. Linguistic Inquiry 20, 365-424. Rizzi, L. (1997). “The fine structure of the left periphery” in L. Haegeman (ed.) Elements of grammar. Dordrecht: Kluwer Academic Publisher: 281-337. —. (2004). “Locality and left periphery” in A. Belletti (ed.) Structure and beyond: The cartography of syntactic structures, Vol. 3. Oxford: Oxford University Press: 223-251. Saito, M. (1985). Some asymmetries in Japanese and their theoretical implications. MIT Dissertation. Cambridge, MA: MITWPL. —. (1989). “Scrambling as semantic vacuous A'-movement” in M. Baltin & A. Kroch (eds.) Alternative conceptions of phrase structure. Chicago: University of Chicago Press: 182-200. —. (1992). “Long distance scrambling in Japanese”. Journal of East Asian Linguistics 1: 69-118. —. (2004). “Japanese scrambling in a comparative perspective” in D. Adger, C. de Cat, & G. Tsulas (eds.) Peripheries : syntactic edges and their effects. Dordrecht: Kluwer Academic Publishers: 143-163. Syntactic Encoding of Topic and Focus in Korean 163 —. (2005). “Further notes on the interpretation of scrambling chains” in H. van Riemsdijk, J. Koster, & H. van der Hulst (eds.) The free word order phenomenon. Berlin, New York: Mouton de Gruyter.: 335-376. Schütze, C. (1997). INFL in child and adult language: Agreement, Case and licensing. Doctoral dissertation, MIT. —. (2001). “On the nature of default Case”. Syntax 4.3: 205.-238. Yoon, J.-M. (1991). The Syntax of A chains: A typological study of ECM and scrambling. Doctoral dissertation, Cornell University. CHAPTER EIGHT SYNDROMES OF MEANING: EXPLORING PATTERNED COUPLING IN A NSW YOUTH JUSTICE CONFERENCE MICHELE ZAPPAVIGNA, PAUL DWYER AND J R MARTIN 1. Introduction This chapter focuses on how two related meanings in a text may be grouped into larger patterns. We refer to the relationship between these meanings as a “coupling” (Martin, 2000) and to patterns of couplings as “syndromes”. One motivation for this research is to contribute to work in Systemic Functional Linguistics (SFL) on modelling instantiation, the meaning potential that is activated in a text. The other main motivation is to offer a qualitative perspective on youth justice conferencing that considers the process of the conference rather than simply the outcome of that conference. Most research has taken the latter approach (Van Ness, Maxwell & Morris, 2001; Palk, Hayes & Prenzler, 1998; Trimboli, 2000), considering, for example, the extent to which conferences are successful in terms of quantitative measures such as reductions in recidivism. This work typically uses surveys or participant observation to draw conclusions. In contrast, we undertake detailed analyses of the discourse that occurs in the conference using small-scale corpora of transcribed talk. Ultimately, the two approaches to inquiry will be complementary. We begin with a brief introduction to youth justice conferencing in New South Wales and a description of the specific conference analysed. An overview of Systemic Functional Linguistics (SFL) and appraisal theory follows, explaining how they model instantiation. The idea that meanings coupled along the cline of instantiation may be involved in larger textual patterns is introduced. We review related ideas from the Syndromes of Meaning 165 literature in the area of “systemic probability”, that is, ideas about how to determine whether particular meanings are related and measure the strength of association. The particular syndrome that we proceed to analyse occurs in the talk of the mothers of the young persons in the conference. 2. Youth Justice Conferencing in New South Wales New South Wales Youth Justice Conferences (YJC) are meetings held as an alternative to formal proceedings in the Childrens’ Court for sentencing a young person (YP) who has committed an offence. They are typically held in a seated circle formation in a PCYC or some other public venue such as a room in a council chambers. The conferences are run by a Convenor who acts as a facilitator and include as active participants the young person, the victim of the crime or a relevant community representative, supporters for both the young person and the victim, and, often, a Youth Liaison Officer, trained in working with young people. The Young Offenders Act 1997 (NSW) which defines the scope and operation of conferencing asserts that conferences should: …deal with alleged offenders in a way that: (i) enables a community based negotiated response to offences involving all the affected parties, and (ii) emphasizes restitution by the offender and the acceptance of responsibility by the offender for his or her behaviour, and (iii) meets the needs of victims and offenders. The text analysed in this chapter is the transcript of a YJC broadcast on ABC Radio National during the Radio Eye program in a story entitled “Offending Behaviour”1. The conference dealt with the vandalising of a vacant house in a housing estate by two teenage boys, referred to as Young Person A (YPA) and Young Person D (YPD). The boys inflicted $50,000 of damage to the property and, after admitting their guilt, took part in a conference that sentenced them to 25 hours voluntary work and participation in sport at a Police Citizen’s Youth Club (PCYC). This chapter focuses on the discourse of the mothers of the two boys. During the conference the mothers spoke about their reactions to the offence and how they felt about themselves, their parenting and their children in light of their sons’ behaviour. Their discourse is of interest in light of claims that parents are “are likely to feel that their parenting skills are on trial in a conference” (Prichard, 2002). Mothers, in particular, may feel that they are held responsible for their child’s actions (Cook, 2006). 166 Chapter Eight Such feelings of culpability appeared to be present, although ambivalent, in the conference analysed in this paper. Thus the mothers’ talk was a likely candidate for displaying patterns of evaluative meaning that would provide an interesting case study for analysing the functioning of couplings and syndromes in spoken discourse. 3. Introducing Systemic Functional Linguistics (SFL) The analyses undertaken in this chapter use a theory of language, Systemic Functional Linguistics (SFL), that is tailored to answering questions about how meanings function within the particular contexts in which they are made. SFL draws its ideas about context from the anthropologist Malinowski who introduced the notion of an ethnographic theory of language. Malinowski (1935) argued that “the real linguistic fact is the full utterance within its context of situation” (p.11). SFL appropriates two broad types of context from Malinowski: the context of culture and the context of situation. The context of culture is a semiotic system, a system of meaning, that may be thought of as providing an environment for language, another semiotic system. When using language, a person makes choices from the options that the culture makes available to them. They make these choices within a particular circumstance, the context of situation. For example, a young person in a YJC makes choices that are influenced by the environment in which they have grown up, often an environment of educational and economic disadvantage (the context of culture). They also make choices that depend upon the particular circumstances that they are in, in this case a seated circle formation requiring verbal communication (the context of situation). As well as being functional, SFL is “systemic” in the sense that it models meaning as selections within system networks. System networks are networks of interrelated options that, instead of being organised as catalogues of structures, are organised paradigmatically in terms of “what could go instead of what” (Halliday & Matthiessen, 2004, p.22). System networks represent these options at increasing levels of delicacy. For example, Figure 1 is the system network for ATTITUDE. The choices possible in this network involve “three semantic regions covering what is traditionally referred to as emotion, ethics and aesthetics” (Martin & White, 2005, p.42). Syndromes of Meaning 167 Figure 1: System network for attitude. Part of this network is the system of AFFECT: language about positive and negative feelings. As the network suggests, meanings about happiness, security and satisfaction are resources that a speaker might deploy when talking about emotions. For example, the following exchange between the Convenor and YPA involves talk about the impact of the crime on the young person’s emotions: Convenor: How did you feel then? YPA: Scared, worried...what me dad would think and that. 168 Chapter Eight Applying the network in Figure 1, the selection underlined above is an example of ATTITUDE: AFFECT: SECURITY. This means that it is language about emotions, specifically those to do with confidence or anxiety. The theory applied in this chapter about how such evaluative meanings function in texts is appraisal theory (Martin & White, 2005). Appraisal theory deals with how interpersonal meanings such as the construction of stances or personae are enacted in language. The three main evaluative systems defined by this theory are ATTITUDE (Figure 1), GRADUATION (Figure 4) and ENGAGEMENT. The first deals with evaluations relating to feelings, judgements and assessments, while the second grades evaluation, and the last expresses the sources of authority in the text. 4. Instantiation The system network introduced in Figure 1 represents the potential choices that a speaker might make, that is, it models “meaning potential”. When we analyse an actual text produced in a particular context, for example, the transcript of a YJC, the data that we annotate is the “instantiation” of that meaning potential. So, in the previous example, scared and worried instantiate meaning potential about AFFECT. SFL theorises the relationship between potential and actual meanings along what is referred to as the “cline of instantiation” (Figure 2). This cline models aggregation of instances (enacted meanings). For example it represents texts types as more generalised instances of texts, in the sense that a text type is a clustering of texts that bear some linguistic similarity. We move up the cline toward the system which is the most generalised instance. In other words, all known instances of texts are part of the system. As the following section will elaborate, coupling between different meanings can be modelled as occurring along this cline. Figure 2: The cline of instantiation (Martin & White, 2005, p.25). Syndromes of Meaning 169 5. Coupling and syndromes The term “syndrome” is most often used in medicine where it refers to a collection of associated symptoms. More generally, a syndrome might be conceived as clusters of associated signs. The relationship between the signs is the way they co-construe more complex meaning. Systemic Functional Linguistics (SFL) has a long history of analysing patterning in texts (Hasan, 1985; Halliday & Matthiessen, 2001; Martin & Rose, 2003). These patterns are typically represented as choices in system networks. However, if we are interested in associations between instantiated meanings, using instantiation in the sense introduced in the previous section, we need to consider how meanings bind across systems. In so doing we enter a further dimension of complexity, that of the co-rendering of patterns, or in other words, patterns of coupling. Conceptualising the association of patterns of multiple variables is difficult. To illustrate the clustering involved in a syndrome we will begin with the most delicate example: coupling. Martin (2000) first introduced the notion of coupling to explain humour “as involving discordant couplings – either between appraisal selections and what is being appraised, or among the appraisal variables themselves” (p.164). He defines coupling as the binding of two meanings at any point along the cline of instantiation (see Martin, in press). In this sense a coupling is not “flat” because we are interested in more than the collocation of lexical items, we are also interested in bindings at different levels of generalization. So for example, bindings across different discourse semantic instances within and across texts, texts types and registers, depending on the scope of linguistic data available (e.g. a single text or a corpus of texts). Using Martin’s (in press) analogy of the double helix formation we might visualise this simple coupling along the cline of instantiation (Figure 3). The double helix structure captures the property that meaning potential does not bind directly into simple couplings but is instantiated through a process that involves interaction with co-text and context, and, is, in this sense, “constantly negotiated”2. As Figure 3 suggests, the discourse semantic space in which a coupling might occur diminishes as the meaning potential available is constrained moving down the cline. 170 Chapter Eight Figure 3: Coupling along the cline of instantiation. Coupling is both syntagmatic, since items that are coupled are likely to occur very near each other logogenetically, and paradigmatic, since patterns of the same type of coupling, for example, couplings of A + B rather than B + C, are likely to reoccur across a text. Following this logic, we might think of a sign in a text as involved two fundamental relationships: association with other signs involved in a particular meaning through coupling, and association with other signs of the same “type”, that is, signs of the same node in a system network. For example, taking the instance introduced in the background section of this chapter, “scared” and “worried” are associated by being part of the AFFECT system. They might also often occur with selections from other systems. For example, affect of this kind might be intensified, so we might see selections such as “very scared”, “extremely worried”. These would represent associations between the system of GRADUATION (Figure 4) and AFFECT (Figure 1). Specifically, a relationship exists between “ATTITUDE: AFFECT: SECURITY: NEGATIVE” and “GRADUATION: FORCE: INTENSIFICATION: QUALITY: NEGATIVE”. This type of association has been theorised by Bednarek (2006) as an “evaluative collocation”. Bednarek adopts a parameter based Syndromes of Meaning 171 theory of evaluation that, while different to a system-based approach, appears concordant with the conceptualisation of coupling suggested here. Figure 4: System network for GRADUATION. 6. Tightly or loosely coupled: A brief history of systemic probability The idea that discourse semantic and lexicogrammatical features may bind with each other in texts precipitates questions about the likelihood of particular couplings and how tendencies in coupling might be modelled. As Nesbitt and Plum (1988) have suggested, the “grammar of language is not only the grammar of what is possible but also a grammar of what is probable” (p.9). Within a single text or a corpus of texts some patternings are more likely than others: …what the corpus is telling us is that all options are not equal; there are tendencies. And such tendencies are uncovered from what speakers and writers, recorded in a corpus, as a representative set, are observed to be doing. (Tucker, 2006, p.91) What is coupled, or “co-instantiated”, is paradigmatic in the sense that it depends upon the systemic probabilities that are defined in the relevant 172 Chapter Eight system networks, but it is also syntagmatic as it becomes the product of the process of instantiation. For example, consider the following generic system network with four complementary systems that each involve two choices (Figure 5): Figure 5: A generic system network. In this example a coupling of E + G is the product of the two systemic probabilities: (0.1 x 0.5) = 0.05, that is, a probability of 5% that this particular coupling will occur. The probability that a particular cluster or syndrome will occur is the product of the constituent systemic probabilities. So for a syndrome involving F, G, I, and L this would be: 0.9 x 0.5 x 0.1 x 0.5 = 2.25% Syndromes of Meaning 173 However, if we look at Figure 5, we might say that there is a 5% chance of seeing G+I (0.5 x 0.1) and a 45% chance of seeing G+J (0.5 x 0.9). But, if we know more about the actual instantiation it could be true that every time you see G, you will also see I . In this scenario there would be more G+I and less G+J to round it all out (Table 1). Table 1: An example of an instantiation of the generic system network Instance 1 2 3 4 5 6 7 8 9 10 number Couple G G G G G H H H H H I J J J J J J J J J As Table 1 shows, the probabilities that would seem to be instantiated here are: G+I = 10% G+J = 40% H + J = 50% B + I = 0% These probabilities do not conform to those defined in the system network. This is an example where additional information about the relationships is required, and an example of why we need to engage with statistical measures such as correlation when talking about instantiation. Within corpus linguistics, measures of association that are commonly used to assess the less complicated domain of lexical collocation include t scores, mutual information and log-likelihood. Current work to measure more complex types of patterning occurs in the domain of computational linguistics. Research in the sub-discipline of text categorisation has suggested that clusterings of systemic features may be used to automatically group texts into genres or other categorisation schemes. Machine learning on “functional lexical features” has been used in the area of stylistic text classification to differentiate texts in terms of, for example, the kind of appraisal language that they employ (Argamon, Whitelaw, Hota, Garg & Levitan, 2007). “Systemic features”, that is, lexical items that are the most delicate manifestation of particular system networks, have been used in tasks such as financial scam classification where they have been employed to separate documents into registers based on 174 Chapter Eight variance in levels of interpersonal distance in texts (Whitelaw, Patrick, & Herke-Couchman, 2006). Couplings are analogue rather than digital. In other words, a coupling is not an “on/off” option and may be represented as weighted. Something is strongly coupled if it frequently co-occurs and loosely coupled if the logogenetic patterning is sparse. We cannot determine the strength of a coupling without an understanding of the co-text. This understanding may be the qualitative judgement of the linguist or the combination of this judgement with some specification of the tendencies in the text. A syndrome is present in a text when particular permutations and combinations are more likely than others. A permutation is a combination where ordering is important. For example, information about the logogenesis of the text in Figure 1 can be captured if we considered when the coupling occurs as the text unfolds. In this text there appears to be more coupling in the beginning phase than the conclusion. The permutations and combinations are likely to be independent for a text from a different genre or within a corpus of texts from multiple, distinct genres. These other texts are likely to display their own particular fingerprint of coupling. Figure 6: Considering the relative position of a coupling in a text. However, making the corpus “representative” is clearly problematic and involves foregrounding the assumptions that have been made during corpus design. Our capacity to model and measure such tendencies, is however, limited. If the modelling technique is corpus-based, severe restrictions are encountered due to the limited features, typically purely lexical items, that can be automatically detected in corpora. Manual analysis may be used to guide the various, though limited, kinds of automatable analyses of corpora that are possible (Matthiessen, 2006). Syndromes of Meaning 175 Even if we adopt a more qualitative approach we confront the problem of how to effectively explore the complexity without a tool for visualising the couplings that we have manually annotated in the text. Within SFL, work on probabilistic profiling of grammar was begun by Halliday (1991) in a quantitative study that argued that the selection of features in a system is either equi-probable or skewed. According to Halliday (1995), modelling tendency with the theoretical construct known as probability “is just the technicalizing of "modality" from everyday grammar” and our ability to comment on such tendency “is what makes it possible for natural language to function as metalanguage” (p.21). In this tradition, Matthiessen (2006) argued that “systemic profiles” produced using analyses of corpora offer a way of representing the patterning of particular texts, text types and the overall systemic potential of language. These profiles are “counts of selections with the system of language…at any point along the cline of instantiation” (Matthiessen 2006, p.104-5). Systems may instantiate simultaneously as “systemic intersections” (Matthiessen, 2006, p.114). For example, within a medium-sized corpus, Matthiessen (2006) explored the intersection of the AGENCY and PROCESS TYPE systems, determining the following: This sample suggests very clearly that ‘material’ and ‘effective’ attract one another: the only environment in which ‘effective’ is selected as often as ‘middle’ is that of ‘material’ clauses; and ‘material’ clauses account for roughly 75 per cent of all instances of ‘effective’ clauses (p.121). We might also consider the systems that are likely to repel each other. In this paper we entertain the notion that such attraction may be multifactorial, to use a term from genetics, because it is attributable to both the interaction of two linguistic features and their interaction with the co- text, just as inheritance is attributable to the interaction of two genes with each other and with their environment. 7. A syndrome of ambivalent culpability A syndrome is the recurrent co-instantiation of patterns of linguistic potential. The discourse of the two mothers in the conference enacted a syndrome of ambivalence about their level of responsibility regarding the behaviour of their sons. In other words, the mothers’ talk contained recurrent couplings of certain types of meanings involved in expressing culpability. These included meanings empathising with the young person and meanings asserting a kind of covert blame. The couplings typically involving the following types of evaluative meaning: 176 Chapter Eight • NEGATIVE AFFECT, e.g. “I was worried for him” [A’s Mother] • Speculation about the young person’s actions and mental state, e.g. “I think he sort of doesn't know what to think at the moment.” • Oscillation of EXPANSION and CONTRACTION in terms of ENGAGEMENT, e.g. “I just wanted to know what really went on, but um, when I came in and seen him I knew he was scared...” [A’s Mother] • POSITIVE JUDGEMENT about the young person to counter INVOKED NEGATIVE JUDGEMENT e.g. “…because D's always been such a... he's been a good boy, it's not in character for him to do something like that.”[D’s Mother] Consider, for example, a particular coupling of ideational and evaluative meaning that reoccurs in the discourse of YPA’s mother. As the mother talked about her son’s offence there was a tendency towards NEGATIVE AFFECT coupled with relational processes, that is, verbs that establish abstract relationships between concepts (Table 2). Table 2: Instances of NEGATIVE AFFECT in the talk of A’s Mother (relational processes underlined). I was scared for him I was disappointed He’s a bit confused He’s worried about what his father’s gonna think, what everyone else is gonna think I was worried for him Because I knew he’d be scared I was very angry I know he was scared Plus I was scared for him as well Using the visualisation introduced at the beginning of this chapter we might conceptualise this coupling along the cline of instantiation (Figure 7). As the following diagram suggests, at the most disaggregated end there is a coupling between attributive relational processes and evaluative meanings of disinclination. Viewed from the most aggregated end of the cline this represents a coupling of transitivity and attitude. Syndromes of Meaning 177 Figure 7: Coupling along the cline of instantiation. Extract 1 is a sample of A’s Mother’s discourse that shows this type of coupling of relational processes and negative affect in context. In this extract the mother positions herself as both having access to her son’s affectual experiences in clauses such as, “I knew he was scared”, and, positions herself closed off from knowledge about his motivations in clauses such as, “I just wanted to know what really went on”. As the discourse progresses, sequences of EXPANSION and CONTRACTION construe the mother as ambivalent about her capacity to know her son’s emotional state. When I first got the phone call I was worried for him because I knew he’d be scared. I was very angry, I just wanted to know what really went on, but um, when I came in and seen him I knew he was scared... Extract 1: Coupling of RELATIONAL PROCESSES with AFFECT (relational processes underlined, affect in bold) However, and perhaps more interestingly, the coupling “RELATIONAL PROCESS + AFFECT” is also coupled with MENTAL PROCESSES via projection. For example, A’s mother uses projection to speculate about her son’s thoughts and feelings (Extract 2). 