Situated Meaning: Inside and Outside in Japanese Self, Society, and Language 9780691194455, 9780691608549, 9780691656205

Citation preview

Situated Meaning

Situated Meaning I N S I D E AND O U T S I D E IN J A P A N E S E SELF, S O C I E T Y , AND LANGUAGE Edited by Jane M. Bachnik and Charles J. Quinn,Jr.

PRINCETON

UNIVERSITY

PRINCETON,

NEW

PRESS

JERSEY

Copyrigh t © 1994 by Princeto n Universit y Press Publishe d by Princeto n Universit y Press, 41 William Street , Princeton , Ne w Jersey 08540 In th e Unite d Kingdo m Princeto n Universit y Press, Chichester , West Sussex All Right s Reserved Library of Congress Catalogingin-Pubhcation Data Situate d meanin g inside an d outsid e in Japanes e self, society, an d language / edite d by Jan e Μ Bachmk , Charle s J Quinn,J r ρ cm Include s bibliographica l reference s an d inde x ISBN 0-691-06965- 4 — ISBN 0-691-01538- 4 (pb k ) 1 Japan—Civilization—Philosoph y 2 Japanes e language—Socia l aspect s I Bachnik.Jan e II Quinn , Charle s J ,Jr , 1948DS82 1 S59 1994 952—dc20 93 8956 Publicatio n of thi s boo k ha s bee n aide d by Th e Ohi o Stat e Universit y Institut e for Japanes e Studie s Thi s boo k ha s bee n compose d in Ne w Baskerville Princeto n Universit y Pres s book s are printe d on acid-fre e pape r an d mee t th e guideline s for permanenc e an d durabilit y of th e Committe e on Productio n Guideline s for Book Longevit y of th e Counci l on Librar y Resource s Printe d in th e Unite d State s of Americ a

1 3 5 7 9

10 (Pbk) 1 3 5 7 9 10

8 6 4 2 8 6 4 2

Princeton Legacy Library edition 2019 Paperback ISBN: 978-0-691-60854-9 Hardcover ISBN: 978-0-691-65620-5

To our parents

CONTENTS

Foreword, by Donald L. Brenneis

ix

Preface

xiii

Note on Romanization

xv

Key to Abbreviations and Orthographic Conventions

xvii

Contributors

xix PAR T O N E :

Indexing Self and Social Context

CHAPTE R O N E

I n t r o d u c t i o n : uchi/soto: Challengin g O u r Conceptualization s of Self, Socia l Order , a n d L a n g u a g e Jane M. Bachnik

3

CHAPTE R T WO

T h e T e r m s uchi a n d soto as Window s o n a World CharlesJ. Ouinn,Jr.

38

CHAPTE R THRE E

A Movabl e Self: T h e Linguisti c I n d e x i n g of uchi a n d soto Patricia J. Wetzel

73

CHAPTE R FOU R

I n d e x i n g H i e r a r c h y t h r o u g h J a p a n e s e G e n d e r Relation s Nancy Κ Rosenberger

88

CHAPTE R FIV E

Uchi/soto: Choice s in Directiv e Speec h Acts in J a p a n e s e Robert J. Sukle

113

CHAPTE R SI X

I n d e x i n g Self a n d Societ y in J a p a n e s e Famil y Organizatio n Jane M. Bachnik PAR T T W O :

Failure to Index: Boundary and Social Breakdown

143

Disintegration

C H A P T E R SEVE N

Uchi no kaisha: C o m p a n y as Family ? Dorinne K. Kondo

169

CONTENTS

Vlll

CHAPTER EIGHT

The Battle to Belong: Self-Sacrifice and Self-Fulfillment in the Japanese Family Enterprise Matthews M. Hamabata

192

CHAPTER NINE

When uchi and soto Fell Silent in the Night: Shifting Boundaries in Shiga Naoya's "The Razor" MichaelS Molasky

209

CHAPTER TEN

Uchi/soto: Authority and Intimacy, Hierarchy and Solidarity in Japan Jane M. Bachnik

223

Language as a Form of Life: Clines of Knowledge as Clines of Person

PART THREE:

CHAPTER ELEVEN

Ucht/soto: Tip of a Semiotic Iceberg? 'Inside' and 'Outside' Knowledge in the Grammar of Japanese CharksJ. Quinn,Jr.

247

Index

295

FOREWORD:

SITUATED

MEANING

DONALD L. BRENNEIS PITZER COLLEGE

volume adds a new dimension, both literal and metaphoric, to our understanding of Japan. Turning from the vertical axis of hierarchy and subordination, an organizing trope in much of the literature on Japan, these essays draw our attention to the horizontal, mapping a surprising and stimulating range of cultural practices and orientations in terms of such relational concepts as uchi 'inside' and soto 'outside'. Subde, insightful, and very effectively informed by multiple disciplinary perspectives, Situated Meaning takes us deeply inside fascinating aspects of Japanese language, literature, daily life, and social experience. It also speaks powerfully to the outside, raising broader questions of language, meaning, and cultural analysis. Guiding this volume is the notion ofJapanese social order as situated, that is, that what analysts see as "social order" inheres in the very specific circumstances of its particular enactments. Situated Meaning dramatically extends our understanding of this general insight, which has informed much recent scholarship on Japan. It does so in several ways: first, by exploring the highly significant pairs of relational terms through which "situation" is shaped and construed (for example, uchi/soto 'inside/outside' and omote/ura 'in front/in back'); second, by considering such terms—and the relationships they imply—not as manifestations of a fixed social framework but as "indexical coordinates," deictic locational practices necessarily centered on an "I"; and third, by considering with great subtlety the experiential implications of such a perspective. This volume takes the often assumed general "context" of Japanese society as problematic, forcing attention to a constantly reconstituted and person-centered set of particular relationships. Further, through its subtle and insightful discussions of personal experience—whether in novels, in domestic life, or in business—Situated Meaning successfully conveys the lives of individuals in a complex and remarkably flexible social world. In its focus on the central role of social indexicality Situated Meaning directly challenges the primarily referential view of linguistic and social "meaning," an implicit assumption in much work in linguistics, anthropology, and cultural analysis. Basically, such a view implies the notion of T H I S REMARKABLE

χ

FOREWORD : SITUATE D MEANIN G

an externa l "world" which is referre d to an d reflecte d in language , whethe r it be a physical world ("Thi s is a table") or a social on e ("Thi s is a social superior") . Th e centra l role of languag e lies in conveyin g such messages about an external , nonlinguisti c world; consequently , th e cor e problemati c of languag e lies in ho w well it reflects thi s world. Th e indexical approac h at th e hear t of thi s volum e demand s a new perspec tive, on e takin g th e situationall y locate d relationship s amon g speaker , hearer , an d scen e as a startin g point . Meanin g become s located , rathe r tha n bein g "about " th e external . Socia l life is not , therefore , so muc h reflecte d in linguistic practic e as it is constitute d throug h it, with speec h an d its implication s bein g necessaril y entangle d with situation . "Con text, " th e old warhors e of sociolinguisti c writing, takes on a new flavor here ; it is no t externa l to speec h bu t reciprocall y shape d by it, a poin t centra l to th e essays in a provocativ e new volum e edite d by Durant i an d Goodwi n (1992) . Thi s volum e speaks in stimulatin g ways to anothe r recen t collection , UnwrappingJapan: Society and Culture in AnthropologicalPerspective (Ben Ari et al. 1990). Centra l to thi s boo k is th e importanc e of wrappin g in Japan ; theme s of elegan t presentation , disguise, an d concealmen t are considere d in a variety of Japanes e social practices , from gift-giving to conversation . Jo y Hendry' s contributio n (1990) speaks directl y to th e processe s of wrappin g an d unwrapping , suggesting tha t presentatio n an d an almos t litera l embeddin g of th e "contents, " whethe r of a gift or of talk, may well be mor e significant tha n th e content s themselves . Locatin g th e gift within its wrapping—or th e talk within its honorific s an d allusions—an d penetratin g th e layers of concealmen t as the y shap e an d locat e what the y contai n is central . As with Situated Meaning, th e proces s of linkin g contex t an d message is necessar y an d reciprocal . Thi s boo k clearly offers a great dea l to scholar s of Japan . At th e same time , it speaks very effectively to a rang e of question s at th e hear t of contemporar y wranglings with language , cultur e an d social experience . I want briefly to not e several provocativ e point s at which such theoreti cal an d comparativ e concern s are directl y implicated . First , in takin g issues of social inequalit y as flexible an d situationall y shape d practices , tha t is, one s tha t are constructe d an d negotiate d rathe r tha n operatin g in som e mechanisti c way, Situated Meaning reverses what is usually take n for grante d in studie s of Japa n an d othe r apparentl y hierarchica l societies . Fo r thos e of us workin g in communitie s with explicitly egalitaria n ideologie s (see, for example , th e Pacific island com munitie s represente d in Brennei s an d Myers 1984), th e negotiatio n of social relation s is seen as a given; what is often less commonl y recognize d are thos e moment s in which hierarchica l practic e is consequential . Bach -

FOREWORD: SITUATED MEANING

XI

nik and Quinn's volume makes the assumed problematic for Japan and, in so doing, illuminates the specifics of personal and social negotiation at the heart of quotidian life in an explicitly stratified society. Second, there are remarkable resonances with emerging discussions in the broader western Pacific literature. The work of Frake (1964, 1975) in the southern Philippines prefigured these discussions, bringing questions of social space and its implications for communicative and cultural practice to the fore. Clare Wolfowitz's recent (1991) monograph on Suriname Javanese, for example, challenges the usual "fixed hierarchy" model ofJavanese language style, reorienting an understanding of language use in terms of social space. As with the uchi/soto dimension, the situational elements of interactional practice become central, taking precedence over apparently fixed hierarchical distinctions. Similarly, recent work in Melanesia and Polynesia stresses the role of situation and event in shaping both behavior and local theories of personal experience (Watson-Gegeo and White 1990). Within this broader region, issues of interpretation, social practice, and "emotional" life are being linked in increasingly convincing ways to an implicitly indexical, situation-based perspective. This may represent shared cultural orientations within the western Pacific; it may also reflect, in part at least, an emergent regional discourse among Pacific anthropologists and other scholars working in the area. In contrast with the literature on the Mediterranean and the Middle East, for example, where the paradigm of relatively fixed public and private places—and of their consequences for an interpretation of gender relations—shaped arguments not only about those areas but worldwide, Situated Meaning makes a compelling case for considering space as socially, situationally, and flexibly defined. The implications of this perspective go far beyond Japan. Finally, this book contributes both to a refinement of our understanding of how linguistic and cultural meaning works and to a new view of the linguistic ideologies held by both social scientists and those whose lives they seek to understand. Situated Meaning is one of the most effective discussions of a socially located approach to culture and communication I have read; the themes of deixis and indexicality are brilliandy teased out, providing a productive and finely wrought exploration of Peircean theory. It joins such works as Silverstein's germinal essay on shifters and indexical meaning (1976), Hanks's magisterial consideration of deixis in Yucatan (1990), and Gumperz and Levinson's reconsideration of linguistic relativity (1991), in mapping a new view of linguistic practice and its complex relationship to cultural meaning. Situated Meaning provides new maps of territory long recognized as central within Japanese studies. Household, business, aesthetics, and in-

Xll

FOREWORD. SITUATED MEANING

dividual experience a p p e a r in new a n d revelatory light. Beyond the regionalist audience, however, this work also brings these new maps tog e t h e r with what is relatively u n k n o w n territory for many a n t h r o p o l o gists a n d linguists. Remarkable in its thoughtful attention to the particulars, the volume makes a compelling case for rethinking some of the basic u n d e r s t a n d i n g s central to anthropology. It speaks with grace a n d insight to both the "inside" a n d the "outside" of scholarship in a n d about Japan. REFERENCES

Ben-Ari, Eyal, Brian Moeran, and James Valentine, eds. 1990. UnwrappingJapan: Society and Culture in Anthropological Perspective. Honolulu: University of Hawaii Press. Brenneis, Donald, and Fred R. Myers, eds. 1984. Dangerous Words: Language and Politics in the Pacific. New York: New York University Press. Duranti, Alessandro, and Charles Goodwin, eds. 1992. Rethinking Context. Language as an Interactive Phenomenon. New York: Cambridge University Press. Frake, Charles O. 1964. "How to ask for a drink in Subanun." American Anthropologist 66: 127-32. Frake, Charles O. 1975. "How to enter a Yakan house," in Mary Sanches and Ben G. Blount, eds., Soaocultural Dimensions of Language Use, 25-40. New York: Academic Press. Gumperz,JohnJ., and Stephen C. Levinson. 1991. "Rethinking linguistic relativity." Current Anthropology 32: 613-22. Hanks, William F. 1990. Referential Practice. Language and Lived Space among the Maya. Chicago: University of Chicago Press. Hendry, Joy. 1990. "Humidity, hygiene, or ritual care: Some thoughts on wrapping as a social phenomenon," in Eyal Ben-Ari, Brian Moeran, and James Valentine, eds., UnwrappingJapan: Society and Culture in Anthropological Perspective, 18-35. Honolulu: University of Hawaii Press. Silverstein, Michael. 1976. "Shifters, linguistic categories, and cultural description," in Keith H. Basso and Henry A. Selby, eds., Meaning in Anthropology, 11-56. Albuquerque: University of New Mexico Press. Watson-Gegeo, Karen Ann, and Geoffrey M. White, eds. 1990. Disentangling: Conflict Discourse in Pacific Societies. Stanford: Stanford University Press. Wolfowitz, Clare. 1991. Language Style and Soaal Space. Stylistic Choice in Sunname Javanese. Urbana: University of Illinois Press (Illinois Studies in Anthropology, no. 18).

PREFACE

developed out of a "homestay" the co-editor, Jane Bachnik, made with a Japanese farm family, while traveling in Japan after graduating from college. The purpose of her homestay was to learn Japanese, and in the process of teaching her the language the family members managed to convey something deceptively simple but nonetheless profound: that language is lived. This book is a product of that experience; its focus on 'inside/outside' or uchi/soto is a product of glimpsing how language is lived, and how this lived perspective is equally important for social life. All the contributors to the volume shared a similar insight—that 'inside/outside' coordinates are fundamental to Japanese self, society, and language—and that these coordinates have to be grasped and used in experience. The idea of the volume arose spontaneously from conversations—in the field, in academic corridors, and at meetings—that made us aware we were exploring something that crossed disciplinary boundaries, and made us decide to pursue our ideas in a book project. Charles Quinn became a co-editor with Jane Bachnik because he had completed an extended study of epistemology in Japanese grammar, which complemented Jane Bachnik's project on self and society in the Japanese family. Included in the volume are three specialists in linguistics, three in anthropology, one in sociology, and one in comparative literature. All of us focus on Japan. The volume was conceived as a means of exploring the important parallels we discovered between the use of uchi/soto in language and in social life. Most intriguing to us were the ways in which uchi/soto revealed deep-seated notions of the Japanese social universe—as located in time and space; as encompassing human beings and requiring their perspectives and actions. Yet human beings were also part of an ordering process, so that "structuring," and ordered continuity over time, were equally crucial with human beings in time. In fact, the balancing of these two perspectives—of specific human beings engaged in social situations with a general order that transcends situations is the crux of uchi/ soto organization. This is played out in everyday life in the constant attempts to balance individual will against social norms or obligations; or to weigh self-interest against self-sacrifice for a greater social good. We wanted to portray the centrality of uchi/soto issues to Japanese social participants in their everyday lives. T H I S PROJECT

χιν

PREFAC E

A pane l was organize d as a first step towar d realizin g th e boo k volume , an d Matthew s Hamabata , Patrici a Wetzel, Rober t Sukle, Dorinn e Kondo , Charle s Quinn , an d Jan e Bachni k presente d paper s for "uchi/ soto:Shifts in Languag e an d Socia l Boundaries, " at th e Association for Asian Studie s meeting s in Washington , D.C. , in 1988. Th e pane l served its purpos e of providin g a foru m to hel p us focus ou r ideas. A series of theme s emerged—focusin g o n th e indexin g of uchi/soto an d its relatio n to language , self, an d social organization , includin g th e organizatio n of gender , hierarchy , family, an d family business. An 'underside ' to uchi/ soto indexin g also emerged—in th e form of violation s of appropriat e indexing , which ha d surprisingl y seriou s consequences . All th e paper s from thi s pane l were the n rewritte n an d incorporate d int o th e volume , an d Nanc y Rosenberge r an d Michae l Molask y were invited to contribut e to th e volume . Earl y on , we realize d tha t th e issues raised by uchi/soto linke d ou r volum e to endeavor s outsid e Japan , especially thos e tha t involve performa tive approache s to language , self, an d social life, an d focus on indexica l or 'situated ' meaning . We are grateful to Donal d Brennei s for his intel lectua l suppor t in assisting us in workin g ou t thi s linkage, an d for his suggestions in readin g th e manuscript . We ar e thankfu l to A. L. Becker an d Eleano r H . Jorde n for thei r continuin g exampl e in insistin g tha t linguistic an d social form s have som e relationshi p to social activity, an d to Stanle y Cavell for his lecture s o n Wittgenstein , which unknowingl y shed muc h light on Japanes e society. All of us have Japanes e friends , informants , an d mentors , such as th e family mentione d above, to whom we ar e profoundl y grateful, an d who shed muc h light for us on perspec tives elaborate d in ou r chapters . We are grateful to Margare t Case, Rit a Bernhard , an d Directo r Walter Lippincott , of Princeto n Universit y Pres s for thei r considerabl e suppor t in assisting in th e publicatio n of th e volume . Finally , we owe a deb t of gratitud e to Bradle y M. Richardson , Directo r of th e Institut e for Japa nese Studie s at Th e Ohi o Stat e University , whose timel y assistanc e gave thi s boo k a secon d lease on life. We hop e we have mad e it clear tha t all th e contributor s helpe d in th e creatio n of thi s volume . We than k the m all for makin g thi s a genuinel y collaborativ e intellectua l experience , of which thi s volum e is th e product .

NOTE

ON

ROMANIZATION

WE ARE USIN G a hybrid system of romanizatio n for Japanes e terms . Con sonant s are based on th e Hepbur n system, while vowel lengt h is indi cate d by doublin g th e vowel. Japanes e vowels are pronounce d approximatel y as follows: Th e symbol: approximates : 'a' in 'father ' a i 'i' in 'machin e 'u ' in 'put ' u 'e' in 'bet ' e Ό ' in 'horse ' ο Th e occurrenc e of th e same vowel symbol twice indicate s a lon g vowel, for example , aa, where eac h vowel retain s its origina l qualit y an d length , bu t th e sequenc e is pronounce d as a continuu m (Jorden , Japanese: The Spoken Language, Par t 1 [Ne w Haven : Yale Universit y Press, 1987], p. 2).

KEY TO ABBREVIATIONS AND ORTHOGRAPHIC CONVENTIONS

ACC AFF CAUS COND CONF COP or CP COPtent D DIR EP EXTP FACT G GEN HON HUM I ID IF INF INST LOC LOC„ MID NEG OBJ PAST PForP PFinf PL PL+ POT PROV Q QUOT

ACCusative AFFirmative (i.e., opposite of NEG) CAUSative CONDitional (-tara form) CONFirmative sentence particle ne COPula COPula, tentative (daroo) Directive, Imperative DIRectional particle e 'to, toward' Extended Predicate (/clause + no (da)/): Jorden, Japanese: The Spoken Language, part 1, 1987 Extended Predicate (same as EP) modal suffix of established fact ("recall" -ken) Gerund (= a perfective infinitive) GENitive HONorific (T) HUMble (i) Imperfective IDentifying focus (= "nominative" ga) Inclusive Focus (particle mo) INFinitive INSTrumental LOCative (particle ni) LOCative of accompaniment (particle to) MIDdle voice (inflecting suffix -(r)areru) NEGative OBJect PAST tense (now-extinct inflecting suffix -ki) PerFective (-ta form) PerFective infinitive (= the "gerund", V-te) PLural suffix (-ra, -tachi) Polite PLural (-kata) POTential PROVisional (-(r)eba) Question particle (ka) QUOTative to (< L O C J

XV111

R RF S SP SUBJ SUP TOP UI V CAPS ' ' () [] //

ABBREVIATIONS AND CONVENTIONS

Representative (-ton form) Restrictive Focus (particle wa) Stem (= verbal infinitive) Sentence Particle (e.g., 310, ne, etc.) SUBJect Suppositional modal TOPic Unique Identification (focus particle koso) Verb emphasis, usually contrastive translation of Japanese optional meaning not signaled in original to be taken as a unit (e.g., grammatical construction, formula, etc.)

CONTRIBUTORS

JAN E Μ. BACHNI K Associate Professo r of Anthropolog y Th e Universit y of Nort h Carolin a at Chape l Hil l MATTHEW S M.

HAMABAT A

Dea n of Haverfor d College DORINN E K. KOND O MacArthu r Associate Professo r of Women' s Studie s an d Anthropolog y Pomon a College M I C H A E L S.

MOLASK Y

Ph.D . Candidat e in Japanes e Literatur e Universit y of Chicag o CHARLE S J. QUINN , J R . Associate Professo r of Japanes e Th e Ohi o Stat e Universit y N A N C Y R.

ROSENBERGE R

Assistant Professo r of Anthropolog y Orego n Stat e Universit y ROBER T J. SUKL E

Senio r Lecture r in Japanese ; Director , FALCO N Japanes e Progra m Cornel l Universit y PATRICI A J. W E T Z E L

Professo r of Japanes e Portlan d Stat e Universit y

Chapter 1 INTRODUCTION: UCHI/SOTO:

CHALLENGING

O U R C O N C E P T U A L I Z A T I O N S O F SELF, SOCIAL ORDER, A N D L A N G U A G E J A N E M.

BACHNIK

INSIDE and outside have far more dimensions than appear at first glance. For example, in English, whether one is "looking in" or "looking out," depends on where one is looking from. Whether one "zooms in" or "out" depends on the direction of the camera lens in relation to oneself. "Reaching in" may refer to putting one's arm inside a container, "outreach" is a metaphorical arm outside one's normal sphere of influence. We manage a rather complex set of directional vantage points, in the course of everyday conversations (which may include take-outs, sit-ins, infields, and outpatients), and we do this without taking much notice. 1 In this volume we focus on this taken-for-granted process in the everyday use of 'inside' and 'outside'—uchi and soto—in Japanese society. Although usually unnoticed, these distinctions are crucially important: uchi/soto is a major organizational focus for Japanese self, social life, and language. In uchi/soto, the 'inside/outside' orientations are also specifically linked with another set of meanings, denoting "self and "society." Thus the organizing of both self and society can be viewed as situating meaning, through the indexing of inside and outside orientations. But inside/outside distinctions are not limited to Japanese; they are used by people in every society to situate meaning, as illustrated in the above English examples of "looking," "zooming," "reaching." These directional movements are part of a broad system of basic orientations through which all of us—in every known language and society—constantly locate ourselves in relation to the world. Through them we define not only our physical orientations in space but our social and psychic orientations as well. Inside and outside, like a drop of water in a pool, move in ever-widening circles to encompass a broad series of issues, both inside and outside Japan. More specifically, approaching the organization of self and social order as indexical, or situated, meaning addresses longstanding issues in the formulation ofJapanese social life—and links these to theoretical

4

JANE Μ. BACHNI K

issues tha t are importan t beyon d Japan . Thes e includ e (1) definin g th e organizatio n of a Japanes e self, lon g characterize d as "relational " (Araki 1973), (2) approachin g Japanes e social orde r as "situationall y defined " (Lebr a 1976), "organic " (Smit h 1983), an d "shifting" (Do i 1986). Thes e issues, in turn , can be relate d to : (1) th e "interdependent " self, acknowledge d as havin g a worldwide existenc e (Marku s an d Kitayama 1991); (2) perspective s on contex t as a central , organizin g focus for social life (Durant i an d Goodwi n 1992); an d (3) relationship s between self an d society, as mutuall y constitutive , such as Bourdieu' s perspective on "habitus " an d "practice " (1977, 1990); Giddens ' focus on agenc y an d structur e (1979, 1984); Merleau-Pont y (1962) an d Hei degger's (1962) focus on social life as "lived," an d C. S. Peirce' s formula tion s of indexe s an d indexica l meanin g (1931-1958) . Thi s volum e will take u p all thre e sets of issues, an d will focus particularl y on indexe s an d indexica l meaning . Joy Hendry' s argumen t on "wrapping" in Japanes e society (1993, 1990, 1989) provide s an excellen t illustratio n of indexica l or situate d meaning . Hendr y point s to th e extrem e car e an d attentio n th e Japanes e take in wrappin g objects—from grocerie s to gifts—and she draws paral lels from th e wrappin g of materia l object s to wrappin g in language , in th e frequen t use of polit e an d respectfu l languag e (keigo). Architectur e wraps as well: "[PJape r roo m partition s still commonl y foun d in Japa nese house s ma y be describe d as wrappin g architectura l space an d even a smallish hous e will have inne r an d oute r layers to which visitors are permitte d entr y accordin g to thei r social proximit y an d status " (Hendr y 1989, 627). Ritua l wrapping , exemplifie d in th e heavy rop e tha t wraps a shint o shrine , can be linke d to transition s from th e mundan e to th e transcendent , or in th e othe r direction , from th e highly polit e an d forma l to th e mundane . Wrappin g of th e bod y also takes plac e throug h donnin g th e multipl e layers of garment s involved in wearin g th e forma l Japanes e kimono . Wrappin g pervade s Japanes e society; bu t how doe s thi s hel p us to characteriz e th e Japanese ? I n thi s regard Hendr y caution s tha t attemptin g to penetrat e a labyrint h of layers to reac h th e con ten t behin d th e form is to utterl y miss th e point. 2 At issue her e is n o less tha n th e impenetrabilit y by which Japa n ha s lon g been characterized . Wrappin g is muc h mor e tha n a "cover-up " of th e meanin g inheren t in objects, social activities, an d linguistic commu nication . Two approache s to meanin g are at issue: on e tha t views th e meanin g of gifts, language , social order , an d ritual s in term s of a "center " or "core " beneat h th e wrapping , an d a secon d approac h tha t focuses on th e activity of wrappin g itself, an d th e relation s wrappin g defines between social participant s an d thei r context . I n th e first approac h wrappin g obfuscate s content , as pape r covers th e gift, an d mus t therefor e be remove d an d discarde d to reac h th e gift,

SELF , SOCIA L ORDER , AND LANGUAG E

5

as cor e meanin g But Hendry' s poin t is tha t for Japanese , wrappin g as a form of polit e language , ritual , dress, architecture , hierarchy , an d so on—is itself closely linke d to meanin g Thi s approac h regard s meanin g no t as th e cor e behin d th e wrappin g bu t as linke d with th e wrappin g itself, no t as a "thing"—a gift, a speec h "level," a "layer" of clothing—bu t "as an almos t litera l embeddin g of th e 'contents, ' whethe r of a gift or of talk [that ] may well be mor e significant tha n th e content s themselves " (Brenneis , ρ χ) Wrappin g focuses on th e proces s of locatin g th e social participant s in th e social context , an d of delineating the context itself in terms of the "wrapping" Wrappin g can also be closely linke d to inside / outside , an d to indexica l perspective s on meanin g In thi s volum e we focus, no t on the inside , as a mean s of penetratin g th e cor e of Japanes e society, bu t on uchi/soto as uncoverin g th e proces s of indexin g tha t is crucia l to th e delineatio n of a "situated " social order—an d a relationa l self—both highly embedde d in social contex t Throug h thi s volum e we attemp t a shift of focus in a larger sense We conside r tha t graspin g th e significanc e of uchi/soto insideJapa n links th e volum e to importan t issues outside Japa n Viewed as a society where indexin g is an organizingfocus for self, society, an d language , Japanes e society shoul d provid e us with th e opportunit y for a sustaine d perspectiv e on an "ordering " tha t is closely tied to th e way in which participant s locat e themselve s in contex t I n proposin g tha t bot h self an d social orde r are reciprocally—an d mdexically—define d in Japan , we are also proposin g tha t th e organizatio n of a "contextuall y defined " self an d an "interactionally " approache d contex t can be linked as two sides of the same coin Th e sustaine d focus on indexin g "self" an d "social order, " as well as th e relationshi p between th e contextuall y define d self an d th e contextu ally define d social order , ar e therefor e highly significant beyon d Japa n as well At stake her e is a constitutivel y oriente d approac h towar d social life In othe r words, th e organizatio n of self o r society is no t reducibl e to a "core " of pattern s or structures , or even a "core " of perso n tha t is foun d "behind " social contexts , like th e object lies "behind " its wrappings , an d social orde r "behind " its participant s No r is orde r itself discarde d as a concep t Instead , th e approac h to orde r is embedde d in social context , an d include s th e proces s by which participant s constitut e social situations , and thereby partiapate in a dynamic that includes the mutual process of their constituting and being constituted by social order Thi s proces s is th e order , an d thi s orde r include s th e organizatio n of self an d society, since it is mutuall y constitutiv e of bot h Thi s dynami c speaks to "practice " (Bourdie u 1977, 1990), to relationship s between agenc y an d structur e (Gidden s 1979, 1984), to phenomenologica l perspective s (Heidegge r 1962, Merleau-Pont y 1962), to Vygotsky's perspectiv e o n languag e an d consciousnes s as bot h lodged withm a matri x of social activity (1962,

6

JANE Μ BACHNI K

1978) , a n d Bakhti n (1981) a n d Volosinov's (1973) perspective s o n lan g u a ge as e m b e d d e d withi n a matri x of h u m a n interaction , as well as formulation s o n indexica l m e a n i n g by Peirc e (1931-1958) , J a k o b s o n (1960) , Lyon s (1977) , Silverstein (1976) , O c h s 1990, a n d H a n k s 1990

