Japanese Applied Linguistics: Discourse and Social Perspectives 9781474211871, 9780826489616

This book showcases recent developments in the field of Japanese applied linguistics. It covers a wide range of current

217 64 3MB

English Pages [364] Year 2008

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Japanese Applied Linguistics: Discourse and Social Perspectives
 9781474211871, 9780826489616

Citation preview

SFK FM

SFK ICBK027

May 7, 2008

23:7

Char Count=

Contributors

Haruko Minegishi Cook is Professor at the University of Hawai‘i-Manoa, where she teaches sociolinguistics and Japanese. Her main research interests include language socialization, discourse analysis, and pragmatics. She has published a number of articles on the Japanese sentence-final particles and honorifics. Her book on style shifts in dinnertime conversation between JFL learners and their host families will be published in 2008. Kimberly Jones is Associate Professor at the University of Arizona, where she teaches Japanese language and linguistics. Her research interests include sociolinguistics, discourse and grammar, and language acquisition and attrition. Her research on sociolinguistics and on the connections between discourse studies and pedagogy has been published in the journals Language in Society, Japanese Language and Literature, and Gengogaku, among others. Yasuko Kanno is Associate Professor of TESOL in the College of Education at Temple University, Philadelphia. She is the author of Negotiating Bilingual and Bicultural Identities: Japanese Returnees betwixt Two Worlds (2003) and Language and Education in Japan: Unequal Access to Bilingualism (2008). Ryuko Kubota is a second language teacher and teacher educator, currently teaching as a professor at the University of North Carolina at Chapel Hill. Her research interests include culture in second language teaching, contrastive rhetoric, critical pedagogies, and critical multicultural education. Her publications appear in such journals as Foreign Language Annals, Journal of Second Language Writing, TESOL Quarterly, and World Englishes. Yoshiko Matsumoto is Associate Professor at Stanford University. Her research covers a wide range of topics in linguistic pragmatics including pragmatics of complex noun phrases, zero-anaphora, honorifics, politeness theories, speech acts, and the relations among language, gender, and age. She is the author of Noun-Modifying Constructions in Japanese and has published in a number of journals and edited volumes.

SFK FM

SFK ICBK027

viii

May 7, 2008

23:7

Char Count=

Contributors

Junko Mori is Associate Professor of Japanese language and linguistics at the University of Wisconsin-Madison and a recipient of the ACTFL/MLJ Paul Pimsleur Award for Research in Foreign Language Education. Her research interests center on the application of the methodological framework of conversation analysis to the study of talk-in-interaction involving first and second language speakers of Japanese. Kanae Nakamura is a doctoral student in the Department of East Asian Languages and Literature at the University of Wisconsin-Madison. Her dissertation examines the interplay between the syntactic structure and the ways of negotiating agreement/disagreement in opinion-proffering turns in Japanese conversation. She is currently teaching as a lecturer in the Department of Japanese at Tamkang University in Taiwan. Amy Snyder Ohta is Associate Professor at the University of Washington. She is the author of Second Language Acquisition Processes in the Classroom: Learning Japanese, and has published articles in journals including Applied Linguistics, System, Japanese Language and Literature, and Journal of Pragmatics, as well as participating in edited volumes related to sociocultural theory, classroom interaction, interlanguage pragmatics, and Japanese pedagogy. Misao Okada is Associate Professor at Hokusei Gakuen University, Sapporo, Japan. Her research focuses on language and gender, analysis of multimodal organization of action, and learning and teaching as interactional processes embedded within social practices. Her publications include an article in Language in Society and book chapters on conflict among Japanese speakers and usages of the Japanese sentence-final particle ne. Shigeko Okamoto is Professor of Linguistics at California State University, Fresno. She received a Ph.D. in Linguistics from the University of California, Berkeley (1985). Her areas of specialty include sociolinguistics, discourse analysis, and pragmatics. Her recent publications focus on Japanese linguistic diversity as it relates to gender, standard and regional Japanese, politeness/honorifics, and the role of ideology in language practice. Tsuyoshi Ono is Associate Professor at the University of Alberta, where he directs the Spoken Discourse Research Studio. He specializes in discourse and grammar, examining various aspects of Japanese morphosyntax in conversation data. He is currently engaged in two projects with researchers both in Japan and the US: a corpus of spoken Japanese and a documentation of an endangered language, Ikema, spoken on several remote islands in Okinawa, Japan. Patricia J. Wetzel is Professor of Japanese at Portland State University. Her primary research is in Japanese linguistic anthropology and linguistic

SFK FM

SFK ICBK027

May 7, 2008

23:7

Char Count=

Contributors

ix

pragmatics, with a focus on keigo (polite language). Her publications include Keigo in Modern Japan (University of Hawai‘i Press, 2004) and various articles that have appeared in Language in Society, Language Variation and Change, Multilingua, and elsewhere. She is past president of the Association of Teachers of Japanese. Dina Rudolph Yoshimi is Associate Professor of Japanese in the Department of East Asian Languages and Literatures at the University of Hawai‘i-Manoa. Her research interests include Japanese L2 acquisition and pedagogy with a focus on the pragmatics of everyday conversation. She has coordinated several summer institutes on L2 pragmatics and has authored several articles on pragmatics and the JFL/JSL learner. Lindsay Amthor Yotsukura is Associate Professor of Japanese at the University of Maryland, College Park. Her research focuses on Japanese pragmatics and discourse analysis from a cross-cultural perspective. Recent publications include her book, Negotiating Moves: Problem Presentation and Resolution in Japanese Business Discourse, and a chapter in Dialogue with Bakhtin on Second and Foreign Language Learning: New Perspectives.

SFK FM

SFK ICBK027

May 7, 2008

23:7

Char Count=

Acknowledgements

We enjoyed the process of putting this volume together thanks to the participation and support of many people. First, we appreciate the contributions of our chapter authors. They each contributed original scholarship to this book and we enjoyed the collaborative process of working with each author. Throughout the process we have been grateful for cooperation, shared and divergent perspectives, creative engagement, constructive criticism, intellectual rigor, and friendly interaction. We appreciate the time taken in creating the chapters and discussing them via email, telephone, and face-to-face meetings. We learned a tremendous amount from the research within these pages and through the editing process; for this we thank Yoshi, Kim, Kanae, Haruko, Pat, Shigeko, Misao, Yoshiko, Lindsay, Yasuko, Dina, and Ryuko. We have long thought about a volume like this. Continuum’s Jennifer Lovel and Colleen Coalter helped us to realize our visions by helping this project develop from its earliest stages. Upon Jenny’s departure, the baton was passed on to Gurdeep Mattu and we are grateful for his guidance and expertise. We would also like to thank Mutsuko Endo Hudson, who introduced us to Continuum. On a personal note, we received invaluable assistance from our husbands, George and Kaoru. Both provided helpful feedback and support during this project. We would also like to thank each other. Prior to this volume, we had never collaborated but merely enjoyed some common academic interests and a friendly acquaintance when we occasionally met at conferences. Working together, we discovered that we have compatible working styles and complementary skills. We have appreciated each other’s collaborative engagement and prompt responsiveness. It has been a pleasure working on this volume together. Finally, we thank you as a reader of this research. Whether a newcomer or an old hand, familiar with Japanese or not, we appreciate your readership

SFK FM

SFK ICBK027

May 7, 2008

23:7

Char Count=

Acknowledgements

xi

and look forward to the conversations and scholarship that this volume may inspire. We look to you to fill the gaps that we have left. For each gap you notice, we would be grateful if you would find inspiration for research or your own editorship of a volume that does what we were unable to do. Thank you for joining us in building the future of our field.

SFK FM

SFK ICBK027

May 7, 2008

23:7

Char Count=

Abbreviations, Transcription Conventions, and a Note Concerning Romanization

Abbreviations CONJ COP IP NEG O S Q QT TOP TENT

conjunctive particle the copula be interactional particle negation object marker subject marker question marker quotative marker topic marker tentative form

Transcription Conventions These conventions primarily originate from the transcription system developed by Gail Jefferson that is commonly used in conversation analysis (CA). The chapter authors use this system differently depending on their theoretical frameworks and analytical focuses.

[ //

onset of overlapping talk and/or gesture the point where the next turn starts, overlapping the current turn ] end of overlapping talk and/or gesture (0.0) length of silence in tenths of a second (.) micro-pause less than two tenths of a second underlining higher pitch CAPS increased volume

SFK FM

SFK ICBK027

May 7, 2008

23:7

Char Count=

Abbreviations

:, :: word-word = ? / . /, ↑/↓ ! ( ) (word)

(( )) hh .hh (h) > < < > ◦

◦◦



◦◦

bold 

xiii

lengthened syllable cut-off; glottal stop connection with the previous morpheme “latched” utterances rising/falling/continuing intonation a rise/fall in pitch animated tone, not necessarily an exclamation unintelligible transcriber’s candidate hearings / when used in English translation, indicates supplementary translation analyst’s description, including nonvocal conduct audible outbreath audible inbreath laughter within a word increase in tempo decrease in tempo reduced volume substantially reduced volume target item analyzed target line analyzed

A Note Concerning Romanization This volume adopts the Hepburn system of romanizing Japanese. Some chapter authors use the traditional Hepburn system which indicates long vowels o and u with a macron, while others repeat the vowel to show long vowels, as in oo and uu. These choices represent common practices in the various subfields of Applied Linguistics.

SFK FM

SFK ICBK027

May 7, 2008

23:7

Char Count=

SFK int

SFK ICBK027

April 21, 2008

17:1

Char Count=

Introduction

Junko Mori and Amy Snyder Ohta

Francis Xavier, one of the earliest missionaries to Japan in the sixteenth century, called Japanese “the devil’s own tongue.” This tale is often cited to illustrate just how foreign the Japanese language can be to Western ears and how challenging it is for them to learn. Despite this rather discouraging description, there has been a dramatic increase in second language speakers of Japanese. In 2003, the number of Japanese language learners outside Japan reached 2,356,745, which is 18.5 times higher than the number reported in 1979.1 Ministry of Justice statistics indicate that at the end of 2005, the number of registered aliens in Japan surpassed two million for the first time in history.2 While this is just 1.57 percent of Japan’s total population, it also shows a 47.7 percent increase over a decade. Such remarkable demographic changes can influence people’s perspectives toward the language and its speakers. “To speak and to use the Japanese language is to be a Japanese; to be a Japanese is to speak and to use the Japanese language,” said Miller (1977: 71), describing a common sentiment held among Japanese in the 1970s. Thirty years hence, we live in a different reality. Facing waves of globalization, while some may cling to, or reassert, the legend of Japanese uniqueness and homogeneity, others question its validity (e.g., Befu 2001; Donahue 2002; Sugimoto 1997; Yoshino 1992). The complexity and dynamics of Japanese language use and acquisition can be understood only by examining the intersection of ideologies, policies, and actual practices. This volume, consisting of twelve chapters of original research, aims to explore diverse issues surrounding first and second language speakers of Japanese today.

SFK int

SFK ICBK027

2

April 21, 2008

17:1

Char Count=

Introduction

The Nature and Scope of Applied Linguistics What is Applied Linguistics? Conceptualization of this volume inevitably started with this question. Recently, a flurry of books on applied linguistics and special issues of journals exploring the current state of the field have appeared, addressing this question (e.g., Cook 2003; Davis and Elder 2005; Gass and Makoni 2004; Grabe 2000; Kaplan 2002; Schmitt 2002; Seidlhofer 2003). Here are some examples of how the editors and authors of these volumes define the field: “Applied linguistics” is using what we know about (a) language, (b) how it is learned and (c) how it is used, in order to achieve some purpose or solve some problem in the real world. (Schmitt and Celce-Murcia 2002: 1) Applied Linguistics: The academic discipline concerned with the relation of knowledge about language to decision making in the practical world. (Cook 2003: 125) . . . the focus of applied linguistics is on trying to resolve language-based problems that people encounter in the real world, whether they be learners, teachers, supervisors, academics, lawyers, service providers, those who need social services, test takers, policy developers, dictionary makers, translators, or a whole range of business clients. (Grabe 2002: 9) A consensus among these thinkers appears to be that applied linguistics considers real-world problems, aiming to contribute to their solution. However, the content of these volumes suggests different understandings as to the kinds of problems to be addressed and approaches to be taken. When the field of applied linguistics emerged in the late 1940s, it focused on the problems associated with language teaching and teacher education, especially those related to English as a second or foreign language (ESL/EFL). Remnants of this are evident today as some academic programs in applied linguistics remain in departments of English and some publishers subsume applied linguistics under the category of ESL/EFL-related works. However, applied linguistics is not limited to the English language or to language classrooms, as Grabe’s description, cited above, shows. Indeed, it goes beyond educational contexts. Applied linguistics includes areas such as automatic translation, forensic linguistics, speech therapy, sign languages, or communications in business, medical, or legal professions, among others, as demonstrated by the titles of presentations given at the recent International Association of Applied Linguistics (AILA) conference.3

SFK int

SFK ICBK027

April 21, 2008

17:1

Char Count=

Japanese Applied Linguistics

3

As the real-world problems explored in applied linguistics diversify, the range of knowledge, theories, and approaches that applied linguists use has also increased. In its infancy, applied linguistics was considered a mere consumer of the findings of theoretical linguistics (Corder 1973). Now it is informed by disciplines ranging from anthropology, psychology, and sociology, to computer science and neuroscience. Applied linguistics today also intends to inform various academic and professional communities whose problems it studies, rather than just consuming knowledge generated by other disciplines. In such interdisciplinary inquiries, however, researchers may not all call themselves applied linguists. Those situated in the crossovers between the fields that compose applied linguistics may, depending on their training, identify themselves as anthropologists, psychologists, or sociologists; this also makes it challenging to identify who exactly the applied linguists are. In the meantime, as applied linguistics has developed as a decidedly interdisciplinary field with all of the trappings of an independent academic discipline, subfields of linguistics have also emerged. Corpus linguistics, discourse functional linguistics, systemic functional linguistics, sociolinguistics, and a more recent trend of interactional linguistics, for instance, all conduct empirical studies of language use. Because these studies also take a contextual and holistic view of language, considering language in action instead of treating it as an autonomous system, they are closely related to the mission of applied linguistics. This further makes it difficult to determine whether a study falls into one of the hyphenated linguistic fields or into applied linguistics. Discussing the overlap between applied linguistics and other areas of linguistics, Cook (2003, 2005) characterizes applied linguistics as offering principles and theories of solution or intervention concerning real-world problems, while the latter does not. Edmondson (2005) also asserts that a study does not qualify as applied linguistics just because it investigates language use. While this is one way to classify different types of studies, it may be an ideal promoted by some leaders in the field rather than a defining characteristic of the field; in fact, not all applied linguistic studies offer solutions or methods of interventions for the problems identified. Indeed, Bygate (2005: 572), in reviewing four articles featured in the special issue of Applied Linguistics entitled “Applied linguistics and real-world issues,” recognizes a range of approaches taken among them: some serve as “a source of insight and information in a complex world,” while others “point to more precise spaces for action and to more explicit purposes.” In summary, applied linguistics has transformed itself over the last several decades by expanding its interests and becoming a truly interdisciplinary and multidisciplinary field. Thus, depending on one’s view of how various

SFK int

SFK ICBK027

4

April 21, 2008

17:1

Char Count=

Introduction

related fields should be mapped, and where in the map one situates oneself, the terrain of applied linguistics and its products can be defined differently.

“Applied Linguistics in Japan” versus “Applied Linguistics of the Japanese Language” While applied linguistics considers itself an international field, our summary above was informed primarily by works written by American, British, or Australian authors and published in the US and UK. Given that the field originated in these countries and centered around the ESL/EFL field, it is not surprising that the majority of applied linguistics research has been conducted in Anglo-American or ESL/EFL contexts. In fact, the Anglocentrism of the field has been a matter of discussion and dissatisfaction among applied linguists. Recently, a greater diversity of traditions has begun to gain much awaited recognition in the field. For instance, AILA celebrated its fortieth anniversary with a volume focusing on applied linguistics as an international discipline and featuring authors from different regions of the world (Gass and Makoni 2004). In this spirit, the present volume works to expand the field’s regional scope with its focus on Japanese applied linguistics. It offers Japanese specialists as well as applied linguists in general an opportunity to explore how theoretical and methodological frameworks widely adopted in applied linguistic inquiries manifest themselves in studies of the Japanese language. To clarify the nature of this volume and its contributions, we will provide a brief historical overview of how applied linguistics has developed in Japan as well as among scholars who take a particular interest in Japanese language use and acquisition. One strand of applied linguistics in Japan, as elsewhere, developed out of the ESL/EFL field and many studies involving Japanese learners of English have been conducted. According to Block (1996, 1997), Japan ranks as the fourth most common origin (after the US, Canada, and the UK) of articles that appeared in four major journals in the field, Applied Linguistics, Language Learning, TESOL Quarterly, and Studies in Second Language Acquisition (Lessard-Clouston 1998). That the Japan Association of Applied Linguistics (JAAL), AILA’s Japan affiliate since 1984, is housed within the Japan Association of College English Teachers (JACET) also symbolizes this developmental history.4 In fact, some of the Japanese authors in this volume were introduced to applied linguistics through backgrounds in EFL education. In the meantime, studies concerning Japanese as a foreign or second language (JFL/JSL) have developed in parallel as numbers of JFL/JSL learners and speakers have risen. The Nihongo Ky¯oiku Gakkai (The Society

SFK int

SFK ICBK027

April 21, 2008

17:1

Char Count=

Japanese Applied Linguistics

5

for Teaching Japanese as a Foreign Language) was founded in 1962 and comparable associations were also established in neighboring East Asian countries and in North America, Australia, and Europe. However, the number of JFL/JSL studies appearing in publications written in English, thereby participating in the mainstream discourse of applied linguistics, is still limited; the majority of these originate in North America and Australia, or are written by authors who received graduate training in those countries but reside in Japan. Along with these post World War II developments, we should also consider how pre-war activities have shaped the contemporary Japanese language. Critical questions concerning Japanese language emerged as Japan confronted the West and began a process of modernization and industrialization in the latter half of the nineteenth century. Those questions included which version of Japanese should be “Standard Japanese” and how the orthography could be simplified to increase literacy and improve accessibility to public documents. The Japanese government established the Kokugo Shingikai (National Language Policy Board) in 19345 to address these problems and set policies and standards for spoken and written Japanese. The Kokugo Shingikai continues functioning today (see Wetzel, Okamoto, and Kubota, Chapters 4, 5, and 12, respectively, in this volume, for more details). After World War II, in 1948, the Kokuritsu Kokugo Kenkyuujo (National Language Research Institute)6 was also established “to conduct scientific surveys on the Japanese language and the linguistic life of the Japanese people in order to create reliable bases for rationalization of the Japanese language” (National Language Research Institute 1998: 1). Although these activities were independent of the development of applied linguistics in the West, these government-initiated activities were informed by scholarly activities that were “applied linguistic” in nature vis-`a-vis the field’s most recent and inclusive definition. They have influenced both first and second language educational materials, affected public perceptions and beliefs regarding different variations of the language, and impacted the topics and research methods pursued by Japanese applied linguists. Thus applied linguists are continually interested in the historical formation of language polices and ideologies that interact with the daily life of language users. The chapters in this volume explore various aspects of the Japanese language and a range of issues concerning its use and acquisition by accounting for the language’s particular historical, cultural, or linguistic features while incorporating theories and methodologies developed in the West. All the authors received training in North America and their academic backgrounds and affiliations include applied linguistics, education, Japanese linguistics, and linguistics. Such diversity means that all may not

SFK int

SFK ICBK027

6

April 21, 2008

17:1

Char Count=

Introduction

immediately identify themselves as applied linguists, yet all have presented research at conferences of the American Association of Applied Linguistics (AAAL) or AILA. Together, this volume aims to enhance dialogue between Japanese language specialists and applied linguists in general.

The Chapters This volume has four parts, featuring renewed understandings of language in action (Part One), language ideologies, their formation, and their influence on identity development (Part Two), speakers of Japanese as an additional language in various settings (Part Three), and critical approaches to foreign language pedagogy (Part Four). The first half of the volume focuses on interactions among people who speak Japanese as a first language or public discourse taking place in Japan, while the second half focuses on Japanese as a second or additional language. These halves are implicitly and explicitly interrelated; the findings of the first group of studies contributes to our understanding of the experiences and challenges encountered by JFL/JSL users as they learn Japanese and become socialized into new communities, while the second group of studies addresses themes from the first as they explore issues related to second language acquisition and socialization.7 Part One, Reexamination of language in action, contains three chapters. In Chapter 1, Ono and Jones discuss the importance of functional linguistics, or what has more recently been called usage-based linguistics (e.g., Bybee 2006; Bybee and Hopper 2001; Giv´on 1979; Hopper and Thompson 1984; Langacker 1987, 1991), and exemplifies its contribution by reexamining tara, nara, ba, and to, forms that have been considered as a set of conditionals. Traditionally, it has been assumed that a speaker chooses one form from this set based on various factors and productive rules. However, through an analysis of the frequency and use of these forms in corpora of naturally occurring conversations, Ono and Jones demonstrate that they can be characterized as more lexical than grammatical and do not seem to form a set for speakers. Their findings are important for language instruction, prompting a critical analysis of current approaches to teaching these forms, which are primarily informed by introspection and invented examples and do not fully reflect how the language is actually used in real life. Chapter 2, written by Mori and Nakamura, continues the focus on Japanese spoken discourse showcasing an approach called interactional linguistics, which has emerged in the intersection between conversation analysis, functional linguistics, and linguistic anthropology (e.g., Ochs, Schegloff, and Thompson 1996; Selting and Couper-Kuhlen 2001). Interactional linguistics emphasizes the importance of replicating the moment-by-moment

SFK int

SFK ICBK027

April 21, 2008

17:1

Char Count=

Japanese Applied Linguistics

7

progress of turns at talk experienced by interactional participants as well as tracking non-verbal behaviors such as gaze shifts and gestures. Mori and Nakamura pay special attention to how speakers negotiating opinions project or equivocate turn-completion points by manipulating linguistic forms that have dual syntactic status. Their findings prompt a reconsideration of stereotypical characterizations that Japanese ways of speaking are indirect or non-confrontational, and show the viability of an approach that explicates Japanese talk-in-interaction by considering interrelationships between linguistic structures and social actions. Cook’s analysis of the Japanese plain form, presented in Chapter 3, also examines linguistic forms in use, concerning how they are situated in, and at the same time recreate, interactional contexts. She examines how speakers organize and sustain interactional activities via linguistic and non-linguistic resources, and how “the naked plain form” (i.e., non-honorific form without an affect key) conveys distinct meanings in these different activities. Cook’s analyses draw upon the work of Silverstein (1976) on indexicality, Gumperz (1982) on contextualization cues, as well as Duranti and Goodwin (1992) on the re-conceptualization of contexts. Her analysis of the language used in a Japanese elementary school classroom in Tokyo demonstrates how fourth graders, as first language speakers of Japanese, have already acquired the subtle ways of selecting different speech styles for different purposes in different contexts, the mastery of which presents a tremendous challenge for adult JFL learners. The complexity of spontaneously making an appropriate choice of speech styles is echoed in later chapters by Wetzel (Chapter 4) and Yotsukura (Chapter 9) although they approach the issue through different pathways. Part Two, Ideologies, diversities, and identities, takes political and sociolinguistic dimensions into consideration in their research on Japanese language use. In Chapter 4, Wetzel presents a historical account of keigo, often translated as “honorific language,” “linguistic politeness and formality,” or “polite forms/style,”8 and the development of language ideology (Cameron 1995; Fairclough 1989, 1992, 1995; Joseph and Taylor 1990). Wetzel begins with the language standardization movement of the Meiji era (1863–1911), when the term keigo was first coined, and ends in early 2007. Recent public discourse reveals how keigo is seen as an important cultural artifact and a societal necessity, while at the same time being viewed as a complex system whose misuse is on the rise. Attempting to provide guidance for proper use, in 2006, the government introduced a new analysis of keigo that replaces the popularly accepted three-way division—sonkeigo (‘honorific language’), kenj¯ogo (‘humble language’), and teineigo (‘formal language’)—by introducing two different types of kenj¯ogo and adding the category of bikago (‘beautified language’). Wetzel’s analysis of varying

SFK int

SFK ICBK027

8

April 21, 2008

17:1

Char Count=

Introduction

reactions to this newest document demonstrates the competing ideologies of keigo that exist in Japanese society today. Okamoto’s discussion of “regional” versus “standard” Japanese in Chapter 5 also relates to Japan’s linguistic standardization movement. In the process of Japan’s modernization, Standard Japanese, developed based on language used by educated elites in Tokyo, was promoted by enforcing language policies, most significantly through the educational system. The development of mass communication and transportation further accelerated its spread. As a result, many Japanese speakers today are considered to be bi-dialectal, code-switching between regional and standard varieties depending on the situation (e.g., Inoue 1988; Long 1996; Miyake 1995; Sanada 1996; Shibata 1988). However, Okamoto challenges the assumptions behind such descriptions, namely the existence of regional dialects as distinct varieties of Japanese. Through a close analysis of conversations recorded in various contexts, Okamoto demonstrates how a speaker residing in the Kansai region mixes elements of regional and standard variations in a complex manner, even within a sentence or a phrase. She argues that regional and standard forms are variants, and that variant choice indexes not only regionality but also formality. Her discussion refers to micro-level discourse and interactional contexts as well as macro-level social and historical contexts. Okada’s work in Chapter 6, which addresses issues concerning gender and language, also relates to the linguistic standardization movement discussed in the previous two chapters. As well documented by Okamoto and Smith (2004) and Inoue (2006), among others, gender ideology assumes that distinct men’s and women’s languages developed along with the spread of Standard Japanese. While recent research has suggested that gendered differences may be less pronounced (e.g., Okamoto 1995), public discourse presents competing voices that promote, accept, or grieve this diminishing difference, revealing varying ideological beliefs among Japanese. Against this background, and following a recent trend of investigating language used by diverse populations of women, Okada introduces data that feature a female boxing trainer coaching male boxers. She analyzes directives the trainer used, including an imperative form that has been labeled as “strongly masculine.” Rather than immediately linking these forms to a particular gender, Okada shows the importance of reexamining these forms in relation to the sequential development of talk and co-occurring multimodal resources, including gaze and body movements. In the context of sports, the imperative form, for instance, conveys the immediacy of actions that need to be performed at precise moments. Thus, selection of particular forms moves beyond gender to the establishment of “professional vision” (Goodwin 2003).

SFK int

SFK ICBK027

April 21, 2008

17:1

Char Count=

Japanese Applied Linguistics

9

In Chapter 7, Matsumoto explores the issue of identity construction through talk by focusing on elderly Japanese women, a rapidly growing population in Japan. Expanding earlier studies on elderly speakers’ “painful self-disclosure (PSD)” (Coupland, Coupland, and Grainger 1991; Coupland, Coupland, Giles, and Henwood 1991), Matsumoto analyzes how Japanese elderly women perform PSD in interactions with peers and the researcher. Qualitative analysis demonstrates how these women continue to adjust their established adult personae by acknowledging the changing reality associated with the aging process and how their identities are discursively co-constructed with the interlocutors, who may or may not share similar experiences and perspectives. They illustrate elderly identity as one of (or in) transition while highlighting identities other than “being elderly” that are important for defining who they are. The three chapters composing Part Three, Japanese as an additional language, return to applied linguistics’ origins by considering issues related to Japanese language learning and teaching. These studies portray second language learners and speakers in and out of classrooms. In Chapter 8, Ohta examines novice JFL learners at a North American university, paying special attention to occurrences of laughter in classroom interaction and how laughter functions in the discourse. Grounding her research in a sociocultural perspective on foreign language learning and development (Lantolf and Thorne 2006), Ohta posits that laughter is simultaneously social interactive, affective, and cognitive. Her findings show how laughter reveals learner mental processes. Laughter occurred during ludic language play (Broner and Tarone 2001) including humor and joking, during production of katakana English (English words pronounced using Japanese phonological rules), when classmates made errors, and during production difficulties. Showing the influence of conversation analytic methodology (Glenn 2003), Ohta also considers the placement of laughter. Results demonstrate how laughter shows second language comprehension, with analysis of placement revealing learners’ facility at parsing Japanese phrasal and sentence structure. Namely, laughter is not just about affect, but careful contextual examination of this affect-laden paralinguistic device provides insight into second language developmental processes. Moving outside of the classroom, Chapter 9 by Yotsukura examines the Japanese language used by first and second language speakers to do toiawase (‘inquiry’) telephone calls to educational institutions in Japan. The second language speakers studied by Yotsukura were beyond the novice level examined by Ohta and were using Japanese for their own life purposes to gain information for themselves or others. Using the notion of speech genres proposed by Bakhtin (1986) and building on her own earlier work on business calls (Yotsukura 2003), Yotsukura describes the sequential moves, forms,

SFK int

SFK ICBK027

10

April 21, 2008

17:1

Char Count=

Introduction

and styles that make toiawase calls a distinctive genre. Proficient speakers organized their calls by using maeoki (‘prefatory statements’); further they utilized the proper forms and styles associated with each particular move. In this context of institutional calls, appropriate use of keigo and maintenance of formality also constitute crucial elements. Yotsukura’s data reveal notable differences among first language speakers of Japanese that indicate their facility with this particular genre. The data also show distinctively “non-native” features observed in some calls made by JSL speakers. In Chapter 10, Kanno introduces another distinct population of JSL learners and speakers, children of immigrants and migrant workers. Along with the increase in the number of registered foreigners in Japan, the number of students in public schools who need JSL instruction has nearly doubled in the last decade. Through the theoretical lens of transnationalism (Basch, et al. 1994; Portes, et al. 1999) and imagined communities (Anderson 1991; Kanno and Norton 2003), Kanno examines policies and practices of two elementary schools that are known as “model schools” for language minority education. In a country like Japan that has fostered the hegemony of linguistic and cultural homogeneity, the rising trend of immigrants’ maintenance of transnational ties can present significant challenges. Through case studies of the two schools, Kanno reports that the Japanese educational system tends to classify language minority students into two groups, those staying permanently in Japan and those residing temporarily in Japan. By doing so, it places these students either in or out of the existing nationalist education framework and approaches this new reality of growing diversity without revising its fundamental educational policies. Finally, Part Four, Critical reflection on language pedagogy, presents two chapters, which reflect on issues raised in the preceding chapters and propose new perspectives and approaches to be brought into the second language classroom. In Chapter 11, Yoshimi considers ways of scaffolding learners’ developing second language competence by explicitly encouraging them to draw upon existing competencies: interactional, institutional and sociocultural, or communicative. Her proposal is based on recent studies which promote the renewed appreciation of “competencies as resources” that enable learning activities (e.g., Lee 2006; Mondada and Pekarek Doehler 2004; Sullivan 2000). Yoshimi’s participants were enrolled in two classes of beginning level JFL at a North American university. Students in the one class were instructed in a conventional method valuing fluency and target-like production of a limited instructed repertoire of communicative practices modeled after textbook dialogs. On the other hand, those in the experimental group, while exposed to the same instructional materials, were encouraged to bring their own experiences, intuitions, and knowledge into the JFL classroom. Comparison of students’ performances at the end

SFK int

SFK ICBK027

April 21, 2008

17:1

Char Count=

Japanese Applied Linguistics

11

of the semester demonstrated how different pedagogical principles may socialize them into different practices and how viable it is to embrace the learner’s existing competence for development of extended use of Japanese in classroom settings. Chapter 12 by Kubota concludes the volume by discussing how critical applied linguistics (e.g., Norton and Toohey 2004; Pennycook 2001) can be applied to examine three main constructs in teaching and learning Japanese, namely Japanese language, people, and culture. While critical perspectives have been nurtured in the field of ESL/EFL for some time, in the field of JFL education, discussions of critical pedagogies and critical approaches to teaching language and culture are just emerging (e.g., Carroll 2005; Heinrich 2005; Kubota 1996, 2003; Ohara, et al. 2001). Many of the chapters in this volume discuss social, historical, and political factors contributing to the formation of language and educational ideologies or examine actual language use of diverse populations of Japanese speakers which intricately interact with such ideologies, but Kubota explores how such information can be brought into critical reflection of instructional practices and materials. She also situates our understanding of Japanese language, people, and culture in Japan’s imperial and colonial history, and recommends that Japanese language educators recognize the discursive nature of these constructs. She reminds us that no matter what kind of stance or action language teaching professionals take, the choice unavoidably reflects some kind of ideology, and at the same time contributes to the preservation of the very ideology.

Conclusion Given the heterogeneous nature of Japanese applied linguistics, it is impossible to present the breadth and depth of this field in one volume and we quickly gave up the idea of providing a comprehensive review of the field. Instead, aiming to establish a balance between representing the field’s diversity and maintaining coherence, we chose these topics to reflect prominent issues surrounding first and second language speakers of Japanese today. We also hoped that these chapters would offer some innovative perspectives to the field of applied linguistics. As the subtitle indicates, the volume focuses on discourse and social perspectives rather than cognitive or psycholinguistic perspectives. There are many other issues we were not able to cover that deserve serious attention (literacy and written language, technology and language use, heritage language education, to name a few);9 at the same time, each topic included in this volume can be expanded into a single book-length manuscript. Within our own limitations as authors and

SFK int

SFK ICBK027

12

April 21, 2008

17:1

Char Count=

Introduction

editors, we hope that we have provided a window to critical reflection on language use, ideology, policy, and pedagogy and renewed appreciation of how discourse and social approaches can be adopted to investigate these issues. Our aim has also been to speak to accomplished scholars as well as newcomers to the field of applied linguistics, or those who are unfamiliar with issues associated with Japanese language. To accomplish this, the authors discuss the theoretical and methodological frameworks for their studies, as well as writing with an eye to providing examples illustrating how these were realized in their research. If our work motivates further reading in a particular area or prompts new research to advance our understanding of these issues and solve unresolved problems, we would consider our goal to have been achieved.

