Cross-Cultural Pragmatics
W DE
G
Cross-Cultural Pragmatics The Semantics of Human Interaction Second edition by
An...
236 downloads
1747 Views
32MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
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
Some examples of complex speech genres
195
an application, it is sufficient to fill out the form). The assumption is that the addressees will be guided by explicit criteria and that they have no 'arbitrary power over individuals. Typically, a podanie starts with a phrase such as Uprzejmie prosze ... or Niniejszym zwracam si~ z uprzejma prosbfl ... 'I ask politely', 'hereby I address you with a polite request (for a favor)'. But it would be odd to couch an application in some such terms, because a person who is writing an application is applying, not asking for or requesting. (For the difference between ask for and request see Wierzbicka 1987. The Polish words prosic (verb) and prosba (noun) are in fact closer to ask for than to request.) It is also worth noting that while an application may be 'unsuccessful' it can hardly be 'refused' or 'rejected'. Furthermore, the author of an unsuccessful application may be even thanked by the addressee for applying ('Thank you for your application ... '). But it is unthinkable that the bureaucrat to whom a podanie is addresseed could 'thank' the writer. Similarly, applications can be, and often are, 'invited' (for example 'applications from suitably qualified persons are invited ... '); but it is unthinkable that podania could be 'invited'. The reason is that people handling applications can be seen as wanting people to apply, but in the case of a podanie, the writer is (or was) seen as a 'petitioner', a 'supplicant', not as someone who can be seen as cooperating with the addressee. This discussion leads us to the following explications: podanie I say: I want something (X) to happen to me I know: it cannot happen if you don't say you want it to happen I say this because I want you to say you want it to happen I don't know if you will do it I know many people say things like this to you I know you don't have to do things that people want you to do application I say: I want something (X) to happen to me I know: it cannot happen if you don't say you want it to happen I think many people may want the same I think you want some people to say to you that they want it I say this because I want you to say you want it to happen to me I know you cannot do it if I don't say some things about me
196
Speech acts and speech genres across languages and cultures
I say these things here because of this I think you will say if you want it to happen to me Of course, every system can be abused, but the language-specific taxonomy of speech genres reflects what is seen as the typical situation. Contrasting concepts such as application and podanie reflect what native speakers perceive as forms of life typical of their society.
4. Conclusion Nearly three centuries ago, John Locke wrote: A moderate skill in different languages will easily satisfy one of the truth of this, it being so obvious to observe great store of words in one language which have not any that answer them in another. Which plainly shows that those of one country, by their customs and manner of life, have found occasion to make several complex ideas, and given names to them, which others never collected into specific ideas ... Nay, if we look a little more nearly into this matter, and exactly compare different languages, we shall find that, though they have words which in translations and dictionaries are supposed to answer one another, yet there is scarce one of ten amongst the names of complex ideas ... that stands for the same precise idea which the word does that in dictionaries it is rendered by. (Locke 1959 [1690], 2:48)
I believe that few areas of language illustrate the truth of Locke's statement better and more revealingly than the area of speech acts and speech genres. Language-specific terms for speech acts and speech genres present linguistically codified modes of social interaction. Precise semantic analysis of such terms should, it seems to me, provide a new source of insight and rigour in cross-cultural studies.
Chapter 6
The semantics of illocutionary forces
1. Are illocotionary forces indeterminate? Common sense suggests that human life consists, to a very large extent, of a variety of speech acts. From morning to night, people ask, answer, quarrel, argue, promise, boast, scold, complain, nag, praise, thank, confide, reproach, hint - and so on and so on. Moreover, from morning to night, they seek to interpret (consciously or subconsciously) what other people are doing when they speak, that is, what kinds of speech acts they are performing. Virtually every time someone opens his or her mouth in our presence we need to interpret their utterance as this or that kind of speech act. Was it a threat? Or just a warning? Was this a suggestion or a request? Was this a criticism or just a casual remark? Was this a hint? The idea that in speaking people perform different kinds of acts, and that the semantic and/or syntactic structure of an utterance may depend on the kind of act being performed, is very old. It goes back at least as far as the Stoics. 4 In modern times, too, similar ideas have been put forward by a number of different thinkers - notably, by Josef Schachter (1935) in his Prolegomena to a critical grammar, by Ludwig Wittgenstein (1953) in his Philosophical investigations, and by Mixail Baxtin (1952) in his Speech genres. But it was of course J.L. Austin (1962) who expressed and developed this idea in a way which attracted the attention of present-day linguists, and through whose work it was introduced into modern linguistic theory. As a result of Austin's influence, a new model of speech emerged in linguistics, according to which whenever people speak they produce sequences of identifiable speech acts: they greet, they ask, they invite, they suggest, they warn, they apologise, and so on. In fact, it was suggested (by 'generative semanticists', for example Ross 1970) that every sentence contains in its semantic structure (which is also its underlying syntactic structure), a clause that identifies the nature of the speech act performed by means of the sentence. For example, a sentence such as:
198
The semantics of illocutionary forces
What time is it? was said to be derived from I ask you what time it is, and a sentence such as Come here! was said to be derived from I order you to come here. Actually, this 'new' model of analysis, too, was a reinvention of a mode of analysis developed many centuries earlier - notably, by Peter Abelard in the twelfth century, by Roger Bacon in the thirteenth century, or by Paul of Venice in the fourteenth century. 5 The close parallels between the speech act analysis advanced by linguists in the second half of the twentieth century, and that developed by their mediaeval predecessors, are quite fascinating, and it is strange that they are never mentioned in contemporary linguistic writings on speech acts. This is not the place, however, to comment any further on these parallels, or on the historical lessons which could be drawn from them. But contemporary syntactic theories tend to have a rather limited life span. After a few years of vigorous activity, 'generative semantics' lost confidence in itself, and disintegrated, and the newly reinvented idea of deriving illocutionary forces from underlying illocutionary verbs lost most of its adherents. Nonetheless the view that the illocutionary force of an utterance is a part of its semantic structure has survived - at least to the extent of being still regarded as a worthwhile target for attacks. Such attacks are increasingly frequent and increasingly vigorous (cf. for example Bach Harnish 1982; Leech 1983). The attackers point out that most utterances can't be identified, unambiguously, as instances of a particular speech act. For example, Come here! can be an order but it can also be a request. This gun is loaded can be a warning, but it can also be a statement of fact - or a threat. How are you? can be a question but it can also be a greeting. And so on. The so-called performative analysis forced the analyst to reconstruct in each case one explicit performative verb, but in fact - it is now claimed - most utterances are indeterminate. In other words, when we hear people speak, we may understand perfectly well what they are saying but we usually don't know what exactly they are doing: warning? threatening? boasting? suggesting? promising? We can of course make guesses as to the speaker's illocutionary intentions, but we can never be sure if these guesses are correct. Sometimes the situation and the context can make it fairly clear what the speaker is doing (for example, that he is threatening rather than warning), but even then (it is claimed) our interpretation is guided by 'pragmatic considerations' rather than by linguistic clues.
Are illocutionary forces indeterminate?
199
And what applies to ordinary speakers and listeners applies also to linguists. As linguists - so the argument goes - we cannot identify the illocutionary force of any utterances that we may wish to analyse. Consequently, illocutionary forces are outside the province of linguistics; they are a concern of pragmatics, but not of syntax or semantics. For example, a linguist qua linguist cannot distinguish between expressions such as Go to London! and expressions such as Go to hell! (both 'commands'), or between How are you? and How old are you? (both 'questions'). 1 disagree with this view, and 1 believe that it misrepresents the nature of human communication. I believe that when we listen to other people we more often than not do know what they are doing, and we know it, to a large extent, due to unmistakable linguistic clues. Intonation no doubt plays an important role in this respect (see below, section 7.6), but even leaving intonation aside, we still must recognise the presence of innumerable linguistic indicators of illocutionary force. 'Pragmatic' guesses playa part, too, but if we had to rely on guesses, human communication would be much less successful, much less effective than it in fact is. What is unsuccessful and ineffective are the models in terms of which linguists have tried to analyse illocutionary forces. But the solution is to abandon those inadequate models, and to replace them with better ones, rather than to wash one's hands of any illocutionary analysis, and to throw it out of linguistics altogether. This is, then, the main thesis of this chapter: the supposed indeterminacy of illocutionary forces is largely an artefact of inadequate syntactic and semantic analyses.
1.1. Illocutionary forces as bundles of components Consider again a sentence such as: Come here! If we analyse it as 'I order you to come here' then we are indeed over-specifying its illocutionary force. The speaker may have been 'asking' rather than 'ordering', or may have been simply 'telling' the addressee what to do. By spelling out the illocutionary force as 'I order you' rather than as 'I ask you' or 'I tell you', the linguist would be acting in an arbitrary and, one might say, irresponsible fashion. Thus, when Leech (1983: 175) claims that illocutionary force "is more subtle than can be easily accommodated by our everyday vocabulary of speech act verbs" and that it cannot "be
200
The semantics of illocutionary forces
adequately captured by reference to such categories as offers, suggestions, and statements" (1983:156), one can only agree with him. In my view, however, it doesn't follow from this that illocutionary force is necessarily 'indeterminate' or that "it must be studied in part in non-categorical, scalar terms" (Leech 1983: 175). We can't tell if an imperative utterance such as Come here! stands for an order, or a request, or a command, but we can tell that it conveys the idea which can be spelt out as 'I say: I want you to come here'. The point is that orders, commands, and requests have something in common (and of course this is why they can all be enacted by means of the same grammatical category: the imperative). We don't have to say that Come here! is an order (which would be arbitrary), or that it is ambiguous between an order, a command and a request. Instead, we can extract the semantic common denominator of these different interpretations. This can be done precisely by means of the formula 'I say: I want you to do it'. Contextual or suprasegmental clues may provide additional information, but the core is signalled by the construction itself. The illocutionary force of the utterance in question is quite 'determinate', but it can only be captured in a framework which operates with sufficiently fine-grained components. What applies to the alleged 'indeterminacy' of illocutionary forces applies also to their alleged 'scalar variability': both are artefacts of an inadequate analytical framework. Consider, for example, the following statement (Leech 1983: 175): "The difference between 'ordering' and 'requesting' is partly a matter of the scale of optionality (how much choice is given to h), and the difference between 'requesting' and 'offering' is a matter of the cost-benefit scale (how far is A to the cost/ benefit of s/h)." (h stands for hearer, s for speaker, and A for act.) But clearly, the difference between 'order' and 'request' or between 'request' and 'offer' can be represented by means of discrete illocutionary components. There is no need to invoke any 'scalar variability'. An order includes a component which can be stated, roughly, as follows: 'I think you have to do what I say I want you to do'; a request includes the component 'I think: you don't have to do what I say I want you to do'; an offer includes components such as 'I think you might want this', 'I think this would be good for you'. (For a detailed discussion of these and over two hundred other speech acts, see Wierzbicka 1987.) Discussing 'whimperatives' such as Could you be quiet?, Bach and Harnish (1982) consider two possible analyses: (1) The sentence is ambiguous between a question and a request (Sadock 1970); (2) The
Are illocutionary forces indeterminate?
201
sentence is both a question and a request (Searle 1975). They conclude that "the data suggest that whimperatives pattern in part like questions, in requiring verbal responses, and in part like imperatives, in requiring action for compliance. The conventionality thesis [i.e. Searle's analysis], which has them being both, is better able to accommodate such facts than the ambiguity thesis [i.e. Sadock's thesis] is." (Bach - Harnish 1982:186). In other words, the choice is this: should mules be regarded as being sometimes horses and sometimes donkeys or as being always both horses and donkeys? Not surprisingly, both analyses run into insuperable difficulties. As a result, Bach and Hamish feel forced to abandon them both and to propose a third one. What is it? That mules are always ... horses! (i.e. that whimperatives are always questions!) But since the obstinate mules don't want to behave like horses and, for example, since they take a pre-verbal (clause-internal) please, which questions generally don't do: Could you please be quiet? *How old please are you?
they therefore get condemned as 'bad horses', malformed horses, misbehaving horses; and whimperatives, such as Could you please be quiet! get condemned as ungrammatical! What I propose is this: mules are not sometimes horses and sometimes donkeys; and they are not both horses and donkeys; they are neither horses nor donkeys; they are mules. Being mules, they are similar to both horses and donkeys. In describing mules we must show in what respects they are similar to horses and in what respects they are similar to donkeys. In what follows, I will discuss a number of syntactic constructions which encapsulate specific illocutionary forces, trying to demonstrate that although these forces can't be stated by means of single speech act verbs, they can be stated with full precision by means of bundles of components (cf. Wierzbicka 1972, 1977, 1980; cf. also Chapter 5 above). I would add that by proposing a mode of analysis which requires decomposition of illocutionary forces into illocutionary components I am not proposing that the apparatus of linguistic description should be further complicated or added to. On the contrary. Decomposition of speech acts into illocutionary components is a necessary part of linguistics anyway - if lexicography has anything to do with linguistics at all. To state the meaning of verbs such as order, request, warn, threaten or
202
The semantics of illocutionary forces
offer we have to isolate their semantic components anyway. I suggest the following mode of analysis (for justification and discussion see Wierzbicka 1972, 1977, 1980, 1987; see also Chapter 5, section 1): I I I I
order you to do this (X) say: I want you to do this (X) say this because I want you to do it think: you have to do it because of this you will do it because of this
I ask you to do this (X)
I say: I want you to do this (X) it will be good for me I say this because I want you to do it I think: you don't have to do it I don't know if you will do it I suggest that you do this (X)
I say: I think it may be good if you do this (X) I say this because I want you to think about it I think: I don't know if you will want to do it To decompose the verb order into its semantic components is the same thing as to analyse the illocutionary force of the speech act 'order' (pace Searle 1979). There would seem to be little point, therefore, in assuring ourselves that we as linguists don't have to engage in an analysis of illocutionary forces because somebody else (a conversational analyst) will do it for us. We have to do it, as long as lexical semantics is a part of linguistics. And if decomposition of illocutionary forces into components solves at the same time the insuperable syntactic and semantic difficulties that the 'performative hypothesis' ran into, this is a sheer bonus.
1.2. Illustration: the discrete and determinate character of 'whimperatives' Consider, for example, the following characteristic remark (Comrie 1984b:281): " ... 'Would you open the door' is normally going to be used
Are illocutionary forces indeterminate?
203
as a request, perhaps even as a command. The usual response would be to carry out that request. But notice that a possible response is 'No'." Comrie's hesitation between 'request' and 'command' makes it sound as if the pattern in question were somehow 'indeterminate'. In fact, however, the apparent indeterminacy stems entirely from the inappropriateness of the descriptive categories used, and - as Comrie's subsequent discussion suggests - the illocutionary force of the Would you pattern is quite specific and includes a mixture of 'directive' and 'interrogative' components. This can be illustrated with the following exchange: Standing on the porch, before she rang the bell, Miss Ellis took out a comb. 'Would you try to pull this through your hair?' Gilly shook her head. 'Can't.' 'Oh, come on, Gilly - ' 'No. Can't comb my hair. I'm going for the Guinness Recordfor uncombed hair.' 'Gilly, for pete's sake ... ' (GH)6 Evidently, the speaker does want something, and her subsequent protestations (Oh, come on, Gilly - , for pete's sake ...) make this abundantly clear. At the same time, the directive is phrased tentatively, it shows respect for the addressee's personal autonomy, it solicits a verbal response (as well as an action), and implicitly acknowledges the possibility of obstacles, counter-indications, and the like ('would you do it if - if I said I wanted it?'). To show the full force of an illocutionary pattern such as Would you, we need to reveal all the illocutionary components encapsulated in it. This calls for a painstaking qualitative analysis, taking into account both the form and the communicative range of the pattern in question. It is an illusion to think that instead of engaging in a qualitative analysis of this kind we can simply draw some 'continuum' of illocutionary forces, such as that advocated by Giv6n (1984:249 and 1989:153) as the "continuum ... between prototypical imperative and interrogative speech acts": Most prototypical imperative a. Pass the salt! b. Please pass the salt. c. Pass the salt, would you please? d. Would you please pass the salt?
204
The semantics of illocutionary forces
e. Could you please pass the salt? f. Can you pass the salt? g. Do you see the salt? h. Is there any salt around? i. Was there any salt there? Most prototypical interrogative
According to Givan (1989: 154), three socio-psychological dimensions underlie this continuum: "(a) the power (authority) gradient between the speaker and the addressee, (b) the degree of the speaker's ignorance concerning a state of affairs about which he wishes to learn, and (c) the degree of the speaker's sense of urgency or determination vis-a-vis the attempted manipulation. All these parameters are scalar, thus representing a multidimensional space." Quite apart from the fact that these three hypothetical dimensions are undefined, ad hoc, and arbitrary (for example, why 'urgency'? does the bare imperative always imply 'urgency'?) their 'scalar' character is clearly an artefact of Givan's presentation. For example, the difference between Pass the salt and Please pass the salt is perfectly discrete, and if Givan's methodological framework doesn't provide him with adequate tools for spelling it out, it is a feature of the framework, not of the data under discussion. Being an artefact of Givan's analysis, rather than a reflection of the nature of linguistic communication, the 'scalar description' of illocutionary forces could nonetheless, in principle, serve some useful purpose in applied linguistics. I doubt, however, that it does that either. Consider, for example, the situation of a Russian learner of English who is told that the Would you pattern is number 4 from the top on the continuum from the 'most imperative' to the 'most interrogative' sentences. What use would this be to him or her? As Comrie points out, Russian doesn't use the Will you or Can you patterns in the way English does, and the same applies to the Would you pattern: ... in English one polite way of getting someone to do something is by asking a yes/no question using either some form of 'will' or some form of 'can'. In other languages, that's not conventionalised. If you tried it in Russian, the reaction would be 'What's this guy trying to do?' (Comrie 1984b:282)
Givan did not put the Will you and Can you patterns on this continuum, so one doesn't know what slots he would have assigned to them: 3 and 4? 4 and 5? perhaps 3 and 5? Whichever numbers he might decide to
Are illocutionary forces indeterminate?
205
assign to these patterns, their value to a language learner would be limited, and they could not substitute for a qualitative description of the meanings and uses of the construction types. Furthermore, it would not always be easy (or possible) to rank different types of utterances 'from the most imperative to the most interrogative'. For example, Giv6n places Would you above Could you, and presumably he would opt for placing Will you over Would you. But how would one rank, for example, Will you please with respect to Would you (without please), or Would you please with respect to Could you (without please)? Incidentally, Comrie's comment linking some forms of will and can whimperatives with politeness is also misleading, given the fully felicitous nature of English sentences such as Will you please shut up! or Will you bloody well hurry up! (Williamson 1974:56; cf. Chapter 2 above). But if we reveal the different configurations of illocutionary patterns, this may be practically helpful, as well as theoretically illuminating. In particular, it will show which patterns expect a verbal response and which don't; which patterns assume (according to the speaker) that the addressee 'has to do' what the speaker wants and which don't; which show that the speaker feels bold enough to say 'I want you to do it' and which reveal the speaker's reluctance to go on record as having said this. And so on. Consider, for example, the relationship between the three English constructions: the bare imperative, the Will you pattern, and the Would you pattern. The semantic structure of the bare imperative (of action verbs) can be represented as follows: Do (a) (b) (c)
this (X) I say: I want you to do this (X) I say this because I want you to do it I think: you will do it because of this
This is closely related to the explication proposed earlier for orders, but there is one important difference: while both orders and bare imperatives confidently expect that the speaker's 'want' will be complied with, orders include one additional component: an assumption that the addressee has to do what the speaker wants. For example, if one shouts to someone unaware of a sudden danger: Step back!, this is not an order, and there is no assumption that they have to comply with the speaker's will, but there is an expectation that they will obey.
206
The semantics of illocutionary forces
The semantic structure of the two 'whimperatives' can be portrayed as follows: Will you do X? (e.g. Will you please shut up? Will you pass the salt?) (a) 1 say: 1 want you to do X (b) 1 say this because 1 want you to do it (c) 1 don't know if you will do it (d) 1 want you to say if you will do it Would you do X? (e.g. Would you try to put this through your hair?) (a) 1 say: 1 would want you to do X (b) 1 say this because 1 want you to do it (c) 1 don't know if you would do it if 1 said 1 wanted you to do it (d) 1 want you to say if you will do it
Component (b), which spells out the illocutionary purpose, is the same in both cases ('I say this because 1 want you to do it'), and it links the two patterns in question with bare imperatives (Do X). Component (d), which solicits a verbal response (in addition to a non-verbal one) is also the same in both cases. But the most overt component, which can be called the dictum (a), is different in each case: in the Will you pattern it is the same as in the case of the straight imperative ('I say: 1 want you to do it'), whereas in the Would you pattern it appears to be more tentative ('I say: 1 would want you to do it'). Furthermore, while neither of the two patterns is totally confident about the outcome, the Will you pattern constitutes a more forceful attempt to ensure the desired outcome. This difference is reflected not only in the (a) component (the dictum) but also in the ignorative component (c): compare 'I don't know if you will do it' with 'I don't know if you would do it if 1 said 1 wanted you to do it'. This analysis accounts, 1 think, for both the similarities and differences in the way the two patterns are used. For example, according to my informants the sentence Would you bloody well hurry up? is less felicitous than Will you bloody well hurry up? - presumably, because the forceful, uninhibited nature of the verb shut up can be seen as clashing with the conditional and rather tentative character of the Would you pattern. 1 might add that when my research assistant leaves timesheets for me to sign she usually puts on them a pencilled line "A(nna), would you sign these?", and never "A, will you sign these?", which
Are illocutionary forces indeterminate?
207
would sound (in my judgement and in that of my informants) more peremptory and less courteous. Furthermore, the analysis proposed here accounts also for the formal features of the patterns in question, and in particular, for their links with both imperative and interrogative constructions; and with tagged imperatives such as Do X, will you? and Do X, would you?, which will be discussed later. Another closely related pattern is Can't you ... ?, which suggests even more irritation and dissatisfaction, as the example below illustrates (note also the use of just and stupid): 'Can I come in?' 'No!' shrieked Gilly, then snatched open the door. 'Can't you leave me alone for one stupid minute?' (OR) Can't you just leave me alone? (OH)
I propose for this pattern the following explication: Can't you do X? I see you are not doing something that you should be doing I feel something bad because of this I think you should know: this is bad I say: I want you to say if you can't do this thing I think you can do it and don't want to do it I say this because I want to say what I think and what I feel
The form of the patterns described above is not arbitrary, and it can be partly understood in terms of 'conversational implicatures' and 'natural logic' because it reflects some of the components of the semantic (illocutionary) structure. It is, however, language-specific, and it is partly 'arbitrary' (that is, unpredictable from a cross-linguistic perspective), because it doesn't reflect overtly all the components of the illocutionary structure.
1.3. Syntax and illocutionary force Consider utterances such as the following ones (Schreiber 1972:326; Levinson 1983:256; Davison 1975:163). Frankly, I lied. John's at Sue's house, because his car's outside.
208
The semantics of illocutionary forces
Proponents of the performative hypothesis have often tried to account for the syntax of such sentences by claiming that the adverb frankly and the because clause modify the underlying performative verb tell. However, as Levinson pointed out: ... there are significant semantic difficulties here. ... it is simply not clear that the meanings of the relevant adverbs are indeed parallel in the explicit performative, the (allegedly) implicit performative and the reported performative usages: (50) I tell you frankly you're a swine. (51) Frankly, you're a swine. (52) John told Bill frankly that he was a swine. (Levinson 1983:255)
In prefacing one's sentence with frankly, as in (51), the speaker is not commenting on her own comment while delivering it, as (50) would suggest. Rather, she is saying two things, first preparing the addressee for what is to follow: I now want to say something frankly I say: you're a swine As for sentences with a because clause: it is clear that the because-clause here does not in fact modify any implicit I state or I claim, but rather an understood I know, as made explicit in (59): (58) I state John's at Sue's house because his car's outside. (59) I know John's at Sue's house because his car's outside. (Levinson 1983:257)
(Noticing John's car outside the speaker infers, in her own mind, that John is at Sue's house. Her stating that John is at Sue's house follows this inference, and may be due to a number of causes. The direct causal links hold between noticing that A and saying that B.) Perhaps the clearest evidence of the inability of the performative hypothesis to deal with the kind of data it was originally meant to account for comes from metalinguistic comments such as To sum up, To change the subject or To cut a long story short. As Mittwoch (1977: 183) pointed out, "it would be counter-intuitive to derive (29b) [To change the subject] from: I tell you this in order to change the subject." One says To change the subject to alert the addressee that a new topic is being introduced; but what one says about this new topic may be said
Are illocutionary forces indeterminate?
209
in order to inform the addressee, to express a sudden thought, and so on - not necessarily in order to change the subject. I quite agree with Levinson, Mittwoch and others, that the attempts to explain the interaction between syntax and illocutionary force on the basis of the performative hypothesis were unsuccessful. But the alternative accounts, proposed by the opponents of this hypothesis such as Leech (1983) or Bach - Hamish (1982), are hardly more convincing. What this alternative account suggests is, to my mind, quite incredible: that all sentences which signal overtly some aspect of their illocutionary force, and which can't, therefore, be explained in terms of 'autonomous' (non-illocutionary) syntax, are simply ungrammatical! They are of course perfectly acceptable, and they are used all the time, but since our favourite grammatical theories can't generate them, then they are ungrammatical. Because "usability is not grammaticality, and acquiring a use does not turn the ill-formed into the well-formed" (Bach Harnish 1982:225). In other words, if there is a conflict between grammatical theory and plain facts - then too bad for the facts! Among facts thus condemned by the grammatical theory are all the 'whimperatives', all the 'queclaratives', all sentences with so-called 'style disjuncts': in fact, it seems, the bulk of conversational English. Thus it is 'ungrammatical' to say any of the following utterances (Leech 1983:193-194; Bach - Hamish 1982:219,230): Can you please close the window? Frankly, you bore me. Who gives a damn about that? Why don't you be quiet? No smoking. Two coffees, please. Shut the window, can't you? As Leech (1983: 195) puts it, "such sentences do not lend themselves to generalisation in any kind of grammatical framework, being essentially exceptions to general rules". For my part, I would like to ask: what is so sacrosanct about those 'general rules'? If they don't fit the facts shouldn't we re-examine the 'rules', rather than condemn the facts? In fact, it is not true that facts of this kind can't be accommodated in any kind of grammatical framework. They can be accommodated in a framework which derives surface structures from semantic structures couched in terms of illocutionary components. For example, while it is
210
The semantics of illocutionary forces
true that sentences with the illocutionary adjunct frankly can't be plausibly derived from I tell you frankly, they can be plausibly derived from the following structure: 1 now want to say something frankly 1 say: X An infinitive clause such as to change the subject can't be plausibly derived from I tell you this in order to change the subject, but it can be plausibly derived from a structure where it will modify one of the illocutionary components: 'I say this because ... " 'I know this because ... " 'I think this because ... " and so on. Generally speaking, the main mechanism which operates in deriving surface structures from semantic structures is that of abbreviation, or 'deletion'. A 'whimperative' sentence, which combines in its surface structure imperative and interrogative features, must be derived from an underlying structure which contains, among others, components such as 'I want you to do something', and 'I want you to say something'. But the full meaning of a particular illocutionary form (for example the full meaning of Why don't you do X? or of How about X?, or of Why do X?) is very complex, and it is language-specific. The view that a sentence in a frame such as Why don't you is ungrammatical, and that its force can be guessed on the basis of "the general principles of rational, purposive human behaviour" (Leech 1983:195), is singularly unhelpful, and indeed perverse - especially from the point of view of a second language learner in whose native language a literal equivalent of this frame can't be used with anything like the force that it has in English. An immigrant learning English might guess that an utterance in the frame Why don't you will have an interrogative component in its meaning, but can't possibly guess its full meaning. This full meaning has to be stated for the learner. It is surely the responsibility of the linguist who wants to describe English to say what this meaning is.
More whimperative constructions
211
2. More whimperative constructions 2.1. Why don't you do X (tomorrow)? A sentence such as Why don't you play tennis any more? can be a straightforward question. If, however, a sentence in the frame Why don't you refers to a specific (non-habitual) action and has a future time reference, as in: Why don't you go and see a doctor tomorrow?
then the sentence cannot be simply a question: it must convey the assumption that it would be a good thing for the addressee to do the thing mentioned. Green (1975:127) has pointed out that the sentence: Why don't you be quiet! is an unambiguous 'whimperative', whereas the sentence: Why aren't you quiet? is an unambiguous question. I think that this observation can be generalised: not just be, which differs from the present tense form are, but any infinitive combined with why don't you will be interpreted as a 'whimperative', provided that the sentence has a specific time reference. In a particular context, it is often fairly clear that a sentence in the Why don't you frame is in fact meant as an invitation, or as an offer, or as a suggestion, or a request, for example: 'Did you do division with fractions at Hollywood Gardens?' Gilly shook her head. ... 'Why don't you bring your chair up to my desk and we'll work on it?' (GH) [offer of help] Why don't you bring me a wild one, Miss Gilly? I need to wake up that fifty-year-old senior citizen I've got for a son. (GH) [request for a necktie] Why don't you just go up and finish your nap, Mr. Randolph? I feel bad waking you up like this. (GH) [solicitous suggestion] Tell you what, why don't you stay the weekend? (GP) [suggestion/invitation]
212
The semantics of illocutionary forces
But even in such relatively clear cases, a measure of 'indeterminacy' remains. For example, the invitation and the offer cited above could also be reported with the verb suggest rather than with invite and offer. It is important to realise, however, that this 'indeterminacy' is a function of the descriptive framework which forces the analyst to choose between invite, offer, suggest and request. Utterances in the frame Why don't you do X? can be reported in a number of different ways because English has a number of illocutionary verbs which are compatible with the illocutionary force encoded in this construction. In choosing a particular verb (request, suggest, or whatever) the reporter imposes a certain interpretation on the original utterance, and can choose one of a number of interpretations compatible with the force signalled by Why don't you. In doing so, the reporter adds to what is encoded in the construction itself. This is, then, where the 'indeterminacy' lies: in the range of possible interpretations, which can be signalled by a range of reporting verbs. But the force of Why don't you is quite determinate. It can be stated as follows: Why don't you do X? (a) I say: I want you to say, if you can, why you/we shouldn't do X (b) I think you can't (say why) (c) I think it will be a good thing if you/we do it (d) I say this because I would want you to do it
Since it is the task of linguistics to pair surface structures with the meanings encoded in them, it is plainly the task of the linguist to spell out, in particular, the meaning of a frame such as Why don't you - and, I believe, this task can be carried out in the way indicated above. The formula postulated here can be validated in terms of meaning, in terms of form and in terms of use. The meaning can be verified intuitively, since the metalanguage used in the formula is derived from natural language. The form can be explained as derived by mere deletions from the underlying structure. (Of course, to say that the rules linking semantic representations to surface forms are for the most part 'mere deletions' doesn't mean that they don't have to be described, and accounted for in terms of more or less general principles.) And various aspects of use can be accounted for in terms of individual components either present or absent in the postulated semantic structure. For example, the fact that the frame in question can't take a pre-verbal please can be explained by the absence from the underlying structure of the
More whimperative constructions
213
component 'I want you to do it', which would be posited for 'whimperatives' such as Could you do X? but not for 'impositives' such as Why don't you do X: Could you please be quiet. *Why don't you please be quiet.
It is particularly interesting to note that, although more tentative than a straight imperative, the Why don't you pattern does not have to be particularly 'polite'. For example, it is perfectly felicitous in curses, such as Why don't you all go to hell! (Hibberd 1974:199). But a curse of this kind - contrasted with the imperative Go to hell! - suggests a somewhat impotent exasperation rather than self-confident anger. (For further discussion of the formula proposed here see the following subsections.)
2.2. Why do X?
Now consider sentences such as the celebrated: Why paint your house purple?
(Gordon - Lakoff 1975). The long controversy over whether a sentence of this kind is a question or a criticism seems to me rather futile. In fact, it partakes of the nature of both question and criticism, and its illocutionary structure simply can't be captured accurately in terms of global categories such as ask and criticise. What is needed are more fine-grained components, such as: Why do X? (a) I say: I want you to say, if you can, why you/we should do X (b) I think you can't (say why) (c) I think: X is not a good thing (to do) (d) I say this because I want to say what I think
I maintain that the syntactic construction Why do X? always carries these components with it, not only in those sentences where the content supports this interpretation but also in those where it doesn't. For example, the sentence: Why paint your house white?
214
The semantics of illocutionary forces
implies exactly the same attitude as the sentence with the word purple in lieu of white. This means that the construction itself provides unambiguous grammatical clues to the illocutionary force. But this force cannot be stated as simply 'I ask' or 'I criticise', or even as 'I ask and I criticise' (a la Sadock 1974). It can only be stated in terms of a set of illocutionary components. It is worth adding that the construction in question has a more or less symmetrical counterpart of the form Why not do X?, as in the following example: Why not have him come here? (Why go all the way to his place?)
The illocutionary force of this 'twin' construction differs from that of Why do X? in terms of evaluation and also in terms of illocutionary purpose. Why do X? implies that X is not a good thing to do; Why not do X?, on the other hand, implies that X is a good thing to do. The former is a kind of tentative criticism; the latter, a kind of tentative suggestion. The former can refer to faits accomplis, but the latter has a future orientation. For example, if a house has already been painted, and cannot be repainted, one can still say Why paint the house purple?
But if one says: Why not paint it purple?
one is implying that it is not too late to carry out the 'good idea' in question. Searle (1975:69) insists that a sentence in the frame Why not do X? can be a straightforward question 'without necessarily being also a suggestion'. As I see it, it is true that sentences of this kind don't have to be suggestions, but it is not true that they can be straightforward questions, because they will always contain some component such as 'I think that it would be a good thing to do X'. I suggest that the illocutionary force of the Why not frame can be represented as follows: Why not do X? (a) I say: I want you to say, if you can, why you/we shouldn't do X (b) I think you can't (say why) (c) I think it would be a good thing if you/we did it (d) I say this because I want you to think about it and to say what you think about it
More whimperative constructions
215
Comparing the explications for the frames Why don't you do X? and Why not do X? one will see that they share a number of components, and that they are both compatible with the meaning encoded in the verb suggest. One will also see that they differ in ways which make the former, but not the latter frame, compatible with the meaning of request. Thus, the first three components of this formula (a, b and c) are the same as those posited for the frame Why don't you? But the component (d) postulated for Why don't you? ('I would want you to do it') has been omitted from the explications of Why not do X?, to show that this time the speaker presents himself as totally disinterested. The illocutionary purpose of each of the two constructions is also different: in the case of Why don't you? sentences the speaker is attempting to cause the action to happen (if the addressee is in favour); but in the case of Why not do X? sentences the speaker is merely trying to cause the addressee to make up his/her mind.
2.3. How about X? How about sentences are usually used for making suggestions, offers or invitations (but not, for example, orders or instructions): How about a drink? How about a movie? How about you and me doing a little red-hot reading after supper? (GH) How about dinner at my place? How about going to Sydney?
One can't claim, however, that sentences of this kind can be represented as derived from or equivalent to I suggest (that we see) a movie, I offer you a drink or I invite you to dinner, because it would involve a great deal of arbitrariness to decide which performative verb to use in each case. And to suggest that all How about sentences are multiply ambiguous (not simply ambiguous but multiply ambiguous) is almost tantamount to admitting their indeterminacy. But if one admits this then one is left with an analysis devoid of explanatory power; because if How about sentences are either multiply ambiguous or indeterminate then why is their range of possible illocutionary forces so strangely limited? Why can't they be used, for example, for ordering, commanding, begging, imploring, or giving permission?
216
The semantics of illocutionary forces
The 'mystery' is easily solved, however. If How about sentences can be used for suggesting, offering or inviting, but not for ordering or begging, it is because the former three acts are compatible with the components encoded in English by the How about construction. These components can be spelt out as follows: How about doing X? (a) I think you might want to do X (b) I say: I want to know 'how you would feel about it' (c) I would want it to happen if you wanted it (d) I don't know if you would want to do it (e) I say this because I want you to think about it and to say what you think about it The action is presented as potentially desirable, but the speaker is not as confident on this point as in the case of Why don't you sentences. Furthermore, the speaker doesn't advocate the action as desirable from his/her own point of view, but rather invites the addressee to form and express a view. The speaker is not saying, therefore, 'I think it would be a good thing if it happened', but rather, 'I think you might want to do it'. There is no assumption, therefore, that the action should be carried out. Nonetheless, the speaker does indicate a personal interest in the action, but as in the case of Why don't you sentences, this personal interest is expressed in a conditional form ('I would want it to happen if you wanted it'). A How about sentence is more 'open-minded' or open-ended, therefore, than a Why don't you one. The speaker wants to hear if the addressee wants to carry out what has been proposed as potentially desirable, but is not pressing the addressee in any way. (It is interesting to note in this connection that while the sentence Why don't you all go to hell! is felicitous, the sentence How about you all going to hell! would be comical.) Since the How about pattern is used commonly for offers and suggestions it might seem that it presents the action in question as 'good for the addressee', and that its explication should therefore include a component of the form 'I think it would/might be good for you'. But in fact, this pattern can also be used for requests (though not for orders), as in the following examples: 'How about giving me a hand with this salad?' 'No.' 'Oh.' (GH)
More whimperative constructions
217
Supper's' bout ready. How about going next door and getting Mr. Randolph? He eats here nights. The word No was just about to pop out of Gilly's mouth, but one look at Trotter's eyes, and she decided to save her fights for something more important. (GH)
The important feature of such requests is that they are counting on the addressee's willing cooperation. This is compatible with the component 'I think you might want to do X', but not with 'I think it would/might be good for you.' In an insightful discussion of How about sentences in English, Shopen (1974:794) argues that the pattern in question "should be viewed as having status in the grammar in its own right". I would entirely agree with this claim, in view of the unique illocutionary force signalled by this pattern. I could not fully agree, however, with the concomitant claim that "there is no non-elliptical source ... from which this pattern could be derived that has the same semantic properties". An explication of the kind proposed above can be regarded as just such a 'non-elliptical source' - at least in the sense that it constitutes a non-elliptical paraphrase (or 'paralocution', cf. Boguslawski 1981 a) which spells out the unique illocutionary force of the How about pattern. To say this is not to deny that the expression How about can also be used in 'pure questions'. But as Shopen rightly recognised, apart from its use in pure questions it can also be used with an illocutionary force different from that of questions. I suggest that this illocutionary force can be fully spelt out in a non-elliptical formula which takes the form of a unique series of illocutionary components (unique, and yet sufficiently overlapping the formula for 'pure questions' as to account for the formal link between them). It should be added that the 'non-elliptical' formula proposed here includes elements other than those postulated as universal semantic primitives (for example 'how', 'feel'); and also, that phrases such as 'feel about' do not belong to the proposed semantic metalanguage. In this sense, the explication sketched here cannot be regarded as the true 'non-elliptical source' of the expression in question. I have kept, however, the phrasing 'how you would feel about it' to show how the 'how about' pattern can be seen as related to (and as an abbreviated form of) a 'deeper' (intermediate) level of semantic representation (cf. Wierzbicka, in press b). The paralocution 'how you feel about it' is still semi-idio-
218
The semantics of illocutionary forces
matic, and at a deeper level of analysis it should (and could) be replaced with a non-idiomatic configuration of universal semantic elements.
3. Additional remarks on the explication of illocutionary forces Several formulae explicating illocutionary forces have been introduced so far, and many more are going to follow. It may be useful to add a few comments on the methodology of semantic explications at this stage, for the reader may still feel puzzled: how does one know what components should be posited for a given construction? and how does one prove that the components proposed have not been arbitrarily chosen? I reply that one proceeds, essentially, by trial and error. The goal is this: to propose for each construction a minimal set of components which will jointly account for all the aspects of its use. As a research strategy, a contrastive approach is most fruitful: if one tries to model several closely related constructions at the same time, trying to capture both the similarities and the differences between them, one has the best chance of providing the most accurate 'portrait' of each illocutionary force. However, the ultimate goal consists in explicating, as fully and accurately as possible, each illocutionary force in its own right - not in providing some sort of differential schema for the whole lot. Above all, one must resist the temptation of imposing some 'system' on the empirical reality of illocutionary forces, which can be as capricious, idiosyncratic and asymmetric as any other semantic structures. To be sure, a great deal of symmetry and order is also likely to be discovered, but the exact proportion of symmetry and asymmetry, of 'system' and idiosyncracy has to be revealed through empirical research, not decided about on an a priori basis. For example, if we decide to posit the component 'I think you might want to do it' for the frame How about (How about a beer, How about a movie, and so on) but not to posit it for the frame Why don't you (Why don't you shut up), it would be a grave error to posit for the latter frame a negative component 'I don't think you might want to do it', or 'I don't say that 1 think that you might want to do it'. There is an essential difference between the absence of a component and the presence of its opposite.
Selected conversational strategies
219
Semantic explications are difficult to justify fully (within a single study, or even a single volume), because ultimately, the only way to justify them is by refuting alternative formulations. Each explication proposed in the present chapter has gone through a number of versions, and in each case dozens of competitors have been considered and discarded. This is of course no guarantee of perfection. But it would be an illusion to imagine that there is (or could be) some simple, mechanical procedure for evaluating semantic explications. The only way to challenge them is to engage in the laborious process of devising competing analyses, and of defending them, point by point, against attempts at refutation. The present study of various sentence types and grammatical forms, and the illocutionary forces they encode, may be seen as a first step in this process. I try to propose for each construction a set of (illocutionary) semantic components which can account for its use. Once proposed, they are accessible to discussion, revision, refutation or verification.
4. Selected conversational strategies This section examines several examples of conversational frames used for different illocutionary purposes: making suggestions, introducing information, and indicating the speaker's attitude or opinion. An explication of the illocutionary force of each construction is proposed.
4.1. Tell you what, Sf
Acts similar to suggestions can also be signalled in English by means of the frame (I) tell you what, as in the following example, (quoted earlier for a different purpose): Tell you what, why don't you stay the weekend (GP)
An utterance in this frame could be reported with the verb suggest ('he suggested that she stay the weekend'), just as Why don't you utterances and How about utterances can. In fact, in the example given above Tell you what and Why don't you are used jointly.
220
The semantics of illocutionary forces
But Tell you what implies that the idea has just occurred to the speaker, whereas How about or Why don't you could voice an idea which the speaker has been nurturing for a long time. Furthermore, Tell you what implies that the sudden idea which has just occurred to the speaker will constitute a solution to a problem of current relevance. The word what seems to echo a question which has been occupying the interlocutors' minds (what should we do?). It offers an answer to that implicit question: (I) (will) tell you what I think (we should do). No such implications are included in the formally similar expression I (will) tell you something. The reference to the first person should of course be interpreted broadly. As pointed out to me by Tim Shopen, one could say, for example: Tell you what, she could try auditing a course.
But even addressee I think what (for
here, the implication seems to be that the speaker and the are somehow responsible for the person in question. the following formula can be proposed for the frame Tell you a different analysis, see Fillmore 1984):
Tell you what, Sf I think I now know what you/we should do because I now think something (I didn't think it before) I want you to know it I say: I think it will be good if you/we do something (S) I say this because I want you/us to do it if you think the same
4.2. Do you know, S? Do you know, he has started a new poem? (NI) Do you know sentences are related to Tell you what sentences, in so far as both types signal 'news', and moreover, news that the addressee can be expected to be interested in. In the case of Tell you what sentences, the 'news' is directly relevant to the addressee, because the speaker has a 'good idea' concerning the addressee or someone close to the addressee ('it would be good if you/we did X'). No such direct relevance is implied by Do you know. What is implied by the latter frame is simply interest: 'I think you would want to know this'. The full semantic formula would read:
Selected conversational strategies
221
Do you know, S? I say: I want to say something to you (S) I think you would want to know it I think you wouldn't have thought this I say this because I want you to know it I think you will say something because of this I want you to say something because of this
An interesting formal feature of Do you know sentences used in this sense is that they don't allow an explicit complementiser: Do you know THAT S indicates that the addressee may already know the news. Do you know, I had a dream about you last night. ?Do you know that I had a dream about you last night.
Sentences in the form Do you know that may also introduce interesting and unexpected news, but they have to have an 'ignorative' component 'I don't know ... ' (cf. Wierzbicka 1980:315): Do you know that X?! => 'I don't know if you know it'
The contrast in illocutionary force between Do you know (*that) S? and Do you know that S? can be illustrated with the following pair of examples: Do you know, for years I used to dream that he'd caught us in bed together? You and me. Even after we were married. (GP) 'It's absolutely disgusting!' 'Do you know, Denise', Gavin said, 'that the foolish folk of Staunton might well, and in a deeply depressing majority, find it more disgusting that your delightful nipples are visible through your charming jersey.' (GP)
In the first example, the speaker knows that the addressee cannot possibly know the content of the complement clause, and he introduces this clause by Do you know, without that; but in the second, the speaker cannot have the same certitude, and uses the variant with that (Do you know that). What holds for the frame Do you know that holds also, mutatis mutandis, for the frame Did you know that. A sentence such as:
222
The semantics of illocutionary forces
Did you know that the Spanish anarchists once passed a resolution saying that any woman who excited a man's desire had a moral obligation to satisfy it? (GP)
could be meant as a piece of news, rather than as a genuine question, but it would still have an 'ignorative' component, 'I don't know if you know it'. But the same sentence introduced by the frame Do you know (*that) would imply that the speaker assumes the addressee doesn't know. And a piece of news which couldn't possibly be known to the addressee would have to be introduced by Do you know (*that), not by Did you know that: *Did you know that I had a dream about you last night?
4.3. Don't tell me Sf
Eleanor: (angrily). Michael! Don't tell me you're becoming jealous of John again! (WE) Unexpected news can also be indicated by the frame Don't tell me. This time, however, what is said is new to the speaker, not to the addressee. This means that the speaker discovers something new and unexpected, and voices this discovery, with disbelief and disapproval. Usually, the discovery concerns the addressee,. and is based on something that the addressee has said, (as in the example above), but it doesn't have to be so. For example, catching a glimpse of a female acquaintance one might exclaim to somebody else: Don't tell me she's pregnant again!
Similarly, having planned an excursion and having made no provision for bad weather one could exclaim: (Oh no!) Don't tell me it's raining!
But one would have to be a masochist, as well as a pessimist, to be able to exclaim: Don't tell me the weather's fine!
As a first approximation, 1 propose to spell out this illocutionary force by means of the follolwing formula:
Selected conversational strategies
223
Don't tell me S! I see something now I think I have to think S because of this I don't want to think this I wouldn't have thought I would see this I feel something because of this I think one cannot say that this is not a bad thing I want you to say something because of this I say: I don't want you to tell me S I say this because I want to say what I think
The discovery, the disbelief, the negative evaluation and the consequent feeling link the expression Don't tell me S with the exclamation Oh my God (see below, section 20). But Don't tell me, unlike Oh my God, leaves some room for doubt, and it includes an appeal to the addressee: 'I want you to say something because of this', 'I don't want you to tell me S'.
4.4. How many times have I told you (not) to do Xl Even stronger disapproval than that embodied in Don't tell me is of course signalled by expressions of direct rebuke (reproach, reprimand etc.) such as How many times have I told you (not) to do X. In this case, the disapproval has to be directed at the addressee (rather than at a third person), and it is accompanied by an attempt to make the addressee feel bad, because of what he or she has done. The full illocutionary force can be spelt out as follows: How many times have I told you not to do X! I see that you did X I think you know that you shouldn't do this because I told you many times I say: I want you to say how many times it was, if you can I think you can't, because it was very many times I feel something bad because of this I say this because I want you to feel something bad
The expression 'I feel something bad', included in this explication and in many others, sounds no doubt somewhat naive and unidiomatic, but if we specified the 'bad feeling' in question as displeasure, annoyance,
224
The semantics of illocutionary forces
irritation, anger or anything else, we would be acting arbitrarily. On the other hand, to say that the component in question is 'indeterminate' (between annoyance, anger, etc.) would be just as unjustified as to say that How many times have I told you as a whole is indeterminate between rebuke, reproach, reprimand, scolding and whatever.
4.5. Who's talking about doing X? Who is talking about getting married? You talk as if marriage was the only alternative. Don't you ever listen? (GP)
This is another construction expressing the speaker's annoyance with the addressee. In this case, the annoyance (or, more generally, 'bad feeling') must be caused by something that the addressee has just said, and the nature of the offense is quite specific (although it is not specified on the surface of the sentence): the addressee has attributed to the speaker an intention which in fact the speaker doesn't have. The expression Who's talking about (or of) doing X? is used as a kind of angry disclaimer, protesting the addressee's mistaken assumption. The full illocutionary force can be spelt out as follows: Who's talking about doing X? I think you think that I want to do X I think you think this because I said something now I feel something bad because of this I say: I don't know who is talking of doing X I say this because I want you to say it (who) if you can I think you can't I don't want you to think that I want to do X
5. Tag questions 5.1. Tags with declarative sentences To take another set of examples, I will tum now to a number of different tag question constructions (cf. Chapter 2, section 3.3). First, let us consider declarative sentences with an opposite polarity tag such as:
Tag questions
225
Maria is Italian, isn't she?
At first sight, it might seem that sentences of this kind express the speaker's uncertainty, and seek confirmation from someone who might know what the truth is. But clearly, it is not knowledge or truth that the speaker is seeking in examples such as the following ones: Maria is very nice, isn't she? This is delicious, isn't it? Lovely day, isn't it?
In earlier work, I have tried to account for the apparently different functions of different tag questions by assigning to them slightly different types of explications, and by phrasing some of these explications in terms of what one knows or doesn't know, and others, in terms of what one thinks. But further consideration of a wide range of examples has convinced me that the distinction between tag questions seeking factual confirmation and tag questions checking sameness of opinion was not always feasible. Consider, for example, the following exchange: 'Mr. Randolph got enough books to start a public library, haven't you, Mr. Randolph?' 'Well, I do have a few', he chuckled. (GH)
In this case, the tag question refers to a factual matter, not to a matter of opinion; but what the speaker is seeking is not information or verification. Tag questions of this kind can therefore be represented neither in terms of 'I don't know - you may know', nor in terms of 'I think this - I want to know if you think the same'. On the other hand, a different formula, couched in terms of 'saying' rather than 'knowing' or 'thinking', can accommodate all confirmation-seeking tag questions. I suggest, therefore, an explication along the following lines: S, [opposite-polarity tag]? (e.g. Maria is Italian, isn't she? Maria is lovely, isn't she? Mr. Randolph got enough books to start a public library, haven't you, Mr. Randolph?) I say: S [Maria is Italian; Maria is very nice; Mr. Randolph got enough books to start a public library] I think you would say the same
226
The semantics of illocutionary forces
I know: you may not want to say the same I want you to say if you would say the same I think you will say you would say the same As for matching-polarity tags, such as: Sally is pregnant, is she? You have bought a house, have you?
Cattell (1973) has shown that they usually echo the interlocutor's earlier utterances. For example, if John asks Harry about the meaning of a Russian sentence and Harry supplies that meaning, John could echo Harry's explanation saying: It means 'necessity is the mother of invention', does it?
He could not respond by saying: It means 'necessity is the mother of invention', doesn't it?
I would add, however, that matching polarity tags can also be used as an interpretation, rather than as an echo of the interlocutor's utterance, as in the following exchanges: Hannah: I know, but couldn't you just accept it as security for a few days' stay here? Maxine: You're completely broke, are you? (NI) Shannon: They can't go back in toooowwwwn! - whew - ... Are they getting out of the bus? Maxine: You're going to pieces, are you? (NI) One can account for both the echoing and the interpreting function of matching polarity tags by means of the following formula: S, [matching-polarity tag]? I say: S (e.g. you are completely broke) I say this because of what I see/hear I don't want to say that I know this I want you to say that I can say I know this
The speaker is not really checking whether she has heard correctly (often, this would be impossible to doubt). Rather, she is checking her interpretation of what she has heard - or pretending, for whatever reason, that she is checking it (to be sarcastic, to savour a piece of
Tag questions
227
news, to give herself a chance to reflect on what she has just heard, to encourage the interlocutor to elaborate, and so on). In fact, it seems that although matching-polarity tags are typically used in response to speech, they can also be used in response to something that one can see looking at the addressee. To allow for this possibility, the second component in the explication has been phrased as 'I say this because of what I see/hear' rather than 'I say this because of what you say'.
5.2. Tags with imperative sentences Turning now to tag questions added to an imperative, I should like to point out a wide range of illocutionary forces encoded in such structures - illocutionary forces which are fully 'determined' and which, nonetheless, cannot be expressed by means of simple performative verbs. Compare, for example, the following three constructions: a. Sit down, will you? b. Sit down, won't you? c. Sit down, can't you? These three patterns are of course closely related to the three 'interrogative' patterns discussed earlier: a. Will you sit down? b. Won't you sit down? c. Can't you sit down? But despite this close relationship, the imperative patterns cannot be fully reduced to the interrogative ones (or vice versa) and they merit a separate discussion, and a separate set of explications. In some circumstances, the three imperative constructions in question (will you?, won't you?, and can't you?) may seem to be interchangeable, but the force of each is quite distinct. In fact, while it is easy to imagine situations where either (a) and (b) or (a) and (c) are interchangeable, it is difficult to do so with respect to (b) and (c). Wishing to be very polite to a distinguished visitor, one would probably say (b) rather than (a), and one would certainly avoid (c). (c) implies that the addressee should have already performed the action, that it is a bad thing s/he didn't, and that the speaker feels some impatience or even irritation because of that.
228
The semantics of illocutionary forces
(b) and (a) are close to each other, but they differ in their implications concerning the beneficiary. Won't you implies that the action is seen as something that the addressee can be expected to want to do, whereas will you implies that it is seen as something wanted by the speaker. For example (as pointed out to me by Emmie O'Neail), if the addressee is pacing restlessly up and down the room, getting on the speaker's nerves, the speaker will be more likely to say Sit down, will you rather than Sit down, won't you. Typically, won't you is used in situations when the action is seen as beneficial for the addressee, not for the speaker. Since, however, won't you can also be used in utterances such as: Give me a hand, won't you?
I have formulated the relevant component as 'I think you might want to do it', rather than 'I think that it would be good for you to do it'. But of course will you, unlike can't you, doesn't always imply negative feelings. It is often used simply to convey a request. It can also be used to convey instructions and directions. Won't you, on the other hand, is frequently used in offers and in invitations. For example: Pass me a piece of toast, will you, Gilly? (GH) Make the beds, will you? (GH) Sit down for a minute, won't you? (GH) Come back for another little visit, won't you? (GH)
This association of will you with requests and of won't you with offers tallies very well with the idea that will you presents the action as desirable from the speaker's point of view, whereas won't you presents it as something that the speaker thinks the addressee would want, or would be willing to do. I would add that the combination of an imperative with will you constitutes a fairly confident request and that it is often used in asymmetrical relationships. The confident, self-assured nature of this frame can be accounted for by the component 'I think you will do it'. The fact that the combination of an imperative with won't you sounds more tentative, can be accounted for by the component 'I don't know if you will do it'. Sit down, will you? I say: I want you to do X (sit down) I don't know if you will do it
Tag questions
229
I think you will do it I say this because I want you to say that you will do it, and do it Sit down, won't you? I say: I want you to do X if you want to do it I don't know if you will do it I think you might want to do it I say this because I want you to say if you want to do it and do it if you want to do it Sit down, can't you? I say: I want you to do X I see you are not doing it I feel something bad because of this I want you to say if you can't do it I think you can't say it (that you can't do it) I think you can do X and don't want to do it I think you should know that this is bad I say this because I want you to feel something bad and to do X because of this
5.3. Why can't you (do X)!
Sentences such as Why can't you understand and be generous - be just! (WE)
are very close in force to imperatives followed by Why can't you: Why can't you leave me alone! Leave me alone, why can't you!