178 Chapter Eight He's a bit confused and I think he sort of doesn't know what to think at the moment. Like he doesn't want to upset me, he's worried about what his father's gonna think, what everyone else is gonna think. Extract 2: A's Mother speculating about her son's thoughts and feelings This further coupling is shown in Figure 8. This diagram also should serve as an indication of the inadequacy of two-dimensions for representing associations between more than two variables. If we perhaps, for example, invoked a metaphor of volume then multiple couplings might appear like a cone shape along the cline (Caple, 2007). You can picture this by imagining the dotted lines spinning around the cline to form a cone shape (Figure 9). In this visualisation, multiple, related instances are contained within the volume of the cone and a cross-sectional slice represents a particular level of aggregation. Figure 8: Multiple couplings along the cline of instantiation. Syndromes of Meaning 179 Figure 9: Visualising couplings of multiple instances as a cone. The couplings in the mothers’ talk introduced thus far co-occurs with additional couplings that fall within a syndrome of talk about responsibility. For example, YPA’s mother displayed a tendency to positively evaluate her son. This appeared as an attempt to dispel the notion that he was a bad person and that her parenting, by implication, was at fault (Extract 3). ...it has had a deep effect on him. Like he's um...he's been a lot more helpful around the house and a lot better within himself, his attitude. And I think this is a good thing what's happening today. Extract 3: A's Mum positively evaluating her son The amount of understanding that the mothers chose to express about the inner workings of their sons was finely balanced. Presenting too much knowledge about their motivations might implicate the mothers in the criminal actions, however, presenting too little knowledge, or no empathy for the mental state of their sons might leave the mothers open to accusations of being a poor parent. The use of ENGAGEMENT resources such as “just” by both mothers seemed to function in the text to enact this balance. It was used as a means of tempering3 the type and extent of knowledge that the mothers “confessed” (Table 2). “Just” functions as a form of CONTRACTION within the ENGAGEMENT system. The engagement system deals with meanings relating to intertextuality, that is, about the sources of authority in the text. CONTRACTION functions to reduce the 180 Chapter Eight scope of a proposition. For example, the comment “I just wanted to know what really went on”, distances an imagined voice saying “you wanted to know much more than what really went on”. Table 3: Instances of “just” in the talk of the mothers I just said why? Why? I just couldn't understand why. I just think it will be good for him. I just wanted to know what really went on I think it’s just admitting that your children have made a mistake And I s'pose it's just a whole lot of mixed feelings from sorrow and sadness They’re both good kids in their own way just …and I hope for better things for them and The various instances of coupling outlined in this section form a cluster of meanings involved in the mothers’ construal of their sense of culpability for the crime that their sons have committed. The clustering appears to suggest an ambivalence about that responsibility that is manifest as a kind of defensiveness about the propriety of the two young persons. The cluster might be visualised as the syndrome in Figure 10, where each circle represents a tendency to instantiate features of that particular node in the relevant system network. The lines connecting the circles are intended to suggest the interrelatedness of the meanings in the cluster but they might be omitted for visual clarity. The limitation of a 2D drawing of this kind is that the unfolding of these types of meanings cannot be represented. A preferable representation would be a 3D swarm of instances where the relative distance between instances was a representation of the tightness or looseness of their coupling. Syndromes of Meaning 181 Engagement: Expand & Graduation: Contract tone down Relational process Affect: Mental Insecurity Process (Projection) Judgement: Social sanction Figure 10: A syndrome of ambivalent responsibility. 8. Conclusion Modelling instantiation is a current focus of systemic functional theory. Understanding how meaning potential is enacted in texts necessarily raises questions about how those instantiated meanings are related to each other in their functional context. System networks have provided a powerful means to visualise choice and meaning potential, however, they are unable to represent the multiple relations co-instantiated in a syndrome. Consider, for example, Figure 11 which summarises some of the couplings discussed in this chapter. The relationships marked with a dotted line show coupling as relating simultaneous systems. The systems are simultaneous in the sense that they are linked by an “and” relation based on the choices that have been made in the text. Relationships exist both within systems (e.g. within process type) and between systems (e.g. between process type and attitude). 182 Chapter Eight Figure 11: An example of binding within and between simultaneous systems. The dotted lines are required on the system network to represent associations that form between instances when the meaning potential laid out in the network is enacted. We might think of this as the network “disappearing” as the meaning potential is realised, and the dotted lines representing associations between actual items of meaning that are chosen. Determining the best way to represent a cluster of this kind is the topic of a future paper. The analysis undertaken in this paper is qualitative in the sense that the linguist has selected couplings that appear to be prominent in creating the patterning of particular meanings. Given an annotated corpus of adequate size and representativeness, it may be possible to automatically detect couplings defined by certain parameters. This type of corpus would be useful for assessing the validity of claims, Syndromes of Meaning 183 such as those made in this paper, that have been made on the basis of a small sample of linguistic evidence. Notes 1. This transcript is publicly available at http://www.abc.net.au/rn/arts/radioeye/crime/epis4.htm. 2. Knight, N. (2007) Personal communication. 3. For an account of tempering see Zappavigna, Dwyer & Martin (2007). References Argamon, S., Whitelaw, C.C., Hota, S., Garg, N., & Levitan, S. (2007). Stylistic Text Classification using Functional Lexical Features. Journal of the American Society for Information Science and Technology, 58(6): 802-822. Bednarek, M. (2006). Evaluation in Media Discourse. London; New York: Continuum. Braithwaite, J. (2003). Restorative Justice and Criminal Justice: Competing or Reconcilable Paradigms? In A. Von Hirsch, J. Roberts, A.E Bottoms, K. Roach & M. Schiff (Eds.) Studies in Penal Theory & Penal Ethics (195-217). Oxford: Hart Publishing. Caple, H. (2007). Instantiation, affiliation and the multimodal news story. Multimodal Discourse Analysis Research Seminar, 12 October. Sydney, University of Technology. Cook, K. (2006). Doing difference and accountability in restorative justice conferences. Theoretical Criminology 10(1): 107-124. Fry, B. (1997). Organic Information Design. Unpublished Doctoral dissertation, Massachusetts Institute of Technology. Halliday, M.A.K. (1991). Corpus Studies and Probabilistic Grammar. In K. Aijimer & B. Altenberg (Eds.) English Corpus Linguistics: Studies in Honour of Jan Svartvik (30-43). Harlow: Longman. Halliday, M.A.K. (1995). Fuzzy grammatics: a systemic functional approach to fuzziness innatural language. International Joint Conference of the Fourth IEEE International Conference on Fuzzy Systems and The Second International Fuzzy Engineering Symposium, Vol. 1 (9-26). Yokohama, Japan. Halliday, M.A.K. & Matthiessen, C.M.I.M. (2001). Construing Experience Through Meaning: A Language-based Approach to Cognition. NY: Cassel. 184 Chapter Eight Halliday, M.A.K. & Matthiessen, C.M.I.M. (2004). An Introduction to Functional Grammar, 3rd ed. London: Arnold. Hasan, R. (1985). Linguistics, Language and Verbal Art. Waurn Ponds, Victoria: Deakin University Press. Hoey, M. (2005). Lexical Priming: A New Theory of Words and Language. New York: Routledge. Knight, N.K. (2007). Personal communication. Sydney. Malinowski, B. (1935 [1991]). Coral Gardens and Their Magic: A Study of the Methods of Tilling the Soil, Volume II. London: Allen and Unwin. Martin, J. (2000). Beyond Exchange: Appraisal Systems in English. In S. Hunston & G. Thompson (Eds.) Evaluation in Text: Authorial Stance and the Construction of Discourse (142-175). Oxford: Oxford University Press. Martin, J.R. (in press) Tenderness: realisation and individuation in a Botswanan town. In N. Nørgaard (Ed.) Odense Working Papers in Language and Communication (Special Issue of Papers from 34th International Systemic Functional Congress), Odense. Martin, J. & Rose, D. (2003). Working with discourse: Meaning beyond the clause. New York : Continuum . Martin, J. & White, P. (2005). The Language of Evaluation: Appraisal in English. London and New York: Palgrave. Matthiessen, C. (2006). Frequency profiles of some basic grammatical systems: an interim report. In G. Thompson & S. Hunston (Eds.) System and Corpus: Exploring connections (103-142). London: Equinox. Nesbitt, C. & Plum, G. (1988). Probabilities in a systemic functional grammar: the clause complex in English. In R. Fawcett & D. Young (Eds.) New developments in systemic linguistics, Vol. 2. (8-9) London and New York: Pinter Publishers.Grammar, 3rd ed. London: Arnold Palk, G., Hayes, H., & Prenzler, T. (1998). Restorative Justice and Community Conferencing: Summary of Findings from a Pilot Study. Current Issues in Criminal Justice 10(2): 125-137. Prichard, J. (2002). Parent-Child Dynamics in Community Conferences - Some Questions for Reintegrative Shaming, Practice and Restorative Justice. The Australian & New Zealand journal of criminology 35(3): 330-346. Trimboli, L. (2000). An Evaluation of the NSW Youth Justice Conferencing Scheme. Sydney: NSW Bureau of Crime Statistics and Research. Syndromes of Meaning 185 Tucker. (2006). Systemic Incorporation: On the Relationship between Corpus and Systemic Functional Grammar. In G. Thompson & S. Hunston (Eds.) System and Corpus: Exploring Connections (81-102). London: Equinox. Van Ness, D., Maxwell, G., & Morris, A. (2001). Introducing restorative justice. In Morris, & G. Maxwell (Eds.) Restorative justice for juveniles: Conferencing, mediation and circles (3-16). Oxford: Hart Publishing. Whitelaw, C., Patrick, J., & Herke-Couchman, M. (2006). Identifying Interpersonal Distance using Systemic Features. In J. Shanahan, Y. Qu & J. Wiebe (Eds.) Computing Attitude and Affect in Text: Theory and Applications (199-214). Dordrecht: Springer. Zappavigna, M., Dwyer, P., & Martin, J. (2007). Just, like, guilty...sort of: The rhetoric of tempered admission in youth justice conferencing. Paper presented at the Australian Systemic Functional Linguistics Congress, Woollongong. PART II: APPLICATIONS AND VARIATION CHAPTER NINE DEMYTHOLOGISING CLT: WANTED – A REORIENTATION FOR TEACHERS IN THE 21ST CENTURY ANNE BURNS 1. Introduction In this chapter, I argue that we need to hold the major precepts of CLT up to the lens of teaching as the “relentlessly local” (Levinson & Holland, 1996). My argument is that for deep understanding of what is effective in teaching and learning we need to study closely the dynamics of life in schools and classrooms, to understand the processes of teaching and learning within them and the intricate relationships upon which these processes are forged. To support my case I present data from recent research in Australian adult English language classrooms, which illustrate the complexity of the local and suggest what can be learned about practical language teaching from studying classrooms as localised environments. 2. What is Communicative Language Teaching (CLT)? Despite over thirty years of adherence to the concept and its spread universally, there are still no coherent or agreed definitions of CLT. Such assumptions that exist about CLT are broad, vague and contested and include the following kinds of assertions: • Language learning is enhanced through a focus on communicative purpose (typically oral) • Communicative activities should be authentic and meaningful • The focus should be on language use rather than usage • Learner autonomy, responsibility and risk-taking are major goals Demythologising CLT 189 • Fluency takes precedence over accuracy • The teacher should facilitate and not control learning Advice such as the following, taken from a website for aspiring language teachers captures some of these major assumptions: Teachers in communicative classrooms will find themselves talking less and listening more - becoming active facilitators of their students learning…The teacher sets up the exercise, but because the students’ performance is the goal, the teacher must step back and observe, sometimes acting as referee or monitor. A classroom during a communicative activity is far from quiet, however. The students do most of the speaking, and frequently the scene of a classroom during a communicative exercise is active, with students leaving their seats to complete a task. Because of the increased responsibility to participate, students may find they gain confidence in using the target language in general. Students are more responsible managers of their own learning. Such advice is rarely helpful as it provides little by way of concrete practical teaching strategies or development of skills in teaching language. In a recent paper, Mangubhai, Marland, Dashwood and Son (2007) propose that the range of concepts used in the literature to describe CLT and make recommendations for its use may not be well understood or serve the practical needs of language teachers. They argue that this situation arises primarily from definitional ambiguity (Nattinger, 1984) and inconsistent representations of its meaning (Harmer, 2003). Extreme and conflicting positional statements about strong and weak versions of CLT (Whitley, 1993) add to the confusion, while the lack of a basis in research derived from actual teaching practice and teacher knowledge (Bartels, 2005) compounds the problem. There is a consistent theme from the literature over 20 years that many language teachers are unsure what CLT means and how to implement it (Duquette, 1995). Empirical studies of language classrooms and meta-analyses of this research tend to support these contentions. Thornbury (1998) notes that observational studies of language classrooms show CLT to be a “chimera”. He argues that many classrooms reflect hybridised versions of traditional and CLT approaches and that manifestations of CLT are in general “not only weak but very weak” (p.110). Reviewing CLT classroom research over 20 years, Karavas-Doukas (1996) concludes that classes in which CLT is observed to be effectively used are rare and that more traditional non-CLT teaching approaches remain the norm (see also Sato, 2002). 190 Chapter Nine Bax (2003) argues that approaches to CLT suffer from being insufficiently positioned within the local and contextual, a stance that has also been increasingly supported by the literature from scholars in Asian contexts who argue that CLT is underpinned by Euro-centric concepts (Holliday, 1994) that do not reflect local understandings and norms (e.g. Burnaby & Sun, 1989; Shamin, 1996; Li, 1998; Canagarajah, 1999; Ramanathan, 1999; Rao, 2002). In this situation and in the face of such evidence, the focus on CLT as the “norm” for ELT practice can only lead to the deskilling of teachers and more calls at the policy level for teacher accountability. 3. Refocusing from CLT to ELT? It can be argued that CLT falls within a progressive movement in education generally over the last four decades which has replaced the language of education with the language of learning (Biesta 2006). The discourse of learning is manifested through the increasingly consumerist relationship within which educational provision is positioned as a commodity: the learner becomes a consumer or customer; the teacher is the supplier of learning commodities; and education is readily available for purchase. Learning is foregrounded over teaching through such terms as the learner-centred curriculum, and the lifelong learner or subsumed with learning as a flipside of the same coin in expressions like teaching and learning or teaching-learning. In the light of these recent educational reorientations, Biesta (2006) contends: …One of the key educational responsibilities is that of providing opportunities for individuals to come into the world… coming into the world is definitely not about self-expression. It is about entering the social fabric and is therefore thoroughly relational. It requires first and foremost the creation of situations in which learners are able and are allowed to respond…. It is about responding and therefore also being responsible for what and who is ‘other’….To respond is as much about activity, about saying and doing as it is about passivity: listening, waiting, being attentive…(p.28). Biesta’s arguments echo those of Walkerdine (1984) who outlines how, within what are termed progressive approaches, the focus has shifted from teaching to learning. She argues that in the learning approach the teacher is expected to provide an educational environment in which learning occurs and teaching becomes invisible (compare the notion of the teacher as facilitator promulgated by the website excerpt above). Demythologising CLT 191 Progressive approaches have been critiqued for their consequences for students, particularly those who bring fewer learning resources and support to the learning process (e.g. Bourne, 2003). One of the central criticisms is the failure of progressive approaches to meet the needs of disadvantaged students, who are not part of the dominant culture, as Delpit (1988) explains: Some children come to school with more accoutrements of the culture of power already in place – ‘cultural capital’ as some critical theorists refer to it – some with less. Many liberal educators hold that the primary goal of education is for children to become autonomous, to develop fully while they are in the classroom setting without having arbitrary, outside standards forced upon them. This is a very reasonable goal for people whose children are already participants in the culture of power and who have internalized its codes. But parents who don’t function within that culture often want something else. It’s not that they disagree with the former aim, it’s just that they want something more. They want to ensure that the school provides their children with discourse patterns, interactional styles and spoken and written language codes that will allow them success in the larger society (p.285). Delpit goes on to argue that in many ways traditional approaches were more successful and effective in educating black children in the US than later progressive approaches, as the curriculum was less hidden, textbooks supported learning, teachers concentrated on teaching, and teachers (not learners) took responsibility for ensuring that learners had a sense of being able to learn. Another criticism of a progressive model of education which “reinforces the inequalities of access” (Feez, 1995) is that the curriculum framework within which students work remains hidden to students, as Feez goes on to explain: It is simply that in progressive pedagogies, the way these inequalities are perpetuated becomes invisible. Learners’ individuality and freedom may be more highly valued in progressive classrooms, but during and at the end of their course of study learners are still assessed against the standards of the dominant culture … although classrooms are more pleasant, what is actually expected of learners in order for them to be successful is not made explicit ... progressive classrooms tend to reinforce existing social inequalities of opportunity because it seems that it is the learner, rather than the educational institution, who is to be blamed for failure in such benevolent and rich learning environments (p.9). 192 Chapter Nine If language education is to become more effective and “educationally responsible” as Biesta argues what shifts need to be made to refocus the lens towards teaching? Vygotsky (1978) argues that learning is essentially a social activity and that teaching, rather than being invisible in the learning process, is central to it. Learning constitutes an apprenticeship into the social environment – the “social fabric” in Biesta’s terms – and no matter where it occurs or through what media, it involves a teacher, thinking about instructional events “to intervene and short-circuit the learning process” (Widdowson, 1990). The notion of effective language teaching (ELT) would hold, also in line with Bax’s (2003) arguments, that teaching needs to be systematically and finely responsive to local conditions within which deep understanding of student characteristics and needs, positive, not deficit, positioning of student learning, continuing development of student language, and building of confidence and interest in using the language outside the classroom are paramount. Thus, rather than focusing on whether language teaching “conforms” to ambiguous norms of CLT, research and practice should increasingly turn attention to what constitutes effective language teaching seen through the lens of the “relentlessly local” (Levinson & Holland, 1996). In an attempt to provide a basis for considering the nature and characteristics of localised effective language teaching, I draw upon a recent study within classrooms in the Adult Migrant English Program (AMEP) (Burns & de Silva Joyce, 2008). 4. The context and methodology The AMEP was originally established (1949) as part of the settlement program of the Commonwealth Government of Australia with a focus on English language learning that would enable new immigrants and refugees to participate in Australian society. Most recent policy directives have been towards preparing students more directly for participation in the workforce. The curriculum, the Certificates in Spoken and Written English (CSWE), organised at three levels from beginner to intermediate and which is used nationally, reflects learning goals in general outcome statements, but it is the AMEP teacher who contextualises them through pedagogical classroom discourse and interventions. The research positioned itself in the light of surveys conducted by AMEP providers that have indicated that many incoming immigrant learners are not familiar with CLT approaches and often feel confused by what they perceive as randomised activities and unstructured syllabus design. In addition, the overall take-up of legislated AMEP entitlement (510 hours) relative to potential enrolment is low with concerns being expressed at policy and Demythologising CLT 193 institutional levels about what is achievable, especially for low level and special needs students (Burns & de Silva Joyce, 2007). The goals of the research were to investigate the characteristics and features of effective teaching in this context, and more specifically, the role of the teacher in framing the language learning experiences of students and the way students responded to teaching approaches. Over six months, observations took place in 11 classrooms and involved 180 students: Table 1: Observations of AMEP Classrooms Location of observations Class level and Date of observation student numbers NSW AMES CSWE III: 14 May 2006 AMES Victoria CSWE I: 18 July 2006 CSWE II: 18 CSWE III: 19 TAFE Queensland Pre-CSWE: 13 September 2006 CSWE II: 10 CSWE III: 23 CDU Darwin CSWE II: 13 November 2006 CSWE III: 18 TAFE Tasmania Pre-CSWE: 19 November 2006 CSWE II: 15 The researchers spent one and a half to two hours in each classroom, during which time field notes were made on classroom layout, class profile, teacher and student activities and teaching techniques. The observations were audio-recorded and samples of materials used and produced during the lesson were collected. Post-observation interviews were conducted with volunteer focus groups of 4-8 students and with individual teachers. In cases where students’ English was considered not sufficiently developed for interview, recorded translations of the interview protocols were used. All focus group and interview responses were audio- recorded. 194 Chapter Nine 5. Analytical frameworks In investigating effective teaching practices which support adult immigrant student language development, we adopted the analytical frameworks proposed by Bernstein (e.g. 1996) more commonly used by educators in non-language educational sectors, to understand the central role of the teacher in language classrooms. Bernstein’s work concentrated on understanding “the relationship among political economy, family, language and schooling and differences in learning among the social classes” (Sadovnik, 2001, p.687). He was concerned about the loss of student potential within educational contexts, especially students from the working class. Bernstein’s sociological theory of education was based on several key concepts, which can be used to investigate all aspects of educational processes from the level of educational institutions to the level of classroom interaction. Primarily, four of these concepts were used to analyse the pedagogic practices in the classrooms we observed. In each case, the relevance of the concept to AMEP provision follows. 1 Individual enhancement Bernstein saw enhancement as “a condition for experiencing boundaries, be they social, intellectual or personal, not as prisons, or stereotypes, but as tension points condensing the past and opening possible futures ... it is the right to the means of critical understanding and to new possibilities” (Bernstein, 1996, p.7). For AMEP students English is the means to “possible futures” in a new country. “English is seen as providing access to community membership and such membership can be high stakes indeed for individual learners … [and] high stakes for the community itself in terms of being an inclusive community where capacity for “participatory association” (Somers, 1993, p.587) is available to all” (de Silva Joyce & Hood, forthcoming). Surveys of AMEP students identify language learning as centrally important to settlement and place the language classroom at the centre of language learning. 2 Pedagogic device Bernstein was interested in the question: “[A]re there general principles underlying the transformation of knowledge into pedagogic communication, whether the knowledge is intellectual, practical, expressive, or official knowledge or local knowledge?” Demythologising CLT 195 (Bernstein, 1996, p.39). In exploring this question he developed the concept of the pedagogic device, which mediates between the “field of potential pedagogic meanings and what emerges as pedagogic communication” (Davis, 2004, p.45). For Bernstein, “[P]edagogy is a sustained process whereby somebody(s) acquires new forms or develops existing forms of conduct, knowledge, practice and criteria, from somebody(s) or something deemed to be an appropriate provider and evaluator” (Bernstein, 1999, p.259). In the case of second language learning in the AMEP the classroom mediates real world discourse into a “mediated, virtual or imaginary” (Bernstein, 1996, p.47) discourse. 