Anna Wierzbicka, Cross Cultural Pragmatics

Cross-Cultural Pragmatics

W DE

G

Cross-Cultural Pragmatics The Semantics of Human Interaction Second edition by

Anna Wierzbicka

Mouton de Gruyter Berlin New York

2003

Mouton de Gruyter (formerly Mouton, The Hague) is a Division of Walter de Gruyter GmbH & Co. KG, Berlin. The first edition was published in 1991 as volume 53 of the series Trends in Linguistics. Studies and Monographs @ Printed on acid-free paper which falls

within the guidelines of the ANSI to ensure permanence and durability.

ISBN 3-11-017769-2 Bibliographic information published by Die Deutsche Bibliothek

Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available in the Internet at .

© Copyright 1991, 2003 by Walter de Gruyter GmbH & Co. KG, 10785 Berlin All rights reserved, including those of translation into foreign languages. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Printing: Gerike GmbH, Berlin. - Binding: Liideritz & Bauer-GmbH, Berlin. Cover design: Sigurd Wendland, Berlin. Printed in Germany.

Introduction to the second edition

I am very happy to see the demand for a new edition of my 1991 CrossCultural Pragmatics - the Semantics of Human Interaction. I am also happy to be able to say, in 2003, that since this book was first published the field of cross-cultural pragmatics has advanced enormously; and furthermore, that this progress has not only not made my 1991 CrossCultural Pragmatics dated, but that, on the contrary, its tenets and its overall approach have been essentially vindicated. A decade ago, the "pragmatic" scene was still largely dominated by the search for the "universals of politeness" and for the "universal maxims of conversation". The widely accepted paradigms were those of Brown and Levinson's (1978, 1987) theory of politeness, which affirmed "pan-cultural interpretability of politeness phenomena" (1978: 288), and Grice's (1975) theory of conversation, which posited a number of universal conversational principles. It is heartening to see to what extent the situation has now changed. In the nineteen eighties, and well into the nineties, the idea that interpersonal interaction is governed, to a large extent, by norms which are culture-specific and which reflect cultural values cherished by a particular society went against the grain of what was generally accepted at the time, and successive conferences of the Inernational Pragmatic Association and other similar occasions were dominated by studies seeking to confirm Grice's "maxims" and Brown and Levinson's "universals of politeness" in this or that new area, and this or that new language. In 1978, Brown and Levinson set out "to describe and account for what is in the light of current theory a most remarkable phenomenon. This is the extraordinary parallelism in the linguistic minutiae of the utterances with which people choose to express themselves in quite unrelated languages and cultures" (Brown and Levinson 1978: 60). A quarter of a century later, it is increasingly widely accepted that this "extraordinary parallelism" was largely an illusion due to that "light of current theory". (If you set out to show that everything can be described in terms of "negative and positive face" you may indeed find that everything can be so described.) What is seen as more remarkable today, is the extent of cross-linguistic and cross-cultural differences in ways of speaking. Brown and Levinson (1978: 61) described it as their goal "to rebut the once-fash-

VI

Introduction to the second edition

ionable doctrine of cultural relativity in the field of interaction" and "to show that superficial diversities can emerge from underlying universal principles and are satisfactorily accounted for only in relation to them". Their major conclusion was that "interactional systematics are based largely on universal principles" (1978: 288). Today, it is increasingly accepted that those diversities in ways of speaking and interacting are not superficial at all and that they can be accounted for, above all, in terms of different cultural attitudes and values; and the "cultural relativity in the field of interaction" is increasingly seen as a reality and an important subject for investigation. When in 1983 I presented, at the monthly meeting of the Sydney Linguistic Circle, a paper entitled "Different cultures, different languages, different speech acts: English vs. Polish" (Wierzbicka 1985), in which I argued that the supposedly universal maxims and principles of "politeness" were in fact rooted in Anglo culture, my ideas were regarded as heretical. When I argued, in particular, that the "freedom from imposition", which Brown and Levinson (1978: 66) saw as one of the most important guiding principles of human interaction, was in fact an Anglo cultural value, and that the avoidance of "flat imperative sentences", which Searle (1975: 69) attributed to the "ordinary (human, A. W.) conversational requirements of politeness", did not reflect "universal principles of politeness" but rather, expressed special concerns of modern Anglo culture, my claims were confidently dismissed. As a matter of fact, it was the hostile and dismissive reaction of that audience which was for me the initial stimulus for engaging in a long-term campaign against what I saw as a misguided orthodoxy of that time. From the perspective of the intervening years, I must be grateful for the negative reaction of that Sydney audience to a paper which became the nucleus of my 1991 Cross-Cultural Pragmatics. I am even more grateful, however, to other linguists, who in that inhospitable post-Gricean climate were also raising their voices in defence of culture as a key factor determining ways of speaking, and in particular, to those who ventured to link language-specific ways of speaking with different cultural values. To mention just a few scholars, whom I saw in those early years, and whom I see now, as "comrades-in-arms": Ho-min Sohn, the author of a pioneering study "Intercultural communication and cognitive values" (1983); Tamar Katriel, the author of Talking Straight: Dugri speech in Israeli Sabra culture (1986); Yoshiko Matsumoto, the author of "Reexamination of the universality of face: politeness phenomena in Japanese" (1988); James Matisoff, the author of Blessings, Curses, Hopes, and Fears:

Introduction to the second edition

VB

Psycho-Ostensive Expressions in Yiddish (1979); Thomas Kochman, the author of Black and White Styles in Conflict (1981); Sachiko Ide, the author of a study on the Japanese value of wakimae or discernment (1989); Donal Carbaugh, the author of Talking American (1990); and closer to home, my colleagues: Cliff Goddard, whose numerous publications are listed in the References; Jean Harkins, the author of Bridging Two Worlds: Aboriginal English and Cross-Cultural Understanding (1994); Felix Ameka, the author of studies on Ghanaian conversational routines and the editor of a volume on interjections (see the References); and Michael Clyne, the author of Intercultural Communication at Work: Cultural Values in Discourse (1994). Last but not least, I would like to mention the important role of two open-minded and cross-culturally alive journals: Jacob Mey's Journal of Pragmatics, and Marcelo Dascal's Pragmatics and Cognition. Outside linguistics, there were of course anthropologists who did not give in to the superficial and anti-cultural universalism of the time and who continued to focus on the language-particulars and to probe the links between ways of speaking, ways of thinking, ways of feeling and ways of living. To mention just a few names and works, particularly important from a linguistic point of view: Catherine Lutz, the author of the classic book Unnatural Emotions: Everyday sentiments on a Micronesian atoll and their challenge to Western theory (1988); Richard Shweder, the founder of "cultural psychology" and the author of Thinking Through Cultures - Expeditions in Cultural Psychology (1991); Dorothy Holland and Naomi Quinn, the editors of Cultural Models in Language and Thought (1987); and Roy D'Andrade and Claudia Strauss, the authors of Human Motives and Cultural Models (1992). I would also like to mention here two journals which I see as especially important: Ethos and Culture and Psychology. There were also some philosophers who started to question the pragmatic theories of Grice, Griceans, and "neo-Griceans" from a philosophical as well as cross-linguistic point of view. In particular, Wayne Davis (1998) has argued in a book-length critique that "the Gricean theory has been barren" and that "the illusion of understanding provided by the Gricean theory has only served to stifle inquiry" (Davis 1998: 3). "The Gricean explanation of common implicatures" is, Davis argued, "undermined by the existence of nonuniversal implicature conventions" (Davis 1998: 183). For example, Grice and his followers (e.g. Levinson 1983) have claimed that the correct interpretation of a tautology like War is war can

VIll

Introduction to the second edition

be calculated from some universal maxims of conversation. Davis points out (with reference to Chapter 10 of my Cross-Cultural Pragmatics) that this claim is refuted by the observation that such tautologies receive different interpretations in different cultures, and he concludes: "The moral is clear. Generalized tautology implicatures ... are not explained by Gricean Maxims" (Davis 1998: 46). In a similar context, Davis (1998: 168) quotes and endorses my own observation that "from the outset, studies in speech acts have suffered from an astonishing ethnocentrism" (Wierzbicka 1985: 145). Since the decline of the Gricean paradigm, which, as Davis puts it, has only served to stifle inquiry, defines to a large extent the difference in the context between this second edition of my Cross-Cultural Pragmatics and the 1991 one, I hope I can be forgiven for quoting at some length Davis' historical account, including his comments on my own work. Many of the criticisms I present have been known for some time. But the import and seriousness of the defects individually and collectively have not been widely appreciated, and the problems have had little impact on the general acceptance of Gricean theory. The best known critics of the Gricean theory have either expressed confidence that solutions would be found within the Gricean framework (Harnish 1976) or presented alternative theories with similar defects (Sperber and Wilson 1986).... Only one author (Wierzbicka 1987) has argued that the conception is fundamentally flawed. (Davis 1998: 3)

I hasten to add that Davis has reserved some critical comments for me too, and that I will quote these later. What matters at this point is the historical record, which the reader of this second edition is entitled to know. From the historical, as well as theoretical, point of view, it is important to note that a powerful impulse for the rise of cross-cultural pragmatics in the last decade came from the growing field of studies focussed on cross-cultural (or inter-cultural) communication. I have quoted in the 1991 book Deborah Tannen's (1986: 30) statement that "the future of the earth depends on cross-cultural communication". At a time when every year millions of people cross the borders, not only between countries but also between languages, and when more and more people of many different cultural backgrounds have to live together in modern multi-ethnic and multi-cultural societies., it is increasingly evident that research into differences between cultural norms associated with different languages is essential for peaceful co-existence, mutual tolerance, necessary understanding in the work-place and in other walks of life in the increasingly "global" and yet in many places increasingly diversified world.

Introduction to the second edition

IX

The once popular assumption that the "principles of politeness" are essentially the same everywhere and can be described in terms of "universal maxims" such as those listed in Leech (1983: 132) flies in the face of reality as experienced by millions of ordinary people - refugees, immigrants, the children of immigrants, caught between their parents and the society at large, cross-cultural families and their children, and also by monolingual "stay-at-homes" who suddenly find themselves living in societies which are ethnically, culturally and linguistically diverse. In addition to their obvious untruth in relation to daily experiences of millions of people, the supposed "universals of politeness" and the supposed "universal principles of conversation" are clearly of no use in the practical task of furthering cross-cultural communication. When, for example, a well-meaning, liberal Anglo-Australian says of her Chinese neighbours that "they are very good neighbours - but they are so rude ... for example, they said to me: cut down that branch - we don't want it on our side of the fence" (Canberra 2002), if we as linguists tell her and others like her that the principles of politeness are essentially the same everywhere (recall Brown and Levinson's tenet of "pan-cultural interpretability of politeness phenomena" quoted earlier), we can only confirm her in her view that the Chinese neighbours are very rude (cf. Clyne 1994). The tremendous practical importance of identifying, and describing, the culture-specific norms of "politeness" and, more generally, norms of interpersonal interaction, has been increasingly recognized by the field of language teaching. In this field, too, the realization grew steadily over the last decade or so that "Grice's Razor", which extols the economical virtues of concentrating on the supposed universality of the "underlying principles" and which cuts off "unnecessary" culture-specific explanations, spells out a disaster for the students' communicative competence and their ability to survive socially in the milieu of their "other" language. As Kramsch (1993) puts it in her book Context and Culture in Language Teaching: If ... language is seen as social practice, culture becomes the very core of language teaching. Cultural awareness must then be viewed both as enabling language proficiency and as being the outcome of reflection on language proficiency.... Once we recognize that language use is indissociable from the creation and transmission of culture, we have to deal with a variety of cultures. (Kramsch 1993: 89)

A key question for Kramsch and many other contemporary theorists and practitioners of language teaching aimed at communicative competence is this: "How can a foreign way of viewing the world be taught via an

x

Introduction to the second edition

educational culture which is itself the product of native conceptions and values?" (Kramsch 1993: 9). Clearly, it is not a question that Grice's Razor or the supposedly universal notion of "positive" or "negative" "face" can help answer. My own long campaign against the fictitious and harmful "universals of politeness" and "universal principles of human conversation" is rooted in my own experience as a "language migrant" (to use a term introduced by Mary Besemeres, 1998 and 2002) - from Polish into English, especially academic English, in which I have written many books and articles, and also, into Australian English, which has been my daily linguistic environment for thirty years. I have described this experience in some detail in an article entitled "The double life of a bilingual - a cross-cultural perspective" (Wierzbicka 1997b). On a very small scale, this article illustrates an important new aspect of cross-cultural pragmatics as it has evolved over the last decade or so: the new alliance between, on the one hand, linguistic pragmatics, based on "hard linguistic evidence" and rigorous linguistic analysis, and, on the other, the new field of study focused on the "soft data" of personal experience of cross-cultural and cross-linguistic living (cf. Besemeres 2002; Dalziell 2002). I have referred to my own cross-linguistic and cross-cultural experience in a number of publications, both before and after the 1991 Cross-Cultural Pragmatics. Here, I will permit myself to adduce several long quotes from that 1997 cross-cultural memoir, which deliberately takes a personal rather than "objective" perspective. I believe that such a personal perspective legitimizes the insistence with which proponents of cross-cultural pragmatics have been challenging, in the last decade or so, the earlier paradigm. Commenting on my life in Australia, to which I emigrated from Poland in 1972 (having married an Australian) I wrote: I had to start learning new "cultural scripts" to live by, and in the process I became aware of the old "cultural scripts" which had governed my life hitherto. I also became aware, in the process, of the reality of "cultural scripts" and their importance to the way one lives one's life, to the image one projects, and even to one's personal identity. For example, when I was talking on the phone, from Australia, to my mother in Poland (15,000 km away), with my voice loud and excited, carrying much further than is customary in an Anglo conversation, my husband would signal to me: 'Don't shout!' For a long time, this perplexed and confused me: to me, this 'shouting' and this 'excitement' was an inherent part of my personality. Gradually, I came to realise that this very personality was in part culturally constituted. (Wierzbicka 1997b: 119)

Introduction to the second edition

Xl

The realization of the close links betweeen my ways of speaking, my personality and my Polishness raised for me the question that countless other immigrants are constantly confronted with: to what extent was it desirable, or necessary, to change myself in deference to my new cultural context? Early in our life together, my husband objected to my too frequent - in his view - use of the expression of course. At first, this puzzled me, but eventually it dawned on me that using of course as broadly as its Polish counterpart oczywiscie is normally used would imply that the interlocutor has overlooked something obvious. In the Polish 'confrontational' style of interaction such an implication is perfectly acceptable, and it is fully consistent with the use of such conversational particles such as, for example, przeciei ('but obviously can't you see?'). In mainstream Anglo culture, however, there is much more emphasis on 'tact', on avoiding direct clashes, and there are hardly any confrontational particles comprarable with those mentioned above. Of course does exist, but even oj' course tends to be used more in agreement than in disagreement (e. g. 'Could you do X for me?' - 'Of course'). Years later, my bilingual daughter Mary told me that the Polish conversational expression alei oczytviscie: 'but-EMPHATIC of course' (which I would often replicate in English as 'but of course') struck her as especially 'foreign' from an Anglo cultural point of view; and my close friend and collaborator Cliff Goddard pointed out, tongue in cheek, that my most common way of addressing him (in English) was 'But Cliff ... '. (Wierzbicka 1997b: 119)

Thus, I had to learn to avoid overusing not only of course but also many other expressions dictated by my Polish "cultural scripts"; and in my working life at an Anglo university this restraint proved invaluable, indeed essential. I had to learn to 'calm down', to become less 'sharp' and less 'blunt', less 'excitable', less 'extreme' in my judgements, more 'tactful' in their expression. I had to learn the use of Anglo understatement (instead of more hyperbolic and more emphatic Polish ways of speaking). I had to avoid sounding 'dogmatic', 'argumentative', 'emotional'. (There were lapses, of course.) Like the Polish-American writer Eva Hoffman (1989) I had to learn the use of English expressions such as 'on the one hand ... , on the other hand', 'well yes', 'well no', or 'that's true, but on the other hand'. Thus, I was learning new ways of speaking, new patterns of communication, new modes of social interaction. I was learning the Anglo rules of turntaking ('let me finish!', 'I haven't finished yet!'). I was learning not to use the imperative ('Do X!') in my daily interaction with people and to replace it with a broad range of interrogative devices ('Would you do X?' 'Could you do X?' 'Would you mind doing X?' 'How about doing X?' 'Why don't you do X?' 'Why not do X?', and so on). (Wierzbicka 1997b: 119-120)

xu

Introduction to the second edition

As I discussed in that 1997 memoir, these were not just changes in the patterns of communication, these were also change 'in my personality. I was becoming a different person, both in the context of my cross-cultural family and in the context of my work as a university teacher. Students' course assessment questionnaires have often thrown light on my cultural dilemmas. Thus, while often very positive and praising my 'enthusiasm', for a long time they also often included critical accents referring to my 'intensity', 'passion" and 'lack of detachment'. I was coming from a language and culture system (Polish) where the very word beznamif2tny (lit. 'dispassionate') has negative connotations, but I was lecturing in a language (English) where the word dispassionate implies praise while the word emotional has negative connotations. I had to learn, then, to lecture more like a 'spokesman' and less like an 'advocate' (in Kochman 1981 terms). I had to learn to become less 'emotional' and more 'dispassionate' (at least in public speaking and in academic writing). (Wierzbicka 1997b: 120)

And yet, while I saw some cultural adaptation as necessary I did not want to adapt too much; I felt instinctively that the social benefits of such an adaptation needed to be balanced against the personal cost involved in it. There were therefore limits to my malleability as a 'culturally constituted self'. There were English modes of interaction that I never learnt to use - because I couldn't and because I wouldn't: they went too much against the grain of that 'culturally constituted self'. For example, there was the 'How are you' game: 'How are you?' - 'I'm fine, how are you?'; there were weather-related conversational openings ('Lovely day isn't it?' - 'Isn't it beautiful?'). There were also 'white lies' and 'small talk' (the latter celebrated in a poem by the Polish poet and professor of Slavic literatures at Harvard University, Stanislaw Baranczak). The acute discomfort that such conversational routines were causing me led me to understand the value attached by Polish culture to 'spontaneity', to saying what one really thinks, to talking about what one is really interested in, to showing what one really feels. It also led me to contemplate the function of such linguistic lubricants in Anglo social interaction. Why was it that Polish had no words or expressions corresponding to 'white lies' or 'small talk'? Why was it that English had no words or expressions corresponding to basic Polish particles and 'conversational signposts' such as przeciei, alei ('but can't you see?') alei skg,die (lit. 'but where from?' i.e. where did you get that idea?), or skg,die znowu ('but where from again?') - all expressions indicating vigorous disagreement, but quite acceptable in friendly interaction in Polish?

Introduction to the second edition

XlII

As I meditated on my experience, and as I discussed it with other immigrants, I developed a strong theoretical interest in the problems of cross-cultural understanding and a -deep conviction that the universalist theories of human interaction dominant of the time were fundamentally flawed. Clearly, the rules for 'friendly' and socially acceptable interaction in Polish and in English were different. Consequently, I could never believe in the "universal maxims of politeness" and in the universal "logic of conversation" promulgated in influential works such as Grice (1975), Leech (1983) or Brown and Levinson (1978, 1987). I knew from personal experience, and from two decades of meditating on that experience, that Polish "maxims of politeness" and the Polish rules of "conversational logic" were different from the Anglo ones. (Wierzbicka 1997b: 120)

As these quotes make clear, the personal knowledge derived from such personal experience was not purely theoretical: above all, it was practical. I had no doubt that the insistence on cultural differences was not only theoretically justified (because these differences were real) but also that acknowledging them, and above all, describing them, was vitally important for the practical purposes of cross-cultural communication and understanding; and in the case of people like myself, of daily living. The 1991 edition of my Cross-Cultural Pragmatics was an attempt to challenge the Gricean and Brown-and-Levinsonian paradigms, and to expose the anglocentric character of various supposedly universal maxims, principles and concepts (including the key concept of "face", which was the linchpin of Brown and Levinson's theory of "politeness"). Twelve years later it can be said that tide has changed and it may seem unnecessary and unkind to press the same charges again. In response I would say that, first of all, many linguists who are out of touch with the developments in the fields of intercultural communication and language teaching are not aware of this change of tide and assume that the Gricean and neo-Gricean paradigms are still held by "those in the know" in the same esteem as they once were. But there is also another reason why some of the old charges still need to be pressed. This second reason has to do with the fact that, paradoxically, while the universalist pragmatic frameworks developed in the seventies were gradually losing their appeal, the program of actually describing the different ways of speaking and thinking linked with different cultures continued to encounter a great deal of resistance and criticism. As the differences between cultures and subcultures were increasingly celebrated, there was also a growing suspicion of any generalizations as

XIV

Introduction to the second edition

to what exactly these differences might be. Diversity was seen as beautiful but also as inherently elusive and indescribable. With the growing emphasis on diversity, the view gradually developed that diversity was everywhere, and that while those differences could and should be celebrated they could not be described. Thus, in many quarters, there developed a great fear of the notion of culture (especially, "a culture"), and attempts to identify any differences between particular cultures came to be seen as "static culturologies" (cf. Darnell 1994). For example, the anthropologist Eric Wolf, writing of "the heterogeneity and the historically changing interconnectedness of cultures" (Wolf 1994: 5), argued that "notions of a common cultural structure underlying all this differentiation sound a bit too much like a little cultural homonuculeus built into everyone through the process of socialization" (Wolf 1994: 6). Another anthropologist, Immanuel Wallenstein, spoke in the same vein in his commentary on Wolf's paper, for example: "races, cultures, and peoples are not essences. They have no fixed contours. They have no self-evident content. Thus, we are all members of multiple, indeed myriad, 'groups' - crosscutting, overlapping, and ever-evolving" (Wallenstein 1994: 5; for discussion, see Wierzbicka 1997a). There can be no quarrel with the claims that "cultures are not essences", that "cultures are not monads", and that "cultures have no fixed contours". But to conclude from this that cultures cannot be discussed, described, and compared at all - because they have no substance at all would be a spectacular case of throwing the baby out with the bath water. It would also be a conclusion denying the subjective experience of immigrants, and, as I have argued in detail elsewhere (Wierzbicka, forthcoming), one going against their vital interests. To deny the validity of the notion of culture-specific cultural patterns (including "Anglo" cultural patterns) is to place the values of political correctness above the interests of socially disadvantaged individuals and groups. At this point, it will be apposite to return to Davis' (1998) critical comment on my own work, to which I have alluded earlier. Characteristically, this comment refers especially to my remarks on Anglo culture. To quote: To the extent that norms for polite, cooperative, efficient communication vary from culture to culture, so should implicature conventions. Thus Wierzbicka (1985) offered the "heavy restrictions on the use of the imperative in English and the wide range of use of interrogative forms in performing acts other than questions" as "striking linguistic reflexes" of the Anglo-Saxon cultural tradition, one that "places special emphasis on the rights and on the autonomy

Introduction to the second edition

xv

of every individual, which abhors interference in other people's affairs," and so on. She observed that languages such as Polish, used by speakers with opposed cultural traditions, have different conventions involving imperatives and interrogatives. The fact that Wierzbicka is fighting ethnocentrisn1 with cultural stereotypes does not diminish her point. (Davis 1998: 174-175)

The fear of "cultural stereotypes" has been as great an obstacle in the development of cross-cultural pragmatics as has the fear of "essentialism" and the "reification" of cultures. Giving in to this fear, Davis seems to be doing something analogous to what he himself criticized Brown and Levinson for, when he said that they "note the evidence but insist the 'underlying principles' are universal, derivable from universal face assumptions and rationality" (Davis 1998: 167). Similarly, Davis notes the evidence concerning the language-specific character of pragmatic conventions but he rejects off-hand any possible links between different pragmatic conventions and different cultural attitudes and values. He accepts that those conventions are not universal and he himself calls for "historical and sociolinguistic research ... which did not and could not arise when the Gricean theory held sway" (Davis 1998: 3). At the same time, however, he feels compelled to dismiss cross-cultural generalizations as "stereotyping" . Yet from the point of view of effective cross-cultural understanding and intercultural comlllunication it is essential not only to know what the conventions of a given society are but also how they are related to cultural values. For example, the Chinese immigrants in Canberra need to be told not only to be careful with the imperative when speaking to their Anglo neighbours, but also, why the imperative (e. g. "cut down that branch - we don't want it on our side of the fence") can be perceived as offensive and "rude" in Australia. Similarly, the Anglo-Australians need to be told not only that they should be "tolerant" to their Chinese neighbours, but also, that their own imperative-avoiding conventions reflect special historically-shaped concerns of their own culture rather than any natural and universal principles of politeness. With the increasing domination of English in the world, both Anglos and non-Anglos need to learn about various Anglo "cultural scripts". To try to describe these scripts, and to explain the values reflected in them, is not to indulge in stereotyping, but on the contrary, it is to help Anglos to overcome their inclination to stereotype Chinese (or, for that matter, Polish) immigrants as "rude", while at the same time helping the immigrants to better fit in, socially, and to improve their lives. As more and more often noted by bilingual and bicultural theorists such as, for exam-

XVI

Introduction to the second edition

pIe, Young Yun Kim (2001), for millions of people in the modern world cultural adaptation is necessary for survival; and liberal monocultural Anglos fixated on fighting "stereotypes" are not helping the cause of that adaptation and of increased inter-cultural understanding. In this context, I would like to emphasize again the new light thrown on problems of cross-cultural pragmatics by the new field of studies focussing on the experience of bilingual and bicultural persons, and in particular, on the immigrant experience. It is becoming more and more obvious to those concerned with cross-cultural understanding that in addition to objective methods usually employed in social sciences (data collection, statistical tables, diagrams, and so on), the voices of flesh-andblood people crossing linguistic and cultural boundaries need also to be taken into account. "The immigrant experience of having to 'translate oneself' from one's mother tongue into a foreign language and losing part of oneself in the process" (Besemeres 2002: 9) can expose what Davis (1998) calls the stifling effect of universalistic accounts of human conversation better than many scrupulous objective studies of linguistic competence or behaviour. It can also show more clearly than purely theoretical debates that cultures are real and that they can influence and even shape people's lives and people's selves. If this or that theoretical framework is not helpful in describing cultural differences in ways of speaking, thinking and feeling, it can only blame itself for its irrelevance to cross-cultural understanding, intercultural communication, language teaching, and what John Locke called "human understanding" in general. In my 1991 Cross-Cultural Pragmatics I did seek to describe and compare different cultures, and I did use expressions like "Vietnamese culture", "Japanese culture", "Anglo-American culture", "Polish culture", and so on. Given the potential for misunderstanding that such terms carry with them I would now prefer to avoid them, as far as possible, and to use instead terms like "cultural patterns", and especially, "cultural scripts". Both these terms were used in the 1991 text of the book, but along with, say, "Japanese cultural scripts" or "Anglo cultural scripts" I was also using quite freely terms like "Japanese culture" or "Anglo culture". Given present-day sensitivities, it will be in order to warn the reader explicitly that by using such terms I did not mean to imply that I see those cultures as immutable essences, self-contained monads, or "bounded, coherent and timeless systems of meaning" (Strauss and Quinn 1997: 3). Rather, I was using such terms as convenient abbreviations, referring to complexes of shared understandings or, as colleagues and I have been calling them for years, "cultural scripts". To quote Strauss and Quinn (1997) again:

Introduction to the second edition

XVll

Our experiences in our own and other societies keep reminding us that some understandings are widely shared among members of a social group, surprisingly resistant to change in the thinking of individuals, broadly applicable across different contexts of their lives, powerfully motivating sources of their action, and remarkably stable over succeeding generations. (Strauss and Quinn 1997: 3)

In the twelve years which have elapsed between the first and the present edition of his book, colleagues and I have been increasingly moving from the language of "cultures" to that of "cultural scripts". Since we have never thought of cultures as "timeless monads", this is above all a change in the style of exposition. The formulae included in this book under headings like "Polish culture" or "Japanese culture" would now be presented explicitly as "cultural scripts". Although this would be only a change in presentation, not in substance, it would be an important change. Since for logistic reasons this change is not being made in the text of this book, the reader of this second edition is asked to bear this point in mind: this book is not seeking to describe whole cultures, let alone to imply that these cultures are immutable, but rather, to articulate certain specific "cultural scripts". At the same time, I would like to point out to the reader that since the publication of the first edition, the idea of "cultural scripts", implicit in this book, has come into its own as a full-fledged theory - the theory of cultural scripts, which has by now resulted in many descriptive studies, across many languages and cultures (or "lingua-cultures", cf. Attinasi and Friedrich 1995). Since the idea of cultural scripts has now been developed into a theory of cross-cultural pragmatics, inter-cultural communication and indeed cross-cultural understanding in general, the reader of this second edition may wish to follow up the development of this theory and its applications in descriptive studies. For this, the starred references listed at the end of this introduction may be particularly useful. The theory of cultural scripts is an offshoot of the NSM (Natural Semantic Metalanguage) theory, on which all the analyses in this book are based. In a nutshell, this theory postulates that semantic analysis should be based on empirically established universal human concepts, that is, simple concepts realized in all languages as words or word-like elements, such as GOOD and BAD, KNOW, THINK, WANT and SAY, DO and HAPPEN and fifty or so others. In relation to cross-cultural pragmatics, this means that cultural norms of speaking should be formulated neither in technical or semi-technical English terms like "formal" and "informal" or "direct" and "indirect", nor in terms of English folk categories like

XVl11

Introduction to the second edition

"apology", "compliment", "sarcasm", "understatement" and so on, but rather in terms of simple words which have equivalents in all languages, such as those mentioned above (in small capitals). The use of such concepts can free us from what Goddard (2002c, in press a, b, c) calls "terminological ethnocentrism" and give us a neutral, culture-independent metalanguage for describing different cultural norms. At the same time, the use of such concepts allows us to capture the native speaker's point of view, without distorting it through the application of descriptive tools rooted in the English language or Anglo academic culture. On this point, NSM-based approach to cross-cultural pragmatics differs radically from that characteristic of works like Blum-Kulka et al.'s (1989) Cross-Cultural Pragmatics: Requests and Apologies or Kasper and Blum-Kulka's (1993) Interlanguage Pragmatics. While the works in this tradition must be appreciated for their attention to cultural differences reflected in ways of speaking, they cannot escape the charge of terminological, and not only terminological, ethnocentrism. Given that words like requests and apologies stand for conceptual artefacts of the English language, using them as analytical tools inevitably involves imposing an Anglo perspective on other languages and cultures. To describe ways of speaking across languages and cultures in terms of folk categories encoded in English is like describing English talk in terms of Japanese, Hebrew or Russian folk categories (e.g. the Japanese wakimae, cf. Ide 1989; the Hebrew dugri, cf. Katriel 1986; or the Russian vran'e, cf. Wierzbicka in press). But of course nobody would dream of describing English in such terms. The unshakable conviction shared by so many semanticists and pragmaticists that it is all right to try to describe all languages through English terms untranslatable into the language of speakers whose ways of thinking those terms are supposed to explain and illuminate shows the same astonishing anglocentrism as the Gricean and post-Gricean maxims, principles, and "conversational postulates" (cf. Gordon and Lakoff 1975) once did. By contrast, words like good and bad or say, think, know and want, which as evidence suggests have morpho-lexical exponents in all languages, free us from an Anglo perspective, while allowing us at the same time to retain a mini-lexicon of sixty or so English words as a practical lingua franca for articulating different culture-specific conventions, norms and values. Judging from some reviews, and some other responses to the first edition of this book which have been reported to me, I was understood by

Introduction to the second edition

XIX

some readers of that edition as claiming that "semantics should swallow pragmatics". This is a misunderstanding that the theory of cultural scripts should effectively dispel. What I did and do claim is, first, that a great many subjective and attitudinal meanings are indeed semantically encoded, and second, that since all observations on language use have to be themselves formulated in some language, their descriptive and explanatory power depends on the adequacy of that (meta-)language. For example, claims that in many societies people are guided in their ways of speaking by principles like "don't impose" or "be relevant" depend on the English words ilnpose and relevant, which have no equivalents in other languages. To say that speakers of those other languages are deeply concerned about some values which - "as it happens" - can only be formulated in English means to give English a curiously privileged position in humankind's mental world. (To quote my colleague Cliff Goddard's ironic comment on such methodological practices, "thank God for English!".) The theory of cultural scripts rejects those practices, and seeks to formulate norms, values and principles of language use in words which, unlike impose or relevant, have equivalents in all other languages, that is words which can be said to stand for universal human concepts. These "universal words" (or word-like elements) are the same words in which semantically encoded meanings can also be explicated. The terminological distinction between "explications" and "cultural scripts" can help clarify the boundary between those aspects of language use which are semantically encoded and those which are not. Not everything is semantically encoded but everything can be described in universal human concepts. For example, "pragmatic" meanings encoded in "diminutives" like doggie and birdie, in interjections like wow! or gee!, or in tautologies like }var is wa~ can be explicated in those concepts; and cultural norms which are not encoded in any particular expressions can be articulated in those concepts as a culture's "cultural scripts". The main point is that neither conceptual artefacts encoded in the English language nor Anglo "cultural scripts" can be legitimately used as analytical tools for the interpretation of language use throughout the world. The use of the Natural Semantic Metalanguage can free us from such ethnocentrism and enable us to capture, in every case, the cultural insider's point of view, while at the same time making that point of view intelligible to the outsider. Both explications and "cultural scripts" seek to articulate, in a rigorous yet intelligible way, shared cultural representations. To quote from Enfield (2000):

xx

Introduction to the second edition The very idea of the English language is a cultural and metalinguistic artefact. So when we work with categories like English or Lao, this must be kept in mind. And the same goes for 'Anglo' or 'Lao' culture. What we are really talking about is some set of cultural representations - private representations which are carried, assumed-to-be-carried and assumed-to-be-assumed-to-becarried - among some carrier group.... if we really want to characterize what cultural representations unite groups of people, we had better start with the cultural representation in question, and ask what group of people are united by their sharing it, rather than starting with some group ... and asking what cultural representations are shared among members. (Enfield 2000: 57)

To "start with the cultural representations" we need to have a framework within which such representations can be identified - "from a native speaker's point of view" (cf. Geertz 1984) and yet through concepts accessible to cultural outsiders as well. The NSM theory, with its set of empirically established universal human concepts, provides such a framework. The search for universals is of course important, but it must go in the right direction. This book is based on the assumption that what is universal are the conceptual building blocks which we find in a tangible form in all languages, and not some putative principles of "natural logic", "conversation" or "politeness". It is important to point out to the reader of this second edition that the NSM semantic theory has developed considerably since the publication of the first edition - largely as a result of the theoretical as well as empirical input from Cliff Goddard. Goddard himself has commented on this development as follows: In the thirty years since the publication of Semantic Primitives in 1972, the mode of operation of the NSM research program has been akin to that of socalled "normal science" (cf. Kuhn 1970; Lakatos 1970, 1978). There has been internal consensus on the hard core of fundamental goals and assumptions the quest to identify the indefinable semantic elements in natural language and to use these as a basis for a "self-explanatory" system of meaning representation. On the other hand, a number of auxiliary hypotheses have been revised or replaced in the light of empirical work and the "model NSM" has passed through a series of progressive refinements and expansions. (Goddard 2002a, vol. 2: 314)

The expansions mentioned in the last sentence include the development of the theory of cultural scripts and the new field of "ethnopragmatics" (Goddard 2002b, in press a, b, c, and forthcoming), and the "refinements" - the enlarged set of the universal semantic primes (roughly, the double of that outlined in the first edition) and the construction of a more or less complete model of universal grammar, presented in our re-

Introduction to the second edition

XXI

cent edited book Meaning and Universal Grammar (Goddard and Wierzbicka 2002). If the present book were to be rewritten in the light of these developments, the formulae included in it would be refined. Since the new expanded set of universal human concepts constitutes the major outcome of the NSM research over the last decade or so and may be of interest to the reader of this second edition, I will include the current set in Chapter 1, alongside with the 1991 version. The doubling of the inventory of universal semantic primes must of course be seen not only as a "refinement" but also as a major development. In his insightful and generally very positive review of the 1991 Cross-Cultural Pragmatics James Matisoff (1996) has expressed some scepticism with regard to the explanatory power of an inventory of only 27 elements, as it was at the time. The doubling of this set in more recent NSM work vindicates Matisoff's scepticism. At the same time, I would like to point out that most of the "new" post-1991 set of primes belong to semantic domains which are less relevant to cross-cultural pragmatics than the old ones, and also, that the actual analyses in the 1991 edition of this book rely on more than 27 elements, although those additional elements were regarded at the time as semantic "molecules" rather than as semantic "atoms". Among the new primes which are relevant to many "cultural scripts", the most important no doubt is TRUE (cf. e. g. Wierzbicka in press). In any case, I would encourage all those interested in adopting the NSM framework for their own work on cross-cultural pragmatics or indeed on any other aspect of language and culture to consult also our 2002 edited book Meaning and Universal Grammar (Goddard and Wierzbicka 2002). In an article entitled "Cross-Cultural Literacy: A National Priority", Luce and Smith (1987) wrote: "Cross-cultural literacy" means that our citizenry knows how culture influences perceptions and actions. It no longer accepts cultural stereotypes and cliches about other nations. It recognizes that American culture takes its place beside other national cultures as one contruct within the spectrum of human societies. Most importantly, cross-cultural literacy requires that Americans know how to read the cultural cues of other nations and decode their meaning. Within this decade, cross-cultural communications skill will become increasingly an indispensable tool for every citizen. Cross-cultural literacy must be a priority on our national agenda as we approach the end of the decade of the 1980s and near the 21 st century. (Luce and Smith 1987: 4)

If "cross-cultural literacy" was justly seen as a priority in 1987, it is all the more so in the post-September-l1 th world of 2003 - and not only as

XXll

Introduction to the second edition

a priority for the national agenda of the United States but also in Europe and in many other parts of the world. Cultural stereotypes and cliches are indeed no longer acceptable, but a wide-spread cross-cultural literacy n1ust be seen as more important a goal than ever. The NSM semantic theory based on universal human concepts offers a framework within which the "cultural scripts" of different nations and different "linguacultures" can be effectively articulated, taught and explained. Canberra, January 2003 Anna Wierzbicka

References [The articles on "cultural scripts" are marked with an asterisk] Ameka, Felix 1994 Areal conversational routines and cross-cultural communication in a multilingual society. In: H. Piirschel (ed.), Intercultural Communication, 441-469. Bern: Peter Lang. 1992 Interjections: The universal yet neglected part of speech. Journal of Pragmatics 18 (2/3): 101-118. Attinasi, John and Paul Friedrich 1995 Dialogic breakthrough: Catalysis and synthesis in life-changing dialogue. In: Bruce Mannheim and Dennis Tedlock (eds.), The Dialogic Emergence of Culture, 33-53. Urbana: University of Illinois Press. Besemeres, Mary 1998 Language and self in cross-cultural autobiography: Eva Hoffman's "Lost in Translation". Canadian Slavonic Papers 40: 3-4. 2002 Translating one's self: Language and seljhood in crosscultural autobiography. Oxford: Peter Lang. Blum-Kulka, Shoshana, J. House, and Gabriele Kasper (eds.) 1989 Cross-cultural Pragn1atics: Requests and Apologies. Norwood, N J: Ablex. Brown, Penelope and Stephen Levinson 1978 Universals in language usage: politeness phenomena. In: Esther Goody (ed.), Questions and Politeness: Strategies in Social Interaction, 56- 31 O. Cambridge: Cambridge University Press. 1978 Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Carbaugh, Donal 1988 Talking American. Cultural Discourses on DONOHUE. Norwood, New Jersey: Ablex Publishing Corporation.

Introduction to the second edition

XXlll

Clyne, Michael 1994 Intercultural Communication at Work: Cultural values in discourse. Cambridge: Cambridge University Press. Dalziell, Rosamund (ed.) 2002 Selves Crossing Cultures: Autobiography and Globalization. Melbourne: Australian Scholarly Publishing. D'Andrade, Roy and Claudia Strauss 1992 Hun1an Motives and Cultural Models. Cambridge: Cambridge University Press. Davis, Wayne A 1998 Implicature: Intention, Convention, and Principle in the Failure of Gricean Theory. Cambridge: Cambridge University Press. Darnell, Regna 1994 Comments on Wolf's "Perilous ideas: Race, culture and people." Current Anthropology 35.1: 7-8. Enfield, Nick J. 2000 The theory of cultural logic: how individuals combine social intelligence with semiotics to create and maintain cultural meaning. Cultural Dynamics. 12.1: 35-64. Geertz, Clifford From the Native's Point of View: On the Nature of Anthropologi1984 cal Understanding. In: Richard A. Shweder and Robert A. LeVine (eds.), Culture Theory: Essays on Mind, Self, and Emotion. 123136. Cambridge: Cambridge University Press. Goddard, Cliff 1992 Traditional Yankunytjatjara ways of speaking: A semantic perspective. Australian Journal of Linguistics. 12.1: 93 - 122. 1996 The social emotions of Malay (Bahasa Melayu). Ethos. 24.3: 426-464. *1997 Cultural values and cultural scripts of Malay (Bahasa Melayu). Journal of Praglnatics 27.2: 183-201. *2000 Cultural scripts and communicative style in Malay (Bahasa Melayu). Anthropological Linguistics 42.1: 81 -106. 2001 Cultural semantics and intercultural communication. In: D. Killick, M. Perry and A. Phipps (eds.), Poetics and Praxis of Languages and Intercultural Con1munication, 33 -44. (Proceedings of the conference at Leeds Metropolitan University December 1999). Glasgow, Scotland: University of Glasgow French and German Publications. 2002a The on-going development of the NSM research program. In: C. Goddard and A. Wierzbicka (eds.), Meaning and Universal Grammar: Theory and En1pirical Findings. vol. 2: 301 - 322. Amsterdam: John Benjamins.

XXIV

Introduction to the second edition

2002b

Ethnosyntax, ethnopragmatics, sign-functions, and culture. In: N. J. Enfield (ed.), Ethnosyntax: Explorations in Grammar and Culture, 52-73. Oxford: Oxford University Press. 2002c Overcoming terminological ethnocentrism. lIAS Newsletter 27, 28. Leiden, The Netherlands: International Institute for Asian Studies. in press a An ethnopragmatic perspective on active metaphors. Journal 0.[ Pragmatics. *in press b "Cultural Scripts": A new medium for ethnopragmatic instruction. In: Michael Achard and Susanne Niemeier (eds.), Cognitive Linguistics, Second Language Acquisition, and Foreign Language Teaching. Berlin/ New York: Mouton de Gruyter. *in press c Directive speech-acts in Malay: An ethnopragmatic perspective. In: Christine Beal (ed.), Les Cahiers de Praxematique. *Forthc. Introduction to Cliff Goddard (ed.), Ethnopragn1atics: Understanding discourse in cultural context. Berlin/New York: Mouton de Gruyter. Goddard, Cliff and Anna Wierzbicka 1997 Discourse and Culture. In: Teun A. van Dijk (ed.), Discourse as Social Interaction, 231 - 257. vol. II of Discourse: A Multidisciplinary Introduction. London: Sage Publications. Goddard, Cliff and Anna Wierzbicka (eds.) 2002 Meaning and Universal Grammar: Theory and Empirical Findings. 2 vols. Amsterdam: John Benjamins. Gordon, David and George Lakoff 1975 Conversational Postulates. In: Peter Cole and Jerry L. Morgan (eds.), 83-106. Syntax and Semantics 3: Speech Acts. New York: Academic Press. Grice, H. P. 1975 Logic and conversation. In: Peter Cole and Jerry L. Morgan (eds.), Syntax and Semantics 3: Speech Acts, 41-58. New York: Academic Press. Harkins, Jean 1994 Bridging TIvo Worlds: Aboriginal English and cross-cultural understanding. Brisbane: University of Queensland Press. Harnish R. M. 1976 Logical form and implicature. In: T. G. Bever, J. J. Katz, and J. Langendoen (eds.), An Integrated Theory of Linguistic Ability, 313392. New York: Thomas Crowell. Hoffman, Eva 1989 Lost in Translation: A Life in a nelV language. London: Heinemann. Holland, Dorothy and Naomi Quinn 1987 Cultural Models in Language and Thought. Cambridge: Cambridge University Press.

Introduction to the second edition

xxv

Ide, Sachiko 1989 Formal forms and discernment. Multilingua 8: 223-248. Kasper, Gabriele and Shoshana Blum-Kulka 1993 Interlanguage Pragmatics. Oxford: Oxford University Press. Katriel, Tamar 1986 Talking Straight: Dugri Speech in Israeli Sabra Culture. Cambridge: Cambridge University Press. Kim, Young Yun 2001 Becoming Intercultural: An Integrative Theory of Communication and Cross-Cultural Adaptation. Thousand Oaks, CA: Sage Publications. Kochman, Thomas 1981 Black and White Styles in Con.flict. Chicago: University of Chicago Press. Kramsch, Claire 1993 Context and Culture in Language Teaching. New York: Oxford University Press. Kuhn, Thomas 1970 The Structure of Scientific Revolutions. 2nd ed. Chicago: Chicago University Press. Lakatos, Imre 1970 Falsification and the methodology of scientific research programmes. In: I. Lakatos and A. Musgrave (eds.), Criticism and the Grolvth of Knowledge, 91-196. Cambridge: Cambridge University Press. [Republished in J. Worrall and G. Currie (eds.), Imre Lakatos: Philosophical Papers, vol. 1: 8- 101. Cambridge: Cambridge University Press.] 1978 Introduction: Science and pseudoscience. In: J. Worrall and G. Currie (eds.), Imre Lakatos: Philosophical Papers, vol. 1: 1-7. Calnbridge: Cambridge University Press. Leech, Geoffrey 1983 Principles of Pragmatics. London: Longn1an. Levinson, Stephen 1983 Pragnlatics. Cambridge: Cambridge University Press. Luce, Louise Fiber and Elise C. Smith 1987 Cross-cultural literacy: a national priority. In: Luce and Smith (eds.), Toward Internationalism, 3-10. Cambridge: Newbury House Publishers. Lutz, Catherine 1988 Unnatural Emotions: Everyday Sentinlents on a Micronesian Atoll and their Challenge to Western Theory. Chicago: University of Chicago Press. Matisoff, James 1979 Blessings, Curses, Hopes, and Fears: Psycho-Ostensive Expressions in Yiddish. Philadelphia: Institute for the Study of Human Issues.

XXVI

Introduction to the second edition

Matsumoto, Yoshiko 1988 Reexamination of the universality of face: politeness phenomena in Japanese. Journal of Pragmatics 12: 403-426. Shweder, Richard 1991 Thinking Through Cultures: Expeditions in Cultural Psychology. Cambridge, MA: Harvard University Press. Searle, John Indirect speech acts. In: Peter Cole and Jerry Morgan (eds.), Syntax 1975 and Semantics 3: Speech acts, 59-82. New York: Academic Press. Sohn, Ho-min 1983 Intercultural communication in cognitive values: Americans and Koreans. Language and Linguistics (Seoul) 9: 93 - 136. Sperber, Dan and Deirdre Wilson 1986 Relevance: Conzmunication and cognition. Cambridge, MA: Harvard University Press. Strauss, Claudia and Naomi Quinn 1997 A Cognitive Theory of Cultural Meaning. Cambridge: Cambridge University Press. Tannen, Deborah 1986 That's Not What I Meant! How conversational style nzakes or breaks relationships. New York: Ballantine. Wallenstein, Immanuel 1994 Comments on Wolf's 'Perilous ideas: race, culture and people'. Current Anthropology. 35.1: 9-10. Wierzbicka, Anna 1985 Different cultures, different languages, different speech acts: English vs Polish. Journal of Pragmatics 9: 145-178. Boys Will Be Boys: "Radical Semantics" vs. "Radical Pragmatics." 1987 Language 63: 95-114. "Cultural Scripts": a new approach to the study of cross-cultural *1994a communication. In: Martin Piitz (ed.), Language Contact and Language Conflict. 69-88. Amsterdam/Philadelphia: John Benjamins. *1994b Emotion, language and 'cultural scripts'. In: Shinobu Kitayama and Hazel Rose Markus (eds.), Enlotion and Culture: Enlpirical studies oj'mutual influence. 130-198. Washington D.C.: American Psychological Association. *1996a Contrastive sociolinguistics and the theory of "cultural scripts". In: Marlis Hellinger and Ulrich Ammon (eds.), Contrastive Sociolinguistics, 313 - 344. Berlin/New York: Mouton de Gruyter. *1996b Japanese cultural scripts: cultural psychology and "cultural grammar". Ethos 24.3: 527 - 555. 1997a Understanding Cultures through their Key Words: English, Russian, Polish, German, Japanese. New York: Oxford University Press.

Introduction to the second edition

1997b

*1998 *2002

XXVll

The double life of a bilingual: a cross-cultural perspective. In: Michael Bond (ed.), Working at the Interface of Cultures: Eighteen Lives in Social Science, 113-25. London: Routledge. Gern1an cultural scripts: public signs as a key to social attitudes and cultural values. Discourse and Society 9.2: 241 - 282. Australian cultural scripts - bloody revisited. Journal of Pragn1atics 34.9: 1167-1209. Russian cultural scripts. Ethos. The English Language: Meaning, History and Culture.