Notes 1. Japan Foundation survey information was retrieved from www.jpf.go. jp/j/japan j/news/0407/07-01.html on February 11, 2007. 2. Information retrieved from www.moj.go.jp/PRESS/060530-1/0605301.html on February 11, 2007. This number includes registered aliens who do not speak Japanese as well as those who speak Japanese as their first language. 3. Expansion of topics covered under applied linguistics continues; however, studies related to language education still constitute the major body of applied linguistic research and some introductory books (e.g., Schmitt 2002) focus on language teaching and learning. 4. The Japanese Association for Language Teaching (JALT), the Japan affiliate of TESOL and IATFL, is another major organization in Japan. 5. Its predecessors include Kokugo Ch¯osakai (National Language Investigative Group) established in 1899 and Kokugo Ch¯osa Iinkai (National Language Research Council) established in 1902 (Wetzel 2004: 46). 6. The English name was changed to the National Institute for Japanese Language in 2001. 7. Recent collections of articles edited by Hatasa (2003) and Yoshitomi, Umino, and Negishi (2006), on the other hand, exclusively address issues concerning second language acquisition and pedagogy. 8. These English terms appear in Wetzel (2004), who subsequently discusses the difficulty of representing keigo and the complex linguistic phenomena the term refers to in simple English terms. 9. See Gottlieb (1995, 2005), for instance, on script reform, literacy, and technology and language use, and Kondo-Brown (2005), Matsunaga (2003), or Shibata (2004) on heritage language education.

SFK int

SFK ICBK027

April 21, 2008

17:1

Char Count=

Japanese Applied Linguistics

13

References Anderson, B. (1991) Imagined Communities: Reflections on the Origin and Spread of Nationalism (rev. ed.). London: Verso. Bakhtin, M. M. (1986) Speech Genres and Other Late Essays. Austin: University of Texas Press. Basch, L., Glick Schiller, N., and Szanton Blanc, C. (1994) Nations Unbound: Transnational Projects, Postcolonial Predicaments, and Deterritorialized Nationstates. Langhorne, PA: Gordon and Breach. Befu, H. (2001) Hegemony of Homogeneity: An Anthropological Analysis of “Nihonjinron.” Melbourne: Trans Pacific Press. Block, D. (1996) “Not so fast! Some thoughts on theory culling, relativism, accepted findings and the heart and soul of SLA.” Applied Linguistics 17(1), 65–83. Block, D. (1997) “Publishing patterns and McDonaldization.” IATEFL Newsletter 136, 12–15. Broner, M. A. and Tarone, E. E. (2001) “Is it fun? Language play in a fifthgrade Spanish immersion classroom.” Modern Language Journal 85(3), 493–525. Bybee, J. (2006) “From usage to grammar: The mind’s response to repetition.” Language 82(4), 711–733. Bybee, J. and Hopper, P. (2001) Frequency and the Emergence of Linguis- tic Structure. Amsterdam: John Benjamins. Bygate, M. (2005) “Applied linguistics: A pragmatic discipline, a generic discipline?” Applied Linguistics 26(4), 568–581. Cameron, D. (1995) Verbal Hygiene. London: Routledge. Carroll, T. (2005) “Beyond keigo: Smooth communication and the expression of respect in Japanese as a foreign language.” International Journal of Sociology of Language 175/176, 233–247. Cook, G. (2003) Applied Linguistics. Oxford: Oxford University Press. Cook, G. (2005) “Calm seas or troubled waters? Transitions, definitions and disagreements in applied linguistics.” International Journal of Applied Linguistics 15(3), 282–301. Corder, S. P. (1973) Introducing Applied Linguistics. Harmondsworth: Penguin. Coupland, J., Coupland, N., and Grainger, K. (1991) “Intergenerational discourse: Contextual versions of aging and elderliness.” Ageing and Society 11, 189–208. Coupland, J., Coupland, N., Giles, H., and Henwood, K. (1991) “Formulating age: Dimensions of age identity in elderly talk.” Discourse Processes 14, 87–106. Davis, A. and Elder, C. (eds.) (2005) The Handbook of Applied Linguistics. Malden, MA: Blackwell.

SFK int

SFK ICBK027

14

April 21, 2008

17:1

Char Count=

Introduction

Donahue, R. T. (ed.) (2002) Exploring Japaneseness: On Japanese Enactments of Culture and Consciousness. Westport, CT: Ablex. Duranti, A. and Goodwin, C. (1992) “Rethinking context: An introduction.” In A. Duranti and C. Goodwin (eds.), Rethinking Context. Cambridge: Cambridge University Press, pp. 1–42. Edmondson, W. (2005) “Prejudice and practice in applied linguistics.” International Journal of Applied Linguistics 15(3), 389–398. Fairclough, N. (1989) Language and Power. London: Longman. Fairclough, N. (1992) Discourse and Social Change. London: Polity Press. Fairclough, N. (1995) Critical Discourse Analysis. London: Longman. Gass, S. and Makoni, S. (eds.) (2004) “World applied linguistics.” AILA Review 17. ´ T. (1979) On Understanding Grammar . New York: Academic Press. Givon, Glenn, P. (2003) Laughter in Interaction. Cambridge: Cambridge University Press. Goodwin, C. (2003) “The semiotic body in its environment.” In J. Coupland and R. Gwyn (eds.), Discourse, the Body, and Identity. Basingstoke, Hants.: Palgrave Macmillan, pp. 19–42. Gottlieb, N. (1995) Kanji Politics: Language Policy and Japanese Script. London; New York: Kegan Paul International. Gottlieb, N. (2005) Language and Society in Japan. Cambridge: Cambridge University Press. Grabe, W. (ed.) (2000) “Applied linguistics as an emerging discipline.” Annual Review of Applied Linguistics 20. Grabe, W. (2002) “Applied linguistics: An emerging discipline for the twenty-first century.” In R. B. Kaplan (ed.), The Oxford Handbook of Applied Linguistics. New York: Oxford University Press. Gumperz, J. (1982) Discourse Strategies. Cambridge: Cambridge University Press. Hatasa, Y. (ed.) (2003) An Invitation to Second Language Acquisition Research in Japanese. Tokyo: Kuroshio Publishers. Heinrich, P. (2005) “Language ideology in JFL textbooks.” International Journal of Sociology of Language 175/176, 213–232. Hopper, P. J. and Thompson, S. A. (1984) “The discourse basis for lexical categories in universal grammar.” Language 60(4), 703– 752. Inoue, F. (1988) “Dialect image and new dialect forms.” Tokyo University of Foreign Languages, Area and Culture Studies 38, 13–22. Inoue, M. (2006) Vicarious Language: Gender and Linguistic Modernity in Japan. Berkeley, CA: University of California Press. Joseph, J. E. and Taylor, T. J. (eds.) (1990) Ideologies of Language. London: Routledge.

SFK int

SFK ICBK027

April 21, 2008

17:1

Char Count=

Japanese Applied Linguistics

15

Kanno, Y. and Norton, B. (2003) “Imagined communities and educational possibilities: Introduction.” Journal of Language, Identity, and Education 2(4), 241–249. Kaplan, R. B. (ed.) (2002) The Oxford Handbook of Applied Linguistics. New York: Oxford University Press. Kondo-Brown, K. (2005) “Differences in language skills: Heritage language learner subgroups and foreign language learners.” Modern Language Journal 89(4), 563–581. Kubota, R. (1996) “Nihongo ky¯oiku ni okeru hihan ky¯oiku, hihan teki yomi kaki ky¯oiku [Critical pedagogy and critical literacy in teaching Japanese].” Japanese-Language Education around the Globe 6, 35–48. Kubota, R. (eds.) (2003) “Special Issue: Sociocultural issues in teaching Japanese: Critical approaches.” Japanese Language and Literature 37. Langacker, R. W. (1987) Foundations of Cognitive Grammar. Vol. I: Theoretical Prerequisites. Stanford, CA: Stanford University Press. Langacker, R. W. (1991) Foundations of Cognitive Grammar. Vol. II: Descriptive Applications. Stanford, CA: Stanford University Press. Lantolf, J. P. and Thorne, S. L. (2006) Sociocultural Theory and the Genesis of Second Language Development. New York: Oxford University Press. Lee, Y. (2006) “Towards a respecification of communicative competence: Condition of L2 instruction or its objective?” Applied Linguistics 27(3), 349–376. Lessard-Clouston, M. (1998) “Perspectives on language learning and teaching in Japan: An introduction.” Language, Culture and Curriculum 11(1), 1–8. Long, D. (1996) “Quasi-standard as a linguistic concept.” American Speech 71, 118–135. Matsunaga, S. (2003) “Instructional needs of college-level learners of Japanese as a heritage language: Performance-based analyses.” Heritage Language Journal 1(1) PDF document retrieved from www. international.uda. edu/languages/heritagelanguages/journal/article.asp?parentid=3600 on October 29, 2007. Miller, R.A. (1977) The Japanese Language in Contemporary Japan: Some Sociolinguistic Observations. Washington: American Enterprise Institute for Public Policy Research. Miyake, Y. (1995) “A dialect in the face of the standard: A Japanese case study.” Proceedings of the 21st Annual Meeting of the Berkeley Linguistics Society, 217–225. Mondada, L. and Pekarek Doehler, S. (2004) “Second language acquisition as situated practice: Task accomplishment in the French second language classroom.” Modern Language Journal (88), 501–518.

SFK int

SFK ICBK027

16

April 21, 2008

17:1

Char Count=

Introduction

National Language Research Institute (1998) An Introduction to the National Language Institute: A Sketch of its Achievement Fourth Edition. PDF document retrieved from www.kokken.go.jp/english/en/ on February 11, 2007. Norton, B. and Toohey, K. (eds.) (2004) Critical Pedagogies and Language Learning. Cambridge: Cambridge University Press. Ochs, E., Schegloff, E. A., and Thompson, S. A. (eds.) (1996) Interaction and Grammar . Cambridge: Cambridge University Press. Ohara, Y, Saft, S., and Crookes, G. (2001) “Toward a feminist critical pedagogy in a beginning Japanese-as-a-foreign-language class.” Japanese Language and Literature 35(2), 105–133. Okamoto, S. (1995) “‘Tasteless’ Japanese: less ‘feminine’ speech among young Japanese women.” In K. Hall. and M. Bucholtz (eds.), Gender Articulated: Language and the Socially Constructed Self . New York: Routledge, pp. 297–325. Okamoto, S. and Smith, J. S. (eds.) (2004) Japanese Language, Gender, and Ideology: Cultural Models and Real People. New York: Oxford University Press. Pennycook, A. (2001) Critical Applied Linguistics: A Critical Introduction. Mahwah, NJ: Lawrence Erlbaum Associates. Portes, A., Guarnizo, L. E., and Landolf, P. (1999) “The study of transnationalism: Pitfalls and promises of an emergent research field.” Ethnic and Racial Studies 22(2), 217–237. Sanada, S. (1996) Chiikigo no dainamizumu: Kansai hen [Dynamism of ¯ u. Regional Languages: The Case of Kansai]. Tokyo: Of¯ Schmitt, N. (2002) An Introduction to Applied Linguistics. London: Arnold. Schmitt, N. and Celce-Murcia, M. (2002) “An overview of applied linguistics.” In N. Schmitt (ed.), An Introduction to Applied Linguistics. London: Edward Arnold, pp. 1–18. Seidlhhofer, B. (ed.) (2003) Controversies in Applied Linguistics. Oxford: Oxford University Press. Selting, M. and Couper-Kuhlen, E. (eds.) (2001) Studies in Interactional Linguistics. Amsterdam: John Benjamins. Shibata, T. (1988) Ikite iru Nihongo: Hoogen tansaku [Living Japanese: Exploring dialects]. Tokyo: Kodansha. Shibata, S. (2004) “The effects of Japanese heritage language maintenance on scholastic verbal and academic achievement in English.” Foreign Language Annals 37(2), 224–231. Silverstein, M. (1976) “Shifters, linguistic categories, and cultural description.” In K. Basso and H. Selby (eds.), Meaning in Anthropology. Albuquerque: University of New Mexico Press, pp. 11–55. Sugimoto, Y. (1997) An Introduction to Japanese Society. Cambridge: Cambridge University Press.

SFK int

SFK ICBK027

April 21, 2008

17:1

Char Count=

Japanese Applied Linguistics

17

Sullivan, P. N. (2000) “Playfulness as mediation in communicative language teaching in a Vietnamese classroom.” In J. P. Lantolf (ed.), Sociocultural Theory and Second Language Learning. Oxford: Oxford University Press, pp. 115–131. Wetzel, P. (2004) Keigo in Modern Japan: Polite Language from Meiji to the Present. Honolulu: University of Hawai‘i Press. Yoshino, K. (1992) Cultural Nationalism in Contemporary Japan: A Sociological Enquiry. London: Routledge. Yoshitomi, A., Umino, T., and Negishi, M. (eds.) (2006) Readings in Second Language Pedagogy and Second Language Acquisition: In Japanese Context. Amsterdam: John Benjamins. Yotsukura, L.A. (2003) Negotiating Moves: Problem Presentation and Resolution in Japanese Business Discourse. Oxford: Elsevier Science.

SFK int

SFK ICBK027

April 21, 2008

17:1

Char Count=

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Chapter 1

Conversation and Grammar: Approaching So-Called Conditionals in Japanese

Tsuyoshi Ono and Kimberly Jones

Introduction1 For a number of years, we have been interested in what the empirical study of language use might have to offer not only linguists, but also specialists in second language acquisition and teaching. As we observed the diversity of how the Japanese language is used by speakers interacting in real-life contexts, it became clear that what we saw was frequently at odds with the sentences described in much linguistic research, with descriptions of the Japanese language found in textbooks for nonnative speakers, and with pervasive ideologies about the Japanese language that are held by many people, be they linguists, language teachers, or lay people. As the analysis of language based on naturally occurring discourse continues to progress, it seems appropriate that we reexamine our textbooks, curricula, and classroom practices to consider whether what we teach and do in our language classes reflects language practices in the real world. As one example of a mismatch between ideology and actual practice, in her earlier work on Japanese conflict talk, Jones (1990, 1993, 1995) found that despite the prevalent belief that Japanese speakers do not explicitly engage in conflict talk, it was not difficult to find examples of quite explicit talk expressing and negotiating conflicts that arose between participants in conversations. However, whether because of the ideology of a harmonious Japan, or simply because language textbooks in general tend to portray a smoothly functioning world free of interpersonal conflict, students of Japanese are rarely taught how to engage in disagreement (see Mori and Nakamura, Chapter 2). The experience of encountering a relatively

SFK ch01

SFK ICBK027

22

April 20, 2008

10:24

Char Count=

Conversation and Grammar

advanced nonnative speaker who thought that a common discourse marker of disagreement, saa, soo deshoo ka “hmm, I wonder if that is the case,” conveyed an inclination to agree with one’s interlocutor convinced Jones that students would benefit from being taught that Japanese speakers do at times explicitly discuss the inevitable conflicts that arise between them, and that there are specific forms that they typically use to do so. Research presented in a number of the chapters in this book also looks at ideologies about language use that are commonly referred to in Japanese language pedagogy. Chapters that fall in this category include those that consider speech styles (Cook, Chapter 3), keigo (Wetzel, Chapter 4), and “gendered” language (Okada, Chapter 6), all issues that are typically addressed by teachers and textbooks of Japanese. By taking an empirical look at how speakers in real-life contexts actually use these and other ways of speaking, as the chapters in the first half of this book do, we can develop a more informed pedagogy and avoid teaching students a stereotypical and/or inaccurate version of Japanese. Nor is it only more global ideologies about ways of speaking and about what can or cannot be discussed that turn out to be inaccurate. In another study connecting discourse and language pedagogy, we discovered a number of ways textbook dialogues fail to accurately reflect natural speech (Ono and Jones 2001; Jones and Ono 2005). Textbook dialogues generally focus on exchanging information and are typically comprised of neat pairs of complete sentences, often with a large amount of new information in one sentence. Information exchange, however, is only one of many possible functions of interpersonal talk, and in real conversations, only a limited amount of new information is typically introduced in each utterance (see Chafe 1987, 1994). Perhaps even more importantly, textbook dialogues typically fail to include many of the linguistic devices that Japanese speakers use frequently in order to clarify what they are saying and to confirm mutual comprehension—devices such as repetition, repair, postposing, interactional particles, backchannels, fillers, and lengthening. All these aspects of language use can be taught, so here again we can see clear applications that follow once we understand more about the nature of language as it is used in interactions. In another study that considers the gap between textbook and classroom speech on the one hand, and naturally occurring social interactions on the other, Mori (2005) questions the way dooshite (‘why’) questions are introduced in Japanese language classes. Textbooks introduce these questions with the aim of teaching a particular question structure. However, a dooshite question can be interpreted as challenging something that another person has said or done. Thus, although such questions may seem acceptable in classroom language practice, they can be problematic when used in actual

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

23

social interactions. As Mori points out, it is crucial that we integrate discourse and sociolinguistic perspectives with the teaching of grammar. These are only a few of the relevant studies that have explored how studies based on empirical data can help us better understand the reality of how speakers use language, and thus help us provide a more accurate picture of language use to students. While some of our previous studies have tended to look at more global issues such as language ideologies or discourse organization, in this chapter, we want to discuss how what has traditionally been called “grammar” (or “morphosyntax”) can be studied in actual use, thus extending our previous work to a consideration of how an understanding of actual language use can inform our teaching of specific grammatical structures, and indeed, our understanding of the very nature of human language. In order to do this, we will first discuss the theoretical assumptions underlying our approach to the study of language, which is a discourse-functional approach, and mention some typical basic assumptions about human language that such an approach might call into question. The assumptions that we question underlie much work in linguistics that does not examine spoken language. They represent an earlier and dominant tradition that has been advocated by the linguist Noam Chomsky and his followers since the late 1950s. These assumptions have permeated the field of linguistics for the past several decades without having really been tested to see whether they are a good fit for what we can observe in actual language use. In this chapter, we start afresh with a set of perspectives which are often at odds with the traditional assumptions, perspectives that seem to us to be a better fit with what speakers do in everyday talk. We should note that some of the alternative perspectives we discuss have already been discussed and advocated with various degrees of detail by other researchers, especially those who take an approach to linguistics often called functional linguistics, such as Giv´on (1979), Hopper and Thompson (1980, 1984), Langacker (1987, 1991), Chafe (1994), Ochs, Schegloff, and Thompson (1996), Tomasello (1998, 2003), Bybee and Hopper (2001), and Bybee (2006). This alternative tradition came about in the 1970s as a reaction to the dominant tradition mentioned above. What will be presented here is essentially our own version of functional linguistics, more recently called usage-based linguistics.2 We feel that the insights offered by a usagebased approach to linguistics, stemming as they do from what we observe in everyday talk, have valuable implications for what a model of human language should be like. In the second part of the chapter, we will examine so-called “conditionals” in Japanese as a case study to illustrate the way in which grammar can be studied based on actual language use.3 We hope that this case study will

SFK ch01

SFK ICBK027

24

April 20, 2008

10:24

Char Count=

Conversation and Grammar

show the reader how we might go about studying everyday talk in order to understand and represent human language, and why such an understanding might prove valuable not only to linguists, but also to language teachers and their students.

Theoretical Underpinnings Centrality of everyday talk: Data and methodology It is well accepted among linguists that spoken language is the fundamental form of human language. There are several reasons for this. Spoken language is shared by the great majority of people in the world. With the exception of sign languages, it is part of every natural language. Without being explicitly taught to do so, we learn to speak and participate in speech activities that are part of everyday interactions, such as greeting, chatting, joking, and telling stories. Obviously, speakers have different skill levels in these verbal activities, but the fact remains that all speakers learn to perform them, barring some sort of disability that inhibits language acquisition. Writing, another form of language, on the other hand, is much more specialized. For one thing, a great majority of world languages do not have a writing system. For languages which have a writing system, the written language has to be explicitly taught to most people and often requires years of practice to acquire. And even in a language like Japanese, in which there is a long tradition of writing and a great majority of the speakers can write, writing still seems to be a marked form of language. That is, though Japanese speakers living in Japan are unlikely to get through a day without seeing some written materials, for most speakers, the amount of written language they actually produce on a given day is likely to be far less than the amount of spoken language they produce. For the majority of people, the spoken language constitutes the greater part of their linguistic life. Due to the centrality of spoken language described above, linguists have always been interested in finding out what makes spoken language possible. This is partly motivated by a goal shared by many linguists: to identify language universals, features common among all the languages in the world. The basic idea is that since spoken language is so fundamental and is shared by practically all languages, an examination of spoken language should help us identify language universals. Following this line of thought, linguists have attempted to understand the ability (often called competence or knowledge) that allows speakers to produce actual spoken language (often called performance or use). In order to do this, linguists who are native speakers of the languages they are examining have typically used constructed sentences

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

25

for their data, consulting their own intuitions to decide whether those sentences are grammatical or not. The work by Chomsky and his followers represents this dominant tradition. However, our approach is instead to examine everyday talk, the fundamental form of language. More precisely, we perform close investigations of recorded talk and its transcripts. Our basic method is inductive in that we examine that recorded talk and transcripts in order to propose hypotheses about what speakers of the language may know. It should be mentioned, however, that in practice we often begin the investigation with re-examining previously discussed and/or assumed categories, rules, or phenomena. So the basic procedure has been, for example, to start with something like the following: “Let us see what ‘conditionals’ are like in everyday talk. Are they really used in everyday talk? If so, how are they used?” This has been a typical procedure mainly because it is not easy to begin a study completely from scratch.4 But this has produced results that question many of the categories and rules that have been previously assumed to exist. So by adopting this procedure, we have learned that we have to deal with traditional categories and assumptions very carefully, always testing them against what people actually do. As the reader can see, the approach described above involves observation and description. At this stage, much of its work is still in a hypothesis-building stage, a stage that it is necessary to go through before we can engage in further theorization. However, we hope it is clear from the discussion so far that the sort of research represented by this chapter is a type of theoretical morphosyntax that uses actual spoken language as its data, unlike the standard methodology of a more traditional approach to morphosyntax. It should also be pointed out that, unlike the traditional approach, much of this line of research has not been formalized. This may be an outcome of the sentiment shared by its practitioners that the amount of knowledge we have about everyday talk is still very limited and formal modeling is premature. Having discussed the general type of data we use and how we typically approach a new study, we should also address the issue of sampling. This is an ongoing challenge. There are some existing corpora, but these tend not to be suitable. The National Institute for Japanese Language has produced a spoken corpus, the Corpus of Spontaneous Japanese. The majority of what is included in that corpus is interviews, lectures, and read speech, so it can be used to investigate those particular types of language use, but it is not particularly helpful for investigations of more conversational Japanese. The Linguistic Data Consortium has also produced two corpora of Japanese telephone conversations, CALLHOME (phone calls between family members) and CALLFRIEND (phone calls between friends). These are good sources of data for the study of those particular types of everyday talk, although

SFK ch01

SFK ICBK027

26

April 20, 2008

10:24

Char Count=

Conversation and Grammar

the transcription for CALLFRIEND in particular is not always as reliable as might be desired. For children’s language use, resources include CHILDES (Child Language Data Exchange System). Simply put, the data available for the kind of work described above is still very limited. Individual researchers, including ourselves, typically base their studies on their personal collections of recordings and transcripts, so the sample size tends to be rather small. One of our goals as Japanese linguists is to contribute to an understanding of the nature of the language in general. This rather lofty goal, however, obviously requires a large enough set of data to be a good representative of what Japanese speakers do in their everyday interactions. At the time of writing this chapter, no such corpus exists. In terms of corpora, the problem of sample size needs to be kept in mind in any study that we undertake or read. Because of this, we feel that it is most important for researchers to strive to amass and share more data at every opportunity so that we will eventually be able to build a large corpus (or perhaps multiple smaller corpora, each representative of different types of talk) that will allow us to make general statements about the whole language with more confidence.5 In addition to the sample size, we also need to consider the composition of the sample. As mentioned above, different corpora contain different types of talk. Depending on what aspects of the Japanese language we hope to examine, we will need to look at different genres of talk that vary in aspects such as the purpose of talk and the formality of the talk. And if we hope to make generalizations about the Japanese language as a whole, we need to be sure that our sample represents a diverse range of speakers, such as speakers of different ages, genders, and regional dialects. This is an extremely difficult yet crucial question that we have to address at some point in our attempt to learn the nature of Japanese in general. Clearly, a large amount of collective research effort will be needed in order to address this issue. Some Standard Assumptions In this section, we would like to discuss some closely related assumptions that underlie most traditional understandings of grammar, but that we believe our data calls into question. We hope that our data will suggest new ways of understanding the nature of grammar and human language and that these new perspectives will prove fruitful to both linguistics and language pedagogy. These issues demand a much more extensive treatment than we are able to provide here. We would like to consider this brief discussion as a small step toward synthesizing these related issues with a larger goal of coming up with a model of human language reflecting

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

27

actual language use. We hope our brief treatment of these issues provides a foundation for further research. Modularity

This is the assumption (or hypothesis) found in much linguistic research from the past several decades, that human language consists of various discreet components, such as phonology or syntax, and thus, that it can be studied one component at a time. Thus many studies, particularly in morphosyntax, have been conducted by focusing on only one area, typically using information only from that area. However, a number of more recent studies have in fact questioned this long-held assumption. Linguists working from this alternative perspective have pointed out that many of the utterances observed in spoken language are actually more or less fixed. Thus, the lexicon and grammar might not be clearly separable as has been assumed. This suggests a different way of conceptualizing language (Langacker 1987, 1991; Erman and Warren 2000; Bybee and Hopper 2001; Bybee 2006). Based on this understanding, we will examine Japanese conditionals without limiting ourselves to looking only at grammatical evidence. Discreet categories and (binary) features

Another common assumption in linguistics is the idea that linguistic categories are discreet. So, for instance, part of speech categories such as “noun” and “verb” have typically been assumed to be grammatically distinct from each other. These sorts of categories have also been defined using discreet (very often binary) features. However, we do not yet know whether these assumptions are warranted, so we need to examine feature-based discreet categorizations in order to see if they actually fit with how human language operates. In fact, some research in the past several decades that has considered both grammar and meaning/function has suggested a rather different type of categorization in which categories have been shown to be defined based on prototypes and have “fuzzy edges,” with given forms fitting into the categories to a greater or lesser degree, based on how closely they resemble the prototypes for those categories. (See Rosch 1978, Lakoff 1977, Hopper and Thompson 1980, 1984; Langacker 1987, 1991.) That is, categories in language may instead be non-discreet and non-feature based. Interestingly, the non-discreetness of categories seems to be related to a fundamental characteristic of human language to which we now turn our attention: change. Synchronic grammar

Another assumption of convenience in linguistics is the idea that our goal is the description and representation of synchronic grammar—that is, of what

SFK ch01

SFK ICBK027

April 20, 2008

28

10:24

Char Count=

Conversation and Grammar

speakers of the language know at the present moment. In reality, of course, we know that everyday talk is full of variation that defies clear categorization. In particular, it includes examples in which category boundaries are fuzzy rather than discreet, or, perhaps more precisely, in which utterances look more or less as if they belong to different categories. An example of this would be Nakayama and Ichihashi-Nakayama’s (1997) work on the form kedo (‘but’), showing that kedo clauses often exhibit an intermediate status between subordinate clauses (in which kedo functions to link a subordinate clause and a main clause or to indicate a presumed main clause that is not overtly expressed) and main clauses (in which kedo functions more like a final particle). Adopting a diachronic approach to grammar allows us to have a better understanding of this situation: typical examples in everyday talk exhibit various degrees of categoriality and cannot easily be fitted into traditional categories, but that is a natural outcome of ongoing change in the language. We need a model that can handle this essential feature of human language, and some researchers have in fact attempted to address this issue (Langacker 1987, 1991; Hopper 1998). Single grammar

This is perhaps not an overtly stated assumption. Nevertheless, there is a tendency to conceptualize language in terms of a single system that is responsible for various different types of linguistic skills, including both spoken and written language. So, for instance, all of the various forms that are considered conditionals in Japanese are treated together, with the implication being that they belong to the same component of the grammar. Even if researchers find mode- or genre-specific uses of these forms, such as differences between spoken and written language, for instance, they tend to maintain the assumption that there is a single grammar, simply noting the modes or genres of language in which these forms are used. However, we could instead envision treating grammar as a phenomenon that is less unified than generally assumed. For instance, mode- or genre-specific variations could be due to having two or more relatively distinct grammatical systems. Shoichi Iwasaki has recently been working on a proposal for a multiple grammar hypothesis in order to grapple with this issue (e.g., Iwasaki 2006). Rules

Another common assumption about language is that utterances can be accounted for by a finite set of productive rules. Another possible scenario, however, is that utterances are often comprised of memorized expressions and/or variations that are newly created for the occasion, but that are nonetheless based on memorized expressions. Under this hypothesis, the role of rules might actually be much more limited than typically assumed,

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

29

and (semi-)fixed phrases such as idioms, set phrases, and collocations, and variations on those fixed phrases, might account for much of the linguistic activity that comprises everyday talk. A number of scholars have advocated this position (Pawley and Syder 1983; Erman and Warren 2000; Bybee and Hopper 2001; Bybee 2006; Wray 2005). Simplicity/economy

An assumption that is related to the last few points is that the adequacy of a proposed grammar should be judged based on economy and simplicity. That is, the simpler the proposed grammar and the more examples it can account for, the better. Here we would simply suggest that we need data that shows that human behaviors in general, and language in particular, are structured economically. Competing forms

Finally, it is common in both linguistics and language pedagogy to treat certain sets of linguistic forms as if they are forms that are related to each other, perhaps members of the same group, and then to try to explain why one form is used in a certain situation while another form is used in another situation. In Japanese, for example, linguists, language teachers, and students alike tend to consider the so-called topic marker wa and the so-called subject marker ga together and to attempt to distinguish between the contexts that are appropriate for the use of each particle. The same could be said of other forms as well, such as active and passive verbs, epistemic/evidential forms soo da, yoo da, mitai da, and rashii, and the conditional forms that we consider in the third section of this chapter. It is interesting that this attitude is observed even in some of our own research that examines discourse data, as in: Discourse-functional approaches to grammar have two goals. The first goal is a descriptive one: given the richness of the grammatical resources languages typically have for expressing the ‘same’ content, how do speakers choose among them? That is, what are the functions of the grammatical and lexical alternations of a language? We can ask, for instance, how speakers choose between a full noun phrase and a pronoun, or between two alternative orders for subject and verb. (Cumming and Ono 1997: 112) Against this commonly held view, Pawley and Syder (1983) demonstrate that only certain of the “competing” forms in English may be appropriate in a particular context and/or in a certain genre. Further, as found in the quoted passage above, competing forms are generally grouped together based on semantic (or propositional) similarities. However, we might legitimately

SFK ch01

SFK ICBK027

30

April 20, 2008

10:24

Char Count=

Conversation and Grammar

ask why we should necessarily start with semantics? Why not pragmatic similarities, for instance? If our goal is to capture units that are real to speakers of the language, it might make more sense to start with a set of forms that are used similarly in actual speech contexts.

Conditionals As a case study illustrating the type of approach described above, and as a way of examining whether the assumptions we have laid out are necessarily warranted, in this section we examine what might be considered a set of competing forms par excellence, forms that have been called conditionals in Japanese: tara, nara, ba, and to. Linguists have traditionally attempted to show the grammatical characteristics associated with each of these forms, to explain how they differ from each other semantically, and to delineate the conditions under which they can or cannot be used (e.g., Kuno 1973; Akatsuka 1985; Hasunuma 1987). In Japanese textbooks as well, we find that these forms are often introduced together and the characterization of each form is typically contrasted with other conditional forms (e.g., Tohsaku 1995b, Hatasa et al. 2000). Examples (1a) through (1e) illustrate how two or more of these forms have traditionally been examined together, suggesting that they are conceived of as a set in the grammar of Japanese.