I N S I D E A N D O U T S I D E AS O R I E N T A T I O N S T O T H E W O R L D

I will no w briefly explai n ' i n s i d e / o u t s i d e ' — a n d t h e J a p a n e s e counter part s uchi/soto—in m o r e detai l T h e m e a n i n g of insid e a n d outsid e is r e m a r k a b l e for two reason s it is largely assumed , a n d it is practical , o r situate d I n o t h e r word s we u n d e r s t a n d insid e a n d outside , n o t intellec tually, b u t largely t h r o u g h use , a n d we use thes e t e r m s constantl y in En glish As Mar k J o h n s o n trace s o u t t h e first few m i n u t e s o f a wakin g da y You wake out of a dee p sleep an d pee r out from beneat h th e covers into your roo m You graduall y emerg e out of your stupor , pull yourself out from unde r th e covers, clim b into your robe , stretc h out your limbs an d walk in a daze out of th e bedroo m an d into th e bathroo m You look in th e mirro r an d see your face starin g out at you You reac h into th e medicin e cabinet , take out th e toothpaste , squeez e out som e toothpaste , pu t th e toothbrus h into your mouth , brush your teet h in a hurry , an d rinse out your mout h At breakfast you perfor m a hos t of furthe r ιη-out moves—pourin g out th e coffee, settin g out th e dishes, puttin g th e toast in th e toaster , spreadin g out th e jam on th e toast , an d on an d on Onc e you are mor e awake you migh t even get lost in th e newspaper , migh t ente r into a conversation , which leads to your speakin g owion some topi c (1987, 31-32 ) "You" in thi s scenari o is t h e r e a d e r w h o empatheticall y take s t h e positio n of t h e wakin g p e r s o n I n s i d e / o u t s i d e ar e define d by t h e m o v e m e n t of o u r bodie s in space , as eviden t in you r e m p a t h e t i c imaginin g of pull in g yoursel f out from u n d e r t h e covers, climbin g int o you r r o b e , stretch in g out you r limbs , a n d walkin g out of t h e b e d r o o m a n d into t h e bath r o o m I n s i d e / o u t s i d e also involves m o r e abstrac t nonspatia l relation s (Johnso n 1987, 32) , suc h as e n t e r i n g m t o a conversatio n As indicate d above , t h e J a p a n e s e have specifically linke d 'inside ' a n d 'outside ' t o m e a n i n g s t h a t specify " s e lf a n d "societ y " T h e s e multipl e m e a n i n g s ar e r e p r e s e n t e d in p a i r e d sets o f terms , whic h i n c l u d e uchi 'inside ' versus soto 'outside' , ura Ίη-back' , 'wha t is k e p t h i d d e n fro m o t h e r s ' versus omote 'in front' , 'surfac e a p p e a r a n c e ' (Do i 1986), ninjoo ' t h e worl d of persona l feelings' versus gin 'socia l obligation' , honne ' t h e i n n e r life o f feelings' versus tatemae ' t h e surfac e world o f socia l obligations ' ( H a m a b a t a 1990, 134) N o r is thi s list exhaustiv e 3 Do i (1973b , 1986) point s o u t tha t t h e m e a n i n g s of eac h of thes e p a i r e d t e r m s overlaps , so t h a t aspect s of self cluste r at o n e pole , as uchi,

SELF, SOCIAL ORDER AND LANGUAGE

7

mnjoo, ura, and honne, while aspects of social life cluster at the other pole, as soto, omote, gin, and tatemae Doi's observation is important, since it directly links aspects of self and society with degrees of insidedness and outsidedness These double sets of terms have been widely observed and discussed as occurring throughout Japanese society, for example, in political hierarchy (Ishida 1984), large enterprise organization (Gerlach 1993, Hamabata 1990), small enterprise organization (Kondo 1990), family/household organization (Bachnik, in press), marriage (Edwards 1989), gender (Rosenberger this volume, Kondo 1990, Hamabata 1990), health and illness (Ohnuki-Tierney 1984), and religion (Hardacre 1986) It is also worth noting that several Japanese specialists have considered these terms as organizational "keys" to various realms of society Thus Doi uses omote and ura as central concepts for the organization of a Japanese self, which is characterized by situational shifting between the two modes 4 Lebra (1976) considers uchi/soto and omote/ura as central to her treatment of "situationahsm" in Japanese culture and behavior Nakane utilizes le/uchi (ha 'field' or 'frame') as a central organizational facet of Japanese society, according to which 'Japanese stress situational position in a particular frame" (1970, 2) 5 In fact, Nakane states that the corporate group based on 'frame'—from the household to baseball teams and political factions—is "the basic pnnaple on which Japanese society is built" (emphasis mine) (1970, 7) In this volume we focus especially on uchi/soto (although we include the other paired terms as well) We propose that the significance of uchi/soto extends beyond the directional coordinates of 'inside' and 'outside', and that, specifically, these terms link the directional coordinates with self, society, and language, moreover, they provide an organizational dynamic for this linkage To put this another way, we propose that the universally defined orientations for inside/outside are linked with culturally defined perspectives for self, society, and language in Japan Consequently, the organizations of all three have striking parallels, derived from the directional orientations of inside/outside Moreover, the directional coordinates of uchi/soto are basic to the other paired sets of terms as well, making uchi/soto the most fundamental of all the terms UCHI/SOTO

VIEWING THE RECORD

Consistent problems exist with the present delineation of uchi/soto and the other paired terms It is important to examine these issues, in order to specify how this volume constitutes a shift in perspective from those that precede it In a nutshell, we agree with the virtually unanimous

8

JANE Μ BACHNI K

characterizatio n ofJapanes e self an d social life as "contextua l " Ou r goal is to illuminat e mor e specifically how the organization of context is closely linked to the organization of both self and society Th e difficulties involved in thi s focus can be examine d throug h five sets of issues Th e first problemati c issue involves th e Japanes e social order , which ha s been linke d to "relativity," "situationahsm, " an d "situationa l shifting, " all of which are regarde d as basic "structura l principles " of Japa nese society (Nakan e 1970, Lebr a 1976) Smit h put s thi s succinctl y "[This] is a social orde r of a decidedl y curiou s kind " It is bot h ordere d an d highly situationa l "In so diffuse a system, where coul d authorit y possibly he? " (1983, 47-48 ) Mor e precisely, term s such as "situationahsm, " "situationa l ethic, " or "situationa l positio n in a frame " (Nakan e 1970, 2) are deepl y problem atic as "structura l principle s " Th e latte r shoul d conve y genera l mean ings tha t carr y beyon d situations , instea d thes e term s conve y tha t ethics , morality , an d possibly even social orde r are define d by th e situatio n Th e term s seem patentl y circula r ho w can the y explain anythin g abou t con text if the y require contex t to mea n tha t the y mean ' (Bachm k 1989, 240) 6 Th e secon d problemati c issue, hierarchy , is also linke d closely with social orde r Hierarch y is widely regarde d as on e of th e centra l organiz ing parameter s in Japanes e society In on e of th e mos t influenua l book s on moder n Japa n Nakan e posits a single rankin g order , define d by a "vertical principle " tha t exists irrespectiv e of situation s (1970, x, 29) Th e result is a rigidity, a fixedness tha t Nakan e fully acknowledge s in what she calls th e "rigid vertica l system " As a "structura l principle, " th e "vertical principle " shoul d conve y genera l meaning s abou t hierarch y in specific context s Yet, accordin g to Nakane , hierarchica l rankin g (accomplishe d via parameter s such as age, sex, an d status ) must be grasped from the situational context (1970, 30) She also note s tha t rankin g orde r is perceive d from an egocentric perspectiv e an d consistentl y uses th e ter m revive to qualify rankin g orde r 'Th e relative ranking s are thu s centre d on ego an d everyon e is place d in a refotive locu s within th e firmly establishe d vertica l system" (1970, 28, emphasi s mine ) Rathe r tha n applyin g a "principle " to defin e conduc t in situations , hierarch y seems to work in reverse, displayin g th e same difficulties as "situationahsm " an d "social order " above Thu s situationa l parameters , formulate d by relationship s between participants , seem essentia l to defin e hierarch y appropriately 7 Th e thir d problemati c issue in th e delineatio n of Japanes e self an d social orde r concern s Doi' s commen t tha t "poin t of view" is importan t for bot h uchi/soto an d omote/ura "[SJinc e soto an d uchi are differen t for eac h individual , what is soto for on e perso n may becom e uchi for a perso n include d in tha t soto Clearly, th e former' s omote become s th e

SELF, SOCIAL ORDER, AND LANGUAGE

9

latter's ura" (1986, 29). Uchi involves a point of view, as Doi notes, and this viewpoint "locates" social life in space and time. But how can such a point of view be accounted for in the organization of self and social order? Furthermore, if point of view is important to uchi organization, Nakane's definition of uchi (ba or 'frame') as a "structural principle," becomes problematic. If Japanese use uchi (ba) to "stress situational position in a particular frame" (1970, 2), how can uchi (and ba) as "principles" ignore the point of view embedded in the frame? How can we reconcile the fact that Nakane's definition of this "basic principle"—as a "structural principle"—excludes context, while uchi (ba) revolves around the organization of context? (Bachnik 1989). The fourth problematic issue involves the characteristic of shifting, which is closely related to point of view discussed above. Doi's statement that "[S]oto and uchi are different for each individual, [so that] what is soto for one person may become uchi for a person included in that soto" (1986, 29), includes a shifting of perspective. But Doi also relates shifting to the organization of self, when he comments that the organization of self does not hinge on omote and ura per se, but on the ability to ship, between them. "The ease with which one shifts from omote to ura and back again . . . is regarded as the measure of social maturity. . . . [One's] integrity rests upon the complete mastery of omote and ura" (1973b, 259). Shifting also occurs frequently in language, in the use of register (speech "levels" or politeness) and in social organization, where Lebra (1976) notes the same kind of shifting, and "relativity." But how do Japanese gauge the shifts from omote to ura and back again to demonstrate their "social maturity"? How do they gauge the shifts of language register, both within and between contexts? How can shifting be included in the organization of self and social order? Many of the terms used to define and describe Japanese society, including "situationalism," "relational self," "social relativism," evidence a frustrating circularity. These terms inform us that specific situations are key features of Japanese social order. Yet the terms themselves—as general labels abstracted from context—contradict what they tell us we need to know about specific situations in the way (or how) they inform us about these situations. The fifth problematic issue involves the Japanese self, commonly discussed as highly contextual, "relational" (Araki 1973), and social; meaning that it is defined in interaction with self and other. "There are no fixed points, either [for] 'self or 'other'. . . (Smith 1983, 77). [B]oth . . . can be expressed only in relational terms" (Smith 1983, 49). Person terms are also highly unstable in Japanese (Fischer 1964; Suzuki 1978, 1977, 1973; Bachnik 1982; Wetzel 1984). Problems with the organiza-

10

JANE Μ. BACHNI K

tio n of Japanes e self include most of the issues noted above: thu s self (1) resemble s th e social orde r in bein g "situationall y denned" ; (2) is "embedded " an d encompasse s a "poin t of view"; (3) is characterize d by shifting, bot h in th e instabilit y of perso n terms , an d th e lack of "fixed points. " Kond o note s tha t a "relational " self challenge s th e unit y an d coherenc e of th e Western subject, alon g with th e dualisti c perspective s tha t creat e distinction s between "self an d "(social) world" (1990, 34). But if so, doesn' t "self the n shar e th e same kind of circularit y as term s such as "situationalism " note d above? Ho w can we identif y self analyti cally, as self, withou t losing th e very aspect s of contex t tha t seem crucia l to its identity ? D o th e very connotation s of self, with its sense of genera l identit y independen t from social context , contradic t what we nee d to kno w abou t th e relatednes s of self to contex t for Japanese ? And if so, how do we get at th e organizatio n of "self withou t eithe r denyin g th e existenc e of what we are lookin g for, or becomin g embedde d in a con tradiction-in-terms ? All of th e five issues outline d above presen t differen t facets of th e same problem . Th e "relational " self, "situated " social order , "vertical principle, " embedde d "point s of view," an d "shifting" all pose th e same genera l questio n for th e organizatio n of Japanes e self, society, an d language use: namely , ho w can thes e aspect s of contex t tha t are particular , situated , embedded—o r identifie d with "uniqueness"—be identifie d with genera l "principles, " or pattern s tha t transcend context ? DELINEATIN G TH E ISSUE S IN

UCHI/SOTO

As mentione d before , two ways of perceivin g meanin g are at issue here . A semanti c (referential , ideational , or propositional ) construa l of mean ing views language (an d social life) in term s of what is communicated , rathe r tha n a set of contextua l relationship s tha t locat e thi s particula r referenc e in time an d space . Th e latte r construa l of meanin g addresse s th e questio n of how somethin g is communicate d (throug h indexical , or pragmati c meaning) , as well as what. Fo r example , all of th e in/ou t moves in th e wake-u p scenari o gave indexica l informatio n as th e perso n waking u p was reoriente d in myria d ways to tim e an d space. We can speak of th e first meanin g perspectiv e discussed above as semanti c or propositional—wher e th e semanti c value of a sign is con sidere d to have th e meanin g it ha s apar t from contextua l factors . Thi s follows a traditio n in logic an d linguistic semantics , in which meanin g can be treate d as an inheren t propert y of a propositio n regardles s of its contex t of use. Th e secon d kin d of meanin g is pragmatic , whereb y a relationshi p is stipulate d between referenc e an d context , detailin g spatiotempora l contiguitie s (suc h as distance) . Her e a referenc e is relate d

SELF, SOCIAL ORDER, AND LANGUAGE

11

to its context of use. Note that this use of "pragmatic" differs from that in everyday use, and is derived from Peirce's semiotics; here pragmatic meaning is dependent on context, since meaning is derived from relating something to its context. 8 I emphasize that it is impossible to use language without describing both general meaning and relating what we describe to social context; the foci of both these construals must be taken into account. The question here is how these perspectives are prioritized. Pragmatic meaning is far less obvious to the English-speaker than semantic meaning, although this has nothing to do with the existence or nonexistence of indexes. Rather, it has to do with what Silverstein calls an "ideology" of language; meaning assumptions—explicit and implicit—about the nature of language that determine its usage and the attitudes taken toward it (Crapanzano 1982, 181). The strong tendency to focus on reference (on words like chair, table, and television), rather than words like I, has a long history (Silverstein 1979; Crapanzano 1982), and "a linguistic ideology, going back at least to Aristotle, that gives priority to the naming, referential, denotative function of language (over such other language functions as the indexical)" (Crapanzano 1982, 181). Indexes have been treated by linguists largely as a residual category: "All of our analytic techniques and formal descriptive machinery have been designed for referential signs, which contribute to referential utterances in referential speech events" (Silverstein 1976, 15). Yet Silverstein argues that cultural meaning should be viewed as pragmatic, rather than semantic meaning (1976, 54); and that " 'pragmatic' analysis of speech behavior—in the tradition extending from Peirce to Jakobson—allows us to describe the real linkage of language to culture, and perhaps the most important aspect of the 'meaning' of speech" (1976,11-12). In this volume we present considerable evidence that pragmatic meaning is closely linked to cultural meaning for the Japanese. This can be illustrated, for example, by the relationships between indexical meanings such as 'inside/outside' and the meanings for self and society contained in the paired terms, uchi/soto, omote/ura, tatemae/honne, and giri/ninjoo. Links with indexing can also be made for the cultural focus on wrapping delineated by Hendry (1993, 1990) and Ben-Ari, et al. (1990). Even the common definitions of the Japanese as "preoccupied" with relationships (Lebra 1976), and involved in a group 'frame' (ba) or context (Nakane 1970), can be linked to indexing. Peirce considered that indexes signal spatiotemporal relations between two points in a context; so that indexes are instrumental in creating contexts. Indexical signs are thus denned by two important parameters: (1) relationships, linking the

12

JANE Μ. BACHNI K

inde x with an interpretiv e "ground " (Hank s 1992); an d (2) context , which is partiall y constitute d by such relationships . Lebra' s statement — " [T] h e Japanes e are extremel y sensitive to an d concerne d abou t social interactio n an d relationships " (1976, 2)—and Nakane' s identificatio n of th e Japanes e grou p organizatio n as based on "situationa l positio n in a given frame" (1970, 1) identif y the same two parameters, namel y relation ships an d context , as cruciall y importan t in Japanes e society. Thu s th e Japanes e "preoccupation " with social relationship s may actuall y be a preoccupatio n with relationship s tha t are indexicall y defined ; grou p organizatio n based on "situationa l positio n in a given 'frame ' " may actu ally refer to th e grou p as a contex t constitute d throug h th e proces s of indexing . We can also ask whethe r "situationalism, " an d "situationa l positio n within a frame " are term s referrin g to indexing , an d thu s pragmati c meanin g rathe r tha n "structura l principles" ; an d whethe r embedde d "point s of view" are example s of deixis an d deicti c referenc e points , rathe r tha n "extrem e relativism. " Is "shifting" anothe r exampl e of pragmaticall y define d meaning ? Are th e "vertical principle " (an d th e "horizonta l principle" ) example s of indexes , rathe r tha n structura l principles ? To addres s thes e issues, we nee d to elaborat e furthe r th e notion s of indexing , deixis, an d pragmati c meaning . Indexe s have bee n briefly defined above as signaling spatiotempora l relation s between two point s in a context . Th e relationship s between th e point s are actual : thu s smok e is an inde x of fire; a ra p on th e doo r is an inde x of someon e seeking entr y (Peirc e 1931-1958 , vol. 3). "Peirce' s ter m 'indexical ' . . . [is] based upo n th e idea of identification , or drawin g attentio n to , by pointing " (Lyon s 1977, 2:637). Peirc e furthe r specifies tha t indexe s do no t nam e objects; the y are used to identif y an d measur e degree s an d magnitudes . Thermometers , rules, scales, an d wind gauges work by indexing, an d all thes e example s includ e th e importan t specificatio n tha t "gauging" or pointin g mus t be carrie d ou t in relatio n to a locate d referenc e point . Ordinar y discours e is also organize d throug h indexing , which is a majo r functio n of deicti c devices such as tense , demonstratives , persona l pronouns , an d particularl y th e pronou n I. I n English discours e I function s as the indexica l referenc e point—th e zero poin t from which all th e conversationa l indexe s are anchored . I thu s point s to feature s of th e surroundin g context , such as here an d over there, an d "anchors " such pointin g by providin g a locatio n in referenc e to which th e pointer s can be understood . Thi s kin d of indexin g is known as deixis. Jesperso n ([1924] 1965: 219) andjakobso n (1957) have identifie d th e deicti c term s discussed above, such as I, you, here, there, now, then, an d so forth , as "shifters," becaus e th e referent s of thes e expression s con -

SELF, SOCIAL ORDER, AND LANGUAGE

13

stantly shift as the relationship between utterance and context changes— for example, when the person identified as I changes as the speaker changes. Thus if we cannot trace the speaker who shouts "I'm here!" in a cavernous, echoing building, we cannot understand where here is. The discussion above may seem elementary, but this is because indexing is so familiar to us that we accomplish it without thinking much about it. Moreover, indexing occurs not only with thermometers, rulers, and wind gauges, but throughout social life as well. Thus, for example, choices of politeness register or hierarchical ranking or kinship distance can index the social status of the speaker or the social relationship between speaker and addressee. Yet this indexing is not a mechanical kind of process. As Ochs notes: "Indexical relations are more complex than one-to-one mappings between linguistic forms and contextual features" (1990, 293). One reason for this is that "[indexical relations] cannot be fully understood without additional mappings—between a particular contextual dimension and sets of linguistic forms, and between a particular linguistic form and several contextual dimensions" (Ochs 1990, 293). Thus single linguistic forms—for example, in honorific register, can index a variety of sociocultural dimensions in context, including emotional and social relationships between speaker and addressee, or speaker and referent. Hendry indicates how this works in wrapping: "Appropriate attire may well correlate with other elements of wrapping, so that in a formal (ly wrapped) environment when participants in an event are wearing kimono or even formal Western dress, people who would otherwise use very little polite language to each other will adopt a more formal level and gifts presented would also need to be properly wrapped. On the other hand, informal occasions, such as those associated with sporting activities . . . may reduce the level of formality in speech, too" (1989, 628). But indexical relations are more complex than one-to-one mappings for other reasons as well. Thus language socializes not only through relying on explicit practices, or propositional content; by far "the greatest part of sociocultural information is keyed impliatly, through language use" (Ochs 1990, 291). This information is conveyed by the manner in which the utterances are delivered, as messages about the message that are conveyed along with the direct message (Bateson 1972). Thus the vast majority of these messages about acts and activities, identities and relationships, feelings and beliefs must be inferred (Ochs 1990, 291). For such reasons Ochs considers language socialization to rest on a theory of indexicality, and Lave's focus on studies of apprenticeship and children's math practice (1990, 310) also considers learning (and culture) to rest on pragmatic meaning.

14

JANE Μ. BACHNI K

Th e implication s of a cultura l focus on pragmati c versus semanti c meanin g are worth commentin g on . While semanti c (o r referential ) meanin g is a way of focusin g on cultur e as an accumulatio n of lexical, or genera l knowledge (as "what"), pragmati c meanin g focuses on establishing genera l meanin g throug h cultura l practice s tha t are learned , an d carrie d on , throug h a proces s of doin g (o r "how"). In thi s sense, pragmati c meanin g form s a kind of "ground " (th e habitualize d practices ) for which referenc e is a "figure," althoug h thi s is a bit to o tidy. Yet th e two facets of genera l meanin g establishe d throug h habitualize d practic e an d opennes s are bot h crucia l to pragmati c meaning . Sudnow' s ethnog raph y on learnin g to play improvise d jazz (1978) aptl y illustrate s ho w such practice s are a combinatio n of "the way we do things, " an d open ness or improvisation . Finally , th e creativ e or performativ e aspec t of meanin g expressed in practica l understandin g also help s to structur e th e ongoin g events. Fo r example , Silverstein remark s tha t in som e cases th e speec h signal is th e only overt sign of th e contextua l parameter , an d "Unde r thes e circum stance s th e indexica l toke n in speec h perform s its greatest apparen t work, seemin g to be th e very mediu m throug h which th e relevan t aspec t of th e contex t is mad e to 'exist'" (1976, 34) . 9 Th e way in which practica l understandin g constitute s event s include s th e ongoin g relationshi p between th e social participant s an d th e struc turin g of thei r social contex t (Gumper z 1982, 1992), which takes plac e constantly , in social interaction . At issue are two perspective s towar d social life, exemplifyin g th e two perspective s towar d meanin g discussed above, an d illustrate d by th e two perspective s towar d learnin g reporte d by Lave: on e in which th e subject matte r envelop s th e learne r in a "world" tha t is empt y of context ; th e othe r in which th e child' s under standin g gives significanc e to an d envelop s th e subject matter . Th e first view make s it seem as if th e proble m contain s th e learner , an d result s from viewing schoolin g "as th e institutiona l site for decon textualizin g knowledge [an d formal , or 'context-free ' learning ] so that , abstracted , it may becom e genera l an d henc e generalizable , an d there fore transferabl e to situation s of use in th e 'real ' world" (Lave 1990, 310). Th e secon d view—which includes , for example , apprenticeship — result s from viewing learnin g as "generate d in practice , in situation s whose specific characteristic s are par t of practic e as it unfolds " (Lave 1990, 310-11) . In thi s view th e learner' s understandin g encompasse s an d gives meanin g to th e subject matter ; thu s th e learner , and the context in which the learning is accomplished, envelop s th e subject matte r an d become s an integral part of the process of learning. Th e two perspective s on social life paralle l thos e on learning , with a bit of modification . I n th e first, or "context-free " perspective , th e organi -

SELF, SOCIAL ORDER, AND LANGUAGE

15

zation of "social order," and "self," are abstracted from situations, identified as structures, patterns, traits, and so forth, and in turn utilized to identify a general, context-free "self and "social order." The circularity in this process comes from failing to acknowledge the facets of situation that are inevitably embedded in both (Bateson 1972). The second view, in which self and social life are defined in practice, considers the context and the process by which each is constituted as integral to the organization of both self and social order. This is a complex dynamic, for social life is both structured and open; participants face ambiguity, potential misunderstanding, and conflict in the practice of social life; yet the habitual nature of the practices, and their associated "mappings," also constitute the social "world"—and even the selves—of the participants. It is worth elaborating briefly on the issues of semantic versus pragmatic meaning in relation to "self." Both "person" and "self are regarded as comprised of identities that are bounded off from the fluidity of context, and are therefore fixed and stable. The "fixedness" of self is supported by naming and, in particular, by proper names, just as the identity of person is supported by personal pronouns in English and other Indo-European languages. Here the organization of person is closely linked to that of self, since person is identified via referential indexes (as I, you, etc.), differentiated from context and viewed as a stable entity. Person and personal pronouns are also based on the individuals the locus of discourse since, in Indo-European languages, the distinction between I and you in first- and second-person pronouns is that between self and not-self (Bachnik, in press). 10 The identification of self through referential meaning makes it difficult to view self as constituted in practice, since the very practices by which the "fixed" self is identified, for example, through internal attributes such as traits, abilities, motives and values (Markus and Kitayama 1991), proper names and personal pronouns, incorporate sharply defined distinctions between self and not-self. Such distinctions are also sharply demarcated along inside/outside dimensions. Japanese terms for selfand person—like those of many other languages in the world—are unstable over time, constantly shifting and contextually dependent in use.11 This is true even for the usage of proper names (Smith 1983), and suggests that the distinctions between self and social context are not drawn as sharply for Japanese as they are for us—that self is not viewed as "fixed." Another way of viewing the relation between self and context is via a cline, or gradient, which extends from "self to "world." Becker and Oka have considered person in language in terms of such a cline extending from the speaker and ordering language in terms of degree of distance from the speaker. Rather than considering

16

JANE Μ. BACHNI K

p e r s o n by way o f referentia l indexe s explicitl y specifyin g p e r s o n , thei r perspectiv e allows a b r o a d s p e c t r u m of linguisti c c o m m u n i c a t i o n t o b e i n c l u d e d in t h e definitio n of p e r s o n . I t is w o r t h p o i n t i n g o u t tha t Becke r a n d O k a (1974, 229) conside r t h e "clin e of p e r s o n " as central : [P]erhap s the centra l threa d in th e semanti c structur e of all language s is th e clin e of person , an orderin g of linguistic form s accordin g to thei r distanc e from th e speaker . Between th e subjective, pointed , specific pronomi na l Τ an d th e objective, generi c commo n noun , between thes e pole s the words of all languages . . . are ordere d an d categorize d accordin g to thei r distance—spatial , temporal , social, biological , an d metaphorical—fro m th e first person , th e speaker , [emphasi s mine ] Thi s clin e ca n e x t e n d b e y o n d languag e t o t h e c o m m u n i c a t i o n of distanc e in socia l c o n d u c t (Leac h 1964; Bachni k 1 9 8 6 ) — i n c l u d i n g distinc tion s of formalit y a n d variou s kind s of w r a p p i n g behavio r discusse d by H e n d r y . Suc h a clin e ca n b e linke d t o pragmatic , r a t h e r t h a n proposi tiona l m e a n i n g , p r o d u c e d t h r o u g h a proces s o f d o i n g socia l activity, whic h woul d defin e self, o r p e r s o n , t h r o u g h t h e c o m m u n i c a t i o n of distanc e gradient s linkin g t h e speake r t o a context . As b o t h Becke r a n d O k a a n d Lave' s foci indicate , pragmati c m e a n i n g define s a differen t perspectiv e o n self a n d socia l life t h a n semantic , o r propositional , m e a n i n g . Pragmati c m e a n i n g ca n even b e viewed as a constitutiv e dynami c betwee n self a n d socia l life, in whic h eac h consti tute s a n d is constitute d by t h e other . T h u s t h e "socially defined " self a n d a perspectiv e o n structur e t h a t include s "agency " ca n b e viewed as two sides of t h e sam e coin . We p r o p o s e t h a t thi s dynami c is centra l t o Japa nes e self a n d socia l o r d e r as well, a n d t h a t it is exemplifie d in t h e organi zatio n of uchi/soto. A p p r o a c h i n g t h e organizatio n of self a n d societ y t h r o u g h pragmati c m e a n i n g ca n explai n t h e h e r e t o f o r e p r o b l e m a t i c characterizatio n of b o t h self a n d societ y for J a p a n e s e as "relational, " "shifting, " a n d "situational" ; a n d it allows all five of t h e p r o b l e m a t i c issues c o n c e r n i n g t h e organizatio n of self a n d societ y raise d above t o b e addressed . Consequently , we a r g u e t h a t t h e ideolog y of languag e (Silverstei n 1985) " t h a t gives priorit y t o t h e n a m i n g , referential , denotativ e functio n of languag e (ove r suc h o t h e r languag e function s as t h e i n d e x i c a l ) " ( C r a p a n z a n o 1982, 181) , a n d t h a t make s i n d e x i n g i n t o a residua l cate gory, needs to be turned around. As exemplifie d in t h e J a p a n e s e case , indexica l m e a n i n g shoul d b e t h e p r i m a r y focus , so t h a t referential , d e n o tativ e function s of l a n g u a g e — o r , t h e " p a t t e r n s " a n d " o r d e r i n g " of socia l life—can b e viewed as i n d e x e d by socia l participant s interactin g in social context . Specifically , a p p r o a c h i n g J a p a n e s e societ y t h r o u g h indexica l mean in g suggests t h e following:

SELF, SOCIAL ORDER, AND LANGUAGE

17

1. T h e organization of self should b e a p p r o a c h e d t h r o u g h its linkage to social context. Rather t h a n viewing self, o r person, t h r o u g h a set of specified terms, which differentiate self from context, as a "fixed" entity, self should b e a p p r o a c h e d t h r o u g h a cline of distance from self to "world," e x t e n d i n g Becker a n d Oka's c o n c e p t to social life as well as language. H e r e self would b e redefined in relation to situation. 2. T h e organization of a "situationally defined social o r d e r " must b e a p p r o a c h e d t h r o u g h linguistic c o m m u n i c a t i o n a n d social practice in context, r a t h e r t h a n principles, n o r m s , structures, a n d so on, abstracted from use (Silverstein 1976). Moreover, hierarchy must b e r e e x a m i n e d a n d defined in terms of social practice in context, r a t h e r t h a n t h r o u g h indexes such as "vertical" a n d "horizontal," which lose their points of reference when abstracted from context. 3. "Points of view" a n d "shifting" can b e included in t h e organization of social life. H e r e deixis, which has primarily b e e n associated with language, should b e investigated for its relationships to t h e organization of b o t h self a n d social order. 4. T h e dynamic by which self a n d social o r d e r are mutually constitutive can also b e investigated t h r o u g h t h e organizational dynamic evident in viewing self, social life, a n d language as g e n e r a t e d in practice. This dynamic speaks to issues that are extremely i m p o r t a n t for conceptualizating social order, including t h e way in which social participants b o t h structure a n d are structured by their social context, a n d t h e way in which factors such as contingency, agency, process, chaos, a n d o p e n n e s s can be integral to the conceptualization of order.