*in press Forthc. Wolf, Eric R. 1994 Perilous ideas: race, culture and people. Current Anthropology 35.1: 1-7.

AcknowledgelTIents

I would like to express my gratitude to colleagues who at different times have discussed with me problems explored in this book, and who have offered comments on the earlier versions of some of the chapters, and in particular, Felix Ameka, Andrzej Bogus}awski, Cliff Goddard, Jean Harkins, Igor Mel'cuk, and Tim Shopen. I am particularly grateful to Jean Harkins, who worked as my research assistant, and who made innumerable valuable suggestions, as well as providing expert and thorough editorial assistance. I would also like to thank Mrs. Ellalene Seymour for her expert and patient typing of the successive drafts. In addition to several completely new chapters and other new work, this volume includes also some chapters which had their starting point in some articles published earlier. Although the contents of these chapters is largely new (in length alone, none of the older articles comes to much more than one third of the corresponding chapter), I would like to thank the publishers of those earlier articles for their permission to make use of the material contained in them: 1985: "A semantic metalanguage for a cross-cultural comparison of speech acts and speech genres", Language in Society 14:491-514; "Different cultures, different languages, different speech acts: Polish vs. English", Journal of Pragmatics 9:145-178. 1986: "A semantic metalanguage for the description and comparison of illocutionary meanings", Journal of Pragmatics 10:67-107; "Italian reduplication: cross-cultural pragmatics and illocutionary semantics", Linguistics 24:287-315; "Precision in vagueness: the semantics of English 'approximatives"', Journal of Pragmatics 10:597-614; "The semantics of quantitative particles in Polish and in English", in: Boguslawski - Bojar, 175-189. 1987: "Boys will be boys: 'radical semantics' vs. 'radical pragmatics"', Language 63:95-114 (by permission of the Linguistic Society of America). 1990: "The semantics of interjections", Journal of Pragmatics 14 (special issue on interjections, ed. by F. Ameka). Canberra, September 1990

Anna Wierzbicka

Contents

Introduction to the second edition Acknowledgements Chapter 1 Introduction: semantics and pragmatics 1. Language as a tool of human interaction 2. Different cultures and different modes of interaction 3. Pragmatics - the study of human interaction 4. The natural semantic metalanguage 5. The need for a universal perspective on meaning 6. The uniqueness of every linguistic system 7. The problem of polysemy 8. Semantic equivalence vs. pragmatic equivalence 9. Universal grammatical patterns 10. Semantics versus pragmatics: different approaches 10.1. 'Complementarism' 10.2. 'Pragmaticism' 10.3. 'Semanticism' 10.4. A fourth approach: two pragmatics 11. Description of contents

v XXIX

1 1 2 5 6 9 10 11 12 14 15 16 17 18 18 20

Chapter 2 Different cultures, different languages, different speech acts

25

1. Preliminary examples and discussion 2. Interpretive hypothesis 3. Case studies 3.1. Advice 3.2. Requests 3.3. Tags 3.4. Opinions 3.5. Exclamations 4. Cultural values reflected in speech acts 4.1. Lexical evidence 4.2. Objectivism as a cultural value 4.3. Cordiality as a cultural value

27 30 31 31 32 37 41 45 47 47 49 50

xxxii

Contents

4.4. Courtesy as a ~ultural value 5. Theoretical implications 6. Practical implications

56 59 64

Chapter 3 Cross-cultural pragmatics and different cultural values

67

' Self-assertion' 1.1. 'Self-assertion' in Japanese and in English 1.2. 'Self-assertion' in black and white American English 1.3. Spontaneity, autonomy, and tum-taking: English vs. Japanese 1.4. 'Spontaneous self-assertion' vs. 'regulated self-assertion' : black English vs. white English vs. Japanese 1.5. 'Self-assertion' as personal display: black English vs. white English 1.6. 'Self-assertion' and 'good interpersonal relations' 'Directness' 2.1. American culture vs. Israeli culture 2.2. 'Indirectness' in Japanese 2.3. Greek culture and American culture 2.4. 'Indirectness' and 'dissimulation' in Javanese Further illustrations: same labels, different values 3.1. 'Intimacy' 3.2. 'Closeness' 3.3. 'Informality' 3.4. 'Harmony' 3.5. 'Sincerity' Different attitudes to emotions 4.1. Polish culture 4.2. Jewish culture 4.3. American black culture 4.4. Japanese culture 4.5. Javanese culture Conclusion

72 72 78

1.

2.

3.

4.

5.

Chapter 4 Describing conversational routines 1. Conversational analysis: linguistic or non-linguistic pragmatics?

80

82 84 85 88 89 93 95 100 104 105 108 111 113 115 121 121 122 123 126 128 129 131 131

Contents

2.

'Compliment response' routines 2.1. Upgrades 2.2. Contrastive opposites 2.3. Scaled-down agreements 2.4. Downgrades 2.5. Reassignment of praise 2.6. Returns 3. 'Compliment responses' in different cultures 4. Conclusion Chapter 5 Speech acts and speech genres across languages and cultures 1. A framework for analysing a culture's 'forms of talk' 1.1. The importance of folk labels 1.2. Two approaches 1.3. Some examples: English vs. Japanese 1.4. Another example: English vs. Walmatjari 1.5. The elimination of vicious circles 1.6. Evidence for the proposed formulae 1.7. The first-person format 1.8. The problem of other minds 2. Some Australian speech-act verbs 2.1.

2.2. 2.3. 2.4. 2.5.

Chiack (chyack) Yarn Shout Doh Whinge

XXX111

136 138 139 140 140 142 142 143 147

149 149 150 150 152 158 161 161 162 164 165 165 170 173

177 180

3. Some examples of complex speech genres 3.1. The black English dozens 3.2. The Hebrew 'dugri talk' 3.3. The Polish kawaf 3.4. The Polish podanie 4. Conclusion

185 188

Chapter 6 The semantics of illocutionary forces

197

1. Are illocutionary forces indeterminate? 1.1. Illocutionary forces as bundles of components 1.2. Illustration: the discrete and determinate character of 'whimperatives'

183 183

192

196

197 199 202

XXXIV

2.

3. 4.

5.

6. 7.

8. 9.

Contents

1.3. Syntax and illocutionary force More whimperative constructions 2.1. Why don't you do X (tomorrow)? 2.2. Why do X? 2.3. How about X? Additional remarks on the explication of illocutionary forces Selected conversational strategies 4.1. Tell you what, S! 4.2. Do you know, S? 4.3. Don't tell me S! 4.4. How many times have I told you (not) to do X! 4.5. Who's talking about doing X? Tag questions 5.1. Tags with declarative sentences 5.2. Tags with imperative sentences 5.3. Why can't you (do X)! 5.4. OK? Personal abuse or praise: You X! Illocutionary forces of grammatical and other categories 7.1. Modal verbs 7.2. Mental verbs 7.3. Particles and conjunctions 7.4. Interjections 7.5. Fixed expressions 7.6. Intonation Comparing illocutionary forces across languages Conclusion

Chapter 7 Italian reduplication: its meaning and it~ cultural significance 1. 2. 3. 4. 5. 6. 7. 8.

Italian reduplication: preliminary discussion Discourse and illocutionary grammar The illocutionary force of clausal repetition The illocutionary force of Italian reduplication Clausal repetition as a means of 'intensification' The absolute superlative in Italian and in English Illocutionary grammar and cultural style Conclusion

207 211 211 213 215 218 219 219 220 222 223 224 224 224 227 229 231 232 235 235 238 240 243 245 247 248 252

255 255 258 260 263 268 270 276 282

Contents

Chapter 8 Interjections across cultures 1. Preliminary discussion 1.1. Interjections: physis and thesis ('nature' and 'convention') 1.2. Defining the concept of 'interjection' 1.3. Types of interjections 2. Volitive interjections 2.1. Interjections directed at animals 2.2. Interjections directed at people 2.2.1. The 'I want silence' group 2.2.2. The 'I don't want you in this place' group 2.2.3. The 'I want you to jump' group 2.2.4. The 'urging' group 2.2.5. The 'communication over distance' group 3. Emotive interjections 3.1. Interjections of 'disgust' and similar feelings 3.1.1. The Polish fu and the English yuk 3.1.2. The Russian fu 3.1.3. The Polish fe 3.1.4. The Yiddish feh 3.1.5. The Polish tfu and the Russian t'fu 3.1.6. 'Disgust' and bodily gestures 3.1.7. 'Disgust' and sound symbolism 3.2. 'General purpose' interjections 3.2.1. The Polish oj 3.2.2. The Russian oj 3.2.3. Ochs and achs 4. Cognitive interjections 4.1. The Polish aha and Russian aga 4.2. The Polish oho 4.3. The Polish 0 4.4. The English oh-oh 4.5. The Russian ogo 5. Conclusion Chapter 9 Particles and illocutionary meanings 1. English quantitative particles 1.1. Non-approximative particles: only, merely and just

xxxv 285 285 285 290 291 292 292 293 293 296 298 298 300 302 302 302 304 306 308 310 313 315 317 318 322 323 326 326 331 333 334 334 337 341 345 346

XXXVI

Contents

1.1.1. Only 1.1.2. Merely 1.1.3. lust 1.2. English approximative particles 1.2.1. Around and about 1.2.2. Approximately 1.2.3. Roughly 1.2.4. Almost and nearly English temporal particles Polish temporal particles 3.1. lui and jeszcze 3.2. Dopiero Polish quantitative particles 4.1. Non-approximative particles 4.1.1. Tylko 4.1.2. Ai 4.1.3. Zaledwie 4.1.4. Ledwie 4.2. Polish approximative particles 4.2.1. o malo nie 4.2.2. Niemal and prawie 4.2.3. Blisko Conclusion

346 348 350 354 355 358 360 361 367 371 371 376 379 379 379 380 381 382 384 384 385 388 389

Chapter 10 Boys will be boys: even 'truisms' are culture-specific

391

2. 3.

4.

5.

1. The meaning of tautologies 1.1. Gricean maxims: universal or language-specific? 1.2. Problems in interpreting implicatures 1.3. Context as an excuse for analytical failure 2. English nominal tautologies: semantic representations 2.1. 'Realism' in human affairs 2.2. Tolerance for human nature 2.3. Tolerance at 'special times' 2.4. The limits of tolerance 2.5. Seeing through superficial differences 2.6. Recognising an irreducible difference 2.7. Tautologies of value 2.8. Tautologies of obligation

391 392 397 400 403 404 405 408 410 411 413 414 419

Contents

3. Some comparisons from Chinese and Japanese 3.1. Chinese concessive tautologies 3.2. 'Irreducible difference', Chinese style 3.3. Chinese tautologies of unreserved praise 3.4. Japanese tautologies of 'a matter of course' 3.5. Japanese tautologies of irrelevance 4. Verbal tautologies 4.1. Future events 4.2. The immutability of the past 5. Is there a semantic invariant? 6. The deceptive form of English tautological constructions 7. The culture-specific content of tautological patterns 8. Conclusion

XXXVll

423 423 426 427 429 430 431 431 434 439 444 446 448

Chapter 11 Conclusion: semantics as a key to cross-cultural pragmatics

453

Notes

457

Bibliography

461

Subject and name index

487

Index of words and phrases

497

Chapter 1

Introduction: semantics and praglllatics

The fate of the earth depends on cross-cultural communication. Deborah Tannen (1986:30)

1. Language as a tool of human interaction This book is devoted to the study of language as a tool of human interaction. It investigates various kinds of meanings which can be conveyed in language (not in one language, but in different languages of the world) - meanings which involve the interaction between the speaker and the hearer. It could be argued, of course, that all meanings involve interaction between the speaker and the hearer: whether we talk about colours, animals, children, love, the fate of the universe, or even pure mathematics, we use language as a tool of social interaction. In some sense this is true. Nonetheless, there are words which involve directly the concepts of 'I' and 'you', and interaction between 'I' and 'you', and there are others which do not. Similarly, there are grammatical categories, and grammatical constructions, which involve these concepts directly, and there are others which do not. For example, the English words blue and yellow make no reference to the speaker, the addressee, or the relationship between them; on the other hand, words such as darling, bastard, already, yuk, thanks, or goodbye do. Similarly, grammatical categories such as singular and plural number (dog vs. dogs) or masculine and feminine gender (for example, la lille 'girl' vs. Ie garfon 'boy' in French) do not involve the speaker, the addressee, or the relationship between them; whereas categories such as diminutives (doggie vs. dog), augmentatives (for example, problemon, problemazo 'big problem' vs. problema 'problem' in Spanish) or honorifics (for example, otaku 'esteemed house' vs. ie 'house' in Japanese) do.

2

Introduction: semantics and pragmatics

At the level of grammatical constructions, the choice between an (a) and a so-called 'whimperative ' (b):

imper~tive

a. Sign this. b. Would you sign this. involves directly the relationship between the speaker and the addressee, whereas the choice between a relative clause (a) and a participial construction (b) does not: a. The boy who was sitting in front ... b. The boy sitting in front ... This book, then, deals with words, categories, constructions, and linguistic routines which involve interpersonal interaction, that is, which involve, more or less directly, you and me. It is a book about you and me, and about the different modes of interaction between you and me, and, more particularly, between me and you (that is, between the speaker and the hearer); and about cultural values and cultural norms which shape these different modes of interaction.

2. Different cultures and different modes of interaction There are many different possible modes of interaction between you and me, between me and you. They depend partly on what you and I feel and want at any particular time; but they depend also on who you and I are - both as individuals and as members of particular social, cultural, and ethnic groups. For example, if you and I are Japanese our interaction will be different than it would be if we were both Americans or Russians. And if we were both Americans, the prevailing modes of our interaction would probably depend on whether we were white or black, Jewish or non-Jewish, and so on. Consider, for example, a typical Australian utterance such as Silly old bugger!, recently used in public, in front of the television cameras, by the Australian Prime Minister, Mr. Bob Hawke, during a meet-the-public session, when he was goaded by an old-age pensioner about high parliamentary salaries. One has to know a good deal about Australian culture and society (cf. Chapter 5) to interpret correctly the communicative value

Different cultures and different modes of interaction

3

of this remark. In particular, one has to understand the link between the common use of 'b-words' such as bugger, bastard, and bloody (cf. Baker 1966:201) and the core Australian values of 'roughness', 'antisentimentality', 'sincerity' and so on (cf. Renwick 1980; Wierzbicka, to appear, chap. 11). Similarly, one has to appreciate the core Australian values of 'mateship', 'toughness', 'anti-verbosity', 'anti-emotionality' and so on, to appreciate the attitudes expressed in characteristic Australian greeting e'xchanges (Bowles 1986:37; cf. Chapter 4): G'day, mate, owyagowin? Nobbad. Owsyerself? (Or: earn complain.)

In some cases, culture-specific modes of interaction have their own folk names (cf. Chapter 5). This is the case with Black English speech events such as 'rapping' or 'sounding', which can be illustrated with the following characteristic utterances (Kochman 1972): Baby, you're fine enough to make me spend my rent money. (A 'rap' from a man to a woman.) Baby, I sho' dig your mellow action. (Another e/xample of 'rapping' to a woman.) Yo mama is so bowlegged, she looks like a bite out of a donut. (A 'sound' from a schoolboy to another schoolboy.)

But this is not necessarily always the case. Consider, for example, the following conversation, from a short story by the Jewish-American writer, Bernard Malamud: [When he knocked, the door was opened by a thin, asthmatic, grey-haired woman, in felt slippers.] 'Yes?' she said, expecting nothing. She listened without listening . He could have sworn he had seen her, too, before but knew it was an illusion. 'Salzman-does he live here? Pinye Salzman,' he said, 'the matchmaker?' She stared at him a long minute. 'Of course.' He felt embarrassed. 'Is he in?' 'No.' Her mouth, though left open, offered nothing more. 'The matter is urgent. Can you tell me where his office is?' 'In the air.' She pointed upward. 'You mean he has no office?' Leo asked. 'In his socks.' (... )

4

Introduction: semantics and pragmatics 'Where is he?' he insisted. 'I've got to see your husband.' At length she answered, 'So who knows where he is? Every time he thinks a new thought he runs to a different place. Go home, he will find you.' 'Tell him Leo Finkle.' She gave no sign she had heard.

(Malamud 1958:210-211) The story is written in English, and it includes no unusual or nonstandard words, but the ways of speaking and of interacting reflected here are those characteristic of Yiddish, not of (mainstream) American English. Note in particular the use of No and Of course, the bare imperatives Tell him and Go home, the rhetorical question Who knows?, the irony, the wry humour, the bluntness and the gruffness (for discussion, see Chapter 3 below). And one last group of examples - English translations of typical Yiddish blessings and curses (Matisoff 1979): A lament to you, are you crazy or just feeble-minded? Oh, you should be healthy, what a mess you've made here! May he live - but not long. A black year on her, all day long she chewed my ear off with trivia! My wife - must she live? - gave it away to him for nothing. His son-in-law - may he grow like an onion with his head in the earth - sold it to me. Maybe my mother-in-law is going to visit us the day after tomorrow, may the evil hour not come! All such utterances encode important interactional meanings. This book explores such meanings, and their cultural significance, and offers a framework within which they can be described in an illuminating and rigorous way.

Pragmatics -

3. Pragmatics -

the study of human interaction

5

the study of human interaction

The discipline studying linguistic interaction between 'I' and 'you' is called pragmatics, and the present book is a work in pragmatics. It differs, however, from other works in pragmatics in so far as it is also a work in semantics - not in the sense that some chapters of the book are devoted to pragmatics, and others, to semantics, but in the sense that pragmatics is approached here as a part, or an aspect, of semantics; and this is the major theoretical novum of the present approach. I will explain what I mean by means of an example. Let us consider first the words question and ask, sentences (questions) such as What time is it?, so-called 'indirect questions' such as I don't know what time it is, and so-called pre-questions, such as Do you know what time it is? Traditionally, the word question would be described in a dictionary, the sentence type illustrated by What time is it? would be discussed in a chapter of a grammar devoted to 'interrogative constructions', and the type illustrated by I don't know what time it is in a chapter of a grammar devoted to 'indirect questions', whereas expressions such as Do you know, Did you know or You know would be discussed (if at all) in some works on 'discourse strategies', 'discourse markers', or on 'organisation of conversation'. Thus, these different descriptions of words, grammatical constructions, and 'pragmatic devices' would be discussed in totally different types of works, and in totally different frameworks - as if they had nothing in common whatsoever. In fact, however, they are of course closely related. They all involve crucially the concepts of 'knowing', 'not knowing', and 'saying'; and they all involve the concepts of 'you' and 'I'. They all involve some semantic components such as 'I don't know' or 'you don't know', 'I say' and 'I want you to say', 'I want to know' or 'I want you to know'. All these are 'interactional' (or 'pragmatic') meanings. To understand human interaction we have to understand 'interactional' meanings expressed in speech; and we have to have suitable analytical tools for identifying and describing such meanings. In the past, analytical tools of this kind were sorely lacking. Quite apart from the compartmentalisation of linguistic descriptions, which made it impossible to even raise the question of the semantics of human interaction, there were simply no adequate tools for describing any kind of interactional meanings. Standard lexicographic descriptions of

6

Introduction: semantics and pragmatics

words such as question or ask illustrate rather well the general level of precision and clarity prevailing in the description of such meanings. For example, Longman's ambitious Dictionary of the English Language (LDOTEL 1984), which, according to its jacket blurb, "provides unrivalled access to contemporary English and the way it is used", offers us the following definitions: a command or interrogative expression used to elicit information or a response interrogative - an interrogative utterance, a question command the act of commanding response an act of responding to write or speak in reply (to) respond (to) reply to respond in words or writing

question

All such explanations of interactional meanings (like, incidentally, those of any other meanings) are, clearly, totally circular. But it is an illusion to think that circularity of this kind is exclusively a feature of dictionaries (which are, after all, modest practical reference works), whereas scholarly literature on language use is somehow different. It is not different. It relies on various more or less technical-sounding labels (such as, for example, 'face', 'distance', 'indirectness', 'solidarity', 'intimacy', 'formality', and so on), which are never defined; or if they are defined, they are defined in ways which prove, sooner or later, to be just as circular and obscure as traditional dictionary definitions. Furthermore, they are defined in terms which are language-specific (usually, Englishspecific), and which provide no language-independent, universal perspective on the meanings expressed in linguistic interaction.

4. The natural semantic metalanguage To compare meanings one has to be able to state them. To state the meaning of a word, an expression or a construction, one needs a semantic metalanguage. To compare meanings expressed in different languages and different cultures, one needs a semantic metalanguage independent, in essence, of any particular language or culture - and yet accessible and open to interpretation through any language.

The natural semantic metalanguage

7

1 propose for this purpose a 'natural semantic metalanguage', based on a hypothetical system of universal semantic primitives, which my colleagues and 1 have developed over more than two decades (see, in particular, Boguslawski 1966, 1972, 1975, 1981a,b, 1989; Wierzbicka 1972, 1980, 1987, 1988, 1989a,b; Goddard 1989a,b); and this is the metalanguage employed in the present book. This means that 1 will try to state the meanings under consideration in terms of simple and intuitively understandable sentences in natural language. This, 1 believe, will ensure that the proposed semantic explications will be immediately verifiable and intuitively revealing. But the subset of natural language in which the explications are formulated is highly restricted, standardised, and to a large extent language-independent (that is, isomorphic to equivalent subsets of other natural languages). For this reason, the natural language used in the explications - a kind of highly reduced 'basic English ' - can be viewed as a formal semantic metalanguage. The metalanguage applied in the present work is, so to speak, carved out of natural language - any natural language. For practical reasons, the version of the metalanguage employed here is carved out of English, but it could be just as easily carved out of Russian, Latin, Japanese, or Swahili, because it is based, by and large, on what 1 believe to be the universal core of natural languages. For example, if 1 say in an explication: 'I want', 1 mean something that could be just as easily represented as 'ja xocu' (Russian) or 'ego volo' (Latin). The expression 'I want' is used here, therefore, not as part of the 'normal' English language, but as part of the English-based version of the universal semantic metalanguage. The metalanguage in question is a technical, artificial language, not a natural language; nonetheless, it is appropriate and illuminating, 1 think, to call it a 'natural semantic metalanguage', (cf. Goddard 1989a,b), because it is derived entirely from natural language and because it can be understood via natural language without any additional arbitrary signs and conventions. Arbitrary signs and conventions are not allowed in this metalanguage, because their meaning would have to be explained - and these explanations, in their turn, would not be intelligible unless they were couched in immediately understandable natural language. (On the other hand, it is allowed to use 'iconic' conventions, such as spatial arrangement of components, the use of separate lines for different chunks of meanings, and the like.)

8

Introduction: semantics and pragmatics

Addendum to Chapter 1 The 1991 version of the Natural Semantic Metalanguage included only 27 hypothetical semantic primes, although, as mentioned earlier, a number of auxiliary concepts, regarded at the time as semantic "molecules" rather than "atoms" were also used in the explications. The current, greatly expanded, set includes 60 or so elements. Both versions, the 1991 and the current one, are given below. * Universal semantic primes (1991 version) Pronouns I you someone something

Determiners this the same two all

Classifiers kind of part of

Adjectives good bad

Verbs want don't want say think know do happen

Modals can if/imagine

Place/Time place time after (before) above (under)

Linkers like because

Universal semantic primes (2003 version) Substantives: I, YOU, SOMEONE, PEOPLE, SOMETHING/THING, Determiners: THIS, THE SAME, OTHER Quantifiers: ONE, TWO, SOME, ALL, MUCH/MANY Evaluators: GOOD, BAD Descriptors: BIG, SMALL Mental predicates: THINK, KNOW, WANT, FEEL, SEE, HEAR Speech: SAY, WORDS, TRUE Actions, events and movement: DO, HAPPEN, MOVE Existence and possession: THERE IS, HAVE Life and death: LIVE, DIE Time: WHEN/TIME, NOW, BEFORE, AFTER, A LONG TIME, A

BODY

SHORT TIME, FOR SOME

TIME, MOMENT

Space:

WHERE/PLACE, HERE, ABOVE, BELOW, FAR, NEAR, SIDE, INSIDE, TOUCHING (CONTACT)

Logical concepts: NOT, MAYBE, CAN, BECAUSE, Intensifier, augmentor: VERY, MORE Taxonomy, partonomy: KIND OF, PART OF Similarity: LIKE

IF

* This paragraph replaces an outdated paragraph from the 1991 edition. This is the only paragraph in the text of the first edition replaced with a new version. The only other change in the 1991 text is the addition of the 2003 table of universal semantic primes, on the same page.

The need for a universal perspective on meaning

9

5. The need for a universal perspective on meaning It is impossible for a human being to study anything - be it cultures, language, animals, or stones - from a totally extra-cultural point of view. As scholars, we remain within a certain culture, and we are inevitably guided by certain principles and certain ideals which we know are not necessarily shared by the entire human race. We must also rely on certain initial concepts; we cannot start our inquiry in a complete conceptual vacuum. It is important, however, that as our inquiry proceeds, we try to distinguish what in our conceptual apparatus is determined by the specific features of the culture to which we happen to belong, and what can be, with some justification, regarded as simply human. Trying to explore both the universal and the culture-specific aspects of meaning, we should beware of using concepts provided by our own culture or by our own scholarly tradition as culture-free analytical tools (cf. Lutz 1985). As human beings, we cannot place ourselves outside all cultures. This does not mean, however, that if we want to study cultures other than our own all we can do is to describe them through the prism of our own culture, and therefore to distort them. We can find a point of view which is universal and culture-independent; but we must look for such a point of view not outside all human cultures, (because we cannot place ourselves outside them), but within our own culture, or within any other culture that we are intimately familiar with. To achieve this, we must learn to separate within a culture its idiosyncratic aspects from its universal aspects. We must learn to find 'human nature' within every particular culture. This is necessary not only for the purpose of studying 'human nature' but also for the purpose of studying the idiosyncractic aspects of any culture that we may be interested in. To study different cultures in their culture-specific features we need a universal perspective; and we need a culture-independent analytical framework. We can find such a framework in universal human concepts, that is in concepts which are inherent in any human language. If we proceed in this way, we can study any human culture without the danger of distorting it by applying to it a framework alien to it; and we can aim both at describing it 'truthfully', and at understanding it. We cannot understand a distant culture 'in its own terms' without understanding it at the same time in our own terms. What we need for

10

Introduction: semantics and pragmatics

real 'human understanding' is to find terms which would be both 'theirs' and 'ours'. We need to find shared terms; that is, universal concepts. I suggest that we can find such concepts in the 'universal alphabet of human thoughts' (Leibniz 1903 :430), that is, in the indefinable (Le. semantically simple) words and morphemes of natural language, (such as I, you, someone, something, this, think, say, want, or do), which can be found, it seems, in all the languages of the world.

6. The uniqueness of every linguistic system Every language is a self-contained system and, in a sense, no words or constructions of one language can have absolute equivalents in another. The idea that there might be some linguistic elements which are universal in the sense of having absolute equivalents in all the languages of the world is of course all the more fanciful. However, as soon as we abandon the notion of absolute equivalents and absolute universals, we are free to investigate the idea of partial equivalents and partial universals; and if the former notion is sterile and useless, the latter idea is fruitful and necessary. What I mean by 'partial universals' is this. Within a particular language, every element belongs to a unique network of elements, and occupies a particular place in a unique network of relationships. When we compare two, or more, languages we cannot expect to find identical networks of relationships. We can, nonetheless, expect to find certain correspondences. To put it differently, although every language has its own unique structure and its own unique lexicon (embodying unique semantic configurations), nonetheless there are certain areas of languages which can be regarded as mutually isomorphic (some examples are given in the sections which follow). It is this (limited) isomorphism in grammar and in the lexicon that gives sense to the notion of semantic universals. The metalanguage employed in the present book is based on such putative universals.

The problem of polysemy

11

7. The problem of polysemy The search for lexical universals may seem to be a purely empirical task: laborious, to be sure, but relatively straightforward. In fact, however, the presence or absence of a word for a given concept cannot be established by any mechanical, checklist method. The search is empirical, but it also has necessarily an analytical dimension. Above all, there is the problem of polysemy. For example, 1 have postulated 'you' and 'I' as universal semantic primitives, but what I mean by 'you' is 'you SG' ('thou'), rather than 'you PL' or 'you SGjPL'. Yet one doesn't have to look further than modem English to find a language which doesn't seem to have a word for 'thou'. To maintain the claim that 'thou' is a lexical universal we would have to posit polysemy for the word you: (1) 'you SG', (2) 'you PL'. Initially, this seems an unattractive solution, but I think there are good reasons for accepting it. Polysemy is a fact of life, and basic, everyday words are particularly likely to be polysemous (cf. Zipf 1949). For example, say is polysemous between its abstract sense, which ignores the physical medium of expression (for example What did he say in his letter?, The fool said in his heart: there is no God), and its more specific sense, which refers to oral speech only. Know is polysemous between the two senses which are distinguished in French as savoir and connaftre, and in German as wissen and kennen (cf. I know that this is not true vs. I know this man). It goes without saying that polysemy must never be postulated lightly, and that it has always to be justified on language-internal grounds; but to reject polysemy in a dogmatic and a priori fashion is just as foolish as to postulate it without justification. In the case of the English word you, I think its polysemy can be justified on the basis of the distinction between the forms yourself and yourselves; the choice between yourself and yourselves is determined by the choice between youse and you pL (cf. youse yourself vs. you pL yourselves). There is nothing surprising in the fact that one word may have two meanings, one indefinable and one definable. It is more surprising if one word appears to have two different indefinable meanings. In fact, however, the evidence available so far suggests that there are no languages in the world which would use the same word for 'you' and 'I'. More generally, there appear to be no languages in the world which wouldn't have special (separate) words for these two vital concepts.

12

Introduction: semantics and pragmatics

8. Semantic equivalence vs. pragmatic equivalence If there are scholars who - like the ordinary monolingual person believe that most words in one language have exact semantic equivalents in other languages, there are also those who believe that no words in one language can have exact equivalents in many other languages, let alone in all the languages of the world. For example, they say, there are languages which have no personal pronouns, no words for 'you' or 'I'. Japanese is sometimes cited as an example of this. In my view, this is a fallacy. For cultural reasons, Japanese speakers try to avoid the use of personal pronouns (cf. Bamlund 1975b; Suzuki 1986) and the language has developed a wealth of devices that allow its speakers to avoid such overt reference without producing any misunderstandings. For example, there are certain verbs in Japanese (so-called honorific verbs) which are never used with respect to the speaker; and there are 'humble', selfdeprecating verbs, which are never used with respect to the addressee; the use of such verbs often sufficiently identifies the person spoken about and the person addressed as to make an overt reference to 'you' and 'I' unnecessary. But the words for 'you' and 'I' do exist, and can be used when it is necessary or desired. It is also true that many languages, especially Southeast Asian languages, have developed a number of elaborate substitutes for 'you' and 'I', and that in many circumstances it is more appropriate to use some such substitute than the barest, the most basic pronoun. For example, in a polite conversation in Thai, the use of the basic words for 'you' and 'I' would sound outrageously crude and inappropriate. Instead, various self-deprecating expressions would be used for 'I', and various deferential expressions for 'you'. Many of the expressions which stand for 'I' refer to the speaker's hair, crown of the head, top of the head, and the like, and many of the expressions which stand for 'you' refer to the addressee's feet, soles of the feet, or even to the dust underneath his feet, the idea being that the speaker is putting the most valued and respected part of his own body, the head, at the same level as the lowest, the least honourable part of the addressee's body (cf. Cooke 1968; Palakomkul 1975). But this does not mean that Thai has no personal pronouns, no basic words for 'you' and 'I'. A language may not make a distinction which would correspond to that between the words 'he' and 'she', and in fact many languages, for example Turkish, have just one word for 'he' and 'she', undifferentiated

Semantic equivalence vs. pragmatic equivalence

13

for sex. But no known language fails to make a distinction between the speaker and the addressee, i.e. between 'you' and 'I'. This does not mean that the range of use of the words for 'you' and 'I' is the same in all languages. For example, in Japanese, the word ore, which Japanese English dictionaries gloss as 'I', has a range of use incomparably more narrow than the word I has in English. Thus, in a recent study of the use of the first and second person pronouns (Kurokawa 1972), it was found that none of the women in the sample used ore, whereas 90% of the men did - along with boku (100%), watashi (80%), watakushi (50%) and atashi (80%). It was also found that "the pronoun ore 'I' is often used among male adult speakers only in such very informal occasions as between two close friends and at home. It is not an exaggeration to say that in many elementary schools the use of this pronoun ore is discouraged by the teacher. ... This pronoun is almost never introduced in texts for an elementary, or an intermediate, Japanese course for English speaking students." (Kurokawa 1972:231). The survey also shows that "men use ore more frequently when talking with their wives than when talking with their parents: 44% versus 33%" (1972:232). What does ore mean, then? It may be considered 'rude' for a child to use ore to other children at school, but ore cannot mean 'I + disrespect', because if it did it would not be permissible for a man to use it when speaking to his parents. This suggests that ore means simply 'I' - and that there are no invariant semantic components which could be always attributed to it other than 'I'. The heavy restrictions on its use must therefore be attributed to cultural rather than semantic factors. In a society where references to oneself are in many situations expected to be accompanied by expressions of humility or deference, a bare 'I' becomes pragmatically marked, and it must be interpreted as either very intimate or very rude. But this pragmatic markedness should not be confused with demonstrable semantic complexity. Above all, it should be pointed out that words such as the Japanese ore 'I' or kimi 'you' (or French tu, or German du), cannot be further defined within the languages to which they belong. Even if someone insisted that words of this kind can be defined via English, for example, along the following lines: ore - 'I; I don't have to show respect for you' kimi (tu, du) - 'you; I don't have to show respect for you'

14

Introduction: semantics and pragmatics

explications of this kind could not be translated into Japanese, French, or German without regressus ad infinitum (for what words would be used for you and I in the explication?). We have to conclude, therefore, that words of this kind are true semantic primitives of the languages in question. To say that they are not semantic primitives, but that their inherent complexity can be shown only via definitions phrased in English, not in the languages to which they belong, would be a case of blatant ethnocentrism. Since, however, these primitives (of the Japanese, French, or German language) can be matched semantically across language boundaries, we can acknowledge their analogous (indefinable) position within the language systems to which they belong by calling them universal semantic primitives, and by equating them in semantic explications - despite the huge cultural differences reflected in their different frequency and different range of use.

9. Universal grammatical patterns But if the supposed lexical universals are embedded, in each language, in language-specific grammatical patterns, can they really be matched and identified cross-linguistically? In any case, words or morphemes by themselves cannot really express any meanings: they can only contribute in a certain way to the meaning expressed by a sentence. If we want to identify meanings cross-linguistically we must look not for isolated lexical items but for commensurable lexical items used in commensurable sentences. This means that we must look not only for commensurable lexical items but also for commensurable grammatical patterns. It seems clear that the great majority of grammatical patterns of any given language are language-specific. It is possible, however, that there are also some patterns which are universal. In fact, if cross-cultural understanding is possible at all, despite the colossal variation in language structures, there must be some common core of 'human understanding', and this common core must rely not only on some shared or matching lexical items but also on some shared, or matching, grammatical patterns in which those shared lexical items can be used. To put it differently, there must be some 'atomic sentences' (cf. Russell 1962), or 'kernel sentences' (cf. Chomsky 1957), which can be said in any language, and which can be matched across language bounda-

Semantics versus pragmatics: different approaches

15

ries. The grammar of these 'atomic sentences' must consist in the possible distribution patterns of the 'atomic elements' (that is, the lexical indefinables). Trying to discover those patterns we should look, therefore, at the lexical indefinables themselves, and try to see what their possibilities of co-occurrence might be. In searching for universal grammatical patterns, therefore, we should not look for any universals of form; rather, we should look for universals of combinability. The search for such simple and 'language-independent' grammatical patterns has begun fairly recently and is still in its early stages (cf. Wierzbicka 1988, in press a, b). The explications proposed in the present work employ a kind of reduced (English) syntax which is relatively simple and relatively language-independent, without being simple or universal in any absolute sense. Above all, I try to rely on simple clauses rather than on complex sentences and to avoid participial constructions, relative clauses, nominalisations, and other similar pieces of complex, language-specific syntactic machinery. I do not, however, try in this work to go as far as possible in the direction of simplicity and universality, because this would often increase the length of explications and make them more difficult to read. I aim at a compromise between simplicity and universality on the one hand and the reader's convenience on the other.

10. Semantics versus pragmatics: different approaches Leech (1983 :6) distinguishes three different ways of viewing the relationship between semantics and pragmatics, which he summarises usefully in the form of three diagrams, shown in Figure 1. He labels the three approaches 'semanticism' (A), 'complementarism' (B), and 'pragmaticism' (C). The classical Morrisian (1938) position, which divides the study of sign systems into syntax, semantics and pragmatics, is an instance of 'complementarism'. The philosophical tradition in the study of language which started with Wittgenstein's Philosophical Investigations (1953) and which urged, 'Don't ask for meaning, ask for use', is an instance of 'pragmaticism'. The 'generative semantics' of the early 1970s, which tried to present the illocutionary force of an utterance as part of its semantic structure, can be said to have represented 'semanticism' (see

16

Introduction: semantics and pragmatics A

B

c

r---------, semantics

r----------, IL (pragmatics) .JI

semantics

pragmatics

IL (semantics) -JI

pragmatics

Figure 1. Three views of the relationship between semantics and pragmatics

the articles in Cole - Morgan 1975). All three of these approaches present serious difficulties, which will be discussed briefly below.

10.1. 'Complementarism' Morris wanted to separate the relations between signs and 'reality' from the relations between signs and their users. But the very nature of natural language is such that it doesn't separate extralinguistic reality from the psychological and social world of language users. Language is an integrated system, where everything 'conspires' to convey meaning: words, grammatical constructions and various 'illocutionary' devices (including intonation). Accordingly, one might argue that linguistics falls naturally into three parts, which could be called lexical semantics, grammatical semantics, and illocutionary semantics. A Morrisian division of the study of signs into semantics, syntax, and pragmatics may make good sense with respect to some artificial sign systems, but it makes no sense with respect to natural language, whose syntactic and morphological devices (as well as illocutionary devices) are themselves carriers of meaning. In natural language, meaning consists in human interpretation of the world. It is subjective, it is anthropocentric, it reflects predominant cultural concerns and culture-specific modes of social interaction as much as any objective features of the

Semantics versus pragmatics: different approaches

17

world 'as such'. 'Pragmatic (attitudinal) meanings' are inextricably intertwined in natural languages with meanings based on 'denotational conditions' (see for example Wierzbicka 1980, 1987; see also Paduceva 1985). Since the meanings conveyed in natural language are inherently subjective and anthropocentric, they cannot be neatly divided into 'referential' and 'pragmatic', or 'denotational' and 'attitudinal'. What is needed, therefore, is a unified semantic framework, equally suitable for describing the meaning of 'cultural kinds' (such as cup and mug in English, or sake in Japanese), 'natural kinds' (such as cat and dog in English, or nezumi 'rat/mouse' in Japanese), interactional verbs (such as promise, vow, or pledge in English, or materit'sja 'mother-swear' in Russian), and so on. All such meanings are culture-specific, subjective, and anthropocentric (see Wierzbicka 1985a,b, 1987), 'referential' and 'pragmatic' at the same time. For example, Leech's 'complementarist' position forces him to analyse illocutionary forces such as requesting, promising, and ordering under 'pragmatics', and the meaning of verbs such as request, promise, and order, under 'semantics', as if the two tasks had nothing in common, and as if the so-called illocutionary force of requesting, promising, or ordering wasn't simply a function of the English verbs request, promise, and order.

10.2.

'Pragmaticism'

The approach that Leech has called 'pragmaticism' has perhaps more to offer, because it creates no artificial gulf between 'pragmatic meanings' and 'denotational meanings' and recognises the anthropocentric nature of natural language, where 'man' (the language user) is truly a measure of all things, and where 'objective' aspects of meaning are inextricably linked with 'subjective' and interactional ones. Yet 'pragmaticism', too, proves very hard to apply fruitfully when it comes to actual description of meanings, especially in a cross-cultural perspective, because it has no rigorous framework for description and comparison, no firm grid in terms of which the endless vagaries of language use can be rigorously analysed and interpreted. To try to describe language use without such a grid is like trying to describe phonological systems of different languages without having a universal phonetic alphabet of any sort. Not surprisingly, many linguists accustomed to high standards of rigour in domains such as phonology,

18

Introduction: semantics and pragmatics

syntax, or historical linguistics reject linguistic articles and books based on the philosophy of 'pragmaticism' as 'woolly', 'waffly', and arbitrary.

10.3. 'Semanticism' In the present writer's view, the approach which Leech calls 'semanticism' has much more to offer to the study of meaning in natural language, because it can provide it with a firm basis and can allow it to combine insight with rigour. Natural language is a system for conveying meaning, and any integration of linguistic science can be achieved only on the basis of meaning. The fact that a well-known linguistic school which advocated a 'radically semantic' position (' generative semantics ') has failed, and has acknowledged its defeat (see Newmeyer 1980:167-173; Lakoff 1986: 584-585), doesn't mean that there is something inherently wrong with a 'radically semantic' orientation as such. One cannot describe and compare meanings in a non-arbitrary way without a well-justified set of (candidates for) universal semantic primitives. Generative semanticists didn't strive to discover such a set (although they did like to refer, in the abstract, to some unidentified 'atomic predicates '). One can argue that this was the main cause of their failure (in pragmatics, and in semantics in general), not their 'radically semantic' approach. What they lacked was a methodology which would lend coherence and unity to the field of semantics, and which would define a well-justified boundary around it. Linguistic semantics and linguistic pragmatics are one. What applies to colour semantics, kinship semantics, speech-act semantics, to the semantics of natural kinds, cultural kinds, emotions, and so on applies also to the semantics of interpersonal attitudes.

10.4. A fourth approach: two pragmatics But can all aspects of pragmatics be handled by means of a universal semantic framework, the same framework which can also be used for all other areas of meaning? Probably nobody would want to go so far as to claim that. The term 'pragmatics' has been applied to a very wide and heterogeneous range of phenomena, including 'conversational analysis', 'linguistic etiquette', 'acquisition of communicative competence', and so on. In fact, many

Semantics versus pragmatics: different approaches

19

scholars have suggested that 'pragmatics' is no more than a wastepaper basket, where everything that has to do with language but which cannot be treated rigorously is thrown. This position gives 'pragmatics' a very broad scope indeed, but it leaves the 'core linguistics' greatly impoverished and deprived of a component which is essential to a coherent and integrated description of linguistic competence. In my view, the only possible solution to this dilemma is to recognise that there are two pragmatics, differing from one another not so much in subject matter as in methodology. There is a linguistic pragmatics, which can form a part of a coherent, integrated description of linguistic competence, and there is another pragmatics, or other pragmatics (in the plural): a domain or domains of the sociologist, the psychologist, the ethnomethodologist, the literary scholar, and so on. As Hugo Schuchardt (1972:67) pointed out, the unity of a scholarly discipline is created by its coherent methodology, not by any inherent unity of the subject matter. Pragmatics is, up to a point, an integral part of linguistics, and the boundary between linguistic pragmatics and nonlinguistic pragmatics is determined by the stretching capacities of a coherent unified linguistic framework. Attitudinal meanings can be treated in the same descriptive framework as any other kinds of meaning. They can therefore be regarded as belonging to semantics and, ipso facto, to 'core' linguistics. There is no gulf between linguistic pragmatics and linguistic semantics; on the contrary, linguistic pragmatics can be fruitfully seen as part of linguistic semantics. But there is a gulf between linguistic pragmatics and various other, heterogeneous, considerations of language use. This leads us to propose a fourth diagram, shown in Figure 2, in addition to the three proposed by Leech. This diagram represents a 'radically semantic' approach to meaning, with so-called 'pragmatic meanings' being treated in exactly the same way, and being described in exactly the same framework, as any other kind of meaning. But this doesn't mean that anything that has ever been called 'pragmatics' could, or should, be swallowed by semantics.

20

Introduction: semantics and pragmatics j:::)

I'Q

r----------7------ €t/)Q --~-----. Semantics tics

Non -linguisti~ pragmatics

~--------~----------------

Figure 2. A 'radically semantic' approach to meaning

11. Description of contents Chapter 2, 'Different cultures, different languages, different speech acts', discusses a number of differences between two languages, English and Polish, in the area of speech acts, and links these differences with different cultural norms and cultural assumptions. It is shown that English, as compared with Polish, places heavy restrictions on the use of the imperative and makes extensive use of interrogative and conditional forms. Features of English which have been claimed to be due to universal principles of politeness are shown to be language-specific and due to specific cultural norms and cultural traditions. Linguistic differences are shown to be associated with cultural values such as individualism and respect for personal autonomy in the case of English, and cordiality in the case of Polish. Furthermore, certain characteristic features of Australian English are discussed and illustrated, and are shown to reflect

Description of contents

21

some features of the Australian national ethos. Implications for the theory of speech acts and for intercultural communication are discussed. In particular, certain influential theories of speech acts, based largely on English (in particular, Searle's theory) are shown to be ethnocentric and dangerous in their potential social effects. Chapter 3, 'Cross-cultural pragmatics and different cultural values', uses a much wider range of examples (in particular, from Japanese, Black American English, Yiddish, and Hebrew), to show that differences in the ways of speaking associated with different languages are profound and systematic, and that they reflect, and can be explained in terms of, independently established differences in cultural traditions, cultural values, and cultural priorities. It demonstrates the anglocentrism of supposedly universal 'maxims' of human conversational behaviour of the kind put forward by Grice (1975) or Leech (1983). It also shows how progress in cross-cultural pragmatics has been hampered by the use of inadequate conceptual tools: in particular, of unanalysed, obscure and protean global labels such as 'directness', 'self-assertion', 'distance', 'intimacy', 'solidarity', 'harmony', 'informality', and so on, which have led to paradoxical and contradictory conclusions; and it proposes a method whereby different communicative styles can be clarified in terms of 'cultural scripts' written in the metalanguage of universal semantic primitives. Chapter 4, 'Describing conversational routines', shows that while considerable effort has gone into the description and comparison of conversational routines associated with different languages and different cultures, much less has been achieved in this important area than might have been - because not enough thought has been given to the vital question of a standardised and 'culture-free' metalanguage in which such comparisons could be fruitfully carried out. To show how the use of the natural semantic metalanguage can facilitate this task I examine, in particular, a number of generalisations suggested in Pomerantz's (1978) paper 'Responses to compliments', and I show how these generalisations could be reformulated to make them both clear and verifiable. I also examine a number of other conversational routines, trying to show how the use of the natural semantic metalanguage can bring a new level of rigour to conversational analysis, and can free it from ethnocentric bias. Chapter 5, 'Speech acts and speech genres across languages and cultures', discusses a number of speech acts and speech genres from English, Polish, Japanese, Hebrew, and Walmatjari (an Australian Aboriginal language), approaching them through the words which name

22

Introduction: semantics and pragmatics

them (that is to say, through their folk labels). It is claimed that folk names of speech acts and speech genres provide an important source of insight into the communicative styles most characteristic of a given society, and reflect salient features of the culture associated with a given language; and that to fully exploit this source one must carry out a rigorous semantic analysis of such names, and express the results in a culture-independent semantic metalanguage. This is shown in detail through the semantic analysis of a group of Australian English speechact verbs, together with a discussion of traditional Australian values and the Australian national ethos. Chapter 6, 'The semantics of illocutionary forces', examines a wide range of English constructions and expressions encoding certain modes of interpersonal interaction, and spells out their meaning (or their 'illocutionary force'). For example, different types of tag questions and different types of 'interrogative directives' ('whimperatives ') are discussed, and both the similarities and differences between them are made explicit. Here as in the other chapters of the book, the analysis takes the form of decomposition of illocutionary forces into their components, which are formulated in the natural semantic metalanguage. It is argued that the decomposition of illocutionary forces illustrated in this chapter offers a safe path between the Scylla of the 'performative hypothesis' (which has proved to be empiricially inadequate and theoretically unjustifiable) and the Charybdis of the 'autonomous grammar', which tries to divorce the study of language structure from the study of language use. Chapter 7, 'Italian reduplication: its meaning and its cultural significance', constitutes a case study of one culture-specific pragmatic device: the Italian 'reduplication' (for example bella bella 'beautiful beautiful '), examined against the background of various other 'intensification devices', such as, for example, the absolute superlative (for example I am most grateful). It is demonstrated that subtle pragmatic meanings such as those conveyed in Italian reduplication can be identified and distinguished from other, related meanings if ad hoc impressionistic comments are replaced with rigorous semantic explications; and it is shown how a semantic metalanguage derived from natural language can be used for that purpose. It is also argued that syntactic reduplication belongs to a system of pragmatic devices which reflect, jointly, some characteristic features of Italian communicative style. More generally, it is argued that illocutionary grammar can be linked directly with 'cultural style', and that cross-cultural pragmatics can gain considerably in both

Description of contents

23

insight and rigour if its problems are translated into the framework of illocutionary semantics. Chapter 8, 'Interjections across cultures', argues that interjections like any other linguistic elements - have meanings of their own, and that these meanings can be identified and captured in the natural semantic metalanguage. A number of interjections from English, Polish, Russian, and Yiddish are discussed, and rigorous semantic formulae are proposed which can explain both the similarities and the differences in their range of use. For example, the English interjection yukI is compared and contrasted with its nearest Polish and Russian counterparts Jul, Je I, tful and t'Jul It is shown that while the meaning of interjections cannot be adequately captured in terms of emotion words such as disgust, it can be captured in terms of more fine-grained components, closer to the level of universal semantic primitives. The role of sound symbolism in the functioning of interjections is discussed, and the possibility of reflecting this symbolism in semantic formulae is explored. Chapter 9, 'Particles and illocutionary meanings', examines a number of English and Polish particles, quantitative (for example only, merely, just) and temporal (for example already, still, yet), and in each case offers a paraphrase in natural semantic metalanguage, substitutable in context for the particle itself. Special attention is given to 'approximative' particles, such as almost, around, about, or at least. It is shown how the 'radically pragmatic' approach to the study of such particles, advocated by Sadock (1981) and others, fails to account for the range of their use. It is demonstrated that even the vaguest 'hedges' and 'approximatives' (for example roughly and approximately) can be given rigorous, and yet intuitively clear, semantic explications, which can explain their uses, and the differences in the use of closely related particles, both within a language and between different languages. Chapter 10, 'Boys will be boys: even 'truisms' are culture-specific', develops more fully a critique of a 'Gricean' or 'radically pragmatic' approach to language use. Evidence against this approach is drawn mainly from the area of colloquial 'tautologies' such as War is war or A promise is a promise, which have often been adduced, by Grice and by others, in support of such a 'radically pragmatic' approach to language use. The chapter shows that such 'tautological constructions' are partly conventional and language-specific, and that each such construction has a specific meaning, which cannot be fully predicted in terms of any universal pragmatic maxims. It is argued that the attitudinal meanings conveyed by various tautological constructions and by similar linguistic

24

Introduction: semantics and pragmatics

devices can be stated in rigorous and yet self-explanatory semantic formulae. 'Radical pragmatics' is rejected as a blind alley, and an integrated approach to language structure and language use is proposed, based on a coherent semantic theory, capable of representing 'objective' and 'subjective' aspects of meaning in a unified framework. Chapter 11, 'Conclusion: semantics as a key to cross-cultural pragmatics', recapitulates the main features of the approach to the study of human interaction advanced in the present book, stressing in particular its universal, 'culture-free' perspective, and its 'multicultural', culturespecific, content. It highlights the theoretical and methodological novelty of the book, its empirical orientation, and its potential for use in language teaching and in the teaching of cross-cultural understanding and cross-cultural communication.