It might even be suggested that the two types may have the same force, and that they may be derived from the same underlying structure (cf. Sadock 1974). But in fact the two constructions, though fairly close, are distinct, and must be defined differently. The imperative construction is restricted to the second person, and it constitutes an attempt to influence the addressee. The interrogative construction allows third-person as well as second-person targets, and since we can't influence by speech someone who doesn't hear us, this construction can't constitute an attempt to influence other people. For example, the sentences:
230
The semantics of illocutionary forces
Why can't the English learn how to speak! Why can't a woman be more like a man!
are not attempts to amend the ways of the English, or of women. Both constructions state a certain negative perception (W is not doing X), both express a critical judgment which the speaker thinks should be shared by the culprits (I think W should know that this is bad), both attribute the culprits' failure to do what they should be doing to ill-will, not to inability (I think that W can do X and doesn't want to do it), and both appeal, rhetorically, for an explanation, an explanation which, in the speaker's view, can't be forthcoming (I want someone to say, if they can, why W can't do it; I assume nobody could). But the imperative construction expresses the speaker's will to change the situation, as well as a desire to make the addressee 'feel bad' (I say this because I want you to feel something bad and to do X because of this). In the interrogative construction, the illocutionary purpose seems to consist merely in expressing one's thoughts. Moreover, the imperative construction conveys a current perception (I see that W is not doing X). By contrast, the interrogative construction states an opinion, which could easily be a 'repeat'. For example, it seems likely that a person who says Why can't the English learn how to speak! once, would say it many times. It is appropriate, therefore, to phrase the relevant component of the interrogative construction as 'I say: W is not doing something that W should be doing'. This tallies well with the general impression that Why can't they do X is primarily a kind of criticism, irritated but powerless, whereas Do X, why can't you! is primarily a kind of directive, aimed at correcting an unsatisfactory situation. Do X, why can't you! I say: I want you to do X I see you are not doing X I feel something bad because of this I think you should know this is bad I want you to say, if you can, why you can't do X I think you can't say why I think you can do X and don't want to do it I say this because I want you to feel something bad and to do X because of this Why can't W do X! I think: W is not doing what W should be doing (X)
Tag questions
I I I I I I
231
feel something bad because of this think W should know this is bad say: I want somebody to say, if they can, why W can't do it think nobody can say it think W can do X and doesn't want to do it say this because I want to say what I think about Wand what I feel because of it
5.4. OK? OK? is an ubiquitous English tag, and yet little or no attention has been given to it in the literature on the English language. When it accompanies an imperative its value may seem to be fairly close to that of will you: Wait a minute, will you? Wait a minute, OK?
But in fact, OK? can't always be substituted for will you, as the following contrast in acceptability shows: Shut up, will you (will-ya)? ?Shut up, OK?
There is, so to speak, more to OK? than to will you? Will you is checking merely the addressee's willingness to do something, whereas OK? is also appealing to the addressee's judgement, and is soliciting something like agreement, as well as an indication of willingness to comply. There is also an implication that the speaker's judgement, as well as will, is involved. For example: Gilly, give Maime Trotter half a chance, OK? (GH) You need anything, honey, just let Trotter know, OK? (GH)
In both cases, the speaker proposes a course of action as something 'good', as well as 'wanted', and is checking the speaker's understanding and acceptance of what is being proposed. Consider also the following example: Then she thought better of it. 'You do it, William Ernest, OK?' (GH)
232
The semantics of illocutionary forces
Here, will you? could not be substituted for 'OK?' - presumably, because the contrastive stress on you shows that the addressee is being invited to consider a new variant of the situation. Will you?, it seems, can only be directed at the addressee's willingness, not at his or her thought processes. Most importantly, OK?, in contrast to will you? can also be used with declarative sentences (with a subject other than 'you'), for example: I'll stop by for you every day, OK? (*will you) (GH)
The condition seems to be that future or future/present tense, not past tense, is used: *1 stopped by for you every day, OK? I'm leaving it here, OK?
To account for all these features of OK? I would propose the following explication: OK? I say: I want this to happen I think it will be a good thing I know: it will not happen if you don't want it I want you to say if you want it I say this because I want it to happen
6. Personal abuse or praise: You X! Utterances such as You filthy swine! You fool! You dickhead! You idiot!
are so common in English that a linguistic theory which would be unable to analyse them and which would consequently rule them out as 'ungrammatical' would, I think, have to be regarded as a curiosum. Utterances of this kind convey with perfect clarity an identifiable illocutionary force, or rather, three alternative illocutionary forces, depending on the semantic class to which 'X' belongs.
Personal abuse or praise: You XI
233
a. You liar! b. You angel! You darling! c. You beauty! To see that these three different types cannot be collapsed under one broader category, consider the following regarded by most speakers as unacceptable:
*you student! *you killer! ?You wog!
Types (b) and (c) can be said to constitute minor types, but (a) is a major category of utterances, with a wide range of 'X's' and a high frequency of use. 'X' (in category a) has to fall into a general category of names of person which describe a person as someone who habitually does something bad and which convey a negative feeling. Student does not qualify because it doesn't refer to any activity regarded as bad, killer because it has no negative feeling built into it, and wog because it doesn't identify any particular vice. Used with an appropriate modifier, however, wog becomes perfectly acceptable: You dirty wog! You stupid wog!
The full illocutionary force of this category (a) can be spelt out as follows: You X [neg.] ! I see that you did something very bad I feel something bad towards you because of this I want to say something bad about you because of this I say: you are a bad kind of person one has to feel something bad towards someone like you I say this because I want to say what I think about it and what I feel because of it I think you should feel something bad because of this
Expressions such as you fool! or you monster! can of course be used jocularly and affectionately (I love you, you fool! I miss you, you little monster!), but such uses exploit the inherently pejorative character of words such as fool or monster.
234
The semantics of illocutionary forces
The positive subtype of You X! utterances (b) is not entirely symmetrical with respect to the negative one, as can be seen from the fact that in many speakers' judgement words of high praise, such as saviour or saint, are not fully felicitous in it. What is fully felicitous is words which combine unspecified and hyperbolic praise with affection. An utterance such as You angel! implies that the addressee has done something good for the speaker, not something good in general, but specifically, good for the speaker. As a result, the speaker wants to say something good about the addressee - and is 'lost for words' ('I can't think of something good enough to say about you'). Consequently, no specific good qualities can be mentioned, and the praise intended can be phrased only in terms of general 'lovability' ('you are a person towards whom one must feel good feelings'). A claim of 'lovability' must be distinguished from a purely subjective expression of individual affection, such as love or sweetheart. One can't say:
*you love! (cf. You angel!) Love used as a term of address means 'my love'. It expresses a relationship between the speaker and the addressee, and does not describe any generallovability, as angel and darling do. The illocutionary force of the you X subtype can be spelt out as follows: You X [pos.]! I see that you did something very good for me I feel something good towards you because of this I wouldn't have thought anyone would do this I want to say something good about you because of this I don't know what I can say I say: you are more than good one has to feel something good towards you I say this because I want to say what I think about it and what I feel because of it [I think you should feel something good because of this]
Finally, subtype (c) is confined to the expression You beauty! and its variants such as You little beauty!, You bloody beauty! and You beaut! In this subtype, you doesn't have to refer to a person but it has to refer to an agent or an action which delights the speaker and surpasses his/her expectations. For example, one couldn't say You beaut! or You bloody
Illocutionary forces of grammatical and other categories
235
beauty! at the sight of a beautiful sunset, although one could say it at the sight of a beautiful horse, or at the sight of a beautiful jump performed by a horse. In this subtype, too, the speaker can't find words good enough to describe what s/he sees, and s/he feels 'something good' because of it; but this time, the 'good feeling' doesn't have to be directed towards the agent ('I feel something good towards you because of this '). Furthermore, the exclamation You beaut! is not addressed to the agent. For example, it can well be uttered by someone watching a football match on television. It is not expected, therefore, to have any impact on the agent. The other two sub-types on the other hand, do seem to include a component aimed at influencing the addressee ('I think you should feel something bad/good because of this'). The semantic formula for You beauty! and its variants reads: You beauty! 1 see that something good happened because someone did something I feel something good because of this I wouldn't have thought I would see this 1 want to say something good about it because of this I don't know what I can say I say: this is beautiful one has to feel something good because of this I say this because I want to say what I feel because of it
7. Illocutionary forces of grammatical and other categories 7.1. Modal verbs Another category of clues to illocutionary force comprises modal verbs, used as in the following exchanges (whose dialectal aspect is irrelevant in the present context): Marthy: How old' II she be now? Chris: She must be - lat me see py Yo! (AC)
she must be twenty year ole,
236
The semantics of illocutionary forces
The combination of the modal will with the adverb now shows clearly that in this sentence will does not indicate future tense. What it does indicate, in this particular sentence, is that the speaker is inviting the addressee to make an estimate. Most generally, it indicates a thinking process, leading a person from not being able to say the right thing about something to being able to do so, with some confidence: Johnny: Where's it from? Larry (after a glance): St. Paul. That'll be in Minnesota, I'm thinkin'. (Ae) Quite clearly, the speaker is not informing the addressee that 8t. Paul is in Minnesota, and he is not exactly inferring it either. There is no performative verb in English which would capture exactly the illocutionary force of such sentences. Nonetheless, its force can be clearly stated: X will be P
I I I I I
don't say I know this [X is P] think I'll know it if I think about it want to think about it think I know it now say: [ ]
A question such as How old'll she be now? invites an answer based on current thinking effort: I think you will know· it if you think about it I want you to think about it The explication has temporal order built into it, referring as it does to a passage of time. It shows that the modal will is related to the future will, and that in fact the meaning of the latter is contained in the meaning of the former. (Consequently, the explication is not circular, as it reduces the modal will to the temporal will. It is not my purpose to explicate the latter in the present context.) As for the modal must (She must be at least forty), it indicates a similar, but not identical illocutionary force. It is less speculative, and it doesn't refer to time spent in thinking. Looking at the portrait of an unknown woman, one would say: She must be very beautiful.
Illocutionary forces of grammatical and other categories
237
but hardly ?She will be very beautiful.
Since in this particular situation the judgment is clearly based on what one can see (the portrait), not on a process of thinking, will is inappropriate. Of course the sentence is fine if will is interpreted in the future sense 'she will be beautiful in the future', but not if it is interpreted in the modal sense 'she will be beautiful now'. The illocutionary force of must sentences can be stated as follows: X must be P
I I I I
say: I think I can say this about X don't say I know it think one must think this say this because I want to say what I think
This is of course a much more tentative force than that of / must say sentences. The modal must in sentences of the X must be P type is not derived from an underlying 'I must say' component. Rather, it is derived from 'I think one must think this'. In saying 'X must be P' the speaker stops short of asserting that X is P and the explication doesn't include a component of the form 'I say: X is P'. But an explication of the frame 'I must say: X is P' would have to include such a component. For example, if the speaker is telling us a story about some unknown woman, we could respond by saying 'She must be very beautiful', but not by saying 'I must say, she is very beautiful' - because the latter sentence would imply a personal assessment of the woman's beauty, which is impossible in the case of someone unknown. Compare also the following two sentences: This must be the milkman. ?/ must say, this is the milkman.
The first sentence expresses a rather confident guess and it is of course perfectly acceptable; but the second sentence is odd, because it suggests a personal judgment concerning a non-factual matter, whereas the proposition introduced in this frame is in fact factual. The personal judgment in question appears to be expressed reluctantly but without the slightest hesitation as to its validity. This can be portrayed as follows: I must say, she is very beautiful (?this is the milkman) I say: she is very beautiful
238
The semantics of illocutionary forces
I don't say it because I want to say it I must say it if I want to say what I think I say it because I want to say what I think The non-factual character of the frame 'I must say' is accounted for by the combination of a confident assertion (I say: she is very beautiful) with an indication of a subjective basis of this assertion ('I say this because I want to say what I think') and by the absence of the disclaimer 'I don't say I know it'.
7.2. Mental verbs The illocutionary force of an utterance can be identified quite clearly by a 'mental' verb used in it (present tense, first person singular), which identifies the speaker's mental state associated with that utterance. (Cf. Urmson 1963; Ross 1973.) A sentence introduced by a frame such as I think, I hope, I fear, I reckon, I suppose, I gather, or I remember, without a complementiser, makes it quite clear that the speaker is not informing the addressee, not announcing anything, and not even telling the addressee anything, but that s/he is rather expressing his/ her thoughts. You have a young lady, I understand. (GP) I gather Donald has a girl too, is that right? (GP) Your wife's working, I gather. (GP)
Thus, the illocutionary purpose of an utterance introduced in a firstperson mental frame (without a complementiser) seems to be: I say this because I want to say what I think Given this illocutionary purpose, it is perhaps not necessary to include in the explication an explicit disclaimer: 'I say this not because I want other people to think this'. If the speaker shows clearly that what s/he wants is to express his/her thoughts, then it is more or less implied that s/he is not motivated by a desire to influence other people. It seems, however, that a mental illocutionary clause serves as a disclaimer in a different sense, and this other disclaiming function should perhaps be portrayed explicitly in the semantic formula. A person who says:
Illocutionary forces of grammatical and other categories
239
I believe he's quite a decent fellow. I hope she comes.
indicates that s/he does not want to commit himself/herself to what s/he says. S/he indicates that s/he is not prepared to say (to 'assert') the proposition in question. For this reason, if an utterance such as He's quite a decent fellow can be represented as: I say: he's quite a decent fellow and is compatible with the illocutionary purpose: I say this because I want you to know it an utterance such as: I believe he's quite a decent fellow should, I think, be represented as: I say: I believe this [he's quite a decent fellow] I don't want to say that I know this I say this because I want to say what I think The sentence in the I believe frame would not be reported as, for example: She claimed that he was quite a decent fellow.
It could only be reported as: She said she thought (believed) he was quite a decent fellow.
This mode of reporting confirms the supposition that in sentences in the frame I believe, the complement of I believe is not embedded directly under I say, and that I believe is so embedded. I offer here, for comparison, explications of the I fear and I gather frames: S, I fear (e.g. Too late now, Don, I fear) I say: I think this [it is too late now] this is bad I don't want to say I know this I say this because I want to say what I think S, I gather (e.g. His wife's working, I gather) I say: I think this [his wife is working] I don't want to say I know this
240
The semantics of illocutionary forces
I think this because of what I hear/see I say this because I want to say what I think (In the last example I have replaced you and your with he and his, because you frames have their own peculiarities, which have to be discussed separately.)
7.3. Particles and conjunctions As the ancient and medieval theorists of language well realised, the illocutionary force of an utterance is often signalled by particles, conjunctions and interjections. For example, the Greek particle hos 'how' was regarded by the Stoics as a marker of what they called thaumastikon, the lekton of admiration or astonishment (see Nuchelmans 1973:63; see also note 4), and the Latin particle utinam 'would it' was described by Paul of Venice as nota optandi, the mark of a wish (Nuchelmans 1973:48). In a similar vein, Austin suggested that "we may use the particle 'still' with the force of 'I insist that'; we use 'therefore' with the force of 'I conclude that'; we use 'although' with the force 'I concede that'." (Austin 1962:75). Clearly, there is also a close relationship between yes and I agree, between no and I disagree, or between and and I add. Nonetheless, no particle or conjunction has exactly the same meaning as a performative verb. In fact, particles and conjunctions usually specify only some parts of the illocutionary force, whereas a performative verb specifies all of it. Consequently, particles and conjunctions, in contrast to performative verbs, can be combined with a fairly wide range of other devices partially specifying the illocutionary force. Thus, one can say: So you've just landed? (AC) It'll get him all right, then? (AC)
using so or then in the context of a question, but one could hardly use a phrase such as I infer or I conclude in this way. In this section I will discuss these two words, so and then, in some detail, to show how the proposed method of analysis is applied to discourse particles and conjunctions of this kind. (Cf. also Chapter 9 below on particles.) Halliday - Hasan (1976:257) suggest that when so and then are used as causal conjunctions they indicate "the speaker's reasoning process: 'I
Il.locutionary forces of grammatical and other categories
241
conclude from what you say (or other evidence)' - compare expressions such as I gather." I think that this explication is basically correct, but that it has to be refined. For one thing, so is not identical in meaning with then, so two slightly different formulae are needed rather than one. For another, the verb conclude is too intellectual to portray the impact of so. Concluding means not so much saying something on the basis of the evidence available, as saying something on the basis of an examination of the evidence. For this reason, concluding is based on arguments, on reasoning, not directly on evidence. But so refers directly to the evidence perceived. For example, in the first scene of 'The night of the iguana' (Williams 1961) one of the protagonists, Shannon, is told by another, Maxine, that the latter's husband, Fred, is dead. A few moments later Shannon resumes the topic, saying: So Fred is dead?
Clearly, Shannon is not concluding that Fred is dead, because she has been informed of it. Her attitude can't be portrayed, therefore, as: 'I conclude from what you say that Fred is dead'. It can be portrayed, however, as follows: I think I can now say I know 'Fred is dead' I think this because of what you say If so is combined with the interrogative intonation, as in the example under discussion, another component is signalled in addition: I want you to say something more about it and possibly even one more: I want us to talk about it (i.e. to say some things about it to one another) Thus, a full explication could perhaps read as follows: So S? (e.g. So Fred is dead?) I think I can now say I know S because of what you say I want you to say that I can say it I want you to say something more about it I want us to say some things about it to one another
242
The semantics of illocutionary forces
Then differs from so in a number of respects. Unlike so, it doesn't refer to the immediate, audible evidence. For this reason, it can be used in more hypothetical contexts, where so is out of place, as in the following examples repeated here after Halliday - Hasan (1976:258): 'And what does it live on?' 'Weak tea with cream in it.' A new difficulty came into Alice's head. 'Supposing it couldn't find any?' she suggested. 'Then [*so] it would die, of course.' Then can also be used in making a judgement, i.e. in assessing a situation. 'Have some wine', the March Hare said in an encouraging tone. Alice looked all around the table, but there was nothing on it but tea. '[ don't see any wine' , she remarked. 'There isn't any', said the March Hare. 'Then [?so] it wasn't very civil of you to offer it', said Alice angrily.
All these differences in usage between the causal so and the causal then can be predicted, I think, from the following two explications: so: I think I can now say I know this because of what you say then: I say this because of what you said now So, unlike then, is indeed closely related to the expression [ gather, and it wouldn't make much sense to use [ gather in expressing an evaluation:
?[ gather that it wasn't very civil of you to offer it. Both so and [ gather indicate that the speaker is trying to say a true sentence on the basis of what s/he perceives; by contrast, then indicates that the speaker is saying something because of what has been said before. But the semantic differences between so and then can't be equated with the differences between gather, conclude, infer, or other illocutionary verbs. For example, gather, though closer to so than it is to then, is less confident and less directly linked to evidence than so is. Having heard a few minutes earlier, from Maxine, that Fred is dead, one couldn't say to Maxine: [ gather that Fred is dead.
Illocutionary forces of grammatical and other categories
243
Each verb, and each particle, has to be described separately, and though many components recur, each verb and each particle constitutes a unique configuration of components. (See Goddard 1979; Wierzbicka 1976.)
7.4. Interjections Interjections express a feeling or a 'want' on the speaker's part. They have, therefore, their own illocutionary force, which can be described in terms of components such as 'I feel X', 'I want Y'. Since, however, they typically combine with other utterances into larger wholes, and since their illocutionary force must be compatible with that of the co-utterance, they often serve as important clues identifying the illocutionary force of the combined utterance as a whole. (A more comprehensive study of interjections is contained in Chapter 8 below.) For example, in the utterance: Ah, my God, are they still in the bus? (NI)
the part are they still in the bus? will not be interpreted as a pure question, despite its interrogative structure, because the meaning of the interjection Ah (or Oh) my God ensures that what follows is to be interpreted as a sudden realisation: Oh my God, Sf 1 think 1 now know something bad is happening (S) 1 wouldn't have thought this would happen 1 feel something bad because of this 1 say this in this way because of this
The related interjection for God's sake encodes a slightly different attitude: it implies that although something bad is going on, the speaker intends to stop it; and that as s/he is not in control of the situation, s/he will attempt to stop it via an action of the addressee. Hence the coutterance will probably include an imperative, and, moreover, a bare imperative, unembellished by any interrogative devices, such as tag questions, why-don't-you's, etc. The illocutionary force of an utterance in the For God's sake frame can be spelt out as follows: For God's sake, Sf 1 see something bad in happening
244
The semantics of illocutionary forces
I I I I I I
feel something bad because of this don't want this to happen want you to do something want you to do it now say this in this way because I can't do anything else think that you have to do it because of this
Using a more conventional (but less precise) language, one might say that for God's sake signals a negative judgment, negative feelings, frustration, helplessness, a sense of urgency, an appeal to the addressee, a feeling of not being in control, combined with a desire to obtain one's goal no matter what, a sense of desperation. As a final example, let us consider the interjection gee. Often described in dictionaries as an interjection of surprise, gee has in fact a more complex meaning; and 'surprise' is sometimes an inappropriate gloss, as well as an inaccurate one: Gee, Gee, Gee, Gee,
that's a nice dress! I was scared for a moment I killed you! (AC) you look like you had it! (NI) wasn't I sick of it - and of them! (AC)
The first example expresses appreciation or admiration. In the second the speaker re-lives a recently experienced fear. The third expresses a mild pity. In the last, the speaker re-lives a past disgust. It is not surprise, then, which constitutes the semantic invariant of gee. The real invariant, however, is related to surprise in so far as it combines an emotional component with the idea of unexpectedness: Gee! I am thinking about X I feel something because of this I wouldn't have thought I would see (feel) this
It is easy to check that this set of components does fit all the examples of utterances with gee cited above. Generally speaking, it seems that while many interjections and other illocutionary devices encode an emotion, the nature of this emotion is never very specific. If we tried to include in the semantic representation a name of the emotion (such as surprise, anger, irritation, frustration, etc.) we would be overspecifying the emotion conventionally conveyed - just as we would be over-specifying the illocutionary force of a
Illocutionary forces of grammatical and other categories
245
syntactic construction by reconstructing a speech act verb). The level of specificity of emotions encoded in various illocutionary devices doesn't seem to go beyond 'I feel something', or 'I feel something bad/good' or 'I feel something bad/good towards you'. Additional information about the kind of emotion involved is conveyed implicitly, and neither can nor should be specified in the semantic representation. Returning to gee, I would add that besides having its own illocutionary force, it functions also as an important clue to the illocutionary force of the co-utterance. The point is that an utterance framed by gee is not used for informing, stating facts, reporting, reminding, warning, and countless other purposes which can be served by declarative sentences. Gee signals that the force of the co-utterance is this: I say this because I now think this In fact, gee shares this function with many other interjections. For example, the sentence: I left the oven on.
could be reported as 'she informed him that ', 'she confessed that ... ', 'she admitted that ... ', 'she stated that ', and so on. But when framed by Oh, my God, it could not be so reported. Oh my God! I left the oven on!
In this frame, the co-utterance expresses a spontaneous thought: I say this because I now think this
7.5. Fixed expressions In addition to lexical devices such as conjunctions, particles and interjections, and to syntactic constructions, English (and presumably every language) contains a large number of fixed expressions, encoding a variety of illocutionary forces. In fact no sharp line divides these 'fixed expressions' from productive lexical and grammatical resources. For example, expressions such as Oh, my God! or For God's sake! can be described either as 'fixed expressions' or as 'complex interjections', and both descriptions would be equally apt. Similarly, expressions such as Why don't you or How about can be described either as markers of specific syntactic constructions or as
246
The semantics of illocutionary forces
'fixed expressions'. Since all 'fixed expressions' have their 'syntactic fields' (i.e. more or less limited spheres of use), they can be seen as pertaining to syntax as well as to the lexicon. The clearest examples of fixed expressions carrying their own illocutionary force come from the area of greetings and leave taking. Roughly speaking, one can describe the force of expressions such as Good morning or Good afternoon in terms of the verb greet. But no similar verb exists in contemporary English which would correspond to the expressions Good-bye or Good night. The illocutionary force of the latter expressions, then, can be stated only in terms of a set of components. For Good-bye, this can be done as follows: Good-bye I think you and I both know: after now we will not be in the same place we will not be able to say things to one another (any more) because of this I want to say something to you because of this I want to say something of the kind that people say to one another at a time like this I say this because I know I should say something like this Expressions such as Good morning or Good night are of course very frequently used, and it is hard to overlook them in an over-all description of language use. But English, and presumably other languages as well, have countless other expressions, which while less frequently used, are equally clear in their illocutionary force. I will illustrate this claim with two examples: How dare you! and Go (and) jump in the lake! For reasons of space, I won't discuss these expressions, but I will simply propose semantic formulae to represent their illocutionary force. How dare you! I see (perceive) this: you are doing something bad I don't want you to do this I wouldn't have thought you would do it I say: I don't know 'how you can dare' to do this I feel something bad because of this I say this because I want you to know what I think and what I feel because of this
Illocutionary forces of grammatical and other categories
247
Go (and) jump in the lake! I know that you want me to do something I think you know that I don't want to do it I will not do it I think you want me to think: if I don't do it something bad can happen to you I say: I don't want to think about this I say: if something bad happens to you I will not think/feel something bad because of this I say this because I don't want to have to say more about it
I will confine myself to just two comments on these explications. It will be noticed that the surface form of the expression How dare you! incorporates fragments of the proposed underlying form, whereas the expression Go and jump in the lake! does not. This means, that, for example, the lexical item dare occurs in the fixed expression under consideration in the same meaning which it carries elsewhere: the fixed expression adds additional components to the meaning encoded by its constituents, but the usual meaning of those constituents is preserved. On the other hand, the expression Go and jump in the lake! can be said to be more deeply 'idiomatic', because the usual meaning of its lexical constituents is not preserved here; accordingly, words such as jump or lake are not mentioned in the proposed explication at all.
7.6. Intonation It goes without saying that no survey of the illocutionary devices of natural language could be complete if it failed to mention intonation. The present chapter has no ambition of providing a complete survey (a solid book would be necessary for that). Nonetheless, I feel that even here intonation must be at least mentioned. It seems indubitable that intonation plays a fundamental role in the area of illocutionary meanings. Limitations of space (as well as of my competence) preclude any serious discussion of this problem here. I would like, however, to illustrate this role of intonation with a short quote from what seems to me to be a highly competent and illuminating source of information on the subject (Deakin 1981). With respect to 'positive interrogative exclamations', Deakin says:
248
The semantics of illocutionary forces
Utterances like 'is she beautiful' with rhythmically repeated rising-falling pitch glides and lengthening on each syllable (especially the last) are not taken as meaning an inquiry, or even that the addressee is expected to reply. Rhetorical questions like 'is this the way we should be doing things' with 'high' pitch and 'tense' voicing are not taken as inquiries, or as expecting an answer (although the answer 'no' is consistent with them). Similarly, 'can you pass the salt' uttered with low rising intonation is generally taken as a request, not an inquiry. The addressee does not have to reply, unless he does not comply with the request. He can readily comply and say nothing, although it is more usual to reply as well. (Deakin 1981 :57)
It is true that the best authorities on intonation, such as Bolinger (1982), caution against too hasty attempts to link specific intonational features with specific kinds of speech acts. But while Bolinger's strictures are probably fully justified with respect to his particular targets, cautious and careful semantic analysis of intonation of the kind attempted by Deakin seems to me not only justifiable but urgently needed. Just as the illocutionary force conveyed by grammatical and lexical means can't be adequately portrayed in global terms (for example, by means of English speech act verbs), neither can the illocutionary force conveyed by means of intonation. Arguably, however, it can be portrayed by means of more subtle analytical tools. Deakin (1981) goes so far as to postulate specific illocutionary meanings for the basic English tones, and for the falling and rising heads, using for that purpose the natural semantic metalanguage based on semantic primitives. His analysis seems to me illuminating and plausible.
8. Comparing illocutionary forces across languages The method of analysis illustrated in this chapter allows us, I think, to compare illocutionary forces encoded in different constructions in a precise and illuminating way. It allows us also to compare, in a precise and illuminating way, attitudinal and interactional meanings encoded in different languages, and correlated with other cultural and social differences intuitively more obvious perhaps but harder to document. In this section, I will illustrate this point by comparing two illocution-
Comparing illocutionary forces across languages
249
ary devices allowing the speakers to express anger (or something like anger) - one in English, and one in Yiddish. In English - in particular, in Australian English - people often express their anger by means of the transitive subjectless pattern: 'V you!', for example: Damn you! Blast you! Fuck you!
Of course, the pattern in question does not belong to what is called polite conversation. It is not, nonetheless, a marginal and negligible feature of English speech. On the contrary, its role in the universe of English discourse can be seen as significant. Formally, this pattern appears to violate the general rule that the 'deleted' subject of an imperative is 'you'. In this case, 'you' is the object, not, as one would expect, the subject, and this reversal contributes to the expressive force of the expressions in question. The set of verbs which can occur in this pattern is very restricted. For example, one cannot say: *Murder you! *Strangle you!
The underlying generalisation seems to be this: the speaker not only wants someone 'to do something bad to the addresse', but he also wants 'to say something bad', that is, to break some verbal taboo. This can be represented as follows: V you! (a) I feel something bad towards you (b) I want to say something bad because of this (c) I say: I want someone to do something bad to you (d) I don't want to say more
Component (a) shows the speaker's ill feeling towards the addressee, component (b) indicates the impulse to break a taboo, and component (c) spells out the 'bad action' which the addressee .should undergo; component (d) shows that the whole utterance is perceived as short and somehow dismissive. It is very interesting to compare this English pattern with the characteristic Yiddish pattern described by Matisoff (1979:68):
250
The semantics of illocutionary forces
Ver geharget! 'Be (lit. become) murdered! Ver dershtikt! 'Be (lit. become) choked!' Ver geshvoln! 'Become swollen!'
This Yiddish pattern, too, expresses the speaker's anger (ill feelings) towards the addressee, and a desire that something bad should happen to him. However, it seems less 'active' and more 'wishful' (helpless?) than the corresponding English expressions, in so far as it is not restricted to transitive action verbs; the speaker wants something bad to happen to the addressee, but not necessarily something bad to be done to the addressee; and there is no desire, on the speaker's part, to break a verbal taboo. We can represent the meaning involved as follows: Be murdered! Be choked! Be swollen! (a) I feel something bad towards you (b) I want to say something because of this (c) I say: I want something bad to happen to you
It is interesting to note that in Yiddish the pattern in question appears to be restricted to the 'bad feelings' towards the addressee, whereas the English pattern can be extended to the third person as well: Damn him! Stuff him!
In Yiddish, bad feelings towards third person are normally expressed in a somewhat less violent manner, with the auxiliary verb zol 'let/should' rather than with an imperative of the verb vern 'become'. For example (Matisoff 1979:66-68): Platsn zol er! 'May he explode!' A fayer zol im ontsindn! 'Maya fire ignite him!'
The 'milder' character of such third person curses is manifested also in their ability to be used in complex sentences, as parenthetical clauses, or as main clauses, for example (Matisoff 1979:61-71):
Comparing illocutionary forces across languages
251
Mayn man, zol er geshosn vern, hot nekhtn 6ngevorn a sahk gelt. 'My husband, may he be shot, lost a lot of money yesterday.' Mayn shviger, klog veys ir, hot a beyze tsung. 'My mother-in-law, maya lament be known to her, has a wicked tongue.' A shvarts yor oyf ir, a gantsn tog hot zi mir gehakt a tshaynik vegn kleynikaytn! 'A black year on her, all day long she chewed my ear off with trivia. ' OJ, zol im nor az6y rinen fun noz, vi's rint mir fun der kvalpen! 'Oh, may his nose only leak on him the way this fountain-pen leaks on me!'
It is also significant, needless to say, that Yiddish third person curses are often not only long and flowery but also humourous, for example (Matisoff 1979:66, 70): Vos es hot zikh mir gekholemt di nakht un letste nakht, zol zikh 6yslozn tsu dayn kop un layb un lebn! 'May my nightmares of the last two nights let themselves loose onto your head and body and life!' A ziser toyt zol er hobn - a trok mit tsuker zol im iberforn! 'May he have a sweet death - run over by a sugar truck!'
For sentences of this kind (third person curses with zol) I would propose the following explication: my father-in-law, maya disease enter his gums (a) I feel something bad towards this person (b) I want to say something because of this (c) I say: I would want something bad to happen to this person
The distinction between 'I want' and 'I would want' is not without its problems, but it captures well, it seems to me, the difference in the speaker's attitude ('milder' and more tentative in the case of a third person than in the case of the addressee). In English, however, no similar distinction is drawn between Damn you! and Damn him! It is worth noting that the Yiddish curses play a more central role in Yiddish discourse than the English pattern discussed at the outset. The English 'Verb you!' pattern is based on a conscious violation of a
252
The semantics of illocutionary forces
taboo; its role in English discourse is therefore, naturally, restricted. The same can be said about other forms of verbal abuse available in English. Traditional Jewish culture, however, allows much more free expression of 'bad feelings' directed at other people; and the language offers suitable devices for doing so, devices which are more generally available, and which are compatible with a wider range of genres and registers. The use of a language-independent semantic metalanguage based on natural language enables us to describe the illocutionary forces encoded in different languages in considerable detail, and to compare them both intra- and inter-linguistically - in a precise and verifiable manner. Precise and verifiable linguistic comparisons of this kind can constitute, in tum, a reliable basis for cultural comparisons, which in the past were often based on speculations or on ad hoc, unsystematic, observations on linguistic differences.
9. Conclusion I hope to have shown that English and, presumably, any other language, possesses a whole range of devices which convey well defined illocutionary forces. Since these devices are largely language-specific, their force cannot be calculated on the basis of any universal pragmatic maxims, Gricean or non-Gricean. Needless to say, even leaving aside the intonation, the devices discussed in this chapter constitute only a small part of the language's illocutionary resources. They should suffice, however, to establish the point that the alleged enormous indeterminacy of illocutionary forces is largely an illusion, born out of an inadequate analytical model. When instead of trying to squeeze every conceivable utterance into a pigeon-hole created by a speech act verb we analyse the illocutionary force of each utterance into individual components, it emerges that language provides numerous unmistakable illocutionary clues, which enable the listeners - and the linguists - to identify illocutionary forces with considerable precision (on a sub-conscious level in the case of listeners, and on a conscious level in the case of linguists). (For numerous subtle analyses of illocutionary clues in Russian, see Paduceva 1985.)
Conclusion
253
To return to the question of 'indeterminacy', I think the answer must be that when people speak to us we usually can detect what they are doing, i.e. what acts they are performing - or at least we can do it to a much higher degree than has been assumed. Illocutionary forces are not nearly as indeterminate as has been maintained. But the exact level of their determinacy is an empirical matter which has to be established on an empirical basis. A preliminary study of a few samples of dialogue taken from plays, as well as from transcripts of natural conversation, seems to suggest that, generally speaking, we know fairly well what the speakers are doing: the samples examined swarm with clear illocutionary clues, whose force can be spelt out in rigorous semantic formulae. But to show that this is the case a separate study is required. The main purpose of the present chapter consists in proposing a framework within which illocutionary forces can be analysed at all - with precision and without arbitrariness. It has also been a purpose of this chapter to show that the decomposition of illocutionary forces into illocutionary components offers solutions to problems which have led a number of linguists to abandon the performative analysis and which have driven them to a desperate position where they are forced to condemn the bulk of conversational English as 'ungrammatical'. The performative analysis, which met the problems of language use head on and which indeed 'took the bull by the horns' (Bach - Hamish 1982:225), was, in my view, on the right track (and in fact, on the track indicated by the great medieval thinkers such as Peter Abelard or Roger Bacon). It is true that in the form in which it was proposed by generative semanticists this hypothesis encountered insuperable problems. But the problems encountered by the more recent theories, which once again dig a gulf between grammar and language use, are even worse. Decomposition of illocutionary forces allows us to preserve the common sense position and not to have to condemn conversational English as 'ungrammatical'. It allows us to account for language use, and to make sense of grammar in terms of people's communicative needs and illocutionary intentions. It allows us to relate syntactic surface structures to meaning, and to validate claims about meaning with observations about the actual language use. The problem of 'what people do with words' can of course be studied from a number of different points of view, not only from a purely linguistic point of view such as that advocated here. In particular,
254
The semantics of illocutionary forces
conversational analysis of the kind practised by sociologists (see for example Sudnow 1972; Schenkein 1978; Psathas 1979) has, in my view, a great deal to offer in this respect. But when sociologists interpret, for example, English tag questions as 'exit techniques' (cf. Sacks Schegloff - Jefferson 1978:30), their analysis is on an entirely different plane from that proposed in this work. A sociological study of conversation can't represent an alternative to illocutionary linguistics; rather, the two must complement each other. I would also venture to suggest that while the former should be of great interest to the latter, the logic of their relationship is such that the former should build on the latter, rather than vice versa. I think that the hopes of linguists who look to sociology for solutions to linguistic problems (cf. for example Levinson 1983) are bound to be disappointed.
Chapter 7
Italian reduplication: its meaning and its cultural significance
In this chapter, I want to look at the relationship between semantics and pragmatics through the prism of one illocutionary device used in one particular language. The language is Italian, and the device is that of 'syntactic reduplication'. The problem is conveniently 'simple', and so allows a reasonably deep probe within the confines of one short chapter. At the same time, it is sufficiently multifaceted to illustrate a wider range of fundamental theoretical issues such as meaning versus implicature, grammar versus rhetoric, autonomous grammar versus an integrated theory of linguistic communication, the boundary between semantics and pragmatics, iconicity versus arbitrariness, or the relationship between culture (in the broad sense) and linguistic structure.
1. Italian reduplication: preliminary discussion In Italian, it is very common to reduplicate adjectives, adverbs, and adverbial expressions, for, roughly speaking, expressive purposes. I call the device in question reduplication rather than repetition, because I see a functional difference (as well as a similarity) between pauseless expressions such as adagio adagio 'slowly slowly' and expressions such as adagio, adagio where a comma signals the presence of a pause. At the same time, I call this phenomenon 'syntactic reduplication' rather than just 'reduplication', because the process in question operates on words rather than on morphemes, so that one can say, for example, not only adagio adagio, but also adagino adagino (where -in- is a diminutive suffix). Like many other languages, Italian also has some frozen expressions based on reduplication, but I say nothing about these here, since this chapter is concerned only with reduplication as a productive illocutionary device. Grandgent (1908:32) notes the emergence of what I call syntactic reduplication in Vulgar Latin: "Repetition for intensive
256
Italian reduplication: its meaning and its cultural significance
effect is not uncommon in late writers: Commodian, malum malum ... bene bene, bonis bonis, fortis fortis, malus malus, etc." Typically, reduplications of this kind are translated into English by means of the 'intensifier' very. For example, the Italian expressions below would usually be rendered in English by the expressions in the right-hand column: bella bella duro duro zitto zitto adagio adagio in fretta in fretta
'very 'very 'very 'very 'very
beautiful' hard' quiet(ly)' slowly' hurriedly'
But the possible range of use of Italian reduplication is much wider than the range of use of the word very in English. For example, the expression neri neri ('black black') would probably not be translated into English as very black. Compare, for example, the following sentence: Due occhi, neri neri anch' essi, si fissavano tallora in viso aIle persone, con un'investigazione superba ... (Manzoni 1972:235)
with its two English translations (by different translators): Sometimes she would fix two very dark eyes on another's face with a piercing look of haughty investigation ... (Manzoni 1914, 1:154) A pair of eyes - jet black, too - would sometimes fasten on people's faces with an air of haughty curiosity ... (Manzoni 1968: 116)
The first translator does use the word very, but simultaneously substitutes dark for black; the second translator translates neri neri as jet black. Even more remarkably, however, syntactic reduplication can be used in Italian in contexts where no 'qualities', gradable or otherwise, are spoken of, as in the following examples: ... se no, lascio Ie mie scuse, e me ne vo diritto diritto a casa mia. (Manzoni 1972:578) '... if not, [' II leave my excuses, and go straight off back home.' (Manzoni 1968:323)
Italian reduplication: preliminary discussion
257
Di grazia, ... un po' di luogo, un pochino; appena appena da poter passare. (Manzoni 1972:344) 'Please, gentlemen, ... a little room, just a very little - just enough to let us pass.' (Manzoni 1968:183) se rimaneva Ii' in ginocchio, ancora per qualche momento, quasi quasi gli chiedevo scusa io, che m' abbia ammazzato il fratello. (Manzoni 1972: 119) 'If he'd stayed down on his knees a moment longer, I'd almost have got to the point of asking his forgiveness myself, for having killed my brother for me.' (Manzoni 1968:52) . .. e che mi faccia la carita di venir da noi poverette, subito subito. (Manzoni 1972:95) , ... and would he do us poor folk the kindness of coming to us straight away.' (Manzoni 1968:39)
This shows that whatever the function of Italian reduplication may be, it is not the same as that of 'intensifiers' such as the English word very or its Italian equivalent molto. To say that bella bella means 'very beautiful' or that leggera leggera means 'very light' is not only inaccurate but could also be misleading, as it encourages false predictions. It leads one to predict, for example, that the Italian expressions subito subito, quasi quasi, appena appena or diritto diritto should be as ungrammatical as the English expressions 'very at once', 'very almost', 'very barely' ('very just'), or 'very straight' ('I'll go very straight home') would be. In fact, however, the Italian expressions in question are perfectly grammatical and perfectly felicitous (unlike their alleged English counterparts). Italian grammars usually characterise the function of reduplication (raddoppiamento) as 'intensification' (l' intensificazione). For example, Lepschy - Lepschy (1984:103) write: "L'intensificazione di un aggettivo si puo ottenere, oltre che combinandolo con un avverbio, anche attraverso la ripetizione: una stanza molto piccola 0 piccola piccola." [The intensification of an adjective can also be achieved, in addition to combining it with an adverb, by repetition: 'a very small room', or 'small small'.] But again, one has to say that this characterisation is misleading because it suggests that the 'repetition' is interchangeable with the adverb molto. But one cannot say molto subito 'very at once' as one says molto piccola 'very small', even though one can say both subito subito and piccola piccola.
258
Italian reduplication: its meaning and its cultural significance
One could always say, of course, that the function of the Italian reduplication is that of 'emphasis' - as people often do when speaking of devices whose exact force they are unable to state. But this would hardly have much explanatory value, as 'emphasis' is also invoked as an explanation for a whole range of other devices, in Italian and, it seems, in every other language which has ever been described (heavy stress is used 'for emphasis', particles are used 'for emphasis', repetition is used 'for emphasis', and so on). As I see it, the reduplication illustrated with expressions such as bella bella or subito subito is a characteristically Italian illocutionary device, whose exact function and force is revealed neither by means of rough translation equivalents from other languages (such as the English word very) nor by means of vague and opaque labels such as 'emphasis'. It can only be revealed by means of a semantic formula which would fit all the contexts where the device in question can be used - and only those contexts. (When I say 'characteristically Italian' I do not mean 'uniquely Italian', cf. Triandaphyllidis' (1975:653) account of a similar device in Modem Greek.)
2. Discourse and illocutionary grammar I would like to put forward a general hypothesis to the effect that 'illocutionary grammar' is born in the ethnography of spoken discourse (see Hymes 1962). What I mean is this. Every language has its own set of language-specific illocutionary devices, encoding specific illocutionary meanings. A set of this kind can be called the 'illocutionary grammar' of a given language. In addition, there are universal or near-universal illocutionary devices. It goes without saying that languages differ from one another in their 'illocutionary grammars'; but they also differ from one another in the relative importance they give to this or that universal device, and in the relative frequency with which this or that universal device is used in a given language. For example, it seems likely that in all languages which have the imperative as a grammatical category, imperatives can be repeated 'for emphasis'. Thus, one can say in English: Come in, come in! Run, rabbit, run! Similarly, one can say in Japanese: Kaere, kaere! 'Go away, go
Discourse and illocutionary grammar
259
away!' In Italian, one might say: Parla, parla! 'Speak, speak!' Scappa, scappa! 'Run away, run away!' I think that repetition of this sort, which I shall henceforth call 'clausal repetition', is different in function from the repetition of the entire speech act, usually signalled in writing by the repetition of the same 'final' punctuation mark (a full stop, an exclamation mark, or a question mark) rather than by the use of a comma; for example, sequences such as: All right. All right., Mary! Mary! seem quite different in function from their counterparts which would be transcribed with a comma: All right, all right., Mary, Mary! What I want to suggest in the present context is that in Italian, the use of 'clausal repetition' seems to have a much wider scope, and a much greater importance, than it has in English. To give some substance to this claim, let me point out that in the English translations of Italian novels numerous instances of 'clausal repetition' (of imperatives, as well as other types of constituents) are left out, or modified so that the overall level of direct repetition is reduced. Consider, for example, the following intances of such a change: Be.ne, bene, parleremo. (Manzoni 1972:131) 'Very well, we'll have our talk.' (Manzoni 1968:59) Parla, parla ... (Manzoni 1972:160) 'Go on, speak out ...' (Manzoni 1968:75) Vedra, vedra ... (Manzoni 1972:287) 'He'll see - he'll just see ... ' (Manzoni 1968:148) Era indietro, indietro. (Manzoni 1972:134) 'Behind-hand, very much behind-hand ... ' (Manzoni 1968:60) Ma senta, ma senta 'But listen, do listen
(Manzoni 1972:90) ' (Manzoni 1968:36)
Ma ascolti, ma ascolti, ma ascolti. (Manzoni 1972:134) 'Listen, listen.' (Manzoni 1968:61)
Whatever the exact meaning of the 'clausal repetition' is, it would appear that both the speakers of English and the speakers of Italian feel on occasion a need to convey it (assuming that the meaning conveyed is in both cases similar), but it also seems clear that the speakers of Italian feel this need more often, and that the meaning in question plays a more important role in Italian discourse than it does in English discourse.
260
Italian reduplication: its meaning and its cultural significance
What I would like to suggest is that there may be a link between the enormous role of 'clausal repetition' in Italian discourse and the existence of the illocutionary device of 'syntactic reduplication' in Italian grammar. It seems likely that the pragmatic meanings associated with 'clausal repetition' have led, through wide use, to the emergence of a new grammatical category, a language-specific grammatical device ('syntactic reduplication'). In support of this suggestion, I would adduce the fact that in many cases the same words seem to be among those most frequently used in both repetition and reduplication: Presto, presto! Presto presto! 'Quickly, quickly!' Adagio, adagio! Adagio adagio! 'Slowly, slowly!'
3. The illocutionary force of clausal repetition In English, the repetition of an imperative, or of an adverb which can be interpreted as modifying an imperative, tends to be interpreted as a directive urging the addressee to act immediately. Natural-sounding utterances such as Come in, come in! Stop it, stop it! Wait, wait! Look, look! Quickly, quickly!
all seem to imply a message which can be spelt out as 'I want you to do something NOW'. Long-term directives sound less natural when repeated:
?Look after yourself, look after yourself! ?Write to us, write to us. One could say, of course: Do write to us! You must write to us! Don't forget to write to us!
The illocutionary force of clausal repetition
261
but the repetition:
Write, write! inevitably introduces a note of urgency and the implication: 'do it right away'. Furthermore, expressions which are compatible with a message of urgency without necessarily implying it tend to acquire such implications when repeated. For example, the expression all right can be used to express simple agreement or acceptance; but all right, all right sounds as if the speaker was trying to cut short the interlocutor's speech ('stop saying it; I want you to do it now; I have already agreed, so there is no need for you to go on talking about it'). In Italian, clausal repetition does not imply urgency, and so it is not restricted to contexts where a message of urgency is appropriate. It conveys a desire to influence the addressee ('I want you to do X') but not necessarily to urge the addressee to do something quickly ('I want you to do X now'). For example, these sentences:
In prigione, in prigione! (Manzoni 1972:347) In una grotta, in una grotta; lontano da costoro. (Manzoni 1972:353) convey long term goals, not urgent commands; it is appropriate, therefore, that in the English translations of Manzoni' s novel they have been rendered in a modified form, without the perfect symmetry of the Italian original:
To prison, yes, to prison! (Manzoni 1968:184) A cave, a cave for me! Far from all this rabble. (Manzoni 1968: 188) A fully symmetrical repetition would introduce a note of urgency, absent from the Italian original. But 'clausal repetition' can also be used - in both Italian and English - in contexts where no goals (other than illocutionary ones) are indicated, as for example in the following dialogue:
"Ben arrivato, ben arrivato!" "Ben trovati." "Avete fatto buon viaggio?" "Bonissimo; e voi altri, come state?" "Bene, bene." (Manzoni 1972:407)
262
Italian reduplication: its meaning and its cultural significance
'Welcome, welcome!' 'Well met.' 'Have you had a good journey?' 'Excellent. And how are you all?' 'Fine, fine.' (Manzoni 1968:223)
How do utterances such as 'welcome, welcome' or 'fine, fine' differ in their communicative import from their hypothetical non-iterative counterparts 'welcome' and 'fine'? 1 think one important difference lies in the degree of the speaker's commitment to the utterance. Hearing 'X!' one can always remain unsure whether the speaker really meant what he said, but hearing 'X, X!' one should be able to rest assured that what was said was really meant (or so the speaker implies). For example, when one says Thank you one expresses gratitude, but it is possible to say it coldly, in a way which will signal to the addressee that quite possibly no gratitude was really felt. But if one says Thank you, thank you, one signals to the addressee that one does mean what one says. For this reason, one cannot say it coldly. (One can say it impatiently, dismissing the interlocutor, but not coldly.) Of course when one says it warmly, one can do so insincerely, pretending that one is grateful when in fact one is not, but a cold or hostile tone is incompatible with the semantic component encoded in the repetition as such. Thus, whatever the illocutionary purpose of the first occurrence of the repeated item might be (such as, 'I say it because 1 want you to know it'), the illocutionary purpose of the second occurrence is distinct and perhaps always the same: 1 say it another time because 1 want you to know that 'I mean it' Accordingly, the following semantic representation can be proposed: Come in, come in! Look, look! All right, all right! (a) 1 say: I want X to happen (b) I want it to happen now (c) 1 know: you can think: 1 say this, 1 don't think this (d) 1 want you to know: 1 think this (e) I say this another time because of this (f) I feel something when I say it
The illocutionary force of Italian reduplication
263
Component (b), which expresses something like urgency ('I want it to happen now') is particularly easy to detect in imperative utterances, such as Look, look! or Wait, wait!, where it refers to specific actions of the addressee. It may be somewhat harder to detect in non-imperative utterances, such as All right, all right! or No, no!, in which the speaker may be seeking to influence the addressee's state of mind rather than to prompt the addressee to any external action. But the impatience or eagerness of such utterances shows that a reference to time ('now') is present in them, too. The assurance of something like sincerity ('you may think that I say this, [but that] I don't think this; I want you to know that I think this') is easy to detect in utterances which express good feelings towards the addressee, for example in Welcome, welcome! or Thank you, thank you!, and also in some imperative utterances, such as, for example, Come in, come in! But can one really talk of 'sincerity' in the case of utterances such as Look, look! or Wait, wait! which do not seem to similarly engage the speaker's feelings? Probably not. Nonetheless, in urgent imperative utterances of this kind it is also very important for the speaker to get across the message: 'I mean it'; this can be interpreted, depending on the context, either as 'I feel something (good towards you) - I mean it' or as 'I want you to do it now - I mean it' (i.e. I mean now, not after now). The proposed explication seems to fit all of these cases.
4. The illocutionary force of Italian reduplication What is the difference in communicative import between instances of clausal repetition, such as adagio, adagio 'slowly, slowly' and instances of syntactic reduplication such as adagio adagio? And what is the difference between 'syntactic reduplication' and the 'intensification' marked by words such as molto 'very'? To begin with the first of these questions, I suggest that in both cases (adagio, adagio and adagio adagio) the speaker insists on the truth or validity of what is said. In the case of clausal repetition, the nature of this insistence is unspecified: the speaker may be dismissing possible doubts regarding the sincerity, seriousness, accuracy, or some other aspect of the utterance. In the case of reduplication, however, the nature of this insistence is quite specific: roughly speaking, it regards the
264
Italian reduplication: its meaning and its cultural significance
accuracy of the utterance. In calling someone' s eyes neri neri the speaker insists that these eyes were 'really' black, literally black; that their color was not close to black but exactly black; that no exaggeration is involved. Saying subito subito the speaker insists that in saying 'at once' what is meant is literally 'at once', and not something more or less close to 'at once'; again, the reduplication implies that there is no exaggeration in what is said. Saying 'barely barely' the speaker insists that in saying 'barely' there is no exaggeration. Furthermore, I conjecture that, unlike clausal repetition, reduplication involves a single speech act. For this reason, I have assigned to reduplication a single component in the frame 'I say', and a single illocutionary purpose. In clausal repetition, the speaker performs a particular speech act (such as a request) with its own illocutionary purpose (such as, 'I say it because I want you to do it'), and then repeats the utterance, with a new illocutionary purpose ('I say it another time because 1 want you to know that I mean it'). In reduplication, no similar split seems to occur. The prosodic unity of the reduplicated utterance mirrors, it seems to me, its illocutionary unity, and this unity is reflected in the proposed semantic representation. Turning now to a comparison between the reduplication and the intensification signaled by the word molto 'very', I would first of all reiterate the observation that molto - unlike reduplication - is restricted to gradable qualities. In essence, to say that someone is molto bella 'very beautiful' or molto gentile 'very nice' is to say that the person in question is more beautiful (more nice) than one could imagine on hearing that she was simply beautiful (or nice). In the case of words which don't refer to gradables the notion of 'more X' is simply not applicable. One can't say *molto quasi 'very almost', just as one can't say *piu quasi 'more almost'. What one can do in the case of non-gradables is stress that they are being used responsibly, accurately, strictly. By repeating a word ('XX') the speaker draws attention to that word, and insists on its strict correspondence with what is meant ('I mean X, not something a little different from X'). The fact that reduplicated expressions often cooccur with the adverb proprio 'really, truly' (E proprio bianca bianca 'It is really white-white') is significant in this connection. Of course one could still say that, when applied to gradable qualitative adjectives and adverbs such as leggero 'light' or adagio 'slowly', the reduplication signals a 'high degree' ('very light', 'very slowly'). But there is no need to postulate a separate meaning for this use of redupli-
The illocutionary force of Italian reduplication
265
cation. Rather, we should say that the connotation of 'high degree' is due to an implicature, calculable from the combined effect of the reduplication and the invariant meaning of the base. For example, if adagio adagio means something like 'the word adagio ("slowly") fits this situation perfectly; in saying adagio I am not exaggerating', then it is natural to infer that the process in question was in fact very slow. But the general formula: 'no exaggeration' fits both qualitative and nonqualitative uses of reduplication. There is no need, therefore, to analyse this device as polysemous. Furthermore, even in the case of gradable adjectives, reduplication doesn't always produce an effect similar to that of molto 'very'. For example (as Anna Ravano has pointed out to me), there was some years ago a film entitled Un borghese piccolo piccolo 'A small-small citizen'. The hero was a 'perfectly ordinary' man, who became involved in extraordinary events. Clearly, the expression piccolo piccolo wasn't used to mean 'very small', but to mean 'truly small, truly ordinary'. The stress was not on a high degree of the feature in question, but on the accuracy or validity of the word used. The difference in function between reduplication (bella bella 'beautiful beautiful') and 'intensification' (molto bella 'very beautiful') becomes even clearer when one considers instances in which reduplication is applied to nouns rather than adjectives. For example, Lepschy Lepschy (1984:103) write: "Con i nomi l'intensificazione (0 meglio un'identificazione della qualita autentica) si puo ottenere anche col radoppiamento: caffe caffe, cioe caffe vero e non un surrogato." [With respect to nouns, intensification (or, better, the identification of an authentic quality) can also be achieved by means of 'doubling': caffe caffe 'coffee coffee' is true coffee and not a surrogate.] Further examples of nominal reduplication include brodo brodo 'broth broth', i.e. genuine broth, and lana lana 'wool wool', i.e. genuine wool. The formula offered by Lepschy and Lepschy ('the identification of an authentic quality') is, I think, insightful. But the implication that the device of 'doubling' (reduplication) has two different functions ('intensification' in the case of adjectives and 'identification of a true quality' in the case of nouns) is unfortunate. In fact, the function of the reduplication is the same in both cases. But to capture it, we need a unitary formula different from that used by Lepschy and Lepschy. Moreover, we need a formula which would account not only for nouns such as caffe caffe and for adjectives such as piccola piccola, but also for adverbs and adver.bial expressions where no 'quality' is referred to, such as, for
266
Italian reduplication: its meaning and its cultural significance
example, subito subito 'at once at once' or quasi quasi 'almost almost'. Before proceeding to tentative semantic representations, I would like to draw attention to yet another difference between the two patterns under discussion. I believe that expressions such as bella bella - unlike expressions such as molto bella 'very beautiful' - contain an emotional component, which can perhaps be represented roughly as 'I feel something thinking about it'. For example, when one compares utterances like these:
Venga subito subito. 'Come at once at once.' b Come straight away. b". Come at once - I mean at once!
a. l
•
one is bound to notice the highly expressive, emotional tone of (a), in contrast to (b and (b"). It seems impossible to pronounce (a) with the neutral, detached prosody with which (b') or (b") could be uttered. The fact that the reduplication can be applied to purely descriptive adjectives, such as nero 'black', bianca 'white', piccolo 'small', or fisso 'attentive' doesn't undermine the hypothesis that the pattern contains an emotional component. Examining utterances where expressions such as neri neri, bianca bianca or duro duro are actually used, it is usually easy to detect in the context clear clues to the emotional undertones. For example, when one of the heroes of Manzoni' s novel quoted above undergoes a great spiritual crisis and tosses and turns in his bed, unable to sleep, it is small wonder that it seems to him that his bed has become duro duro 'hard hard' and his blankets, pesanti pesanti 'heavy heavy'. Similarly, it is small wonder that in one of the most dramatic moments of the story, an accidental witness, most intrigued, follows the amazing dialogue fisso fisso 'attentively attentively'. In another dramatic moment, the hero, trying to escape from the police, wants to cross a river in a fisherman's boat, without being noticed by anybody. Naturally, therefore, he addresses the fisherman in a voice which is leggera leggera 'light light', 'soft soft'. Examples can be multiplied. They all show, however, that the reduplication adds an emotional dimension to the utterance. In some cases (such as 'come at once at once') this aspect of meaning can easily be captured by means of the component l
)
I feel something thinking about it (I feel something saying it?)