3 Classification Classification is about boundaries between one category and another: “where we have strong classification, the rule is: things must be kept apart. Where we have weak classification, the rule is: things must be brought together” (Bernstein, 1996, p.26). For Bernstein, “classifications, strong or weak, always carry power relations” (p. 21) and it is important to analyse what is happening in classrooms to ensure that “the inescapable authority of the teacher” (Bourne, 2004, p.65) is not masked and is used to develop collaborative practice. The key question is: “[I]n whose interest is the apartness, and in whose interest is the new togetherness and the new integration?” (Bernstein, 1996, p.26). In the second language classroom one of the main categorical divisions is the four macroskills: reading, writing, speaking and listening. CLT tends to advocate integrating (blurring?) these skills across activities, which can lead students to lose the intended focus because of weak classification. Integration of skills may also lead to non-achievement; for example, students complete an activity involving speaking skills but the reading load related to the activity is beyond reading capabilities. Also common to CLT classroom expectations is weak classification between learning language, learning through language and learning about language (Halliday, 1979). The AMEP classroom involves learning the language as a resource to use in social contexts (learning language), learning about the culture and society in which learners are settling (learning through language) and consciously learning about aspects of the language such as phonology, spelling, grammar and 196 Chapter Nine registers (learning about language). CLT classrooms may confuse the expectations of students about the role of the teacher and the learner. The division of labour can become blurred (Bernstein, 1996, p.29) when students are asked to make decisions about content and to negotiate activities when they have little knowledge of the new field of learning, limited learning skills and previous unsuccessful learning experiences. Having someone support and assist them along the learning pathway - which may eventually lead to individual decision-making and responsibility for learning – is vital. Multi-modal signals, such as “changes of place, pace and deportment” (Bourne, 2004, p.67) also establish changing relationships between teacher and students within lessons. 4 Framing “[F]raming is related to the transmission of knowledge through pedagogic practices” (Sadovnik, 1995, p.8). It is about the relationship between the transmitter (teacher) and acquirers (students). “Where framing is strong, the transmitter has explicit control over the selection, sequence, pacing, criteria and social base. Where the framing is weak, the acquirer has apparent control” Bernstein (1996, p.27). For Bernstein strong framing develops “a visible pedagogy where rules of instructional and regulative discourse are explicit” while weak framing develops “invisible pedagogic practice where the rules of regulative and instructional discourse are implicit and largely unknown to the acquirer” (p. 28). Bernstein distinguished between regulative discourse and instructional discourse in the classroom. Regulative discourse includes “expectations about how conduct, character and manner order the relationship between teacher and students” (Arnot & Reay, 2004, p.135), through rules of how people engage with one another. Instructional discourse refers to “the selection, sequence, pacing and criteria of knowledge” (p. 135). In AMEP classrooms the teacher is largely responsible, within the general outcomes and criteria of the curriculum, for the content of units of work, sequencing of content, pacing of lessons and creation of the social base of the classroom. The needs-based development of syllabuses for specific groups of learners gives apparent control to the students over content, but it is the teacher who ultimately decides what will be taught and makes the decisions about how the classroom operates. External factors can affect the strength of Demythologising CLT 197 framing within the classroom, for example, the need to meet the learning outcomes of the curriculum. However, in the AMEP there are no syllabus documents to dictate the content and sequencing of courses. The curriculum is text-based, thereby creating a text-based orientation to course design, but the teacher is left to select content in which to contextualise the curriculum outcomes and to decide on the type, sequencing and pacing of classroom activities. In classrooms where the decisions of the teacher are explicit and visible, student will be conscious of classroom expectations, “since the rules of legitimate expectations are made clearly available to the acquirer [student] by the transmitter [teacher]” (Sadovnik, 1995, p.13). However, where the teacher’s decisions are implicit, the students must create their own criteria for classroom events and “the teacher’s role is as a facilitator not a transmitter” (p. 13). Visible pedagogy operates with strong classification and framing and is explicit in acknowledging responsibility for taking up a position of authority (Bourne, 2004, p.65). According to Vygotsky (1978), all students bring a history of learning to the classroom. The issues at stake for AMEP students, in classrooms strongly oriented to CLT, are whether their educational experiences have prepared them for implicit pedagogic practices and whether they have the time available to learn language only by participating in communicative activities and without explicit teaching about language as system. 6. Findings of the study Using the theoretical concepts outlined above, we focused our observations on identifying instances of visible pedagogy that would lead to a framework for developing the concept of effective local pedagogy. In this section, I summarise the examples of such pedagogy derived from the classrooms we observed, the teacher interviews and the student focus groups. Here I focus on four major areas. 6.1 Teachers’ philosophy of teaching and expectations of students are made clear In class, teachers spent time outlining in detail their personal approaches to teaching and learning in their classrooms. Interview data revealed that these philosophies were universally derived from their experiences of teaching adult immigrants with the express purpose of equipping them to function in their lives beyond the classroom. Thus the 198 Chapter Nine concept of individual enhancement leading to the potential for future entry into “the fabric” of Australian society (Bernstein, 1996; Biesta, 2006) underpinned teaching practices. Further, teaching and learning roles were not invisibly assumed in these classrooms. For example, one teacher we observed took her students through a series of activities deliberately focused on different roles that might be adopted in the classroom by teachers and learners, asked her students about their experiences of and responses to these roles, and then negotiated with them the roles that they saw as preferable in her classroom for effective learning. Thus, her students were both apprenticed into (Bourne, 2004) and had directly laid- out experiences of role expectations in her classroom (Arnot & Reay, 2004). Students were also inducted explicitly into ways to participate discoursally, for example, being expected to request appropriate clarification from each other without reference to the teacher or being exposed to explicit models of how to interact through group work. Great emphasis was placed on responsibility for storing, using and repacking materials that were portrayed as a valuable resource owned by all class members. Consequently, teachers created visible pedagogical devices through which students were scaffolded into new forms of knowledge, discourse and behaviour (Muller, Davies & Morais, 2004). To focus the students on the goals of the course, teachers displayed their whole program for that term on notice-boards that students could continually refer to, including the topics to be covered and dates. Copies were provided for students to file in folders which were also referred to at the beginning of each lesson to position the current lesson within its overall curriculum progression. Similarly, curriculum outcomes, especially focusing on the learning outcomes students were expected to achieve as part of the current curriculum plan, were displayed and provided to students. In cases where students were at low beginning levels, programs and outcomes were translated into the languages of the students in the class. Where available, bilingual resources were used to explain verbally to beginning students with little or no literacy. Thus, in the most effective classrooms we observed, curriculum boundaries were established and enacted through strong classification Such practices reinforced students’ capacity for individual enhancement and participatory association (Somers, 1993) within the larger practices of the educational institution. Demythologising CLT 199 6.2 Movements across and within the curriculum are explicitly signalled At the beginning of lessons, teachers referred students to the overall curriculum program and explained where they were in that program. In addition, they wrote up the plan for that day on the board. As each activity making up the plan was concluded, the teacher referred back to it and ticked off the activity on the list, drawing students’ attention to the completion of the activity. During the lesson itself, the teacher did not commence the lesson or the next activity until the attention of all the students was fully focused on him/her. Typically teachers then outlined or modelled the activity first or demonstrated with the assistance of students; the teachers’ focus was on evaluating the students’ comprehension of what the activity entailed, which was realised discoursally by frequent clarification checks and requests to various students to re-tell the activity to others. Teachers also deliberately used various classroom proxemics (cf. Bourne, 2004) to focus attention and emphasise goals, for example by moving to the front of the class when changing stages of the lesson, standing closer to students requesting or providing information, joining groups undertaking various activities, or physically modelling responses when students expressed uncertainty. The preparation, use and handling of resources and materials was signalled, for example through direct requests to open books, turn to specific pages, write dates on materials, place pages in folders, or place unnecessary material elsewhere. Often the teachers’ movements and actions in relation to activities were explained before they occurred in order to prepare students for what was to come. For example, students were explicitly told that the teacher was about to play a tape twice, to put down their pens and to listen; the activity did not then commence until all students had focused their attention. Just as the teacher had overviewed the lesson plan at the beginning of the lesson, the plan was reviewed at the end. Where planned activities had not taken place, teachers stated explicitly the reasons and what should be done by the teacher and students, for example that they had not had time and that students should complete certain activities at home for checking by the teacher in the next lesson. The selection of such activities was not arbitrary but focused on consolidation that could be done without teacher assistance. Finally, in relation to lesson closure, typically the teachers made clear statements that the lesson had now finished and that it was time for the students to leave. Collectively such cross-curriculum movements constituted practices and processes that created strong boundaries or classifications between the 200 Chapter Nine macro and micro tension points (Bernstein, 1996) of whole lessons, lesson stages, activity preparation and the activity itself. The teacher’s role remained authoritative (Bourne, 2004) even when the primary focus of pedagogical action was not on the teacher, as distinctions between regulative and instrumental discourse were explicitly framed (Sadovnik, 2001; Arnot & Reay, 2004). 6.3 Movements across and within activities are signalled In addition to signalling attention at the macro-level of the curriculum processes, the teachers also signalled at the micro-level, within activities. In this respect, teachers named and emphasised the specific macroskill focus of the activity and whether the learning goal was reading, writing, speaking or listening. They also alerted students to the specific skills within the overall macroskill that the activity was aimed to help them develop, for example, identifying key ideas in a reading text or applying the discourse structures learned in a speaking task. Further, teachers explained the purpose of the activity and the nature of the specific language skill being developed, such as extending vocabulary related to a particular field in preparation for writing. For each activity, and then the component of the activity, teachers drew the students’ attention to the steps for completing the task, checked students’ understanding and if relevant linked the task procedures to other similar tasks already undertaken in class. Some teachers reconsolidated the intended learning purpose by displaying activity sheets on an OHP and modelling what students needed to do before asking them to proceed. Where the teacher then saw signs of misunderstanding as the task proceeded, the activity was halted and re-explained, or more competent students were asked to retell the activity or provide assistance in the mother tongue. In these various respects the framing of regulative and instructional discourse contributed to mediating the transmission of knowledge among teachers and students, which did not just occur at the lesson and activity level but unfolded as reiterative and multilayered modalities (Sadovnik, 1995; Bernstein, 1996). 6.4 Classroom processes are explicitly linked to classroom contexts In the most effective classrooms, the physical spaces in which the lessons took place interconnected with the teaching processes that went on in them. Materials supporting student learning and displaying student achievement were plentiful, made visible, and actively incorporated into Demythologising CLT 201 lessons. Apart from the curriculum and program outlines, goals and learning outcomes mentioned, materials offering important and frequently used reference points for learning were available. Teachers displayed visual resources such as alphabets, phonetic alphabets, maps of students countries of origin, photographs of known landmarks, personal photographs, names and autobiographies of students, posters of vocabulary already learned, models of written and spoken genres highlighted with the major language features amongst many others. Texts produced by students by the end of the lesson were pinned up for comparison, re-reading, review and revisiting in subsequent lessons, as were dated samples of handouts and materials used that day. In these classrooms students were thus, constantly surrounded by examples of the language they were learning, expected to achieve and had produced. The effective language classroom was thus a place of connected multimodal signals where the physical, behavioural and pedagogic resources interrelated in productive and authoritative ways (Bernstein, 1996; Bourne, 2004) and where pedagogical resources of language (Halliday, 1979) were visibly differentiated. 7. Concluding remarks These classrooms, where curriculum and pedagogic frameworks were highly visible, were not places where communicative activity for its own sake drove the teaching or whether the teacher played a non-interventive facilitative role. Rather the most effective teachers were authoritative teachers who spent substantial time focusing on one text or activity typically not moving on until they were satisfied with student achievement. They tolerated long periods of silence in the classroom enabling students to absorb new knowledge, worked within repetitious and identified frameworks which provided students with familiarity, routine and security, presented new language through already recognisable tasks and activities, repeated lesson stages to enable students to absorb new language and introduced new ways of operating in the classroom through a staged and scaffolded process and only when students had had an opportunity to review or practice already learned material. Thus the primacy of doing, which as Widdowson has pointed out has typically characterised CLT, was not a central consideration in these classrooms. As Widdowson (1990) explains: The natural learning problem is this. It turns out that learners do not readily infer knowledge of the language system from their communicative activities. The grammar, which they must obviously acquire somehow as a necessary resource for use, proves elusive. So quite often the situation 202 Chapter Nine arises where learners acquire a fairly patchy and imperfect repertoire of performance which is not supported by an underlying competence, Their doing does not seem to lead naturally to knowing, as has been optimistically assumed (p.161). So far the dominance of CLT still seems assured in the world of language teaching and, moreover, the norms of CLT are still being adopted into policy decisions for curriculum reform in many international locations. However, there is evidence that the kinds of indicators of quality effective teaching for language learners outlined above are supported by other recent research. In the New Zealand context, Franken and McComish (2003) draw on their observations to argue that the following need to be present for NESB learners in schools: • Learners should be given explicit and focused instruction on all aspects of language • They should be given opportunities to use language in extended contexts, as well as comprehend language • They need opportunities to interact on meaningful material, especially with a teacher • They need very extensive language learning opportunities • Goals should not focus on correctness at particular points, which cannot be guaranteed to be in line with a learner's readiness • Opportunities to engage with the same material need to be allowed for many times, therefore a linear approach to language development or language teaching content is unlikely to be successful • Goals for L2 learners should not be limited to performance in easier contexts, or on easier objectives. Similarly, Roberts et al (2008) summarise the features of effective practice from a large scale study carried out in the UK, the Effective Pedagogies Project (see also Roberts & Baynham, 2006): • Clear planning, both strategic and on-line, with sufficient time to explore and exploit activities, and explicit framing and metalanguage • The ‘bricoleur-teacher’ who creatively and inventively assembles materials and activities: a ‘principled eclecticism’ • Extended ‘talk from within’ combined with planned and on the spot form- focused work • Collaborative group work where socially distributed knowledge is managed and exploited • Using the classroom as a place to learn from and for the outside world, including the socio-cultural knowledge leant through language and its use in communication outside the classroom • Constant revisiting and reworking of the linguistic items in different contexts Demythologising CLT 203 • Safe and fun learning The notion of CLT has served the language learning community since the late 1970s. Importantly, it has drawn attention to the need in language teaching to recognise what people do with language in social situations as well as how language is formed and structured. But the question remains of whether CLT now continues to serve the profession well in the 21st century. My argument in this chapter has been that it is time to look to alternative criteria and to reorient the focus on to effective teaching that will serve language learners productively in localised contexts. Notes 1. The content of this chapter owes much to a previous unpublished paper prepared jointly with Helen de Silva Joyce. I am grateful for her permission to use this material. 2. My thanks go to Brian Paltridge, University of Sydney, for his insightful comments on a previous draft of this chapter. References Arnot, M. & Reay, D. (2004). The Framing of Pedagogic Discourse: Regulating order in classroom learning. In J. Muller, B. Davies & A. Morais (Eds.) Reading Bernstein. Researching Bernstein (135-150). London: Routledge Falmer. Bartels, N. (2005). Applied Linguistics and Language Teacher Education: What we Know. In N. Bartels (Ed.) Applied Linguistics and Language Teacher Education (405-425). Boston: Kluwer Academic Publishers. Bax, S. (2003). The End of CLT: A context approach to language teaching. ELT Journal 57: 278-287. Bernstein, B. (1996). Pedagogy, Symbolic Control and Identity. London: Taylor and Francis. Biesta, G. (2006). Beyond Learning: Democratic learning for a human future. Boulder, Col.: Paradigm Publishers. Bourne, J. (2003). Vertical Discourse: The role of the teacher in the transmission and acquisition of decontextualised language. European Educational Research Journal 2 (4): 496-521. Bourne, J. (2004). Towards A ‘Radical Visible Pedagogy. In J. Muller, B. Davies & A. Morais (Eds.) Reading Bernstein. Reading Bernstein. Researching Bernstein (61-74). London: Routledge Falmer. Burnaby, B. & Sun, Y. (1989). Chinese Teachers’ views of Western Language Teaching: Context informs paradigm. TESOL Quarterly 23: 219-238. 204 Chapter Nine Burns, A. & de Silva Joyce, H. (2007). Adult ESL programs in Australia. Prospect, 22(3): 5-17. Burns, A. & de Silva Joyce, H. (2008). Clearly Teaching: Teaching in action for teachers. Sydney: Macquarie University. Canagarajah, S. (1999). Resisting Linguistic Imperialism in English Teaching. Oxford: Oxford University Press. Davis, Z. (2004). The Debt to Pleasure: The subject and knowledge in pedagogic Discourse. In J. Muller, B. Davies & A. Morais (Eds.) Reading Bernstein. Researching Bernstein (44-57). London: Routledge Falmer. de Silva Joyce, H. & Hood, S. (forthcoming). English for Community Membership. In D. Belcher (Ed.) Teaching Language Purposefully: English for Specific Purposes in Theory and Practice. Cambridge: Cambridge University Press. Delpit, L. (1988). The Silenced Dialogue: Power and pedagogy in educating other people’s children. Harvard Educational Review 58(3): 280-298. Duquette, G. (Ed.) (1995). Second Language Practice. Classroom strategies for developing communicative competence. Clevedon: Multilingual Matters. Feez, S. (1995). Systemic Functional Linguistics And Its Application In Australian Language Education: A short history. Interchange 27: 8-11 Franken, M. & McComish, J. (2003). Improving English Language Outcomes for Students Receiving ESOL Services in New Zealand Schools, with a Particular Focus on New Immigrants. Retrieved 8 February 2008 from http://www.educationcounts.govt.nz/publications/schooling/5707 Halliday, M.A.K. (1979). Occasional Paper No.1. Curriculum Development Centre: Language Development Project. Canberra: Curriculum Development Centre. Harmer, J. (2003). Popular Culture, Methods and Context. ELT Journal 57(3): 288-294. Holliday, A. (1994). Appropriate Methodology and Social Context. New York: Cambridge University Press. Karavas-Doukas, E. (1996). Using Attitude Scales to Investigate Teachers’ Attitudes to the Communicative Approach. ELT Journal 50(3): 187- 198. Levinson, B. & Holland, D. (1996). The Cultural Production of the Educated Person: An introduction. In B. Levinson, D. Foley & D. Holland (Eds.) The Cultural Production of the Educated Person: Critical ethnographies of schooling and local practice (1-54). Buffalo, N.Y.: State University of New York Press. Demythologising CLT 205 Li, D. (1998). It’s Always More Difficult Thank You Plan Or Imagine: Teachers’ perceived difficulties in introducing the communicative approach in South Korea. TESOL Quarterly, 32(4): 677-704. Mangubhai, F., et al. (2007). Framing Communicative Language Teaching for Better Teacher Understanding. Issues in Educational Research, 17(1): 85-106. Muller, J., B. Davies & Morais, A. (Eds.) (2004). Reading Bernstein. Researching Bernstein. London: Routledge Falmer. Nattinger, J. R. (1984). Communicative Language Teaching: A new metaphor. TESOL Quarterly 18(3): 391-407. Ramanathan, V. (1999). “English Is Here to Stay”: A critical look at institutional and educational practices in English. TESOL Quarterly 33(2): 211-232. Rao, Z. (2002). Chinese Students' Perceptions of Communicative and Non-communicative Activities in EFL Classrooms. System 30(1): 85- 105. Roberts, C. & Baynham, M. (Eds.) (2006). Where Talk is Work: The social contexts of adult ESOL classrooms. Special Issue. Linguistics and Education 17(1). Roberts, C., et al. (forthcoming). Adult ESOL In The UK: Policy and research. In A. Burns & H. de Silva Joyce (Eds.) Special Issue. Prospect 22(3). Sadovnik, A. (Ed.) (1995). Knowledge and Pedagogy: The sociology of Basil Bernstein. Norwood, NJ: Ablex. —. (2001). Profiles Of Famous Educators: Basil Bernstein, 1924-2000. Prospects: The Quarterly Review of Comparative Education XXXI (4): 607-620. Sato, K. (2002). Practical Understandings of CLT and Teacher Development. In S. Savignon (Ed.) Interpreting Communicative Language Teaching: Contexts and concerns in teacher education (1- 27). New Haven & London: Yale University Press. Shamim, F. (1996). Learner Resistance to Innovation in Classroom Methodology. In H. Coleman (Ed.) Society and the Language Classroom (105-121). Cambridge: Cambridge University Press. Somers, M. R. (1993). Citizenship And The Place Of The Public Sphere: Law, community and political culture in the transition to democracy. American Sociological Review 58: 587-620. Thornbury, S. (1998). Comments on Marianne Celce-Murcia, Zoltán Dörnyei, and Sarah Thurrell’s Direct Approaches in L2 Instruction: A turning point? TESOL Quarterly 32(1): 109-115. Vygotsky, L. (1978). Mind in Society. Cambridge, MA: MIT Press. 