Chapter 2

Different cultures, different languages, different speech acts

From the outset, studies in speech acts have suffered from an astonishing ethnocentrism, and to a considerable degree they continue to do so. Consider, for example, the following assertion: "When people make requests, they tend to make them indirectly. They generally avoid imperatives like Tell me the time, which are direct requests, in preference for questions like Can you tell me the time? or assertions like [' m trying to find out what time it is, which are indirect requests." (Clark - Schunk 1980:111) It is clear that these authors have based their observations on English alone; they take it for granted that what seems to hold for the speakers of English must hold for 'people generally'. Another author writes: The focus of this chapter is on the situational conventions that influence how people make, understand, and remember requests. I will argue that people's knowledge of particular social situations results in certain requests being seen as conventional. ... My starting point will be to show how social contexts constrain the ways in which people comprehend indirect requests .... I will sketch a new proposal that specifies how the structure of social situations directly determines the surface forms used by speakers in making requests. (Gibbs 1985:98)

This author seems to be quite unaware that there are people other than speakers of English; consequently, he doesn't even suspect that 'surface forms used by speakers in making requests' may differ from language to language, and that if they do differ then they cannot be 'directly' determined by 'social situations'. Throughout this chapter, I will try to show that statements such as those quoted above are based on an ethnocentric illusion: it is not people in general who behave in the ways described, it is the speakers of English. Presumably, the ethnocentric bias characteristic of speech act studies is largely due to their origin in linguistic philosophy rather than in linguistics proper (see below, section 5). Nonetheless, statements mistak-

26

Different cultures, different languages, different speech acts

ing Anglo-Saxon conversational conventions for 'human behaviour' in general abound also in linguistic literature. I will quote just one more characteristic example: "Every language makes available the same set of strategies - semantic formulas - for performing a given speech act. ... if one can request, for example, in one language by asking the hearer about his ability to do the act (Can you do that?), by expressing one's desire for the hearer to do the act (I'd really appreciate if you'd do that), . .. then these same semantic formulas - strategies - are available to the speakers of every other language." (Fraser - Rintell - Walters 1980:78-79). These authors are not unaware of some crosslinguistic differences in this respect, but they dismiss them as 'minimal'. Such preconceptions could probably be seriously dented by reference to almost any language. Here, I shall be drawing mainly upon illustrative material from Polish and from Australian English. But even if one limits the task at hand to comparing selected speech acts from only two languages, the topic is still vast and couldn't be treated exhaustively in anyone work. The cultural norms reflected in speech acts differ not only from one language to another, but also from one regional and social variety to another. There are considerable differences between Australian English and American English, between mainstream American English and American Black English, between middle-class English and working-class English, and so on. There is also a great deal of variation within Polish. Nonetheless, there is also a remarkable amount of uniformity within English, as there is within Polish. It goes without saying that the differences between English and Polish discussed in this chapter could, and should, be studied in a much more thorough and systematic way than has been done here. But to do so, one would have to devote a whole book to the subject, or one would have to limit one's field of vision to a strip so narrow that one would have no grounds for reaching the generalisations which in my view explain phenomena of the kind discussed here. The present overview was compiled as a pilot study. I believe, however, that even in its present form it amply demonstrates that different cultures find expression in different systems of speech acts, and that different speech acts become entrenched, and, to some extent, codified in different languages.

Preliminary examples and discussion

27

1. Preliminary examples and dicussion At a meeting of a Polish organisation in Australia a distinguished Australian guest is introduced. Let us call her Mrs. Vanessa Smith. One of the Polish hosts greets the visitor cordially and offers her a seat of honour with these words: Mrs. Vanessa! Please! Sit! Sit!

The word Mrs. is used here as a substitute for the Polish word pani, which (unlike Mrs.) can very well be combined with first names. What is more interesting about the phrasing of the offer is the use of the short imperative Sit!, which makes the utterance sound like a command, and in fact like a command addressed to a dog. The phrase Sit down! would sound less inappropriate, but in the context in question it would not be very felicitous either: it still would not sound like an offer, let alone a cordial and deferential one. A very informal offer could be phrased as Have a seat, with imperative mood, but not with an action verb in imperative mood. More formal offers would normally take an interrogative form: Will you sit down? Won't you sit down? Would you like to sit down? Sit down, won't you?

In fact, even very informal offers are often performed in English by means of sentences in the interrogative form: Sure you wouldn't like a beer? (Hibberd 1974:218) Like a swig at the milk? (Hibberd 1974:213)

Significantly, English has developed some special grammatical devices in which the interrogative form is normally used not for asking but for making an offer, a suggestion or a proposal, especially the form How about a NP?: How about a beer? (Buzo 1979:64) How about a bottle? (Hibberd 1974:187)

In Polish, How about utterances have to be rendered in a form indistinguishable from that of genuine questions (except of course for the intonation) :

28

Different cultures, different languages, different speech acts

Moze si~ ezegos napijesz? 'Perhaps you will drink something?'

A further difference between Polish and English concerns the literal content of interrogative offers. In English, a tentative offer (even a very informal one) tends to refer to the addressee's desires and opinions: Like a swig at the milk? (Hibberd 1974:213) Sure you wouldn't like a bash at some? (Hibberd 1974:214)

The phrasing of such offers implies that the speaker is not trying to impose his will on the addressee, but is merely trying to find out what the addressee himself wants and thinks. In Polish, literal equivalents of offers of this kind would sound inappropriate. The English question Are you sure?, so often addressed by hosts to their guests, sounds comical to the Polish ear: it breaks the unwritten law of Polish hospitality, according to which the host does not try to establish the guest's wishes as far as eating and drinking is concerned but tries to get the guest to eat and drink as much as possible (and more). A hospitable Polish host will not take 'No' for an answer; he assumes that the addressee can have some more, and that it would be good for him or her to have some more, and therefore that his or her resistance (which is likely to be due to politeness) should be disregarded. A reference to the addressee's desire for food is as inappropriate in an offer as a reference to his or her certainty. Sentences such as: Miatbys oehotf na piwo? 'Would you like a beer?'

would be interpreted as questions rather than as offers. It would not be good manners to reveal to the host that one feels like having a beer; the social convention requires the host to prevail upon the guest, to behave as if he or she was forcing the guest to eat and drink, regardless of the guest's desires, and certainly regardless of the guest's expressed desires, which would be simply dismissed. The typical dialogue would be:

Proszf bardzo! Jeszcze

troszk~!

Ale juz nie mogf(! Ale koniecznie!

'Please! A little more!' 'But I can't!' 'But you must!' (literally: 'But necessarily! ')

Preliminary examples and discussion

29

What applies to offers applies also, to some extent, to invitations. For example, in English a man can say to a woman: Would you like to come to the pub tomorrow night with me and Davo? (Buzo 1979:60) Would you like to come out with me one night this week? (Hibberd 1974:214) Hey, you wouldn't like to come to dinner tonight, would you? (Hibberd 1974:193)

In Polish, literal translations of such utterances would make very poor invitations. A sentence in the frame: Czy mialabys ochotf 'Would you like to

? ?'

sounds like a genuine question, not like an invitation or a proposal. If a man wants to ask a woman out, it would sound presumptuous for him to express overtly an assumption that she 'would like' to do it. Rather, he should show that he would like to go out with her, and seek her consent. One would say: Moiebysmy poszli do kina? 'Perhaps we would go to the cinema?' (implied: if I asked you)

rather than: Czy mialabys ochot~ pojsc ze mng do kina? 'Would you like to go to the cinema with me?'

A tentative and self-effacing invitation such as the following one: Say, uh, I don't suppose you'd like to come and have lunch with me, would you? (Buzo 1974:44)

could not be translated literally into Polish without losing its intended illocutionary force: Powiedz, hm, nie przypuszczam, iebys miata ochote zjfsc lunch ze mnf/, co?

The sentence sounds bizarre, but if it could be used at all it would be used as a genuine question, not as an invitation or proposal. A question of this kind could of course be interpreted as a prelude to an invitation, but it would have to be reported as he asked me whether, not as he invited me to. Clearly, one factor responsible for this difference is the

30

Different cultures, different languages, different speech acts

principle of 'polite pessimism', characteristic of Anglo-Saxon culture (cf. Brown - Levinson 1978:134-135), but absent from Polish culture.

2. Interpretive hypothesis Of course, Polish is not alone among European languages in differing from English in the ways indicated above. On the contrary, it is English which seems to differ from most other European languages along these lines. Many of the observations made in the present chapter would also apply to Russian, Serbo-Croatian, Spanish and many other languages. It is English which seems to have developed a particularly rich syste.m of devices reflecting a characteristically Anglo-Saxon cultural tradition: a tradition which places special emphasis on the rights and on the autonomy of every individual, which abhors interference in other people's affairs (It's none of my business), which is tolerant of individual idiosyncrasies and peculiarities, which respects everyone's privacy, which approves of compromises and disapproves of dogmatism of any kind. The heavy restrictions on the use of the imperative in English and the wide range of use of interrogative forms in performing acts other than questions, constitute striking linguistic reflexes of this socio-cultural attitude. In English, the imperative is mostly used in commands and in orders. Other kinds of directives (i.e., of speech acts through which the speaker attempts to cause the addressee to do something), tend to avoid the imperative or to combine it with an interrogative and/or conditional form. (For certain important qualifications to this overall tendency, see Lakoff 1972; Ervin-Tripp 1976.) At least this is how English strikes native speakers of a language like Polish, where the bare imperative is used on a much wider scale. It is interesting to note that from a different cultural perspective English may be seen as a language favouring, rather than shunning, the use of imperative. This is, in particular, how English appears to speakers of Japanese. For example, Higa (1972:53) notes the wide use of the imperative in the English advertising language and points out that, for example, the Japanese sign corresponding to the ubiquitous English Drink Coca-Cola! would read Coca Cola 0 nomimasho! (Literally, 'We will drink Coca Cola! ') rather than the imperative Coca Cola 0 nome! Similarly, Matsumoto (1988:420) points out that in Japanese recipes or instructions

Case studies

31

an imperative would be avoided, whereas in English recipes or instructions it is quite common. It should be noted, however, that advertisements and recipes are, first, anonymous, and second, directed at an imaginary addressee, not at a particular individual. What Anglo-Saxon culture abhors is the impression that one individual is trying to impose his or her will upon another individual. In the case of 'public speech acts' such as advertisements or recipes this danger does not arise, and the imperative is not felt to be offensive. In Polish, however, 'private' speech acts, directed from one person to another, can also use the imperative, and they do not rely on interrogative devices in this area either. In what follows, I will consider a number of areas where Polish, and other languages, differ from English along the lines suggested here, specifically: advice, requests, tag questions, opinions, and exclamations.

3. Case studies 3.1. Advice In a language like Polish, advice is typically offered in the form of an imperative: fa ci radz~ powiedz mu prawd~. 'I advise you: tell him the truth.'

In English advice would normally be formulated more tentatively: If I were you I would tell him the truth. Tell him the truth - I would. Why don't you tell him the truth? I think it would be best. Why not tell him the truth? I think that might be best. Maybe you ought to tell him the truth? Do you think it might be a good idea to tell him the truth? All these utterances could be reported in English using the verb advise (She advised me to tell him the truth). But their literal Polish equivalents would not be reported using the verb radzic 'advise'. Normally, only utterances in the imperative mood or utterances with the verb radzit used performatively could be so reported:

32

Different cultures, different languages, different speech acts Radz~

ci, zebys mu powiedzial prawd~. 'I advise you to tell him the truth.'

It is also worth noting that the English verb advise is seldom used performatively in ordinary speech: the phrase I advise you sounds very stiff and formal; by contrast, its Polish equivalent ja ci radz~ sounds perfectly colloquial and is frequently heard in everyday conversations.

3.2. Requests In English, if the speaker wants to get the addressee to do something and does not assume that he could force the addressee to do it, the speaker would normally not use a bare imperative. Speech acts which could be reported by means of the verbs request or ask (to) frequently have an interrogative or an interrogative-cum-conditional form, as in the following examples (all from Green 1975:107-.130): Will you close the door please? Will you close the window please. Will you please take our aluminium cans to the Recycling Centre. Would you take out the garbage please. Would you get me a glass of water. Would you mind closing the window. Would you like to set the table now. Won't you close the window please. Do you want to set the table now? Why don't you clean up that mess. Do you want to get me a scotch. Why don't you be nice to your brother for a change. Why don't you be quiet. Why don't you be a honey and start dinner now.

Not a single one of these utterances could be translated literally into Polish and used as a request. In particular, literal equivalents of sentences in the frame Why don't you would be interpreted as a combination of a question and a criticism, rather like utterances based on the modal Why do it are in English (Why paint your house purple?) (See Gordon - Lakoff 1975:96; cf. also Wierzbicka 1988:28.) In fact, a sentence such as:

Case studies

33

Dlaczego nie zamkniesz okna? (Literally) 'Why don't you close the window?'

would imply unreasonable and stubborn behaviour on the part of the addressee ('why haven't you done what was obviously the right thing to do - you should have done it long ago; I can't see any excuse for your failure to have done it'). The corresponding English sentence could also be interpreted in this way, but it doesn't have to be. In particular, as pointed out to me by Jane Simpson (p.c.), the contracted from Why' n' tcha suggests a request rather than a question. It is worth noting in this connection that English has developed some special devices for expressing requests and other directives in a partly interrogative style, especially the expression Why don't you be (ADJ) , which can hardly be used for genuine questions. As pointed out in Green (1975:127), the sentence Why aren't you quiet? can be a genuine question, but the sentence Why don't you be quiet?! cannot. Thus, the construction Why don't you be (ADJ)? has an interrogative form, and an interrogative component in its meaning, but is specialised in speech acts other than questions. Characteristically, Polish has no similar constructions. Since in Polish the use of interrogative forms outside the domain of questions is very limited, and since the interrogative form is not culturally valued as a means of performing directives, there was, so to speak, no cultural need to develop special interrogative devices for performing speech acts other than questions, and in particular, for performing directives. As for literal equivalents of sentences in the frame Won't you, such as: Nie zamkniesz okna? 'Won't you close the window?'

they would be interpreted as surprised questions (not necessarily critical questions, but surprised questions). They would invite both an answer and an explanation ('You are not going to do it? That's strange; I wonder why?'). The difference between English and Polish in this respect becomes particularly clear in cases of transference. For example, my daughters, who are bilingual, but who live in an English-speaking environment, often phrase their Polish requests interrogatively (or did when they were younger): Mamo, czy podasz mi chusteczk~? 'Mum, will you give me a Kleenex?'

34

Different cultures, different languages, different speech acts

This sounds very odd to me, and I tend to correct them, urging them to use the imperative (with the word prosz~ 'please') instead. To an English speaker, this might look like an attempt to teach one's child to be impolite. But in Polish, politeness is not linked with an avoidance of imperative, and with the use of interrogative devices, as it is in English. The expression Would you mind has simply no equivalent in Polish. I do not wish to imply, however, that Polish never uses the interrogative form in requests. It does, but in comparison with English, the possibilities are heavily restricted. Thus, one could perform requests, or acts closely related to requests, by ostensibly 'asking' about the addressee's ability to do something, or about his or her goodness (or kindness): Czy mogfbys 'Could you

? ?'

Czy bylbys tak dobry, zeby ... ? 'Would you be so good as to ?' Czy byf(a)by Pan(i) laskaw(a) ? 'Would you be so kind/gracious as to ... ?'

But one could not ask people to do something by using literal Polish equivalents of the phrases Would you do it, Won't you do it, Why don't you do it, Do you want to do it or Would you like to do it. Pseudo-questions which ostensibly inquire about the addressee's desire and which in fact are to be interpreted as requests (Would you like to, Do you want to) seem particularly odd and amusing from a Polish point of view, as transparent acts of what looks like naive hypocrisy. But it is not just the range of acceptable interrogative devices which distinguishes Polish directives from the English ones. Differences in function are at least as striking. Thus, in Polish interrogative directives sound formal and elaborately polite. They are also tentative, lacking in confidence. One would use them when one is genuinely not sure whether the addressee would do what is requested. Moreover, they could not be used in anger (unless sarcastically) and they are incompatible with the use of swear words. In Australian English, however, both the interrogative and the interrogative-cum-conditional forms are frequently used in speech acts which could be reported by means of the verbs order to, command or tell to, and they are perfectly compatible with verbal abuse and verbal violence, as the following examples demonstrate:

Case studies

35

Can't you shut up? (Hibberd 1974:228) Why don't you shut your mouth? (Hibberd 1974:228) Will someone put the fucking idiot out of his misery? (Williamson 1974:48) Will you bloody well hurry up! (Williamson 1974:56) For Christ's sake, will you get lost. (Williamson 1974:191) Why don't you shut up? (Buzo 1979:37)

Andrew (to Irene, very angry): Will you please go to bed? (Williamson 1974:197) Could you try and find the source of that smell before then, and could you possibly put your apple cores and orange peel in the bin for the next few days? (After a pause, loudly) And could you bloody well shit in the hole for a change? (Williamson 1974:7)

In fact, the interrogative form in English has reached the stage of being so thoroughly dissociated from the language of courtesy and respect that it can well be used in pure swear phrases, where the speaker forcefully expresses his feelings apparently without attempting to get the addressee to do anything, as in the following example: Why don't you all go to hell! (Hibberd 1974:199)

This shows particularly clearly that the English predilection for the interrogative form in human interaction, and the heavy restrictions which English places on the use of the imperative, cannot be explained simply in terms of politeness. After all, Polish, too, has its polite and extrapolite ways of speaking, and has developed a repertoire of politeness devices. What is at issue is not politeness as such, but the interpretation of what is socially acceptable in a given culture. For example, Australian culture is highly tolerant of swearing. Swear words are often used to express strong feelings and not only negative but also positive feelings, as in the following examples: Stork: Not bloody bad, is it? Clyde: It's a bloody beauty. (Williamson 1974:18) Bloody good music! (Buzo 1979:30)

36

Different cultures, different languages, different speech acts

There is no longer any widely shared taboo against swear words in 'polite conversation', for example in conversation with ladies about music. On the other hand, there is evidently a strong reluctance to use bare imperatives - not only in polite conversation, but even in not-sopolite conversation. The implicit cultural assumption reflected in English speech seems to be this: everyone has the right to their own feelings, their own wishes, their own opinions. If I want to show my own feelings, my own wishes, my own opinions, it is all right, but if I want to influence somebody else's actions, I must acknowledge the fact that they, too, may have their feelings, wishes or opinions, and that these do not have to coincide with mine. It is interesting to note that the flat imperative, which in English cultural tradition can be felt to be more offensive than swearing, in Polish constitutes one of the milder, softer options in issuing directives. When the speaker gets really angry with the addressee, the speaker will often avoid the imperative and resort to 'stronger' devices, in particular the bare infinitive: Nie pokazywac mi

si~

tuta}!

'Not to show oneself to me here!' (i.e. 'You are not to come here.') Wynosic

si~

stqd!

'To get away from here!' (i.e. 'Get away from here! ') Zabierac

si~

stgd!

'To take oneself off from here!' (i.e. 'Off with you! ') In the examples above (taken from Andrzej Wajda's film "Moralnosc pani Dulskiej", based on a number of Gabriela Zapolska's plays), the verbs chosen (wynosic si~, zabierac si~) are offensive and pejorative, but especially offensive is the impersonal syntactic construction, with the infinitive used instead of the more neutral imperative. The impersonal infinitive seems to annihilate the addressee as a person (the absence of a mention of the addressee in the sentence being an icon of his/her 'nonexistence'): it implies that the addressee is not worthy to be addressed as an individual human being, and that the speaker does not wish to establish any 'I-you' relationship with him/her. In particular, the speaker excludes the possibility of any reply from the addressee. The infinitive signals: 'No discussion' ('there is no person here whom I would regard as a potential interlocutor, for example, as someone who could refuse or decline to do as I say').

Case studies

37

By contrast, the English interrogative directives explicitly invite a verbal response, as well as a non-verbal one (Okay, All right, Sure, and the like), and thus indicate that the speaker views the addressee as an autonomous person, with his or her own free will, who can always decline to comply. The imperative is neutral in this respect: it neither precludes nor invites a verbal response. Partly for this reason, no doubt, it is favoured in Polish and disfavoured in English. I would add that the infinitive construction is by no means restricted to contexts where the speaker is angry. It can also be used simply to assert one's authority; for example it can be used by parents who wish to sound stern, as in the following example:

Isc

Macie parasol? prosto - nie oglpdac si~. skromnosc - skarb dziewcz{!cia. (Zapolska 1978:30)

Pami~tac:

'Do you have the umbrella? (To) go straight - not to look around. (To) remember: modesty is a girl's treasure.' When the speaker wants to be more polite while still wishing to signal coldness and a lack of intimacy, the infinitive can be used in combination with a performatively used verb: Prosz~ si~

do tego nie mieszac. (Zapolska 1978:108)

'I ask not to interfere.' Prosz~

-

prosz~

powiedziec,

prosz~ si~

nie krepowac. (from the

film "Moralnosc pani Dulskiej") 'I ask - I ask to say, I ask not to be embarrassed.' In a sense, the infinitive directive functions as a distance-building device in Polish, just as an interrogative directive does in English. But in Anglo-Saxon culture, distance is a positive cultural value, associated with respect for the autonomy of the individual. By contrast, in Polish culture it is associated with hostility and alienation.

3.3. Tags The deep-rooted habit of acknowledging possible differences between individual points of view is particularly clearly reflected in the English tag questions. Seen from a Polish point of view, English speech is characterised by an all-pervasive presence of tag questions, highly diversified in form and function. Essentially, Polish has only five or

38

Different cultures, different languages, different speech acts

six words which can be used as tags: prawda? 'true?', nie? 'no?', tak? 'yes?', co? 'what?', dobrze? 'good', and nieprawdaz? 'not true?' (slightly archaic). These are comparable to the English tags okay?, right?, and eh? (this last one frequently encountered in Australia). If these five or six Polish words were used nearly as often as English tag questions are, Polish speech would sound grotesquely repetitive. The English strategy of using auxiliary verbs - any auxiliary verbs, in any combinations of moods, tenses and persons - as tags, ensures great formal variety of tag questions. Expressions such as did he, was she, have you, aren't they and so on may all have the same function, but the sheer variety of their form allows them to be used much more frequently than the five Polish tag words could be used. But the differences between the English and the Polish systems of tag questions go much further than that. The topic is vast and obviously cannot be treated exhaustively here (see Chapter 6, section 5 on the illocutionary force of tag questions). Let me simply make a few observations. As has often been noted, English imperatives allow not one tag but several, each with a slightly different function: Close Close Close Close Close Close Close

the the the the the the the

door, door, door, door, door, door, door,

will you? won't you? could you? can't you? why don't you? why can't you? would you?

In Polish, all these different tags would have to be rendered by means of a single one: dobrze? 'well (good)?': Zamknij drzwi, dobrze?

Semantically, the Polish tag corresponds most closely to the English will you, the tag which assumes and expects compliance. The sentence Sit down, will you? is more confident, more self-assured than Sit down, won't you?, and the sentence Shut up, will you? sounds much more natural than Shut up, won't you? Shut up, won't you could of course be used sarcastically, but the sarcasm would exploit the effect of the semantic and stylistic clash between the forcefulness of shut up and the tentativeness of won't you.

Case studies

39

In contrast to won't you, will you can be used very widely, for example in orders and commands, as well as in requests, and it is compatible with the use of swear words: Look at this bloody ring, will you? (Williamson 1974:58) So just move out, will you? (Buzo 1979:73) (said by a wife throwing her husband out of their house)

In Polish in similar circumstances a bare imperative would normally be used, unembellished by any tag whatsoever. There are many other kinds of contexts where a tag question would be used in English but not in Polish. In particular, English negative questions with an opposite polarity would normally be translated into Polish without a tag: I don't suppose you've seen Hammo around, have you? (Buzo 1979:79) Nie widziales przypadkiem Hammo? (literally: 'You haven't seen Hammo by any chance?') You are not having a go at me, are you? (Buzo 1979:11) ezy ty si~ przypadkiem ze mnie nie nabijasz? (literally: 'You are not having a go at me by any chance?') You haven't heard anything about me, have you: Any sort of ... rumours, have you? (Buzo 1979:64) Nie slyszeliscie przypadkiem czegos 0 mnie? Jakichs ... plotek? (literally: 'You haven't heard anything about me, by any chance? Any rumours?')

Another situation where a tag question sounds plausible in English but not in Polish can be illustrated with the following utterance: I've made a bloody fool of myself, haven't /? (Williamson 1974:48)

The speaker discovers something about himself that he supposes the addressees have been aware of all along. In Polish, a plausible thing to say in a case like this would be widz{? 'I see', without a tag: Widz{?, ie si~ zachowalem jak duren! (?co, ?prawda, ?tak, ?nie, etc.) 'I see I have acted like a fool!' (?what, ?true, ?yes, ?no, etc.)

40

Different cultures, different languages, different speech acts

Again, I am not suggesting that tag questions are always used in English out of consideration for other people or out of politeness. In fact, they can be combined with accusations, insinuations and abuse, as in the following examples: Well. We have become a sour old stick, haven't we? (Williamson 1974:195) What? You've changed your mind again, have you? (Williamson 1974:198) You are a smart little prick, aren't you. (Williamson 1974:192) You've engineered this whole deal, haven't you? (Williamson 1974:193) You'd rather I was still over there, wouldn't you? (Williamson 1974:187)

In cases like these, one would not use a tag in Polish. In Polish the use of tags is, by and large, restricted to situations when the speaker really expects confirmation. In English, however, tag questions have come to be so ubiquitous, and they have developed into such a complex and elastic system, that their links with politeness, cooperation and social harmony have become quite tenuous. Often, they are used as a tool of confrontation, challenge, putdown, verbal violence and verbal abuse. The very fact that tag questions have come to play such a major role in English seems to reflect the same cultural attitudes which have led to the expansion of interrogative forms elsewhere, and to the restrictions on the use of the imperative, the same emphasis on possible differences of opinion, of point of view. Basically, tag questions express an expectation that the addressee will agree with the speaker, but the very need to voice this expectation again and again signals constant awareness of a possibility of differences. The range of contexts and situations where speakers of Polish would invite confirmation is not nearly as wide, precisely because Polish cultural tradition does not foster constant attention to other people's 'voices', other people's points of view, and tolerates forceful expression of personal views and personal feelings without any consideration for other people's views and feelings. In fact, the basic Polish tag, prawda? 'true?', presents the speaker's point of view not as a point of view but as an objective 'truth'; and it doesn't seek agreement but an acknowledgement of this 'truth'.

Case studies

41

Needless to say, it would be good if the observations ventured above could be supported with text counts. So far, I have not undertaken any large-scale counts of this kind. But to give the reader some idea of the order of differences let me say, on the basis of a perusal of a large anthology of Polish plays and of several volumes of Australian plays by different authors, that one can easily get through fifty or more pages of Polish plays without encountering a single tag, while in Australian plays one can seldom get through five pages without encountering one, and often one finds several on one page. I would like to stress, however, that apart from quantitative differences suggested here, which require statistical validation, there are also some indubitable qualitative differences. As a particularly clear example I would mention chains of tag questions, characteristic of English conversation but impossible in Polish. I quote a dialogue which I heard not long ago at a bus stop in Canberra: A: Lovely shoes, aren't they? B: Aren't they nice? A: Lovely, aren't they?

One might say that in exchanges of this kind the interlocutors are no longer seeking confirmation, but rather are, so to speak, celebrating a ritual of social harmony based on anti-dogmatism and religiously respected freedom of judgement and right to one's own opinion. Similarly, the difference between the 'opinion-oriented' English tag ('I think you would say the same; I don't know if you would say the same') and the 'truth-oriented' Polish tag ('true?') is a matter of structure, not of frequency, and needs no statistical validation.

3.4. Opinions In Polish, opinions are typically expressed fairly forcefully, and in everyday speech they tend not to be distinguished formally from statements of fact. One tends to say: To dobrze. To niedobrze. 'That's good.' 'That's bad.'

as one says: 'That's white', 'That's black', in situations where in English one would say: I like it, I don't like it, or even I think I like it.

42

Different cultures, different languages, different speech acts

As mentioned above, this difference is manifested in the structure of Polish tag questions. One says in Polish, literally: 'She is nice (terrific), true?' as if being nice or terrific or not were a matter of truth. In English, one might say: She is Italian, right?

but hardly ?She is nice, right? ?? She is terrific, right?

But in Polish, the same tag, prawda 'true', would be used in both cases. In Polish, one seldom presents one's opinions as just opinions (rather than as 'the truth'), and one seldom prefaces them with expressions such as I think, I believe or in my view. Expressions of this kind exist of course (ja Sfldz~, ja mysl~, moim zdaniem, ja uwaiam) , but their use is much more restricted than the use of their English equivalents. In particular, Polish has no word which would correspond to the English word reckon, which is used very widely in working class speech, especially in Australia, in non-intellectual contexts, and which has no intellectual pretentions. Translating utterances with I reckon into Polish one would often have to leave it out, since all the conceivable Polish equivalents would sound too intellectual, too cerebral, and simply would not fit the context. For example: Gibbo: I reckon it's the spaghetti they eat. Drives them round the bend after a while. (Buzo 1974:37) Jacko: (smiling) You know, Robbo, I reckon you'd have to be about three hundred to have done all the things you reckon you've done. (Buzo 1974:51) Polish expressions such as Sf/,dze, mysl~ or uwaiam would sound as inappropriate in these contexts as the expressions I believe or in my view would be in English. Similarly, the expression I guess, commonly used in American English, is very colloquial, and it has no similarly colloquial counterparts in Polish. In situations when in English one says, for example: I guess it's true.

Case studies

43

in Polish one would say simply: To prawda. 'This is true.'

Drazdauskiene (1981) notes that expressions such as I think, I believe, I suppose or I don't think are used much more often in English than they are in Lithuanian. She suggests, basically correctly, I think, that they signal "diminished assurance and therefore courteous detachment and optional treatment of the subject matter" (1981 :57), and a desire not to put one's view bluntly, and not to sound too abrupt or quarrelsome. I don't agree, however, with her interpretation of this difference: "This leads to a conclusion of the principal differential feature of English and Lithuanian which is that in the familiar register English is verbally more courteous and less straightforward than Lithuanian." (1981:60-61). In my view, it is ethnocentric to say that Lithuanian is less courteous than English (or, for a Lithuanian author, ethnocentric a rebours): simply, the rules of courtesy are different in each language. Furthermore, the significance of the English norm in question should be seen as a reflection of a deeper cultural attitude. English speakers tend to use expressions such as I think or I reckon even in those situations in which they evidently don't wish to be courteous, as in the following exchange: Gibbo: Shows how much you know. Those back room boys work harder than any of us. Jacko: Ar bulls. I reckon it'd be a pretty soft cop being a back room boy. (Buzo 1974:20) As a different manifestation of the same cultural difference I would mention the English preference for a hedged expression of opinions and evaluations, and the Polish tendency to express opinions in strong terms, and without any hedges whatsoever. Consider, for example, the following exchange: Norm: Well, you see, Ahmed, I'm all alone now, since my good wife Beryl passed away to the heaven above. Ahmed: I'm very sorry to hear that, Norm, you must feel rather lonely. (Buzo 1979:15) In Polish, one would not say anything like 'rather lonely'. Instead, one would say bardzo samotny 'very lonely' or strasznie samotny 'terribly lonely'. Similarly, if someone' s wife should kick him out of their house, to live there with another man, it would be very odd to comment

44

Different cultures, different languages, different speech acts

on this situation in Polish using a term such as rather, as in the following passage: Richard: Bentley: Richard: Bentley:

Tell me, how's your lovely wife? I don't know. She's living with Simmo in our home unit. Bad luck. Yes, it is, rather. (Buzo 1979:64)

In English, hedged opinions go hand in hand with hedged, indirect questions, suggestions or requests. People avoid making 'direct', forceful comments as they avoid asking 'direct', forceful questions or making 'direct', forceful requests. They hedge, and an expression such as rather or sort of often fulfills a function similar to that of conditional and interrogative devices. In fact, lexical hedges of this kind often co-occur with grammatical devices such as the conditional and the interrogative form, as in the following examples: Richard: (to Sandy) Could you sort of ... put in a good word to Simmo about me? (Buzo 1979:42) Jacko: Oh, Pammy's a nice enough kid in her own way. But you're sort of different. I mean, there's a lot more to you, I'd say. I mean, now don't get me wrong, I'm not trying ... well, all I said was, how about coming to lunch? (Buzo 1974:44) Translating this last passage into Polish, one would have to leave out several of the hedges. There is no way of saying I mean in Polish, in any case no way of differentiating I mean from I' d say; there is no particle in Polish which would correspond to well (cf. Wierzbicka 1976); and there is no equivalent for sort of (except perhaps for jakas/jakos, but this is closer to somehow than to sort of: the emphasis is on the speaker's inability to describe the quality in question, not on a lack of full commitment to what is said). Thus, English is fond of understatement and of hedges; by contrast, Polish tends to overstate (for emphasis) rather than understate. When I translate my own writings from Polish into English, I find myself removing words such as totally, utterly, extremely or always, or replacing them with words and expressions such as rather, somewhat, tends to, or frequently; and vice versa.

Case studies

45

3.5. Exclamations The notion that English is fond of understatement is of course commonplace. Sometimes, however, the validity of this notion is disputed. For example, it was questioned by Drazdauskiene (1981:66), who noticed that strong positive stereotypical exclamations such as How lovely! or Isn't it lovely! are much more common in English speech than they are in Lithuanian speech. I would say that the same observation would apply to Polish: Polish, like Lithuanian, makes frequent use of negative (critical) exclamations but not of positive, enthusiastic ones. I would point out, however, that the English understatement applies to spontaneous opinions and feelings, not to opinions or feelings which are presumed to be shared. The stereotypical exclamations discussed by Drazdauskiene typically express enthusiastic appreciation for something which the speaker presumes to be shared by the addressee. They often sound exaggerated and insincere, and they certainly don't sound dogmatic. The speaker is not bluntly stating hislher own view, disregarding any potential dissent; on the contrary, he (or, according to the stereotype, she) is eager to agree with the addressee. It is of course highly significant that, as mentioned earlier, the stereotypical exclamations often take an interrogative form (Isn't that lovely?) or are followed by a symmetrical question asking for confirmation (How wonderful! Isn't that wonderful?) Drazdauskiene suggests that the difference between English and Lithuanian with respect to the use of stereotyped positive exclamations may be related to the fact that Lithuanians are reserved and restrained (and this view, expressed by a Lithuanian, certainly agrees with the Polish stereotype of Lithuanians). But Poles, unlike Lithuanians, are not regarded as restrained or reserved, and yet in this particular respect they seem to be closer to Lithuanians than to speakers of English. I suggest that exclamations under discussion do not point to any lack of emotional restraint on the part of the speakers of English. On the contrary: they are a conventional device aimed at 'being nice' to the addressee rather than any spontaneous and unrestrained outburst of the heart. In English, exclamations can take not only an affirmative and positive form, as in: How nice!

but also (especially in what tends to be regarded as more typically feminine speech) an interrogative-negative one, as in the utterance:

46

Different cultures, different languages, different speech acts

Isn't he marvellous! (Buzo 1979:41)

Thus, the function of such exclamations is similar to that of tag questions with an opposite polarity: Terrible place, isn't it?

Negative-interrogative exclamations do not always have an interrogative intonation, and do not always invite confirmation. Often, they are used simply to express the speaker's feeling, and are followed by a positive statement from the speaker rather than by a pause to be filled by the addressee: Bentley: Isn't she a sweetie? a real darling. (Buzo 1979:45) Sundra: Wasn't that funny? That was the funniest thing I've ever heard. (Buzo 1974:114) Sundra: Isn't that nice of them? I think that's very nice of them. (Buzo 1974:115) Sundra: Isn't that wonderful? I think that's wonderful. (Buzo 1974: 115) However, even when interrogative-negative exclamations are not used as a truly dialogic device they still signal (at least in a perfunctory way) an interest in what the addressee would say; they acknowledge the possibility that the addressee could say the opposite (even though the speaker regards this as unlikely) and symbolically seek confirmation. The speaker expects agreement, but does not take this agreement for granted, and 'graciously' leaves the addressees the opportunity to express their point of view, too. All this may of course be purely perfunctory, purely conventional, but the convention is there, and it has its own cultural significance. Characteristically, in Polish there is no similar convention. Exclamations always take a positive form: Jak gJupo! 'How stupid!' Wspaniale! 'Wonderful! '

The interrogative form would be interpreted as a genuine question.

Cultural values reflected in speech acts

47

4. Cultural values reflected in speech acts 4.1. Lexical evidence The cultural differences between English and Polish discussed here have also innumerable lexical reflexes. I will mention two of them here. One is the presence in the English lexicon of the word privacy, which has no equivalent in Polish, nor, apparently, in other European languages. In fact, the concept of privacy seems to be a characteristically AngloSaxon one. The word privacy is a very common one, frequently used in everyday speech, and it clearly reflects one of the central values of Anglo-Saxon culture. To have privacy means, roughly, 'to be able to do certain things unobserved by other people, as everyone would want to and need to'. The cultural assumption embodied in this concept is very characteristic: it is assumed that every individual would want, so to speak, to have a little wall around him/her, at least part of the time, and that this is perfectly natural, and very important. One is tempted to speculate, in this connection, that the absence of an intimate T-form of address (in the sense of Brown - Gilman 1972), which sets English apart from other European languages, is a reflex of the same attitude. The English you is of course very democratic, it is a great social equaliser, but it can also be seen as a distance-building device. This is not to say that the meaning of the English word you is analogous to that of a V-form in a language which does have a T-V contrast. But I think that in the absence of such a contrast the form you can't convey the intimacy signalled by the choice of a T-form. An intimate form allows the speaker to get psychologically close to the addressee, to penetrate the wall surrounding each individual. The English you keeps everybody at a distance. In Anglo-Saxon culture non-sexual body contact is heavily restricted, as compared, for example, with Slavic and Mediterranean cultures: people seldom touch one another, hug one another, kiss one another, or seldom even shake hands (see Triandis Triandis 1960). They also physically keep at a considerable distance from one another, as compared, for example, with Slavs (cf. for example Monahan 1983). The absence of an intimate T-form reflects and fosters the culturally expected psychological distance between individuals, the general need for psychological and physical 'privacy'. One might add that the cultural taboo on 'personal remarks' characteristic of Anglo-Saxon culture, and the existence in English of the set

48

Different cultures, different languages, different speech acts

expression personal remarks, with its negative implications, can be seen as another strategy for building a little protective wall around every individual. There is no similar expression in Polish, and there is no similar taboo in Polish culture. Of course English doesn't have the elaborate distance-building deferential devices of Far Eastern languages such as Japanese, Korean, Javanese or Thai, either. It is interesting to note that from the perspective of languages of this kind English may appear as a language highly sensitive to intimacy. (Cf. for example Hijirida - Sohn 1986:391.) But this is an illusion. American (and generally English) address forms such as Bob, Jim, Tom or Kate, have nothing to do with intimacy. It is true that they imply less 'distance' than, for example, Dr Smith, but they don't come anywhere near the potential intimacy of the Polish ty, or the French tu, or even the Japanese (2SG) kimi. They imply informality and friendliness, not intimacy. Intimacy implies an especially close personal relationship between the speaker and the addressee; and English has no devices to convey that. For example, at an Australian university a head of department, or a dean, may send a memo to all the members of the department or the faculty, signing it with a first-name form: Bob, or Bob Johnson; and at a meeting of a university committee, the members from different departments may introduce themselves to each other as Bob (Johnson) and Kate (Brown), and start addressing each other as Bob and Kate, from the very first meeting. This has nothing to do with intimacy. They can be friendly, informal, and familiar, but they are not claiming, in this way, any 'special relationship' with the addressee. (For further discussion, see Chapter 3, section 3.) The universal English you is of course less 'distant' than the deferential Japanese third person forms of address such as sensei 'teacher', or than the deferential Polish third person form of address such as Pan Profesor 'Mr Professor'; but it is also far less 'intimate' than the Japanese kimi or the Polish ty. Being the great equaliser, the English you keeps everybody at a distance - not a great distance, but a distance; and it doesn't allow anybody to come really close. The second lexical difference between English and Polish that I would like to comment on concerns the concept embodied in the English word compromise (in the sense of mutual concessions) and its Polish counterpart, kompromis. In English, the word is essentially neutral, and if it has any value connotations they would tend to be positive rather than negative. By contrast, the Polish word tends to be used with negative connotations. In any case, lexical and phraseological derivates

Cultural values reflected in speech acts

49

of kompromis unquestionably embody value judgments. Thus, pojse na kompromis 'accept a compromise' suggests a moral weakness, a deplorable lack of firmness, a sell-out of values. The adjective bezkompromisowy 'without compromise' (said of someone who would never accept a compromise) is emphatically positive: it is a word of high praise, like heroic, noble or immaculate. Thus, in the Polish cultural tradition, holding firmly to one's beliefs and making no concessions to those of others is a valued and desirable attitude. In the Anglo-Saxon tradition, similar attitudes would be regarded as dogmatic and inflexible, and would be viewed with disapproval. In fact, the word inflexible and its Polish literal counterpart nieugi~ty provide another example of the same kind: the English word has negative connotations, whereas the Polish one is highly positive. Polish has also the word niezJomny 'unbreakable', which is also a term of praise and has no equivalent in English. (For further discussion see Wierzbicka, to appear, chap. 6.)

4.2. Objectivism as a cultural value The complex of cultural attitudes which conditions every individual to be constantly aware of other people, other voices, other points of view, to see oneself as one individual among many, all of them equally entitled to their psychological space, their autonomy, their own peculiarities and eccentricities, leads to objectivism and anti-dogmaticism being regarded as important social and cultural values. I would venture to suggest that this objectivism may be reflected in peculiarly English ways of referring to oneself, and to one's own country, as it were from an external point of view. This can be illustrated with the characteristic expression this country (commented on in Doroszewski 1938:120). In Polish, it would be inconceivable to refer to one's own native land, one's ojczyzna 'fatherland' as ten kraj 'this country', as if it were one among many countries, where one just happens to be at a particular time. The Polish expression ten kraj could only be used with respect to a foreign country; if it was used with reference to one's own country it would mark the speaker as a psychological emigre. Similarly, in English it is possible to refer to one's own nation as this nation, especially in an elevated rhetorical style. In Polish one would say in similar circumstances nasz nar6d 'our nation' (cf. Nasz narod jak lawa, z wierzchu zimna i martwa, sucha i plugawa, 'Our nation is like

50

Different cultures, different languages, different speech acts

lava, at the surface cold and lifeless, dry and repellent,' Mickiewicz 1955 :21 0). To say ten nar6d 'this nation', would indicate a complete lack of identification with one's nation; to use this expression one would have to psychologically leave one's own nation. As a further example, consider the English expression (the) same here, referring to oneself, as in the following dialogue: Michael: I might just have a small claret. Carmel: Same here. (Williamson 1974:155-156) In Polish, the literal translation of same here or the same here would be simply incomprehensible (as a way of identifying the speaker). It seems to me that this inclination to look at oneself from outside, to be conscious of the existence of many different points of view, all of them equally valid (at least potentially), fits in very well with the other characteristic features of English speech described here.

4.3. Cordiality as a cultural value Throughout this chapter, the emphasis has been mainly on Anglo-Saxon cultural values, reflected in the English language. Polish has been presented mostly in negative terms, as lacking certain devices characteristic of English. I would like to say a few words to redress the balance. It would be ridiculous to suggest that English speech acts reflect certain cultural values whereas Polish speech. acts reflect nothing but an absence of those values. It goes without saying that in fact Polish reflects values characteristic of Polish culture. From an English speaker's point of view, Polish ways of speaking may appear to reflect dogmatism, lack of consideration for other people, inflexibility, a tendency to be bossy, a tendency to interfere, and so on. On the other hand, from a Polish speaker's point of view, English ways of speaking may be seen as reflecting a lack of warmth, a lack of spontaneity, a lack of sincerity. The central place of warmth, of affection, in Slavic as well as in Mediterranean cultures, is reflected, among other things, in the rich systems of expressive derivation, and in particular in the highly developed systems of diminutives, involving not only nouns, but also adjectives and adverbs. By contrast, in English, productive diminutive derivation hardly exists at all, despite the existence of isolated baby forms such as handies, doggie or birdie (one can say girlie but not *mannie; auntie but not *unclie, horsie but not *goatie, and so on).

Cultural values reflected in speech acts

51

The central role of 'warmth', of affection, in Polish culture (and in Slavic culture in general) is evidenced above all in the expressive derivation of personal names (which goes much further than anything one can find, for example, in Italian or Spanish). The topic is vast, and cannot be discussed here in detail. Let me just mention that one personal name, for example Anna or Maria, can have in Polish as many as ten different derivates, all commonly used with respect to the same person, each of them implying a slightly different emotional attitude, and 'emotional mood'. For example: Anna: Ania, Anka, Aneczka, Anusia, Anuska, Anusienka, Anulka, Anuchna; Anusifltko Maria: Marysia, Marysienka, Maryska, Marysiuchna, Marychna, Marys, Marysiulka, Marycha, Marysifltko

This is quite apart from a variety of forms such as Maryla, Mania, Marynia, Maryna, etc. (all from Maria), which are usually chosen from, for a particular person, on a more permanent basis. (For a detailed discussion of the semantics of expressive forms of names in Polish and in Russian, see Wierzbicka, to appear, chap. 7.) I would suggest that there are many subtle ways in which expressive derivation interacts with speech acts. The topic deserves a separate study. In this chapter, I will mention just two examples of this interaction. In Polish, warm hospitality is expressed as much by the use of diminutives as it is by the 'hectoring' style of offers and suggestions. Characteristically, the food items offered to the guest are often referred to by the host by their diminutive names. Thus, instead of asking: Would you like some more herring? Are you sure?

one might say in Polish: Wei jeszcze sledzika! Koniecznie! 'Take some more dear-little-herring-(DIM)! You must!'