The illocutionary force of Italian reduplication
267
In others, the identity of the person who 'feels something' is less clear: is it the hero who feels something, or is it the narrator, or both? I am not sure. It seems to me, however, that if we wish to give a unitary account of the use of the reduplication we should perhaps assume that it is the speaker's feeling which is relevant to the pattern. In a story in which the narrator merely empathises with the hero, the hero's emotions may be primary and the narrator's secondary, but I am inclined to think that it is the narrator's emotion which is signalled directly by the reduplication. 1 suggest that the full illocutionary force of the reduplication can be portrayed along the following lines: [Her eyes were] neri neri 'black black'; [He was] ricco ricco 'rich rich' (a) I say: X (her eyes were black; he was rich) (b) 1 know: you can think: I say 'X', I think: 'something like X' (c) I want you to know: I think 'X', not 'something like X' (d) I say this like this because of this (e) I feel something because of this [I want you to come] subito subito 'at once at once' (a) I say: I want X (I want you to come at once) (b) 1 know: you can think: 1 say: 'X', 1 think 'something like X' I want you to know: I think 'X', not 'something like X' (d) I say this like this because of this (e) I feel something because of this Instead of saying 'I say this like this because of this' in component (d) (a phrasing suggested by Jean Harkins) one might consider an alternative phrasing: 'I say this another time because of this'. It is not clear to me, at this stage, which of these phrasings is more adequate. As for the reduplicated nouns, such as caffe caffe 'genuine coffee', we can assign to them virtually the same explication, without, however, the emotional component (e): caffe caffe 'coffee coffee' (a) 1 say: X (coffee)
268
Italian reduplication: its meaning and its cultural significance
(b) I know: you can think: I say 'X', I think 'something like X' (c) I want you to know: I think 'X', not 'something like X' (d) I say this like this because of this And for the 'intensification' marked by molto 'very', I suggested that its pragmatic force could be spelt out along the following lines (see Wierzbicka 1972:86):
E molto
ricco 'He is very rich' (a) I say: he is X (rich) (b) I want to say more than X
I have come to think, however, that 'very' may be a better candidate for the status of a universal semantic primitive than 'more'. If this is correct, then no further explication of words like molto or very would be necessary. In any case, I don't posit for this pattern an emotional component ('I feel something thinking about it'), as I did for the syntactic reduplication, because 'intensifiers' such as molto and very can be used in contexts which exclude emotionally loaded words or expressions: Objectively speaking, she is very beautiful. ?Objectively speaking, she is gorgeous. ?Objectively speaking, she is most attractive. (See 6,7 below.)
5. Clausal repetition as a means of 'intensification' Bolinger (1972:90) has shown that repetition can be used in English as a means of 'intensification', analogous to words such as very or extremely. He adduced, among others, the following examples (which I think should be qualified as gushing exaggerations): That's very, very interesting. They were quite, quite willing to accept. She's a tiny, tiny baby. It was a big, big bear. He's a wonderful, wonderful person. I carefully, carefully put it down.
Clausal repetition as a means of 'intensification'
269
As Bolinger points out, lexical iteration is accompanied here by prosodic intensification (which Bolinger shows by spacing). It seems to me that in sentences of this kind the repetition is, so to speak, intraclausal, and that its function is quite different from what I have called clausal repetition, exemplified by sentences such as: Come in, come in. Thank you, thank you. Fine, fine.
In the case of clausal repetition, the speaker repeats a whole clause (which may of course consist of just one word). In the case of intraclausal repetition, the speaker repeats a word. In the first case, the speaker insists on the validity of what is said. In the second case, the speaker insists that the WORD in question is well-chosen. Intraclausal repetition is of course closer in function to syntactic reduplication than is clausal repetition. Nonetheless, there are some important differences between the two (intraclausal repetition and reduplication). For one thing, intraclausal repetition seems to introduce an additional speech act, analogous, in a sense, to a parenthetical clause: a. That's very, very interesting. b. That's very interesting; I say: very (is the right word here). a. I carefully, carefully put it down. b. I carefully put it down; mark the word: carefully. By contrast, Italian expressions such as neri neri seem to be used in unitary speech acts (as expressions such as jet black are used in English). Secondly, in English, intraclausal repetition seems to be confined mostly to intensifiers (such as very and quite) and to words which have an expressive component built into them. Purely descriptive words don't lend themselves equally well to such use: ?She was a small, small baby. ?It was a large, large bear. *? He's an intelligent, intelligent person.
The expression It's a small, small world (Jane Simpson, p.c.) sounds better than a small, small baby, precisely because the former, in my view, encodes an emotional component, and is usually uttered with a wordly-wise smile or some other expression of feeling. Expressions such as neri neri or in fretta in fretta indicate that there is no similar restriction in Italian.
270
Italian reduplication: its meaning and its cultural significance
Third, English intraclausal repetition is restricted to contexts compatible with the intended meaning of 'intensification'. One can hardly say in English ?/ almost, almost apologised to him.
?Make just, just enough room for us to be able to pass.
Above all, however, it must be stressed that syntactic reduplication is an illocutionary device which Italian has grammaticalised; intraclausal repetition, however, may well be a rhetorical device available (though not necessarily widely utilised) in any language. The exact meaning, and the exact range of use, of syntactic reduplication cannot be calculated on the basis of any universal principles of human behavior (such as Grice's maxims). It is part and parcel of Italian grammar, which doesn't have counterparts in English, Japanese, or even French. Of course it cannot be denied that the basis of this device is largely iconic, and certainly reduplication is known to convey related meanings in very many unrelated languages of the world (see, for example, Moravcsik 1978). But reduplication can also be put to entirely different uses (see, for example, Hayakawa 1985; Wilkins 1984; for a broad survey, see Mel'cuk, in press). The exact force of Italian reduplication is at least partly a matter of language-specific conventions, and so it cannot be regarded as due to 'implicature' rather than meaning.
6. The absolute superlative in Italian and in English Before turning to the cultural significance of 'syntactic reduplication', I think it in order to discuss briefly another grammatical device which is also characteristic of Italian and whose pragmatic force is closely related to that of 'raddoppiamento'. Italian grammarians call this device superlativo assoluto, the 'absolute superlative'. It can be illustrated with forms such as devotissimo 'extremely devoted', bianchissimo 'extremely white', or velocissimo 'extremely fast'. The superlativo assoluto continues, historically, the old Latin superlative. But from a synchronic point of view, it is a separate grammatical category, distinct from the (Italian) superlative; that is, from what Italian grammarians call superlativo relativo:
The absolute superlative in Italian and in English
comparativo: superlativo relativo: superlativo assoluto:
271
bello piu bello il piu bello, bellissimo
As one grammarian (Fochi 1966: 168) puts it: "quando, per esempio, diciamo che 'Giulio e generosissimo' (superlativo assoluto) non consideriamo quanto siano generosi, al suo confronto, Tizio, Caio e via dicendo: osserviamo tale qualita in lui solo, e ci basta affermare che egli la possiede in grado molto alto." [When we say, for example, that Giulio is 'most generous', we are not comparing Giulio' s generosity with that of Tizio, Caio, and so on: we are considering the quality exclusively in him, and all we want is to affirm that he possesses it in a very high degree.] Italian grammars commonly describe the 'absolute superlative' as a device equivalent to reduplication, and some even extend the same label (superlativo assoluto) to both of these devices (see, for example, Fochi 1966:168; Kaczynski 1964:106). In fact, however, there are important differences between the two devices (as well as similarities). For one thing, the absolute superlative in the strict sense of the term is, like molto 'very', restricted to qualities, and to qualities regarded as gradable. One cannot say, for example, subitissimo as one says subito subito 'at once at once'. For another, ·the absolute superlative is not meant to convey accuracy. Normally, it involves a self-evident exaggeration; and this exaggeration is functional, in view of the speaker's emotional attitude. For example, if one describes a drink as una bevanda agrissima 'most bitter' (Fochi 1966: 170) or if one describes an apple as una mela asprissima 'most sour', one is not pretending to be accurate: the very exaggeration serves to highlight the speaker's displeasure. On the other hand, the reduplication does lay a claim to precision; and for this very reason, it is inappropriate in purely emotional contexts, where no descriptive content is conveyed. For example, the form carissimo 'dearest' is used in Italian very frequently; but the hypothetical form carD carD sounds comical. A repetition with a comma intonation: caro, carD sounds all right, of course (cf. in English I want you to meet a dear, dear friend of mine) but a reduplicated carD carD sounds odd. Similarly, forms such as illustrissimo 'most illustrious', or obbligatissimo 'most obliged' are commonplace; but forms such as illustre illustre or obbligato obbligato sound peculiar. This peculiar effect has nothing to do with the sincerity, or otherwise, of the alleged emotion. When the emotion expressed is purely a matter of social convention there is still
272
Italian reduplication: its meaning and its cultural significance
a clash between the purely subjective, non-descriptive content of the adjective and the purported stress on precision conveyed by the reduplication; as a result, forms such as obbligato obbligato or illustre illustre sound as odd as caro caro. To capture the relevant semantic relationships, I would propose a semantic representation along the following lines: [She is] gentillissima 'extremely nice'; [It is] velocissimo 'extremely fast' (a) I say: it is very X (b) no one/nothing could be more X (c) I feel something thinking about it The comparative component (b) could be further explicated as follows (cf. Wierzbicka 1971): no one/nothing could be more X = if one can't say this of this person (thing) one can't say this of anybody (anything) One could say that English, too, possesses a category of an 'absolute superlative', and that it can be illustrated with expressions such as most kind, most generous, most helpful, most unpleasant, and so on. It must be pointed out, however, that in English the use of this category is very restricted. For example, we may be able to call someone 'most kind', but we couldn't call a thing 'most white' or 'most long' (in the absolute superlative sense). But in Italian, the forms such as lunghissimo or bianchissimo are perfectly acceptable. One generalisation which suggests itself is that in English, the pattern 'most Adj.' cannot be applied to purely descriptive adjectives, that it applies only to terms of assessment ('good' or 'bad'). However, to say this is not sufficient, as the following contrasts in acceptability show: He was most helpful. It was most kind of you. She is most attractive. I am most grateful. She has a most attractive personality. It was most effective. It was a most generous offer. It was most disappointing.
?He is most good-looking.
?She is most beautiful. ? ?She is most healthy. ?He was most pleased. ??She has most regular features. ?It was most elegant. ??It was a most silly play. ??It was most bad.
The absolute superlative in Italian and in English
It was most frightening. It was most ingenious.
273
? ?It was most deafening. ? ?He is most modest.
Studying contrasts of this kind, I am led to believe that the pattern 'most Adj.' refers to an effect that people, and human acts, have on other people. In fact, even combinations of the word most with 'human propensity' adjectives such as kind, generous, or helpful (see Dixon 1977) sound most natural (in the relative, not the absolute, superlative sense!) in sentences referring to specific acts of kindness or generosity, rather than in abstract 'character references'. In the two columns below, sentences on the left sound, I think, more felicitous than those on the right: It was most kind of you. He was most generous (to us, in his dealings with us). He was most unpleasant (to us).
John is most kind. Mary is most generous. Max is most unpleasant.
If the adjective in the absolute superlative construction refers to a person rather than to an act, it sounds best in the construction which Bolinger (1977:141-142) calls 'the ergative of (it was Adj. of you/him), which implies an evaluation of a person in relation to some specific act performed by him or her. However, expressions such as most attractive, which don't imply actions, demonstrate that interaction between people is not a sine qua non of this construction; an effect of a person on other people is sufficient for the pattern to be able to be used. Near-synonyms such as attractive and beautiful or good-looking are particularly instructive in this respect: a person's good looks may leave other people cold; but a person's attractiveness, by definition, cannot. Similarly, it is significant that most grateful sounds much better than most pleased - presumably, because the word grateful refers to human interaction and implies an emotional reaction to another person, whereas the word pleased has no such implications. However, one cannot neatly divide the set of English adjectives into those which always can, and those which never can, occur in the absolute superlative construction. The real constraint is semantic, not lexical: it is the meaning of the entire sentence, not just of the adjective as such, which determines the degree of the sentence's acceptability. For example, of the two sentences below: a. She is most beautiful. b. He gave me a most beautiful necklace.
274
Italian reduplication: its meaning and its cultural significance
(b) is more acceptable than (a), presumably because it refers to human interaction and implies an emotional reaction to it (gratitude). Quirk et al. (1972:287) assert that a sentence such as She is most beautiful "can only mean she is extremely beautiful and not that she is more beautiful than all the others". But if an expression such as most beautiful really didn't mean more than the expression extremely beautiful then we would have no explanation for the gushing, exaggerated character of the former, in contrast to the latter. I think that Jespersen's observation that expressions such as most beautiful are exaggerated should not be lost sight of. And we do lose sight of it if we say that most means no more than extremely. It is not an accident that the absolute superlative shares, to some extent, the morphology of the relative (comparative) superlative. A comparison is there, in both cases. The differences lie elsewhere: in the nature of the comparison, and in the illocutionary purpose of the utterance. If I say: Mary is the most attractive girl in her class, I am comparing Mary to a specific, finite set; and I am informing the addressee that I set her above all the other members of that set. By contrast, if I say: Mary is most attractive, I am comparing Mary to an imaginary set of competitors; and if I am setting her above those imaginary competitors it is not in order to inform the addressee ('I say it because I want you to know it') but in order to express my emotional reaction to my perception of Mary ('I feel something because of it'). Quirk et al. (1972:287) also say that "absolute most is restricted as to the adjectives with which it occurs, perhaps premodifying only those expressing subjective rather than objective attitudes ... : She is most unhappy; *She is most tall." I think that the basic point is correct, but that the way the restriction is formulated is clearly not sufficient. I presume that what the authors really have in mind is not so much a contrast between 'subjective attitudes' and 'objective attitudes' as a contrast between attitudes and objective characteristics (clearly, being tall is not an 'objective attitude'). But the statement they offer does not explain why there is a difference in acceptability between, say, unhappy and sad, or unhappy and pleased, or unhappy and happy: She is most unhappy. ?She is most angry. *She is most happy.
I think the clue to these differences in acceptability lies partly in the emotional effect that a person's 'subjective attitude' exerts on other
The absolute superlative in Italian and in English
275
people. Seeing a person who is obviously very angry we may fail to 'feel something because of that'; and another person's extreme happiness is, unfortunately, unlikely to echo in the hearts of the bystanders; but seeing someone who is obviously very unhappy, we are less likely to remain totally indifferent. It seems to me that the reason why the range of use of the English 'absolute superlative' is so much more restricted than that of its Italian counterpart is that the meaning encoded in the English construction is much more specific than that encoded in its Italian counterpart. X-issimo seems to imply, 'it is more X than anything or anyone that one could imagine', but X-est or most X seem to imply more: 'it is more X than anyone or anything that one could imagine', and also 'this is good/bad for someone'. One might say, then, that the Italian absolute superlative constitutes a grammatical device which enables the speakers of Italian to perform a kind of expressive overstatement any time, that is to say regardless of the nature of the qualities spoken of, whereas English is much more restrictive in this respect. Another, related, difference concerns the nature of the emotional component embodied by both the Italian and the English construction. In English, what is involved is, typically, an emotional reaction to a human act, an act which can be evaluated. (Another English quasi-superlative construction, which doesn't take the definite article, seems to be restricted to the expression of emotions: My deepest sympathy, With best wishes, With warmest regards.) If it is not an act, then it is a perception, followed by an assessment and combined with an emotional reaction. In Italian, the nature of the emotional component is not similarly restricted. Forms such as velocissimo 'extremely fast' or nuovissimo 'extremely new', 'completely new' can be used in abstract descriptions of objects, where there is no question of a direct reaction to an act, or even to a perception. The difference in question is subtle, but I think we should at least try to capture it in the semantic representations of the two constructions, to account as accurately as possible for the differences in the range of their use. I would propose the following way of portraying this difference: You are most generous; I am most grateful; She is most attractive => I feel something because of this (i.e. because of what you have done, because of the way she looks, etc.)
276
Italian reduplication: its meaning and its cultural significance
E nuovissimo (velocissimo, bianchissimo, etc.) => I feel something thinking about it (i.e. thinking of how new it is, etc.) The two full explications would then compare as follows:
E
velocissimo/bianchissimo (It is 'most fast', 'most white') (a) I say: it is very X (b) nothing could be more X (i.e. if one cannot say 'X' of this one cannot say it of anything) (c) I feel something thinking about it
It is most ingenious/unpleasant. (a) I say: it is very X (b) nothing could be more X (i.e. if one cannot say 'X' of this one cannot say it of anything) (c) I think: this is good/bad for someone (d) I feel something because of this
7. Illocutionary grammar and cultural style Different cultures favour different styles of social interaction, and illocutionary grammars tend to reflect cultural differences of this kind. For example, the fact that in Anglo-Saxon culture respect for a person's autonomy occupies a high position in the hierarchy of values is reflected in the great importance given in this culture to pragmatic values such as 'tact', 'non-interference' and 'anti-dogmaticism'; this, in turn, is reflected in the exuberant growth of interrogative and quasi-interrogative devices in the grammar of English (see Goody 1978; Chapter 2 above). The importance of the illocutionary strategy of understatement in English speech (see Hubler 1983) also has its obvious roots in AngloSaxon culture. The fact that in English, understatement can be used even in those situations where the speaker wishes to speak in the strongest possible terms, is highly significant in this respect. For example, one can say in English that a crime was 'rather horrible', or that a performance was 'pretty awful', or that a student is 'fairly enthusiastic' - all expres-
Illocutionary grammar and cultural style
277
sions which cannot be literally translated into languages such as Polish or Italian, whose speakers prefer emphatic overstatement to cautious understatement (see Chapter 2). For example, expressions such as abbastanza orribile 'rather horrible' or piuttosto orrendo 'pretty awful' sound rather ludicrous in Italian. Of course one cannot identify Anglo-Saxon culture with the English language. There are many cultural divisions within the English-speaking world, and English is spoken today as a first language by many groups who do not belong to the Anglo-Saxon cultural tradition. But a language reflects past traditions, as well as the living culture. Since the Anglo-Saxon tradition is, or has been for a long time, the dominant one in English-speaking communities, it is little wonder that this tradition has left a strong imprint on the English language. But it should also be pointed out that sooner or later, historical and cultural diversity within the English-speaking world does find its reflection in linguistic diversification. For example, the English spoken by American Blacks differs from the mainstream in ways which are culturally revealing (see, for example, Abrahams 1970, 1974; Kochman 1972; Mitchell-Kernan 1971, 1972; see also Chapter 3 above). And Australian English has developed many distinctive features which reflect Australian history, culture, and ethos (see Wierzbicka 1986b; Harkins 1988; Chapter 5 above). I would like to suggest that syntactic reduplication and the absolute superlative belong to a system of illocutionary devices which reflect, jointly, certain characteristic features of Italian culture and, in particular, of the Italian styIe of social interaction. In my view, the absolute superlative constitutes, in a sense, an antithesis of understatement. Discussing expressions such as most kind and most ingenious, Jespersen (1965, 7:395) attributes them to a 'universal tendency to exaggerate'. Leech (1983: 147), too, regards hyperbole as a 'natural tendency of human speech'. However, different cultures differ in the extent to which they encourage, or discourage, this 'natural human tendency'. The contrast between the English and Italian absolute superlative constructions, provides, I think, a good illustration of the more general differences between the two cultural traditions and cultural styles. The device of syntactic reduplication, too, can be regarded as the antithesis of understatement, and as a case of emotional overstatement. But concepts such as 'understatement', 'hyperbole', or 'litotes' have been taken over directly from traditional rhetoric, and, although they are useful as hints and points of orientation, they are nonetheless vague and imprecise, as most such traditional notions have had to be.
278
Italian reduplication: its meaning and its cultural significance
The use of explicit formulae, which force us to be rigorous and explicit, helps us to recognise that expressions such as neri neri or duro duro don't necessarily involve 'exaggeration' in the sense in which expressions such as most ingenious or bellissima do. The absolute superlative can be plausibly interpreted as meaning 'it is more X than one could imagine'. The speaker is fully aware of saying 'a little more than what is strictly true', and is not trying to deceive the addressee on that score. The speaker assumes that it will be clear to the addressee that 'a little more than what is true' is said in order to convey an emotional attitude to the state of affairs spoken of. In the case of syntactic reduplication, the speaker is doing something different. This time, he does insist that what he says is not even a little different from what is true. And yet this strategy, so different from that of emotional exaggeration, nonetheless has a similar pragmatic effect (that of 'overstatement', or 'anti-understatement'). I think that in order to understand this apparent paradox, it is useful to compare expressions such as bella bella not only with expressions such as bellissima but also with English expressions such as 'rather pretty'. The person who says (see also Leech's 1983:148 examples):
She's rather pretty. We're rather proud of it. Actually, I'm rather good at it. is trying hard not to say 'very pretty', 'very proud', or 'very good' and clearly conveys this message to the addressee ('I don't want to say more than 'pretty', 'proud', 'good'.') The addressee can infer that the speaker thinks more than he says ('he says that he doesn't want to say more than X - presumably, he thinks more than X'). In this sense, the English strategy in question can indeed be interpreted as an understatement. It is worth noting also that utterances of this kind are often introduced with an apology: I don't want to boast, but .... The speaker who says bellissima or velocissimo does say more than what the bare adjective would convey, and clearly conveys his intention of doing so ('I want to say more than X'). This, however, is not different from the effect conveyed by 'intensifiers' such as molto or very. The real 'overstatement' comes, I think, in the implicit comparison: 'more than 'one could imagine'. The person who says bella bella doesn't imply 'more than beautiful', let alone 'more beautiful than one could imagine'. But his behaviour, too, is (though in a different sense) the opposite of that of somebody who
Illocutionary grammar and cultural style
279
says 'rather pretty'. The person who says bella bella insists on the absolute validity of his words; the person who says 'rather pretty' deliberately refrains from doing so. The first one emphatically dots the i, confident of the absolute validity of what he says and ready to assert himself to the full, without 'tactfully' anticipating the possibility of different views. The person who says 'rather pretty' doesn't want to dot the i, doesn't want to insist on the absolute validity of what he says, and does wish to leave room for other points of view. Of course Anglo-Saxons, too, can and do 'overstate', when they wish to do so - both in the sense of saying more than they really mean, and in the sense of insisting on the absolute validity, and the absolute accuracy, of what they are saying (see Jespersen 1965,7:395; Sapir 1949:145; Leech 1983:145; Brown - Levinson 1978:224). But the fact that, unlike Italian, English has no grammatical devices for doing so suggests the existence of a cultural difference - presumably, the same cultural difference which is also reflected in the different scopes of the English and Italian absolute superlatives. I would add that the emotional nature of expressions such as bellissima or bella bella seems as significant, from a cultural point of view, as their 'emphatic' and 'overstated' character. As I have argued in Chapter 2, there is a link between the rich system of expressive derivation (diminutives, augmentatives, and the like) in languages such as the South Romance or Slavic ones, and the uninhibited display of emotions characteristic of Mediterranean and Slavic cultures. There is also, I believe, a link between the virtual absence of expressive derivation in English and the taboo on overt displays of emotion in Anglo-Saxon culture. In English, utterances like You're too, too kind, She gave me the most beautiful ring, He's a dear, dear man (Jane Simpson, p.c.) are associated with a speech style ascribed to rich women, private-school girls, homosexuals, actors, etc., who are popularly supposed to engage in public displays of emotion, especially affection, or hysteria. And these are thought of as insincere. The fact that both syntactic reduplication and the absolute superlative, so characteristic of Italian, embody an emotional component, is, I think, another manifestation of the same cultural difference. But while an uninhibited display of emotions paralleled by a rich system of expressive derivation is as characteristic of, say, Russian or Polish as it is of Italian (in fact, more so), devices such as syntactic reduplication don't have any counterpart in Slavic languages - certainly not on the same scale. Russian and Polish are at least as 'emotional' as
280
Italian reduplication: its meaning and its cultural significance
Italian; and when compared with English, they must be characterised as favouring 'overstatement' rather than 'understatement' (at least in the expression of opinions). Nonetheless, they don't have, or don't make much use of, pragmatic devices such as syntactic reduplication. As for the absolute superlative, Polish doesn't have it, whereas Russian does (see Mel'cuk, in press). This difference can be seen as related to the fact that derivation of expressive diminutives goes much further in Russian than it does in Polish (for example in Russian, but not in Polish, diminutive suffixes can be added to relative adjectives and adverbs, ~uch as pervyj 'first', pravyj 'right', or casto 'often'; the same holds for many categories of nouns). Furthermore Russian, in contrast to Polish, has at least some expressions analogous to reduplicated expressions in Italian, such as cut' -cut' 'a bit' (lit. 'a bit a bit'), net-net 'rarely and unpredictably' (lit. 'no-no'), vot-vot 'is on the point of' (lit. 'there-there'). Genuine syntactic reduplication is very rarely used in Russian, but it is not impossible, as the following examples demonstrate: No vy ne mozete ze menja seitat' za devocku, za malen' kujumalen' kuju devocku, posle moego pis' ma s takoju glupoju sutkoj! (Dostoevskij 1976:167) 'But you can't consider me as a child, a little girl, after that silly joke!' (Dostoevsky 1974:186) fa dumala on takoj ueenyj, akademik, a on vdrug tak gorjacogorjaco ... (Dostoevskij 1976: 178) 'I thought he was so learned, such a savant, and all of a sudden he behaved so warmly ... ' (Dostoevsky 1974:198) Ona vidimo cego-to stydilas' i, kak vsegda pri etom byvaet, bystro-bystro zagovorila sovsem 0 postoronnem. (Dostoevskij 1976:195) 'She was evidently ashamed of something, and, as people always do in such cases, she began immediately talking of other things.' (Dostoevsky 1974:217).
Can one identify any specifically Italian cultural features which might explain the role of syntactic reduplication in Italian speech? Looking for such features, we might recall the 'theatrical quality' of Italian life (Barzini 1964:73), the "importance of spectacle", "the extraordinary animation ... , the expressive faces, the revealing gesticulation", and "the noise", which are "among everybody's first superficial impressions in Italy, anywhere in Italy, in the north as well as in the south, in big cities
Illocutionary grammar and cultural style
281
as well as in decrepit and miserable hamlets forgotten by history" (Barzini 1964:66). The point about 'noise' is perhaps particularly relevant in the present context. Barzini writes: The noise is usually deafening. People chat, whistle, swear, sing, curse, cry, howl, weep, call to each other and shout, carrying on elaborate discussions or delicate negotiations. Mothers murmur endearing baby words to their little children and ask bystanders to be witnesses to their darlings' charm and pigheadedness. Other mothers call their sons from top-storey windows with voices carrying to the next province. Bells clang with deep bronze notes from the top of the belfry above, drowning every other sound. Somebody is always practising the cornet or the trombone. At times the same popular song or famous operatic aria comes apparently from everywhere, from radios in every shop, from the open windows of apartments, from under the tables in the cafe, from the pockets of clients, from the abdomens of passing housewives. Vespas, cars, motorcycles, trucks go by with roaring engines. The air is in fact filled with so much noise that one must usually talk in a very loud voice to be understood, thereby increasing the total uproar. Lovers sometimes have to whisper 'I love you' to each other in the tones of newsboys selling the afternoon papers. Italians on their death beds, in rooms facing especially noisy squares or streets, are known to have renounced leaving their last wishes and advice to weeping relatives, being too weak to make themselves heard. It is, however, a gay and happy noise, magnified by the stone walls, the absence of greenery, the narrowness of the streets. It goes on from dawn to the small hours of the night, when the last strollers stop under your bedroom window to debate a fine point of politics or the personality of a common friend, both speaking at the same time at the top of their voices. (Barzini 1964:59-60)
What applies to 'noise' applies also to facial expressions and to gestures: What makes all such scenes more intensely fascinating is perhaps the transparency of Italian faces. Conversations can be followed at a distance by merely watching the changing expressions of those taking part in them. You can read joy, sorrow, hope, anger, relief, boredom, despair, love, and disappointment as easily as large-printed words on a wall poster. ... Then there are the gestures. Italian gestures are justly famous. Indeed Italians use them more abundantly, efficiently, and imaginatively than other people. They employ them to emphasise or clarify whatever is said, to suggest words and meanings it is not prudent to express with words, sometimes simply to convey a message at great distance, where the voice could not carry. ...
282
Italian reduplication: its meaning and its cultural significance
A few gestures are as arbitrary and conventional as the deaf and dumb alphabet or the sign language of American Indians. Most of them, however, are based on natural and instinctive movements, common to the majority of men, certainly common to all Western men, elaborated, intensified, stylised, sharpened, made into art. Like all great traditional arts, this one too can generally be understood by the inexperienced at first sight. (Barzini 1964:61-62)
It seems to me that this 'loudness', this animation, this display of Italian life go a long way towards explaining the relevance of a pragmatic device such as syntactic reduplication to Italian culture. Just as Italian gestures, facial expressions, and vocal productions, can be seen as 'exaggerated', hyperbolical, emphatic, and dramatic (as well as emotional), so can Italian pragmatic devices. Sounds, gestures, and facial expressions are 'overdone' - for purposes of self-expression, of theatrical display, and of dramatic effect. (It is also worth recalling, in this connection, the pragmatic principle of 'interest', posited by Leech 1983:146 in connection with overstatement.) Linguistic devices such as the 'absolute superlative' and the syntactic reduplication can be seen as part and parcel of this characteristically Italian cultural styIe.
8. Conclusion It is not my present purpose to discuss in any detail the use of the 'absolute superlative', or any other illocutionary devices related to syntactic reduplication. Rather, I want to reiterate the general point, that illocutionary devices characteristic of a language are not mutually independent, but tend to form networks of 'conspiracies' aimed at common cultural targets. At one time, scholars referred to such conspiracies using old-fashioned labels such as Sprachgeist, the 'spirit of language' (see, for example, Humboldt 1903; Vossler 1904, 1925; Spitzer 1928). When a sociological and anthropological orientation replaced the earlier philosophical and psychological bias, linguists started to talk about 'language as a guide to social reality' (Sapir 1921; Whorf 1956) rather than about 'language as an expression of Volksgeist'. Later, the notion of 'cognitive style' became an acceptable way of referring to more or less the same
Conclusion
283
sorts of phenomena (see Hymes 1961). These days, the preferred conceptual umbrella is that of 'cross-cultural pragmatics' (for example, Pride 1985). I am not suggesting that these changes in ways of speaking are purely superficial, and that they don't reflect any deeper changes in concerns, assumptions, and methodologies. I believe, however, that it is important to see the present concerns in the area of what is now called crosscultural pragmatics in their proper historical perspective: to recognise the continuity of the tradition, and to learn how to learn from past insights as well as from past mistakes; and, above all, to sharpen our methodological tools so that 'cross-cultural pragmatics' will come to represent a real rather than a purely nominal progress with respect to the writings of our predecessors who concerned themselves with similar problems a century or half a century ago. It seems to me that real methodological progress can be achieved by translating the problems of cross-cultural pragmatics into the language of illocutionary semantics. What our predecessors lacked was, above all, methodological rigor and conceptual discipline. They lacked a rigorous analytical framework which would allow them to study both the similarities and the differences between the languages compared (and, for that matter, between related constructions within one language) with a clear sense of purpose and with clear standards of precision. Today, in a post-structuralist and post-Chomskyan era, it is widely felt that new standards of explicitness and rigour (if not of formalisation) are called for, in this area of linguistics as in others. But rather than try to develop and to sharpen new methodological tools which would allow them to meet the required standards, many linguists prefer to abandon the vital questions concerning links between language and culture altogether. Certainly, by avoiding such questions they avoid many errors and blunders which one might commit when one ventures into this 'unsafe' territory. But while preventing themselves from committing many errors and blunders they are also preventing themselves from discussing worthwhile questions and perhaps reaching worthwhile insights. They are, in other words, narrowing the horizons of linguistics and making it less exciting and less relevant to vital human concerns. Certainly, any discussion of the relation between illocutionary grammar and cultural style should be carried out in an extremely careful and cautious way. I suggest that the use of a semantic metalanguage suitable for a standardised description of 'pragmatic' and illocutionary
284
Italian reduplication: its meaning and its cultural significance
meanings can provide a partial answer to the question of how to combine rigour with insight in this tantalising area.
Chapter 8
Interjections across cultures
Pnin sighed in Russian: Och-och-och! (Nabokov 1983:105)
1. Preliminary discussion 1.1. Interjections: phjsis and thesis ('nature' and 'convention') Serge Karcevski (1969 [1941]: 196) starts his pioneering 'Introduction to the study of interjections' with the following quote from a novel by Alexandre Dumas: Aha! - s' ecria-t-il en portugais. 'Aha! - he exclaimed in Portuguese.'
Karcevski notes that this sentence has an unintentional humorous effect, and he attributes this effect to the fact that "one doesn't doubt for a moment that we have to do here with 'natural' language, which doesn't have to be learnt". But in fact, Karcevski points out, "to have the right to think that, one should first engage in systematic investigations of the functioning of interjections in all the remotest corners of the Earth". The need for such systematic investigations is of course undeniable. But even without such investigations we know - as Karcevski was very well aware - that interjections differ considerably from language to language. In fact, far from being universal and 'natural' signs which don't have to be learnt, interjections are often among the most characteristic peculiarities of individual cultures. Rosten (1968:26) makes this point about the Yiddish interjection nu, which, he says, "is the word most frequently used (aside from 'oy' [another interjection] and the articles) in speaking Yiddish". According to Rosten, "Nu is so very Yiddish an interjection that it has become the one word which can identify a Jew. In fact, it is sometimes used just that way, i.e. instead of
286
Interjections across cultures
asking, 'Are you Jewish?', one can say, 'Nu?' (The answer is likely to be 'Nu-nu. ')" This suggests that interjections are based largely on culture-specific conventions, not on any universal 'laws of nature', or at least that some mixture of the two is usually involved. In her charming Body book Claire Rayner writes: Smells travel in the air, and go into your nose when you breathe.... Your brain thinks about the smell. . .. If it is a bad smell your brain makes you move your nose away from it. It tells you to wrinkle your nose, or cover it with your hand. It sometimes makes you try to blow away the air which has the bad smell in it. That's why people say 'Pooh!' to bad smells. It is a blowing-away sort of word. (Rayner 1978:21)
This sounds quite convincing, but it doesn't change the fact that speakers of languages other than English don't say Pooh! when confronted with a bad smell, although they might say something similar. For example, in German they say Pfui! and in Polish, Fu! (See below, section 3.) Clearly, there are some similarities, but there are also differences. Generally speaking, interjections often show remarkable similarities across language and culture boundaries, but these similarities are unpredictable and have to be learnt just as much as the differences. Often, there are no similarities at all. For example, the English interjections gee and wow have no equivalents or near-equivalents in Polish (my own native language), and a Polish immigrant in an English-speaking country has no way of guessing what they mean and how they are to be used. On the other hand, the English interjections ha or ow do have their counterparts in Polish (similar in both form and meaning). But Polish immigrants cannot predict that, and have to learn these interjections as much as they have to learn gee and wow. Usually, dictionaries recognise this unpredictable and conventional character of interjections by including them (or some of them) and trying to define them. But the definitions they offer are not of the kind that could help anyone to learn how to use them. For example, if LDOTEL (1984) says that in English, wow is "used to express strong feeling (e.g. pleasure or surprise)" and that gee is "used to express surprise or enthusiasm"; it gives the reader no clue whatsoever as to how these two interjections differ in use from one another (and what is, for example, the difference between a gee of surprise and a wow of surprise). Similarly, when it says that ah is "used to express delight, relief, regret or contempt" and that phew is "used to express relief, amazement or
Preliminary discussion
287
exhaustion" it gives no indication as to the difference between an ah of relief and a phew of relief. Furthermore, no definitions of the kind provided by LDOTEL (or by other conventional dictionaries) would teach the reader in what situations wow, gee, ah or phew would be inappropriate. And yet there is no way a language learner could guess this, without instruction, or without prolonged immersion in an English speaking environment.
But if interjections do not belong to any 'natural language' and have to be learnt, why is it that they sound comical if reported in direct discourse? The first observation which should be made at this point is that direct discourse in general is somewhat difficult to report in the frame 'X said in language L'. It seems a little odd to say (or to write) the following: ? ?Good morning - she said in Russian. ??How nice to see you - she said in French. ?It is getting late - she whispered to him in German. ?Who is this? - he asked her in Italian.
The reason for this slightly odd effect is that (as I have suggested in Wierzbicka 1974) direct discourse represents something like play-acting; the addressee is invited to imagine that the 'reporter' imitates the original speaker's utterance, and therefore that he reports jointly the what and the how. On the other hand, the phrase 'in language L' indicates that the reported speech separates the what from the how, and that what is being reported is the content but not the manner. Nonetheless it seems clear that some types of utterances are more felicitous in the kind of reported speech under discussion than others. In particular, exclamations involving complex expressions such as Good heavens! are somewhat more felicitous in this context than primary (global) interjections such as Oops!, Ouch! or Wow!; lexically productive exclamations such as How nice! or what a woman! sound a little more felicitous than lexically restricted ones such as good heavens!; and full questions or declarative sentences sound more felicitous than exclamations of any kind. For example, informants tend to agree to the following gradation in acceptability: ?It is time to go - she said in Russian. ??What a woman! - she said in Russian. ?? ?Good heavens! - she said in Russian. ??? ?Wow! - she said in Russian.
288
Interjections across cultures
The reason why any combination of direct discourse with a reference to translating is a little odd, has been suggested earlier. The reason why primary interjections sound particularly odd in such circumstances has to do, I think, with their peculiar semantic structure. The direct discourse suggests that what is being reported is not only the content (the what) of the original utterance but also the how. The phrase 'in language L' indicates that in this particular case the how refers not to the actual words but, presumably, to the illocutionary force of the original utterance. But perhaps primary interjections do not have any illocutionary force which could be separated from their actual form. Haiman (1989:156) points out the deep semiotic difference between utterances such as (a) and (b) below: a. YukI b. I feel disgust. and (endorsing my own analysis in Wierzbicka 1973) he links the first type with BUhler's (1933) Ausdrucksfunktion (expressive function of language), and the second, with BUhler's Darstellungsfunktion (symbolic or representational function of language). The essence of this difference could perhaps be represented along the following lines: 7 YukI = I feel disgusted I feel disgusted! = I say: 1 feel disgusted I say this because I want to say what I feel
Similarly: Ow! (or: Ouch!) I feel pain
=
I feel pain = I say: 1 feel pain I say this because I want to say what I feel
If these partial explications are valid, primary interjections have no illocutionary force at all, because they include neither an 'I say' component nor an illocutionary purpose (an 'I say this because ... ' component). Nonetheless, they do have a meaning. Having no illocutionary components (no components in the frame 'I say ... ') they are
Preliminary discussion
289
not speech acts. Rather, they are what one might call vocal gestures. (Cf. Goffman 1981:78-123.) But they are neither universal nor meaningless. On the contrary, they are language-specific, and they are meaningful. Furthermore, we don't have to be content to hint vaguely at what the meaning of a given interjection might be, as dictionaries usually do, for example: "wow interjection used to express strong feeling (e.g. pleasure or surprise)" or "oops - interjection used to express mild apology, esp. for carelessness" or "ah - interjection used to express surprise, triumph, derision or amused discovery" (LDOTEL 1984). They have a semantic invariant, and this invariant can be revealed and formulated in such a way as to account precisely for their range of use. This can be done without relying on language-specific terms. such as 'disgust' (for yuk), 'impressed' (for wow) or 'pain' (for ouch), which have to be explicated themselves if they are to be meaningfully compared with emotion terms available in other languages, and which in any case don't correspond exactly to the meaning of any interjections. We can capture the subtlest shades of meaning encoded in interjections relying exclusively on universal or near-universal concepts such as 'good' and 'bad', 'do' and 'happen', 'want', 'know', 'say', or 'think'. (Cf. Wierzbicka 1972, 1980, 1985c, 1987, 1988.)8 For example, the English interjections oops and wow can be explicated as follows (cf. Goffman 1981:102 and 108): oops 1 now know: 9 1 did something bad something bad happened because of that 1 didn't want it (to happen) I would not want someone to think that it is very bad (I feel something because of that) wow 1 now know something 1 wouldn't have thought 1 would know it I think: it is very good lO (I wouldn't have thought it could be like that) 1 feel something because of that
290
Interjections across cultures
1.2. Defining the concept of 'interjection' An interjection can be defined as a linguistic sign (1) which can be used on its own, (2) which expresses a specifiable meaning, (3) which does not include other signs (with a specifiable meaning), (4) which is not homophonous with another lexical item that would be perceived as semantically related to it, and (5) which refers to the speaker's current mental state or mental act (for example I feel ... , I want ... , I think ... , I know ... ) . By these criteria, exclamations such as Good Lord!, Good heavens!, Christ! or Hell! are not interjections, whereas those like gee, wow, oops or ha are. (Cf. Goffman' s 1981 :99 definition of what he calls "response cries": "exclamatory interjections which are not full-fledged words".) One class of signs for which the definition proposed above is somewhat problematic includes onomotopoeic signs such as hyc (jump) or b~c (fall) in Polish, which are meant to depict an action in a semi-iconic way and which are used as substitutes for predicates, for example: Kot hyc! - Z okna na ziemi~. (Grodzienska 1970:16) 'The cat hyc (i.e. jumped) - from the window onto the floor.' Jak to slonko ujrzalo, Tak si~ glosno zasmialo, Tak si~ wzi~lo pod boki, Ai b~c! - z chmury wysokiej. 'When the sun saw that, It burst out laughing, And was so delighted, It [fell] - crunch! - from its place in the clouds.'
Iconic 'depictives' such as hyc or b~c are somewhat problematic because in a sense they can occur as constituents of larger constructions and because they are often homophonous with global signs to which they are semantically related. For example, the Polish word hop can be used either as a depictive representing a jump (and stressing, unlike hyc, its vertical aspect), or as a global utterance urging the addressee to jump. If we wish to categorise the urging hop as an interjection, its relationship with the depictive hop presents a problem, in view of condition (4) proposed above. If we lift this condition, however, we will be no longer able to distinguish between 'primary' exclamations such as wow or gee and 'secondary' exclamations such as Christ! or Hell!
Preliminary discussion
291
A possible solution to this difficulty would be to rephrase condition (4) in such a way as to allow for homophony with certain categories of signs (such as, for example, depictives, or particles) but not with others (such as nouns, verbs, or adjectives). Tentatively, we might propose the following: (4) which is not homophonous with another lexical item whose meaning would be included in its own meaning (that is, in the meaning of the putative interjection). If we adopted this phrasing of condition (4), we would have to conclude that exclamations such as Christ!, Hell! or Damn! mayor may not tum out to be interjections, depending on what turns out to be a justifiable semantic formula for a given exclamation. For example, if it can be shown that a semantic formula for the exclamation Hell! should include the noun hell, then this exclamation would not be categorised as an interjection; but if its explication should not include the noun hell, then it would be so categorised. This does not seem to be an unacceptable conclusion.
1.3. Types of interjections If interjections have to refer to the speaker's mental state, or mental act, then they can be classified on the basis of the exact nature of that state or act. Thus, we could establish the following classes of interjections: (1) emotive ones (those which have in their meaning the component 'I feel something'); (2) volitive ones (those which have in their meaning the component 'I want something' and which do not have the component 'I feel something'; (3) cognitive ones (those which have in their meaning the component 'I think something' or 'I know something' and which have neither the emotive component 'I feel something' nor the volitive component 'I want something'. Other classifications of interjections can of course be proposed, based on different criteria. For the present purposes, however, the division of interjections into emotive, volitive and cognitive ones is useful, and the following discussion will be based, essentially, on this division. It must be pointed out, however, that this division cannot be adhered to very strictly because of close semantic links between interjections of different types. In particular, it is often the case that a cognitive interjection has a homophonous emotive one, in which case a comparison of the two meanings served by the same form is clearly called for.
292
Interjections across cultures
2. Volitive interjections In Polish (and probably in many other languages as well), volitive interjections can be divided into two classes of unequal importance: a minor class of interjections directed at animals, and a major class of interjections directed at human beings. I will start with the first. (In the rest of this chapter, interjections are cited without exclamation marks, even though many of them have an obligatory emotive intonation.)
2.1. Interjections directed at animals In Polish, reduplicated expressions such as cip-cip-cip, kici-kici-kici or tas-tas-tas can all be assigned the same basic semantic formula: 'I want you to come here'. The differences between them consist only in the class of addressees: cip-cip-cip kici-kici-kici tas-tas-tas
I want you (chickens) to come here I want you (cat) to come here I want you (rabbits) to come here
There is also at least one interjection (directed at non-humans) with the opposite kind of meaning: 'I want you to go away from here'. The word in question is sio (or a sio), and it corresponds rather closely to the English shoo. Usually sio, like shoo, is directed at flies, or at birds, and its meaning can perhaps be represented as 'I want you to flyaway from here'. Some interjections with astonishingly specific meanings are aimed at horses: 1 want you (horse/s) to move forward (cf. English gee-up) 1 want you (horse/s) to stop hejta 1 want you (horse/s) to move to the right wista I want you (horse/s) to move to the left
wio prr
Poland's hunting traditions are reflected in an interjection aimed at dogs (and, by metaphorical extension, applied also sometimes to people):
huzia
1 want you (dog) to do something bad to this (these) creature(s) (cf. English sic-em)
Volitive interjections
293
Volitive interjections directed at animals are no doubt very common in different languages of the world. For example, Karcevski (1969: 198) mentions that "to stop a horse both the Russians and the Finns shout 'tpru', that is to say, a long bilabial r". Goddard (1987) provides some examples from the Australian Aboriginal language Pitjantjatjara: t}u-t}uu wii
'here boy' (a call to dogs) 'whoa!' (command to horse or command to pull up)
2.2. Interjections directed at people Nearly all items in this class can be assigned the basic meaning: 'I want (don't want) you to do something now'. Each item, however, contains in addition some more specific components. The whole class can be divided (somewhat arbitrarily) into the following subclasses: 1. The 'I want silence' group (sza, pst, cii); 2. The 'I don't want you in this place' group (won, precz, sio, wara);
3. The 'I want you to jump' group (hop, hopla); 4. The 'urging' group (nuie, he), he}ie); 5. The 'communication over distance' group (hop hop, hallo, ahoj);
6. The 'I give it to you' group (na). I will briefly discuss some of these groups, trying to capture the invariant of each interjection and illustrating it with some examples. In some cases, 1 will compare the Polish interjections with their nearest (phonological or semantic) counterparts in English or in Russian.
2.2.1. The 'I want silence' group Polish has at least three different interjections aimed, roughly speaking, at imposing silence: sza [Sn], pst [pst] and cii [tci:]. Sza (probably derived from cisza 'silence', as well as motivated by sound symbolism) can be used, for example, when one has put one's children to bed and when one wants to warn them not to talk any more, whether aloud or in whispers, after one has left the room. There is a
294
Interjections across cultures
well-known saying urging children to be quiet: Cicho, dzieci, cicho sza 'quiet, children, quiet, sza'. (Cf. the English shush.) Pst is somewhat conspiratorial. It could be said at a meeting or in a classroom to someone who is whispering and who doesn't want to be overheard by someone other than the intended addressee. The interjection seeks to warn the speaker that he can be overheard. It could also be used when a secret is being discussed (for example a present which is meant to be a surprise) and when the person approaches who is not suppose'd to hear about it. Cii! (derived no doubt from cicho 'quiet', but also perceived as motivated by sound symbolism) can be said, for example, to one's family, if one is trying to talk on the telephone and they are conducting a conversation, or making some other noise, in the same room. But pst or sza would never be used in such a situation. The crucial differences are these. A person who says either pst or cii tries to be quiet himself, whereas a person who says sza seeks to impose silence on others but not on himself. Thus, one can call or shout sza, but one cannot call or shout pst or cii: Sza! sza! - zawolal na kup~ dzieci b~bnig,cych lyikami w pusty, blaszany rondel. (Reymont) (SIP) 'Sza! sza!, he called at the crowd of children drumming their spoons on an empty tin pan.'
This may be related to the fact that (as noted by Rosten 1968:327 with respect to the homophonous Yiddish shah, no doubt related to the Polish sza) sza is normally directed at a group of people rather than at an individual person. But pst and cii can have, and perhaps usually have, an individual addressee. A person who says cii tries to prevent any noise (it could be whispers, but it could also be footsteps, or some rustling noise); and it is presupposed that some effort directed at not being heard is already being made. A person who says pst tries to prevent audible speech, rather than any other kind of noise; and may be seeking to impose silence on some particular topic (for example a surprise party) rather than absolute silence, as cii does. The different though related meanings of the three interjections in question can be represented as follows: sza I don't want anyone of you to say anything now
Volitive interjections
295
I don't want people to hear you now you know you have to do what I say pst (with a finger on one's mouth) I don't want you to say anything now I think someone other than I could hear you say it I don't want someone other than me to hear you say it I don't want someone other than you to hear me say this
cii I don't want you to say or do anything now that people could hear I think they could hear you now I don't want someone other than you to hear me say this Sza, pronounced fB, resembles of course, both in form and in meaning, the English interjection sh, "often used", according to LDOCE (1978), "in prolonged or reduplicated form, to urge or command silence". As its different form suggests, however, the English sh is 'quieter' than the Polish sza: sza can be said quite loudly itself, whereas the English sh cannot be said very loudly, and seeks to include the speaker in the imposition of silence. The explications of sza and sh can therefore be differentiated as follows: sza => I don't want people to hear you now sh => I don't want people to hear anything here now
The Polish pst brings to mind, of course, the English psst. But here, too, there is a difference in meaning, if not in form. The English psst suggests that the speaker wants to say something to the addressee (unbeknown to other people). For example, in English one might say psst to draw somebody's attention to a beautiful bird, without frightening the bird. But the Polish pst would never be used like that. It could be used to prevent the other person from speaking (in a way that the bird could hear), but not to draw that person's attention to the bird. Accordingly, the explication of the English psst should include one additional component: 'I want to say something to you now': psst (English) I want to say something to you now
296
Interjections across cultures
1 don't want someone other than you to hear it I don't want someone other than you to hear me say this As these explications show, both the English psst and the Polish pst have a 'conspiratorial' quality; but the English psst indicates that the speaker wants to say something (in a private and 'conspiratorial' manner), whereas the Polish pst indicates that the speaker wants to prevent the addressee from saying something that an outsider could hear. A similar interjection (pst, pss or psst) exists also in Russian, where it is used, however, quite differently. The Academy dictionary of Russian (SSRLJa 1950-65) defines it as follows: "it is an onomatopoetic interjection used to express displeasure, disapproval, warning etc." For example: Mne ne ponravilsja glavnokomandujuseij - meloenyj celovek, glupyj ... Siroty - nul' ... Johann Hensen proiznes prezritel' no Psst! (A. Tolstoy) (SSRLJa) 'I was not impressed by the commander-in-chief - a petty man, stupid ... No breadth whatsoever! ... Johann Hensen uttered contemptuously: Psst! '
Clearly, the English and the Polish pst have nothing to do with expressing displeasure or disapproval, and at the level of semantic analysis at which one operates with global terms of this kind the Russian pst must be seen as entirely unrelated to the English or Polish one. If, however, we attempt to capture the invariant of the Russian pst in terms of components formulated in the universal semantic metalanguage we do discover some shared elements of meaning: pst (Russian) 1 think something bad about this I don't want to say now other things about it (I feel something bad thinking about it)
It appears that despite the considerable differences between the Polish, the English, and the Russian pst, they all can be assigned the component 'I don't want (you) to say something'.
2.2.2. The 'I don't want you in this place' group Polish has at least three different interjections (directed at people) with the general meaning 'I don't want you to be here' or 'I don't want you to be in this place': won, precz and wara.
Volitive interjections
297
Won and precz both imply 'bad feelings', but won is also contemptuous ('I think something bad about you, you are like an animal'). Precz can be used as a semi-adverb as well as an interjection (idi precz! 'go away! ') but won is normally used only as an interjection; presumably, its contemptuous character constitutes an obstacle to its use as an adverb. As for precz, it usually conveys hostility (though not contempt), but it can also be used with 'bad feeling' aimed at something other than the addressee, as in Adam Mickiewicz's poem addressed to his lost love: Precz z mej pami~ci! 'Go away from my memory!'
It is inconceivable that won could be substituted here for precz. As for wara, it is related to the verb warczec, 'growl', and it evokes the image of a growling dog, showing a bad feeling and warning the addressee to stay away from a place. Examples for won and wara are: Ryknfl/ basem - Won! - potem jeszcze raz, glosniej - Poszed/ won! (Zukrowski) (SIP) 'He roared: - Won! (Get out of here!), and then once again, louder 'Be gone, Won!' Koziel, kt6ry tam wlasnie przyszedl wody szukac: Ej - krzyknfll z g6ry - Ej, ty ryzy kudla, wara od ir6dla. I hop w d61. (Mickiewicz) (SIP) 'A billy-goat that had come in search of water shouted: Ej, ej, you gingertail, keep away (wara) from the spring. And hop off down the hill.' [Note the use of wara, ej and hop.]
The three interjections may be explicated as follows: precz ('go away') I don't want you to be here I want you to go away from here (now) I feel something bad thinking about it won ('get out!') (cf. English g'wan) I don't want you to be here I want you to go away from here (now) I think something bad about you (you are like an animal) I feel something bad towards you
298
Interjections across cultures
wara (gesturing with a finger) I don't want you to come to this/that place (where X is) I feel something bad thinking about it I will do something bad towards you if you do it
2.2.3. The 'I want you to jump' group Both hop and hopla can be used not only as volitive interjections but also as expressions accompanying a jump and, ostensibly, depicting it in what is perceived as an onomatopoeic manner (as in the example from Adam Mickiewicz's poem, quoted under wara). hop I want you to jump (now) (or: to move quickly like someone who jumps) hopla I want you to jump over this thing (now)
2.2.4. The 'urging' group Nuie is usually used with an imperative or an infinitive (in the function of a command), but it can also be used on its own, or with a form of address. It indicates that the addressee should hurry. For example: Nuie, nuie! Wstawaj, dose spoczywac! (Gruszecki) (SIP) 'Come on, come on! Up you get, that's enough lazing about!' Nuie szelmy, jese mu dae, nim pacierz minie, bo lby pourywam! (Sienkiewicz) (SIP) 'Come on you scoundrels, give him some food in a wink, or I'll pull your heads off.'