INDEXING: INSIDE AND O U T S I D E OF JAPAN

In reorienting t h e focus o n J a p a n e s e self, social life, a n d language, each of the four issues outlined above must be related beyond J a p a n . I will now briefly spell o u t four kinds of foci that d o this. 1. The "relational" or "interdependent" self is increasingly reported to be widespread, existing in virtually every area of the world, as acknowledged by psychologists (Markus and Kitayama 1991). 2. Incorporating "context" as a central, organizing focus for social life, has become a subject of attention, for example, in the essays collected in Duranti and Goodwin (1992). 3. Increasing concern has been expressed about the importance of incorporating "emotional aspects" of self-organization in a way that is integral to "social" aspects (Lutz 1988; White 1992), and of relating deictic organization to social organization (Hanks 1990, 1992). 4. There is considerable theoretical interest in including social participants anchored in time in the organization of society, and in developing rela-

18

JANE Μ. BACHNI K

tionship s between self and society, as mutuall y constitutive . Example s include Gidden s (1979, 1984); Bourdie u (1977, 1990); (Merleau-Pont y 1962; Heidegge r 1962); Vygotsky (1962, 1978); Bakhtin (1981), Volosinov (1973); and Peirc e (1931-1958) . I will now briefly elaborat e these issues. Th e organizatio n of a "relational, " contextuall y define d "social" self is hardl y peculia r to Japa n (Brune r 1990; Shotte r an d Gerge n 1989). Fo r example , in a recen t review articl e in psychology, Marku s an d Kitayama delineat e an "interdependent " self as representativ e no t onl y of Japa n an d othe r Asian culture s bu t of Africa, Latin-America , Islan d Pacific cultures , an d souther n Europ e as well (1991, 224-25) . Marku s an d Kitayam a not e strikin g contrast s between "interdependent " construal s of self an d tha t of th e "independent " self, "the so-calle d Western view of th e individua l as an independent , self-contained , autonomou s entity " (1991, 224). 12 Specifically, th e "interdependent " conception s of individualit y insist on th e "fundamenta l relatednes s of individual s to eac h other, " so tha t th e organizatio n of an "interdependent " self "canno t be properl y characterize d as a bounde d whole, for it change s structur e with th e natur e of th e particula r social context " (Marku s an d Kitayam a 1991, 224—27), an d therefor e involves constan t shifting. No t th e individual, but th e relationship, may be viewed as "the functiona l uni t of consciou s reflection " (Marku s an d Kitayam a 1991, 226) (emphasi s in original) . Th e "interdependent " self describe d her e strongl y resemble s th e Jap anes e self; moreover , th e characteristic s attribute d to it link th e "inter dependent " self to indexica l meanin g as well. Thu s self is characterize d by embeddednes s in contex t an d "shifting," so tha t relationship s between individual s an d interactivel y define d meanin g are prioritize d over th e individua l self an d "private" meaning . Moreover , as describe d earlier, inside an d outsid e constitut e basic orientation s in social life, by which all of us defin e th e world in relatio n to ou r embodie d selves. The y are par t of a huma n orientatio n to th e world tha t is largely practica l an d assumed . The y are also par t of a wider set of spatiotempora l orientation s tha t includ e up/down , front/bac k (necessar y for uprigh t posture) , an d left/righ t (Strau s 1967). Thes e orientation s are extremel y basic to huma n existence , an d are foun d in every known languag e an d society. Becaus e of th e commonnes s of bot h th e inside/outsid e orientation s an d th e "interdependent " self, it shoul d no t be surprisin g tha t term s which resembl e uchi/soto, an d in fact relat e self an d societ y to an inside / outsid e axis, are importan t organizationa l parameter s in othe r societie s besides Japan . Fo r example , Shor e (1982) explore s th e relationshi p between self an d society in Samoa , communicate d in th e categorie s of aga, actio n tha t is controlle d an d appropriat e to socially define d statuses, an d amio, actio n tha t is willful an d aggressive. I n Java, Geert z an d Keele r describ e th e Javanes e as movin g on two axes: between th e realm s of ex-

SELF, SOCIAL ORDER, AND LANGUAGE

19

ternal actions (lair) and inner experience (batin), and between rough and impolite (kasar) and pure and polished speech and behavior (alus) (Geertz 1960, 232-33; Keeler 1987). R. Rosaldo (1980) describes the Philippine Ilongot as shifting between beya and liget—"between a state of sociality and one of opposition and withdrawal, between a self at ease with its environment and one that stands apart" (1980, 44). And Abu-Lughod (1986) describes Egyptian Bedouins in terms of two aspects of self and society, one embedded in honor that is closely related to power, social constraint, and the "outside realm" of males; the other characterized by autonomy, deference, and a strong tradition of emotion and romantic love expressed in the "inside realm" of females through spoken poetry. In fact, Andrew Lock (1982, 32-33) speaks of two prominent axes in society, which move respectively between inner and outer and control and lack of control. At the same time a series of efforts have focused on reframing issues of emotion and feeling, to move them from residual categories to become a central focus in the organization of self, politics, and social life (M. Rosaldo 1984; White and Kirkpatrick 1985; Lutz 1988; Lutz and Abu-Lughod 1990; White 1992). It is worth noting that these issues are consistently cast within a rubric of inner and outer (as, for example, White's 1992 title). It is also worth examining M. Rosaldo's perceptive discussion of these issues: "What is important here is, first, the claim that meaning is a fact of public life and, second, the view that cultural patterns—social facts— provide a template for all human action, growth, and understanding" (1984, 140). Rosaldo indicates two problems central to social theory, each the inverse of the other. First is the separation of social order (as "structure," "principle," and so on) from persons interacting in social life. "Culture is, then, always richer than the traits recorded in ethnographers' accounts because its truth resides not in explicit formulations of the rituals of daily life but in the practices of persons who in acting take for granted an account of who they are and how to understand their fellows' moves" (140). Conversely, the divorce of self from relationships to others divorces one from anchorage in social context as well. In other words, immaculate distinctions between "inner" and "outer" spheres create Cartesian distinctions between mind and world, private and public, and emotional and social. "Thus, for ethnographers in the field, a set of rules that tells them what the natives do can never show them how and why a people's deeds make psychological sense because the sense of action ultimately depends upon one's embeddedness within a particular sociocultural milieu" (M. Rosaldo 1984, 140; emphasis mine). The issues of "public meaning" and "embeddedness within a particular sociocultural milieu" have been raised as major philosophical issues

20

JANE Μ. BACHNI K

by Wittgenstei n (1958) , phenomenologist s such as Heidegge r (1962) an d Merleau-Pont y (1962) , an d ethnomethodologist s influence d by Husser l an d Schut z (includin g Garfinke l 1967 an d Cicoure l 1973). Thes e issues have also been raised recentl y in perspective s on contex t as performativel y constitute d (Durant i an d Goodwi n 1992; Durant i an d Brennei s 1986; Hank s 1990, 1992). "Meanin g is collectivel y define d o n th e basis of recognize d (an d sometime s restated ) social relationships " (Durant i 1986, 241). Simila r issues are involved in recen t attempt s at locatin g th e event s of talk, an d of "self in th e broade r spectru m of com municativ e practice , including , importantly , politic s (Brennei s an d Myers 1984; White an d Kirkpatric k 1985; White an d Watson-Gege o 1990). Th e author s in thes e collecte d volume s focus on th e interactiv e role of situation an d event in shapin g bot h behavio r an d local theorie s of persona l experience . Two broa d sets of issues ar e involved here : (1) th e widesprea d existenc e of an "interdependent " self, an d of pattern s of relatin g inside / outsid e to self/societ y in a rang e of culture s besides Japan ; an d (2) th e relatio n of meanin g to context , as "public, " define d in a "we relation ship. " Such meanin g is define d throug h social practice , as "cooperativ e work between speake r an d hearer" ; an d exhibite d in linguistic commu nicatio n an d share d huma n interaction . Thes e two issues, in turn , can be linke d to th e separatio n of self from social life discussed by M. Rosaldo . Th e proble m ha s two sides: th e separatio n of social orde r (as "structure, " "principle" ) from person s interact ing in social life; an d th e divorce of self from social relationships , by a lack of anchorag e in social context . Th e doubl e proble m actuall y result s from a single deficiency : lack of relation of either self or structure to context. In fact, th e organizatio n of an "interdependent " self, by th e "fundamenta l relatednes s of individual s to each other, " bear s certai n marke d similaritie s to th e organizatio n of contextua l meanin g define d in use, throug h a "we relationship " an d share d huma n interaction . Eac h is defined throug h huma n relationships , throug h meanin g tha t moves between "us," in actio n an d in context . At issue her e is a perspectiv e o n "self an d "context " as two sides of th e same coin , eac h bein g th e rect o or verso for th e other . Th e difficulty is tha t we d o no t ordinaril y con sider "self an d "context " as a single subject—moreover , we do no t even pursu e the m within th e same academi c disciplines . Thi s is becaus e academi c distinction s themselve s reflect a series of pervasive dualisms , as Lut z point s out , "includ[ing ] especially th e shar p oppositio n of th e indi vidual to th e social which is reflecte d in th e existenc e of th e separat e discipline s of psychology an d anthropology ; th e concer n with a subjective versus an objective realit y . . . th e analyti c separatio n of though t an d action " (Lut z 1985,38) .

SELF, SOCIAL ORDER, AND LANGUAGE

21

At stake in the distinction between independent and interdependent selves is more than a mere categorization. Rather, these terms represent a set of dualisms formed around a radical separation between inside and outside that developed out of the scientific revolution in Europe. The separation of the physical universe, with its laws of gravity and motion, from the social universe, and ultimately even from the divine (Burtt 1932; Koyre 1957), effectively eliminated the social world from the cosmological universe, which was entirely physical, or "natural." Thus the ordering of the Newtonian universe was based on a delineation of dualities that separated both nature and self from the social world. These dualities included individual/social order; mind/world; ideal/material; objective/subjective; "outside" world/"inner" self; constancy/flux; unity/ multiplicity. Nor is this list exhaustive. These dualities have had a considerable impact on the way we view theory and everyday life (Luckmann 1973). Both the "autonomous" individual and the separation of individual (and consciousness) from "structure" or social order can be related to these perspectives on dualities. The Japanese perspectives on self and social order, which were considerably influenced by China, form a distinct counterpoint to those of the West and reveal a perspective on human nature that defines society as profoundly human, and self as quintessentially social. In the Japanese reading of Confucianism, "the centerpiece of the cosmos is human society and its manifold human relationships" (Smith 1983, 103). Moreover, the cosmos is moral as well as social, since there is "no recognized separation of the moral order from the actual" (Hall 1968, 29, cited in Smith 1983, 25). Both morality and social order are found in the phenomenal, actual world (Nakamura 1964, 362). Thus constancy is found in the flux of human existence; order in the midst of human feelings. Nevertheless, focusing on the difference between these two perspectives should not make us overlook a crucial similarity: both focus on tensions basic to human existence in all societies (Evens 1990). The relationship between self and society speaks to a fundamental paradox, that each and every social formation is at the same time a multiplicity and a unity (Stark 1962). Social life is carried on by concrete individuals, existing in time and space, yet cannot be ultimately reduced to the individual self, since society includes dimensions they do not originate, in the form of shared traditions (including those that produce and define the self as self). By the same token, social life is not reducible to shared traditions or normative order, since individuals define situations that are inherently ambiguous; make choices that have consequences; and attempt to remake their traditions. The tensions that exist between individual and society are ontological and are widely manifested in balancing such conflicting demands as self-expression versus self-sacrifice,

22

JANE Μ. BACHNI K

individua l desire versus social obligation , an d even choice s of forma l 13 versus informa l mode s of expression. In approachin g thes e tension s between self an d society, two perspec tives can be formulated , which in fact paralle l th e two formulation s of self—as independen t an d interdependent—an d th e two kind s of mean ing—semanti c an d pragmatic—outline d above. In othe r words, we can approac h self an d societ y as distinc t entities , resultin g in an analyti c separatio n of self from social order , which differentiate s th e "subjective" aspect s of self from th e "objective" aspect s of politica l econom y an d social order . Anothe r focus is also possible, which goes beyon d mer e shifting from on e pola r dualit y to th e other : to conside r bot h self an d society by focusin g o n th e relationship between them . Thi s relationshi p include s "the significanc e of th e social in constitutin g th e self, an d th e significanc e of th e self in bringin g share d social constraint s to bea r on actua l events " (Bachnik , in press). Thi s alternat e focus replicate s th e relation ship between th e "interdependent " self an d contextuall y define d approache s to meanin g at anothe r level, by focusin g o n th e relationshi p between self an d society as ontologica l dualities. 14 Considerabl e attentio n in Western philosoph y an d social theor y in th e past centur y ha s focused on reorientin g th e dominanc e of dualisti c legacies, leadin g to attempt s to develop a dynami c of a constitutiv e self an d social order , outline d above, from a numbe r of theoretica l perspec tives. Thes e includ e Nietzsche' s discussion of th e "eterna l return, " as well as th e theoretica l perspective s of "agency" an d its relatio n to social structur e (Gidden s 1979, 1984); conceptualization s of "practice " an d "ΛαδίίΜί " (Bourdie u 1977, 1990); phenomenologica l approache s to being-in-the-worl d (especiall y Heidegge r 1962); lived perspective s on tim e an d space (Merleau-Pont y 1962); Vygotsky's focus on languag e an d consciousnes s as lodged within a matri x of social activity (1962, 1978); Bakhti n (1981) an d Volosinov's (1973) perspective s o n th e relatio n between language an d context ; an d approache s to indexica l or pragmati c meaning , from Peirc e (1931-1958) . Yet th e subject matte r of such theo retica l approache s mandate s a sustaine d focus on self, social life, an d languag e in context— a focus on theor y and practic e tha t can be pro vided by examinin g uchi/soto in th e Japanes e context . INSIDE/OUTSID E AS BASIC HUMA N ORIENTATION S

We can no w relat e th e 'inside/outside ' dynamic s of uchi/soto to th e inde pendent/interdependen t approache s to th e organizatio n of "self" an d th e relationship s between self an d social orde r presente d above. All th e comple x issues concernin g relationship s of self to context , an d self to society discussed above are embedde d in th e organizatio n of uchi/soto. Th e issues can be approache d on thre e levels. I t is necessar y to clarify

SELF, SOCIAL ORDER AND LANGUAGE

23

here that "levels" do not refer to a stacked, nested set of dependent relations, but literally to issues stemming from different "levels" of social reality These levels of issues include (1) the process of indexing that produces an "interdependent" self, a "situated social order," and creates and sustains the interactional context, (2) the constitutive dynamic of the relations between self and society, in which each mutually constitutes and is constituted by the other, and (3) the delineation of relationships between what have largely been considered as polar dualities in the West These include individual/society, subjective/objective, emotions/social order, "inner" mind/"outside" world, unity/diversity, empincal/ideal, constancy/flux, content/form, to name a few Any instance involving uchi/soto—no matter how seemingly simple—encom passes all these levels Thus simple situations are themselves "wrapped" so that they have much more meaning than appears at first glance Moreover, uchi/soto addresses virtually the entire range of issues involving the relationship of self and social life in time and their continuity over time The implications of uchi/soto are that we must thoroughly reformulate our conceptualizations of self and social order, to acknowledge the facets of Japanese organization that otherwise resist classification What is perhaps most central to uchi/soto organization is its incompatibility with the notion of an external "world" which Brenneis notes "is referred to and reflected in language whether it be a physical world ('This is a table') or a social one ('This is a social superior')" (p x) The focus on an "external world" creates a perspective of social life as a scripted stage in which human beings perform the equivalent of "reading" lines or "following" norms But a focus on indexes like "this" requires a location in a particular situation, with relationships among speaker, hearer, and scene as a starting point Such a focus also defines a perspective that inverts the "external world" focus—where people interacting in context do not "reflect" a world already created, but participate in the ongoing dynamic of a social world literally constituted in action But by the same token, the lack of compatibility of uchi/soto with the notion of an external "world" must be extended to its lack of compatibility with the notions of "self as a bounded, inner, reflective psychological essence The bounded self reflects an "internal" world that is distinct from an "external" world, thereby replicating polar distinctions between self and world As such self is constantly differentiated from the world as a distinct entity, rather than being defined through relationships to the world via indexes such as deference and formality/informality distinctions It is necessary to spell out more carefully how indexes work, and the parallel ways in which uchi/soto works, in order to proceed to delineate the organization of this volume We approach the range of issues devel-

24

JANE Μ. BACHNI K

ope d above throug h Peirce' s formulatio n of indexica l signs, an d his particula r focus on pragmatis m as a movemen t between dualities . Peirc e develope d th e inde x as par t of a comprehensiv e philosophica l system, which is of specific interes t becaus e of his concer n with escapin g th e Cartesia n legacy of mind/matte r an d subject/objec t dualism . Hi s system include d bot h a set of phenomenologica l categorie s an d an equall y comprehensiv e sign system, based on thre e irreducibl e sets of relation ships, which were in tur n relate d to thre e kind s of signs: th e icon , index , an d symbol. 15 Eac h of Peirce' s thre e differen t kind s of signs—icon, index , an d symbol—is characterize d by differen t relation s of th e sign to its referent : "resemblance , contiguity , an d associatio n by convention , respectively" (Singe r 1980, 491). An iconi c sign resemble s its referen t (in th e form of a diagra m or an image, which may even momentaril y mak e us forget th e distinctio n between th e real an d th e copy) . Example s of icon s rang e from th e eucharisti c brea d an d wine to maps , diagrams , word orde r in a phrase , an d onomatopoeti c utterance s such as "quack quack. " An index ical sign has a direc t relatio n to its object an d qualifies its referen t by really occurrin g alon g with it (Quinn , persona l communication , 1992). Smok e an d fire, fever an d mercur y in a thermometer , a pointin g finger an d what it point s to , an d deictic s in language are all example s of index ical signs. Peirc e define s symbols as triad s becaus e the y have a join t relationshi p between th e object denote d an d th e mind . Th e relationshi p is established by conventio n (e.g., th e word dog) an d th e relationshi p is arbitrary. Example s of symbols are noun s such as "dog," "woman, " or "tree, " as well as thing s we acknowledg e as symbols, like th e America n flag (which , as a diagram , is also iconic) . Fo r Peirce , symbols are th e residua l class of signs, where neithe r physical similarit y no r contextua l contiguit y hol d between sign vehicle an d entit y signaled. Indexe s are highly contextual , signaling spatiotempora l relation s between two point s in a context . Indexe s do no t nam e objects; the y are used to identif y an d measur e degrees, magnitudes , an d number s of specific observations . 'Th e [reading s registere d by] barometer , plum b line , spirit level, weathe r vane, pendulum , an d photomete r are cite d by Peirc e as respectiv e indice s of observed specific pressure , vertica l an d horizonta l directions , wind direction , gravity, an d star brightness " (Singe r 1980, 490) used to identif y an d measur e degrees, magnitudes , an d number s of specific observations . What make s all thes e indexica l is th e zero poin t or referenc e poin t from which the y are "placed " (o r "read"). 1 6 Thu s indexe s work by gauging position s alon g a scale. Thi s can oper ate , for example , between high an d low (in gauging wind velocity); ho t an d cold (a thermometer) ; shor t an d tall (a measurin g rod) ; fat an d

SELF, SOCIAL ORDER, AND LANGUAGE

25

thin (a scale), and so on. But these points are always gauged against a located reference point. The logic involved in indexing relates the terms at each end of the scale inversely to each other. Being hot is inversely related to being cold; so that being hot means being not cold. Moreover, hot is defined according to degrees along the temperature scale, so that each degree of being hotter varies inversely (and precisely) with the corresponding degree of not being colder. The closer the reading is to 220° on a Fahrenheit Scale, the hotter it is, and the less cold (because the farther from zero) it can be. We can now relate this discussion of indexes to uchi/soto organization. For Japanese 'inside/outside' operates as a basic scale or axis, along which relationships can be indexed via degrees of "more" or "less." Virtually any kind of communication can be mapped in relation to this scale, including bowing, gift-giving, politeness or formality in speech, social space, choice of topic, and dress. In fact, virtually all of Ochs' list of pure indexes are included in the focus of what is indexed by the Japanese; meaning that a wide range of phenomena are indexed, including (but not limited to) social distance, status, relationships, settings, topics, affective and epistemological stances of social participants. Moreover, this entire variety of communication is broadly organized along an axis that can be related to Becker and Oka's distance "cline of person." The axis is produced by indexing self, in relation to society, as well as the converse: society in relation to self. The communication that produces this axis is organized by means of linking 'self with 'inside' at one pole, and 'society' with 'outside' at the opposite pole of the cline. Any number of qualities can then be associated with the 'self/inside' pole (including, for example, engaged, intimate, spontaneous), while the opposite qualities (such as detached, disciplined, ordered) are associated with the 'society/outside' pole. The social participant must "gauge" the appropriateness of a particular characteristic (such as selfrestraint), which also inversely indexes the opposing characteristic (spontaneity). Linked in this way, intimacy and socially required discipline are inversely related so that the more discipline (or self-restraint) that is gauged as appropriate, the less intimacy that can be expressed. To put this another way, one is constantly constraining or expressing 'self, in relation to the degree of social constraint, or relaxation of constraint, that is perceived appropriate. 'Self is inversely indexed vis-a-vis 'social order' in the same manner. Consequently, the gauging of self expression along the person cline involves a double dynamic in which the degree of self-expression is inversely related to a lack of social constraint. As mentioned before, uchi has at least two basic meanings as listed in the dictionary, Sanseido's 1982 Kokugo-jiten (Wetzel 1984, 9). One meaning for uchi is 'inside', which is juxtaposed to 'outside' or soto. This

26

JANE Μ BACHNI K

can be relate d to th e continuu m outline d above. Th e othe r meanin g for uchi is 'we', 'us', Ou r group' , 'me' , 'my', 'my group' , T . Uchi here is also identifie d as a ter m tha t mean s te (translate d as 'family', or 'house hold') , an d tha t specifically signifies on e aspec t of th e le—the living grou p of househol d members—thi s generation . In thi s sense uchi is locate d in social life (as 'we', 'ou r group' , 'thi s generation') . Uchi ha s been closely linke d to Japanes e grou p organizatio n (Nakan e 1970). But uchi is no t "the " group , bu t "our " group ; anothe r term , otaku, is used for "your" group . Th e embedde d viewpoint of uchi can thu s be linke d to deixis: uchi is a deicti c ancho r point . Th e English I is anothe r deicti c ancho r point . Thi s mean s tha t I function s in discours e as th e indexica l referenc e point—th e zero-poin t from which all th e conversationa l indexe s are anchored . Fo r example , to un derstan d "Tha t one! " two point s are necessary : an indexica l referenc e ('that') , an d an ancho r poin t from which on e locate s 'that ' ou t of all th e othe r possibilities on th e horizon . Anothe r way of puttin g thi s is in term s of a figure/ground relationshi p (Hank s 1992). 'That ' is mad e compre hensibl e (as a figure) by th e relationshi p of th e speaker' s ancho r poin t which "backgrounds " th e statemen t an d make s th e referenc e compre hensible . If we canno t find th e referenc e point—th e equivalen t of "you are here " on a map—we literally canno t "read " th e ma p indexically . 'That ' indexe s a spatia l relationshi p between th e speake r an d a person or object , while 'now ' direct s th e attentio n of th e addresse e to a particula r poin t in time. 'Now ' also presume s an axis—of tempora l orientatio n which in English span s from 'now ' to 'then' , bu t may also includ e 'thre e days ago', 'last week', as well as 'nex t week' an d 'thre e days from now' . Verb tens e indicate s a tempora l axis tha t require s two point s as well: th e speaker' s positio n at th e tim e of th e utterance , an d th e tim e h e or she is pointin g ou t ("I saw it yesterday") (Fillmor e 1975). I thu s anchor s th e discourse , allowing th e participant s to follow ou t th e pointers . Th e proces s by which I define s thes e spatiotempora l relationship s by anchorin g the m is called deixis, from th e Gree k word deiknumi 't o point' , or to 'indicate'. 1 7 Joh n Lyons' s definitio n of deixis is a hand y one : "the locatio n an d identificatio n of persons , objects, events, processe s an d activitie s talke d about , or referre d to , in relatio n to th e spatiotempora l contex t create d an d sustaine d by th e act of utteranc e an d th e participatio n in it, typically, of a single speake r an d at least on e addressee " (1977,2:637) . As a deicti c referenc e poin t it is extremel y significant tha t uchi is not an individual , bu t a speake r heated within a collectivity (so tha t th e mean ing shifts contextuall y from an 'us' at th e uppe r limit , to a 'me ' at th e lower) . Althoug h it is recognize d tha t all language s have deictic s (Ben veniste 1971; Kurylowicz 1972; Lyon s 1977) little in-dept h descriptio n of

SELF , SOCIA L ORDER , AN D LANGUAG E

27

actua l usage exists (Levinso n 1983; Harr e an d Muhlhausle r 1990). More over, attempt s at comparativ e account s of deixis have often assume d to o man y of th e deicti c feature s of English an d Indo-Europea n languages, particularl y tha t of th e individua l referenc e point , Τ (Wetzel 1984). While uchi can signify th e individua l Τ as well as th e collective 'we', th e contextua l usage of th e same ter m for bot h indicate s clearly tha t th e individua l is identifie d within th e backdro p of th e collectivity. This in no way eliminates the individual, bu t merel y specifies tha t th e forme r are always identifie d within a social context . Moreover , uchi is also closely iden tified with th e le, th e basic "unit " of social organization—th e family. Uchi as a collectivel y define d deicti c ancho r poin t therefor e raises importan t question s abou t th e relationshi p between deixis an d social organization , specifically thos e of family, small groups , an d work organizations . Soto provide s a relationa l contras t with uchi, so tha t eac h is define d in relatio n to th e other . Fro m th e vantage poin t of th e deicti c ancho r uchi an d soto may be reckone d from differen t "point s of view," just as Do i says, for what is outsid e (soto) for on e perso n become s inside (uchi) for a perso n include d in tha t outsid e (which to tha t perso n is inside [uchi]) (1986, 29). Th e locatednes s of a "poin t of view" in uchi result s in differen t orientation s to th e 'inside/outside ' coordinates . Thi s gives a basis for bot h th e "relational " aspec t of meanin g an d th e constan t shiftin g involved in creatin g it. Yet inside an d outsid e are linked , an d Do i make s thi s clear for omote/ura: "althoug h omote an d ura are clearly distin guished conceptually , the y are in fact closely related . Withou t omote ther e is n o ura, withou t ura n o omote—they are literally two sides (aspects ) of a single entity " (1986, 26). Althoug h othe r paire d term s exist in additio n to uchi/soto, includin g omote/ura, gin/ninjoo, an d tatemae/honne, which can also functio n as indexes, uchi/soto is mor e basic in delineatin g indexica l organization , for two reasons . First , uchi as 'us', 'ou r group' , an d so forth , is explicitly associate d with th e deicti c ancho r point , which is no t th e case for th e 'inside ' ter m of th e othe r sets. Second , uchi/soto is explicitly linke d to th e axis indexe d by inside/outsid e distinctions , throug h gauging degrees of insidednes s an d outsidedness . Fo r thes e reason s we focus on uchi/soto, rathe r tha n th e othe r paire d term s as identifyin g th e organiza tio n of pragmatic , or indexica l meaning . FORMULATIO N OF ISSUE S I N T H I S VOLUM E

We can no w relat e th e two basic meaning s of uchi in th e discussion above to defin e a basic paradig m for th e organizatio n of self an d society in Japan . Uchi as 'we', 'us', Ou r group' , 'me' , 'my', 'my group' , Τ is a deicti c ancho r poin t tha t "anchors " th e collectivity, giving it an embed -

28

JANE Μ BACHNI K

de d "poin t of view." Uchi as 'inside' , assume s a juxtapositio n with 'out side' o r soto, since uchi is define d in relatio n to soto. Th e 'inside ' mem bers constantl y defin e uchi/soto relationships , by gauging insidednes s an d outsideness . Th e juxtapositio n create s a continuu m of inversely relate d positions , gauged by th e indexin g of degree s of insidednes s versus degree s of outsidedness , an d by th e inverse relationshi p between 'inside' an d 'outside' , so tha t th e mor e inside somethin g is gauged to be, th e less outsid e it can be, an d vice versa. As th e locu s of th e "self," uchi is thu s linke d to th e organization of self, which is define d within a collectivity, even in moder n Japa n (Rosen berger 1992; Bachni k 1992a, 1992b; Kond o 1990; Hamabat a 1990; Edwards 1989; Smit h 1983; Lebr a 1976; Nakan e 1970). Becaus e th e deicti c locu s revolves aroun d engagement , uchi involves its member s in th e indexin g of thei r participatio n in social life. Juxtapose d to uchi is soto, an d a cluste r of characteristic s such as 'detached' , 'disciplined' , 'ordered' . Thes e represen t th e publi c 'outer ' 'ordered ' aspects , which characteriz e social life, an d also "index" it as 'abstract ' o r 'detached ' in relatio n to th e contextual , engaged , "located " placemen t of self. To examin e thes e meaning s mor e closely, the y accoun t for bot h th e 'inside/outside ' axis an d a tiltin g of th e axis, becaus e uchi as 'we', 'us' , Ou r group' , 'me' , 'my', T , as a deicti c referenc e poin t is therefor e an chored , while soto is abstract , objective, an d unanchored . Th e "tilt" or bias of th e axis thu s anchor s it at on e en d {uchi) in a specific social situation . Soto is detached , an d nonspecific—i n othe r words, general . Thu s th e relationshi p between uchi/soto also allows it to move between th e occupied , embedde d context , an d th e mor e genera l focus tha t carrie s beyon d context . Th e "tilted " axis, then , create s two sets of inverte d relationships : between self/societ y an d engagement/detachment . These , in turn , allow uchi/soto to be relate d to all thre e of th e levels previously outline d (p . 23). Keepin g in min d tha t all levels are encompasse d in each uchi/ soto distinction , a discussion of th e levels mus t procee d in a circle . We begin with th e mos t basic o r thir d level—the proces s of relatin g onto logical dualities , such as self/societ y an d engagement/detachment — carrie d ou t by indexin g the m inversely in relatio n to eac h other . Thi s relationshi p is continuall y bein g redefine d in interactiona l contexts , where a variety of communicatio n is indexe d alon g a distanc e cline , the n mapped , by additiona l indexing , for example , in term s of degree s of formality/informality . Th e entir e rang e of thi s communicatio n is cru cial in definin g self, which is thereb y relate d in multipl e ways to context ; thi s communicatio n mutuall y define s bot h a "relational " self, an d a "contextuall y defined " social order . Bot h self an d social orde r ar e thu s