(1a)

tomodachi ga dekireba/dekitara gakkoo mo friends S be-made-BA/TARA school also tanoshiku naru deshoo enjoyable become TENT ‘If/when s/he makes some friends, school will probably become more enjoyable.’ (Jacobsen 1992: 139)6

(1b)

asu Tokyo ni iku nara/ ∗ ittara, issho ni tomorrow to go NARA went-TARA together tsurete itte kudasai taking go please ‘Please take me with you if you are going to go to Tokyo tomorrow.’ (Kuno 1973: 177)

(1c)

sonna kurai tokoro de hon o yon-dara/yomu to/?yome-ba such dark place at book O read-TARA/TO/BA me o waruku-shimasu yo eye O bad do IP ‘If (you) read a book in such a dark place, (you) are going to make your eyes (go) bad.’ (Hasunuma 1987: 3)

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

(1d)

31

konshuu no doyoobi isogashiku-nakat-tara uchi this.week of Saturday busy -NEG -TARA house ni kimas-en ka to come-NEG Q ‘if you are not busy this Saturday, would you like to come to my house?’ (Hatasa et al. 2000: 214) konshuu no doyoobi isogashiku-nai nara uchi this.week of Saturday busy -NEG NARA house ni kimas-en ka to come-NEG Q ‘Since you’re not busy this Saturday, would you like to come to my house?’ (Hatasa et al. 2000: 214)

(1e)

Conditional (. . . to, ∼tara, ∼ba) ii/yokatta + noni It would be good if . . . , I wish it would happen that . . . , It would have been good if . . . , I wish it would have happened that . . . (Tohsaku 1995b:417)

In a more recent series of work, Akatsuka, Clancy, and Strauss examine conditionals in actual discourse data (e.g., Akatsuka 1997; Clancy et al. 1997; Akatsuka and Strauss 2000). They explore the meanings, discourse functions, acquisition, and history of these forms and some of their findings are clearly compatible with what we will discuss below. Akatsuka, Clancy, and Strauss, however, look at conditionals as the object of their research, considering tara, nara, ba, and to to constitute a set of conditional forms, and do not detail the frequency of these forms’ occurrence or explore what they might mean in relation to the representation of Japanese grammar as a whole. We have tried to build on their work by going on to consider the frequency with which these forms are used, the sorts of patterns we find in how they are used, and what light they shed on the nature of Japanese grammar and how we might represent it. In our chapter, we would like to change perspectives and take a corpusbased approach to tara, nara, ba, and to. Specifically, we use a small-scale corpus of everyday talk to examine aspects such as these forms’ frequency and productivity in order to gain an understanding of the grammar of socalled conditionals. Our corpus consists of 28 audio-recorded spontaneous informal conversations of mostly Standard Japanese (see the chapters in this volume by Okamoto (Chapter 5) and Kubota (Chapter 12) for discussion regarding “Standard Japanese”).7 The conversations total about 3 hours of talk and roughly 5,000 clauses. Each conversation involves between two and five participants who are family members, couples, and/or friends. The speakers range in age from approximately 15 to 65.

SFK ch01

SFK ICBK027

April 20, 2008

32

10:24

Char Count=

Conversation and Grammar Table 1.1 Conditional forms found in the corpus

tara

to

ba

nara

Total

166 48.0%

88 25.4%

84 24.3%

8 2.3%

346 100%

We will start our investigation by focusing on the tokens of tara, nara, ba, to, and related forms (e.g., kya) found in the corpus. Our approach is to examine how these forms are used in everyday talk in order to explore what Japanese speakers may know about these forms and how that knowledge may be organized. Frequency of conditional forms We found 346 tokens of tara, nara, ba, to and related forms in the corpus. As shown in Table 1.1, we further found that the distribution of these forms was highly skewed.8 In particular, nara is extremely rare;9 it seems that is not really ‘competing’ with the other forms in this sort of casual conversation. That is, forms that have typically been treated as a set by linguists may not actually be a set for the speakers of the language, or if they are somehow members of the same set, one of those members may be a very marginal member in conversational Japanese. Fixedness As shown in Table 1.2, when we looked at the utterances in which these forms were used, we found that more than half of the conditional forms in our data (52.5 percent) are associated with various degrees of fixedness. These examples include lexicalized expressions, idioms, set phrases, and collocations that do not seem to be produced based on regular syntactic rules. We call this type “(semi-)fixed conditionals,” as opposed to “rule-based conditionals,” which can be understood to be based on regular syntactic rules. The prevalence of (semi-)fixed conditionals raises the possibility that fixed linguistic expressions play at least as important a role as syntactic rules in everyday talk, suggesting that we need to pay far more attention to this type of language in future research.11 The (semi-)fixed conditionals are not simply a random set of expressions that are memorized by speakers. Instead, they are associated with various types of internal structure and fixedness and involve several subtypes. In the remainder of this section, we will go over some examples illustrating these subtypes, the frequency of which can be seen in Table 1.3.

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

33

Japanese Applied Linguistics Table 1.2 (Semi-)fixed and rule-based conditional

(Semi-)fixed conditionals

Rule-based conditionals

Total

181 52.5%

164 47.5%

34510 100%

Table 1.3 Subtypes of conditionals

Semi-fixed Conjunctions expressions

-ba positive/ -kya negative

Topic Suggestions marking Total

47 26.0%

51 28.2%

7 3.9%

34 18.8%

42 23.2%

181 100.1%12

Conjunctions and other (semi-)fixed expressions Quite a few instances of conditional forms (47 out of 346 tokens, or 13.6%) are found as part of lexicalized conjunctions, as illustrated in examples (2a) through (2d). Reduced versions of some of these forms are also found in our corpus, and these are presented together with the corresponding full form below. These reduced forms can be understood as reflecting further change in the language.

(2a)

soo-shi -tara soshitara, hoshitara, tara so -do -TARA ‘then’

(2b)

soo-suru-to sosuto so -do -TO ‘in that case/then’

(2c)

dat -tara COP-TARA ‘in that case/then’

(2d)

sore -nara that -NARA ‘then’

honnara

Some of these examples are listed in dictionaries,13 and their lexicalized status can be further seen in the fact that the conditional form found

SFK ch01

SFK ICBK027

34

April 20, 2008

10:24

Char Count=

Conversation and Grammar

in each example cannot be freely interchanged with other supposedly competing forms. That is, replacing tara and to in (2a) and (2b) with ba and nara, as in soosureba and soosurunara, for instance, does not produce conjunctions.14 One might suggest that although examples such as (2a)–(2d) may include conditional forms, they are not functioning as conditionals, so they should not enter into a discussion of conditionals. But that type of thinking prevents us from asking the question of why conditional forms commonly end up being lexicalized as conjunctions. The prevalence of this type of lexicalization in fact demands an explanation, particularly because it might reflect some general nature of Japanese or of human language. We need a theory of human language that can represent this window into diachronicity in synchronic data. There are also 34 cases of other (semi-)fixed expressions that involve conditional forms, as in examples (3a) through (3f).15

(3a)

(3b)

(3c)

(3d)

(3e)

(3f)

tatoe -ba compare-BA ‘For example’ soo-ie -ba so -say-BA ‘Now that you mention it’ hyotto shi -tara by.chance do -TARA ‘Maybe’ moshika shi -tara if do -TARA ‘Maybe’ yat -tara yamer-are -nai do -TARA stop -potential-NEG ‘Once you (start) do(ing) it, you can’t stop.’ Musashiya mo shira -nakere-ba Uomasa mo shira -nai Musashiya also know-NEG -BA Uomasa also know-NEG ‘If (I) also don’t know Musashiya, (I) also don’t know Uomasa/ (I) don’t know anything about either Musashiya or Uomasa.’

Similar to the case of the lexicalized conjunctions discussed above, some of these (semi-) fixed expressions are found in dictionaries. Further, replacing ba in sooieba in (3b) with to, tara, or nara would result in a different meaning; the expression would not mean ‘now that you mention it.’ Similarly, while tara in (3d) may be replaced with to as moshika suruto, the use of either ba

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

35

or nara in the same example does not seem to work as well: moshika sureba and moshika surunara sound odd. It should be noted that some of these expressions are not completely fixed. For instance, based on yattara yamerarenai (‘Once you (start) do(ing) it, you can’t stop’) in (3e), you can rather easily say tabetara yamerarenai (‘Once you (start) eat(ing), you can’t stop’), mitara yamerarenai (‘Once you (start) look(ing), you can’t stop’) or even tabedashitara yamerarenai (‘Once you start eating, you can’t stop’). That is, while parts of these expressions may be fixed, they also involve open slots in which various items are inserted. Bybee (2006) calls this type of unit a “construction” and highlights its centrality in actual discourse, suggesting that it should play a major role in our theorization of human language. ∼ba positive We also found 26 examples involving ba in utterances expressing some sort of positive outcome, as in examples (4a) through (4c).

(4a)

jibun no shigoto dake yatte-re -ba ii self of work only do -stative-BA good ‘It’s good if (I) do only (my) own work/(I) only need to do (my) own work.’

(4b)

kure -ba yokat-ta noni Kumi-chan come-BA good-past despite Kumi-chan ‘Even though it would have been good if (you) had come, Kumi-chan/(You) should have come, Kumi-chan.’

(4c)

Maabin ni itte oke -ba betsuni mondai Marvin to say in.advance-BA particularly problem wa nai TOP exist.NEG ‘There won’t be any particular problem, if (we) tell Marvin in advance.’

In these examples, the predicate of the main clause expresses a positive outcome (e.g., ‘good’ and ‘no particular problem’) resulting from the situation described in the conditional clause. We call these types of examples ‘∼ba positive.’ It is interesting that this use has been fixed to the extent that some examples are associated with a meaning that has further developed from its original meaning. So for example, (4b) actually means something more like ‘(you) should have . . .’ (the second translation), which appears to have derived from its more literal meaning ‘it would have been good if . . .’ (the first translation).

SFK ch01

SFK ICBK027

April 20, 2008

36

10:24

Char Count=

Conversation and Grammar

Example (4d) further illustrates the fixed status of ‘∼ba positive’ whereby the utterance is associated with a positive outcome even without the main clause overtly expressing it.

(4d) The speaker first says she might look for a boyfriend and then says ii hito ga ire -ba na good person S exist-BA IP ‘(It would be good) if (I) had a good guy/(I wish) (I) had a good guy.’ That is, such examples show that ‘∼ba positive’ has been established as a category in the mind of Japanese speakers to the extent that it is associated with a positive outcome even when the utterance does not explicitly state that positive outcome. ∼kya negative As a counterpart to ‘∼ba positive’, we also found 25 examples of ‘∼kya negative’, in that kya is a form of ba and is associated with a negative outcome, as in examples (5a) through (5d).16 Similar examples have been extensively discussed in Akatsuka (1997), Akatsuka and Strauss (2000), and Clancy et al. (1997).

(5a)

ogora-na -kya ikenai deshoo treat -NEG-BA bad TENT ‘It would be bad if (you) don’t pay (your girlfriend’s way)/(You) must treat (your girlfriend).’

(5b)

gyooseki age -na -kya dame da na results increase-NEG-BA bad COP IP ‘It’s bad if (you) don’t come up with results/(You) must come up with results.’

(5c)

chanto shi-na -kya iya properly do-NEG-BA bad ‘It’s bad if (I) don’t do (it) right/(I) have to do (it) right.’

(5d)

mukoo ni akuseputo sare -nake-rya over.there in accept do.passive-NEG-BA imi ga nai meaning S exist.NEG ‘It’s meaningless if (your work) isn’t accepted in the West.’

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

37

The main clause in each of these examples expresses a negative outcome (e.g., ‘bad’ and ‘meaningless’) resulting from the situation described in the conditional clause. Similar to ‘∼ba positive’, ‘∼kya negative’ has been fixed to the extent that some examples are associated with a meaning that has further developed from its original meaning. So, for example, a better translation for (5a) is ‘(you) must . . .’, which appears to have derived from its more literal meaning ‘it would be bad if . . .’. Also as with ‘∼ba positive’, with ‘∼kya negative’ we again find examples in which the utterance indicates a negative outcome even though it lacks a main clause that overtly expresses that outcome:

(5e)

henji hayaku dasa-na -kya reply quickly send-NEG-BA ‘(I) have to send (her) a reply right away.’

As we saw earlier regarding ‘∼ba positive’, examples such as these are good evidence that ‘∼kya negative’ has been (semi-)fixed and established as a category in the minds of Japanese speakers. It seems reasonable to speculate that examples such as (4d) and (5e) derive from the frequent association between a particular form (ba or kya) with a particular meaning (positive or negative). That is, the frequent use of ‘∼ba positive’ and ‘∼kya negative’ may have resulted in a situation in which a particular meaning is evoked in the mind of speakers even though only the first clause is explicitly expressed.

Suggestions We also found seven other examples involving only the conditional clause that express a suggestion, as in (6a) through (6c):17

(6a)

ki -tara come-TARA ‘Why don’t you come (over)?’

(6b)

hokoten itte mire-ba pedestrian.paradise go try -BA ‘Why don’t you try going to a “pedestrian paradise” (an area where the street has been blocked off for pedestrians)?’

(6c)

higoro no koodoo o ne moo chotto everyday of behavior O IP emphatic a.little

SFK ch01

SFK ICBK027

April 20, 2008

38

10:24

Char Count=

Conversation and Grammar

jimini suru to restrained do TO ‘(You should) make (your) everyday demeanor a little more restrained.’ Similar to the ‘∼ba positive’ (4d) and ‘∼kya negative’ (5e) examples that we saw above, the conditional clause is used alone in examples (6a) through (6c). The situation described in the clause is understood as a suggestion even though the suggestion is not made overtly, which suggests again that this type of suggestion is a (semi-)fixed expression and constitutes a category for speakers. Topic marking The connection between conditionals and topic marking has been noted by previous researchers as well (Haiman 1978; Akatsuka 1986; Jacobsen 1992).18 We found 42 examples in which the conditional clause is functioning similarly to wa and tte, both of which can function as topic markers, as in:

(7a)

sore da to takai that COP TO expensive ‘That’s expensive.’

sore wa takai that TOP expensive ‘That’s expensive.’ (constructed)

In this example, the conditional clause involving a copula serves a function similar to topic marking, as demonstrated by the constructed example with the topic marker wa, given in the right column. A more literal translation of (7a) may be ‘If (it) is that, (it’s) expensive’. It seems reasonable to suggest that the conditional clause ‘if (it) is that’ has been re-analyzed as a way to indicate a topic, as in ‘That, (it’s) expensive/That’s expensive.’ Example (7b) is another topic-marking example of a conditional involving a copula.

(7b)

hokoten dat -tara bando mo yatte-ru shi pedestrian.paradise COP-TARA band also do -stative and ‘As for “pedestrian paradises”, bands are also playing and . . .’

Other predicates commonly found in topic-marking examples of conditionals are the verbs iu (‘say’) and naru (‘become’), as in (7c) and (7d).

(7c)

nengu tte it -tara okome da yo tax QT say-TARA rice COP IP ‘What you call “land tax” is rice.’ (Land taxes were paid in rice.)

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

(7d)

39

kuruma toka naru to sutereo to onaji de car like become TO stereo with same and dakara mania tte yuu no ga iru deshoo so maniac QT say nominalizer S exist TENT ‘When it comes to things like cars,/As for cars, like stereos, there are those called “maniac.”’ (Cars, like stereos, have people who are really crazy about them.)

The connection between similar English verbs and topic marking may be seen in expressions such as ‘speaking of . . .’ and ‘when it comes to . . .’ which are used for a similar function. It should be also noted that most topic-marking uses of conditionals involve either tara or to (39 out of 42 examples), most commonly occurring with the predicates mentioned above: a copula or the verbs iu (‘say’) or naru (‘become’). We have thus examined a number of (semi-)fixed expressions involving so-called conditional forms. They include several different subtypes associated with different types and degrees of fixedness. The sheer number of such examples suggests that studying conditional forms only from a grammatical-rules perspective will miss a great deal of how speakers use these forms. Rule-based uses As was shown in Table 1.2, close to half (47.5 percent) of the conditional forms used in our data can be understood based on rules, and these we will term “rule-based conditionals.” (8a) and (8b) are examples of this type.

(8a)

demo Amerika ni i -chau to tsuyoku natte kuru but America in be-end.up TO strong become come ‘But if (Japanese women) end up staying in America, (they) become strong.’

(8b)

chotto kii -ta konaida Yoosuke to hanashi-tara a.little hear-past the.other.day Yoosuke with speak -TARA ‘(I) heard a little, when (I) spoke with Yoosuke the other day.’

Our examination of rule-based conditionals has also revealed several very interesting findings. First, it is known that, cross-linguistically, so-called conditional forms often denote a temporal meaning as in (8b), and in fact, as shown in Table 1.4, we found that rule-based conditionals in our data were actually more frequently associated with temporal meanings, making the term ‘conditional’ seem somewhat of a misnomer.

SFK ch01

SFK ICBK027

April 20, 2008

40

10:24

Char Count=

Conversation and Grammar Table 1.4 Rule-based conditionals and temporal or conditional meanings

Temporal

Conditional

Temporal/ Conditional 19

Total

89 54.3%

68 41.5%

7 4.3%

164 100.1%

Table 1.5 Rule-based conditionals: Frequency of tara and to

tara

to

ba

nara

Total

90 54.9%

56 34.1%

16 9.8%

2 1.2%

164 100%

Second, our data reveals that the overwhelming majority (89 percent) of rule-based conditionals involve either tara or to, as seen in Table 1.5. As the table shows, compared to tara and to, ba is not common, and nara is extremely rare. It seems that if any of these forms should be considered as ‘competing’ forms in the grammar of Japanese speakers, the competition is between tara and to. We saw in Table 1.1 that ba is used almost as frequently as to in our data, at 24.3 percent and 25.4 percent of the total, respectively. However, Table 1.5 shows that the rule-based use of ba is not common. This demonstrates that ba is used mostly in the (semi-)fixed expressions that we saw in the last section. Third, it is customary that Japanese conditionals are illustrated in the literature with examples in which the conditional clause is followed by the main clause as seen in (8a) above and in examples (9a) through (9d):

(9a)

watashi ga Hayashi-san dat -tara, Gibson-san ni I S Hayashi-san COP-TARA Gibson-san to puropoozu shimasu propose do ‘If I were Mr. Hayashi, I would propose to Ms. Gibson.’ (Tohsaku 1995a: 446)

(9b)

anata ga kuruma de ike-ba, watashi mo kuruma de iku you S car by go-BA I also car by go ‘If you go by car, I will, too.’ (Jorden with Noda 1990: 93)

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

41

Japanese Applied Linguistics Table 1.6 Conditional + main and main + conditional

Conditional + main

Main + conditional

Other

Total

29 17.7%

3 1.8%

132 80.5%

164 100%

(9c)

yasui (no) nara kau wa yo cheap (nominalizer) NARA buy IP IP ‘If they’re cheap, I will buy (some).’ (Hatasa et al. 2000: 213)

(9d)

sono kissaten ni iku to, Chin -san ga i -ta that coffee.shop to go TO Chin-san S be-past ‘When I went to the coffee shop, Ms. Chin was there.’ (Tohsaku 1995b: 47)

Interestingly, as Table 1.6 shows, such a configuration is actually very rare in our data. Even including examples in which the conditional clause follows the main clause, as in (8b), examples consisting simply of two clauses actually account for less than 20 percent of the rule-based conditionals in our data. Instead, more than 80 percent of the rule-based conditionals occur in a string of clauses produced within speakers’ currently unfolding turns and/or embedded in a larger utterance, often as quotes, as in (10a) and (10b).

(10a) 1 A: Ai-chan shitteru tte iwa-re -[te] Ai-chan know QT say-passive-and ‘“Ai, do (you) know (about it)?” (I) was asked,’ 2

K:

[un] ‘mhm’

3

A: nani toka it -tara what like say-TARA ‘when (I) said like “what?”,’

4

K: un ‘mhm’

5

A: atashi wa jitsu wa kinoo kii -te ne I TOP actually TOP yesterday hear-and IP ‘I actually heard yesterday and’

SFK ch01

SFK ICBK027

April 20, 2008

42

10:24

Char Count=

Conversation and Grammar

6

sugoi bikkuri shi-chat -ta n da awfully surprised do-end.up-past nominalizer COP

7

kedo toka i -tte but like say-and ‘was really surprised but,” (she) said and’

In example (10a), the tara clause occurs with three clauses marked with te. Example (10b) involves to.

(10b) 1 K: n demo kochakocha aru to mhm but bits.and.pieces exist TO ‘Well but if (I) have (classes) split up at various times (throughout the week)’ 2

moo motto ippai shi-na -kya ikenai kara emphatic more a.lot do-not-BA bad so ‘(I) have to do a lot more so’

3

sonobun is -shuu-kan ni ik -kai toka da kara that.degree one-week-period in one-time like COP so ‘instead it’s like once a week so’

4

M: n a mhm oh ‘mhm, oh!’

5

K: naga[i deshoo] long TENT ‘(it) is long.’

6

M:

7

K:

[kurasu ga is -shuu]-kan ni ik[-kai tte] class S one-week -period in one-time QT ‘(You’re saying you have) class once a week.’ [n ]n ‘mhm mhm’

In this example, the to- clause appears with clauses marked with kara. As shown in (10a) and (10b), the rule-based conditional clauses with tara and to are typically found in a sequence of clauses marked with such forms as te, kara, and kedo. This type of sequence of multiple clauses is called ‘clause chaining’ and is found cross-linguistically, especially in

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

43

predicate-final languages like Japanese, in which a number of different forms are used to chain clauses (Haiman and Munro 1983). Clause chaining is commonly observed in Japanese everyday talk and seems to be employed when the speaker continues talking while keeping the current turn. The prevalence of examples such as (10a) and (10b) above suggest that tara and to should be characterized primarily as part of a set of clause-chaining devices used to maintain talk in spontaneous speech (Ono and Iwasaki 2002; Iwasaki and Ono 2007).

Conclusion We have seen that the various Japanese forms traditionally known as conditionals are highly skewed as far as how frequently they occur. In addition, they are more often used in (semi-)fixed expressions than has been recognized previously and thus are less rule-oriented than has been assumed in the past. Finally, the more rule-oriented or “grammatical” uses of conditionals are less “sentence-oriented” than previously assumed, occurring only infrequently in the two-clause (conditional clause plus main clause) sentences that have been assumed to be the canonical conditional structures. Our findings lead us to question whether language is best understood as being modular and comprised of forms that can be divided into discrete categories, and to question the degree to which everyday talk is carried out by speakers’ relying on productive rules, forming clauses while choosing between competing forms. So-called conditional forms tara, nara, ba, and to, which might be thought of as quintessential competing forms, do not seem to form a set for speakers. The actual use of these forms can be characterized as much more lexical than has been assumed previously, and there are no clear boundaries dividing more grammatical and productive uses from semi-fixed expressions and completely lexicalized conjunctions. We have thus seen that in order to account for the behavior of so-called conditional forms in Japanese, we may need to reconsider some of the traditional assumptions about language that we outlined earlier in this chapter with new understandings that are suggested by the behavior of actual speakers. To the extent that our findings may be surprising, they underscore the importance of examining so-called performance data in our attempts to represent speakers’ knowledge. An examination of everyday talk is the first step we must take if our goal is to understand the nature of human language. Such an examination is also crucial if our goal is to understand the details of how particular forms from specific languages are used—an understanding that is vital for language teachers who hope to impart to their students

SFK ch01

SFK ICBK027

44

April 20, 2008

10:24

Char Count=

Conversation and Grammar

an accurate picture of how a language is used in a myriad of real-world contexts. Whether native speakers or nonnative speakers of a language, our assumptions about the language are likely to be inaccurate if not based on the observation and analysis of actual language use. What then are the implications of our study for second language researchers and teachers? As far as the Japanese conditional forms in particular are concerned, we would argue that for many types of use, teachers should not take a rule-based perspective, but rather, should introduce specific fixed uses or common patterns without referring to other conditional forms that a traditional perspective might view as possible in the same context. Considering the frequency of the various (semi-)fixed uses, as shown in Table 1.3, will give us an idea of which are the most important to introduce for teaching informal spoken Japanese. Work based on a larger corpus is needed to verify these frequencies, of course, and there is also a need for similar work based on corpora of more formal spoken Japanese and of written Japanese. This study, at any rate, suggests that if we want to enhance our students’ ability to engage in everyday conversation, we should introduce them to, and have them practice using, the most common conjunctions, the conditional phrases most often used for the topic-marking function, and the ‘∼ba positive’ and ‘∼kya negative’ patterns. We should also take a look at the category of other semi-fixed expressions to see what uses are common and consider teaching those as well. Teaching these few uses would insure that our students could produce examples similar to the overwhelming majority of the (semi-)fixed uses of conditional forms found in our data. We should note that there are cases in which we might want to introduce patterns that are not so frequent in actual interaction. For example, suggestions were not very common among the (semi-)fixed expressions in our data. This may be because making suggestions to others is potentially problematic in social interactions. That is not to say that we should not teach our students these ways of making suggestions, however. For one thing, as relative novices in the target culture, they may be likely to be the target of well-meaning suggestions about how they should behave, and so it is important for them to recognize that conditional forms can function in this way. When we do teach this use of conditionals, though, we should caution students about the potential social pitfalls of making suggestions to others. In teaching more rule-governed uses of conditionals, which comprised 47.5 percent of the occurrences in our study, we might introduce tara, to, and ba, but then focus classroom language practice on the use of tara and to, since they make up all but a small percentage of such conditionals. Given a limited amount of instructional time, it is probably less important to focus on the production of rule-governed ∼ba conditional phrases, which are rare in everyday talk.

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

45

Other less common uses of conditional forms, whether rule-governed or (semi-)fixed, can be taught as they occur. For example, the meaning of nara can be taught whenever it is naturally encountered, whether that be in spoken or written Japanese. In most cases, there is probably no reason to have students practice producing it. If they need to produce more formal or spoken Japanese or written Japanese, and if indeed an examination of corpora of those sorts of language shows that nara is commonly used in those contexts, students could then be encouraged to use the form as appropriate. Finally, in addition to using frequency of occurrence in everyday talk to guide our choice of forms to present to our students and have them practice, we should also consider the structures in which those forms typically occur. This study suggests that the traditional view of conditional forms as occurring primarily in two-clause sentences does not reflect actual use. Students will benefit from being exposed to examples of typical Japanese talk so that they can learn how these forms typically function as clause-chaining devices in naturally occurring talk. This study of Japanese conditional forms is merely one example of how a discourse functional approach can be helpful to applied linguistics, but the implications are far-reaching. For every form or “set” of forms that we look at from this perspective, we are sure to find ways to improve both our overall understanding of language and specific aspects of language pedagogy. Similar areas in which we see potential applications for this approach to the Japanese language include (but are definitely not limited to!) the use of various particles, such as wa and ga; active versus passive voice of verbs; different types of nominal reference, such as full noun phrase, pronoun, or no explicit reference at all; different word order types; the nominalizers koto, mono, and no;20 and the epistemic/evidential forms soo da, yoo da, mitai da, and rashii. Observing how these are used in actual interactions and then applying those observations has the potential to make our language teaching both more accurate and more efficient. At every level of language, from more global beliefs about what is appropriate to say and how to say it, to our assumptions about how sentences and discourses are typically structured, to traditional beliefs about grammar and vocabulary, ideologies about language have the potential to blind us to its real nature. Looking at actual language use with an open mind can help us overcome some of those blind spots.

Notes 1. We would like to thank the editors of this volume, Junko Mori and Amy Snyder Ohta, for their patience and help at each stage of writing

SFK ch01

SFK ICBK027

46

2.

3.

4.

5.

6.

7. 8. 9.

April 20, 2008

10:24

Char Count=

Conversation and Grammar and revising this chapter. Although we alone are responsible for any remaining shortcomings in the chapter, their editorial suggestions were invaluable in helping us frame the chapter for readers from the field of applied linguistics. The first author would also like to thank Shoichi Iwasaki, Ritva Laury, and Sandy Thompson for discussing with him some of the issues addressed in the chapter and inspiring him to take on the current project. Please note that the literature on Japanese inspired by this alternative tradition is found starting in Kuno (1973) and Shibatani (1990), and more recently, for instance, in many of the papers published in the Japanese/Korean linguistics volumes published by CSLI. We originally presented an analysis of this data at the Third International Conference on Practical Linguistics of Japanese (ICPLJ3), held at San Francisco State University in March, 2002, and at the 1st Second Language Acquisition and Teaching Interdisciplinary Roundtable, held at the University of Arizona in April, 2002. We then published a Japanese version of the analysis in a volume of papers from the ICPLJ conference, Linguistics and Japanese Language Education IV (Ono and Jones 2005). We wish to thank participants at both of those conferences, the anonymous reviewers of the paper, and the editor of Linguistics and Japanese Language Education IV , Masaahiko Minami, for their helpful comments on that paper. Fresh examinations of data made without taking traditional categories and earlier findings into consideration (unless those findings are based on data from spoken language), and made without taking a priori theoretical orientations, might actually be just what is needed for the approach we are advocating here. As an effort to address the sample size problem, Ono and several collaborators are currently working to develop a large-scale corpus of audio- and video-recordings and transcriptions of everyday Japanese interactions. Throughout the chapter, we use the Hepburn system of romanization. Examples cited from published work that uses the Kunrei system have been altered to correspond to Hepburn romanization. See the appendix for abbreviations used in glosses. We modified some of the abbreviations used in the examples cited from previously published work for the sake of consistency. There were occasional uses of non-standard Japanese in the corpus. Variant forms (e.g., kya in chanto shinakya iya (‘It’s bad if (I) don’t do (it) right.’)) are grouped together with their base form (ba) in this table. A similar skewed distribution is reported in Clancy et al. (1997: 26).

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

47

10. One of the original 346 conditionals could not be coded, as the surrounding material was inaudible. 11. Hayes and Shinzato (2001), which discusses the grammaticized use of tara, is a good example of this sort of work. 12. Total not equal to 100 percent due to rounding. 13. Determination of fixedness is difficult partly because we are dealing with the degree of fixedness mirroring ongoing change. Further, most dictionaries only list lexicalized forms used in written language, tending not to deal with forms found only in spoken language. We chose to be conservative in our determination of fixedness by selecting only relatively clear cases. Further investigation of this topic is needed. 14. Soosureba might sound good to some speakers. We think that it is because of its frequent use in another (semi-)fixed expression discussed later in this section (i.e., ∼ba positive). In any case, the point here is that, unlike the lexicalized conjunctions sooshitara and soosuruto, soosureba and soosurunara do not function as conjuctions. 15. Other examples of this type include: dotchikatte iuto, kyokutanni iuto, ikinari iwareruto, soo iwareruto, soo iwarereba, soo iwarete mireba, dekirunara, moshi yokattara, nainara naitte iyaa ii jan. 16. Kya is said to have been derived by going through the following steps. All of these forms are still used in present-day Japanese: nakere-ba → nakerya → nakya NEG -BA 17. Similar examples are discussed in Hayes and Shinzato (2001). 18. Using constructed data, however, Jacobsen (1992) argues that there is no connection between conditionals and topic marking. 19. It was not possible to tell whether a temporal or a conditional meaning was intended in these examples. 20. For example, Maynard’s (1997) study of koto and no would be helpful to Japanese language teachers who are trying to help their students learn to use these forms. In addition, the Gengogaku to Nihongo Kyooiku [Linguistics and Japanese Language Education] series published by Kuroshio contains many relevant articles.

References Akatsuka, N. (1985) “Conditionals and epistemic scale.” Language 61, 625– 639. Akatsuka, N. (1986) “Conditionals are discourse-bound.” In E. C. Traugott, A. ter Meulen, J. S. Reilly, and C. A. Ferguson (eds.), On Conditionals. Cambridge: Cambridge University Press, pp. 333– 351.

SFK ch01

SFK ICBK027

48

April 20, 2008

10:24

Char Count=

Conversation and Grammar

Akatsuka, N. (1997) “Negative conditionality, subjectification, and conditional reasoning.” In A. Athanasiadou and R. Dirven (eds.), On Conditionals Again. Amsterdam: John Benjamins, pp. 323– 354. Akatsuka, N. M. and Strauss, S. (2000) “Counterfactual reasoning and desirability.” In E. Couper-Kuhlen and B. Kortmann (eds.), Cause-conditionconcession-contrast: Cognitive and Discourse Perspectives. Berlin: Mouton, pp. 205–234. Bybee, J. (2006) “From usage to grammar: The mind’s response to repetition.” Language 82(4), 711–733. Bybee, J. and Hopper, P. (2001) Frequency and the Emergence of Linguistic Structure. Amsterdam: John Benjamins. Chafe, W. (1987) “Cognitive constraints on information flow.” In R. Tomlin (ed.), Coherence and Grounding in Discourse. Amsterdam: John Benjamins, pp. 21–51. Chafe, W. (1994) Discourse, Consciousness, and Time: The Flow and Displacement of Conscious Experience in Speaking and Writing. Chicago: University of Chicago Press. Clancy, P. M., Akatsuka, N., and Strauss, S. (1997) “Deontic modality and conditionality in discourse: A cross-linguistic study of adult speech to young children.” In A. Kamio (ed.), Directions in Functional Linguistics. Amsterdam: John Benjamins, pp. 19–57. Cumming, S. and Ono, T. (1997) “Discourse and grammar.” In T. A. van Dijk (ed.), Discourse as Structure and Process. London: Sage, pp. 112– 137. Erman, Br. and Warren, B. (2000) “The idiom principle and the open choice principle.” Text 20, 29–62. ´ T. (1979) On Understanding Grammar. New York: Academic Press. Givon, Haiman, J. (1978) “Conditionals are topics.” Language 54, 564–589. Haiman, J. and Munro, P. (eds.) (1983) Switch-Reference and Universal Grammar. Amsterdam: John Benjamins. Hasunuma, A. (1987) “Jookenbun ni okeru nichijooteki suiron: ‘tewa’ to ‘ba’ no sentaku yooin o megutte [Common inferences in conditional clause: On the factors behind the choice between ‘tewa’ and ‘ba’].” Kokugogaku 150,1–14. Hatasa, Y. A., Hatasa, K., and Makino, S. (2000) Nakama 2: Japanese Communication, Culture, Context. Boston: Houghton Mifflin. Hayes, S. T. and Shinzato, R. (2001) “Jooken no setsuzokujoshi kara danwa(taijinkinoo no joshi e: tara, ttara no bunpooka [From conditional conjunctive particles to discourse(interactional particles: The grammaticization of tara(ttara].” In M. Minami and Y. S. Alam (eds.), Gengogaku to Nihongokyooiku II: New Directions in Applied Linguistics of

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

49

Japanese. [Linguistics and Japanese Language Education II]. Tokyo: Kuroshio, pp. 127–142. Hopper, P. J. (1998) “Emergent grammar.” In M. Tomasello (ed.), The New Psychology of Language: Cognitive and Functional Approaches to Language Structure. Mahwah, NJ: Lawrence Erlbaum, pp. 155–175. Hopper, P. J. and Thompson, S. A. (1980) “Transitivity in grammar and discourse.” Language 56(2), 251–299. Hopper, P. J. and Thompson, S. A. (1984) “The discourse basis for lexical categories in universal grammar.” Language 60(4), 703–752. Iwasaki, S. (2006) “The ‘multiple-grammar’ hypothesis: Case studies of Japanese and English.” Invited talk given at the Association of English Linguistics meeting. Tokyo University, November 4–5, 2006. Iwasaki, S. and Ono, T. (2007) “‘Sokujibun’ to ‘hisokujibun’: Gengogaku no hoohooron to kiseigainen [‘Spontaneous sentence’ and ‘processed sentence’: Methodology and categories in linguistics].” In S. Kushida, T. Sadanobu, and Y. Den (eds.), Bun to Hatsuwa [Sentence and Utterance]. Tokyo: Hituzi Shob¯o, pp. 135–157. Jacobsen, W. M. (1992). “Are conditionals topics? The Japanese case.” In D. Brentari, G. N. Larson, and L. A. Macleod (eds.), The Joy of Grammar: A Festschrift in Honor of James D. McCawley. Amsterdam: John Benjamins, pp. 131–160. Jones, K. (1990) “Conflict in Japanese conversation.” Doctoral dissertation, University of Michigan. Jones, K. (1993) “Nihongo no konfurikuto-ji no hanashiai [Negotiating conflicts in Japanese].” Nihongogaku 12(4), 68–74. Jones, K. (1995) “Masked negotiation in a Japanese work setting.” In A. Firth (ed.), The Discourse of Negotiation: Studies of Language in the Workplace. Oxford: Pergamon Press, pp. 141–158. Jones, K. and Ono, T. (2005) “Discourse-centered approaches to Japanese language pedagogy.” Japanese Language and Literature 39, 237– 254. Jorden, E. H. with Noda, M. (1990) Japanese: The Spoken Language, Part 3. New Haven, CT: Yale University Press. Kuno, S. (1973) The Structure of the Japanese Language. Cambridge, MA: The MIT Press. Lakoff, G. (1977) “Linguistic gestalts.” Proceedings of the Annual Meeting of the Chicago Linguistic Society 13, 236–87. Langacker, R. W. (1987) Foundations of Cognitive Grammar. Vol. I: Theoretical Prerequisites. Stanford: Stanford University Press. Langacker, R. W. (1991) Foundations of Cognitive Grammar. Vol. II: Descriptive Applications. Stanford: Stanford University Press.