Its meaning can be represented as follows: nuie I want you to do it now I don't want you to do it after now Hej is also trying to mobilise the addressee to action, but it doesn't imply that the addressee is (or can be expected to be) slow, and it has an
Volitive interjections
299
additional function of an attention-getter. Thus, trying to urge the addressee to get up quickly, one would say nuie rather than hej; but calling to someone who is some distance away and who doesn't know that he is being addressed, one would say hej! rather than nuie. Some examples: Hej! z drogi! ... wolal na kupig,cych si~ ludzi. (Sieroszewski) (SIP) 'Hej! Off the road! ... he shouted at the gathering of people.' He}! he}! czlowieku! a gdzie to idziecie? Zblfldziliscie - nie t~dy droga. (Zmorski) (SIP) 'Hej! hej! You there! Where are you going? You've taken a wrong turn. That's not the way.' He}, szynkarko, w6dki, w6dki! (Krasicki) (SIP) 'He}, waitress, vodka, vodka!'
The meaning of hej can be represented as follows: hej I want to say something I want you to hear it I want you to do something now He} has also another use as an emotive interjection; and the two uses are closely related, but this point will not be discussed here. Hejie (formally, a combination of hej and the particle ie) urges the addressee to do something quickly, but it adds to this volitive aspect a component of 'merriness' and it implies an invitation to a 'merry action'. It is not clear whether it carries the other ('attention-getting') components of he}. For example: Hejie do roboty, bierzcie iywo mloty. (Lenartowicz) (SIP) 'Hejie let's get to work, grab your hammers.' 'He}ie! w g6r~ czasze! (Zielinski) (SIP) 'He}ie! Drain your glasses!'
The meaning of he}ie may be represented as follows: hejie I want you to do something now I think you will feel something good when you are doing it
300
Interjections across cultures
2.2.5. The 'communication over distance' group In this group, the most interesting item is hop-hop, which is used to call people loudly, over considerable distance (especially in a forest, but not necessarily so). For example: Hop, hop! ... hop! hop! ... - rozlegalo si~ po lesie. Wiele razy powtarzaly si~ te okrzyki bez odpowiedzi. (Skiba) (SJP) 'Hop, hop! ... - there resounded a cheerful call from the direction of the tower.' Hop, hop, panie, ezy tu mieszka pan Grelowicz? (Breza) (SJP) 'Hop, hop, does Mr Grelowicz live here?'
The meaning of the Polish hop hop can be portrayed as follows: hop hop I know that you are far away from me I want you to hear me I want you to say something
What is particularly interesting about this interjection is its relationship to comparable interjections in other languages. Thus, the Russian au [au] appears to have a somewhat different range of use, stressing visibility and contact rather than distance. For example, one can call au when entering a friend's flat or house (if the friend cannot be seen); but one would never say hop hop in this situation. SRJa defines au as "a call that people use in order to find, or not to lose, one another", and SSRLJa defines it as a "reciprocal call used by people who can't see one another". For example: Kogda odna uxodila v sad, to drugaja uze stojala na terrase i, gljadja na derev'ja, oklikala: "au, Zenja!" iii "Mamocka, gde ty?" (Cexov) (SSRLJa) 'When one went off into the wood, the other would always stand on the porch, gaze at the trees and callout: "au, Zenja!" or "Mummy, where are you?'" Masa: fa pojdu poiscu ego. Arkadina: Pozalujsta, milaja. [Masa (idet v levo)}: Au! Konstantin Gavrilovic! ... Au! (Cexov) (SRJa) 'Masa: I'll go and look for him.
Volitive interjections
301
Arkadina: Please do, my dear. [Masa (goes to the door)] : Au! Konstantin Gavrilovic! ... Au!' The combination "Au! Konstantin Gavrilovic!" suggests that one can call au! without assuming that one is very far from the target person. But in Polish, hop hop does imply the assumption that the target person is far away and can only be reached by shouting. Furthermore, au implies that the speaker doesn't know where the addressee is, but this is not necessarily the case with hop hop (cf. the last example of hop hop where the speaker and the addressee can in fact see each other). It is also interesting to compare the Polish hop hop and the Russian au with the Australian English interjection cooee (apparently from an Aboriginal word imitating the sound of a bird), used by people to communicate over distance in the bush. Cooee is similar to au (and different from hop hop) in its stress on visibility, as it could not be used by people who see one another. It is different from au, however, in so far as it implies that both the speaker and the addressee don't know where the other person is (since they are both walking in the bush); whereas in the case of au, the addressee may know where the speaker is (as in the first Chekhov example above). Finally, au could be uttered when entering a friend's house (if the house is large and if one thinks that the friend is somewhere in the house but cannot see him or her); but one could never use cooee or hop hop like that (except in jest). cooee I know we are now in a kind of place where people can't see one another (if they are not in the same part of that place) I can't see you I think you are far away I want to know where you are I want you to know where I am I say this in this way because I want you to hear me au I can't see you I want to know where you are I want you to hear me
302
Interjections across cultures
3. Emotive interjections 3.1. Interjections of 'disgust' and similar feelings Haiman (1989: 157) calls the English interjection yuk "a languagespecific verbalisation ... of a universal gesture of revulsion". He supports his contention that yuk is a "language-specific verbalisation" with a reference to the German interjection pfui and to the Dakota interjection xox. Apparently, however, he sees only the form of these interjections as language-specific. Their meaning is seen in terms of "a universal gesture of revulsion". But is this meaning really universal? Since emotive interjections are less conventional than lexical terms of emotion, it is perhaps natural to expect that their meanings may be less language-specific than terms such as the English anxiety, the German Schadenfreude, the Polish tesknota or the Russian xandra. Nevertheless, it is an illusion, I think, to believe that there is for example 'a universal gesture of revulsion'. For example, Polish does not have an exact equivalent of yuk. On the other hand, it has three other interjections: fu, fe, and tfu, which could be linked, loosely speaking, with something like disgust, but which nonetheless differ from one another and differ from yuk.
3.1.1. The Polishfu and the English yuk The Polish interjection fu has a homophonous counterpart in Russian, but the meaning of these two fus is not the same. Karcevski (1969: 198) glosses the Russian fu as an interjection of "repulsion oifactive", that is, 'olfactory revulsion'. This description (whose adequacy with respect to Russian we will consider later) would fit the Polish interjection fu to some extent, as it is quite likely to be used when one is suddenly confronted with a bad smell. However, it can also be used in other circumstances, as for example when one sees a person licking off somebody else's plate. One might also say fu when one discovers in one's refrigerator some item of decaying food, or when one is invited for the first time to eat snails. On the other hand, one would not say fu if a repulsive creature such as a spitfire caterpillar or some particularly repellent worm landed on one's arm, or when one saw a squashed slug on the footpath - both situations when one could easily say yuk.
Emotive interjections
303
The generalisation seems to be this: one says fu when confronted with a 'disgusting' smell, 'disgusting' food or with 'disgusting' eating habits, that is, when one's nose or mouth comes close to something perceived as 'bad' and when one feels something bad because of that and wants to avoid the offensive contact (or closeness). But if a worm landed on one's nose one would not say fu, presumably because the bad feeling caused by this contact is only accidentally linked with the nose (having a repulsive worm land on one's nose does not seem different from having it land on one's cheek). Fu seems to be restricted to feelings which are necessarily defined in terms of nose or mouth - that is, to situations of introducing something 'repulsive' into one's nose or one's mouth. To account for the range of use of the Polish fu, I would propose (as a first approximation) the following explication: fu (Polish)
I now know something about something in this place I feel something bad in my body because of that I feel like someone who thinks: I don't want this thing to come to be in my nose or in my mouth I think other people would feel the same The component 'I think that other people would feel the same' is intended to capture the speaker's notion that there is something 'objectively' disgusting or repugnant about the situation. The subcomponent 'in my body' is intended to show that this interjection suggests a physical reaction on the part of the speaker, and not, for example, moral, intellectual or aesthetic disgust. In the case of yuk, the prototype is no doubt physical, too. LDOTEL, which describes yuk (yuck) as an "interjection of disgust, probably imitation of the noise of retching", may well be right in this respect. Nonetheless many of my informants (especially younger ones) affirm that they use this word not only to express a physical reaction (to· a physically disgusting or revolting object), but also in the presence of a non-physical stimulus; for example, some students say that they would say yuki when given by their lecturer an assignment which they strongly dislike. II This disagreement among informants may reflect some sociolinguistic variation: the use of yuki is spreading and its range of use appears to be widening. The following quote from The Australian newspaper (25.10.88:14) shows that - at least in Australia - yuk in an abstract (non-physical) sense can now appear in writing and even in
304
Interjections across cultures
print: "FEMINISTS - YUK! As a woman I feel ashamed" (from a letter to the Editor expressing the reader's reaction to new guidelines for gender-inclusive language, called on the same page "Sheila-Speak"). In the case of the Polishfu the presence of a 'bodily component' is indubitable. For example, one could never say fu in response to some new set of guidelines, or to an intellectual task such as a university assignment. 12 Yuk differs in this respect from another (perhaps older) English interjection which is used to express disgust: phew (which may be segmentally homophonous with the phew of relief mentioned earlier but which has a different tone pattern, sometimes represented as pee-yew and the like). The 'phew of disgust' is linked specifically with what Karcevski called 'olfactory revulsion', and - like the German pfui mentioned by Haiman (1989) - it refers clearly to a bad smell. It is nonetheless perceived as semantically similar to yuk. For example, in an advertisement for garbage bags frequently shown on Australian television, a cuckoo jumping out of a clock and seeing a heap of garbage utters consecutively phew and yuk. Tentatively, I would propose for yuk the following explication: yuk I think: this is bad I feel something bad because of that I think other people would feel the same I feel like someone who thinks: I don't want to be in the same place as this
3.1.2. The Russianfu If the Polish fu is similar to the English phew (the phew of disgust) in being clearly related to bodily sensations, although not necessarily olfactory ones, it differs in this respect from the Russian fu, which, as mentioned earlier, was glossed by Karcevski (1969:198) as an interjection of 'olfactory revulsion', but which in fact can express a purely mental (moral, aesthetic etc.) reaction. Consider for example the following passage: ... ej xotelos' pojti v spal' nju Verocki, sobrat'sja k nej pod odejalo i prilaskat' ee ... No ona otgonjala eti uzasnye mysli. Fu! kak uiasno! Kak kakaja-nibud' lesbijanka! Gadost' kakaja! (Suslov 1982: 188)
Emotive interjections
305
' ... she was tempted to go to Verocka's bedroom, to sneak under her covers and to cuddle her ... But she would chase such awful thoughts away, in horror. Fu! How horrible! Like some lesbian! How disgusting!' One might say that in this case the speaker's disgust is almost physical, but it is caused by perceiving 'disgusting' thoughts in one's head, not by perceiving 'disgusting' smells or sights. The Polish fu could never be used in a similar context, and the English yuk would be unlikely to be used in this way. In fact, the Russian fu can also be used in situations which are even further removed from any physical disgust or revulsion. The Academy dictionary of Russian (SSRLJa) glosses fu as an interjection of "contempt, annoyance, revulsion (otvrascenie) etc.", mentioning revulsion as only one of the possibilities, whereas Dal's dictionary (1882) doesn't mention it at all, glossing fu as "an interjection of contempt or annoyance (mezdometie prezrenija, dosady)". The examples cited in these dictionaries refer to human behaviour evaluated as 'stupid', or to irritating events concerning the speaker. For example: Fu, kak vy bestolkovy!
00.
Sejcas ze pozovite eOtix niscix! (Kuprin)
(SSRLJa) 'Fu, how stupid you are! ... Call these beggars in at once!' Fu, kakoe skvernoe pero! - zakrical Sumnov, udariv v dosade im po stolu. (Dostoevskij) (SSRLJa) 'Fu, what an awful pen! - shouted Sumnov, angrily hitting it
against the table.' To account for the non-physical character of the Russian fu, we can perhaps explicate it along the following lines: fu (Russian)
I I I I I
now know/imagine something think: this is bad feel something bad because of that think that other people would feel the same feel like someone who thinks: I don't want this
If we wanted to take into account Karcevski' s suggestion that the Russian however tenuous or metaphorical - with the nose, we could rephrase the last component of this explication as follows: 'I don't want this thing to be near my nose'; this would not
fu does have some link -
306
Interjections across cultures
be incompatible with the examples adduced above because moral or aesthetic 'disgust' could be conceptualised in terms of a physical prototype. It is important, however, not to include in this explication the component 'in the body', which we have assigned earlier to the Polish fu.
3.1.3. The Polishfe Unlike the Polishfu, the Polishfe is similar to the Russianfu in not being restricted to physical sensations, although it differs from it in other respects. Roughly speaking, fe expresses a mild moral disgust. It is always a reaction to human behaviour, and it could never be used, like the Russian fu, with respect to an annoying object (for example a pen). The feeling expressed is 'mild', or rather it is expressed in a 'mild' manner, because the speaker's purpose seems to be didactic rather than purely expressive: by expressing a negative emotional reaction to the addressee's behaviour the speaker is trying to shame the addressee, and to cause him to correct his behaviour. Typically, fe is used by adults speaking to small children, but it can also be used, for example, by a mild-mannered woman flirtatiously scolding a man. SJP glosses fe as "an interjection expressing disgust, displeasure, reprimand", but this somewhat heavy-handed gloss is followed by a much more helpful comment: "it means something similar to 'you should be ashamed of yourself! that's not nice! '''. Some examples: Rzepa takze go zagadnie: Fe! Niedobrze! Fe! Nieladnie! Jak pan moze, Panie pomidorze? (Brzechwa 1983:23) 'The turnip will also reprove him: Fe! That's bad! Fe! That's not nice! How can you, Mr Tomato?' Fe! Jak pan moie mowic takie rzeczy! (Sienkiewicz) (SJP) 'Fe! How can you (polite form) say such things!' Fe, Sewerynku, wstydi si~, zlym bratem jestes. (Bliziilski) (SJP) 'Fe, Jimmy (Sevie), you should be ashamed of yourself, you're a bad brother.'
Emotive interjections
307
In fact, the most common collocation is fe, nietadnie 'fe, that's not nice', as in the following characteristic examples from children's poems: Fe, nie/adnie! Fe, k/arnczucha! (Brzechwa 1983:9) 'Fe, that's not nice! Fe, you're a little liar!' Fe, nieladnie! ktoz tak k/arnie? Zaraz si~ poskarz~ rnarnie! (Brzechwa 1983:9) 'Fe, that's not nice! Who lies like this? I'll tell mum about it!'
Other common collocations involving fe are 'fe! how can you', or 'fe! you should be ashamed of yourself!', as in the examples above. But fe can also occur on its own, as in the following children's poem: Brudasek A ten piesek Brys, co si~ nie chcial kg,pac dzis, rna na lapkach kurz i piasek, wi~c spac pojdzie jak brudasek. Fe! (Szelburg-Zarembina 1970:60)
'Dirty little grub. That little dog Spottie, who didn't want to have a bath today, has sand and dust on his little paws, so he'll go to sleep like a dirty little grub. Fe!' It is interesting to note in passing that there is a similar-sounding interjection of disgust in Danish and Swedish (Anne Dineen and Jean Harkins, p.c.), which has a similarly didactic purpose: fy, and which appears to be used, prototypically, with reference to toilet training. However, while a small child who has soiled himself may well be reprimanded in Danish or Swedish with a disgusted fy, the Polish fe would be much less likely to be used in such an elementary situation, as it appeals to higher-level social and moral rules. One might say that Ie appeals to the addressee's sense of shame, whereasfy seeks to activate a Pavlovian reflex of elementary disgust. fy I now know that you have done something bad
308
Interjections across cultures
1 feel something bad because of that I think that other people would feel the same I want you to feel something bad because of that I don't want you to do this fe
I I I I
now know that you have done something bad feel something thinking about it think that other people would say the same think you know that if someone does something like this it is bad I want you to feel something bad thinking about it
Fe is a somewhat genteel or 'refined' thing to say, which appeals to the addressee's sense of shame and knowledge of social rules, suppressing, as it were, an impulse of disgust. For this reason, I have phrased the feeling itself more vaguely as 'I feel something thinking about it' rather than as 'I feel something bad because of that'. (For further discussion of this point, see section 3.1.6.) On the other hand, I have included in the explication of fe a sentence referring to the addressee's understanding of certain general rules of civilised behaviour ('I think you know that if someone does this it is bad'), which 1 have not included in the definition of the more elementary fy.
3.1.4. The Yiddishfeh In a Russian-Jewish popular song, called 'A Soviet Jew's prayer' or 'A Jewish marseillaise', the Yiddish interjection fe is used as an overt signal of Jewishness: Otrecemsja ot starogo mira ... Nam ne nuzno zlatogo kumira - Fe! 'Let's renounce the old world ... We don't need a golden calf - Fe!' Fe represents here the Soviet Jews' mock-identification with 'Soviet' values and their mock-rejection of, and mock-disgust for, any other values (in particular, capitalist values). According to Rosten (1968:115), "Feh! is the Yiddish replacement for exclamatory expressions of disgust such as 'Phew!', 'Pee-oo!', 'U g h"., 'Ph ooey., I' 'E cc. h" an d 'P~-h I' In sayIng . 'F e., h" you may Jrr.....
Emotive interjections
309
bare the teeth and wrinkle the nose, in visible reinforcement of the meaning". Rosten says that he "once wrote an entire story to illustrate the puissance of this incomparable expletive", and he lists (among others) the following situations "in which Feh! may serve as the perfect utterance": 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
Smelling a rotten egg. Passing an open sewer. Inhaling Los Angeles smog. Driving past the sulfur pits that fringe New York in New Jersey. Whiffing a rotten fish. Depicting a beatnik with mare's-nest hair. Summarising a political position you detest. Appraising the honor or benevolence of an enemy. Contemplating an operation for haemorrhoids. Delineating the character of the paskudnyak who ran off with your wife. Recounting how a soprano murdered an aria. Depicting a hangover.
It seems likely that the Yiddish feh comes from the Polish fe (or possibly the other way around). But it is clear that the range of use of the Yiddishfeh is much wider, and that in fact it is at least as wide as that of the English yuk. It is not, however, as wide as the last point on Rosten's list would suggest: "portraying strongly negative feelings about any sight, event, person, crisis, experience or emotion". As Rosten is well aware, Yiddish has some other interjections expressing "strongly negative feelings", in particular oy vay, for which feh is by no means always substitutable. The crucial difference between these two interjections seems to lie in a more 'disinterested' and more 'descriptive' character of feh. Oy vay implies: 'something bad is happening to me'; feh implies: 'this is bad' without the element of personal involvement 'to me' and without the dynamic element 'is happening'. But feh is not merely lacking in personal involvement: it positively claims that something is bad 'objectively', and that other people, too, would feel something bad because of that. For example, if a brick fell from a roof and hit one on the head, one could not exclaim Feh! To account for this 'public' character of feh I would posit for it, once again, the component 'I think that
310
Interjections across cultures
other people would say the same'. This leads to the following partial explication of feh (cf. section 3.1.6): feh
1 think: this is bad 1 feel something bad because of that 1 think other people would say the same 1 think other people would feel the same In view of the wide range of use of feh, I haven't included in this explication a reference to the speaker's body. It is interesting, however, to note Rosten's observation on the 'wrinkling of the nose' and the 'baring of one's teeth', which he thinks reinforce the meaning of feh. I presume that these gestures indicate a mild version of olfactory and oral repugnance: the speaker behaves as if he was trying to minimise the contact of his nose and his lips with some offensive substance. These gestures of repugnance are not as violent, however, as to suggest an image of a person trying to get rid of some such substance from their nose, mouth or throat. Leaving aside the question of any references to body parts and bodily gestures, the formula proposed above for feh is virtually identical with that proposed earlier for yuk - except for the component 'I think that other people would say the same', which presents feh as being more of a judgement than yuk is, and which links it with the highly judgemental Polish fee Both feh and yuk involve a judgement and a gut reaction, to some extent, but in yuk the latter element appears to be stronger, and it seems more marked in abstract contexts (such as, for example, recounting how a soprano murdered an aria) than in the case of feh.
3.1.5. The Polish tfu and the Russian t'fu Turning now to the Polish interjection tfu and to the Russian interjection t'fu, we must note, first of all, that they both indicate the action of spitting. Karcevski makes this point about the Russian t'fu, which he links with the Russian expression naplevat' (lit. 'to spit'), and which he also compares with the German expression Ich spucke drauf! 'I spit on this'. As pointed out by Karcevski, the Russian t'fu can be used as a substitute for naplevat', and can function as the main verb.
Emotive interjections
Mne to-me
na on
eto this:ACC
311
nap/evat' ! to.spit
=
fa t'fu na eto! I:NOM t'fu on that:ACC 'I spit on this (that's how I feel about it)!'
But in Russian, naplevat' and, consequently, t'fu, means more than an outsider could infer from the information that both these words symbolise the action of spitting: spitting itself is a symbolic action with its own culture-specific meaning. In Russian, naplevat' '(I can) spit (on that)' expresses a contemptuous indifference. 'I don't care', the speaker seems to be saying, adding to this a defiantly coarse touch. T'fu as a substitute for naplevat' can mean the same. For example: A naplevat' mne na nego, on mne nipocem! (SSRLJa) 'I can spit on him, I don't need him!' Mne na eto vase zoloto - t'fu! I ona dejstivitel' no pljunula sebe pod nogi. (Polevoj) (SSRLJa) 'I can t'fu (spit) at that gold of yours! And she actually spat in front of her feet.'
This sense of t'fu (and of naplevat') can be explicated as follows: t'fu (Russian) I don't care about this I feel something bad thinking about it I feel like someone who wants to do this: [spit]
The expression I don't care used in this explication can be further explicated along the following lines: I don't want to think about this I don't want someone to think that I want this The link between the symbolic act of spitting and an attitude of contemptuous indifference is particularly clear in the following passages: - Pojdem, - govorit, - posmotrim tvoi ramy. - U menja, - govorju, - urok dolzen byt'. - Da pljun' ty, - govorit, na urok, raz takoe delo. (Goljavkin 1968:110) "'Let's go", he says, "and have a look at your frames."
312
Interjections across cultures
"I can't", I say, "I am due for a lesson." "Stuff the lesson" [lit. 'spit on the lesson'], he says.' SSRLJa posits also another meaning of t'fu, asserting that "it is also used to express displeasure, annoyance, disappointment, etc." This alleged second meaning is illustrated with examples such as the following one: Zato u madam Bubnovoj ... - T'fu ty so svoej Bubnovoj! Aleksandra Semenovna vybezala v velicajsem negodovanii. (Dostoevskij) (SSRLJa) 'On the other hand, as for Mrs. Bubnova ... "T'fu [I spit] on you and your Mrs. Bubnova!" and Aleksandra Semenovna ran out extremely indignant.'
It appears, however, that examples of this kind fit very well the formula postulated for naplevat' and for the corresponding use of t'fu, and that the postulated polysemy of t'fu is not justified. On the other hand, the Polish tfu is used differently, although it, too, symbolises the action of spitting. Roughly speaking, it expresses not personal indifference (contemptuous indifference), but contempt and moral disgust. SJP cites (among others) the following examples: Tfu! wstyd! warcholilismy si~ okrutnie, a teraz trzeba przed cnotfl i niewinnoscifl swiecic. (Sienkiewicz) (SJP) 'Tfu! shame on us! we behaved like oafs and rascals, and now we have to answer for our behaviour before the virtuous and the innocent! ' Tfu! Mospanie Hrabia, czy Wasc zb6jca? (Mickiewicz) (SJP) 'Tfu! Count, are you a robber?'
The meaning encoded in the Polish tfu appears to be this: tfu (Polish) I think: X did something bad I wouldn't have thought that someone like X could do that I feel something bad thinking about it I think that other people would feel the same (I feel like someone who wants to do this: [spit] )
It is interesting in this connection to recall Darwin's comments on the meaning of spitting: Spitting seems an almost universal sign of contempt or disgust; and spitting obviously represents the rejection of anything offensive from the
Emotive interjections
313
mouth. Shakespeare makes the Duke of Norfolk say, 'I spit at him - call him a slanderous coward and a villain'. So, again, Falstaff says, 'Tell thee what, Hal, - if 1 tell thee a lie, spit in my face'. Leichhardt remarks that the Australians 'interrupted their speeches by spitting, and uttering a noise like pooh! pooh! apparently expressive of their disgust'. And Captain Burton speaks of certain negroes 'spitting with disgust upon the ground'. Captain Speedy informs me that this is likewise the case with the Abyssinians. Mr. Geach says that with the Malays of Malacca the expression of disgust 'answers to spitting from the mouth'; and with the Fuegians, according to Bridges, 'to spit at one is the highest mark of contempt'. (Darwin 1955:260)
The examples adduced, however, show that spitting does not have the same meaning in different cultures. Disgust, contempt, disdain, and moral indignation do indeed have something in common, but the differences between them are no less real than the similarities.
3.1.6. 'Disgust' and bodily gestures Concluding our discussion of selected interjections of 'disgust and similar feelings' we will note that in most cases they appear to be linked with certain physical gestures, and that this may well be the reason why they can be perceived as 'natural' (that is, non-arbitrary). Thus, the Polish tfu and the Russian t'fu are perceived as imitations of an act of spitting (although their meaning is by no means the same). The English yuk can be perceived as an imitation of the sound of retching. The English phew, the German pfui and the Scandinavian fy can be thought to imitate an attempt to breathe out of one's nose a repulsive smell (although again, their meaning is by no means the same). The Polish and Russianfu can be perceived as associated with 'olfactory or oral revulsion', although the use of the Russian fu, in contrast to that of the Polish one, is not restricted to physical sensations. The close phonetic resemblance between fu and the interjections symbolising the act of spitting (tfu and t'fu) supports, I think, the idea that the Polish and Russian fu is associated with an olfactory/oral revulsion, rather than with a purely olfactory one. Despite the different range of use of the interjections under discussion (which of course must be reflected in adequate definitions), one might consider including in their definitions a physical prototype. This could be done as follows:
314
Interjections across cultures
tfu, t'fu ('oral repugnance') => I feel like someone who thinks: I don't want this to be in my mouth I want it to go out of my mouth phew, pfui ('nasal repugnance') =>
I feel like someone who thinks: I don't want this to be in/near my nose I want it to go away from my nose fu, fy ('nasal/oral repugnance') =>
I feel like someone who thinks: I don't want this to be in my nose or in my mouth I want it to go away from my nose/mouth yuk ('pharyngal repugnance') =>
I feel like someone who thinks: I don't want this to be in my throat I want it to go out of my throat feh, fe ('nasal/labial repugnance') =>
I feel like someone who thinks: I don't want my nose or my lips to be in the same place as this It might be added that the Polish fe could also be thought of as a euphemistic version of 'nasal/oral repugnance'. The consonant may be seen as a symbol of something like repugnance whereas the vowel can be perceived in terms of a 'deliberate' vowel change: the speaker wants to express his (or, more typically, her) repugnance in a softened and 'genteel' fashion, that is to say, without really sounding like someone who gives way to their bodily impulses. This could be represented as follows: fe =>
I feel like someone who thinks: I don't want my nose or my lips to be in the same place as this I don't want to say it like this Components referring to specific body parts and bodily gestures would facilitate, I think, an inquiry into the sound symbolism of interjections in a cross-linguistic perspective. In view, however, of the exploratory nature of the present chapter, I don't want to argue strongly for their inclusion in the semantic formulae; rather, I wish at this stage to offer
Emotive interjections
315
them only for the reader's consideration, together with the following quote from Darwin: As the sensation of disgust primarily arises in connection with the act of eating or tasting, it is natural that its expression should consist chiefly in movements round the mouth. But as disgust also causes annoyance, it is generally accompanied by a frown, and often by gestures as if to push away or to guard oneself against the offensive object. ... With respect to the face, moderate disgust is exhibited in various ways; by the mouth being widely opened, as if to let an offensive morsel drop out; by spitting; by blowing out of the protruded lips; or by a sound as of clearing the throat. Such guttural sounds are written ach or ugh; and their utterance is sometimes accompanied by a shudder, the arms being pressed close to the sides and the shoulders raised in the same manner as when horror is experienced. Extreme disgust is expressed by movements round the mouth identical with those preparatory to the act of vomiting. The mouth is opened widely, with the upper lip strongly retracted, which wrinkles the sides of the nose, and with the lower lip protruded and everted as much as possible. This latter movement requires the contraction of the muscles which draw downwards the corners of the mouth. (Darwin 1955:257-258)
3.1.7. 'Disgust' and sound symbolism We have examined a number of interjections expressing, loosely speaking, disgust or similar feelings, and we have seen that they are not necessarily phonetically similar to one another. For example, the English yuk is clearly not similar to the German pfui or to the English phew or pooh. At the same time, it is easy to see that many - though not all interjections of 'disgust' and similar feelings are phonetically similar to one another: they tend to consist of a bilabial or labiodental voiceless fricative (or a bilabial plosive), followed by a close vowel. This particular phonetic structure can be understood as an icon of a 'natural' oral or nasal gesture: an attempt to blow away a bad smell or to get rid of some undesirable stuff from one's mouth, and to do so without opening one's mouth widely (which might allow some undesirable substance to get in). The link between disgust-type feelings and the kind of phonetic structure described above may seem so 'natural' and so self-explanatory that it is difficult for a native speaker of English, Polish, Russian, German or Yiddish not to be surprised when one finds out that, for example, in Greek feu (phonetically nearly homophonous with the English phew of
316
Interjections across cultures
disgust) is an interjection of 'grief and anger' rather than disgust or repugnance (cf. Kinchin-Smith - Melluish 1966:33). Facts of this kind suggest that it would probably be crude and naive to say, for example, that in interjections the sound [f] is linked, crosslinguistically, with the meaning of 'disgust'. We might of course try to cover ourselves by insisting that we are talking not about 'disgust' as such but about 'disgust and similar feelings'. But are anger and grief expressed in the Greek feu similar to those expressed in the Polish fu or to those expressed in the Yiddish feh? If we do not spell out the meaning in question more precisely the claim becomes meaningless. One hypothesis which suggests itself is this: a physical gesture of oral or nasal rejection may be linked, cross-linguistically, with the semantic component 'I don't want this', or possibly, with a combination of components: 'this is bad' and 'I don't want this'. It needs hardly to be pointed out that the component 'I don't want this' may be as compatible with 'anger' or 'grief' (that is to say, with feelings of the kind encoded in the Greek feu) as it is with disgust, repugnance or revulsion. It is clear, however, that the presence of this component does not always lead to some interjection whose form indicates 'oral or nasal rejection'; and when it does, usually additional components are needed - different ones in different languages. Furthermore, it is likely that not all languages have interjections whose meanings include the components 'this is bad' or 'I don't want it'. But these are all, of course, empirical questions, which must await further investigation. To quote Darwin once more: We have now seen that scorn, disdain, contempt, and disgust are expressed in many different ways, by movements of the features, and by various gestures; and that these are the same throughout the world. They all consist of actions representing the rejection or exclusion of some real object which we dislike or abhor, but which does not excite in us certain other strong emotions, such as rage or terror; and through the force of habit and association similar actions are performed, whenever any analogous sensation arises in our minds. (Darwin 1955:261)
"The rejection of some real object which we dislike or abhor" can be said to correspond to the postulated semantic components: 'I don't want it' and 'this is bad'. The proviso that the stimulus in question doesn't arouse in us "certain other strong emotions, such as rage or terror" seems to point to a certain 'disinterestedness' of the feelings in question. There is an important difference between thinking 'this is bad' and thinking
Emotive interjections
317
'this is bad for me'; concepts such as fear, frustration or anger imply the latter, whereas concepts such as disgust or contempt imply the former. One could perhaps venture to hypothesise that - "throughout the world" - symbolic actions of oral and nasal rejection (including interjections) are associated with the thoughts 'this is bad', not with the thought 'this is bad for me'; but this hypothesis, too, requires empirical verification.
3.2. 'General purpose' interjections According to most dictionaries, many interjections encode no more than an unspecified emotion, an emotion of any kind. For example, LDOTEL describes the English oh as "an interjection used to express an emotion (e.g. surprise, pain, or desire)", and Webster (1973) describes the English ah as "an exclamation expressive of pain, surprise, pity, compassion, complaint, contempt, dislike, joy, exultation, etc. according to the manner of utterance". Other writers on the subject of interjections, too, often insist that this or that interjection can express any feeling whatsoever, depending on the context and intonation. For example, Rosten (1968: 274) lists no fewer than twenty-nine emotions which, he thinks, can be expressed by the Yiddish interjection oy (I omit his examples): (1) Simple surprise, (2) Startledness, (3) Small fear, (4) Minor sadness, (5) Contentment, (6) Joy, (7) Euphoria, (8) Relief, reassurance, (9) Uncertainty, (10) Apprehension, (11) Awe, (12) Astonishment, (13) Indignation, (14) Irritation, (15) Irony, (16) Pain (moderate), (17) Pain (serious), (18) Revulsion, (19) Anguish, (20) Dismay, (21) Despair, (22) Regret, (23) Lamentation, (24) Shock, (25) Outrage, (26) Horror, (27) Stupefaction, (28) Flabbergastation, (29) At-the-end-of-one' swittedness. But neither long lists of this kind nor short statements such as 'interjection X expresses an emotion (any emo"tion)' account satisfactorily for the fact that different interjections have different values. For example, Rosten illustrates the different values of the Yiddish interjections oy and ah as follows: When you jump into cold water you cry oy! Then, enjoying it, say a-aah. When you commit a sin, you revel in the pleasure, a-aah; then, realising what you've done, you cry oy! (Rosten 1968:274)
318
Interjections across cultures
He concludes: "Oy, accordingly, can be used to express: ... " and he lists the twenty-nine different emotions cited above. But, clearly, the illustration provided suggests something quite different from a list of twentynine different emotions, including joy, euphoria and so on, and in fact it can hardly be reconciled with such a list. I do not deny that oy can be combined with all those different emotions; but surely, it must also have a meaning of its own, a meaning different from the meaning of ah and much more specific than simply 'any emotion whatsoever'. I believe, however, that to capture that kind of meaning we must look not for some suitable emotion term or terms but for an underlying thought - a thought which would be consistent with the entire range of emotions compatible with a given interjection. To show how this can be done I will start with the Polish oj, homophonous with the Yiddish oy and closely related to it, but much more restricted in use.
3.2.1. The Polish oj
OJ is one of the commonest Polish interjections. SIP describes it as "an interjection strengthening the utterance, expressing pain, regret, fear, admiration, resignation, confirmation, etc." This description suggests a virtually unlimited range of use and no particular cognitive content. In fact, however, the range of oj is limited, and its cognitive content is quite specific, although difficult to spell out in words. The most common collocations involving oj are probably these: oj, niedobrze 'oj, that's bad', oj, boli! 'oj, it hurts!' and oj, nie 'oj, no'. These common collocations suggest clearly a negative bias. Intriguingly, however, oj can also be used sometimes in a positive context; in particular, it can be used as a 'moan' of admiration or delight. Searching for a possible underlying thought which would be compatible with all these different clues I am inclined to propose the following tentative formula: I feel something thinking about it I feel like someone who thinks: things can happen to people that people don't want This combination of components presents the speaker as feeling rather helpless and powerless. This is not necessarily a 'bad feeling', but it is more likely to be linked to pain, regret, apprehension or resignation than to positive feelings. When it is linked to positive feelings it
Emotive interjections
319
sounds as if the speaker was surprised at his good fortune. For example, a sentence such as
OJ, jak tu pi~knie! 'OJ, how beautiful this place is!' sounds a bit like a moan, a whine, or a squeak of delight. In searching for the elusive invariant of this interjection we can proceed as follows. First, we can note that it is often used to draw attention to one's pain (as in the common complaint oj boli 'oj, it hurts'); second, that it is often used in other 'negative' contexts, such as a situation of regret, apprehension or resignation; third, that it is never associated with violent negative feelings, such as anger or despair; and fourth, that when it expresses positive feelings, such as delight or admiration, it has a somewhat squeaky or whiny character, as if a little mouse or a little bird were expressing their feelings. One can't imagine, for example, a lion having the kind of feelings which could express themselves in an oj. The common theme seems to be something like weakness or helplessness. For example, oj nie 'oj no' sounds like a whine rather than like a protest, refusal or rejection. When the speaker feels fully in control of the situation, oj can hardly be used: a. OJ, niedobrze. 'OJ, that's bad.' b. *OJ, swietnie. ,OJ, that's excellent.' c. OJ, jak tu pi~knie! 'OJ, how beautiful this place is!' Sentence (a) above is very natural because it expresses a negative evaluation and suggests that the speaker is not in control of the situation; sentence (b) suggests the opposite, and is infelicitous; sentence (c) expresses a positive evaluation, but does not suggest that the speaker may be in control, and it is felicitous. The speaker's delight conveyed in (c) is coloured with surprise and it is quite compatible with an attitude expecting something bad (rather than something good). Most characteristically, however, oj is linked with helpless negative feelings, as in the following examples:
OJ, nic dobrego z tej dziewczyny nie ,OJ, this girl will come to no good.'
b~dzie!
(Gomulicki) (SJP)
320
Interjections across cultures
OJ, ja nieszcz~sny, w jak~z p6jd~ drog~? Kocham dziewczyn~, a wzifl,c jej nie mog~! (Grudzinski) (SJP) 'OJ, how unhappy I am; where should I go? I love a girl, and I cannot take her!'
Consider also the following examples from children's poems: OJ, na niebo wyszly chmurki! (Szelburg-Zarembina 1970: 12) 'OJ, clouds have appeared in the sky!' OJ, niedobrze - mysli swierszczyk To nie iarty z zimnym sniegiem! (Grodzienska 1970:30) ,OJ, that's bad - thinks the Cricket Cold snow - that's serious!' OJ, to klopot wielki! Te male literki Mylfl, si~ Jasiowi jut: czy to A, czy to B pr6tno pojfl,C chce, nie pami~ta ani rusz. (Grodzienska 1970:84) 'OJ, that's a worry! those little letters are hard to tell apart: is it an A, is it a B? Johnny can't understand it, he can't get it right.'
Kladzie mama koteczka do cieplego 16ieczka. Bierze syna za rfl,czk~. OJ, masz kotku gorfl,czk~! (Grodzienska 1970:81) 'Mum puts her kitten to bed-DIM. She takes him by the pawDIM. - OJ my kitten, you have a temperature!' OJ, ratunku! slonko tonie! S10nko tonie! Co to (Grodzienska 1970:10) 'OJ, help! The sun-DIM is sinking! What will happen?'
b~dzie?
The negative bias of oj, which is not strong enough to prevent it from co-occurring with positive feelings, is even stronger in the longer variant of this interjection: ojej, (and in its childish derivate ojejku), which has a somewhat exasperated tone. Another fairly clear difference between ojej and oj lies in the more spontaneous character of ojej, which has to be linked with a current (and, typically, sudden) thought. For example, one could hardly replace oj with ojej in the sentence: OJ, [*ojej) ja
nieszcz~sny! ...
kocham
dziewczyn~
a
wzi~c
jej nie
mog~!
'OJ, how unhappy I am! ... I love a girl and I cannot take her!'
Emotive interjections
321
which implies a long term reflection. By contrast, ole} is typically used as a comment on what is happening at the moment, as in the following examples: O}e}, alei lecicie, ledwom was zlapal. - zawolal ostatkiem tchu. (Brzoza) (SIP) 'Ole}, you're racing along. I scarcely caught up with you, he panted.' O}e}, tak si~ bo}~, ieby si~ nie pomylic! - wzdychala Marysia. (Swierszczyk) (SIP) 'Ole}, I'm so scared that I could make a mistake! - Marysia was sighing. '
Both the similarity and the difference between ojej and o} are well illustrated in the following passage, where o}ej is linked with a thunderstorm and oj, with a shower. (Partly, it is no doubt a matter of the right number of syllables; but it is also a matter of semantic plausibility.) Idzie chmura - ciemna, bura, duia. Oje}! Bedzie burza! Szumi(!, gn(! si~, skrzypi(! drzewa. OJ! B~dzie ulewa! (Szelburg-Zarembina 1970:22) 'A cloud is coming - dark, brown, big. Ojej! There will be a thunderstorm! Trees are rustling, bending, creaking. OJ! There will be a shower!'
To account for both the similarities and the differences between ojej and o} the following explication can be proposed for ole}. o}ej I now think: things can happen that I don't want I feel something bad because of that
This formula differs from that assigned to oj by the element 'bad' in the last component, by the direct reference to 'I' in the second component, and by the added element of immediate thought: 'now' in the first component.
322
Interjections across cultures
3.2.2. The Russian oj It is interesting to note that a very similar interjection (also spelt oj) exists in Russian, and that it has a semantic value very similar to that of the Polish oj. For example: OJ! Petr !vanovic, Petr !vanovic, nastupili na nogu! (Gogol') (SSRLJa) 'OJ! Petr Ivanovic, Petr Ivanovic, you stepped on my toe!' OJ, strasno! Ne govorite, pozalujsta, ne prodolzajte. (Ostrovskij) (SSRLJa) 'OJ, that's scary! Please, don't speak, don't go on.' Mamocka, oj, kak xoroso! (Gor'kij) (SSRLJa) 'Mummy, oj, that's excellent!' [lit. 'how good! ']
It appears, however, that the range of use of the Russian oj is wider, and that its frequency is higher than that of the Polish oj. This is reflected, for example, in the existence of the verb ojkat' (to say oj) in Russian. Polish has no corresponding verb derived from oj, although it does have, for example, the verb achac (to say ach) derived from the common Polish interjection ach. (Russian, too, has a correspnding verb axat' / axnut', as well as oxat' /oxnut': 'to say ax' and 'to say ox'). The wider range of use of the Russian oj than of the Polish oj can be illustrated with the following example: Potom, podojdja k divanu, ... s udivleniem skazala: - OJ, mjagko. (Simonov) (SSRLJa) 'Then, coming up to the couch, ... she said with surprise: - OJ, soft. '
In Polish, one would not use oj in a similar sentence, although one could exclaim, for example: OJ, jak mi~kko! 'OJ, how soft!' mi~kko!
(cf. OJ, niedobrze!) 'OJ, soft! (cf. OJ, (that's) bad!)' ?OJ,
The difference seems to be that in Polish positive evaluation can be linked with oj only if it is sufficiently extreme to warrant something like a squeak of delight. The Russian oj, on the other hand, can convey an
Emotive interjections
323
element of surprise, being linked with a sudden thought, ('I now think X'), and can therefore be applied to relatively trivial situations. The differences between the two ojs can perhaps be represented as follows:
oj (Polish) 1 think: things can happen to people that people don't want 1 feel something thinking about it oj (Russian) 1 now think (know?) something 1 wouldn't have thought that 1 would think (know?) this 1 think: things can happen to people that people don't want 1 feel something because of that The phrasing of the 'emotive' component is meant to reflect the greater immediacy of the Russian oj, which sounds as if it was less controlled than the Polish oj. This spontaneous, uncontrolled character of the Russian oj seems particularly clearly reflected in the verb ojknut', 'to utter oj'.
3.2.3.
Ochs and achs
Och and ach, too, may appear to be all-purpose interjections in Polish. SIP defines ach in very general terms: "an interjection expressing a lively movement of emotion (lit. 'feeling')" and for och it suggests virtually the same, although the phrasing is somewhat different: "an interjection expressing a lively movement of emotion (wzruszenie, lit. 'being moved'), excitement, pain, impatience etc." As one could expect, however, the two are in fact not identical, with och sounding even more 'emotional' than ach, and being more likely to be seen as an exaggerated outpouring of emotion. But here as elsewhere, the apparent difference in strength is in fact due to an underlying difference in quality: there are some contexts when - regardless of 'strength' - och is more felicitous than ach, or vice versa. For example, och tak 'och yes' could not be used to check whether one has understood the interlocutor correctly, whereas ach tak 'ach yes' could very well be so used: A: 'He is leaving.' B: Ach tak? / *Och tak?
324
Interjections across cultures
In some contexts, both ach and och can be used, but with a clear difference in meaning. For example, ach nie 'ach no' is likely to be used to correct an error on the part of the addressee, whereas och nie 'och no' suggests, rather, that the speaker is rejecting some proposal. An affirmative ach tak could be used when one hears something interesting and unexpected, as one says in English is that so. But an affirmative och tak cannot be used in that way. A: 'He is leaving.' B: Ach tak! / *Och tak!
On the other hand, an affirmative och tak! (but not ach tak) could be used in reply to an eagerly awaited invitation or proposal: A: 'Would you like to come?' B: Och tak! / *Ach tak!
What these clues suggest is that ach has something to do with knowledge (or with coming to know something), whereas och involves the speaker's wanting rather than knowing. This difference between the two can be represented as follows: ach I feel something (now) because I know something och I feel something (now) because I want something These explications explain, I think, why one can use ach - but not och - to ask for confirmation: it makes sense to ask the addressee to confirm a fact but not to ask him to confirm that I want what I want. However, the explications sketched above are lacking some components which would account for the ironic use of nouns and verbs derived from these interjections. This ironic use of ach and och can be illustrated with the following examples: wpadaj{l w achy i w ochy na widok ponczochy. (Tuwim) (SJP) 'they lapse into oohs and aahs (achs and ochs) at the sight of stockings. ' Przybyla jakas niemiecka Korinna, pelna ach6w. (Chopin) (SJP) 'There arrived a German woman called Corinna emitting nothing but aahs (achs).'
Emotive interjections
325
Wrocili nareszcie! i w kilku powiatach zakipialo jak w garnku na odglos tej nowiny; a bliskie s(lsiedztwo ochalo i achalo z podium, sluchaj(lc opowiadania pana Groszka 0 cudach ich rezydencji i 0 trybie ich zycia. (Plug) (SJP) 'At last they returned! And in several districts, the news evoked a furious excitement; their neighbours on the podium oohed and aahed, as they heard Mr Groszek's stories about the marvels of their residence and their way of life. '
The point is that of all the Polish interjections, these two more than any other are perceived as 'loud' and rhetorical, as if the speaker was trying to publicise his or her emotions and to attract to them other people's attention. For example, they seem much 'louder' (both phonetically and semantically) than the English interjections ooh and aah, which would normally be used in English translations of Polish sentences with och and ach. This 'loudness' is no doubt due to the highly audible final fricative consonant, which the English interjections in question don't have. It should be noted that Polish has also purely vocalic interjections o and a, which also correspond, roughly speaking, to the English oh and ah, and which also differ in 'loudness' from och and ach. To account for this 'loud' character of ach and och, I would postulate for them the following additional component: 'I want someone to know about this'. OJ is not perceived as 'loud' and an utterance opened with oj would be much more likely to be represented in writing with a comma than with an exclamation mark. It is much more common for ach than for oj to be represented in writing with an exclamation mark, and even more common, for och, as in the following examples: Och! zawolola z dobrze udanym zachwytem. - Pani co dzien piekniejsza. (W. Kowalski) (SJP) "'Och!", she exclaimed, with well-feigned delight. "You are more beautiful every day.'" Kamionkowa wydala okrzyk: och! i zemdlona padla na ziemi~. (Dygasinski) (SJP) 'Mrs. Kamionek gave a cry: och! and fell in a faint on the ground.' Och! jak bolesnie, och! jak bolesnie, ie dzien wczorajszy nigdy nie wskrzesnie! (Syrokomla) (SJP) 'Och! how painful, och! how painful, That yesterday can never be restored!'
326
Interjections across cultures
It is worth noting that virtually homophonous interjections ox and ax exist also in Russian, where they appear to be used with meanings very similar to the Polish ach and och. A more detailed discussion of these interjections, however, is beyond the scope of the present study.
4. Cognitive interjections Cognitive interjections are global (unanalysable) expressions which express the speaker's mental state without reference to feeling or wanting. Three examples from Polish, aha, oho and 0, will be discussed below and will be compared with some related interjections from Russian and English.
4.1. The Polish aha and Russian aga SIP defines aha as "an interjection which expresses recalling, confirming, satisfaction, irony". In fact, aha by itself never signals satisfaction or irony, but once the invariant of this interjection has been captured it is easy to see why such attitudes can easily 'attach themselves' to it, in a suitable context. 'Recalling' and 'confirming' are more directly relevant to that invariant, but no more so than 'recognising' or 'understanding'. What all of these different uses have in common is a kind of sudden realisation, which can be captured in the following invariant: 'now I know it'. The following examples demonstrate, I think, the semantic correspondence between the Polish aha and the expression 'now I know it'.
o
czym mysmy mowili? ... Aha! Otoi chcialem was spytac ... (Iackiewicz) (SIP) 'What were we talking about? Aha! [oh, that's right - now I know it] I wanted to ask you ' Aha, jui wszystko rozumiem. (Prus) (SIP) 'Aha [now I know it], I understand everything now.' Aha! ... a tuscie zlodzieje! (W. Boguslawski) (SIP) 'Aha! [now I know it] ... here you are, you thieves!'
Cognitive interjections
327
PowtarzaJ z przechwaJk{l. Aha! Nie na moje wyszlo? (Orkan) (SJP) 'He was repeating boastfully: Aha! [now I know it] Wasn't I right all along?'
I do not maintain, however, that the expression 'now I know it' captures the full meaning of aha. As we will see below, Polish has also another interjection (oho) to which the same (or a very similar) component could be attributed and which nonetheless differs from aha in its range of use. The main difference is due to the fact that in oho the knowledge is due to perception, whereas in aha it is due either to the understanding of something that the interlocutor has said, or to a sudden realisation (that is, to a thought), or to a sudden discovery confirming one's thoughts or suspicions. As a generalisation which would cover all these different possibilities and which would nonetheless exclude cases where oho rather than aha would be said, I propose the following: aha (Polish) I now know it I have thought about it
Understanding, recall, realisation (through a thought process), confirmation of one's suspicions - all these different situations where aha is likely to be used, are compatible with this formula. Situations where oho can be used but aha cannot, appear to be incompatible with it. It is interesting to note that the corresponding Russian interjection, aga, is used to indicate 'agreement or confirmation' (SRJa). For example: - My tebja na poruki voz'mem - skazal Fidel' - ja pogovorju s muzikami. - Aga, pogovori. (Dovlatov 1977: 135) "'We'll vouch for you", said Fidel, "I'll talk with the blokes." "Aga, (you) talk (then)." , - Mal' ciki, - sprosila Nadja, - vy nemnogo coknutye? - Aga, - govorju, - my psixi. Kukareku. (Dovlatov 1977:136137) "'Boys", asked Nadja, "are you both nuts?" "Aga", I said, "we're psychos. Cock-a-doodle-do!" ,
328
Interjections across cultures
... ponjal? Miska obradovalsja: - Aga, ponjal! (Dragunskij 1968:207) , " ... got it?" Miska (answered) joyfully: "Aga, I got it.'" - fa, mama, sejcas byka s" est' mogu. Ona ulybnulas'. - Zivogo byka? - skazala ona. - Aga, - skazal ja. (Dragunskij 1968:154) '''Mum, I could now eat a bull." She smiled. "A live bull?" she said. "Aga", I said.'
The Polish aha could normally not be used in corresponding Polish sentences. As a first approximation, we could represent the meaning of the Russian aga as follows: I think: you want to know if I want to say the same (as you now say) you can know: I want to say the same It is true that aga can also be used to show that one has come to understand something, as the following example clearly shows: - Xorosaja skola, - skazal on. - Kakaja skola? - ne ponjal ja. - Xudozestvennaja, - skazal on. - Aga, - skazal ja. Xotja vse ravno ne ponjal. (Dragunskij 1968:83) '''A good school", he said. "What school?" I couldn't understand him. "An art school", he said. "Aga", I said, although I still couldn't understand.'
It appears, however, that this use of aga to show a sudden understanding can be interpreted by the same semantic formula: presumably, the first speaker hopes that the addressee understood him, and wants to know if the addressee would say the same; and the addressee assures him that this indeed is the case: 'you can know: I want to say the same (I have understood).' Interestingly, the same (or homophonous) Russian
Cognitive interjections
329
interjection aga can also be used to express "a successful guess, joyful surprise, derision, etc." (SRJa), as in the following example: Teper' ty nasi Aga, drozi§'. (Puskin) (SRJa) 'Now you belong to us! Aga, you're trembling.'
This seems very similar to the triumphant and derisive use of the Polish aha, as in the Polish examples quoted earlier: Aha! 'Aha!
a tuscie zlodzieje! so here you are, you thieves!'
Powtarzal z przechwalkfl. Aha! Nie na moje wyszlo? 'He repeated with satisfaction: Aha! Wasn't I right all along?'
However, in the case of the Polish 'triumr~ant aha', the formula proposed for the purely cognitive Polish aha seems to fit, too (perhaps with an additional component 'I feel something because of it'). By contrast, in the case of the Russian 'triumphant aga(2)' the formula proposed for the purely cognitive aga(l) does not fit, and despite the intriguing semantic links between the two agas, perhaps we should agree with the dictionaries of Russian (cf. SRJa and SSRLJa) which treat them as different lexical items: aga l , (roughly, agreement/confirmation) and aga 2 (roughly, satisfaction) The formula 'I now know', which was proposed for the Polish aha, might seem to fit the use of the Russian aga in sentences such as (a) below, but not in sentences like (b): a. Aga! Stiva! Oblonskij! Vot ion! - pocti vsegda s radostnoj ulybkoj govorili, vstrecajas' s nim. (Tolstoj) (SSRLJa) "'Aga! Stiva! Oblonsky! There he is!", they would say almost invariably with a happy smile whenever they met him.' b. Aga! - dognal tebja! postoj! - kricit naezdnik udaloj. (Puskin) (SSRLJa) "'Aga! I've caught you! One moment!" shouted the bold horseman.'
The formula which would fit examples of both kinds (a and b) seems to be this: aga 2 now I can say it I feel something good because of that
330
Interjections across cultures
The Polish aha implies that the speaker discovers some kind of missing link, that is, comes to know something that a moment earlier he didn't know; the Russian aga 2 implies that the speaker comes to be able to say something, which a moment earlier he wasn't able to say. To capture the link between aga l (the aga of agreement) and aga 2 we could now try to rephrase the formula assigned earlier to aga l as follows: aga l I think you would want to say: 'you and I say the same (about it)' you can say it now Interestingly, the formula assigned above to the Russian aga 2 seems to fit also the English interjection aha, which, according to LDOTEL, is "used to express surprise, triumph, derision, or amused discovery". Triumph, derision, or amused discovery are all perfectly consistent with the formula: now I can say it I feel something good because of that It is true that LDOTEL mentions also surprise, which may have little to do with this formula. As LDOTEL itself recognises, surprise can also be expressed by other interjections, for example by gee and wow, whose meaning is of course different from that of aha. It could be argued that aha by itself does not signal surprise, although it may be compatible with it; but perhaps the English aha should have a third component referring to something like surprise: aha (English) now I can say it I feel something good because of that one wouldn't have thought I could say it It should be noted that the English aha cannot be used as, roughly speaking, a sign of affirmation ('yes'), as the Russian aga l can. In this respect, the English aha is similar to the Polish aha. But the English aha and the Polish aha differ from one another in another respect. The Polish aha appears to be used much more widely and much more frequently than the English aha. This difference in range and frequency is linked, I think, with the fact that (as suggested by LDOTEL) the English aha carries an emotional component (represented in my formula as 'I feel
Cognitive interjections
331
something good because of that'), whereas the Polish aha can be used as a purely cognitive interjection: now I know it I have thought about it An emotional component ('I feel something [good?] because of that') can of course attach itself to this cognitive meaning and can be signalled by the tone of voice, but it is not part of the semantic invariant of the Polish aha.