SELF, SOCIAL ORDER, AND LANGUAGE

29

mutually denned by means of the interactional context. The mutual definition of self and social order constitutes level 1. Level 2 is constituted by the tilting of the axis and the resulting relationship between engagement and detachment. Through this relationship, the participants (as agents) are involved in the constitution of social life. The tilting of the axis is what allows a different organizational level of relationships to be indexed: linking concrete with abstract, and the specific context with what is carried beyond that context. This elaborates the dynamic through which self and social order are mutually constituted, for "selves" are constituted by an "ordering" which they themselves do not create, and "social order" encompasses the qualities of existence in time, openness, and ambiguity, which the linkage with self brings. The indexically defined dynamic by which self and social order are mutually constituted forms the second level of self and society in uchi/soto organization. Level 3, the most basic level, revolves around the relationships between ontological dualities. Since these dualities are gauged from an anchored perspective in social life, they also provide a means of indexing "perspective" itself. Perspectives are gauged by epistemic distinctions; meaning that degrees of sharedness/nonsharedness can be indexed, differentiating between knowledge that is familiar, known, experienced in common—or uchi—versus the unfamiliar, unknown, not experienced in common—or soto. Such distinctions differentiate my/ our cultural perspective (as uchi) from yours (as soto), thus enabling indexing of degrees of cultural shared/nonsharedness by gauging degrees of uchi/soto from the perspective of each deictic anchor point. This perspective on indexing addresses a conceptual problem arising from a dualistic focus, revolving around difficulties in portraying culture in terms of unity and multiplicity (which parallel the difficulties of relating individual and society). At the third level, then, uchi/soto defines relationships between unity and diversity which allow dualistic perspectives that have crept into theoretical approaches—namely, culture-as-unity versus cultural relativism—to be addressed by the native indexing of inverse relationships between cultural unity and diversity, which parallel the way in which self and society are indexed. This volume is organized in three parts, which exemplify the three levels of uchi/soto organization discussed above. In each part we attempt a shift in perspective, to reformulate the organization of Japanese self, society, and language, by focusing on indexical, or situated meaning. The volume is organized as a series that builds; although each chapter can stand alone, within the volume each is linked to build closely on the discoveries of the previous chapter (s). Moreover, the close relationship

30

JANE Μ BACHNI K

between indexica l meanin g an d contex t make s th e inclusio n of detaile d explication s of context s essentia l to th e volum e Narrative s are useful becaus e the y evoke situatedness , as n o detache d summar y can , by engaging th e reade r in th e narrator' s uchi Quantitauv e or systemic analyses provid e a soto kind of complemen t to this, with thei r abstracte d an d generalize d views Yet thes e quantitativ e analyses mus t also addres s con text (uchi) just as situate d narrative s mus t addres s generalit y (soto) Th e contributor s in par t 1, "Indexin g Self an d Social Contexts, " elaborat e th e first level of uchi/soto organizatio n by focusin g on th e role of indexin g uchi/soto in th e constitutio n of a variety of interactiona l con texts in language an d social life Th e theme s in par t 1 includ e th e delin eatio n of uchi as a deicti c ancho r point , bot h in language an d family/ househol d orgamzauon , an d detaile d empirica l exploration s of how participant s constitut e thei r constantl y shifting social context s throug h th e indexin g of relationship s between self an d other , includin g guest/host , husband/wife , vender/custome r Thes e paper s begin th e delineatio n of self an d social order , includin g dimension s of gende r an d hierarchy , an d develop importan t relationship s between th e organizatio n of deicti c ancho r point , indexing , agency, an d practic e in social life Just as th e paper s in par t 1 focus on th e mutua l constructio n of self an d social context , th e paper s in par t 2, "Failur e to Inde x Boundar y Disintegratio n an d Socia l Breakdown, " explor e th e secon d level of uchi/ soto organization—th e dynami c by which self an d social orde r are mu tually constitutiv e But becaus e th e constructio n of self an d social orde r is muc h mor e embedde d an d difficult to access tha n tha t of social con text, th e paper s in thi s sectio n all focus on conflict s producin g breakdown s an d destructio n of social orde r an d even self The y presen t a new focus which relate s Japanes e conflic t to th e proces s of indexin g Eac h of th e chapter s in par t 2 examine s a mistak e or failure to inde x appropri ately, an d explore s th e conflict-producin g imbalance s tha t result , in a variety of family-busines s settings, rangin g from large to very small scale In eac h instanc e th e destructio n result s from abuse of persona l power, an d lack of acknowledgin g collectivel y define d goals—or, in othe r words, from imbalance s in th e indexin g of self in relatio n to society Thes e conflict s represen t failures on th e par t of social participant s to act appropriatel y as agent s in th e constitutio n of social life, an d th e conse quence s are disastrousl y real, includin g interpersona l tensions , family disintegration , business bankrupcy , an d self-destructio n Par t 3, "Language as a For m of Life Chne s of Knowledg e as Chne s of Person, " explore s issues in th e thir d level of uchi/soto organization , of indexin g relationship s between dualism s Thi s sectio n put s togethe r th e theme s of th e volum e by demonstratin g th e remarkabl e consistenc y between th e organizatio n of uchi/soto in languag e an d in social life Her e

SELF, SOCIAL ORDER, AND LANGUAGE

31

epistemic distinctions basic to J a p a n e s e are explored, including the indexing of perspectives o n knowledge as uchi/soto ranging from uchi as firsthand, engaged, detailed, a n d individuated, or 'inside', to soto as s e c o n d h a n d , detached, c o m m u n i c a t e d from others, a n d generic, or epistemically 'outside' This section develops a c h n e of engaged a n d d e t a c h e d perspectives, considered to be the most widely indexed epistemological distinction in the language This c h n e is t h e n linked to the c h n e developed for self a n d social life in parts 1 a n d 2, tying the volume together, a n d bringing the r e a d e r full circle back to the b e g i n n i n g

NOTES

1 I am very grateful to the Institute for the Arts and Humanities, and a Pogue Fellowship, both from the University of North Carolina at Chapel Hill, for support while working on this chapter I also wish to thank my co-editor Charles Qmnn for his insightful help, both theoretical and editorial, in reading and discussing the several drafts of this chapter Any remaining shortcomings are my own 2 Wrapping is also the focus of a collected volume by Ben-Ari, Moeran, and Valentine, eds (UnwrappingJapan, Manchester University Press, 1990 ) 3 The list of paired terms also includes ooyake-goto 'public matters' and watakushi-goto 'private matters', hare 'sacred, extraordinary, formal' and ke 'profane, ordinary, informal' For additional discussion of these paired terms see Bachnik (1992a, 1992b, 1989, and in press) 4 In fact, the Japanese title of Doi's work, translated as The Anatomy of Self, is Omote to Ura or "Omote and Ura " Uchi/soto, which Doi regards as "corresponding" to the distinction of omote/ura, is introduced even earlier, in his work on the concept of amae (1973a, 40-44) 5 Nakane equates uchi with te, as a colloquial form of te (1970, 7) She regards te 'household' as a corporate group, and 'frame', which is the English translation for ba, as a "criterion which sets a boundary and gives a common basis to a set of individuals who are located or involved in it" (1970, 1) Ba is also translated as 'location', or 'field' in physics Problems with these definitions will be discussed in this chapter and in chapter 6 6 The specific contradiction is brought out clearly in Lebra, who states "First, the Japanese distinguish one situation from another according to the di chotomy of uchi and soto [Then they] differentiate their behavior by whether the situation is defined as uchi or soto (Lebra 1976, 112) Here the terms provide an underlying or structural organization for the empirical situation But Lebra also notes that "the essential point is that the uchi/soto distinction is drawn not by structure but by constantly varying situations" (1976, 112) This makes her explanation circular, "since she first specifies that uchi/soto defines the situation, but then argues that the terms are defined by the situation" (Bachnik 1989, 240) 7 In fact, Nakane notes that the "vertical principle" is not calculated by a

32

JANE Μ BACHNI K

single criteria , bu t by multipl e social axes, includin g relative age, sex, an d statu s Smit h (1983) an d Marain i (1975) also characteriz e bot h Japanes e hierarch y an d social orde r as "diffuse " The y link "diffuseness" to social relationship s "it is throug h th e measure d an d considere d respons e of all, as givers in some an d receiver s in othe r relationships , tha t th e social orde r is in par t maintained " (Smit h 1983, 47) 8 Fo r furthe r discussion of th e scope of pragmatics , which ha s a numbe r of distinc t usages, see Levinson (1983) an d Silverstein (1976) 9 "Certainly , th e English indexica l pronoun s I/we an d you (vs he/she/it/they) perfor m thi s creativ e functio n in boundin g off th e persona e of th e speec h event itself (Silverstein 1976, 34) 10 Boas point s ou t "Logically, ou r thre e person s of th e pronou n are based on th e two concept s of self an d not-self , th e secon d of which is subdivided , accordin g to th e need s of speech , int o th e two concept s of th e perso n addresse d an d th e perso n spoke n of," or addressed/proximat e versus referre d to/dista l (in Forchheime r 1953, 5) Greenber g state s "All language s have pronomina l categorie s involving at least thre e person s an d two numbers " (1963, 96) But Becker an d Oka disagree "I is no t I, you is no t you, an d we is no t we from on e languag e to th e next " (1974, 230) 11 Th e instabilit y of Japanes e "pronouns " ha s lon g bee n a subject of discus sion, bot h in linguistic s an d anthropolog y (Befu an d Norbec k 1958, Fische r 1964, Neustupn y 1977, Suzuk i 1978, 1977, 1976, Smit h 1983, Wetzel 1984, Kond o 1990) 12 Marku s an d Kitayam a acknowledg e tha t difference s exist amon g th e "interdependent " construal s of self amon g th e above cultures , thei r poin t is tha t conceptualizatio n within th e interdependen t categor y is derived alon g similar premises , which allows the m to be juxtapose d to th e "independent " construal s 13 I do no t mea n to imply tha t th e "independent " individua l exists in all societies, bu t rathe r tha t "even in societie s where relationa l tension s are empha sized, a social bein g is someho w differentiate d from th e social whole And even in societie s like th e Unite d State s where th e individua l is highly differentiated , relationship s between self an d othe r are bot h explicitly an d implicitl y acknowl edged in communication " (Bachnik , in press) 14 Th e ter m ontologtcal dualism is from T M S Evens, η d , I am indebte d to him for th e distincUon s mad e in thi s paragrap h 15 Th e triadi c structur e of th e semioti c sign is significant , for it allows Peirc e to combin e a theor y of significatio n with a theor y of communicatio n Thi s is in marke d contras t to Saussure' s semiology, "which define s th e sign-functio n as a dyadic relatio n of signifier an d signified tha t dispense s with bot h independen t object s an d subjects" (Singe r 1980, 491), yieldin g a view of cultur e as "hermeti cally sealed" (Danie l 1984, 15) Fo r Peirc e th e triadi c relatio n of sign, object , an d interpretan t include s th e speaker an d interprete r as well, makin g th e relation dialogica l 16 I wish to acknowledg e Charle s Quinn' s contributio n formulatin g thi s section on Pence' s signs 17 I am grateful to Steven Klein (persona l communication , 1990) for thi s informatio n

SELF, SOCIAL ORDER, AND LANGUAGE

33

REFERENCES

Abu-Lughod, Lila. 1986. Veiled Sentiments. Berkeley: University of California Press. Araki, Hiroyuki. 1973. Nihonpn no koodooyoshiki (Japanese behavioral patterns). Tokyo: Kodansha. Bachnik, Jane. 1982. "Deixis and Self/Other Reference in Japanese Discourse." Working Papers in Soaolinguistics 99: 1-36. Austin, Texas: Southwest Educational Development Laboratory. 1986. 'Time, Space and Person in Japanese Relationships," in J. Hendry and J. Webber, eds., Interpreting Japanese Society: Anthropological Approaches, 4975. Oxford: JASO. 1989. "Omote/ura: Indexes and the Organization of Self and Society in Japan," in C. Calhoun, ed., Comparative Social Research 11: 239-62. Greenwich, Conn.: JAI Press. 1992a. "The Two 'Faces' of Self and Society in Japan." Ethos 20 (1): 3-32. 1992b. "Kejime: Indexing Self and Social Life in Japan," in N. Rosenberger, ed., Japanese Sense of Self, 152-72. Cambridge: Cambridge University Press. In press. Family, Self, and Society in Contemporary Japan. Berkeley: University of California Press. Bakhtin, Mikhail. 1981. The Dialogic Imagination: Four Essays. Translated by C. Emerson and M. Holquist. Austin: University of Texas Press. Bateson, Gregory. 1972. Steps to an Ecology of Mind. New York: Ballantine Books. Becker, A. L., and I Gusti Ngurah Oka. 1974. "Person in Kawi: Exploration of an Elementary Semantic Dimension." Oceanic Linguistics 13 (1 and 2): 22955. Befu, Harumi, and Edward Norbeck. 1958. "Japanese Usages of Terms of Relationship." Southwestern Journal of Anthropology 14: 66-86. Ben-Ari, Eyal, Brian Moeran, James Valentine, eds. 1990. Unwrapping Japan. Manchester: Manchester University Press. Benveniste, Emile. 1971. Problems in General Linguistics. Coral Gables: University of Miami Press (Miami Linguistic Series No. 8). Brenneis, Donald L., and Fred R. Myers, eds. 1984. Dangerous Words: Language and Politics in the Pacific. New York: New York University Press. Bourdieu, Pierre. 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press. 1990. The Logic of Practice. Stanford: Stanford University Press. Bruner, Jerome S. 1990. Acts of Meaning. Cambridge, Mass.: Harvard University Press. Burtt, E. A. 1932. The Metaphysical Foundations of Modern Science. Garden City, N.Y.: Doubleday. Cicourel, Aaron V. 1973. Cognitive Sociology: Language and Meaning in Social Interaction. Harmondsworth: Penguin. Crapanzano, Vincent. 1982. "The Self, the Third, and Desire," in Benjamin Lee, ed., Psychosocial Theories of the Self, 179-206. New York: Plenum.

34

JANE Μ. BACHNI K

Daniel , Ε. Valentine . 1984. Fluid Signs: Being a Person the Tamil Way. Berkeley: Universit y of Californi a Press. Doi , Takeo . 1973a. The Anatomy of Dependence. Tokyo: Kodansh a International . 1973b. Omote an d Ura: concept s derived from th e Japanes e 2-fold struc tur e of consciousness . Journal of Nervous and Mental Disease 155 (4) : 258-61 . 1986. The Anatomy of Self. Tokyo: Kodansh a International . Duranti , Alessandro . 1986. "The Audienc e as Co-Author : An Introduction. " Text 6 (3):239-47 . Duranti , Alessandro , an d Donal d Brenneis , eds. 1986. The Audience as Co-Author, Special Issue of Text 6 ( 3 ) . Duranti , Alessandro , an d Charle s Goodwin , eds. 1992. Rethinking Context: Languageas an Interactive Phenomenon. Cambridge : Cambridg e Universit y Press. Edwards , Walter. 1989. Modern Japan through Its Weddings. Stanford : Stanfor d Universit y Press. Evens, Τ. M. S. 1990. Transcendence in Society: The Comparative Study of Conflict and Sacrifice in Social Movements. Greenwich , Conn. : JAI Press. n.d . Anthropology as Ethics or Anti-Dualism: Reason and Human Agency in a Postmodern Universe. Unpublishe d manuscript . Fillmore , C.J . 1975. Santa Cruz Lectures ofDeixis (1971). Bloomington : Indian a Universit y Linguistic s Club . Fischer , Joh n L. 1964. "Words for self an d othe r in some Japanes e families." American Anthropologist66 (6) : 115-26 . Forchheimer , Paul . 1953. The Category of Person in Language. Berlin : Walter de Gruyter . Garfinkel , Harold . 1967. Studies in Ethnomethodology. Englewoo d Cliffs, N.J. : Pren tice-Hall . Geertz , Clifford. 1960. The Religion of Java. Chicago : Universit y of Chicag o Press. Gerlach , Michael . 1993. Alliance Capitalism: The Soaal Organization of Japanese Business. Berkeley: Universit y of Californi a Press. Giddens , Anthony . 1979. Central Problems in Soaal Theory: Action, Structure and Contradiction in Soaal Analysis. Berkeley an d Los Angeles: Universit y of California Press. 1984. The Constitution of Society. Berkeley: Universit y of Californi a Press. Greenberg , Josep h H . 1963. UniversaL· of Language. Cambridge , Mass.: MI T Press. Gumperz , Joh n J. 1982. Discourse Strategies. Cambridge : Cambridg e Universit y Press. 1992. "Contextualizatio n an d Understanding, " in A. Durant i an d C. Goodwin , eds., Rethinking Context, 229-52 . Cambridge : Cambridg e Univer sity Press. Hall , Joh n W. 1968. "A Monarc h for Moder n Japan, " in R. E. Ward, ed., Political Development in Modern Japan. Princeton , N.J. : Princeto n Universit y Press. Hamabata , Matthews . 1990. Crested Kimono: Power and Love in theJapanese Business Family. Ithaca : Cornel l Universit y Press. Hanks , William F. 1990. Referential Practice: Language and Lived Space among the Maya. Chicago : Universit y of Chicag o Press. 1992. "Th e Indexica l Groun d of Deicti c Reference, " in A. Durant i an d

SELF , SOCIA L ORDER , AN D LANGUAG E

35

C Goodwin , eds , Rethinking Context, 43-7 6 Cambridg e Cambridg e Univer sity Press Hardacre , Hele n 1986 Kurozumikyoo and the New Religions ofJapan Princeton , Ν J Princeto n Universit y Pres s Harre , Rom , an d Pete r Miihlhausle r 1990 Pronouns and People The Linguistic Construction of Social and Personal Identity Oxford Basil Blackwell Heidegger , Marti n 1962 Being and Time Translate d by Joh n MacQuarn e an d Edwar d Robinso n Ne w York Harpe r an d Row Sein und Zeit Tubinge n Neo mariu s Verlag, 1926 Hendry , Joy 1989 "To Wrap or No t to Wrap Politenes s an d Penetratio n in Eth nographi c Inquir y " Man (N S ) 24 (4) 620-3 5 1990 "Humidity , Hygiene , or Ritua l Car e Som e Thought s on Wrappin g as a Socia l Phenomenon, " in Ε Ben-Ari , Β Moeran , an d J Valentine , eds , UnwrappingJapan Mancheste r Mancheste r Universit y Press 1993 Wrapping Culture Politeness, Presentation and Power in Japan and Other Societies Oxford Oxford Universit y Pres s Ishida , Takesh i 1984 "Conflic t an d It s Accomodatio n Omote-ura an d Uchi soto Relations, " in Ε Krauss, Τ Ρ Rohlen , an d Ρ G Steinhoff , eds , Conflict in Japan, 16-3 8 Honolul u Universit y of Hawai i Pres s Jakobson , Roma n 1957 "Shifters, Verbal Categories , an d th e Russian Verb," 1-14 Cambridge , Mass Russian Languag e Projec t Departmen t of Slavic Lan guages an d Literatures , Harvar d Universit y 1960 "Concludin g Statemen t Linguistic s an d Poetics, " in Τ Sebeok , ed , Style in Language, 350-7 3 Cambridge , Mass MI T Pres s Jesperson , Ott o [1924] 1965 The Phibsophy of Grammar Ne w York W W Norto n Johnson , Mar k 1987 The Body in the Mind Chicag o Universit y of Chicag o Press Keeler , Ward 1987 Javanese Shadow Plays, Javanese Selves Princeton , Ν J Prince ton Universit y Press Kondo , Donnn e 1990 CraftingSelves Power, Gender, and Discourses of Identity in a Japanese Workplace Chicag o Universit y of Chicag o Pres s Koyre, Alexandr e 1957 From the Closed World to the Infinite Universe Baltimor e Th e John s Hopkin s Press Kurylowicz, J 1972 "UniversauxLmguistiques,"in L Heilman , ed , Proceedings of the Eleventh International Congressof Linguists Bologn a Lave, Jea n 1990 "Th e Cultur e of Acquisitio n an d th e Practic e of Understand ing, " in G Herdt , R Shweder , an d J Stigler, eds , Cultural Psychology Essays on Comparative Human Development, 309-2 7 Cambridg e Cambridg e Universit y Pres s Leach , Edmun d 1964 "Anthropologica l Aspects of Languag e Anima l Catego ries an d Verbal Abuse," in Ε Η Lenneberg , ed , New Directions in the Study of Language, 23-6 3 Cambridge , Mass MI T Pres s Lebra , Takie 1976 Japanese Patterns of Behavior Honolul u Universit y of Hawai i Pres s Levinson , Stephe n C 1983 Pragmatics Cambridg e Cambridg e Universit y Press Lock , Andre w 1982 "Universal e in Huma n Conception, " in Ρ Heela s an d

36

JANE Μ BACHNI K

A. Lock , eds., Indigenous Psychologies: The Anthropology of the Self. London : Academi c Press. Luckmann , Thomas . 1973. "Philosophy , Science , an d Everyday Life," in M. Na tanson , ed. , Phenomenology and the Social Sciences, Vol. 1, 143-85 . Lutz , Catherin e A. 1985. "Ethnopsycholog y Compare d to What? Explainin g Behavior an d Consciousnes s Amon g th e Ifaluk, " in G. M. White a n d j . Kirkpatrick , eds., Person, Self, and Experience: Exploring Pacific Ethnopsychologies. Berkeley: Universit y of Californi a Press. 1988. Unnatural Emotions. Chicago : Universit y of Chicag o Press. Lutz , Catherin e Α., an d Lila Abu-Lughod . 1990. Language and the Politics ofEmotion. Cambridge : Cambridg e Universit y Press. Lyons, John . 1977. Semantics, Vols. 1 an d 2. Cambridge : Cambridg e Universit y Press. Maraini , Fosco . 1975. "Japan an d th e Future : Som e Suggestion s from Nihonjinron Literature, " in Giann i Fodella , ed., Social Structures and Economic Dynamics in Japan up to 1980, 15-77 . Milan : Institut e of Economi c an d Socia l Studie s for East Asia, Luigi Boccon i University . Markus , Haze l Rose , an d Shinob u Kitayama . 1991. "Cultur e an d th e Self: Impli cation s for Cognition , Emotio n an d Motivation. " Psychological Review 98 (2) : 224-53 . Merleau-Ponty , Maurice . 1962. Phenomenology of Perception. Translate d by Coli n Smith . London : Routledg e an d Kegan Paul . Nakamura , Hajime . 1964. Ways of Thinking of Eastern Peoples: India, China, Tibet, Japan. Honolulu : East-Wes t Cente r Press. Nakane , Chie . 1970. Japanese Society. Berkeley an d Los Angeles: Universit y of Californi a Press. Neustupny , J. V. 1977. "The Variabihty of Japanes e Honorifics, " in Proceedings of the Symposium on Japanese Soaolinguistics, 125-46 . Honolulu : Universit y of Hawaii . Ochs , Elinor . 1990. "Indexicalit y an d Socialization, " in G. Herdt , R. Shweder , an d J. Stigler, eds., Cultural Psychology: Essays on Comparative Human Development, 287-307 . Cambridge : Cambridg e Universit y Press. Ohnuki-Tierney , Emiko . 1984. Illness and Culture in ContemporaryJapan: An Anthropological View. Cambridge : Cambridg e Universit y Press. Peirce , Charle s Sanders . 1931-1958 . Collected Papers, Vols. 1-8. Edite d by C. Harts horn e an d P. Weiss. Cambridge , Mass.: Harvar d Universit y Press. Rosaldo , Michell e Z. 1984. 'Towar d an Anthropolog y of Self an d Feeling, " in R. Shwede r an d R. Levine, eds., Culture Theory: Essays on Mind, Self and Emotion. Cambridge : Cambridg e Universit y Press. Rosaldo , Renato . 1980. Ilongot Headhunting: A Study in History and Society. Stan ford: Stanfor d Universit y Press. Rosenberger , Nanc y R., ed. 1992. Japanese Sense of Self. Cambridge : Cambridg e Universit y Press. Shore , Bradd . 1982. Sala'ilua: A Samoan Mystery. Ne w York: Columbi a Universit y Press. Shotter , John , an d Kennet h J. Gergen , eds. 1989. Texts of Identity. London : Sage. Silverstein, Michael . 1976. "Shifters, Linguisti c Categories , an d Cultura l Descrip -

SELF , SOCIA L ORDER , AND LANGUAG E

37

tion, " in K. Basso an d H . Selby, eds., Meaning in Anthropology, 11-55 . Albuquer que : Universit y of Ne w Mexic o Press. 1979. "Language , Structure , an d Linguisti c Ideology, " in P. R. Klyne et al., eds., The Elements: A Parasession on Linguistic Units and Levels, 193-247 . Chi cago: Chicag o Linguisti c Society. Singer, Milton . 1980. "Signs of th e Self: An Exploratio n in Semioti c Anthropol ogy." American Anthropologist82: 485-507 . Smith , Rober t J. 1983. Japanese Society: Tradition, Self and the Social Order. Cam bridge: Cambridg e Universit y Press. Stark , Werner . 1962. The Fundamental Forms of Social Thought. London : Roudedg e an d Kegan Paul . Straus , Erwin . 1967. "On Anosognosia, " in E. W. Strau s an d R. M. Griffith , eds., The Phenomenology of Will and Action, 103—26. Duquesne : Duquesn e Universit y Press. Sudnow , David . 1978. Ways of the Hand: The Organization of Improvised Conduct. Cambridge , Mass.: Harvar d Universit y Press. Suzuki , Takao . 1973. Japanese and the Japanese: Words in Culture. Translate d by Akira Miura . Tokyo : Kodansh a International . 1976. "Language an d Behavio r in Japan : Th e Conceptualizatio n of In terpersona l Relationships. "Japan Quarterly 23: 255-66 . 1977. "Hito as a self-specifier an d otaku, kare an d kanojo as other-specifi ers, " in Symposium on Japanese Soaohnguistics, 195-204 . Honolulu : Universit y of Hawaii . 1978. Words in Context: A Japanese Perspective on Language and Culture. Translate d by Akira Miura . Tokyo: Kodansh a International . Volosinov, Valintin Nikolaevic . 1973. Marxism and the Philosophy of Language. Translate d by L. Matejk a an d I. R. Titunik . Ne w York: Semina r Press. Origina l publicatio n in Russian , 1929. Vygotsky, Lev Semenovich . 1962. Thought and Language. Translate d by E. Hanf man n an d G . Vakar. Cambridge , Mass.: MI T Press. 1978. Mind in Society. Edite d by M. Cole , V. John-Steiner , S. Scribner , an d E. Souberman . Cambridge , Mass.: Harvar d Universit y Press. Wetzel, Patrici a Jean . 1984. Uti and Soto (In-Group and Out-Group): SoaalDeixis in Japanese. Ph.D . dissertation , Departmen t of Linguistics , Cornel l University . White, Geoffre y M. 1992. "Emotion s Insid e Out : Th e Anthropolog y of Affect," in M. Lewis an d J. Haviland , eds., Handbook of Emotion. Ne w York: Guilford . White, Geoffre y M., an d Kare n Ann Watson-Gegeo , eds. 1990. Disentangling: Conflict Discourse in Pacific Societies. Stanford : Stanfor d Universit y Press. White, Geoffre y M. , an d Joh n Kirkpatrick , eds. 1985. Person, Self, and Experience: Exploring Pacific Ethnopsychologies. Berkeley: Universit y of Californi a Press. Wittgenstein , Ludwig. 1958. Phihsophical Investigations. Translate d by G . Ε. M. Anscombe . Ne w York: Macmillan .