SFK ch01

SFK ICBK027

50

April 20, 2008

10:24

Char Count=

Conversation and Grammar

Maynard, S. K. (1997) “Shifting contexts: The sociolinguistic significance of nominalization in Japanese television news.” Language in Society 26, 381–399. Mori, J. (2005) “Why not why? The teaching of grammar, discourse, and sociolinguistic and cross-cultural perspectives.” Japanese Language and Literature 39, 255–289. Nakayama, T. and Ichihashi-Nakayama, K. (1997) “Japanese kedo: Discourse genre and grammaticization.” In H. Sohn and J. Haig (eds.), Japanese(Korean Linguistics 6. Stanford, CA: CSLI Publications, pp. 607– 618. Ochs, E., Schegloff, E. A., and Thompson, S. A. (eds.) (1996) Interaction and Grammar . Cambridge: Cambridge University Press. Ono, T. and Iwasaki, S. (2002) “Toward an understanding of ‘sentence’ in spoken Japanese discourse: Clause-combining and online mechanisms.” In K. Kataoka and S. Ide (eds.), Bunka, Intaa-akushon, Gengo [Culture, Interaction, and Language]. Tokyo: Hituzi Shob¯o, pp. 103–131. Ono, T. and Jones, K. (2001) “Kaiwa ni okeru ninchiteki sokumen to washakan no soogo sayoo: Nihongo kyooiku e no teian” [The cognitively based and interactionally situated nature of talk: Suggestions to Japanese language pedagogy]. In M. Minami and Y. S. Alam (eds.), Gengogaku to Nihongo Kyooiku II: New Directions in Applied Linguistics of Japanese [Linguistics and Japanese Language Education II]. Tokyo: Kuroshio, pp. 181–196. Ono, T. and Jones, K. (2005) “Bunpoo kisoku no shiyoo to keishiki no sentaku no jissai: Kaiwa ni okeru “jookenbun” kansatsu kara” [Grammatical rules and competing forms: conditionals in Japanese conversation.] In M. Minami (ed.), Gengogaku to Nihongo Kyooiku IV [Linguistics and Japanese Language Education IV]. Tokyo: Kuroshio, pp. 73–85. Pawley, A. and Syder, F. H. (1983) “Two puzzles for linguistic theory: Nativelike selection and nativelike fluency.” In J. C. Richards and R.W. Schmidt (eds.), Language and Communication. London: Longman, pp. 189– 226. Rosch, E. (1978) “Principles of categorization.” In E. Rosch and B. B. Lloyd (eds.), Cognition and Categorization. Hillsdale, NJ: Erlbaum, pp. 27– 47. Shibatani, M. (1990) The Languages of Japan. Cambridge: Cambridge University Press. Tohsaku, Y. (1995a) Yookoso! An Invitation to Contemporary Japanese. New York: McGraw-Hill. Tohsaku, Y. (1995b) Yookoso! Continuing with Contemporary Japanese. New York: McGraw-Hill.

SFK ch01

SFK ICBK027

April 20, 2008

10:24

Char Count=

Japanese Applied Linguistics

51

Tomasello, M. (ed.) (1998) The New Psychology of Language: Cognitive and Functional Approaches to Language Structure Volume 1. Mahwah, NJ: Lawrence Erlbaum. Tomasello, M. (ed.) (2003) The New Psychology of Language: Cognitive and Functional Approaches to Language Structure Volume 2. Mahwah, NJ: Lawrence Erlbaum. Wray, A. (2005) Formulaic Language and the Lexicon. Cambridge: Cambridge University Press.

SFK ch02

SFK ICBK027

April 21, 2008

17:6

Char Count=

Chapter 2

Negotiating Agreement and Disagreement in Japanese: An Analysis of Designedly Ambiguous Turn Completion Points

Junko Mori and Kanae Nakamura

Introduction1 A common theme advanced by comparative studies of Japanese and American communication styles and strategies can be summed up as “Americans are independent and individualistic, and use explicit language, while Japanese are interdependent and group-oriented, and use implicit language” according to Laura Miller (1999: 138). A number of pragmatic studies, in particular those conducted in the 1970s through the 1990s, compared Japanese and American English speakers and produced these portrayals of two distinct cultural groups. The ways in which speakers negotiate their opinions and stances have been one of the practices examined in such studies (e.g., Lebra 1976; Watanabe 1993; Yamada 1997). These studies tend to start with the assumption that the speakers’ national or cultural affiliation is the most critical factor for their behaviors. Consequently, their findings have often reaffirmed the images created by the ideological tradition of nihonjin-ron (or theory of Japanese uniqueness) (Befu 2001), which has also produced controversial influences upon Japanese as a foreign or second language education (Heinrich 2005; see Kubota, Chapter 12 in this volume for further details) as well. Critical voices such as Miller’s that question the essentialization of cultural differences have become more pronounced since the mid 1990s (e.g., Higgins 2007; Meeuwis 1994; Miller 1999; Moerman 1996; Rose 1996). These critics have advocated the necessity of undertaking systematic

SFK ch02

SFK ICBK027

April 21, 2008

17:6

Char Count=

Japanese Applied Linguistics

53

analyses of dynamic interactional processes rather than hastily linking what is observed in the data with pre-existing ideological or cultural assumptions. Studies such as Jones (1990), Miller (1994a, 1994b, 2000), and Saft (2001, 2004), for instance, demonstrate how the styles commonly thought of as “typically Japanese” can be seen as a myth that does not explain the diversity of various individuals performing different actions in different situations. These researchers emphasize the importance of detailed descriptions of situated interactional practices while discouraging the use of ideological and cultural beliefs as sources of explanations for the observed practices; they instead strive to uncover what exact observable phenomena might have contributed to the establishment of such beliefs as outcomes. To illustrate this viewpoint, the current chapter examines excerpts of talk-in-interaction in which Japanese speakers negotiate agreement and disagreement when stating their opinions and stances; it explicates how these participants carefully attend to their co-participants’ verbal and non-verbal behaviors and incorporate the results of their observations into the ongoing construction of their own talk. This study also situates itself in a growing number of conversation analytic studies that explore the linguistic structures of Japanese as resources for conducting particular actions in interaction (Hayashi 2001, 2003, 2004a, 2004b, 2005; Hayashi and Mori 1998; Hayashi et al. 2002; Iwasaki 2007; Lerner and Takagi 1999; Mori 1999, 2006; Nakamura 2005; Tanaka 1999, 2000, 2001a, 2001b, 2005, among others). The sociological methodology of conversation analysis (CA), which has inspired and guided these studies, was initially developed based on studies of American or British English conversations. By applying CA to Japanese data, these studies have attested that the fundamental organizational features of interactions explicated by previous CA studies appear to hold in Japanese as well; at the same time, findings show how linguistic resources specific to Japanese may alter the subtle ways in which turn-taking and other interactional moves are realized. That is, these studies consider how what has been described as communication styles and strategies specific to a particular cultural group may be linked to the linguistic resources available for the speakers of the particular language. Following these studies, the current study investigates the intricate relationship between the linguistic structures of Japanese and the opportunities for inspecting emerging alignment, or the lack thereof, which these linguistic resources make available for the speakers. More specifically, it illustrates how Japanese speakers utilize the predicate-final sentence structure of Japanese and some linguistic items suitable for making the completion of their turns at talk designedly ambiguous, when monitoring the co-participants’ incipient reactions and pursuing their agreement.

SFK ch02

SFK ICBK027

54

April 21, 2008

17:6

Char Count=

Negotiating Agreement and Disagreement

In the following, we will first review some of the essential, and interrelated, concepts of CA, and the basic structures of Japanese, which constitute the critical foundations for the analysis presented in the subsequent section. The analysis section introduces two excerpts, which illustrate the different developments of opinion negotiations, and examines the occurrences of designedly ambiguous turn completion points in these cases. The concluding remarks address the implications such a microanalysis of talk-in-interaction can offer to the understanding of the interrelationship of language, interaction, and culture, and to the teaching and learning of Japanese as a second language.

Projection and Equivocation of a Turn Completion Point One of the fundamental questions that CA researchers pose when analyzing talk-in-interaction is how participants in an interaction determine when the current speaker’s talk ends and a next speaker’s talk is expected to begin. They pose this question not solely as an analyst’s interest, but as a practical concern of those engaging in talk-in-interaction. The fact that the participants take turns at talk often with no gap or overlap between them indicates that they have some shared understanding of normative patterns concerning when the transition of speakership is expected to occur; the components of this shared understanding is what the following CA studies have worked to explicate. The seminal paper by Sacks et al. (1974) and subsequent studies by Auer (1996), Ford (2004), Ford et al. (1996), Ford and Thompson (1996), Schegloff (1996), and Selting (2000), among others, discuss what kinds of resources enable participants to identify a “transition relevance place (TRP).” A consensus among these studies is that “turn-constructional units (TCUs)” are recognized with reference to syntax, prosody, and action implicated by the sequential context. That is, TCUs may consist of various syntactic unit types such as sentential, clausal, phrasal, or lexical, depending on where in a sequence of talk the turn occurs and whether or not the particular syntactic unit completes an action conditionally relevant to that particular position in the sequence (Sacks et al. 1974, further explanation provided later in this section). Further, prosody (i.e., falling, rising, or continuing contours, sound stretch, rush through, etc.) is also considered to serve as a resource for nominating or disqualifying a particular syntactic completion point as a TRP (e.g., Couper-Kuhlen and Ford 2004; Selting 2000). In addition, non-linguistic behaviors such as gaze, posture, and gesture, in coordination with linguistic structures, have been observed to provide participants with crucial cues for organizing turns at talk (e.g.,

SFK ch02

SFK ICBK027

April 21, 2008

17:6

Char Count=

Japanese Applied Linguistics

55

Fox 1999; Goodwin 1981, 1986, 2000; Lerner 2003; Streeck 1993, 1994, 2002). What is important to keep in mind is that unlike various linguistic units that can be defined independently of their use, TCUs cannot be determined without taking the contingency inherent in interaction into consideration (e.g., Ford 2004; Schegloff 1996). Namely, at a possible completion of a turn, the transition to a next speaker can be anticipated, but does not need to be accomplished. Depending on the recipients’ uptake or the lack thereof, a turn that appears to be completed once can be extended through further incremental additions of talk (e.g., Ford et al. 1996, 2002). Further, the different resources serving for the construction and recognition of TCUs may be deployed precisely to equivocate their possible end points, especially when the participants are handling such delicate practices as negotiating agreement and disagreement or demonstrating politeness (e.g., Ford 2004).

Structures of Japanese and their implication for turn-taking operations As mentioned earlier, recent studies have attested that the fundamental mechanisms involved in turn-taking operations appear to hold across languages, while the linguistic structures of each language alter the set of resources available for speakers to accomplish the construction, projection, and recognition of TCUs. In the case of Japanese, a Subject-Object-Verb language, for instance, the basic word order, which is quite different from that of English, a Subject-Verb-Object language, drastically changes the temporal development of the current turn and the projectability of a completion point, as extensively reported by Tanaka (1999, 2000). For instance, (1a) through (1d) below, introduced by Tanaka (1999: 125– 126), illustrate the progression of syntactic development of talk that the recipients of this talk experience and its consequence for their projection of a turn completion point:

(1a)

K: moo netsu ga moo sagatta already temperature S already fell ‘already the temperature had already fallen . . .’

Given the predicate-final orientation, the verb sagatta (‘fell’) can be considered as completing the sentential unit.

(1b)

K: moo netsu ga moo sagatta to already temperature S already fell QT ‘that already the temperature had already fallen . . .’

SFK ch02

SFK ICBK027

56

April 21, 2008

17:6

Char Count=

Negotiating Agreement and Disagreement

The quotative particle to occurring after the verb sagatta (‘fell’), however, retroactively re-specifies the preceding portion to be the quoted content and indicates that the unit is still in progress.

(1c)

K: moo netsu ga moo sagatta to omot already temperature S already fell QT thought ‘ (I) thought that already the temperature had already fallen . . .’

The occurrence of the next verb omot (‘thought’) again raises the possibility that it marks the completion of the unit.

(1d)

K: moo netsu ga moo sagatta to omot tara already temperature S already fell QT thought CONJ ‘when (I) thought that already the temperature had already fallen . . .’

And yet, the conditional particle tara (see Ono and Jones, Chapter 1 in this volume) again transforms the preceding unit to be a so-called subordinate clause, which projects more talk that corresponds to the main clause to come. While these syntactic features of Japanese may present recipients with possible challenges for projecting or recognizing a TRP in a timely fashion, other resources such as prosody, sequential context (further explained later in the next section), and non-verbal behaviors seem to provide them with clues as to how the remaining portion of the turn might proceed.2 Conversely, grammatical features associated with delayed projectability in Japanese can actually be viewed as useful devices for making a turn’s completion designedly ambiguous when accomplishing certain social actions. This phenomenon is what the current study aims to document. More specifically, the cases introduced in the following section demonstrate how linguistic items which can be viewed either as unit-internal items or as grammaticalized unit-final items, including mitaina, yoona, and tte yuu, among others, serve to equivocate if the current turn completes after the item or continues from there on. Mitaina and yoona are often translated into such English expressions as ‘to appear,’ ‘to look,’ ‘to resemble,’ and ‘to be like’ and are generally considered to take the form of [X mitaina/yoona Y], in which X encompasses some kind of descriptor and Y comprises a noun being described. Likewise, tte yuu, the combination of the quotative particle tte and the verb yuu (‘to say’), frequently occurs in the form of [X tte yuu Y], in which X indicates some content that is quoted and linked

SFK ch02

SFK ICBK027

April 21, 2008

17:6

Char Count=

Japanese Applied Linguistics

57

to a noun or a nominalizer in the following Y. While such usages of these items are omnipresent across various genres of discourse, the sentence/ utterance-final usages of these expressions have been recognized in casual oral as well as written interactions,3 especially among younger Japanese speakers according to Lauwereyns (2002). Maynard (2005) and Suzuki (1995), for instance, specifically examine the sentence/ utterance-final use of mitaina. Although these studies do not discuss the sentence/utterancefinal use of yoona, it can be treated as the end of a unit of talk much like mitaina, as demonstrated in an excerpt to be introduced later. As for tte yuu, as far as we know, there is no study that discusses its sentence/utterance-final use, but the grammaticalized sentence/utterance-final use of the quotative particle tte or to has been studied by Hayashi (1997) and Okamoto (1995) and there are cases in which tte yuu is also treated as the end of a TCU.4 The following analysis demonstrates how the dual syntactic status of these sorts of items, in coordination with other resources such as prosody, gaze, and gesture, play a significant role in creating a space to examine the co-participant’s incipient reaction. Opinion negotiation sequences and preference organization As touched upon earlier, in addition to syntax, prosody and non-verbal cues, the participants’ shared understanding of the sequential organization of turns critically contributes to the accomplishment of projection and recognition of TRPs. Namely, the participants closely monitor and analyze each TCU to see what action may be enacted through it and anticipate what possible actions may be made relevant in the subsequent turn (cf. Schegloff 2007). The most basic, pervasive sequence organization observed in conversational interaction is an adjacency pair, which is composed of two adjacent turns, each of which is uttered by different speakers. The notion of adjacency pair does not merely concern the order in which the first pair part (FPP) and the second pair part (SPP) occur, but it features the relationship between the two turns: FPP sets up the conditional relevance as to what types of SPP should follow (e.g., question and answer, request and acceptance/refusal, assessment and agreement/disagreement, etc.); if such an SPP is produced next, it is heard as responsive to the FPP; if it is not produced, on the other hand, its non-occurrence is heard as noticeably absent. CA studies have also explicated how a set of actions anticipated as SPP for a particular FPP present asymmetrical designs that distinguish preferred responses from dispreferred responses (see Pomerantz (1984a) and Sacks (1987) for more discussion on preference organization and see also Mori (1999) for its application to Japanese data). The notion of preference

SFK ch02

SFK ICBK027

58

April 21, 2008

17:6

Char Count=

Negotiating Agreement and Disagreement

discussed here does not refer to the individual participants’ psychological states, but rather it concerns the organization of differential turn shapes and the sequential consequences associated with each type of response. Generally speaking, preferred turns, typically associated with socially affiliative responses such as agreement, tend to be delivered with a minimized gap and in a straightforward manner, whereas dispreferred turns, typically associated with disaffiliative responses such as disagreement, tend to be delayed by a pause, a preface displaying hesitation, or an initiation of repair, and in a mitigated or qualified manner.5 Such structural asymmetries between preferred and dispreferred turn formats provide the FPP speakers a clue to determine the recipients’ reaction before it becomes fully-fledged; the speakers can then redesign their talk if necessary (e.g., Davidson 1984; Mori 1999; Pomerantz 1984b; Sacks 1987). What has been discussed so far, as applied to the case of opinion negotiation sequences, can be schematized as follows:

1st position A: Proffering of an opinion or stance (FPP) 2nd position B: No uptake, delayed and/or minimal claim of agreement (SPP) 3rd position A: Reiteration, extension, or modification of the viewpoint stated in FPP Namely, upon a speaker’s proffering of his or her opinion or stance toward a given issue, which constitutes FPP, delivery of agreement or disagreement becomes conditionally relevant as its SPP. If the slot for SPP is occupied by no uptake or a delayed and/or minimal claim of agreement, the FPP speaker may attempt to pursue a better alignment in the subsequent turn by reiterating, extending, or modifying the opinion or stance expressed in FPP. In contrast, fully fledged agreement delivered as an SPP tends to be initiated promptly and forthrightly, and also followed by substantiation of the claimed agreement, i.e., upgrading and/or giving an independent support for the opinion or stance stated by the FPP speaker. Such an agreement delivered in the preferred turn format, then, occasions the FPP speaker’s acknowledgement or confirmation of the SPP speaker’s agreement as a relevant action, as schematized below:

1st position A: Proffering of an opinion or stance (FPP) 2nd position B: Agreement (claim + substantiation) (SPP) 3rd position A: Acknowledgment/Confirmation (Third Turn Receipt)

SFK ch02

SFK ICBK027

April 21, 2008

17:6

Char Count=

Japanese Applied Linguistics

59

The FPP speaker’s confirmation of the SPP speaker’s agreement achieves mutual alignment between the two parties, which can possibly bring the current sequence to its closing. On the other hand, if the FPP speaker does not appreciate the SPP speaker’s substantiation, the FPP speaker may redo the proffering of an opinion or stance to attain the SPP’s correct understanding. Another possible scenario is that the SPP speaker, who does not receive the FPP speaker’s immediate confirmation of the substantiation he or she offered, may extend it in a way similar to the pursuit of agreement described above. The following analytical section will introduce two sequences of opinion negotiation illustrating how the projection and equivocation of a turn completion point is practiced. To be more precise, it investigates how the linguistic items that have dual syntactic status, as well as some other accompanying devices are used, during the pursuit of agreement in SPP and the pursuit of confirmation as a third turn receipt.

Data Analysis The excerpts examined in this section are extracted from a database that consists of ten interactions among native speakers of Japanese (a total length of 6.6 hours). The participants in the following excerpts, who knew each other well as friends, were asked to read an essay introduced in an intermediate level Japanese language textbook (Miura and McGloin 1994), which describes struggles experienced by Japanese women today, and to discuss their own perspectives. Because of the procedure used for the solicitation of opinion negotiations and the existence of a video camera, the data here might not be considered entirely “natural.” However, we believe that fine details of the participants’ behaviors captured in video recordings provide us with richer information about what the participants experience during the interactions than audio-recordings or field notes. Further, opinion exchange prompted by an assigned reading is a frequently employed task in Japanese as a foreign or second language classrooms; examining the ways in which these Japanese speakers respond to such an instruction eventually informs us as to how instructions for second language learners should be designed and how their performance should be assessed. The two cases demonstrate how the participants attend to the ongoing development of each other’s talk in order to design their own contributions accordingly. In Excerpt (2), Makoto (male speaker) and Fumiko (female speaker) move away from what is written in the text and discuss their own persectives on gender roles and inequalities. Fumiko only produces delayed, minimal uptakes of Makoto’s presentation of men’s

SFK ch02

SFK ICBK027

60

April 21, 2008

17:6

Char Count=

Negotiating Agreement and Disagreement

perspectives, if at all, and Makoto pursues Fumiko’s more elaborate response. On the other hand, in Excerpt (3) Taiji (male speaker) and Asako (female speaker) stick to the text and engage in a critique of the author’s viewpoint. Unlike Fumiko, Asako delivers her agreement with Taiji’s critiques of the text in a timely (but possibly premature) fashion and substantiates it, but her substantiation does not get Taiji’s confirmation until their third trial.6 Pursuit of agreement as SPP The first excerpt to be examined here demonstrates how Makoto develops his argument by carefully monitoring Fumiko’s reaction to his developing talk. As just mentioned, Makoto and Fumiko in Excerpt (2) discuss their own perspectives toward the issue of gender inequalities, emphasizing hardships experienced by their respective genders. The third participant, Toru, minimally participates in the discussion in this segment, aligning with neither one of his co-participants. In line 1 through 11, Fumiko states that wives usually have to take some time off from their work after childbirth in order to take care of infants and that women tend to experience challenges when they try to go back to the workplace. Both of the male participants acknowledge the point raised by Fumiko (lines 13 and 14), and Makoto claims his qualified agreement (line 15). Starting in line 17, however, Makoto delivers a counter argument stating that the time off from work after childbirth can be seen as an opportunity to reflect on one’s lifestyle, whereas men, who are generally considered responsible for supporting their families, cannot easily consider changing careers.

(2a) F = Fumiko, M = Makoto, T = Toru 1 F: soo yuu baai yappari- (.) okusan ga:: (.) 2 ikujikyuuka de::, ‘In such a case, typically (.) it is wives (.) who take a childcare leave and’ 3 M: un un ‘uh huh uh huh’ 4 F: (.) nagai taamu‘(.) for a long term-’ 5 M: un [un ‘uh huh uh huh’ 6 F: [yasumu koto ni nan desho[::? ‘they end up being away from work’ 7 M: [u:::n ‘uh hu:::h’

SFK ch02

SFK ICBK027

April 21, 2008

17:6

Char Count=

Japanese Applied Linguistics

8 9 10 11 12 13 14 15 16 17 18 19

20 21 22 23 24 25

26 27 28 29 30 31

(2.0) F: [soo naru to (.) nee?= ‘If that happens (.) you see?’ M: [sore wa: ‘That is’ F: = [fukki suru no ga shindokattari toka.= ‘Like returning to work gets hard.’ M: [sore- un ‘that, uh huh’ M: =u::n. ‘uh hu::h’ T: un. ‘uh huh’ M: sore wa soo da to omoo. ‘I think that’s right.’ (1.2) M: dakara koso: gyakuni ie ba onna no hito wa chansu de atte:, .hh e:to (.) soko de ikkai ikuji de zettai ikkai (.) YAsumu.= ‘That’s why, if you look at the reverse side, women have a chance and, .hh well (.) once they definitely take a leave for child-care once then.’ =sukoshi. [sukoshi (no jikan). [◦ sukunakutomo.◦ ‘A little bit. For a short while. At least.’ F: [◦◦ n◦◦ [◦◦ n◦◦ ‘uh huh’ ‘uh huh’ (.) M: de sono toki ni:: (0.4) jibun: no (kyoria) o hikidoo yatte tsunagete iku no ka. >aruiwa< doo yatte henka sasete[iku no ka. ‘Then at that time (0.4) how do they maintain their career? Or how do they change it?’ F: [◦◦ n◦◦ ‘uh huh’ (1.1) M: soo yuu chansu ga aru, ‘such a chance exists,’ F: ◦◦ n◦◦ ‘uh huh’ (0.6) M: ◦ n ja nai kana.◦ =GYAKUNI ieba sa otoko tte sa (0.4) ‘doesn’t it?= On the contrary, as for men (0.4)’

61

SFK ch02

SFK ICBK027

62

April 21, 2008

17:6

Char Count=

Negotiating Agreement and Disagreement

In lines 17 to 18, Makoto first presents an upshot of his counter argument that women have chansu (‘chance’). He then elaborates on his argument by explaining what he means by it. At the end of line 25, Makoto’s elaboration comes to the point at which what he means by “chance” becomes understandable. At this moment, Makoto gazes at Fumiko and produces a slight head nod that seems to indicate that he expects, or at least observes, Fumiko’s reaction. Fumiko, however, only produces a slight head nod and hardly an audible token, ◦◦ n◦◦ . After a short silence in line 27, Makoto reiterates the upshot of his argument stated at the beginning of this turn. The verb that comes at the end of line 28, aru (‘exist’), can possibly mark a syntactic as well as action completion of the turn, and thus this point arguably becomes another possible completion where Fumiko’s response would be relevant. However, here Makoto produces the verb with a slightly rising continuative contour, thereby making it ambiguous whether or not he is soliciting Fumiko’s response or continuing. While producing the verb aru, he brings his gaze, which was once moved away from Fumiko at the beginning of line 28, back to her, only to find another minimal reaction from Fumiko. In line 31, then, Makoto further extends his turn in a very soft voice by adding the sentence-final tag-like expression, n ja nai kana? (‘doesn’t it?’), which clearly marks a syntactic and prosodic completion of his turn. At this possible TRP, however, he immediately produces the next unit of talk in a significantly louder voice. In the additional unit of talk starting in line 31, Makoto describes men’s disadvantages. This additional talk shifting the focus of his argument seems to be prompted by the lack of Fumiko’s indication of her alignment observed so far at critical points in the ongoing development of his turn. Further, in describing the general societal expectation (lines 31–33), Makoto inserts a self-qualification that admits that there are some exceptions to this tendency, which serves to preempt possible challenges that can be raised by the co-participants (Mori 1999). This last attempt by Makoto in fact successfully attains Fumiko’s qualified, and yet substantial, sympathetic response.

(2b) 31 32 33

M: ◦ n ja nai kana.◦ =GYAKUNI ieba sa otoko tte sa (0.4) maa shakai tsuunenjoo ippanni wa? (0.6) gyaku mo aru to omoo kedo ippanni wa (0.5) moo kaseede:: chanto: (0.4) ‘doesn’t it? =On the contrary, as for men (0.4) well, by common sense generally speaking? (0.6) I think there are exceptions, but generally (0.5) men have to earn money and properly (0.4)’

SFK ch02

SFK ICBK027

April 21, 2008

17:6

Char Count=

Japanese Applied Linguistics

34

F:

35

M:

36 37

F:

38

M:

39 40

F:

41

M:

42 43

M:

44

F:

63

◦◦ ◦◦

n ‘uh huh’ (0.8) kazoku kuwashite ikanakya ikenai. ‘(0.8) support their families.’ (0.8) maa [ne. ‘Well, yeah.’ [◦ >ikenaisoo suru to< (.) karugarushiku, ‘ =if so (.) lightly’ [(seekatsu-) ‘Life-’ chotto tenshoku shite miyoo ◦ ka toka.◦ ‘trying to change career or something like that.’ (1.0) [soo yuu no nakanaka (.) dekinai. ‘(Men) can’t easily do such things.’ [sore wa::: tsurai toko da na:: ◦ to wa omoo kedo [ne.◦ ‘That is a tough thing, I think, but.’

Makoto’s description of the societal expectation regarding men’s assuming financial responsibility comes to a syntactic, prosodic, and action completion in line 35. At this moment, Makoto’s hands, which have been brought up to create a series of gestures starting in line 32,7 are brought down to their “home position” (Sacks and Schegloff 2002). This hand movement also marks a completion of the delivery of his argument. During the silence in line 36, Makoto and Fumiko sustain mutual gaze, and in line 37, Fumiko finally produces a delayed, qualified token of acknowledgment, maa ne, which is still more substantial than ◦◦ n ◦◦ produced earlier. As soon as Makoto hears this qualified token of acknowledgement, he confirms the existence of such a societal expectation. Further, without giving Fumiko an opportunity to respond, he quickly continues to discuss how such an expectation prevents men from considering career changes with ease (lines 39 and 41). Towards the end of line 41, Makoto produces toka, which indicates the preceding utterance to be providing an example, in a falling contour, and gazes at Fumiko during the following silence while producing slight head nods. The adverb karugarushiku (‘lightly’) in line 39 projects a negative ending and therefore what comes after toka can be considered predictable. These features again arguably make this point a possible completion of

SFK ch02

SFK ICBK027

64

April 21, 2008

17:6

Char Count=

Negotiating Agreement and Disagreement

Makoto’s turn and a place where Fumiko’s response is relevant. But the syntactic structure and the placement of his hand, which is out of home position to produce additional gesture, keep the possibility of continuation open. And indeed after a short pause, Makoto continues to develop his talk in line 43. Makoto’s reference to men’s difficulty in making career changes caused by the societal expectation of financial responsibility placed on them indeed meets Fumiko’s qualified (with the use of particles such as wa or kedo indicating contrastive meaning) and delayed (if we take the toka to be marking a completion) claim of agreement, or expression of sympathy (line 44). Upon securing Fumiko’s qualified and yet slightly more affirmative response, Makoto reiterates the last portion of his prior talk, which was overlapped with Fumiko’s response, while changing the predicate from dekinai (‘cannot do’) to shizurai (‘hard to do’) as if incorporating the adjective included in Fumiko’s response, tsurai (‘hard’). Following this third turn confirmation, Makoto adds on another concluding remark in which he compares men’s hardships with women’s (line 47 and 48).

(2c) 44 45 46 47 48 49

50

F: [sore wa::: tsurai toko da na:: ◦ to wa omoo kedo [ne.◦ ‘That is a tough thing, I think, but.’ M: [u::n. ‘uh hu::h’ (0.5) M: nakanaka booken ga shizurai.=.hh SOO iu imi de otoko no hoo ga kurushii tte yuu (0.7) me(h)n datte na(h)ki ni shi mo arazu? ‘It’s hard to take a risk.=.hh In that sense, men’s life is harder (0.7) it’s not that there isn’t such a perspective?’ F: tada (.) [atashi no tomodachi no ree de iku [to ne? ‘But (.) in the case of my friend? . . .’

As Makoto’s talk reaches the adjective kurushii (‘hard’) (line 48), which constitutes the core of his argument, Makoto brings his gaze, which was moved away from Fumiko at the beginning of this added component (at the middle of line 47), back to Fumiko while moving his head forward toward Fumiko as if adding emphasis on the adjective. After the quotative expression tte yuu, an item that has the dual syntactic status, Makoto pauses while maintaining mutual gaze with Fumiko. As Fumiko does not provide any response to this version of Makoto’s argument, Makoto continues his talk by adding the noun men (‘aspect’), which retroactively treats the preceding item to be

SFK ch02

SFK ICBK027

April 21, 2008

17:6

Char Count=

Japanese Applied Linguistics

65

continuative, and the final expressions with double negatives, na(h)ki ni shi mo arazu (‘it isn’t that there is not’), which significantly weakens his assertion. This last part of talk was produced with some laughter and a smile, which is also reciprocated by Fumiko’s smile. In the following turn, Fumiko, without providing any response to Makoto’s comparative statement, shifts the topic and starts delivering a story in which a couple she knows decides not to marry as they considered their marriage would unfairly inconvenience the woman. The examination of different stages of negotiation shown in Excerpt (2) reveals how the speaker, in pursuit of the co-participant’s affirmative response, makes efficient use of the linguistic structures that equivocate the completion or continuation of the current talk.