4.2. The Polish oho According to SIP, oho is "an interjection strengthening an utterance" and either (a) "expressing admiration, approval, surprise" or (b) "emphasising a fact which has just been stated and which is usually evaluated as negative". In this section I will limit myself to the non-emotive (i.e., b) use of oho. SIP is on the right track, I think, when it links oho with "a fact which has just been stated". It is important to add, however, that in this case the speaker notices the fact rather than realises (or recalls) it, as in the case of aha. For example, one could say oho - but not aha on hearing remote thunder; one could also say oho in response to something the interlocutor says, but not to indicate comprehension (in which case aha would be appropriate). Oho would indicate that the speaker treats the interlocutor's utterance as an observational datum. For example, it may convey the idea 'now I see that you are afraid of your wife'. The main difference, then, is that aha links knowledge with thinking, whereas oho links it with perception or observation. To avoid using in the semantic formula the complex and language-specific term 'perceive' ('I perceive something') we can represent this link with observation as follows: 'I think anyone who is here now can know it'. aha I now know it I have thought about it oho I now know something I think anyone who is here now can know it
332
Interjections across cultures
As these formulae suggest, there is also another, related difference between oho and aha, which has to do with the contrast between 'old information' and 'new information'. In the case of oho what is perceived is 'new information'. In the case of aha the knowledge is new (in the sense of being just acquired), but its content may have been 'old' (the speaker may have suspected this content to be true before, or may have known it and forgotten it). Hence the contrast between 'something' in the explication of oho and 'it' in the explication of aha. Furthermore, as pointed out by SIP, oho appears to have a negative bias. This could be represented by means of the component 'I think this is bad', but this might be too strong, as the negative evaluation is only hinted at rather than clearly conveyed. A better phrasing of the relevant component is therefore 'I think this can be bad'. This brings us to the following formula oho now 1 know something (X) 1 think anyone who is here now can know it I think this (X) can be bad
There is, however, one further respect in which this formula is perhaps not entirely satisfactory, which is revealed by the comparison between oho and the English expression: oh-oh (to be discussed shortly). The point is that oh-oh (unlike aha) is something that the speaker might be muttering to himself; by contrast, oho is deliberately 'loud', as if the speaker was noting a 'public' fact and possibly trying to draw other people's attention to it. This suggests the following further amendment to the formula proposed: oho now I know something (X) 1 think anyone who is here now can know it 1 want someone to know it 1 think this (X) can be bad
The reader is invited to check the adequacy of this formula against the following examples: Przyjrza/a si~ uwazniej Paw/owi, kt6ry siedziaf osowialy i milczg,cy. - Oho - zauwazyla - jestes dzis nie w humorze. (Brandys) (SIP)
Cognitive interjections
333
'She looked more closely at Pawel, who sat silent and owlish. "Oho", she remarked, "you're not in a good mood today." , Oho, juz ci~ przyszla zona za leb bierze. (Przynski) (SIP) 'Oho, your future wife's already got you by the throat.' Oho, na mnie jui czas! (looking at a watch). Za pi~tnascie minut capstrzyk ... (Gruszecki) (SIP) 'Oho, I have to go. In a quarter of an hour we've got a demo ... ' Oho! znowu moraly! B~dzie gderal dzien caly! (Rodoe) (SIP) 'Oho! More sermons! He'll be complaining all day!
It could be added that oho is clearly related to the particle/interjection 0 and to the emotive interjection hoho. For reasons of space I will restrict myself here to looking briefly at the cognitive (non-emotive) use of o.
4.3. The Polish
0
The demonstrative (non-emotive) Polish 0 (quite different from the emotive/vocative 0) can be regarded as a modal particle rather than an interjection, because it rarely is used on its own. It can, however, be so used (with a gesture). As pointed out by SIP, "the interjection (particle) 0 can accompany the gesture of pointing to something in space, and drawing special attention to it". For example: Tu szlo }akies wo}sko: 0, slady po czolgach. (Broniewska) (SIP) 'Some soldiers have been past here: 0, here are marks left by tanks. ' Wilgot, zimno, 0, tu lezy koc, a pan nawet si~ nie okry}e. (Szaniawki) (SIP) 'It's damp and cold, 0, and here's a blanket and you won't even wrap up.' 0, mucha leci do koni. (Galczynski) (SIP) '0, there's a fly flying towards the horses.' Niechie pan doktor bedzie laskaw powiesic futerko tuta} ... 0 tuta}. (Prus) (SIP) 'Doctor, would you like to hang your coat here ... 0 just here.'
334
Interjections across cultures
The following semantic formula can account for all these (and other similar) examples: o I think anyone who is here now can know it I want you to know it (I do this [gesture] because of that) Like oho, 0 is 'public' because it refers to a public fact, accessible to any observer ('I think anyone who is here now can know it'). At the same time, however, it is aimed at a specific addressee ('I want you to know it'), whereas oho is directed at someone/anyone rather than at 'you' ('I want someone to know it'). Unlike oho, 0 does not imply any evaluation (any 'I think this can be bad').
4.4. The English oh-oh On the other hand, a negative component is clearly included in the English interjection oh-oh (uh-oh), which is used when one perceives that something bad and unforeseen is going to happen to someone (for example, a runner is going to fall into a hole of which the speaker is aware but the runner is not). The meaning of this interjection can be represented as follows: oh-oh I now know something I think something bad will happen now I would not want someone to think that it will be very bad As pointed out earlier, this is related to the Polish oho, but it is more 'private' as it lacks the components 'I think anyone who is here now can know it' and 'I want someone to know it', and is more definitely 'bad', though not very bad ('something bad will happen' versus 'this can be bad').
4.5. The Russian ogo In the Russian translation of Franklin Folsom's children's book entitled The language book, the meaning of the Russian interjection ogo is explained with reference to an imaginary prehistoric scene:
Cognitive interjections
335
'Ux! - a prehistoric hunter might have exclaimed throwing down the deer which he had dragged into his cave. And the members of his family started, perhaps, to repeat this sound every time after some hard work. Ogo! the hunter's daughter might have said when she saw what a big deer her father had caught.' (Folsom 1963:25)
The meaning of the Russian interjections ux and ogo suggested in this passage seems to be this: ux I can now say that I did something that I had to do I felt something bad when I did it I don't have to do it now I feel something good because of that ogo I now know something I think this is more than one would have thought (I think this is good) (I feel something because of that) Ux does not belong to the domain under discussion and will not be discussed here any further. As for ogo, the formula proposed above on the basis of one imaginary example seems to fit quite well the examples adduced in the dictionaries of Russian - except for the positive component 'I think this is good', which apparently is not part of the invariant. For example, it is compatible with the first example below, but not with the second: Peresmotrju narocno, skol' ko u menja deneg. Eto ot sud'i trista; eto ot poclmejstera trista, fest'sot, sem'sot, vosem'sot ... devjat'sot ... Ogo! za tysjacu perevalilo! (Gogol') (SRJa) 'I'll take a look how much money I've got. Here's three hundred from the judge; and from the postmaster another three hundred, six, seven, eight ... nine hundred ... Ogo! That's already more than a thousand!' Pal' ba byla ne dal'se, kak za verstu. Vse nastorozilis'. - Ogo! Ne dremljut japoncy! - skazal Sancer, nervno ozivljajas'. (Veresaev) (SSRLJa) 'There was shooting, not more than a verst away. Everyone became alert. - Ogo! The Japanese are pretty active! - said Sancer, stirring nervously.'
336
Interjections across cultures
If we also exclude from the hypothetical invariant the emotive component 'I feel something because of that', we are left with two components: I now know something I think this is more than one would have thought The examples of the Russian ogo adduced so far would also be compatible with the component of 'perception', which we have included in the explication of the Polish oho ('I think anyone who is here now can know it'). But ogo can be used much more widely than the Polish oho partly, no doubt, because it does not require a reference to current perception. Consider, for example, the following examples: Kostik sprosil: - A razve byvajut xorosie bolezni? - Ogo, skazal ja, - skol' ko xoces'. (Dragunskij 1968:28) 'Kostik asked: "Can there be good illnesses?" "Ogo", I said, "plenty." ,
The Polish oho could not be used in a similar context - partly, no doubt, because the speaker is referring here to his knowledge but not to his perception. In fact, examples of this kind suggest also that the knowledge implied by ogo should perhaps be represented as 'I know' rather than as 'I now know'. In the case of the Polish oho, the speaker's knowledge has the form of a realisation: 'I now know'. The Russian ogo, too, is often used in the situation of a sudden realisation, but apparently it can also be used to express knowledge acquired earlier. This brings us to the following semantic formula: ogo I know something I think this is more than one would have imagined
Or, perhaps: ogo I can now say something because I know something I think this is more than one would have imagined
Conclusion
337
5. Conclusion I have tried to show that interjections - like any other linguistic elements - have their meaning, and that this meaning can be identified and captured in rigorous semantic formulae. Formulae of this kind can explain the range of use of a given interjection, and can account for the differences in the use of different interjections. Are the interjections themselves identifiable (by native speakers) on the basis of the proposed formulae? Emotive and cognitive interjections seem to differ in this respect from the volitive ones. Volitive interjections such as shoo or psst in English can be easily identified on the basis of the semantic formulae assigned to them. On the other hand, emotive and cognitive explications such as the Polish oj, och, aha, or oho, are harder to identify, and a native speaker may be sometimes at a loss trying to link a semantic formula to an interjection which it is supposed to explicate. In this respect, interjections may be different from verbs or adjectives (and perhaps more similar to modal particles). However, if formulae of the kind proposed here are compared with traditional descriptions of interjections (along the lines of "a word which can be used to express different emotions depending on the context and on the intonation") their higher explanatory value seems to be evident: they try, at least, to account for the differences in the use and in the value of different interjections, and they try to discover in each case an invariant which would explain the unique range of use of every individual interjection. The fact that the route back from the explication to the word may be longer and more difficult to travel than it usually is in the case of major lexical classes, should of course be noted, and its implications should be explored. Perhaps we should conclude from it that different types of linguistic signs have different psychological status. After all, an interjection. is an equivalent of a full sentence. Perhaps a mental act encoded globally in one phonologically tiny word is generally harder to recognise (reconstruct) on a conscious level than an act encoded in a more articulated linguistic expression? Perhaps a global sign such as an interjection is in some sense more 'automatic' than a non-global sign, such as a verb or an adjective? We should also ask if the 'semi-automatic' character of (primary) interjections is not related to their partial phonological motivation. It
338
Interjections across cultures
seems clear that in the domain of interjections certain types of sounds tend to be linked, cross-linguistically, with certain types of meanings. Correlations of this kind have often been suggested in the literature, but they could not be seriously investigated before the question of the semantics of interjections was placed, in a serious way, on the linguistic agenda. Here as elsewhere the absence of a language-independent semantic metalanguage has been a major stumbling block. It is easy enough to say, casually, that if the German interjection pfui, the Danish interjection fy or the Russian interjection fu all express disgust, then sound symbolism is probably involved. But if one considers that 'disgust' itself is just an English word, without an exact equivalent in German, Danish or Russian (see Wierzbicka 1986c; to appear, chap. 3), it becomes clear that it is not a language-independent descriptive category, which could be fruitfully used for a cross-linguistic study of correlations between form and meaning. On the other hand, operating with universal or near-universal semantic chunks such as 'good' and 'bad', 'feel' and 'think', 'know' and 'hear' we can at least start to investigate the meaning of interjections in a cross-cultural perspective. Once we have begun to model their meanings in a relatively culture-free semantic metalanguage, we can start to investigate the extent to which sound symbolism plays a role in the functioning of interjections. We can start to explore the universal and the culture-specific themes in the semantics of interjections, and the interplay between the two. We can also start to explore, and to document, different 'emotive styles' associated with different cultures and reflected in language-specific systems of interjections. For example, it can hardly be an accident that in a culture whose most prominent speech acts include laments and impotent curses - the Jewish-Yiddish culture (cf. Matisoff 1979) - the most prominent interjection appears to be oy vay (oy veh), a linguistic 'symptom' of distress and helplessness, whose meaning can be represented as follows: oy vay
I I I I
now know that something bad happened would want to do something because of that can't do anything feel something bad because of that
It appears that languages differ in the kinds of emotions which they have found worthy of encoding in special interjections. For example, some languages appear to have special interjections in the domain of
Conclusion
339
fear, others in the domain of anger, and yet others in the domain of sadness and distress. In addition to qualitative differences of this kind, there are also important quantitative differences. Firstly, some languages appear to have much larger sets of primary interjections than others. Secondly, the interjections used in different languages differ greatly in frequency. This last point, though generally very difficult to document, can be illustrated with reference to English and Russian. For example, in Zasorina's (1977) megacorpus of one million running words of Russian, the interjection t'fu was counted 19 times, and fu, 23 times, whereas in Kucera - Francis' (1968) comparable corpus of American English, interjections such as yuk, phew or pooh were not listed once. Compare also the following data regarding the most common emotive interjections of both languages: Russian oj ox
ax
19 83 212
English oh ah
119 22
Striking as these numbers are, they don't show the full extent of the difference, because Russian has one more common interjection, 0, whose frequency cannot be assessed on the basis of the available data: o is also a preposition in Russian, corresponding to the English about, and to some uses of the English of In Zasorina's data, the word 0 has the frequency of 4156, and we don't know how many of these belong to the interjection. Generally speaking, one would expect that in societies which discourage a spontaneous and uninhibited show of emotions, the use of primary interjections would be more limited than in those where emotions are shown freely and where expressive behaviour is valued rather than discouraged. The contrast between English and Russian illustrated above is a good case in point (cf. Wierzbicka, to appear). But of course in this area as in others, much research is needed before any firm generalisations can be reached. Cross-cultural research in the area of emotion concepts lexicalised in different languages has just begun (cf. for example Levy 1973; Lutz 1982, 1988; Solomon 1984; Rosaldo 1980; Gerber 1985; Wierzbicka, to appear, chaps. 3, 4). Cross-cultural research in the area of emotion symptoms lexicalised in different languages in the form of interjections, is, one can hope, about to begin.
Chapter 9
Particles and illocutionary meanings
There are few aspects of any language which reflect the culture of a given speech community better than its particles. Particles are very often highly idiosyncratic: 'untranslatable' in the sense that no exact equivalents can be found in other languages. They are ubiquitous, and their frequency in ordinary speech is particularly high. Their meaning is crucial to the interaction mediated by speech; they express the speaker's attitude towards the addressee or towards the situation spoken about, his assumptions, his intentions, his emotions. If learners of a language failed to master the meaning of its particles, their communicative competence would be drastically impaired. The meanings embodied in particles are often remarkably complex. Though these meanings can perhaps be expressed, in one way or another, in any language, they are often so complex that if a particular language does not provide any abbreviatory devices (like particles) the speakers are effectively discouraged from ever expressing them. As Hymes (1974a:1450-1451) put it, although "one could come to render anything in any language, given sufficient time and trouble", nonetheless "it is not the case that one can 'say anything in any language' if conditions of acceptability and cost, as are always present in real situations, are admitted". For this reason, particles, which provide generally accepted ways of expressing complex pragmatic meanings at minimal cost, play an essential role in co-determining the range of behavioural styles that a given language makes available to its speakers. The meaning of particles is often excruciatingly hard to state. Not that very many assiduous attempts have been undertaken towards that end; on the contrary, until very recently, the meaning of particles has seldom attracted linguists' attention. Even in semantics, it would be difficult to point to a more grossly neglected area. As John Locke said three hundred years ago: This part of grammar has been perhaps as much neglected as some others over-diligently cultivated. It is easy for men to write, one after another, of cases and genders, moods and tenses, gerunds and supines: in these and the like there has been great diligence used; and particles themselves, in
342
Particles and illocutionary meanings
some languages, have been, with great show of exactness, ranked into their several orders. But though prepositions and conjunctions, etc. [original emphasis] are names well known in grammar, and the particles contained under them carefully ranked into their distinct subdivisions; yet he who would show the right use of particles, and what significancy and force they have, must take a little more pains, enter into their own thoughts, and observe nicely the several postures of his mind in discoursing. (Locke 1690, 2:99)
Fortunately, the situation described by Locke is now beginning to change, and particles have finally started to attract serious attention. In particular, there has been a wave of serious publications devoted to particles in German and Russian, both particle-rich languages (see, for example, Kemme 1979; Weydt 1969; Weydt et al. 1983 for German; Boguslavskij 1985; Nikolaeva 1985; Universite de Paris-7 1986 for Russian; for a pioneering study of particles in a non-Indo-European language, see Ameka, to appear). Nonetheless, I believe that Locke's comments on the importance of particles, and on the need for their investigation, are still worth recalling; as are his remarks on the methodology suitable for this purpose. Neither is it enough, for the explaining of these words, to render them, as is usual in dictionaries, by words of another tongue which come nearest to their signification; for what is meant by them is commonly as hard to be understood in one as another language. They are all marks of some action or intimation of the mind; and therefore to understand them rightly, the several views, postures, stands, turns, limitations, and exceptions, and several other thoughts of the mind, for which we have either none or very deficient names, are diligently to be studied. (Locke 1690, 2:99)
The role that Locke attributes to particles corresponds to what in current terminology are known as 'illocutionary forces'. I have argued on more than one occasion (see Wierzbicka 1972, 1976, 1980) that the only possible way to represent these illocutionary forces accurately is to decompose them: illocutionary forces are bundles of assumptions, intentions and other more or less elementary 'postures' and 'turns' of the mind. I think that there is a profound insight in Locke's assertion that particles represent actions (actions of the mind), and that they are, therefore, abbreviations for whole sentences: The instances I have given in this one [particle] may give occasion to reflect on their use and force in language, and lead us into the contemplation of several actions of our minds in discoursing, which it has found a
Particles and illocutionary meanings
343
way to intimate to others by these particles, some whereof constantly, and others in certain constructions, have the sense of a whole sentence contained in them. (Locke 1690,2:100)
If a particle condenses in itself an entire sentence, then the proper way of stating its meaning must be to reconstruct this sentence. This conclusion seems to follow automatically from Locke's discussion. Yet Locke himself did not state it explicitly, and in his analysis of his selected examples he was content simply to make some general comments on the function of that particular particle. It was Leibniz who first applied to particles the principle of substitutability (the crux of Leibnizian semantics in general): if a particle contains in itself a condensed sentence, then to state the meaning of this particle one has to reconstruct this sentence in extenso. The possibility (or otherwise) of substituting the reconstructed sentence for the particle in question provides an empirical test of the adequacy of the proposed explication. For a proper explanation of the particles it is not sufficient to make an abstract explication ... ; but we must proceed to a paraphrase which may be substituted in its place, as the definition may be put in the place of the thing defined. When we have striven to seek and to determine these suitable paraphrases [original emphasis], in all the particles so far as they are susceptible of them, we shall have regulated their significations. (Leibniz 1949:366-367)
The task of reconstructing one sentence which would correspond to all the varied uses of a particle may seem unfeasible. In fact, however, there is no reason to assume a priori that all the different uses of a particle correspond to just one sentence (and just one mental 'posture'). Some particles may well have more than one meaning. But they will not each have 'countless' different meanings. To quote Leibniz again: Scholars have attempted to make special treatises upon the particles of the Latin, Greek, and Hebrew; and Strauchius, a celebrated jurisconsult, has published a book upon the use of particles in jurisprudence, where their significance is of no small consequence. We ordinarily find, however, that it is rather by means of examples and synonyms that they attempt to explain them, than by distant notions. Further we can not always find a general or formal signification for them ... which would satisfy all the examples; but notwithstanding this we can always reduce all the uses of a word to a definite number of significations. And this is what should be done. (Leibniz 1949:365-366)
344
Particles and illocutionary meanings
The idea that the meaning of particles can be adequately elucidated by means of examples, synonyms and translation equivalents, would probably not be seriously defended by many linguists as a valid theoretical option, although the practice in question is still widespread. But the notion that the meaning of particles can be adequately elucidated by means of abstract formulae, totally unsubstitutable for the particles themselves, continues to be put forward as a theoretical program, and continues, I would add, to fail to produce empirically adequate clues to the use of the particles discussed. (For illustrations and discussion, see Wierzbicka 1986a.) I think Leibniz was right: one can always reduce all the uses of a particle to a definite number of significations (stated as paraphrases capable of being substituted for the particles in various contexts); and this is what should be done. On the other hand, the Leibnizian approach to the study of particles has produced in the course of the last decade a number of empirically based studies, which have revealed the meanings encoded in a number of particles from a wide range of languages, and which have achieved a degree of predictive power undreamt of in the more traditional approaches to the study of particles. (Cf. in particular Ameka 1986; Boguslawski 1986; Goddard 1979, 1986; Grochowski 1986; Harkins 1986; Wilkins 1986.) The purpose of this chapter is to demonstrate this method of analysis by elucidating the meanings of a variety of particles in English and Polish. This will be done in the following order: English 'quantitative' particles (first non-approximative ones: only, mostly and just, and then approximative ones such as about, around, and almost); English 'temporal' particles (such as already, still, and yet); Polish 'temporal' particles with similar - but different - meanings; and Polish 'quantitative' particles (first non-approximative, then approximative ones), also with meanings similar to, but different from, those of the English particles. In all cases, paraphrases will be sought which would account correctly _for the observable similarities, and differences, in the range of use.
English quantitative particles
345
1. English quantitative particles English has a number of particles used for talking mainly, or exclusively, about quantities. These particles rely crucially on components such as 'no more', 'no less', 'not much', 'it could be more' or 'it could be less'. Some of these particles, and related expressions, purport to give precise information, whereas others are used as estimates and approximations. Compare, for example, the expressions as many as and more or less. The meaning of the former can be represented as follows:
as many as, as much as it is this one could think: it is less it is not less The meaning of the latter would at first seem to be representable along the following lines:
more or less it could be this it could be a little more than this it could be a little less than this It should be pointed out, however, that while as many (much) as is really restricted to quantitative contexts, more or less is not. Thus, one can say not only more or less 50 but also it is more or less the same. This suggests that while etymologically, or perhaps metaphorically, more or less is related to 'more' and 'less', synchronically and literally it might be better represented in terms of 'different' rather than 'more/less' :
more or less it could be this it could be a little different from this it couldn't be very different from this This example illustrates the tendency of quantitative particles to develop metaphorical extensions, and the difficulty one encounters in trying to separate quantitative senses from non-quantitative ones. In what follows, I shall largely ignore this problem, trying to focus, above all, on the similarities and differences between individual particles and particle-like expressions.
346
Particles and illocutionary meanings
1.1. Non-approximative particles: only, merely and just
1.1.1. Only
As a first approximation, the meaning of only can be represented as follows: only it is this (many/much) it is no more one could think: it is more
For example: only fifteen, only a dollar it is this (many/much): [15, $1] not more one could think: it would be more
In some contexts, the phrase 'not another' appears to correspond to only better than 'not more', as in examples like: Only Socrates runs. (cf. Goddard 1979) Only Nescafe gives you that fresh roasted flavour.
and it seems justified to suggest a second meaning, along the following lines: Only Socrates runs. this (person, thing, place ... ) not another one could think: it would be another
It seems to me, however, that in cases of this kind the identification is due to the definite noun phrase, whereas only itself implies, even here, the component 'no more'. For example, consider the following exchanges:
a. - Who's there? -Me. - You and who else? - Only me.
English quantitative particles
347
b. - Who's there? - It's only me. In (a), only me implies 'no more people', whereas in (b) it appears to be concerned exclusively with identification, not with the number of people ('me, not somebody else'). At the same time, it should be noted that even in (b), 'only me' implies a kind of limitation and (in Sapir's terms) 'downward grading': 'it's me, not somebody else, nothing more than me, you don't have to worry'. This reassuring ring of the phrase 'only me' can be accounted for if we posit for this use, too, the component 'no more than this' (in addition to 'not another'), although in this case more does not refer to a number and it may said to be meant metaphorically. It is an open question whether such a metaphorical interpretation of 'no more' should be regarded as a separate meaning; it should be recognised, however, that a single explication of only in terms of 'no other' would not be sufficient, and that all uses of only appear to require something along the lines of 'no more'. The same 'downward grading' quality of only is evident in its adverbial uses, for example: I was only joking. (cf. ??/ was only speaking seriously.)
Clearly, the speaker doesn't mean here 'I was doing one thing, no more', and the paraphrase 'I was joking, nothing else' seems more apposite. But here, too, the phrase 'nothing else' (or 'not something other than this ') by itself would not account for the downplaying implication of the sentence (cf. 'it was no more than a joke'). My tentative conclusion is that while the uses of only considered here can apparently be reduced to the formula 'no more; one could think: it is more', this formula is open to two rather different interpretations. In clearly quantitative contexts, for example in combination with numerals, only has no 'down-playing' quality, because its component 'no more' is taken in a literal sense. If, however, no quantifier is present (for example, in combination with proper names), only can be interpreted in two different ways: either as 'this person (thing, etc.); no more persons' or as, roughly speaking, 'this person; no one (nothing) else; nothing more'.
348
Particles and illocutionary meanings
1.1.2. Merely Merely is closely related to only, but it appears to have two additional components, which can be roughly described as 'this is not much' and 'this is not something important'. For example, the phrase 'merely 50 cents' implies, like only 50 cents, that the sum in question was 'no more than 50 cents and one could expect it would be more'. In addition, however, it dismisses the sum in question as something small and unimportant, which only doesn't do. If only can be said to be 'restrictive' or 'limiting', merely can be said to be also 'minimising' and 'depreciatory' or 'dismissive'. To begin with the 'minimising' aspect of merely, it can be represented as 'not much', and thus contrasted with the 'limiting' character of only, which we have represented in terms of 'not more'. Compare, for example, the following sentences:
a. She gave him only two thirds of the sum he expected. b. She gave him merely two thirds of the sum he expected. Sentence (b) sounds odd because 'two thirds' doesn't sound like 'not much'; but (a) is fine because only does not imply 'not much', only 'not more' (in this case, not the whole sum). Turning now to the 'dismissive' or 'depreciatory' aspect of merely we can note that by calling someone's words 'mere words', 'mere promises' or 'mere conjectures' one is not only 'minimising' their value ('that's not much'), but also dismissing them as unimportant, not worth thinking about. For example, making a suggestion or putting forward a hypothesis one could well add, modestly: This is only a suggestion. This is only my hypothesis.
but it would be rather self-defeating to say: This is merely a suggestion. This is merely my hypothesis.
Similarly, the sentence She's merely a child could well be used in a situation when one was trying to dismiss a girl's words as unimportant or unreliable. In a situation when one was trying to protect a child from excessive burdens one would be more likely to say She's only a child (or She's just a child).
English quantitative particles
349
It is worth noting that merely, in contrast to only (and just), can hardly combine with an imperative (whether explicit or implicit): Bring just (only, *merely) two. Just a moment! *Merely a moment!
The imperative implies 'I want X', but merely implies 'X is not worth bothering about'; hence the semantic clash. Finally, one can say, in reply to the question Who was there?: Only John and me. Only me. Just John and me. Just me.
but hardly: ?Merely John and me. ?Merely me.
On the other hand, the sentences It's only (/just) a game and It's merely a game are both felicitous, and easily interpretable. Only a game implies 'no more than a game' and merely a game dismisses the game in question as something unimportant. I suggest, then, that merely can combine only with words or expressions which can be interpreted as 'cultural synonyms' of 'unimportant', 'non-serious', 'non-real' or 'small', and as opposed to something 'important', 'serious', 'real' or 'big'. For example: This is merely a game (not serious business). She's merely a child (not an adult). Mere words/promises/conjectures (not deeds, not certainties). Merely a fraction (not the whole thing).
And so on. As a first approximation, we could consider, therefore, the following semantic formula: merely no more than this this is not much one doesn't have to think much about it
350
1.1.3.
Particles and illocutionary meanings
Just
According to Lee (1987), the English particle just has as many as four different meanings: a 'depreciatory' meaning, a 'restrictive' meaning, a 'specificatory' meaning, and an 'emphatic' meaning. In my view, the claim that just is polysemous is perhaps justified, but the number of meanings posited is too high, and the labels (as well as accompanying descriptions) do not show what these meanings really are, how they are mutually related, or how they differ from those of related particles such as merely or only. After all, one could equally well say that merely is depreciatory, or that only is restrictive; but this wouldn't explain how just, merely and only differ from one another, or why their ranges of uses are different. Let us start with two examples of the kind that Lee calls ,depreciatory' : They're not serious -
just a nuisance.
- Have you had any chest pain? - Just a little bit. Not as much as I had before.
According to Lee (1987:378), in examples of this kind "the speaker uses the particle to minimise the significance of some process. ... In many cases in this corpus [of doctor-patient conversations] a particular process is explicitly downplayed by comparison with some other process". In my terms, the meaning in question can be formulated as follows: just nothing else (= not something other than this) this is not much one could think: it is (would be) more
The label 'depreciatory' is misleading, because it suggests a negative evaluative component, which is certainly not there (cf. for example Let's have dinner together - just the two of us); and so is the label 'downplaying', which seems to suggest something unimportant. But the term 'minimise' is roughly correct, and it corresponds in essence to the proposed formula. For example, the component 'not much' assigned to just (but not to only) explains the following contrast: Only 47 people came. ?Just 47 people came.
English quantitative particles
351
Only 47 implies that more than 47 could be expected, but it doesn't imply that 47 is 'not much'; just 47 does imply this and this implication makes the sentence sound rather odd, because if the number of people is small enough for them to be individually counted, then 47 would appear to be a lot. Similarly: She gave him only two thirds of the sum that he expected. ?She gave him just two thirds of the sum that he expected.
At the same time, we should note that just, unlike merely, has not only a 'minimising' but also a 'precise' ring to it: just this implies not only 'this is not much' but also 'nothing other than this', 'precisely this'. It is partly for this reason, I think, that the phrase: Just the two of us sounds intimate and cosy: it is not just a question of a small number, but also of the identity of those included (there will be no outsiders, nobody other than the two of us). Generally speaking, if only is perfectly neutral, and merely is depreciative, just easily lends itself to mildly positive (reassuring, defensive, apologetic, even praising) interpretations. The reason is that while 'small' can be easily viewed as unimportant, it can also be viewed as desirable ('small is beautiful', 'small is safe', and so on). In the absence of a depreciative component ('one doesn't have to think much about this '), and in the presence of an identifying one ('no other than this'), the minimising component 'not much' embodied in just lends itself easily to positive interpretations. Lee (1987:387) illustrates what he calls the 'restrictive' meaning with sentences such as these: You can get a B grade just for that answer. [' II just get you to take your shirt off. Just close your eyes.
It seems to me that sentences of this kind fit perfectly well the formula proposed above, and that there is, therefore, no need to assign to the just used in them a separate, 'restrictive' meaning. Lee himself is aware that there are no firm grounds for distinguishing the two hypothetical meanings, but he blames this on "the indeterminate nature of the boundary between these two categories" (1987:387). A precise semantic formula such as the one posited here allows us to dispense with the alleged polysemy and reduce the 'depreciatory' and 'restrictive' senses of just to one. For example:
352
Particles and illocutionary meanings
I' II just get you to take your shirt off. I'll get you to take your shirt off nothing else this is not much one could think: it would be more
The 'specificatory' sense of just is illustrated with sentences like these: It gets itchy just under the eyebrows. He'd just come out of hospital.
According to Lee (1987:389), in sentences of this kind a locative or temporal expression refers to an area which is sharply defined at one end but undefined at the other, and "the function of just ... appears to be to identify only that marginal phase ... close to the sharply specified [part of the] area". It seems to me that this is an insightful observation. The ideas of 'not much' and 'not some other' appear to be relevant here, too, but their interpretation is somewhat different. For example, just under the eyebrows implies 'not much under the eyebrows', so much so that the place in question can be thought of as the same (no other) place as where the eyebrows are. Similarly, just after he'd come out of hospital implies 'very little (not much) after' - so little that the time in question can be thought of as the same as (not other than) when he came out of hospital. Since places and times can be thought of as areas and periods, as well as 'points', two different points in either space or time can be thought of as 'the same place' or 'the same time'. For example: just under the eyebrows where the eyebrows are not in another place it is under (the eyebrows) not much (not much under) one could think: it would be more just after he'd come out of hospital when he came out of hospital not at some other time it was after (he'd come out of hospital) not much (not much after) one could think: it would be more
English quantitative particles
353
In a sense, then, this use of just does indeed 'specify' a certain place or time - but not in the sense in which the particle right (as in it was right here) specifies a certain place. The 'specificatory' just 'specifies' a place or time by ignoring (downplaying, minimising) the distance between two points in space or time, and this is how it is related to the 'depreciatory/restrictive' just. The so-called 'emphatic' just occurs in sentences such as the followIng ones: It was just incredible. I just can't keep going. I just can't understand it.
Lee (1987:394) comments: "Here we have moved a long way from the depreciatory just. In fact, the particle appears to have precisely the opposite effect - to emphasise the expression with which it enters into construction". And yet, Lee feels that there are close links between these different and apparently opposed senses of just. It seems to me that precise semantic formulae phrased in the natural semantic metalanguage can show what these links are. In the two sentences below: I just don't like it I just can't understand it.
it is clearly counterintuitive to say that just is used in two very different, indeed opposed, meanings. Intuitively, the link between these two alleged senses seems very clear, and it can be stated as, roughly: this is all I can say about it (e.g. I don't like/can't understand it) I can't say anything else I know: this is not much one could expect more But if we put it like this, it transpires that, in essence at least, the same semantic formula which was proposed earlier for the so-called 'depreciatory', 'restrictive' and 'specificatory' senses of just applies also, mutatis mutandis, to the so-called 'emphatic' just: It is just incredible. It is just impossible. I just can't understand it.
354
Particles and illocutionary meanings
just 1 can't say anything other than this (e.g. it's incredible/impossible, I can't understand it) (I know: ) this is not much one could think: it would be more
The 'emphatic' sense of just appears to occur always in combination with the idea of impossibility: I just can't, it is just impossible, it is just unbelievable (i.e. one just can't believe it), it is just gorgeous (i.e. 'I can't say anything other than this', 'I can't say another word'), and so on. The phrase (I) just can't ... implies, generally, that 'I can't say anything other than this; this is not much, but 1 can't'. The implication is that the speaker is overwhelmed and lost for words; hence the impression that the message is, in the speaker's eyes, 'overwhelming', and that the speaker is 'emphasising' it, not 'downplaying' it. But it seems that this effect can be accounted for in terms of essentially the same semantic formula which has been assigned to the other uses of just.
1.2. English approximative particles There is no sharp line separating 'quantitative' particles from 'approximative' ones, because a particle, or particle-like expression, can be 'quantitative' and 'approximative' at the same time. This applies, in particular, to the particle-like expressions at least, at the most, no less, and no more, as used in the following sentences: At least fifty people were there. The dining-room seats forty at (the) most. No less than fifty chairs were crammed into the tiny room. No more than fifty people came to the plenary session. at least it is not less than this it could be more one could think: it would be less at (the) most it is not more than this it could be less one could think: it would be more
English quantitative particles
355
no less it is not less than this it could be more one could think: it would be less no more it is no more than this it could be less one could think: it would be more Like only, all of these expressions refer not only to quantifiers but also to expectations, and they imply that the actual quantifiers are different from the expected ones. At the same time, however, they imply that the speaker is making an estimate, whereas only implies precise information. It is important to stress that no less and no more act as approximatives only when they are not followed by any further quantification. It is by no means self-contradictory to say, for example, no less and no more than twelve. But this does not mean that no less on its own does not imply 'it could be more', or that no more on its own does not imply 'it could be less'. In this section, I shall consider a number of 'approximative' words and expressions, including around, about, approximately, roughly, almost, and nearly, trying to reveal the similarities and the differences between them.
1.2.1. Around and about Around 50 people came. / About 50 people came. Come around five 0' clock. / Come about five 0' clock. Dictionaries usually 'define' around via about (and/or via approximately), and vice versa. And indeed, in many contexts, the two words seem interchangeable. A rough gloss for all three is suggested by another 'approximative' expression, which can often - though by no means always - be substituted for them: more or less.
Fifty people came, more or less. (? )Come at five 0' clock, more or less. But do around and about really mean the same? And what do they mean?
356
Particles and illocutionary meanings
Sadock (1981), having discussed about In some detail, formulates his main conclusions as follows: I suggest that about be given the following definition: A sentence of the form about P is true just in case P is a quantitative proposition and there is a possible world not very different from the real world in which P is true. (Sadock 1981 :267)
But this definition cannot be regarded as correct, if only for the following reason: about is not restricted to 'quantitative propositions', and in fact it differs in this respect from around. Consider, for example, the following dialogue:
- Is the food there any better than in our college? - About the same. / *Around the same. I propose that this difference between the purely 'quantitative' around and the 'quantitative' or 'qualitative' about can be accounted for by defining around in terms of 'more' and 'less', and about, in terms of 'different'. As a first approximation, I would propose this:
around it could be this it could be a little more it could be a little less about it could be this it could be something a little different from this Sadock (1981:262) has also claimed that all English 'approximatives' have the conversational implicature 'not P' , i.e. 'not exactly P'. It seems to me, however, that this is simply not true. By saying around twenty or about twenty the speaker does not wish to imply that the actual figure was different from twenty. He implies that the actual figure may be a little different from twenty, but not that it has to be different. This is why in the proposed explications of around and about I have included the component 'it could be this'. But there are also other differences between around and about. One of them consists in the fact that around, but not about, is able to apply to a whole area, or period, surrounding a point in space or time, as in the following sentence:
Hats of this kind were worn in Paris around 1880.
English quantitative particles
357
If we replaced around with about, the sentence would imply that what is meant is one particular year, not a few years, as in the case of around 1880. In fact, about can hardly apply to a period at all; rather, it is used as a guess, an estimate, a 'shot' at one particular point in time:
Hats of this kind first appeared in Paris (in) about 1880. To account for this difference between about and around, we could differentiate the relevant components as follows:
around it could be this it could be a little more than this it could be a little less than this about it could be this if not this, it could be something a little different from this This means that about applies to one point: 'this', or to some other point in the area nearby; but around can, in principle, apply to a number of points in the area around the reference point. Of course such an extended interpretation can be excluded by a particular context, such as around 50 people or about 50 people, where the difference in question gets neutralised, but given an appropriate context the extended interpretation is always available for around, and not for about. For example:
The next few days were very hot, so we returned to our pattern of resting during the hottest part of the day and around [/?about] midnight ... (Facey 1981:144) Another difference between around and about has to do with the idea of 'rounding' encoded in the former, but not in the latter. While one would be unlikely to say either ?around 87 people or ?about 87 people, it is easier to say about 27 or 28 people than around 27 or 28 people, and easier still to say about 6 or 7 people than around 6 or 7 people. In fact, around 6 or 7 people sounds as if one were rounding some fractional number, such as six and a half, or six and three quarters. To account for this difference between around and about, I would postulate the following component for around: I say X, not something a little different from X because it is easy to think of X
358
Particles and illocutionary meanings
No such component would be postulated for about. The fact that one is unlikely to say about 87 people can be explained without positing any such component: it is simply a bizarre thing to do to mention such a high number as a rough estimate; if the figure was based on counting then there would be no need for about, and if it was not based on counting then the estimate would probably be made in tens rather than in single units (e.g. about eighty or ninety people). But when one says about 6 or 7 people one is trying to be accurate as far as possible, and no process of 'rounding' is implied. In sum, the proposed tentative formulae for around and about read as follows: around it could be this it could be a little more than this it could be a little less than this it couldn't be much more than this it couldn't be much less than this I say this (number), not another (number), because it is easy to think of this (number) about it could be this if not this, it could be something a little different from this it couldn't be something very different from this
1.2.2. Approximately Approximately is similar to around in implying a process of 'rounding'. One would be unlikely to say: ?Approximately 7 people came. ?Approximately 17 or 18 people came.
A feature which separates approximately from both around and about is its greater degree of abstraction and conceptual complexity (reflected in its stylistic 'bookishness' and formality). For example, approximately can apply to relationships between sizes or dimensions rather than to straight numbers: This line is approximately (/*around) twice as long as that. This block is approximately (/*around) three times as long as it is wide.
English quantitative particles
359
The action was repeated five times, at approximately (I*around) equal intervals.
Being a more 'learned' and abstract word, approximately suggests also a degree of control and mental discipline. It doesn't sound like a rough estimate, or a rough 'shot', which could reflect a lack of concern for accuracy, but rather, like a purposeful device, revealing a respect for precision even at times when the speaker feels precision is not called for, or not possible. Needless to say, it is not easy to reflect such subtleties in semantic formulae. As a first approximation, the following formula may be considered: approximately if it is more than this, it is not much more than this if it is less than this, it is not much less than this I say this word, not another word, because it is easy to think of this word I don't want to say 'it is this' because I know it is something a little different from this
It is interesting here to compare approximately with exactly, which may seem to be a kind of 'anti-approximative', symmetrically opposite in meaning to approximately. In fact, however, there is an important difference between the two (in addition to the obvious similarities): exactly, like about, can appear in purely qualitative contexts: It is exactly the same.
whereas approximately cannot: ?It is approximately the same.
To account for this fact, we can assign to exactly the following formula: exactly one can say this I know: someone can think: 'it is not this, it is something a little different from this' I don't want anyone to think this one can say this, not something a little different from this
360
Particles and illocutionary meanings
1.2.3. Roughly
Dictionaries (for example both Webster 1977 and OED 1933) equate approximately and roughly with one another. For example, Webster (1977) says: "rough - approximate (a rough guess)". However, in the very example provided to illustrate this presumed identity, approximate couldn't be substituted for rough (*an approximate guess). Similarly, one c~n have a rough estimate but not an *approximate estimate. Generally speaking, roughly implies a quick action, and a willingness to sacrifice precision in favour of convenience, ease and simplicity. Approximate has no such connotations, and on the contrary, while signalling a lack of accuracy it manages to convey a respect for accuracy. Another important difference between roughly on the one hand and approximately (and, for that matter, around) on the other, has to do with the qualitative, non-quantitative character of roughly. The sentence: This block is roughly twice as long as it is wide.
sounds fine, but the following two sound a little less felicitous: It was roughly five 0' clock. There were roughly 50 people present.
Furthermore, roughly can apply to purely qualitative matters, as in the following sentence: My idea, roughly, is this. Approximately is not restricted to numbers and quantities, but it does seem to be restricted to matters of form, where an accurate reproduction of all the parts is possible. For example, one can say, I think: The meaning of the word X can be stated, approximately, as follows.
if one assumes that there is a unique optimal semantic formula which one may try to approximate, better or worse. One can hardly say, however: ?My idea is, approximately, this. Approximate(ly) implies that accuracy is possible, rough(ly) doesn't. One can do a rough sketch for a future painting, but not an approximate sketch. Finally, roughly is much further from the target than approximately: while approximately implies that what is said is no more than a
English quantitative particles
361
little different from what is true, roughly implies a difference that can be, and probably is, quite considerable. For example, a rough sketch is in some respects like the real thing, but it is in all probability more than a little different from it. roughly it is not this it is something like this I say this because it is easy to say this I want to say something that one can say quickly
1.2.4. Almost and nearly Almost seems to have two rather different uses which can be called, very roughly, 'gradual' and 'non-gradual'. These two uses can be illustrated with sentences like these two: He is almost bald/blind. He almost killed her.
(gradual) (non-gradual)
Almost} implies, rather like about, that one is describing a certain state or situation in a way which is slightly inaccurate. Almost2 occurs in sentences referring to various 'narrow escapes'. In the former case, omitting almost would make the sentence a little different from what is true, while in the latter, it would make it patently false. For example, the difference between 'being bald' and 'being almost bald' is slight, whereas the difference between 'dying' and 'almost dying' is very considerable. The labels 'gradable' (or 'gradual') and 'non-gradable', (or 'non-gradual '), however, are not particularly apposite, since the former applies also to numerical contexts, as in almost twenty, where strictly speaking there is nothing 'gradable' or 'gradual'. But it is not labels that are important but semantic formulae. If we assume that there are two distinct (though of course related) senses of almost we could propose for them the following two explications: almost} (a) one can't say this (b) if I said this it would be no more than a little different from what one can say
362
Particles and illocutionary meanings
almost2 (a) one can't say: this happened (b) if something was no more than a little different from what it was one could say this
It is worth noting here that in some languages, for example Polish (see section 4.2 below), there are two separate lexical items corresponding to these two different uses of almost: On jest prawie lysy. 'He is almost bald.' On 0 malo jej nie zabil. 'He almost killed her.' (lit. 'By a little he didn't kill her. ')
Since, however, English doesn't distinguish lexically the two senses corresponding to prawie and 0 malo nie, good methodology requires that we should at least make an effort to find a unitary semantic formula for the two uses, or for the common core of these two uses. The following rough formula suggests itself as a possibility: if something was no more than a little different from what it is (was) one could say this In support of such a unified formula, one could point out that while there is more than a little difference between being killed and being 'almost killed', there may well be no more than a little difference between a shot that kills and a shot that 'almost kills' (but misses); or between a rage that leads to a killing and a rage that almost leads to a killing. However, assuming that we have reduced the two apparently different senses of almost to one, we still have one very serious problem to resolve: we have to explain why almost - in contrast to around, about and approximately - seems to approximate the target, so to speak, from below, not from both sides. An expression such as around 50 can refer to situations where the actual figure was more than 50 (say, 52 or 53), whereas almost 50 cannot be so used. This difference between almost and around could of course be easily accounted for if we posited for almost one additional component: it is a little less than this But this simple solution runs into difficulties when one considers sentences such as:
English quantitative particles
363
Almost nothing was left. Almost nobody came. Even if we did posit two different meanings for almost, it would not help us, since in these sentences almost is not used in the sense of narrow escape; clearly, it is not almostz but almost 1. Sadock (1981) has argued against the inclusion of the component 'less than X' in the semantic representation of almost, and I think that the acceptability of expressions such as almost nobody and almost nothing supports his position in this respect. On the other hand, his conclusion that almost means virtually the same thing as about and around (roughly, 'not very different from what it is ') can hardly be accepted, in view of the radically different interpretations of expressions such as almost 50 (definitely no more than 50 and in fact less than 50) and around 50 (could be a little more than 50). Sadock tried to account for these differences in interpretations in terms of the allegedly quantitative nature of about. We have seen, however, that about is not restricted to quantitative contexts any more than almost is. At the same time, around, which is restricted to quantitative contexts, is similar to about, in implying a range on both sides of the target point.
around 20
could be 20 could be a little less could be a little more
about 20
could be 20 could be a little less could be a little more
almost 20
a little less than 20 couldn't be 20 couldn't be a little more
I believe that the clue to the dilemma is provided in Paduceva's (1985:74) observation that almost (or rather, its Russian near-counterpart pocti) includes a negative component as well as a positive one. A sentence such as
Almost 50 people came to the party. conveys the idea that one couldn't say (truly) of any 50 people that 'these people came to the party'. The sentence:
364
Particles and illocutionary meanings
Around (about) 50 people came to the party. doesn't imply that. I suggest that it is this negative component of almost which creates the impression that in numerical contexts almost implies 'less than'. Around and about do not have such a component, and this is why they never appear to imply 'less than'. As a final point concerning almost, it should be mentioned that like around (and unlike about), it implies a process of 'rounding'. One can hardly say: ?Almost seven people came although one can very well say: Almost twenty people came. Almost requires, therefore, the 'rounding' component: I say this because it is easy to think of this Sadock (1981 :267) asserts that "with the exception of a few idioms such as about to and just about, almost can occur wherever about can". But in fact one can hardly substitute almost for about in a sentence such as:
About six or seven people came. The 'rounding' component postulated here for almost accounts for this difference. Sadock notes also that if, say, the actual number of demonstrators on a particular occasion was 950, then the sentence:
Almost 1000 demonstrators picketed. seems truer than:
Almost 990 demonstrators picketed. The 'rounding' component posited here explains such facts. '1000' may well seem the closest number bigger than 950 that could be chosen on the grounds of being 'easy to think of' (in comparison with '950'). But there is no reason why '990' should be thought of in such terms. However, unlike around, almost is not in any way restricted to numbers, and unlike approximately, it is not restricted to formal relationships. One can say:
This block is almost twice as long as it is wide. but one can also say:
At such moments, she seemed almost pretty.
English quantitative particles
365
whereas one can't say: At such moments, she seemed *approximately (*around, *about) pretty. Interestingly, almost differs also in this respect from nearly, which is otherwise analogous to almost in many respects: *At such moments, she seemed nearly pretty. The fact that one can say: We are nearly (*around, *about, *approximately) there. suggests that nearly is not restricted to numbers or numerical relationships. I presume that the reason why one can't say nearly pretty is that nearly is so to speak 'upward graded' (cf. Sapir 1949), in other words, it is viewed in terms of 'adding more of the same'. But it is not clear what exactly would have to be added to make someone pretty if they are 'almost pretty'. To put it differently, nearly appears to refer to a process, whereas almost can be purely stative. For example, one would be more likely to say (a) than (b):
a. When I first saw her she was almost naked. b. When I first saw her she was nearly naked. The phrase 'nearly naked' is not impossible but it implies that one watches the process of undressing, as in: By that time, she (the stripper) was nearly naked. To account for this processual character of nearly we could posit for it the component: if a little more of the same happened one could say this But nearly is also 'upward graded' in a different sense, which can be illustrated with the following contrasts: nearly everything / *nearly nothing nearly everyone / *nearly no one almost everyone / almost no one almost everything / almost nothing
366
Particles and illocutionary meanings
To account for these contrasts we can (as a first approximation) postulate for nearly the component 'it is a little less than this', which, as we saw, cannot be assigned to almost. Thus, the following two semantic formulae can be proposed for almost and nearly:
almost one can't say: it is this if something was no more than a little different from what it is one could say this I say this like this because it is easy to say it like this nearly one can't say: it is this if no more than a little more of the same happened one could say this it is a little less than this The analysis of almost proposed here is very different from the 'radical pragmatic' one proposed in Sadock (1981). According to Sadock, "the meaning of almost is such as to make a statement of the form almost P true just in case there is a possible world in which P is true that is not very different from the real world" (1981:258-259). Sadock believes that all the aspects of the use of almost can be explained in terms of this meagre definition strengthened only by 'heavy doses of Gricean pragmatics'. But Sadock's definition would apply to nearly as much as to almost. It would also apply to about and to many other 'approximatives' in English and in other languages. So how can such a definition, even combined with the heaviest possible doses of Gricean pragmatics, account for the differences in the range of use of all such expressions? I contend that what is needed to account for all such differences, as well as similarities, in the use of related expressions, is a 'radically semantic' approach, not a 'radically pragmatic' one.
English temporal particles
367
2. English temporal particles English has a number of temporal particles, related to one another in rather intricate, and certainly intriguing ways. There is a considerable literature devoted to these particles or to their counterparts in other languages, in particular, in German, Polish, and Russian (cf. e.g. Bankowski 1971, 1975a,b, 1976, 1977; Doherty 1973; Grochowski 1986; Konig 1977; Moiseev 1978; Pasicki 1976; Shetter 1966; Traugott Waterhouse 1969). I do not survey this literature here, partly for reasons of space, and partly because, with the exception of Grochowski (1986), it doesn't propose any explicatory semantic formulae which would attempt to account for the range of use of the particles under discussion. Two of these particles, already and still, appear to be mutually symmetrical, at least in the sense that they both refer to an expected change, and relate this change to an expected time: in the case of already, the change occurred, and it occurred before the expected time, whereas in the case of still, the change did not occur before the expected time, and in fact, it did not occur at all. He still hasn't come. it (the situation) is the same (as before this time) one could think: something would happen it would not be the same because of this it would happen before this time it did not happen before this time He has already come. it (the situation) is not the same (as before this time) one could think: something would happen it would not be the same because of this it would happen after this time it happened before this time
In the case of sentences with stative predicates the reference to an event is implicit, but it is still there; for example (cf. Traugott Waterhouse 1969:302): He is still young (*old). he is the same (as before this time)
368
Particles and illocutionary meanings
one could think: something would happen he would not be the same because of this it would happen before this time it didn't happen before this time He is already old (*young). he is not the same (as before this time) one could think: something would happen he would not be the same because of this it would happen after this time it happened before this time
The word 'it' in the last two lines of these explications refers to some expected change. Consequently, these explications explain why the starred variants are not felicitous: it makes sense to refer to an expected change from 'not old' to 'old' or from 'young' to 'not young' but not the other way around. Still is also closely related to yet, as they both refer to an expected change which hasn't happened, or possibly hasn't happened. But in the case of still, the change was expected to happen before a certain time, whereas in the case of yet it was expected to happen at some unspecified time, not necessarily before the reference time. For example: It still hasn't happened. it (the situation) is the same one could think: something would happen it would not be the same because of this it would happen before this time it didn't happen before this time It hasn't happened yet. it didn't happen before this time one could think: it would happen one can think: it will happen
It is easier to interpret the sentence with still than the sentence with yet as impatient; and it is easier to interpret the sentence with yet than the sentence with still as confident: if the change was expected to occur before now, and hasn't, the speaker may well feel impatient (and perhaps
English temporal particles
369
even doubtful as to whether the change will occur at all); if the change was expected to occur at some time, not necessarily before now, there is no reason to feel impatient, and there is every reason to be confident that it will occur later. Another interesting difference between yet and still has to do with the fact that yet occurs, as a rule, with sentences which are either negative or interrogative, whereas still is also used in affirmative declarative sentences. This difference is related to the fact that still indicates the continuation of a state ('it is the same as before') whereas yet does not have this component: He still hasn't come. it (the situation) is the same it didn't happen it didn't happen before this time He hasn't come yet. it didn't happen it didn't happen before this time
This explains the following contrast in acceptability: Has he come yet? *Has he still come?
The starred sentence is, semantically, self-contradictory, because the speaker appears to be asking two incompatible questions at the same time: a. is the situation the same (as before now)? b. did he come (before now)? But 'his' coming would mean that the situation is not the same. Yet does not refer to any sameness of the situation, and consequently, the unstarred sentence above is not self-contradictory; it implies only (b), not (a). One feature which all three particles discussed here share is their relative character: none of them allows for the event to be 'dated', they only relate the (unspecified) time of the event to the reference time. Consequently, they don't co-occur with simple past (of action/event verbs): He still hasn't come. He has already come. He hasn't come yet.
370
Particles and illocutionary meanings
*He still came on Friday. (possible with a non-temporal meaning of still) *He came already on Friday. *He didn't come yet on Friday.
On the other hand, events can be dated with the particle only, which, I suggest, has a separate temporal meaning in addition to those considered earlier. This temporal meaning can be seen in sentences such as these: He came back only after he'd squandered all his money. He came back only when he realised that he wouldn't win.
This sense may be called onlYt (for 'temporal'). It can be spelt out as follows: onlYt it happened at the time when something else happened it didn't happen before this time one could think: it would happen before this time
The first component of this explication shows that the event is 'dated', but it also shows that it is 'dated' by reference to some other event. If we said, instead, 'it happened at this time', we would fail to account for the fact that sentences such as the following one are not acceptable (in the relevant sense): ?He came only at five 0' clock. (cf. He didn't come until five 0' clock.)
In a sense, therefore, all the temporal particles discussed here are relative, though in different senses: still, already and yet are relative in the sense that they don't specify the time of the event, but only situate it with respect to the time of reference; onlYt is relative in the sense that while it does specify the time of the event it specifies it not in absolute terms but in relation to some other event. To see the meanings of the English temporal particles more clearly I shall compare them now with their closest counterparts in another language: Polish, with additional mention of their counterparts in Yiddish and German.
Polish temporal particles
371
3. Polish temporal particles 3.1. Jut and jeszcze Leo Rosten (1968:xv) has noted that the following phrases can be frequently heard in American English, presumably due to Yiddish influence: All right already. This I need yet?