Chapter 2 TH E TERM S UCHI ΑΝΌ

SOTO AS

WINDOW S ON A WORL D CHARLE S J . QUINN , J R . Fuku wa uchi1 Oni wa soto' 'Goo d fortun e inside ' Demon s outside' '

EDITORS ' INTRODUCTIO N

Charle s Q u i n n explore s a wide r a n g e of expression s t h a t use t h e wor d uchi 'inside' , a n d th e wor d soto 'outside' , o r b o t h , fro m t h e earlies t ext a n t text s dow n t o t h e p r e s e n t Distinction s draw n over t h e centurie s with t h e word s uchi a n d soto hav e s h a p e d a b r o a d r a n g e o f d o m a i n s of h u m a n e x p e r i e n c e , fro m t h e spatia l a n d t e m p o r a l t o t h e social, psycho logical, a n d even grammatica l I n d o m a i n afte r d o m a i n , uchi/soto reap pear s as a flexible, lived schema , whereb y a b o u n d a r y tacitl y o r explicitl y separate s a n insid e fro m a n outsid e While t h e particula r " c o n t e n t " o f thes e inside s a n d outside s varies with t h e d o m a i n , a n d while d o m a i n s c o m e a n d go, th e J a p a n e s e proclivit y t o structur e e x p e r i e n c e in thi s way ha s r e m a i n e d r a t h e r c o n s t a n t N o t every use o f th e wor d uchi ha s a sotobase d c o u n t e r p a r t a n d , in fact, th e lexical evidenc e suggests a t e n d e n c y to elaborat e t h e world m o r e in t e r m s o f uchi t h a n o f soto Suc h skewing is p e r h a p s onl y t o b e expected , given t h e h u m a n t e n d e n c y t o scrutiniz e a n d elaborat e th e nearb y in greate r detai l t h a n th e distan t I n th e aggregate , uchi expression s d e p i c t a world t h a t is ENCLOSED , CONCAVE , I N DOORS , FAMILY, LINEAL , "US, " SHARED , FAMILIAR , INFORMAL , PRIVATE , E X P E R I E N C E D , K N O W N , I N C O N T R O L , SACRED , a n d PRIMAR Y

Soto expres -

sions , by contrast , see t h e worl d as OPEN , CONVEX , OUT-DOORS , N O N -

FAMILY , EXTRALINEAL , "THEM, " NO T SHARED , UNFAMILIAR , FORMAL , PUB LIC , OBSERVED , UNKNOWN , UNCONTROLLED , PROFANE , a n d SECONDAR Y As a lexical resource , uchi a n d soto expression s ar e a t o n c e t h e historica l p r o d u c t s of a p a t t e r n e d way of living a n d i n t e r p r e t a t i o n s adaptabl e for p a t t e r n i n g life ane w T h e s e sam e o p p o s e d values p e r m e a t e t h e usage o f verbs t h a t assum e a n ' i n s i d e / o u t s i d e ' frame , suc h as transitiv e ireru 'insert ' a n d dasu ' p u t / tak e o u t ' , o r intransitiv e hairu ' e n t e r ' a n d deru ' e m e r g e ' T h e s e verbs ar e

UCHI AND SOTO AS WINDOWS ON A WORLD

39

typically used in relation to 'insides' that are known, private, controlled, and primary, and 'outsides' that are unknown, public, outside one's control, and secondary. Moreover, the opposed values or conceptual contrasts associated with the words uchi/soto and 'in/out' verbs like hairu/deru and ireru/dasu are congruent with epistemological distinctions indexed by a number of basic grammatical devices, as Quinn argues in the volume's concluding chapter. The lexical evidence Quinn presents and analyzes in this chapter suggests that Japanese people have long practiced an uchi/soto lifeway, or habitus, to orient themselves and others, in the business of their daily lives; the habitus continues to be remade somewhat in each occasion it shapes. By elaborating the range of experience that has been construed as uchi/soto, this chapter sets the stage for the remaining chapters of part 1, each of which delineates particulars of contexts inhabited as 'insides' and 'outsides'. THE TERMS UCHI AND SOTO AS WINDOWS ON A WORLD explores the ways in which uchi/soto 'inside/outside' orientations are expressed in words, in particular, the more established combinations in which the words uchi and soto have been used. That there is social behavior understood by the Japanese to be patterned in an uchi/ soto kind of way is abundantly evident in the vocabulary of their language. The lexical evidence indicates that uchi/soto orientations constitute a lifeway, a socially learned way of construing, approaching, and moving through one's world, in domains of experience as different as perception and interpersonal relations. The uchi/soto lifeway is thus a kind of habitus, or "system of acquired dispositions functioning on the practical level as categories of perception and assessment or as classificatory principles as well as being the organizing principles of action . . . constituting the social agent" (Bourdieu 1990,13). 2 As the wide range of lived contexts surveyed below suggests, the uchi/soto habitus is anything but a rigid, fixed, template, but rather is adapted and remade somewhat in every particular encounter it informs. Focusing on the words uchi and soto as a lexical resource not only facilitates access to the concepts that go by those names, but will also complement the other essays in this collection, by providing a glimpse at the most clearly labeled practice of the 'inside/outside' habit, the widespread use of words that mean just that. If the more ethnographic essays in this collection, with their experiential detail, are stories that illustrate the lived, indexical quality of what is tagged "uchi/soto," the present chapter is a summary about the tags themselves. It asks what varieties of experience each term is used to label and how different uses might be T H I S ESSAY1

40

CHARLES J

QUINN.JR

related to one another. It surveys informally, but at some length, the varieties of other words with which ucht and soto combine, in combinations that are both lexical (i.e., compounding) and collocational (separate words in a single, somewhat idiomatic construction),' as items in a larger web of other words. Under the working assumption that "the meaning of a word is its use in language" (Wittgenstein 1958, 20), it sketches an outline of such uchi/soto expressions as have emerged and continue to evolve. Their place in the lexicon looks to be as broadly and deeply established as the orientational habit is in social life. The expression of this orientational habit with the words uchi and soto is in fact one guarantee of that habitus's continued survival, for once people incorporate such terms into the physical, social, and psychological activities they engage in, use of those terms is no longer simply reflective of such behavior, but constitutive of it as well. The expressions in which we find the word uchi, the word soto, or both, are orientational. With these expressions, people get a fix on the world: themselves, other people, reports; in space, in time, in relation to other people; in the waking world and in their dreams, ad infinitum. They also use these expressions to project an orientation onto other people and things as they relate in space, in time, and with one another, even to the extent that there are grammatical constructions that have employed the words uchi and (so) to. Like other words, the words uchi and soto are at once the historical products of a patterned way of living and tools for patterning life anew. The striking fact about them is the breadth of domains of human experience in which each is used, from the spatial and physical to the social and psychological. As we shall see, ucht has been used more widely and variously than soto.4 Whichever of these experiential domains is described with the terms uchi and soto, the crucial feature is that of "in relation to." Like other basic orientations, such as "up/down," "front/back," "on/off," or "deep/ shallow," the uchi/soto opposition is grounded in "the fact that we have bodies of the sort we have, and that they function as they do in our physical environment" (Lakoff and Johnson 1980, 14). While the uchi/ soto habitus cannot be reduced to the fact of embodied existence, there is no denying the body's informing role in basic orientational schemata. Just how true this is for speakers of English, for example, is illustrated in a telling way by the following passage, an account of "only a few of the many m-out orientations that might occur in the first few minutes of an ordinary day" (Johnson 1987, 31-32): You wake out of a deep sleep and peer OMifrom beneath the covers mtoyour room. You gradually emerge out of your stupor, pull yourself out from under the covers, climb into your robe, stretch owiyour limbs, and walk in a daze out of the bedroom and into the bathroom. You look in the mirror and

UCHI AN D SOTO AS WINDOW S O N A WORL D

41

see your face starin g out at you You reac h into th e medicin e cabinet , take out th e toothpaste , squeez e out some toothpaste , pu t th e toothbrus h into your mouth , brush your teet h in a hurry , an d rinse out your mout h At breakfast you perfor m a hos t of furthe r ιη-out moves—pourin g out th e coffee, settin g out th e dishes, puttin g th e toast in th e toaster , spreadin g out th e jam on th e toast , an d on an d on Onc e you are mor e awake you migh t even get lost in th e newspaper , migh t ente r into a conversation , which leads to your speakin g out on som e topi c As J o h n s o n notes , som e of thes e use s of in a n d out involve physica l orien tatio n in space , while o t h e r s hav e t o d o with nonspatia l relations , suc h as e n t e r i n g into a conversatio n Bu t w h e t h e r t h e d o m a i n is physical , social , o r psychological , essentia l t o all thes e expression s is som e establishin g of relation s a m o n g physica l objects , a m o n g socia l entities , a m o n g event s T h e r e is n o " i n " a n d n o " o u t " e x c e p t in relation t o som e "this " o r "that, " som e " h e r e " o r " t h e r e , " t h a t is, t o som e context Wha t is m o r e , t h e rela tio n implie s a b o u n d a r y of som e kind , insofa r as it indicate s a n " m s i d e " a n d a n "outsid e " T h e b o u n d a r y ma y b e crosse d (i n t h e m o r e dynami c predicates , suc h as " p o u r out t h e coffee") o r it ma y simpl y separat e t h e observer fro m t h e observe d (a s in t h e m o r e stati c predicates , suc h as "peer in g out," "into t h e r o o m " ) T h e sam e essential s of relatio n a n d b o u n d a r y ar e implici t in t h e e p i t h e t t h a t provide s t h e subtitl e t o thi s essay Metaphor , a c c o r d i n g t o Aristotle , is "th e applicatio n t o o n e t h i n g of a n a m e b e l o n g i n g t o a n o t h e r " {Poetics 1457b) Thi s definitio n entail s n o absolut e oppositio n of m e t a p h o r i c a l m e a n i n g t o litera l m e a n i n g , a n d it is quit e c o m p a t i b l e with a view t h a t take s litera l m e a n i n g as simpl y conventionalized , " d e n a t u r e d " m e t a p h o r As Dwigh t Bolmge r ha s re m a r k e d , an y e x p e r i e n c e t h a t is n o t expresse d in t e r m s u n i q u e t o itself is known , in som e sense , metaphoricall y A n d insofa r as n o sign is identi ca l with its r e f e r e n t — s o lon g as "th e m a p is n o t t h e t e r r i t o r y " — t h e use of an y sign is t o som e e x t e n t m e t a p h o r i c a l U n d e r s t o o d in thi s sense , th e m e t a p h o r i c a l ac t is o f interes t t o u s becaus e it is t h e mos t c o m m o n way in whic h word s c o m e t o participat e in ne w m e a n i n g s Man y expres sion s like t h e Englis h o n e s above with " i n ( t o ) , " "ou t (of) " a n d thos e in J a p a n e s e t h a t we shal l e x a m i n e , suc h as uchiki 'shy' a n d soto-wanhki 'rea l discount' , ca n b e u n d e r s t o o d as p r o d u c t s of m e t a p h o r , 6 insofa r as a wor d o r p h r a s e t h a t ha s f u n c t i o n e d in o n e d o m a i n of e x p e r i e n c e (say, spatial , social ) is applie d in a n o t h e r real m (e g , psychological , eco n o m i c ) I n fact , it is well d o c u m e n t e d t h a t n a m i n g in ne w semanti c fields, suc h as p o r t i o n s of a c o m p u t e r disk, a n d less directl y perceptibl e aspect s of o u r e x p e r i e n c e , suc h as subatomi c particles , is c o m m o n l y d o n e by b o r r o w i n g fro m m o r e familia r fields a n d m o r e perceptibl e p h e n o m e n a 6 N o t onl y ar e t h e r e a g o o d m a n y uchi a n d soto expression s t o b e f o u n d outsid e t h e real m of t h e physica l a n d spatial , b u t mos t of t h e

42

CHARLES J. QUINN.JR.

basic ones are in evidence in the oldest extant written records. Others were clearly added as new domains emerged, such as those relating to financial transactions. The social domain is particularly rich in uchi and soto expressions, which place another person 'inside' or 'outside' the same bounded area as oneself. To judge from usage, the smallest social uchi is the individual; the largest, entities such as the Japanese nation, "the major industrial nations," or even planet earth. But social life is still but one room in a larger household, in which the words uchi and soto serve to structure a variety of experiential domains. While it is well nigh impossible to single out points in time when the use of uchi or soto was extended to a new dimension, it is nevertheless significant that the same terms are found in such a wide variety of semantic fields. This in itself bears testimony to the proclivity of the Japanese to construe much of their experience in ways that have 'insides', 'outsides' and, implicitly, boundaries. The uchi/soto orientation does not play out in the same way in every domain of meaning, but these are the three features we can expect to be shared, the aspects of uchi and soto on which analogies and extensions across domains will be based. If a survey is limited, as this one is, to the record preserved in dictionaries, it is important that those consulted go beyond the listing of meanings as definitions, identify distinct contexts of use, and illustrate these with examples from authentic sources. Attested contexts of use provide a surer foothold than explanations alone when one is reaching hard for contemporary meanings, but there is nevertheless no discounting their selectivity.7 In the largest unabridged Japanese dictionary, while there are approximately 405 entries for lexical (including compounds) and collocational uses of the noun uchi, there are but 157 listed for soto (including eight for its earlier form to). The evidence suggests that ever since the earliest written texts, uchi has been used in a greater variety of contexts than (so)to. There are many more expressions based on the application of uchi to nonspatial domains, such as the social and psychological, than there are for (so)to. The relative morphonological stability over time of the words themselves suggests that uchi has remained the more solidly established of the pair: while soto is developed from an earlier form, to, which is found in Nara (710-784) and Heian (794-1192) period texts, uchi has apparently always been uchi. This asymmetry in the degree to which each term has endured, developed, and been extended to new contexts of use is probably a consequence of the natural human proclivity to develop and elaborate that which is nearby and accessible, and to leave unelaborated the removed, distant, or inaccessible. Indeed, that which is significant in our biological and social lives is to a large extent that which is close enough to impinge on us. When the two are mentioned together, whether the scope is as large as a proverb or as narrow as a compound word, uchi invariably precedes

UCHI AND SOTO AS WINDOWS ON A WORLD

43

soto, as "in" precedes "out" in English "ins and outs," as "near" precedes "far," omote 'front' precedes ura 'back', and so on. The order in which uchi and soto appear together is an example of what Cooper and Ross (1975) have called the "Me First Principle," whereby semantically related but contrasted elements are frozen into a syntactic order that reflects a proximate-distal semantics. There are also phrasings which come in both an uchi version and a soto version, such as that illustrated by the epithet quoted at the outset of this chapter, or in such pairs of complementary terms as uchi-mawari 'inner [train line] loop' and sotomawari 'outer loop' or uchi-sen 'concave drawknife' and soto-sen 'convex drawknife'.8 Any beginning apparent to an observer will of course have its own beginnings, somehow inaccessable to the observer, and thus not recognized as such. The earliest attested tokens of uchi available today appear in texts compiled in the eighth century, and imply established usage that antedates the texts. A wide range of uses for uchi is evident by the tenth century, but for all we know, this too may have already been long established, common practice. One of the larger dictionaries of earlier stages of the Japanese language lists ten separate meanings for the word uchi, and cites an illustrative example or two for each, in effect claiming that the word was used in ten different ways. The earliest of these uses are from the eighth century collection of poetry, Manyooshuu (completed 759), while others date from the ninth, tenth, and eleventh centuries.9 Even by the rough gauge of dictionary entries, by the eleventh century uchi is attested in all of the primary domains in which it is used today: spatial, temporal, quantitative, social, and psychological. It appears unlikely that there was a time in recorded history when uchi was restricted to one, two, or even three domains. While to 'outside' is not attested in the numbers that uchi is, it too was also used in a variety of domains, such as tokata 'away, other direction' (= to 'outside' + kata 'direction'); tozama Outside alternative, other direction' (= to + sama 'appearance, style'); tosaburai 'outer guards' (= to + saburai 'serving man'); or totsukuni 'foreign country' (to + genitive tsu + kuni 'land'). There is a construction /verb-negative + to/ 'before [verb], while-not-[verb]' attested in Manyooshuu (yo nofuke-nu TO ni 'before the night grows late'), which appears to have contrasted with /verb-affirmative + uchi/ 'during [verb], while [verb]' (tamakiharu UCHI 'interval while alive'). Furthermore, to is used contrastively with uchi in a variety of domains of human experience, spatial and social, from the eighth through the fourteenth centuries. Some uses straddle two domains, as when a physical barrier enforces a social one. Uchi and to were also used together contrastively in a single phrase or clause from the earliest texts on, in a wide variety of genres, written down as early as 759. Although from the time of the earliest texts there

44

CHARLE S J QUINN,J R

were several words tha t coul d be considere d to be semanticall y oppose d to uchi (e.g., yoso 'other' , hoka 'outside') , onl y to appear s to have bee n paire d with it as systematicall y as soto is today . Even so, paire d contrastiv e uchi/to expression s ar e no t in evidenc e to anywher e nea r th e exten t we observe toda y of uchi/soto. While th e domain s of experienc e in which th e word uchi was used ru n th e gamu t from spatia l to social to psychological, muc h as today , th e ter m to was probabl y no t nearl y so widely used as eithe r its contemporar y uchi was or its descendan t soto is. At th e very least, it is clear tha t as a handl e o n variou s aspect s of experience , th e concep t of uchi ha s bee n employe d in Japa n from well before th e first writte n records . While th e cor e constellatio n of meaning s expressed with uchi is in evidenc e from early on , th e rang e an d productiv ity of its later contrastiv e pairin g with (so)to seems less in evidence . TURNIN G A PHRAS E INSID E OR O U T

Th e uses of uchi an d to atteste d from th e early texts are , in term s of sign types, indexical or symbolic, dependin g o n th e actua l contex t of use. 10 Thi s remain s just as tru e today . Th e compoun d words (e.g., uchi-mawan 'inne r [train ] loop' ) an d construction s (san-nm no uchi 'amon g th e thre e [people]' , hon oyonde iru uchi 'while readin g a book' ) tha t includ e uchi and/o r soto are mos t often used symbolically, which is to say as com mo n noun s meanin g an 'X kin d of inside ' or a 'Ykind of outside' , an d are no t anchore d in th e speaker' s tim e or plac e of utterin g them . Com poun d noun s in which uchi or soto specifies a secon d noun , such as uchimawan, represen t a lexicalized kin d of deixis, in which groun d zero is som e plac e or tim e othe r tha n th e speaker' s here-and-now , an d is fixed by th e semanti c fram e implici t in tha t word or phrase . If, on th e othe r hand , uchi or soto is th e hea d nou n modifie d by a clause, as in hon oyonde iru uchi, th e clause identifie s th e zero point . Such phrase s are commonl y used to groun d an act or event describe d in a superordinat e predicate , as in hon ο yonde iru uchi ni omotdashita ' [I] remembere d [it] while [I] was reading' . But insofar as tha t groun d zero is no t th e momen t of speakin g or writing, th e expression doe s no t signify in a basic indexica l way. Th e 'inside ' or 'outside ' in som e of thes e expression s can be an chore d in th e speaker' s own location , an d thu s used indexically, bu t thi s is a matte r of th e occasio n of use, an d is no t a featur e of th e words an d phrase s themselves . Sinc e it is th e centra l contentio n of thi s volum e tha t muc h social behavior in Japa n is indexed in an uchi/soto schema , th e questio n naturall y arises as to why we nee d to pain t a pictur e of th e symbolic uses of th e same schema . On e reaso n is th e genera l theoretica l poin t tha t Peirce' s thre e sign types (indexes , icons , an d symbols) are bette r understoo d as perspectives, modalities, or uses tha n as unchanging , invarian t entities . A

UCHI AND SOTO AS WINDOW S ON A WORLD

45

particula r word, or sign vehicle, may be used in differen t modalitie s at differen t times , so tha t a word tha t was a motivate d icon for a communit y in th e past may com e to functio n as an unmotivated , arbitrar y symbol for thei r descendants . Thi s is true , for example , of man y hanji 'Chines e characters' , th e componen t part s of which , while onc e transparentl y iconi c ("pictographic") , are n o longe r so. Fo r Americans , th e Star s an d Stripe s was onc e a diagrammati c icon , in tha t th e number s of stars an d stripe s correspon d to th e number s of presen t an d origina l states, bu t for man y toda y it is simply a symbol of th e Unite d States . Conversely , erstwhile symbols often com e to be used indexically . Several of th e so-calle d persona l pronoun s of present-da y Japanese , for example , are symbolsturned-indexes . Th e provenance s of words like wata(ku)shi Τ (formerl y 'personal , private') , boku Τ (formerl y 'humbl e servant') , kimi 'you' (formerl y 'gentleman , prince') , kisama 'you [rud e address] ' (formerl y 'exalte d person') , are fairly well known , bu t th e proces s by which thes e erstwhile symbols cam e to functio n indexicall y seems little appreciated. 11 Th e mai n reaso n for examinin g th e symbolic side of uchi/soto, however, is becaus e a catalo g of thes e uses will suggest just ho w man y aspect s of Japanes e life are routinel y an d overtly symbolized as 'insides ' an d 'outsides' . As several of th e early example s show, symbolic uses of thes e two term s have been ther e all along . An examinatio n of th e lexical evidenc e will no w reveal th e exten t to which Mc/ii/soto-oriente d indexica l behavio r ha s been informe d an d supporte d by symbolic uses of th e words uchi an d soto. Th e lists includ e a numbe r of uses of eac h tha t did no t survive th e thre e centurie s precedin g thi s one , bu t I shall no t paus e to identif y them , since my mai n concer n ha s bee n to suggest, however roughly, th e rang e of domain s in which th e distinctio n has served an orientationa l functio n in a tradition , or culture , over time . Even thos e uses tha t are n o longe r au couran t are nonetheles s par t of th e story of ho w Japanes e society is patterned , just as th e heartwoo d or knot s of a tre e contribut e to its presen t an d futur e form . Th e following taxonom y of uchi-based words an d phrase s is representativ e of what som e Japanes e lexicographer s have seen as warrantin g separat e entrie s an d example s of usage, n o more , n o less; th e validity of conclusion s drawn her e is nec essarily constraine d by thi s fact. UCHI: A CATALO G OF USE S BY DOMAI N

Th e spatia l domai n refers simply to bounde d areas, physically separat e from thei r surroundings . (See Uchi, Table 1.) Th e words an d phrase s listed describ e acts tha t occu r or condition s tha t obtai n in some relatio n to tha t bounde d space: it can be 'entered ' (uchi e hairu), 'worked on ' (uchisakuji) , experience d as 'dark ' (uchikura, presumabl y becaus e n o light can 'enter') , or it can serve as th e origin or groun d from which

UCHI , TABLE 1

SpaUa l uchi e havru 'go inside ' uchi kara kagi ο kakeru 'lock from inside ' uchi kara hiraku 'ope n from inside ' uchikura 'dar k inside ' uchisakuji 'interio r work, repair ' uchiniku 'endoplasm ' uchipasu 'insid e calipers ' ucktnon 'inne r measurement s (from inne r edge to inne r edge) ' uchizura 'interio r surface ' ('at-hom e visage') uchimawan 'inne r (beltway, train ) loop ' uchisarugaku 'no h performe d indoors ' uchiburo 'indoo r bath ' uchiniwa 'inne r garden , courtyard ' uchishirasu 'whit e pebble s insid e th e apro n skirtin g a no h stage' uchichootsugai 'inne r hinge s (o n bod y armor) ' uchibon 'inne r moat ' uchipoketto 'insid e pocket ' uchndo 'insid e well' uchiniwa 'courtyard , enclose d garden ' uchigooshi 'insid e lattice ' Uchi Mooko 'Inne r Mongolia ' uchisen 'conve x drawknife ' (name d for its concave cut ) uchvmata 'pigeo n toed ' uchigama 'insid e of legs turne d inward ' uchihachimonji 'insid e of legs turne d inward ' uchuanzun 'inne r par t of ankl e area ' uchigun 'hollowe d inside (sculpture) ' uchtkatagin 'cu t with graver facing inside of cut ' (meta l engraving) uchiumi 'bay, gulF uchitsu 'inlet ' uchiyama 'settle d mountain s (= aroun d one' s village)' uchimuki 'facin g inward ' uchigake 'insid e leg sumoo throw ' (Ther e are man y mor e such terms , an d som e are used in juudoo as well.) uchibutokoro 'inne r "pocket " in folds of kimono' uchisumi 'inne r corners ' (seatin g position s at tea ceremony ) uchisetsuin 'inne r toilet ' (for a tea room ) uchiroji 'inne r garden ' (for a tea room ) uchiban 'pape r or clot h stretche d inside a frame ' uchibe 'insid e part ' uchimisu 'inne r boxes at kahuki theater ' uchimiyoshi 'inne r timber s of bow of boat ' uchidoi 'inne r (concealed ) roo f gutter ' uchiwaku 'insid e frame ' uchisogi 'cuttin g inward (tip s of protrudin g shrin e rafters)' , 'shrin e rafter s cu t inward on thei r tips' Note: Other s listed in "Uchi with Soto, Table 1: Spatial. "

UCHI AND SOTO AS WINDOWS ON A WORLD

47

such acts as 'opening' or 'locking' the boundary (a door or window) are staged (uchi kara hiraku/kakeru). An uchinori 'inside measurement' describes the shortest distance between two bounding edges. Uchinori and uchipasu 'inside calipers' are so named because of their focus on this interval between bounding edges. The image of a container with a boundary separating an interior from an exterior is clearly evident in expressions of this domain. The other domains are also configured along the lines of a container schema, but they differ from these spatial uses in that while they assume an enclosed space, they are about more than that, too, in ways peculiar to each. The four compound words beginning with uchibori 'inner moat' are so named for the position of the second element in each compound, uchi-X. X here is implicitly located within a larger, bounded frame. An uchibori is so called because it is inside, with the castle it encircles, in relation to a sotobori 'outside moat'. The deictic zero point is thus the castle. An uchipoketto 'inside pocket' is a pocket that is enclosed or attached inside a piece of clothing, which is construed as a container with contrasting inside and outside; the deictic anchor point for this expression is the wearer, who is 'inside' the clothing. Uchiido is a 'well' 'inside' the household enclosure, and an uchiniwa is an 'inside garden', enclosed in the same space. Uchi Mooko 'Inner Mongolia' is south of the Gobi Desert and closer to China, the zero point implicit in this name; Soto Mooko 'Outer Mongolia' was further out from the same center, beyond Inner Mongolia. In keeping with its name, Uchi Mooko remains under control of China, but Soto Mooko is now its own 'People's Republic' (Mongoru jinmin kyoowakoku). Such names of things, activities, conditions, and places are instances of wc/ii-as-symbol; they come with an understood zero point that is not necessarily dependent on the speaker/ writer's own, present-time zero point. Other words used as symbols in the spatial domain specify the shape of some object. They fully or partially describe an outline with a containing shape, a bounded space with an inside and an outside. An uchisen 'convex drawknife', for example, is named for the con cave hollow it cuts in wood; the knife creates a shape that approximates the canonical container, since the hollow forms a partially enclosed space. Uchimata 'inward-turned legs' is one of several terms that the Japanese have devised to describe the stance assumed when the toes of both feet point inward, thus creating a partial enclosure. An uchiumi 'bay', like an uchitsu 'inlet', is a body of tidal water that is surrounded to some extent by land, and thus enclosed (as sotoumi 'open sea', for example, is not). Uchiguri 'hollowed-out carving' is likewise named for its enclosed shape. All of these compounds are instances of aspect-for-whole naming, or synecdoche. In this way, acts, states and objects have been named for the way in which they approximate, in their own particular manners, the canonical

48

CHARLE S J. QUINN , JR . UCHI , TABLE 2

Social, Domesti c uchi de asobu 'play at home/indoors ' uchigirai 'be loat h to stay indoors ; perso n with such a disposition ' uchi ο motsu 'maintai n a household ' uchi ο deru 'leave home ' uchi ο akete iru 'be away from home ' kane ο uchi e ireru 'contribut e (financially ) at home ' uchi e kaeru 'retur n home ' uchi no hito '[my ] husband ' (lit., 'insid e person' ) uchi no mono 'person/peopl e in [my] family' typically used deicticall y uchi no shujin ' [my] husband ' uchi ' [ I ] ' (female , primaril y Western Japan ) uchigenkan 'side entranc e (no t main ; for househol d members) ' uchigi 'casua l clothin g (for wear at home) ' uchiburo 'bat h take n at home ' uchigoshirae 'hom e made ' uchimago 'grandchil d who is chil d of successor couple ' uchi-kerai 'longtime/principa l vassal' uchi-shuu 'househol d employees ' uchi-shigoto 'piec e work don e at home ' uchi-kosaku 'tenan t farmin g (by servant or otherwis e in-servic e people) ' uchi no kaisha 'ou r firm (corporation) '

uchi/soto p a t t e r n : enclosed , encircled , inwardl y curved , a n d / o r close r t o a center . T h e flexible s c h e m a of a container , with its b o u n d a r y a n d out side, is basic, b u t is a p p r o x i m a t e d t o varyin g degrees , in differen t ways, d e p e n d i n g o n t h e d o m a i n a n d o t h e r c o n c e p t ( s ) b r o u g h t t o b e a r in a n individua l lexical expression . Le t u s t u r n n e x t t o a somewha t less physically enclose d k i n d of space . (Se e Uchi, Tabl e 2.) U n d e r thi s s e c o n d d o m a i n , social/domestic , ar e listed word s a n d phrase s tha t expres s aspect s o f t h e b o u n d e d space s in whic h p e o p l e shelte r themselve s a n d a n c h o r thei r socia l lives. S o m e o f thes e uses, suc h as uchi de asobu 'pla y a t h o m e / i n d o o r s ' , uchigirai 'b e loat h t o stay i n d o o r s ; p e r s o n with suc h a disposition' , o r uchigenkan 'insid e = side / b a c k / n o n - m a i n e n t r a n c e ' , m i g h t also b e place d u n d e r t h e spatia l cate gor y o f Tabl e 1, sinc e the y relat e t o som e locate d aspec t of t h e uchi. Bu t the y also relat e t o t h e n o t i o n o f a home, n o t j u st an y ' i n d o o r s ' , b u t in d o o r s in a h o u s e in whic h a family dwells; n o t j u st an y 'insid e e n t r a n c e ' , b u t o n e use d by t h e m e m b e r s of t h a t h o u s e h o l d a n d thei r familiars , while outsider s ar e g r e e t e d a n d a d m i t t e d a t t h e omote genkan 'fron t en t r a n c e ' . I t is in thi s fram e of referenc e t h a t phrase s like uchi no hito 'm y h u s b a n d ' ('perso n of t h e h o u s e h o l d ' ) a n d uchi ο motsu ' m a i n t a i n a