Pursuit of confirmation as a third turn receipt In contrast to Excerpt (2), Excerpt (3) demonstrates cases in which a fullyfledged agreeing turn is delivered in a timely fashion. What becomes an issue in this excerpt is that the SPP speaker’s substantiations demonstrating her understanding of the FPP speaker’s viewpoint are not confirmed by the FPP speaker in the third position. In this case, then, a phenomenon similar to what is observed in the previous section occurs in the extension of SPP turns in pursuit of confirmation as a third turn receipt. As mentioned earlier, Taiji and Asako in this excerpt are discussing their critical reading of the text provided. A recurrent pattern observed in this segment, thus, is to first locate the portion of the text to be commented on, and then to read aloud the selected portion, and finally to comment on what is written there. For instance in Excerpt (3a), Taiji locates the portion of text in lines 3, 4, and 6, and upon securing Asako’s recognition (line 7), he reads the portion aloud in lines 9 and 10. He then states his interpretation of the portion in lines 10, 11, and 14: that is, he is not sure if the author sufficiently discusses the men’s responsibility for the lack of change in their attitudes and behaviors. Immediately after Taiji produces yoona (at the end of line 14), whose syntactic status for marking a turn completion is ambiguous, Asako initiates her response, i.e., a claim of agreement followed by the substantiation of her understanding of the text (line 15).

(3a) A = Asako, T = Taiji 1 T: ma toriaezu sa:[: (1.5) ‘Well, anyway, (1.5)’ 2 A: [n::: ‘uh hu::h’ 3 T: ma demo i- soo ka::. demo jidai ga- ichiban ichi- ichi-

SFK ch02

SFK ICBK027

April 21, 2008

66

Char Count=

Negotiating Agreement and Disagreement

4

5

A:

6

T:

7

A:

8 9 10 11

17:6

T:

12

A:

13 14

T:

15 16 17

A:

18

T:

19

A:

20

T:

21

A:

ichiban ue no sa::[: koko ni ju- juu::ichi ni ro-(.) ‘Well but, I see. But “the time is,” the most, mos- most top column, here eleven, twelve, six (.)’ [n:: ‘uh huh’ nanagyoome?= ‘seventeenth line?’ =n:::. ‘uh huh’ (0.4) tch jidai ga kawatte kita noni otoko ga kawatte kurenai node tte itte kkara ichioo otoko ga kawatte nai tte koto ni taishite:::, ‘Since (the author) writes, “Although time is changing, men haven’t changed,” and so in a way about the fact that men haven’t changed,’ n::: ‘uh huh’ (0.2) monku wa = ‘(the author) seems to complain or doesn’t?’ =soo. itteru::, (0.2) ichioo itteru n da kedo::, (0.4) jaa doo suru ka tte yuu toki ni:::, (0.3) no:: sajesshon ga e- (0.3) ee:: nan- e [onna no hito ni sase: (.) ru no, ‘Yeah. He’s complaining. (0.2) he in a way complains, but (0.4) then at the time when he discusses what we should do, (0.3) the suggestion is, “Wha- (0.3) Oh what? Would you make women do’ [n::: ‘uh huh’ sore made ◦ mitai na◦ = ‘even that”? or something like that’ =>DE SA [saigo no-< ichiban saigo no saigo tte sa::[:, ‘And then the last, the final, as for the final ending,’ [◦ ki ga◦ shimashita. ‘is what (I) felt.’

Asako acknowledges that the author problematizes the fact that men have not changed, but criticizes him for not clearly presenting a solution for the situation while having women take on the entire burden. Asako moves her head toward Taiji at the self-repair of the particles in the middle of

SFK ch02

SFK ICBK027

April 21, 2008

17:6

Char Count=

Japanese Applied Linguistics

67

line 16, as if examining Taiji’s reaction to the following part of her turn. During this time, however, Taiji keeps his gaze on the text. And as Asako produces some speech perturbations in line 17, Taiji produces the minimal token n::: (line 18) thereby letting Asako continue. Taiji could have joined in the composition of a shared critique at this point, had he considered Asako’s turn to actually be in alignment with his own perspective. Asako’s subsequent behaviors, in fact, indicate that she might have sensed the lack of Taiji’s alignment: that is, she brings her gaze back to the text as she utters the verb sase: (‘make (someone) do’) and reduces the volume of her talk toward the end of line 19. The expression mitaina (‘like’) at the end of line 19 is also one of those items that have dual syntactic status. While Asako continues her talk, which then comes to a definitive syntactic, prosodic, and action completion (line 21), Taiji indeed takes it as a place where he can initiates his turn (line 20). This is where he, the FPP speaker, could produce his confirmation of Asako’s SPP response. However, what actually happens is not a confirmation of Asako’s point, but a continuation of his own argument referring to a different portion of the text. Here again, Taiji first locates the portion of the text (line 20), and after securing Asako’s recognition (line 22), he reads the portion aloud (lines 23–26). In line 28, Taiji begins to deliver his critique, which meets once again Asako’s rather prompt, or seemingly premature, claim of agreement (line 29).

(3b) 20 T: =>DE SA [saigo no-< ichiban saigo no saigo tte sa::[:, ‘And then the last, the final, as for the final ending,’ 21 A: [◦ ki ga◦ shimashita. ‘is what (I) felt.’ 22 A: [n::: ‘uh huh’ 23 T: jibun no ikikata o erabu no wa- erabu no wa kojin 24 no jiyuu desu.=hoka no hito no toyakaku iu koto 25 dewa arimasen. .hh demo (0.8) ichibu no hito- onna ga 26 dinkusu ya shinguru ni natte mo tte kaite aru jan? ‘“It’s up to an individual to choose, choose what kind of life she lives. It’s not something other people should censor. But (0.8) even if some women become DINKS ((Double Income No Kids)) or single,” that’s what’s written, right?’ 27 A: n:: ‘uh huh’ 28 T: karuku yappari hit- (.) are da yo ne.=kono (0.3) ‘Lightly, you know, den- (.) doing that, right? This (0.3)’

SFK ch02

SFK ICBK027

68

29 30 31 32 33

34 35 36 37 38 39 40

41 42 43 44 45 46

April 21, 2008

17:6

Char Count=

Negotiating Agreement and Disagreement

A: ea:: soo desu ne.= ‘Oh, that’s right.’ T: =hite- hitee shiteru yo ne.= ‘deny- denying, isn’t he?’ A: = karuku hitee shite masu ne.=kore kanari:. da- kanari kono .hh (0.5) ano:: atashi mitaina hito(h):mo(h)[::(0.6) ‘He’s lightly denying. Considerably. Considerably this (0.5) well, someone like me also (0.6)’ T: [n:: ‘uh huh’ A: WA::, (1.3) ji- kojin no jiyuu da keredomo::, ‘has (1.3) individual freedom, but,’ T: soo soo. ‘Yes yes.’ A: demo ss. sugoku ha- koo:: (.) shuhu mo shite::, ‘But ss. very ha- well (.) being a housewife and,’ T: n:: ‘uh huh’ A: ano:: nanda, kodomo:: sodatete:, (0.3) kaisha ni mo itteru- (0.8) onna no hito ga ichiban yappari ‘well, what to say, raising children and, (0.3) working for a company (0.8) such women are the most, as expected,’ T: n:: ‘uh-huh’ A: erai n- (1.0) erai n desu yo ◦ tte yuu::,◦ ‘greatest (1.0) are greatest, that’s,’ T: ◦ n::◦ ‘uh huh’ A: koto desu yo ne? ‘what is said, isn’t it?’ (1.6) A: ◦ tashikani.◦ = ‘Certainly.’

The adverb karuku (‘lightly’) used in Taiji’s turn in line 28 suggests that he is going to provide some negative evaluation. The distal demonstrative pronoun are used in line 28 is considered to serve as a placeholder when the speaker momentarily experiences trouble producing a word (Hayashi 2004b; Kitano 1999). The use of the placeholder enables the speaker to continue producing and completing the turn despite the problematic word. Thus, it is not out of question for Asako to initiate her claim of agreement in

SFK ch02

SFK ICBK027

April 21, 2008

17:6

Char Count=

Japanese Applied Linguistics

69

line 29. However, this response in line 29 can be also seen as “premature,” as Taiji’s turn in line 28 has not yet disclosed the substance of his critique. In line 30, Taiji restates his critique of the author, who lightly denies something, but what that something is has not yet been clarified. Nevertheless, Asako, once again, immediately delivers her claim of agreement in line 31 and upgrades the degree of the negative evaluation by changing the adverb karuku (‘lightly’) to kanari (‘considerably’). She then continues to extend her turn by elaborating on her understanding of the text. Asako first produces the prefatory clause marked by -keredomo that reiterates what the text says, and this draws Taiji’s soo soo (‘yes yes’) (line 36), which is typically used to confirm that the co-participant’s immediately preceding talk is in alignment with the soo producer’s perspective (Kushida 2006). On the other hand, as her turn moves into the main clause, which is expected to deliver her understanding of Taiji’s intended critique, Taiji’s reactions return to the production of the minimal token n::, which does not necessarily indicate the token producer’s confirmation of the co-participant’s talk being in full alignment with his or hers (Kushida 2006). As Asako utters hito ga (‘person’ + subject marker) in line 40, she brings her gaze up from the text, and looks at Taiji as she utters the adverbial phrase ichiban yappari (‘most as expected’) (line 40). As discussed by Goodwin and Goodwin (1992), Lerner and Takagi (1999), Tanaka (2001a), among others, these sorts of adverbial expressions strongly project a forthcoming assessment in this particular environment, and therefore if Taiji were in strong alignment with Asako, he could join in the production of the assessment at this point. However, Taiji only produces a minimal token while keeping his gaze on the text (line 41). Perhaps in response to this lack of strong uptake by Taiji, Asako brings her gaze back down to the text as she continues her utterance delivering the assessment, erai (‘great’). She then prolongs her turn by pausing and repeating the adjective, adding the copula, the final particle, and the quotative expression tte yuu, providing various opportunities for Taiji to join in her evaluative statements. As mentioned earlier, tte yuu is also an item whose syntactic status for marking a turn completion is ambiguous, and as she produces this in a soft voice, she looks up again at Taiji, but finds him looking down at the text and producing the minimal token (line 43). As she returns her gaze back to the text, Asako finally completes her turn in line 44 with rising intonation soliciting Taiji’s reaction. Not receiving any third turn confirmation from Taiji, Asako produces tashikani (‘certainly‘), another item, which often serves as a claim of agreement placed in SPP, in a soft voice (line 46). Taiji’s third attempt to deliver his critique of the text begins in line 47. Once again, he identifies the portion of the text to be criticized (lines 51, 53, 54), which is actually the same portion mentioned in the first attempt. But this time, the target is narrowed down to the use of the single word kurenai

SFK ch02

SFK ICBK027

70

April 21, 2008

17:6

Char Count=

Negotiating Agreement and Disagreement

(‘not give (favor)’) in the text. This time, Asako does not necessarily deliver an immediate claim of agreement, but she finally provides an appropriate substantiation of the critique originating in Taiji, which meets his third turn confirmation (line 82).

(3c) 47 48 49 50 51 52 53 54 55

56 57 58 59 60 61 62 63 64

65

T: =ATO SA:: ‘Also,’ A: u::n= ‘uh huh’ T: = ore cho- s:uggoi komakai kedo:[:, ‘I’m lit- very picky, but’ A: [n:: ‘uh huh’ T: kono juunanagyoome no::,= ‘this seventeenth line,’ A: = n::= ‘uh huh’ T: =jidai ga kawatte kita noni otoko ga kawatte kurenai >tte yuu< kono kurenai tte yuu no ga sugoi hikkakaru.= ‘The term kurenai in the passage “Although time is changing, men haven’t change-kurenai” makes me feel very uncomfortable.’ A: =He heh heh heh heh ‘He heh heh heh heh’ T: otoko wa kawa- [nande otoko wa kawaranai no! ‘Men don’t chan- why don’t they change!’ A: [a hah hah hah hah ‘a hah hah hah hah’ (1.1) ((Asako is still laughing silently.)) A: .h nanka(h).h [(.) tashikani nanka= ‘.h like.h (.) certainly, like’ T: [dakara (.) ‘So (.)’ A: =[tashikani soo ◦ desu ne.◦ uh(h)m ‘That’s certainly right. uh(m)m.’ T: [dakara::(.) onegaishiTE::, >sono< kawatte morau tte yuu? ‘So (.) (Women) beg and receive men’s favor of changing themselves?’ A: N::N. ‘UH HUH’

SFK ch02

SFK ICBK027

April 21, 2008

17:6

Char Count=

Japanese Applied Linguistics

66 67 68 69

70 71

72 73 74 75

76 77 78 79 80 81

82 83 84 85 86 87

(0.3) T: so[no ‘that’ A: [SOO desu.=koozoo ga sonna kanji desu yo [ne. ‘Yeah. The structure is like that, isn’t it?’ T: [n::n. ‘uh huh’ A: .h otoko ga- wa kawa- (0.5) tt: (0.3) kurerutanonde mo kawatte kurenai shi::, ‘Since men chan- (0.5)ge (0.3) gives-, men wouldn’t change even though women beg and,’ T: n::= ‘uh huh’ A: =dakara motto motto onna no hito ga:, koo yuu otoko no hito ni:, kawatte- (.) moraeru yooni::, [doryoku shimashoo >mitaina kanji ja nai su ka.↑soo ↓desu ka.< [a: ↑nichiyoobi:, ↓nichiyoobi:, ‘Is that right? Um: Sunda:y, Sunda:y,’ 11 T: [n. ‘Yeah.’  12 S: eiga: demo: mimasen ka? ‘ won’t you go to the movies with me?’ Consistent with the dialogs modeled in the instructional materials, the student produces the availability checking step (line 3) and manages the tester’s confirmation check regarding the day (line 4) by producing a confirmatory hai (‘yes’) in line 5. The tester then responds to the initial query about her availability with hima desu yo? (‘I’m free, but why do you want to know?/tell me why you’re asking’), a response that both asserts that she is available while at the same time signaling— through the use of rising intonation—a request for more information from the student about why her (the tester’s) availability is a matter of interest for the student.7 The student’s response in line 7, however, addresses only the informing function of the tester’s turn: her use of a! marks the receipt of the new information from the tester, and the subsequent ii desu ne projects a shared alignment toward the tester’s

SFK ch11

SFK ICBK027

310

April 21, 2008

15:27

Char Count=

Learner Competence as a Resource

availability as “a good thing” that is mutually recognizable to both parties as such. From the perspective of the instructional materials, the learner’s production in line 7 is clearly “off-script.” Indeed, no other student in the control group produced this type of turn at this juncture in conjunction with completing Task 2. The move, a positive evaluation of some aspect of the tester’s response, is relevant to the activity-in-progress (i.e., inviting a classmate out for a meal) in that it functions as a positive orientation toward the tester/tester’s response, and a rapport-building move with the tester. The student, then, appears to be activating a dimension of her learner competence. Yet despite the fact that this turn serves to build rapport between the two participants (in their role-played roles as classmates) in an unscripted and authentic manner, as formulated, this turn does not adequately address the communicative demands placed on the student by the tester’s preceding turn, i.e., to provide the basis for the student’s interest in the tester’s availability “on Sunday.” Owing to the ambiguity inherent in the formulation of the student’s turn in line 7, the student “withholds” precisely the information the tester implicitly requests through her use of rising intonation in line 6. This inadequacy of the student’s response is reflected in the tester’s continued pursuit of the “missing information” (line 8, a (dooshita no)?). The student, however, does not address the tester’s pursuit of this agenda. The tester’s elicitation of additional information is, at first, ignored by the student: she talks over the tester’s turn-in-progress, producing de! (‘so’) in an animated manner, a move which marks her transition to the next step in the invitation sequence (line 9).8 Then, after the tester completes her query (line 8), the student gives only the most cursory acknowledgement of the tester’s turn-at-talk: she produces the neutral formulaic acknowledgement soo desu ka (‘Is that right?’) with rapid speech, and then immediately moves to the final step of the (instructed) invitation sequence, the proffering of the invitation itself, again talking over the instructor’s turn (a minimal response to her soo desu ka). In sum, learner competence is evident in this excerpt, both in the learner’s innovative production of the aligning move ii desu ne, and in the use of the neutral acknowledgement token soo desu ka to provide a minimal uptake on the turn produced by the tester in line 8. Yet, these displays of learner competence do little to enrich the learner’s negotiation and presentation of her “speaking self” vis-`a-vis the assessment-based task of extending an invitation to a classmate. Instead, the student’s invoking of learner competence in this excerpt serves predominantly to sustain her interactional footing as the testee in an assessment activity, moving toward the completion of the obligatory task at hand (“Invite your classmate . . .”): from line 7 until the

SFK ch11

SFK ICBK027

April 21, 2008

15:27

Char Count=

Japanese Applied Linguistics

311

proffering of the invitation at the end of the excerpt, the student pursues the completion of her invitation with minimal deviation from the script modeled in the instructional materials. Moreover, the way the student positions herself in this role is collaboratively accomplished through her interaction with the tester. The tester does not hold the student accountable for her rather abrupt treatment of the tester’s talk, but instead, allows the somewhat awkward interactional moment to pass uncommented upon once it is evident that the student is “on-track” with respect to accomplishing the target activity of proffering the invitation. Thus, while the tester’s responses in lines 6 and 8 provide the student with opportunities to collaboratively coconstruct a less scripted, more interactionally negotiated exchange between peer familiars—in terms of the negotiation of discourse roles and identities through information sharing and stance indexing, among other resources (Tracy and Naughton 2000:75), the student’s responses to these turns structure her participation more along the lines of a student intent on performing the target dialog without getting sidetracked or derailed by the tester’s “off-script” interactional moves. Experimental instructional treatment: Valuing learner use of learner competence9 The experimental group used the same materials as the control group, however the use of the materials was supported by two pedagogical principles that were reinforced over the course of the semester. First, during conversational practices, the instructor emphasized that learners must pay attention to what they were doing when they were saying some particular utterance in Japanese. The notion of “doing” was constructed with reference to the organization of (sequence, opening and closing moves, transitional moves), orientation toward (stance, positionality, social identities), and accomplishment of (actions, preferences, response behaviors, particular linguistic resources) the target activity. This first principle was pedagogically reinforced by a series of consciousness-raising activities and instructional practices directed at training the student to draw on the resources afforded by learner competence as a basis for managing participation in role-played interactions. Core activities included teacher-led discussions of L1 and L2 interactional practices as well as the development of a learnergenerated/teacher-moderated metalanguage for talking about everyday spoken interaction.10 Second, there was a constant call to the learners to “say what you want to say,” that is, to draw on their learner competence in ways that supported their efforts to participate in Japanese language interactions that “felt right” and thereby validated their sense of a “speaking self” (cf. LoCastro’s (2001:83) discussion of the importance of taking learners’

SFK ch11

SFK ICBK027

312

April 21, 2008

15:27

Char Count=

Learner Competence as a Resource

individual variations into account with respect to the learning of L2 pragmatics). The instructor monitored these aspects of production for appropriateness, interjecting with confirmations of the appropriateness of learner innovations, alternate suggestions for accomplishing a particular innovation, or, on occasion, explanations of “why it can’t be done that way in Japanese.” Experimental group data: Rapport building, shifting activity, managing tester responses Although the assessment sessions for the experimental group covered the same course materials as those covered by the control group, differences in the assessment procedures that arose from the pedagogical emphasis on learner competence are salient and, arguably, significant. These differences are evident from the initial presentation of the fundamental premise of the role-play. Consistent with the pedagogical emphasis on establishing a sense of a “speaking self” (see above) when engaging in conversational interaction, the instruction sheet for the experimental group specified that the classmate—role-played by the tester (i.e., the instructor)—is someone with whom “you do not have a close relationship, but whom you would be interested in getting to know better.” By specifying this aspect of the participants’ situated social identities, the instructor generated the expectation that the talk during the oral assessment would be constructed between peers who are familiar, but not necessarily close familiars; moreover, the positionality of the learner’s role supports him/her taking a slightly more outgoing conversational stance toward the “classmate”, as role-played by the tester. A second difference is the use of an outline of the expected flow of the role-play interaction as shown below (as opposed to the card-by-card procedure used with the control group): Meet (unexpectedly) and greet classmate, and chat Extend an invitation to study together/eat out/go out –Ask preferences about time, location, meeting place At the {location: restaurant, movie theater, etc.} –Talk about ordering (i.e., what is good, what will you have) –Order –Chat This approach was developed in an effort to make assessment procedures consistent with the scenario-based instructional practices used with the experimental group. Further, because the goal was to assess student ability to participate in the guided role-play interaction using the language learned in class in an

SFK ch11

SFK ICBK027

April 21, 2008

15:27

Char Count=

Japanese Applied Linguistics

313

unscripted manner (i.e., without reliance on rote production of instructed dialogs), immediately prior to the beginning of the assessment session, the student was given a prompt regarding the nature of the invitation he/she was to proffer; the student was then required to incorporate the prompt into the subsequent role-play interaction. The prompts ranged from a flyer for an actual on-campus activity (Prompt: “Invite me to this activity.”) to the more general cue, “Invite me to study for the final exam with you.”11 The use of an instructor-generated prompt might, at first, appear to work against the goal of promoting learner access to and reliance on learner competence as well as against the pedagogical principle of having the student “say what you want” (see discussion of instructional approach above). To the contrary, the inclusion of these last-minute instructor-generated prompts, prevented students from generating and carrying out a pre-planned agenda or a previously rehearsed script, and also guaranteed that at least one aspect of the interaction would be consistent with the “spontaneous” (i.e., as purported in the role-play description) nature of the encounter between the student and the classmate.

Data Analysis The examination of the experimental group data focuses on the extent to which learner language use reflects an ability to manage the organization of, orientation toward, and accomplishment of the tasks set forth on the assessment session instruction sheet in ways that extend beyond the use of the scripts and linguistic resources in the instructional materials as presented therein. The data reveal that while many of the component acts of the invitation sequences produced by the experimental group were accomplished with the linguistic strategies modeled in the instructional materials (i.e., the use of -masen ka? to proffer an invitation), the interactional work undertaken by these students to position themselves in the role of inviting peer (i.e., the negotiation of a friendly rapport with the tester, the collaborative achievement of shifts between chatting and the invitation activity, and the management of unscripted tester responses in conjunction with both the chatting activity and the invitation activity) demonstrates that the learners make effective and productive use of a broad range of innovative and situationally appropriate conversational moves and interactional stances in conjunction with their participation in and joint accomplishment of the role-played assessment activity. In Excerpt (3), the student, who had no prior experience learning Japanese before enrolling in this class, engages in considerable interactional work to establish a friendly rapport with the tester (in her role-played role as classmate) prior to moving to the invitation activity.

SFK ch11

SFK ICBK027

314

April 21, 2008

15:27

Char Count=

Learner Competence as a Resource

(3) ((The participants have just finished talking about what classes they are taking this term.))  1 S: arubaito: (.) shite iru n desu ka? ‘Are you: (.) working a part-time job?’ 2 T: ano, kongakki desu ka? ‘You mean this semester?’ 3 S: e. ‘Yeah.’ 4 T: ee. ano:, watashi ga Taun andu Kantrii Saafu Shoppu de= 5 =[arubaito shiteru n desu kedo ‘yeah. um:, I work at Town and Country Surf Shop, so’  6 S: [a-h-h-h soo ‘o:h-h-h’  7 a soo desu ka! [.h] Ara- Ara Moana um Mooru desu ka? = ‘oh really! [.h] at Ala- Ala Moana um Mall?’ 8 T: [e.] ‘yeah’ 9 S: [desu !ne?! ‘Right?!’ 10 T: [soo desu ne! ‘Right!’ 11 so. Ara Moana [( ) Ara Moana sentaa desu. ‘Right. Ala Moana ( ) Ala Moana Center.’  12 S: [desu ne-h-h. ‘ri-h-h-ght. (at Ala Moana)’  13 oh um watashi wa: (.) um (.) arubaito sh:ite iru n desu. ‘Oh um I: (.) um (.) work part-time.’ 14 T: a Ben san mo arubaito ga aru n de[su ka? ‘Oh you also have a part-time job?’ 15 S: [e. ‘yeah. In contrast to the question used to initiate the topic about the classmate’s courses (not shown here: kongakki wa, um nani o totte imasu ka?), the learner initiates this subsequent topic with an n desu ka question (line 1), which marks a more conversational tone (Yoshimi 2001), and an orientation toward engaging in the topic in a more involved manner. His uptake (line 6) of the classmate’s response (line 4–5) is timed in a manner that shows close attention to the details of the classmate’s talk, an expected listener behavior (Ikeda 2004). The classmate’s response, which continues past the completion point predicted by the learner (line 6) is taken up a second time (line 7) with an interest marker (produced in an animated

SFK ch11

SFK ICBK027

April 21, 2008

15:27

Char Count=

Japanese Applied Linguistics

315

manner) and a follow-up question, Ara Moana Mooru desu ka?, which serves to sustain the topic and provide an additional index of the student’s involvement in the tester’s talk. Although the formulation of the follow-up question is both grammatically and pragmatically appropriate to its function in the flow of the interaction, the student self-repairs his utterance, shifting from a ka-marked question (which is neutral with respect to speaker-hearer alignment) to a ne-marked confirmatory question (line 9), which signals a shared epistemic stance toward the topic at hand and interactional alignment with the tester. While the motivation for this shift in footing, partially repeated in line 12, is not immediately evident, the subsequent talk reveals that this marking of alignment (lines 9 and 12) provides the interactional groundwork for the student’s next conversational move: a topic expansion (line 13) that makes his own part-time work experience at a retail establishment at the same mall the next topic of talk (not shown). The slight laughover on the repetition of desu ne (line 12) may, in retrospect, be viewed as a form of private speech, arising as the learner orients to the inherent pragmatic value of his aligning stance as facilitative of a next conversational move. The change of state marker (“oh”) to initiate his next turn (line 13), as well as the initial disfluency of the turn as he works to bring the topic expansion to the floor, suggests that this planning occurs spontaneously. The above analysis demonstrates that the student does not take the relative social closeness inherent in the role-played roles of classmates for granted. Instead, the student uses linguistic and discourse resources (i.e., follow-up questions, the interactional markers ne and n desu, and the strategic noticing of shared information) to actively co-construct a context in which shared experience and joint alignment can be established and sustained, and can be used to reduce the social distance between the two classmates who are still getting to know each other. Jucker and Smith (1996:15) refer to this as the establishment of common ground and argue that this type of phatic talk has the critical role of keeping the communication channel open and enabling the participants to maintain a feeling of intimacy. In Excerpt (4), which follows closely on Excerpt (3) above, the talk on the topic of part-time work continues for several turns (lines 1–10) before reaching closure (line 11). In the talk between these two excerpts (not shown), the student has shared information about where he works, and both the student and tester have discussed the respective locations of their workplaces in the large multistoried mall. The tester continues to pursue the topic (lines 1, 2, 4, 6, and 8), but the student’s responses are increasingly minimal, and there is an absence of markers of both alignment and engagement. The absence of a contentful response from the student in line 7, and the nomination of a candidate response for the student by the tester

SFK ch11

SFK ICBK027

316

April 21, 2008

15:27

Char Count=

Learner Competence as a Resource

(line 8) followed by a brief (and highly uncharacteristic) codeswitch by the student (line 9) reflect a dwindling engagement by the student in topic development, possibly owing to the student’s having reached the limits of his communicative abilities on this topic. At this point, the student is faced with the task of moving to the next “requirement” of the assessment, inviting the tester to join him to study for the final exam. As will become evident in the discussion of this excerpt, owing to the chatting activity that, by design, preceded the proffering of the invitation to the classmate, there was no way for a student to predict or plan the moment in which he/she would be able to initiate the invitation activity nor the form that that activity might take. Moreover, since practice in inserting an invitation into an ongoing conversational interaction was not a central component of course coverage, the tester was effectively relying on the students’ ability to effectively do so by drawing on their experience as socially mature interlocutors, that is, their learner competence.

(4) 1 T: de, ano, doo desu ka? sono shigoto‘So, um, how is it? your job-’ 2 T: [ano sono arubai[to ‘um your part-time work’ 3 S: [um [maamaa. ‘Um’ ‘so-so’ 4 T: maamaa desu ka ‘So-so?’ 5 S: taihen desu. ‘It’s tough.’ 6 T: nihonjin kimasu ka? yoku kimasu ka? ‘Do Japanese come? Do you get a lot of (Japanese nationals)?’ 7 S: um ‘Um’ 8 T: tokidoki? ‘Sometimes?’ 9 S: yeah. [ee ee. tokidoki. ‘Yeah. Yeah yeah. Sometimes.’ 10 T: [watashi mo ‘Me, too/my place, too.’ 11 (.)  12 S: um. um. suiyoobi no gogakki, juuyokka, um um Um. Um. Wednesday May 14th , um um 13 tesuto ga aru n desu. ‘There is a test.’

SFK ch11

SFK ICBK027

April 21, 2008

15:27

Char Count=

Japanese Applied Linguistics

 14  15 16 17 18

317

T: a, soo ↓desu [ka. ‘Oh, is that right.’ S: [ee- desu yo ne. ‘We have a test don’t we.’ T: ( )- soo ne! ( ) soo desu ne, kimatsu ‘( )-that’s right! ( ) that’s right, the final’ shiken desu ne. ‘exam.’ S: ee um. issho ni benkyoo shimashoo ka12 ‘Yeah um. Wanna study together?’

Unprompted by the tester, he initiates a frame for the invitation activity by nominating the final exam for their class as the next topic of talk (lines 12–13)—although the identity of the exam is not explicitly mentioned, it is potentially recoverable based on the specificity of the student’s formulation of the timeframe of the event (gogakki is a misproduction of gogatsu). The move itself is positioned to accomplish a pragmatic function similar to the preinvitation turn marked by n desu kedo in the instructed invitation sequence. However, the form of the turn fails to index the shared epistemic stance that would have rendered this turn interpretable as a pre-announcement of a pending invitation. The tester responds with an interest marker (line 14), taking up the student’s turn as constructed—i.e., as a declarative statement about a domain unfamiliar to the tester that is being presented as a next topic of conversation, and thereby holding the student accountable to the pragmatic meaning conveyed (rather than the pragmatic meaning that might have been inferred from the shared institutional knowledge about “what the student is supposed to do next” in the exam). As the tester produces the interest marker, the student orients to the missed alignment with the tester,13 and, in a partial overlap with the tester, repairs his prior utterance (line 15), reformulating it as a pre-announcement. The timing of the student’s move reflects both his careful monitoring of the tester’s response and his awareness of what conversational work must be accomplished at this moment in order to maintain an interactional space within which the invitation activity can be effectively and economically pursued. While his knowledge of Japanese linguistic resources (i.e., reaction tokens, interactional markers) clearly contributes to the success of the student’s repair, it is clear that his ability to manage the contingencies of such interactional moments (i.e., recognizing a moment of miscommunication and initiating repair in a timely manner) draws on a socially mature knowledge of discourse organization, a knowledge that, I am arguing, is derived from his learner competence.