Phrases of the same kind can be heard in the speech of Polish immigrants in English-speaking countries, presumably for the same reason: Polish, like Yiddish, has temporal particles which appear to be very close to already, still, and yet, and which are, consciously or subconsciously, equated with them in the speakers' minds, whereas in fact the semantic structures encoded in these Polish or Yiddish particles differ, in some ways, from those encoded in the English ones (cf. Pasicki 1976). The Polish particles in question are jUi, often translatable as already, and jeszcze, often translatable as yet or still. For example: On jut tam jest. 'He is already (jut) there.' On jeszcze tam jest. 'He is still (jeszcze) there.' On jeszcze nie przyszedl. 'He hasn't come yet (jeszcze).' On jut przyszedl. 'He has already (jut) come.'
But there are also many types of contexts where jut cannot be translated as already, and where jeszcze cannot be translated as either yet or still. For example: Jut czy jeszcze? 'Jui or jeszcze?' (cf. *Already or still? *Already or yet?)
372
Particles and illocutionary meanings
What the Polish question really means can be stated as follows: 'No more or some more?' i.e. '(is this) enough or (do you want) some more?' This is not to imply that jut can always be translated as no more or jeszeze as some more: jui and jeszeze are, essentially, temporal particles, and they can be used only in contexts which have something to do with time (or which can be interpreted as having something to do with time). Nonetheless, the fact that a virtually context-free sentence such as Jui ezy jeszcze? means something like 'Some more or no more?' is highly instructive: it suggests that the two particles include components crucially involving the notion of 'more'. Similarly, jeszcze would be translated into English as more in the following context: Jeszeze raz! 'Once more.' (lit. 'jeszcze once')
Compare, also, the following context, where jeszeze would normally be translated by even rather than by still: Zosia jest bardzo ladna, ale Rasia jest jeszeze ladniejsza. 'Sophie is very pretty, but Barbara is even prettier.' Jeszeze seems to imply here, roughly: 'more; one wouldn't expect to hear this after what I said before'. This is, then, the essence of my hypothesis: jut and jeszeze differ from already and still in so far as the former set is based on the idea of 'more', whereas the latter is not. As a first approximation, I would propose an explication along the following lines: On tam jeszeze jest. ('He is jeszcze there. ') one could think: it would not be the same as before this time it is the same as before this time it is more of the same
A very similar, virtually identical explication applies also to those sentences where jeszcze has to be translated by means of even rather than still: Zosia jest bardzo ladna, a Rasia jest jeszeze ladniejsza. ('Sophie is very pretty, and Barbara is even prettier. ')
Polish temporal particles
373
(I say: ) Sophie is very pretty one could think: 1 wouldn't say the same about Barbara as 1 said before (about Sophie) 1 want to say the same about Barbara more of the same On juz tam jest. ('He jui is there. ') one could think: it would be the same as before this time it is not the same as before this time it is not the same any more
Consider, now, a sentence like the one cited by Rosten: Jui dobrze. 'Jui all right.' / 'All right already.'
This sentence could be used, for example, in soothing a child: Here, here. That's enough. It is all right now. ("All right already." )
The Polish sentence implies: it is not the same any more (so there is no need for any further crying) But while a paraphrase in terms of 'not any more' makes sense here, one phrased in terms of 'it happened before this time' would not, and this is, 1 suggest, why already in its standard-English meaning cannot be used here. Consider now the other sentence cited by Rosten: This I need yet? which can be seen as an attempted English rendering of sentences such as the following Polish one: Tego mi jeszcze potrzeba? 'This 1 need jeszcze?' The sentence is sarcastic and it implies: 'do 1 need this, on top of everything else?' or 'do 1 need any more (implied: misfortunes, troubles)? do 1 need this (trouble)?' Again, the sentence This I need yet?
374
Particles and illocutionary meanings
sounds odd in standard English partly because yet doesn't imply 'more', which is clearly what the speaker assumes it does. In addition, yet carries the implication: 'one can think: this will happen', which doesn't make sense in the present context. But the Polish sentence is felicitous, because the implications of jeszcze do make sense in this context: Tego mi jeszcze potrzeba? (implied: Do I need more (troubles)?) one could think: it would not be the same as before this time it is the same as before this time it is more of the same (trouble)
I must admit, however, that while the picture presented here appears to be coherent and to have considerable explanatory power, there are some other uses of jui and even more of jeszcze, which remain puzzling and which don't seem to fit this picture. What I have in mind is, above all, the use of these particles in 'dated' sentences such as the following ones: Stalo si~ to jui w pifltek. 'It happened "already" on Friday.' (i.e. it happened as early as Friday). Stalo si~ to jeszcze w pifltek. 'It happened "still" on Friday.' (i.e. it happened as early as Friday).
One puzzling feature of the sentences above is that both jui and jeszcze seem to mean here the same ('as early as'); and yet the feel of the two sentences is quite different. I would suggest that the difference consists here largely in the speaker's perspective on the event: jui suggests a point of view predating the event, whereas jeszcze - a present point of view, that is (in this case) one post-dating it. Roughly speaking, then, jui should be glossed here not as 'as early as' but as 'as quickly as', whereas jeszcze should indeed be glossed as 'as early as'. This distinction appears to be supported by the use of these particles in 'dated' future sentences: Stanie si~ to jui jutro (?jui w tym tygodniu). 'It will happen jui tomorrow (?jui this week).' (i.e. this event will come very quickly)
Polish temporal particles
375
Stanie si~ to jeszcze w tym tygodniu (?jeszcze jutro). 'It will happen jeszcze this week' (?jeszcze tomorrow).' (i.e. it will happen very early)
But why is it that in future dated sentences jut requires precise dating whereas jeszcze appears to require a reference to periods (which include 'now')? At this stage, I am not able to offer paraphrases which would account correctly for all aspects of the use of these particles. It is interesting to note that all the uses of the Polish particles jut and jeszcze discussed here are shared also by the German particles schon and noch. Even the two Yiddish-English sentences quoted earlier have their exact counterparts in German: All right already! Schon gut! This I need yet! Das fehlt mir noch!
And yet the meanings encapsulated in the German particles schon and noch must be slightly different from those encoded in the Polish particles jui and jeszcze, because in some kinds of contexts the former can be used and the latter not. For example: a. Du wirst noch sehen! Jeszcze zobaczysz! 'You will see noch/jeszcze (yet)!' b. Du wirst schon sehen! *Jui zobaczysz! 'You will see schon/*jui!' c. Das wird dir schon leid tun! Jeszcze tego potalujesz! 'You will regret this schon/jeszcze (yet)!' d. Das wird dir noch leid tun! *Jui tego poialujesz! 'You will regret this noch/*jut!' Finally, as pointed out to me by Gerda Smith (p.c.), in German, noch and schon can co-occur, whereas jui and jeszcze cannot co-occur, any more than already and still or yet can:
376
Particles and illocutionary meanings
Das wird dir schon noch leid tun! *Jeszcze jui tego pozafujesz! 'You will regret this schon noch!' *You will regret this yet already!
I do not attempt here to construct explications which would account for both the similarities and the differences between these German particles and the others discussed here, although the task is most challenging and tempting.
3.2. Dopiero Dopiero is another intriguing Polish particle, which seems to lend itself to diametrically opposed interpretations: Przyszla dopiero 0 czwartej. 'She came only at four o'clock.' (i.e. late) Jest dopiero czwarta. 'It is only four o'clock.' (i.e. early)
Similarly: Dopiero po uplywie kilkunastu minut ukazal si~ we drzwiach mfody chlopiec. CZeromski) (SIP) 'Only after about fifteen minutes a young boy appeared at the door.' Dopiero si6dma. Do dziewifltej jeszcze dwie godziny. (Morcinek) (SIP) 'It's only seven o'clock. Two more hours are left before nine o'clock. '
It is very implausible, however, that the same particle would have two diametrically opposed meanings, 'late' and 'early'. One interpretation ('early') arises in sentences 'dating' the reference time itself; the other ('late') arises in sentences dating the events spoken of. A unified semantic analysis can be sought along the following lines: it happened at this time it didn't happen before this time one could think: it would happen before this time
Polish temporal particles
377
It is easy to see how the formula sketched above (which is in fact rather similar to that proposed by Grochowski 1986: 102) can apply to sentences referring to specific events, for example: Stalo si~ to dopiero w pifltek. ('It happened dopiero (not before, as late as) on Friday.') it happened at this time it didn't happen before this time one could think: (it would happen) before this time
It is not so easy to see how the same formula could apply to sentences expressing nothing but time. It seems to me, however, that the same formula can be seen as applying here, too, if we assume that the speaker views time as moving in a certain direction. For example: It is dopiero (no later than) noon. (i.e. it is noon now) noon came now it didn't come before now one could think: it would come before now Of course, objectively speaking, noon cannot 'come' sooner or later, since time always 'moves' at the same speed; but in the speaker's subjective impression, time can move fast or slowly, and a particular point in time (e.g. noon), can come sooner or later than one would (subjectively) expect. I would suggest, nonetheless, that although all the uses of dopiero have a common core, polysemy may have to be postulated to account for its use in those numerical contexts where dopiero clearly implies 'no more'. For example: Mam dopiero dwoje dzieci. ('I have dopiero two children. ') I have two children no more (so far) one could think: it would be more one could think: it would happen before this time it didn't happen before this time
Consider also the common idiomatic expression dopiero co 'just a moment ago' (lit. 'dopiero what'), where dopiero seems to imply a very short time. But the explication proposed earlier fits this use of dopiero, too.
378
Particles and illocutionary meanings
- Kiedy to si~ stalo? ('When did it happen?') - Dopiero co. it happened now (a moment ago, no more) one could think: it is more one could think: it happened before this time it didn't happen before this time
It is easy to see that the Polish particle dopiero corresponds, to some extent, to that use of the English particle only which I have singled out as onlYt" In the case of only, however, we could perhaps accommodate frame-setting sentences such as It is only four
0' clock.
under the basic, general sense of only ('no more than'). In the case of dopiero, this cannot be done because dopiero always implies a sequence of time, even in the purely numerical contexts such as 'dopiero two children'. Furthermore, we noted that onlYt cannot be used in sentences dating events without reference to other events:
?H e came only at four
0' clock.
(i.e. very late)
To account for this restriction on the use of onlYt' I have postulated for onlYt the following component: it happened when something else happened Since no similar restriction applies to dopiero, the relevant component of dopiero has been phrased differently: 'it happened at this time'.
Polish quantitative particles
379
4. Polish quantitative particles 4.1. Non-approximative particles
4.1.1. Tylko Tylko corresponds fairly closely to the English particle only and its
meanings can perhaps be represented by the same semantic formulae: Przyszlo tylko pifC os6b.
'Only five people came.' tylko l
= onlYl
no more than this one could think: it would be more On tylko robi to na zlosc.
'He's only doing it to annoy us.' To tylko ja.
'It's only me.' tylko 2 = onlY2
this no other than this one could think: it would be more than this Admittedly, we have seen earlier that only and tylko can't always be substituted for one another. Thus, only can be used in sentences referring to the passage of time, but tylko can't: It's only five 0' clock! *Jest tylko piflta!
If I have, nonetheless, posited the same semantic formulae for tylko and only, it is on the assumption that only has a third (temporal) meaning, which tylko doesn't have. In support of the suggestion that only has one extra meaning, ambiguous sentences such as the following one can be adduced: He only did it when it was clear that he wouldn't win.
380
Particles and illocutionary meanings
On one interpretation, only implies 'at no other times' and can lated into Polish as tylko. On another interpretation, however, a more likely one) it implies 'no earlier than', and cannot be in Polish as tylko. Instead, a different Polish particle has to (dopiero):
be trans(perhaps rendered be used:
Zrobil to dopiero wtedy, kiedy bylo jasne, ie nie wygra. 'He did it only (no sooner than) when it was clear that he wouldn't win.'
4.1.2. At The particle ai is, in some ways, an opposite of tylko. If tylko implies that 'one could think that it would be more', ai implies that 'one could think that it would be less'. Often, to replace at with tylko one would have to replace, at the same time, much with little, or for a long time with for a short time, and vice versa. Wizyta trwala dlugo, bo ai do p6znego wieczora. (Orzeszkowa) (SJP) 'The visit lasted for a long time, in fact right up to the late evening.'
We could try, therefore, to posit for ai a meaning which could be a mirror image of that posited for tylko: no less than this one could think: it would be less However, although symmetrical patterns of this kind are always appealing to the analyst, symmetry should not be cherished over and above empirical accuracy. It should be pointed out, therefore, that there are contexts where the differences in the behaviour of ai and tylko can't be explained in terms of the neat formulae sketched in above. In particular, these formulae don't explain why ai can't very well co-occur with an imperative, whereas tylko can: Daj mi tylko pi~c! 'Give me only five!' ?Daj mi at pi~c! 'Give me as many as five!'
Polish quantitative particles
381
I think that this difference between at and tylko may be connected with the fact that at seems more emphathic, more expressive, more, so to speak, 'surprised' than tylko. Tylko, like only, is fully compatible with a purely descriptive, objective tone. But at isn't. To account for this fact, it seems more appropriate to formulate the relevant component of at as follows: one would have thought: it would be less The symmetrical component: 'one would have thought: it would be more' would not be suitable for tylko, or for only, which don't sound 'surprised' but merely 'cautious'. I posit, therefore, the following semantic formula: at this not less than this this is much one would have thought: it would be less
English has no particle corresponding to at. The closest to it is the particle-like expression as many as (as much as), but at is not similarly restricted to numbers and measurable quantities. For example, one cannot say in English: ?The visit lasted as long as up to the late evening.
whereas as we have seen, at can very well be used in this kind of context.
4.1.3.
~aledvvie
~aledvvie
combines the roles of a particle and of a conjunction. It occurs either in concessive clauses or in temporal ones, with slightly different force in each case. The concessive use can be illustrated by the following sentence: Nie skartyl si~ na przem~czenie, choc od paru dni sypial zaledvvie po par~ godzin na dob~. (L. Bartelski) (SIP) 'He didn't complain of fatigue, although for the last few days he had been sleeping only (zaledvvie) a few hours per day.'
382
Particles and illocutionary meanings
The semantic contribution of zaledwie in such sentences appears to be this: zaledwie this no more (time) than this this is not much one would think: it would be more
Usually, the 'substance' whose quantity is being estimated is time. I have put the word time in parentheses, however, to indicate that sentences with zaledwie referring to other 'substances' can also be encountered and can be accepted (at least by some native speakers): Dala mu zaledwie pi~c dolarow (ale on si~ nie obrazil). She gave him only five dollars (but he didn't take offense).
The concessive context improves, in my judgement, the acceptability of the sentence, but again there seems to be some variation in this area. In temporal clauses, non-concessive zaledwie signals an even smaller amount of time ('next to nothing', 'hardly any') than it does in concessive ones. Zaledwie wzifllem pioro do r~ki, zapukal ktos do drzwi. (Jan Lam) (SJP) 'I had hardly picked up my pen when somebody knocked on the door'.
In sentences of this kind zaledwie signals an almost direct succession of events. It also conveys the idea that the quick succession of events is unexpected. An explication of the temporal meaning of zaledwie (zaledwie 2 ) is not attempted here.
4.1.4. Ledwie Ledwie is closely related semantically, as well as morphologically, to zaledwie, and both can often be rendered in English as hardly. Yet the two are not always interchangeable, and the semantic contribution of each is distinct. In particular, ledwie is not restricted to time, in the way zaledwie is (in the modem usage). The use of ledwie can be illustrated with the following sentences:
Polish quantitative particles
383
Ledwi~
sie wcisn~lam. 'I just managed to squeeze in.'
Ledwie go poznalam. 'I could barely recognise him.' Ledwie mi wystarczylo. 'I had barely enough.' Ledwie implies that there was 'just enough' of something for X to happen, and that had there been even a tiny bit less of it, X wouldn't have happened. For example, had the friend changed just a little more than he had the speaker wouldn't have recognised him. Had the speaker had even a tiny bit less money than he had it wouldn't have been enough. Had there been even a very little less room than there was the speaker wouldn't have managed to squeeze in. And so on. The link with zaledwie, especially with zaledwie 2 , is evident. In the case of zaledwie 2 the margin of time is 'next to nothing', and in the case of ledwie the margin of whatever is 'next to nothing'. It can even be the margin of time, as in the sentence: Ledwie zdflzylam. 'I just made it.' (i.e. 'I barely made it. ')
But ledwie can be used in a simple sentence, where it doesn't refer to the relationship between two events X and Y. Zaledwie 2 can't be used like that: *Zaledwie zdflzylam. Zaledwie 2 stresses the minimal difference in time between two events. By contrast, ledwie stresses the minimal 'distance' between the happening and the non-happening of an event. To say that something ledwie happened is close to saying that it nearly (almost) didn't happen. Ledwie zdflzylam na poci{lg. 'I just made it to the train.' = 'I nearly missed the train.'
Furthermore, ledwie seems to imply that if the event in question didn't happen it would have been unfortunate, undesirable. For example, one can't say: *Ledwie si~ spoiniles. 'You just missed it.'
384
Particles and illocutionary meanings
as one can say: Ledwie zdflzyles. 'You just made it.'
In fact, ledwie typically co-occurs with 'verbs of success', such as zdpzyc 'make it on time', wystarczyc 'suffice', zdac 'pass an exam, successfully', udac si~ 'manage', wytrzymac 'bear' or wycipgnpc si~ z czegos 'pull through'. The following semantic formula can be proposed for ledwie: ledwie if something was no more than a little different X would have happened this would be bad
4.2. Polish approximative particles
4.2.1. 0 malo nie The expression 0 malo nie (literally 'by little not', i.e. 'but for a little margin it wouldn't have happened') is frequently interchangeable with ledwie - not in the same sentences but in the same situations: Ledwie zdflzylam na poci{lg. 'I barely made it to the train.'
o malo si~ nie sp6inilam na pociflg. 'I nearly missed the train.' However, ledwie can be applied to an ongoing situation as well as to an event, whereas 0 malo nie (like nearly) applies only to momentary events. Ledwie mi wystarcza. 'I can barely make ends meet.' *0 maio mi nie wystarcza. 'I nearly can't make ends meet.'
To account for this 'momentary' implication of the expression 0 malo nie 1 have formulated its second component as 'X would have happened at that moment'.
Polish quantitative particles
385
Finally, 0 malo nie shares the 'evaluative' character of ledwie (both imply that a disaster, major or minor, has barely been averted). Hence the need for the component 'this would be bad'. The overall formula for 0 malo nie might read: o malo nie if it was no more than a little more X would have happened at that moment this would be bad
4.2.2. Niemal and prawie Niemal (nie+mal, lit. 'not+little') may seem to be just a variant of 0 malo nie, alongside perhaps, omal and nieomal ('not+by+little'). In fact, however, niemal differs from both 0 malo nie and omal (if not from nieomal) in several important ways. First, niemal refers to static states of affairs rather than to events or other dynamic situations. Thus, one can say: Odziez pOS10W byla tak skromna, ie graniczyta niemal z ubostwem. (Mieczyslaw Gomulicki) (SIP) 'The clothing of the envoys was so modest that it almost (niemal) bordered on poverty.' But one could not say: *Niemal si~ spoznilam na poci{lg. 'I nearly (niemal) missed the train.' Furthermore, niemal is free of the negative connotations of omal and malo nie, and can be used in positive contexts:
0
Byla bardzo urocza w swym niemal dziecinstwie, a juz kobiecosci. (Zeromski) (SIP) 'She seemed almost (niemal) a little girl, and already a woman, and was very appealing.' It should be noted that despite what its morphology suggests, niemal can refer to a minimal difference in quality, as well as to a minimal difference in quantity. In other words, what is at issue is not so much the idea that 'if it was a little more then it would be true to say X' as the idea
386
Particles and illocutionary meanings
that 'if it was no more than a little different then it would be true to say X'. For example, one can say: S(l niemal jednakowe. 'they are nearly (niemal) identical.'
Purely quantitative particles, such as blisko, cannot be used like that (see the section below). Prawie is extremely close to niemal, and the two are almost interchangeable. The difference between them may seem purely stylistic, niemal being slightly less colloquial. Nonetheless even here the synonymy is probably not complete. There are contexts where prawie sounds much better than niemal, and evidently not for stylistic reasons. For example: Jui prawie (?niemal) skonczylam. 'I have already almost finished.' Dala mu prawie (?niemal) sto dolarow. 'She gave him almost one hundred dollars.' Przyszlo prawie (??niemal) 20 os6b. 'Almost 20 people came.'
The fact that niemal is least appropriate in numerical and otherwise factual contexts suggests that it has something to do with the speaker's evaluation. Prawie is not similarly restricted. To account for this (slight) difference 1 would phrase the explication of prawie in terms of what 'one can say', in contrast to the explication of niemal, which has been phrased in terms of what 'I want to say'. As a first approximation, then, the following can be proposed: prawie one can't say this if something was no more than a little different one could say this niemal I don't want to say this if something was no more than a little different 1 would say this
However, if these explications were entirely correct, prawie would be identical in meaning to the English particle almost. But in fact, prawie
Polish quantitative particles
387
and almost cannot be identical in meaning, since prawie, unlike almost, cannot be used in sentences which would be made patently false by the deletion of the particle; for example: She almost killed him.
To translate this sentence into Polish, one would have to use 0 malo nie, not prawie. Prawie can only be used in sentences where the deletion of the particle would change the sentence from true to 'almost true', not to patently false; and the same applies to niemal. To account for this important difference between prawie and almost I would propose that their explications can be differentiated as follows: almost one can't say this if something was no more than a little different one could say this prawie one can't say this if one said this it would be no more than a little different from what one can say
For example, the sentence: Sp prawie jednakowe. 'They are almost identical.' implies that while one cannot say 'they are identical', if one did say this it would be no more than a little different from what is true. One cannot say, however: *Prawie go zabila. 'She prawie killed him.'
because this would imply that in saying zabiia go ('she killed him') one would be saying something no more than a little different from what is true (that is, from what one can say). The English sentence: She almost killed him.
is felicitous, because it doesn't imply this. Rather, it implies that if something happened that was no more than a little different from what did happen, it would be true to say 'she killed him'.
388
Particles and illocutionary meanings
4.2.3. Blisko Blisko (literally 'nearly') is indeed very similar to the English particle nearly, and differs from prawie in much the same way as nearly differs from almost. However, blisko differs from nearly, as prawie differs from almost, in not being able to be used in sentences when the deletion of the particle would make the sentence glaringly false: I nearly died of exhaustion. *Blisko umarlem z wyczerpania.
As for the relationship between blisko and prawie, blisko has a narrower range of use. In quantitative contexts, both blisko and prawie can be used: Przyszlo blisko (prawie) sto os6b. 'Nearly/almost one hundred people came.'
In qualitative contexts, however, prawie is still appropriate, but blisko is not: Sfl prawie (* blisko) takie same.
'They are almost (nearly) identical.' As the last example shows, blisko differs in this respect from nearly, as well as from prawie and almost. Consider also the following contrasts: Prawie (*blisko) nikt nie przyszedl. 'Almost nobody came.' Jestesmy prawie (*blisko) na miejscu. 'We are almost (/nearly) there.' 'Jut prawie (*blisko) skonczylismy. 'We have almost (/nearly) finished.
Evidently, blisko should be assigned a semantic formula phrased in terms of 'a little less' rather than just 'a little different'. I would propose the following: blisko it is not this it is a little less than this
Conclusion
389
if one said: 'it is this' it would be no more than a little different from what one can say
5. Conclusion I hope to have shown that even subtle nuances of the meaning of particles can be captured in paraphrases whose empirical adequacy can be verified by substitution in context, as envisaged by Leibniz. The data discussed in the present chapter suggest, it seems to me, that particles offer a particularly fruitful field for structural semantic analysis. The contrast between the volume of literature which has been devoted to just a few, always the same, semantic fields, such as colours and kinship, and the paucity of literature devoted to other highly structured fields, such as particles, seems to me quite remarkable. Examining the citations for a number of particles in a large historical dictionary, such as SIP, I am also struck by the clarity with which such a body of citations reflects on-going semantic change, and by the relative ease with which semantic mutations in the area of particles can be captured in the form of paraphrases of the kind employed here. I conclude that far from being one of the most idiosyncratic and 'fuzzy' areas of the lexicon, particles offer in fact an excellent example of a highly structured semantic domain, and constitute a rewarding field for methodological experimentation.
Chapter 10
Boys will be boys: even 'truisllls' are culture-specific
1. The meaning of tautologies What is the meaning of English sentences like Boys will be boys, Boys are boys, Kids are kids, or Business is business? One popular answer is that such sentences (roughly, X is X) are patent tautologies, and so necessarily true. Their meaning - which is identified with their 'logical form' - can be informally stated as follows: 'For every entity of which it is true to say that it is a boy, it is true to say that it is a boy.' ("A tautology is a symbolic sentence whose truth value is T with respect to every possible assignment. For instance, P -> P is a tautology." Kalish - Montague 1964:74.) That is, it is assumed that the syntactic structure N is N is exactly equivalent to the logical formula p = p. In fact, of course, sentences of this kind convey more. As Levinson (1983:125) puts it, the implication is: "That's the kind of unruly behaviour you would expect from boys." But, according to the pragmatic explanation, this implication is a conversational implicature, calculable from Grice's (1975:45) maxim of Quantity: roughly, "Make your contribution no more, and no less, informative than required." I want to argue against this account, and against the whole vision of linguistics which goes with it. The question of how to interpret sentences such as Boys will be boys may seem minor and unimportant. I think, however, that the consequences of one's stand on this point are far-reaching; they determine one's entire idea of linguistics, its boundaries, its capacities, and its responsibilities. In choosing the expression Boys will be boys as the title of this chapter, I do not wish to imply that this is a paradigm example. It is a frozen expression, whereas the chapter is concerned above all with the productive pattern NP i be NPj" Still, the 'proverb' has been frequently discussed in connection with linguistic 'tautologies' and Gricean maxims, and it has a greater evocative and mnemonic value than, say, A promise is a promise or A man is a man. Although there are of course
392
Boys will be boys: even 'truisms' are culture-specific
other types of tautological utterances, the linguistic literature on 'tautologies' refers largely to sentences of the form X is X. While the construction type considered in this chapter is a minor one, it should perhaps be noted that tautological constructions are not at all rare in colloquial English speech. For example, in three successive episodes (of 25 minutes each) of the American television program 'Bewitched', I counted no fewer than seven occurrences. Some cultural aspects of what things are so 'obviously' true that they can be stated in the form of a 'tautology' will also be considered here.
1.1. Gricean maxims: universal or language-specific? My first objection to the Gricean account of 'tautologies' is that it suggests that the import and use of such constructions should be calculable from some universal, language-independent principles. In fact, however, it is not. For one thing, some English 'tautological constructions' have no literal counterparts which can be used or interpreted in many other languages. Some English tautological constructions do have literal counterparts in other languages, which are used, however, with a different communicative import. This suggests that, in each case, the communicative import is conventionally encoded in a given construction, and is not calculable from any language-independent pragmatic maxims. In saying this, I am not arguing against the validity or the significance of language-independent pragmatic maxims like those posited by Grice. I am arguing only against the use to which such maxims have been put in much current literature on linguistic pragmatics, and in particular against attempts to explain the use of English 'tautological constructions' exclusively in terms of universal pragmatic principles, as 'radical pragmaticists' try to do (cf. for example Cole 1981). Instead, I am advocating what might be called a 'radically semantic' approach to the task: I argue that the constructions in question have a language-specific meaning, and that this meaning must be spelt out in appropriate semantic representations. For example, sentences of the kind adduced in the title of this chapter. are not used in French, German, or Russian. French sentences like Lesgarr;ons sont les (des?) gar~ons ' (The) boys are (the) boys', or Les garr;ons seront les (des) gar~ons 'The boys will be (the) boys', would be simply incomprehensible to French speakers. Conceivably,
The meaning of tautologies
393
one could be understood if one said Les gar(;ons seront toujours les (des) garc;ons 'Boys will always be boys' (cf. Bally 1952: 17), but even this would be puzzling. Similarly, in German one would not say Knaben sind Knaben 'Boys are boys', or Knaben werden Knaben sein 'Boys will be boys.' If foreigners did use such sentences, wishing to convey the messages of their literal English counterparts, they might not be understood, as the 'implicature' of the tautological constructions in question would not be understood. The sentence Knaben bleiben (immer) Knaben 'Boys remain (always) boys' is more readily interpretable; but here it is doubtful if its literal English equivalent would be similarly interpreted. Russian has three copula constructions: one with eto, one with est', and one with zero, but none of these would be used to translate the relevant English sentences: ??Mal' tiki eto/est' mal' tiki.
'Boys are boys.' ??Mal'ciki mal'ciki. 'Boys (are) boys.' ??Mal'ciki budut mal'tiki (mal'tikami). 'Boys will be boys.' ??Deti eto/est' deti. 'Kids are kids.'
To translate the English sentences, one would use a particle: (Cego ty xoces'?) Oni ze mal' ciki. (what you want) they PRT boys '(What do you expect?) They are boys.' (Cego ty xotel?) Oni ze deti. (what you wanted) they PRT children '(What do you want?) They are children.'
It should be pointed out, however, that the Russian particle ie is not used specifically for the purpose expressed by English sentences like Kids are kids. Rather, it appears in a variety of constructions, to indicate, roughly, that something should be obvious to the addressee. It is thus similar to the German particle doch (cf. Rath 1975; Sekiguchi 1977), though the two are by no means fully equivalent. However, 'tautological constructions' with action nouns do exist in Russian. For example, one can say, as Bulat Okudzava does in a popular
394
Boys will be boys: even 'truisms' are culture-specific
song, Rabota est' rabota, rabota est' vsegda, 'Work is work, there is always work.' One can also say Vojna est' vojna, 'War is war', but the intended meaning of the Russian expression, and its range of use, would be different from those of its literal English counterparts. For example, in Grossman's novel Zizn' i sud' ba 'Life and fate', one of the heroes uses this expression to express his determination to fight fiercely: This isn't the moment for me to start talking about internationalism or class conscience. What matters is to mobilise the fury of the masses against the enemy.... There's no place for Christian humanitarianism now.... now the Germans have attacked the homeland of workers and peasants. War's war! (Vojna est' vojna!) They deserve what they get. (Grossman 1980:153; 1985:237)
In the English translation above, the tautological expression War's war sounds somewhat incongruous, because this is not how this expression is used in English (see section 2.1 below). Interestingly, in English one can say That's life, but not *Life is life. On the other hand, in Russian, where the tautological pattern in question affirms that one has to do something (rather than put up with something), the expression Zizn' est' zizn', 'Life is life', is perfectly possible. For example: The old woman sitting next to him, the mother of the wife he had loved and now lost forever, kissed him on the head and said: 'It doesn't matter, Stepan my dear, it doesn't matter. It's life. CZizn' est' zizn' .)' (Grossman 1980:605; 1985:867)
The Russian original implies that 'life has to go on', but the English translation implies something different: a sober, 'realistic', worldly-wise attitude to life. Furthermore, in French one can hardly say La guerre est la guerre 'War is war'; to express a similar idea, one would say C' est la guerre 'That's war', just as one says C' est la vie 'That's life'; or La guerre, c' est la guerre, literally 'The war, that's the war'. To widen the sphere of comparison, in Polish one cannot say *Chlopcy to b~d(l chlopcy 'Boys will be boys', or ??Chlopcy to (S{l) chlopcy 'Boys are boys'; however, one can say (Jednak) co Paryi to Paryz '(However) what (is) Paris this (is) Paris', or (Jednak) co Europa to Europa '(However) what (is) Europe this (is) Europe.' I invite the 'radical pragmaticists' to work out the communicative import of this construction. For those readers who acknowledge that they cannot
The meaning of tautologies
395
work it out for themselves, here is an approximate answer. The Co X to X construction ('What is X is X') implies that there is something uniquely good about X, and that the speaker feels he must admit it. (No matter what virtue one may find in other comparable things, one must admit that X is uniquely good.) The construction is used with respect to referents which are well known, and which are widely regarded as uniquely good. Given traditional Polish attitudes, Paris qualifies par excellence; but any other well known symbol could also be so mentioned, in grudging recognition of its widely-acclaimed superiority. Hence, for example, the following contrasts: (Jednak) co Wal~sa to Wal~sa! ? ?(Jednak) co Barbara to Barbara. (Jednak) co rodzina to rodzina. [rodzina 'family'] ??(Jednak) co kuzyni to kuzyni. [kuzyni 'distant cousins']
The Polish construction Co X to X can be seen as related, in certain respects, to the English pattern illustrated by the sentence used recently as an opening statement in a television interview by a Ku Klux Klan leader: White is white. In such sentences, the speaker stresses the unique quality of something which must be accepted because it cannot be expected to change. But qualities like 'whiteness' are seen as belonging to certain contrastive sets; so, by stressing their uniqueness, the speaker emphasises the irreducible difference between the members of the set. The uniqueness is not interpreted here as superiority, but merely as a reason for an irreducible contrast, which must be accepted as such. The attitude encoded in this construction is highlighted in the familiar passage from Kipling's 'Ballad of East and West': East is East and West is West and never the twain shall meet.
In the film A passage to India, based on E. M. Forster's novel, Kipling's saying is used in a reduced version, as East is East; this version seems parallel to White is white (with the second member of the contrastive set being only implied.) In the Polish construction, no finite sets of comparable elements are implied, and so the notion of uniqueness is not tantamount to one of irreducible difference; instead, it is tantamount to a notion of irreducible superiority (with respect to any conceivable competitors).
396
Boys will be boys: even 'truisms' are culture-specific
It is interesting to note that a different but somewhat related 'tautology of admiration' exists in Korean; one can say Apoci-ja apoci ida 'Father is father' (where -ja means, roughly, 'certainly'), or Colsu-ja Colsu ida 'Colsu [a person's name] is Colsu', to express one's admiration for the person referred to. However, Korean also has a tautological construction which is used to express disapproval. For example, seeing a person crying without any good reason, one might say to him, with a scornful laugh, No-to no ida 'You are you' (where -to means, roughly, 'also'). In English (or in Polish), in these circumstances, one would have to say something quite different; for example, It's just like you (that's you allover) - crying for no reason at all. (lowe the information on Korean to an unpublished paper by Gi-hyun Shin.) As a final example, consider the following Japanese sentences, which come from unpublished papers by Itsuo Harasawa (1985) and Deborah Field (1988): Makeru toki wa makeru yo. 'When (I) lose (a game), (I) lose (it).' Kare datte, kekkonsuru toki wa kekkonsuru yo. 'Even he, when (he) marries, (he) marries.' Okoru toki wa okoru. 'When (he) gets angry (he) gets angry.'
Can native speakers of English work out the 'implicatures' of such sentences? Several speakers of English whom I have asked have offered the following: 'When he loses the game, he loses it badly'; 'When he gets married, he will get married in a spectacular way.' 'When he gets angry, he gets really very angry'. Clearly, these guesses are based on the interpretation of English tautologies like the following: "Four cakes! Gee!" said fern. "When we give a party, we give a party," said Susan grandly. (Montgomery 1980:73)
But in fact these guesses turn out to be incorrect. According to Harasawa, the Japanese construction really means that something regarded as quite impossible is actually possible. It would seem that this meaning is language-specific, and cannot be calculated solely on the basis of any Gricean maxims (or 'post-Gricean' ones, cf. Atlas 1984).
The meaning of tautologies
397
Again, I do not wish to deny that the various 'tautological constructions' used in different languages have something in common, and that they may be partly explained in terms of some language-independent principles such as Grice's maxim of quantity. I would insist, however, that their use cannot be fully accounted for in such terms. Of course, most facts of grammar can be partly explained in terms of languageindependent principles - functional, perceptual, logical etc. If linguistic descriptions of particular languages were to be purged of everything that could be so dealt with, then little would be left for the linguist to write about. When philosophers write about the constructions of their native language as if their use were fully determined by the general laws of human reason, or by 'normal' rules of conversational cooperation, they can perhaps be excused: it is not the job of philosophers to compare different languages, or to be aware of both the differences and the similarities between them. But when reputable linguists eagerly and enthusiastically adopt the philosophers' illusions, the situation begins to look like a historical aberration. One can understand Givan's exasperation on this score, expressed in the following outburst (whether or not one agrees with his proposed solution): Nobody would deny the stimulating effect that a first reading of Grice's 'Logic and conversation' [1975] may produce. But to base an entire boomlet, indeed a fad, on this rather limited construction of the pragmatic agenda in terms of Grice's 'maxims', from which all else is presumably derived as deus ex machina, is the climax of in-group folly.' (Giv6n 1983: 154) (Cf. also Boguslawski 1981a.)
1.2. Problems in interpreting implicatures According to Levinson (1983:124), among others, a sentence like Boys are boys is necessarily true. I dispute the validity of this statement, which reflects a mistaken belief that the sentence under discussion is factual. It is clearly not: it expresses a certain attitude, and attitudes can hardly be called 'true' or 'false'. Roughly speaking, it is a call for tolerance, an injunction; and it is no more 'true' than the Ten Commandments, or maxims like Time is money or The early bird gets the worm. Since the attitude in question cannot be fully worked out on the basis of any language-independent principles, it must be regarded as the language-specific meaning of the sentence in question. It is the responsi-
398
Boys will be boys: even 'truisms' are culture-specific
bility of the linguist to identify this construction, and to spell out the meaning encoded in it. To the best of my knowledge, neither task has been undertaken by any of the numerous linguists who have written about tautological sentences in a Gricean vein. A foreigner trying to learn English will have to be taught some rules for correctly using various tautological constructions. It would be perfidious in the extreme merely to draw his attention to the Gricean maxims, and then to leave it to him to work out for himself the permissible range of use of these constructions. An interpretive formula like that quoted above - by which Boys are boys implicates something like 'That's the kind of unruly behaviour you would expect from boys' - says both too much and too little. On the one hand, the epithet 'unruly' is over-specific and arbitrary. On the other hand, the formula is not specific enough, in not spelling out the speaker's tolerant and indulgent attitude. Consequently, it doesn't predict that the following sentences are odd (the behaviour in question being too 'bad' to be treated indulgently): ?Sadists are sadists. ?Rapists are rapists. ?Nazis are Nazis.
The subtle difference in attitude between the sentence Boys will be boys and sentences such as Business is business is implicitly acknowledged in the glosses offered by Cowie's (1976) Dictionary of current idiomatic English. According to this dictionary, the former saying is offered as an 'excuse', whereas the latter constitutes an attempt at 'justifying oneself' (or, indeed, others). I think this differentiation reflects a subtle insight, of the kind that radical pragmaticists or other Griceans prevent themselves from being able to capture. The tolerant and indulgent attitude is even more transparent in the pattern with the modal will: Boys will be boys. Students will be students. *Wars will be wars. *Business will be business.
The pattern with will implies that the nature of the human beings in question cannot be repressed; one shouldn't try to change boys (etc.) because it won't work anyway: whatever one may do, boys will still behave like boys, so the wisest course is just to let them be.
The meaning of tautologies
399
Levinson's formula is not nearly as unsatisfactory as many other similar formulae offered as statements of various 'conversational implicatures', but it still illustrates the detrimental effect of the 'radical pragmatics' doctrine upon the analysis: when linguists assume that the communicative import of a construction is a matter of implicature, rather than meaning, they then seem not to care about how, exactly, this import is to be stated. In a sense, radical pragmatists can't afford the luxury of stating the differences in the communicative import between different tautological constructions, either across language boundaries or within a single language. By assuming an absence of meaning (other than that equivalent to the 'logical form'), they are also forced to assume an absence of polysemy; thus they must either posit a single formula which may be too broad to predict the exact range of use, or else must be rather vague about the details of the alleged implicature. If a methodology requires us to devise a semantic formula (or a few formulae) for each construction, it forces us to be explicit and precise. As a result, differences as well as similarities come to light which would otherwise be missed - or which may be vaguely attributed to differences in 'context', with no precise generalisations being offered or sought. An illustration of this last failing is provided in the next section. Before we turn to this illustration, however, let it be pointed out that an eagerness to make the Gricean maxims work at all costs can be also detrimental to the analysis from a purely formal point of view. Sentences such as Boys will be boys provide a good case in point. For isn't it strange that several writers who have discussed such sentences have followed Grice (1975:52) in focussing exclusively on the less idiomatic variant Boys are boys and have failed even to note the existence of the variant with will? It is hard not to regard this omission as 'purposeful', on an unconscious level. Boys are boys looks like a tautology, so it may seem to be explainable in terms of the 'maxim of quantity'. But Boys will be boys is not a tautology: normally people assume that boys will not always be boys (but will become men). How is it then, that such an obvious non-tautology can have almost the same 'conversational implicature' as a 'tautology' (Boys are boys)? Both the form and the use of the version with will seem to undermine the validity of the Gricean account. As a result, this version is simply left out of the account altogether! I am certainly not suggesting that this is a case of intellectual dishonesty, but I am suggesting that the omission in question is a Freudian omission, so to speak.
400
Boys will be boys: even 'truisms' are culture-specific
1.3. Context as an excuse for analytical failure It will be useful to take, as a starting point, a longish passage from Levinson: The uttering of simple and obvious tautologies should, in principle, have absolutely no communicative import. However, utterances of (38)-(40) and the like can in fact convey a great deal: (38) War is war. (39) Either John will come or he won't. (40) If he does it, he does it. Note that these, by virtue of their logical forms (respectively: Vx [W(x) -> W(x)]; P V ~ p; p -> p) are necessarily true; ergo they share the same truth conditions, and the differences we feel to lie between them, as well as their communicative import, must be almost entirely due to their pragmatic implications. An account of how they come to have communicative significance, and different communicative significances, can be given in terms of the flouting of the maxim of Quantity. Since this requires that speakers be informative, the asserting of tautologies blatantly violates it. Therefore, if the assumption that the speaker is actually cooperating is to be preserved, some informative inference must be made. Thus in the case of (38) it might be 'Terrible things always happen in war, that's its nature and it's no good lamenting that particular disaster'; in the case of (39) it might be 'Calm down, there's no point in worrying about whether he's going to come because there's nothing we can do about it'; and in the case of (40) it might be 'It's no concern of ours.' Clearly these share a dismissive or topic-closing quality, but the details of what is implicated will depend upon the particular context of utterance. (Incidentally, exactly how the appropriate implicatures in these cases are to be predicted remains quite unclear, although the maxim of Relevance would presumably playa crucial role.) (Levinson 1983: 110-111)
To Levinson's credit, he does point out that "exactly how the appropriate implicatures in these cases are to be predicted remains quite unclear". He still pins his hopes, however, on the context of the utterance. I believe that such hopes are bound to be disappointed. Utterances like War is war or Boys will be boys are remarkably contextindependent in their force, as Levinson himself tacitly recognises by trying to spell out their 'implicatures' without asking the reader to imagine any particular context. Grice too (1975:52) discusses the 'conversational implicatures' of such sentences without invoking any particular context. In fact, various dictionaries of 'sayings', proverbs,
The meaning. of tautologies
401
and idiomatic phrases have often listed such expressions, and have also explained their meanings, as best they could, without appealing to any particular contexts (cf. Bartlett 1980; Cowie 1976; Stevenson 1949). The lexical meaning of the word war may indeed influence the 'implicature' of the saying War is war; but this doesn't mean that the construction itself is similar in meaning to Boys are (will be) boys. Both constructions signal a 'philosophical' attitude, a kind of acceptance of bad things which cannot be prevented from happening; but the plural-human construction also signals an indulgent attitude, while the singular-abstract one does not. The element will, which can show up in the sentence as an alternant of are, is not accidental either: it reflects both a reference to the future (you can predict how boys will behave from their nature) and to a characteristic human 'obstinacy' or a tendency to persist in behaving in certain ways, no matter how undesirable from other people's point of view. This will is related to the future will; but it is also related (synchronically) to the noun will and to the modal used in sentences referring to habitual (stubborn) 'undesirable' behaviour (He will smoke while we're still eating). It is worth recalling in this connection Jespersen's (1965) remarks on the use of will, as well as his examples illustrating the semantic links between human will, human nature, habitual behaviour, and a tendency to stubborn persistence in what might seem foolish: Another connected transition is a consequence of the fact that what one does willingly, one is apt to do frequently. Hence will ... comes to be the expression of a habit, especially a habit which is a consequence of one's character or natural disposition ... In the present tense, it does not seem usual in the first person ... , in the second it is often emotionally coloured: 'You will smoke all day long - and then complain of a sore throat!' ... If will is emphasised, obstinacy may be meant: Gammer 102 fooles will be fooles styli! boys will be boys ... (Jespersen 1965,4:240-241)
Once 'human' tautologies, such as Boys are boys, are distinguished from abstract ones like War is war, the appearance of will in some 'tautological' sentences, but not others, begins to make sense. Turning now to sentences like Either John will come or he won't, I would agree with the gist of Levinson's interpretation ('Calm down, there is no point in worrying about whether he's going to come because there is nothing we can do about it. '). But if it is stated in these words, one can't see how this particular 'implicature' is related to the 'implica-
402
Boys will be boys: even 'truisms' are culture-specific
tures' of the other tautological constructions considered. If one says that all the different tautological constructions express a certain philosophical acceptance of undesirable events, one will be more or less right - but only more or less. For example, the nuance of tolerance/indulgence conveyed by Boys are boys and Boys will be boys is lost. (No such nuance is conveyed in 'fatalistic sentences' like Either John will come or he won't.) One might try to defend the claim that the differences between the various 'tautological' constructions can be predicted from the context in the following way. If the sentence refers to future events, the implication is that future events are unknown and must remain unknown (for the present), so there is no point in speculating about them. If the sentence refers to past events, the implication is that what has already happened can't be changed, so it is not worth dwelling on. If the reference is not to any past event, but specifically to a past action by the speaker himself, the implication may be that of determination and refusal to change one's position, as in Pilate's utterance: What I have written I have written. If a sentence has a generic meaning, and predicates something about a 'species', then the implication will be that the nature of an individual is determined in some respects by the nature of the species, and so cannot be changed - and moreover that it should be excused, to the extent that 'bad' behaviour of an individual is determined by the nature of the species. Indeed, the Spanish sentence Que sera, sera conveys not only a philosophical acceptance of what cannot be changed, but also a dismissal of useless speculations about the future. The Polish saying Co bylo to bylo 'What was (has been), this was (has been)' implies that past events cannot be changed, so there is no point in dwelling on them. In fact, this saying has a more explicit version, which has the status of a proverb: Co bylo a nie jest nie pisze sit? w rejestr 'What has been but is not shouldn't be put on record.' Nevertheless, and despite appearances, the exact meaning of sentences which express a fatalistic attitude to future events (like the Spanish Que sera, sera) can also vary from language to language. For example, the best rendering of the Spanish sentence into Polish would be Co rna bye to b~dzie 'What is to be, that will be', and not Co bedzi~ to bedzi~ 'What will be will be'. The latter sentence is used in Polish to express one's determination to act, regardless of possible negative consequences; thus it can be used by a soldier before a battle or by a student before an exam. Characteristically, it is often followed by the proverb Raz kozie
English nominal tautologies: semantic representations
403
smierc 'A goat has to die once', or by the saying Co ma wisiec nie utonie 'What is meant to hang will not drown', both of which express a similar sentiment of reckless determination. The Spanish sentence can be used by an idle person who doesn't want to worry about the future, but its literal Polish equivalent would not be so used. Facts of this kind show, it seems to me, that the use of 'tautological constructions' and 'tautological sayings' in different languages is partly conventional and language-specific - even though it can also be explained, to a large extent, by language-independent 'Gricean' principles.
2. English nominal tautologies: semantic representations My main claim concerning English nominal tautologies is this: English has not one, but many, productive tautological patterns conforming to the formula (ART) N j be (ART) Nj" If we want to state the meaning of these patterns accurately, we must recognise this plurality of types, and state the meaning of each one separately. In some cases, formal clues enable us to separate the different types. In other cases, the form of two different tautological patterns is the same, yet they cannot be collapsed under a simple semantic representation because their meanings differ in ways which cannot be accounted for in terms of context or lexical differences. For this reason, some tautological sentences are ambiguous. Thus A mother is a mother can mean either that a mother can always be expected to act in a motherly way (even if she seems different from other mothers), or else that one has obligations toward one's mother. Similarly, A steak is a steak can mean either that there isn't much difference in value between one steak and another (one is neither much better nor much worse than another), or else that all steaks are undeniably and reliably things of high value. From a formal point of view, the following tautological constructions (among others) can be distinguished in English:
is Nabstr N p1 are N p1 N p1 will be N p1 A N is a N Nabstr
War is war; *Wars are wars, *Wars will be wars. Kids are kids; *The kids are the kids. Boys will be boys; *A boy will be a boy. A party is a party; *The party is the party.
404
Boys will be boys: even 'truisms' are culture-specific
The N is the N The law is the law; *The war is the war. N l is N l (and N 2 is N 2 ) East is East, and West is West.
From a semantic point of view, however, the list of English nominal tautological patterns (which I will call sub-constructions) would present a different picture, as shown by the survey of sub-constructions offered below. While this survey is not presented as an exhaustive one, it covers most of the commonly used types. I should add here that, although it seems appropriate and fruitful to speak of English tautological sentences in terms of a set of subconstructions, we should also note that some such sentences have the status of more or less set phrases, for example Fair is fair; Enough is enough; A deal is a deal; Business is business. However, it is important to recognise that they are not idioms, but rather particularly frequent tokens of productive tautological patterns. Even the sentence Boys will be boys, which must be regarded as the focal member of the class to which it belongs, is not really idiomatic, since its meaning is strictly parallel to that of other sentences based on the same pattern, for example Students will be students or Teenagers will be teenagers. Even if such sentences are seen as variations on the theme of Boys will be boys, one cannot indulge in such substitutions with true idioms, such as He kicked the bucket.
2.1. 'Realism' in human affairs A 'realistic' attitude toward complex human activities is expressed by the following syntactic formula: N abstr is N abstr
Examples: War is war. Politics is politics. Business is business. *Wind is wind. *Sneezing is sneezing. *Wars are wars.
This sub-construction seems to be restricted to complex human activities, and apparently to those which involve human interaction. This is perhaps linked to the complex character of the activity - which can be seen as a special 'way of life', or as a world apart - and to 'inevitable' negative aspects of this activity, which must be understood and tolerated.
English nominal tautologies: semantic representations
405
From a formal point of view, this sub-construction is marked by an absence of articles (as well as by the singular number). One can say A war is a war; but the meaning is different from that of War is war (see 2.8 below). The meaning encoded in the present sub-construction can be represented in the form of a number of interrelated components. These include a reference to a supposed truism that complex activities of the specified kind must have some undesirable consequences. The nature of these consequences is viewed as well known, so that it would be superfluous to spell them out. Further, there is a call for acceptance of those undesirable consequences: since they are inevitable, there is no point in getting oneself into a negative emotional state every time one observes them. The 'nature' of the activity is such that it necessarily entails the undesirable consequences in question. Hence, the War is war type may be explicated as follows: War is war (a) everyone knows: when people do things of this kind something bad can happen to other people because of this (b) I know: someone can think: this is bad it should not be like this (c) I think: one should not think this (d) one should know: it is always the same (when people do things of this kind) it cannot be not like this (e) I don't want to think: this is bad (f) I don't want to feel something bad because of this
2.2. Tolerance for human nature A tolerant, and also 'realistic', attitude toward human nature is expressed by the following syntactic formula: N hum.pI are N hum.pI.
Examples: Boys are boys. Kids are kids. Women are women. Children are children. They are there to be put up with. (Pascal 1981:34).
406
Boys will be boys: even 'truisms' are culture-specific
The meaning of this sub-construction is of course very similar to the preceding one, and one might well be tempted to assign it the same semantic representation. But a semantic formula obliges one to be explicit, and this in tum forces one to pay attention to detail and to subtleties. When this is done, certain clear differences seem to emerge. First, the component 'bad' seems applicable to the 'human activity' type, but not to the 'human nature' type. Wars are said to be 'horrible'; politics is a 'dirty business'; the world of business is 'ruthless'. But when one says that Kids are kids, one doesn't wish to imply anything truly 'bad' about children: they may be noisy, boisterous, unruly, tiresome, but not 'bad'. Accordingly, I would differentiate the relevant components of the semantic representations as follows: (a) everyone knows: when people do things of this kind something bad can happen to other people because of this (a') everyone knows: people of this kind do things like this one would want them not to do things like this Furthermore, I would posit for the 'tautologies of human nature' an additional component, which highlights their benign, indulgent tone: (g) I think: people of this kind are not bad All the other components are essentially the same: the realisation that other people may complain about the undesirable aspects of the situations discussed, the conviction that one should not do that, the recognition of the predictability and the inevitability of those undesirable aspects, and the call for calm and acceptance. But given the differences pointed out above, this call for acceptance will be interpreted differently in each case. In the case of complex human activities such as war it will have the flavour of sober, wordly-wise resignation; whereas in the case of the 'tautologies of human nature', it will sound like a call for indulgence and tolerance. This is linked with the fact that while in the case of human activity the undesirable aspects are seen as a matter of (so to speak) grim necessity, in the human nature type they are seen, rather, as a matter of human weakness. 13 Compare: (d) it cannot be not like this (d') they cannot be not like this
English nominal tautologies: semantic representations
407
This brings us to the following over-all formula for the Kids are kids type: Kids are kids (a) everyone knows: people of this kind do things like this one would want them not to do things like this (b) I know: someone can think: this is bad it should not be like this (c) I think: one should not think this (d) one should know: all people of this kind are like this they cannot be not like this (e) I don't want to think: this is bad (f) I don't want to feel something bad because of this (g) I think: people of this kind are not bad
The indulgent and patronising attitude of human-nature tautologies is particularly pronounced in the subtype with will (Boys will be boys), where the 'immutability' of the phenomenon is presented as due to the willful and uncontrollable spontaneity of a non-serious and pleasureseeking species. The following set of sub-components can be postulated for this subtype: (d) one should know: all people (creatures?) of this kind are the same they want to do things like this because they want to feel something good they will do them because of that they cannot be not like this The indulgent attitude encoded in this construction comes across very clearly in the section entitled "Dogs will be dogs" in Christina Johnson's (1978) book In praise of dogs in Australia. The section, which is devoted to dog exercise, opens as follows: "Australia's extensive beaches, ample parklands and wide open spaces offer ideal opportrunities for dogs to revel in exuberant romps, wild gallops and invigorating swims." It ends on the same note: "Frivolous though they sometimes seem, these romps will not only keep the dog fit; they are essential to his mental well-being." (Johnson 1978:32).
408
Boys will be boys: even 'truisms' are culture-specific
The expression "Dogs will be dogs" is of course jocular, but it is instructive in so far as it highlights two important semantic aspects of the category in question: 'unruliness' (mentioned also in Levinson's offthe-cuff paraphrase) and fun-loving. The relevance of these two features is also corroborated by some other types of English nominal tautologies, to which we will turn next. But first, a formula for the Boys will be boys type: Boys will be boys (a) everyone knows: people of this kind do things like this one would want them not to do things like this (b) I know: someone can think: this is bad they should not do it (c) I think: one should not think this (d) one should know: all people of this kind are the same they want to do things like this because they want to feel something good they will do them because of that they cannot not do them (e) I don't want to think: this is bad (f) I don't want to feel something bad because of this (g) I think: people of this kind are not bad
2.3. Tolerance at 'special times' Syntactic formula: A N is a N
English seems to have a productive tautological pattern which expresses a forebearing attitude towards pleasurable activities taking place during 'special', privileged times, and on special, privileged occasions, such as holidays, birthdays, parties and so on. One can say, for example: A picnic's a picnic. (Nesbitt 1980:222) A party is a party. A holiday is a holiday.
English nominal tautologies: semantic representations
409
A honeymoon is a honeymoon. A game is a game.
One can hardly say, however: ?A Monday is a Monday. ?An autumn is an autumn. ?A morning is a morning.