The diminutive praises the quality of the food and minimises the quantity pushed onto the guest's plate. The speaker insinuates: 'don't resist! it is a small thing I'm asking you to do - and a good thing!' The target of the praise is in fact vague: the praise seems to embrace the food, the guest, and the action of the guest desired by the host. The diminutive and the imperative work hand in hand in the cordial, solicitous attempt to get the guest to eat more. Certainly, the cultural style of such offers is very different from that of Would you like some more?, but the difference

52

Different cultures, different languages, different speech acts

cannot be described in terms of politeness. Rather, it has to be described in terms of different cultural traditions, and, ultimately, different hierarchies of values. If one's own view of what is good for another person does not coincide with the view of that person, Anglo-Saxon culture requires that one should rather respect the other person's wishes (i.e., autonomy) than to do what we think is good for the person; Polish culture tends to resolve the dilemma in the opposite way. A similar dilemma is involved in leave-taking behaviour: if the guests indicate that they are about to leave, should one let them go or should one try to prevent them from leaving? In Anglo-Saxon culture, one usually lets them go, acknowledging in this way their autonomy and 'self-determination'. In Polish culture, however, such behaviour would be seen as cold and uncaring; usually, therefore, one tries to prevent the guests from leaving, since a display of warmth towards the addressees is perceived as more important than a display of respect for their autonomy. An Anglo-American or Australian host, therefore, would normally thank the guests for coming and let them go, whereas a Polish host would insist that the guests must stay longer, and would shower them with 'you must's and with warm diminutives at the same time:

Ale jeszcze troszeczk~! Ale koniecznie! 'But [stay] a little-DIM more! But you must! As a third example of the interaction between diminutives and illocutionary strategies I will mention requests. In Polish, a request formulated in the imperative mood would often be softened by means of a diminutive. Thus, while it would be more natural for a wife to use an imperative than an interrogative-cum-conditional request when speaking to her husband, she would be likely to soften that imperative by a double diminutive form of his name (as well as by intonation):

Jureczku, daj mi papierosa! 'George-DIM-DIM, give me a cigarette!' An indirect interrogative request would be less appropriate in this situation because 'interrogativity' in directives is a distance-building device: there is an implicit conflict between intimacy and affection on the one hand and complete mutual independence on the other. (If I ask you to do something for me, and if I think we are close, I will assume that you will do what I want you to do; to show that I don't know if

Cultural values reflected in speech acts

53

you'll do it is to acknowledge your independence, but also, your 'distance' from me.) Similarly, in speaking to a child one would be unlikely to use an interrogative request (could you, would you be so good as to). Normally, one would use an imperative. But this imperative would be likely to be softened not only with a multiply diminutive form of the name, but also with numerous other diminutives, on nouns, adjectives, adverbs, and occasionally some other parts of speech: Monisienko, jedz zupk~! 'Monica-DIM-DIM, eat your soup-DIM!' Jedz pr~ciutko! 'Eat quickly-DIM!' Zjedz wszysciutko! 'Eat it all-DIM up!'

Rich systems of diminutives seem to playa crucial role in cultures in which emotions in general and affection in particular is expected to be shown overtly. Anglo-Saxon culture does not encourage unrestrained display of emotions. In adult English speech diminutives (even those few diminutives which English does have) feel out of place, just as non-erotic kissing and hugging feels more often than not out of place. It is fascinating to note, in this connection, that in comparison with say Japanese culture, Anglo-Saxon culture in general and American culture in particular emerges as one which greatly encourages physical expressiveness. Barnlund (1975a:445) reports a "dramatic contrast between the [American and Japanese] cultures" in this respect. "Touching behavior is reported nearly twice as often in all categories and with all persons by Americans as by Japanese." (1975a:452). On the other hand, American students of Russia and things Russian are amazed by the amount of touching, kissing and hugging which visibly takes place among the Russians (cf. Smith 1976:136; Monahan 1983). From a Polish perspective, Anglo-Saxon culture in general (including American culture) seems as restrained in physical expressiveness as Japanese culture seems to Americans. Most observers seem to agree that the Poles are not quite as effusive as the Russians, but, for example, kissing, hand-kissing and hand-shaking in greetings take place on a daily basis. The overtones which the word emotional has acquired in English are a good illustration of the disapproval of public display of emotions,

54

Different cultures, different languages, different speech acts

characteristic of Anglo-Saxon culture. Frequently this word is used with negative connotations, but even when it is not it implies at least 'an unexpected and somewhat embarrassing display of emotions' . For example, when an abducted baby was returned, after two days, to his mother, who thought she would never see him again, the Australian reporter (ABC News, 24.8.1983) described the mother's behaviour as 'emotional' ("The baby was reunited with his emotional mother"). In this particular context, the word emotional is not used as a criticism, since the mother's 'emotional state' is apparently seen as something that can be understood and excused. Nonetheless, from a Polish speaker's point of view, the very need to mention and to excuse the mother's emotion would seem odd: it would not occur to one that a mother could do other than show emotion in a situation of that kind. As pointed out by Lutz (1986:290-301) the "widely shared American [and, I would add, Anglo-Saxon in general - A.W.] ethnotheory of basically Protestant European, middle class background" identifies "emotion primarily with irrationality, subjectivity, the chaotic and other negative characteristics". "One of the most pervasive cultural assumptions about the emotional is that it is antithetical to reason or rationality"; "emotions are fundamentally devalued ... as irrational, physical, unintentional, weak, biased, and female"; "emotions tend predominately [sic] to lead to erroneous judgements and hence senseless or irrational actions. ... people tend to see emotion as a disruption of, or barrier to, the rational understanding of events. To label someone emotional is often to question the validity, and more, the very sense of what they are saying." Not so in Polish culture. In the romantic poetry which played a fundamental role in shaping Polish national ethos, serce 'heart' is opposed to the scientist's szkielko i oko 'magnifying glass and eye', as a source of 'live truth' versus the domain of 'dead truths', and this opposition has retained an important place in the Polish ethnotheory. The fact that the Polish counterpart of the English word emotional, that is, uczuciowy, has positive connotations, reflects this. Uczuciowy does not designate someone who shows emotion (because there is no cultural expectation that feelings would or should not be shown), but rather someone who possesses rich and strong emotions (seen as a 'good thing'). It must be stressed, however, that the Anglo-Saxon taboo on 'emotions' does not concern all feelings to the same degree. For example, as mentioned earlier, in Australian culture it is quite all right to swear, that is to show 'strong', 'masculine' feelings. What is not all right is to show, without restraint, 'weak', 'soft', 'feminine' emotions, such as tenderness.

Cultural values reflected in speech acts

55

Lutz (1986:299) points to the Anglo-Saxon (she says, American) distinction between emotions seen as typical of, and forgivable in, women, and those which can be expected of men. "American cultural belief does not deny that men may become emotional; it does, however, engender expectations that men will experience only certain types of emotion, notably anger. Women are expected to experience the entire range of emotions more frequently and deeply, with the possible exception of anger". In Australian culture, which highly values 'toughness' and antisentimentality, and where the word bloody is the main vehicle for expressing emotions (both negative and positive ones), any display of 'soft', 'feminine' emotions is particularly abhorred. (Cf. Wierzbicka, to appear, chap. 11.) It is worth noting in this connection that characteristically Australian abbreviations, such as mozzies (mosquitoes), mushies (mushrooms), prezzies (presents), barbie (barbecue), lippie (lipstick), or sunnies (sunglasses), which are often referred to as diminutives, in fact are not really diminutives and have a function quite different from the main function of diminutives (although it is of course a simplification to speak of diminutives as if they had only one function). Formally, they differ from English diminutives because they are abbreviations: baby words such as birdie, fishie or doggie add a diminutive suffix to the full form of the base word, but words such as barbie or lippie add a suffix to a truncated form of the base word. Semantically, they differ from diminutives in expressing, essentially, not endearment but good humour. The core meaning of true diminutives (such as doggie) can perhaps be represented as follows (cf. Wierzbicka (1980, 1984; to appear): doggie =>1 I think: this is something small like you are someone small I feel something good towards you because of this, when I say something about this to you I feel something good towards it

The core meaning of Australian abbreviations with the suffix -ie is different. I would represent it as follows (cf. Wierzbicka 1984; to appear): mozzies => I think: this is something small I think: you think the same when I say something about this to you I feel something good

56

Different cultures, different languages, different speech acts

Thus, calling mosquitoes mozzies, the speaker is good-humouredly dismissing the problem; he thinks of mozzies as small-(but not endearing), and expects that the addressee would share this attitude. As I have suggested elsewhere, the semantic complex explicated above reflects many characteristic features of the Australian ethos: anti-sentimentality, jocular cynicism, a tendency to knock things down to size, 'mateship', good-natured humour, love of informality and dislike for 'long words' (Slavic or Romance diminutives are typically much longer than the base words, but Australian abbreviations are normally shorter than the base words, and Australians feel that this formal brevity is somehow functional) . As another linguistic reflex of the same Australian attitudes, and in particular of the Australian non-sentimental good humour, I would mention the quintessentially Australian expression no worries, which permeates Australian speech and which serves a wide range of illocutionary forces. The casual optimism encapsulated in this expression and also in the Australian abbreviations is something quite different from the warmth of Slavic diminutives.

4.4. Courtesy as a cultural value I think it is important to add that while Polish culture shares one major theme of Slavic culture in general, cordiality, it combines it with a different one: courtesy, in the sense of a somewhat ceremonial show of respect for every individual person (and especially for women). There is in Polish culture, alongside cordiality and spontaneity, an element of ceremony, of somewhat ritualised courtesy and chivalry. The Polish custom of kissing a lady's hand (by men) is a characteristic example of this: vigorous warm kisses on both cheeks signal cordiality, but one kiss on a lady's hand signals both cordiality and ceremonial courtesy. Courtesy is not in conflict with cordiality, but it imposes on it certain ritual forms, a certain ceremoniality. The courtesy aspect of the Polish savoir vivre is manifested particularly clearly in forms of address. 2 As mentioned earlier, in English everybody (except perhaps the Queen) can be addressed in the same way, as you. In Polish, one always distinguishes the intimate ty 'thou' from the courteous pan/pani 'sir', 'madam' (with the verb in third person singular). The English you is democratic, the same for everyone; it lacks

Cultural values reflected in speech acts

57

both the (potential) intimacy of the Polish form ty and the courteousness of the Polish forms pan/pani. This link between courtesy and cordiality is interesting to note because it seems to be, typologically, rather unusual. Ceremony and ritual may seem to be antithetical to spontaneity and 'emotionality', and cultures which favour the former usually restrict the latter. The Japanese and Javanese cultures are cases in point (see Benedict 1947; Lebra 1976; Smith 1983 on Japanese culture; Geertz 1976 on Javanese). But Polish culture distinguishes sharply between spontaneity and emotionality on the one hand and informality on the other. Like Japanese, Polish is very fond of titles, and the list of titles commonly used goes far beyond the 'Doctor', 'Professor' or 'Father', commonly used in English. For example, one says commonly Panie Dyrektorze 'Mr. Director', Panie Naczelniku 'Mr. Head', Panie Iniynierze 'Mr. Engineer', Panie Magistrze 'Mr. MA-holder' (usually said to a pharmacist, who holds an MA in pharmacy), Panie Mecenasie 'Mr. Barrister', and so on. But unlike in Japanese, in Polish the 'language of respect' doesn't involve humility and self-abasement: one pays respect to the status and rank of the addressee without ever lowering oneself. Furthermore, this respect for the addressee is commonly combined in Polish with cordiality and affection. The compatibility between courtesy and cordiality is best seen in forms of address or of personal reference which combine formal titles pan 'Mr.', pani 'Mrs.' and panna 'Miss' with affectionate diminutive forms of personal names, such as Panie Mareczku 'Mr Mark-DIM' or Pani Basienko 'Mrs Barbie-DIM'. Polish dislikes informality (which is so characteristic, for example, of Australian English), and it encourages the use of titles even between 'equals' who know each other very well, and who have known each other for years (for example, between workmates). At the same time, however, the formality of such forms of address does not prevent the show of emotion, and affectionate diminutives of first names are freely combined with titles, as they are with hand-kissing. Polish differs in this respect from Russian, which has also a wealth of devices for showing emotion, but which is not similarly rich in devices for showing courtesy, and which links affection with informality. To show respect, courtesy, and non-intimacy one uses in Russian a combination of full first name and patronymic, and normally the patronymic cannot be combined with an affectionate diminutive. (Cf. Wierzbicka, to appear, chaps. 7,8.)

58

Different cultures, different languages, different speech acts

Apart from names and names with patronymics, Russian has two basic forms of address: ty 'thou' and vy 'you' (PL), one of which signals 'intimacy' and the other 'distance'. But the Russian form vy doesn't correspond exactly to the Polish forms pan/pani, because it signals only 'distance', not courtesy. In Warsaw shops one sometimes encounters Russian women tourists, delighted and amused to be addressed as pani 'Madam' - a form which they perceive as quaintly courteous and ancien regime. The absence of a special courtesy value in the Russian form vy makes it suitable for use among party apparat and police as well as among ordinary people. In communist Poland, police and party apparat avoided the forms pan/pani, whose 'aura' didn't fit the communist ideology. Characteristically, the communist regime in Poland attempted for many years to eradicate these forms, replacing them with wy (on the Russian model). (Cf. Davies 1981,2:581.) These efforts, however, proved futile. The fact that in the documentary film 'Workers 1980' the representatives of the Government, talking to Lech Wal~sa and other representatives of the workers, used publicly the forms pan/pani, was widely commented on in Poland, as a kind of symbolic recognition of the defeat of efforts aiming at eradicating the Polish tradition of courtesy. Following on Brown - Gilman (1972), different forms of address such as ty vs. pan/pani in Polish are usually described in terms of 'power and solidarity' (see, however, Ervin-Tripp 1974). I would suggest, however, that as far as Polish is concerned, it is more illuminating to refer here to cultural values such as intimacy and courtesy. The forms pan/ pani differ from the so-called V-forms of languages such as Russian in having positive courtesy built into them. The form wy (second person plural), favoured by the communist regime, carried with it implications of impersonal equality, as well as distance. To the Polish ear, it sounded cold, impersonal and discourteous. It de-emphasised personal ties (either intimate, signalled by ty, or based on mutual respect, signalled by pan/ pani) in favour of equality derived from membership in a collectivity. Pan/pani, on the other hand, is non-intimate, but it is also courteous and personal. I presume that the 'personal' character of pan/pani is due partly to its singular form, and possibly also to its sex differentiation, whereas the 'impersonal' character of the form wy is due partly to its plural and genderless form. Polish courtesy stresses respect for every individual as an individual, and is highly sex-conscious. The collectivist and genderless ring of the form wy is jarring, in that tradition.

Theoretical implications

59

One might add that, in communist Poland, the officially-supported form wy co-occurred with 'collectivist' vocatives and appellatives such as towarzyszu, towarzysz 'comrade', and, to a lesser extent, obywatelu, obywatel 'citizen', and was no doubt interpreted in conjunction with, or against the background of, Soviet-style forms. The appeal to Sovietstyle equality conveyed by the official wy was backed by an explicit or implicit reference to a collectivity of 'comrades', that is, ideologically committed equals. In Polish dialects, the form wy has a different origin and a different function: it is opposed to ty only (not to ty on the one hand and to pan/ pani on the other), and it expresses not equality but respect. Significantly, it doesn't co-occur there with any collectivist and ideologically loaded forms of address such as towarzysz 'comrade'. Rather, it cooccurs with terms referring to the addressee's personal status, such as 'mother' or 'uncle', or with first names, usually in a 'dignified', nondiminutive form. I would add that the contrast between the courteous, Polish-style form pan/pani and the impersonal, Soviet-style form wy is something that Poles are acutely aware of and often comment on. To illustrate this general awareness of the semantic implications of the two forms, I quote a characteristic passage from an essay which was published in the leading Polish emigre monthly, Kultura: When the Russians speak of us ironically as te polskie pany ['those Polish gentlemen'], the connotations are of culture rather than class. The gentry as a class has long since ceased to exist, but we are still 'gentry' because we didn't submit to Soviet attempts at 'Gleichschaltung', at 'comradising' us, and the form wy ['you PL'] didn't take. In communist Poland the only contrast really felt is that between panowie ['gentry', but also 'misters'] and those who are generally referred to as oni ['they', i.e. the regime people, the new ruling class]. (Schrett 1984:7)

5. Theoretical implications In the literature on speech acts, English conversational strategies discussed here are frequently interpreted as a manifestation of a universal 'natural logic' (Gordon - Lakoff 1975), a universal 'logic of conversation' (Grice 1975) or universal rules of politeness (Searle 1975). In the

60

Different cultures, different languages, different speech acts

light of the facts discussed in this chapter this line of interpretation must be seen as highly ethnocentric. Consider for example, the following statement: . .. ordinary conversational requirements of politeness normally make it awkward to issue flat imperative sentences (e.g. Leave the room) or explicit performatives (e.g. I order you to leave the room), and we therefore seek to find indirect means to our illocutionary ends (e.g. I wonder if you would mind leaving the room). In directives, politeness is the chief motivation for indirectness. (Searle 1975:64)

I hope I have shown that it is an illusion to think that 'ordinary conversational requirements of politeness make it awkward to issue flat imperative sentences'. It is not an 'ordinary' requirement, it is an English requirement. Similarly, the rule that 'it is awkward to issue explicit performatives' is an English conversational requirement, not a universal one. The awkwardness of the utterance quoted at the outset: Please! Sit! Sit!

stems precisely from the fact that Polish does not share the two conversational requirements mentioned by Searle as 'ordinary'. (Please is a rough equivalent of the Polish word prosz~, which literally means I ask. In P.olish speech, the performative form meaning 'I ask' is simply ubiquitous, even more so than the highly colloquial form radz~ ci 'I advise you'.) Furthermore, considering sentences such as: Will you bloody well hurry up? Why don't you shut your mouth?

one wonders how much explanatory force can be attributed to the claim that 'politeness is the chief motivation for indirectness', even if one limits this claim to English. (Cf. Ervin-Tripp 1976:59-61.) From the data discussed in this chapter, it emerges that what is at issue is neither universal rules of politeness nor even English-specific rules of politeness. What is really at issue is English conversational strategies, and Anglo-Saxon cultural values. In an interesting study of politeness markers in English and German, House - Kasper (1981:184) have observed that "on the whole, the German speakers selected more direct levels for both complaint and request acts". The authors comment on this difference as follows: "From an etic standpoint, then, the behavior of the German speakers may well

Theoretical implications

61

be considered impolite by reference to an English norm; however, from an emic standpoint, which is the one we would prefer here, one would simply claim that the differential behavior displayed by the German and English speakers may be a reflection of the fact that the two cultural systems are organised differently, and that, for example, a level 6 complaint in the German culture is not necessarily comparable to a level 6 complaint in the English culture, because the value of each is derived from the value it has relative to the remaining levels, and their frequency and modality of use in the particular cultural system." (1981: 184). To my mind, however, the significance of the differences observed goes much deeper than that. What is at issue is not just different ways of expressing politeness, but different cultural values. As I see it, the crucial fact is that different pragmatic norms reflect different hierarchies of values characteristic of different cultures. Commenting on the form Can you, Searle (1975:74-75) says: "Firstly, X does not presume to know about Y's abilities, as he would if he issued an imperative sentence; and, secondly, the form gives - or at least appears to give - Y the option of refusing, since a yes-no question allows no as a possible answer. Hence, compliance can be made to appear a free act rather than obeying a command." This is all true and insightful. It is an illusion, however, to think that the norms referred to in this passage have the same weight in all cultures. Searle is not unaware that "there are differences in the indirect speech forms from one language to another", but he regards such differences as idiomatic, due to accidental variation (1975:76). He explains: The mechanisms are not peculiar to this language or that, but at the same time the standard forms from one language will not always maintain their indirect speech potential when translated from one language to another. ... within the class of idiomatic sentences, some forms tend to become entrenched as conventional devices for indirect speech acts. In the case of directives, in which politeness is the chief motivation for the indirect forms, certain forms are conventionally used as polite requests. Which kinds of forms are selected will, in all likelihood, vary from one language to another. (Searle 1975:76-77)

But this makes it sound as if the variation were more or less random and accidental, whereas the general mechanisms were universal. In fact, as I have tried to show, specific differences between languages in the area of so-called 'indirect' speech acts are motivated, to a considerable degree, by differences in cultural norms and cultural assumptions,

62

Different cultures, different languages, different speech acts

and the general mechanisms themselves are culture-specific. (Cf. Hollos - Beeman 1978:353-354.) This is not to deny that the generalisations suggested in works such as Grice (1975), Gordon - Lakoff (1975) or Searle (1975) provide useful insights into mechanisms of language use. It is important, however, that generalisations of this kind should not be seen as absolute. 'Natural logic' provides a considerable range of options. The choices embodied in individual languages reflect not only 'natural logic' , and not only a combination of 'natural logic' with historical accidents. They reflect also what Gumperz (1982:182) aptly calls 'cultural logic'. Searle insists that interrogative English sentences such as: C an you pass the salt(?) Would you pass me the salt(?) Will you pass me the salt(?)

are not ambiguous (between question and request), but that by virtue of their meaning they are simply questions (even when they are uttered with intonation characteristic of directives, cf. Searle 1975:69). If they are interpreted as requests, that is by virtue of the hearers' "general powers of rationality and inference" (Searle 1979: 176). But to say this is to imply that speakers of languages such as Polish are sadly lacking those 'powers of rationality and inference'. Poles learning English must be taught the potential ambiguity of Would you sentences, or Why don't you sentences, just as they must be taught the polysemy of the word bank. Searle might say that what they have to be taught is not meaning but 'conventions of usage' (cf. Searle 1975:76). But this distinction between meaning and conventions of usage becomes meaningless if the ignorance of the relevant 'conventions of usage' leads not just to un-idiomatic speech but to simple misunderstanding of what Searle himself would recognise as meaning. For example, if Polish newcomers to Australia interpret sentences such as: How about a beer? Why don',t you come and have lunch with us?

as genuine questions, rather than as an offer and an invitation, they are making a semantic error just as much as when they interpret the utterance How do you do? as a genuine question. It is essential to recognise that what is involved is not any differences in 'powers of rationality and inference', but differences in 'cultural logic', encoded in language:

Theoretical implications

63

The fact that two speakers whose sentences are quite grammatical can differ radically in their interpretation of each other's verbal strategies indicates that conversational management does rest on linguistic knowledge. But to find out what that knowledge is we must abandon the existing views of communication which draw a basic distinction between cultural or social knowledge on the one hand and linguistic signalling processes on the other. We cannot regard meaning as the output of non-linear processing in which sounds are mapped into morphemes, clauses and sentences by application of the grammatical and semantic rules of sentence-level linguistic analysis, and look at social norms as extralinguistic forces which merely determine how and under what conditions such meaning units are used. (Gumperz 1982:185-186)

I would add that descriptions of 'cultural logic', to be helpful, must be done in fairly specific terms. It is worth noting in this connection that in numerous studies written by Western scholars and concerning nonWestern cultures epithets such as 'direct' or 'blunt' are used to refer to the Anglo-Saxon cultural norms, whereas, by contrast, the other cultures studied often appear to value 'indirectness' (cf. for example Geertz 1976; Eades 1982). In the present study, the reverse is the case: by comparison with Polish, the English ways of speaking appear to be highly 'indirect'. This shows, however, that terms such as 'directness' or 'indirectness' are much too general, much too vague to be really safe in cross-cultural studies, unless the specific nature of a given cultural norm is spelt out. The present study shows that English cultural norms (as compared with Polish norms) favour 'indirectness' in acts aiming at bringing about an action from the addressee. On the other hand, studies such as Eades (1982), Sansom (1980) or Abrahams (1976) show that Anglo-Saxon cultural norms (as compared with Australian Aboriginal norms, or with Black American norms) encourage 'directness' in seeking information from the addressee. Evidently, the Anglo-Saxon principle of non-interference, which accounts for the heavy restrictions on the use of the imperative, doesn't extend to questions (I don't mean 'personal questions', but questions in general) - presumably, because information is seen in Anglo-Saxon culture as a free and public good. In fact, the restrictions on the use of the imperative seem to be compensated by a tremendous expansion of interrogative devices. Similarly, Geertz (1976:240-248) stresses the 'indirection' and 'dissimulation' characteristic of Javanese culture, and contrasts these features with those characteristic of American culture. According to Geertz's classical study, Javanese culture favours "beating about the

64

Different cultures, different languages, different speech acts

bush", "not saying what is on one's mind', "unwillingness to face issues in their naked truth", "never saying what one really thinks", avoiding "gratuitous truths", "never showing one's real feelings directly" and so on. Clearly, all these forms of 'indirection' are rather different from those cultivated in Anglo-Saxon culture (especially, the dissimulation of truth). It seems to me, therefore, that it is very important to try to link language-specific norms of interaction with specific cultural values, such as autonomy of the individual and anti-dogmaticism in AngloSaxon culture or cordiality and warmth in Polish culture. The issues involved are of fundamental importance, and they merit a more general discussion; I attempt to undertake such a discussion in Chapter 3.

6. Practical implications In a multi-ethnic country like Australia, or like the United States, the problem of speech acts and of their cultural significance is not a purely academic one. It is a problem of immense practical significance. As long as it is widely assumed that English conversational routines reflect what is 'ordinary', 'normal', 'natural' and 'logical', the prospects for cultural understanding between immigrants and the Anglo-Saxon population are not particularly bright. Anglo-Saxon institutions such as schools, courts or government departments, as well as the streets and 'market places' are, inevitably, an arena of cultural clashes and cultural misunderstandings. If immigrants who speak passable English tend to utter flat imperatives, they are likely to be seen as rude or boorish. If they fail to respond to pieces of elaborate 'indirection', they are likely to be seen as uncooperative, or dumb. Elaborate indirectness accompanied by juicy swearing can be as confusing to an immigrant as the directness, forcefulness and 'emotionality' of some immigrants can be offensive and irritating to an 'Anglo'. Anglo-Saxon doctors and nurses (as Jane Simpson has pointed out to me) are accustomed to thinking that pain should be borne stoically, and that one should only cry in real extremity. Therefore they are unsympathetic to people who complain, cry and scream at pains which can be considered minor, behaviour acceptable to Italians and Greeks. This can lead to very unsympathetic treatment by doctors and nurses, and to a

Practical implications

65

general idea that Mediterranean peoples are cowardly because they complain about things that only hysterical cowardly Anglo-Saxons would mention. I have heard similar comments from Australian nurses, during two seminars on linguistic problems of immigrants which I gave to nurses in two Canberra hospitals in 1983. A number of nurses commented on the unsympathetic attitude of Anglo-Saxon doctors towards immigrant women screaming in childbirth, and on the fact that often injections are administered merely to stop the screaming. An immigrant woman who screams, cries or complains, is seen as hysterical or unbalanced. The taboo on showing pain is clearly related to the taboo on showing emotions. Obviously, cultural clashes of this kind cannot be completely eliminated, but they can be minimised by enlightened, well-planned multicultural education. It seems clear that a linguistic study of culturespecific speech acts and speech styles has a great deal to contribute in this domain.

Chapter 3

Cross-cultural pragmatics and different cultural values

Anyone who has lived for a long time in two different countries knows that in different countries people speak in different ways - not only because they use different linguistic codes, involving different lexicons and different grammars, but also because their ways of using the codes are different. Some of these differences are so stable and so systematic that one cannot always draw a line between different codes and different ways of using the code; or between different 'grammars' and different 'ethnographies of speaking' (cf. Hymes 1962). The extent of the differences between different societies and different language communities in their ways of speaking is often underestimated in the literature dealing with language use. In particular, theories of speech acts and of conversational logic associated with, or following from, the work of philosophers such as John Searle (1969, 1979) and Paul Grice (1975, 1981) have tended to assume that the ways of speaking characteristic of mainstream white American English represent 'the normal human ways of speaking', and that, apart from minor variations, they can be expected to be the same as those prevalent in any other human society. But this is of course an ethnocentric illusion. The search for universals in language usage at the expense of culturespecifics is also a feature of the influential study of 'politeness phenomena' by Brown - Levinson (1978; revised edition 1987). There would of course be nothing wrong in focussing on universals rather than on culture-specific aspects of language usage - if the search for universals is undertaken from a truly universalist, culture-independent position. But as a number of recent studies have shown, the basic conceptual tools introduced and relied on by Brown· and Levinson (in particular, the notion of 'face') have in fact a strong anglocentric bias (cf. for example Matsumoto 1988; Katriel 1986; Tannen 1984; Wierzbicka 1985a, b). Brown - Levinson see two principles as the most important ones in human interaction: 'avoidance of imposition' ('negative face') and 'approval of the other person', which they exemplify with the English compliment What lovely roses! ('positive face'). But their very choice

68

Cross-cultural pragmatics and different cultural values

of these particular parameters reflects clearly the authors' culturespecific (anglocentric) perspective. The same charge of anglocentrism can be made with respect to various other supposedly universal 'maxims' and principles of human conversational behaviour and interaction, which have been advanced in the literature. Consider, for example, Leech's (1983:132) maxims of 'modesty' and of 'approbation': Approbation maxim (a) Minimise dispraise of other; [(b) Maximise praise of other.] Modesty maxim (a) Minimise praise of self; [(b) Maximise dispraise of self.] Leech is aware that the weight of maxims such as these may vary from culture to culture, but he assumes that apart from quantitative differences they are in essence universally valid. In fact, however, empirical evidence suggests that this is simply not true. For example, Kochman (1981) has shown that in Black American culture the norm of 'modesty' does not apply, and that self-praise is not viewed negatively at all. Kochman mentions in this connection the title of Mohammed Ali's autobiography: I am the greatest, and he discusses the significance of Black folk categories such as 'rapping', 'grandstanding', and 'showboating' (I return to this matter in section 1.5 below). Similarly, Mizutani - Mizutani (1987) show that 'approbation' or 'praise of other' is not encouraged in Japanese culture; and they devote a whole section (1987:45-46) to "refraining from direct praise". Likewise, Honna - Hoffer (1989:74) point out that 'praise of other' is seen as arrogant and presumptuous in Japanese culture, where "even when [the speaker] has to or wants to express his praise for persons within his circle, he often begins with a phrase such as 'I don't really mean to praise ... ' or 'I know it is too presumptuous to praise ... '. By so doing he tries to give the impression that he is not really an arrogant person." It is not true, then, that all human societies view 'praise of self' negatively, and 'praise of other' positively. The same applies to the supposedly universal maxims of harmony: "minimise disagreement, maximise agreement" (Leech 1983: 132). For example, as Schiffrin (1984) has shown, Jewish culture displays a clear preference for disagreement: in this culture, people show their involvement with other people and their interest in other people by saying 'no' rather than 'yes'. In Jewish culture, argument is valued as a form of

Cross-cultural pragmatics and different cultural values

69

sociability, and it is disagreement rather than agreement that is seen as something that brings people closer together (see section 2.1 below). It is, then, an anglocentric illusion to think that all cultures value agreement more than disagreement, discourage self-praise, encourage praise of other, and view 'imposition' as the main sin in social interaction. The last decade has witnessed a growing reaction against this kind of misguided universalism, a reaction which has led to the emergence of a new field and a new direction in language studies associated with the term 'cross-cultural pragmatics' (cf. for example Abrahams 1976; Ameka 1987; Eades 1982; Goddard 1985; Harkins 1988; Hijirida Sohn 1986; Katriel 1986; Kochman 1981; Mizutani - Mizutani 1987; Ochs 1976; Schiffrin 1984; Sohn 1983; Tannen 1981a; Wierzbicka 1985a, b). The main ideas which have informed and illuminated this new direction in the study of language are these: (1) In different societies, and different communities, people speak differently. (2) These differences in ways of speaking are profound and systematic. (3) These differences reflect different cultural values, or at least different hierarchies of values. (4) Different ways of speaking, different communicative styles, can be explained and made sense of, in terms of independently established different cultural values and cultural priorities. These four points are, in my view, of fundamental importance - not only from the point of view of our knowledge and understanding of the world, but also from a practical, social point of view; and in particular, from the point of view of cross-cultural understanding in a multi-ethnic society such as the United States or Australia. Consider, for example, the situation of Australians of Anglo-Saxon or Anglo-Celtic background who note that some immigrants behave verbally in what appear to be strange, unfamiliar ways. For example, they seem to shout and scream for no reason at all, they interrupt other people, they start heated arguments for no apparent reason, they speak in what is perceived as a blunt, dogmatic and bossy way, they flatly assert their opinions and flatly contradict other people, and so on. If 'strange' and possibly offensive behaviour of this kind can be explained, and made sense of, in terms of independently understandable

70

Cross-cultural pragmatics and different cultural values

cultural values, serious social and interpersonal problems can be resolved, and serious conflicts prevented or alleviated. Of course, not all problems can be solved in this way: if there is a real conflict in underlying values, mere explaining will not help. But in many cases, perhaps in most cases, what is involved is not a real conflict in values but a difference in the hierarchy of values; and when this is the case, explaining can help. It can only help, however, if it is done in a way which is intelligible to the target audience. And this is, I believe, where cross-cultural pragmatics often fails. Even the most enlightened studies in cross-cultural pragmatics (such as for example Kochman 1981; Sohn 1983; Lebra 1976) tend to explain different cultural priorities associated with different languages (or different dialects and sociolects) in ways which are not, and which cannot be, comprehensible to people of different cultural backgrounds. The crux of the matter lies in the language in which the explanations are couched. What usually happens is that researchers in cross-cultural pragmatics try to explain differences in the ways of speaking in terms of values such as 'directness' or 'indirectness', 'solidarity', 'spontaneity', 'sincerity', 'social harmony', 'cordiality', 'self-assertion', 'intimacy', 'self-expression', and so on, without explaining what they mean by these terms, and using them as if they were self-explanatory. But if one compares the ways in which different writers use these terms, it becomes obvious that they don't mean the same things for everyone. In fact, the intended meanings are often not only different but mutually incompatible. As a result, the same ways of speaking are described by some authors as 'direct' and by others as 'indirect'; as a manifestation of 'self-assertion' or an absence of 'self-assertion'; as an expression of individuality or suppression of individuality. This leads to total confusion, and to an absence of any consensus, even on the most basic points. For example, in the literature on Japanese culture and society, Japanese ways of speaking are often described as 'indirect' and are contrasted with the English ways of speaking, which are supposed to be more 'direct'. It is also claimed, or even assumed, that English ways of speaking are characterised by a high degree of self-assertion, whereas in Japanese self-assertion is avoided and suppressed. It is also said that English ways of speaking reflect high regard for sincerity and spontaneity, whereas Japanese ways of speaking discourage sincerity and spontaneity, preferring to them courtesy and consideration for others.

Cross-cultural pragmatics and different cultural values

71

On the other hand, in the literature on (American) Black English, the 'standard' (white) English is presented in the opposite way. Here, it is said, and even assumed, that standard English is 'indirect' rather than 'direct', that it avoids self-assertion, and that it discourages sincerity and spontaneity. It is Black English which is said to be 'direct', and to favour self-assertion, sincerity and spontaneity. Similarly, in the literature on Jewish culture, on the Yiddish language and also on Israeli Hebrew, Yiddish and Hebrew are presented as 'direct', as bent on selfexpression and self-assertion, and as favouring sincerity and spontaneity, whereas English is presented as associated with the suppression of all these values. At first, one might think that conflicting assertions of this kind are due simply to differences of degree: perhaps English (that is, standard white English) is more 'direct' or more 'self-assertive' than Japanese but less so than Black English or than Israeli Hebrew. But when one examines the data adduced in support of the conflicting generalisations, one discovers that this is not the case, and that in fact the differences referred to are qualitative rather than quantitative. For example, what is called 'self-assertion' in the studies of Black English is not the same thing that is usually meant by this term in the studies of Japanese; and the same applies to 'self-expression', 'sincerity', 'spontaneity', 'solidarity' and so on. I conclude from this that labels of this kind are simply not helpful in the elucidation of cultural differences. Labels of this kind are semitechnical and obscure at the same time. They are used differently by different writers because they have no clear or self-evident meaning. They are also highly anglocentric, as they have no exact equivalents in other languages. For example, Japanese has no words corresponding to sincerity. The two Japanese words which are usually translated as 'sincerity', magokoro and makoto, mean in fact something very different from sincerity, as Ruth Benedict (1947) among others has clearly demonstrated. Nor does Japanese - or, for that matter, Polish, Italian, French or Russian - have a word for self-assertion. It seems obvious that if we want to compare "different cultures in terms of their true basic values, and if we want to do it in a way that would help us to understand those cultures, we should try to do it not in terms of our own conceptual artefacts (such as the English terms self-assertion or sincerity) but in terms of concepts which may be relevant to those other cultures as well - that is, in terms of concepts which are relatively, if not absolutely, universal. We should also try to do it in

72

Cross-cultural pragmatics and different cultural values

terms of concepts which are intuitively clear and intuitively verifiable, and which th~refore will not be used differently by different scholars and in different cultural contexts. This may seem a tall order, but I submit that it can be done if we rely on such simple and universal or near-universal concepts as want, say, know, think, good and bad. In this chapter I shall try to demonstrate the explanatory value of this approach by examining a number of parameters which are widely relied on in the literature, seeking to clarify the sources of confusion, and to reveal the real differences between languages obscured by the use of confusing and inconsistently applied labels.

1. 'Self-assertion' 1.1. 'Self-assertion' in Japanese and in English From a Japanese point of view, Western culture in general and AngloAmerican culture in particular can be seen as dominated by 'self-assertion'. For example, Lebra (1976:257) contrasts "the Western model based on the complex of individuality, autonomy, equality, rationality, aggression, and self-assertion" with "the traditional [Japanese] complex of collectivism, interdependence, superordination-subordination, empathy, sentimentality, introspection, and self-denial". Similarly, Suzuki (1986) emphasises the Japanese tendency to avoid 'self-assertion' and the difficulties which this creates for the Japanese in contact with Westerners: We, used to assimilation and dependency, expect to project ourselves onto the other, and expect him to empathise with us. We have great difficulty with the idea that so long as our addressee is not Japanese we can't expect to have our position understood without strong self-assertion. But establishing our own viewpoint or position before out addressee has understood is not our forte ... So when Japanese, who aren't good at foreign languages, don't show their true ability in international conferences and scholarly meetings, it is less because of their language skills than because of the weak development of the will to express themselves linguistically to sufficient degree. It lies furthermore in the underdeveloped ability to stand apart from the position taken by another and at least assert oneself to the extent of saying, 'This is where I stand at this moment.' (Suzuki 1986: 157)

'Self-assertion'

73

On the other hand, when Kochman (1981:29) compares "the capacities and inclinations of whites and blacks [in America] to assert themselves" he sees the whites (that is, the members of the mainstream Anglo-American culture) as less able, and less inclined, to assert themselves. According to Kochman, "black culture allows its members considerably greater freedom to assert and express themselves than does white culture". He illustrates this claim, among other things, with the different attitudes of white and black culture towards boasting and bragging: "White boasting and bragging also contrasts with black practice with respect to the etiquette governing self-assertion. As white culture restricts individual self-assertion generally, it requires that individuals be governed by the norms of modesty when characterising their performance" (1981 :69). Thus, according to Kochman, white Anglo-American culture restricts individual self-assertion, whereas accord,ing to Lebra or Suzuki, the same white Anglo-American culture strongly encourages individual self-assertion. Who is right and who is wrong? My view is that both sides are right in what they are trying to say, but that they both fail to say it clearly and unambiguously. Both sides use the same label 'self-assertion', but they don't define it, and in fact they mean something quite different by it. The main difference between Japanese and mainstream English in the area under discussion can be represented in terms of certain clearly specifiable underlying conceptual structures. These structures are, above all, these two: Japanese Anglo-American

don't say: 'I want this', 'I don't want this' do say: 'I want this', 'I don't want this'

Japanese culture discourages people from saying clearly what they want and what they don't want, whereas Anglo-Saxon culture, on the contrary, encourages them to do so. In a similar vein, Japanese culture discourages people from expressing clearly their wishes, their preferences, and their desires (what they would or wouldn't like or want), whereas Anglo-Saxon culture encourages them to do so: Japanese Anglo-American

don't say: 'I would/wouldn't like (want) this' do say: 'I would/wouldn't like (want) this'

74

Cross-cultural pragmatics and different cultural values

Furthermore, Japanese culture, in contrast to Anglo-American culture, discourages clear and unequivocal expression of personal opinions;

Japanese Anglo-American

don't say: 'I think this / 1 don't think this' do say: 'I think this / 1 don't think this'

As pointed out by Smith (1983:44-45), "the Japanese are at pains to avoid contention and confrontation ... much of the definition of a 'good person' involves restraint in the expression of personal desires and opinions". This restraint manifests one of the greatest Japanese cultural values, called enryo, a word usually translated as 'restraint' or 'reserve'. "One way to express enryo is to avoid giving opinions and to sidestep choices when they are offered. As a matter of fact, choices are less often offered in Japan than in the United States." (Smith 1983:83-84) Smith quotes in this connection Japanese psychiatrist Takeo Doi' s account of the strain he experienced on a visit to the United States, where he was constantly offered choices: Another thing that made me nervous was the custom whereby an American host will ask a guest, before a meal, whether he would prefer a strong or a soft drink. Then, if the guest asks for liquor, he will ask him whether, for example, he prefers scotch or bourbon. When the guest has made this decision, he next has to give instructions as to how much he wishes to drink, and how he wants it served. With the main meal, fortunately, one has only to eat what one is served, but once it is over one has to choose whether to take coffee or tea, and - in even greater detail - whether one wants it with sugar, milk, and so on.... I could not have cared less. (Doi 1973:12)

Smith comments: The strain must have been considerable, for in Japan, by contrast, the host, having carefully considered what is most likely to please this particular guest, will simply place before him a succession of an overwhelming number of items of food and drink, all of which he is urged to consume, in the standard phrase, 'without enryo'. It is incumbent on the guest to eat and drink at least part of everything offered him, whether or not he likes the particular item, in order not to give offence by appearing to rebuke his host for miscalculating what would please him. (Smith 1983:84) Since Japanese culture places a taboo on direct expression of one's wants, it is also culturally inappropriate to ask other people directly what they want. Mizutani and Mizutani explain:

'Self-assertion'

75

Asking someone' s wishes directly is also impolite In Japan. Saying things like *Nani-o tabetai-desu-ka. (What do you want to eat?) *Nani-ga hoshii-desu-ka. (What do you want to have?) should be limited to one's family or close friends. ... To be polite, one should ask for instructions rather than directly inquire into someone' s wishes. Thus, saying: Mado-o akemashoo-ka. (Shall I open the window?) is more appropriate than *Mado-o akete-moraita-desu-ka. (Would you like me to open the window?)

(Mizutani -

Mizutani 1987:49)

The same cultural constraint prevents people in Japan from clearly stating their preferences, even in response to direct questions. Many Japanese, when asked about their convenience, decline to state it, saying instead, for example: Itsu-demo kekkoo-desu. (Any time will do.) Doko-demo kekkoo-desu. (Any place will be all right with me.) Nan-demo kamaimasen. (Anything will be all right with me.) (Mizutani - Mizutani 1987:117-118)

"In actuality one cannot always agree to what another person wishes, and one will then have to state one's own convenience anyway, but it is regarded as childish to immediately start stating one's own convenience when asked." (1987:118) What applies to the expression of one's wants applies also the expression of one's opinions. This, too, comes under the value of enryo. Lebra (1976:29) writes: "Pressure for conformity often results in a type of selfrestraint called enryo, refraining from expressing disagreement with whatever appears to be the majority opinion." But "the virtue of enryo, 'self-restraint', is exercised not only to respond to group pressure for conformity but to avoid causing displeasure for others, regardless of their group membership ... The imposition of self-restraint to avoid hurting Alter's feelings ... can reach an extreme that reveals immaturity even to most Japanese. The individual may acquiesce in the face of an intrusion on his rights or autonomy only because he is reluctant to offend another person by claiming his right." (Lebra 1976:41-42) I believe that the English concept of self-assertion is just as confusing and unhelpful when applied to Japanese culture, as the Japanese concept of enryo would be if applied to Anglo-American culture. On the other

76

Cross-cultural pragmatics and different cultural values

hand, the concept of enryo provides an essential key to understanding Japanese culture. But to be able to use this key, we must first understand what this concept really means; and we cannot understand it by trying to translate it into English cultural concepts such as reserve, restraint, modesty, or self-effacement. We can only understand it if we translate it into culture-independent, universal or near-universal concepts such as want, think, say, good or bad. This can be done in the following way: enryo X thinks: I can't say to this person: I want this, I don't want this I think this, I don't think this someone can feel something bad because of this X doesn't say it because of this X doesn't do some things because of this

Difficulties experienced by the Japanese in dealings with Americans (of the kind described by Doi) highlight the fact that no similar value is embodied in Anglo-American culture. On the contrary, in English one is expected to say clearly and unequivocally what one wants, what one would like, or what one thinks. If that is what is meant by 'self-assertion', then uninhibited self-assertion is indeed allowed and encouraged in mainstream Anglo-American culture - as long as it doesn't come into conflict with another cherished value of the culture, that is, personal autonomy. This means that while one is allowed to say, in principle, 'I want X', one is not allowed to say freely: I want you to do X since in this case, the speaker's right to 'self-assertion' would come into conflict with the addressee's right to personal autonomy. This is why in English the use of the bare imperative is very limited, and why directives tend to take an interrogative or semi-interrogative form in English. This means that in English there is a strong cultural constraint on saying to other people something that would amount to 'I want you to do X'. Instead, one is expected to combine this component with some other components, which would recognise the addressee's personal autonomy, for example:

'Self-assertion'

77

I want you to do X I don't know if you will do it I want you to say if you will do it This or a similar combination of components can be realised in English by means of interrogative-directive devices (sometimes called 'whimperatives ') such as: Would you do X? Will you do X? Could you do X? Can you do X? Why don't you do X?

and so on. By contrast, in many other languages, for example Polish (Chapter 2 above), Russian (Comrie 1984a), Hebrew (Blum-Kulka Olshtain 1984; Blum-Kulka - Danet - Gherson 1985), Italian (Bates 1976), and Hungarian (Hollos - Beeman 1978), the bare imperative is used much more freely, and the use of interrogative structures in directives is much more limited. In fact, even in Japanese, the use of interrogative structures in directives is more limited than in English (see for example Matsumoto 1988). This does not mean that Japanese encourages the use of the bare imperative any more than English does. But in Japanese, the important thing is to show deference and to acknowledge one's dependence on other people rather than to avoid imposition. As Matsumoto (1988) rightly points out, non-imposition based on individual rights is an Anglo-Saxon (or Anglo-American), not a universal value. For example, in Japanese it is very polite to start interaction with other people by uttering 'direct' requests, such as Doozo yoroshiku onegaisimasu. (lit.) 'I ask you to please treat me well.' Musume 0 doozo yoroshiku onegaisimasu. (lit.) 'I ask you to please treat/take care of my daughter well.'

Matsumoto (1988 :410) observes that in utterances of this kind the speakers "in indicating that they, or someone closely related to them, are someone who needs to be taken care of by the addressee, humble themselves and place themselves in a lower position. This is certainly typical of deferential behaviour. The speech act in question, however, is a direct request; thus, an imposition. ... it is an honour to be asked to

78

Cross-cultural pragmatics and different cultural values

take care of someone in that it indicates that one is regarded as holding a higher position in the society." This means that, in many situations, it is easier to say 'I want you to do X' than 'I want to do X' - as long as one acknowledges one's dependence on the addressee: I I I I

want you to do X know that you don't have to do it say: it will be good for me if you do it think: you will do it because of this

In English, if one wants the addressee to do something, it is important to acknowledge the addressee's autonomy by inviting them to say whether or not they will comply with the request. Hence, the proliferation and the frequency of 'whimperatives' in English. In Japanese, interrogative directive devices or 'whimperatives' exist, too, but their scope is much narrower than in English (cf. Matsumoto 1988; Kageyama Tamori 1976). Instead, there is in Japanese a proliferation of devices acknowledging dependence on other people, and deference to other people. Hence, the basic way of making requests in Japanese involves not 'whimperatives' (i.e. quasi-interrogative structures) but dependenceacknowledging devices (usually combined with expressions of respect): V-te kudasai. 'Give me (please) the favour of doing V.' 'I feel respect towards you.'

Even speaking to a child one would usually phrase a request in terms of 'favours', although the expression of respect would be omitted: V-te kure. 'Give me (please) the favour of doing V.'