206 Chapter Nine Walkerdine, V. (1984). Developmental Psychology And The Learner- Centred Pedagogy: The insertion of Piaget into early education. In J. Henriques, W. Hollway, C. Urwin, C. Venn & V. Walkerdine (Eds). Changing the subject (153-202). London: Methuen. Whitley, M. S. (1993). Communicative Language Teaching: An incomplete revolution. Foreign Language Annals 26(2): 137-154. Widdowson, H. (1990). Aspects of Language Teaching. Oxford: Oxford University Press. CHAPTER TEN FINE-TUNING DISCOURSE IN THAI EFL ACADEMIC AND ELECTRONIC BULLETIN BOARD WRITING MONTRI TANGPIJAIKUL 1. Introduction Fine-tuning utterances appropriately through the use of modality and degree of intensity in English is of paramount importance for the development of pragmatic competence because learners will not be able to express themselves precisely the way they want without the ability to fine- tune their utterances subtly. However, conveying interpersonal meaning in English through the use of modality and degree of intensity has always been a problem for EFL learners. This may be because learners do not have enough chance to exercise their repertoire of these linguistic features on an everyday basis to express themselves naturally in less formal contexts than the classroom settings. As a result, their utterances may sound too direct, and even abrupt to the ears of the English native speakers. Studies such as Altman (1982), Bhandhufalck (1983), Gibbons and Markwick-Smith (1992) and Karkkainen (1992) show that these linguistic features are generally underused by non-native learners of English. De Cock, Granger, Leech and McEnery (1998) also finds that a possible underuse of some pragmatic markers such as "sort of" and "I mean" among the non-native speakers was correlated with their lack of fluency in speech. Online bulletin board writing is interpersonal by nature as it shares some features that are typical of speech (Crystal, 2006, p.31-32); hence, the writing is informal and more carefree. In other words, it is similar to the genre of personal letters which is characterized by the use of first and second person pronouns, while academic writing is more informational and less interpersonal (Biber, 1988, p.128). As modality is said to be 208 Chapter Ten found more in informal contexts than formal ones (Holmes, 1983, p.110), it is of interest to find out whether online bulletin board writing would allow language learners to exercise their repertoire of modality in English more than formal academic writing. This is the issue addressed in this study. 2. Purpose of the study The aim of this research is to establish whether students use modal and intensifying elements more in electronic bulletin board than in academic essays. If the frequency of modal and intensifying elements appears greater in online bulletin board writing, then its use as a communicative writing medium would appear to provide the learners with an opportunity to exercise their repertoire of modality in English. The findings in this research are expected to raise awareness of this area of learners’ command of modal and intensifying elements in English. 3. Definition of terms “Bulletin board writing” refers to asynchronous free-style writing in the online electronic bulletin board. In this mode of writing, students can post their messages in many turns in responding to the assigned topics or to interact with other students who also participate in the same group discussion through, for example, agreeing, disagreeing, supporting, and acknowledging. “Academic writing” refers to the learners’ controlled format writing. This form of writing is more like essays or compositions which serve as an activity to give students the chance to develop their skills required for their future writing in the academic genre. Since the aim of this activity is to develop the students’ skills in writing well-organized essays by helping them develop well structured paragraphs, it is different from the generally understood concept of academic writing which aims to report research findings in academic journals or theses. 4. Analytical framework The linguistic features related to the area of modality and degree of intensity are amazingly broad; moreover, the existing frameworks in these two areas, which are mainly based on English native speakers’ performance, are diverse both in their definitions and classifications, Fine-Tuning Discourse in Thai EFL 209 depending on whether the concept is viewed grammatically or functionally. Therefore, for the benefit of this study on the performance of Thai learners of English, a clear analytical framework to be used for the analysis of the present research data is needed. The following analytical framework used in this study is an attempt to present three dimensions where modal and intensifying elements in English can be viewed together in grammatical, functional and semantic categories, though the semantic lens is the main focus in the presentation of results in this study. The proposed framework mainly combines Halliday and Matthiessen’s (2004) framework of modality, which includes both the core modal auxiliaries and the metaphorical lexical variants of modal devices, with that of grammatical categories as presented in Quirk, Greenbaum, Leech, and Svartvik (1985), Biber, Johansson, Leech, Conrad, and Finegan (1999), and Carter and McCarthy (2006). The list is then supplemented by the research literatures on modality including Coates (1983), Perkins (1983), Palmer (1990), and those on the notion of doubt and certainty including Holmes (1983, 1988), Hyland (1996), and Hyland and Milton (1997). To avoid long and redundant description of the terms, these modal and intensifying linguistic features, shown in Figure 1, are referred to in this study in a general sense as “fine-tuning devices” or “FTDs”. They are called such because of the feature they share in helping speakers fine-tune their meanings in the way of moderating or reinforcing the degree of doubt and certainty, degree of commitment, and degree of inclination. This study includes both the lexical and grammatical devices. Lexical devices refer to adverbials such as MAYBE, PROBABLY, IN MY OPINION, I THINK, while grammatical devices refer to the core modal auxiliaries including WILL, WOULD, SHALL, SHOULD, COULD, MAY, MIGHT, MUST. Even though CAN may have developed the meaning of possibility (Coates, 1995), in most cases, the meaning falls under “capability” under Eggins and Slade’s (1997, p. 107) or “capacity” in Martin and White’s (2005, p. 54) framework of modality. Because of its primary sense as conveying the meaning of ability, rather than possibility, this study does not include it in the notion of fine-tuning discourse. In functional categories, this study includes the four functions of the fine-tuning devices: modality, intensity, usuality, and perspective as presented in Figure 1. Modality covers the meaning of probability (including possibility and certainty) which can be realized by the use of modal auxiliaries, adverbs and copular verbs other than “be”. Intensity includes what Quirk, Greenbaum, Leech and Svartvik (1985) call “intensifiers” (including downtoners and amplifiers). This function also falls under the notion of amplification which is one of the key concepts in 210 Chapter Ten the appraisal framework (Martin & Rose, 2003, p.25). Usuality is also included as one of the fine-tuning elements as this function is part of Halliday’s (1994) concept of modality. However, as the inclusion of usuality in the modality framework has not been generally shared by all grammarians, this function is then put separately in this study so that it can be viewed in its own right. Adverbial phrases and clauses conveying speakers’ perspective or viewpoints are also included in Biber, Johansson, Leech, Conrad and Finegan (1999) under the label “epistemic stance adverbials”, which is in line with the notion of doubt and certainty. They are, thus, included in this study. Functional Grammatical Semantic Categories: Categories Categories Fine-tuning Devices (FTDs) Hedging Boosting Committal Inclination Modality Modal verbs may, will, should, will, might, shall must, want to, could, have to, would like to would had better Modal adverbs maybe, surely, probably, certainly, possibly definitely Copulars other tend, than ‘be’ seem, appear Intensity Degree quite, very, adverbs rather, highly, a bit strongly Usuality Adverbs of often, always, frequency usually, never hardly Perspective Clauses and I think, I’m sure, phrases to me, It’s true… for me, Figure 1: Analytical framework of fine-tuning devices (FTDs) used in this study Fine-Tuning Discourse in Thai EFL 211 While many fine-tuning devices are categorically identifiable in terms of functions, the functional distinction has not always been possible as there are some cases where words can fall under more than one category (Hillier, 2004, p.76). For example, REALLY can be viewed as conveying epistemic meaning signalling the speaker’s judgment referring to the factuality of what is said or the truth of a proposition as in “she is really pregnant” or simply be used as an amplifier in reinforcing the degree of intensity as in “she is really nice”. Thus, viewing its functional categorization alone, without context may be problematic. Because of this, it is also useful to look at them in terms of semantics, where both uses of REALLY can be viewed in the guise of boosting function. Simultaneously, fine-tuning devices from different grammatical categories can be viewed in the same semantic category. For example, possibility meaning can be expressed through modal auxiliaries such as in ‘You may be right’ or through adverbs as in ‘You are probably right’, so both of them fall under hedging function and can be considered as hedges. WILL and CERTAINLY convey similar degree of certainty, so they both fall under boosting function and can be regarded as boosters. Through the semantic lens, these fine-tuning devices can be viewed under four categories: (1) ones that allow learners to attenuate their meanings (hedging), (2) ones that allow them to reinforce meanings (boosting), (3) ones that express speakers’ obligation and necessity (committal), and (4) ones that express speakers’ desire or volition (inclination). These four semantic functions are important for the natural use of language as they help fine-tuning modal and interpersonal meanings that speakers want to convey. While hedging allows learners to express affective meaning in terms of negative politeness, e.g. “You might be wrong, I think”, boosters allow them to intensify or exaggerate their meanings in terms of positive politeness, e.g. “That was a truly delicious meal” (Hinkel, 2005, p.31). Downtoners such as A LITTLE, SOMEWHAT, and RATHER also allow speakers to express their low commitment to the propositional content, hence, functioning as hedges, while amplifiers such as STRONGLY, TOTALLY, and EXTREMELY allow them to express high commitment to the proposition, hence, functioning as boosters. The criteria for classification of intensifiers were adapted from Quirk et al. (1985, p.589- 602). In addition, committal is an important concept in fine-tuning discourse as it conveys the deontic sense signalling human control over the situation or the proposition, mainly to convey the speaker’s obligation or suggestion, and these are realized, for example, by the use of modal 212 Chapter Ten auxiliaries such as SHOULD as in “the university should provide more funds” or MUST as in “I must go now”. The term “committal” itself is chosen to combine the two compulsory meanings, one in which the speakers commit themselves into doing something (obligation), and one in which the speakers are committed by external force (necessity). For example, in expressing obligation, one can say strongly that “I have to leave now” or opt for a more moderate choice by saying “I should leave now”, while in necessity, one can say strongly that “The university must provide more funds”, or choose to moderate its degree of necessity by replacing MUST with SHOULD. Inclination is also one of the key elements in Halliday’s (1994) framework of modality. It allows speakers to show their desire or preference mainly through the use of semi modal such as WANT TO as in “Today, I want to tell you that …” The function of use can also be fine- tuned by writers’ or speakers’ moderating their wants through the use of WOULD LIKE TO instead of WANT TO to express polite preference. At the same time, their wants can also be reinforced by some uses of WILL and SHALL. According to Biber, Conrad and Leech (2002, p.176), some uses of WILL and SHALL + dynamic verbs, when used with human subjects, can convey volition or intention, rather than prediction. In this function, they can be replaced by WANT TO such seen in “I will talk about…” which is similar in meaning to “I want to talk about …”. The bordering meanings of prediction and volition in the use of WILL and SHALL have long been discussed in the literature and researchers (e.g. Palmer, 1990, p.134-135) agree that the distinction between the two meanings cannot always be made clear. This study simply separates one function where WILL and SHALL could be read as conveying volition or intention as in, for example, “Now, I will talk about…” or “Today, I shall write about…” In this context, speakers seem to express the intention to deliver their messages rather than to predict their own future actions. 5. Research methodology 5.1 Subjects The participants were Thai second year undergraduate students majoring in English at Kasetsart university in Bangkok. The participants were from two sections of a writing course (Writing II). The two parallel sections were taught by different teachers but used the same course material, criteria of assessment and course syllabus. Both classes met twice a week for one and a half hours in a regular classroom. The sample Fine-Tuning Discourse in Thai EFL 213 subjects comprised the population of 39 students, with 35 females and 4 males. 5.2 Data Collection In their Writing II course, a website was set up by the researcher and students were asked to write in response to three discussion topics on the online bulletin board in the web site. They were also asked to submit three academic writings on topics similar to the ones online. Topics in online bulletin board were posted once every 4 weeks, at about the same time as when the paralleled topics in academic writing were assigned. In this way, learners could decide on their own which task they wanted to perform first. The following topics were adjusted from the students’ coursebook Introduction to Academic Writing (Oshima & Hogue, 1997). They are shown as follows: Topics assigned in academic writing 1. Write about the most frightening movie or TV program you have ever seen. 2. Write about the changes that a modern invention has already caused or will cause you in the future. 3. People’s personalities are reflected in their hairstyles and clothing. Do you agree or disagree? Topics posted in online bulletin board writing 1. Write about the most frightening experience you have ever had. 2. Write about the effect that the internet has already caused or will cause you in the future. 3. People’s personalities are reflected in the way they live their lives. Do you agree or disagree? All writing activities were done outside of the classroom. Previous to this class, students had passed the three English Foundation I, II, III, and Writing I courses. Most students had had little experience writing formal academic essays and although all of them had basic background in composition writing from high school, these students making the transition from secondary to tertiary education level were still limited in academic discourse competence. Outside the classroom, they had access to the Internet in many campus areas. In the faculty alone, there were two areas 214 Chapter Ten for using the Internet. The university also provided access via a home wired-internet connection and at the University IT centre. Since students varied in their computer skills, students from the two sections were given a training session on how to participate in online bulletin board discussion in the course web site. 5.3 Data analysis After collecting the two types of writing–bulletin board writing (BB corpus) and academic writing (ACAD corpus), the two corpora were examined and all the fine-tuning devices that appeared in them were tagged. The two corpora were roughly equal in size with approximately 58,000 words in each. After the modal devices were annotated, a computer concordancing program (ConcGram Concordancer v. 2005) was used to count frequency of FTDs. 6. Results The following section provides a preliminary quantitative view on learners’ repertoire of fine-tuning devices in the two writing modes. These devices are presented in four semantic categories based on the framework shown in Figure 1 (hedging, boosting, committal, and inclination). Table 1 shows that hedges are found more in the online bulletin board than in academic essays in total (791 v. 532 occurrences). Even though modal auxiliaries and copulars other than “be” are found more in academic essays than in the bulletin board in total, the difference is slight. In fact, it is lexical devices that play an important role in hedging discourse in the online bulletin board. One piece of evidence supporting this is that three out of the four probability adverbs are found more in the bulletin board than in student essays, particularly MAYBE which is used more than twice as often in the bulletin board as in academic essays (66 v. 30 occurrences). Similarly, most downtoners are found more in bulletin board than in student academic essays with obvious cases such as RATHER, A LITTLE, and A (LITTLE) BIT. Approximators such as ABOUT, AROUND, APPROXIMATELY, and NEARLY, on the contrary, are found more in academic writing. Note also the high frequency of AROUND in academic essays (13 occurrences), as compared to bulletin board where only two occurrences are found. Fine-Tuning Discourse in Thai EFL 215 Table 1: Hedging (numbers represent frequency of occurrence) ACAD BB ACAD BB Modal auxiliaries Adverbs of frequency could 17 16 barely 0 2 may 64 68 commonly 2 1 might 33 33 generally 4 4 would 69 64 hardly 1 2 183 181 mainly 3 3 normally 4 5 Adverbs of probability occasionally 2 4 maybe 30 66 often 24 21 perhaps 2 7 rarely 0 9 possibly 4 5 seldom 1 4 probably 17 15 sometimes 47 100 53 93 usually 50 29 138 184 Copulars other than ‘be’ appear 0 3 Comment phrase/clauses seem 17 12 According to 1 5 tend 6 3 For me 10 44 23 18 I guess 1 4 I suppose 1 2 Downtoners I think 3 14 a (little) bit 5 19 In my opinion 6 34 a little 3 13 In my view 4 37 about 29 25 To me 2 25 almost 23 35 28 165 approximately 2 0 around 13 2 nearly 3 2 Total 532 791 pretty 2 5 quite 25 36 rather 1 8 somewhat 1 5 107 150 Note: In Table 1, the abbreviation ACAD stands for academic writing, while BB stands for online bulletin board writing. 216 Chapter Ten Even though the total frequency suggests that adverbs of frequency are generally used more in the bulletin board than in academic writing, the numbers fluctuate, depending on the individual items. For instance, while adverbs of frequency such as COMMONLY, OFTEN, and USUALLY are found more in academic writing, most others such as RARELY, HARDLY, NORMALLY, OCCASIONALLY, and SELDOM are used slightly more in the bulletin board. The more frequent use of adverbs of frequency in the bulletin board than in academic essays is salient in cases such as SOMETIMES (100 v. 47 occurrences) and RARELY (9 v. 0 occurrences). The other two adverbs of frequency (GENERALLY and MAINLY) are found to be used equally in the two modes. Comment phrases and clauses which are used to convey probability in terms of students’ viewpoints are used much more frequently in the bulletin board. In comparison with other categories, these comment phrases and clauses show the greatest difference in number of occurrences between the two writing modes. All of them are used much more in the online mode than in the essays. While the learners seem to attenuate their meanings more in the bulletin board than in academic essays through hedges, they also tend to reinforce or even exaggerate them more in this mode at the same time. This can be generally seen throughout all of the five grammatical categories shown in Table 2. The total number of fine-tuning devices used in conveying the boosting function in the bulletin board almost doubles that in academic essays. Fine-Tuning Discourse in Thai EFL 217 Table 2: Boosting (numbers represent frequency of occurrence) ACAD BB ACAD BB Modal auxiliaries Amplifiers be going to 13 27 absolutely 7 25 will 193 290 apparently 0 1 shall 0 2 clearly 1 6 206 319 completely 7 2 entirely 1 3 Adverbs of certainty extremely 12 7 actually 9 44 fully 6 2 certainly 4 11 greatly 0 3 definitely 2 7 highly 4 4 for sure 2 7 obviously 2 5 in fact 3 11 perfectly 2 1 in reality 0 3 really 40 91 indeed 13 1 strongly 1 7 of course 1 12 totally 4 15 really 3 10 truly 1 8 surely 6 13 88 180 truly 1 2 undoubtedly 0 4 Opinion clause anchors 44 125 … certain (that) 3 4 … clear (that) 2 2 Adverbs of frequency … sure (that) 8 14 always 82 113 … true (that) 2 5 never 45 51 15 25 127 164 Total 480 813 Grammatical devices such as WILL and BE GOING TO are obviously used much more in the bulletin board than in academic writing. SHALL is not found at all in the academic mode, and is found only twice in the bulletin board, one in a persuasive tag “shall we?”, and the other in the strong sense of WILL, conveying certainty. Adverbs which are used to express certainty or confidence are also found more in the bulletin board than in academic writing, with an exception in the case of INDEED which is found more in academic (13 occurrences) than in the bulletin board (1 218 Chapter Ten occurrence). Two adverbs of frequency (ALWAYS and NEVER) are also used more in the bulletin board writing than in academic writing. In line with other grammatical categories, most amplifiers are also found more in the bulletin board. In fact, in this category, only FULLY and EXTREMELY are used more in academic writing. Similarly, most adjectives conveying certainty which are used in the clause anchor construction such as in “I am sure that…” or “It is true that…” are also generally found more in the bulletin board than in academic essays, except for CLEAR which is found twice equally both in the online mode and academic essays. Table 3: Committal (numbers represent frequency of occurrence) ACAD BB be supposed to 0 2 had better 0 2 have to 119 172 must 46 42 need to 15 20 ought to 0 1 should 66 102 Total 246 341 Modal devices which are used to express committal are found more frequently in the bulletin board than in academic essays with the difference in frequency of 341 and 246 occurrences in total. Even though there is an exception in the case of MUST, which is found slightly more in the academic essays than in the online bulletin board, the difference in frequency of MUST being used in the two modes is slight (46 v. 42 occurrences). While HAVE TO is the most frequently found device in the two modes, with SHOULD, MUST, and NEED TO (in respectively diminishing degrees) being used less frequently, OUGHT TO, HAD BETTER, and BE SUPPOSED TO are rarely used. Fine-Tuning Discourse in Thai EFL 219 Table 4: Inclination (numbers represent frequency of occurrence) ACAD BB (I / You) want to 37 129 (I / You) will 0 40 (I / You) would like to 3 19 Total 40 188 Modal devices which are used to express personal preference or volition are clearly found much more in the bulletin board (188 times in total) than in academic writing (40 times in total), with WANT TO being used the most. Most uses of WANT TO and WOULD LIKE TO are found in the context such as in “I want to say that… “ The use of WILL in conveying this function such as in “Now, I will write about…” does not appear at all in student academic essays. Figure 2: Distribution of fine-tuning devices in four semantic categories The above findings in the four semantic categories can be viewed together in Figure 2 which shows the amount of fine-tuning devices being used to convey four semantic realizations. It is evident that the hedges are 220 Chapter Ten found more in the online bulletin board writing than in student academic essays. The difference is even more evident in the numbers of boosters in online bulletin board which almost double those found in academic writing. In general, learners seem to fine-tune their meanings more in bulletin board than in academic essays, both in attenuating and reinforcing their utterances. This may be because learners are aware of the fact that online writing is more casual as it shares many features of speech (Crystal, 2006, p.31-32), so they tend to transfer fine-tuning devices that are typically found in speech into their writing. As for expressing committal, learners seem to use slightly more modal devices in the online than the offline mode. This is supported by the evidence that 6 out of 7 deontic modal auxiliaries are found more in the bulletin board. The findings also show that while HAVE TO and SHOULD seem to be dominant in learners’ repertoire of modality, OUGHT TO and HAD BETTER are scarcely found. The difference in expressing committal between the two modes, however, is not as great as those found in the areas of hedging and boosting. Similarly, inclination is also found much more in online bulletin board than in academic essays. This is not at all surprising as expressing personal desire is one of the commonly found features in speech. According to Biber (1988, p.128), there is more personal involvement in conversational genres such as personal letters and spontaneous speech than academic prose. 