(unless in a totally different sense, to be discussed below: 'one Monday is not better or worse than another'). The privileges associated with 'special' times such as holidays, parties, etc., may entail inconveniences for other people, but such inconveniences have to be excused and put up with Gust as inconveniences due to the weakness of human nature have to be put up with). One cannot say, however: *A picnic will be a picnic, or *A holiday will be a holiday, as one can say: Boys will be boys, because in the case of picnics, holidays and such, the need for tolerance is based on the 'special' character of a certain time, not on a foreseeable character of a 'species' of (willful) human beings. A picnic is a picnic (a) everyone knows: at times of this kind people want to feel something good because of this they want to do some things they wouldn't do these things at other times (b) I know: someone can think: this is bad it should not be like this (c) I think: one should not think this (d) one should know: it is always like this (at times of this kind) it cannot be not like this (e) I don't want to think: this is bad (f) I don't want to feel something bad because of this (g) I think: this is not bad
Neither can one say: *Picnics are picnics, *Parties are parties, *Honeymoons are honeymoons. Since the stress is here on the 'special' rather than on the general and the predictable, the singular is apparently more appropriate for the category in question; and it can in fact be seen as an icon of this aspect of meaning. But since the 'special' character of
410
Boys will be boys: even 'truisms' are culture-specific
times, and situations, such as picnics, parties, or honeymoons, consists in the suspension of normal 'rules' of life, and since this suspension of rules is linked with a licence for pleasure-seeking activities, this type is in fact closely related to the Boys will be boys type.
2.4. The limits of tolerance The English saying: Enough is enough.
may seem to express a sentiment radically different from that expressed in sayings referring to human nature, such as Boys will be boys. The latter is tolerant and indulgent, the former stern and implacable. In fact, however, the two types are closer than they seem, as they share a common theme: that of tolerance. In saying Enough is enough (or similar utterances like A joke's a joke (but that's enough))the speaker is not being stem and implacable in the sense in which he would be if he said, for example, Stop it at once! Rather, he indicates that while he recognises the need for tolerance and while he is quite prepared to be indulgent for a time, now the limits of his tolerance have been reached. Accordingly, the semantic representation of the 'stem and implacable' sentence Enough is enough is in fact closely related to that of the indulgent sentences such as A picnic is a picnic or Boys will be boys: Enough is enough (a) everyone knows: sometimes people want to feel something good because of this they want to do some things they wouldn't do these things at other times (b) I know: someone (you?) was doing something like this for some time (c) I know: someone can think: this is not bad (d) I think: one should not think this (e) one should know: one can do it for some time one should not do it for a long time (f) I want this person not to do it any more now
English nominal tautologies: semantic representations
411
2.5. Seeing through superficial differences Syntactic formula: A N hum is a N hum In the BBC television series 'Fawlty Towers' one of the characters says at one point: A man is a man. The attitude to men conveyed in this particular utterance is far from tolerant and indulgent. The speaker, a woman, has just discovered a dead man in a cupboard, and feels frightened; another woman tries to reassure her, pointing out that a dead man can do her no harm; this, however, is countered with the somber tautology quoted above. It is interesting to note that the deep distrust towards men (even dead men) has to be expressed in the singular (A man is a man, and not Men are men), whereas patronising tolerance is always expressed in the plural (Kids are kids, not A kid is a kid). The singular focusses on the serious harm which is to be expected from an individual and which should not be dismissed lightly. The plural refers to peccadilloes which have already been perceived and which should be dismissed, because they are trivial, and because they are not under the individual's control. A number of eighteenth and nineteenth century collections of English proverbs include one or another version of the following: A man is a man, tho' he have but a hose upon his head Stevenson (1949:1511) offers the following comment: "Some commentators think that the meaning of this proverb is not clear, but it must mean that a man's a man, whatever his attire, even if his poverty compels him to wear a stocking for a cap." Obviously, the attitude to men conveyed by the proverb is different from that conveyed in the 'Fawlty Towers' quote. One says, in effect: 'don't trust any man'; the other says 'don't despise any man'. But it is not difficult to find a common thread linking the communicative import of the proverb and of the homophonous sentence used in 'Fawlty Towers'. Both imply that 'despite differences in appearance, one man is not significantly different from others; in essence, all men are the same'. An examination of other English tautologies encoding a similar message ('despite appearances, all Xs are the same') shows that in different collocations and in different contexts this message can be combined with a number of different attitudes. For example:
412
Boys will be boys: even 'truisms' are culture-specific
man is a man (and not to be trusted, or, not to be despised). priest is a priest (and can be trusted). mother is a mother (and can be expected to care). lawyer is a lawyer (and can be relied upon as an expert). A woman is a woman (and can be expected to show, at some stage, a 'feminine weakness', even if she seems generally tough).
A A A A
This list of attitudes and expectations is of course far from exhaustive, and it is doubtful whether an exhaustive list of this kind could ever be compiled. It seems sounder, therefore, to assume that the different attitudes and expectations involved are not part of the meaning of the construction, and to look for a general formula, which would include only what all tautologies of this kind have in common. Clearly, what is needed is a formula which would fit the whole range of possible expectations, and which nonetheless wouldn't over-predict (cf. Boguslawski 1981a:l02). Consider, for example, a situation where a girl giggles in a 'typically girlish' fashion. The comment: (Ah, well), girls are girls would imply that this is precisely what one could expect of any member of the category 'girl'. On the other hand, the comment: (Ah, well), a girl is a girl, if acceptable at all, would be seen as implying that one would not have expected such 'girlish' behaviour of this particular girl, although one would expect it of girls generally; and the comment would be taken as a vindication of the general validity of the stereotype. On the other hand, if a usually tomboyish girl exhibits some behaviour regarded as typically girlish, one could say A girl is a girl. The difference between the singular and the plural pattern is, then, this: while both patterns endorse a certain stereotype, the 'singular' one acknowledges that in some particular case the appearances seem to be to the contrary and yet reaffirms the stereotype despite the appearances. By contrast, the 'plural' one simply reaffirms the stereotype, without any references to deceptive appearances. In addition, the 'plural' type implies, as we have seen, benign tolerance for human weakness. But there is no suggestion of benign tolerance in the singular type. This brings us to the following semantic formula for the A man is a man (or A girl is a girl) type: A man is a man (a) everyone knows: one can say: people of this kind are like this
English nominal tautologies: semantic representations
413
(b) I know: someone can think: this person is not like other people of this kind (c) I think: one should not think this (d) one should know: all people of this kind are the same a person of this kind can be a little different from other people of this kind they cannot be very different
2.6. Recognising an irreducible difference Syntactic formula: N 1 is N 1 (and N 2 is N 2)
As mentioned earlier, sentences such as: East is East (and West is West). White is white (and black is black).
refer to what the speaker presents as an irreducible difference, which follows from the unique nature of each member of the set, and which has to be accepted as irreducible. As a rule, the sets in question are binary, and the two members can be seen as opposites. Formally, tautologies of this kind are distinguished by their complex character: they are not merely tautologies but double tautologies; and although the second part can be omitted, its presence is implied by the first part. Consider also a parallel Polish example: - Co pan wlasciwie rozumie, panie Stefanie, pod srowem borowka? pyta pan Mareczek. - Borowka - odpowiada wiolonczelista - to czarna Jagoda. A dla pana to co? - Borowka to borowka a czarna Jagoda to czarna Jagoda. (Rymkiewicz 1984:9)
'''What exactly do you mean by the word borowka, Stefan?" Mareczek asked. "Borowka," the violinist replied, "means (a) blueberry. And for you, what does it mean?" "(A) borowka is (a) borowka, and (a) blueberry is (a) blueberry.'"
414
Boys will be boys: even 'truisms' are culture-specific
This sub-construction can be explicated as follows: East is East (and West is West) (a) everyone knows: one thing/person is not the same as another (b) I know: someone can think: if one can say something about X one can say it about Y (c) I think: one should not think this (d) one should know: X is not like Y one thing/person can be like another one thing/person cannot be the same as another
It might be added that a similar (though perhaps not identical) message can be conveyed in sentences referring to individual people. For example, in the ABC television production 'Bewitched' (30.5.1985) one of the heroes, Larry, suspects that two young women called Samantha and Serena are in fact the same person, and he is glad to discover, while speaking to one of them, that You are you and she is she.
In the same program, somebody else asserts that Samantha is Samantha and Serena is Serena.
In a similar vein, in C. S. Lewis' Screwtape letters the senior devil makes the following assertion: The whole philosophy of Hell rests on recognition of the axiom that one thing is not another thing, and, especially, that one self is not another self. My good is my good and your good is yours. (Lewis 1946:92).
The question of the relation between sentences of this kind and those referring to finite (binary?) sets, such as East is East, is an interesting one, but I will not pursue it here any further.
2.7. Tautologies of value English has a number of tautological constructions focussing on the value of certain things (and also people). I will discuss here four such subconstructions.
English nominal tautologies: semantic representations
415
The first construction emphasises the low value of certain categories of things (and people). As an example, consider Kipling's sentence (from 'The Betrothed') quoted in Bartlett (1980): A woman is only a woman, but a good cigar is a smoke.
At first, one might think that the meaning of the expression A woman is only a woman is no different from the meaning of the expression A woman is a woman, except for the additional semantic content of only. But in fact, the variant with a 'belittling', minimising particle such as only lacks what appears to be a crucial feature of the variant without such a particle: the idea that 'despite appearances, this particular X is not different from other Xs'. Thus in Kipling's sentence the stress is not on 'some particular woman being no different from other women'; rather, it is on the low value of the entire class of women (as compared with cigars). Consider also the 'belittling' intention of the tautologies in the popular song from the film Casablanca: You must remember this: A kiss is just a kiss, A smile is just a smile ...
Without some particle such as just or only, the same tautologies would imply a different, and, in a sense, opposite attitude. Sentences such as A kiss is a kiss, A smile is a smile would normally be interpreted as implying undeniable commitment, value or significance. But the variant with just asserts that all kisses (or smiles) are fairly insignificant, not that this particular kiss (or smile) is not different in value from others. Thus, in the semantic representation of tautologies with minimising particles, the component 'I know: someone can think: this X is not like other Xs of this kind' would not be appropriate. Consider in tum the sentence: A point is a point, which I heard uttered at an examiners' meeting at the AND (on 27 June, 1985). The speaker's intention was, clearly, to imply that 'every point [for a semester unit completed by a student] counts, and the importance of a single point should not be underrated'. A similar sentiment is conveyed in the following sentence: 'Must find it,' Fatty was saying. 'A shilling is a shilling.' (Blyton 1980:57)
416
Boys will be boys: even 'truisms' are culture-specific
Here, too, the speakers do not seem to be concerned with an apparent difference between one particular point and all other points, or between one particular shilling and all the other shillings. Rather, they are concerned with the value of any single point (or shilling); and they insist that this value - though small - is nonetheless not negligible. It should be noted that one could hardly use a plural in this sense: ? Points are points, ? Shillings are shillings. On the other hand, a plural can probably be used in a related sub-construction, which is used to remind people of the undeniable value of things which are not being despised but whose value could nonetheless be occasionally forgotten. Thus, one can say: A steak is a steak. A Mercedes is a Mercedes. Money is money. Eggs are eggs.
Some tautological utterances can be ambiguous in this respect. For example, in saying: An egg is an egg, the speaker may mean either 'every small thing counts' or 'don't forget the well-known value of eggs'. Another related sub-construction requires an adverbial modifier of time or space and allows the form Ns were Ns. For example: In those days, women were women. In Japan, women are still women. In those days, children were children: they played, they studied, and they did as they were told.
The adverbial modifier implies a contrast, and the tautological construction conveys approval, not indulgence. Again, one must note that the 'well-known value' tautologies don't seem to be necessarily concerned with differences (however superficial) between one thing and another (of the same kind). For example, in saying: A steak is a steak, the speaker doesn't seem to be claiming that 'despite appearances, this particular steak is no different from others'. Rather, he is emphasising the undeniable value of all steaks. Finally, English has a tautology which expresses indifference to individual differences, and which stresses the interchangeability and the equal value of things (within a kind). Consider, for example, the following exchanges:
English nominal tautologies: semantic representations
417
- How was the party? - Oh - a party is a party. - Do you want Nescafe or Maxwell House? - It doesn't matter. Coffee is coffee.
The stress on the identical value of all Xs conveyed in sentences of this kind makes them seem extremely close to the pattern A N hum is a N hum • And yet I believe that there is a difference between the two patterns. Speakers who utter such tautologies sound blase, wordlywise, somewhat cynical. They want to be seen as people who don't get unnecessarily enthusiastic or over-excited; they 'know' that usually things are neither very good nor very bad; for example, A party is a party: nothing to get excited about. To give a somewhat jocular illustration: a recent issue of Australian Natural History (The Australian Museum, 1989, 22:540) reported new evidence on animals' responsiveness to individual faces. When presented with a series of slides, "the sheep responded particularly to pictures of species with horns (an important signal of sex and status), and the bigger the horns the more excited they became". The picture accompanying the article bears the caption: "A sheep is a sheep ... but not to a sheep." The message is clear: individual sheep do differ in perceived value (to other sheep) - contrary to the conventional wisdom embodied in the tautology. In essence, the difference is this: one type of tautology refers to 'nature' and the other, to 'value'. For example, in saying: A doctor is a doctor.
a speaker may be either referring to the supposed 'common nature' of all doctors or to their 'equal value'. The intended message may be either: (1) this particular doctor may seem different from others, but this is deceptive, at the level of essential characteristics all doctors are the same, and sooner or later their common nature will show; or (2) one particular doctor may seem better (or worse) than others, but in fact they are all the same: they do have some value, but one should not put too much faith in them, and one should not attach too much importance to the choice; this one is as good as any other. A third possibility is the 'undeniable value' one: all doctors are highly qualified professionals and can be relied upon. Naturally, ambiguities of this kind arise mainly with human nouns, since it is mainly in the case of people (or animals) that one would want to appeal to their nature. A sentence such as: A party is a party is
418
Boys will be boys: even 'truisms' are culture-specific
ambiguous, too, but here the choice is not between 'value' and 'nature', but between 'value' and 'special time'; compare: - How was the party? - [shrugs] A party is a party. It's noisy, but (after all) a party is a party!
Below, I will sketch in tentative semantic representations of the four tautologies of value which have been singled out here. A point is a point (a) everyone knows: things of this kind are good (b) I know: someone can think: this thing is not like other things of this kind it is something very small (c) I think: one should not think this (d) one should know: all things of this kind are the same they are all good a thing of this kind can be small it cannot be not good A kiss is just a kiss (a) everyone knows: things of this kind are like something small (b) I know: someone can think: this thing is not like other things of this kind it is something very good (c) I think: one should not think this (d) one should know: all things of this kind are the same they are like something small one thing of this kind can be a little less like something small than another it cannot be not like something small A party is a party; Coffee is coffee; A girl is a girl (a) everyone knows: things of this kind are not very good, not very bad (b) I know: someone can think: this thing is not like other things of this kind it is very good, or very bad (c) I think: one should not think this
English nominal tautologies: semantic representations
419
(d) one should know: all things of this kind are the same one thing of this kind can be a little better or a little worse than another it cannot be very good or very bad A steak is a steak; A Mercedes is a Mercedes (a) everyone knows: things of this kind are good (b) I know: someone can think: this one is not good (c) I think: one should not think this (d) one should know: all things of this kind are the same they are all good one thing of this kind can be a little better or a little worse than another it cannot be not good
2.8. Tautologies of obligation This sub-construction is expressed by the syntactic formula: (ART) N is (ART) N One use of tautological constructions, particularly widespread in different languages, is concerned with obligations, and, more broadly, with rules of human behaviour. Generally speaking, if a noun embodies a modal meaning such as 'One should do X', then the pattern (ART) N is (ART) N implies that the obligations in question must be fulfilled, even if one prefers not to do so. Thus one can say The law is the law, as the Australian judge Michael Kirby said recently ,on the ABC radio current affairs program 'PM'. Such sentences are extremely common in English. To quote a few characteristic examples which I have recently heard or read: A rule is a rule. A bet is a bet. A promise is a promise. (Doyle 1981:417,51.3) A deal is a deal. A test is a test. An agreement is an agreement. (Doyle 1981:141)
420
Boys will be boys: even 'truisms' are culture-specific
Usually, this type can be recognised by the presence of a noun referring to contractual undertakings such as bet, promise, deal, law etc. But one can also use human relational terms such as father to refer to generally accepted role obligations. Thus one can say A father is a father, meaning that one has certain obligations toward one's father which should be fulfilled, without any implication that this father may seem to be different from other fathers. The speaker may mean that in this particular case there seem to be some extenuating circumstances (such as, for example, one's illness, or one's recent marriage) but that the obligation 'remains an obligation' and has to be carried out. As mentioned earlier, somebody might even want to say A war is a war, meaning that one must carry out one's duty with respect to a war. The two tautological expressions A war is a war and War is war differ in their 'implicatures', and are not interchangeable. However, when the meaning of a noun clearly implies an obligation, then mass nouns without an article can be used in the sense under discussion; for example Duty is duty. Such sentences don't imply that promises, bets, the law, or duty 'cause bad things to happen', like wars or politics (cf. 2.1 above); rather, they imply that certain rules of human behaviour require compliance, regardless of their unpleasant or inconvenient consequences. It is also interesting to note that tautologies of obligation normally require a singular form of the noun. One can hardly say ?Bets are bets (cf. A bet is a bet), or ?Deals are deals (cf. A deal is a deal). A sentence like Promises are promises seems to imply that 'Promises are no more than promises, and can't always be relied on' - which is very different from what we imply by A promise is a promise. It seems that plural nouns can be used in the relevant sense only if they refer to items which normally occur in sets, rather than singly: Rules are rules; Regulations are regulations. This is parallel to the extension of the singular patterns to fit the other kinds of tautologies which normally preclude the plural: A girl is a girl. (absolute generalisation - or indifference) Girls are girls. (tolerance for human nature) *Girls are girls. (impossible as an absolute generalisation) Twins are twins. (possible as an absolute generalisation) A party is a party. (indifference - or absolute generalisation) *Parties are parties. (impossible as an expression of indifference) Beans are beans. (possible as an expression of indifference)
English nominal tautologies: semantic representations
421
However, the ability to co-occur with the definite article (or with a definite possessive) does seem to distinguish 'tautologies of obligation' from the other tautologies which normally require the singular. Thus Neighbours are neighbours can be used as an expression of indifference, in response to the question Do you like your new neighbours? But the sentence My neighbours are my neighbours conveys some sense of obligation (perhaps, one must help one's neighbours, or one must be loyal to one's neighbours). Compare also the following sentences and their possible interpretations: a. Fathers are fathers. b. Your father is your father. c. A husband is a husband. Example (a) has only one possible interpretation: tolerance for human nature. Example (b) also seems to have only one interpretation: obligation (one must fulfil one's obligations toward one's father). Example (c) has as many as four interpretations: obligation (one must fulfil one's obligations toward one's husband), appreciation (everyone knows that there is something good about having a husband), indifference (one husband is neither better nor worse than another), and absolute generalisation (all husbands are essentially the same; one knows what to expect from them). I conclude that, for both semantic and syntactic reasons, tautologies of obligation cannot be subsumed under any other type, and must be recognised as a separate tautological sub-construction. The meaning encoded in such sentences can be spelt out as follows: The law is the law; A promise is a promise; A father is a father (a) everyone knows: all people have to do things of this kind (b) I know: someone can think: this time this person doesn't have to do this thing (c) I think: one should not think this (d) one should know: all things of this kind are the same one thing of this kind can be a little different from other things of this kind one cannot not do this thing because of this It is interesting to note the links between 'tautologies of obligation' such as The law is the law and 'tautologies of grim necessity' such as
422
Boys will be boys: even 'truisms' are culture-specific
War is war, links which could be summarised in the two key notions: 'cannot' and 'don't want'. For example, wars have aspects that people 'don't want'; and these aspects 'cannot' be avoided. Similarly, the law may force people to do things that they 'don't want' to do; and yet they 'cannot' not do them. It would be easy to phrase the explications of the two types in identical terms, and thus to reduce the two types to one. I think, however, that this would be a false economy. The minimal pair mentioned earlier: War is war. ('realism, grim necessity') A war is a war. ('obligation') shows that in fact the two types are truly distinct in English (although they would not necessarily be distinguished in some other language). I would add, however, that although the tautologies of obligation occur in many different languages, and can be presumed to express an attitude which is largely language-independent, in some languages it is counterbalanced by the existence of constructions whose communicative import is quite different. For example, in Polish one can say not only:
Prawo jest prawem. 'The law is the law.' as one does in English, but also:
Prawo prawem, ale trzeba jakos iyc. 'The law [is] the law, but one has to live somehow.' thus excusing lawlessness rather than endorsing the validity of laws. A sentence of this kind indicates that while the speaker accepts the immutable validity of laws (or whatever rules are referred to) theoretically, in practice he expects that people will find some ways around those rules, and that this, too, should be understood and excused. This use of tautological expressions, which indicates a dismissal of certain phenomena rather than acceptance, seems to be alien to English. But in Polish, tautological expressions are often used in such a dismissing way. For example, the Polish expression:
tarty iartami. 'Jokes (NOM.PL) [are] jokes (INSTR.PL).' translates into English as 'jokes aside' (as does its Russian equivalent Sutki sutkami) and not as A joke is a joke. It indicates neither indulgence nor impatience but simply a lighthearted dismissal ('jokes are fine, but
Some comparisons from Chinese and Japanese
423
let's now speak seriously'). Similarly, an expression such as Prawo prawem ('the law is the law, but ... ') doesn't challenge the validity of laws; it merely 'sets them aside' ('laws are all very well, but ... ').
3. Some comparisons from Chinese and Japanese 3.1. Chinese concessive tautologies Chinese has some tautological patterns which seem more or less similar to the English ones; but it also has patterns which are distinctly different. (My discussion of Chinese tautologies is based on data and analysis provided by Luo 1988.) One type which has no equivalent in English is what one might call 'concessive' tautologies. They can be illustrated with the following examples: Qtnqi
dao
shi
qfnqi,
(jiushi
bu
tai
qfn).
relatives PRT are relatives (but not very close) 'Relatives though they are, they are not very close.' Jingzi dao
shi jingzi
(jiushi zhaoren bu
qfng
chu).
mirror PRT is mirror (but person not clear see) 'Though it is a mirror, it does not reflect the image clearly.' Nanren shi nanren, (ki queshiio nanren wei). man is man (but lacks manhood) 'Although he is a man, he lacks manliness.'
As these examples show, the 'concessive' tautology has the form of a subordinate clause, which precedes the main clause. The subordinate clause states an 'undeniable truth' but the main clause contradicts this truth with respect to a specific instance: since this particular entity (X) belongs to a certain kind (X), one might expect that it will have certain properties, generally seen as characteristic of that kind; and yet, the speaker points out, this particular X (X) doesn't have the properties in question. Tentatively, I would represent the meaning of this type of tautology (as described by Luo) along the following lines:
424
Boys will be boys: even 'truisms' are culture-specific
Nanren shi nanren (a) everyone knows: one can say: people/things of this kind are all the same (b) I know: because of this someone can think: this one is like this (c) I think: one should not think this (d) one should know: a thing of this kind can be not like this this one is not like this
The function of this pattern can be seen as the opposite of that of the common tautologies of the English language. In English, most nominal tautologies emphasise the sameness of the members of a given category; they acknowledge that one member may differ from another in some superficial characteristics but they stress the commonality of the essential features. By contrast, the Chinese concessive tautology acknowledges the sameness of some characteristics (presumably, the definitional ones), but emphasises individual differences; it can be seen, therefore, as a warning against absolute generalisations. If the English tautologies of several kinds defend and reinforce the stereotype (A man is a man, Boys will be boys, A Mercedes is a Mercedes), the Chinese concessive tautologies warn against an excessive confidence in stereotypes. This is not to say that Chinese doesn't allow the use of tautologies for making absolute generalisations and for underscoring stereotypes. It does, as in the following sentences (which usually include a modal adverb): Nuren z6ng shi nuren. 'Women are (always) women.' Miigu6ren daodi shi meigu6ren. 'Americans are Americans, (after all).' Qinqi zhongjiu shi qinqi. 'Relatives are relatives (after all).' Lang z6ng shi lang 'Wolves are (always) wolves.'
Interestingly, the range of possible interpretations for such generalisations appears to be even wider in Chinese than it is in English, as they can apply not only to 'human weakness' or to 'human/animal nature',
Some comparisons from Chinese and Japanese
425
but also to posItIve aspects of human relations. Thus, it seems less natural to say in English than in Chinese: ?Relatives are relatives. ?Friends are friends.
?A relative is a relative ?A friend is a friend.
meaning that these relationships are invariably good and invariably reliable, although the exact range of use of such tautologies requires further investigation. But while the English 'generalising' tautologies do have some counterparts in Chinese, the Chinese concessive tautologies do not seem to have any counterparts in English. On the contrary, all English nominal tautologies seem to have a 'final', 'topic-closing quality' about them, epitomised in the common English tautological expression That's that. By contrast, nominal tautologies in Chinese - and, I might add, in many other languages of the world - do not seem to have such a 'topic-closing' quality. This is why they can often be used as a premise in a concessive or quasi-concessive sentence of one kind or another, and it is very interesting to see what kinds of 'undeniable truths' can be undermined in a given language (and culture) in this way. For example, it is interesting to compare Chinese concessive tautologies with Polish and Russian concessive tautologies mentioned earlier, such as the following: Prawo prawem, ale law:NOM.SG law:INSTR.SG, but 'The law is the law, but ... ' Przyjain Druiba friendship:NOM.SG 'Friendship is all very
(Polish) .
przyjaini{l, druzboj, friendship:INSTR.SG, well, but ... '
ale no but
(Polish) (Russian) .
Polish and Russian tautologies of this kind can be said to be iconic: the identity of the subject and the predicate nominal represents the 'undeniable truth' aspect of the sentence, whereas the absence of the copula and the instrumental (rather than nominative) case of the predicate nominal de-emphasises that identity and casts a doubt upon its absolute validity. It should be pointed out, however, that Polish and Russian concessive tautologies of this kind are narrower in scope than the Chinese concessive tautologies illustrated earlier, being restricted to abstract concepts such as 'joke', 'friendship', 'law', or 'promise'; unlike
426
Boys will be boys: even 'truisms' are culture-specific
the Chinese type, they cannot apply to concrete individual entities like a person or a mirror. If the Chinese concessive tautologies warn against absolute generalisations, the Polish and Russian ones call for a flexibility of attitudes.
3.2. 'Irreducible difference', Chinese style Like English (see 2.6 above), Chinese has nominal tautologies which proclaim an 'irreducible difference' between two categories or two entities. For example: Dtng shi ding, mao shi mao. 'Ding is ding, mao is mao (one should keep ding from mao),' (where ding represents 'a Heavenly stem' and mao, 'an Earthly branch'). Shfsi shi shfsi, sishf shi sishf. 'Fourteen is fourteen, forty is forty.' Tii shi ta, wo shi wo. 'He is he, I am I.'
However, as pointed out by Luo, the meaning of such tautologies is not quite the same as that of their closest counterparts in English. In the Chinese examples, unlike the English ones, the speaker combines the statement of an 'irreducible difference' with a moral judgment: 'X and Yare not the same; this is good; one shouldn't try to change it'. In the English examples, the speaker points to an 'irreducible difference' as to something that should be recognised, without implying, however, that it is good and that it should be preserved. The similarities and the differences between the two meanings (the English and the Chinese ones) can be portrayed as follows (C for Chinese, E for English): Ding shi ding, mao shi mao (C) cf. East is East and West is West (E) (C/E) everyone knows: one thing/person is not the same as another (C/E) I know: someone can think: if one can say something about X, one can say it about Y
Some comparisons from Chinese and Japanese
427
(C/E) I think: one should not think this (C/E) one should know: X is not like Y (E) one thing can be like another (E) one thing cannot be the same as another (C) this is good (C) one should not think: X can be like Y According to Luo (1988), Chinese allows also for multiple (rather than binary) tautologies of 'irreducible difference'. He illustrates this with a famous (copula-less) example which epitomises Confucius' social doctrine: Jun jiin, chen chen, fil fil, Zl Z1. 'A ruler is a ruler, a minister is a minister, a father is a father, and a son is a son.' Luo (1988) comments: "Written more than two millennia ago, this doctrine was accepted as infallible law and has influenced Chinese ideology ever since: what Confucius wants to express is that all the social hierarchies must be kept distinct and should not be altered or confused." Arthur Waley's rendering of Confucius' tenet (quoted by Luo) supports Luo' s interpretation of this pattern: Let the prince be a prince, the minister, a minister, the father, a father, and the son, a son. It supports also the components included in the explication of the Chinese pattern (in contrast to the English one): (X is not like Y) this is good one should not think: X can be like Y
3.3. Chinese tautologies of unreserved praise The last type of Chinese tautologies to be mentioned here is very different from anything that we have seen in English. It conveys praise and admiration towards an entity whose all parts or all aspects are seen as genuinely good. Typical examples are:
428
Boys will be boys: even 'truisms' are culture-specific
(Kan ren jia nei tiaojian:) caidian shi caidian, bingxiang shi bingxiang, xiyiji shi xiyiji. '(Just look at what they have:) colour TVs, fridges, washing machines, (all kinds of things)!' Tao shi tao, ZI shi Zi, xing shi xing, (guo yuan Zi shenme dou you). 'Peaches, plums, apricots - there are all kinds of fruit in the orchard.' Td yan zhenhao, yiinshen shi yiinshen, shenduan shi shenduan, banxiang shi banxiang. 'S/he performs brilliantly - rich in expression, elegant in posture and marvellous in appearance.'
I am not going to propose a full explication for this type of tautology, but I would like to sketch, and to discuss briefly, some possible components. First, there seems to be a component of 'total praise': 'one can say all good things about this.' Possibly, there is an implication that 'everyone knows' what the relevant aspects to be assessed would be for the category in question. For example, a good orchard would be expected to have peaches, plums, and apricots, all of good quality. Furthermore, all the aspects which are enumerated are individually assessed against an imaginary model, perhaps again with reference to common knowledge (everyone knows what 'real' peaches should be like - these peaches are like this; and so on). Finally, if Luo is right, this pattern expresses also the speaker's admiration for the genuine quality of the individual aspects of the entity in question and for the 'all-roundness' of its perfection. Some of the motives mentioned above (for example, 'everyone knows') correspond to those which we have established for the nominal tautologies of the English language; but some are clearly different. It seems obvious that the exact semantic value of this last type of Chinese tautology cannot be worked out on the basis of any 'natural logic': it is language- and culture-specific. Japanese has a number of nominal tautologies which correspond, more or less, to some of the types distinguished here for English. However, it has also some forms without any counterpart in English. One such type ('what seems impossible is really possible ') was already mentioned in the Introduction. I will now discuss briefly two further ones (on the basis of data and analysis provided by Field 1988).
Some comparisons from Chinese and Japanese
429
3.4. Japanese tautologies of 'a matter of course' Tautologies of 'a matter of course' (Field's term) can be illustrated with the following examples: Keireki personal history
ga SBJ
keireki pers.hist.
dakara ... COP-BECAUSE
Toshi age
ga SBJ
toshi age
dakara ... COP-BECAUSE
Kao face
ga SBJ
kao face
dakara ... COP-BECAUSE
The general pattern is, then: N.J ga N.J dakara ...
As pointed out by Field, in tautologies of this kind the subject particle ga cannot be replaced with the topic particle wa. This feature distinguishes the type in question from other types of nominal tautologies, which normally take wa, not ga, for example: Yakusoku wa yakusoku da. rule TOP rule COP 'Rules are rules.' (and have to be obeyed) Gaijin wa gaijin da. foreigner TOP foreigner COP 'Foreigners are foreigners.' (implied: criticism and resignation)
Tautologies of 'a matter of course' imply that the reason of some state of affairs is self-explanatory. They could be, therefore, compared with the English expression: N being what it is ... For example: Toshi ga toshi dakara ... age SBJ age COP-BECAUSE 'His age being what it is ... ' (it's no wonder he's puffing by the time he reaches the fourth floor)
I am not going to propose a full explication for this type of tautology. One familiar component, however, is fairly easy to discern: 'everyone knows'. Another familiar component which may come to mind is 'all Ns are the same': for example, 'all old men are the same (in the relevant
430
Boys will be boys: even 'truisms' are cUlture-specific
respect)'. But this can hardly be applicable to strictly individual circumstances, such as somebody's 'personal history'. A better guess might be: 'it would be the same for anyone (of that age, with that kind of background, with such a face, and so on).' But the matter requires further investigation.
3.5. Japanese tautologies of irrelevance Tautologies of irrelevance can be illustrated with the following sentences:
Hito person
wa TOP
Otto wa husband TOP
hito person
da. COP
otto da. husband COP
Field (1988:9) observes that Hito wa hito da could be "said by a teenager to her mother who refuses to let her go to a rock concert because the other mothers in the neighbourhood are doing the same", and it would mean that the teenager is implying to her mother that she wants the other mothers kept out of the argument; that other people are irrelevant to the issue at hand. Otto wa otto da "could be uttered in a situation where a group of housewives are discussing whether or not they should still remain friends with a Mrs. Tanaka whose husband has just been gaoled for fraud. The housewife who utters the tautology implies that Mrs. Tanaka's husband should be kept out of the issue - because the husband is 'bad' does not mean that Mrs. Tanaka is bad too." As a first approximation, the meaning of such tautologies can be represented as follows:
Hito wa hito da (a) everyone knows: people are not the same different people can do different things (b) I know: someone can say: if one can say something about one person then one can say it about another because of this, one should think about other people
Verbal tautologies
431
(c) I think: one should not think this (d) I don't want to think now about other people As this formula indicates, the Japanese tautologies of irrelevance are closely related to the English tautologies of 'irreducible difference'. In fact, sentences such as Hito wa hito da can be seen as functional equivalents of English sentences like Samantha is Samantha and you are you. But the English sentence emphasises not only the difference but also the uniqueness of the entities in question (hence the common use of proper names or personal pronouns in this type of tautology). By contrast, the Japanese tautologies of irrelevance do not refer to people's unique features or unique situations; rather, they simply stress that people are not the same, and more specifically, that different people can do different things.
4. Verbal tautologies The bulk of this chapter has been devoted to various kinds of nominal tautologies, and limitations of space preclude a similarly detailed discussion of what might be called verbal tautologies. However, before any general discussion of a possible invariant or invariants of all tautologies can be fruitfully undertaken, one needs to look briefly at at least some verbal tautologies - from English, and not only from English. This is the purpose of this section.
4.1. Future events Tautologies referring to future events often have a rather fatalistic ring about them. A somewhat fatalistic acceptance of future events (including misfortunes) is encoded, for example, in the Spanish saying Que sera sera ('what will be will be'), whose meaning can be explicated as follows: Que sera sera I know: we don't know what will happen to us I know: something bad can happen to me I don't want to think about it
432
Boys will be boys: even 'truisms' are culture-specific
I don't want to feel something bad because of it I know: I cannot think: 'if I do something these bad things will not happen to me' The first of these components is reflected in the song: Que sera sera, Whatever will be will be, The future's not ours to see, Que sera sera.
Tautologies of this kind ('whatever will be will be ') seem to be very common in different languages of the world. But here, again, one should beware of hasty generalisations and one should not mistake similarities for identities. For example, as pointed out earlier, the literal Polish counterpart of the Spanish saying conveys also the idea of personal risk, and an attitude of foolhardy bravery. The Polish saying requires, therefore, some additional and different semantic components. I propose the following: Co b~dzie to b~dzie I want to do something I will do it because of this I know: something bad can happen to me because of this I don't want to think about it
The need for such different components highlights the culture-specific twist which Polish culture gives to the conventional and (one might think) universal wisdom. This is entirely consistent with the characteristic Polish ethos, which has been described and documented elsewhere on the basis of other kinds of evidence, both linguistic and non-linguistic (see Wierzbicka, to appear, chaps. 2 and 5). A different kind of tautology referring to future events is illustrated by the English sentence mentioned by Levinson (1983: 111): Either John will come or he won't.
Levinson's gloss, quoted earlier, reads: "Calm down, there's no point in worrying about whether he's going to come because there's nothing we can do about it." But though sound in essence, this gloss is clearly ad-hoc and arbitrary in phrasing, which makes any comparison of this particular pattern with other tautological patterns impossible. Using the
Verbal tautologies
433
controlled semantic metalanguage employed in the present work, we can propose the following explication: Either he will come or he won't I know: it can happen: he will come it can happen: he will not come I know: someone can think: this is bad someone can feel something bad because of this I don't want to think about it I don't want to feel something bad because of it I know: one of these things will happen I know: I cannot think: 'if I do something it will not happen' This is not nearly as 'fatalistic' as the Spanish saying. To begin with, the speaker is talking about future events which are probably 'undesirable' rather than 'bad', and which in any case are not presented as directly affecting the speaker (or the addressee). There is no question here, then, of 'something bad happening to me'. Furthermore, there is no reference here to the (supposed) 'unknowability of the future in general' (no 'we don't know what will happen to us' component). Finally, there is some positive certainty, from which the speaker can take some satisfaction, or at least some reassurance: 'I know: one of these things will happen'. What is unknown and unpredictable to us is, above all, other people's future actions; but these can, presumably, be planned and controlled by the individuals involved. All in all, the English pattern seems more compatible with the view that human beings have some control over their future; not over other people's actions, because this is their (other people's) domain of freedom; but there is some ground for calm in the thought that every person has control over their own actions; and also, that to a logicallyminded person there are some certainties in life ('I know: one of these things will happen'). This quiet certainty derived from a rational, logical thinking ('either it will happen or it won't') can also be interpreted as a kind of control over the world. This is different from the more passive, more indolent tone of Que sera sera.
434
Boys will be boys: even 'truisms' are culture-specific
4.2. The immutability of the past Examples: Let bygones be bygones. Co bylo to by/o. (Polish) 'What has been has been.'
Tautologies expressing an acceptance of past events may seem to be a mirror image of those expressing an acceptance of things to come. And indeed, when one compares the English saying Let bygones be bygones with the Spanish saying Que sera sera, one has to acknowledge a number of close parallels. In both cases, the events referred to must be seen as 'bad'; in both cases, they must be seen as unchangeable; and in both cases the speaker refuses to think about the bad events in question and to let them generate 'bad feelings'. But the parallel is far from perfect, and the details of the interpretation are by no means predictable. In particular, the sentence Let bygones be bygones must refer to 'bad things' which have occurred between the interlocutors, that is to say, to a past conflict between them. Nothing of the kind is implied by tautologies referring to future events. Furthermore, the formula Let bygones be bygones conveys an active call for peace. The full meaning of Let bygones be bygones can perhaps be represented as follows: Let bygones be bygones I know: we can think: 'some time before now this person did something bad to me' we can feel something bad because of this I don't want us to think about it any more I don't want us to feel anything bad because of it any more we know: it happened before now we cannot think: 'if we do something these bad things will not happen' I think we should not think about it any more
In the speaker's view, the past is immutable. There is no point in worrying about it because nothing can be done about it, anyway. It cannot be thought that 'if we do something, these bad things will not happen': they have already happened.
Verbal tautologies
435
In his call for peace, the speaker does not reiterate any accusations. Presumably, it is the addressee who has done something bad to the speaker (or so the speaker sees it); but the message is tactfully diffused so as not to point a finger at the addressee; it is: 'we can think: someone did something bad to me' rather than 'I know: you did something bad to me'. Again, it appears that exhortations of this kind ('let's forget the past') are very common in different languages of the world; and coming across yet another saying such as the following Chinese one (Luo 1988:12), provokes a feeling of deja vu: Guoqu past
shi is
guoqu, past
xianzai present
shi xianzai. is present
It must be noted once more, however, that superficially analogous sayings in other languages may convey meanings different from that encoded in the English one. For example, the Serbian saying: Sto je bilo bilo je, 'What has been has been', can convey a nostalgic longing for good events of the past which have irrevocably vanished. It is often used with reference to a past love, as in the popular song: Sto je bilo bilo je, Vratiti se neee ... 'What has been has been, It will not return ... '
Apparently, the same attitude can be conveyed by the Russian saying eto bylo to bylo ('what has been [that] has been'), as its use in Bulat Okudzava's poem Lunin v Zabajkal' e illustrates: Neuzto eto bylo to by/o? I gvardija vas pozabyla, i daze ne snites' vy ej ... (Okudzava 1976:47) 'Is it possible that what has been has been? And that the regiment has forgotten you, and that they don't even have dreams about you any longer ... '
The literal English translation (What has been that has been) is not only unidiomatic but is simply puzzling and hard to interpret to native speakers. And the literal Italian translation is normally used in yet another sense, illustrated by the popular love song:
436
Boys will be boys: even 'truisms' are culture-specific
Quel ch' e stato e stato, Non importa s' efinita ... 'What has been has been, It doesn't matter if it is finished ... '
In this use, the speaker affirms the immutable value of things past ('nothing can rob one of the treasure of one's past good experiences'). The equivalent Polish sentence, Co byio to byio, is normally used to dismiss thoughts about some past events which were 'bad'. To use it in a sense parallel to that of its Italian counterpart one would have to mark it in a special way, preceding it with the particle conjunction ale 'but': Ale co byio to byio. 'But what has been that has been.' (i.e. 'but nothing can erase the good things that have happened to us')
Being used in this way (with an initial ale) the sentence is an instance of a more comprehensive Polish construction referring to past events, which can be illustrated with the following sentences: Ale cosmy si~ najedli tosmy si~ najedli. 'But what we have eaten, that we have eaten.' Ale coscie widzieli toscie widzieli. 'But what you have seen, that you have seen.'
The construction in question (marked formally with the particle/ conjunction ale) is quite productive, its range of use being restricted only by its meaning. Roughly speaking, this meaning can be characterised as follows: 'despite all the bad things that have happened to person X, X should recognise and relish the gain which these events have also caused; X should recognise that this gain has a lasting value, which nothing can take away from X'. Typically, the events in question concern the speaker and/or the addressee, but a third person use is not totally excluded. What does seem to be required is a pronominal, or rather a zero-anaphoric reference to the experiencer. The meaning conveyed by this construction can be represented, roughly, as follows: Ale co S to S I know: some bad things happened to us
Verbal tautologies
437
we could feel something bad because of this I think: we should not think about it any more we should not feel something bad because of it we know: some good things happened to us at the same time I think: we should think about these good things we should feel something good because of this we know: it happened before now we cannot think: 'these good things will not happen to us' Thus, the (perceived) immutability of the past can be invoked as a justification or source of support for different emotional attitudes: for resignation, for nostalgia, but also for a comforting certainty; for generous peace-making, or for resolutely looking ahead, without dwelling fruitlessly on past wrongs or past mistakes. Attitudes of this kind can become culturally entrenched and linguistically encoded. When they are dressed up, in a particular language and culture, as unquestionable 'universal wisdom', they can, one must presume, become further entrenched, and may become an important feature of a given cultural tradition. As one last example of this kind let us consider the English tautology: What's done is done (exemplified in a cartoon in AND postgraduate newsletter Antitheses 1989, 1.4:1). This expression, as far as I know, doesn't have any (tautological) counterparts in the language of Eastern or Southern Europe. The meaning of this tautology can be represented as follows: What's done is done I know: someone did something bad someone can think: this is bad someone can feel something bad because of it I think: we should not think about it any more we should not feel something bad because of it we know: it happened before now we cannot think: 'if someone does something this bad thing will not happen'
438
Boys will be boys: even 'truisms' are culture-specific
It seems clear that this is yet another example of a pragmatic and rationalistic folk philosophy which is associated with the English language: the speaker takes stock of the situation, briskly acknowledges some past mistakes, and resolutely looks ahead, trusting in logic, experience, and action. I suspect that while the use of tautological constructions with respect to past and future events shows striking cross-linguistic similarity, languages differ more in the way they use such constructions with respect to the present. Consider, for example, the awkwardness of the English translations of the Italian tautologies in the following exchange from Manzoni's novel 'The betrothed': 'Tonio! non mi riconosci?' 'A chi la tocca, la tocca', rispose Tonio ... 'L' hai addosso eh? povero Tonio; ma non mi riconosci piu? ' 'A chi la tocca la tocca', replieD quello ... (Manzoni 1972:794). Bohn's translation of this passage reads as follows: 'Tonio! don't you know me?' 'Whoever has got it, has got it [the plague] ... ' 'It's on you, eh? poor Tonio: but don't you know me again?' 'Whoever has got it has got it', replied he ... (Manzoni 1914, 2:615). In Colqhoun's translation, the same passage reads: 'Tonio! Don't you recognise me? 'What comes to one, comes to one', replied Tonio ... 'You've got it [the plague] on you, eh?' Poor Tonio; but don't you recognise me any more?' 'What comes to one, comes to one,' replied the other .... (Manzoni 1968:457). It appears that the only pattern which can be used in English to express acceptance of the present is the pseudo-conditional (If S, S) one: If you've got it you've got it. If you have freckles you have freckles. If it rains it rains.
Is there a semantic invariant?
439
5. Is there a semantic invariant? I have argued that the English tautological expressions don't all have the same inherent value, and that their different 'implications' cannot be fully attributed to context. It goes without saying that the similarities between different tautological constructions are no less real. It is much easier, however, to assert that such similarities exist, or to hint at their nature by means of vague or semi-metaphorical cover terms, than to establish with some rigour what the common components of English 'tautologies' might be. Levinson mentions "a dismissive and topic-closing quality" as a shared feature of tautological constructions of different kinds. To make these hints clearer, let's assume that the 'topic-closing quality' can be spelt out in the form of the following semantic component: I don't want to talk about it any more. One could argue with some plausibility that a component of this kind is indeed present in the meaning of some tautological expressions, such as, for example, Enough is enough and That's that. A wish for no more to be said on the topic seems associated with the tautologies in the following Italian examples, and their English translations: "Ma se non ne voglio saper nulla di quelle cose", diceva. "Quante volte 10 devo ripetere, che quel che e andato e andato?" (Manzoni 1972:735) "But what if I don't want to hear anything more about these things?'" he kept on saying. "How many times must I go on repeating that what's gone has gone?" (Manzoni 1968:416) "Chetatevi un po' ", disse don Abbondio; "che gia Ie chiachiere non servono a nulla. Quel ch' e fatto e fatto: ci siamo, bisogna starci." (Manzoni 1972:723) "Keep quiet, will you, for a bit?" said Don Abbondio. "Chattering's no use now. What's done is done. Here we are, and here we must stay. It's all in the hands of Providence ... " (Manzoni 1968:407) But it seems doubtful that all tautological expressions can be interpreted along those lines. For example, when the woman in 'Fawlty Towers' points out that A man is a man, she is indeed 'dismissing' her friend's
440
Boys will be boys: even 'truisms' are culture-specific
attempt to reassure her, but is she also conveying the idea 'I don't want to talk about it any more' ? And even the expression which may seem to be 'topic-closing' par excellence, That's that, doesn't always seem to be interpretable as 'topic-closing'. For example, in the BBC television program 'All Creatures Great and Small', a veterinarian examines a little girl's dog which has been run over by a car, and then says gently to the little girl: I'm afraid that's that. Certainly, there is a 'finality' about the vet's verdict, but can his attitude be described as 'I don't want to talk about it any more' ? Presumably not. Finally, when we say Boys will be boys do we really want to say 'I don't want to talk about it any more'? Consider the following example: "Just look at them swilling all that beer, and telling dirty jokes! It's enough to make you sick." "Oh, I don't know, you must let them off the leash a bit now and then and remember that boys will be boys, poor things." (Cowie 1976, 2:75)
It is not at all clear that the second speaker is trying to close the subject. The word 'dismissive' fits hislher attitude better, but again, what exactly does this label mean? If we interpret it in the sense 'I don't want to worry about it' or 'I think we shouldn't worry about it', then these interpretive formulae would indeed seem to fit several other tautological expressions. But would they fit, for example, sayings such as The law is the law? One can of course say that this saying, too, is 'dismissive', because here the speaker is dismissing somebody's objections to a particular law, or somebody's attempts to get around a law. But surely, in this case, 'dismissive' does not mean 'I don't want to talk about it any more'; and if it doesn't mean that, then what does it mean? The same question would apply to another idea which is sometimes mentioned in connection with tautological constructions - the idea of 'acceptance'. Speaking loosely, one can easily agree that by saying War is war or Kids are kids or Whatever'll be will be, the speaker 'accepts' the situation. But in what sense does he accept it? For a number of tautological expressions, I have proposed the component 'I don't want to feel something bad because of this'. But when the vague term 'acceptance' is replaced with a clear semantic formula which can be checked against the actual use of different expressions, it becomes clear that the formula in question is not equally appropriate in all cases. For example, when someone says A snake is a snake, can
Is there a semantic invariant?
441
this be reasonably interpreted as conveying 'acceptance' in the sense of 'one shouldn't feel something bad because of that' ? Presumably not. In fact, it would seem that the component of 'acceptance' in the sense just explained (roughly, 'there is no point in getting upset about it') is never present in nominal tautologies with a singular noun and indefinite article. For example, despite all the differences in their respective interpretations, none of the tautologies below seems to imply acceptance: A Mercedes is a Mercedes (one should know the value of Mercedes cars) A point is a point (one shouldn't undervalue it just because it's small) A teacher is a teacher (one should respect teachers) A kiss is just a kiss (one shouldn't overvalue a kiss) A promise is a promise (one shouldn't break a promise) A party is a party (one shouldn't overestimate the differences between one party and another)
On the other hand, acceptance is not tied to a single formal pattern, since it is compatible with plural nouns (without an article): Boys will be boys; with singular nouns referring to complex activities and unaccompanied by an article: War is war; with singular nouns referring to unique denotata: John is John (OK, he has done that bad thing again, but you know John, he's always like that; he has to be accepted as he is, because he cannot be changed); with verbal tautologies of different kinds: If it's raining it's raining, Whatever will be will be, and so on. Perhaps the most plausible candidate for a common denominator of the different tautological expressions commonly used in English is something like this: this cannot be changed (or: this cannot be not like this) This seems to fit expressions conveying distrust such as A snake is a snake, as well as expressions conveying tolerance or indulgence such as Kids are kids or Boys will be boys. It also seems to fit expressions referring to the future, such as Whatever will be will be, those referring to the past, such as Let bygones be bygones, and those referring to activities, such as War is war. It also fits expressions like Enough is enough, which indicate the speaker's determination not to budge, as well as the expression That's that (whether it signals the same determination not to budge or whether it refers to an irreversible event such as death).
442
Boys will be boys: even 'truisms' are culture-specific
Furthermore, it appears to fit well certain other types of sentences which haven't thus far been mentioned in this chapter, such as When you're out of form, you're out of form.
(a remark made by commentator Bill Lorry during a recent televised cricket match, referring to the well-known cricket player, David Gower, who had been having a bad run with the bat). Consider also the following example from a recent newspaper article: In the Brussels stadium we saw a display of lethal stupidity. Meanwhile, the dead are dead and are past wondering why they had to be killed. (The Weekend Australian 1-2.6.85)
In the context of this article, the tautological expression is neither topic-closing nor dismissive. But the message 'this cannot be changed' is clearly conveyed. An added bonus of an analysis in terms of 'this cannot be changed' is that it offers a possibility of spelling out something like the idea of 'acceptance' in a way which may seem to be applicable to all tautological expressions, including those which convey distrust: one should know: this cannot be changed The frame 'one should know' combined with the component 'this cannot be changed' permits us to express an idea akin to acceptance without conveying connotations incompatible with any specific use to which the tautological pattern is put in English. Nonetheless I have not used the formula 'this can't be changed' in all the explications proposed in this chapter, because I am not fully convinced that it is the optimal one in each particular case, even if it is the optimal 'compromise' between all the different constructions. For example, if one says I don't care what brand -
coffee is coffee.
does one really mean 'this cannot be changed'? Doesn't one mean, rather, something like 'it makes no difference - all Xs (all kinds of coffee) are always the same' ? On the face of it, the idea of 'immutability', of 'this cannot be changed', seems also to fit rather well the three-part patterns, presumably derived from Gertrude Stein's famous sentence A rose is a rose is a rose. For example, the composer Tristram Kerry, to a question about works in an electronic mode, replied (on ABC-FM radio, 10.6.85): "A
Is there a semantic invariant?
443
composition is a composItIon is a composition". Presumably, in saying so Kerry wished to emphasise the continuity between traditional composition and electronic composition. The formula 'this cannot be changed' seems to fit quite well, but does it really capture in the best possible way the communicative import of the utterance? Wouldn't it be more illuminating and more accurate if we spelt it out along the following lines: 'in essence, all Xs are the same' or: 'no matter how Xs differ from one another, there is something about them all that is the same; and this is the most important thing about them'. Consider also the following sentence: It had been hard, for they were links of friendship rather than authority which bound him, but 'chains are always chains, even when made of flowers'. (Tolstoy 1983:258)
Is the idea here that 'this cannot be changed' ? Or is it, rather, that 'Xs can be different from one another in some ways, but in essence they are all the same' ? Another plausible candidate for a semantic invariant could be formulated along the following lines: this cannot be denied (one cannot say: no, it is not like this) Possibly, the restriction on the use of modifiers in tautological constructions (?? Little boys will be little boys, ???A hasty, insincere kiss is a hasty, insincere kiss) could be explained in these terms: if the predicate repeats a bare noun, the identity of the subject and the predicate is obvious, and so it must be obvious to everyone that the claim made by the speaker must be valid. The use of a more complex noun phrase could obscure the identity of the subject and the predicate, and so it would no longer be 'obvious' that what the speaker is saying cannot be rejected. The foregoing discussion of a possible semantic invariant of tautological constructions can be summarised in the form of the following four hypothetical components, which I propose as a starting point for further investigation: (a) I don't want to say more things about this ('dismissive quality') (b) it cannot be not like this ('immutability', 'it cannot be changed')
444
Boys will be boys: even 'truisms' are culture-specific
(c) all things of this kind are the same ('identity') (d) one cannot say: no ('undeniability') For English nominal tautologies, a more specific set of four recurring components can be tentatively proposed: (a) (b) (c) (d)
everyone knows ... I know: someone can think ... I think: one should not think this one should know ...
I believe that the problem of a possible semantic invariant of tautological expressions of different kinds cannot be seriously studied without a set of tentative semantic representations of individual expressions or constructions. The formulae proposed in this chapter can be discussed, criticised and revised. In the process, the problem of a hypothetical semantic invariant will, I think, get a chance of being clarified, if not actually solved.
6. The deceptive form of English tautological constructions I think one reason why English tautological constructions have been so widely mistaken for universal devices fully interpretable on the basis of Grice's maxim of quantity, is that their form is deceptively simple, and deceptively similar to logical tautologies. For example, Brown - Levinson (1978:225) draw a distinction between what they call 'conventionalised tautologies' and other tautologies (presumably, non-conventionalised ones). 'Conventionalised tautologies' are illustrated with the Tamil sentence: A vaar-aam avaaru. 'He they say is he.' (i.e. 'big deal him')
By contrast, English tautologies such as: War is war. Boys will be boys.
The deceptive form of English tautological constructions
445
are treated as non-conventionalised. According to Brown and Levinson, by uttering such tautologies, "5 [the speaker] encourages H [the hearer] to look for an informative interpretation of the non-informative utterance." (1978:225). But aren't these English-language tautologies in fact just as 'conventionalised' as the Tamil one? Is it not an ethnocentric illusion to regard English tautologies as 'natural', readily interpretable, based on universal maxims of conversational behaviour, rather than on conventions of the English language? Isn't the difference between these two utterances: War is war. Uustification) A war is a war. (obligation)
largely conventional? Aren't the differences between these expressions: Boys will be boys. Knaben bleiben immer Knaben. (lit. 'Boys remain always boys') Oni ze mal' tiki. (lit. 'They PRT boys') A father is a father. Un pere est toujours un pere. (lit. 'A father is always a father'; cf. Bally 1952: 17)
due to language-specific grammatical conventions? It is indeed remarkable that English nominal tautologies tend to use the very simple formal pattern: N is N (with only minor variations), similar to logical formulae such as p is p. Other languages often use more complex patterns, which include particles, modal verbs, adverbs of time such as immer or toujours ('always') and so on. But is the English formal pattern really more 'natural', less conventionalised, than the complex patterns used by other languages? And isn't it the case that, despite appearances, English, too, has a number of different formal patterns, associated with different interpretations (N is N, A N is a N, Ns are Ns, Ns will be Ns, and so on)? If the analyses contained in this chapter do something to counteract the spread of such illusions, its purpose will have been achieved.