1.2. 'Self-assertion' in black and white American English When we turn to the comparisons between what have been called black and white speech styles in America, we see that the term 'self-assertion' stands here for rather different features of verbal behaviour than those to which it usually refers in the literature contrasting English with Japanese. For example, Kochman writes:

'Self-assertion'

79

Black culture values individually regulated self-assertion. It also values spontaneous expression of feeling. As a result, black cultural events typically encourage and even require individuals to behave in an assertive/ expressive manner, as in such black speech events as rapping and signifying ... and, as I am claiming here, argument. (Kochman 1981 :29-30)

Similarly, when white American culture is described in terms of 'selfrestraint', this word doesn't stand for the same thing for which it stands in the literature on Japanese culture. Another example from Kochman: White culture values the ability of individuals to rein in their impulses. White cultural events do not allow for individually initiated self-assertion or the spontaneous expression of feeling. Rather, self-assertion occurs as a social entitlement, a prerogative of one's higher status or, as with turn-taking, something granted and regulated by an empowered authority. And even when granted, it is a low-keyed assertion, showing detachment, modesty, understatement. ... 'Showing off', which would represent individually initiated (unauthorised) self-assertion and more unrestrained self-expression, is viewed negatively within white culture. Black culture, on the other hand, views showing off - in black idiom stylin' out, showboating, grandstanding - positively.... Because white culture requires that individuals check those impulses that come from within, whites become able practitioners of self-restraint. However, this practice has an inhibiting effect on their ability to be spontaneously self-assertive. (Kochman 1981:30)

Clearly, 'the ability of individuals to rein in their impulses' is something quite different from the ability to say clearly what one thinks, what one wants to do, or what one's preferences are. If the Japanese 'selfrestraint' consists mainly in refraining from saying 'I want X', the white Anglo-American 'self-restraint' consists largely in refraining from saying now what I want now and from saying what I think the moment I think it. The very principle of turn-taking, regarded as fundamental in AngloAmerican culture, forces the individual speakers to 'rein in' their impulses to some extent. In black culture - as in Jewish culture (cf. Tannen 1981 b) - different speakers are allowed to speak all at once, to overlap with one another and to interrupt one another, to share in this way excitement, interest, and mutual involvement, and to maintain a continuous flow of uninhibited communication and self-expression. But this is not a difference between saying and not saying 'I want X'. Rather, it is a difference between saying it at once and saying it at what one sees as an appropriate moment.

80

Cross-cultural pragmatics and different cultural values

1.3. Spontaneity, autonomy, and turn-taking: English vs. Japanese This doesn't mean that white Anglo-American culture generally discourages spontaneous self-expression. Rather, it discourages it to the extent to which spontaneous self-expression might come into conflict with the principle of everyone's personal autonomy: one can express oneself spontaneously, if this doesn't infringe on what is seen as other people's right to speak without interruptions and without interference from other people. It is interesting to note in this connection that in the literature comparing Japanese culture with mainstream Anglo-American culture, the latter is usually said to encourage rather than discourage 'spontaneity'. For example, the author of a study comparing Japanese and American educational materials writes: Is the expression of spontaneous feelings encouraged or discouraged? ... Japanese teachers are advised to discourage students from expressing impulsive thought and emotional opinions ... The result may be the establishing of two identities, one functioning on the communicative level and the other known only to ego. A damper is based on the potential for shared excitement. There are, of course, inhibitors in the United States. The difference is a matter of degree. (Lanham 1986:294)

But I don't think it is a matter of degree. Rather, it is a matter of different cultural priorities. In Japanese culture, the overriding cultural principle seems to be constant caution not to offend or not to hurt other people (and also to avoid embarrassment for oneself which could follow from this); that is, an attitude which can be portrayed as follows: if I do X someone could feel something bad because of this I don't want this The Anglo-American principle of personal autonomy can be represented as follows: everyone can say:

'I want this', 'I don't want this' 'I think this', 'I don't think this' one can't say to someone: 'you have to do X because I want it' 'you can't do X because I don't want it'

The Anglo-American principle of tum-taking can be seen as a manifestation of this more general principle of personal autonomy, and of a more

'Self-assertion'

81

general respect for the rights of every individual. The principle of tum-taking can be represented as follows: someone is saying something now I can't say something at the same time I can say something after this It is interesting to note that Japanese culture doesn't observe the same principle of turn-taking. On the contrary, since Japanese culture values interdependence more highly than autonomy, in Japanese conversation utterances are expected to be, to a large extent, a collective work of the speaker and the addressee, or, more generally, of different speakers. This is done, in particular, by means of 'response words', that is, of what is called in Japanese aizuchi, a word which likens Japanese conversation to the work of two swordsmiths hammering a blade in tum. Mizutani Mizutani (1987:18-19) write:"The word ai means 'doing something together' ... ; tsuchi means 'a hammer' .... Two people talking and frequently exchanging response words is thus likened to the way two swordsmiths hammer on a blade. In Japanese conversation, the listener constantly helps the speaker with aizuchi ... - the roles of the speaker and the listener are not completely separated." Mizutani and Mizutani stress that aizuchi are absolutely essential to Japanese conversation and they support this with a startling statistic: 'The average number of aizuchi per minute is ... from 12 to 26, according to the study made by one of the authors." (1987:20) This is a striking manifestation of the Japanese value of interdependence, which is just the opposite of the Anglo-American principle of personal autonomy. The same applies to the Japanese conversational principle of leaving sentences unfinished so that the addressee can complete them. As Mizutani - Mizutani (1987:27) describe it, in Japanese, "leaving a part of the sentence unsaid so that the listener can supplement it is often more considerate and polite than just going ahead and completing one's own sentence.... always completing one's own sentences can sound as if one is refusing to let the other person participate in completing a sentence which might better be completed by two people". The attitude reflected in Japanese conversational style can be portrayed as follows: I want to say something now I think you know what I want to say I think you would say the same

82

Cross-cultural pragmatics and different cultural values

I think I can say part of it, you can say another part of it I think this will be good Thus, if the Anglo-American conversational principle of turn-taking reflects the cultural value of personal autonomy, the Japanese conversational principle of 'collective sentence production' reflects the Japanese cultural values of interdependence, co-operation, and 'groupism'.

1.4. 'Spontaneous self-assertion' vs. 'regulated self-assertion': black English vs. white English vs. Japanese Returning now to black English, we note that although it too rejects the tum-taking model, it doesn't reject it in favour of the conversational co-operation and interdependence characteristic of Japanese. On the contrary, it rejects it in favour of what Kochman (1981) calls spontaneous or impulsive self-assertion and self-expression, that is to say, in favour of some values which are contrary to the Japanese ethos. It is interesting to note that Kochman describes the contrast between black English and white English in this respect using the same pair of terms that, for example, Barnlund (1975b:35) uses to describe the contrast between white English and Japanese: 'regulated' vs. 'spontaneous'. Thus, according to Kochman, black English is 'spontaneous' and white English 'regulated', whereas according to Bamlund, Japanese is 'regulated' and English (that is to say, white English) is 'spontaneous'. But this means that the same white English that from a Japanese perspective is seen as 'spontaneous' and 'not regulated', is seen from a black perspective as 'regulated' and 'not spontaneous'. The 'regulated' character of white English means, roughly speaking, that while one can express one's thoughts, wants, and feelings, one is expected to observe certain rules in doing so; in particular, one is expected not to interrupt other people, and not to speak at the same time as other people. This constrains one's spontaneity, to some extent, but it doesn't constrain one's freedom of self-expression. On the other hand, in Japanese one is expected to be much more circumspect in expressing one's thoughts, one's wants, and one's feelings. It is not only a question of when to express them, but whether one should express them at all; Japanese discourse can be said to be 'regulated' with respect to what to say, not just when to say it. When the Japanese self is described as a "guarded self' (for example by Barnlund

'Self-assertion'

83

1975b: 112), reference is made, in the first place, to what is said, and in particular, to the care Japanese speakers take "to prevent overexposure of inner selves" (1975b:112). Bamlund illustrates this 'restraint' in selfdisclosure with striking statistical data, showing enormous differences between Americans and Japanese in the range of topic they are prepared to talk about, and also in the range of persons to whom they are prepared to reveal their thoughts and their opinions. As for when, the important thing is not so much not to overlap with other people, as to premeditate what one is going to say in order to avoid saying something which could hurt or offend somebody, or which could embarrass the speaker him/herself. Thus, Barnlund (1975b:131) describes Japanese communication as "a three-act play: 'Premeditation', 'Rehearsal', and 'Performance"'. One can see why the terms 'regulated' and 'non-spontaneous' can come to mind in this connection, but clearly this cannot be the same thing as Kochman has in mind when he describes white American English as 'regulated' and 'non-spontaneous'. This shows, once again, that labels such as 'regulated' or 'spontaneous' are not self-explanatory, just as ' self-assertion' and 'self-expression' are not self-explanatory, and are used by different writers to apply to different phenomena, and to different cultural norms. On the other hand, semantic formulae couched in terms of universal semantic primitives can be both precise and self-explanatory. I propose the following: Black American culture I want/think/feel something now I want to say it ('self-assertion', 'self-expression') I want to say it now ('spontaneity') White Anglo-American culture I want/think/feel something I want to say it ('self-assertion', 'self-expression') I cannot say it now because someone else is saying something now ('autonomy', 'tum-taking') Japanese culture I can't say: I want/I think/I feel something someone could feel something bad because of this if I want to say something I have to think about it before I say it

84

Cross-cultural pragmatics and different cultural values

1.5. 'Self-assertion' as personal display: black English vs. white English But the alleged 'self-assertiveness', attributed by Kochman and others to American black culture, has other features, which are reflected in characteristic black styles and genres such as sty/in' out, showboating, and grannin' (grandstanding) (Kochman 1981). Each of these concepts deserves detailed analysis, which cannot be undertaken here. All that I can do in the present context is to point out to some characteristic cultural features which are manifested in these and other similar folk-concepts. The black so-called self-assertion consists largely in an uninhibited desire to draw attention to oneself, and to behave, verbally and nonverbally, in ways which would ensure this. As a first approximation, this can be represented as follows: I want people to think about me now I want to do something because of this now In addition to this general desire for attention, however, there is also the more specific desire for admiration, or rather, for admiring attention - a desire which in black culture is viewed positively, not negatively. This is clearly visible, for example, in black boasting, bragging, and overt exultation and jubilation over one's success. For example, Kochman (1981 :72) cites a television interview with some black basketball players, who had just won a championship basketball game. "One of the main players of the team, asked to comment on their opponents, was serious at first, talking about 'playing hard and matching us height for height', etc. However, he ended up with the exultant and self-congratulatory 'But we were just too good for them!' " As a first approximation, we could portray this attitude as follows: I know: I can do good things other people can't do the same I feel something good because of this I want people to think good things about me because of this It is important to recognise, however, that in black culture self-aggrandisement of this kind has a somewhat theatrical quality, and that it is meant partly as public entertainment. To reflect this vital aspect of black 'self-aggrandisement' one important component has to be added to the formula sketched above:

'Self-assertion'

85

I know: I can do good things other people can't do the same I feel something good because of this I want people to think good things about me because of this I say this because I want people to feel something good Kochman (1981:73) points out that in black culture, boasting is interpreted "not as an unwarranted and uncouth claim to superiority but as humour"; or as Reisman (1974:60) puts it, as "the assertion of oneself, the making of one's noise, which depends not so much on the specific content of the boast as on the fact that it is made -loudly - at all". The expression 'assertion of oneself' appears here again, but, again, the context makes it clear that it is not the same 'assertion of oneself' which the literature on Japanese language and culture attributes to mainstream Anglo-American culture.

1.6. 'Self-assertion' and 'good interpersonal relations' One might hypothesise that all cultures cherish and seek to promote 'good relations' among people. But different cultures interpret this goal differently, and they seek to implement it in different ways; and these different interpretations are reflected in different 'ethnographies of speaking'. In Japanese culture, the prevailing conceptual formula is this: if I do/say something someone could feel something bad because of this I don't want this I have to think about it before I do it This is why Japanese culture can be seen as a 'culture of anticipatory perception' and a 'culture of consideration' (Suzuki 1986:157), a culture bent on preventing displeasure. Lebra (1976:41) remarks: "One should note how often in speech the Japanese refer to the need not to cause meiwaku, 'trouble', for another person, not to be in his way, and not to hurt his feelings. In actual behaviour, too, they tend to be circumspect and reserved, so as not to offend other people." In black American culture there is no similar emphasis on preventing displeasure, and, consequently, there is no emphasis on 'self-restraint'. On the contrary, black culture encourages uninhibited spontaneous self-

86

Cross-cultural pragmatics and different cultural values

expression. At the same time, however, it is a culture where self-expression and self-display is seen as conducive to 'good feelings', not only in the speaker but also in other people; and as a means to promote shared excitement, shared fun, shared interest, and shared 'colour'. A black English derogatory term for white people is 'grey': white people are seen as 'grey' not only because of the colour of their skin, but because of what is perceived as their 'lifelessness', their 'moderate impassioned behaviour', their lack of spontaneous emotionality, their 'reining in of their impulses'. (Johnson 1972:144-145) In white Anglo-American culture, the main emphasis is not on preventing displeasure, or on spontaneous and uninhibited self-expression, or on generating good feelings among one's 'audience', but on personal autonomy (for everyone), on non-imposition, and non-interference. It is a culture which encourages everyone to say freely - at the right time what they want and what they think, and (in a characteristic phrase) to 'agree to disagree'. Thus, while one can presume that all cultures cherish and seek to promote 'good relations' among people, it is not true that, for example, both American culture and Japanese culture cherish 'warm and cordial' relations among people, as asserted, for example, by Lanham (1986:293). A cultural emphasis on interpersonal warmth (in private relations) can be said to be characteristic of Russian culture (cf. for example Smith 1983), but not of American or Japanese culture. Such an emphasis is reflected, for example, in the extraordinary wealth of Russian expressive derivation, and in particular, in the abundance of hypocoristic forms of Russian names (see Wierzbicka, to appear, chap. 7). Japanese culture can be said to encourage empathy, consideration, and avoidance of hurting others, but not warmth or cordiality, as is shown by the extraordinary wealth, and wide use, of devices encoding 'apologies', 'quasi-apologies', 'preventive apologies', 'grateful apologies', and so on (see for example Mizutani - Mizutani 1987; Coulmas 1981). The virtual absence of linguistic devices encoding 'warmth' (in sharp contrast with the wealth of devices encoding 'respect', 'deference', and the like), points in the same direction. The relatively small degree of physical contact and physical intimacy between people in Japanese society provides further evidence for this (cf. Barnlund 1975b:106-108). American culture encourages a generalised friendly attitude to people, including strangers. But this, too, is different from the personalised affection displayed, for example, in Russian hypocoristic names. The American generalised friendliness can be seen in the common phrase

'Self-assertion'

87

Have a nice day! (often addressed to complete strangers, sometimes even displayed on badges on the uniforms of shop assistants, or on taxi windows). These three different cultural emphases in the interpretation of 'good interpersonal relations' can be represented as follows: 'Warmth', of the kind associated with Russian or Polish culture: I feel something good towards you 'Considerateness' and empathy, of the kind associated with Japanese culture: I don't want someone to feel something bad 'General friendliness', of the kind associated with American culture: I want everyone to feel something good

Needless to say, the formulae sketched above are not meant to capture all the different aspects of different cultural attitudes to emotions. For example, for Japanese culture we might also posit the following rule: I don't want to say what I feel whereas for Russian or Polish culture we might postulate the opposite norm: I want to say what I feel On the other hand, it would not be justified to posit for Japanese culture the rule which seems to prevail in Javanese society, especially among the Javanese gentry (prijaji): I don't want people to know what I feel For example, Geertz (1976:247) writes of the Javanese: "One often hears people say in praise of someone that 'one can never tell how he feels inside by how he behaves on the outside' "; and he speaks of "the nearly absolute requirement never to show one's feelings directly, especially to a guest" (1976:246). (See section 2.4 below.) In Japan the norm seems to be different: not 'I should conceal what I feel' but 'I should not verbalise what I feel'; that is, not 'I don't want people to know what I feel' but 'I don't want to say what I feel'. The whole Japanese emphasis on empathy, on omoiyari (cf. Lebra 1976:3849) shows that Japanese culture does not discourage an interest in other people's emotions; quite the contrary. But it does discourage verbal

88

Cross-cultural pragmatics and different cultural values

expression of emotions. We could formulate, therefore, the following, fuller, set of Japanese cultural norms related to emotions: (1) I don't want someone to feel something bad (2) I don't want to say what I feel (3) I should know what this person can feel this person doesn't have to say it

2. 'Directness' The terms 'directness' and 'indirectness' are often used in linguistic descriptions as if they were self-explanatory. In fact, however, they are applied to totally different phenomena, which are shaped by totally different values. The confusion which surrounds this notion is linked with the widely accepted distinction between so-called 'direct' and 'indirect' speech acts, and in particular, between imperatives and the so-called whimperatives. Thus, it is widely assumed that if one says to somebody Close the door! this is a 'direct' speech act, whereas if one says Could you close the door? or Would you mind closing the door? this is an 'indirect' speech act. But although these particular examples may seem clear, it is by no means clear how the distinction in question should be applied to other phenomena and to other languages. Thus, in many languages, for example, in Russian, Polish, Thai, or Japanese, the imperative is often combined with various particles, some of them somewhat impatient, others rather friendly, some of them described as 'softening' the directive, others as, on the contrary, making it harsher or more peremptory, and so on. Are such combinations of the imperative with a particle 'direct' or 'indirect' speech acts? There is no general principle which would allow us to answer this question. I suggest, therefore, that the whole distinction between 'direct' and 'indirect' speech acts should be abandoned - at least until some clear definition of these terms is provided; and also, that the distinction between 'direct' and 'indirect' ways of speaking in general should be abandoned, and that the different phenomena associated with these labels should be individually examined. I believe that when this is done, the confusion surrounding these concepts can be cleared, and some clear

'Directness'

89

cultural explanations for the cross-linguistic differences associated with these terms can be provided.

2.1. American culture vs. Israeli culture According to Blum-Kulka - Danet - Gherson (1985: 133), "viewed from a cross-cultural perspective, the general level of directness in Israeli society is probably relatively very high". What exactly is meant by this 'high level' of 'directness'? One clear example is provided by the wide use of bare imperatives in social interaction, including public interaction: (Passenger to driver: on the bus) Passenger A: ptax et hadelet, nehag (Open the door, driver.) (No response.) Passenger B: nexag, delet axorit. (Driver, rear door.) (Compliance.) (Blum-Kulka - Danet - Gherson 1985:129) Presumably, in English, an interrogative-directive device (could you or would you) would be used in a similar situation, and the authors appear to regard this as a clear case of directness vs. indirectness. In asking directions from a stranger on the street, the standard procedure for English is an 'attention-getter' (Excuse me... ) and the form Can/could you tell me ... ? (Blum-Kulka 1982:46). But in Hebrew, the standard procedure is a 'direct request for information' ('Where is the railway station?'). What does 'directness' mean in cases of this kind? I think it means that in Hebrew one can say rather freely something that means: I want you to do (say) X whereas in English, generally speaking, one is not expected to say this without at the same time acknowledging the addressee's personal autonomy: I want you to do X I don't know if you will do it

90

Cross-cultural pragmatics and different cultural values

Hence the combination of the imperative with some interrogative features in common English directives. Why should the American and the Israeli cultures differ in this way? According to the above-mentioned source: One possible explanation for this high level of directness is an ideologicalhistorical one: The early settlers of Palestine were guided by an ideology of egalitarism, which frowned on all manifestations of possible discrimination between people, including a show of deference in speech. ... It is against this background that one should consider the directness of presentday Israeli society ... (Blum-Kulka - Danet - Gherson 1985:133-134)

But this explanation is hardly convincing, given the egalitarian ethos of North America: surely, American culture doesn't encourage manifestations of discrimination among people, or 'a show of deference', either (cf. de Tocqueville 1953). The same authors (1985: 137) also offer another explanation: " ... these findings can be interpreted as reflecting the distinct, culture-specific interactional style of Israeli society. The low value attached to social distance, manifested in language by a relatively high level of directness, suggests that the interactional style of this society is basically solidarity politeness oriented." I think that this observation is more to the point in comparing Hebrew with English, but, unfortunately, terms such as 'social distance' or 'solidarity politeness' are not self-explanatory either. Trying to really understand the cultural values in question, we could propose for Israeli Hebrew the following formula: we can all say to one another: 'I want you to do this' we will not feel something bad towards one another because of this Since in Israeli Hebrew one can also freely express one's 'diswants' (for example, in refusals, disagreements, and so on) the formula above should probably be expanded so as to include 'I don't want' as well as 'I want'. For example, Blum-Kulka observes in an earlier work: Generally speaking, Israeli society seems to allow for even more directness in social interaction than the American one .... It is not uncommon to hear people around a conference table in Israel disagreeing with each other bluntly (saying things like ata to'e 'You're wrong', or 10 naxon! 'Not true! '). Such directness in a similar setting in American society would probably be considered rude. Similarly, refusal is often expressed in Israel by a curt ' No'; the same 10 (no) can also be heard as a response to requests

iDirectness'

91

phrased as requests for information ('Do you have such and such?') in shops, hotels, and restaurants, a habit that probably contributes to the popular view about Israelis' 'lack of politeness'. (Blum-Kulka 1982:30-31)

We can portray the Israeli attitudes in question as follows: we can all say to one another: 'I want this', 'I don't want this', 'I think this', 'I don't think this' we will not feel something bad (towards one another?) because of this In Anglo-American culture, too, one can say fairly freely what one wants, what one doesn't want, and what one thinks, but one is not expected to be similarly 'blunt' about it, because it is as important in this culture to acknowledge everyone's right to independence and personal autonomy as to exercise one's own right to self-expression. Furthermore, in Anglo-American culture there is no emphasis on 'we' (corresponding to the cultural value of 'solidarity' in Israeli culture); rather, there is a strong emphasis on every individual's separate and autonomous 'I'. This is sometimes described in terms of 'rugged individualism' as opposed to an 'ethos of solidarity' (cf. for example Arensberg - Niehoff 1975); but there are many ways to be 'individualistic' and many ways to be 'nonindividualistic' or 'anti-individualistic'. For example, the Israeli ethos of 'solidarity' (cf. Katriel 1986) is different from, though related to, the Australian ethos of 'mateship' (cf. Wierzbicka 1986b); and it is certainly different from the Japanese ethos of 'dependence' and 'groupism' (cf. for example Lebra 1976; Smith 1983). Here as elsewhere, therefore, in spelling out cultural values it is safer to rely on explicit semantic formulae than on undefined and protean global labels such as 'directness', 'individualism', 'solidarity' or 'collectivism'. We can portray the Anglo-American cultural assumption in question as follows: I think: I can say: 'I want this', 'I think this' I know: other people don't have to want the same/think the same no one can say: 'I want you to want this', 'I want you to think this' I have not included in this formula the component 'I don't want this', because Anglo-American culture does impose certain inhibitions on the expression of 'diswants' and doesn't encourage open confrontation. In Hebrew, and in Jewish tradition in general, open confrontation is

92

Cross-cultural pragmatics and different cultural values

encouraged and cherished, as a reflection of spontaneity, closeness and mutual trust. As the Jewish writer Sholom Aleichem put it (quoted in Myerhoff 1978:188): "We fight to keep warm. That's how we survive." (cf. also Schriffrin 1984). In Anglo-American culture, however, 'direct confrontation' is avoided in the interests of social harmony between independent individuals. In view of the emphasis on individualism and on everyone's personal autonomy, 'closeness' is cherished in this culture less than 'harmony'. In saying this, I am contradicting the view of many Japanese scholars, who see Japanese culture as a culture of 'harmony' and Anglo-American culture as one which positively encourages 'direct contention and confrontation'. But this just shows, once again, that global labels such as 'harmony' are used by different writers in different senses. The fact of the matter is that, as pointed out by Blum-Kulka (1982:30-31) or by Levenston (1970), in England or in America it is not common to hear people around a conference table disagree with one another by saying 'you're wrong' or 'that's not true'; in fact, it is not common to use such phrases in everyday conversation either. Anglo-American tradition encourages people to say 'I don't think so' rather than 'you are wrong'. Japanese culture discourages people even from saying 'I don't think so'. But we cannot accurately account for all such differences in terms of labels such as 'harmony', 'directness', or 'confrontation'. Blum-Kulka (1982:30-31) mentions that it is not common in English to express refusal by saying 'No' as one does in Hebrew, or to say 'No' in response to a request for information (for example in shops, hotels, and restaurants): 'Do you have such and such?'. In English, when someone indicates that they want something from us we are free to say 'No', but not to say just 'No'. The label 'directness' is not helpful in describing this aspect of the Anglo-American ethnography of speaking, though one can use here, more illuminatingly, the label 'bluntness'. (It should be noted, however, that 'bluntness', though clearer here than 'directness', is not self-explanatory either, and that for example Geertz (1976:245) attributes 'bluntness' to Anglo-American culture, contrasting it in this way with Javanese culture.) 'Bluntness' in saying 'no' is viewed positively in Israeli culture but not in Anglo-American culture. These different attitudes to 'bluntness' in saying 'No' can be represented as follows: Anglo-American culture I say: No

'Directness'

93

I don't want you to feel something bad because of this I will say something more about it because of this Israeli culture I say: No I think I don't have to say anything more about it

In Japanese culture, the norm seems to be to avoid saying 'No' altogether (in particular, to refuse an offer or a request, to express disagreement, and so on). Thus, Nakane (1970:35) notes: " ... one would prefer to be silent than utter such words as 'no' or 'I disagree'. The avoidance of such open and bald negative expressions is rooted in the fear that it might disrupt the harmony and order of the group". This norm can be represented as follows: Japanese culture I can't say: No I will say something else because of this

Barnlund (1975b) explicitly compares the Japanese with the Americans in this respect: Anyone who has observed groups of Japanese or Americans talking together is aware at once of certain peculiarities in their habits of speech. In one group everyone bows and exchanges personal cards. When they speak they do so quietly, often in the form of understatements. Rarely does one hear a belligerent or unequivocal 'No.' ... In the other group, they all shake hands as they begin a conversation. 'No' is heard at least as often or more often than ' Yes'. ... Arguments are heated, issues often polarised. (Barnlund 1975b:26-27)

But if the difference between the Americans and the Japanese is presented in such a polarised manner, it is hard to see how the same Americans can appear to the Israelis as people who, in contrast to themselves, avoid saying 'No'. It seems to me that the semantic formulae proposed here allow us to paint a clearer and more coherent overall picture.

2.2. 'Indirectness' in Japanese According to Mizutani - Mizutani (1987), Honna - Hoffer (1989), and many other writers on Japanese language and culture, it is extremely important when talking politely in Japanese 'to sound indirect'. But what does one do in Japanese 'to sound indirect'?

94

Cross-cultural pragmatics and different cultural values

First of all, one doesn't say what one wants; instead, one sends 'implicit messages', expecting that the addressee will respond to them: The speaker thus often makes indirect requests, and the listener also responds to implicit messages: this makes the indirect development of speech possible. For instance, a man, usually a superior, will come into the room and say: Kyoo-wa iya-ni atsui-nee. (It's awfully hot today, isn't it?) And one of his men will say hai ['yes', respectful], and hurry to open the window or turn on the air conditioner. He may even apologise saying: Doomo ki-ga tsukimasen-de ... (I'm sorry I didn't notice.) ... many Japanese seem to find pleasure in being with someone who understands them very well and so will sense their wishes and act to realise them without being asked.

(Mizutani -

Mizutani 1987:36)

The attitude manifested in speech behaviour of this kind can be represented as follows: I I I I

want something don't want to say this will say something else because of this think this person will know what I want

A different phenomenon, also described in the literature in terms of 'indirectness', has to do with deliberate lack of precision and lack of specificity in the identification of referents, or in using numbers: In social situations the Japanese like to refer to numbers or amounts in a non-specific way. For instance, when buying apples they will often say: Mittsu-hodo/gurai/bakari kudasai. (Please give me about three of them.) instead of saying Mittsu kudasai.

(Mizutani -

Mizutani 1987:33)

Furthermore, in making proposals or suggestions, the Japanese tend to refer to things with indirect expressions like demo and nado (and others). For example:

Ocha-demo nomimasen-ka. 'How about having some tea?' [lit. 'or something?'] Eiga-demo mimashoo-ka.

{Directness'

95

'How about going to a movie?' [lit. 'How about seeing a movie or something?'] (Mizutani - Mizutani 1987:34) Similarly:

A: Mada jikan-ga aru-n-desu kedo. 'I have some time to kill.' B: la, zasshi-demo yondara doo-desu-ka. 'Then, why don't you read a magazine or something?' (Mizutani - Mizutani 1987:34) In such situations, ocha-demo or eega-demo are preferred to ocha-o or eega-o because they let the listener choose among several possibilities. This deliberate use of non-specific reference and non-specific numeral expressions can be portrayed as follows: I say: I would want something like this I don't want to say: 'I want this' It is not difficult to recognise here again the Japanese value of enryo, discussed earlier - a value which is quite different from the AngloAmerican value of personal autonomy. But if all the different phenomena in question are described by means of the same label 'indirectness' then the different cultural values involved cannot be revealed, and the generalisations made in individual works devoted to comparisons of two cultures do not seem to make sense in a broader cross-cultural perspective. As a particularly striking example of the resulting confusion I now tum to the question of 'which culture encourages more "indirectness" Greek or American?'

2.3. Greek culture and American culture Consider first the following statement, fairly characteristic of the way the concept of 'indirectness' tends to be used in the literature on crosscultural pragmatics: Though languages provide their speakers with explicit, direct ways for achieving communicative ends, in day-to-day communication speakers seem to prefer indirect ways. In making a request to a secretary, for

96

Cross-cultural pragmatics and different cultural values

example, people are more likely to say things like 'Could you do it' or 'Would you mind doing it' than the simple 'Do it'. (Blum-Kulka 1982:30)

The writer of the above knows very well that the generalisation in question applies not to 'people in general' but mainly to Anglo-Saxons, and that, for example, it doesn't apply to the Israelis. But this doesn't prevent her from formulating it as if it in fact applied to 'people in general'. Furthermore, the illustration provided makes it clear that what the author has in mind is the phenomenon of 'whimperatives', directives phrased interrogatively; but the generalisation is couched in terms of 'indirect ways of speaking' - as if it were enough to mention the 'whimperatives' to explain what one means by 'indirect ways of speaking' in general. Blum-Kulka (1982:30) proceeds then to make the important and, I think, perfectly valid point that "one major factor that can influence the application of such principles can be the general 'ethos' of one society as compared to another one". But having said this, she says something rather startling, that is, that "Greek social norms, for example (Tannen [1981a]), require a much higher level of indirectness in social interaction than American ones" (Blum-Kulka 1982:30). This statement might lead one to believe that if in Israel one tends to say 'Do it!' more widely than one does in America, in America one tends to say 'Do it!' more widely than one does in Greece; and that, conversely, if in America one tends to say 'Would you' or 'Could you' in many situations in which in Israel one would say simply 'Do it!', in Greece one tends to say 'Would you' or 'Could you' in many situations in which in America one would say simply 'Do it!'. But is this believable? Surely not. In fact, a claim of this kind would seem to go against everything one knows about Mediterranean culture generally, and about the Greek culture more specifically. In particular, the characterisation of Greek culture as 'indirect' or as 'more indirect' than American culture, seems to be incompatible with the results of behavioural studies devoted specifically to the Greek national character, and of behavioural differences between Greeks and Americans, such as Triandis - Vassiliou (1972). For example, according to this study, typical Greek behaviour shows characteristics that an American will interpret as arrogance, dogmatism, and attempts to appear all-knowing and all-powerful.

'Directness'

97

The characterisation of Greek culture as 'indirect' also goes against the expectation that Greece and Middle East (including Israel) might share some cultural values, and some features of their ethnography of speaking (cf. Tannen - Oztek 1977; Matisoff 1979) rather than being at the opposite poles of a scale, with Anglo-Saxon ways of speaking in the middle: 'direct' 'intermediate' 'indirect'

Israel England and North America Greece

One can only wonder where Japan would appear on a scale of this kind? Below Greece, perhaps? And (American) black English? Above Israel? I believe that here as elsewhere, scales are misleading and confusing if they are not preceeded by rigorous qualitative analysis. If one examines the data in Blum-Kulka's source of information on Greek culture (Tannen 1981a), it transpires that the so-called Greek 'indirectness' applies to phenomena quite different from the use of whimperatives; and the whole puzzling story of 'Greek indirectness' versus 'American directness' begins to make sense. What Tannen did was to present a number of informants (some Americans, some Greeks, and some Greek-Americans) with a written questionnaire, which begins by presenting an exchange between a wife and a husband: John's having a party. Wanna go? Wife: Husband: Okay.

Two paraphrases are then presented, and respondents are asked to indicate which they believe the husband meant when he said okay: (1-I) My wife wants to go to this party, since she asked. I'll go to make her happy. ['indirect'] (I-D) My wife is asking if I want to go to a party. I feel like going, so I'll say yes. ['direct'] Tannen's results are clear and interesting: "A comparison of the percentage of respondents in the three groups who opted for paraphrase 1-1 turns out looking much like a continuum, with Greeks the most likely to take the indirect interpretation, Americans the least likely, and GreekAmericans in the middle, somewhat closer to Greeks." (Tannen 1981a:229).

98

Cross-cultural pragmatics and different cultural values

Although Tannen herself describes her study as dealing with 'modes of indirectness', she is generally careful to point out that she is dealing only with one specific context: a negotiation between husband and wife about whether to go to a party. Nonetheless, some of her comments could be seen as inviting the kind of over-generalisation expressed in Blum-Kulka's account of her study. For example, she reports that "an American-born woman of Greek grandparents ... commented that she tends to be indirect because she picked it up from her mother, who was influenced by her own mother (i.e. the grandmother born in Greece)" (Tannen 1981a:235). Similarly, she quotes another personal testimony which she calls "most eloquent": "that of a professional man living in New York City, whose grandparents were from Greece. He seemed fully assimilated, did not speak Greek, had not been raised in a Greek neighbourhood, and had few Greek friends. In filling out the questionnaire, he chose 1-1, the initial indirect interpretation. In later discussion he said that the notion of indirectness 'rang such a bell'." (1981a:235) This really could lead one to believe that Greek culture is somehow generally 'indirect', certainly more so than American culture. But what does this really mean? All that Tannen has really shown is that Greek couples seem to be more attuned to one another's unexpressed wishes than American couples are, and more ready to guess one another's unexpressed wishes, whereas American couples seem to rely more on explicit verbalisations of wishes. In fact, some of Tannen's comments suggest that in Greek culture it is the woman who is generally expected to guess, and to comply with, her father's, or her husband's, unexpressed wishes: For example, a Greek woman of about 65 told me that before she had married she had to ask her father's permission before doing anything. She noted that of course he never explicitly denied her permission. If she asked, for example, whether or not she should go to a dance, and he answered, (1) An thes, pas. ('If you want, you can go.') she knew that she could not go. If he really meant that she could go, he would say, (2) Ne. Na pas. ('Yes. You should go.') ... This informant added that her husband responds to her requests in the same way. She therefore agrees to do what he prefers without expecting him to express his preference directly. (Tannen 1981a:224-225)

But if this is all there is to it, is it enough to draw the conclusion that "Greek social norms ... require a much higher level of indirectness in social interaction than American ones" (Blum-Kulka 1982:30)? It seems

'Directness'

99

to me that a conclusion of this kind is unwarranted and misleading. On the other hand, Tannen's data suggest the following cultural norm, which seems to be quite credible, clear, and meaningful: I I I I

want something don't have to say this think this person will know what I want think she will do it because of this

It is particularly interesting to note here the difference between the Japanese general enryo ('reserve, self-restraint'): I want something I don't want to say this and the Greek (male, typically) self-confidence: I want something I don't have to say this (I think she will do it anyway) It is also relevant to mention the importance of the division between 'in-group' and 'out-group' in Greek culture, and the great intimacy and closeness prevailing within the 'in-group'. Triandis - Vassiliou (1972:304) speak in this connection of the existence of an "extremely tightly knit family and an 'ingroup' that provides protection, social insurance, and a warm and relaxing environment; in short, a haven from the larger world". In a warm, intimate environment of this kind one doesn't have to rely on overt, verbal expression of one's needs, wishes and desires. As for Anglo-American culture, Tannen's findings are perfectly consistent with the general Anglo-American emphasis on everyone's personal autonomy and on the individualism prevailing even within the family: Anglo-American culture encourages people to say, clearly and explicitly, what they want and what they think. Apparently, American spouses, too, rely less on wordless communication, and more on clear self-expression. Possibly, this implies less of a feeling of 'oneness' between the spouses, and a greater emphasis on each spouse's individuality, unpredictability, and personal autonomy. All this is consistent with what we otherwise know of Anglo-American cultural values. The term 'indirectness' doesn't really help us here. In fact, it is rather an obstacle to understanding.

100

Cross-cultural pragmatics and different cultural values

2.4. 'Indirectness' and 'dissimulation' in Javanese According to Geertz (1976:244), indirectness or 'indirection' is a major theme of Javanese behaviour. Geertz illustrates this feature with the proverb 'to look north and hit south'. He also mentions the fact that old-time kijajis (Koranic teachers) never explicitly informed people they were wrong, but told little stories from which the listeners could get the point less painfully. "One must get the rasa of what people are saying, the real content, informants are always emphasising, because alus people (i.e. civilised people) often don't like to say what is on their minds." 'Indirectness' as described above is closely related to another Javanese cultural norm, that is, to what Geertz calls 'dissimulation or pretence', or what the Javanese themselves call efok-efok. "The characteristic quality of efok-efok, in contrast to our patterns of dissemblance, is not merely that it is far more prevalent and that it is largely approved ... but that it need not have any obvious justification, being merely gratuitous .... In general, polite Javanese avoid gratuitous truths." (1976:245-246). Thus, Geertz quotes the following definition of efokefok, offered by an informant: He said: Suppose I go off south and you see me go. Later my son asks you: 'Do you know where my father went?' And you say no, e{ok-e{ok you don't know. I asked him why should I e{ok-efok, as there seemed to be no reason for lying, and he said, 'Dh, you just e{ok-efok. You don't have to have a reason.' (Geertz 1976:246)

This general cultural norm of concealment, of not saying, not telling people any 'gratuitous truths', applies in particular to the truth about one's personal feelings: The same sort of pattern is involved in the nearly absolute requirement never to show one's real feelings directly, especially to a guest. Any kind of negative feeling towards another must be dissimulated.... Strong positive feelings are also supposed to be hidden except in very intimate situations. The effort is to keep a steady level of very mild positive affect in interpersonal relations, an e{ok-e!ok warmth behind which all real feelings can be effectively concealed. (Geertz 1976:246)

What applies to feelings applies also to wishes: one should conceal one's wishes and one's intentions, particularly if they are in conflict with other people's wishes or desires. For example:

tDirectness'

101

... one must call out to any passerby one knows inviting him to stop in, even though he may be the last person on earth you wish to see. One must refuse food (unless the host persists in offering it) even if one is dying of hunger ... One should never refuse outright people's requests to do something for them; rather, one merely agrees even if one has no intention of going through with whatever it is, and then one never gets around to doing it, putting the petitioner off with various etok-etok excuses, until he realises at last that one was not serious in the first place. (Geertz 1976:246-247)

Apparently, what applies to feelings and to wishes, applies also to thoughts. Geertz (1976:247) quotes a village politician on this point, who began his speech as follows: "No one ever says what he really thinks. People always e{ok-erok when dealing with other people. 1 too never say what 1 really think, and you can't tell how 1 feel about things by what 1 say." The reluctance to express one's feelings, wants, and thoughts links Javanese culture with Japanese cultural norms described earlier; but the element of concealment, of conscious 'dissimulation', seems to be specifically Javanese. We can portray this 'dissimulation' as follows: 1 don't want to say: 1 feel X/I want X/I think X/I know X 1 don't want people to know what 1 feel/want/think/know The more specific norm proscribing explicit requests can be portrayed as follows: 1 can't say to someone: 'I want you to do X' someone could feel something bad because of this 1 have to say something else The norm proscribing explicit refusals can be portrayed along similar lines: if someone says to me:'1 want you to do X' 1 can't say: 'I don't want to do it' someone could feel something bad because of this 1 have to say something else 1 don't have to do it because of this The avoidance of providing 'gratuitous information' can be represented as follows: if someone says to me: 'you know something'

102

Cross-cultural pragmatics and different cultural values

'I want you to say it' I can't say: 'I don't want to do it' I can say something else I don't have to do it because of this The general principle of e!ok-e{ok can perhaps be formulated as follows: I don't want to say what I think/know I don't have to say this I can say something else In Western culture, saying what one thinks tends to be seen as everyone's right, and saying what one knows, as everyone's obligation (although there are of course limits to this). Generally speaking, then, questions can be freely asked and answers cannot be freely withheld (cf. Eades 1982 and the references quoted by her). These attitudes can be portrayed as follows: (1) I can say what I think (2) I can say to people: 'you know something' 'I want to know it' (I can think: they have to say it) (3) if someone says to me: 'you know something' 'I want to know it' I have to say it In many non-Western cultures, however, and in particular in Javanese, culture, a different norm prevails, which can be portrayed as follows: if someone says to me: 'you know something' 'I want to know it' I don't have to say it As shown by Eades (1982), in Australian Aboriginal culture one wouldn't even assume that one has the right to ask; on the contrary, the opposite norm prevails: I can't say to people: 'you know something' 'I want to know it'

'Directness'

103

Many writers have tried to explain cultural differences of this kind, pointing to different cultural attitudes to knowledge, questions, and information (cf. Eades 1982; Abrahams 1976; Sansom 1980; Keen 1978; Harris 1984; Goody 1978). While accepting their explanations, I would like to add to them an additional one: different cultural attitudes to truth. European culture has traditionally placed a great premium not only on 'knowing' but also on saying what one knows, that is, what is knowable (or true). Other cultures may value knowledge without valuing verbal articulation of knowledge. For example, Japanese culture is said to value intuitive knowledge and to mistrust verbalised, articulated knowledge (cf. for example Bamlund 1975b; Lebra 1976). It is interesting to note in this connection that while all languages appear to have a word corresponding to know, many languages do not have a word corresponding to true (cf. Hill 1985). Some languages have a word for something like lying (to another person), without having a word like true which combines in its meaning 'knowing' and 'saying' (that is, 'saying what one can know'), without any reference to interpersonal relations, as in the case of 'lying' (cf. Lutz 1985:73). In fact, even in English the word truth didn't always have the impersonal and objective ring which it has now. As Hughes (1988:61-62) observes, "The central and fascinating point in the semantic history of truth is that it evolves from being a private commitment to a publicly assessed quality. The form of word even changes, so that troth, the private form, can, by the proof of arms, be asserted above even the claims of evidence or testimony, if need arises. (This mediaevalised form of truth is, of course, virtually the opposite of the modem notion, which is factual, demonstrable and essentially impersonal.)" European culture, however, elevated the truth (first the private, personal 'truth', and then the public, impersonal truth) to a particularly high place among generally accepted ideals; and 'truth' can be seen as opposed to both 'lying' and 'concealment', to both saying what one knows is not true and not saying what one knows is true. The cultural norms in question can be represented, roughly, as follows: it is bad to say what is not true it is good to say what is true It might be added that modem Anglo-American culture appears to be more 'pragmatic' in its attitude to truth than European culture. This is reflected, for example, in the concept of 'a white lie', which doesn't seem to have any equivalents in German, French, Italian, or Polish

104

Cross-cultural pragmatics and different cultural values

(see section 3.5 below). Cultural attitudes to conscientiousness, punctuality, or reliability may indeed differ along the lines suggested by Max Weber (very roughly, between Protestant northern Europe, plus its American extension, and the Catholic rest, cf. Weber 1968); but the attitudes to pragmatic, 'white' lies may be divided along rather different lines, with, roughly speaking, continental Europe on one side of the dividing line and the more 'pragmatic' Anglo-American culture on the other. This modified, Anglo-American attitude to truth can be represented, very roughly, as follows: it is usually bad to say what is not true sometimes it is good to say what is not true if nothing bad can happen to anyone because of this As one early Anglo-Saxon put it: "Use not to lie, for that is unhonest; speak not every truth, for that is unneedful; yes, in time and place, a harmless lie is a great deal better than a hurtful truth." (Roger Ascham, 1550, quoted by Stevenson 1946:2058). But the norm discouraging nottruth (whether in an absolute or in a modified, 'pragmatic' form) is by no means universal. In particular, the Javanese principle of etok-etok allows one both not to say what one knows is true and also to say what one knows is not true. Perceived cultural advantages involved in such an attitude may include 'tranquillity', 'harmony', smooth and peaceful interpersonal relations ('I don't want to feel something bad', 'I don't want someone to feel something bad'), and so on.

3. Further illustrations: same labels, different values In this section, I discuss in a more summary way the use of five other global labels, which are generally believed to stand for identifiable cultural values, but which in fact are used to refer to different attitudes and different ways of speaking. I try to uncover the real differences in cultural values, concealed and obscured by such inconsistently and arbitrarily applied terms. The labels in question are: 'intimacy', 'closeness' (contrasted with 'distance'), 'informality' (contrasted with 'formality'), 'harmony', and 'sincerity'.

Further illustrations: same labels, different values

105

3.1. 'Intimacy' It is widely believed that different cultures differ in the importance they give to "intimacy' as a social value. For example, according to Hijirida - Sohn (1986:390) American culture gives this value a high priority, whereas in Japanese and Korean culture other values (for example, respect for rank and status) by far 'overrule' intimacy as a cultural norm. The claim that in American English 'intimacy' overrules rank or social status, whereas the opposite is true of Japanese and Korean, is perhaps not hard to believe, even without any precise definition of 'intimacy'. But when the authors make a more general claim, attributing to Americans an "extreme sensitivity toward the intimacy variable" (1986:391), we cannot go along with this without asking what exactly is meant by 'intimacy', and how this 'sensitivity to intimacy' is assessed. In fact, in my own analysis of Anglo-American culture as compared with Polish culture (cf. Wierzbicka 1985b; see also Chapter 2 above) or with Russian culture (cf. Wierzbicka, to appear), I have reached conclusions very different from those suggested by Hijirida and Sohn. From a Polish, or Russian, point of view, Anglo-American culture is not 'sensitive to intimacy' at all. What, then, is 'intimacy'? If we were to rely on the everyday meaning of the word intimacy (and what else can we rely on?), we could define the concept as follows: Intimacy refers to a readiness to reveal to some particular persons some aspects of one's personality and of one's inner world that one conceals from other people; a readiness based on personal trust and on personal 'good feelings'. This last proviso is necessary because although one might disclose one's secret fears or worries to a doctor or to a psychoanalyst this doesn't qualify as 'intimacy': to count as intimacy, selfdisclosure has to be based on an assumption of personal good feelings. This can be represented as follows: intimacy X thinks: I feel something I want to say it to someone I can say it to Y I feel something good towards Y Y feels something good towards me I can say it to Y because of this I can't say it to other people X says it to Y because of this

106

Cross-cultural pragmatics and different cultural values

Barnlund (1975b), who as we have seen has shown that Americans are more prone to self-disclosure than the Japanese, has concluded from this that the former value intimacy more than the latter: The Americans will tend to cultivate physical as well as verbal intimacy. Since the aim is to seek more complete expression of the inner self, Americans may not only disclose more fully verbally, but may try to utilise as many channels of communication as possible. For this reason they may display greater physical animation and engage in a higher frequency of physical contact during conversation. Touch, as one of the more intimate forms of interaction, may be more encouraged and more accepted. (Barnlund 1975b:38)

But even touch ceases to be 'intimate' if it is applied indiscriminately. A handshake may indeed be more revealing than a bow, but it is not necessarily more intimate. If intimacy could be reduced to self-disclosure, the claim that Americans are more given to intimacy than the Japanese could be sustained. But although intimacy is indeed related to self-disclosure, it cannot be reduced to it. To count as intimacy, self-disclosure has to be selective (in terms of the addressee), and this selectiveness has to be based on personal affection. In my view, a culture where one basic term of address, 'you', is used indiscriminately to everyone, cannot be regarded as one which attaches a great importance to the value of intimacy. If anything, it is extremely difficult to be intimate in English, because of this universal 'you', that is, because of the absence of any 'intimate' forms of address. There are of course nicknames, and so-called affectionate nicknames (for example Bob and Bobby for Robert, Kate and Katie for Katherine); but are these truly instruments of intimacy? Hijirida - Sohn (1986:391) think that they are. They write: "The tendency of Americans to upgrade address forms (from FN to TLN) toward a person they are angry at, structural differentiation of FN into FFN, Nn, and ANn in E[nglish] and the productive use of them all reflect the extreme sensitivity Americans have toward the intimacy variable." But I don't think this is right. It is not 'intimate' in English to call somebody John rather than Dr. Brown, and 'nicknames' such as Bob or Tim are no more intimate than John. As for so-called affectionate nicknames, such as Bobby or Timmy, they are not intimate but child-oriented; they can be affectionate, but affection is not the same thing as intimacy, particularly if it is an affection associated with the adult-child style of interaction. (For fuller analysis of forms of address and names see Wierzbicka, to appear, chaps. 7,8.)