7. Discussion and Teaching Implications Since English is taught as a foreign language in Thailand, Thai learners lack the opportunity to use English interactively outside of the classroom. English is taught simply as another subject in the school curriculum, so Thai students have very little opportunity to interact with others in English in their daily lives (Vitayapirak, 2001, p.1). The lack of enough exposure to natural use results in the learners being unaware of their inappropriate use of the language. Students are often marked for their grammatical correctness, but not appropriateness, so Thai students are very much concerned with how to produce grammatical sentences, without much attention given to whether the sentences are produced appropriately. Their use of language can thus be grammatically correct, but it may sound inappropriate or even impolite to the ears of the native speakers. This may be because they are not aware of the significance of how their utterances can be moderated or fine-tuned appropriately. The skill in fine-tuning can only be developed through authentic interaction with real participants not only through classroom exercises Fine-Tuning Discourse in Thai EFL 221 where students are secluded from the real world. The communication in online bulletin board writing is more interactive, so learners can have the chance to produce their own sentences as well as to exercise their repertoire of vocabulary in which they have been taught in the classroom. According to Swain’s (1985) output hypothesis, it is important that learners use the language and interact with others so that repertoire of vocabulary in which they have been taught through formal teaching can be exercised, developed, and later acquired. Before introducing online bulletin board writing as a communicative activity, teachers should teach their students how to choose the most appropriate fine-tuning devices; how to assert their commitment and support their claims; as well as how to moderate them when they are uncertain about their propositions. These findings suggest that there is a need for teachers to familiarize their students with the list of fine-tuning devices. For example, students should be informed that there are many adverbs which can be used to convey degree of probability, so that they do not have to rely on only few limited items and use them repeatedly. The findings in this study show that the students used MAYBE much more than PROBABLY, while British and American native speakers tend to use them fairly equally (Biber et al., 1999, p.869). The idea of using one form rather than many others seems to correspond with what Thomas (1983) has noted in that “L2 learners select one from many options of modal verbs and stick with it for their uses in all contexts” (p.103). Larsen- Freeman and Long (1991) agree, saying that “Learners will use only those aspects in which they have the most confidence” (p.26). Students should be made aware that if they use the same items too many times, their writing may sound redundant and lexically limited. These findings also indicate that having students write freely online among them without guidance may lead to the students’ incorrect use of modal and intensifying elements. Therefore, online bulletin board writing should be followed up with some activities such as a class discussion to survey some of the problems that students have when they perform the task. In other words, students should be given the opportunity to discuss in class with their teachers the difficulties they have in expressing themselves or conveying their meanings through these fine-tuning devices, and teachers should also be able to give them some explanation and feedback as to how their utterances could be expressed more appropriately: how to convey degree of doubt and certainty, how to commit themselves in terms of obligation and necessity, as well as how to express their desire or preference appropriately in real communication. 222 Chapter Ten Failure to provide feedback or correction could result in the fine-tuning elements being used inappropriately as found in this study. For example, the students’ use of expressions such as IN MY POINT OF VIEW is less likely to be used by English native speakers as FROM MY POINT OF VIEW , which is more preferable. Students should also be informed that FROM MY POINT OF VIEW should not always be used in all cases where IN MY OPINION is used (Swan, 2005, p.434). Another example is MAYBE, an epistemic modal adverb, which has often been confused with the use of modal auxiliary MAY plus copula “be”. For example, “That may be true” is sometimes written by learners as “That maybe true”, while sentence such as “Maybe, that is true” is written as “May be, that is true.” To improve the pragmatic competence in conveying doubt and certainty in English, it is not only the quantity and the variety of the devices being used, but also the ability to use of these modal devices correctly and appropriately in different genres that matter. 8. Conclusion This study proposes an analytical framework to be used for the study of modal and intensifying elements in English for Thai EFL learners’ writings in the two modes: online electronic bulletin board writing and academic essays. Based on the existing and well-established concepts of modality and degree of intensity, the framework distinguishes between four semantic categories of hedging, boosting, committal, and inclination. This study also provides a preliminary view on Thai learners’ repertoire of modal and intensifying elements in English. It is found that bulletin board writing can be used as a communicative activity as it plays a significant role in giving learners a chance to practice using modal and intensifying elements that are common in speech. The activity can help develop learners’ ability to express their meanings more subtly, showing the level of confidence or modesty through their utterances. However, simply counting modal items, regardless of their functions or how they are used in context may not provide a complete picture of learners’ commands of fine- tuning discourse in English. Therefore, a more in-depth study needs to be done at the micro pragmatic level to gain more insight into the learners’ fine-tuning discourse in English. This preliminary study can only attempt to answer the question in a very limited way; ultimately, the chapter’s more important task is to open up more pressing questions regarding the use of modality by non-native learners of English. Fine-Tuning Discourse in Thai EFL 223 References Altman, R. (1982). Interlanguage Modality. Paper presented at the 57th Annual Meeting of the Linguistic Society of America, San Diego, California. Bhandhufalck, F. (1983). Contrastive Analysis of Modals in English and Thai. Pasaa 13, 50-61. Biber, D. (1988). Variation Across Speech and Writing. Cambridge: Cambridge University Press. Biber, D., Johansson, S., Leech, G., Conrad, S. & Finegan, E. (1999). Longman Grammar of Spoken and Written English. London: Longman. Biber, D., Conrad, S. & Leech, G. (2002). Longman Student Grammar of Spoken and Written English. Essex: Pearson Education. Carter, R., & McCarthy, M. (2006). Cambridge Grammar of English. Cambridge: Cambridge University Press. Coates, J. (1983). The Semantics of the Modal Auxiliaries. London: Croom Helm. —. (1995). The Expression of Root and Epistemic Possibility in English. In J. Bybee & S. Fleischman (Eds.) Modality and Grammar in Discourse (503-517). Amsterdam and Philadelphia: John Benjamins:. Crystal, D. (2006). Language and the Internet. Cambridge: Cambridge University Press. De Cock, S., Granger, S., Leech, G. & McEnery, T. (1998). An Automated Approach to the Phrasicon of EFL Learners. In S. Granger (Ed.) Learner English on Computer (67-79). New York: Longman. Eggins, S. & Slade, D. (1997). Analyzing Casual Conversation. London: Equinox Publishing Ltd. Gibbons, J. & Markwick-Smith, V. (1992). Exploring the Use of a Systemic Semantic Description. International Journal of Applied Linguistics 2(1): 36-49. Halliday, M. A. K. (1994). An Introduction to Functional Grammar. London: Arnold. Halliday, M. A. K. & Matthiessen, C.M.I.M. (2004). An Introduction to Functional Grammar. London: Arnold. Hillier, H. (2004). Analysing Real Texts. Research studies in modern English language. New York: Palgrave Macmillan. Hinkel, E. (2005). Hedging, inflating, and persuading in L2 academic writing. Applied Language Learning 15(1&2): 29-53. 224 Chapter Ten Holmes, J. (1983). Speaking English with the Appropriate Degree of Conviction. In C. Brumfit (Ed.) Learning and teaching languages for communication (100-113). London: Center for Information on Language Teaching and Research for British Association for Applied Linguistics. —. (1988). Doubt and Certainty in ESL Textbooks. Applied Linguistics 9(1): 21-44. Hyland, K. (1996). Nurturing Hedges in the ESP Curriculum. System 24(4): 477-490. Hyland, K. & Milton, J. (1997). Qualification and Certainty in L1 and L2 Student’s Writing. Journal of Second Language Writing 6(2): 183-205. Karkkainen, E. (1992). Modality as a Strategy in Interaction. Epistemic modality in the language of native and non-native speakers of English. Pragmatics and Language Learning 3: 197-216. Larsen-Freeman, D. & Long, M.H. (1991). An Introduction to Second Language Acquisition Research. London: Longman. Martin, J. R. & Rose, D. (2003). Working with Discourse. New York: Continuum. Martin, J. R. & White, P. R. R. (2005). The Language of Evaluation. New York: Palgrave Macmillan. Oshima, A., & Hogue, A. (1997). Introduction to Academic Writing. Reading, MA: Addison-Wesley. Palmer, F. R. (1990). Modality and the English Modals. London: Longman. Perkins, M. R. (1983). Modal Expressions in English. London: Frances Pinter. Quirk, R., Greenbaum, G., Leech, G. & Svartvik, J. (1985). Comprehensive Grammar of the English Language. New York: Longman. Swain, M. (1985). Communicative Competence. Some roles of comprehensible input and comprehensible output in its development. In S. Gass & C. Madden (Eds.) Input in Second Language Acquisition (235-253). Rowley, Mass.: Newbury House. Swan, M. (2005). Practical English Usage. Oxford: Oxford University Press. Thomas, J. (1983). Cross-cultural Pragmatic Failure. Applied Linguistics 4(2): 91-112. Vitayapirak, J. (2001). A corpus-based approach to ESP lexicography the case study of English for Thai computer science students. Unpublished PhD dissertation King Mongkut's Institute of Technology, Ladkrabang. CHAPTER ELEVEN THE SEMANTICS OF GRADUATION: EXAMINING ESL LEARNERS’ USE OF GRADUATION OVER TIME CAROLINE LIPOVSKY AND AHMAR MAHBOOB 1. Introduction Additional language acquisition (ALL) research attempts to 1) describe learners’ language and 2) explain the acquisition process. ALL research, which is now in its fourth decade, was grounded in work in L1 acquisition in the 1960s. This L1 acquisition research examined how children acquired grammatical structures and how (and in what order) specific morphemes were acquired. Building their work on this tradition, early ALL researchers compared learners’ L1 and L2 and then posited where and why learners might have problems in learning an L2. However, by the late 1970’s it was becoming apparent that learners’ L2 development could not always be predicted by L1 because learners speaking various L1s shared developmental patterns—consequently new theories of SLA needed to account for the similarities of L2 patterns across learners from various L1 backgrounds. Researchers also realised that there was no single order of acquisition. Realising these limitations of early ALL research, researchers moved on to explore other dimensions of language acquisition. Some of these approaches, such as work by Lydia White (2003) drew heavily on Universal Grammar theories as their underlying framework. Other, more recent work takes a different approach and adopts (among others) Vygotskian theories of learning to explain additional language acquisition (Lantolf 2000). These socio-cultural theories of SLA are currently on the rise and focus on how learning occurs during social interaction. A different aspect of ALL research that emerged in the 1990s focuses on acquisition of pragmatics (e.g. Bardovi-Harlig and Hartford 1993, 226 Chapter Eleven Kasper and Schmidt 1996). Researchers examining the development of learners’ pragmatic systems argued that while learners may develop adequate syntactic patterns, they may not know how and where to use which structures—leading to pragmatic errors. A sizeable proportion of this research on pragmatics uses DCT (discourse completion tasks) data to explore how learners develop their ability to act appropriately when engaged in specific speech acts such as apologising, requesting, complaining, etc. These researchers focused on both specific lexical/grammatical items (such as the role of modals in development of pragmatics) as well as larger chunks in learners’ development of their pragmatic competence. In general, current work on SLA—regardless of these variations— adopts theories of language that separate language structure from its functions and then focuses on either the structure or the function of learner language. This work also largely focuses on lexical and sentence level features and does not include an analysis of larger texts. The present study adopts a Systemic Functional Linguistics (SFL) approach to studying learner language. SFL, based on work by M.A.K. Halliday, views language in terms of inter-related systems of choices that are available for expressing meaning. One of the tenets of this theory is that language structures cannot be idealised nor studied without looking at the circumstances of their use. In this paper, we attempt to show that SFL (and, more specifically, SFL’s Appraisal Framework) can contribute to ALL research by showing how language function and structure are dependent on each other and how learners’ ability to graduate evolves over time. The present study evolves from our research on ESL learners’ attitudes towards their native and non-native English speaking teachers (NESTs vs. NNESTs) (Mahboob and Lipovsky 2007). In this first project, we examined the language used by learners when discussing their native and non-native teachers. Drawing on the Appraisal theory, we examined a set of essays written by 19 Japanese high school students who had attended an ESL programme in the United States. Using the Appraisal framework, we analysed the linguistic resources used by these students when evaluating their NESTs and NNESTs, and compared the results of this analysis with previously used thematic analysis of similar data. We found that Appraisal analysis offered the tools to develop a much more detailed understanding of the students’ attitudes than previously used thematic analyses (Mahboob 2004). Following this study, we decided to examine learner essays to examine their use of Graduation over time (Graduation and other The Semantics of Graduation 227 SFL terms used here will be explained in the section on the ‘Theoretical Framework’). 2. Methodology The data for this project come from essays written by 19 Japanese high school students attending a 4-month ESL orientation program at the beginning of a year-long study abroad programme in the United States. The programme was taught by two TESOL professionals: one an NEST and another an NNEST. Program participants were asked to write a diagnostic essay at the beginning (T1) and end (T2) of the program on the topic: Some students think that only native speakers can be good language teachers. Others think that non-natives can also be efficient teachers. What is your opinion about this issue? Please feel free to provide details and examples. These essays were collected with two goals in mind: 1) to explore any shift in students’ perceptions towards native and non-native English speaking TESOL professionals, and 2) to evaluate students’ writing and grammar. The present study contributes to this latter goal. In specific, it explores how learners use various linguistic tools available to them to appraise NESTs and NNESTs and how their repertoire of tools for graduating their appraisals expands over time. Results of the study show how learners are able to exploit a wider array of graduation tools over time. It is possible to examine this as learners’ ability to appropriate tools available in the cultural reservoir of the learned language potential and making it part of their linguistic repertoire. In this paper, we will first briefly outline the Appraisal framework and describe more specifically the Graduation network by sharing examples from our data set. We will then share the results of our analysis and point out some of the differences in learners’ use of Graduation over the two data sets (T1 and T2) collected at an interval of 4 months. 3. Theoretical framework The system of Appraisal deals with evaluative meanings. It describes the linguistic means by which people align themselves with their readers/listeners, how they share their feelings and values with others and how they intensify or downgrade these feelings, hence the three systems of “Engagement”, “Attitude” and “Graduation” (Figure 1). In this chapter we 228 Chapter Eleven will focus on Attitude and Graduation, which are described in more detail below. monogloss ENGAGEMENT heterogloss AFFECT JUDGEMENT ATTITUDE APPRECIATION FORCE GRADUATION FOCUS Figure 1: Overview of Appraisal resources (adapted from Martin and Rose 2003) The system of Attitude (see Martin 2000, White 2002, Martin and Rose 2003, Martin and White 2005, and White 2005 for further descriptions) involves all types of evaluative assessments, either positive or negative. It describes how individuals share their feelings with others, assess people’s behaviours, and evaluate things and performances, hence the three categories of Affect, Judgement and Appreciation, as in the following examples (Attitudes are in italics): • Affect: I like this teacher. • Judgement: He is competent. • Appreciation: His classes are interesting. The Semantics of Graduation 229 Attitudinal meanings are gradable, so they can be amplified or, on the contrary, downgraded. Note that the Attitude system, as presented in Figure 1, is a discourse semantic system. This means that Attitude can be expressed through a range of grammatical structures (Hood 2004). Since explicit Attitude can be construed as a quality (e.g. “a competent teacher”), a nominalized quality or entity (e.g. “the advantage of…”), or as a process (e.g. “enhance”), there is a strong correlation in the realization of Graduation with particular grammatical structures, such as pre-modifying intensifiers (e.g. “a very competent teacher”), pre-modifications of a nominalized quality (e.g. “the main advantage of…”), or circumstances of manner (e.g. “strongly enhance”). However, these grammatical realizations for grading Attitudes are not inherent to the system, and other realizations can be expected, as detailed below. There are two kinds of resources for scaling attitudinal meanings. The first kind involves sharpening or softening semantic categories that are a priori not scalable, as in the following example from our data: (1) Non-natives teachers support us, when we have homesich [-AFF] or something [GRA: Focus]. #4T21 (2) Teacher, my host family and my friends told me real [GRA: Focus/+APP] English! #14T2 In example (1), “or something” tempers the specificity of homesickness and seems to extend teachers’ support to students suffering from other conditions. In example (2) on the other hand, the student’s use of “real” seems to scale the English that she learnt in Japan and in America, implying that the English that she spoke in Japan was inferior to the one that she learnt while in the United States. This type of graduation is referred to as “Focus” (see Figure 2). sharpen real English, etc FOCUS soften homesick or something, etc Figure 2: Network of choices in Focus (adapted from Martin and White 2005) 230 Chapter Eleven The second kind of graduation resource is referred to as “Force”. Force involves degrees in intensity, quantity and enhancement. Intensification may scale qualities, as in example (3): (3) When I was first year of high school there was a English teacher [NNEST]. I think she is very [GRA: Force: intensity] smart [+JUD] because she knew whatever I want know. #3T2 (4) They became great [GRA/+JUD] speakers of the specific language even they are not native speakers. #19T2 In (3), the pre-modifying intensifier “very” amplifies the teacher’s smartness—conversely, the teacher could be judged as “somewhat smart”. “Smart” is intensified through a grammatical item with no referential meaning (i.e. “very”), whereas in (4) the intensifier is fused within a lexical item that also serves a semantic function—”great” could be unpacked as “very” + “good”. Examples 5-7 below show realizations of intensification through repetition (5 and 6) and punctuation (7). (5) He [NNEST] studied very very very [GRA: Force: intensity] hard [+JUD]. So. He is a very very [GRA: Force: intensity] good [+JUD] English teacher for us. #12T1 (6) She is not native speaker, but she studied and studied [GRA: Force: intensity/+JUD] English in order to be a good teacher. #15T1 (7) I always try my best [+JUD] !!!! [GRA: Force: Intensity]. #12T1 Quantification is another way to grade Attitudes. It applies to entities, as shown in these examples from our data: (8) I think, native speakers know many [GRA: Force: quantity: amount] word [t, +JUD]. #2T2 (9) He [NNEST] has many many [GRA: Force: quantity: amount] experiences and knowleges [+JUD]. #1T2 (10) Non-natives teachers teach me how to learn second language, how to make friend in the country I don’t know anything, and many other things [GRA: Force: quantity: amount] [t, +JUD]. #4T2 Most of our examples so far have involved “inscribed” or explicit Attitude (see Martin 2000). Example (8) however shows how Attitude can be evoked through the grading of non-attitudinal meanings (see Hood The Semantics of Graduation 231 2004). The word “many” which is associated to “word” encourages an evaluative reading, highlighting native speakers’ privileged knowledge, compared with non-native speakers, in their own tongue. In the same manner, the listing structure in (10) evokes the student’s positive attitude towards NNESTs, as the multiplicity of NNESTs’ skills attests their teaching competence. This is further reinforced by the implication of other skills with the words “and many other things”.2 There are other options for grading as quantity. Extent refers to something’s scope or distance in time or space, as in the following examples: (11) She [NNEST] said she have studied English so hard for 10 years [GRA: Force: Extent: Scope: Time] [+JUD]. #8T2 (12) She [NNEST] has never talked native English speakers. “I tried to speak to foreign people when I found them at the station, park and even at the hospital [GRA: Force: Extent: Scope: Space] [t, +JUD].” She said. #15T2 In (11), the prepositional phrase “for 10 years” evokes Attitude by emphasizing the teacher’s efforts to learn English, whereas the listing of places in (12) implies her Tenacity (this is further implied by the counter- expectancy “even at the hospital”). Another way of graduating links processes to degrees of speed, strength, effort and so on, e.g.: (13) They [NNESTs] know which word we learned fast [GRA: Force: enhancement: vigour] [t, +JUD]. #2T2 (14) He [NNEST] taught me easily [GRA: Force: enhancement: vigour] [t, +JUD]. #3T2 In (13) and (14), enhancement is encoded into the circumstance of manner. Lastly, processes can be enhanced as frequency, as in the following examples: (15) When she [NNEST] came back to Japan, she studied more and went to America again [GRA: Force: enhancement: frequency] [t, +JUD]. She returned it many times [GRA: Force: enhancement: frequency] [t, +JUD. #1T1 232 Chapter Eleven (16) She [NNEST] wrote a letter everyday [GRA: Force: enhancement: frequency] to her host family [t, +JUD]. #1T2 In examples (15) and (16), the Adjuncts “again”, “many times” and “everyday” emphasize the NNEST’s tenacity in her study of the English language. Figure 3 sums up the network of choices in Force. Note that the essays contain no instances of quantification for distance in time/space. intensity very smart, great speakers, very very good, studied and studied amount many experiences time quantity for 10 years FORCE scope space extent at the station, park… distance time space vigour he taught me easily enhancement frequency she went to America again she returned it many times Figure 3: Network of choices in Force (adapted from Hood 2004) The two sets of data collected for this study were coded for all occurrences of Graduation and these were categorized using the framework described above. In the following section we will focus on some of the key findings of this analysis. 4. Findings In analyzing the data, we became interested in three types of resources exploited by the learners for graduating their attitudes towards their N/NESTs: intensification, quantification and enhancement. The Semantics of Graduation 233 In our data, students chose to intensify their attitudes through exclamatives, intensifiers, repetitions, and a combination of intensifier and repetition. The following are some examples of such intensification: Exclamatives: I always try my best!!!! #12T1 Intensifiers: Adverbs - He [NNEST] taught me easily. #3T2 Comparatives - Her pronusation [NEST’s] was much better than Japanese teachers’. #18T2 Superlatives - I really respect her the best in the world because she had never given up her dream and she kept studying. #1T2 Repetitions: She [NNEST] studied and studied English in order to be a good teacher. #15T1 Combination of intensifier and repetition: He [NNEST] is a very very good English teacher for us (#12T1) or “She [NNEST] studied English harder and harder” (#15T2). These examples highlight how intensification was realized via isolated lexemes (e.g. “very smart”) and repetition (e.g. “very very smart”), rather than semantic infusion (e.g. “brilliant”). Secondly, learners used non- figurative graduation (e.g. “very clear”), rather than figurative (e.g. “crystal clear”not used in the essays). Lastly, they graduated through grammatical realization (e.g. “very smart”), rather than lexical realization (e.g. “amazingly smart”not used in the essays). When quantifying, learners used pre-modifications of a nominalized quality rather than quantification carried by the noun head itself. Thus, they wrote “He [NNEST] has many many experiences and knowleges” (pre-modification), rather than “He has a profusion of experiences and knowleges” (examples not found in our data). Lastly, when enhancing, participants in this study tended to use circumstances of manner rather than infusion of the verb. Thus, they wrote “we learned fast” (circumstance of manner), rather than “we soaked up…”. 