SFK ch11

SFK ICBK027

318

April 21, 2008

15:27

Char Count=

Learner Competence as a Resource

Discussion: Expanding the Possibilities in Language Assessment In the examples above, I have demonstrated that, consistent with findings reported in the literature on student participation in classroom instructional activities, learner competence is also evident in student performance in high stakes assessment activities. However, the results of this study make it clear that, in the context of such high stakes assessment activities, students access their learner competence in decidedly different ways and with decidedly differing interactional outcomes depending on the organization of the assessment activity and the nature of the classroom training received by the student. The learners who received instruction that (a) raised their awareness of speaking as “doing”, (b) provided them with resources for talking about speaking as “doing”, and (c) encouraged them to “say what you want to say” as a means of developing their sense of themselves as speakers of Japanese drew upon the resources afforded them through their learner competence in ways that heightened their ability to manage both the organization of the target tasks, and the stances and social identities they negotiated with the tester in the course of engaging in those tasks. In contrast, students who received instruction that focused on mastery of a subset of the patterns and practices presented in the course materials as a measure of L2 achievement were far less likely to draw upon the resources afforded them by their learner competence in conjunction with either of these two dimensions of the exam. Moreover, even in the moments when the control group students did draw upon their learner competence, their resulting talk was oriented predominantly toward their participation in the interaction as a relatively scripted accomplishment of the task. While we may attribute these differences in participation to the distinct instructional conditions in which the students were taught, we must also consider the possibility that these differences arise, at least in part, from a far simpler condition: the organization of the assessment activity itself. It is clear from the data that the students from the experimental groups engage more extensively in the creating and negotiating of social identities, in coconstructing topic shift and topic maintenance, and in displaying active and involved listenership. On the other hand, it is also readily evident from a comparison of the task descriptions that these students are asked to do more. Specifically, the control group is asked to “invite” and “negotiate the details” of an invitation with a “classmate,” while the experimental group is asked to “greet, chat, and invite” a classmate with whom “you do not have a close relationship, but whom you would be interested in getting to know better.” The wording and organization of the control group assessment tasks highlight in a very narrow manner conversational patterns, practices, and identities that are modeled in the textbook. Students are effectively being

SFK ch11

SFK ICBK027

April 21, 2008

15:27

Char Count=

Japanese Applied Linguistics

319

asked to reproduce those dimensions of the textbook models as a means of displaying their mastery of them. What is problematic is that, in so doing, they are also, perhaps inadvertently, reproducing the faceless interlocutors and decontextualized talk of the textbook. In other words, by making the student accountable for Japanese language use as modeled in the instructional materials, the student’s performance is effectively constrained by the very real and unavoidable limitations of textbook talk. This is further illustrated by a second feature of all six of the control group students’ performances: the use of a vocative (e.g., (a,) Kado-san(!)) as a conversational opener for Task 2. The instructions for Task 2 provide no explicit requirement to “greet” the tester, yet each control group student initiated participation in the Task 2 invitation activity with a vocative, produced with an intonation that effectively rendered it a conversational opener. Notably, the vocative as opener was modeled repeatedly in the instructional materials. It was also modeled by the tester (albeit with additional variable features, such as the use of the attention-getter ne ne (‘hey’)) in her initiation of the “decline the invitation” activity of Task 1. This commonality across instructor and student talk in the control group provides further evidence that there was a shared understanding of “how to perform the assessment tasks” by the members of the control group, and that this standard of performance was grounded in the textbook models. It is important to note that, if there were no implicit textbook-based constraint on the control group students’ performances, then there would have been nothing to preclude these students’ spontaneous insertion of other components of talk (derived from their learner competence), including chatting components such as those seen among the experimental group students, in their performance of the assessment tasks. The absence of such displays of competence is consistent with the claims presented above that access to learner competence is mitigated against in FL classroom contexts, and especially in assessment contexts. The results from this study suggest that both the instructional approach and the organization of the assessment activity have been collaboratively constructed by the control group instructor and her students to limit production to a certain set of “in-bounds” ways of speaking. These students, then, are able to invoke learner competence to do more than is explicitly asked on the task card only when the additional “doing” has been modeled (and effectively scripted for them) in the instructional materials, or when invoking non-scripted learner competence (such as seen in Excerpt (2)) does not lead them too far afield from the scripted domain they are expected to (re-)produce. From this perspective, it becomes clear that one way to elicit more creative and diverse student performances in high stakes assessment activities is to develop assessment materials that require the student to produce a

SFK ch11

SFK ICBK027

320

April 21, 2008

15:27

Char Count=

Learner Competence as a Resource

performance that is more elaborated than the talk modeled in the instructional materials, and provide classroom instruction and practice opportunities that enable the student to develop an understanding of how to engage in the assessment activities in this less constrained manner of participation. As the student performances in Excerpts (3) and (4) above have illustrated, the domain of learner competence provides a rich and readily available pool of resources for guiding and supporting this enhanced participation. Student access to these resources may be facilitated by awareness-raising activities conducted as part of classroom FL instruction, but, importantly, it is the individual learner, rather than the homogenized textbook, that determines how and when this access will be realized. This ensures that the student, rather than the instructor (or the instructional materials), is accountable for investing her participation in the assessment activity with a sense of her own voice and communicative style, that is, what I have referred to above as the “speaking self.” Arguably, it is precisely differences in the students’ sense of accountability that underlies the distinct outcomes presented above. As students in the control group organize their participation in the assessment activity in ways that reflect their orientation to being accountable to the textbook standard, so too do their invocations of learner competence—the student in Excerpt (2) who manages unscripted aspects of talk and manages to stay “on track” in producing the scripted invitation sequence and the consistent use of the vocative to initiate participation in Task 2—reflect that orientation. Similarly, the invocations of learner competence by students in the experimental group occur in precisely those places where the contingencies of the interaction are not modeled in the textbook materials and, consequently, instructional support for the accomplishment of the required interactional outcomes of the assessment scenario is most tenuous. Unscripted student use of a diverse range of interactional markers (yo, ne, yo ne, n desu, and ja) to manage these critical interactional moments occur in precisely those places where the student initiates discourse organizational moves that will enable him/her to enlist the tester in moving toward the joint accomplishment of some required dimension of the assessment activity (i.e., the establishing of common ground with an as-yet-unfamiliar peer, the proffering of an invitation, the pursuit of the acceptance of an invitation). These dimensions of the students’ performances reflect the instructional emphasis (in the experimental group) on understanding language use as “doing”, where “doings” such as establishing a social relationship, organizing the flow of talk, and accomplishing a task were all construed in terms of “doing with” or “doing to” someone. The student’s talk was not held accountable to a stable text-bound standard, but rather to a socially competent, inherently variable, feeling and thinking “other person.”

SFK ch11

SFK ICBK027

April 21, 2008

15:27

Char Count=

Japanese Applied Linguistics

321

Conclusion In this study, I have argued that a domain of learner competence, evidence of which has been widely noted in the literature on learner interaction in foreign language pedagogical activities, can be recruited as a valuable resource for learner development and language use in high stakes assessment activities. Using an experimental instructional approach and assessment practices that encouraged students to draw upon their learner competence, I demonstrated that this approach can facilitate learner’s L2 use in diverse and creative ways in the context of assessment-based role-play activities in the beginning JFL classroom. Through a comparison of control and experimental groups that were exposed to the same instructional materials, but that were instructed with differing emphases on being accountable to L2 use, I showed that student invocations of learner competence were constrained by the valued and expected pedagogical practices of the two instructional settings, respectively. Specifically, students in the experimental group took up a greater variety of discourse roles, and undertook a greater range of conversational actions and interactional stances as they managed the flow of phatic chat and information exchange associated with the assessment tasks. In sum, this study has shown that foreign language classroom instruction can target learner competence as a viable and significant resource for enhancing learner L2 development. By teaching students to draw upon their personal experiences, intuitions, and knowledge in conjunction with participation in foreign language classroom activities, we can train them to be accountable to their interlocutors in ways that move beyond L2 use as institutionally sanctioned “play”, and build toward their perception of themselves as “real” L2 users.

Notes 1. I am indebted to Junko Mori and Amy Ohta for their invaluable feedback on earlier versions of this chapter. I would also like to thank Robert Huey, Kazutoh Ishida, Tomoko Iwai, Emi Murayama, and Asuka Suzuki for their insights and assistance at various stages of this work. Finally, I would like to acknowledge the support of the UH-Manoa NRCEA during the early stages of this work. 2. The “ecological deficiencies” of the interactional setting in contexts of L2 oral proficiency examinations, and the very considerable impacts they have on the naturalness and organization of talk produced in these contexts are explored at length in Alderson and Banerjee (2002), McNamara and Roever (2006), and Young and He (1998).

SFK ch11

SFK ICBK027

322

April 21, 2008

15:27

Char Count=

Learner Competence as a Resource

3. During the pause, the instructor presents the card for Task 2 to the student and the student reads the instructions on the card before speaking. 4. The name provided on the task card is the instructor’s actual surname. This practice may blur the distinction between the instructor interacting in her role as tester and interacting in her role as “your classmate.” “Kado” is used here as a stand-in pseudonym. 5. The other student (with some high school JFL background) from the control group also attempted to produce the availability checking step prior to proffering the invitation, but produced neutral requests for information instead: Availability check: ashita nani o shimasu ka. ‘What will you do tomorrow?’ Invitation: Asat- asat- uh, eiga o- um mima- mimash- mimasu ↑ka (.) asatte desu? ‘Day after- day after- uh, a movie- um see- saw- will you see a movie? (.) the day after tomorrow?’ Notably, the tester did not hold the student accountable for the pragmatic functions of his turns (i.e., information requests) but responded to both turns as if they accomplished the pragmatic functions required for the student’s completion of Task 2; this reflects the mutual reliance of the participants on the scripting provided by the task cards to provide an interpretive frame that enables them to avoid communicative breakdown and possible non-completion of the requirements of the assessment. 6. It is worth noting that L2 learners’ use of the sentence-final particle ne in assessment turns such as ii desu ne has been demonstrated to be subject to pragmatically anomalous use by beginning, American Englishspeaking learners of Japanese (Yoshimi 1999). While the basis for many of the anomalous uses observed by Yoshimi (1999) was the learner’s failure to attend to the epistemic constraints on the use of the particle, there was also an example where the learner’s failure to use the proper aspectual form of the predicate created the basis for the anomalous use (Yoshimi 1999:1520). Here, a yokatta, an utterance that would signal the speaker’s orientation toward and investment in the current activity-inprogress, that is, the invitation sequence, would be more consistent with the sequential positioning of the learner’s turn as an assessment. 7. I am indebted to Asuka Suzuki for her valuable insights regarding this example. Any errors in the analysis remain my own. 8. The learner’s use of de to shift the invitation-in-progress into the next step is also pragmatically anomalous; this substitution of de for jya is a common pragmatic error among beginning JFL learners (Junko Mori, personal communication).

SFK ch11

SFK ICBK027

April 21, 2008

15:27

Char Count=

Japanese Applied Linguistics

323

9. A more detailed presentation of these instructional procedures and their theoretical underpinnings were presented in Yoshimi (2003), and will be provided in Yoshimi (in preparation). I note that although these instructional procedures were developed independently, some components of this approach incidentally share some features of “phases” or “steps” in pragmatics-focused instructional procedures developed in work by Barraja-Rohan (1997) and Huth and Taleghani-Nikazm (2006). 10. It is my contention (Yoshimi in preparation) that the perceived difficulty of instructing learners in everyday conversational practices arises from the absence of a shared way of “talking about talk.” Institutionally validated (“It’s in the textbook.”) and pedagogically valued (“Today we will learn about {passive sentences, intransitive verbs, etc.}.”) terms for grammatical functions and parts of speech, often gleaned from the jargon of linguists, provide a source of shared terms and concepts that render these domains (i.e., “grammar” and “vocabulary”) relatively “easy to teach.” If we establish a shared set of terms for talking about the domains of conversational organization and self-presentation (e.g., stance, identity construction), they will likewise frame these domains as institutionally valid, pedagogically valued, and, consequently, “easy to teach.” 11. Flyers for on-campus activities, or activities learners had mentioned during the semester were selected to maintain an ecological perspective (van Lier 2002) on the incorporation of the prompts. 12. Having established shared reference with the tester (line 14), the student proffers the invitation using a form that reflects very low social distance and strong solidarity with the tester (in the role-played role of classmate). Students were instructed not to use this form unless there was strong certainty that the invitee would accept the invitation; the extensive rapport-building work that precedes this invitation arguably supports the student’s use of this more abrupt form of invitation. 13. Although the lexical difference between an interest marker and a marker of shared alignment is not evident until the turn-completion point, there is a difference in the height of the initial pitch of the intonation contour—the interest marker has a lower initial pitch—that would cue the nature of the projected turn.

References Alderson, J. C. and Banerjee, J. (2002) “Language testing and assessment (Part 2).” Language Teaching 35, 79–113.

SFK ch11

SFK ICBK027

324

April 21, 2008

15:27

Char Count=

Learner Competence as a Resource

Bannink, A. (2002) “Negotiating the paradoxes of spontaneous talk in advanced L2 classes.” In C. Kramsch (ed.), Language Acquisition and Language Socialization. London: Continuum, pp. 266–288. Barraja-Rohan, A. (1997) “Teaching conversation and sociocultural norms with conversation analysis.” Teaching Languages, Teaching Culture 14/15, 71–88. Beach, W. (1993) “Transitional regularities for ‘casual’ “Okay” usages.” Journal of Pragmatics 19, 325–52. Belz, J. A. (2002) “The myth of the deficient communicator.” Language Teaching Research 6, 59–82. Davies, C. E. (1998) “Maintaining American face in the Korean oral exam: Reflections on the power of cross-cultural context.” In R. Young and A. W. He (eds.), Talking and Testing: Discourse Approaches to the Assessment of Oral Proficiency. Amsterdam: John Benjamins, pp. 271–296. DiNitto, R. (2000) “Can collaboration be unsuccessful? A sociocultural analysis of classroom setting and Japanese L2 performance in group tasks.” Journal of the Association of Teachers of Japanese 34, 179–210. Donato, R. (2000) “Sociocultural contributions to understanding the foreign and second language classroom.” In J. P. Lantolf (ed.), Sociocultural Theory and Second Language Learning. Oxford: Oxford University Press, pp 27–50. Gardner, R. (2007) “The right connections: Acknowledging epistemic progression in talk.” Language in Society 36, 319–341. Huth, T. and Taleghani-Nikazm, C. (2006) “How can insights from conversation analysis be directly applied to teaching L2 pragmatics?” Language Teaching Research 10(1), 53–79. Ikeda, K. (2004) “‘Listenership’ in Japanese: An examination of overlapping listener response.” University of Hawai‘i-Manoa: Second Language Teaching and Curriculum Center. Accessed June 6, 2007 at www.nflrc.hawaii.edu/networks/NW32.pdf. Jucker, A. H. and Smith, S. W. (1996) “Explicit and implicit ways of enhancing common ground in conversations.” Pragmatics 6(1), 1–18. Kasper, G. (2004) “Participant orientations in German conversation-forlearning.” Modern Language Journal 88(4), 551–567. Lantolf, J. P. and Genung, P. B. (2002) “‘I’d rather switch than fight’: An activity-theoretic study of power, success, and failure in a foreign language classroom.” In C. Kramsch (ed.), Language Acquisition and Language Socialization. London: Continuum, pp. 175–196. Lee, Y. (2006) “Towards a respecification of communicative competence: Condition of L2 instruction or its objective?” Applied Linguistics 27(3), 349–376.

SFK ch11

SFK ICBK027

April 21, 2008

15:27

Char Count=

Japanese Applied Linguistics

325

Lim, D. (1996) “Cross-cultural interaction and classroom discourse: A study of foreign language classroom culture.” Unpublished M.A. thesis. University of Hawai‘i-Manoa, Honolulu. LoCastro, V. (2001) “Individual differences in second language acquisition: Attitudes, learner subjectivity, and L2 pragmatic norms.” System 29, 69– 89. McCarthy, M. and O’Keeffe, A. (2004) “Research in the teaching of speaking.” Annual Review of Applied Linguistics 24, 26–43. McNamara, T. and Roever, C. (2006) “The social dimension of proficiency: How testable is it?” Language Learning 56(s2), 43–79. Mondada, L. and Pekarek Doehler, S. (2004) “Second language acquisition as situated practice: Task accomplishment in the French second language classroom.” Modern Language Journal 88(4), 501–518. Mori, J. (2002) “Task design, plan, and development of talk-in-interaction: An analysis of a small group activity in a Japanese language classroom.” Applied Linguistics 23(3), 323–347. Mori, J. (2005) “Why not why? The teaching of grammar, discourse, sociolinguistic and cross-cultural perspectives.” Japanese Language and Literature 39(3), 255–289. Nakahama, Y., Tyler, A. and van Lier, L. (2001) “Negotiation of meaning in conversational and information gap activities: A comparative discourse analysis.” TESOL Quarterly 35, 377–405. National Standards in Foreign Language Education Project (1996) National Standards for Foreign Language Learning: Preparing for the 21st Century. Yonkers, N.Y.: American Council on the Teaching of Foreign Languages. Ochs, E. (1993) “Constructing social identity: A language socialization perspective.” Research on Language and Social Interaction 26(3), 287–306. Ochs, E. (2002) “Becoming a speaker of culture.” In C. Kramsch (ed.), Language Acquisition and Language Socialization. London: Continuum, pp. 99–120. Seedhouse, P. (2005) “Task as research construct.” Language Learning 55(3), 533–570. Shimazu, M. (2000) “Co-construction of classroom identity in Japanese as a foreign language classrooms.” Unpublished M.A. thesis. University of Hawai‘i-Manoa, Honolulu. Storch, N. and Wigglesworth, G. (2003) “Is there a role for the use of L1 in an L2 setting?” TESOL Quarterly 37(4), 760–770. Sullivan, P.N. (2000) “Playfulness as mediation in communicative language teaching in a Vietnamese classroom.” In J. P. Lantolf (ed.), Sociocultural Theory and Second Language Learning. Oxford: Oxford University Press, pp. 115–131.

SFK ch11

SFK ICBK027

326

April 21, 2008

15:27

Char Count=

Learner Competence as a Resource

Tracy, K. and Naughton, J. M. (2000) “Institutional identity-work: a better lens.” In J. Coupland (ed.), Small Talk. Harlow: Pearson Education, pp. 62–83. van Lier, L. (1988) The Classroom and the Language Learner: Ethnography and Second-Language Classroom Research. London: Longman. van Lier, L. (2000) “From input to affordance: Social-interactive learning from an ecological perspective.” In J. P. Lantolf (ed.), Sociocultural Theory and Second Language Learning. Oxford: Oxford University Press, pp. 245– 260. van Lier, L. (2002) “An ecological-semiotic perspective on language and linguistics.” In C. Kramsch (ed.), Language Acquisition and Language Socialization. London: Continuum, pp. 140–164. Yoshimi, D. R. (1999) “L1 language socialization as a variable in the linguistic expression of affect by L2 learners of Japanese.” Journal of Pragmatics 31, 1513–1525. Yoshimi, D. R. (2001) “Culture/Usage notes.” In T. Iwai, Y. Wada, J. Haig, and D. Yoshimi (eds.), Japanese for Oral Communication: A New Approach to Japanese Language and Culture. Department of East Asian Languages and Literatures, University of Hawai‘i-Manoa, Honolulu. Yoshimi, D. R. (2003) “Lectures presented to the 2003 Japanese Summer Institute on ‘Pragmatics in the JFL classroom: Advanced Training’.” University of Hawai‘i-Manoa, Honolulu. August 4–7, 2003. Yoshimi, D. R. (in preparation) “Developing interactional competence in the Japanese as a foreign language classroom” (tentative title). Manuscript. University of Hawai‘i-Manoa, Honolulu. Young, R. and He, A. W. (eds.) (1998) Talking and Testing: Discourse Approaches to the Assessment of Oral Proficiency. Amsterdam: John Benjamins.

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Chapter 12

Critical Approaches to Teaching Japanese Language and Culture

Ryuko Kubota

Introduction Teaching and learning Japanese as a second/foreign language is affiliated with larger fields such as linguistics, applied linguistics, and education. These fields have taken a critical turn during the past 20 years or so. The recent critical turn in the field of teaching English to speakers of other languages (TESOL), for instance, is described by Kumaravadivelu (2006a: 70):1 During the 1990s, the TESOL profession took a decidedly critical turn. It is probably one of the last academic disciplines in the field of humanities and social sciences to go critical. Simply put, the critical turn is about connecting the word with the world. It is about recognizing language as ideology, not just as system. It is about extending the educational space to the social, cultural, and political dynamics of language use, not just limiting it to phonological, syntactic, and pragmatic domains of language usage. . . . It is about creating the cultural forms and interested knowledge that give meaning to lived experiences of teachers and learners. Likewise, foreign language education in North America has begun to embrace critical perspectives by discussing topics such as sociopolitical issues in foreign language instruction, critical pedagogies, constructed foreignness, linguistic and cultural norms (e.g., Castilian Spanish vs. Chicano Spanish), and so on (e.g., Kubota et al. 2003; Leeman 2005; Reagan and Osborn 2002; Osborn 2000).2 In the field of Japanese language teaching and learning, such critical perspectives remain quite marginal in North

SFK ch12

SFK ICBK027

328

April 21, 2008

15:29

Char Count=

Critical Approaches to Teaching Japanese

America, although there have been some discussions of critical pedagogies, critical approaches to teaching culture, and reevaluations of gendered and honorific expressions (e.g., Kubota 1996, 2003; Matsumoto and Okamoto 2003; Okamoto and Shibamoto Smith 2004; Okamoto and Siegal 2003; Tai 2003). Critical approaches provide renewed perspectives on language, culture, and identities and provide tools for viewing these categories as dynamic and diverse rather than static and unitary. Yet there are multiple interpretations and approaches to critical applied linguistics. Some of them can be called postfoundational approaches—namely, postmodernism, poststructuralism, and postcolonialism (see St. Pierre and Pillow 2000). These approaches question the established knowledge constructed by positivism, enlightenment, and Eurocentrism which seek to discover universal truths through scientific approaches to research. Critical scholarship has thrown skepticism on taken-for-granted views of the world, revealed how essentialized knowledge about racial, cultural, class, and gender difference is constructed by discourse and power (rather than existing a priori), and sought possibilities for recreating equitable social relations and alternative ways of using language. Focusing on critical applied linguistics, this chapter aims: (1) to introduce different orientations and key issues of this inquiry through an overview of different interpretations of being critical and (2) to explore how three key topics in Japanese language teaching—language, ethnicity, and culture— can be conceptualized critically by drawing on some of the concepts and frameworks of critical applied linguistics. But first, I present a recent online debate among Japanese language professionals as an example that manifests dominant and yet conflicting assumptions about language.

“Mispronounced Japanese Words” Recently, there was a series of lively discussions on SenseiOnline, a listserv for teachers of Japanese language and culture, on the subject of “mispronounced Japanese words.” The individual who posted the original inquiry was in the United States, preparing a brief presentation on conventional pronunciation of Japanese words (e.g., Kyoto, Hiroshima), and asked the list subscribers to provide more examples, although the author never mentioned the audience of the presentation (e.g., students learning Japanese or general public).3 Of over 30 responses, the majority thought that the project was intriguing and offered examples ranging from words commonly heard in English (e.g., karaoke, origami, futon) and phrases (e.g., konnichi wa) to proper nouns, including product names (e.g., Pokemon, Toyota, Nissan) and personal names (e.g., Aoki, Naomi). A few individuals, however, pointed

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Japanese Applied Linguistics

329

out that these words are in fact English when they are used in English sentences, arguing that they are “correctly pronounced English words that come from Japan where they are pronounced differently.” They argued that pronunciation should vary depending on the audience (i.e., English speaker with no knowledge of Japanese vs. Japanese speaker) and that insisting on the original pronunciation is likely to result in negative reactions from the local community of speakers. Nonetheless, others insisted that words such as karaoke are indeed Japanese words and they should be pronounced using the Japanese standard no matter where the speaker is located. Furthermore, words of foreign origin in any language should sound as close to native sounds as they can. However, contradictions became apparent in the argument supporting preservation of the original pronunciation; as one of the subscribers pointed out, this argument leads to the conclusion that, contrary to the instructional practice in the Japanese language classroom, pronouncing “hamburger” as hanb¯ag¯a is incorrect in contexts where Japanese is spoken because it mispronounces the original English word. Moreover, if we take this position faithfully, karaoke, a word shortened from kara (‘empty’) o¯kesutora (‘orchestra’), should be pronounced as “kara orchestra” because o¯kesutora comes from English. Although respondents had opposing opinions, they seemed to share a feeling of annoyance attached to “mispronounced Japanese words”; those who supported the situational pronunciation still offered some examples. This example demonstrates two different views of language: normative and relativist. The normative view supported the universality of the original pronunciation regardless of the context, whereas the relativist view recognized the fluid nature of standards across languages. These two views differ in the level of perceived rigidity of the linguistic norm, although both presuppose the existence of a certain norm within a language. Yet, as explored in the following sections, there is another view that problematizes the very concepts of norm and language.

Critical Applied Linguistics In order to understand what critical applied linguistics is about, it is necessary to examine what “critical” means. The term “critical” carries different meanings in different inquiry areas within applied linguistics and related fields. Pennycook (2004) delineates four different understandings of the word “critical” in applied linguistics: (1) critical thinking, (2) social relevance (focus on social categories and contexts), (3) emancipatory modernism (critique of ideology and politics as seen in critical discourse analysis or CDA), and (4) problematizing practice (critical applied linguistics

SFK ch12

SFK ICBK027

330

April 21, 2008

15:29

Char Count=

Critical Approaches to Teaching Japanese

informed by postfoundational thoughts). These varied understandings of the term “critical” generate theoretically and politically different stances. First, teaching that seeks to foster critical thinking skills is concerned with the development of cognitive ability of individuals to engage in so-called higher-order thinking skills such as application, analysis, synthesis, and evaluation. In this understanding, there is no direct link between being critical and being able to critique sociopolitical dimensions of society and relations of power. Second, the focus on social contexts and social categories, as seen in inquiries on the relationship between language teaching/learning and issues of gender, regional or contextual differences, and so forth is often based on liberal pluralism and constructivism. While this focus may or may not view the relationship between language and society as problematic, CDA, the third approach, explicitly questions how language constructs and legitimates social inequalities. Through critical analysis of texts, CDA scrutinizes and uncovers unequal relations of power that structure and reproduce various forms of social injustice. Yet, following a neo-Marxist approach to science, CDA’s ideology critique is based on the assumption that ideologies reflected in texts are false consciousness or a misled understanding of reality and that the goal of analysis is to replace ideology with objective reality (Pennycook 2001, 2004). Another area of inquiry that falls into the third category is language policy and planning, which investigates issues related to standardization of language, designation of official language(s), and policies on the media of instruction (see Ricento 2006; Shohamy 2006; Spolsky 2004; Wright 2004) as well as linguistic imperialism (Phillipson 1992) and linguistic human rights (Skutnabb-Kangas 1998). However, there is nothing inherently critical about such scholarly investigations; what qualifies this scholarship as critical is the exploration of how power is exercised to produce differential access and privilege (Pennycook 2004). Moreover, from the point of view of the fourth approach (i.e., the perspective of problematizing practice), critiques of linguistic imperialism and linguistic genocide tend to see the relationship between domination and subordination as fixed. A less deterministic approach is to view language users as capable of appropriating the language of power to express voices of resistance and to perform identities (cf. Butler 1990), rather than simply being dominated by it. While the third approach tends to attract attention in current critical inquires in applied linguistics, the fourth approach—critical applied linguistics as problematizing practice—draws on postfoundational inquiries that reject the existence of objective, transcendent, and universal truths. These inquiries also acknowledge the fluid and plural nature of language, culture, and identities, while problematizing power, knowledge, and discourse. This approach explores how identities and agencies are performed,

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Japanese Applied Linguistics

331

rather than determined by closed categories of language, gender, ethnicity, and sexual identities, and how a plurality of meanings can be achieved in social, educational, and political contexts. It also explores how language can be appropriated and bent to express alternative meanings and identities. At the core, this approach problematizes, not just examines, the relationship between language and society for the purpose of social critique and transformation. As such, it raises “critical questions to do with access, power, disparity, desire, difference, and resistance” (Pennycook 2004: 797) and echoes the postcolonial project of problematizing and recasting canonical knowledge and master narratives constructed and perpetuated by the hegemony of colonialism (Luke 2004). In this view, the very concept of language itself needs to be problematized. While the system of language has typically been described and analyzed within a positivistic framework which assumes a singular objective reality, a postmodern framework views language as a social and political construction, shifting the definition and boundary of a certain language in time and place (Reagan 2004). In other words, “the specific demarcation of distinct languages is fundamentally an arbitrary one” (Reagan 2004: 45). The fourth approach to critical applied linguistics, however, risks falling into plain relativism, approving all views as equally valid, or endorsing another dogmatic commitment to a single perspective. Thus, advocates of this approach emphasize the importance of situated ethics as well as constant self-reflection through problematizing all versions of knowledge including the ones supporting critical applied linguistics. Of these four approaches, the first one on critical thinking and the second one with a mere focus on sociocultural elements involve little political commitment in their inquiries, while the other two approaches share an explicit sociopolitical focus despite significant differences between them. All in all, what can be regarded as a critical approach calls into question how language and identities are related to larger political processes.4 One major focus of applied linguistics research is on language teaching and learning. How do these critical approaches inform practice? In the next section, I will briefly discuss critical pedagogies.

Critical Pedagogies Perhaps the most influential figure associated with critical pedagogies is Paulo Freire, a Brazilian educator/scholar who engaged in literacy campaigns for the poor in Brazil and other parts of the world. In short, critical pedagogies aim to raise students’ awareness of their own oppressed or privileged position, which is constructed and maintained by larger sociopolitical

SFK ch12

SFK ICBK027

332

April 21, 2008

15:29

Char Count=

Critical Approaches to Teaching Japanese

and economic relations of power, and to empower them to become agents for transforming inequitable conditions. Critical awareness is to be raised through problem posing and dialogues to read both the word and the world (Freire and Macedo 1987) rather than through what Freire (1998) calls banking education which simply transmits knowledge from teachers to the empty heads of students. Freirean critical pedagogy has merged with the Marxist approach to education promoted in the United States by scholars such as Michael Apple, Henry Giroux, Joe Kincheloe, Donaldo Macedo, Peter McLaren, Ira Shor, Christine Sleeter, and others. Central tenets of critical pedagogy are discussed by Kincheloe (2005). Underlying critical pedagogy is the belief that teaching and learning are inherently political; that is, no educational practices are free of politics, including those attempting to minimize or neutralize the political dimensions of teaching and learning. Curriculum, instruction, and educational policies often legitimate certain knowledge that perpetuates unequal relations of power. Critical pedagogies seek to establish social justice and equality, aiming to eradicate victimizations caused by racial, class, gender, and other kinds of differences. In order to do this, teachers as intellectuals need to understand the complexity of the context in which they work and seek self-reflective and situated critical practice by “always searching for new voices that may have been excluded by the dominant culture or by critical pedagogy itself ” (Kincheloe 2005: 24). Principles of critical pedagogy have been adopted in other inquiry fields within education. For instance, the field of multicultural education has critically reflected on its liberal apolitical tendency to celebrate cultural differences in a superficial manner, as seen in the “heroes and holidays” approach. Critical multicultural education, thus, addresses how and why differences are constructed, essentialized, and exploited to create “us versus them” dichotomies and marginalize or privilege certain groups of people (Nieto 2004). This is a form of anti-racist education for all students and should permeate the entire curriculum. This perspective is quite relevant to the general field of second language education in which essentialized taken-for-granted cultural and linguistic differences need to be seriously scrutinized (Kubota 2004). Another area of inquiry is critical literacy. Synthesizing issues in critical academic literacy for English as a second language (ESL) learners, Canagarajah (2005) comments that critical scholarship has explored how students negotiate standard written language to express their own voice and how they question their own assumptions about what they read and write. The dynamic relationship between language and the act of reading and writing questions the uniform application of a teaching approach that ignores local situations and student needs. Popular instructional strategies such as the process approach to writing (as well as learner-strategy

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Japanese Applied Linguistics

333

training, task-based approach, and so on) may do a disservice to students when they are implemented without considering local conditions. In TESOL, the so-called post-method conditions have shifted the conceptualization of teaching methods from overdetermined prescriptions to more situationally relevant practice (Kumaravadivelu 2006b). Critiquing the imposed service role of teaching ESL for academic purposes at universities, Benesch (2001) questions the pragmatic orientation, which gives the needs of academic disciplines the highest priority, and advocates for seeking pedagogical goals of English for academic purposes in its own right and analyzing not only the needs but also the rights of the students. These topics are relevant to teaching Japanese. For instance, some have questioned the unmodified application of teaching methods developed for ESL or other foreign languages to teaching Japanese (Kubota 1998; Walton 1991). The service role of teaching English for academic purposes parallels the situation of Japanese language instruction at universities or high schools (i.e., perceived role of teaching language for the ultimate purpose of preparing students to read literature or enroll in university Japanese language classes). In sum, there are multiple approaches to critical engagement in applied linguistics, although not all qualify as critical because some lack an explicit focus on the sociopolitical dimension of language and culture. Some critical approaches engage in ideology critique but operate within a scientific paradigm, whereas others are based on postfoundational thoughts which question the existence of scientific truths. The postfoundational approaches further deconstruct taken-for-granted knowledge in relation to discourse and power, exploring multiple and fluid meanings of culture, language, and identities. As applied to teaching, critical pedagogies advocate problem posing and dialogues for raising awareness of everyday social conditions that produce domination/subordination and seek ways to transform such conditions. These conceptual frameworks can apply to inquiries into Japanese language teaching and learning. Of many inquiry topics, the subsequent sections will focus on three important themes that require critical reflection: namely, Japanese language, people, and culture.

Rethinking Japanese Language Recall the example of mispronounced Japanese words—there were two opposing positions: the normative/purist view and the relativist/situational view. The normative position assumes that one correct standard way of pronouncing a word in the original language should be used universally across other languages, whereas the relativist position acknowledges language shifts caused by borrowing and recognizes a different norm in another

SFK ch12

SFK ICBK027

334

April 21, 2008

15:29

Char Count=

Critical Approaches to Teaching Japanese

language. Yet another view would question the relativist assumption that a borrowed word should still follow the conventional pronunciation specified in the dictionary of that language (e.g., “carry-oh-key” for karaoke). This critique pays attention to diversity within the language as well as to the creative or performative potential of language use, while problematizing the modernist notion of the unitary language system and norm. Although the third position is aligned with critical applied linguistics, the predominant view among professionals falls under either of the first two positions that assume a certain type of norm. Teaching Japanese or any other language is influenced by normatism, which sets and maintains certain norms of various social practices. One realizes that the type of language taught as the model is a very specific variety when posing such questions as: Which variety of Japanese (regional, native/nonnative, etc.) is the ideal model for accent, vocabulary, and expressions? With regard to honorifics and gendered expressions, for instance, is there any gap between the way people actually use the language and the way the language is presented in instruction and in teaching materials? These questions reveal that the language presumably spoken by a native speaker of Tokyo dialect is assigned as the instructional model in textbooks and audio/video tapes. We also find that language variation is often taught in an essentialist way—for example, the difference between male speech and female speech is presented as static and dichotomous. Given the diverse nature of language, the choice of such specific models is not made arbitrarily. When asked the rationale for choosing Standard Japanese, many Japanese language professionals would respond that the standard language is acceptable and useful in most situations, thus learning how to use it correctly and appropriately would be beneficial for the learners and, in fact, that is what the students want. The support for the standard language of the native speaker is observed in the following situations I have personally encountered: hiring instructors (e.g., “We are concerned about Ms. X’s accent”—note Ms. X could be a native or nonnative speaker), rationalizing the current practice (e.g., “Learning how to speak and write as the natives do benefits the learner’s career goals”), casually talking about learning Japanese with nonprofessionals (e.g., “Japanese is such a difficult language. It’s best to learn it from a native speaker, isn’t it?”). It is important to understand, however, that this linguistic norm does not exist a priori; rather, it has been constructed historically, politically, and ideologically. Scholarship on Japanese language education has indeed called into question the historical and sociopolitical dimensions of the Japanese language norm. The following discussions will focus on critiques in two areas: (1) standard language and (2) invention of kokugo and nihongo.