Further illustrations: same labels, different values

107

I conclude that there is no linguistic evidence for the claim that English is particularly sensitive to intimacy. On the contrary, English, with its absence of any 'intimate' form of address, seems to be particularly insensitive to it. On the other hand, there is massive evidence for the importance of intimacy in Slavic languages. This evidence takes the form, above all, of enormous differentiation of expressive forms of personal names, such as, for example, Katja, Katen' ka, Katjusen' ka, Katecka, Katik and so on for Katerfna, or Vanja, Vanecka, Vanjusa, Vanjuska, Vanjusecka, and so on, for Ivan (see Wierzbicka, to appear). With respect to Polish, one can argue that the value of intimacy is even enhanced by the wide use of titles and other linguistic devices keyed to rank and status, since this increases the differentiation of personal relations. Hijirida - Sohn (1986:389) state that "both Japanese and Koreans, being extremely status-conscious, are eager to give and receive powerladen titles in daily interpersonal encounters", and they link this with the low value of intimacy in Korea and Japan. But Poles, too, are extremely status-conscious, and are eager to give and receive titles in daily interpersonal encounters; they also value a degree of formality and ritualised courtesy. At the same time, however, Poles place a high value on intimacy, and the wide range of possibilities between, say, Pani Professor ('Mrs Professor', with a third person form of the verb) and various intimate forms of expressive derivation of names, enhances the value of intimacy enjoyed with those special people with whom one chooses to share it. It is an illusion, then, to think that an egalitarian ethos, such as that prevailing in Anglo-American culture, leads necessarily to an increase in intimacy, or that a culture sensitive to status distinctions is necessarily inimical to intimacy. Once again, terminological confusion leads here to conceptual confusion, clearly visible, for example, in the following passage from an otherwise subtle and insightful study: Of the three societies under comparison, American is least sensitive to power variables, as evidenced in both the patterns and usages of honorifics. This seems to be due to their egalitarian value orientation. As a result, solidarity variables like intimacy and casualness prevail, although groupness [sic] solidarity is the last thing for Americans to give heed to due probably to their strong individualistic value orientation. (Hijirida Sohn 1986:383)

American society is described here as one dominated by 'solidarity variables like intimacy and casualness', although at the same time

108

Cross-cultural pragmatics and different cultural values

solidarity is said to be 'the last thing for Americans to give heed to'. This is confusing and self-contradictory. We can clear this confusion if we stop using undefined labels such as 'intimacy', 'solidarity' or 'casualness', and start using instead precise and self-explanatory semantic formulae couched in terms of universal semantic primitives.

3.2. 'Closeness' Speaking metaphorically, intimacy implies closeness - another variable which often comes up in discussions devoted to cross-cultural pragmatics. But what is 'closeness' in interpersonal relations and how does one assess it? Social psychology has developed various measures that can be used to assess 'social distance' (cf. for example Triandis - Triandis 1960; Bogardus 1933), but these have to do with relations between groups, not between individuals, and cannot be transferred to the study of interpersonal relations. The concept of 'distance' in interpersonal relations is heavily relied on by a number of writers on linguistic pragmatics, and in particular, by Brown - Levinson (1978); but it is never defined, and it is treated as if it was self-explanatory. At best, it is elucidated by means of examples. For example, Brown and Levinson assert that: only D[istance] varies in the following two sentences: (1) Excuse me, ltvould you by any chance have the time? (2) Got the time, mate? Our intuitions are that (1) would be used where (in S's perception) S[peaker] and H[earer] were distant (strangers from different parts, say), and (2) where Sand H were close (either known to each other, or perceptibly 'similar' in social terms). D, then, is the only variable in our formula that changes from (1) to (2) ... (Brown ~ Levinson 1978:85)

But this is baffling and unhelpful. Two university professors are presumably 'similar in social terms', but it doesn't follow from this that they would be likely to exchange phrases like Got the time, mate? On the other hand, two young male hitchhikers may well address one another in this way even if they are 'strangers from different parts'. Similarly baffling and unhelpful is the further claim that'D' is held constant in the following utterances: (3) Excuse me, sir, would it be all right if I smoke? (4) Mind if I smoke? (Brown - Levinson 1978:85)

Further illustrations: same labels, different values

109

To justify this claim, Brown and Levinson appeal to their 'intuitions', but this just shows how unreliable and idiosyncratic such 'intuitions' about abstract and semi-technical concepts like 'distance' may be. It seems to me that if we are to rely on the everyday use of the word close (as applied to human relations) we would have to say that closeness has to do with interpersonal 'knowledge' as well as interpersonal feelings: two people are said to be 'close' if they know one another very well, and have 'good feelings' for one another. This is similar to intimacy, but it is not the same thing. For example, a mother can be said to be very 'close' to her daughter, but it would be a little odd to say that a mother is 'intimate' with her daughter. A mother who is 'close' to her daughter knows a great deal about the daughter - about her 'hidden' thoughts, fears, hopes, desires, and so on. Speech doesn't seem essential to the idea of 'closeness', but mutual knowledge, and the willingness to let one another know what is happening inside us, does seem to be essential. Sometimes, 'closeness' may even reduce the need for verbal self-disclosure: if two people are very 'close' they may each know how the other person feels without overt speech, by a kind of empathy. But not all 'empathy' manifests 'closeness'; 'closeness' being a permanent (long-term) feature of a relationship, based on mutual good feelings. Tentatively: closeness ('X and Yare close to one another') X and Y know: we feel something good towards one another because of this each of them thinks of the other: I want to know what this person feels/thinks/wants I want this person to know what I feel/think/want because of this, each of them can know what the other feels/ thinks/wants when other people can't

To let someone become close to us means to trust them enough, and to feel enough affection (or 'good feelings') for them, to allow them to know us really well - better than other people know us. This may be seen as dangerous, because knowing us so well the other person will probably be able to hurt us. There will also be more opportunities for clashes, for mutual hurt, for open conflict. It may be safer not to get too 'close' - if one values peace, harmony, absence of conflict and absence of mutual hurt.

110

Cross-cultural pragmatics and different cultural values

Not all cultures, therefore, encourage closeness, certainly not to the same degree. For example, if I think something bad about you (for example that you look awful, or that you have done something bad) I have the option of telling you this or of concealing this thought from you. If I do tell you you may be hurt or offended, but at least you will know what I think, and you will know that I am interested in your actions and your appearance. Telling you could promote our closeness. Not telling you is more likely to promote harmony. In a situation like this Polish culture, or Russian culture, would tend to opt for telling (that is, for closeness), and Anglo-American culture, for not telling (that is, for harmony). Or suppose that I have done, or want to do, something that I think you would disapprove of. Should I tell you or not? If I tell you, this will promote our closeness, but it will disrupt our harmony and peacefulness. You may feel something bad because of this, you may feel angry, you may express your disapproval, and you may make me angry and upset. If I don't tell you, there will be no ill-feeling, but we will not be close. Again, in a situation like this, Polish culture, or Russian culture, would probably opt for telling, and Anglo-American culture for not telling. The attitude of a person who cherishes and seeks closeness with another person can be portrayed as follows: I I I I

want you to know what I feel/think/want know that you can feel something bad because of this know that I can feel something bad because of this want you to know it because I know that you feel something good towards me I think you know that I feel something good towards you

If one wanted to put a global label on this attitude, one might suggest 'self-disclosure', or 'openness', but this would be misleading. As we have seen, Barnlund (1975a) interprets his findings concerning American culture in terms of 'self-disclosure', but clearly, the attitude he is talking about is quite different from that portrayed here. In the 'self-disclosure' discussed by Barnlund, the stress is on 'self', on 'I', on saying what 1 think. In the 'closeness' discussed here the stress is on the relationship between 'I' and 'you', on good feelings between 'I' and 'you', and on a desire to continue and to promote a special relationship between us two, even at the cost of hurt and conflict.

Further illustrations: same labels, different values

111

Needless to say the 'closeness' portrayed here has little to do with the attitude reflected in utterances such as Got the time, mate? Mind if I smoke?

Utterances of this kind could be described as informal or casual (among other things), but if they were said to reflect either 'intimacy' or 'closeness', one would have to say that the words 'intimacy' and 'closeness' are being used in some technical sense, not in the everyday sense, and that without clear definitions such use of these words obscures, rather than clarifies, the attitudes involved.

3.3. 'Informality' Informality is a cultural attitude which, as we have seen, is frequently confused with intimacy or closeness. In Australia, when one rings a travel agency, one will often hear a response including the travel clerk's first name, for example: American Express, Cathy speaking.

If one were to believe Hijirida - Sohn (1986) one might conclude that this travel clerk expresses intimacy or closeness towards her customers. I have argued, however, that intimacy involves a 'special relationship' between two people, which certainly does not apply in the present case: the travel clerk cannot be claiming a 'special relationship' with every anonymous caller. Nor can she be claiming deep personal knowledge of the addressee, associated, as I have argued, with 'closeness'. What is signalled by her self-presentation, then, is neither 'intimacy' nor 'closeness', but rather characteristic Australian 'informality' - the same informality which, for example, Australian university students express by addressing their lecturers by their first name, or which Australian public servants express by addressing their colleagues, and most of their superiors, by their first names. What is the meaning of this near-universal Australian 'informality'? I think the essence of 'informality' (at least as practised in Australia) lies in the purposeful rejection of any overt show of respect, with implications of familiarity, friendliness, and equality. Thus, by saying Cathy speaking the travel clerk is inviting the anonymous callers to treat her as if they knew her well, to assume that she 'feels something good

112

Cross-cultural pragmatics and different cultural values

towards all callers, the present caller included', and that there is no need to show overt respect towards her (for example, by calling her Miss, Mrs, or Ms). A university lecturer or a branch head who invites his or her students or subordinates to address him or her as Bob or Jane conveys a similar attitude. Very roughly: (a) you don't have to 'show overt respect for me' (b) I want you to speak to me as people do when they think: (c) we know one another well (d) we feel something good towards one another (e) we can speak to one another in the same way Component (c) of this explication implies familiarity, component (d), mutual 'good feelings', and component (e), egalitarianism. Component (b) shows that the speaker doesn't really have to know the addressee, to have personal good feelings towards the addressee, or to claim full equality and full symmetry in his or her relation with the addressee (for example, the travel clerk may well call the addressee Mrs Brown or Dr Smith while calling herself Cathy). By using one's first name, or the addressee's first name, the speaker is evoking a certain prototype of human relations (spelled out in the components (c), (d), and (e)), and this is, I suggest, the essence of 'informality'. In addition, however, 'informality' has to be opposed to 'formality'; this is reflected in the following, additional component: (f) I know: people can't always speak like this to other people 'Formality' is not always associated with hierarchical human relations and with anti-egalitarianism. For example, in Australia, at formal meetings of a university faculty, everybody speaks in a very 'formal' way, without dissociating themselves thereby from the Australian ethos of super-egalitarianism. In Polish culture, titles of respect are used widely, and mutually: 'informality' is not valued in the way it is in Australia; yet this relative 'formality' is linked with a democratic, relatively egalitarian ethos (cf. Davies 1984:331-336). On the other hand, in 'vertical' societies such as Korea or Japan (cf. Nakane 1972), the value placed on social hierarchy is closely linked with value placed on 'formality'. Hence, from a Korean or Japanese perspective, the 'informality' of the Australian or American culture may seem to be linked to their egalitarianism even more closely than it really is.

Further illustrations: same labels, different values

113

In fact, 'informality' does tend to be linked with egalitarianism, and 'hierarchy' does tend to be linked with 'formality'; but none of these links is straightforward, and none of them can be understood outside the whole complex of other cultural norms and values of a given society. Above all, the norms themselves have to be well understood and carefully defined.

3.4. 'Harmony' In the discussion of 'closeness' I have used repeatedly the word 'harmony' - another key word used widely in discussions of cross-cultural pragmatics. But of course 'harmony' is no more self-explanatory than ,self-assertion', 'indirection', 'intimacy' or 'closeness', and if one doesn't say what one means by it in a particular context, it can be as misleading as the other widely used global labels. In the literature, this word, too, has been used in many different and mutually incompatible senses. For example, while both Anglo-American and Japanese cultures can be said, and have been said, to value 'harmony', it is clear the Anglo-American culture doesn't aim at 'harmony' in the sense in which Japanese culture does; in particular, it doesn't aim at sameness, or apparent sameness, of thoughts. Patricia Clancy describes this Japanese view of harmony as follows: The Japanese reliance upon indirection is consistent with their attitude towards verbal conflict. As Barnlund points out, in Japan conversation is 'a way of creating and reinforcing the emotional ties that bind people together' with the aim of social harmony. Therefore, overt expression of conflicting opinions is taboo. Even conference participants ... , in contrast to their argumentative American counterparts, tend to express their views tentatively, in anticipation of possible retraction or qualification depending upon how they are received; they try to feel out the positions of their colleagues, seeking a common ground for establishing unanimity (Barnlund 1975[b]; Doi 1974). . .. Individuals may hold their own view, but, in the interests of group harmony, should not express it if it conflicts with the opinion of others. (Clancy 1986:215)

The following comment offered by Clancy (with reference to Doi 1974) sums up the Japanese approach to harmony in a particularly striking way:

114

Cross-cultural pragmatics and different cultural values

Since Japanese is a left-branching verb-final language, with negation appearing as a verb suffix, speakers may negate a sentence at the last moment, depending upon the addressee's expression. (Clancy 1986:214)

This attitude, which is certainly different from the Anglo-American idea of harmony, can be portrayed as follows: when someone says something I can't say: 'I don't think the same' someone could feel something bad because of this when people say: 'we all think the same' it is good As Clancy mentions above, Barnlund (1975b) links the attitude portrayed here with the aim of 'creating and reinforcing the emotional ties that bind people together'. This sounds rather like the Polish or Russian ideal of 'closeness'. In fact, however, the attitudes involved are almost diametrically opposed. In Slavic culture, saying 'I don't think the same' is seen as promoting rather than jeopardising 'closeness'; and 'causing people to feel something bad' (now) can be seen as promoting 'closeness' in the long run. Anglo-American attitudes to 'harmony' and 'closeness' are different again. When considered from a Slavic or East European point of view, Anglo-American culture must be said to be oriented to 'harmony' rather than 'closeness', but certainly not to the kind of 'harmony' sought in Japanese culture. Obviously, Anglo-American culture does not discourage people from saying 'I don't think the same'. It does, however, discourage them from saying 'what you think is bad', 'I don't want you to think this', 'I think something bad about you', and so on. Furthermore, it doesn't encourage people to say things which are likely to cause the addressee to 'feel something bad' (not even temporarily, in the interests of long-term 'closeness'). Anglo-American attitudes to unanimity and 'harmony' were spelled out earlier: I can say what I think you can say what you think we don't have to think the same this is good (no one has to feel something bad because of this) Polish attitudes to 'harmony' and 'closeness' are epitomised in the proverb

Further illustrations: same labels, different values

115

Kto si~ czubi ten si~ [ubi. 'People who peck one another on the head (like fighting birds) like one another.' As this proverb suggests, it is not only difference of opinions which is valued in Polish culture, but a forcefully, pointedly, and painfully expressed difference. This attitude can be portrayed as follows: I I I I

want to say what I think know: you can feel something bad because of this don't want not to say it because of this want you to know what I think

Once again, we must conclude that global labels such as 'harmomy' or 'distance' obscure rather than clarify the real differences between different cultures and different ethnographies of speaking.

3.5. 'Sincerity' The problem of 'closeness' in interpersonal relations is closely related to the problem of sincerity. It has often been said that in modem Western culture sincerity has emerged as one of the core values. For example, Trilling writes: If sincerity is the avoidance of being false to any man through being true to one's own self, we can see that this state of personal existence is not to be attained without the most arduous effort. And yet at a certain point in history certain men and classes of men conceived that the making of this effort was of supreme importance in the moral life, and the value they attached to the enterprise of sincerity became a salient, perhaps a definitive, characteristic of Western culture for some four hundred years. (Trilling 1972:5-6)

This may be so - but what does this crucial norm of 'sincerity' really mean? Trilling (1972:2) offers the following definition: "The word as we now use it refers primarily to a congruence between avowal and actual feeling". We could translate this definition into the following formula: if I don't feel X I shouldn't say 'I feel X' Is it true that the norm spelled out above is an important feature of Western culture? More specifically, is it true that it is an important feature of Anglo-American culture?

116

Cross-cultural pragmatics and different cultural values

It is interesting to note that the subjective experience of EasternEuropean immigrants in English-speaking countries often leads them to the opposite conclusion. In particular, Eastern European immigrants often complain about what they perceive as the 'insincerity' of English conversational routines, above all, of conversational openings such as How are you?, Nice to see you, Lovely day, isn't it, and so on. (Cf. Drazdauskiene 1981). The perceived 'insincerity' of the 'How are you?' routine consists both in the belief that the speaker doesn't really want to know how the addressee feels and is expecting the addressee to reply positively ('Fine, thank you', 'Very well, thank you', 'Not too bad') regardless of how the addressee really feels. Consequently, common positive answers ('Fine, thank you') are felt to be generally insincere, and the whole game is perceived as an exercise in shared insincerity. I can add to this my personal testimony as an immigrant and a bilingual: After seventeen years of living in Australia, I still find the pseudo-question How are you? a perplexing one, since my own cultural impulse is to try to reply sincerely, which I know I am not supposed to do. When I recently failed to reply promptly to this question, helplessly searching for words, my interlocutor laughed at me: 'Come on, this is not such a difficult question'. But to me, it is a difficult question, and I know that I share this difficulty with thousands of other East European immigrants in Australia and in America. I can't believe, therefore, that Anglo-American culture really cherishes and promotes the norm that Trilling attributes to it: 'if I don't feel X I shouldn't say "I feel X"'. On the contrary, I think that Slavic and Eastern European culture promotes this norm, and that by doing so it comes into conflict with Anglo-American culture. But this is not to say that I don't recognise the validity of what Trilling is trying to say about Western culture (as opposed to what he actually does say). Clearly, what he had in mind was not sincerity (or otherwise) of conversational formulae, but sincerity of certain kinds of self-disclosure. Trilling (1972:5) quotes in this connection Matthew Arnold's "wistful statement of the difficulty, perhaps even impossiblity, of locating the own self': Below the surface-stream, shallow and light, Of what we say we feel - below the stream, As light, of what we think we feel - there flows With noiseless current strong, obscure and deep, The central stream of what we feel indeed.

Further illustrations: same labels, different values

117

Matthew Arnold called the hidden self the 'best self', but, Trilling asks, 'is it the own self?' In Trilling's (1972:5) view, if there is anything deep down in me which corresponds to 'the archetype of human being', that is, to the 'mankind's best self', this is not my sole self: "I know that it coexists with another self which is less good in the public moral way but which, by very reason of its culpability, might be regarded as more peculiarly mine. So Hawthorne thought: 'Be true! Be true! Be true! Show freely to the world, if not your worst, yet some trait by which the worst may be inferred.' " This brings us, I think, much closer to the real meaning of 'sincerity' in European culture. It is not a question of never saying that one feels something that one doesn't feel; rather, it is a question of knowing what one really feels (including feelings that reveal something bad about oneself) and of being able to disclose those real feelings (especially those which show something bad about oneself) 'to the world'. Every human being is unique, and uniquely interesting because of this. We shouldn't try to appear 'good' to other people. Rather, we should try to reveal 'to the world' our uniqueness, and this involves, above all else, our 'badness': because our 'badness' is more original, and more interesting, than our 'goodness'. The cultural injunctions in question can be formulated as follows: I don't know what I feel I want to know it when I know it I want to say it I want people to know it I think that people can think something bad about me because of this I don't want not to say it because of this I believe that the attitude spelt out above may indeed be, as Trilling says, 'a salient, perhaps a definitive characteristic of Western culture', linked closely with the birth of Western individualism, with the emergence and growing significance of mirrors, self-portraits, diaries, autobiographies, 'confessions', introspection, and so on. My point is that we cannot capture, or identify, this characteristic by means of some global term such as 'sincerity'. Contemporary Anglo-American culture doesn't seem to place any premium on 'never saying that one feels something that one doesn't feel'. On the contrary: the routines of human interaction reflected in the English language encourage saying that one feels something good when

118

Cross-cultural pragmatics and different cultural values

one doesn't feel anything good. This is manifested, in a spectacular way, not only in the How are you? routine, but also in the conventions of letter-writing: the opening phrase Dear Sir expresses a good feeling towards an addressee who may be a complete stranger, and so does the closing phrase Yours sincerely. Phrases of this kind cannot be used in other European languages, certainly not in Slavic languages, which do not allow any formalised expression of clearly non-existent 'good feelings' (and which encourage expression of existing 'bad feelings '). Trilling (1972:3) opens his discussion of 'sincerity' in Western culture with a quote from Hamlet: This above all: to thine own self be true And it doth follow, as the night the day, Thou canst not then be false to any man.

It would appear, however, that in modem times the idea of 'being true to oneself' has become disassociated from that of 'never being false to any man'. This may be linked with the shift of emphasis from 'sincerity' to 'authenticity', which as Trilling points out, has taken place in modem times. A very considerable originative power had once been claimed for sincerity, but nothing to match the marvellous generative force that our modem judgment assigns to authenticity ... Still, before authenticity had come along to suggest the deficiencies of sincerity and to usurp its place in our esteem, sincerity stood high in the cultural firmament and had dominion over men's imagination of how they ought to be. (Trilling 1972: 12).

It seems to me that in modern times the two ideas linked by Shakespeare in the passage from Hamlet have become dissociated: the idea of 'being true to oneself' developed into something like that 'authenticity' discussed by Trilling, whereas the idea of 'not being false to any man' has given way to a modern Anglo-American virtue of social harmony, based on 'distance' and on avoidance of interpersonal clashes. The virtue of 'authenticity' has to do with the notion of 'self', and of a true, genuine, and uninhibited expression of one's self. It does not involve the relation between 'I' and 'you'. As far as the relation between 'I' and 'you' is concerned the emphasis seems to have shifted from 'sincerity' to the avoidance of clashes, to smooth, well-greased, harmonious social interaction. Conventional expressions and conventional routines such as Dear Mr X, How are you?, Lovely to see you, Nice to have met you, Lovely day, isn't it, and so on, provide the oil for such harmonious social interaction. The expansion of such expressions fits

Further illustrations: same labels, different values

119

in logically with the modern Anglo-American constraints on direct confrontation, direct clashes, direct criticisms, direct 'personal remarks' - features which are allowed and promoted in other cultures, for example, in Jewish culture (cf. Schiffrin 1984) or in Black American culture (cf. for example Kochman 1981), in the interest of cultural values such as 'closeness', 'spontaneity', 'animation', or 'emotional intensity', which are given in these cultures priority over 'social harmony'. This is why, for example, one doesn't say freely in (white) English, 'You are wrong', as one does in Hebrew (cf. Schiffrin 1984) or 'You're crazy', as one does in Black English (cf. Kochman 1981:46). Of course some 'Anglos' do say fairly freely things like Rubbish! or even Bullshit!. In particular, Bullshit! (as well as You bastard!) is widely used in conversational Australian English. Phrases of this kind, however, derive their force and their popularity partly from the sense that one is violating a social constraint. In using phrases of this kind, the speaker defies a social constraint, and exploits it for an expressive purpose; indirectly, therefore, he (sometimes, she) acknowledges the existence of this constraint in the society at large. Generally speaking, in mainstream Anglo-American culture one has to be rather careful as to what one says about 'you' (because one prefers to avoid confrontation, preserve harmony, avoid the impression of imposing or interfering, and so on); at the same time, one can be less circumspect in saying things about oneself, although here, too, there are various constraints and restrictions, such as the constraint on 'bragging' or the constraint on the expression of one's bad feelings towards the addressee, or the constraint on 'emotional displays'. The general norm, then, can be portrayed as follows: I can say what I think/want/feel other people can say what they think/want/feel Some of the constraints on this general norm can be formulated as follows: (1) I can't say: I am good I can do things that other people can't (people would think something bad about me because of this) (2) I can't say: I feel something bad towards you I think something bad about you

120

Cross-cultural pragmatics and different cultural values

(3) I can't say: I want you to do something that you don't want to do (4) I can't always say what I feel (people could think something bad about me because of this) (5) I can't say things when someone else is saying something One might say, then, that a curious paradox is involved in the position of 'sincerity' as a cultural value in modem Anglo-American culture. On the one hand, as Trilling says, the very word sincerer ly) has come to have an air of insincerity about it - and, yet, as Goldstein - Tamura (1975) point out, 'Anglos' (in contrast to the Japanese) go to great trouble to sound sincere. To achieve this, they seek to express their feelings in a personalised way, in contrast to the Japanese, who rely on standard forms and do not view 'cliches' or ready-made formulae in a negative way: To the American, the Japanese method of standard messages, such as 'Congratulations' with only a name, the presentation of a gift with a standard phrase, a refusal with a standard phrase before acceptance ... may seem very bare indeed and perhaps somewhat insincere. (Goldstein Tamura 1975:91) [emphasis added] The American guest expressing thanks to his host at the end of dinner has no ... standard form, but rather makes use of a variety of possibilities generally emphasising the success of the meal, with or without an expression of thanks, such as 'Thank you for the delicious dinner' or 'What a delicious meal that was' (more informal) or 'Boy, that was great!' (colloquial-slang), each said with appropriate intonation to express sincerity. (Goldstein - Tamura 1975:72) [emphasis added]

This search for a personalised expression of, so to speak, predictable feelings, seems to reflect a tension between the value of authenticity, of 'being true to oneself', and the search for friendly, harmonious relations with other people; between the desire to express one's 'real self' ('this is what I feel/want/think') and the desire to have friendly interpersonal relations with other people and to ensure that 'everyone feels something good'. The desire to have friendly relations with other people may lead one to say things which do not correspond to what one really feels and thinks. The awareness of this, and the value placed on both 'harmony' and selfexpression, may lead to an attitude which can be portrayed as follows:

Different attitudes to emotions

121

(a) I say: I feel something (b) I say this because I feel this (c) I know: you can think that I say this because I think I should say it (d) I don't want you to think this (e) I say this because I feel this In Japanese culture, there is no room for this kind of attitude because the emphasis is very largely on saying what one thinks one should say, not on saying what one really feels. Hence, there is no perceived need to use a personalised form 'to ring true'. . .. the American speaker makes a personal connection to the hearer [sic] while at the same time expressing his own personality in the arrangement of words he chooses to use. The Japanese speaker, having the form at his disposal, shows chiefly his awareness of his obligation by using the verbal form at the appropriate level at the appropriate time. (Goldstein - Tamura 1975:80)

4. Different attitudes to emotions Different cultures take different attitudes to emotions and these different attitudes to emotions influence, to a considerable degree, the ways people speak. (See for example Lutz 1986, 1988; Wierzbicka, to appear.) Differences of this kind cannot be satisfactorily explained by means of any global labels such as 'emotional' or 'anti-emotional', 'expressive' or 'non-expressive'. They can, however, be made clear by means of semantic explications. In what follows, I will try to present thumb-nail sketches of several cultures, considered from the point of view of their characteristic attitudes to emotions.

4.1. Polish culture Like other Slavic cultures, Polish culture values what might be called uninhibited emotional expression: I want to say what I feel

122

Cross-cultural pragmatics and different cultural values

This includes both good feelings and bad feelings: I feel something good/bad I want to say it As mentioned earlier, it values in particular expression of good feelings towards the addressee: I feel something good towards you (I want you to know it) and it offers for this purpose a wealth of linguistic resources, such as a very rich system of hypocoristic forms of personal names, and also a rich set of terms of endearment. The latter point is illustrated by Polish terms widely used in everyday speech, particularly in speech directed at children: ptaszku 'dear little bird', kotku 'dear little cat', sloneczko 'dear little sun', iabko 'dear little frog', skarbie 'treasure', z/otko 'dear little gold', and so on (cf. Wierzbicka, to appear). Many other features of the Polish ethnography of speaking can be explained in terms of this cultural attitude. For example, cordial imperatives and 'impositives' ('have some more', 'you must have some more', 'you must stay a little longer', and so on) are clearly related to it (cf. Chapter 2 above). Polish principle of 'cordiality' I feel something good towards you I want good things to happen to you I want to be with you

4.2. Jewish culture Emotional self-expression was also highly valued in traditional (East European) Jewish culture, as described, for example, by Matisoff (1979). In this culture, however, good and bad feelings were generally expressed by means of good and bad wishes. Hence the tremendous importance of curses and blessings in Yiddish speech. This characteristically Jewish style of emotional expressiveness is well illustrated in the following passage: There are as many types of curses as there are people cursing, but the hardest to explain is the mother cursing her child. The child may be crying

Different attitudes to emotions

123

because he is hungry. The mother bursts out, 'Eat, eat, eat. All you want to do is eat. May the worms eat you. May the earth open up and swallow you alive.' This mother loves her child, she is only pouring out the bitterness that's in her heart in the only way she knows. But in translation she sounds like a monster. (Butwin 1958:9)

A few further examples of Jewish 'wishes' expressing the speaker's feeling (see Matisoff 1979): Governor Reagan, may he be erased, isn't giving any raise this year to my son the professor, a health to him. A black year on her, all day long she chewed my ear off with trivia. My mother-in-law, maya lament be known to her, has a wicked tongue. My wife - must she live? - gave it away to him for nothing.

Matisoff (1979:86) offers the following comment, which I believe expresses a deep insight: "Especially in the case of curses, the formulas may serve a purely therapeutic function. They are convenient, conventionalised ways of letting off steam - releasing bursts of psychic energy which might otherwise remain hopelessly bottled up ... " Following this insight, we can represent the pragmatic principle in question as follows (the bracketed component is optional): Jewish expressive curses X thinks of person Y (X thinks: person Y did something bad) X feels something because of this X wants to say something because of this X says: I want something bad to happen to Y

4.3. American black culture Uninhibited emotional self-expression is also characteristic of American black culture, as opposed to white culture. In this culture, however, there is no emphasis on 'good feelings towards the addressee', and there is no tradition of expressive wishes. Among several characteristic features of this culture which emerge from the rich literature on the subject I will single out the 'intense', 'emotional' character of black speech: the

124

Cross-cultural pragmatics and different cultural values

'animation', the 'heated tone' of discussions, the 'lack of detachment' in stating one's opinions and expressing one's thoughts, even on abstract, intellectual topics; the distrust of a deliberately dispassionate and detached mode of discussion favoured by white Anglo-American culture. Kochman writes: But they [blacks] have another reason to misinterpret (and distrust) the dispassionate and detached mode that whites use to engage in debate. It resembles the mode that blacks themselves use when they are fronting: that is, consciously suppressing what they truly feel or believe. As one black student put it, 'That's when I'm lyin'.' Fronting generally occurs in black/ white encounters when blacks perceive a risk factor and they decide it would be more prudent to keep silent than to speak. (Kochman 1981:22)

Black culture values, then, and promotes the following attitude: I think something I feel something because of this I want to say it White Anglo-American culture values and promotes what one might call loosely the opposite attitude: I think something I want to say it I don't feel anything because of it The two cultural norms in question could also be represented as follows: Black American X thinks something X wants to say it one can see that X feels something because of this people think: this is good Anglo-American X thinks something X wants to say it one can't see that X feels anything because of this people think: this is good

Furthermore, for blacks, views are inseparable from values and values are closely linked with emotional involvement. Consequently, "blacks present their views as advocates. They take a position and show that they care about this position" (Kochman 1981 :20) - and they care about

Different attitudes to emotions

125

it because they think it is good. By contrast, whites tend to present their ideas as spokesmen, not advocates. "How deeply a person cares about or believes in the idea is considered irrelevant to its fundamental value. . .. Whites believe that caring about one's own ideas, like the infatuation of scientists with their own hypothesis, will make them less receptive to opposing ideas." (1981:21). Kochman speaks in this connection of the separation of 'truth' and 'belief' in Anglo-American culture, and he links the norms of dispassionate, neutral objectivity, and of detachment from one's ideas with the desire to discover 'the real truth' (which is seen as involving only sentences, not people). He points out that in this culture the merits of an idea are seen as intrinsic to the idea itself, and that emotional involvement with ideas is seen as something that can only prevent people from being able to assess their intrinsic values.

Black American I think something I feel something because of this I think it is good to think this I want other people to think this Anglo-American I think something I don't feel anything because of this I know other people don't have to think the same I want to say what I think I want other people to think about it I want to know what other people think about it The conviction that our ideas are good and the attitude of emotional attachment to them leads in black culture to what Kochman (1981:23) calls 'dynamic opposition', an attitude which is perceived as a unifying rather than a divisive force. "Whites attempt to minimise dynamic opposition within the persuasive process because such confrontation, or struggle, is seen as divisive. Blacks, however, see such struggle as unifying ... It signifies caring about something enough to want to struggle for it." I think that to account for this 'dynamic opposition' and for its 'unifying force' we could add to the formulae sketched above the following components:

Black American I know that you don't think the same

126

Cross-cultural pragmatics and different cultural values

I think this is bad I feel something because of this I want us to think the same Anglo-American I know that you don't think the same I don't think this is bad I don't feel anything because of this I think: we don't have to think the same

4.4. Japanese culture In Japanese culture, as we have seen, the prevailing norm with respect to emotions is this: I don't want someone to feel something bad This is manifested in countless ways in the Japanese ethnography of speaking, but perhaps most spectacularly in the omnipresence of apologies and quasi-apologies in Japanese speech (cf. Coulmas 1981; cf. also Mizutani - Mizutani 1987). The importance of apologies in Japanese culture is epitomised in the fact that in the Japanese version of Little Red Riding-Hood the wolf has to appear at the end with tears in his eyes asking for forgiveness (Lanham 1986:290). The theme of indebtedness which pervades Japanese social interaction is related to this omnipresence of apologies, and also to the lack of boundaries between acts which from a Western perspective would be interpreted as apologies and thanks. Roughly: (1) I did something (that was bad for you) I think you could feel something bad because of this I feel something bad because of this (2) You did something good for me I didn't do something like this for you I feel something bad because of this Thus, both the constant fear that someone may feel something bad because of us and the constant awareness of unrepaid good things that other people have done to us lead to the humble expression of our own 'guilt': I did/didn't do something (to/for you)

Different attitudes to emotions

127

I feel something bad because of this In addition, Japanese culture places very high value on empathy, on anticipating what other people might feel. It is this high sensitivity to other people's (unexpressed) feelings that causes the Japanese to 'mask' and conceal their feelings: In social interaction, Japanese people generally are expected to restrain, if not suppress, the strong or direct expression of emotion. Those who cannot control their emotion are considered to be immature as human beings. Strong expression (verbal or nonverbal) of such negative emotions as anger, disgust, or contempt could embarrass other people. Direct expression of sorrow or fear could cause feelings of insecurity in other people. Expression of even happiness should be controlled so that it does not displease other people. The best way to comply with this social code of behavior is to utilise masking techniques. Thus, Japanese people, although unaware, frequently display apparent lack of a meaningful facial expression, often referred to as 'inscrutable' by Western people. It is an attempt to neutralise strong emotions to avoid displeasure or embarrassment on the part of other people. (Hanna - Hoffer 1989:88-90)

This can be represented as follows: I don't want to say what I feel someone could feel something bad because of this This focus on empathy is manifested not only in the constant attempts to avoid anything that might hurt or offend the addressee, but also in the attempts to anticipate and guess other people's unexpressed needs and wishes (cf. Lebra 1976). The Japanese principle of 'empathy' and 'consideration' X thinks: if I do something (Y) this person can feel something bad/good because of this I will not/will do it because of this this person doesn't have to say anything

This stress on empathy is linked to the Japanese reluctance to verbalise feelings at all - due largely to the fear that by doing so one may hurt or offend other people, but also to the conviction that feelings should be expressed, and understood, without words; and furthermore, that feelings cannot be really expressed by words. Goldstein and Tamura

128

Cross-cultural pragmatics and different cultural values

comment in this connection: In general, ... Americans ... tend more to the feeling that they can express sentiments directly and personally through the medium of language.... To the Japanese, on the other hand, 'No word conveys sorrow; you must look at the color of the eyes'. (Goldstein - Tamura 1976:92-93)

The attitude epitomised in this proverb can be portrayed, roughly, as follows: I don't want to say what I feel one can't say what one feels

4.5. Javanese culture As we have seen earlier, not saying what one feels is also highly valued in Javanese culture. Here, however, the motivations seem to be rather different than in Japan: it is not so much the belief that feelings cannot be expressed in words, or the preference for wordless empathy, or the consideration for other people's feelings, but rather, a desire to protect one's own equanimity and peace of mind, which could be threatened by an overt expression of feeling. Thus, Geertz writes: If one can calm one's most inward feelings (by being trima, sabar, and iklas), ... one can build a wall around them; one will be able both to conceal them from others and to protect them from outside disturbance. The refinement of inner feeling has thus two aspects: the direct internal attempt to control one's emotions represented by trima, sabar, and iklas; and, secondly, an external attempt to build a wall around them that will protect them. On the one hand, one engages in an inward discipline, and on the other in an outward defence. (Geertz 1976:241)

This can be reflected in the formula: I don't want people to know what I feel It would seem, then, that while neither the Japanese nor the Javanese want to say what they feel, the Javanese, in addition, don't even want others to know what they feel; and they want to restrain not only the external expression of feelings, but also internal emotional experience (an attitude reminiscent of the Stoic apatheia 'freedom from emotional disturbance', see Wierzbicka, to appear, chap. 6).

Conclusion

129

The management of one's emotional economy becomes one's primary concern, in terms of which all else is ultimately rationalised. The spiritually enlightened man guards his psychological equilibrium well and makes a constant effort to maintain its placid stability. His proximate aim is emotional quiescence, for passion is kasar feeling, fit only for children, animals, peasants, and foreigners. His ultimate aim, which this quiescence makes possible, is gnosis, the direct comprehension of the ultimate rasa. To feel all is to understand all. Paradoxically, it is also to feel nothing ... Emotional equanimity, a certain flatness of affect, is, then, the prized psychological state, the mark of the truly alus [refined] character. (Geertz 1976:239-240)

This would suggest an attitude which can be portrayed as follows: I want to feel the same all the time I think if I do something this will happen I think I can do this

5. Conclusion For intercultural understanding, "More than mere contact is essential. People must become capable of empathy, of being able to project themselves into the assumptive world, the cultural unconscious, of an alien culture. Yet this is a formidable task unless there are ways to introduce people to the assumptive world of others" (Bamlund 1975b:140). I have tried to show that there are ways to do this. We cannot enter the 'assumptive world of others' if we try to rely on culture-specific, complex, and obscure concepts such as 'directness', 'self-assertion', 'solidarity' or 'harmony'; but one can do it if we rely, instead, on lexical universals such as want, think, say, or know. Ruth Benedict (quoted in Barnlund 1975b:140) wrote: One of the handicaps of the twentieth century is that we still have the vaguest and most biased notions, not only of what makes Japan a nation of Japanese but of what makes the United States a nation of Americans, France a nation of Frenchmen, and Russia a nation of Russians. Lacking this knowledge, each country misunderstands the other. (Benedict 1947: 13)

What applies to different nations, applies also to different ethnic groups in a multiethnic society. What makes Japan a nation of Japanese, or

130

Cross-cultural pragmatics and different cultural values

Russia a nation of Russians is reflected - more clearly than anywhere else - in the ways the Japanese or the Russians speak. And the ways in which they speak can be summarised in clearly and universally accessible formulae couched in the natural semantic metalanguage.

Chapter 4

Describing conversational routines

It is a truism to say that different cultures, and subcultures, have different conversational routines, and that it is important that those different routines should be carefully studied, analysed and described. But the ways in which this self-evident program of research should be implemented are by no means clear or generally agreed upon. In this chapter I shall argue that despite the considerable effort which has gone into the description of conversational routines, much less has been achieved in this important area than might have been - because not enough thought has been given to the vital question of a metalanguage in which such analysis can be fruitfully carried out. To show how a suitable metalanguage can facilitate the description and comparison of conversational routines, I examine a number of generalisations suggested, or hinted at, in Anita Pomerantz's (1978) interesting paper on responses to compliments. I try to show why in the present form these generalisations are neither clear nor verifiable, and I propose ways of reformulating them which could make them clear and verifiable. I try to show how the use of the proposed metalanguage makes such reformulation possible, and how it enables us to describe conversational routines used in different societies in a way which can be illuminating, rigorous and free of ethnocentric bias. (For another attempt along similar lines, see Ameka 1987.)

1. Conversational analysis: linguistic or non-linguistic pragmatics? Many conversational routines in many cultures are lexicalised and/or grammaticalised, that is to say, they consist in uttering in certain situations certain phrases, or using certain constructions, which encode certain language-specific interactional meanings. It seems clear that

132

Describing conversational routines

meanings of this kind have to be revealed and described - like any other kind of meanings. This is the task of linguistic pragmatics. For example, English conversations are often started with the conventional phrase How are you? Leech (1983:198) comments on the meaning of this phrase by quoting a couplet: Don't tell your friends about your indigestion: 'How are you!' is a greeting, not a question. But as this couplet implicitly recognises, How are you? is not just a greeting, but a kind of cross between a greeting, a question, and an invitation for the addressee to say something about their current state something that is expected to be short and 'good' rather than long and 'bad'. But this kind of description, though useful as a starting point, is very imprecise and, what is worse, it is inherently ethnocentric. English words such as greeting, question and invitation belong to the English folk-taxonomy of speech acts and have no exact equivalents in other languages, so they cannot possibly be regarded as useful analytical tools for cross-cultural comparison. On the other hand, useful tools of this kind can be found in relatively simple words such as say, want, know, someone, something, good or bad, which have their semantic equivalents in all (or nearly all) languages of the world. Using such simple and (relatively) culture-free tools, we can formulate the meaning of the English phrase How are you? along the following lines: How are you? (a) I know: we can now say things to one another (because we have come to be in the same place) (b) I want to say something to you because of this of the kind that people say to one another when they come to be in the same place (c) I want you to know: I feel something good towards you (d) I say: I want to know 'how you are now' (e) I want you to say something because of this (f) 1 want you to say: 'I am well' (g) 1 think you will say something like this (h) I think we will feel something good because of this

Component (a) of this formula shows that the phrase in question is a conversation opener, or a potential conversation opener. Component (b) shows that the phrase constitutes an established linguistic routine

Conversational analysis: linguistic or non-linguistic pragmatics?

133

employed in such circumstances. Component (c) shows the friendly character of the phrase (absent, for example, from the otherwise comparable phrase Good-bye, cf. Wierzbicka 1987:224). Component (d) shows the speaker's (real or pretended) interest in the addressee's well-being. Component (e) indicates that How are you? - like a question - obliges the addressee to make a verbal response. Component (f) indicates what kind of response is expected (a positive one); and it may be taken to indicate the speaker's wish that the addressee should be well (I want you to say that you are well because 1 want to know that you are well, because 1 want you to be well). Component (g) shows the speaker's optimistic expectation that the answer will be positive (and betrays at the same time a reluctance to hear a negative one). Component (h) suggests that a positive answer will be a 'pleasure' to both interlocutors and hints that this shared pleasure will be conducive to social harmony between them. A response to How are you? involves typically three steps (the first being obligatory and the other two optional). First, one answers the interrogative element of the utterance by saying something like 'I am well' or 'I want to say: 1 am well (but 1 can't)'; second, one thanks the interlocutor (or rather, one says thank you, or thanks); and third, one reciprocates the act (And how are you?, And yourself?, etc.) Of these three steps, the second seems largely lexicalised, that is, determined not only semantically but also lexically. One cannot replace the expression thank you with, for example, the expression thanks a lot: A: How are you? B: Very well, thanks *a lot.

Step three is determined in (some aspects of) its meaning, not in its form: B: And you? / And yourself? / How about yourself?

Step one, too, is lexically 'free'. Semantically, the preferred response is some phrase which means, essentially, 'I am very well'. Using for a moment the kind of metalanguage employed by Pomerantz, we might say that the addressee can be expected to answer in 'strongly positive terms' or, if this is felt to be impossible, at least to avoid answering in 'strongly negative terms'. A sincere and spontaneous positive self-report ('I am well') will tend to be 'upgraded' to something like I'm very well or I'm fine, and a sincere and spontaneous negative self-report ('I am not well') will tend to be 'toned down' (or is it 'up'?) to something like Not very well, I'm afraid.

134

Describing conversational routines

Emphatically negative answers like Rotten or Lousy! are possible, but they are felt to violate the normal routine, and they tend to be delivered in a jocularly defiant manner, implying: 'I know this is not what I am expected to say'. It must be acknowledged, however, that while the phrase How are you? (as a conventional conversational opening) is part of the English language and should be included in an adequate dictionary of English (along with Good morning, Hello! and Hi!), the responses to this phrase are not similarly conventionalised, and should not be similarly listed in a dictionary. The range of such responses, and the normal strategies for formulating them, should of course be described, as an important part of the communicative competence of English speakers, but they would lie outside the boundary between linguistic pragmatics and nonlinguistic pragmatics. What I suggest, nevertheless, is that in a sense, proper methods of semantic description are equally relevant to both types of pragmatics. In particular, both types need a justified and 'culture-free' semantic metalanguage. It is not only the meaning of set phrases such as How are you? which has to be stated in a meaningful and well-justified semantic metalanguage, but also the meaning of loose descriptive expressions, such as 'an upgrading strategy', which despite their technical ring are metaphorical and have no clear meaning, and are therefore not empirically verifiable. As a starting point for a rigorous description and for possible verification, I propose the following generalisation. A reply to a How are you? has to take into account the following set of assumptions: (a) (b) (c) (d)

I know that you want me to say something good I know that you don't want me to say something bad I think that you think I will say something very good I think that you think I will not say something very bad

The first speaker's presumed wishes and expectations can of course be violated in the addressee's response, but they cannot be ignored: the addressee must realise that any such violation will be seen as a departure from the norm, and will be interpreted accordingly. The range of expected responses generated by these guidelines can be described as follows: (A) I want to say something very good [Very well; Fine; etc.]

Conversational analysis: linguistic or non-linguistic pragmatics?

135

(B) I want to say something good [Good; I'm well; etc.] (C) I can't say something very good I don't want to say something very bad [Not too bad; I'm OK; etc.] (D) I can't say something good I don't want to say something very bad [Not too good; Not very well; etc.] In addition to these expected conventional responses there is also the slightly humorous strategy mentioned earlier, which involves a conscious violation of the convention: (E) I don't want to say what you think I will say I don't want to say what you want me to say I want to say what I think [Rotten; Lousy; Terrible; etc.] One final strategy which should perhaps be recognised as a separate possibility could be represented as follows: (F) I don't want to say something good I don't want to say something bad I don't want you to think I say something that I don't think I can say something good [Not bad; etc.] On the surface, this type of response may be difficult to distinguish from type (C), but the intention behind each of them is different. Type (F) implies that one is well but that one doesn't want to sound insincere and gushing; it can be interpreted, therefore, as a kind of understatement. Type (C), on the other hand, implies that one is not well, but that one doesn't want to complain; it can be interpreted, therefore, as a kind of overstatement. As pointed out by Sharon Henschke (p.c.), the potential ambiguity can often be resolved by means of an interjection, and/or intonation. The utterance: Oh, not (too) bad.

will be interpreted as implying 'I am not (quite) well' (type C), whereas a bright and cheerful Not bad, or Not too bad, (accompanied by a grin) will be taken to imply 'I am well'.

136

Describing conversational routines

It appears that different social groups of speakers of English tend to favour different response strategies. For example, I would hypothesise that strategy (A) is used more frequently by women than by men; and also, that it is used more frequently by Americans than by Australians; and that the reverse would be true of strategies (E) and (F) (cf. Renwick 1980). These are of course matters for empirical investigation. But they could not be empirically investigated if we didn't frame the initial hypotheses in a rigorous and (at least relatively) culture-free manner.

2. 'Compliment response' routines Turning now to Pomerantz's (1978) 'responses to compliments', we must ask, first of all, what exactly she means by 'compliments', for it seems fairly clear that she is not using this word in the ordinary sense. For example, she includes among 'compliments' utterances such as the following ones: B: Well anyway nice talking to you A: Nice talkin to you honey (Pomerantz 1978: 107) Furthermore, she often appears to be using interchangeably words such as 'compliments', 'praise' or 'credit', which in the English folktaxonomy of kinds of speech acts stand of course for different categories (cf. Wierzbicka 1987). I presume that what Pomerantz really has in mind is a class of speech acts which can be characterised in terms of the following semantic component: I want to say something good about you She suggests that the addressee's problem consists in responding to such an utterance in a polite and 'supportive' manner while at the same time avoiding explicit or implicit 'self-praise'. We can reformulate this by portraying the addressee's (expected) attitude as follows: I don't want to say something good about myself What can an addressee do to implement these conflicting guidelines without being impolite or 'uncooperative'?

'Compliment response' routines

137

To answer this question, Pomerantz introduces a number of theoretical constructs such as 'agreements' and 'disagreements', 'upgrades' and 'downgrades', and 'referent shifts'. These are introduced by means of illustrations rather than any clear definitions. It appears, however, that the essence of what is intended can be captured in the following formulae: 'agreements' A:X B: I think the same 'disagreements' A:X B: I don't think the same 'upgrades' A: X is good B: X is very good 'downgrades' A: X is very good B: X is good 'referent shifts' A: I want to say something (Y) about X B: I want to say it (Y) about something other than X

One cannot be sure, of course, that these formulae do in fact correspond to what Pomerantz had in mind, because those intentions have only been hinted at in loose and metaphorical terms. They could be, however, easily revised and amended if this proved necessary or desirable. The important thing is that they are explicit, and that they force the analyst to be explicit and to make clear analytical decisions. For example, a term such as 'upgrade' may seem intuitively intelligible, but it doesn't make it clear whether it is meant to apply only to a substitution of 'very good' for 'good', or whether it is also meant to stand for a substitution of 'very bad' for 'bad', or 'good' for 'not bad', and so on. Trying in this way to tentatively translate Pomerantz's vague and metaphorical constructs into a controlled semantic metalanguage, I now proceed to consider some of her suggested generalisations in more detail.