4.1 Comparison of T1 and T2 essays In this section, we will look at some of the differences in learners’ use of resources to graduate over time. To start, it should also be noted that the T2 essays are much longer than the T1 essays: about 70% longer in average. Given this difference in the length of the essays, the raw number 234 Chapter Eleven of Graduation forms in the data cannot be used in understanding learner language. Therefore, in order to observe any differences between T1 and T2, we have presented the data in terms of percentages. In the same manner, in order to check for the repetitive use of specific forms, we chose to present the data by type rather than by token. In comparing learner data over time, we noted an increased use of Graduation (both Force and Focus) in absolute terms (almost twice as many items). However, given the relative length of the essays, students used Graduation just as much: 5.0% for T2, compared with 4.6% for T1 essays (see Appendix A). Having said this, there are some interesting differences as well: there was a rise in the use of items of Focus as compared to Force, there was more use of quantification and enhancement (vigour), and learners used infusion at T2 (there was no use of infusion at T1). One of the key differences between T1 and T2 is in the use of items of Focus and Force. Table 1 shows the comparative use of Focus and Force at T1 and T2 and reveals that while the relative instantiations of force decreased over time, there was an increase in the use of Focus. Table 1: Comparative use of Focus and Force (T1/T2) T1 T2 Instantiation of Force 84% 73% Instantiation of Focus 16% 27% A closer look at the instantiations of Focus shows that learners used more infusion (e.g. “great”, “goes on”, “kept”, “mastered”) in the essays at T2 than T1. There are eight (8) tokens of infusion in data for T2 compared to only three (3) tokens for T1. A proportional study of the different kinds of Force was also carried out to understand learners’ use. Figures 4a & 4b below show an increase in Graduation through quantification (amount + extent: 43% in T2 vs. 37% for T1) and enhancement (vigour + frequency: 20% in T2 vs. 18% in T1) as compared to intensification, which shows a decrease (37% in T2 vs. 45% in T1). Within these categories, quantification (amount) increased proportionally, as did enhancement (vigour). The Semantics of Graduation 235 Enhancement:  vigour Intensification 5% 45% Enhancement:  frequency 13% Quantification:  extent: scope:  time Quantification:  17% amount 20% Figure 4a: Force by type (T1) Enhancement:  Intensification vigour 37% 13% Enhancement:  frequency 7% Quantification:  Quantification:  extent: scope:  amount time 28% 15% Figure 4b: Force by type (T2) A discrete analysis of the different kinds of Force shows differences within intensification, quantification, and enhancement as well. In terms of intensification, learners used more comparatives (e.g. ‘better’, ‘easier’, ‘more’), superlatives (e.g. ‘best’, ’easiest’, ‘most’) and hedges (e.g. ‘a little’, ‘little’, ‘a little bit’) at T2 as compared to T1. Table 2 below gives a numerical summary of these differences. Table 2: Summary of intensification (T1/T2) T1 T2 Comparatives 8 19 Superlatives 6 9 Hedges 0 4 236 Chapter Eleven In addition to a numerical comparison of the means of intensification, a proportional examination of intensification shows an increase in students’ use of comparatives (19% in T2 essays vs. 11% in T1 essays) and superlatives (19% in T2 essays vs. 16% in T1 essays), and a decreased use of repetition combined with intensifiers (e.g. “very very”) (6% in T2 essays vs. 16% in T1 essays). These shifts can be seen by comparing the distribution of the means of intensification at T1 and T2 in Figure 5a & 5b. Intensifier  Exclamative (comparative) 5% 11% Infusion Intensifier/  repetition 5% 16% Repetition 5% Intensifier  (superlative) 16% Intensifier 42% Figure 5a: Means of intensifying by type (T1) Intensifier  Exclamative (comparative) 6% 19% Infusion 6% Repetition Intensifier/  6% repetition 6% Intensifier  Intensifier (superlative) 38% 19% Figure 5b: Means of intensifying by type (T2) The data also show that learners started using listing as one of the ways to quantify items over the four-month study period. There were no instances of listing at T1 and five (5) at T2. This rise reflected the general increase in the use of quantification by our participants over time. Another difference in the increase in the use of quantification is in “quantification as amount” (e.g. “a lot of”, “many”, etc.) – learners used three (3) times as The Semantics of Graduation 237 much “quantification as amount” in their T2 essays as compared with T1 essays. Finally, T2 essays contain more instances of enhancement (e.g. ‘clearly’, ‘easily’, ’fast’, perfectly’, ‘well’): 8 for T2 vs. 2 for T1. In conducting our analysis, we considered specific lexical items used for graduating. Figures 6 and 7 below indicate lexemes that were used twice or more in each corpus of essays. These two figures highlight an increase in the second set of essays in quantification through non-specific numeration (e.g. “many”, “more”, “a lot of”) and listings (compare with expressions such as “for example” or “and so on” in the T1 set of essays), as well as an increase in the use of superlatives (e.g. “best”, “most”) and enhancement through adverbs (e.g. “really”, “clearly”, “easily”). The figures also highlight an increase in infusion (e.g. “great”, “kept”). Figure 6: Instantiations of Force (T1) Figure 7: Instantiations of Force (T2) 238 Chapter Eleven 5. Expanding resources: some preliminary hypotheses The results discussed in the previous section highlight the linguistic resources used by the participants in this study to graduate their beliefs about NESTs and NNESTs in written essays. The results also show the differences in learners’ use of various resources as they developed their language skills over a period of four months. In considering these differences in learners’ use of Graduation over time, we have developed a set of four preliminary hypotheses that may explain learners’ expansion of their repertoire of resources used for Graduation. Hypothesis 1: In incorporating Graduation into their language, beginning additional language learners appear to favor system choices that are syntactically less complex. The evidence for this comes from our observation that learners prefer structures that are syntactically less complex, e.g. they realize intensification through isolated lexemes (e.g. “very good”) or repetition (e.g. “very very good”) rather than semantic infusion (e.g. “great”). Hypothesis 2: In incorporating Graduation into their language, beginning additional language learners appear to favor system choices that can be extended to other choices within the system network. For instance, our data show that once learners start using repetition within intensification (e.g. “very very”), they may extend the use of repetition to quantification (e.g. “many many”). Hypothesis 3: In incorporating Graduation into their language, beginning additional language learners appear to favor grammatical realizations rather than lexical realizations. For example, learners use grammatical realizations such as “very smart” rather than lexical realizations such as “amazingly smart”. In fact, our data did not contain any examples of lexical realization. This suggests that lexical realization of Graduation is acquired at an advanced stage of language acquisition. Hypothesis 4: In incorporating Graduation into their language, beginning additional language learners appear to favor realizations that are non- figurative (e.g. “very clear”) rather than figurative (“crystal clear”). Again, our data only contained examples of non-figurative realization, suggesting that figurative realization of Graduation is acquired at a later stage. The Semantics of Graduation 239 These four hypotheses are supported by the data collected for the study, but cannot be generalized at the moment. Additional data need to be collected and further research done to evaluate these hypotheses. 6. Summary and conclusion One function that language plays is to express and share our opinions and beliefs about things and people. In SFL, this function is realized through the Appraisal system. One aspect of the Appraisal system is Graduation – resources for scaling attitudinal meanings. In this study, we examined 19 English language learners’ use of graduation at two points in time (with a gap of four months). The results of the study led us to present four initial hypotheses. These hypotheses suggest that learners (1) favor network choices that are syntactically less complex (2) favor realization of network choices that can be extended to other choices within the network, (3) favor grammatical realizations rather than lexical realizations and (4) favor realizations that are non-figurative rather than figurative. The findings of this study – and the four hypotheses – are a starting point for further research on learners’ use of Graduation. Given that no prior research is available in this area, the purpose of this chapter was to develop an initial understanding of learners’ use of Graduation. We hope that additional research can follow up on this work and evaluate the findings presented here. Notes 1. The extracts from the essays are shown as written by the students. ‘T1’ refers to the first set of essays, ‘T2’ to the second set. Numbers refer to students. The coding for Attitudes is indicated in the brackets. ‘AFF’ stands for ‘AFFECT’, ‘JUD’ for ‘JUDGEMENT’, ‘APP’ for ‘APPRECIATION’ and ‘GRA’ for ‘GRADUATION’. ‘+’ indicates a positive Attitude whereas ‘-’ indicates a negative Attitude. The letter ‘t’ for ‘token’ indicates an evoked or non-explicit Attitude. Graduations are marked in bold. 2. These examples also highlight how an evaluative reading of the text relies on the surrounding text (see Hood 2004). 240 Chapter Eleven References Bardovi-Harlig, K. & Hartford, B. (1993). Learning the rules of academic talk: A longitudinal study of pragmatic development. Studies in Second Language Acquisition 15, 279-304. Hood, S. (2004). Appraising Research: Taking a stance in academic writing. Unpublished PhD thesis, University of Technology, Sydney. Kasper G. & Schmidt, R. (1996). Developmental issues in interlanguage pragmatics. Studies in Second Language Acquisition, 18, 149-169. Lantolf, J. (2000). Sociocultural theory and second language learning. Oxford: Oxford University Press. Mahboob, A. (2004). Native or nonnative: What do the students think? In L. D. Kamhi-Stein (Ed.) Learning and Teaching from Experience: Perspectives on Nonnative English-Speaking Professionals (121–147). Ann Arbor: University of Michigan Press. Mahboob, A. & Lipovsky, C. (2007). Examining attitudes towards NESTs and NNESTs: a comparison of a thematic vs. an appraisal analysis. In C. Gitsaki (Ed.) Language and Languages: Global and Local Tensions (292-306). Newcastle, UK: Cambridge Scholars Publishing. Martin, J.R. (2000). Beyond Exchange: APPRAISAL Systems in English. In S. Hunston & G. Thompson (Eds.) Evaluation in text: authorial stance and the construction of discourse (142–175). Oxford: Oxford University Press. Martin, J.R. & Rose, D. (2003). Working with discourse: Meaning beyond the clause. London New York: Continuum. Martin, J.R. & White, P.R.R. (2005). The language of evaluation: Appraisal in English. New York: Palgrave Macmillan. White, L. (2003). Second language acquisition and Universal Grammar. New York: Cambridge University Press. White, P.R.R. (2002). Appraisal. In J. Verschueren, J. Oestman, J. Blommaert & C. Bulcaen (Eds.) Handbook of Pragmatics (1-26). Amsterdam/Philadelphia: John Benjamins Publ.Co. White, P.R.R. (2005). The Appraisal web site, from www.grammatics.com/appraisal. Accessed January 6, 2008. The Semantics of Graduation 241 APPENDIX A Student No. of Grad. % No. of Grad. % words T1 words T2 T1 T2 #1 134 8 5.9% 235 13 5.5% #2 126 2 1.6% 221 7 3.2% #3 69 7 10.1% 257 15 5.8% #4 82 3 3.6% 157 8 5.1% #5 174 7 4.0% 170 3 1.8% #6 103 3 2.9% 215 10 4.7% #7 101 4 3.9% 139 7 5.0% #8 187 7 3.7% 232 11 4.7% #9 112 2 1.8% 250 11 4.4% #10 60 0 - 84 3 3.6% #11 41 2 4.9% 128 6 4.7% #12 194 17 8.7% 230 17 7.4% #13 50 2 4.0% 179 21 11.7% #14 134 9 6.7% 159 7 4.4% #15 209 9 4.3% 309 13 4.2% #17 75 5 6.7% 154 8 5.2% #18 105 5 4.8% 352 15 4.3% #19 167 9 5.4% 260 11 4.2% #20 191 6 3.1% 189 10 5.3% TOTAL 2,314 107 4.6% 3,920 196 5.0% CHAPTER TWELVE ANALYSIS OF JAPANESE SPOKEN BY ELDERLY TAIWANESE: WORD USAGE, PARTICLE USAGES, AND PREDICATE FORMS MASUMI KAI 1. Introduction Taiwan was a Japanese colony from 1895 to 1945. During this period Japanese was the official language of Taiwan and education in school was conducted in Japanese. Elementary education was widespread during the Japanese colonial period. According to Copper (1990) the enrolment numbers were as high as 81% for males and 61% for females by 1943. This high rate of education implied that a large proportion of the population could speak and write Japanese. After WWII, Taiwan was taken over by Chiang Kai-shek's Nationalist Party, and the official language of the island changed to Mandarin. Today, even though more than 60 years have passed since the end of the Japanese colonial period, there are still many elderly Taiwanese who speak Japanese. This chapter examines data from conversations with four of these elderly Taiwanese people and looks at the characteristics of their Japanese language. More specifically, the chapter investigates the characteristics of their Japanese and analyses any grammatical variations between their Japanese and that spoken in Japan now. As I demonstrate in this chapter, the Japanese spoken by elderly Taiwanese, which I shall call Taiwan-Japanese here, is different from modern Japanese currently spoken in Japan. However, it has unique characteristics which can distinguish it as a dialect of Japanese. Taiwan- Japanese is now on the verge of dying out, therefore it needs to be recorded and analysed by linguists. There is numerous research discussing the educational system and textbooks used during the colonial periods. There is also research analysing Japanese words remaining in the Analysis of Japanese Spoken by Elderly Taiwanese 243 Taiwanese language, Min-nan.1 However, there do not seem to be any linguistic analyses of conversations by elderly Taiwanese. In this sense, I believe this chapter provides important data. 2. Methods of Analysis and Subjects’ Background The data for this study was collected through interviews with four Japanese-speaking elderly Taiwanese in 2004. The data consists of one 30-minute conversation per person. Transcriptions were made in the UC, Santa Barbara style format (see Appendix for abbreviations). The analysis of the data is based on the comparison of the Japanese spoken by these four participants to modern Japanese. Before looking at the four participants, it would be useful to briefly consider the ethnic and linguistic history of Taiwan. There are three main ethnic groups in Taiwan: 1) Chinese who came from the southern part of China during the period starting from the 17th century to the end of the Qin dynasty (1895) (Taiwanese), 2) Chinese who came from China with Chang Kai-shek’s Nationalist party after WWII (Mainlander), 3) the early inhabitants (Malay-Polynesian) which include ten major tribes. According to Wakabayashi et. al. (1990), the Taiwanese, who are early Chinese immigrants,account for 84.3% of the island’s population; the Mainlanders, who immigrated to Taiwan after WWII, represent 14%; and early inhabitants, 2 1.7 %, as of 1987. The mother tongue of the Taiwanese is Min-nan. By saying Taiwanese, this also includes Hakka people. They speak Hakka, but they can usually understand Min-nan. The Mainlanders are mostly monolingual Mandarin speakers. The early inhabitants have their own tribal languages and basically do not speak Mandarin nor Min-nan. People who received a Japanese education during the colonial period are Taiwanese or early inhabitants. My four subjects are Taiwanese. Taiwanese generally see the Japanese colonial period in a positive light, which caused most of them to retain their Japanese (see Kai 1997 for a discussion of assessments of the Japanese language and Japan by elderly Taiwanese, ways of retaining Japanese, and frequency of use of Japanese in their daily life from questionnaires collected from over 800 elderly Taiwanese). At the time of the interview one subject, A, was 82. The three other subjects, B, C, and D were 80 years old. Three subjects (A, B, C) received education through middle school and one subject (D) received higher education up through medical school in Japan. Subject D had used Japanese often for business or with Japanese friends, but he had not had a chance to use it for a couple of years leading up to the time of the 244 Chapter Twelve interview. He was suffering from dementia and was staying at home. Subjects A, B and C use simple Japanese words mixed with Min-nan when they meet with their friends of the same age. Subject B was also using Japanese words at home in conversations with her husband. The four subjects still could manage Japanese despite the fact that they did not use it frequently anymore. However, their Japanese differs from modern Japanese in that there are certain structures used by them that a modern Japanese speaker would not use. The Japanese of elderly Taiwanese, the Taiwan-Japanese, deviates from modern Japanese and has characteristics of a dialect of Japanese, which is now almost extinct. 3. Types of Deviations There were several types of grammatical deviations from standard modern Japanese seen in the four subjects’ Japanese. See the Table 1 below. Table 1: Type of deviations Subjects A B C D Type of Deviations lexical borrowings/expansions 61 58 31 31 tense or aspects 18 15 10 0 particle 18 7 7 0 demonstratives 13 1 5 0 verb/adjective inflection 10 4 2 0 final particle 1 0 0 0 The Japanese of subject D, who received a higher education, demonstrated fewer deviations. They were mainly of the lexical type – lexical borrowings/expansion. In the Japanese spoken by the other three subjects, however, there were several types of grammatical deviations as seen above. Among these deviations, the most frequently occurring type was lexical borrowing or expansions (for the English variation, see Thomason and Kaufman 1988, Kortmann & Schneider 2004). There are three varieties of this type in the subjects’ Japanese: (a) the subject used a word which is understandable but is not used in modern Japanese, (b) the subject expanded a word class, such as using a noun as a verb or vice visa, (c) the subject overgeneralised and created a new word. Analysis of Japanese Spoken by Elderly Taiwanese 245 A tense/aspect deviation is when the subject did not use the past tense or aspect form. In Min-nan there is no past tense. There are only 3 particles and no verb or adjective inflections, or final particles. Japanese has three distinguished demonstrative pronouns, kore (something close to the speaker), sore (something close to the hearer), and are (something far from both the speaker and the hearer). Min-nan has only two types of demonstrative pronouns, like ‘this’ and ‘that’ in English. The subjects showed deviations in the usage of these items. In this chapter, I will discuss the first type, lexical borrowing/extensions, and the third type, particle usage, of deviations shown in the table 1. 4. Lexical Borrowing or Expansions First, I discuss the most frequently occurring deviations: lexical borrowing or expansions. As I mentioned earlier, there were three varieties of this type. Of the examples below,3 Example (1) and Example (2) are (a)-type: the subject used a word which is understandable but is not used in modern Japanese. In (1), subject B meant to say “(there is a) difference” but B used 差別 (sabetsu) which means “discrimination”. This may be a result of transfer from their first language, Min-nan. In Min-nan, 差別 means difference, but the direct translation of that letter to Japanese means discrimination. (1) B: toshiyori no iu Nihongo to, wakai hito no iu Nihongo old:people of say Japanese:language with young people of say Japanese:language wa ne, sono Nihonjin ga% to Nihonjin ga iu, kotoba TOP FP that Japanese:people SUB with Japanese:people SUB say words → nioite mo sabetsu ga aru-n desu yo. regarding also discrimination SUB exist-NOM COP FP ‘As far as Japanese used by old people and Japanese used by young people, Japanese has differences.’ In Example (2), subject C intended to say “talk”, which is hanasu in Japanese, but C used hanashiatteru instead, which means “be discussing”. This type of lexical borrowing was widely seen in my data of Taiwan- Japanese. (2) 1 C: sorede ano=, Nihongo mo= watashi=, soo ne, therefore DM Japanese:language also I let:me:see FP itsumo shujin to ne=, always husband with FP 246 Chapter Twelve ‘Therefore, well, with my husband 2 INT: un. DM → 3 C: hanashiatteru. discussing I always talk in Japanese, too.’ Examples (3) and (4) below exhibit a (b)-type deviation: the expansion of word class. In Example (3), man (full age) is a noun but subject A used it as a verb. In Example (4), kanshin (interest) is a noun but subject D used it as a verb. (3) 1 INT: a, Taiwan no kazoe kata de [hachijyuu-san-sai]. DM of calculation way by eighty-three-years:old ‘Oh, by way of Taiwanese calculation, (you are) 83 years old.’ 2 A: [hachijyuu-san-sai]. eighty-three-years:old ‘(I am) 83 years old.’ 3 INT: ha=[2=2]. DM 4 A: [2hai2]. Nihon no kazoe% kazoeta-- nen% yes Japanese of calculation → mada, man% man-shite-nai. not:yet full:age-doing-NEG ‘Yes. By way of Japanese calculation, (I am) not yet completely (83 years old).’ (4) 1 D: kondo anta,.. Manshuu-ryoodo ni, this:time you Manchuria-territory to 2 INT: un. DM → 3 D: motto kanshinse-nyaikan wa na. more be:interested-have:to FP FP ‘Now you should be more interested in the Manchuria territory.’ Analysis of Japanese Spoken by Elderly Taiwanese 247 The next example (5) below is a (c)-type deviation: the subject overgeneralised and created a new word. Example (5) shows the regularization of this type by subject D. In Japanese, some nouns can be changed to a sentence adverb by adding ni or teki ni. (e.g.) jissai (actual) Æ jissai-ni (actually) keizai (economic) Æ keizai-teki-ni (economically) Deciding whether a noun needs teki or teki ni to be an adverb cannot be determined by morphological rule. This is lexically determined. Subject D added teki ni to all nouns to make them adverbs. In modern Japanese, “periodically” is teiki-teki-ni as in line 7. “Frequently” is hinpan-ni, without teki, but subject D used hinpan-teki-ni as in line 5. (5) 1 INT: i% ima wa mo=, ano=, rajio=-- rajio yatteru-tte now TOP anymore DM radio radio doing-QUO yutte-ta-desho? saying-PAST-COP:TAG ‘Now lately, radio--, you said you were doing radio, didn’t you? 2 INT: rajio de=, nanka=. radio INS something ‘(You were doing) something on the radio.’ 3 D: ... e... are wa ima=, DM that TOP now ‘That is now, 4 INT: mou yattei-masen? anymore doing-NEG ‘(You) are not doing (that) anymore?’ → 5 D: ee.. sou *hinpan-teki-ni wa yattei-masen kedo. yes such frequency-ADV-ADV TOP doing-NEG though ‘Yes, (I) am not doing (radio) so frequently (now).’ 6 INT: ha==n. DM → 7 D: he\. he\. teiki-teki-ni wa ne. yeah yeah periodic-ADV-ADV TOP FP ‘Yeah. Yeah. Not regularly.’ 248 Chapter Twelve Aside from these few deviations, however, the subject D’s Japanese and intonation is almost identical to that of a native speaker of modern Japanese. This is in sharp contrast to the other three subjects, who had several types of deviations (as seen in Table 1) and also exhibited a unique intonation. In the next section, I will discuss another type of deviation: particle usage. 5. Particle Usage The deviations of particle usage were the third most commonly occurring characteristics in the Japanese of the four subjects. In modern standard Japanese, grammatical relations are indicated by case markers in Table 2: Table 2: Case markers in modern standard Japanese wa topic marker de locative case marker for an ga subject case marker action verb o object case marker e or ni directive case marker no noun connector de instrumental case marker ni ‘in’ ‘on’ ‘at’ Min-nan does not have particles with the exception of a locative marker and two noun connectors. For this reason, the subjects can be expected to demonstrate particle deviations. However, as Table 3 below indicates, the four subjects did not show many deviations, contrary to expectations. Table 3: Particle usage Same Ellipsis Usage as Optional Obligatory Deviations Total Modern (per Total) Ellipsis Ellipsis Japanese (per Total) (per Total) (per Total) D 324 16 2 0 342 (94.7%) (4.7%) (0.6%) (0%) C 273 67 5 7 352 (77.6%) (19%) (1.4%) (1.9%) B 487 40 2 7 536 (90.9%) (7.5%) (0.3%) (1.3%) 318 51 2 18 389 A (81.7%) (13.1%) (0.5%) (4.6%) Analysis of Japanese Spoken by Elderly Taiwanese 249 The subjects used about 350 particles on average in the same manner as a modern Japanese speaker, as indicated in the first column. They also omitted particles. In modern standard Japanese conversation some particles can be omitted, especially the topic marker wa and object marker o. That is an optional ellipsis. Certain types of conversational structures cannot have particles. These are obligatory ellipsis. They occur mostly in interrogative sentences with a plain form predicate4 and in exclamatory sentences (Onoe 1987, Kai 1991). The table above shows that the percentage of deviations of particle usages was low: less than 5% for everyone. The distribution of particle deviations is shown in Table 4 below: Table 4: Types of particle deviations D C B A wa TOP 0 0 1 4 wa NEG 0 0 0 0 ga SUB 0 1 1 1 ga OBJ 0 0 0 0 o OBJ 0 1 0 4 ni or e DIR 0 1 2 4 ni Time 0 0 0 0 de LOC 0 1 1 1 no ‘of’ 0 1 1 0 kara 'from' 0 0 0 0 to 'with' 0 0 0 0 un-certain 0 0 0 1 no particle 0 2 0 3 Total 0 7 6 18 Wa NEG indicates the topic marker wa which needs to be used with a negative predicate. Uncertain means that multiple particles can be recovered. No particle refers to the times the subject used a particle in a structure which cannot have a particle in modern Japanese. The following are some examples of the deviations for brief reference. The words in the parenthesis ( ) are the particles that are used in modern Japanese. 