UCHI AN D SOTO AS WINDOW S O N A WORL D

49

household ' have thei r significance . Som e of th e activitie s tha t character ize an uchi as a family or a hom e are describe d by th e language collocate d with thi s word: playing there , managin g it, leaving, returning , belongin g to it, an d so on . Ther e is thu s a tendenc y to trea t uchi as a thing , an entit y tha t can be 'held ' (motsu) an d 'left empty ' (akeru), a locatio n to which on e 'returns ' (kaeru), from which on e 'departs ' (deru 'separates') , or int o which on e 'deposit s money ' (o-kane ο ireru). Th e homel y images of uchigi 'casua l clothing ' an d uchigoshirae 'hom e made' , like th e proverbs at th e en d of Table 3, remin d us tha t uchi is a space to which peopl e can retrea t an d (as an America n migh t say) "be themselves, " free from th e restraint s an d obligation s of nondomesti c society outside . Thes e kind s of feelings are expressed in man y of th e phrase s of th e psychological domain , which is take n u p next . In th e social/domesti c domain , then , we have th e uchi tha t is succeede d to an d managed , departe d from , returne d to , hande d down to th e next generation , an d so on . While it is often no t withou t physical attributes , such as a share d dwelling or land , uchi as social space is struc ture d by certai n roles an d activitie s its inhabitant s perform , a poin t mad e in detai l by several essays in thi s volume . Th e last five item s on thi s list, beginnin g with th e phras e uchi no kaisha 'ou r company/firm' , remin d us tha t household s are corporat e enterprise s an d tha t member ship may be ha d in a variety of roles othe r tha n thos e on e is bor n to . Thes e phrase s thu s provid e a bridge to th e commerce-relate d expressions listed in Table 4 below, th e domai n of social/nondomestic . But first, let us explor e th e psychologica l domain . In th e experienc e of such invisible phenomen a as feelings, attitudes , an d beliefs, metapho r an d metonym y provid e th e mean s for th e expression of som e rathe r basic meanings . (See Uchi, Table 3.) A futokoro is th e space, or natura l pocket , between th e oute r an d inne r layers of a kimono, an d th e uchibutokoro 'inne r futokoro' is th e space between th e inne r layer of kimono an d th e wearer' s chest . Th e futokoro is reache d int o mos t easily by th e right hand , which enter s just over th e hear t (kokoro) an d above th e stomac h {hard}, seats of one' s tru e feelings an d thoughts . If metonym y is based on contiguity , futokoro was in just th e right plac e to be applie d in th e expression of feelings an d thoughts . Th e words futokoro an d uchibutokoro have presumabl y bee n extende d int o th e psychologica l domai n becaus e what is inside thi s pocke t is accessible onl y to th e wearer of th e clothing , an d hidde n from peopl e outside . Thus , on e can 'conceal ' (shinobaseru) informatio n in th e uchibutokoro or suffer someone' s 'seein g into ' it (misukasareru). Uchikabuto 'inne r helmet ' refers to 'inside ' th e oute r shell of th e kabuto 'helmet ' tha t on e wears to protec t one' s head , or seat of strategi c planning . To have th e 'inside ' of one' s 'helmet ' 'seen into' , then , is to have one' s martia l in-

50

CHARLE S J. QUINN , JR . UCHI , TABLE 3

Psychological/Attitudina l uchibutokoro 'ope n space unde r innermos t layer of kimono', 'one' s inne r feelings, thoughts ' uchibutokoro ni shmobaseru 'concea l evil intent ' uchibutokoro ο misukasareru 'be seen through ' uchikabuto ο misukasareru 'have one' s designs seen through ' (lit., 'be seen throug h inside th e helmet' ) uchiki(mono) 'retiring , withdraw n (person) ' uchi ni kaenmite, yamashiku nai 'have a clear conscience ' (lit., 'lookin g within , nothin g to be ashame d of ) uchi hadaka de mo soto nishiki 'even if nake d at home , brocad e outside ' uchi-hadakan no soto-subon 'bossy at home , mee k outside ' uchi-benkei no soto-beso 'a lion at home , a bellybutto n (= th e opposite ) outside ' uchi-hirogan no soto-suboman 'expansive/outgoing/boss y at home , closed tight when outside ' sotozura wa yoi no desu ga, uchizura wa warm ' [S/he ] is affable outside, ' (lit., 'outsid e demeano r is good') , 'bu t is difficult at home ' (lit., 'at-hom e demeano r is bad') . soto-mizu no uchi hirogan 'naive of th e world, expansive/outgoing/boss y at home ' tent s o r scheme s f o u n d out . Castin g t h e psychologica l in t e r m s o f cloth in g a n d o t h e r item s in whic h o n e ' s bod y is enclose d provide s a way t o spea k of o n e ' s i n n e r feelings a n d attitude s as thing s t h a t ca n b e 'con cealed ' o r 'see n i n t o ' . T h e proverb s t h a t c o n c l u d e thi s list of use s in t h e psychologica l do m a i n contras t 'inside ' attitude s a n d persona l c h a r a c t e r with thei r 'out side ' c o u n t e r p a r t s , a n d suggest again t h a t uchi is a n enclose d spac e w h e r e o n e ca n a d m i t t o hi s o r h e r i n n e r feelings a n d "have a b a d atti t u d e . " Differen t c o m p o n e n t s of t h e spatia l s c h e m a of uchi/soto t h u s c o m e t o b e associate d with t h e m e a n i n g s a n d values o f differen t do mains . T h e word s soto-subon, soto-beso, a n d soto-suboman ar e c o m p o u n d s tha t describ e a p e r s o n ' s behavio r owfeide t h e h o m e (soto-) in inverte d imagery : subori a n d suboman describ e inwardl y p u c k e r e d shapes , as doe s -beso ( < heso 'navel') . Thi s imag e is a m e t a p h o r for a retiring , withdraw n personality , a n d is in eac h of thes e idiom s c o n t r a s t e d sharpl y with t h e sam e p e r s o n ' s behavio r insid e t h e h o m e , viz., -hadakari 'very expressive, " l o u d " ', -Benkei (a large, courageous , a n d volubl e folk-hero) , a n d -hirogan 'expansiveness' . T h e ironi c h u m o r in thes e proverb s follows fro m th e way the y refram e values assume d for socia l 'insides ' a n d 'outsides' , so t h a t uchi's o p e n n e s s is brash , a n d soto's qualit y of close d u p is timidity . Uchi-based expression s use d in t h e n e x t d o m a i n sho w clearl y t h a t

UCHI AND SOTO AS WINDOW S ON A WORLD

51

UCHI , TABLE 4

Social, Nondomesti c uchikata 'your wife', 'living quarter s (bac k of sho p in same building) ' (lit , 'inside direction' ) uchiguruma 'ricksha w for private use' uchiuchi de 'in private uchimaku 'inne r workings/insid e dope ' (lit , 'inside ' + 'curtain' ) uchimaku ο shttte iru 'kno w th e inside story' uchimakubanashi ο sum 'give someon e th e lowdown ' uchiwa 'insid e circle , familiars, insiders ' uchiwa ο nozoku 'pee k inside , obtai n inside information ' uchiwa ο abaku 'expose, uncover , blow open ' uchiwa no koto 'privat e matter ' uchiwa dooshi de 'between/amon g fellow insiders ' uchiwa de kaiketsu sum 'solve [a problem ] inside ' (ι e , withou t outsid e medi ation ) uchiwa no hi ο sarakedasu 'wash one' s dirt y line n in public ' uchiwa ni m 'understate ' nanigoto mo uchiwa ni yaru beki 'Whatever on e does, it shoul d be don e within bounds ' uchikenchi 'han (fief)-interna l surveying' uchikemi 'interna l estimat e of cro p yield' uchiwanbiki 'insider' s (l e , banker's ) discount ' uchikehai 'insid e (estimated ) stock pric e (befor e marke t opens) ' uchikashikan 'lendin g an d borrowin g for insider s only' uchikanjoo 'inside r bookkeeping , false receipts ' uchikoo Ίη-hous e proofreading ' uchizata 'lawsuit handle d privately', 'out-of-cour t settlement ' uchimome 'insid e strife' uchigiki 'love poem ' uchi oton no to-medeta 'inferio r in substance , impressive in appearance ' w h e n uchi refer s t o 'family' , tha t 'family' routinel y e x t e n d s b e y o n d t h e n u c l e a r o n e (Se e Uchi Tabl e 4 ) As N a k a n e Chi e p o i n t e d o u t two decade s ago, c o r p o r a t e enterprise s in J a p a n hav e lon g f u n c t i o n e d as a kin d o f uchi, a n d we find consider abl e use of t h e t e r m uchi in commercia l societ y Som e of thes e expres sion s relat e t o b o t h d o m a i n s , suc h as uchikata, literall y ' i n n e r [o f two] s i d e [ s ] ' , whic h was use d in two r a t h e r differen t sense s (1) t o refe r t o a n o t h e r ' s wife (identifie d with t h e 'inside ' of tha t h o u s e h o l d ) , a n d (2) t o refe r t o a dwellin g a t t a c h e d t o a s h o p (th e residenc e was 'inside' , o p p o s e d t o t h e s h o p 'outside' , see Molasky' s c h a p t e r below) I n t h e ricksha w business , t h e r e was also uchiguruma 'ricksha w for privat e ( n o t p u b he ) use' , a n d in publishing , uchikoo Ίη-hous e p r o o f r e a d i n g ' Two o l d e r

52

CHARLES J QUINN.JR

words, uchikenchi '(fief-) internal surveying of acreage' and uchikemi 'internal estimate of crop yield', refer to local matters, close to home, distinct from those involving the national government, the Tokugawa shogunate. The rhetoric of the next two sets of phrases, those built on uchimaku 'inside curtain' and uchiwa 'inside ring/circle', resembles that of uchibutokoro (Table 3, just discussed). Like the futokoro of clothing, maku 'curtain' and wa 'ring/circle' refer to an enclosed space, albeit in different ways. While futokoro names the enclosed space itself, maku and wa are more like kabuto 'helmet' in that they name only the bounding or separating entity. Thus, one can establish the image of an enclosed self by metonymy, as we saw in the use of uchibutokoro to refer to one's innermost feelings, or by synecdoche, in which part of the enclosure (the boundary or divider) stands for the whole. This synecdoche of using the bounding entity to refer to the enclosure (which is then a metaphor for 'inner thoughts') works as well as it does because insofar as someone encounters a boundary, there exists an 'inside' and an 'outside'. Uchi/ soto is one of those "tight" (cf. Rumelhart, McClelland et al. 1986), "basic" (cf. Lakoff 1987) schemata in which the presence of any component strongly predicts the others; put conversely, the components do not occur without one another. The dividing and concealing typically accomplished by 'curtains' and 'rings' or 'circles' of people make uchimaku 'inner curtain' and uchiwa 'inner ring' particularly apt in expressing the proprietary side of information, as something privately familiar to those who are inside either the 'curtain' or the 'ring', but concealed from those on the outside. Uchiwa-dooshi 'fellow insiders' uses the word uchiwa 'inner circle' to refer to individuals who belong there. The last two examples of the uchiwa-based expressions, uchiwa ni IU 'understate' and nanigoto mo uchiwa ni yaru beki 'Whatever one does, it should be done within bounds', utilize the word uchiwa 'inside circle' in an expression of manner to indicate the range or degree of the behavior specified in the final predicate. By these expressions, then, what one 'says' or 'does' is said or done within the bounds of an implicit uchiwa. In the same group of words, there is a series of terms for activities that have to do with producing and marketing goods (-kenchi 'surveying acreage', -kemi 'estimation of crop yield', -wanbiki 'discounting', -kehat 'hint', -kashikan 'lending and borrowing', -kanjoo 'bookkeeping') that are conducted by insiders on their side of the boundary. The uchi- variant of any of these has the common characteristic of being information that is accessible to those engaged in handling it, but not to outsiders. Words in this domain that do not carry a nuance of willful concealment at least denote a sense of the personal or private, as with uchigiki 'love letter' and uchizata 'private (out of court) settlement'.

UCHI AN D SOTO AS WINDOW S O N A WORL D

53

A s u m m a r y of t h e feature s of thes e first fou r domains—spatial , socialdomestic , psychological , a n d social-nondomestic—yield s a composit e imag e of a b o u n d e d are a tha t is enclosed , nearby , accessible , a n d familiar, whic h is n o t generall y visible t o thos e outsid e t h e b o u n d a r y , a n d insid e whic h o n e is free t o feel relaxe d a n d ac t in a relativel y u n c o n straine d m a n n e r . W h e n outside , by contrast , o n e is in less familia r a n d less controllabl e territory , a n d so in m o r e carefu l c o n t r o l over oneself. 1 2 Viewed fro m without , uchi is s h u t off, inaccessible , a n d concealed . Bein g insid e t h e b o u n d a r y gives acces s t o informatio n tha t is familiar , shared , a n d — b e c a u s e it is n o t accessibl e outside—valuable . An 'inside ' positio n is privileged ; it bring s with it a sharin g of t h e spac e with t h e o t h e r s wh o inhabi t it a n d a share d d e t a c h m e n t fro m thos e w h o d o not . All of thes e values ma y b e seen t o follow fro m focusin g o n thi s o r tha t featur e o f t h e spatia l s c h e m a implici t in uchi as it applie s in a particula r d o m a i n . I n tha t all fou r d o m a i n s ar e s t r u c t u r e d with referenc e t o t h e spatia l charac teristic s of uchi, t h e spatia l is basic; t h r o u g h a b d u c t i o n 1 3 a n d analogy , it provide s a m e a n s whereb y socia l a n d psychologica l e x p e r i e n c e ar e con ceived a n d talke d about . T h e imag e s c h e m a associate d with t h e t e r m uchi, with its enclosure , b o u n d a r y , a n d exterior , is of cours e n o t t h e onl y o n e t h a t serves t o structur e thes e d o m a i n s , b u t it is a majo r o n e . UCHI , TABLE 5

Tempora l Interval s pppun no uchi (ni) 'within ten minutes ' ichinichi, futsuka no uchi (ni) 'in a day or two' sono hi no uchi (ni) 'in th e cours e of th e day' kotoshi no uchi (ni) 'som e tim e thi s year' (i.e. , before next year) wakai uchi (ni) 'while young ' hon ο yonde iru uchi (ni) 'while readin g a book ' tokaku sum uchi (ni) 'in th e meantime ' (lit., 'while doin g thi s an d that' ) hi ga denai uchi (ni) 'befor e sunrise ' (lit., 'while th e sun ha s no t risen' ) ichikiro mo ikanai uchi (ni) 'befor e havin g gon e even a kilometer ' sono uchi (ni) 'befor e long' (lit., 'in th e interval' ) hakushu kassai no uchi (ni) 'amids t applause ' hinkon no uchi ni seijm suru 'grow u p in poverty' ichmen no uchi de sankagetsu shika ιηαι 'is absen t bu t thre e month s of th e year'

Le t u s t u r n n o w t o a pai r o f substitutio n frame s base d o n t h e t e r m uchi, delineatio n of b o u n d e d t e m p o r a l interval s (Uchi, Tabl e 5) a n d b o u n d e d sets (Uchi, Tabl e 6) . T h e secon d of thes e cover s j u st a b o u t an y d o m a i n in whic h c o u n t a b l e item s figure, a n d t h u s differs markedl y fro m other , single-domai n tables . If thi s use ca n b e said t o b e l o n g t o a d o m a i n of o u r e x p e r i e n c e , it is t h e cognitiv e o n e of n o n c e categorization . T h e

54

CHARLES J QUINN,JR. UCHI, TABLE 6

Bounded Sets

sono uchi (de) 'among them' (lit, 'being among them') ano sannin no uchi (de) 'of those three people' (lit., 'being among . . .') shitte iru Amenkajin no uchi (de) 'among Americans [I] know' daikoogyookoku no uchi ni haitte iru 'is among the major industrial nations' daihyoosaku no uchi ni hmru 'is among his major works' uchiwake 'breakdown (of particulars)' uchihyoo 'itemized table' gookei gohyakuman-en, sono uchi wa tsugi no toon 'a total of five million yen, the itemized details of which are as follows' expressions listed within " b o u n d e d t e m p o r a l intervals" a n d " b o u n d e d sets" of items are notably less value-laden t h a n those of the first four tables, since the spaces they demarcate can apply equally to virtually any social g r o u p , natural event, or h u m a n activity. These expressions share the function of grouping, clustering, or otherwise j o i n i n g aspects of socially meaningful experience into b o u n d e d units. It is significant that in most of these uses, uchi functions as a n indep e n d e n t lexical item, a n d does n o t form c o m p o u n d words. T h e structure h e r e is periphrastic (or syntactic, that is, words related to o n e ano t h e r ) , n o t lexical or morphotactic ( m o r p h e m e s j o i n e d to form words). T h e r e is good functional reason for these meanings to b e expressed in syntactic constructions rather t h a n lexical c o m p o u n d s : as a substitution frame, a syntactic construction allows great flexibility with regard to what is framed as an interval or set. By contrast, having a separate comp o u n d word for every conceivable interval or set to b e created with uchi would m e a n a r a t h e r large n u m b e r of c o m p o u n d s which, if theoretically possible, would be unwieldy. C o m p o u n d words are n o t substitution frames a n d are n o t a m e n a b l e to regular, habitual recomposition. Indeed, that is what makes t h e m words. Lexicahzation can b e t h o u g h t of as a kind of minting—pressing into a compact, h a r d form—so that the sign so m i n t e d is handily used a n d retains its shape over many uses, in many contexts. W h e n uchi combines with a n o t h e r t e r m to form a comp o u n d lexical item, this is a function of the currency in social discourse of the c o r r e s p o n d i n g c o m p o u n d c o n c e p t as a regularly isolable a n d referrable entity. T h e genius of syntactic expression, o n the o t h e r h a n d , allows for freer substitution of all m a n n e r of units which may in o t h e r contexts b e conceptually u n r e l a t e d to uchi. A looser frame, in which many different kinds of referents can b e freely substituted, works better for h a n d l i n g the vast n u m b e r of concepts that are not so regularly conceived in terms of b o u n d e d containers as to warrant their own uchi-

UCHI AN D SOTO AS WINDOW S O N A WORL D

55

compound . Th e looser th e structure , th e less th e nee d for semanti c rel14 evanc e amon g th e term s tha t are joined. Grammatica l construction s built aroun d th e word uchi take tha t word as syntacti c hea d in a nou n phras e of th e form /specifie r X + uchi/ 'interval-se t X'. Th e precedin g specifier takes th e form of on e of th e thre e majo r predicat e types, viz. nomina l (sono hi no uchi 'interva l of tha t day', sannin no uchi 'set of thre e people') , adjectival (wakai uchi 'interva l of bein g young') , an d verbal (hon ο yonde iru uchi 'interva l of readin g a book') , as well as boun d demonstrativ e determiner s like kono (e.g., sono uchi 'tha t interval-set') . Whethe r it is an interva l or a set tha t is specified is establishe d with th e meanin g of th e specifier and/o r th e larger con text. Thi s meanin g is furthe r clarified in th e fram e within which thi s /specifie r X + uchi/ constructio n is employed , so tha t for intervals , we often find th e constructio n marke d with locative particl e ni 'in' , to specify th e poin t at which somethin g occurs , for example , Yonde iru uchi NI wakatta 'I t cam e clear to m e while I was reading' . While interval s are spaces within which event s occur , sets typically are not . Instead , sets are referre d to as th e rang e within which som e othe r conditio n is attribute d or som e actio n described . A set is often subordi nate d to th e gerun d of th e copula , de, as in Ano sannin no uchi DE, Tanaka ga ichiban ii, literally 'bein g th e set of thos e thre e people , Tanak a is best' (= 'Of th e three , Tanak a is best') . Or, A to Β to C no uchi DE, dore ο toru ? 'Insid e of [= 'among' ] A, B, an d C, which will you take? ' Th e meta pho r implici t in /uchi X + de/ is tha t X is a containe r with an interna l structur e of members , 'given' (de) which , somethin g hold s tru e of or happen s to on e or some of thos e members . Sets can also be 'entered' , as illustrate d by expression s like daihyoosaku no uchi NI hairu 'is amon g [her ] representativ e works'. Th e last domai n listed for uchi, tha t of "partitive, " is actuall y a subtype of th e use tha t demarcate s a bounde d interva l or set. Eac h exampl e her e indicate s a subinterva l or subset of som e implici t larger whole. (See Uchi, Table 7.) All bu t th e last two of thes e example s are compoun d words, shape d for read y use in particula r domains . While mos t of the m are neutra l with regard to social valuation , ther e is such value in th e notio n tha t th e uchi portio n of a paymen t or delivery du e is th e par t mos t unde r one' s control , an d it is mor e basic—a preconditio n for com pletin g th e transaction . Ther e is also value in th e two words from th e no h theate r an d th e incens e ceremony , which emplo y uchi to indicat e tha t thi s subset ('te n incenses' , Ί 0 0 no h plays') is primary , principal , an d mor e importan t tha n other s outsid e it. I n sum, then , objects, events, people , an d informatio n describe d as uchi ten d to be conceive d of as enclosed, nearby, accessible, familiar, an d controlled, in an engaged realm in which informality is th e rule . Tha t which is containe d is valued

56

CHARLES J QUINN.JR. UCHI, TABLE 7

Partitive (Bounded Subintervals/Subsets) uchibarai 'partial payment, money partly paid' uchikm to shitejuuman-en '¥10,000 as down payment' uchiage 'partial payment for goods/of a loan' uchiwatashi 'partial delivery' uchun 'partial payment of a bill' uchigashi 'cash advance' uchigawa 'tributary (of a river)' uchitokumi 'primary ten incenses' (incense ceremony) uchihyakuban 'primary 100 noh plays' juunin no uchi kunin made 'nine out of ten people' kyuuryoo no uchi kara chokin suru 'put part of one's income into savings'

as important, m o r e so than what exists outside. S o m e o n e else's uchiis o n e to which o n e has n o t gained entry, a n d thus soto, typically detached, inaccessible, concealed, a n d uncontrolled. Now that we have seen a bit of the inside story, let us have a look o n the outside. Again, the catalog is n o t exhaustive a n d the categories or d o m a i n s are informally defined. SOTCK A C A T A L O G O F U S E S BY D O M A I N

T h e way in which the t e r m soto is used in the spatial d o m a i n resembles what we observed of uchi: it describes a position relative to some frame or o t h e r entity, or it describes a shape. As with uchi, w h e n soto is t h e first e l e m e n t in a c o m p o u n d n o u n , it indicates the position or shape associated with what the second e l e m e n t names. For example, sotokazan 'exterior decorating' is kazan 'decorating' on the 'outside', as sotomata 'inside of the leg outward' is mata 'inside of the leg' h e l d in an outward-facing posture, a n d sotosen 'concave drawknife' is a sen that creates a convex cut, a surface that resembles p a r t of the 'outside' of a container. T h e first five examples in the list in Soto Table 1 contrast ' i n d o o r s ' to ' o u t d o o r s ' in the context of a house, a n d are p e r h a p s representative of the first uses l e a r n e d by children. Some examples that follow, beginning with soto kara 'from [the] outside', indicate that soto is one's starting point in approaching an uchi; in soto kara nagameta dake da 'I've only seen it from the outside', it is clear that gazing o n s o m e t h i n g from without does n o t measure u p to seeing it from within. Not all these expressions have counterparts in uchi-; there is, for example, n o *uchikasa ' i n n e r penu m b r a ' of heavenly bodies, a n d n o *uchiera 'concealed gills', presumably since these are either contradictory or the u n m a r k e d , normative case, a n d n o t in n e e d of a m o r p h e m e that overtly specifies position.

SOTO , TABLE 1

Spatia l uchi no soto de Outsid e th e house , outdoors ' soto de asobu 'play outdoors ' soto de matte 'Wait outside. ' soto wa samui 'It' s cold outside. ' soto wa moo kurai 'It' s dar k outsid e already. ' ana kara soto e hikidashita ' [We] pulle d it ou t of its den. ' soto kara akeru 'ope n from th e outside ' soto kara to ni kagi ο kakeru 'lock th e doo r from th e outside ' soto kara nagameta dake da 'I've onl y seen it from th e outside. ' sototsunagi 'outdoo r hitchin g (post) ' sotobenjo 'outhouse ' sotonagashi Outdoo r (washing) sink' sotobtraki ni nam 'open s out ' soto-omote 'foldin g [material ] so tha t oute r surface is exposed ' sotoban 'clot h or pape r stretche d over a frame ' sotoage 'outsid e hem ' sotonon 'oute r measurement s (from oute r edge to oute r edge) ' sotopasu 'oute r calipers ' sotobet 'oute r wall' sotogaki 'oute r brush wall' sotogooshi 'oute r lattice ' sotogakoi 'oute r walls' sotojiro 'oute r par t of castle' sotowa 'oute r circle ' sotobon 'oute r moat ' sotorop 'oute r garden ' (te a gardens ) sotowaku 'oute r frame ' sotoi 'oute r boundary , periphery ' sotopoketto 'oute r pocket ' sotobutokoro 'oute r "pocket " in folds of kimono ' sotochootsugai 'oute r hinge s on bod y armor ' sotoguruma 'outsid e wheel(s)' , i.e., 'sidewheeler ' (steamboat ) sotoumi 'ope n (no t enclosed ) sea', 'dee p sea' sotoiwashi 'fish caugh t on ope n sea (off mid-Honshu) ' sotomawan 'oute r (train , beltway) loop ' sotoido 'well locate d off one' s own land ' sotokasa 'oute r penumbr a (of sun, moon) ' sotoera 'expose d gills' sotomomo 'oute r thigh ' sotogama 'insid e of legs facing/expose d outward ' sotomata 'insid e of th e leg outward ' (toe s pointe d outward , standin g 'te n to two') sotogawa 'oute r side' ('othe r party' ) sotoire 'oute r of two neste d containers ' sotokazan 'exterio r decorating ' sotogamae 'outsid e appearance ' sotogakart 'externa l appearance ' sotosen 'concav e drawknife ' (name d for its conve x cut ) sotomuki 'facin g outward ' Note Other s listed in "Uchi with Soto, Table 1. Spatial "

58

CHARLE S J QUINN,J R SOTO , TABLE 2

Social, Domesti c metta ni soto e denai ' [S/he ] almos t never goes ou t [of th e house] ' shokuji wa taitei soto de sumaseru ' [I] usually eat ou t [away from home]' . sotogirai 'is loat h to go ou t [of th e house]' , 'perso n of such a disposition ' soto ο te m suru 'be rarel y at home ' (lit , 'mak e th e outsid e one' s hous e [hold]' ) sotosuzume 'someon e who never stays at home ' (lit., 'outsid e sparrow' ) sotomawan ο yarn 'mak e business round s (outsid e home) ' sotomago 'grandchil d of a daughte r (who lives out of th e household) ' sotoburo 'takin g a bat h outsid e th e home ' SOTO , TABLE 3

Psychological/Attitudina l uchi hadaka de mo soto-nishiki 'even if nake d at home , brocad e outside ' uchi-hadakan no soto-subon 'bossy at home , mee k outside ' uchi-benkei no soto-beso 'a lion at home , a bellybutto n (= th e opposite ) outside ' uchi-hirogan no soto-suboman 'expansive/outgomg/boss y at home , closed tight when outside ' sotozura wayoi no desu ga, uchizura wa warm ' [S/he ] is affable outside ' (lit., 'out side demeano r is good') , 'bu t is difficult at home ' (lit , 'at-hom e demeano r is bad' ) soto-mizu no uchi hirogan 'naive of th e world, expansive/outgoing/boss y at home ' As in t h e list of spatia l uses in Soto Tabl e 1 ( a n d with uchi), t h e fact tha t spac e is so ofte n define d a c c o r d i n g t o socia l criteri a make s for som e overla p a n d b l u r r i n g of d o m a i n s . Sinc e t h e first several example s h e r e tak e t h e h o m e as uchi a n d use soto in contras t t o it, I hav e listed t h e m u n d e r "social/domestic, " despit e t h e fact tha t the y ar e soto, t h a t is, non-uchi expressions . (Se e Soto Tabl e 2.) I n t h e use of sotogirai 'is loat h to go o u t [of t h e h o u s e ] ' t o refer t o t h e person wh o h o l d s t h a t attitude , we hav e a c o m m o n m e t o n y m , in whic h t h e n a m e of t h e attitud e is use d t o refer t o th e p e r s o n wh o hold s t h a t attitude ; thi s ha s alread y c o m e u p in t h e c o r r e s p o n d i n g uchigirai 'is loat h t o stay a t h o m e ' . As thes e exam ple s suggest, soto t o o is use d in t h e expressio n of attitudes . (Se e Soto, Tabl e 3.) T h e s e example s describ e attitude s a n d p e r s o n a l c h a r a c t e r in t e r m s of th e domesti c a n d n o n d o m e s t i c sides of socia l behavior . We hav e alread y discusse d t h e r h e t o r i c of thei r n u a n c e s for uchi in Uchi, Tabl e 3, above . I t is interestin g t h a t every psychological/attitudina l expressio n t h a t use s th e t e r m soto also use s uchi. I n o t h e r words , t h e r e ar e apparentl y n o n o n contrastiv e uses of soto in thi s d o m a i n . I n contras t t o Soto Tabl e 3, n o n e of Soto Tabl e 4's item s save sotogawa

UCHI AND SOTO AS WINDOWS ON A WORLD

59

SOTO, TABLE 4

Social, Nondomestic sotomezurashu 'unaccustomed to going out' sotogawa 'other party' soto goyoo 'business outside, on master's orders' sotoarukt 'wandering about town' sotode 'going out' sotonagashi 'making (guard's) rounds' sotomawan 'make rounds of corporate customers (leave office)' sotodoman 'staying overnight somewhere other than home' sotomachi '««regulated, unofficial redlight district' sotodaka 'taxation on land cultivated afier tax reforms' SOTO, TABLE 5

Partitive (Bounded Subsets) sototokumi 'secondary ten incenses' (incense ceremony) sotohyakuban 'second 100 noh plays'