138

Describing conversational routines

2.1. Upgrades Pomerantz introduces the concept of 'upgrade' as follows: One type of agreement, an upgrade, can be called 'optimal' on sequential grounds. Upgrades are prevalent in environments in which agreements are preferred; they occur in agreement terms and sequences and typically not in combination with disagreements. Upgrading techniques include the incorporation of stronger second evaluation terms, for example ... A: Isn't he cute? B: O::h he::s a::dorable [2] A: She seems like a nice little lady. B: [Awfully nice little person} (Pomerantz 1978:93) [1]

What exactly is an 'upgrade', then? Pomerantz uses the expression 'stronger evaluation terms', but it appears that in fact she means a 'good' evaluation rather than 'bad'. It seems, therefore, that essentially, an 'upgrade' consists in a replacement of 'good' by 'very good'. Since, however, in the quote adduced above 'upgrades' as a conversational strategy are linked with agreements (and are even called a type of agreement) one is led to conclude that the strategy which Pomerantz has in mind should in fact be represented as follows: A: I think X is good B: I think the same I would say X is very good The component 'I think the same' spells out what Pomerantz calls 'agreement', whereas the core of the 'upgrade' consists in replacing the first speaker's 'good' with the combination 'very good'. It is interesting to note in passing that the term 'upgrade' or 'upgrader' is used by other analysts in what appears to be a totally different sense: Request tokens were further analysed for the presence of linguistic elements that serve to mitigate, soften, or 'downgrade' the act, or those that serve the opposite function, that is, 'upgrading' aggravating elements. (Blum-Kulka - Danet - Gherson 1985:119)

The authors regard this 'definition' as self-explanatory, and as sufficiently precise to take it as a basis for counting: "The data showed 94 cases of downgraders and 118 cases of upgraders." (1985:119) This

'Compliment response' routines

139

shows, once again, how unreliable terms of this kind are, even when they are backed with seemingly precise statistical data.

2.2. Contrastive opposites Another conversational strategy singled out by Pomerantz is the 'contrastive opposite'. If an 'upgrade' constitutes 'an optimal agreement' (presumably, in the context of evaluations), a contrastive opposite constitutes 'an optimal disagreement'. Two examples: (1) A: Did she get my card?

B : Yeah she gotcher card. A: Did she t'ink it was terrible? B: No she thought it was very adohrable. (2) A: [ was wondering if [' d ruined yer - weekend by uh B: [No. No. No, [ just loved to have - ...

The examples provided suggest that a 'contrastive opposite' can be defined as follows: A: I think my X is very bad B: I don't think the same I think your X is very good Pomerantz (1978:93) writes: "Contrastive opposites are produced in environments in which disagreements are preferred, for example, subsequent to self-depreciations. ... negative, critical evaluations are followed by positive, complimentary ones." This suggests that 'selfdepreciation' ('my X is bad') is seen as a typical but not a necessary aspect of this routine. A more general formula would be: A: I think X is very bad B: I don't think the same I think X is very good

140

Describing conversational routines

2.3. Scaled-down agreements According to Pomerantz (1978:94), neither upgraded agreements nor contrastive opposites constitute common responses to compliments. What does tend to be used in response to a compliment is a 'scaled-down agreement', and, more specifically, a praise downgrade. For example: (1) A: I've been offered a full scholarship at Berkeley and at

UCLA. B: That's fantastic.

A: Isn't that good. (2) A: Oh it was just beautiful.

B: Well, thank you uh I thought it was quite nice. The intended generalisation seems to be this: There is (in English) a common conversational routine which can be represented as follows: A: I think B: I think I think I don't

something [that one can say] about you is very good the same it is good want to say: very good

2.4. Downgrades Another conversational strategy described by Pomerantz (1978:99) consists in "proposing diminutions of credit", whereby recipients "do not altogether negate or deny prior assertions but rather downgrade the prior terms". Two examples: (1) A: Good shot. B: Not very solid though.

(2) A: By the way I loved yer Christmas card. B: I hadda hard time, but I didn't think they were too good... Here, the generalisation seems to be this: A: I think your X is very good B: I don't think the same I wouldn't say it is very good because something about it (Y) is not good

'Compliment response' routines

141

Pomerantz (1978:100) comments: "Although these compliment responses are not contrastive opposites but are rather diminutions and qualifications of prior praises, they nonetheless are treated as disagreements." The component 'I don't think the same' reflects accurately, I think, the main point of this comment. As for the 'diminution' and 'qualifications', however, it is hard to be sure if the proposed formula corresponds to Pomerantz's intentions, because terms of this kind are so vague and metaphorical that it is by no means clear what exactly they are meant to stand for. The same uncertainty applies to the further proposal: "Subsequent to such disagreements, praise profferers may challenge or disagree with the diminutions and qualifications and reassert praise." (Pomerantz 1978: 100). This is illustrated as follows: [1] A: Good shot.

B: Not very solid (though). A: Ya' get any more solid, you'll be terrific.

[2] A: By the way I loved yer Christmas card. B: I hadda hard time, but I didn't think they were too good, but - finally, A: (Those) were lovely. I thought they were lovely. Pomerantz (1978: 101) generalises: "Recipients downgrade prior praise, and profferers upgrade the prior downgrades." Trying to state what appears to be the intended generalisation, one could suggest the following: A: I think your X is very good B: 1 don't think the same I wouldn't say it is very good (because something about it (Y) is not good) A: I think it (X) is very good Again, it seems clear that what Pomerantz means by 'downgrades' or 'downgraders' is very different from what, for example, Blum-Kulka Danet - Gherson (1985) mean. Yet both she and they regard these terms as self-explanatory, and do not try to define them. This leads to confusion, covered up by a semblance of precision. Semantic formulae of the kind proposed here prevent this kind of confusion.

142

Describing conversational routines

2.5. Reassignment of praise According to Pomerantz, a common type of compliment responses consists in 'referent shifts'. These are defined as follows: AI: A praises B A 2 : B praises other-than-self Two kinds of referent shifts are distinguished: 'reassignment of praise' and 'returns'. The first kind is illustrated with the following example: A: You're a good rower, Honey. B: These are very easy to row. Very light.

In exchanges of this kind, "in responding to a compliment, a recipient may reassign the praise, shifting the credit from himself to an otherthan-self referent" (Pomerantz 1978: 102). Apparently, the intended generalisation is this: A: I think that something about you is very good B: I would want to say something other than this I would say that something about something other than me is very good I have phrased these components in such a way as to avoid any suggestion of disagreement. In particular, I deliberately refrained from saying 'I don't think the same' or even 'I wouldn't say that', because I wanted to reflect correctly the sense of Pomerantz's (1978:105) comment: "The compliment response consisting of a second praise (albeit refocused) is partially supportive of, that is, a partial warrant for or legitimisation of, the prior praise."

2.6. Returns 'Returns', which are said to be particularly frequent in 'openings and closings of interactions', are defined as follows: AI: A compliments B A 2 : B compliments A

'Compliment responses' in different cultures

143

This is illustrated, among others, with the following examples: (1) - Ya' sound (justiz) real nice.

- Yeah you soun' real good too. (2) - Yer looking good. - Great. So'r you.

These examples suggest the following generalisation: A: I I B: I I I

want want want want want

to say something good about you you to feel something good because of this to do the same to you to say something good about you you to feel something good because of this

It appears that Pomerantz wants to extend the category of 'returns' to situations when the interlocutors are implying rather than saying good things about each other; for example: B: Well anyway nice talking to you A: Nice talkin to you honey (Pomerantz 1978: 107)

It seems to me, however, that the formula proposed above does fit examples of this kind, too: even if the speakers don't say any specific good things about one another, they both clearly convey that they would want to say something good about the interlocutor (and would want to make the interlocutor feel something good because of that); and that is all that the formula says.

3. 'Compliment responses' in different cultures Pomerantz opens her study of compliment responses by quoting a letter from a 'perplexed man' to the Los Angeles Times, together with the editorial response: Dear Abby, My wife has a habit of down-grading sincere compliments. If I say, 'Gee, Ron, you look nice in that dress,' her reply is likely to be, 'Do you really think so? It's just a rag my sister gave me.' Or if I tell her she did a

144

Describing conversational routines

great job cleaning up the house, her response might be, 'Well, I guess you haven't seen the kids' room.' I find it hard to understand why she can't accept a compliment without putting herself down. And it hurts me a little. How do you explain it, Abby? Perplexed Dear Perplexed, Your wife lacks self-confidence and feels somewhat embarrassed to accept praise. Don't be hurt. Most people have difficulty accepting compliments with grace. Abby

A crucial point which is missing from Abby's response, however, is that responses to compliments differ from culture to culture, and that within a complex society such as the United States they depend not only on people's character traits, such as 'lack of confidence', but also on their cultural background. It is quite possible that 'Perplexed' 's wife did not lack self-confidence or self-esteem but was simply Jewish, or was of East European or perhaps Chinese or Japanese background. More surprising than Abby's lack of attention to this aspect of the problem, is the fact that Pomerantz herself doesn't mention it once. Instead, commenting on the letter and the editorial response, she observes (1978:81): "A large proportion of compliment responses deviate from the model response of accepting compliments." But this is a strange observation. If a large proportion of compliment responses 'deviate from the model' of accepting compliments then is it not perhaps the model which is inadequate? Shouldn't one rather speak of a number of different models, operating in different cultures and subcultures? The compliment response reported by 'Perplexed' is not a 'deviation from the model', but a typical instance of a different model, prevailing in many cultures other than the American WASP culture. Using the metalanguage proposed here, we could describe the core of the common conversational strategy which perplexed 'Perplexed', Abby and Pomerantz, as follows: A: I think that something about you (your X) is very good B: I don't think the same I think that something about it (my X) is bad The preferred Anglo-Saxon strategy of what Pomerantz calls 'appreciation', and which Abby calls 'accepting compliments with grace', can be represented as follows:

'Compliment responses' in different cultures

145

A: I think that something about you (your X) is very good. B: Thank you [i.e.: I know you say this because you want to do something good for me (you want me to feel something good) I feel something good towards you because of this] It should be noted, however, that this 'appreciation' is largely lexicalised, as it is in responses to How are you? For example, as in the earlier case, if one said in response to a compliment Thanks a lot, this would sound sarcastic rather than 'appreciative'. Yet another model of compliment responses is characteristic of Japanese society, and more particularly, of Japanese women's speech. For example, Mizutani - Mizutani (1987:43) write: "the Japanese ... will never accept a compliment without saying iie ['no']". This response pattern is particularly striking in the light of the general Japanese reluctance to say 'no' under almost any circumstances (cf. Veda's 1974 discussion of sixteen ways to avoid saying 'no' in Japanese). Miller (1967:289-290) offers the following characteristic example: Female version: A: Md, go-rippa na o-niwa de goziimasu wa nee Shibafu ga hirobiro to shite ite, kekko de goziimasu wa nee B: lie, nan desu ka, chitto mo teire ga yukitodokimasen mono de gozaimasu kara, mo, nakanaka itsumo kirei ni shite oku wake ni wa mairimasen no de goziimasu yo. A: A, sai de gozaimasho nee Kore dake o-hiroin de goziimasu kara, hitotori o-teire asobasu no ni datte taihen de gozaimasho nee Demo rna, sore de mo, itsumo yoku o-teire ga yukitodoite irasshaimasu wa. ltsumo honto ni o-kirei de kekko de goziimasu wa. B: lie, chitto mo sonna koto goziimasen wa. A: 'My, what a splendid garden you have there - the lawn is so nice and big, it's certainly wonderful, isn't it!' B: 'Oh no, not at all, we don't take care of it at all any more, so it simply doesn't always look as nice as we would like it to.' A: 'Oh, I don't think so at all - but since it's a big garden, of course it must be quite a tremendous task to take care of it all by yourself; but even so, you certainly do manage to make it look nice all the time; it certainly is nice and pretty any time one sees it.' B: 'No, I'm afraid not, not at all.'

146

Describing conversational routines

According to Miller (1967:290), a male version of the same dialogue could look as follows: Male version: A: Ii niwa da nii? B: Un. A: 'It's a nice garden, isn't it?' B: ['Mm.'] The patterns underlying these two exchanges can be represented as follows: Female version: A: I think something about you (your X) is very good B: I don't think this I think it is not good A: I don't think this I think it is good B: I don't think this I think it is not good Male version: A: I think this X is good B: I think the same

What applies to compliments (especially in women's speech) applies also to praise, especially to praise directed at personal achievements. Thus, Mizutani and Mizutani write: Except among good friends, the Japanese usually deny any praise received from others.... To deny praise of one's skills or abilities, one will say: lie, madamada-desu. (No, I'm not any good at it yet. - lit. No, not yet.) lie, watashi-nanka dame-desu. (Oh, I'm so poor at it. - lit. Such a person as me is no good.) (Mizutani - Mizutani 1987:43)

This pattern can be represented as follows: A: I think you did something very good B: I wouldn't say this I am not good I can't do things well

Conclusion

147

If one's achievements are absolutely undeniable and praise cannot be simply rejected, one will use a slightly different pattern: Maa, nantoka. 'I manage to do it somehow.' Okagesama-de nantoka. 'Thanks to everybody I could manage.' Doo-yara koo-yara. 'Somehow or other I could.' (Mizutani - Mizutani 1987:43)

Similarly, according to another account: In daily life when being praised by others, the person praised often denies the merit by saying something like, 'I don't deserve the praise'. Otherwise, after expressing his/her thanks for the praise, the person adds some words in which he/she emphasises good luck, a favour from others, or the help of his/her surroundings. By so doing, he/she tries to ascribe his/her virtue or achievements to the power of something other than his/her own ability or efforts. (Honna - Hoffer 1989:240)

This pattern can be represented as follows: A: I think: you did something very good B: I know: it is not because I am good (I am not good) I believe that when different conversational routines are represented in this way, that is, if they are modelled in a controlled semantic metalanguage (derived from natural language and yet largely languageindependent), they can be clearly identified and easily compared, both within and across language and culture boundaries.

4. Conclusion Empirical analyses of conversational interaction of the kind undertaken by Pomerantz are, potentially, very important from the point of view of cross-cultural studies, both theoretical and applied. For example, it is of course very important for the immigrant to know what the rules of the

148

Describing conversational routines

different conversational games in the new country are, and how they differ from those prevailing in the old country. But comparisons of this kind cannot be carried out without a culture-independent descriptive framework, and in particular, without a culture-independent semantic metalanguage. In his Introduction to Studies in the organisation of conversational interaction, the volume containing Pomerantz's study, Schenkein (1978:3) wrote: "the descriptions presented here offer promising movement towards an empirically based grammar of natural conversation". I am all in favour of an empirically based study of natural conversation, and I regard Schenkein's volume in general, and Pomerantz's study in particular, as valuable, interesting and important. But to be truly fruitful, empirical studies of course require a well-justified theoretical framework, and adequate descriptions require not only reliable data but also a justifiable metalanguage. I suggest that categories such as 'upgrade', 'downgrade', 'acceptance' or 'rejection', 'supportive action', 'contrastive opposites', or 'returns' may be useful at some stage of analysis, but cannot be relied on as adequate analytical tools. They do not provide an adequate framework for describing and comparing conversational routines within anyone culture (for example, in the mainstream middle-class white American culture); and they are even less adequate as a framework for comparing the organisation of conversational interaction across cultural boundaries. Here as elsewhere in cross-cultural analysis, what is needed is language-independent and 'culture-free' analytical tools; and these can be found neither in English folk-categories such as compliments and credits, agree and disagree, or accept and reject, nor in arbitrarily invented vague and metaphorical labels such as upgrade or downgrade; but in universal (or near-universal) human concepts such as 'good' and 'bad', 'want' and 'know', 'think' or 'say' - that is to say, in the same 'natural semantic metalanguage' in which meanings encoded in different linguistic systems can be described and compared.

Chapter 5

Speech acts and speech genres across languages and cultures

1. A framework for analysing a culture's 'forms of talk' Every culture has its own repertoire of characteristic speech acts and speech genres. As Baxtin (1979:257) pointed out, in speaking as in writing, "we 'pour' our speech into ready-made forms of speech genres ... These forms are given to us in the same way in which our native language is given."3 Following Baxtin, I regard it as essential that complex speech genres and 'speech events' (such as lecture, letter, or gossip) should be treated, on some level, in the same way as simple speech acts (such as question, request, promise, or warning). This is not to deny that terminological and conceptual distinctions such as that between 'speech genre', 'speech event', and 'speech act' may be useful in certain contexts. But it is also important to stress that despite the tremendous variety of speech genres (using this term in Baxtin's sense, as a cover-all term) - in function, structure, and above all in length - they share, in an important sense, the same linguistic nature and require a unified descriptive framework. (The need for making such conceptual distinctions and yet studying the whole range of phenomena in question within a unified framework was pointed out in Hymes 1962.) The idea that different cultures can be studied and compared via their characteristic speech genres has by now become widely accepted, although it is seldom recognised that this statement applies to simple speech acts as much as to complex speech genres. It seems to me, however, that the cross-cultural study of speech genres has been severely hampered by the absence of a culture-independent and languageindependent semantic metalanguage.

150

Speech acts and speech genres across languages and cultures

1.1. The importance of folk labels I believe that one particularly fruitful way of approaching the repertoire of speech acts and speech genres characteristic of a given culture is via their folk names, that is, the special lexical units which have come to encode a culture's view of its most relevant 'forms of talk' (cf. Goffman 1981). This is not to say that all 'language games' (cf. Wittgenstein 1953) played in a given culture will necessarily have their own folk names. Nonetheless, it seems reasonable to assume that, generally speaking, those language games which do have such names are more relevant to a given culture than those which don't. In Gumperz' s (1972: 17) words, "members of all societies recognise certain communicative routines which they view as distinct wholes, separate from other types of discourse, characterised by special rules of speech and nonverbal behaviour." Crucially, "these units often carry special names" (1972: 17). Consequently, "one good ethnographic technique for getting at speech events, as at other categories, is through words which name them" (Hymes 1962: 110).

1.2. Two approaches Speech genres characteristic of a given culture are usually described in one of two ways: either from outside or from inside. If they are described from outside, researchers raise problems such as 'questions in Eskimo', 'commands in Zulu', or 'blessings and curses in Yakut'. If they are described from within, we read about speech genres such as namakke, sunmakke, kormakke in Cuna or rapping and capping in Black English (cf. Sherzer 1974; Abrahams 1970). The dangers of the first approach seem to be evident. English words such as question, command or blessing identify concepts which are language-specific. They embody an English folk taxonomy, which, like all folk taxonomies, is culture-specific. To approach a repertoire of non-English speech genres through a grid of English folk concepts means to risk a biased and ethnocentric description. On the other hand, if one describes speech genres characteristic of a culture from 'within', in terms of concepts such as namakke or rapping, the results may be free of any ethnocentric bias, but they risk being somewhat hard to grasp to the outsiders. Of course, studies such as Sherzer (1974) or Abrahams (1970) do a great deal towards explaining

A framework for analysing a culture's 'forms of talk'

151

the meaning and the illocutionary force of the relevant genres to the outside world, but the fact that these explanations take the form of lengthy studies and that they are never summed up in succinct and intuitively comprehensible formulae, constitutes, it seems to me, a serious obstacle to genuine understanding. The same applies of course to the first approach. Authors of studies such as 'questions in Eskimo' or 'commands in Zulu' or 'insults in Black English' often take great pains to explain that the speech acts or speech genres in question are not really 'questions', 'commands', or 'insults' in our sense of the term, and that they use these English terms only for convenience's sake. It seems to me, however, that since the explanations of what these genres really are also take the form of long studies which are never summed up in succinct and intuitively comprehensible formulae (other than the nearest English folk labels), it is very hard for the reader to grasp their culture-specific essence. To my mind, two things have to be recognised at the same time: First, folk names of speech acts and speech genres are culture-specific and provide an important source of insight into 'communicative routines' most characteristic of a given society; and second, to fully exploit this source, we must carry out rigorous semantic analysis of such names and express the results of this analysis in a culture-independent semantic metalanguage. Searle has claimed that: Illocutionary acts are, so to speak, natural conceptual kinds, and we should no more suppose that our ordinary language verbs carve the conceptual field of illocutions at its semantic joints than we would suppose that our ordinary language expressions for naming and describing plants and animals correspond exactly to the natural biological kinds. (Searle 1979:ix-x)

I would insist, however, that when philosophers of language discuss illocutionary acts such as 'promising' (cf. Searle 1969:54-71), 'requesting', or 'commanding' (cf. Searle 1979:3), they are not discussing any language-independent 'natural kinds'. They are discussing complexes of components which have been singled out from among countless combinations of components occurring in human communication by concepts which are English-specific. It seems fairly obvious that more or less ritualistic speech acts, such as 'baptising', 'exorcising', or 'absolving sins', cannot be viewed as culture-independent natural kinds, analogous to biological natural kinds. In my view, it is an illusion to think that acts such as 'promising', 'order-

152

Speech acts and speech genres across languages and cultures

ing', or 'warning' are any less culture-dependent (cf. Rosaldo 1982; see also Verschueren 1985). The authors of a recent study (Fraser - Rintell - Walters 1980:78) say that their research is based on the following assumption: "Every language makes available to the user the same basic set of speech acts, such as requesting, apologising, declaring, and promising, with the exception of certain culture-specific ritualised acts such as baptising, doubling at bridge, and excommunicating." Underlying this assumption is "the claim that if one language permits an act such as requesting, every other language will. Though there may be certain exceptions as one moves from the basic everyday acts such as requesting to the more culture-specific ones such as baptising, exceptions to this claim have not arisen and do not appear likely to, given what we know today about language" (1980:79). In my view, it would be difficult to base one's research on a less justified assumption.

1.3. Some examples: English vs. Japanese To focus on one example, a concept such as warning is not as languageindependent as the concept of mimosa pudica or felis domesticus is; rather, it is as language-specific as the concept of shrub or bug is. It is a function of the meaning of the English word warning, not a God-given or science-given Urdatum. English speech-act verbs codify a folk taxonomy of speech acts, not some culture-independent, scientific or philosophical taxonomy of modes of human communication. There are many languages which have no exact equivalent of the word warning and which have, instead, words for modes of communication which have no equivalents in English. For example, Japanese has the word satosu, which combines some of the components of the English concept codified in the word warning with some other components: an assumption that the speaker has authority over the addressee, the intention of protecting the addressee from evil, and good feelings toward the addressee (see Nevile 1981). In English, the assumption of authority is encoded in verbs such as order and forbid, but it is never combined (lexically) with the intention to protect. This combination of components: 'I am your superior; I am responsible for you; I don't want you to do anything bad; I care for you', is a characteristic feature of Japanese culture, in which the relationship between a superior and a

A framework for analysing a culture's 'forms of talk'

153

subordinate is likened to that between a parent and a child (see Nakane 1970, 1972; Lebra 1976; Smith 1983), and this feature is reflected in the meaning of the Japanese verb satosu. The fact that English doesn't have any verb which would combine authority, responsibility, and care seems also significant. Using the semantic metalanguage based on universal semantic primitives we can show both the similarities and the differences between different speech acts clearly and explicitly. For example, the relationship between warn and threaten can be represented along the following lines (cf. Wierzbicka 1987):

warn I say: if you do X something bad (Y) may happen to you I think: if you know it you may not do X I say this because I want you to know it threaten I say: if you do X I will do something bad (Y) to you I think: if you know it you may not do X I say this because I want you not to do X The Japanese concept encoded in the word satosu can be represented along the following lines (cf. Nevile 1981; Morimoto 1985):

satosu (a) I say: you should do X (b) it will be bad if you don't do it (c) I say this because I think I should say it (d) I think: you will do it because of this (e) you know that I feel something good towards you (f) I know: I can say things like this to you (g) you can't say things like this to me Component (a) of this explication shows that satosu is similar, in some respects, to the English concept of advise; (b) links it with the English concept of warn and admonish; (c) shows that the speaker feels responsible for the addressee's actions; (d) shows the speaker's confidence in his own influence; (e) spells out the source of this confidence, and of the

154

Speech acts and speech genres across languages and cultures

speaker's interest in the addressee's actions; while (f) and (g) spell out the asymmetrical character of the relationship. The asymmetrical character of satosu may seem to link it with such power-based English words as order or reprimand. But satosu does not refer to the speaker's power over the addressee. Rather, it seeks its legitimisation in the speaker's resonsibility for, and good feelings towards, the addressee, that is, in a parent-like attitude of the superior (speaker) towards the subordinate (addressee). Thus, the concept of satosu is a clear manifestation of the much-discussed Japanese 'paternalism'. In Japan: The boss ... is supposed to have parent-like feelings. Indeed the social expectations of his role often cause him to display overt behavior suggesting such feelings whether they are present in him or not. There are indeed positive fantasies of being protected directed towards superiors. ... This interplay of expectations resembles the expectations of increased responsibility directed towards the eldest son in the traditional family. He is to internalise the sense of responsibility for others under his authority. Responsibility for taking care of personal matters would be considered intrusive by subordinates were they to be exercised in the west. For example, business executives, foremen, or even higher executives in Japan will sometimes act as go-betweens in assuring a proper marriage for one of their subordinates. This is seen as a part of a parent-like responsibility and indeed a type of nurturant concern with the subordinate. (DeVos 1985:159)

As Lebra (1976:51) points out, in prewar Japan even "military units formed pseudofamilies consisting of pseudoparents and pseudochildren. A former officer is quoted as saying, 'The warrant-officer is like a housewife who takes good care of soldiers as a mother, while the companycommander may be likened to a father whose orders are strictly observed but who has the affection of kinship towards his soldiers'." (Lebra is quoting here from Minami 1953:157). If the superior is expected to behave like a parent, the subordinate is expected to assume the role of a child - not so much an obedient child, as a child who knows and can rely on the parent's affection. This, too, is reflected in the lexicon of Japanese speech-act verbs, and most particularly (as pointed out by Morimoto 1985), in the concept of nedaru. But first a couple of additional quotes from the work of social psychologists:

A framework for analysing a culture's 'forms of talk'

155

The 'child' -role player can expect to depend upon the 'parent' -role player for security and protection by appealing to the latter's oyagokoro ('parental sentiment'), which is characterised as warm, benevolent, and nurturant.' (Lebra 1976:51). In the traditional Japanese system there were no 'rights' on the part of the subordinate. The only recourse for subordinates in the past, since they had no contractual relationships, was to hope to induce kindness and benevolence in their superiors. These feelings were induced by invoking potential feelings of nurturance and appreciation from them. This capacity to induce kindness and benevolence in superiors in a manipulative manner is called amaeru in Japanese. It has been very cogently discussed at length by Takeo Doi [1973]. (De Vos 1985:159-160)

As pointed out by Morimoto (1985), Japanese has two verbs usually translated into English as 'ask (for)': tanomu and nedaru; but while tanomu can indeed be regarded as a semantic equivalent of ask, nedaru implies a rather different, characteristically Japanese concept, which reflects the psychology of amae. (For a semantic analysis of the concept of amae/amaeru, see Wierzbicka, to appear, chap. 4.) Nedaru invariably implies intimate relationship and I think there is something more than 'I assume you will want to do it' about this verb. It is used only by a subordinate to a superior with whom they have a close relationship ... and it shows that a speaker knows that a superior has a friendly feeling towards him and that he could take advantage of it. (Morimoto 1985)

Drawing on Morimoto's analysis of nedaru, I would propose the following explication of this concept: nedaru (a) I say: I want you to do something good for me (b) I say this because I want you to do it (c) I know: you don't have to do it (d) I think you will want to do it (e) because I know that you feel something good towards me (f) I know: you can do good things like this for me I can't do good things like this for you Both the English ask (for) and the Japanese tanomu include the first three of these components (a, b, and c), but they do not include (d), a confident expectation that one's wish will be granted because of the addressee's presumed benevolence; (e), an assumption of the addressee's

156

Speech acts and speech genres across languages and cultures

'good feelings'; or (f), an assumption that the relationship is inherently asymmetrical, not so much in terms of power and authority as in terms of nurturance and dependence. In the formulae proposed here, the meaning of both the English verbs warn and threaten and of the Japanese word satosu have been represented in a kind of simplified and standardised English. Nonetheless, these formulae could be readily translated into Japanese, because while words such as warn (or abuse, swear, or curse) do not have Japanese equivalents, semantically simpler words, such as I, want, know, do, happen, bad, or because, do have equivalents - not only in Japanese but also in most other languages of the world. For this reason, the semantic metalanguage used here can be regarded as, in a sense, languageindependent. Coulmas (1981 :70) writes: "The difficulty boils down to the general question of how speech acts can be cross-culturally compared and 'translated'. To treat speech acts such as thanks and apologies as invariable abstract categories is surely a premature stance." I would put it more strongly: It is not merely premature, it is downright ethnocentric. Coulmas (1981:81) elaborates: "After all, 'thanks' and 'apology' are Western words ... But the applicability [to other cultures] of such categories should not be taken for granted. In particular, we should not assume that names of speech acts of individual languages define universal types of speech acts." I couldn't agree more. Unfortunately, Coulmas continues, "With this in mind we can now approach the problem of the cross-cultural comparability of thanks and apologies. ... As regards apologies and thanks, it seems to be a reasonable assumption that they exist as generic speech acts in every speech community" (1981:81). But in fact, Coulmas himself convincingly shows that the concepts encoded in the English words thanks and apology don't really fit Japanese culture. Shouldn't one firmly conclude, therefore, that they are not suitable tools for cross-cultural comparison of speech acts? They may serve a purpose as long as one is trying to show that they don't fit cultures such as Japanese. We can also ask legitimately: What do the Japanese do in those situations in which we would apologise or thank? But to describe speech acts which are characteristic of Japanese culture in positive terms, one needs a metalanguage which would not be derived from English speech-act labels such as thank or apologise. To thank someone means, roughly, to say that we feel something good towards them because of something good they have done for us. To a European, this may seem a perfectly 'natural' and 'normal' response to a

A framework for analysing a culture's 'forms of talk'

157

favour received. But in Japanese culture, with its stress on social hierarchy, and above all, with its stress on obligatory repayment of all favours, this response is much less 'normal' and generally applicable. As Coulmas himself points out, in Japanese culture it is 'normal' to respond to other people's favours in the same way as one responds to one's own transgressions against other people: that is, by saying that one feels 'something bad' rather than 'something good' because of this. Thus, expressions such as sumimasen, literally, 'it never ends' (i.e., I am aware of my 'never-ending indebtedness' to you) are used both in situations calling, from a European point of view, for thanks, and in situations calling for an apology. It is not surprising, therefore, that Japanese doesn't have a verb corresponding to the English verb to thank. The closest word it has is kansha suru (of Chinese origin), but this could never be used with respect to a child thanking her mother for a present, or to a lecturer thanking her student for some service, and in fact, it is virtually restricted to written language (Kaoru Sakurai, p.c.). The idea that regardless of the status, rank, and the type of relationship one can always react to a favour in essentially the same way is alien to Japanese culture, and the absence of a general speech-act verb corresponding to thank reflects this. By translating both the English word thank and the Japanese word kansha suru into the metalanguage of universal semantic primitives we can reveal both the similarities and the differences between them, and in this way we can document cultural differences which may otherwise seem elusive and non-accessible to rigorous analysis. I propose the following (for a slightly different explication of thank, and for discussion, see Wierzbicka 1987): thank (a) I know: you did something good for me (b) I feel something good towards you because of this (c) I say this because I want you to feel something good kansha suru (a) I know: you did something good for me (b) I say: I feel something good towards you because of this (b') I know: I couldn't do something good like this for you (b") I feel something bad because of this (c) I say this because I think I should say it

158

Speech acts and speech genres across languages and cultures

Components (a) and (b) of these explications are the same; but the 'illocutionary purpose' (c) is different in each case, with the Japanese concept stressing a feeling of obligation on the part of the speaker. Most importantly, however, the Japanese concept implies an asymmetrical relationship (component b'), and a feeling of 'unrepayable debt', which links the feeling of quasi-gratitude with something like a feeling of guilt (component b"). It should be added that the expectation that one would find 'generic speech acts' such as thanks and apologies in most (if not all) cultures is completely unfounded. For example, among the Australian Aboriginal Yolngu people: Yolngu never express 'thanks' verbally to each other ... people normally do things for one of two reasons: either because they want to, or else because they have some obligation to fulfil to specific relations. Therefore, the yolngu for whom a balanda [white person] has just done a 'favour', automatically thinks that the balanda gave him the lift in his boat because he wanted to, so there is no need for any expression of thanks .... This analysis of the two reasons why yolngu do anything, is quite consistent with the principles of reciprocity ... because the principle of reciprocity acts within the system of obligations to specific relations. (Harris 1984: 134-135)

On the other hand, kinship-based obligations, which in part explain the absence of 'thanks' in many Australian Aboriginal languages, are reflected in other speech acts, which have no counterparts in European languages, and which have no place in the lexicalised taxonomies of speech acts in a language like English. This point will be considered more closely in the next section.

1.4. Another example: English vs. Walmatjari Every speech act or speech genre constitutes a bundle of illocutionary components: expressed intentions, assumptions, thoughts, feelings, and so on (see also Chapter 6). For example, the act identified in the English folk taxonomy as order (as in 'X ordered Y to do Z') contains, I suggest, the following semantic structure: order (I order you to do X) (a) I say: I want you to do X (b) I say this because I want you to do it

A framework for analysing a culture's 'forms of talk'

159

(c) I think you have to do it because of this (d) I think you will do it because of this The act identified in English by means of the verb ask (as in 'X asked Y to do Z') can be assigned the following semantic structure: ask (a) I (b) I (c) I (d) I

say: I want you to do something good for me (X) say this because I want you to do it think: you don't have to do it don't know if you will do it

But in the Australian language Walmatjari, spoken in Western Australia, there is another characteristic speech act, related to both the English ask (tolfor) and the English order, but equivalent to neither (see Hudson 1985). It is based on kinship rights and obligations, which are so characteristic of Aboriginal society. Thus, Walmatjari has a special word, japirlyung, for what might be called 'kinship-based requests', the idea being that a kinship-based 'request' cannot be refused. But in calling this speech act a 'request', one is, of course, committing an error: an act which, in the speaker's view, cannot be refused is not really a 'request'. To represent this act in a way which would be free of an Anglo-centric bias and yet which would be meaningful to an outsider, one can, I suggest, use universal (or near-universal) elementary (or near-elementary) semantic units, along the following lines: japirlyung (a) 1 say: 1 want you to do something good for me (X) (b) I say it because I want you to do it (c) 1 think: you have to do good things for me (c') I think you know: everyone has to do good things for some other people (because of the way we are related) (d) I think: you will do it because of this

Components (a) and (b) of this explication are the same as components (a) and (b) of the English verb ask (for). But component (d) expresses a confidence that the addressee will comply, associated in English not with ask but with order. The English ask, on the contrary, implies that the speaker doesn't know whether the addressee will comply: 'I don't know whether you will do it' - a component which links ask (tolfor) with questions. But the basis of confidence is quite different in the case of japirlyung than it is in the case of order. Order implies a hierarchical

160

Speech acts and speech genres across languages and cultures

relation: if I say to you that I want you to do something then you have to do it because of this. Japirlyung implies nothing of the sort: it doesn't imply that the addressee is under the speaker's authority or control; it only implies that the addressee has the obligation to do some 'good things' for the speaker because of the way they are related, and it recalls a general system of kinship-based obligations, a system which involves everyone in the community. (For a semantic analysis of the concept of 'relatedness', see Wierzbicka, to appear, chap. 9.) Hudson (1985:71) offers the following example of the use of japirlyung. Japirlyinya parla parri-ngu nganpayi kuyi-purru. asked he-him boy-ERG man meat-PURP 'The boy asked the man to give him meat (because of kinship obligations). '

She contrasts this with the use of the verb jinjinyung, which, like the English order, "implies the authority on the part of the speaker rather than obligation based on kinship" and which "does not include the component 'that is good for me', which is present in japirlyung". For example: Jinjinyinya manya yinparnu-purru. ordered he-them sing-PURP 'He ordered them to begin singing the corroboree.'

As Hudson points out, in this last example "the speaker has authority in the corroboree, and so he is able to order the performers to begin" (1985:71). Thus, the idea of a 'directive' based on authority is present (lexicalised) in both English and Walmatjari; on the other hand, the idea of a 'directive' based on kinship-obligations is present (lexicalised) in Watmaljari but not in English. It might be added that the idea of a directive based neither on authority nor on kinship-obligations (but on an appeal to individual good will) is lexicalised in English but apparently not in Walmatjari. This fits in well with Harris' remarks on the absence of the concept of 'thanks' in Aboriginal culture.

A framework for analysing a culture's 'forms of talk'

161

1.5. The elimination of vicious circles The use of relatively simple terms in the explications, which makes possible a rigorous comparison of concepts (both within one language and crosslinguistically) ensures also the elimination of the vicious circles which have plagued traditional dictionaries in general and dictionaries of synonyms and related words in particular. Traditionally (as pointed out in Chapter 1 above), a word like request has nearly always been described with reference to ask for, and vice versa (along the lines of 'to request' = 'to ask politely for something' and 'to ask for something' = 'to make a simple request'). For example, Hornby et al. (1969) define ask, in the relevant sense, as "request information or service", request as "make a request", and request (n.) as "asking or being asked for something". In my analysis, no speech-act verb can be defined in terms of another speech-act verb. The only verb referring to speech which can occur in the explications is say, which I regard as indefinable and which has the status of a universal semantic primitive. The other words used in my explications do not always have this status, but they are all relatively simple, and for present purposes can be treated as indefinable. The strict separation of the words which are being defined from the small set of relatively simple words which are used for defining prevents vicious circles and ensures that the analysis is real, in Aristotle's sense, that is, that it consists in reducing posteriora to priora (complex and relatively obscure concepts to simpler and relatively clear concepts), not in translating unknowns into unknowns.

1.6. Evidence for the proposed formulae The formulae proposed should predict correctly the entire range of a term's use. Usually, each formula goes through a great many versions, being checked and rechecked with native speakers, before an optimal version is reached - optimal from the point of view of accounting for all the aspects of the term's use. Accounting for the entire range of use means, for me, accounting for both the situations to which the term would be applied and the syntactic environments in which it would be used. Consider, for example, the verbs reveal, confess, and tell. Why can one tell someone a joke but not *reveal someone a secret or *confess

162

Speech acts and speech genres across languages and cultures

someone one's sins? Presumably, in all three cases, the addressee is affected by the action (in that the addressee comes to know something). But reveal and confess differ from tell in their implications with regard to the message. Revealing a secret to somebody crucially affects the secret, as well as the addressee, because the secret comes into the open. Similarly, if I confess my sins to somebody, my sins cease to be secret, they cease to be a burden on my soul, they may even be 'washed away' by an act of absolution. It is quite different with tell. Tell doesn't imply that the message is secret, as reveal does, and it doesn't imply that the message is a guilty one, as confess does. The message of tell can be quite trivial, and it need not to be affected by the speech act. For example, if I tell John a joke, it is likely that this will affect John (he will laugh), but it will not affect the joke. And even if the message is not trivial, for example, if I tell John 'the truth', John will probably be affected, because he will come to know the truth, but there is no implication that 'the truth' in question will be affected, for example, in the sense of becoming public or coming into the open, as a secret does when it is revealed. The relative unimportance of the addressee in comparison with the message implied by verbs such as reveal or confess is reflected in the ability of the addressee phrase to be omitted, as well as in its inability to be 'promoted' to the internal dative position: He revealed that he had spent ten years in jail. He confessed that he had spent ten years in jail. ??He told that he had spent ten years in jail.

In devising semantic explications for the verbs reveal, confess, and tell I have tried to differentiate the formulae accordingly. In this way, semantic analysis is used to explain differences and similarities in syntactic patterning, and syntactic patterning is used as evidence for semantic explications. (For further data and discussion, see Wierzbicka 1988, chap.6.)

1.7. The first-person format One last feature of the analysis offered in the present chapter requires a comment: the first-person (singular, present tense) format of the explications. This first-person mode of analysis is a major and deliberate departure from generally accepted conventions. It reflects my conviction

A framework for analysing a culture's 'forms of talk'

163

that the first-person form of speech-act verbs ('I warn', 'I request', etc.) is semantically simpler than all the other forms, and that expressions such as 'he warned', 'he requested', and so on, are semantically derived from the corresponding first-person expressions (or from their paraphrases). This is so despite the fact that some speech-act verbs (such as, for example, boast or threaten) are never used in the first-person present tense (in a performative sense). I believe that reports such as 'John boasted of X' or 'John threatened to do X' rightly or wrongly attribute to the author of a given speech act a subjective attitude which can only be described adequately in the first-person mode. To see this, it should be enough to consider the fact that the subjective attitudes in question are often expressed simply by intonation (cf. Deakin 1981) and that intonation has inherently a first-person meaning. The intonation can convey 'I am angry', but never 'she is angry'; it can convey 'I want you to do something' or 'I want you to tell me something', but never 'she wants him to do something' or 'she wants him to tell her something'. To describe the meaning of illocutionary forces in a third-person format is tantamount to deriving direct discourse from indirect discourse. Yet it is well known that many languages have no indirect speech and that in children's speech, utterances such as 'Daddy said: what time is it?' 'Mummy said to him: I love you' occur earlier than utterances such as 'Daddy asked what time it was'. 'Mummy told him that she loved him' (cf. Coulmas 1986). 1 am not suggesting that all verbs which can be used in reporting speech have an inherent first-person perspective. For example, verbs such as ejaculate, bark, babble, fume, thunder, mutter, or snap describe speech events from outside, spelling out the impression of an external observer. One would be unlikely to use them in the first person, even in the past tense: 'Too late', she barked / ?/ barked. '/ absolutely forbid it', thundered the Colonel / ?/ thundered.

Verbs of this kind describe the manner of speech, and while they do so partly in terms of emotions which one would assume they would be associated with, they imply nothing about the speaker's illocutionary purpose. They can be roughly paraphrased in the third person, for example:

164

Speech acts and speech genres across languages and cultures

"X", he thundered. = "X", he said, saying it in a way which would remind one of the sound of thunder, as people do when they want to show that they are angry and that they have power.

What real speech-act verbs (complain, boast, warn, order, promise, announce, etc.) imply about the manner of speech is that it must be compatible with the attitude attributed to the speaker; and they directly attribute to the speaker an attitude which can be accurately portrayed only in terms of a first-person perspective. (For a different view on this point, see Boguslawski 1988.)

1.8. The problem of other minds The first-person format of the analysis solves the paradox that in the view of some linguists (for example, Chomsky 1975) makes all semantic analysis of speech acts a futile enterprise. It is quite obvious that speech acts differ from one another in terms of the speakers' subjective attitudes, that is, in terms of their assumptions, intentions, and so on. However, other people's assumptions, intentions, and so on, cannot be observed and ultimately remain unknown to outsiders. How can we then develop a rigorous semantic analysis on the basis of these unknowns? I regard this objection as valid, but only with respect to the conventional third-person format of speech-act analysis. If John warns someone about something, nobody other than John knows what John's real intentions are. This, however, does not detract from the validity of the equation: I I I I

warn you = say: if you do X something bad (Y) may happen to you say this because I want you to know it think if you know it you may not do it

In explicating speech-act verbs in a first-person format we are modelling the attitudes conveyed in first-person expressions (such as 'I warn you') or attributed to the speakers, rightly or wrongly, in third-person reports (such as 'he warned'). Whether the assumptions and intentions expressed in our formulae are sincerely held by people who perform appropriate speech acts is, from a semantic point of view, quite irrelevant. I am not claiming anything about the real intentions of a person who warns, threatens, or requests. I am only claiming that when some-

Some Australian speech-act verbs

165

one says I warn you or Careful! This gun is loaded!, the attitude conveyed can be described in the verb warn, and that when someone says he warned me, that attitude is attributed to the speaker (cf. Skinner 1970; Hymes 1974b:182-183). Similarly, one can very well say: She ordered him to go, but she didn't really want to be obeyed. This means that it would be incorrect to describe a sentence such as 'X ordered Y to do Z' in terms of 'X wanted Y to do Z'. But it would be correct to describe it in terms of X's saying (or otherwise conveying the meaning of): 'I want you to do Z' (cf. Wierzbicka 1974). Semantics is not concerned with people's 'real' (as opposed to conveyed) intentions and assumptions. The task of speech-act analysis consists in modelling in explicit and verifiable formulae the attitudes that people convey in speech by conventional linguistic means (which, of course, include the intonation). The task of semantics in general consists in modelling in explicit and verifiable formulae the meanings which people convey in speech (again, by conventional linguistic means). In what follows, I am going to discuss a number of language-specific speech acts and speech genres, drawn from a few different languages, trying to show that each of them embodies a mode of social interaction characteristic of a particular culture. Informal discussion will be supplemented by rigorous semantic description, which will take the form of explications formulated in the proposed metalanguage of universal semantic primitives. I will start with a fairly extensive discussion of five speech-act verbs which have emerged as key concepts in Australian English. This will be followed by a more summary analysis of a few concepts which stand for more complex speech genres: the black English 'dozen', the Hebrew 'dugri talk', and the Polish kawal and podanie.

2. Some Australian speech-act verbs 2.1. Chiack (chyack)

The Australian English word chiack ['(t)fajrek], allegedly derived from "the cockney pronounciation of 'cheek' - impudent badinage" (Bulletin 1898, in AND), refers to a characteristically Australian form of social interaction and reflects a characteristically Australian form of humour.