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Japanese Applied Linguistics

335

Standard language Most learners and teachers of Japanese probably think that Standard Japanese is the best variety of the language to learn. However, they may not be aware that the so-called standard carries its own cultural and historical baggage. Matsumoto and Okamoto (2003) analyzed Japanese language textbooks commonly used in North America and observed the propensity for presenting so-called Standard Japanese as a single variety and as a neutral model. The problem here is not just the dismissal of the existing rich linguistic diversity but also the concealment of the ideological process of how a certain variety becomes a standard. Matsumoto and Okamoto (2003: 38–39) state: Further, we cannot overlook recent discussions about the ideological basis underlying the standardization of modern Japanese (Lee 1996; Yasuda 1999), which suggest that to regard the “standard” variety taught at school as acceptable to all Japanese is itself a political assessment based on the ideology of fostering standardization. We should be aware that for many speakers of regional varieties of Japanese, the historical memory still remains that this variety was imposed on them as a matter of national language policy by means that included severe punishments for students who used their own dialect, particularly during and before World War II (Yasuda 1999). The development of a standard language has indeed produced supremacy of what is considered to be a legitimate language, excluding and silencing speakers who lack legitimate language competence (Bourdieu 1991). The view of regional dialect speakers as illegitimate teachers is embedded in this discourse. The discourse that privileges the standard form of language as legitimate has lasted from the past to the present. During Japanese imperialism, Japanese language teachers recruited from various regions in Japan to teach in Japanese colonies (i.e., Taiwan and Korea) brought their own regional accents to the classrooms. Terakawa Kishio, who was known for his research on accents and for his theory that regarded the varieties of Japanese spoken by colonized Korean and Taiwanese people as Japanese dialects, argued in 1945 that because of these teachers’ regional accents, correct pronunciation was not taught to the students, resulting in the development of Korean and Taiwanese dialects of Japanese (Yasuda 1999). These teachers’ accents were viewed as incorrect and problematic, just as in the contemporary example, shown earlier about a teacher candidate’s accent. The assumption behind mispronounced Japanese words in the recent example parallels normative thinking observed during the nation building process in modern Japan, when Standard Japanese (hy¯ojungo)—later

SFK ch12

SFK ICBK027

336

April 21, 2008

15:29

Char Count=

Critical Approaches to Teaching Japanese

called common Japanese (ky¯ots¯ugo)—was constructed. During this process, regional dialects were considered obstacles to standardization, something to be remediated and eliminated. The deviant pronunciation of nonnative speakers in the Japanese colonies was seen as error when juxtaposed with the Standard Japanese and the deviant forms of language spoken by these nonnative speakers were regarded as dialects of Japanese (Yasuda 1999). Such a view integrates nonnative speakers’ utterances as part of the Japanese language system albeit erroneous, while defining the Standard Japanese as correct and legitimate and excluding the dialects as illegitimate. A parallel is found between this view and the contemporary perception that pronouncing karaoke as “carry-oh-key” is mispronunciation. While the spread of the Japanese language during the first half of the twentieth century was motivated by imperialism, the contemporary spread of the language is mainly related to Japan’s economic expansion and the rise of globalization in the 1980s. This has prompted some discussions of internationalizing the Japanese language (e.g., Suzuki 1995). However, compared to the recent discussions of English as an international lingua franca which have called into question Anglo-centric linguistic norms or postcolonial appropriation of the former colonial language to express alternative voices, the norm of Standard Japanese has not been scrutinized. Instead, the discussions of internationalizing Japanese tend to focus on sending information about Japan (e.g., culture and society) to the world. For instance, Suzuki (1995) argued that internationalizing Japanese requires valuing the beauty and usefulness of the language, becoming proud of it, and using it as a weapon to express oneself in the international community. Here, there is no attention to the possibility of diversifying and affirming linguistic forms used by nonnative speakers of Japanese; rather the focus is solely on national benefits. As Yasuda (2003) contends, such an argument is rooted in the persistent nationalistic discourse hidden in language policies since the Meiji period—they have placed an emphasis on the nation state over the individual or on civic control over freedom. Clearly, internationalization of Japanese has not been discussed in the framework of internationalizing Japanese (e.g., transforming and reinventing the norms) but rather as the relationship between internationalization and Japanese (Yasuda 2003). The lack of interest in internationalizing Japanese seems to be reflected in the linguistic patriotism as observed in the popularity of best-sellers by Ohno Susumu (Nihongo Rensh¯u Ch¯o ( Japanese Language Practice Book)) and Sait¯o Takashi (Koe ni Dashite Yomitai Nihongo (Japanese that You Want to Read Aloud)) (Komori 2003).5 Komori comments that the popularity of the former was perhaps caused by an increase in the number of foreign residents in Japan who spoke nonstandard Japanese, which raised interest among Japanese speakers in knowing about the correct usage of Japanese.

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Japanese Applied Linguistics

337

Invention of kokugo and nihongo The third view of mispronounced Japanese words questions linguistic norms and the very concept of language itself. While linguistic conventions exist in dictionaries and textbooks, they do not exist a priori but rather are imagined and invented. In other words, conventions did not exist before humans began using the language; rather, they have been invented and described as a closed system that can be distinguished from other systems. It follows that the Japanese language did not exist a priori but has been imagined as an ontological entity with a clear boundary and that this imagination has had a complicit relationship with the construction of the Japanese nation and culture as an imagined community (Lee 1996; Sakai 1994). Indeed the development of the concepts of kokugo (national language) and nihongo (Japanese language) demonstrates this point. The concept of kokugo was developed around the turn of the twentieth century as the language of the nation state that carried ethnic and cultural identity. In his 1894 lecture entitled Kokugo to Kokka to (A National Language and a State), Ueda Kazutoshi promoted the idea that the Japanese consisted of one unitary people as with a family and drew a direct link among the Japanese ethnic nation, Japanese soul, and shared language (see e.g., Tai 1999; Lee 1996; Yasuda 1999). To quote Tai (1999: 507), Ueda “linked this language to the spiritual blood of the Japanese from which the national polity of Japan (kokutai), ‘a moral concept that constituted the very essence of the state,’ drew its primary sustenance.” In this way, kokugo was invented as a language to unify the nation both linguistically with pure correct usage and spiritually with loyalty to the emperor and love of the nation. Kokugo together with the national spirit was imposed on the people in Japanese colonies. While the invention of kokugo was concerned with nation building, nihongo was invented in the course of language planning for Imperial Japan. Around 1940 when Japan expanded its dominance to Southeast Asia to build the Greater East Asia Co-Prosperity Sphere, the imperial government faced a new challenge of developing a different strategy for disseminating the Japanese language and spirit to the peoples in the occupied territories. This was partly due to the independence movement spreading in the region— it was necessary to support emancipation of the colonized from former European colonial languages and at the same time instill the imperial mind through the Japanese language. Consequently, local indigenous languages were supported while, at the same time, the Japanese language was given the role of lingua franca for mutual communication and conducting rituals, such as worshiping the Imperial Palace and hoisting the Japanese imperial flag. Here, the notion of nihongo as the common language of East Asia was

SFK ch12

SFK ICBK027

338

April 21, 2008

15:29

Char Count=

Critical Approaches to Teaching Japanese

invented (e.g., Yasuda 2003). Unlike kokugo, which had replaced the local languages in Japanese colonies so that the consciousness as a member of the nation state (kokumin seishin) could be disseminated, nihongo functioned as a common language of the region to spread the Japanese mind (nihon seishin). In this discourse, simplification of nihongo was proposed in order to enhance the communicative effectiveness, usefulness, and worldliness of Japanese. However, the simplified variety was conceptualized not as a legitimate or complete form of language spoken by nonnative speakers but rather as a form in transition toward pure Standard Japanese (Yasuda 2003). Although kokugo and nihongo had slightly different ideological aims, they both functioned to simultaneously assimilate and alienate the Other— people such as dialect speakers and colonial citizens. That is, the Other is alienated for using the linguistic forms that do not fit the conventions, while being pushed to conform to the conventions. This is demonstrated in the identification of nonnative forms of Japanese spoken in Korea and Taiwan as Japanese dialects. As Sakai (1994) argues, the system of grammar has two functions: producing linguistic correctness which privileges native fluency and rejecting hybrid and nonstandard usages among dialect speakers and nonnative speakers. This brief overview shows how the Japanese language was invented in the process of modernization and imperialism with the goal of unifying the nation and the empire. This demonstrates that the concepts of language and legitimate language are historical and political inventions. The will to establish and maintain a certain linguistic norm, as seen in the case of mispronounced words, needs to be understood in this past-present ideological continuum. Also invented is the concept of nihonjin (‘Japanese people’). The invention of nihonjin calls into question the taken-for-granted notion that language and ethnic identities are closely linked.

Rethinking the Japanese People As seen in the previous discussion, the invention of kokugo and nihongo aimed to ideologically unify the nation/empire by underscoring the inherent relationship among the Japanese language, people, and mind. However, contemporary scholars in many disciplines reject any intrinsic link between language and ethnic/cultural identity (May 2004). Citing Barth (1969), May (2004: 39) states: “primordial” accounts of ethnicity which argue that ethnicity is determined by particular objective cultural characteristics such as language, ancestry and history—what Barth (1969) has described as the “cultural

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Japanese Applied Linguistics

339

stuff” of ethnicity—is rejected out of hand as reified and essentialist. . . . On this view, ethnicity is about social relationships rather than specific cultural properties since “we can assume no simple one-to-one relationship between ethnic units and cultural similarities or differences” (Barth 1969: 14). This skeptical view of ethnicity raises a question about the definition of nihonjin. I will explore this issue from two angles: (1) the question of who is nihonjin and (2) the concept of native speakers.

Who is nihonjin? There is a pervasive and politically charged conception that the Japanese are a mono-ethnic group. Time and again, Japanese politicians make statements to this effect and minority advocacy groups voice their protest. When this notion is examined in a historical context, however, some intriguing discourses can be found. A study on Japanese ethnic identity by Oguma (1995) revealed a historical shift in dominant discourses. The end of World War II marked a drastic shift from a mixed-ethnicity theory to a mono-ethnic theory. As Japan expanded its imperial hegemony to its Asian neighbors around the turn of the twentieth century, it had to face the reality of both dominating and coexisting with local people in the region. The discourse that identified shared ethnic roots with colonized Koreas, for instance, became a convenient argument for ruling the colony through ethnic/cultural assimilation. Specifically, this mixed-ethnicity theory, as opposed to a view that restricts nihonjin to a narrowly defined ethnic group, posited that Japan’s long experience with multiethnic contacts, mixed blood with various ethnic groups in the broader Asian region, and the Emperor’s Korean heritage could expediently rationalize the superiority of the Japanese in their capacity to assimilate the people in the colonies and occupied territories.6 In contrast, postwar discourse has been dominated by a focus on the uniqueness of the Japanese for their mono-ethnic heritage with little contact with other ethnic groups. This presumably contributed to the establishment of a peaceful nation and yet at the same time the underdevelopment of diplomatic or defense capabilities. These conflicting self-images indicate the discursive nature of nihonjin—the idea of who nihonjin is has indeed been invented by competing discourses (i.e., the ways language is used through various media) rather than existing a priori as an objective fact. While the mono-ethnic theory has predominated in postwar Japan, demographics of this period do not support the theory. One rather simple yet intriguing model to show the diversity of nihonjin is proposed by Fukuoka

SFK ch12

SFK ICBK027

April 21, 2008

340

15:29

Char Count=

Critical Approaches to Teaching Japanese

Table 12.1 Typological framework of ‘Japanese’ and ‘non-Japanese’ attributes

Types

1

2

3

4

5

6

7

8

‘Lineage’ ‘Culture’ Nationality

+ + +

+ + −

+ − +

− + +

+ − −

− + −

− − +

− − −

Source: Fukuoka (2000: xxx)

(2000), who presented the typological framework of so-called Japanese people shown in Table 12.1. Although a typology of this kind might lead to overgeneralization and fixed categorization, it nonetheless provides a framework for recognizing diverse groups of people. According to Fukuoka, ‘lineage’ (or Japanese blood) and ‘culture’ are put in inverted commas to indicate that they are constructs rather than absolute reality. The table presents eight types of people: those who fit the widely held image of pure Japanese (Type 1), first-generation Japanese immigrants overseas (Type 2), the Japanese raised abroad such as returnee students (Type 3), naturalized citizens such as ethnic Koreans living in Japan (Type 4), third-generation Japanese emigrants and war orphans abroad (Type 5), assimilated zainichi Koreans (i.e., Korean residents in Japan) with Japanese upbringing (Type 6), indigenous people like the Ainu (Type 7), and so-called foreigners (Type 8). Fukuoka (2000) includes language as part of ‘culture.’ However, language and culture (or ethnicity) do not necessarily overlap as May (2004) suggests. If language were included in this typological framework as a separate category, there would be more variations and shades of the image of nihonjin. Moreover, there are no unitary speakers of Japanese—there is a range of proficiency and linguistic identities, which would make the framework even more complex. One widely held misconception is that nihonjin is a native speaker of Japanese and that the native speaker is the model for teaching and learning the language, a discourse that requires critical evaluation.

The native speaker It is often believed that students should learn to speak and write as native speakers do because the native speaker provides the best model for them. However, the native speaker construct has been seriously called into question. For instance, Kramsch (1997: 359–60) states:

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Japanese Applied Linguistics

341

Despite the spread of postmodern thought in the humanities and in many branches of linguistics and anthropology . . ., this idealization of the native speaker has not been put into question. But native speakers do not always speak according to the rules of their standard national languages; they display regional, occupational, generational, class-related ways of talking that render the notion of a unitary native speaker artificial. Moreover, whereas students can become competent in a new language, they can never become native speakers of it. Why should they disregard their unique mutilingual perspective on the foreign language and on its literature and culture to emulate the idealized monolingual native speaker? In fact, learners do not necessarily want to behave like native speakers, although they may need to know what perceived norms are (Spence-Brown 2001). Kramsch (1997) questions the practice of modeling after the idealized and standardized language used by the imagined native speaker and instead advocates for teaching a foreign language as a social and cultural practice for constructing new meanings and identities, rather than reproducing a predetermined prescriptive system and further reinforcing the norm. Many criticisms of the native speaker fallacy come from scholars in the field of TESOL (e.g., Amin 1999; Amin and Kubota 2004; Braine 1999; BruttGriffler and Samimy 1999; Kamhi-Stein 2004; Leung et al. 1997; Llurda 2005; Phillipson 1992). The past and present global spread of English with a legacy of (neo)colonialism has created many varieties of English spoken by diverse people in the world. As English is increasingly used as an international lingua franca, the nonnative/nonnative dyad in interaction is becoming more common, reconceptualizating the native speaker model in teaching pronunciation (Jenkins 2000). Although the question of who is the native speaker is increasingly becoming contentious, the superiority of the native speaker of English continues to be disseminated through texts and institutional practices. Simultaneously, the image of the illegitimate nonnative speaker prompts discrimination in hiring practices and biases among students (Amin 1999; Golombek and Jordan 2005). Recent scholarship has further moved beyond the linguistic dichotomy of native versus nonnative speakers and focused on how racialization intersects with language in constructing the idea of the legitimate teacher and excluding and marginalizing teachers of color (e.g., Asian American native speaker of English) (Kubota and Lin 2006; Lee in press). Some Japanese language teachers, especially those who teach privileged students in North American contexts, may argue that teaching Japanese as a foreign language is quite different from teaching English and that the native speaker fallacy or racialization would not apply. However, in the field

SFK ch12

SFK ICBK027

342

April 21, 2008

15:29

Char Count=

Critical Approaches to Teaching Japanese

of teaching Japanese, the native speaker model pervasive in many instructional materials constructs the idea of who the ideal speaker is, even though there is a significant level of diversity indicated in the typology offered by Fukuoka (2000). As a direct link among nation, ethnicity, and language becomes more illusive, the native speaker model no longer provides the answer to who the native speaker is in terms of ethnicity and nationality and with whom the students would be using Japanese. It is also necessary to remember that the legacy of Japanese colonialism and persistent racism and racial inequalities in Japan constructs contradictory yet consistent language policies that assimilate outsiders while excluding them at the same time. The notion of the idealized and racialized native speaker corroborates this mechanism—it encourages learners to reproduce the native speaker model while sending the message that one is a native speaker only if he/she is nihonjin with Japanese blood. The concept of native speaker is closely linked to the normative view of not only language and race/ethnicity but also culture. It is often believed that the native speaker is the bearer of authentic Japanese culture. However, the concept of Japanese culture has also been called into question in various fields.

Rethinking Japanese Culture Culture constitutes an important aspect of language teaching and learning. However, in the instructional process, culture tends to be constructed as essentialized fixed entities. In second language education, criticism against essentialist understanding of culture has increased (for teaching English, see e.g., Kubota 1999, 2001; Kumaravadivelu 2003; Pennycook 1998; Spack 1997; for teaching Japanese, see e.g., Heinrich 2005; Kawakami 1999; Kubota 2003; Tai 2003; Kono 2000). When we shift our attention from teaching Japanese to wider discourses on Japanese culture, we find a large amount of discussion of so-called nihonjinron, or theories on the uniqueness of the Japanese, and its criticism. In brief, nihonjinron emerged from both domestic and international interests in explaining Japan’s successful postwar economic reconstruction by focusing on cultural uniqueness, especially social ethnic harmony and homogeneity (Befu 2001; Sugimoto 1997; Yoshino 1992). In nihonjinron, images of Japanese people and culture are often compared only with those of the West, underscoring both positive and negative dimensions of Japanese cultural uniqueness.7 Critics of nihonjinron point out that these cultural images constitute a myth, signifying a discourse that has served the political and economic interests of the dominant group. That is, internationally this

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Japanese Applied Linguistics

343

discourse has been used for political and economic negotiation by explaining Japan’s unique position, while domestically it has functioned as ethnic cheerleading by providing positive self-images and has been exploited to prevent social conflicts by instilling the notion that Japanese culture values social harmony. In short, just like kokugo, nihonjo, and nihonjin have been (re)invented, Japanese culture has also been invented. The discourse of cultural uniqueness has also constructed the uniqueness of Japanese language and communication styles. Attributes such as indirectness, inexplicitness, silence, nonlogicalness, nonverbal and emotional tendencies, and so on corroborate the discourse of cultural harmony and homogeneity (see Kubota 2001). Again these attributes are mostly contrasted with European languages, which are viewed as logical, direct, explicit, and verbal, underscoring the difference. In applied linguistics, contrastive rhetoric research has reinforced such dichotomized and essentialized images of Japanese and English written texts (see Kubota 2002; Kubota and Lehner 2004). This cultural dichotomy reflects a legacy of colonialism and Orientalism (Said 1978), which has drawn rigid boundaries between the colonizer and the colonized in terms of language, race, and civilization and has been internalized by the colonized themselves (cf. Fanon 1967). The dual role of the actual colonizer and the symbolic colonized places Japan in a precarious and paradoxical position. However, it is clear by now that the discourse of cultural, linguistic, and ethnic uniqueness has been exploited to fulfill a political purpose of domination or resistance in multiple ways. The discursive nature of culture problematizes common teaching practices that assume the existence of culture (and language and ethnicity) as a single objectified entity. It also calls into question the commonly used instructional practice of asking students to compare and contrast the target culture with their own culture, because this approach objectifies and essentializes not only the target culture but also their own culture (Harklau 1999). The compare/contrast approach in learning a language has been promoted in the United States as seen in the National Standards (National Standards in Foreign Language Project 1999). In Japan, this approach is becoming popular as teaching nihon jij¯o (‘Japanese culture and issues’) has shifted its emphasis from the colonial teaching of the Japanese mind to helping learners develop intercultural competency or ability to negotiate the Japanese and their own cultures (Hasegawa 1999; Tanaka 2005). As Nishikawa (2002) states, Japanese language teaching is the field in which a large amount of nihonjinron is produced and consumed because of the cyclic demand for cultural information raised by students and teachers. A critical approach to teaching Japanese culture would need to problematize and deessentialize the persistent discourse that defines culture as a unitary, distinct, and objective reality.

SFK ch12

SFK ICBK027

344

April 21, 2008

15:29

Char Count=

Critical Approaches to Teaching Japanese

Toward Critical Pedagogies and Critical Japanese Applied Linguistics The above critical exploration of three main constructs in teaching and learning Japanese—Japanese language, people, and culture—shows that these constructs are all politically, ideologically, and historically invented, rather than reifying objective, unified, or permanent truths. These constructs produce discourses that circulate and generate certain knowledge, social practices, and relations of power. This, however, does not and should not lead to the argument that Japanese language, people, and culture do not exist. Such an argument merely corroborates a modernist dichotomy that views the world as consisting of two opposing binary categories. Rather, it is necessary to recognize the discursive nature of these constructs and continue to question all knowledge as well as its relation to the political and ideological processes that structure, maintain, and transform particular relations of power. It is also necessary to recognize how a certain research framework and assumption might suffer from limitations even if intended to challenge relations of power. For instance, arguments on issues such as linguistic imperialism, linguistic human rights, and language ecology (e.g., Phillipson 1992; SkutnabbKangas 1998) are founded on well-meaning ideology critique and social activism. However, they share the modernist assumption about language and culture as a closed system, human rights as a universal concept, and the relationship between language and ethnicity as static and primordial, minimizing the possibility of using language to create alternative linguistic forms, meanings, identities, and social realities (May 2004; Pennycook 2004). Another example is teaching Japanese culture. I proposed elsewhere four foci (or four Ds) in teaching Japanese culture critically: i.e., (1) descriptive, rather than prescriptive, understanding of culture, (2) diversity within culture, (3) dynamic nature of culture, and (4) discursive construction of culture (Kubota 2003). Focusing on the first three elements would clearly broaden students’ understanding of culture and language. For instance, examining how people from different regions in Japan speak or how ethnic cultures constitute diversity of Japanese culture would challenge the conventional knowledge about standard language or the myth of a mono-ethnic society. However, if such investigations are conducted with a modernist assumption of language and ethnicity with closed (yet multiple) boundaries, these critical analyses will not go far enough. In other words, examining or promoting linguistic or culture diversity in itself would merely create and affirm multiple complete systems of language and culture, each of which has shared rules and regularities. What critical applied linguistics offers is

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Japanese Applied Linguistics

345

a view of language and culture as a living organism that allows creative innovations for various personal, social, and political purposes. This, however, does not mean that the postfoundational critical approach is a panacea or superior paradigm. As Canagarajah (2005: 946) warns, “defining critical pedagogy or research in terms of a static set of rules or axioms, especially by scholarship removed from local practices, would lead to ossified paradigms that constitute a ‘p.c.’ (politically correct) approach.” The postfoundational approach of critical applied linguistics itself needs to constantly be put under scrutiny and reevaluated. At the same time, scholars and professionals need to be aware that a lack of praxis (i.e., committed action and reflection—Freire 1998) can easily turn the “critical” approach into a clich´e without any critical substance. Some Japanese language teachers and students in North America might comment, “Japanese colonialism has little to do with teaching or learning Japanese in North America.” However, colonialism exerts its power not only as a social, political, and economic system but also as a discourse. As reviewed earlier, the discourse of Japanese colonialism promoted the superiority of the Japanese and functioned to instill the Japanese mind into the colonized. Conversely, the postwar discourse or nihonjinron has a complicit relationship with Orientalism, a form of European colonialism, in emphasizing Japanese cultural uniqueness. Thus these discourses construct particular knowledge for students. Given that students will become future leaders in various professional communities, it is important for teachers to recognize their responsibility and role in either maintaining or transforming these discourses. As surveyed earlier, a major aim of critical pedagogies is to recognize the political nature of teaching and learning and transform the power relations that perpetuate social inequalities and injustices. Whether one disagrees or not, history attests the political nature of language, ethnicity, culture, and pedagogy, which constitutes the core of language education. Put differently, politics in teaching and learning Japanese is inescapable. To this effect, Tai (1999: 532) states: As the term Nihongo reemerges, many scholars engaged in Japanese education seem to be indifferent to its colonial history. In this they risk approximating the behavior of the politicians who have angered Japan’s neighbors by their refusal to acknowledge the outrages of imperialism. This being the case, classroom teachers of Japanese must be especially alert to the political significance of our work. This requires an awareness of the evolution of the concept of nihongo in modern Japanese history, lest

SFK ch12

SFK ICBK027

346

April 21, 2008

15:29

Char Count=

Critical Approaches to Teaching Japanese

we devote ourselves blindly to an unrecognized political goal superimposed on our personal objectives. The current political climate in Japan virtually pushes us to suspend our examination of the meaning of Japanese language, ethnicity, and culture. This is exemplified in the recently revised Fundamental Education Law (2006), in which such words as culture, tradition, nation, and homeland are used with no definitions. Upon encountering ostensibly uncontroversial statements, such as that a goal of education should be to value “tradition and culture” or to maintain “traditions,” critical Japanese applied linguists and teachers would question what these phrases really mean. This questioning is especially important because our past history shows a direct link between domestic imperial education policy and the practices of teaching Japanese language for imperial purposes overseas. Whatever stance we take, we are unable to escape politics. Even seemingly innocuous discussions of mispronounced Japanese words reflect contested ideological assumptions about linguistic norms that carry historical legacies. It is important to be aware that Japanese language professionals are indeed linguistic/cultural navigators directing the future of our field and the larger society.

Notes 1. For critical perspectives in the field of TESOL, see for example Benesch (2001), Canagarajah, (1999), Edge (2006), Morgan (1998), Nelson (2006), Norton and Toohey (2004), Pennycook (2001, 2004). 2. For European context, see for example Guilherme (2002). 3. It is difficult to judge the linguistic background of the original author and the respondents, but judging from their names, it seems that the original author was a nonnative speaker of Japanese and the respondents were both native and nonnative speakers. 4. Other inquiry fields within applied linguistics that have been problematized include testing (Shohamy 2001), English for academic purposes (Benesch 2001), and contrastive rhetoric (Kubota and Lehner 2004). 5. Published in 1999, Nihongo Rensh¯u Ch¯o provides nonspecialists with knowledge about the structures of Japanese language. Koe ni Dashite Yomitai Nihongo, published in 2002 is a collection of annotated classical and contemporary literary texts selected for the purpose of oral reading. 6. It is important to mention that there were divergent views about ethnicity during this period. For instance, Watsuji Tetsur¯o and Yanagita Kunio emphasized the mono-ethnicity of the Japanese by focusing on the uniqueness of monsoon climate and rice farming heritage, respectively, which influenced the postwar discourse on mono-ethnic theory (Oguma 1995).

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Japanese Applied Linguistics

347

7. Nihonjinron is represented by such work as Doi (1971), Nakane (1967), and more recently, Davies and Ikeno (2002).

References Amin, N. (1999) “Minority women teachers of ESL: Negotiating white English.” In G. Braine (ed.), Non-Native Educators in English Language Teaching . Mahwah, NJ: Lawrence Erlbaum Associates, pp. 93–104. Amin, N. and Kubota, R. (2004) “Native speaker discourse: Power and resistance in postcolonial teaching of English to speakers of other languages.” In P. Ninnes and S. Mehta (eds.), Re-Imagining Comparative Education: Postfoundational Ideas and Applications for Critical Times. New York: RoutledgeFalmer, pp. 107–127. Barth, F. (1969) “Introduction.” In F. Barth (ed.), Ethnic Groups and Boundaries: The Social Organization of Culture Difference. Boston: Little, Brown and Co., pp. 9–38. Befu, H. (2001) Hegemony of Homogeneity: An Anthropological Analysis of Nihonjinron. Melbourne: Trans Pacific Press. Benesch, S. (2001) Critical English for Academic Purposes: Theory, Politics, and Practice. Mahwah, NJ: Lawrence Erlbaum Associates. Bourdieu, P. (1991) Language and Symbolic Power . Cambridge: Polity Press. Braine, G. (ed.) (1999) Non-native Educators in English Language Teaching . Mahwah, NJ: Lawrence Erlbaum Associates. Brutt-Griffler, J. and Samimy, K. (1999) “Revisiting the colonial in the postcolonial: Critical praxis for nonnative-English-speaking teachers in a TESOL program.” TESOL Quarterly 33, 413–431. Butler, J. (1990) Gender Trouble: Feminism and the Subversion of Identity. London: Routledge. Canagarajah, S. (1999) Resisting Linguistic Imperialism in English Teaching. Oxford: Oxford University Press. Canagarajah, S. (2005) “Critical pedagogy in L2 learning and teaching.” In E. Hinkel (ed.), Handbook of Research in Second Language Teaching and Learning . Mahwah, NJ: Lawrence Erlbaum Associates, pp. 931– 949. Davies, R. J. and Ikeno, O. (2002) The Japanese Mind: Understanding Contemporary Japanese Culture. Boston: Tuttle Publishing. Doi, T. (1971) Amae no K¯oz¯o [Structure of Dependence]. Tokyo: K¯obund¯o. (Translated as The Anatomy of Dependence by John Bester, Tokyo: Kodansha International, 1973.) Edge, J. (ed.) (2006) (Re)Locating TESOL in an Age of Empire. Basingstoke: Palgrave Macmillan.

SFK ch12

SFK ICBK027

348

April 21, 2008

15:29

Char Count=

Critical Approaches to Teaching Japanese

Fanon, F. (1967) Black Skin White Masks. New York: Grove Weidenfeld. Freire, P. (1998) Pedagogy of the Oppressed. New York: Continuum. Freire, P. and Macedo, D. (1987) Literacy: Reading the Word and the World. South Hadley, MA: Bergin and Garvey. Fukuoka, Y. (2000) Lives of Young Koreans in Japan. Melbourne: Trans Pacific Press. Golombek, P. and Jordan, R. J. (2005) “Becoming ‘black lambs’ not ‘parrots’: A poststructuralist orientation to intelligibility and identity.” TESOL Quarterly 39, 513–533. Guilherme, M. (2002) Critical Citizens for an Intercultural World: Foreign Language Education as Cultural Politics. Clevedon, Avon: Multilingual Matters. Harklau, L. (1999) “Representing culture in the ESL writing classroom.” In E. Hinkel (ed.), Culture in Second Language Teaching and Learning. Cambridge: Cambridge University Press, pp. 109–130. Hasegawa, T. (1999) “‘Nihon jij¯o’: Sono rekishiteki tenkai [Japanese culture and issues: Past and present].” Nij¯u Isseiki no ‘Nihonjij¯o’ [‘Japanese Culture and Issues’ for the 21st Century] 1, 4–15. Heinrich, P. (2005) “Language ideology in JFL textbooks.” International Journal of the Sociology of Language 175/176, 213–232. Jenkins, J. (2000) The Phonology of English as an International Language. Oxford: Oxford University Press. Kamhi-Stein, L. (2004) Learning and Teaching from Experience: Perspectives on Nonnative English-Speaking Professionals. Ann Arbor: University of Michigan Press. Kawakami, I. (1999) “Nihon jij¯o ky¯oiku ni okeru bunka no mondai [Japanese culture and issues; Revised approach through ‘culture’].” Nij¯u Isseiki no ‘Nihonjij¯o’ [‘Japanese Culture and Issues’ for the 21st Century] 1, 16–26. Kincheloe, J. (2005) Critical Pedagogy Primer . New York: Peter Lang. Komori, Y. (2003) “Nihongo b¯umu to nashonarizumu [The Japanese language boom and nationalism].” Ky¯oiku 53(7), 4–8. Kono, R. (2000) “Senryaku teki ‘nihonbunka’ hi sonzai ron: ‘nihon jij¯o’ ky¯oiku ni okeru ‘bunka’ no toraekata o megutte [Strategic theory of denying the existence of ‘Japanese culture’: On understanding ‘culture’ in teaching ‘Japanese culture and issues’].” Nij¯u Isseiki no ‘Nihonjij¯o’ [‘Japanese Culture and Issues’ for the 21st Century] 2, 4–15. Kramsch, C. (1997) “The privilege of the nonnative speaker.” PMLA 112(3), 359–369. Kubota, R. (1996) “Nihongo ky¯oiku ni okeru hihan ky¯oiku, hihan teki yomi kaki ky¯oiku [Critical pedagogy and critical literacy in teaching Japanese].” Japanese-Language Education around the Globe 6, 35–48.