250 Chapter Twelve (6) 1. INT: Giransen.. [de] soko de shoogakkoo ni/? then there LOC elementary:school DIR ‘At Giran-prefecture... (you entered) elementary school there?’ 2. A: [hai]. yes ‘Yes.’ 3. A: e=, Gitankan.. genzai no Se% Seesui,.. ki% DM current of ki%, .. Seisui no,.. ano= kooen ga aru no. of DM park SUB there:is FP ‘Well, in Gitankan,..which is currently Seesui,.. there is Seesui park.’ 4. INT: un. DM → 5. A: asoko *ni= (wa), ima,.. totemo= nigiyaka desu that:place in TOP now very lively COP yo. FP ‘That place is now very lively.’ (7) → C: de itsumo tegami *ga (o), ano=, okutte-kuru no yo. then always letter SUB OBJ DM send-come FP FP ‘Then (my teacher) always sent (me) letters.’ (8) 1. B: Daitoowa-sensoo ni moshimo katte-kure-tara, the:Pacific-war in by:any:chance win-BEN-if ‘If (Japan) had won the Pacific War, 2. INT: un. DM 3. B: korya mukashi no hanashi. this:is old:time of story this is the old story.’ 4. INT: un. DM Analysis of Japanese Spoken by Elderly Taiwanese 251 5. B: katte-kure-tara, watashi nanka osoraku ima win-BEN-if I like probably now → Taiwan *de (ni) ora-nai desu yo. LOC in be-NEG COP FP ‘If (Japan) had won (the Pacific War), a person like me probably wouldn’t be in Taiwan now.’ 6. Predicate Forms One characteristic of modern standard Japanese conversation is the use of various morphemes at the ends of sentences. These morphemes create cohesion with a previous utterance. Examples of such morphemes are: final particles which express information territory (Kamio 1990), -n form + copula and kara which express causes or reason, kedo and ga which add information, and the like. These morphemes were seen often in subject D’s Japanese, and sometimes in subject B’s Japanese. In Example (9) below, subject D used kara (because), -n desu, and ga ne (although + final particle). In Example (10), subject B used shi (and) and jya-nai (tag question). These are not grammatically required items. They can be omitted, although if they are, the conversation becomes fragmented and sounds like merely a listing of sentences without any conjunctions. (9) 1 D: Taiwan no hito de ne, Manshuu ni sumikon-da of people COP FP Manchu DIR live:in-PAST → no ga,.. sanjyuuman gurai orimasu kara. NOM SUB 300,000 about exist because ‘Among Taiwanese people, there are about 300,000 who moved to Manchu.’ 2 INT: hu=n. DM → 3 D: .. mukashi wa motto ookat-ta-n-desu ga ne. old:time TOP much many-PAST-NOM-COP although FP ‘In the old days, there were more people (who moved to Manchu) though.’ (10) → 1 B: tokoroga ima, toshi ga toshi da shi=, however now age SUB age COP and ‘But now my age is so high, 252 Chapter Twelve 2 C: @@@. 3 B: iku-to ne,.. man’ichi nani ga atta-baai, soko o go-if FP just:in:case something SUB happen-case that OBJ kooryosshi% shi-[tara ne], yap[2pa2]-- consider -if FP after:all if I go (to Japan), ( I worry) something may happen, I may consider it, then after all-- 4 INT: [u=n]. DM 5 C: [2anta2] mada you still kenkoo de ii [3yo3]. healthy and good FP ‘It is good that you are still healthy.’ 6 B: [3yappa3] ryokoodan to isshoni after:all tourist:group with together → it-ta-hoo-ga, tayorini naru jyanai. go-better-SUB reliable become COP:TAG ‘After all, going with a tourist group isn’t reliable?’ Contrary to subject D and much of subject B’s Japanese, subjects A and C did not use morphemes as often. They simply added final particles directly to the plain form or they omitted the predicate. This omitting of sentence ending morphemes is not a deviation. It can be characterized as one of the unique features of the Taiwan-Japanese. Japanese has two predicate forms. One is a polite form (i.e., masu- or desu-form), and the other is a plain form which is usually used in casual conversation and in formal writing. Simply adding a final particle to the plain form or omitting a predicate does not create enough cohesion in a conversation and it makes utterances sound fragmented to speakers of modern standard Japanese. In Example (11) below, the predicates in line 3 and 7 are just such a case. (11) 1 A: he=,.. sore kara=, DM that after ‘After that, 2 INT: un. DM Analysis of Japanese Spoken by Elderly Taiwanese 253 3 A: Nihon no=,.. okashi no mise ni=, ano= teein Japan of sweets of shop DIR DM clerk toshite, ano=..., s% ano=, nani/.. kozoo ni as DM DM what flunky for → it-ta [no]. go-PAST FP (I) went to a Japanese sweets shop to be a clerk,... what is that called?.. (I) went (there) to be a flunky.’ 4 INT: [hu]=n. DM 5 A: sono toki wa hitotuski ga san-en desu yo. that time TOP one:month SUB 3-yen COP FP ‘At that time, (the salary for) one month was 3-yen.’ 6 A: Nihon no kane de san-en. i=, ichi, ni, san, Japan of money INS 3-yen one two three san-en desu yo. three-yen COP FP ‘3-yen in Japanese money. 1, 2, 3, 3-yen.’ 7 INT: hu=n. DM → 8 A: otoosan ga ne=, totemo yorokon-da yo. father SUB FP very be:glad-PAST FP ‘My father was very glad.' The next examples show omission of predicates. The word in parenthesis indicates a predicate, the subject omitted. (12) 1 INT: a[n]. DM → 2 A: [ma]n nijyussa% eya hachijyuu-ni-sai Ø (desu). full:age twenty no eighty-two-years:old COP ‘(I am) twenty, no, ..eighty two years old exactly.’ 3 INT: a=n. DM 254 Chapter Twelve (13) 1 INT: demo, Taiwan wa, sei kae-nai desyo? but TOP family:name change-NEG COP:TAG ‘But, in Taiwan, (you) do not change your (maiden) name, do you?’ 2 C: mukashi, Nihon-jidai wa, old:time Japan-period TOP ‘Long ago, in the Japanese period, 3 INT: ha\. DM 4 C: yappari, e=,.. otto no sei ni ka% ka%, watashi still DM husband of last:name to I → no jidai wa, otto no sei to onaji sei Ø (desu). of period TOP husband of family:name with same last:name COP in my youth, (we would) change (our name) to our husband’s name, to be the same as our husband’s family.’ 7. Summary This chapter examined the Japanese spoken by elderly Taiwanese who received education in Japanese during the colonial period of Japan. Taiwanese people learned Japanese at a young age; however, their Japanese showed deviations and unique features which are not seen in modern standard Japanese. Taiwan-Japanese exhibits characteristics of a dialect of Japanese which is spoken by elderly Taiwanese in Taiwan. The degree of deviation varied according to the subject’s educational background. Subject D, who received a higher education, had a few deviations, but spoke Japanese in a manner similar to a native speaker of modern Japanese. The deviations were mainly in word usage. This was the most frequently occurring deviation for all of the subjects. The other three subjects, who had only a middle school education exhibited several types of deviations and features. Two types of deviations and one feature were examined: word usage, particle usages, and predicate forms. One of the deviations that occurred most in the subjects’ Japanese was lexical borrowings or expansions. The subjects used a word which is understandable but is not used in modern Japanese, expanded a word class, and overgeneralised and created a new word. Another deviation discussed in this chapter is particle usage. The subjects showed some deviations in particle usage, but not as often as we would expect. Also there was no specific pattern. One of the unique Analysis of Japanese Spoken by Elderly Taiwanese 255 features of the Taiwan-Japanese was the predicate form. It is not a deviation but affects the cohesion of the conversation. Modern Japanese speakers use a lot of sentence ending morphemes to create cohesion. Examples are -n form, kara, kedo, ga, and the like. The use of various kinds of morphemes at the ends of sentences make the conversation sound connected. These morphemes were seen often in subject D and much of subject B’s Japanese, but did not appear often in subject A and C’s Japanese. The lack of sentence morphemes made the conversation sound fragmented and not well connected. Although the number of subjects is limited, the subjects in my data showed similarities and differences in variations in relation to their educational background. There are other deviations that this chapter did not discuss. For future research, more subjects are needed for comparison. It is also important to compare these Taiwanese with people who received Japanese education in other countries, such as Korea and the Micronesian islands. Notes 1. Examples of such are toosan ‘father’, obasan ‘old lady’, lingo ‘apple’ and many others. 2. Min-nan is one of the seven dialects in Chinese. It is mainly used in Fu-jian in China and in Taiwan. 3. In the following examples, INT signifies the interviewer. 4. There are plain and polite forms in Japanese. For an explanation of this, see the next section. 5. According to Chafe (1994), intonation unit (IU) is a speech unit that has an association with a coherent intonation contour. The criteria of delimitation of an IU are (1) pause, (2) final syllable lengthening or slow speech rate at the end of an IU, (3) fast speech rate at the beginning of the next IU, (4) pitch reset. References Copper, J.F. (1990). Taiwan -Nation-State or Province? San Francisco: Westview Press, Inc. Chafe, W. (1994). Discourse, Consciousness, and Time: The Flow and Displacement of Conscious Experience in Speaking and Writing. Chicago: University of Chicago Press. Kai, M. (1991). “Wa wa ikanishite shooryaku kanoo to naruka”. Nihongo Nihon Bunka, Bulletin of the Department for Foreign Students and the Department of Japanese at Osaka University of Foreign Studies 17: 113-128. 256 Chapter Twelve —. (1997). “Taiwanjin roonensoo no gengo seikatu to Nihongo ishiki”. Nihongo Kyooiku 93: 3-13. The Society for Teaching Japanese as a Foreign Language, Japan. Kamio, A. (1990). Joohoo no nawabari riron. Tokyo: Taishukan. Kortmann, B. & E. Schneider (eds.) (2004). A Handbook of Varieties of English. Berlin: Mouton DeGruyter. Onoe, K. (1987). “Shugo ni wa mo ga mo tsukaenai bun nitsuite”, Kokugogaku 150: 46-48. Thomason, S.G. & T.S. Kaufman (1988). Language Contact, Creolization and Genetic Linguistics. Berkeley: University of California Press. Wakabayashi, M. et al (eds.) (1990). Taiwan hyakka. Tokyo: Taishukan. Analysis of Japanese Spoken by Elderly Taiwanese 257 APPENDIX Transcription symbols: - morpheme boundary [ ] overlap . Final of intonation unit 5 ...(N) Long Pause (Seconds) , Continuing of intonation ... Medium Pause unit .. Short Pause ? Appeal (0) Latching -- Truncated intonation unit @ Laughter % Truncated word <@>___</@> while laughing \ Falling pitch (COUGH) cough / Rising pitch (TSK) utterance of a click of = Lengthening the tongue # Indecipherable Syllable Grammatical symbols: SUB subject case marker DM discourse marker like OBJ object case marker ‘hun’, ‘un’, ‘a’. TOP topic marker FP final particle like ‘yo’, ‘ne’ LOC locative case marker Q question word DIR directive case marker TAG tag question INS instrumental case marker CAUS causative COP copula like ‘desu’ NOM nominalizer NEG negative ADV adverbializer PAST past form BEN benefactive QUA quotation HON honorific CHAPTER THIRTEEN WHAT'S THE USE OF LINGUISTICS? MICHAEL WALSH 1. Saving languages Over the last 20 years linguists have become increasingly concerned about the demise of so many of the world's languages and this, in turn, has led to efforts to revitalize supposedly moribund languages. This latter activity is sometimes billed as “saving languages” although the truth of the matter is that this term covers a wide range of sometimes disparate activity. I will not delve into these questions here (cf. Walsh, 2005c; 2007) but focus instead on recent activity in New South Wales (henceforth NSW) (see also Troy & Walsh, 2005; Walsh, 2001; 2002b; 2003; 2005a; 2005b). Until rather recently the received wisdom was that most–if not all– Aboriginal languages of NSW were either extinct, or, at best, moribund. During 1999-2000, a NSW language survey was undertaken which led to the publication of a report: Strong Language, Strong Culture (Hosking et al,. 2000). The survey canvassed opinion mainly from Aboriginal individuals and groups across NSW, basically posing two questions: what is the current situation with regard to Aboriginal languages in NSW, and, what do Aboriginal people want to see happen in the future. It quickly became evident to the research team of which I was a part that almost without exception Aboriginal people were saddened by the demise of their languages and were keen to see their reinstatement if that were possible. It was also apparent that there had been substantial under-reporting of the vitality of these languages, curiously even by Aboriginal people themselves. It would seem that they felt that anything less than full fluency meant that their languages were essentially gone. Therefore we encountered Aboriginal people who would declare that their language was finished, only to burst into apparently fluent conversation with a countryman who happened to walk in the door! So part of the task of the What's the Use of Linguistics? 259 research team became an attempt to demonstrate that the linguistic knowledge within an Aboriginal community could be combined with the documentary records to achieve a degree of language revitalization. It was also clear that the time was right for a concerted effort. From 2001 onwards the Aboriginal Curriculum Unit of the NSW Board of Studies (BoS) worked towards the introduction of an Aboriginal Languages syllabus into NSW schools. From 2002 onwards the NSW Department of Aboriginal Affairs has been developing a NSW Aboriginal Languages Policy. By March 2003 the NSW Aboriginal Language Research and Resource Centre was established to provide a focus, forum and resource base for language revitalization efforts across NSW. In June 2003 the Aboriginal Languages K-10 Syllabus was first published by the NSW BoS (Walsh, 2003). It is worth bearing in mind that this syllabus was akin to other languages syllabi used in NSW schools for languages as diverse as Arabic, Armenian, French and Mandarin. The languages of NSW were being recognized as being worthy of study and having the potential to reach targets for language proficiency similar to community languages with vibrant speech communities. July 30, 2004 saw the official launch of the Aboriginal Languages K-10 Syllabus and the NSW Aboriginal Languages Policy (http://www.daa.nsw.gov.au/policies/langpolicy.html). The latter was the first officially sanctioned state-wide policy for Aboriginal languages in Australia. Within a few years of the take up of the syllabus “Up to 5000 students are learning Aboriginal languages in NSW schools, less than half of whom are indigenous. There are 46 Aboriginal- language programs being delivered statewide, with 41 of those in public schools” (Patty, 2007). Clearly NSW languages were on the move. 1.1 Why bother? A common enough question, particularly from Non-Aboriginal people was, and is, “Why bother?” Our survey indicated that when an Aboriginal community regains its language this is important step towards regaining its identity and by regaining identity there will be a substantial improvement in community health: mental, physical and social. Indeed one Aboriginal leader suggested to us that a minor investment in language revitalization would be amply repaid by the savings in government programmes for Aboriginal people. Included would be less need for the rebuilding of houses, assistance for substance abuse, incarceration and rehabilitation. These linkages (regaining language > regaining identity > regaining health) I found at first startling and then daunting. It seemed amazing that a relatively minor intervention from linguistics could have such a potential 260 Chapter Thirteen impact on people's lives, so it was imperative that the task be undertaken as proficiently as possible. 1.2 How to do it? As mentioned previously, linguists had begun to focus on the process of language revitalization and a useful model that had been developed by Leanne Hinton (2001) through 9 steps: 1. language assessment and planning 2. if language has no speakers, reconstruct it and develop language pedagogy 3. if language has only elderly speakers, document what they know 4. develop 2nd language learning programme for adults (e.g. Master-Apprentice) 5. support and encourage use of endangered language at home and in public 6. develop intensive 2nd language programmes for children, especially in schools 7. use language as primary means of communication at home 8. expand use of the indigenous language into broader local domains 9. where possible, expand language domains into the broader population (p.6). In the NSW situation a significant advance had been made in Step 1 (language assessment and planning) through the Survey and this had been followed up by the NSW Aboriginal Language Research and Resource Centre which monitored progress across the state and sponsored forums for Aboriginal people to assess existing programmes and plan for the future. Particularly significant in the Hinton approach is the ordering: instruction for children is typically to come after instruction for adults. In this way children who are improving their knowledge of their ancestral language(s) at school come home to an environment where their parents and other carers will have at least some knowledge of these languages as well. The Hinton approach (see also Hinton, 2002) is wide ranging and requires expertise in diverse areas: What's the Use of Linguistics? 261 Table 1: Hinton's 9 steps in relation to areas of linguistics Hinton's 9 steps Relevant areas of Relevance of towards language Linguistics and kindred linguists revitalization disciplines 1. language language planning; significant assessment and sociology of language; planning social psychology of language etc 2. reconstruct linguistic engineering major language and develop language language pedagogy; SLA minor pedagogy etc 3. document salvage linguistics; major remaining language linguistic fieldwork etc 4. develop 2nd language documentation3; major language learning language pedagogy; SLA minor programme for adults etc 5. use of endangered language planning; some minor languages at home and linguistic engineering in public 6. develop intensive language documentation; less major 2nd language learning language pedagogy; SLA minor programmes for etc children 7. use language as language planning; some minor primary means of linguistic engineering communication at home 8. expand use of the language planning; some minor indigenous language linguistic engineering into broader local domains 9. where possible, language planning; some minor expand language linguistic engineering domains into the broader population 262 Chapter Thirteen 1.3 What linguists have been doing to assist with saving languages My own role in “saving” languages in NSW has necessarily been quite meagre, if only because I lack the range of expertise to carry out all of the tasks outlined above. It is therefore valuable to consider more broadly what linguists have been doing to assist with saving languages. Linguists and others have been involved in forums for forming positions in regards to policy and planning for Australian Indigenous languages, for example, Federation of Aboriginal and Torres Strait Islander Languages (FATSIL) (www.fatsil.org); the Victorian Aboriginal Corporation for Languages (VACL) (www.vacl.org.au), and, the NSW Languages forum (most recently held 29-31 October 2007). There have also been attempts towards improving skill sets for Indigenous people as in, for example, Sydney University's Master of Indigenous Language Education (http://www.koori.usyd.edu.au/studying/blockmode.shtml), Monash University's Studies in Language Endangerment (http://www.arts.monash.edu.au/linguistics/pglangen/index.php), and the University of Adelaide's Kaurna Language and Language Ecology (http://www.arts.adelaide.edu.au/humanities/linguistics/). Even more broadly substantial progress has been made in developing resources: language learners' guides; archiving; providing expert advice in, for example, Resource Network for Linguistic Diversity (RNLD) (http://www.linguistics.unimelb.edu.au/thieberger/RNLD.html), Technology-Enhanced Language Revitalization (TELR (formerly ILAT)) (http://projects.ltc.arizona.edu/gates/TELR.html); Language Archives Newsletter (http://www.mpi.nl/LAN); Teaching Indigenous Languages (http://www2.nau.edu/~jar/TIL.html) and how-to-do-it guides like Grenoble & Whaley, 2006; Hinton, 2002; Hinton & Hale, 2001. 1.4 What the academy should do to assist linguists who have been assisting with the saving of languages In considering the appropriate involvement of the academy it is instructive to consider what I think of as the Pensalfini plaint: In the academic world, however, the kind of work described above [language revitalization] is considered less important than, and an accidental by-product of, my collecting, analysing and theorizing over linguistic data. When it comes to calculating funding ratios, assessing impact and quality and research, the theoretical article and the reference grammar are valued much more highly than materials, which can be used What's the Use of Linguistics? 263 for the direct benefit of speech communities. For example, in calculating funding to my department, five points are assigned to an original authored scholarly book, one point to a refereed journal article, and absolutely nothing, zero points, to the kind of work described above. Even where academic linguists would like to be of use to the communities from which they draw the resources for their career, the pressures and priorities of the academic world mitigate against this. (Pensalfini, 2004, p.154) As a first step it would be useful to at least minimize the disadvantages for those in the academy involved in saving languages. At present it could well be regarded as a negative career move to spend more than a token amount of time and effort on such activities. 2. Addressing disadvantage experienced by Aboriginal people in the legal system (Walsh, in press) Aboriginal people have experienced disadvantage in the legal system because of linguistic considerations and linguists have made attempts to address such disadvantage. There is a range of variables in connection with Aboriginal witnesses' vulnerability including: gender e.g. women disadvantaged typically and, in particular, Hindmarsh Island Bridge case (further details for this and other factors can be found in Walsh (in press)); age–where an Aboriginal witness may be old enough to have knowledge of some matter but not be considered by the Aboriginal community to be old enough to articulate that knowledge. This can result in an Aboriginal witness appearing to lack traditional knowledge whereas in fact that witness is following traditional practice by not speaking of that knowledge. This sociolinguistic practice, built on a fundamentally different regime for the management of knowledge, is widespread in Aboriginal Australia and well known to specialists but needs to be made apparent in a legal setting. Lawyers dealing with “traditional” Aboriginal witnesses tend to have a reliance on oral tradition so that, perversely, such witnesses can be criticized for reliance on written material. In an Aboriginal land claim that I worked on, some Aboriginal witnesses were asked whether they had read a substantial document that set out ethnographic and historical background to support their case. If they answered Yes, the ultimate response from those opposing their claim was to ask the judge to set aside their testimony because it was “tainted” by this written material instead of being passed on as part of a purely oral tradition. On the other hand, a witness who said No was berated by the cross-examining barrister for their supposed lack of interest in their own heritage! (Walsh, 1995). Apart from the often alien vocabulary of 264 Chapter Thirteen courtroom discourse Aboriginal witnesses have sometimes been put at a disadvantage because there are words which are English in form but with distinctively different meanings. For instance, in Aboriginal English “kill” simply means “hit, strike” whereas Standard Australian English “kill” is often captured by some kind of expression like “kill him dead/killim dead”. Aboriginal English “father” refers not just to one's biological father but also to one's father's brothers and to other individuals whom one might address as “father” notwithstanding that they have no blood tie whatsoever. Again, such differences are well known to specialists but not to those unfamiliar with Aboriginal Australia. For those with minimal training in sociolinguistics the next example should be a reminder of how members of the public (including very experienced legal practitioners) may be baffled by apparently straightforward notions: it concerns multiple linguistic identities and the “problem” of group labelling. In the De Rose Hill native title claim (Monaghan, 2003, p.205ff) difficulties were encountered when an earlier ethnographer’s ascription of certain territory to one group, Antikirinya, appeared to run counter to the claimant group’s self-identification as Yankuntjatjara. A linguist, Cliff Goddard, was engaged on behalf of the claimants to address this and other issues drawing on his long-term knowledge of the area. Monaghan (2003) explains: In both written and oral testimony, Goddard explained to the court that Yankuntjatjara and Antikirinya are terms denoting the same speech variety: but while Yankuntjatjara is a Western Desert speech label, Antikirinya is an exonym of Arandic origin. This enabled, Goddard argued, the same people to use the former term to distinguish themselves from Pitjantjatjara speakers to the west or to use the latter term to distinguish themselves from non-Western Desert groups to the east (p.206). In the end the judge sided with the earlier ethnographer’s position that there were two separate territories, Yankuntjatjara to the west and Antikirinya to the east. This was not merely because of the apparent mismatch between the expert witness’ view and that of the Aboriginal witnesses but particularly because the judge “simply could not accept that the same person could identify as Yankuntjatjara in one context and Antikirinya in another” (Monaghan ,2003, p.207). In land claim proceedings difficulties can be experienced when those creating the transcript have little or no background in Aboriginal languages. For the Lakefield National Park land claim (proceeding under the Queensland Aboriginal Land Act), the transcript originally recorded: “I What's the Use of Linguistics? 