O t h e r party' (which is also used in t h e spatial sense of ' o u t e r side') has a c o r r e s p o n d i n g uchi expression They are all u n i q u e to soto used in t h e social d o m a i n If we contrast t h e "social/nondomestic" list shown in Soto Table 4 with t h e c o r r e s p o n d i n g list for uchi (Uchi, Table 4), we n o t e that many of t h e items that l e n g t h e n t h e uchi list, such as t h e c o m p o u n d s with -maku 'curtain' a n d -wa 'ring, circle', are missing h e r e , n o d o u b t because those expressions are employed in talking a b o u t acts d o n e exclusively for o r by p e o p l e o n t h e inside T h e reason that t h e soto-based expressions listed h e r e lack any uchi counterparts is presumably t h e same principle cited in reference to soto era 'external gills' in Soto Table 1 above In such instances, soto is used to indicate t h e off-norm, m a r k e d case (e g , sotonagashi 'outside sink', sotodoman 'staying overnight outside t h e h o m e ' , sotomachi ' u n r e g u l a t e d redlight district') Apparently n o special t e r m exists for t h e normative, u n m a r k e d c o r r e s p o n d e n t s of these terms, n o equivalent for ' i n d o o r sink', 'sleeping at h o m e ' , o r 'regulated redlight district' T h e two expressions shown in Soto Table 5 corr e s p o n d to uchitokumi ' p r i m a r y t e n incenses' a n d uchihyakuban 'primary o n e h u n d r e d n o h plays' T h e soto- a l t e r n a n t n a m e s t h e category that contains t h e items (incenses, plays) of less i m p o r t a n c e We are left with two d o m a i n s in which t h e t e r m uchi, b u t n o t soto, is attested T h e s e are uchi "temporal intervals" {Uchi, Table 5) a n d " b o u n d e d sets" (Uchi, Table 6), both of which d e p e n d crucially o n t h e quality of b o u n d e d n e s s for their m e a n i n g I have suggested that t h e uchi/soto habitus, like Lakoff a n d J o h n s o n ' s "container" schema, involves

60

CHARLES J QUINN,JR

three identifiable elements: an inside, an outside, and a boundary that separates them. In the domains in which the terms uchi and soto are used, the compound words and phrases in which they figure focus on the inside, the outside, or both, as the base concept warrants. The uchi domains of Uchi Tables 5 and 6, however, depend in their expression not only on the contents of the 'inside' being accessible and familiar, but also on the boundary being in focus. In other words, these particular meanings can be created with the word uchi not only because its contents are accessible and familiar, but because the boundary or limit is so central to its meaning, as it not with soto. For these two sets of uses, the boundary feature of the uchi/soto schema is crucial. That no such concepts are expressed with soto can be explained by the fact that an 'outside' is not bounded to the extent that an 'inside' is, and thus is not so clearly defined as an 'inside' is. To be sure, on one level of analysis, soto entails an uchi, as uchi entails a soto. And, logically speaking, both entail a boundary that separates them. But of the two, uchi depends for its image as what it is—an enclosure—on the presence of a boundary on most sides. A prototypical uchi space is entirely enclosed, bounded in every direction. Soto spaces, on the other hand, have a boundary at the point where they meet a corresponding uchi, but in all other directions, bounding is simply not specified. If it were, we would have an enclosure—an uchi. In terms of boundary, soto is simply not as well defined as uchi. Consequently, for expressing images or notions to which the notion of a boundary is crucial—such as limits, bounded intervals, or sets—the image that accompanies the term soto is less possessed of the very element crucial to indicating those limits. A boundary is definitive for uchi to an extent—literally, on most sides—that it is not for soto. Since the catalogs presented in this chapter depend on just a few dictionaries (albeit one of them unabridged), it is unlikely that uchi-based expressions at any time outnumbered those based on soto with exactly the ratio that emerges here. On the other hand, if even a dictionary survey reveals this kind of discrepancy, it seems likely that, on the whole, uchi has always been the more productive term of the two, by a considerable margin. This is most likely a consequence of the human tendency to cultivate an uchi kind of space in the world, to differentiate, categorize, name, and otherwise detail those parts of the environment in which we are closely engaged—that which is nearby, accessible, over time familiar, knowable, and therefore manipulable and controllable. It is no less than a preferred perspective with proven survival value. Insofar as our bodies are the indexical ground when we first learn to distinguish an inside from an outside, and insofar as taking a perspective that is grounded outside the self—imagining that one is standing in another uchi—is a skill that is later to develop (e.g., deixis that is not

UCHI AND SOTO AS WINDOWS ON A WORLD

61

speaker-based and anaphora), looking outward from an inside has a kind of existential and experiential primacy. For this reason, uchi is more likely as a place to look from, as a person engages and appropriates the world. Insofar as it is in this primal experience that the container schema is first learned and from which it is then projected onto relations in the external world, we identify with an uchi perspective and tend to regard soto as "out there," observable but less familiar, detached from ourselves and what we know best, and less under our control. As a place to stand, uchi is the stable, familiar norm; out in soto, the footing is less sure. The final lists in this catalog of expressions (shown in Uchi with Soto Tables 1-4) summarize cases in which uchi and soto are used together contrastively, with a shared third term. In most cases, the contrastive frame is lexical, as when either uchir or soto- is the variable first morpheme in compound words in which the second element is invariant, for example, uchimawari 'inner (beltway, train) loop' versus sotomawari 'outer (beltway, train) loop', or uchipasu 'inside calipers' versus sotopasu 'outside calipers'. In other instances, they occur as opposed concepts in a single phrase or clause, as in the proverbs that describe people in terms of how they behave 'inside' and 'outside'. As might be expected, many items already surveyed occur in such pairs. Nor, of course, do we find contrastive uchi/soto pairs used to express "temporal intervals" or "bounded sets." UCHI WITH SOTO: A CATALOG OF CONTRASTIVE PAIRS

As observed earlier, the primary significance of either uchi (or soto)15 as the initial term in a compound is to indicate position or shape of the following head. In the case of position, the head in the compound is 'inside' or 'outside' with reference to its opposite. For example, uchibori 'inner moat' is 'inside' the sotobori 'outer moat', and closer to the castle that a hori 'moat' surrounds. In the case of shape as well, the uchi- alternative specifies that its referent (the second element in the compound) faces or is shaped 'inward', in contrast to the 'outward' facing or shaping of the soto- alternative, as in uchikatagiri 'concave cut' and sotokatagiri 'convex cut'. As might be expected, pairs that contrast position outnumber by far those that contrast shape. But compounds that name things outnumber those that name pure positions or orientations, like uchi-/ soto-muki 'facing inward/outward'. (See Uchi with Soto, Table 1.) Why should there be so few contrastive uchi/soto pairs in the social, domestic domain shown in Uchi with Soto, Table 2? There are undoubtedly a few others, but probably few things, people, and behaviors are regarded as equally common to both 'inside' and 'outside' the household.

62

CHARLES J . QUINN, JR. U C H I WITH S O T O , T A B L E 1

Spatial

uchisoto 'inside and outside' doa wa uchisoto ni aku 'the door opens inward and outward' uchi-/soto-zura 'interior/exterior surface', 'at-home visage/outside visage' uchi-/soto-gooshi 'inside/outside lattice' uchi-/soto-mawari 'inner/outer (beltway, train) loop' uchi-/soto-niwa 'inner/outer garden' uchi-/soto-bori 'inner/outer moat' uchi-/soto-gawa 'inner/outer sides', 'our side/their side' uchi-/soto-gake 'inner/outer leg sumoo throw' ( T h e r e are many m o r e such terms, and some are used in juudoo as well.) uchi-/soto-muki 'facing inward/outward' uchi-/soto-shirasu 'white pebbles inside and outside apron skirting the n o h stage' uchi-/soto-chootsugai ' i n n e r and outer hinges (on body a r m o r ) ' uchi-/soto-buro 'indoor/outdoor bath' uchi-/soto-wa ' i n n e r / o u t e r ring, circle' itchi-/soto-m,omo 'inner/outer thigh' uchi-/soto-sanzun 'inner/outer part o f ankle area' uchi-/soto-butokoro 'inner/outer "pocket" in folds o f k i m o n o ' uchi-/soto-sumi 'inner/outer corners' (seating positions at tea ceremony) uchi-/soto-setsuin 'inner/outer toilet' (for a tea r o o m ) uchi-/soto-roji 'inner/outer garden' (for a tea r o o m ) uchi-/soto-bari 'paper or cloth stretched inside/over a frame' uchi-/soto-be 'inside/outside', 'this side/that side' uchi-/soto-misu 'inner/outer boxes at kabuki theater' uchi-/soto-yama 'settled mountain villages/unsettled mountains' uchi-/soto-miyoshi 'inner/outer timbers o f bow o f boat' uchi-/soto-doi ' i n n e r ( c o n c e a l e d ) / o u t e r (visible) r o o f gutter' uchi-/soto-waku 'inside/outside frame' uchi-/soto-katagiri 'cut with graver facing inside/outside o f cut' uchi-/soto-sogi 'cutting inward/outward (tips o f protruding shrine rafters)', 'shrine rafters cut inward/outward on their tips' uchi-/soto-sen 'convex/concave drawknife' uchi-/soto-nori 'measurements to inner/outer edges' uchi-/soto^pasu 'inside/outside calipers' uchi-/soto-mata 'inside o f legs facing inward/exposed outward' uchi-/soto-gama 'inside o f legs facing inward/exposed outward' uchi-/soto-wa 'inside o f legs facing inward/exposed outward' uchi-/soto-hachimonji 'inside o f legs facing inward/exposed outward'

UCHI AND SOTO AS WINDOWS ON A WORLD

63

UCHI WITH SOTO, TABLE 2

Social, Domestic

uchi-/soto-mago 'grandchild = child of household successor/grandchild = child of child who "married out"' uchi /soto-truro 'at home/public bathing' UCHI WITH SOTO, TABLE 3

Social, Nondomestic

ucht-/soto no saburai 'warrior allies from our own ranks and without' uchi /soto-gawa 'our side/their side' uchi-/soto-muki 'for inside/outside consumption' (information) uchi-oton no to-medeta 'inferior in substance, impressive in appearance' UCHI WITH SOTO, TABLE 4

Partitive

ucht /soto-ben 'amount of grain lost in hulling grain/ratio of loss to remainder' uchi-/soto-tokumi 'primary/secondary ten incenses' (incense ceremony) uchi-/soto-hyakuban 'primary/secondary 100 noh plays'

For ucht X contrasted with soto X in t h e psychological a n d attitudinal d o m a i n , please consult Soto, Table 3 (p 58 above) Ucht with Soto, Table 3 also repeats expressions a n d uses already i n t r o d u c e d Finally, there a r e a few uchi/soto contrasts in t h e partitive (See Ucht with Soto, Table 4 ) CONCLUSIONS

If we review t h e concepts that a r e specified, described, o r otherwise modified with t h e terms ucht a n d soto in t h e p r e c e d i n g t h r e e taxonomies of ucht expressions, soto expressions, a n d ucht/soto expressions, certain p a t t e r n s e m e r g e J u d g i n g from t h e concepts ucht participates in expressing, this word has b e e n associated with notions such as FULLY BOUNDED, INDOORS, NEARBY, ENCLOSED, CONCAVE, DARK, DOMESTIC, FAMILY, INTRALINEAL, "us", CASUAL, COMFORTABLE, INFORMAL, FAMILIAR, PRIVATE, INDULGENT, FREE, CONCEALED, SECRET, PRIVILEGED, DETAILED, KNOWN, SHARED, MUTUAL BENEFIT, LOCAL, LIMITED, CONTROLLED, SPECIFIABLE, ENUMERABLE, PART OF LARGER WHOLE, SACRED, SPECIAL, a n d PRIMARY A s u m m a r y review of soto finds it used in t h e c o m p a n y of such c o n c e p t s as O U T S I D E AN ENCLOSURE, O U T D O O R S , O P E N , P R O T R U D -

ING, CONVEX, EXTRALINEAL, NONDOMESTIC, "THEM," SECULAR, PRO-

64

CHARLES J QUINN JR SUMMARY T A B L E

Lexical Uchi vs Soto, Regular C o n c e p t u a l Contrasts UCHI SOTO = INDOORS OUTDOORS CLOSED OPEN EXPERIENCED OBSERVED HIDDEN REVEALED FULLY BOUNDED PARTLY BOUNDED CLEARLY DEFINED LESS CLEARLY DEFINED LIMITED LIMIT-IRRELEVANT SACRED SECULAR (Cf imperial palace vs o u t e r buildings, o u t e r shrine vs i n n e r shrine at Ise, etc ) SELF(-VES)

OTHER(S)

LINEAL FAMILY FAMILIAR "us"

EXTRALINEAL FAMILY

UNFAMILIAR

"THEM"

PRIVATE INCLUDED KNOWN INFORMED

PUBLIC EXCLUDED UNKNOWN UNINFORMED

CONTROLI ED ENGAGED

UNCONTROLLED

DETACHED

EARLY/PRIMARY

LATE/SECONDARY

FANE, REMOVED, EXPOSED, VISIBLE, CUSTOMERS, WELL-BEHAVED, RESTRAINED, MAINTAINING APPEARANCES, "ON THE TOWN", PERIPHERAL, FOREIGN, LESS KNOWN, LESS DETAILED, UNDIFFERENTIATED, UNCONTROLLED, a n d S E C O N D A R Y T h e s e a s s o c i a t i o n s s h o w r e g u l a r c o n t r a s t , a n d c a n b e a r r a y e d as a s i n g l e p r o p o r t i o n t h a t h o l d s a c r o s s a v a r i e t y o f d o m a i n s ( S e e S u m m a r y T a b l e Uchi vs Soto) Listing such c o m p o u n d s , collocations, a n d g r a m m a t i c a l construct i o n s as e x a m i n e d a b o v e h a s its t e d i o u s s i d e , b u t d o i n g so r e v e a l s affinities a n d o p p o s i t i o n s t h a t m i g h t o t h e r w i s e g o u n r e m a r k e d T h e s e a r e t h e s y m b o l i c v a l u e s w e find i n s i d e a n d o u t s i d e t h e b o u n d a r y i n t h e uchi/soto s c h e m a , as r e v e a l e d b y r e p r e s e n t a t i v e u s a g e o f t h e s e t w o c o m p l e m e n tary w o r d s , r e c o r d e d in J a p a n across a variety of lived d o m a i n s since t h e eighth century I t is t h e affinities c o m m o n t o t h e s e v a l u e s t o w h i c h w e t u r n i n c o n c l u s i o n W h a t h a s h a p p e n e d o v e r t h e c e n t u r i e s is t h a t t h e w o r d s i n d i c a tive o f d o m a i n , w h i c h h a v e c o m b i n e d s y n t a g m a t i c a l l y w i t h t h e w o r d uchi o r soto, h a v e t h e r e b y c o m e t o associate paradigmatically with one another I n t h e c o m p a n y t h e y all k e e p w i t h uchi, f o r e x a m p l e , s u c h c o n c e p t s as NEARBY, ENCLOSED, DOMESTIC, "US," CASUAL, FAMILIAR, KNOWN, DE-

UCHI AND SOTO AS WINDOW S ON A WORLD

65

TAILED , SHARED , LOCAL, CONTROLLED , SACRED , an d PRIMAR Y (as listed just above) themselve s constitut e a natural , evolved categor y Thos e element s tha t are used in syntagmati c combinatio n with th e ter m soto, too , cluste r togethe r in paradigmati c relation , with OUTDOO R spaces associate d With OPEN , UNBOUNDED , REVEALED , an d PUBLIC , as well as UNFAMIL IAR, LESS DETAILED , UNINFORMED , UNSHARED , UNCONTROLLED , SECU LAR, EXCLUDED , an d SECONDAR Y Bot h thes e natura l categorie s have emerged , as othe r social structure s do , in a coevolutio n of cultura l tradi tio n an d social praxis Thei r stability is maintaine d where received custom is applie d in meetin g th e need s of new context s Som e of th e item s in th e precedin g catalo g are n o longe r in curren t use, others , yet to be invented , will n o doub t appea r But any new extensio n of th e meaning s of uchi an d soto ha s th e foregoin g association s to fit in with or to conten d against , projection s of eithe r concep t int o new domain s will be facilitate d or hindere d by thi s historicall y grown ne t of relations , which ha s bee n woven, rended , patched , an d rewoven, in lived communitie s Ne w metaphors , metonymies , an d synecdo che s based on uchi or soto will emerg e from somewher e in thi s habitu s and , like th e presen t members , will be create d in abductio n an d anal ogy Some , like thos e of th e domain s "social," "tempora l intervals, " "bounde d sets," an d "partitive, " will be mor e abstract—whic h is to say versatile acros s mor e domains—tha n other s Still mor e remove d from th e spatia l domai n are thos e grammatica l meaning s tha t are structure d in term s of th e uchi/soto schem a Th e Japanes e habitu s of uchi/soto constitute s for Japanes e cultur e what Gregor y Bateso n (1979, 8) ha s called a "patter n which connects, " a patter n of pattern s An uchi/soto metapatter n insinuate s itself when a variety of differen t aspect s of communicativ e behavio r appea r to patter n homologousl y I n biology, of course , homolog y is a sort of anatomica l analogy, in which th e relation s between th e part s of on e organis m resemble th e relation s between correspondin g part s of anothe r organism , "such resemblanc e is considere d to be evidenc e of evolutionar y related ness" (Bateso n 1979, 250) Fo r example , we may speak of a fish's "head " based on th e homolog y we observe between th e relative position s of its eyes an d mout h an d thos e of ou r own eyes an d mouth s Th e fish's eyes, mouth , an d hea d resembl e ou r own less in phenomena l detai l tha n in th e forma l relation s tha t obtai n amon g the m Similarly, in language , we may say tha t th e prepositio n in of "in schoo l thi s year, " "in th e kitchen, " an d "in trouble " is th e same in, despit e thes e difference s in semanti c domai n Th e region expressed betwee n th e locatio n an d th e located , we feel, is someho w th e same , analogou s Likewise with "accusative " (actually a kind of scope ) marke r ο in Hon ο kaesu 'retur n a book' , Hashi ο wataru 'cross a bridge' , an d Ima ο ikiru 'Live th e present ' Semantically ,

66

CHARLES J QUINN.JR

the scope in each clause differs from the others, as undergoer, path, and range, respectively. Users of the language have nevertheless come to categorize each with the same o. In the case of uchi and soto, too, it is the formal relation that is of definitive importance, so that whether the domain is that of the human body, the physical and spatial world, social relations, or an exchange of information, the relations peculiar to each are related to one another analogically. If this is true of a wide variety of domains—such as social interaction and lexico-grammatical structure— the uchi/soto schema qualifies as a pattern that connects other patterns, a metapattern. In its lived quality, we refer to it as habitus. I should finally like to point out that values associated paradigmatically with the words uchi and soto are also regularly expressed in a number of pairs of words that are unrelated to the terms uchi or soto in any etymological sense. The internal structure of these pairs is analogically uchi/soto, for they too assume an inside, an outside, and a dividing boundary and, what is more, associate the same sorts of physical, social, and psychological values just reviewed with their own insides and outsides. Pairs of semantically opposed nouns such as omote 'surface, front' and ura 'inside, back side' are one type of example, which Bachnik and Molasky touch on in this volume. But the phenomenon is hardly limited to nouns, as is evident, for example, in the many expressions based on intransitive verbs hairu 'enter' and deru 'emerge'. A few examples of how hairu and deru are used will suggest how congruent their own opposed uses are with those already observed of the words uchi and soto. Omote-zata ni deru, literally 'get out into front-side treatment', refers to inside or private information becoming publicly known. News, too, will 'emerge' into a public place where it is accessible to all (nyuusu ni deru 'appear in the news', shinbun ni deru 'appear in the newspaper'). To attend a meeting or seminar, or to enter a competition—all public venues—is to deru 'go out' into it. The warning expressed in nanigoto mo uchiwa ni yaru beki 'Better keep within bounds in all things' has a sotofocused counterpart in the well-known proverb deru kui/kugi wa utareru 'the peg/nail that protrudes gets pounded'. The moral is that one should stay unobtrusively inside with everyone else. The proverb is all the more coherent and resonant for its links with the many lived, social contexts that are shaped and known in an uchi/soto kind of way. Conversely, the verb hairu 'go in, enter' is used in expressing many concepts of an uchi sort: entering, belonging, fitting in, and so on. It is employed in nonvolitional contexts to express being included in a set, as we saw in daihyoosaku no uchi ni hairu 'is among [her] representative writings', and also in a volitional sense, as in furo ni hairu 'take a bath' or kurabu ni hairu 'join a club'.

UCHI AN D SOTO AS WINDOW S O N A WORL D

67

Th e same sort of 'inside/outside ' oppositio n can be foun d in th e transitiv e verbs dasu 'put/tak e out ' an d ireru 'pu t in, insert' , which are etymologicall y relate d to intransitiv e deru an d hairu, respectively. Dasu is used to express such meaning s as 'expose ' (hiza ο dasu 'expose one' s knees' , sita ο dasu 'stick ou t one' s tongue') , 'sen d out ' (tegami ο dasu 'mai l a letter') , an d 'pu t out ' (hon ο dasu 'pu t ou t a book' , shmbun ni dasu 'publish [it] in th e newspaper') . Th e commo n featur e her e is put tin g somethin g out int o an are a beyon d one' s body, control , knowledge , uchv. int o publi c view, int o custod y of th e posta l service, an d so on . Th e same verb is also used routinel y in th e sense of 'handin g in ' or 'submit ting ' thing s (e.g., shukudai ο dasu 'han d in one' s homework') , of 'exhibiting' a piec e of art at a show, of 'serving' a certai n dish (e.g., at a restau rant) , or 'producing ' casualitie s (e.g., an earthquake) . Th e similarit y of such concept s to thos e expressed with intransitiv e deru are apparent . And as with deru, thes e few sampl e context s just scratc h th e surface. If dasu remove s thing s to an 'outside' , ireru 'pu t in, insert ' is for gettin g the m 'inside' . Ireru can be used to denot e 'inserting ' one' s han d int o one' s pocket , 'putting ' crea m int o coffee, 'placing ' someon e int o a particula r grou p (as when eithe r 'admitting ' or 'placing ' a child int o a school) , or to express 'accepting ' or 'accedin g to ' a view tha t differs from one' s own (literall y 'lettin g it in') . Ireru participate s in th e expression of inclusion , too , as in volitiona l nakama ni ireru 'let/includ e [some one ] int o one' s circle ' or somewha t differentl y in nonvolitiona l watashi ο irete, go-ηιη kazoku desu 'Includin g myself, it's a family of five'. Ireru is also used when 'paying ' interes t 'into ' an account , an d when 'casting ' a vote 'in ' favor of a candidate . Ireru differs from dasu as hairu differs from deru. Etymologicall y relate d ireru an d hairu refer to event s tha t take an inside as thei r goal, while dasu an d deru denot e departur e from an inside for a goal tha t is outside . Like th e term s uchi an d soto, th e transitiv e member s of thes e two pair s of verbs also participat e in a variety of com pound s an d collocation s tha t play majo r roles in th e languag e of everyday life. As a productiv e suffix, for example , -dasu indicate s inceptiv e aspect , as in shaben-dasu 'star t to talk', an d -ireru, directionality , as in kaki-ireru 'enter , write in, fill out ' an d kai-ireru 'purchase , lay in' . Thes e an d man y mor e expression s like the m bear a family resem blanc e to th e uses of th e term s uchi an d soto surveyed above, an d also to certai n grammatica l distinctions , which are describe d in chapte r 11, th e final chapte r of thi s volume . The y projec t th e same kin d of dua l perspective, includin g th e values associate d with eac h side, ont o a variety of otherwis e unrelate d domains , which rang e from spatial , physical, an d social to abstrac t an d grammatical . Uchi/soto is, I think , as good a nam e as any for thi s family of relate d concepts , which Japanes e peopl e main -

68

CHARLE S J QUINN.J R

tai n by daily use , in thei r socia l lives, particularl y in thei r languag e As we hav e seen in thi s brie f t o u r of uses of t h e t e r m s ucht a n d soto, meta p h o r a n d o t h e r trope s hav e sprea d thi s oppositio n far a n d wide t h r o u g h th e language , fro m c o m m o n n o u n s use d in prototypicall y " n o u n y " ways, t o a n u m b e r o f very basic grammatica l uses, suc h as kyoo no ucht ni 'withi n today ' a n d ikanai ucht ni 'befor e going ' I n t h e day-in , day-ou t activit y of c o m m u n i c a t i o n , whic h is t o say, of p o i n t i n g out , referring , n a m i n g , a n d r e n a m i n g suc h aspect s of o u r exp e r i e n c e as requir e it, p e o p l e ar e constantl y reestablishing , e x t e n d i n g , o r allowin g t o atroph y t h e categorie s with whic h the y pres s t h e flux of p h e n o m e n a l e x p e r i e n c e i n t o c o h e r e n t part s a n d whole s T h e lexico n of a languag e ca n t h u s b e u n d e r s t o o d as t h e distillat e of a culturall y a n d socially n e g o t i a t e d proces s of categorizatio n I t is n o t surprisin g t h a t t h e o n e n t a t i o n a l habitu s of uchi/soto shoul d b e widesprea d in t h e vocabu lar y with whic h J a p a n e s e societ y sustain s itself As I shal l a r g u e in t h e conclusio n t o thi s volume , t h e sam e uchi/soto habitu s also inform s distinction s draw n by t h e m o r e abstrac t structure s of J a p a n e s e g r a m m a r 1 6 A n d thes e lexical a n d grammatica l p h e n o m e n a themselves , of course , i n d e x t h e dynamic , evanescen t d a n c e of socia l life itself

NOTE S

1 Thi s researc h was conducte d in 1990-9 1 unde r suppor t from a Seed Gran t from Th e Ohi o Stat e University , a Fulbngh t Gran t administere d throug h th e Japan- U S Educationa l Commission , an d a Researc h Gran t from th e Socia l Scienc e Researc h Council , for which I am very grateful Thank s to co-edito r Jan e Bachni k an d to my wife Shelley for thei r helpfu l criticism s an d suggestions Th e words of th e epigrap h (Fuhu wa ucht' Oni wa soto1) are shoute d as bean s are tossed ou t of th e house , in an exorcism ritua l tha t is performe d every year on Februar y 3 Fo r a key to abbreviation s an d orthographi c convention s followed in thi s an d th e othe r linguisticall y oriente d chapter s (Wetzel's, Sukle's, an d Quinn' s closin g contribution) , see th e "Key to Abbreviation s an d Orthographi c Conventions, " ρ xvn 2 Bourdie u ha s said tha t h e develope d th e notio n of habitus in orde r to refer to social life as incorporate d in individual s and , at th e same time , avoid "the scientificall y quit e absur d opposiUo n between individua l an d society" (1990, 31) It is "a system of lasting, transposabl e dispositions , a matri x of percep tions , appreciations , an d actions " (1977, 82-83 ) William Hanks' s (1987, 677) characterizatio n of habitu s suggests its relevanc e for ou r projec t Habitu s ha s th e potentia l to generat e homologou s formation s acros s differen t cultura l fields, relating , for example , th e calendar s of cooking , daily tasks, time s of day, an d th e same set of symbolic relation s an d mode s of practic e (rhyth m of execution , sequence , duration , an d so forth) , with respect to which the y are schematicall y equivalen t Bourdie u represent s thes e

UCHI AND SOTO AS WINDOWS ON A WORLD

69

equivalences as variant realizations of a single type of structure, summarized in his synoptic diagrams (Bourdieu 1977, figures 3-9). Being grounded in cultural schemata that recur across distinct fields of action, habitus is logically prior to any actual event of practice. At the same time, it is subject to innovation and strategic manipulation in practice, such that it is a product as well as a resource that changes over time (emphasis added). A similar idea is explored in contexts chemical and biological, as well as social, in Sheldrake 1988. 3. Collocations are phrases and clauses in which a word typically combines with a limited set of other words. Knowing the parts of a collocation is enough to understand it (e.g., "to be worth the trouble, to be worth the bother"), while this is not true of an idiom ("to be worth while"). Cf. Bolinger 1975, 102. 4. Limiting the survey that follows to the terms uchi and soto restricts our study of the lexical expression of the concepts uchi/soto to but one segment of a larger part of the vocabulary. Excluded from consideration, for example, will be SinoJapanese words that express the same concepts by including as one of their formants the character P^ or 9[ , and reading it not as "uchi" or "soto" but as "nai or "gai." Since, unlike the terms uchi and soto, nai and gai are not usually used as independent words, their role in the lexicon is more specialized. 5. In the sense in which George Lakoff and Mark Johnson (1980), among others, use the term, which is to say broadly, in no way restricted to mere rhetorical flourish. From their perspective, the absolute opposition of "literal" meaning to "metaphorical" meaning is the product of too restricted a vision. Certain very common expressions are based on our bodily experience, as we live it, for example, the many based on the metaphors "Up is good," "Down is not," and are thus existentially more basic than others which, while still metaphorical, are perspectivally independent of our bodies, for example, "A project is a building," with entailed foundation, exterior, interior, stories, and so on. 6. See, for example, Bates, Camaioni, and Volterra 1979; Givon 1973, 1979; Lakoff and Johnson 1980; Lakoff 1987; Sweetser 1990; Traugott 1982; Traugott 1989; and Volterra and Antinucci 1979, among others. 7. In such a survey, it is necessary to rely on dictionaries, despite their shortcomings. Since even the newest and largest dictionary tells us nothing about frequency of use, and grows quickly out of date, what it reveals of how a word fits into the communicative ecology of its community will be fragmented and partial, as Aitchison notes (1987, 14). Nevertheless, provided one's aims are less ambitious than accounting for how people learn, know, and use words (Aitchison's "mental lexicon"), dictionaries can help. A survey will do nicely enough if the aim is to outline such records as exist of a word's uses. 8. These are named for the shape of the cut they make: a concave scooping cut for the uchi-sen, and a convex or rounding-off cut for the soto-sen. 9. Earlier attestations were examined in: Manyooshuu (759), Kokinshuu (907), Utsuho monogatan (mid-Heian), Taketon monogatan (early Heian), Toodaiji fuushoo monkoo (early Heian), Sanboo e-kotoba (984), Genji monogatan (ca. 1010), Ookagami (late Heian), Rin'yooshuu (1178), and Heike monogatan (a text oral in origin, probably written down in some form between 1219 and 1242).