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Japanese Applied Linguistics

349

Kubota, R. (1998) “Voices from the margin: Second/Foreign language teaching approaches from minority perspectives.” The Canadian Modern Language Review 54, 394–412. Kubota, R. (1999) “Japanese culture constructed by discourses: Implications for applied linguistics research and ELT.” TESOL Quarterly 33, 9–35. Kubota, R. (2001) “Discursive construction of the images of U.S. classrooms.” TESOL Quarterly 35, 9–38. Kubota, R. (2002) “Japanese identities in written communication: Politics and discourses.” In R. T. Donahue (ed.), Exploring Japaneseness: On Japanese Enactments of Culture and Consciousness. Westport, CT: Ablex, pp. 293–315. Kubota, R. (2003) “Critical teaching of Japanese culture.” Japanese Language and Literature 37, 67–87. Kubota, R. (2004) “Critical multiculturalism and second language education.” In B. Norton and K. Toohey (eds.), Critical Pedagogies and Language Learning. Cambridge: Cambridge University Press, pp. 30–52. Kubota, R. and Lehner, A. (2004) “Toward critical contrastive rhetoric.” Journal of Second Language Writing 13, 7–27. Kubota, R. and Lin, A. (2006) “Race and TESOL: Concepts, research, and future directions.” TESOL Quarterly 40(3), 471–493. Kubota, R., Austin, T. and Saito-Abbott, Y. (2003) “Diversity and inclusion of sociopolitical issues in foreign language classrooms: An exploratory survey.” Foreign Language Annals 36, 12–24. Kumaravadivelu, B. (2003) “Problematizing cultural stereotypes in TESOL.” TESOL Quarterly 37, 709–718. Kumaravadivelu, B. (2006a) “TESOL methods: Changing tracks, challenging trends.” TESOL Quarterly 40, 59–81. Kumaravadivelu, B. (2006b) Understanding Language Teaching: From Method to Post-Method. Mahwah, NJ: Lawrence Erlbaum Associates. Lee, E. (in press) “Race: A third voice in the native/nonnative speaker debate.” TESOL Quarterly. Lee, Y. (1996) Kokugo to iu Shis¯o: Kindai Nihon no Gengo Ninshiki [The Philosophy called Kokugo: Language Awareness in Modern Japan]. Tokyo: Iwanami Shoten. Leeman, J. (2005) “Engaging critical pedagogy: Spanish for Native Speakers.” Foreign Language Annals 38, 35–45. Leung, C., Harris, R. and Rampton, B. (1997) “The idealized native speaker, reified ethnicities and classroom realities.” TESOL Quarterly 31, 543−560. Llurda, E. (ed.) (2005) Non-Native Language Teachers: Perceptions, Challenges and Contributions to the Profession. New York: Springer. Luke, A. (2004) “Two takes on the critical.” In B. Norton and K. Toohey (eds.), Critical Pedagogies and Language Learning. Cambridge: Cambridge University Press, pp. 21–29.

SFK ch12

SFK ICBK027

350

April 21, 2008

15:29

Char Count=

Critical Approaches to Teaching Japanese

Matsumoto, Y. and Okamoto, S. (2003) “The construction of the Japanese language and culture in teaching Japanese as a foreign language.” Japanese Language and Literature 37, 27–48. May, S. (2004) “Rethinking linguistic human rights: Answering questions of identity, essentialism and mobility.” In D. Patrick and J. Freeland (eds.), Language Rights and Language Survival: A Sociolinguistic Explanation. Manchester: St Jerome, pp. 35–53. Morgan, B. (1998) The ESL Classroom: Teaching, Critical Practice, and Community Development. Toronto: University of Toronto Press. Nakane, C. (1967) Tate Shakai no Ningen Kankei: Tan-itsu Shakai no Riron [Human Relations in Vertical Society: A Theory of a Unitary Society]. Tokyo: Kodansha (Translated as Japanese Society, Harmondsworth: Penguin Books, 1973). National Standards in Foreign Language Project (1999) Standards for Foreign Language Learning in the 21st Century. Lawrence, KS: National Standards Report. Nelson, C. D. (2006) “Queer inquiry in language education.” Journal of Language, Identity, and Education 5, 1–9. Nieto, S. (2004) Affirming Diversity: The Sociopolitical Context of Multicultural Education (4th edn.). New York: Addison Wesley Longman. Nishikawa, N. (2002) Sens¯o no Seiki o Koete: Gur¯obaru-ka Jidai no Kokka, Rekishi, Minzoku [Transcending the Century of Wars: Nation, History, and Ethnicity in the Age of Globalization]. Tokyo: Heibonsha. Norton, B. and Toohey, K. (eds.) (2004) Critical Pedagogies and Language Learning. Cambridge: Cambridge University Press. Oguma, E. (1995) Tan-itsu Minzoklu Shinwa no Kigen: “Nihonjin” no Jigaz¯o no Keifu [The Myth of the Homogeneous Nation: A Genealogy of the “Japanese” Self-portrait]. Tokyo: Shin-y¯o-sha. Okamoto, S. and Shibamoto Smith, J.S. (eds.) (2004) Japanese Language, Gender, and Ideology: Cultural Models and Real People. Oxford: Oxford University Press. Okamoto, S. and Siegal, M. (2003) “Toward reconceptualizing the teaching and learning of gendered speech styles in Japanese as a foreign language.” Japanese Language and Literature 37, 49–66. Osborn, T. A. (2000) Critical Reflection and the Foreign Language Classroom. Westport, CT: Bergin and Garvey. Pennycook, A. (1998) English and the Discourses of Colonialism. New York: Routledge. Pennycook, A. (2001) Critical Applied Linguistics: A Critical Introduction. Mahwah, NJ: Lawrence Erlbaum Associates. Pennycook, A. (2004) “Critical applied linguistics.” In A. Davies and C. Elder (eds.), The Handbook of Applied Linguistics. Malden, MA: Blackwell, pp. 784–807.

SFK ch12

SFK ICBK027

April 21, 2008

15:29

Char Count=

Japanese Applied Linguistics

351

Phillipson, R. (1992) Linguistic Imperialism. Oxford: Oxford University Press. Reagan, T. (2004) “Objectification, positivism and language studies: A reconsideration.” Critical Inquiry in Language Studies 1, 41–60. Reagan, T. G. and Osborn, T. A. (2002) The Foreign Language Educator in Society: Toward a Critical Pedagogy. Mahwah, NJ: Lawrence Erlbaum Associates. Ricento, T. (ed.) (2006) An Introduction to Language Policy: Theory and Method. Malden, MA: Blackwell. Said, E. (1978) Orientalism. New York: Pantheon Books. Sakai, N. (1994) “Shizan sareru nihongo, nihonjin [Stillborn Japanese language and people].” Shis¯o 845, 5–35. Shohamy, E. (2001) The Power of Tests: A Critical Perspective on the Uses of Language Tests. London: Longman. Shohamy, E. (2006) Language Policy: Hidden Agendas and New Approaches. London: Routledge. Skutnabb-Kangas, T. (1998) “Human rights and language wrongs: A future for diversity?” Language Sciences 20, 5–27. Spack, R. (1997) “The rhetorical construction of multilingual students.” TESOL Quarterly 31, 765–774. Spence-Brown, R. (2001) “The symbiosis of Japanese studies and Japanese language teaching: A language teaching perspective.” Japanese Studies 21, 77–83. Spolsky, B. (2004) Language Policy. Cambridge: Cambridge University Press. St. Pierre, E. and Pillow, W. (2000) “Introduction: Inquiry among the ruins.” In E. St. Pierre and W. Pillow (eds.), Working the Ruins: Feminist Poststructural Theory and Methods in Education. New York: Routledge, pp. 1–24. Sugimoto, Y. (1997) An Introduction to Japanese Society. Cambridge: Cambridge University Press. Suzuki, T. (1995) Nihongo wa Kokusaigo ni Nariuru ka [Can Japanese Become an International Language?]. Tokyo: K¯odansha. Tai, E. (1999) “Kokugo and colonial education in Taiwan.” Positions 7, 503– 534. Tai, E. (2003) “Rethinking culture, national culture and Japanese culture.” Japanese Language and Literature 37, 1–26. Tanaka, R. (2005) “Sengo no nihongo ky¯oiku ni okeru shis¯oteki ‘renzoku sei’ no mondai: Nihongo ky¯okasho ni miru ‘kokka,’ ‘kokumin,’ ‘gengo,’ ‘bunka’ [Problematic continuity over the postwar period in an ideological realm of Japanese language teaching: Contents analysis of ‘state,’ ‘nation,’ ‘language,’ and ‘culture’ in textbooks].” WEB-ban Riterashiizu [Web Version Literacies] 2(2), 1–9. Walton, A. R. (1991) “Expanding the vision of foreign language education: Enter the less commonly taught languages.” In E. S. Silber (ed.), Critical Issues in Foreign Language Education. New York: Garland Publishing, pp. 160–185.

SFK ch12

SFK ICBK027

352

April 21, 2008

15:29

Char Count=

Critical Approaches to Teaching Japanese

Wright, S. (2004) Language Policy and Language Planning: From Nationalism to Globalisation. New York: Palgrave Macmillan. Yasuda, T. (1999) “Kokugo” to “H¯ogen” no aida: Gengo K¯ochiku no Seijigaku [Between Kokugo and Dialects: Politics of Language Construction]. Kyoto: Jinbun Shoin. Yasuda, T. (2003) Datsu “Nihongo” e no Shiza [Perspectives beyond “the Japanese Language”]. Tokyo: Sangensha. Yoshino, K. (1992) Cultural Nationalism in Contemporary Japan. London: Routledge.

SFK

SFK

ind1

ICBK027

April 21, 2008

16:1

Char Count=

Author Index

Note: Page numbers followed by an ‘n’ indicate notes. 3A Network, 161, 165 Abe, H., 162, 181 Abe, T., 132, 135 Adelsw¨ard, V., 215 Akatsuka, N., 30, 31, 36, 38 Alfonso, A., 251 Anderson, B., 10, 276, 277, 293 Anderson, F., 85, 91, 93, 104n Aronsson, K., 216 Askenasy, J. J. M., 213 Asp, E. D., 238 Atkinson, J. M., 161, 167 Auer, P., 54, 147 Bakhtin, M. M., 9, 245, 247, 264 Ball, C., 132, 137, 149, 150, 155n Basch, L., 10, 275, 277, 290 Bauman, R., 267n Befu, H., 1, 52, 216, 342 Belz, J. A., 216, 303 Bergmann, J., 267n Bergson, H., 213, 214, 217 Berk, L. E., 218 Berkenkotter, C., 245, 264 Biller, L. W., 213 Block, D., 4 Blom, J., 147 Blum-Kulka, S., 267n Bourdieu, P., 126, 335 Briggs, C. L., 267n Broner, M. A., 9, 216

Brown, B., 112, 155n Bunka Shingikai, 117, 120, 121, 122, 123 Bybee, J., 6, 23, 27, 29, 35 Bygate, M., 3 Cameron, D., 7, 111 Carroll, T., 11, 132, 135, 136, 151 Cekaite, A., 216 Celce-Murcia, M., 2 Chafe, W., 22, 23, 104n Chapman, A. J., 214 Clancy, P. M., 31, 36, 46n Clark, P., 116 Clyne, M., 245 Coleman, J. G. J., 216 Comrie, B., 112 Cook, G., 2, 3, 216 Cook, H. M., 85, 86, 87, 94, 155n, 267n Corder, S. P., 3 Cornett, C., 216 Coulthard, R. M., 89 Couper-Kuhlen, E., 6, 54 Coupland, J., 9, 189, 190, 191, 192, 194, 207n Coupland, N., 9, 189, 190, 191, 192, 194, 207n Cumming, S., 29 Davidson, J., 58 Davis, A., 2

SFK

SFK

ind1

ICBK027

April 21, 2008

16:1

Char Count=

354

Author Index

Devereaux, P. G., 214 Diaz, R. M., 218 Donahue, R. T., 1 Drew, P., 89 Duranti, A., 7, 80, 81, 103, 267n Eckert, P., 183, 193, 207n Edmondson, M. S., 214 Edmondson, W., 3 Elder, C., 2 Erikson, E. H., 205, 206 Erman, Br., 27, 29 Ervin-Tripp, S. M., 194, 207n Fairclough, N., 7, 111 Foerster, C., 217, 218, 224 Fogel, A., 214 Foley, E., 213 Foot, H. C., 214 Ford, C. E., 54, 55, 74 Fox, B. A., 55 Fry, W. F., 213 Fukushima, S., 267n Furo, H., 164 Gal, S, 151, 152 Gallois, C., 189 Gass, S., 2, 4, 267n Giacalone Ramat, A, 147 Giddens, A., 113 Giles, H., 9, 189, 190, 191, 192, 194, 207n Ginsberg, G. P., 214 ´ T., 6, 23 Givon, Glenn, P., 9, 215, 222 Glick Schiller, N., 277, 293 Goffman, E., 81, 89 Goldstein, J., 214 Goodwin, C., 7, 8, 55, 69, 80, 81, 103, 161, 166, 167, 175, 177, 181

Goodwin, M. H., 177 Gottlieb, N., 12n, 117 Grabe, W., 2 Grainger, K., 9, 191 Grice, H. P., 193 Gumperz, J., 7, 80, 81, 82, 132, 147, 152, 193, 213 ¨ Gunthner, S., 245, 264, 267n Haiman, J., 38, 43 Hamaguchi, T., 189 Hamilton, H. E., 189, 190 Hanks, W., 80, 267n Harada, S.-I., 83 Hasunuma, A., 30 Hatasa, K., 30, 31, 41 Hatasa, Y., 12n, 30, 31, 41 Hayashi, M., 53, 57, 68, 73, 74n Hayes, S. T., 47n Heinrich, P., 11, 52, 342 Henwood, K., 9, 191 Heritage, J., 89, 104n, 161, 166, 167 Higgins, C., 52 Hill, J. H., 151 Himeno, T., 267 Hinds, J., 84 Hohenstein, C., 245, 265 Hopper, P., 6, 23, 27, 28, 29 Hoshina, K., 116 Hosotani, F., 132, 136 Huckin, T., 245, 264 Hymes, D., 80 Ide, S., 83, 84, 112, 160, 162, 163 Ikeda, K., 215, 314 Ikuta, S., 83, 84, 87 Inoue, F., 8, 132, 135, 136, 138, 139, 152 Inoue, M., 8, 163, 181 Iwasaki, S., 28, 43, 53 Izaki, Y., 250

SFK

SFK

ind1

ICBK027

April 21, 2008

16:1

Char Count=

Author Index Jacobsen, W. M., 30, 38, 47n, 89, 90 Jefferson, G., 95, 101, 214, 215, 230, 236, xii Johnson, M. L., 189 Johnstone, B., 193 Jones, K., 21, 22, 46n, 53, 56 Jorden, E. H., 40, 84, 112 Joseph, J. E., 7, 111 Kabaya, H., 124 Kanazawa, H., 152 Kane, T. R., 214 Kanno, Y., 10, 277, 281 Kaplan, R. B., 2 Kashiwazaki, H., 243, 250, 251, 252, 269n Kasper, G., 267n, 301, 303 Kataria, M., 213 Kawaguchi, Y., 124 Kawakami, K., 213, 342 Kendon, A., 81 Kida, M., 140 Kikuchi, Y., 124 Kindaichi, H., 84 Kitano, H., 68 Knoblauch, H., 267n Kobayashi, T., 134, 135 Kokugo Shingikai, 117, 118 Kokugo-gakkai, 134 Komori, Y., 135, 336 Kondo-Brown, K., 12n Kroskrity, P. V., 114, 151 Kubota, R., 5, 11, 31, 52, 74, 115, 134, 153, 154n, 327, 328, 332, 333, 341, 342, 343, 344, 346n Kumatoridani, T., 267n Kuno, S., 30, 46n Kushida, S., 69 Lakoff, G., 27 Lakoff, R., 81, 164 Lampert, M. D., 194, 207n

355

Langacker, R. W., 6, 23, 27, 28 Lantolf, J. P., 9, 214, 218, 303 Lauwereyns, S., 57 Lee, Y., 10, 116, 135, 151, 154n, 301, 302, 335, 337, 341 Lepper, G., 200 Lerner, G. H., 53, 55, 69, 74, 74n Lessard-Clouston, M., 4 Levinson, S., 80, 112, 193, 247 Li Wei, 147, 148 Linde, C., 206 Long, D., 8, 132, 136 Luckmann, T., 245, 264, 267n Lyons, J., 80 Maeda, I., 138, 139 Mahony, D. L., 213 Makino, S., 84, 162, 163 Malinowski, B., 80 Markee, N., 90 Martin, S., 83, 112 Masuoka, T., 162 Matsumoto, Y., 112, 181, 189, 192, 194, 203, 267n, 328, 335 Matsunaga, S., 12n Mayes, P., 244, 245, 265, 267n Maynard, S., 57, 84, 113, 231, 251, 267n McCafferty, S., 217 McConnell-Ginet, S., 183 McGhee, P. E., 214 McGloin, N., 59, 267n McHoul, A., 90 Medgyes, P., 216 Meeuwis, M., 52 Mehan, H., 89 Miller, C. R., 267n Miller, L., 52, 53 Miller, R. A., 1 Milroy, L., 147 Miura, A., 59 Miyake, Y., 8, 132, 136

SFK

SFK

ind1

ICBK027

April 21, 2008

16:1

Char Count=

356

Author Index

Miyoshi, M., 119 Mizutani, N., 162, 163 Mizutani, O., 162, 163 Moerman, M., 52 Mondada, L., 10, 301, 302 Mori, J., 21, 22, 23, 53, 57, 58, 62, 74n, 103, 238n, 251, 301, 303, 304, 322n Morreall, J., 217 Munro, P., 43 Muysken, P., 147 Nakamizu, E., 267n Nakamura, K., 21, 53, 103 Nakayama, T., 28, 251 Nasir, U. M., 213 National Language Research Institute, 5 Nazikian, F., 87 Neu, J., 267n Neustupny, J. V., 83 Nishizaka, A., 161, 267n Niyekawa, A., 83 Noda, M., 40, 84, 112, 267n Norton, B., 10, 11, 277, 346n Nussbaum, J. F., 189, 207n Nwokay, E., 214 Ochs, E., 6, 23, 85, 104n, 156n, 197, 201 Ohara, Y., 11, 162, 163 Ohta, A. S., 218, 219, 229, 233, 238n Oishi, H., 116, 125 Okamoto, S., 8, 57, 87, 112, 140, 152, 156n, 160, 161, 162, 163, 182, 267n, 269n, 328, 335 Olsher, D., 177 Onishi, T., 135 Ono, T., 22, 29, 43, 46n, 56 Ota, H., 189

Park, Y., 251, 269n Pawley, A., 29 Pekarek Doehler, S., 10, 301, 302 Pennycook, A., 11, 329, 330, 331, 342, 344, 346n Pomerantz, A., 57, 58, 75n Poplack, S., 155n Portes, A., 10, 274, 276 Provine, R. R, 214, 230, 231 Ramsey, S. R., 127n Raskin, V., 214 Ray, Y. T., 267n Reinhardt, 216 Reutzel, E. J., 217, 218 Reynolds, K. A., 160, 164 Rose, K., 27, 52, 267n Ryan, E. B., 207n Sacks, H., 54, 57, 58, 63, 89, 90, 95, 101, 200, 215, 247 Saft, S., 53 Sakaguchi, N., 132, 136 Sakamoto, M., 124 Sanada, S., 8, 116, 132, 134, 135, 136, 139, 149, 151, 154n, 156n Sato, R., 132, 136 Schegloff, E. A., 6, 23, 54, 55, 57, 63, 73, 89, 90, 95, 101, 247, 255, 259 Schmitt, N., 2, 12n Selting, M., 6, 54 Shade, R., 213 Shibamoto, J. S., 162 Shibamoto Smith, J. S., 112, 160, 161, 182, 328 Shibata, S., 12n Shibata, T., 8, 132, 135, 136, 151, 154n Shibatani, M., 46n, 84, 134, 138 Shibuya, K., 136 Shingu, I., 139

SFK

SFK

ind1

ICBK027

April 21, 2008

16:1

Char Count=

Author Index Shinozaki, K., 134, 135 Shinzato, R., 47n Siegal, M., 160, 162, 328 Silverstein, M., 7, 80, 151, 184n, 201 Sinclair, J. M., 89 Smith, J. S., 8, 112, 160, 161, 164, 181, 182, 315, 328 Sroufe, L. A., 213 Stam, G., 217 Stambor, Z., 216 Strauss, S., 31, 36 Streeck, J., 55 Sugimoto, Y., 1, 342 Sullivan, P. N., 10, 301, 302 Sunaoshi, Y., 162, 163, 164, 181 Suzuki, A., 214, 223 Suzuki, M., 161 Suzuki, S., 57, 179, 336 Syder, F. H., 29 Takagi, C., 136 Takagi, T., 53, 69, 74n Takahashi, K., 213, 267n Takano, S., 164 Takubo, Y., 162 Tanaka, H., 53, 55, 69, 231, 251, 267n, 343 Tannen, D., 193, 214 Tarone, E. E., 9, 216 Taylor, T. J., 7, 111 ten Have, P., 89 Thompson, S. A., 6, 23, 27, 54 Thorne, S. L., 9, 214, 218

357

Tohsaku, Y., 30, 31, 40, 41 Tokieda, M., 125 Tomasello, M., 23 Toohey, K., 11, 346n Umino, T., 12n Usami, M., 112 Vygotsky, L. S., 217 Wade, J., 85 Warren, B., 27, 29 Washburn, R. W., 213 Watanabe, S., 52 Waters, E., 277 Wertsch, J., 217 Wetzel, P., 5, 7, 22, 111, 112, 113, 114, 117, 118, 124, 125, 127n, 258, 269n Whitney, W. D., 115 Williams, A., 189 Woolard, K. A., 113, 114, 151 Wray, A., 29 Yamada, H., 52, 116 Yanez, M., 218 Yasuda, T., 135, 151, 154n, 335, 336, 337, 338 Yoneda, M., 135 Yoshino, K., 1, 342 Yoshitomi, A., 12n Yotsukura, L. A., 7, 9, 245, 246, 248, 249, 250, 251, 308

SFK

SFK

ind2

ICBK027

April 21, 2008

17:37

Char Count=

Subject Index

Note: Page numbers in italics indicate figures and tables. Page numbers followed by an ‘n’ indicate notes. ∼ba positive conditional, 35–36 ∼kya negative conditional, 36–37, 47n actual use, 24–45, 32, 33, 40, 41 addressee honorific (masu) form, 82–84, 104n adjacency pairs, 57 see also first pair part, second pair part affect keys, 85 agreement, 60–65 see also interactional linguistics aizuchi (backchannel responses), 231 American versus Japanese communication styles, 52 assessment activities, and learner competencies, 304–320 audience, and pronunciation, 328–329 authoritative positions, women in, 164–165 see also josei-go (feminine linguistic forms) backchannel responses (aizuchi), 231 beautified language (bikago), 124–125 behavioral environment, 81

bikago (beautified language), 124–125 Bunka Shingikai (Culture Board), 119, 119–126, 123 CA (conversation analysis), 53, 167 case studies elderly identity, 193–205 gender variations, 168, 169, 170, 171, 171–183, 174, 175, 180, 181 keigo ideology, 111–127, 119, 123 language minority education, 280–293 laughter, 219–237 learner competencies, 304–317 plain form, 91–103, 94, 96, 101, 102 regional variations, 137–147, 138, 139, 140, 141, 143 toiawase (inquiries), 253–264 turn completion points, 59–72 usage-based linguistics, 24–45, 32, 33, 40, 41 changes, dealing with, 203–205 cheerfulness, use of, 203–205 classrooms classmate errors, 227–229 discourse in, 89–91 happyoo routine, 91–94, 94

SFK

SFK

ind2

ICBK027

April 21, 2008

17:37

Char Count=

Subject Index and laughter, 213–214, 216–218 and learner competencies, 303–304 see also education clausal boundaries, 230–231 code-switching, 132–133, 135–137, 147–149 see also regional variations common language (kyootsuugo). See Standard Japanese (SJ) communication styles, Japanese versus American, 52 competencies as resources, 301–321 completing forms, assumption of, 29–30 conditionals ∼ba positive, 35–36 ∼kya negative, 36–37, 47n case study, 24–45, 32, 33, 40, 41 corpus-based approach, 31–32, 32 fixedness of, 32, 33 introduction of, 30–31 confirmation, pursuit of, 65–72 conflict talk, 21–22 conjunctions, 33–35, 47n contemporary language policy, 115–118 context age, 188–189 authoritative positions, 164–165 classrooms, 81–94, 213–237, 303–304 dimensions of, 89–91 gender variations, 171–183 importance of, 81 regional variations, 137–147 conversation analysis (CA), 53, 167 critical applied linguistics approaches, 329–331 direction of, 344–346

359

cultural affiliation, assumptions based on, 52–53 Culture Board (Bunka Shingikai), 119, 119–126, 123 dansei-go (masculine linguistic forms), 162–163 see also gender variations detached speech style, 84–89, 85, 88 see also plain form dialects. See regional variations disagreement, 60–65 see also interactional linguistics dispreferred versus preferred responses, 57–59 dooshite (‘why’) questions, 22–23 economy, assumption of, 29 education insiders versus outsiders, 289–290 language minority education, 274–293 see also classrooms elderly identity case study, 193–205 dealing with changes, 203–205 previous studies, 190–193 stereotypes, 192–193, 207n transitory nature of, 205–206 elementary schools. See classrooms emancipatory modernism approach to critical applied linguistics, 329–331 exams, accommodation for, 294n expressions, fixedness of, 33–35, 47n extended predicate construction (EP), 250 feminine linguistic forms (josei-go), 162–163 see also gender variations

SFK

SFK

ind2

ICBK027

April 21, 2008

17:37

Char Count=

360

Subject Index

fictive relationships, 113 first pair part (FPP), 57–59, 65–72 fixed conditionals, 32, 33 fixed expressions, 33–35, 47n forgetfulness, disclosure of, 193–205 formal style (teineigo), 112 see also keigo forms, completing, 29–30 four Ds, in teaching Japanese culture, 344–345 functional linguistics, 24–45, 32, 33, 40, 41 Fundamental Education Law (2006), 346 gender variations, 160–183, 168, 169, 170, 171, 174, 175, 180, 181 genres of speech, 244–246 grammar, assumptions of, 26–30 happyoo routine mixed with ordinary conversation, 97–99 plain form case study, 91–94, 94 use of speech styles, 85 honorific language (sonkeigo), 112 see also keigo hoogen bokumetsu movement, 135, 150–153 see also regional variations humble language (kenjoogo), 112 see also keigo humor ludic language play, 220–225 use of, 195, 201–203 see also laughter hyoojungo. See Standard Japanese (SJ)

identity construction, 188–207 crisis, 203 elderly, 190–206 feminine, 162–164 masculine, 162–163 imagined communities, 275–278 see also language minority education immigration versus transnationalism, 275 imperatives, use of, 165–166, 173–181 see also gender variations impersonal form, 104n indexicality, 80, 103–104 informal speech style, 84–89, 85, 88 inquiry phone calls. See toiawase institutional talk, 89–91 see also happyoo routine interactional linguistics, 6–7, 52–74 internationalizing Japanese, 336 intra-speaker variation, 139–142, 141 see also code-switching Japanese culture rethinking, 342–343 teaching, 344–345 Japanese language applied linguistics, 4–6 historical descriptions, 1 policy, 115–118 rethinking, 333–338 Japanese people cultural uniqueness of, 342–343 rethinking, 338–342, 340 Japanese versus American communication styles, 52 JFL (Japanese as a foreign language) learners, 216–218

SFK

SFK

ind2

ICBK027

April 21, 2008

17:37

Char Count=

Subject Index joking. See humor josei-go (feminine linguistic forms), 162–165 see also gender variations JSL (Japanese as a second language) learners, 216–218, 274–293 katakana English, 225–226, 236–237 keigo categories of, 123, 123 contemporary opinion of, 126–127 definition of, 111–113 ideology, 111–127 principles, 118–126, 119 roots of, 115–118 Keigo Sh¯o-iinkai (Keigo Committee), 119, 119–126, 123 kenjoogo (humble language), 112 see also keigo kokugo (national language), 115–116, 337–338 kyootsuugo (common language). See Standard Japanese (SJ) L2 learners, 216–218, 274–293 language, nature of, 73–74 language ideology, definition of, 113–114 language minority education, 274–293 language policy, contemporary, 115–118 laughter case study, 219–237 in classrooms, 213–214 clausal boundaries, 230–231 effects of, 213–214, 236–237 and mutuality, 214 phrasal boundaries, 230–231 placement of, 230–236

361

previous studies, 214–218 punctuation effect, 214–215, 230–231 sentential boundaries, 231–232 sociocultural approach, 217–218 see also humor learner competencies and assessment activities, 304–320 case study, 304–317 and pedagogical activities, 302–304 lexical variants, 134, 139, 140, 142–147, 143 see also code-switching lexicalized conjunctions, 33–35, 47n maeoki utterances, 247–248, 250–253, 252 masculine linguistic forms (dansei-go), 162–163 see also gender variations masu (addressee honorific) form, 82–84, 104n migrant workers, 286–289, 291–292 mispronunciation of Japanese words, 328–329, 335–336, 337 see also laughter morphological variants, 134, 139, 139, 142–147, 143 see also code-switching morphological verbal forms. See speech styles multimodal organization of actions theory, 167 mutuality, and laughter, 214 naked plain form. See plain form national language (kokugo), 115–116, 337–338 native speakers, critique of, 340–341

SFK

SFK

ind2

ICBK027

362

April 21, 2008

17:37

Char Count=

Subject Index

nihongo (Japanese language), invention of, 337–338 see also Standard Japanese (SJ) nihonjin (Japanese people), critique of, 339–340, 340 nihonjinron, critique of, 342–343 normative view of language, 329 off-task talk, 89–91, 95–99, 96 old age. See elderly identity on-task talk, 89–91 ordinary conversation, 89–91, 95–99, 96 Osaka dialect (OD) case study, 133, 137–147, 138, 139, 140, 141, 143 painful self-disclosure (PSD) definition of, 189 effects of, 92 in peer interactions, 195–197 with a younger acquaintance, 197–199 peer interactions, 195–197 phonological variants, 134, 138, 138, 142–147, 143 see also code-switching phrasal boundaries, 230–231 plain form case study, 91–103, 94, 96, 101, 102 functions of, 84–85, 85 interpretation of, 100–103 see also detached speech style; informal speech style polite language. See keigo politeness, 163–165 political significance of teaching Japanese, 344–346 prefatory utterances, 246–248 see also maeoki

preference organization, affect on turn completion points, 57–59 see also first pair part, second pair part preferred versus dispreferred responses, 57–59 presentation routine. See happyoo routine problematizing practice approach to critical applied linguistics, 329–331 production difficulties, 226–227 professional vision, 161, 175, 181–183 pronunciation of Japanese words, 328–329, 335–336, 337 see also laughter prosody, use of, 54 PSD (painful self-disclosure). See painful self-disclosure (PSD) public education. See education punctuation effect, 214–215, 230–231 see also laughter purist view of language, 329 recipient design, 247 recipient laughter, 215 see also laughter regional variations case study, 137–147, 138, 139, 140, 141, 143 hoogen bokumetsu movement, 135, 150–153 previous studies, 134–137 see also code-switching relativist view of language, 329 rule-based conditionals, 39–43, 40, 41 rules, assumption of, 28–29

SFK

SFK

ind2

ICBK027

April 21, 2008

17:37

Char Count=

Subject Index sampling, challenges of, 25–26 scholarship, influence on language policy, 115–118 schools. See classrooms second language learners, 216–218, 274–293 second pair part (SPP), 57–59, 60–65 semi-fixed conditionals, 32, 33 semi-fixed expressions, 33–35, 47n SenseiOnline, 328–329 sentential boundaries, 231–232 sequence organization affect on turn completion points, 57–59 analysis of, 167 sex-linked forms. See gender variations simplicity, assumption of, 29 single grammar, assumption of, 28 situational view of language, 329 social relevance approach to critical applied linguistics, 329–330 sonkeigo (honorific language), 112 see also keigo soosureba, 47n soto/uchi distinction, 113 South American migrant workers, 286–289, 291–292 speaking self, sense of, 311–312 speech genres, 244–246 speech styles masu form, 82–84 plain form, 84–85, 85 research on, 82–89, 85 spoken language, centrality of, 24–30 sprachnation, 115–116

363

Standard Japanese (SJ) critique of, 335–336 hoogen bokumetsu movement, 135, 151–153 invention of, 337–338 rationale for choosing, 334 standardization, 135, 150–153 see also regional variations style switching (sutairu kirikae). See code-switching suggestions, expressions of, 37–38 sutairu kirikae (style switching). See code-switching teineigo (formal style), 112 see also keigo TESOL profession, critical turn, 327 textbook dialogues, inaccuracies of, 22–23 third turn receipt, pursuit of confirmation, 65–72 toiawase case study, 253–264 definition of, 243–244 previous studies, 250–253, 252 structure of, 248–250 theoretical background, 244–250 topic marking, 38–39 transition relevance place (TRP), 54–59 transnationalism, 275–278 see also language minority education turn completion points analysis of, 54–59 case study, 59–72 see also interactional linguistics uchi/soto distinction, 113 usage-based linguistics case study, 24–45, 32, 33, 40, 41

SFK

SFK

ind2

ICBK027

April 21, 2008

17:37

Char Count=

364

Subject Index

variant choice. See code-switching verbal forms. See speech styles vocatives, use of, 319 war orphans, 293n Western scholars, influence on language policy, 115–118

why (‘dooshite’) questions, 22–23 women’s language studies, 112–113 see also josei-go (feminine linguistic forms) word order, affect on turn completion points, 55–57 written language, 24