265 am branded with murder”. What was actually said by the Aboriginal witness was: “I blanta Rirrmerr”, i.e. I belong to Rirrmerr = I am part of the Rirrmerr Aboriginal organization (Rigbsy, 1995 and personal communication 2007–transcript example quoted in Sutton, 1994, p.120). Often the record of the proceedings which contains material not only of potential value to the claimants' case but also having heritage value for the descendants of the claimants has such infelicities scattered through it. In the context of what the witness had been saying the original really does not makes sense at all but through what I refer to as the “tyranny of the transcript” this too often becomes the only record. In this case the judge allowed Rigsby, an anthropologist and linguist of great experience in the region, to listen to the audio recordings and produce a more faithful transcript. More often it is only the official court transcribing service who are privy to these recordings and mistakes which could otherwise have been picked up remain on the record for good. To conclude this brief overview, I would like to provide three examples from my own experience of how knowledge of linguistics can assist Aboriginal people in legal settings. These examples are mainly drawn from land claim proceedings. The first example again arises from non-specialists attempting to capture for the transcript “exotic” words and expressions from Aboriginal languages. In land claim proceedings Aboriginal witnesses inevitably use words and expressions from their own, or neighbouring Aboriginal languages. Particularly frequent are Aboriginal place names and knowledge of these forms an important part of the case on behalf of the Aboriginal claimants. Consider a hypothetical Aboriginal place name: Malandjakitj. For the sake of the transcript someone (usually an anthropologist appearing on behalf of the claimants) is invited to give a spelling. The witness' testimony is temporarily interrupted while the anthropologist spells it out: M-A-L-A-N-D-J-A-K-I-T-J. This hypothetical place name is consistent with phonological patterns readily found in Aboriginal languages: it is neither particularly short, nor particularly long and certainly not particularly difficult. However its effect on the flow of the Aboriginal witness' testimony is profound as there will sometimes be lists of place names which require this laborious spelling out each time a new name is introduced. The effect is that the narrative flow of the witness' evidence is chopped up and may even render the witness reticent about introducing an “exotic” word for fear of the judge or some counsel requesting: “Could we just get Dr Walsh (or whoever) to provide the court with a spelling?” 266 Chapter Thirteen This disruption goes to a wider problem for Aboriginal people in legal settings. The norm for eliciting evidence is through a question and answer technique where the barrister often has a lengthy list of pre-arranged questions to put to the witness. Unfortunately this approach is alien to Aboriginal preferences and has been shown to decrease substantially the quality of their evidence. The way to overcome the problem is to put a wide-ranging question to the witness (e.g. “Tell the judge in your own words what happened”) which results in a continuous narrative (Cooke, 1996). So the challenge for land claim proceedings is to devise a practice whereby the requirement for accuracy in the transcript is satisfied without a need for a laborious and disruptive spelling out of Aboriginal words and expressions and Aboriginal people are allowed to speak in a more culturally suitable way. The solution is simple and obvious but – to my knowledge – rarely practiced. By arrangement with the transcribing service and the court (in particular, the judge) I kept a running list of any Aboriginal words and expressions that cropped up in a day's evidence. That evening I went through the list with the transcribing service and provided any spellings needed. Of course this required a little more work on my part and for the transcribing service but undoubtedly saved a significant amount of the court's time and considerably improved the quality of the Aboriginal evidence as well as their comfort in presenting it (Walsh, 1999). Even lawyers long experienced in dealing with Aboriginal witnesses can benefit from linguistically informed accounts of Aboriginal ways of talking. There is a tendency to think that the Aboriginal witness' understanding and production of English is little different from any Anglo- Australian witness. However it is well known that there are significant differences between Aboriginal interaction and that of Anglo-Australians. Effectively one encounters something rather similar to Standard Australian English morphosyntax but with Aboriginal pragmatics. One prominent example is what has been referred to as “gratuitous concurrence” where an Aboriginal person will respond with an apparent affirmative like “Yes” to some question but actually might not agree with the questioner nor even understand the question. Another difference is that Aboriginal people might not feel a need to answer a question immediately but in a legal setting a timely response to questions is expected and a “long” silence can be damaging to the person's case. Such pragmatic differences are well known to linguists (e.g. Walsh, 1994; 1997a) but may need to be brought to attention of legal practitioners. What's the Use of Linguistics? 267 My third example involves the explication of key concepts. In the context of land claim and native title proceedings lawyers will sometimes quiz Aboriginal witnesses about their degree of knowledge of ancestral languages or–even more of a problem–about other people's proficiency. The latter question goes to another aspect of fundamental differences in the management of knowledge: Aboriginal people may be very reluctant to comment on other people's knowledge even if they are very much aware of it. For them it is more appropriate that each person–except for small children or the mentally incapacitated–should speak for themselves. For the former question we need to examine a key distinction: language owners vs. language users (Walsh, 2002a). For most Aboriginal people a language has owners: it is one part of an Aboriginal person’s birthright along with membership in a land-owning group, often inherited from one’s father but also from a mother, particularly as more non-Aboriginal men had children with Aboriginal women. Crucially, this ownership does not require a person to be a speaker of the language although in earlier times the person would have been fluent not only in their own language (the language they owned) but also in a number of other Aboriginal languages through intermarriage, employment or other encounters with more distant Aboriginal groups. Even for an Aboriginal person with a sizeable linguistic repertoire it is unusual for there to be any doubt about which language they own whether from their own perspective or from Aboriginal people familiar with their area. This will be reflected in statements like, “His/her language is X” notwithstanding that the person in question does not speak X or any other Aboriginal language. Such a person might say: “My language is X, I get that from my mother, but I never got to learn it and I hope to speak it one day.” This is a lifelong property of the person, gained at birth and not at all a matter of choice. Nevertheless Aboriginal witnesses who speak no more than a few words of any Aboriginal language continue to be quizzed about which language(s) they speak and how much when the more appropriate questions to ask in order to establish land ownership are of the kind: “What is your language? How did you get that language?” and “What is her language? How did she get that language?” 3. Language and the Land For Aboriginal people traditional country is an integral part of their identity, so it is not surprising that land and language are interdependent (Walsh, 1997b). In this concluding section I want to draw attention to two ways in which linguistics can be of some use in this connection. 268 Chapter Thirteen 3.1 A little language to keep country accessible Aboriginal people regard their land as being inhabited by spirits of so- called "Dreamtime" entities (who shaped the land originally) and by their ancestors. To that end it is not uncommon for Aboriginal people to address the spirits directly – in their own language: the language of the land. Particularly for certain places of high significance it is customary to inform the spirits when newcomers are approaching the place for the first time. In some instances this can create delays while a person with the right status and knowledge of the language can address the spirits appropriately. Without such a person some places remain closed off: it is contrary to customary law to approach the place without the use of the land's ancestral language. But what is to happen when many Aboriginal languages have fallen into disuse? One strategy is to create a little language so as to keep the language accessible. This is just one more example of linguistic engineering and, provided the relevant Aboriginal community has ultimate control of the process, it is a way that places of high significance can remain open to their communities. 3.2 Dual naming Dual naming is a process whereby the original, Aboriginal place name is reinstated alongside the introduced name. An example from the Sydney area is Dawes Point (at the southern foot of the Sydney Harbour Bridge) which has been dual named to include the original name from the local Dharuk language: Tar-ra. This gives recognition to the original inhabitants of the area and raises public awareness of their language. I have been part of a process that has involved reinstating some 20 Aboriginal place names around Sydney Harbour (Troy & Walsh, to appear). This requires an examination of spellings of the original place names from early written sources and then educated guesswork about what the original names might have sounded like. One then designs a proposed spelling in an attempt to capture the original as closely as possible. These proposed spellings are then presented to the Aboriginal people concerned for consideration and amendment, if necessary. 4. Conclusion These are clearly not the only instances of linguistics as being useful outside of a university classroom or a person’s study. In looking at disadvantage by Aboriginal people in the legal system the relevant areas of What's the Use of Linguistics? 269 linguistics include: phonetics and phonology (Aboriginal English and Australian Indigenous languages); graphology; interactional sociolinguistics; lexical semantics; ethnography of speaking; dialectology; and social psychology of language, while in the section on language and the land some of the relevant areas of linguistics are: phonetics and phonology; graphology; orthography development; as well as interactional sociolinguistics; ethnography of speaking; and social psychology of language. It should be noted that these applications of linguistics in the wider world are quite wide-ranging. They do not rely on some kind of artificial split between applied linguistics and so-called general or theoretical linguistics: instead it is a matter of linguistics applied. Notes 1. An earlier version of this paper was presented at the Free Linguistics Conference, 7 October 2007, University of Sydney. 2. I deliberately avoid the usage ‘Australian Indigenous Languages’, which is inclusive of the languages of the Torres Strait Islands, because I have virtually no expertise in those languages and my remarks in this paper are entirely directed towards Australian Aboriginal languages. 3. “language documentation”: This is a very broad term which includes the whole range of “structural linguistics” as well as comparative linguistics (historical and typological). References Cooke, M. (1996). A different story: narrative versus ‘question and answer’ in Aboriginal evidence. Forensic Linguistics 3(2): 273-288. Grenoble, L. & Whaley, L. (2006). Saving Languages: An Introduction to Language Revitalization. Cambridge: Cambridge University Press. Hinton, L. (2001). Language revitalization: an overview. In L. Hinton & K. Hale (Eds.) The Green Book of Language Revitalization in Practice (3-18). San Diego: Academic Press. Hinton L. (2002). How to Keep Your Language Alive: a Commonsense Approach to One-on-One Language. Berkeley, CA: Heyday. Hinton L. & Hale, K. (2001) The Green Book of Language Revitalization in Practice. San Diego: Academic Press. Hosking, D., Lonsdale, T.J., Troy, J.F. & Walsh, M.J. (2000). Strong Language Strong Culture. New South Wales Strategic Language Study. Final Report and Strategy Action Plan. Canberra: AIATSIS. 270 Chapter Thirteen Monaghan, P. (2003). Laying down the country: Norman B. Tindale and the linguistic construction of the north-west of South Australia. Unpublished PhD thesis, University of Adelaide. Patty, A. (2007). Aboriginal languages help more stay on. Sydney Morning Herald Retrieved 15th September 2007 from www.smh.com.au. Pensalfini, R. (2004). Eulogizing a language: the Ngarnka experience. International Journal of the Sociology of Language 168: 141-156. Rigsby, B. (1995). Lakefield National Park land claim. Paper delivered to the International Association of Forensic Linguists, University of New England. —. (2007). Personal communication. Sydney. Sutton, P. (1994). “About the Gist of What was Said”: Communication in the Context of What Was Said. In F. McKeown (Ed.) Native Title: an Opportunity for Understanding (115-123). Perth, WA: National Native Title Tribunal. Troy, J.F. & Walsh, M.J. (2005). Languages Off Country? Revitalizing the 'Right' Indigenous Languages in the South East of Australia. In N. Crawhall & N. Ostler (Eds.) Creating Outsiders: Endangered Languages, Migration and Marginalization. (Proceedings of Ninth Conference of the Foundation for Endangered Languages, Stellenbosch, South Africa, 18-20 November 2005 (71-81). Bath: Foundation for Endangered Languages. Troy, J.F. & Walsh, M.J. (to appear). Reinstating Aboriginal placenames around Port Jackson and Botany Bay. In H. Koch & L. Hercus (Eds.) Aboriginal Placenames Old and New [working title] (Aboriginal History Monographs). Canberra: ANU e-press. Walsh, M.J. (1994). Interactional styles and the courtroom: an example from Northern Australia. In J. Gibbons (Ed.) Language and the Law (217-233). London: Longman. —. (1995). Tainted evidence’: literacy and traditional knowledge in an Aboriginal land claim. In D. Eades (Ed.) Language in Evidence. Issues Confronting Aboriginal and Multicultural Australia (97-124). Sydney: University of New South Wales Press. —. (1997a). Cross Cultural Communication Problems in Aboriginal Australia. Darwin: North Australia Research Unit. Discussion Paper No.7/1997. —. (1997b). The Land Still Speaks? In D. Rose & A. Clarke (Eds.) Tracking Knowledge in North Australian Landscapes. Studies in Indigenous and Settler Ecological Knowledge Systems (105-119). Darwin: North Australia Research Unit. What's the Use of Linguistics? 271 —. (1999). Interpreting for the transcript: problems in recording land claim proceedings in northern Australia. Forensic Linguistics 6(1): 161-195. —. (2001). A case of language revitalisation in ‘settled’ Australia. Current Issues in Language Planning 2(2&3): 251-258. —. (2002a). Language ownership: a key issue for Native Title. In J. Henderson & D. Nash (Eds.) Linguistics Issues in Native Title Claims (230-244). Canberra: Native Title Research Series, Aboriginal Studies Press. —. (2002b). Teaching NSW’s Indigenous Languages Lessons from Elsewhere. Paper prepared for the Aboriginal Curriculum Unit of the NSW Board of Studies, 27pp. (http://www.boardofstudies.nsw.edu.au/aboriginal_research/pdf_doc/te ach_indig_lang_nsw_walsh.doc). —. (2003). Aboriginal Languages. Mandatory and Elective Courses. K-10 Syllabus. Sydney: Board of Studies, New South Wales. (major author). —. (2005a). Indigenous Languages of Southeast Australia: Revitalization and the Role of Education. Australian Review of Applied Linguistics. 28(2): 1-14. —. (2005b). Learning while revitalizing: Aboriginal languages in New South Wales, Australia. In S. May, M. Franken, & R. Barnard (Eds.) LED2003: Refereed Conference Proceedings of the 1st International Conference on Language, Education and Diversity.Hamilton: Wilf Malcolm Institute of Educational Research, University of Waikato. —. (2005c). Will Indigenous languages survive? Annual Review of Anthropology 34: 293-315. —. (2007) Is saving languages a good investment? Plenary address to Australian Linguistic Society, University of Adelaide. 26 September 2007. —. (in press). Which way? Difficult options for vulnerable witnesses in Australian Aboriginal land claim and native title cases. Journal of English Linguistics, Special issue. ABOUT THE CONTRIBUTORS Monika Bednarek is a Chancellor’s Post-doctoral Research Fellow at the University of Technology, Sydney. She is the author of Evaluation in Media Discourse (2006, Continuum) and Emotion Talk across Corpora (2008, Palgrave Macmillan) and is currently co-editing a volume on multi- modality (forthcoming, Continuum). Anne Burns is a Professor in the Department of Linguistics and the former Dean of the Division of Linguistics and Psychology at Macquarie University in Sydney, where she supervises doctoral students in the Applied Linguistics Program. She has conducted research and published extensively in the areas of migrant education, action and classroom research, spoken discourse, second language literacy, teacher education, and curriculum development. Her most recent book (co-edited with Jill Burton) is Language Teacher Research in Australia and New Zealand (TESOL, 2008). She is the co-editor of Prospect: A Journal of Australian TESOL, book reviews editor for Reading in a Foreign Language, and sits on the Advisory Boards of eight journals in the field of Applied Linguistics. Helen Caple is a doctoral student with the Department of Linguistics at the University of Sydney. She lectures in Media Discourse and Journalism Studies and is a former press photographer. Paul Dwyer returned to University study in the mid-1990s after almost a decade of professional theatre work, mostly as an actor, animateur, director or dramaturg in the field of youth/community theatre. After returning to academia, Paul gained a Diplôme d'Etudes Approfondies in Theatre Studies from the University of Paris (8) and a PhD in Performance Studies from the University of Sydney. Paul’s doctoral research focused on various modifications of, and continuing challenges to, the theory and practice of Augusto Boal's 'Theatre of the oppressed'. He is now just embarking on research into the performative aspects of 'restorative justice' and various kinds of reconciliation ceremony. About the Contributors 273 Masumi Kai is an Associate Professor in the Department of Humanities at the University of Guam, USA. She was an Associate Professor at Okayama University, Japan prior to that. She obtained a Ph.D degree in Language and Culture from Osaka University of Foreign Studies, Japan. She was a visiting scholar at University of California, Santa Barbara and Harvard University. Her research interests include discourse analysis, syntactic analysis of particles, language acquisition, and Japanese education. She has published papers on ellipsis in Japanese, semantic analysis of wa-ga structure, socio-linguistic analysis of Japanese in Taiwan. Hyeran Lee obtained a PhD degree in linguistics from the University of Florida in 1997. Since then she has been working at Kyung Hee Univesity in Korea. She has published papers on reflexives, ellipsis, scrambling, left periphery, and Case in the area of generative syntax. She now belongs to the Department of General Education at Kyung Hee Univesity as an assistant professor. She is also a coordinator of the Institute of Continuing Education at the same university. Naomi Knight is a PhD candidate in the Linguistics Department at the University of Sydney. She is finishing her PhD thesis on conversational humour in the systemic functional linguistic framework, with a focus on the social bonding and discursive identity functions of this phenomenon. Naomi specializes in conversation and discourse analysis, and is also interested in the areas of phonology, pragmatics and sociolinguistics. Additionally,she has worked on research projects in ape language studies involving data of language-competent apes Kanzi and Panbanisha, and completed her honours thesis, entitled 'Semantic Analysis of an Instance of Bonobo-Human Discourse', on the subject. Naomi is the co-creator and co-convener of the International Free Linguistics Conference with Dr. Ahmar Mahboob, and has worked as a casual tutor and guest lecturer at the University of Sydney. Alan Reed Libert was born in 1959 in Los Angeles, CA. He obtained his PhD in linguistics at McGill University, Montreal in 1993; his dissertation title was "On the Distinction between Syntactic and Semantic Case". Alan has been working at the University of Newcastle since 1994. His research interests include adpositions, cases, Turkic languages, and artificial languages. His most recent book is "Ambipositions" (Lincom Europa, Munich). His earlier books include "A Priori Artifiicial Languages" and "Artificial Descendants of Latin". 274 About the Contributors Caroline Lipovsky is a Lecturer in the School of Languages & Cultures at the University of Sydney, Australia. Her research interests include intercultural communication, differences in communicative styles, humour, small talk, negotiation of co-membership and solidarity. Current projects investigate linguistic aspects of self-presentation in job interviews, students’ attitudes towards their native and non-native teachers, and ESL (English as a second language) learners’ interlanguage. Ahmar Mahboob teaches linguistics at the University of Sydney, Australia. He earned his PhD at Indiana University, Bloomington, in 2003, with a dissertation on Status of nonnative English speakers as ESL teachers in the United States. Ahmar has worked in the areas of language policy development, pidgin and creole languages, NNEST studies, English language acquisition, English language teaching and teacher education, World Englishes, pragmatics, and issues surrounding minority languages in South Asia. His recent work has focused on Pakistani English and on languages of Pakistan. Ahmar is the Past President of Indiana TESOL and the Past Chair of the NNEST Caucus in TESOL International. James Martin is Professor of Linguistics at the University of Sydney. His research interests include systemic theory, functional grammar, discourse semantics, register, genre, multimodality and critical discourse analysis, focussing on English and Tagalog - with special reference to the transdisciplinary fields of educational linguistics and social semiotics. Recent publications include Re/Reading the Past (Edited with Ruth Wodak) Benjamins, 2003; Negotiating Heteroglossia (a special issue of Text Edited with Mary Macken-Horarik) Mouton de Gruyter, 2003; Language Typology: a functional perspective (Edited with A Caffarel & C Matthiessen) Benjamins 2004; Interpreting Tragedy: the language of September 11th, , 2001 (a special double issue of Discourse & Society Edited with John Edwards) Sage 2004, The Language of Evaluation (with Peter White) Palgrave 2005; Language, Knowledge and Pedagogy (Edited with Fran Christie) Continuum 2007. He has recently completed, with David Rose, a book on genre (Genre relations: mapping culture, Equinox 2008) and a second edition of Working with Discourse (Continuum 2007). Professor Martin was elected a fellow the Australian Academy of the Humanities in 1998, and awarded a Centenary Medal for his services to Linguistics and Philology in 2003. About the Contributors 275 Alastair Pennycook is Professor of Language in Education at the University of Technology, Sydney. He has published widely on cultural and political implications of the global spread of English, colonialism and language policy, plagiarism and intertextuality, and critical applied linguistics. He is the author of The Cultural Politics of English as an International Language (Longman, 1994), English and the Discourses of Colonialism (Routledge, 1998), Critical Applied Linguistics: A Critical Introduction (Lawrence Erlbaum, 2001), and co-editor, with Sinfree Makoni, of Disinventing and Reconstituting Languages (Multilingual Matters, 2007) and with H. Samy Alim and Awad Ibrahim of Global Linguistic Flows: Hip Hop Cultures, Youth Identities, and the Politics of Language (Routledge, in press). His most recent book on language, globalization and hip hop, Global Englishes and Transcultural Flows (Routledge) was published in 2007. Montri Tangpijaikul holds an M.A. in Applied Linguistics from Bond University. He works in the Department of Foreign Languages, Kasetsart University (Thailand) and now studies in the Department of Linguistics, Macquarie University. His research interests are the EFL learners' online writing and their use of modality and intensifying elements in English. Michael Walsh is an Honorary Associate in the Department of Linguistics at the University of Sydney in Australia. He has been engaged in linguistic fieldwork with Australian Aboriginal groups for 35 years. Outside of strictly linguistic matters he has carried out research or advised on land claims, assessment of Aboriginal witnesses in legal settings and Native Title matters. Michele Zappavigna is an ARC Postdoctoral Fellow in the Department of Linguistics where she is working with Jim Martin and Paul Dwyer on the discourse of Youth Justice Conferencing. Her research interests include functional semiotics, text visualisation and corpus linguistics. She completed her PhD in the School of Information Technologies and is interested in how technology can support the work of applied linguists.