70

CHARLE S J. QUINN,J R

10. See Bachnik' s introductio n to thi s volum e for a mor e detaile d discussion of C. S. Peirce' s semiotic . In Peirce' s conception , a symbol is a sign tha t stand s in a relatio n to its referen t tha t is less obvious tha n it is with th e othe r sign types, icons, an d indexes. Symbolic signs nee d no t be grounde d in th e situatio n in which the y are used, in contras t to indexica l signs (e.g., th e speake r as groun d for "this, " "that, " or he r presen t poin t in tim e as groun d for "now," "then, " etc.) . No r doe s a symbolic sign mimi c in its form perceive d feature s of th e concep t it represents , in contras t to th e icon (e.g., th e vowel lengt h in "loooong " is an iconi c sign; th e word "long" is a symbol) . Th e symbol's abstractio n away from th e particularitie s of occasion-of-us e an d th e signified gives symbolic representa tion its great flexibility, which "can tur n realit y over on its beam end s beyon d what is possible throug h action s or images" (Brune r 1966, 11). 11. Thes e term s inde x mor e tha n grammatica l person , specifically som e aspec t of th e speaker' s social (uchi/soto) relatio n to th e referent . Th e indexica l use was originall y motivate d by th e term' s symbolic value. Othe r term s of persona l referenc e are no t onl y indexica l in origin , bu t extension s from simpler , mor e purel y spatia l deictics , such as anata 'you' (formerl y 'directio n away from speaker an d listener') , omae 'you' (formerl y 'honored-in-front') , or kare 'he ' (formerl y 'th e on e [we know/remove d from speake r an d listener]' ) an d kanojo 'she ' (formerl y 'tha t female [we know/remove d from speake r an d listener]') . And while anata an d omae take th e speake r as groun d zero , the y express an uchi/soto indexing . Omae, onc e a distancin g polit e term , ha s been reinterprete d ironicall y to express intimac y or insult , dependin g on whethe r th e perso n so addresse d is in-grou p or out-grou p to th e speaker . Anata 'you' , earlie r a distal directiona l pronou n ('tha t directio n away from us', equivalen t to present-da y achira) was reinterprete d as an indirec t mean s of addres s an d reference . 12. Rosenberge r (thi s volume ) cites scholarshi p tha t identifie s th e "poles of meanin g for self an d social life" amon g Samoans , Javanese , an d th e Ilongot s (Philippines ) with "context s of bound , concentrate d energ y tha t is outer-ori ente d an d context s of unbound , less controlle d energ y tha t is inner-oriented. " 13. Abduction: in Charle s Peirce' s sense of th e term , th e proces s of graspin g ne w phenomeno n Β by assumin g it can be understoo d in a way similar to tha t whereb y we understan d known phenomeno n A. 14. O n th e significanc e of thi s concep t for morphology , see Bybee 1985. 15. A slash between a hyphenate d uchi- an d soto- indicate s tha t eac h occur s as th e first elemen t in a compoun d word. Fo r example , uchi-/soto-zura mean s ther e exist two semanticall y contrastiv e compounds , uchi-zura an d soto-zura. When n o slash separate s them—fo r example , uchisoto—they are bot h par t of th e same word. 16. Fo r this latte r aspec t of uchi/soto structuring , see chapte r 11, th e final c h a p ter of thi s volume . REFERENCE S

Aitchison , Jean . 1987. Words in the Mind: An Introduction to the Mental Lexicon. Ne w York: Basil Blackwell. Aristotle. 1984. "Poetics, " in The Rhetoric and Poetics of Aristotle. Translate d by In gram Bywater. Ne w York: Th e Moder n Library.

UCHI AND SOTO AS WINDOW S ON A WORLD

71

Bates, Elizabeth , Luigia Camaioni , an d Virginia Volterra 1979 'Th e Acquisitio n of Performative s Prio r to Speech, " in Ε Och s an d Β Schieffehn , eds , Developmental Pragmatics, 111-30 Ne w York Academi c Press Bateson , Gregor y 1979 Mind and Nature a Necessary Unity Ne w York Banta m Books Bohnger , Dwight 1975 Aspects of Language, 2d ed Ne w York Harcour t Brac e Jovanovic h Bourdieu , Pierr e 1977 Outline of a Theory of Practice Ne w York Cambridg e Uni versity Press 1990 In Other Words Stanfor d Stanfor d Universit y Press Bruner , Jerom e S 1966 The Relevance of Education Cambridge , Mass Harvar d Universit y Pres s Bybee,Joa n 1985 Morphology Philadelphi a Joh n Benjamin s Cooper , William Ε , an d Joh n R Ross 1975 "World Order, " in Proceedings of the Annual Meeting of the Chicago Linguistic Society, 63—103 Chicag o Chicag o Linguistic Societ y Givon , Talm y 1973 "Th e Time-axi s Phenomeno n " Language 49 890-92 5 1979 On Understanding Grammar Ne w York Academi c Pres s Hanks , William F 1987 "Discours e Genre s in a Theor y of Practice, " American Ethnologist 14 (4) 668-9 2 Johnson , Mar k 1987 The Body in the Mind Chicag o Universit y of Chicag o Pres s Lakoff, Georg e 1987 Women, Fire and Dangerous Things What Categories Reveal about the Mind Chicag o Universit y of Chicag o Press Lakoff, George , an d Mar k Johnso n 1980 Metaphors We Live By Chicag o Univer sity of Chicag o Pres s Nakada , Nono , Wada Toshimasa , an d Kitahar a Yasuo, eds 1983 Kogo dayiten (Majo r dictionar y of earlie r Japanese ) Tokyo Shoogakka n Rumelhart , David Ε , Jame s L McClelland , an d th e PD P Researc h Grou p 1986 Parallel Distributed Processing Explorations in the Mtcrostructure of Cognition, 2 vols Cambridge , Mass Bradfor d Books/MI T Press Sheldrake , Ruper t 1988 The Presence of the Past Morphic Resonance and the Habits of Nature Ne w York Vintage Books Shoogak u Tosho , ed 1981 Kokugo Dayiten (Majo r dictionar y of th e Japanes e language ) Tokyo Shoogakka n Sweetser, Eve 1990 From Etymology to Pragmatics Metaphorical and Cultural Aspects of Semantic Structure Ne w York Cambridg e Universit y Pres s Traugott , Elizabet h Closs 1982 "Fro m prepositiona l to textua l an d expressive meanings , som e semantic-pragmati c aspect s of grammaticahzauon, " in Per spectives on Historical Linguistics, Winfred Ρ Lehman n an d Yakov Malkiel , eds , 245-7 1 Philadelphi a Joh n Benjamin s 1989 "On th e rise of epistemi c meaning s in English An exampl e of subjectificatio n in semanti c change, " Language 65 31-5 5 Volterra, Virginia, an d Francesc o Antinucc i 1979 "Negado n in Chil d Lan guage a Pragmati c Study, " in Ε Och s an d Β Schieffehn , eds Developmental Pragmatics, 281-30 4 Ne w York Academi c Pres s Wittgenstein , Ludwig 1958 Philosophical Investigations Translate d by G Ε Μ Anscomb e Ne w York Macmilla n

72

CHARLE S J QUIN N JR RELATE D READING S

Bennet , David C 1975 Spatial and Temporal Uses ofEnglish Prepositions An Essay in Stratificational Semantics Londo n Longman s Bruner , Jerom e S 1990 Acts of Meaning Cambridge , Mass Harvar d Universit y Pres s Givon , Talm y 1984 Syntax a Functional-Typological Introduction, Vol 1 Philadel phi a Joh n Benjamin s Greenberg , Josep h Η 1978 Universah of Human Language, Vol 4 Syntax Stan ford Stanfor d Universit y Pres s Grimm , Η 1975 "On th e child' s acquisitio n of semanti c structur e underlyin g th e wordfield of preposition s " Language and Speech 18 (1975) 97-11 9 Labov, William 1973 "Th e Boundarie s of Words an d Thei r Meanings, " in C -J Ν Bailey an d R W Shuy, eds , New Ways of Analyzing Variation in English Washington , D C Georgetow n Universit y Pres s Miller , Georg e A , an d Phili p Ν Johnson-Lair d 1976 Language and Perception Cambridge , Mass Th e Belkna p Pres s of Harvar d Universit y Pres s Ochs , Eleanor , an d Bamb i Schieffelin , eds 1979 Developmental Pragmatics Ne w York Academi c Press Rosch , Eleano r 1975 "Cognitive Representation s of Semanti c Categorie s " Journal of Experimental Psychology General 104 192-23 3 Rosch , Eleanor , an d Β Β Lloyd, eds 1978 Cognition and Categorization Hillsdale , Ν J Lawrenc e Erlbau m Talmy, Leonar d 1978 "Figure an d Groun d in Comple x Sentences, " in J Η Greenberg , ed , Universah of Human Language, Vol 4 Syntax Stanfor d Stan ford Universit y Press 1985 "Lexicahzatio n Pattern s Semanti c Structur e in Lexical Forms, " in Timoth y Shopen , ed , Language Typology and Syntactic Description, Vol 3 Grammatical Categoriesand the Lexicon Cambridg e Cambridg e Universit y Pres s Traugott , Elizabet h Closs 1978 "On th e Expressio n of Spatio-Tempora l Relation s m Language, " in J S Greenberg , ed , Universah of Human Language, Vol 3 Word Structure, 369-40 0 Stanfor d Stanfor d Universit y Press 1985 "Conditiona l Markers, " in J Haiman , ed , Iconicity in Syntax, 289310 Philadelphi a Joh n Benjamin s Tversky, Barbara , an d K. Hemenwa y 1984 "Objects, Parts , an d Categorie s "Jour nal of Experimental Psychology General US 169-9 3

Chapter 3 A MOVABLE SELF: T H E L I N G U I S T I C I N D E X I N G O F UCHI A ND P A T R I C I A J.

SOTO WETZE L

EDITORS ' INTRODUCTIO N

Patrici a Wetzel's semina l pape r challenge s th e assumptio n tha t th e Τ of Indo-Europea n language s is th e universa l deicti c groun d for all discourse . She argue s tha t in Japanes e th e deicti c ancho r poin t or groun d is no t an individua l ego, bu t rathe r a collectivel y define d vantage poin t represente d as uchi. Wetzel focuses on th e implication s of considerin g uchi as a deicti c ancho r point , in allowing for th e developmen t of a mor e coheren t accoun t of th e organizatio n of "person " in Japanese . Wetzel's reformulation s have importan t consequences : she shows th e Indo-Europea n paradig m of first, second , an d thir d person s to be ill suited to describin g perso n in Japanese . Instead , perso n can be consid ere d to be communicate d mor e broadl y tha n by perso n terms . Uchi/soto 'in-group/out-group ' distinction s are muc h mor e crucia l tha n pro noun s for delineatin g perso n in Japanese . Moreover , man y 'in-group / out-group ' distinction s are signaled no t throug h nomina l uchi/soto referenc e bu t in a variety of othe r ways, for example , throug h verbs of giving an d receiving, an d polit e forms. Thu s it is significant tha t referenc e to 'in-group/out-group ' by politenes s markin g on th e verb outnumber s tha t of pronoun s by a rati o of ten to one . Wetzel's reformulation s ope n up a numbe r of importan t new avenue s for investigation . D o th e collectivel y define d deicti c ancho r point , an d th e designatio n of perso n via 'inside/outside ' distinctions , have implica tion s for th e organizatio n of "self"? Th e designatio n of uchi an d soto— who is 'in ' an d who is 'out'—depend s on th e constan t calculation s of participant s in particula r situations . Shifts in thei r calculation s creat e shifts in grou p boundaries , which are fluid, rathe r tha n static , an d there fore differ markedl y from Nakane' s portraya l of rigid boundaries . But what ramification s doe s group-base d deixis have for th e social organization of Japanes e groups? Doe s th e proces s of fluid, constantl y shiftin g grou p boundarie s characteriz e social as well as linguistic aspect s of Japa nese grou p organization ? Thes e question s are pursue d by all th e paper s tha t follow.

74

PATRICIA J. WETZEL

A MOVABLE SELF: THE LINGUISTIC INDEXING OF UCHIANO SOTO UCHI AND SOTO

It has become virtually impossible to speak of Japanese social behavior without reference to (or at least recognition of) the importance of uchi/ soto boundaries. 1 Benedict's 1946 interpretation of the importance of personal ties within Japanese culture initiated Western recognition of Japanese uchi/soto 'in-group/out-group' relationships, and this recognition has been expanded and refined by anthropologists (Nakane 1970; Lebra 1976), psychologists (DeVos 1985; Roland 1988), political scientists (Pye 1985), and others, drawing together much of Japanese behavior that is otherwise inexplicable (at least from a Western perspective). Feelings of group belongingness are nurtured throughout one's lifetime in Japan (DeVos 1985, 165), and are associated with other positive feelings such as amae (Doi 1973; Rosenberger, this volume). Each of the papers in this volume provides important evidence of the centrality of uchi/soto to our analysis of Japanese culture and behavior: parallels between the behavior of family members (Rosenberger) and office workers (Kondo) make sense only in the recognition of uchi/soto; even as simple an interaction as ordering vegetables at a market (Sukle) falls within the analytical scope of uchi/soto. The explanatory power of the uchi/soto distinction cannot be ignored in our analysis of Japanese behavior—including linguistic behavior and, in particular, Japanese deixis. Uchi comes to constitute the central anchor for one's identity in Japan. NAKANE'S ACCOUNT OF GROUP CONSCIOUSNESS

Nakane's pioneer work on Japanese group formation and structure begins with the observation that for the Japanese there is a tendency to emphasize situational position over individual attribute in any given frame: [W]hen ajapanese "faces the outside" (confronts another person) and affixes some position to himself socially he is inclined to give precedence to institution over kind of occupation. . . . In group identification, a frame such as a "company" or "association" is of primary importance; the attribute of the individual is a secondary matter. . . . Such group consciousness and orientation fosters the strength of an institution, and the institutional unit (such as a school or company) is in fact the basis of Japanese social organization. (1970, 2-3)

INDEXING OF UCHI AND

SOTO

75

The main shortcoming of Nakane's work is that it does not go far enough in relating group consciousness to behavior. Group consciousness for Nakane is at most a cognitive category or a matter of ideology that has no real explanatory power when we attempt to apply it to observations of linguistic behavior. Nakane's portrayal of group consciousness results in rigid group boundaries, and this is contradicted by all the papers in this volume. As has been demonstrated again and again, uchi/ soto involves shifting boundaries that at once include and define the self. "There are no fixed points, either 'self or 'other' . . . there is no fixed center from which, in effect, the individual asserts a noncontingent existence" (Smith 1983, 77-81). What this means for our analysis of the language is that we need to reexamine categories that have long been assumed to be axiomatic, in particular those that are said to define deixis. DEIXIS: UCHI AS ANCHOR

The uchi/soto distinction has far-reaching implications for study of the language. Overwhelming evidence from Japanese language and behavior indicates that in Japanese the deictic center is not in fact ego as understood or intended by those of us who share a Western cultural heritage, but rather a situationally dependent uchi with fluid boundaries. This uchi must always include the speaker,2 but its boundaries shift from moment to moment depending on the kinds of sociocultural factors explored in this volume. When we recognize uchi in Japanese we open up the possibility of a very different sort of deictic center from that to which we are accustomed. An early categorization of phenomena that are currently termed dactic was developed byjesperson (1922), who coined the word shifters for elements of language "whose meaning differs according to the situation" (123). Jakobson's (1957) expanded treatment recognizes categories of person, gender, tense, mood, and elements like this/that, here/ there, as deictic. Lyons's (1977) definition of deixis is provisionally adopted here: 'The location and identification of persons, objects, events, processes, and activities being talked about, or referred to, in relation to the spaciotemporal context created and sustained by the act of utterance and participation in it, typically, of a singk speaker [emphasis mine] and at least some addressee" (636). Lyons's definition rests firmly on the notion that ego is the deictic anchor, and much of linguistic as well as philosophical discussion of deixis explores the nature of "I." In Japan however, where the basic unit of society is not the individual self but uchi, the deictic anchor point comes into serious question (Bachnik 1982, 11).

76

PATRICI A J WETZE L

A corollar y of Lyons' s definitio n is th e assumptio n tha t first perso n is primary , an d secon d an d thir d perso n are someho w derivative (see, for example , Benvemst e 1946) Form s tha t encod e perso n in variou s languages have com e unde r intens e scrutin y in th e Western linguistic tradi tio n Th e overridin g concer n with grammatica l perso n is th e reaso n why perso n deixis has been investigate d to th e exclusion of deixis encode d elsewhere in Japanes e PERSO N PERSONA L PRONOUNS , ANAPHOR A AND ELLIPSI S

Whethe r person an d all tha t it implie s is a ter m tha t can be applie d to Japanes e is a questio n tha t ha s so far received little or n o attentio n Japa nese linguistic s follows th e Western lead in assumin g th e centralit y of perso n to description s of Japanes e deixis Thi s approac h ha s resulte d in a multitud e of analyses of Japanes e pronoun s an d pronominahzatio n (Hind s 1978, Kun o 1973, Kurod a 1967, Miyaji 1964, Pen g 1973 ), ana phor a (Farme r 1980, Oshim a 1979, Saito an d Hoj i 1983, Whitma n 1982), an d ellipsis (Clanc y 1980, Hamad a 1983, Hind s 1978, 1980, Kurod a 1967) Th e single mos t importan t argumen t in favor of recognizin g grammatica l perso n within th e Japanes e languag e is tha t th e form s watakusht Τ an d anata 'you' refer uniquel y to speake r an d addresse e Sinc e th e use of bot h of thes e form s depend s on th e participan t roles of th e speake r an d addressee , the y fall within Lyons' s definitio n of deixis Yet th e bound s of perso n as such are problemati c in Japanes e Wolff (1980, 20) mention s uses of watakusht such as watakushi-ntsu 'private' , watakusht sum 'tak e for one' s own use, embezzle' , an d wataku shi no nat 'unselfish , impartial , fair' H e point s ou t tha t in thes e expressions watakusht is no t used for self referenc e But in fact, watakusht is no t used for referenc e at all in thes e example s As a referrin g expression , watakusht can onl y be used for th e first person , an d anata for th e addressee Onl y in nonreferrin g expression s like thos e listed above is watakusht used in th e sense of 'self or 'private ' Grante d tha t ther e are form s reflective of th e participan t roles of speake r an d addressee , as Lyon s suggests are necessar y for person , th e next questio n center s on what we are to mak e of th e remainin g bod y of term s tha t are typically include d in discussion s of Japanes e perso n In th e spoken languag e alone , amon g form s for speake r referenc e Mille r (1967) lists watakusht, watasht, washt, atakusht, atasht, temae, waga hat, boku, ore, ora, otra , an d for addresse e anata, anta, sensei, omae, ktmt, ktsama, an d temae Suzuk i (1978, 98-100 ) add s kochtra to th e list of first-perso n forms,

INDEXIN G O F UCHI AN D SOTO

77

a n d note s t h a t all thes e ar e specialize d uses of definit e description s whos e m e a n i n g s ar e watakushi 'private , personal ' temae 'thi s side o f boku 'your servant ' ore'onesel f kochira 'thi s direction '

anata 'tha t direction ' omae 'front ' kimi 'lord , prince ' kisama 'nobl e person ' sensei 'teache r

Martin' s (1975 , 1075-79 ) list of first-perso n form s include s watakushi (varian t form s watashi, atashi, atakushi, watai, wate, wai, atai, ate, watchi), boku, ore/ora, uchi, jibun, a n d ono/onore while hi s accompanyin g list of second-perso n form s include s anata, omae, kimi, otahu, a n d sochira O n e o u t s t a n d i n g featur e o f thes e lists o f first- a n d second-perso n form s is thei r lac k of a g r e e m e n t as t o wha t is t o b e i n c l u d e d Is sensei ' t e a c h e r ' a second-perso n for m o r is it n o t ? T h e r e ar e n o criteri a offere d for p u t t i n g t o g e t h e r an y of thes e lists Mille r (1967, 341) observe s t h a t "th e multiplicit y o f thes e word s reflect s in a n almos t one-to-on e rati o t h e m a n y levels i n t o whic h J a p a n e s e societ y itself is s t r u c t u r e d " Thi s is remarkabl y simila r t o a situatio n tha t Lyon s (1977 , 641-43 ) speculate s a b o u t t h e possibilit y of a languag e lackin g persona l p r o n o u n s as such , b u t whic h make s use of a "special subset of definit e description s (suc h as 'master' , ' s e r v a n t ' ) " for vocative a n d referrin g expression s T h e s e special nominals , h e allows, "are indirectl y relate d t o p a r t i c i p a n t role s b u t it doe s n o t follow fro m thi s fact t h a t the y ar e persona l p r o n o u n s , o r even t h a t the y grammaticahz e t h e categor y of p e r s o n " Thi s raises t h e additiona l issue of w h e t h e r p e r s o n mus t be e n c o d e d via persona l p r o n o u n s Account s of p e r s o n in J a p a n e s e overwhelmingl y o p e r a t e u n d e r t h e assumptio n t h a t t h e J a p a n e s e form s t h a t e n c o d e per son ca n safely b e t e r m e d pronouns, o r in J a p a n e s e daimeishi (lit 'pro noun' ) A r g u m e n t s over w h e t h e r t h e categor y " p r o n o u n " applie s to J a p a n e s e focu s primaril y o n t h e existenc e of forma l feature s t h a t m i g h t distin guish p e r s o n a l p r o n o u n s fro m o t h e r n o u n s in t h e languag e K u r o d a (1967) c o n c l u d e s t h a t item s suc h as watakushi a n d anata ar e n o t t o b e t e r m e d pronouns, base d o n t h e fact t h a t the y d o n o t hav e characteristi c declension s Contras t Englis h I/me, she/her, a n d he/him with t h e Japa nes e form s watakushi Ί / m e ' , kanojo ' s h e / h e r ' , a n d kare ' h e / h i m ' in t h e followin g e x a m p l e s 7a b 8a b

I called her/hi m She/h e called m e Watakush i ga kanojo/kar e ο yond a Kanojo/kar e ga watakush i ο yond a

78

PATRICI A J WETZE L

Kurod a also note s tha t th e Japanes e form s can be modifie d in th e same way as othe r nouns , tha t th e distributio n of noun s an d pronoun s is identica l in Japanes e Compar e th e following English example s to th e Japanes e 9a b 10a b

the short man *the short he chnsa i hito chnsa i kare

Hind s (1971) argues, in contrast , tha t Japanes e doe s indee d manifes t a separat e class of persona l pronoun s A distinguishin g featur e of thi s class, h e argues, is tha t it is obligator y tha t all its member s ad d a suffix such as ra or tachi in th e plura l Fo r all othe r [+human ] noun s thi s is optional , tha t is, th e sentenc e Kodomo wa byooki desu can be interprete d as 'Th e child or childre n are sick' But Watakushi wa byooki desu can onl y mea n Ί am sick' I n orde r to convey 'We are sick' th e plura l suffix tachi mus t be used Watakushi-tachi wa byooki desu Similarly, for third-perso n forms, kare 'he ' can onl y be singular, while kare-ra is plura l Hind s fails to observe, however, tha t prope r name s in Japanes e follow th e same plurahzatio n patter n as watakushi, tha t is, Tanaka san can onl y mea n 'Mr/M s Tanaka' , it canno t mea n 'th e Tanakas ' In orde r to indi cat e tha t th e entir e Tanak a family is sick th e plura l suffix -tachi is require d Tanaka-san-tachi wa byooki desu Hind s also claim s tha t th e persona l pronoun s canno t be modifie d by demonstrative s such as kono 'this ' an d sono 'that ' as othe r noun s can Marti n (1975, 1067) provide s evidenc e to th e contrar y kono watashi mo 'even thi s perso n who I am' , sono kare wa 'tha t he ' Thus , ther e are n o morphologica l or distributiona l criteri a tha t warran t recognizin g an independen t class of pronoun s in Japanes e Elsewhere , th e search for perso n in Japanes e leads to th e analysis of anaphor a an d ellipsis In term s of sheer use, it ha s bee n observed tha t Japanes e ten d no t to use perso n form s to th e same exten t that , for example , English doe s Thus , in on e dat a sample, 3 ther e were thre e uses of first- or second-perso n form s (two watashi an d on e boku), an d eleven othe r form s of address/referenc e tha t include d predominantl y use of name s an d th e ter m sensei Based on figures provide d by Allen an d Gu y (1974) an d Goffma n (1961), we would expec t upward s of 225 pronoun s pe r 1,000 words in a comparabl e English conversatio n Th e Japanes e dat a do no t approac h thi s figure Ther e is a good dea l of researc h to th e effect tha t th e anaphori c functio n of pronoun s in English correspond s mor e closely to ellipsis in Japa nese (Kurod a 1967, Hind s 1975,1977, 1978,1980, Clanc y 1980) Clanc y (1980, 133) shows tha t Japanes e speaker s used ellipsis in 73 percen t of

INDEXING OF £/67//AND SOTO

79

the places where a n o m i n a l was possible, c o m p a r e d to 20 p e r c e n t in English. She interprets this as evidence that "the distributions of English p r o n o u n s a n d J a p a n e s e ellipsis are m o r e similar t h a n those of n o m i n a l references" (140). Yet underlying the focus of interest o n personal p r o n o u n s , a n a p h o r a , a n d ellipsis is the assumption that the central deictic distinctions of Japanese are the same as those of English a n d Indo-European, a n d that those distinctions are reflected in n o m i n a l elements. It is just such an a p p r o a c h that, I think, fails to d o justice to the system underlying these elements, a n d m o r e seriously misses key generalizations a b o u t t h e m . SOCIAL D E I X I S

O n e of the most recently defined a n d explored linguistic p h e n o m e n a for which sociocultural distinctions play a descriptive role is what Levinson (1983) t e r m s social deixis. Social deixis refers to "the e n c o d i n g of social distinctions that are relevant to participant roles, particularly aspects of the social relationship holding between the speaker a n d the addressee(s) or speaker a n d some referent" (63). 4 Certainly J a p a n e s e elements such as the polite forms (honorific a n d h u m b l e forms) ,5 as well as the verbs of giving a n d receiving, fall within this characterization of social deixis: they e n c o d e social distinctions (uchi/soto) that are relevant to the social relationship holding between the speaker a n d the addressee(s) or speaker a n d some referent. Yet Levinson's account of social deixis is still clouded by an underlying assumption that social deixis is somehow derivative or peripheral to person deixis. T h e speaker (ego) is still assumed to b e the deictic anchor. But compare his definition of social deixis above with his earlier definitions for pragmatics a n d deixis: Pragmatics is the study of those relations between language and context that are grammaticahzed, or encoded in the structure of the language. (9) The single most obvious way in which the relationship between language and context is reflected in the structure of languages themselves is through the phenomenon of deixis. . . . Essentially, deixis concerns the ways in which languages encode or grammaticalize features of the context of utterance or speech event. (54) Are we to c o n c l u d e that deixis is the primary c o n c e r n of pragmatics? Does this c o n f o u n d i n g of deixis with pragmatics d e m o n s t r a t e a lack of focus in what we are to take to be the c o n c e r n of pragmatics? O r of investigations into deictic p h e n o m e n a ? T h e inadequacy of Levinson's analysis b e c o m e s even m o r e evident w h e n h e allows that "it is generally

80

PATRICI A J WETZE L

(bu t no t invariably) tru e tha t deixis is organize d in an egocentri c way" (63) Levinson fails to follow u p on what nonegocentri c deixis would mea n to ou r understandin g of deixis an d ou r analysis of languag e in genera l When for Japanes e we defin e uchi, an d no t ego, as th e deicti c ancho r point , it ha s profoun d implication s for ou r analysis of language In essence , we erase th e division tha t Levinson ha s drawn between deixis an d social deixis LINGUISTI C MANIFESTATION S OF UCHI AND SOTO

Nakan e observes "If on e listen s to a conversatio n between Japanes e on e doe s no t have to wait lon g to hea r th e words uchi-no or uchi dewa referrin g to one' s own peopl e an d work plac e Uchi may mea n th e institutio n as a whole, or it ma y mea n th e departmen t or sectio n to which th e speake r belongs" (1970, 125) Yet nomina l referenc e to uchi is onl y on e of a numbe r of ways in which th e distinctio n between uchi an d soto is signaled in Japanes e Th e even mor e frequen t use of Japanes e verbs of giving an d receiving, as well as polit e form s (honorifi c an d humbl e forms) , provide s a windo w on speakers ' perception s of th e uchi an d soto of th e situatio n in which the y occu r Th e Japanes e dat a sampl e men tione d above 6 containe d forty-thre e instance s of verbs of giving an d receiving an d polit e form s If we recogniz e th e deicti c natur e of th e verbs of giving an d receivin g an d polit e forms, ther e is muc h mor e deicti c referenc e tha n first realize d verbs of giving an d receivin g an d polit e form s outnumbere d perso n form s by mor e tha n ten to on e A few Japanes e grammarian s have trie d to call attentio n at least to th e directiona l natur e of th e polit e form s an d verbs of giving an d receivin g On e of thes e is Mikam i (1970, 149-53 ) who append s a discussion of verbs like kureru, morau, an d ageru to his discussion of th e deicti c an d anaphori c uses of th e demonstrative s Althoug h h e doe s no t overtly refer to thes e verbs as deictic , h e doe s refer to thei r "directiona l nature " (hookoosei) Th e dichotomy , h e says, is between verbs denotin g "for me " versus thos e denotin g "for non-me " (h e uses th e English) , an d h e ιηηο vatively calls th e forme r "centripetal " (kyuushin-teki) an d th e latte r "non centripetal " (hikyuushin-teki) Miyaji (1965) also attend s to th e directiona l natur e of th e verbs of giving an d receivin g in term s of ho w speake r viewpoint interact s with th e grammatica l perso n of th e subject, object , or indirec t object Thi s illustrate s th e tendency , even amon g Japanes e grammarians , to accep t perso n deixis as a poin t of departur e for th e analysis of ho w th e polit e form s an d verbs of giving an d receivin g functio n Bot h Marti n (1964, 408) an d Mille r (1967, 273) analyze th e verbs of giving an d receivin g in term s of th e speaker' s ιη-grou p or uchi—that is,

81

INDEXIN G O F UCHI AN D SOTO

as 'give to uchi' {kureru/kudasaru) , 'give (fro m uchi) t o sotd (ageru/sashiageru), a n d 'get fro m sotd (morau/itadaku) I n effect, we observe t h e set of verbs tha t ca n b e illustrate d as follows ( t h e differenc e betwee n plai n a n d polit e form s is discusse d below) kureru/kudasaru 'give to uchi ageru/sashiageru