166

Speech acts and speech genres across languages and cultures

(The word is highly colloquial and since it belongs, essentially, to spoken rather than written language, its spelling is variable.) Essentially, 'chiacking' consists in saying something bad about the addressee for shared fun. Australians themselves are inclined to see 'chiacking' as one of their favourite national pastimes, and forms of entertainment. Most examples of this word cited by either Wilkes (1978) or AND (1988) refer in fact to a habitual, rather than occasional, activity of 'chiacking'. In examples from these sources, Dawes (1943, in AND) talks about the "Australian passion for handing out chiack", and Hardy's reference to the old chiack indicates that 'chiacking' is very much part of the familiar (and hence positively viewed) Australian way of life: Hullo, hullo, Chilla said, always a bit too keen on the old chiack, especially when it came to Tich's unsuccessful carryings on with the female of the species. (Hardy 1971) (AND)

Other characteristic examples include the following: My mates chyacked me all night. (Australasian Printer's Keepsake, 1885) (AND) Diggers of the Yarra tribe ... like to chiack the Cornstalk variety about 'our' arbour'. (Aussie 1919) (AND) They whooped, they made ribald noises, they chyacked one another. (S. Campion 1944) (AND) They chyacked their sissy mates and their sisters who were forced to attend late afternoon dancing classes. (R. McKie 1977) (AND)

As these examples indicate, 'chiacking' is closely associated with the Australian idea of 'mateship': it is usually done among 'mates', and it is often done reciprocally, and if not reciprocally among mates, then collectively with mates. Usually, the men speak one at a time, making negative remarks about the addressee, while the other men are laughing, so that a group of 'mates' constitutes both a group of participants and an audience, as in the following examples: The circle offrivolous youths who were yelping at and chy-acking him. (Australian Monthly Magazine 1879) (AND) They're always a-poking borack an a-chiackin' hut! (l.A. Barry 1893) (Wilkes 1978)

0'

me over in the

Some Australian speech-act verbs

167

There were several pretty girls in the office, laughing and chiacking the counter clerks. (Henry Lawson 1896) (Wilkes 1978) Don't walk about; it's tirin'; stand at street-corners and spit besides that ther best place ter see life and chyack the girls. (Henry Fletcher 1908) (Wilkes 1978) The milk-carters ... sloshed the milk into the cans, chyacked Dolour about her goggles, and charged out again. (Ruth Park 1948) (Wilkes 1978) The rowdy bodgie youths kept seats near this group, chiacking the buxom, brassy-haired waitress as she rushed around with a tray-load of dishes and lively back-chat. (K.S. Pritchard 1967) (Wilkes 1978)

'Chiacking', then, is very much a shared entertainment, which both expresses and promotes the feeling of 'mateship' among those who jointly engage in it. It is definitely a pleasurable activity, associated with laughter, rowdiness, noise, and good humour. The following examples highlight this aspect of 'chiacking': Pleasant chi-ack in the billets (Action Front, 1940) (AND) They served out hot tea and in a few moments grumbling gave place to 'chiacking',. criticism that a few moments ago had been edged was now good-humoured. (R.H. Knyvett 1918) (AND) Thus ended the relief of Rustenburg, in cheers and laughter and chyacking and sleep. (S. Campion 1944) (AND) They were a vociferous crowd, ruggedly vocal in a loud, chiacking anticipation of the heady joys to come. (E. Lindall 1964) (AND) Other types of humour - chyacking and leg-pulling, sardonic anecdotes, jolliness and exuberance. (Donald Horne 1967) (AND) The groomsmen all red in the face and looking as if they would choke in their stiff white collars, rocked the whole congregation with a desire to chuckle and chiack. (K.S. Pritchard 1948) (AND)

168

Speech acts and speech genres across languages and cultures

Though pleasurable for those who engage in it, the activity of 'chiacking' is by no means always pleasurable for those who are the victims of it. Nonetheless, it is never hostile, and it is expected that it will be borne with good humour: Ironbark' s face was red by this time with all the chyacking he got from the blokes. (D. Stivens 1955) (AND) Next day at lunchtime I got the same chyacking treatment from Gordon's brother Frank. (B . Heslin 1963) (AND) I was always civil to the chaps, for all the chyacking they gave me. (W.H. Suttor 1887) (AND) Tommy Bent ... was a victim of most of the 'chyacking'. (Gadfly 1906) (AND) When their chiacking got too much I would go out and talk to the turkeys. (M. Eldridge 1984) (AND)

What does it mean, then, to chiack somebody? I propose the following analysis of the attitude encoded in this concept: chiack (a) we want to say something bad about you (b) because we want to laugh and to feel something good (c) not because we want you to feel something bad (d) I think you know: one feels something good if one can do this with someone like oneself, like a man with a man

Component (a) indicates that chiacking is a collective activity; component (b) shows that it is done for pleasure and for fun; component (c) shows that it is a good-humoured and good-natured activity, devoid of any hostile intent; and component (d) indicates that chiacking is primarily, though not exclusively, a male activity, either reciprocal (done from man to man) or collectively (done by a group of men), and that it implies 'solidarity' and egalitarianism. The concept of chiacking reflects some of the most characteristic features of Australian culture: sociability, 'mateship', enjoyment of joint activities with one's mates (especially idle activities, such as drinking), male solidarity and male togetherness, associated with displays of 'masculinity' in 'bad language', and so on. The concept of chiacking reflects also the Australian preference for saying 'bad things' rather than 'good things', about people in general and about the addressee in particular -

Some Australian speech-act verbs

169

not because one thinks 'bad things' about them or feels 'bad feelings' towards them, but because of the cultural ideals of roughness, toughness, anti-sentimentality, anti-emotionality, and so on. The link between 'saying something bad' and 'feeling something good' is particularly characteristic. It is the one link which is also manifested in the typically Australian phenomenon of friendly insults (G'day ya old bastard!, cf. Taylor 1976), in the tendency to express enthusiasm by means of swearwords (you bloody beauty!), in the lack of offensive connotations linked with words such as bugger (poor bugger), crap, bullshit, and so on. ... the interesting thing about the Australian attitude to human relationship is the special forms it has to take to avoid coming into conflict with our basic antipathy towards the public expression of sentiment and emotion. Because we are unsentimental and cynical towards the emotions, Australians have to express their social affection in some way which is not on the face of it self-revealing. Thus, there has evolved the principle of 'rubbishing' your mates and chayacking [emphasis added] the stranger. In an atmosphere of reciprocal banter or 'rubbishing' Australians can express mutual affection without running any risk of indecently exposing states of feeling. (Harris 1962:65-66)

As Renwick (1980:22-23) points out, "with regard to personal characteristics, Australian men and women are friendly, humorous, and sardonic (derisive, disdainful, and scornful)". They "express negative feelings and opinions about both situations and people, sometimes about people they are with". They have a tendency "to be personally evaluative and to express negative reactions" (1980:29). In particular, negative remarks play an important role in Australian humour. Renwick observes that "Americans sometimes feel that Australians' humor is ... disrespectful, harsh, and offensive" (1980:29), and he advises Americans as follows: "Stand ready, in a relaxed manner, to be tested. The Australian may challenge you and probe to see if you are a person of substance, someone with a backbone, some steel inside, some depth and character. Practice testing and sparring with the Australian yourself." "Develop personal resilience. Don't be put off by derisive comments, undercutting, and cynicism." (1980:33-35). These comments, and this advice, show deep insight into the traditional Australian ethos - an ethos reflected with particular clarity in the Australian concept of 'chiacking'. The fact that this key word is now disappearing from Australian speech, so that the younger generation of Australians is often unfamiliar with it, reflects some of the changes

170

Speech acts and speech genres across languages and cultures

which this culture is undergoing. As pointed out by many observers of the Australian scene, since World War II there has been a considerable shift in Australia from traditional working-class values to middle-class values. For example, McGregor (1980) reports that according to Gallup polls taken over the past few decades the proportion of the Australian population identifying themselves as working class has declined markedly, while the proportion identifying as middle class has correspondingly increased. The decline of the use of key Australian concepts such as 'chiacking' (and also yarn and shout, to be discussed below) reflects these broader social changes. In particular, while 'chiacking' is still a common Australian activity, the concept of 'chiacking' is already losing some of its salience in the Australian national mentality.

2.2. Yarn Yarn (which can be used in Australia as either a noun or a verb) is another important Australian concept, referring to something like a chat or a talk, but embodying a characteristically Australian way of looking at the activity in question. It is typically used in the phrase have a yarn; for example: They asked the Buxtons to come over to their camp, and have a 'yarn'. (J. Bonwick 1870) (AND) He used to delight in going to travellers' camps to have a 'yarn' with them. (M.A. McManus 1913) (AND)

As these examples indicate, 'having a yam' is often seen as a form of pleasurable sociability. The expectation that 'yarns' generate 'good feelings' is reflected in the common collocation 'a good yarn', which implies a satisfying as well as fairly lengthy (and leisurely) verbal exchange: You are questioned all about home, what brought you out, and all such questions, until what is termed in the colony a good yarn is over, you may then be asked to have a nobler. ('Eye Witness' 1859) (AND) He says he doesn't really want to do any sort of interview, but it doesn't take long to see that deep down, the man likes a good yarn. (Sydney Morning Herald 1986) (AND)

Some Australian speech-act verbs

171

Yarn as a verbal exchange should be distinguished from yarn as a kind of long tale 'spun' out of facts and fantasy for the purpose of companionship: to have a yarn is not the same as to spin a yarn - another favourite Australian speech genre of 'the olden days'. But the slow, relaxed nature of the yarns that people spin (or used to spin) highlights the unhurried, relaxed nature of the yarn that one can have with someone else. But although a 'good yam' (with someone) is normally a long one, a short 'yarn' is also seen as enjoyable, provided that it is leisurely, unhurried, and without a rigidly imposed temporal boundary. This is reflected in the common collocation 'a bit of a yam'. For example: There they all stood and had a bit of a yarn before they came home. (A.A. Smith 1944) (AND)

The pleasurable, sociable, and unhurried character of 'yarning' is highlighted in the following examples: The manager received me with open arms, and we 'yarned' far into the night over the old country. (A.W. Stirling 1884) (AND) I thought it glorious fun smoking our cigars and yarning until overcome by our long drive, we both fell asleep. (S.S. Junr. 1868) (AND)

But 'yarning' is not an idle activity undertaken solely for pleasure and devoid of any serious meaning: By 'yarning', dear reader, I don't mean mere trivial conversation, but hard, solid talk. (M. Clarke 1896) (AND)

'Yams' differ in this respect from 'chats', which are also exercises in pleasurable sociability, but which, by definition, play down the significance of the exchange. Chatting can be idle, but yarning is not seen as idle, whatever the topic, because it suggests a serious need for human contact and for human communication. The following example illustrates well this aspect of 'yam': Some of me old mates from the bush turned up for a beer and a yarn. (A. Buzo 1986) (AND)

Certainly, 'a beer' and 'a yarn' stand here for enjoyable activities, undertaken for pleasure. But the sentence illustrates also well the special importance of such activities in the Australian context, where the distances, the isolation, the loneliness create a special need for human contact,

172

Speech acts and speech genres across languages and cultures

human warmth, and human communication going far beyond the casual, light-weight sociability characteristic of a chat. Like chiacking, yarning or having a yarn are concepts with strong masculine associations. This is another dimension of contrast between a yarn and a chat; in Australia, men would traditionally have a beer and a yam (with their 'mates'), whereas 'ladies' would have a 'cuppa' (a cup of tea) and a chat. These different gender associations may have something to do with different expectations with regard to 'verbal economy': the concept of 'chat' implies 'chattiness', that is, a facility with words, an uninterrupted and easy verbal flow between two people; by contrast, the concept of 'yam' implies a terseness and a background of silence, of isolation, and of a real need for a verbal exchange as a form of scarce human contact, especially with one's friends (one could chat with one's neighbours every day but one could hardly yarn with them every day). This need for 'congenial fellowship' (especially male fellowship) reflected in the concept of yarn is highlighted in this example: It's hard work sinking bores, and after a few months on your own, with no one but a couple abos [Aborigines] to yarn to, you've gotta get stinkin' [drunk] once in a while. (I. Marshall 1962) (AND)

In my English speech act verbs: a semantic dictionary (Wierzbicka 1987) I have posited for chat the following meaning (reproduced here in a somewhat simplified form): chat (a) I (b) I (c) I (d) I

want us to say many different things to one another think you want the same think we will feel something good because of this don't think: these things are important

For yarn, I would propose a similar semantic structure but without the trivialising component (d), and with additional components (a') and (d'), stressing the participants' need for communication as a form of companionship with someone like oneself: yarn (a) I want us to say many different things to one another (a') I want to do it for a long time (b) I think you want the same (c) I think we will feel something good because of this

Some Australian speech-act verbs

173

(d') I think: it is good if a man can do this from time to time with someone like himself Thus the concept of yarn points, indirectly, to the concept of 'mateship', to the importance of shared activities, to the emphasis on human relations rather than on productivity or achievement of external goals, and to the relaxed attitude to time prevailing in Australia. As pointed out by Renwick (1980), the pace of life in Australia is relatively slow (at least in comparison with urban/corporate America): people are less 'task-oriented' and 'future-oriented'; rather, they have a relaxed, 'day-to-day' orientation, they want to enjoy life and enjoy being with others, and are more interested in personal relationships (especially with 'mates ') than in productivity. The characteristically Australian concept of 'yam' reflects and documents these attitudes. It should be noted, however, that like the concept of 'chiacking', this concept too is losing its salience, and that the word yarn is losing ground in Australian English.

2.3. Shout Shouting is a specifically Australian concept, standing for an actIvIty which from early on in the Australian history established itself as one of the most characteristic national customs, remarked on by all observers. For example: Nearly everyone drinks, and the first question on meeting generally is, 'Are you going to shout?', i.e. stand treat. (W. Burrows 1859) (AND) 'A shout' , in the parlance of the Australian bush, is an authority or request to the party in waiting in a public-house to supply the bibulous wants of the companions of the shouter, who of course bears the expense. (C. Munro 1862) (AND) Of all the folly that has ever beset a community, that of shouting has held the ground the longest, and is the most absurd. (Bell's Life in Sydney 1864) (AND) He viewed this 'shouting' mania with disgust. (Bulletin (Sydney) 1892) (AND)

174

Speech acts and speech genres across languages and cultures

As many examples from AND clearly indicate, shouting is definitely linked in Australia with the idea of generosity, and it often is (or was) asymmetrical, as when a man with money 'shouts' drinks for the moneyless 'hands' or even for bystanders: Most peculiar thing to me as the night wore on, and yarn after yarn went around, the old bloke always shouted, and for all hands each time. (Western Champion 1894) (AND) At our approach four miserable derelicts left the stool on the verandah and slouched into the bar on the prospect of a 'shout'. (F.l. Brady 1911) (AND)

In the generally non-competitive and super-egalitarian Australian society, shouting was one domain where one could be freely competitive - competing with other people, as it were, in generosity and in the spirit of companionship: In the Westralian mining towns ... man's class is decided by the number he shouts for ... To shout for the room is common, to shout for the 'house' nothing extraordinary, and if the shouter is 'brassed up' at all, he says: 'Call in them chaps outside'. (Bulletin (Sydney) 1909) (AND) He was also of that species of good Aussie mixers who, if someone 'shouted' a round, would forthwith plonk down a handful of silver to indicate payment for the next round before anyone could raise the first glass. (S. Hope 1956) (AND)

At the same time, however, shouting has strong connotations of reciprocity and turn-taking: It is drink, drink, all day, and swim in it at night. Everyone you meet will 'shout', and you have to 'shout' in return. (Demonax

1873) (AND) You wouldn't expect a man to leave before his shout would you Ben? (M. Paice 1978) (AND)

The expected reciprocity of 'shouting' highlights the link which this concept has with the key Australian value of 'mateship'; and the words shout and mates frequently occur together:

Some Australian speech-act verbs

175

The unbreakable custom that if four or five mates grouped together one started to buy all the drinks, but in the circle everyone had to have his turn. (H.G. Tesher 1977) (AND)

The expectations of reciprocity and turn-taking appear to imply a mutuality and an equality which is hard to reconcile with the frequent asymmetry of shouting illustrated earlier. Trying to solve this apparent paradox I would propose that reciprocity and tum-taking constitute a social convention associated with shouting but are not a necessary part of the concept itself. On the other hand, the idea of drinking companionship (male companionship) is part of the concept: even in those cases when 'shouting' constitutes a one-sided treat and a display of one-sided generosity the notion is still there that it is good and pleasurable for a man to drink with other men and that on such occasions it is good to 'do things' for one's companions and to identify one's own interests with theirs. Thus, although shouting can be done by one individual, it is still analogous to chiacking and to yarning in its celebration of relaxed male companionship, and male solidarity ('mateship'). The following example illustrates clearly this importance of male companionship and solidarity over and above any strict reciprocity: All Merr's mates shouted him at the pub for a week. (A. Garve 1968) (AND)

Thus, the idea of a shout implies not just one invitation to shared drinking but a sequence of such invitations (typically, a sequence of 'rounds ') and it strongly suggests reciprocity and turn-taking without, however, precluding one-sided generosity on the part of one particular person. To account for these facts, I propose to include in the explication of shouting the following component: 'I think someone will say the same after me'. It would be natural to interpret 'someone' as 'someone else', but it can also refer to the speaker himself. shout l (a) I say: I want everyone here to have a drink with me now (b) I will pay for this (c) I think we will all feel something good because of this (d) I think someone will say the same after this (e) I think we all think: it is good if a man does this with other men

176

Speech acts and speech genres across languages and cultures

The non-importance of strict reciprocity in the concept of shouting is highlighted by what AND rightly describes as an extended ('transferred and figurative') sense of this word (roughly, 'treat offered to someone else'). For example: The governor shouted heavy, and gave us all an excellent feed (N. Earle 1861) (AND) I'll shout a trip (first-class) for him from Sydney to Narrandera (Bulletin (Sydney) 1896) (AND) Once or twice a year I 'shout' the boys of an orphanage to the pictures. (R. Comm. Moving Picture Industry 1927) (AND) It's Saturday, and I was wondering if you'd like to have dinner there. It'll be my shout. It goes on the expense account. (D. Middlebrook 1975) (AND)

But even this kind of one-sided shout has implications of shared pleasure, as well as of generosity: a person who shouts a treat for someone else fully expects to share in the target person's enjoyment (if only by enjoying their enjoyment) and thus shows a generous and friendly spirit. (The last example above shows a somewhat jarring mixture of attitudes, and reflects a modem corruption of the pioneer ideal.) What is truly important about the concept of shouting is the idea of being generous with other people in the spirit of solidarity and congenial (male) fellowship. There is no stress on reciprocity in the sense of 'repayment of a debt' (as in the case of Japanese key concepts of on and giri, cf. Benedict 1947; Lebra 1974). One is obliged to drink, to share in the companionship, and to enjoy a relaxed atmosphere of generosity and group-identification, rather than necessarily 'repay' the treat to the very person who has provided it. Reciprocity is at the most hinted at by a general expectation that the recipient would want to do the same (perhaps some other time, with some other people). Accordingly, 1 have posited for shout2 the component 'I think we all think it is good if people do this with other people', which echoes the last component of shout l : 'I think we all think it is good if a man does this with other men'. shout2 (a) 1 say: 1 want to do something good for you (b) 1 will pay for this (c) 1 think you will feel something good because of this (d) 1 think we will all feel something good because of this

Some Australian speech-act verbs

177

(e) I think we all think: it is good if people do this for other people It should be noted that this second, extended sense of shout is in fact growing in use, while the primary sense is declining (together with the social tradition which gave rise to it, and with the social and cultural conditions associated with it).

2.4. Doh If the words chiack, yarn, and shout can be said to affirm and celebrate 'mateship' and 'congenial fellowship' in a positive way, dob in (described by OEDS as "Australian slang") can be said to affirm it and celebrate it as it were in a negative way - by condemning, with contempt, anyone who betrays it. AND defines the meaning of the expression dob in as "to inform upon, to incriminate". But this is not an improvement on the earlier description offered by OEDS: "to betray, inform against". The notion of 'betraying' constitutes a crucial difference between the specifically Australian concept of dobbing and the pan-English concept of informing. On the other hand, the description 'Australian slang', offered by OEDS, is misleading: in Australia, dob in is not slang (restricted to some particular social group), it is simply part of common everyday language, a word which is in general use and which is clearly one of the key words in Australian English. O'Grady (1965) offers the following comment in this connection (using the word dob in): Australians are noted for a deep-seated reluctance to report any fellowcitizen to anyone in a position of authority. Police, bosses, foremen, wives, etc. must do their own detecting. Anybody who 'dobs in' anybody else is a 'bastard' - in the worst sense of the word. (O'Grady 1965:34)

Similarly, Baker (1959:15-16) mentions "a totally unforgiving attitude towards 'rats', 'scabs' and betrayers in general" among the most distinctive features of the 'Australian character'. "The essence of the tradition is loyalty to one's fellows, and the strength of its appeal may be seen in the restraining power of the term 'scab' in an Australian union." (Crawford 1970: 137). According to Ward (1958), quoted in Crawford (1970: 135), "the combination of loyalty to one's fellows with disrespect

178

Speech acts and speech genres across languages and cultures

towards superior orders [and the] enduring disrespect for authority [may be] traced back to the convicts." All this is reflected very clearly in the key word dob in. Some examples: You said you'd go to the police and dob him in unless he coughed up. That's the story isn't it? (J. Waten 1957) (Wilkes 1978) A couple of the Indonesian p.o.w's have dobbed us in. Told the Nips everything. (R. Braddon 1961) (AND) In these two examples, dob in could be in principle replaced with inform on (though not without a significant change in meaning). In the examples which follow, however, inform on could hardly be used at all, since it is not used with respect to strictly personal relations (such as, for example, family relations): Helen stuck on a real act and dobbed me in to Mum, screaming about how I had busted her best doll on purpose. (P. Barton 1981) (AND) Unlike inform on, dob in is derogatory and contemptuous: dobbing is something a decent person cannot possibly do. I shut up and let Ray take all the credit. Couldn't dob him in, could I? (J. O'Grady 1973) (AND) You bitch! Go and dob me in because I gave you a bit of a shove! (Williamson 1972:66) But you feel such a rat to tell on her. To dob her in. (H. F. Brinsmead 1966) (AND) The noun dobber is equally, or even more, contemptuous and derogatory. Don't look at me, you bastards! I'm no bloody dobber! (J. Powers 1973) (AND) The expression 'dobber' was one that I knew implied contempt and was apt to be applied to tale-bearers and informers. (G.A.W. Smith 1977) (AND) One further difference between inform on and dob in is that the latter implies that the agent is definitely hurting the person spoken of whereas the former does not necessarily imply that. In informing, the stress is on

Some Australian speech-act verbs

179

the transmISSIon of (potentially damaging) information, not on interpersonal relations between the speaker and the person spoken of, but in dobbing in, the stress is on interpersonal relations. This semantic difference between the two verbs is reflected in a syntactic one. Dob in treats the victim as a direct object ('to dob someone in') and thus suggests that the agent is 'doing something to' the person dobbed in. By contrast, inform on treats the victim as an oblique object (one cannot 'inform someone on'); this suggests that the agent of informing is not necessarily 'doing something to' the person informed on. It is interesting to note in this connection that dob, too, can be used with the particle on, and that dob on is closer semantically to inform on than dob in is. Inform on, tell on, and dob on all suggest intentional transmission of damaging information without implying that serious harm has already been done, as dob in does. At the same time dob on, which appears to be used mainly by schoolchildren, shares with dob in its contemptuous and derogatory character: evidently, the general Australian contempt for those who break group solidarity and who attempt to side with the authorities against fellow 'subordinates', is an important part of the Australian school ethos, as well as of the Australian ethos in general. I will not try to propose here an explication of dob on, interesting as it is, focussing instead on the more basic concept dob in, used widely right across the whole of Australian society.

dob in I say: person X did something bad I want you to know this I think: you will do something bad to X because of this I know: people like you can do something bad to people like X and me I know: X would think that I wouldn't say this to you I want to say this to you [people would think something bad about this person because of this] [people would feel something bad towards this person because of this] It is worth noting that dob in has also another meaning in Australia: roughly, doing a bad turn to a 'mate' by 'volunteering' for something on his or her behalf. This meaning is related to the first one insofar as it implies saying something about a 'mate' to a person in charge, causing

180

Speech acts and speech genres across languages and cultures

something bad to happen to the 'mate,' and thus violating the expectation of loyalty and mutual support. The main difference between the two meanings consists in the fact that in one case, one says something bad about the mate, whereas in the other, one says something unfounded and embarrassing: namely, that s/he is willing to do something which in fact s/he is not.

2.5. Whinge Whinge [wlnd3] (roughly, 'complain' or 'whine') is clearly one of the key words in Australian English. In other parts of the English-speaking world it is marginal (although not totally unknown); OED qualifies it as 'Scottish and dialectal' (noun) and 'Scottish and northern dialects' (verb), although OEDS hedges this qualification by means of the adverb ,originally' . The marginal character of whinge and its derivates such as whinger outside Australian English is reflected by the following examples: Other local terms for crying ... in Dublin the usual word is 'whinging', hence 'whinger', a term also still used in Cumberland, and occasionally heard in Liverpool. (I. and P. Opie 1959) (OEDS) Touching the query about 'whinger' ..., 'winjer' was accepted slang for 'grumbler' at Q. Uni. [Queensland University] a few years ago, and probably still is. I have seldom heard it elsewhere, and no one who uses it seems to know the derivation. (Bulletin (Sydney) 1934) (OEDS)

The verb whinge, evidently marginal in other varieties of English, in Australia is a household verb. It plays a crucial role in the socialisation of children (Stop whingeing!), and in the formation and transmission of the Australian national ethos. As one observer put it, discussing the relative unimportance of the value of 'success' and the crucial importance of the values of 'tough masculinity', gameness, and resilience in Australian culture: There is little public glorification of success in Australia. The few heroes of heroic occasions (other than those of sport) are remembered for their style rather than for their achievement. The early explorers, Anzac Day: these commemorate comradeship, gameness, exertion of the Will, suffer-

Some Australian speech-act verbs

181

ing in silence. To be game, not to whinge [emphasis added] - that's the thing - rather than some dull success coming from organisation and thought. (Horne 1964:26)

The importance of the concept of 'whingeing' in Australian culture is reflected in a spectacular way in the very common Australian expression 'whingeing Poms' or 'whingeing Pommies' - an expression which shows both the Australian perception of English people and their own Australian self-image: English people are, above all, 'whingers', whereas Australians are, above all, 'non-whingers': The British national pastime of 'grousing' (to use an English phrase) has given rise in Australia to the derisive expression wingeing pommy. (Marshall and Drysdale 1962) (AND) It'll pass a law to give every single wingein bloody Pommie his fare home to England. Back to the smoke and the sun shining ten days a year and shit in the streets. Yer can have it. (T. Keneally 1972) (AND)

Whingeing Poms make me ill. (W.F. Mandle 1974) (AND)

What exactly is whingeing? Clearly, it is a concept closely related to complaining. But, first, complain is neutral, and does not imply any evaluation of the activity in question, whereas whinge is critical and derogatory. Furthermore, complain is purely verbal, whereas whinge suggests something that sounds like an inarticulate animal cry. Being purely verbal, complaining can be seen as fully intentional, whereas whingeing can be seen as only semi-intentional and semi-controlled. Finally, whingeing, like nagging, and unlike complaining, suggests monotonous repetition. In English speech act verbs (Wierzbicka 1987) I posited for complain the following semantic structure (reproduced here in a slightly simplified form): (a) I say: something bad is happening to me (b) I feel something bad because of this (c) I want someone to know about this Whinge appears to attribute to the speaker (the whinger) an analogous, but more elaborated, attitude: whinge (a) I say: something bad is happening to me

182

Speech acts and speech genres across languages and cultures

(b) (b') (c) (d) (e) (f)

I I I I I I

feel something bad because of this can't do anything because of this want someone to know this want someone to do something because of this think no one wants to do anything because of this want to say this many times because of this

Component (b') of this formula suggests a feeling of total baby-like helplessness, (d) indicates passivity and reliance on others, (e) suggests an element of infantile resentment and self-pity, whereas (f) spells out the reliance on the equally infantile 'strategy' of monotonous repetition (as in an infant's crying). Generally, then, the concept of whingeing likens the attitude of those who indulge in it with that of crying babies, and what Australians think of people who behave like crying babies is best expressed in another important Australianism: the noun sook [srok] (adj. sooky). Some examples: (He goes to her and holds her gently ... She sobs a little, but then forces a laugh and leaves him.) Ruby: Well! You'll think I'm a sook. (R.I. Merritt 1975) (AND) Annie felt sick with fear. 'Sookie sook, I'm going to tell on you' , chanted Rosa. (Australian Short Stories 1985) (AND) The girl applied a hefty hip ... and flattened him. Sprawled on the bitumen, he began to howl. 'Bloody sook!' said the girl, disgustedly. (Bulletin 1986) (AND)

As pointed out by Horne (1964:40) among others, Australians are cheerful and practical-minded optimists. They admire toughness, resilience, good humour, and rough 'masculinity' - in hard times as well as in good times. Their folk heroes are Ned Kelly and various other real or legendary 'wild colonial boys', for whom the important thing was not so much to live in comfort and security or to succeed as to: ... die hard, die game, . die fighting, like that wild colonial boy, Jack Dowling, says the ballad, was his name. (a poem by John Manifold, quoted in Ward 1958:217)

Acccording to the same Australian ballad, "'I'll die but not surrender', said the Wild Colonial Boy" (Wannan 1963:17). The present generation of Australians think, it seems, less of 'dying hard' and more of 'having a good time' (on the growing hedonism of

Some examples of complex speech genres

183

Australians see Conway 1971; King 1978); but the contempt for 'sooks' and 'whingers' has remained part and parcel of the present-day Australian ethos. Australian national mottos are still 'no worries' and 'she'll be right'. Australians still admire actions rather than words or ideas. They value practicality and self-reliance. They also assume and approve of mutual reliance of 'mates' on one another (cf. Renwick 1980:16). But any 'babyish' reliance on more powerful 'other people' and 'babyish' indulgence in repeated 'crying' (instead of a search for practical solutions) is totally incompatible with the Australian 'bush ethos' and with whatever remains of it in the modern Australian mentality. The key verb whinge reflects and documents these attitudes. A full study of Australian colloquial speech-act verbs would have to include many more, such as stir, sledge, skite, rouse on, pimp on, earbash, big-note oneself, knock, (w)rap (up), and fang (see Wilkes 1978; AND). I believe, however, that the five that have been discussed here, are particularly representative, and particularly important.

3. Some examples of complex speech genres 3.1. The black English dozens I will start with a genre that is well known, especially through Labov (1972): Black English 'ritual insults'. As Labov and others have pointed out, the genre in question has a number of different folk names, which reflect some regional and, apparently, semantic variation, sounding and playing the dozens being perhaps the most common among them. It is a form of 'street talk', engaged in by black adolescent boys, a kind of verbal contest. As Abrahams (1974:241) points out, speaking of black culture: "Playing ... is an important way in which one distinguishes oneself in public, and engaging in witty verbal exchanges is one important way of playing ... Active verbal performance in the street is one of the main means of asserting one's presence and place." The wit and the verbal virtuosity are exercised mainly through breaking the rules and conventions of the 'respectable' society. The cultural significance of this genre is excellently captured in a passage from Mezzrow and Wolfe, quoted in Abrahams (1974:240):

184

Speech acts and speech genres across languages and cultures

These boys I ran with at The Corner, breathing half-comic prayers at the Tree of Hope, they were the new sophisticates of the race, the jivers, the sweet-talkers, the jawblockers. They spouted at each other like soldiers sharpening their bayonets - what they were sharpening, in all this verbal horseplay, was their wits, the only weapons they had. Their sophistication didn't come out of moldy books and dicty colleges. It came from opening their eyes wide and gunning the world hard ... They were the genius of the people, always on their toes, never missing a trick, asking no favours and taking no guff, not looking for trouble but solid ready for it. Spawned in a social vacuum and hung up in mid-air, they were beginning to build their own culture. Their language was a declaration of independence. (Mezzrow - Wolfe 1969:193-194)

It is not my purpose to try to add anything to the understanding of the ge:lre in question, of which I have no first-hand knowledge. All I would like to do is to propose a semantic formula, constructed in the natural semantic metalanguage, which spells out the illocutionary force of this genre, as I understand it on the basis of studies such as Labov (1972) and Abrahams (1974). It seems to me that a succinct formula of this kind 'sums up' the genre in question in a metalanguage which is essentially culture-independent and thus facilitates cross-cultural comparisons. One characteristic (though uncharacteristically mild) example may be in order: "Your mother so old she got spider webs under her arms" (Labov 1972:312). The semantic formula: (a) I want to say something bad (about your mother) (b) I think: everyone here knows: I don't think this (about your mother) (c) I want to say something that some people would say is bad (d) I say this because I want people here to feel something good (e) and to think something good about me (f) I want people here to think that I can say things that other people can't (g) I think after this you will say something like this (about my mother) (h) I think you will want to say something more bad than this (i) I want you to say it if you can (j) I think we can say things like this to one another because we are becoming men (k) I think we will all feel something good because of this

Some examples of complex speech genres

185

Essentially, the speaker is saying something 'bad' and obscene about the addressee's mother (component a), something that is clearly not true (component b), trying to be daring (c) as well as ingenious (f), inviting the addressee to respond in kind (g) and challenging him to outdo him (h and i). In doing so, he is trying to entertain the nonaddressed participants (d) and to gain their admiration (e), as well as participate in the shared 'fun' himself (k); he is also trying to assert and validate the group's view of themselves as approaching manhood U). It is interesting to note that certain aspects of this genre as explicated here correspond quite closely to certain features of the black English ethnography of speaking, discussed earlier; in particular, I have in mind the positive attitudes to 'boasts' and the desire to entertain the audience (cf. Chapter 3 above). It is also interesting to note some links between the Black English 'dozen' and the Australian English 'chiacking': both are collective male activities undertaken for fun, involving an audience, and consisting in saying offensive things about the addressee, without the intention to offend. But the differences between the two genres are as striking as the similarities. In particular, the 'dozen' stresses verbal virtuosity, whereas the Australian ethos values grunts and discourages eloquence; the 'dozen' is competitive, whereas the 'old chiack' is cooperative (with 'mates' acting as a group rather than as separate and competing individuals); the 'dozen' is obscene and reaffirms the group's rejection of the values and taboos of the society at large, whereas the 'chiack' is not obscene, and reaffirms the prevailing values of the traditional Australian society (while emphasising the distinct character of this society, defined partly by its rejection of, and opposition to, 'Pommy ways').

3.2. The Hebrew 'dugri talk' According to Katriel (1986), a 'dugri talk' is a kind of speech event which plays a crucial role in social interaction in the contemporary Israeli society. It is a kind of frequently enacted social ritual which takes to its logical conclusion the more general 'dugri' mode of speaking, highly valued in the Israeli 'Sabra culture'. Dugri, a loan from Arabic, means, literally, 'straight', and sabra is the name of a native Israeli fruit, a kind of prickly pear. As Oring (1981:24, quoted in Katriel 1986:19) points out, "the sabra fruit is a metaphor for the native personality. Like

186

Speech acts and speech genres across languages and cultures

the prickly pear, the native born is sweet and gentle within, but only to those who understand how to penetrate the tough and thorny exterior." This 'tough and thorny exterior' of typical Sabras manifests itself, above all, in their tendency to 'talk straight (i.e. dugri)'. The sentence "He is dugri" means in Hebrew "that the speaker tends to be direct and straightforward in expressing his non-complimentary thoughts or opinions" (Katriel 1986: 15). There are good historical reasons for the great value attached in Israeli culture to the 'dugri' mode of speech, explored in an illuminating way in Katriel's book (and also in Oring 1981). Katriel (1986:21-22,31) mentions in particular: "the assertiveness cluster, which has been associated with the revolutionary orientation of the Zionist movement encapsulated in the phrase 'the negation of the Diaspora"'; "the Sabra culture's emphasis on simple, manual agricultural labor as a means of getting away from the Diaspora image of the Jew as a luftmensch"; the "rejection of decadent European ways of speaking that involve twisting the forms of speech for the purpose of showing respect"; the substitution of solidarity, cameraderie and a spirit of 'communitas' for respect, hierarchy, and distance-creating courtesy; and so on. According to Katriel, all of the cultural values associated with the Sabra culture (a cult of solidarity, simplicity, sincerity, frankness, 'directness', 'straightforwardness', 'truthfulness', 'assertiveness', and so on) find their best expression in the 'dugri' mode of speech, and are epitomised, in particular, by the 'dugri ritual', that is, by the 'dugri talk'. In native terms, this event is referred to as siha dugrit, a dugri talk. A dugri talk is not just any encounter in which the dugri idiom is employed or in which utterances indexed as dugri are exchanged. A dugri talk is a distinct speech event with a sequential and motivational structure of its own. (Katriel 1986:57).

Katriel offers, among others, this example: "An engineer in his early thirties told me at some length about a dugri talk he initiated with his boss. He started what he described as siha dugrit by declaring: 'I want to speak to you dugri. I don't like the way this department is being run. ,,, Describing a typical example of this genre, Katriel writes: It was a ritual act of confrontation, a ceremony of discord, performed in the culture's legitimising idiom: the idiom in which one's integrity and one's shared cultural world are reaffirmed. The use of dugri speech here, as in all other cases of its ritual enactment, served to counteract what in the Sabra culture is considered the tendency to gloss over interpersonal differences

Some examples of complex speech genres

187

in the service of a false, superficial consensus, a concern with harmony in interpersonal relations at the expense of dealing with basic issues and matters of principle. Despite the discomfort caused by the confrontational tone, the dugri ritual was experienced as a moment of true contact, of unmasking, and was received as both legitimate and appropriate even by participants whose own style was a far cry from dugri speech. (Katriel 1986:58-59)

This general characterisation is very helpful, but it cannot replace a rigorous definition, formulated in a language which would facilitate cross-linguistic and cross-cultural comparisons. I would propose the following: a dugri talk I think something bad about you I want to say this to you I know: you don't think the same I know: if I say this to you you can feel something bad because of this I know: someone can think: I will not say it to you because of this I don't want not to say it because of this I think: it would be bad if I didn't say it to you you are someone like me you and I can say things like this to one another because you and I want the same kind of thing people think: it is good if people can say to one another what they think

It is particularly interesting to compare the ethos reflected in the 'dugri talk' (as represented in this formula) with the Australian ethos, reflected in chiacking, rubbishing, or in friendly insults (G'day ya old bastard). Both Israeli culture and Australian culture can be described, and have been described, as cultures which value 'directness', 'straightforwardness', 'terseness', 'simplicity of speech', 'plain talk', and so on; and as cultures whIch value 'solidarity', 'equality', 'cameraderie', 'cooperation', and so on; as cultures which dislike 'artificial politeness', 'social graces', 'glib (or smooth) talk', 'polite polysyllables', and so on, as cultures which encourage people to express negative reactions, and to say 'unpleasant things' rather than 'pleasant things' to one's addressee.

188

Speech acts and speech genres across languages and cultures

All this is true (in some sense) and yet global descriptions of this kind conceal profound differences which are as real as the similarities. In particular, Australian ethos does not encourage people to express freely their 'bad thoughts' about the addressee. For example, it is one thing to call someone, to their face, 'an old bastard' (making clear, at the same time that one doesn't think this), and another, to tell them that one really thinks bad things about them. Israeli culture encourages the speaker to reveal to other people 'bad thoughts' that he or she has about them; Australian ethos does not encourage that. In Australia, people often say 'bad things' when they think or feel 'good things'. But this is not the kind of 'directness' celebrated in Israel as 'dugri speech'. Australians, too, could be described, with some justifications, as 'prickly pears'. But metaphors, like global labels, are often misleading. Explications couched in semantic primitives force us to be precise, and enable us to identify the real similarities as well as the real differences.

3.3. The Polish kawal There are few speech genres as important in Polish culture as kawaf (plural kawaly). Roughly speaking, kawat is a kind of joke. But Polish lexically recognises other kinds of jokes: dowcip and zart. tart is not necessarily verbal or not only verbal; it may correspond to what is called in English a practical joke. Dowcip is necessarily verbal, and so is kawal. But kawal differs significantly from dowcip in its repeatability and in its ingroupness. Dowcip as a mass noun means 'wit'; a dowcip (as a speech genre) can be used as the nearest translation equivalent of 'witticism': it evokes the idea of verbal creativity. Of course, people can repeat old dowcipy (PL) to one another, but the word itself evokes an original display of wit by some creative individual. A kawai, on the other hand, contains no reference to individual wit. It is conceived of as an anonymous creation of oral culture, as a cultural coin which is meant for general circulation. Every kawal expresses a bit of collective wisdom, collective experience, collective outlook. It promulgates ingroupness, solidarity, social integration vis-a.-vis some outsiders. Thus, the prototypical kawaly (PL) are political. They express national solidarity vis-a.-vis foreign powers: the Nazi occupation during World War II, the Soviet-imposed communist regime in post-war Poland, the foreign partitioning powers in the nineteenth century.

Some examples of complex speech genres

189

In addition to political kawaly, which have played a colossal role in Polish culture, there are also kawaly which correspond to English dirty jokes. These usually express male solidarity and male ingroupness. Thus, kawal is a folk genre stressing ingroupness and wide circulation. One enjoys hearing a kawal not only because of its humourous value but also because of the feeling of belonging it gives one. In contradistinction to dowcipy, kawaly are not valued for their ingenuity or sophistication. Etymologically, kawal is an augmentative (from kawalek 'a piece '), and this augmentative character is still felt, carrying the implication that a kawal has nothing over-sophisticated about it, that it is something 'rough' and 'thick', something that can be widely shared. (A kawal chleba 'a piece-AUG of bread', is a very thick and inelegant piece of bread, but the connotations are positive, not negative; the expression implies a hungry man's point of view.) Thus, the ingroupness of a kawal has nothing elitist about it. The group to which a kawal refers is seen as a very broad and strong one. Nonetheless, a kawal also has a certain conspiratorial quality: the group within which a given kawal can circulate wants to exclude outsiders. I hypothesise that the semantic structure of the word kawal contains the following component: 'I think I can say this to you because we think the same about things of this kind'. The implication is: I can tell you, but there are people to whom I couldn't tell it. This conspiratorial air of kawaly is often felt to be the saving grace of a joke which would be rather poor if it were assessed in terms of its sophistication or elegance. As a genre, kawal is firmly rooted in Polish history, in the specific conditions of Polish life in the nineteenth and twentieth centuries. It is an expression of a counter-culture, subsisting in defiance to the official culture imposed from outside. It reflects the persistent need of the nation for the therapeutic effect of a shared laugh. The presumed wide circulation of kawaiy written into the very structure of the concept kawal fulfils an important social function as an expression of widely felt solidarity vis-a.-vis some powerful outsiders, as a way of keeping up the spirit of defiance, as a means of psychological self-defence of the nation. This psychological role of kawaiy is clearly reflected in the way they have tended to correlate with political change in the country - not only in their content but also in their relative frequency at any given time. For example, as pointed out by Garton Ash (1983) and other observers of the Polish scene, the period of national euphoria due to the emergence of Solidarity in 1980-81 witnessed a marked decrease in the

190

Speech acts and speech genres across languages and cultures

production and circulation of kawaly, whereas the bleak time after the imposition of martial law in December 1981 saw their renewed growth. Since August [1980] the consumption of political jokes, like the consumption of alcohol, had dwindled - people had other, better outlets for their political energy and ingenuity ... (Garton Ash 1983: 106) The incredible claims of official propaganda were swept away in a torrent of popular jokes - the acid political humour which returned with martial law, for this kind of humour is an expression of impotence as much as defiance. Every new demonstration, strike or protest was ascribed by the TV news to tiny groups of Solidarity 'extremists'. The Poles translated this with the 'TV Dictionary': 2 Poles: an illegal gathering 3 Poles: an illegal demonstration 10 million Poles: a handful of extremists. (Garton Ash 1983:271-272)

Normally, a kawal requires some kind of introduction: addressees must realise in advance that what they are going to hear is a kawal. Typically, this introduction comes in the form of a question: 'Do you know this kawal?', (the assumption being that the kawaly circulate widely and are likely to be known to the addressee). By contrast, jokes can come unannounced (for example, in the course of a lecture, a talk, or a serious conversation). Kawaly (like parties, parlor games, or social drinking) are meant to promote pleasant togetherness, that is, they are meant to make the addressee and the speaker feel good together. I think jokes have that function too. Witticisms, on the other hand, don't: they seek the addressee's admiration for the speaker rather than shared pleasure. But even though jokes, like kawaly, are meant to promote a pleasant togetherness, they do not have to express and promote group solidarity. For example, imaginary book titles, such as 'All about dogs, by K. Nine', 'Say your prayers, by Neil Down' or 'The world of vegetables, by R.T. Choke', can be listed in a collection of jokes, but their Polish equivalents could not be listed in a collection of kawaly. The reason is, I think, that while such imaginary titles are funny, they do not appeal to any particular attitudes shared by some particular social group. But kawaly always appeal to such shared attitudes. Finally, a kawal has a point, which is not expressed overtly and which has to be grasped ('got') by the addressee. Typically, jokes, too, have a point to be discovered by the addressee ('Did you get it? I don't get it. ').

Some examples of complex speech genres

191

But in the case of a joke, this doesn't seem to be absolutely necessary. If an adult says to a plump little girl, "I like you a lot, I'll eat you with plum sauce", it is a joke, but one without a point to be discovered, beyond the mere fact that it wasn't really meant and that it was said to cause the little girl to laugh and to feel good. On the other hand, a kawai, with its fictitious little script, always poses a mental task for the addressee. One can joke with a baby, but one can hardly tell a baby a kawal. Of course, there is a difference between joking and telling jokes. One wouldn't tell a baby a joke either. But it seems that when the noun joke is used, the differences between joking and telling jokes are disregarded, and the two concepts are subsumed under one. This broad concept, identified by the noun joke, has no Polish equivalent, presumably due to the importance of the specialised concept kawal and the concomitant restriction of the remaining members of the same semantic field. We can now attempt semantic formulae for kawal and joke. kawal (a) I want to say something to you of the kind that many people say to one another (b) I say: one can know this (X) (c) I think you know that this is not true (Le. that one can't know this) (d) I say this because I want you to laugh (e) I think you know that when I say it I want you to think something that I don't say (f) I think when you think of it you will laugh (g) I think we will both feel something good because of this (h) I think I can say it to you because you and I think the same about things of this kind and feel the same when we think about them

Component (a) of this explication shows that kawa/y have to circulate widely; components (b), (c), (d) and (g) are shared by jokes, and jointly spell out the fictitious character of what is being said, the intended humour, and the intended shared pleasure; component (e) indicates that there is a 'point' that the addressee has to 'get'; and (f) shows that it is this 'point', which has to be reconstructed by the addressee, that is expected to be a source of laughter; component (h) spells out the in-groupness of kawaly and the assumption of shared attitudes.

192

Speech acts and speech genres across languages and cultures

joke (a) (-)

(b) I say: one can know this (X) (c) I think you know that this is not true (i.e. one can't know this) (d) I say this because I want you to laugh (g) I think we will both feel something good because of this

3.4. The Polish podanie Writing in 1983, the British historian Timothy Garton Ash characterised the socio-political situation in post-war Poland in the following words: This regime can accurately be described as 'totalitarian' in the sense that it aspires to total control over every aspect of its citizens' lives, to break every social bond outside its aegis, to destroy what the Enlightenment philosophes called 'civil society'. (Garton Ash 1983:8)

Other western observers of life in communist Poland have made similar remarks. The key words and phrases constantly recurring in this literature include the following: 'total control' (Garton Ash 1983:8), 'bureaucratic controls' (Davies 1981, 2:597), 'the bureaucratic Leviathan' (Hirszowicz 1980), 'petty bureaucrats' (Davies 1981, 2:617), 'party-state bureaucracy' (Kolankiewicz - Lewis 1988:24), 'official lawlessness' and 'petty despots' (Davies 1984:41), and so on. Writing more generally about the Soviet bloc, Davies commented: In the Soviet Bloc, there is no higher authority than the Kremlin. There is no rule of Law above the dictator of the day.... What is more, in relation to the mortals beneath him, his particular mode of the 'dictatorship of the proletariat' can be imitated by all the descending hierarchy of petty despots right down the endless links of the political chain. From the supreme web of the Soviet nomenklatura with the Grand Spider at its centre, web upon tangled web radiates out from the Kremlin into the farthest reaches of the Soviet empire ... Every web has its 'spider'; and the spiders, even the benevolent ones, recognise no Law higher than their own. Such a state of official lawlessness is so alien to Americans and to West Europeans (not to mention the Japanese) that most serious attempts to describe it are instinctively dismissed as flights of fancy. (Davies 1984:40-41).

Some examples of complex speech genres

193

Speaking more specifically about communist Poland, Davies pointed to the profound division between the 'power' (w/adza) and 'society' (spo/eczenstwo), between the bosses and the people; and he commented: ... it is undeniable that what Milovan Djilas named the 'New Class', and what others have called 'The Bureaucratic Leviathan', constitutes the most characteristic feature of the supposedly classless societies of Eastern Europe. ... In Poland, where the regime has virtually no legitimacy, it is the source of common oppression. (Davies 1984:45).

The socio-political realities of life in communist Poland found innumerable reflections in the Polish language (cf. for example Wierzbicka 1990). In the present chapter, I will point to just one such reflection, particularly relevant to the area of speech acts and speech genres: to the central importance of the concept of podanie in everyday life in communist Poland. The monumental Dictionary of the Polish Language (SIP 1958-68) defines the word podanie as follows: "pismo skierowane do wladz z prosb~ 0 cos; petycja", 'a written document addressed to the authorities, asking for something; a petition'; and Skorupka' s (1974) phraseological dictionary of Polish offers a shorter version of the same: "pismo z prosbfl 0 co", 'a written document asking for something'. Podanie, then, is a special genre, involving communication between 'people' and 'authorities'; and this communication consists, invariably, in 'people' asking' authorities' for 'favours' and presenting themselves as dependent on the authorities' good will. Clearly, the word podanie has no equivalent in English - not because in English ordinary people never ask any authorities for 'favours', but because the idea that individuals have to ask authorities for 'favours' is not sufficiently salient in English-speaking countries to have led to the emergence of a special concept, and of a special speech genre. But in Polish, the idea is (or was) very salient, no doubt because ordinary people's lives in communist Poland were dominated, to a considerable degree, by their dependence on the arbitrary decisions of bureaucratic 'despots'. An interesting item of linguistic evidence for this is furnished by the set phrase papier podaniowy, 'paper for writing podania (PL)' the usual way of referring to A4 writing paper (a phrase given a special entry in both SIP and in Skorupka 1974). Skorupka's (1974) phraseological dictionary offers the following common collocations involving podanie: "podanie do ministra, do dyrektora; do dziaru kadr, do s~du", 'a podanie to the minister, to the

194

Speech acts and speech genres across languages and cultures

director; to the Personnel Department, to the court'; "podanie 0 co: podanie 0 stypendium, 0 przyj~cie na wyzsz~ uczelni~", 'a podanie (asking) for a scholarship, for admission to a university'; "odrzucic, przyj~c podanie", 'to reject, to accept a podanie'; "zalatwic podanie (odmownie, pozytywnie)", 'to settle a podanie (by refusal, or positively)' . Thus, a podanie is a written communication from an individual to an institution, asking for something that the institution mayor may not grant, where the response is seen by the petent ('supplicant') as arbitrary and unpredictable, and yet as indispensable to normal conduct of life. Hardly any aspect of people's lives in communist Poland, no matter how trivial, could be conducted without the need to write podania - and to wait for the response, hoping that it might be benevolent. For example, a university student asking for an extension of the deadline for submitting a thesis, or an employee asking for permission to take annual leave at a particular time, submitted a podanie. In similar circumstances in an Anglo-Saxon society, it is often sufficient to write a letter, that is, something that is seen as belonging to the same genre as private letters. The closest Polish equivalent of letter is list, but in communist Poland a student or an employee would never seek official favors or concessions by means of a list. The decision of the institution would not be communicated to the requester in the form of a list either; what that person c