446
Boys will be boys: even 'truisms' are culture-specific
7. The culture-specific content of tautological patterns Tautologies are like gestures or interjections: they seem 'natural', universal, and self-explanatory; and they are all that, to some extent; but they are also conventional, language-specific, and culturally revealing. They manifest both 'natural logic' and 'cultural logic'; and the more 'natural' they appear the more culturally revealing they are. Consider, for example, the gesture of 'nose-tap', widespread in Europe and probably elsewhere as well (see Morris et al. 1979). The nose is naturally associated with smell, and with sniffing, and sniffing can be seen as naturally associated with alertness and with special knowledge. It seems entirely 'natural', therefore, that the nose-tap gesture is indeed associated with meanings of this kind. Nonetheless, the nose-tap does not mean the same in different parts of Europe. In English-speaking and Italian-speaking areas, it is used mainly to indicate a shared secret (with two peaks occurring in Scotland and Sardinia). But, as Morris et al. (1979:220) note, "almost as popular... is the meaning of friendly warning ['Be alert! ']. This was almost non-existent in the north, but was strongly favoured on the Italian mainland, with the peak at Rome and Naples." In addition to this friendly meaning, an unfriendly one - 'You are nosey!' - has also emerged in parts of Europe. "As an accusation of interference, the gesture was almost entirely confined to the English-speaking region, with only two isolated cases elsewhere." (1979:224). This is a perfect illustration of the way how culture-specific concerns make use of 'natural' and universally-available means. In Italian culture, group solidarity and interpersonal involvement play a greater role than desire for personal autonomy, non-imposition, and 'non-interference', whereas in the Anglo-Saxon culture the latter values are known to play a dominant role. The sign in question (the nose-tap) is seen everywhere as 'natural' and self-explanatory; but in fact, its interpretation is shaped partly by the shared cultural traditions, and cultural preoccupations, of a particular community of speakers. Exactly the same applies to tautologies. They are like convenient allpurpose vessels, into which 'basic truths' can be poured - but 'basic truths' are different in different cultures. Consider, for example, the following description of an American Jewish wedding, in a novel by Bernard Malamud:
The culture-specific content of tautological patterns
447
The guests, including the notables, rise, lift their feet, and dance . ... The women serve a feast of chicken with sesame and tomatoes, roasted yams, and palm wine . ... Those who feel like crying, cry. A wedding is a wedding. (Malamud 1971:217)
The form of the tautology in the last line is impeccably English, but its use in this passage is slightly odd, because it doesn't fit into any of the semantic categories which have evolved in the English language (such as absolute generalisation, indifference, obligation, or undeniable value). Apparently, what Malamud is trying to convey is this: 'human life is complex, impenetrable, uncontrollable; its joys must be interwoven with sorrows; it is natural for people to cry, as well as to dance and to eat, at a wedding; such is our experience of life.' But this is Jewish wisdom, embodied in particular in the Yiddish language and lore (cf. Matisoff 1979), not Anglo-Saxon wisdom. As a result, Malamud's tautology sounds like a calque, and a cultural loan, from Yiddish. The same applies to another tautological expression, used by the same writer in the same context. At the same wedding the rabbi addresses the newlyweds in the following words: Now you are man and wife ... I feel like crying, but why should I cry if the Lord says, "Rejoice!" Willie and Irene, listen to me. Oh, what a hard thing is marriage in the best of circumstances. ... All I am saying is the world is imperfect. ... Willie and Irene, to enjoy the pleasures of the body you don't need a college education; but to live together in love is not so easy. Besides love that which preserves marriage is that which preserves life; this is mutual trust, insight into each other, generosity and also character, so that you will do what is not easy to do when you must do it. What else can I tell you, my children? Either you understand or you don't. (Malamud 1971:215-216)
Again, at first sight the verbal tautology in the last line seems impeccably English; and yet this time, too, there is something slightly odd about it. In English, ve~bal tautologies of the 'either-or' type refer, normally, to actions, not to understanding. One can of course say in English: "What do you mean, you're not sure? Either you know or you don't" (either you understand or you don't); but in this case the speaker would be urging the addressee to make a choice. Except for such promptings, however, an 'either-or' English tautology would normally refer to an action. (Either he comes or he doesn't). The point of such tautologies was spelt out earlier: 'one shouldn't waste one's time and energy in useless worry; if one can't change something there is no point in fretting
448
Boys will be boys: even 'truisms' are culture-specific
over it; it is better to turn one's mental energy to something more productive'. This is perfectly in keeping with the pragmatism of Anglo-Saxon culture. But the point of Malamud's tautology is different. The rabbi's experience of life is sad ("The world is imperfect"). In uttering his tautology he is not trying to dismiss his sadness and melancholy. He is resigned to the futility of his own talk, and he recognises its pointlessness, but he is not trying to spare himself from 'uselessly feeling something bad'. (After all, Jews are a people who have for centuries, indeed millennia, remembered, and 'uselessly' lamented in their midnight service, the destruction of their temple two thousand years ago. 'Either you understand or you don't ... '.) One might argue, of course, that Malamud's tautologies are 'culturally odd', not 'linguistically unacceptable'. I wouldn't want to say that, but this is not the point: I have presented in this chapter many examples of tautologies whose meanings couldn't possibly be 'worked out' by speakers of other languages, and which are, therefore, indubitably language-specific, as well as culture-specific. But no matter how one categorises the slight 'deviance' of Malamud's tautologies, they illustrate very well the way how certain formal patterns, which are felt (in a particular culture) to embody unquestionable universal wisdom, can in fact specialise in expressing wisdom which is highly culture-specific. What is 'self-evident' in one culture may be questionable or unacceptable in another. What is seen as a 'fundamental truth' in one culture may be seen as not worth saying in another. What is seen as an 'unquestionably wise attitude' in one culture might be seen as foolishness in another. Perhaps this is, ultimately, the sense of all (most?) tautologies, in all (most?) languages: that they - like proverbs - are supposed to embody 'wisdom'; and, in particular, that they recommend certain attitudes to life as 'unquestionably wise' and as tested by generations of people.
8. Conclusion In describing a specific language (for example, English) as a system, somewhere or other a line will have to be drawn between what is in and what is out, between grammar and 'human nature', between meaning and 'implicature'. Does it matter very much where exactly this line is
Conclusion
449
drawn? Does it matter, for example, if constructions such as Boys will be boys are counted as being 'in' or 'out', i.e. as part of English grammar, whose meaning is language-specific and has to be included in the linguistic description of English as an individual language, or as free collocations whose use is 'calculable' from general principles of human conversation? In my view, it is a matter of fundamental importance. Obviously, this importance doesn't derive from the frequency of use, or the indispensability, of 'tautological sentences' as such (although their cultural importance may be greater than one would initially suspect). It derives from the fact that one's decision on a 'small' point like this has farreaching consequences with respect to one's whole idea of what linguistics is all about, what it is supposed to do, and what it can do. The basic question is this: should grammar be autonomous of pragmatics, or should 'pragmatic meanings' (i.e. some matters to do with the speakers' assumptions, intentions, thoughts and feelings) be accounted for in the same over-all descriptive framework which is used for 'objective' or truth-functional aspects of language use? Many contemporary linguists seem to feel that they are caught in an insoluble dilemma. On the one hand, it has been demonstrated again and again that 'pragmatic' (or 'subjective ') and 'objective' aspects of language use are interrelated and that grammar interacts with illocutionary force and other 'pragmatic' matters. Furthermore, consistent attempts to separate grammar from 'pragmatic meanings' lead to the paradoxical conclusion that a good deal of conversational English is 'ungrammatical'. For example, perfectly ordinary and perfectly acceptable sentences such as Would you please be quiet?, or Why don't you be quiet! have been classified as 'ungrammatical' - simply because 'autonomous' grammar cannot possibly account for their use. (See for example Bach - Harnish 1979.) This outcome would seem to represent a reductio ad absurdum of the autonomous (non-illocutionary) grammar stand. On the other hand, the most influential attempt to develop a framework for an integrated description of grammar and 'pragmatic' phenomena such as illocutionary force, the approach associated with the generative semantics school, has ended in self-acknowledged defeat. This has resulted in widespread disillusionment not only with generative semantics but also with any attempts to integrate a grammatical and a pragmatic description of language. What is to be done, then? To many linguists, the only salvation seems to be offered by a Gricean approach to language. This state of
450
Boys will be boys: even 'truisms' are culture-specific
mind can perhaps be expressed as follows: 'Somebody has to account for language use, but we linguists have now come to realise that we cannot do it. Fortunately, we don't have to feel guilty about it any longer. We now see that it is simply not our responsibility. Another science will do it: a science of human behaviour in general and human conversational behaviour in in particular. Linguists can concentrate on studying language structure. And in fact we can now say that it would be a mistake to confuse language structure with language use. Grammar is one thing, and illocutionary force another; meaning is one thing and 'implicature' another. We neither can nor should try to account for language use. For example, we don't have to try to predict the communicative import of sentences such as Boys are boys or Business is business. Certainly, such matters are not devoid of interest, but they will be taken care of by the science of conversational behaviour.' In Sperber's - Wilson's words (1981:296, on a slightly different but related matter), "Grice's proposal would relieve semantic theory of the problems of defining" pragmatic meanings. After the self-proclaimed collapse of generative semantics, which tried to develop a framework for a semantic analysis of illocutionary meanings, contemporary semantics seems to have lost all confidence in itself. A philosopher who comes and apparently 'relieves the linguist' from the obligation of analysing many kinds of troublesome meanings, and in particular, illocutionary meanings, tends to be treated like a saviour, and to be accorded total faith. In my view, this faith in the omnipotence of Gricean maxims is bound to be disappointed. Neither tautological constructions nor numerous other matters which have been left to the postulated new science based on Gricean maxims, will be automatically taken care of. Ultimately, numerous 'tautological sentences' used in conversational English will simply have to join the ever-growing pile of rejects - i.e. of sentences which are perfectly acceptable but which are 'ungrammatical', because neither Gricean maxims nor an autonomous grammar can account for the limits of their use. (For example, why can one say Boys will be boys but not *War will be war or *Boys might be boys? Why can one say Kids are kids but not ?Bottles are bottles or *Clouds are clouds? Why can one say Enough is enough but not *Much is much? Why can one say A bet is a bet but not *Bets are bets? Of course nobody denies these days that pragmatic information is reflected in some aspects of grammar, and that, consequently, some aspects of grammar cannot be fully accounted for without some reference to the speaker's and hearer's assumptions, feelings, and so on. But so
Conclusion
451
long as it seems possible to regard such reflections as minor and isolated, they can be ignored by the grammarian and the ideal of autonomous grammar can be upheld (even if it has to be done at the cost of discarding some acceptable sentences as 'ungrammatical '). I would argue, however, that pragmatics pervades grammar, that the two intertwine in countless areas, and that the bulk of conversational English can be explained only on the basis of this interaction. (Cf. Bally 1952.) Given that 'conversational meanings' (Gazdar's term, 1979:55) cannot be explained on the basis of universal principles of conversation, can they perhaps be accounted for on the basis of some language-specific science of language use, separate from grammar (as suggested by Morgan 1978)? If we followed this line of reasoning we could say that the use of tautological constructions is determined partly by conventions which are specific to English, but that these conventions are not 'grammatical' in nature, and can be kept apart from a linguistic description of English structure. To my mind, this program is no more realistic than Grice's (or rather, than the program advanced by the linguistic followers of Grice, who often tend to be more Gricean than Grice himself). Attitudinal meanings enter the core of grammar. If we establish any correlation between the imperative (Do it!) and the meaning 'I want you to do it', or between the interrogative structure and the meanings 'I want to know it' or 'I want you to say it', or between tag questions and meanings such as 'I think X, I assume you think the same, I want you to say it', or between exclamations and meanings such as 'I feel something', we are already 'mixing' grammar and pragmatics. But can any empirically adequate grammar ignore such correlations? Once it is recognised that a semantic metalanguage capable of representing both 'objective' and 'subjective' aspects of meaning constitutes a sine qua non of linguistic description then subtle pragmatic meanings like those encoded in tautological constructions cease to present a serious problem. The linguist can recognise that such constructions are at least partly language-specific, and can seek to account for their use without appealing to any deus ex machina in the form of extra-grammatical linguistic conventions: it becomes clear that the relevant meanings can be modelled in the same semantic metalanguage as all other kinds of meaning.
452
Boys will be boys: even 'truisms' are culture-specific
Furthermore, once the language-specific meanings of various 'tautological' constructions are spelt out in a rigorous (and yet selfexplanatory) semantic metalanguage, it becomes possible to study universal tendencies and regularities in the use of such constructions, which are no less real, and no less accessible to precise semantic analysis, than the language-specific uses to which such constructions can be put. Theories of conversation (such as Grice's) can then proceed without any unrealistic burdens being placed on their shoulders. 'Ungrammatical' sentences such as Why don't you be quiet or Would you please be quiet can be rehabilitated as fully grammatical encodings of languagespecific pragmatic meanings. An integrated theory of linguistic description can, once more, be set out as a goal of linguistics - and as a responsibility of which no other sciences can relieve it.
Chapter 11
Conclusion: selDantics as a key to cross-cultural pragmatics
Language is, it is said, a tool of communication. As a statement, this is of course true, but as a definition, it is narrow, one-sided, and inadequate. Language is indeed a tool of communication, but - crucially - it is also a tool of human interaction. By means of language, we express our personality, our thoughts, intentions, desires, and feelings; and by means of language we relate to other people. This book develops a framework for studying language as a tool of human interaction. 1 have tried to show that we interact with other people by expressing certain meanings - and 1 have developed a method for revealing and stating these meanings in a clear, simple, and rigorous way. For example, modes of interaction associated with English words such as reprimand, praise, thank, apologise, threaten, promise, warn, and complain have been analysed by means of simple and intuitively understandable semantic ~omponents such as the following ones: you did something bad someone did something good you did something good for me 1 did something bad to you 1 will do something bad to you 1 will do something good for you something bad may happen to you something bad happened to me
(reprimand) (praise) (thanks) (apology) (threat) (promise) (warning) (complain)
Components of this kind are formulated in a kind of simple, reduced English, and they can be immediately understood (via English). But unlike the great majority of 'normal' English sentences they could be readily translated into other languages, because they rely on very simple grammatical patterns, and on universal, or near-universal, human concepts such as 'you' and 'I', 'good' and 'bad', or 'do' and 'happen', which have their counterparts in all, or nearly all, languages of the world.
454
Conclusion: semantics as a key to cross-cultural pragmatics
Components of this kind, formulated in the language of universal 'semantic primitives', allow us to study human interaction in a systematic and rigorous way, without relying on any technical jargon or on mathematical and logical models, which, when employed in the study of language and language use, often give the impression of scientific rigour but in fact obscure rather than clarify the phenomena under consideration. Human interaction carried out by means of language is shaped and coloured to a very considerable extent by simple meanings such as 'you did something bad/good' or 'I did something bad/good', 'I want something' or 'I don't want this', and so on; but meanings of this kind are obscured, not clarified, when human interaction is discussed in terms of 'negative' or 'positive face' (cf. Brown - Levinson 1978), or 'cooperativeness' (cf. Grice 1975) or 'indirectness' (cf. Searle 1975). Above all, complex terms of the kind usually employed in studies of linguistic interaction are derived from English (academic English), and don't have exact counterparts in other languages of the world. As a result, the description of human interaction carried out by means of such labels is usually remarkably ethnocentric (anglocentric). This is in keeping with the general anglocentric bias of mainstream modern pragmatics, which developed largely under the influence of British and American philosophers of language, who drew their examples almost exclusively from English and who postulated all sorts of 'universal maxims' or 'universal principles' of human interaction on the basis of Anglo-Saxon or Anglo-American cultural norms and expectations. The present book is based on a wide range of data, drawn from a wide range of languages, including English, Italian, Russian, Polish, Yiddish, Hebrew, Japanese, Chinese, Korean, Walmatjari (an Australian Aboriginal language), and many others. It also draws on different varieties of English, including Australian English and Black American English. It rejects 'universal maxims' of human behaviour based on mainstream British or American English, and it refuses to treat descriptive categories based on academic English as a 'culture-free' and language-independent analytical framework. Instead, it relies on a 'natural semantic metalanguage', based on universal or near-universal semantic primitives; and it employs this metalanguage to explore and to analyse a wide variety of cultural norms and cultural traditions from a universal, language-independent perspective. It recognises the enormous variety of 'communicative styles' linked with different human languages and refuses to treat this variety
Conclusion: semantics as a key to cross-cultural pragmatics
455
in terms of deviations from one 'basic', 'natural' model (that associated with mainstream American, or British, English). At the same time, it recognises the underlying unity of human interaction, determined by the universality of concepts such as 'you' and 'I', 'do' and 'happen', 'good' and 'bad', or 'want' and 'know'; and by the universality of semantic components such as 'I want something', 'I know something', 'you know something', 'I don't want this', 'I/you/someone did something bad', 'I don't want this', 'I want you to say something', 'I want you to do something', and so on. Different communicative styles, and different norms of social interaction, are described here in terms which are not only universal (languageand culture-independent), but also simple and easy to understand, to the point of being accessible to children, and to second-language learners (for example, immigrants) with a very limited knowledge of English. It can be hoped, therefore, that the kind of simple metalanguage used here to investigate different modes of human interaction and different interactive strategies associated with different languages can be used not only as a tool for investigating linguistic interaction in different cultural settings, but also as a basis for teaching it; and in particular, as a basis for teaching successful cross-cultural communication.
Notes
1. The arrow (=» indicates an implication, which forms a part of the full meaning of the expression. 2. The key place of forms of address in cross-cultural pragmatics is highlighted by Friederike Braun's (1988) valuable study, Terms of address: problems of patterns and usage in various languages and cultures, which came to my attention after the completion of the present work. 3. Baxtin's study "Problema recevyx zanrov" ['The problem of speech genres'] was published in 1979, but actually written in 1952, so it seems to me that he can be rightly seen as a precursor of the modem Western literature on speech acts and speech genres which flowered in the 1960s and 1970s. In fact, Baxtin's first work on related topics was published under the assumed name of Volosinov as early as 1929. But the 1929 book, interesting and original as it is, does not focus on speech genres, as does the 1952 study. For an English translation of this pioneering work, see Baxtin (1986). It is devoted specifically to speech genres, and it outlines a program which I find particularly congenial. 4. Since most contemporary linguistic writings on speech acts seem to assume that this idea is a modern invention, I think it is in order to offer here a lengthy quotation which will help put the matter in a more realistic historical perspective: Starting from the domain of dialectic and probably utilising already existing divisions of kinds of speech the Stoics developed rather elaborate classifications of complete lekta. From the several passages in which such lists are given we learn that apart from axiomata and the two kinds of questions the following varieties of complete lekta were distinguished. (a) That which is more than an axioma or is like an axioma, for example: 'How beautiful the Parthenon (is)', 'How like to Priam's sons the cowherd (is)'. It differs from an axioma in so far as it includes some emotional reaction, usually expressed by the particle hos ('How'). It is also called thaumastikon, the lekton of admiration or astonishment; as such it may be contrasted with the psektikon, the lekton of censure. (b) Another sort of question, put to oneself in order to express doubt or despair (epaporetikon): 'Are sorrow and life something akin?'. (c) A command (prostaktikon), for instance: 'Come thou hither, 0 lady dear'.
458
Notes
(d) An oath (horkikon, omotikon, epomotikon), for instance: 'By this sceptre', 'Let earth be my witness in this'. (e) A prayer (aratikon, euktikon) , for example: 'Zeus, my father, who rulest from Ida, majestic and mighty, victory grant unto Ajax and crown him with glory and honour'. Sometimes a distinction is drawn between prayer (euktikon) and deprecation or curse (aratikon); an instance of the latter is 'Even as this wine is spilt, so may their brains be spilt earthwards'. (f) An address or greeting (prosagoreutikon), for example: 'Most honoured son of Atreus, lord of the warriors, Agamemnon'. (g) A supposition or assumption (hypothetikon), for instance: 'Let it be supposed that the earth is the centre of the sphere of the sun'. (h) An exhibition of a particular instance (ekthetikon): 'Let this be a straight line'. According to DL VII, 196, Chrysippus had written a treatise about ektheseis. (i) An explanation or elucidation (diasaphetikon). (Nuchelmans 1973:63) 5. This earlier version of the performative analysis is also described by Nuchelmans: If one utters the sentence '0, that the king might come' (utinam rex veniret), the composite thought produced in the hearer, the intellectus of that sentence, is the same as that produced by the sentence 'I wish that the king would come' (volo regem venire or opto, ut rex veniat). If one utters the sentence 'Help me, Peter' (adesto, Petre), either as a command or as a prayer, the intellectus is the same as that of the sentence 'I order you to help me' (praecipio, ut adsis mihi) or of the sentence 'I pray you to help me' (deprecor, ut adsis mihi). The utterance 'Peter!' (0 Petre) produces the same mental counterpart as the sentence 'I call you, Peter'. And the question 'Did Socrates come?' produces the same thought as the sentence 'I ask you if Socrates came'. Similarly, the one-word utterances papae and heu produce the same thought as ego admiror ('I admire (it)') and ego doleo ('I feel grief (at it) '), although in the first two cases the thought is non-composite and in the second two cases the thought is composite. This means that in spite of the differences in force of the vocal utterances, namely between the non-declarative utterance and the statement-making utterance which in each case corresponds to it, there is no difference between the mental counterparts of the non-declarative utterance and the corresponding statement-making utterance. The composite thought produced in the hearer's mind by each of the several non-declarative utterances is the mental counterpart of the propositio that may take its place. (Nuchelmans 1973:148)
Notes
459
6. Examples drawn from popular fiction and plays are indicated as follows: AC: Anna Christie (O'Neill 1960) GH: The great Gillie Hopkins (Paterson 1978) GP: The glittering prizes (Raphael 1977) NI: The night of the iguana (Williams 1961) WE: Welded (O'Neill 1958) 7. Felix Ameka (p.c.) has suggested that 'vocal gestures' such as YukI should be distinguished from 'speech acts' such as I feel disgusted along the following lines: I feel disgusted.
I say: I feel disgusted I want someone to know it I say this because of that YukI I feel disgusted I want someone to know it I do this: [vocal gesture] because of that
I find this an interesting proposal, which requires further investigation. 8. For earlier attempts at semantic analysis of some interjections based on a controlled semantic metalanguage, see Ameka (1987; to appear) and Hill (1985); see also Wierzbicka (1986a). 9. The expression 'I now know' refers to something that has just come to the speaker's attention. It might seem more natural to use the expression 'I perceive' instead. But perceive, unlike know, cannot be presumed to be an indefinable or a lexical universal, and consequently know must be preferred to it, as an element of the semantic metalanguage. 10. Some readers have objected to the suggestion that wow necessarily implies 'something good'. The counterexamples they have offered, however, could all be interpreted as sarcastic or ironic, that it, as relying for their effect on an underlying presumption of 'good'. For example, a parent may say Wow! in response to either I got three As today! or I got three Ds today!, but in the latter case such a response as Wow! would normally be interpreted as sarcastic. 11. The range of use of yuk is not uniform throughout the English-speaking world; it is particularly widely used in Australia. 12. Consider also the following dialogue from a play by a leading Australian playwright: Don: (to Mal) Why in the hell do you get her pregnant every eighteen months? Mal: She won't take the pill. ... Don: (to Mal) Use Silvertex.
460
Notes
Mal: (with distaste) YukI (Williamson 1973:210) 13. As pointed out to me by Professor Werner Winter (p.c.), Latin had a special dictum corresponding to the English 'tautologies of human nature': sunt pueri pueri; pueri puerilia tractant, literally '(male) children are (male) children; children do childish things'. As Professor Winter points out, this dictum makes the relationship between members of the class and their prototypical behaviour beautifully explicit. At the same time, the word puerilia 'childish (things)', clearly implies an indulgent attitude, incompatible with serious condemnation.
Bibliography
Abrahams, Roger 1970 "Rapping and capping: Black talk as art", in: Szwed (ed.), 132142. 1974 "Black talking on the streets", in: Bauman - Sherzer (eds.), 240-262. 1976 Talking black (Rowley, Mass.: Newbury House). Ameka, Felix 1986 The use and meaning of selected particles in Ewe (M.A. thesis, Australian National University). "A comparative analysis of conversational routines in two 1987 languages: English and Ewe", Journal of Pragmatics 11 :299-326. to appear "Agoo can be English, can it? On the semantic representation of linguistic routines", in: David Wilkins (ed.), Conceptual primitives and semantic analysis: A cross-linguistic perspective (Berlin: Mouton de Gruyter). AND 1988 The Australian national dictionary: a dictionary ofAustralianisms on historical principles (W.S. Ramson, ed.) (Melbourne: Oxford University Press). Arensburg, Conrad M. - Arthur H. Niehoff 1975 "American cultural values", in: Spradley - Rynkiewich (eds.), 363-377. Ash, see Garton Ash Atlas, Jay 1984 "Comparative adjectives and adverbials of degree: an introduction to radically radical pragmatics", Linguistics and Philosophy 7:347-377. Austin, John L. 1962 How to do things with words (Oxford: Oxford University Press). Bach, Kent - Robert Harnish 1982 Linguistic communication and speech acts (Cambridge, Mass.: MIT Press). Baker, Sidney J. 1959 The drum: Australian character and slang (Sydney: Currawong). 1966 The Australian language (Sydney: Currawong). Bakhtin, see Baxtin Bally, Charles 1952 Le langage et la vie (3rd expanded edition) (Geneve: Droz).
462
Bibliography
Bankowski, Andrzej 1971 "Jeszcze i jui" [Jeszcze andjui], J(lzyk Polski 51:21-30. 1975a "0 kwantyfikacji j~zykowej i partykule az" [On linguistic quantification and the particle az], Prace Filologiczne 25:183-189. 1975b "Jeszcze przy komparatywach" [Jeszcze with comparatives], Poradnik J~zykowy 3:148-150. 1976 "Opozycja semantyczna partyku} dopiero i jui" [The semantic opposition of the particles dopiero and jui], Prace Filologiczne 26: 13-38. 1977 "Opozycja semantyczna partykul jeszcze i jui" [The semantic opposition of the particles jeszcze and jui], in: Siatkowski (ed.), 7-46. Barnlund, Dean 1975a "Communicative styles in two cultures: Japan and the United States", in: Kendon - Harris - Key (eds.), 427-456. Public and private self in Japan and the United States: communi1975b cative styles of two cultures (Tokyo: Simul). Bartlett, John 1980 Familiar quotations (15th edition) (London: Macmillan). Barzini, Luigi 1964 The Italians (London: Hamish Hamilton). Bates, Elizabeth 1976 Language and context: the acquisition of pragmatics (New York: Academic Press). Bauman, Richard - Joel Sherzer (eds.). 1974 Explorations in the ethnography of speaking (Cambridge: Cambridge University Press). Baxtin, Mixail 1979 "Problema recevyx zanrov" [The problem of speech genres] [Writ[1952-3] ten in 1952-3, published in his Estetika slovesnogo tvorcestva 1979, 237-280 (C.G. Boxarov, ed.)] (Moskva: Iskusstvo). 1986 Speech genres and other late essays [Mikhail Bakhtin] (Vern W. McGee, trans.) (Austin: University of Texas Press). Benedict, Ruth 1947 The chrysanthemum and the sword: patterns of Japanese culture (London: Seeker & Warburg). Blount, Ben (ed.) 1974 Language, culture and society (Cambridge, Mass.: Winthrop). Blum-Kulka, Shoshana 1982 "Learning to say what you mean in a second language: a study of the speech act performance of learners of Hebrew as a second language", Applied Linguistics 3:29-59.
Bibliography
463
Blum-Kulka, Shoshana - Elite Olshtain 1984 "Requests and apologies: a cross-cultural study of speech act realization patterns", Applied Linguistics 5: 196-213. Blum-Kulka, Shoshana - Brenda Danet - Rimona Gherson 1985 "The language of requesting in Israeli society", in: Forgas (ed.), 113-139. Blyton, Enid 1980 The mystery of the Banshee Towers (London: Granada). Bogardus, Emory S. 1933 "A social distance scale", Sociology and Social Research 17: 265-271. Boguslavskij, Igor' M. 1985 Issledovanija po sintaksiceskoj semantike. Sfery dejstvija logiceskich slov [Investigations into syntactic semantics: the spheres of action of logical words] (A.P. Ersov, ed.) (Moskva: Nauka). Bogustawski, Andrzej 1966 Semantyczne pojfcie liczebnika [The semantic concept of the numeral] (Wroclaw: Ossolineum). 1972 "0 wyrazeniach oznaczaj~cych zmian~" [On expressions denoting change], in: Wierzbicka (ed.), 147-149. "On 'the world' ", Linguistica Silesiana 1:63-69. 1975 "Semantic and pragmatic aspects of reference related problems", 1981a in: Danes - Viehweger (eds.), 1-111. "Wissen, Wahrheit, Glauben: zur semantischen Beschaffenheit 1981b des kognitiven Vokabulars", in: Bungarten (ed.), 54-84. "Also from all so: on a set of particles in service of efficient 1986 communication", Journal of Pragmatics 10:615-633. "Glosa do ksi~gi akt6w mowy" [A gloss to the book on speech 1988 acts], Pamiftnik Literacki 79:103-124. 1989 "Knowledge is the lack of lack of knowledge, but what is that lack lack of?", Quaderni di Semantica 10.1: 15-31 (Round table on semantic primitives, 1). Bogus}awski, Andrzej - Bozena Bojar (eds.) 1986 Od kodu do kodu [From code to code] (Warszawa: Panstwowe Wydawnictwo Naukowe) (Festschrift for Olgierd Wojtasiewicz). Bolinger, Dwight 1972 Degree words (The Hague: Mouton). 1977 Meaning and form (London: Longman). 1982 "Intonation and its parts", Language 58:505-533. Bowles, Colin 1986 G'day! Teach yourself Australian in 20 easy lessons (Sydney: Angus & Robertson).
464
Bibliography
Braun, Friederike 1988 Terms of address: problems of patterns and usage in various languages and cultures (Berlin: Mouton de Gruyter). Brown, Penelope - Stephen Levinson 1978 "Universals in language usage: politeness phenomena", in: Goody (ed.), 56-310. 1987 Politeness: some universals in language usage (Cambridge: Cambridge University Press). Brown, Roger - Albert Gilman 1972 "The pronouns of power and solidarity", in: Giglioli (ed.), 252282. Brzechwa, Jan 1983 Brzechwa dzieciom [Brzechwa for children] (Warszawa: Nasza Ksi~garnia).
BUhler, Karl 1933 Sprachtheorie (Jena) [Reproduced 1965, Stuttgart: Fischer]. Bungarten, Theo (ed.) 1981 Wissenschaftsprache: Beitriige zur Methodologie, theoretischen Fundierung und Deskription (MUnchen: Fink). Butwin, Frances 1958 Translator's introduction to Sholom Aleichem [Rabinowitz], The old country (London: Andre Deutsch). Buzo, Alexander 1974 "The front room boys", in: Four Australian plays (Ringwood, Vic.: Penguin). 1979 Norm and Ahmed, Rooted, The Roy Murphy Show (Sydney: Currency Press). Caton, Charles (ed.) 1963 Philosophy and ordinary language (Urbana, Ill.: University of Illinois Press). Cattell, Ray 1973 "Negative transportation and tag questions", Language 49:612639. Chisholm, William S. 1984 Interrogativity: a colloquium on the grammar, typology and pragmatics of questions in seven diverse languages (Amsterdam: John Benjamins). Chomsky, Avram Noam 1957 Syntactic structures (The Hague: Mouton). 1975 Reflections on language (New York: Pantheon). Clancy, Patricia 1986 "The acquisition of communicative style in Japanese", in: Schieffelin - Dchs (eds.), 213-250.
Bibliography
465
Clark, Herbert - Dale Schunk 1980 "Polite responses to polite requests", Cognition 8:111-143. Clyne, Michael (ed.) 1976 Australia talks: essays on the sociology of Australian immigrant and Aboriginal languages (Canberra: Australian National University) (Pacific Linguistics, D23). Cole, Peter (ed.) 1978 Syntax and semantics 9: Pragmatics (New York: Academic Press). 1981 Radical pragmatics (New York: Academic Press). Cole, Peter - Jerry Morgan (eds.) 1975 Syntax and semantics 3: Speech acts (New York: Academic Press). Coleman, Peter (ed.) 1962 Australian civilisation: a symposium (Melbourne: Cheshire). Comrie, Bernard 1984a "Russian", in: Chisholm (ed.), 7-46. 1984b [Plenary session discussion], in: Chisholm (ed.), 255-287. Condon, John C. - Mitsuko Saito (eds.) 1974 Intercultural encounters with Japan: communication - contact and conflict (Tokyo: Simul). Conway, Ronald 1971 The great Australian stupor: an interpretation of the Australian way of life (Melbourne: Sun Books). Cooke, Joseph R. 1968 Pronominal reference in Thai, Burmese, and Vietnamese (Berkeley: University of California Press). Coulmas, Florian 1981 "Poison to your soul: thanks and apologies contrastively viewed", in: Coulmas (ed.), 69-91. Coulmas, Florian (ed.) 1981 Conversational routine: explorations in standardized communication situations and prepatterned speech (The Hague: Mouton). 1986 Direct and indirect speech (Berlin: Mouton de Gruyter). Cowie, Anthony P. - Ronald Mackin 1976-83 Oxford dictionary of current idiomatic English 1-2. (Oxford: Oxford University Press). Crawford, Raymond Maxwell 1970 Australia (London: Hutchinson). Dal', Vladimir 1955 Tolkovyj slovar' zivogo velikorusskogo jazyka [Explanatory [1882] dictionary of the contemporary Great Russian language] 1-4.
466
Bibliography
(Moskva: Gosudarstvennoe izdatel' stvo inostrannyx nacional 'nyx slovarej). Danes, Frantisek - Dieter Viehweger (eds.) 1981 Pragmatische Komponenten der Satzbedeutung (Berlin: Akademie der Wissenschaften der DDR) (Linguistische Studien Reihe A, 91/1 ). Darwin, Charles 1955 The expression of the emotions in man and animals (New York: [1874] Philosophical Library). Davies, Norman 1981 God's playground: a history of Poland in two volumes 1-2. (Oxford: Clarendon Press). 1984 Heart of Europe: a short history of Poland (Oxford: Clarendon Press). Davison, Alice 1975 "Indirect speech acts and what to do with them", in: Cole Morgan (eds.), 143-185. Deakin, Greg 1981 Indirect speech acts and intonation (M.A. thesis, Australian National University). de Tocqueville, Alexis 1953 Democracy in America (Henry Reeve, trans.) (New York: Alfred [1835-40] A. Knopf). DeVos, George 1985 "Dimensions of the self in Japanese culture", in: Marsella DeVos - Hsu (eds.), 141-184. Dixon, Robert M.W. 1977 "Where have all the adjectives gone?", Studies in Language 1: 1980 [Reprinted in his Where have all the adjectives gone? and other essays in syntax and semantics, 1982, 1-62] (The Hague: Mouton). Doherty, Monika 1973 " 'Noch' and 'schon' and their presuppositions", in: Kiefer Ruwet (eds.), 154-177. Doi, Takeo The anatomy of dependence (John Bester, trans.) (Tokyo: Kodan1973 sha). "Some psychological themes in Japanese human relationships", 1974 in: Condon - Saito (eds.), 17-26. Doroszewski, Witold 1938 J~zyk polski w Stanach Zjednoczonych [The Polish language in the United States] (Warszawa: Naklad Towarzystwa Naukowego Warszawskiego).
Bibliography
467
Dostoevskij, Fedor 1976 Brat'ja Karamazovy [The brothers Karamazov] (Polnoe sobranie socinenij [Complete collected works], 14) (Leningrad: Nauka). Dostoevsky, Fedor 1974 The brothers Karamazov (Constance Garnett, trans.) (London: Heinemann). Dovlatov, Sergej "Krug rasskazcikov" [The circle of storytellers], Kontinent 1977 11: 129-140. Doyle, Arthur Conan (Sir) 1981 The complete Sherlock Holmes (London: Seeker & Warburg). Dragunskij, Viktor 1968 Deniskiny rasskazy [Little Denis' stories] (Moskva: Detskaja Literatura). Drazdauskiene, Maria-Liudvika "On stereotypes in conversation", in: Coulmas (ed.), 1981:55-68. 1981 Eades, Diana 1982 "You gotta know how to talk ... : information seeking in SouthEast Queensland Aboriginal society", Australian Journal of Linguistics 2:61-82. Ervin-Tripp, Susan 1974 "Sociolinguistics", in: Blount (ed.), 1974:268-334. 1976 "Is Sybil there? The structure of some American English directives", Language in Society 5:25-66. Facey, Albert B. A fortunate life (Ringwood, Vic.: Penguin). 1981 Field, Deborah 1988 "Tautologies in Japanese", (unpublished paper, Australian National University). Fillmore, Charles 1984 "Remarks on contrastive pragmatics", in: Fisiak (ed.), 119-141. Fishman, Joshua (ed.) 1968 Readings in the sociology of language (The Hague: Mouton). Fisiak, Jacek (ed.) Contrastive linguistics: prospects and problems (Berlin: Mouton). 1984 Fochi, Franco 1966 L' italiano facile [Italian made easy] (Milano: Feltrinelli). Folsom, Franklin 1977 Kniga 0 jazyke [The language book] (Moskva: Progress). [1963] Forgas, Joseph P. (ed.) 1985 Language and social situations (New York: Springer).
468
Bibliography
Fraser, Bruce - Ellen Rintell - Joel Walters "An approach to conducting research on the acquisition of prag1980 matic competence in a second language", in: Larsen-Freeman (ed.),75-91. Gage, William W. (ed.) 1974 Language in its social setting (Washington, DC: Anthropological Society of Washington). Garton Ash, Timothy 1983 The Polish revolution: Solidarity 1980-82 (London: Jonathan Cape). Gazdar, George 1979 Pragmatics: implicature, presupposition and logical form (New York: Academic Press). Geertz, Clifford 1976 The religion of Java (Chicago: Chicago University Press). Gerber, Eleanor 1985 "Rage and obligation: Samoan emotion in conflict", in: White Kirkpatrick (eds.), 121-167. Gibbs, Raymond W., Jr 1985 "Situational conventions and requests", in: Forgas (ed.), 97-110. Giglioli, Pier Paolo (ed.) 1972 Language and social context (Harmondsworth: Penguin). Givan, Talmy 1983 "Review" [of Cole (ed.) 1981], Studies in Language 7:151-155. 1984 "The speech-act continuum", in: Chisholm (ed.), 245-254. 1989 Mind, code, and context: essays in pragmatics (Hillsdale, N.J.: Erlbaum). Gladwin, Thomas - William Sturtevant (eds.) 1962 Anthropology and human behavior (Washington, DC: Anthropological Society of Washington). Goddard, Cliff 1979 "Particles and illocutionary semantics", Papers in Linguistics 12: 185-229. 1985 A grammar of Yankunytjatjara (Alice Springs: Institute for Aboriginal Development). 1986 "The natural semantics of too", Journal of Pragmatics 10:635644. 1989a "Issues in natural semantic metalanguage", Quaderni di Semantica 10.1 :51-64 (Round table on semantic primitives, 1). "The goals and limits of semantic representation", Quaderni 1989b di Semantica 10.2 (Round table on semantic primitives, 2).
Bibliography
469
Goddard, Cliff (ed.) 1987 A basic Pitjantjatjara/Yankunytjatjara to English dictionary (Alice Springs: Institute for Aboriginal Development). Godel, Robert (ed.) 1969 A Geneva School reader in linguistics (Bloomington: Indiana University Press). Goffman, Erving 1981 Forms of talk (Oxford: Blackwell). Goldstein, Bernice - Kyoko Tamura 1975 Japan and America: a comparative study in language and culture (Rutland, Vermont: Tuttle). Goljavkin, Viktor 1968 Ty prixodi k nam, prixodi [Come to us, come] (Leningrad: Detskaja Literatura). Goody, Esther (ed.) 1978 Questions and politeness: strategies in social interaction (Cambridge: Cambridge University Press). Gordon, David - George Lakoff 1975 "Conversational postulates", in: Cole - Morgan (eds.), 83-106. Grandgent, Charles H. 1908 An introduction to Vulgar Latin (Boston: Heath). Green, Georgia 1975 "How to get people to do things with words: the whimperative question", in: Cole - Morgan (eds.), 107-141. Greenberg, Joseph (ed.) 1978 Universals of human language 1-4. (Stanford, Calif.: Stanford University Press). Grice, H. Paul 1975 "Logic and conversation", in: Cole - Morgan (eds.), 41-58. 1981 "Presupposition and conversational implicature", in: Cole (ed.), 183-198. Grochowski, Maciej 1986 Polskie partykuly: skladnia, semantyka, leksykografia [Polish particles: syntax, semantics, lexicography] (Wroctaw: Ossolineum) (Prace Instytutu J~zyka Polskiego PAN 62). Grodzieriska, Wanda 1970 Swierszczowa muzyka [The crickets' music] (Warszawa: Nasza Ksi~garnia).
Gross, Maurice - Morris Halle - Marcel Schtitzenberger (eds.) 1973 The formal analysis of natural languages (The Hague: Mouton). Grossman, Vasily 1980 Zizn' i sud' ba [Life and fate] (Paris: l'Homme). 1985 Life and fate (Robert Chandler, trans.) (London: Fontana).
470
Bibliography
Gumperz, John 1972 "Introduction", in: Gumperz - Hymes (eds.), 1-25. 1982 Discourse strategies (Cambridge: Cambridge University Press). Gumperz, John - Dell Hymes (eds.) 1972 Directions in sociolinguistics (New York: Holt, Rinehart & Winston). Haiman, John 1989 "Alienation in grammar", Studies in Language 13:129-170. Halliday, Michael A.K. - Ruqaiya Hasan 1976 Cohesion in English (London: Longman). Harasawa, Itsuo 1985 "Japanese tautologies", (unpublished paper, Australian National University). Harkins, Jean 1986 "Semantics and the language learner: Warlpiri particles", Journal of Pragmatics 10:559-573. 1988 English as a 'two-way' language in Alice Springs (M.A. thesis, Australian National University). Harris, Max 1962 "Morals and manners", in: Coleman (ed.), 47-67. Harris, Stephen G. 1984 Culture and learning: tradition and education in north-east Arnhem Land (Canberra: Australian Institute of Aboriginal Studies). Hayakawa, Haruko 1985 The semantics of reduplication in Japanese (M.A. thesis, Australian National University). Hibberd, Jack 1974 "Who", "White with wire wheels", in: Four Australian plays (Ringwood, Vic.: Penguin). Higa, Masanori 1972 "The use of the imperative mood in post-war Japan", in: Lebra (ed.), 49-56. Hijirida, Kyoko - Ho-min Sohn 1986 "Cross-cultural patterns of honorifics and sociolinguistic sensitivity to honorific variables: evidence from English, Japanese, and Korean", Papers in Linguistics 19:365-401. Hill, Deborah 1985 The semantics of some interjectional constructions in Australian English (B.A.Hons. thesis, Australian National University).
Bibliography
471
Hirszowicz, Maria 1980 The bureaucratic leviathan: a study in the sociology of communism (Oxford: Martin Robertson). Hollos, Marida - William Beeman 1978 "The development of directives among Norwegian and Hungarian children: an example of communicative style in culture", Language in Society 7:345-355. Honna, Nobuyoki - Bates Hoffer 1989 An English dictionary of Japanese ways of thinking (Tokyo: Yuhikaku). Hornby, Albert Sydney - E.V. Gatenby - H. Wakefield 1969 The advanced learner's dictionary of current English (2nd edition) (Oxford: Oxford University Press). Horne, Donald 1964 The lucky country: Australia in the sixties (Ringwood, Vic.: Penguin). House, Juliane - Gabriele Kasper 1981 "Politeness markers in English and German", in: Coulmas (ed.), 157-185. Hubler, Axel Understatements and hedges in English (Amsterdam: John 1983 Benjamins). Hudson, Joyce 1985 "Selected speech act verbs in Walmatjari", in: Huttar - Gregerson (eds.), 1985:63-83. Hughes, Geoffrey 1988 Words in time: a social history of English vocabulary (Oxford: Blackwell). Humboldt, Wilhelm von 1903-18 Gesammelte Schriften 1-7 (A. Leitzmann, ed.) (Berlin: Behrs). Huttar, George - Kenneth Gregerson (eds.) 1985 Pragmatics in non-Western perspective (Dallas, Texas: Summer Institute of Linguistics). Hymes, Dell H. 1961 "On typology of cognitive styles in language (with examples from Chinookan)", Anthropological Linguistics 3:22-54. "The ethnography of speaking", in: Gladwin - Sturtevant (eds.), 1962 13-53 [Reprinted in Fishman (ed.), 99-138]. "Anthropology and sociology: an overview", in: Sebeok (ed.), 1974a 1445-1475. Foundations in sociolinguistics: an ethnographic approach 1974b (Philadelphia: University of Pennsylvania Press).
472
Bibliography
Jacobs, Roderick - Peter Rosenbaum (eds.) 1970 Readings in English transformational grammar (Waltham, Mass.: Ginn). Jespersen, Otto 1965 A modern English grammar on historical principles 1-7 (London: [1909-49] Allen & Unwin). Johnson, Christina 1978 In praise of dogs in Australia (Adelaide: Rigby). Johnson, Ken 1972 "The vocabulary of race", in: Kochman (ed.), 140-151. Kaczynski, Mieczyslaw 1964 Gramatyka j~zyka wloskiego [Grammar of the Italian language] (Warszawa: Wiedza Powszechna). Kageyama, Taro - Ikuhiro Tamori 1976 "Japanese whimperatives", Papers in Japanese Linguistics 4: 13-53. Kalish, Donald - Richard Montague Logic: techniques of formal reasoning (New York: Harcourt, 1964 Brace). Karcevski, Serge 1969 "Introduction a l't~tude de l'interjection", in: Godel (ed.), 196-212. [1941] Katriel, Tamar 1986 Talking straight: dugri speech in Israeli Sabra culture (Cambridge: Cambridge University Press). Keen, Ian D. 1978 One ceremony, one song: an economy of religious knowledge among the Yolngu of north-east Arnhem Land (Ph.D. thesis, Australian National University). Kemme, Hans-Martin 1979 Ja, denn, doch usw, Die Modalpartikeln im Deutschen (Miinchen: Goethe-Institut). Kendon, Adam - Richard Harris - Mary Ritchie Key (eds.) 1975 Organization of behavior in face-to-face interaction (The Hague: Mouton). Kiefer, Ferenc - Nicholas Ruwet (eds.) 1973 Generative grammar in Europe (Dordrecht: Reidel). Kinchin-Smith, F. - T.W. Melluish 1966 Teach yourself Greek (London: The English Universities Press). King, Jonathan 1978 Waltzing materialism (Sydney: Harper & Row).
Bibliography
473
Kochman, Thomas 1972 "Toward an ethnography of black American speech behavior", in: Kochman (ed.), 241-264. 1981 Black and white styles in conflict (Chicago: University of Chicago Press). Kochman, Thomas (ed.) 1972 Rappin' and stylin' out: communication in urban black America (Urbana: University of Illinois Press). Kolankiewicz, George - Paul Lewis 1988 Poland: politics, economics, and society (London: Pinter). Konig, Ekkehard "Temporal and non-temporal uses of 'noch' and 'schon' in 1977 German", Linguistics and Philosophy 1:173-198. Kucera, Henry - W. Nelson Francis 1968 Computational analysis of present-day American English (Providence, R.I.: Brown University Press). Kurokawa, Shozo 1972 "Japanese terms of address: some usages of the first and second person pronouns", Papers in Japanese Linguistics 1:228-238. Labov, William 1972 "Rules for ritual insults", in: Sudnow (ed.), 120-169 [Reprinted in Kochman (ed.), 265-314, and in Labov, Language in the inner city, 297-353 (Philadelphia: University of Pennsylvania Press)]. Lakoff, George 1986 Women, fire, and dangerous things: what categories reveal about the mind (Chicago: University of Chicago Press). Lakoff, Robin 1972 "Language in context", Language 48:907-927. Lanham, Betty B. 1986 "Ethics and moral precepts taught in schools of Japan and the United States", in: Lebra - Lebra (eds.), 280-296. Larsen-Freeman, Diane (ed.) 1980 Discourse analysis in second language research (Rowley, Mass.: Newbury House). LDOCE 1978 Longman dictionary of contemporary English (Paul Procter, ed.) (London: Longman). LDOTEL 1984 Longman Dictionary of the English Language (Heather Gay, Brian O'Kill, Katherine Seed, Janet Whitcut, eds.) (London: Longman).
474
Bibliography
Lebra, Takie S. 1974 "Reciprocity and the asymmetric principle: an analytic reappraisal of the Japanese concept of On", in: Lebra - Lebra (eds.), 192207. 1976 Japanese patterns of behavior (Honolulu: University Press of Hawaii). Lebra, Takie S. - William P. Lebra (eds.) 1974 Japanese culture and behavior: selected readings (Honolulu: University Press of Hawaii). 1986 Japanese culture and behavior: selected readings (revised edition) (Honolulu: University Press of Hawaii). Lebra, William P. (ed.) 1972 Transcultural research in mental health (Honolulu: University Press of Hawaii). Lee, David 1987 "The semantics of just", Journal of Pragmatics 11 :377-398. Leech, Geoffrey 1983 Principles of pragmatics (London: Longman). Leibniz, Gottfried Wilhelm 1903 Opuscules et fragments inedits de Leibniz (Louis Couturat, ed.) (Paris) [Reprinted 1961, Hildesheim: Georg DIms]. 1949 New essays concerning human understanding (La Salle, Ill.: Open [1704] Court). Lepschy, Anna - Giulio Lepschy 1984 La lingua italiana [The Italian language] (Milano: Bompiani). Levenston, Edward A. 1970 English for Israelis (Tel Aviv: Israeli Universities Press). Levinson, Stephen 1983 Pragmatics (Cambridge: Cambridge University Press). Levy, Robert 1973 Tahitians: mind and experience in the Society Islands (Chicago: University of Chicago Press). Lewis, Clive Staples 1946 The Screwtape letters (London: Geoffrey Bles). Locke, John 1959 [1690] An essay concerning human understanding 1-2. (A.C. Fraser, ed.) (New York: Dover). Luo, Yongxian 1988 "An account of Chinese nominal tautologies", (unpublished paper, Australian National University). Lutz, Catherine 1982 "The domain of emotion words on Ifaluk", American Ethnologist 9:113-128.
Bibliography 1985
475
"Ethnopsychology compared to what? Explaining behavior and consciousness among the Ifaluk", in: White - Kirkpatrick (eds.), 35-79. 1986 "Emotion, thought, and estrangement: emotion as a cultural category", Cultural Anthropology 1: 287-309. 1988 Unnatural emotions: everyday sentiments on a Micronesian atoll and their challenge to Western theory (Chicago: University of Chicago Press). Malamud, Bernard 1958 The magic barrel (New York: Farrar, Straus). 1971 The tenants (New York: Farrar, Straus). Manzoni, Alessandro 1914 The betrothed (Bohn's Libraries, trans.) (London: Bell). 1968 The betrothed (Archibald Colquhoun, trans.) (London: Dent). 1972 I promessi sposi [The betrothed] (Verona: Mondadori). Marsella, Anthony - George DeVos - Francis Hsu (eds.) 1985 Culture and self: Asian and Western perspectives (New York: Tavistock). Matisoff, James 1979 Blessings, curses, hopes, and fears: psycho-ostensive expressions in Yiddish (Philadelphia: Institute for the Study of Human Issues). Matsumoto, Yoshiko "Reexamination of the universality of face: politeness phenomena 1988 in Japanese", Journal of Pragmatics 12:403-426. McGregor, Craig 1980 The Australian people (Sydney: Hodder & Stoughton). Mel'cuk, Igor in press Cours de morphologie generale. (Montreal: Presses Universitaires). Mezzrow, Milton - Bernard Wolfe 1969 Really the blues (Garden City, N.Y.: Doubleday). Mickiewicz, Adam 1955 "Dziady, cz~sc III" [Grandfathers, part 3] (Utwory zebrane [1832] [Collected works]) (Warszawa: Czytelnik). Miller, Roy 1967 The Japanese language (Chicago: University of Chicago Press). Minami, Hiroshi 1953 Nihonjin no shinri [The psychology of the Japanese] (Tokyo: Iwanami). Mitchell-Kernan, Claudia 1971 Language behavior in a black urban community (Berkeley: University of California Press) (Monographs of the Language Behavior Research Laboratory, 2).
476
Bibliography
1972
"Signifying and marking: two Afro-American speech acts", in: Gumperz - Hymes (eds.), 161-179. Mittwoch, Anita 1977 "How to refer to one's own words: speech act modifying adverbials and the performative analysis", Journal of Linguistics 13: 177-189. Mizutani, Osamu - Nobuko Mizutani 1987 How to be polite in Japanese (Tokyo: Japan Times). Moiseev, Aleksandr 1978 "Casticy uze i jelce v sovremennom russkom jazyke" [The particles uze and jesce in contemporary Russian], Slavia Orientalis 27:357-360. Monahan, Barbara 1983 A dictionary of Russian gesture (Ann Arbor, Mich.: Hermitage). Montgomery, Lucy 1980 Anne of Ingleside (London: Angus & Robertson). Moravcsik, Edith 1978 "Reduplicative constructions", in: Greenberg (ed.), 3:297-334. Morgan, Jerry 1978 "Two types of convention in indirect speech acts", in: Cole (ed.), 261-280. Morimoto, Junko 1985 "Some Japanese speech act verbs", (unpublished paper, Australian National University). Morris, Charles W. 1938 "Foundations of the theory of signs", in: Neurath - Carnap Morris (eds.), 77-137. Morris, Desmond - Peter Collett - Peter Marsh - Marie O'Shaughnessy 1979 Gestures: their origins and distribution (London: Jonathan Cape). Myerhoff, Barbara 1978 Number our days (New York: Dutton). Nabokov, Vladimir 1983 Pnin (Ann Arbor: Ardis). Nakane, Chie 1970 Japanese society (London: Weidenfeld & Nicolson). 1972 Human relations in Japan: summary translation of "Tateshakai no ningen kankei" [Personal relations in a vertical society] (Tokyo: Ministry of Foreign Affairs). Nesbitt, Edith 1980 The enchanted castle (Harmondsworth: Penguin). Neurath, Otto - Rudolf Carnap - C.W. Morris (eds.) 1938 International encyclopaedia of unified science (Chicago: Chicago University Press).
Bibliography
477
Nevile, Ann 1981
A comparison of selected speech acts in Japanese and English (B.A.Hons. thesis, Australian National University). Newmeyer, Frederick 1980 Linguistic theory in America: the first quarter-century of transformational generative grammar (New York: Academic Press). Nikolaeva, Tat'jana M. 1985 Funkcji castic v vyskazyvanii: na materiale slavjanskix jazykov [Functions of particles in the utterance: on materials from Slavic languages] (Moskva: Nauka). Nuchelmans, Gabriel 1973 Theories of the proposition: ancient and medieval conceptions of the bearers of truth and reality (Amsterdam: North-Holland). Ochs Keenan, Elinor "The universality of conversational postulates", Language in 1976 Society 5:67-80. OED 1933 The Oxford English dictionary 1-12. (Oxford: Clarendon Press). OEDS 1972-86 A supplement to the Oxford English dictionary 1-4. (R.W. Burchfield, ed.) (Oxford: Clarendon Press). O'Grady, John 1965 Aussie English: an explanation of the Australian idiom (Sydney: Ure Smith). Okudzava, Bulat 1976 Arbat, moj Arbat: stixi i pesni [Arbat, my Arbat: poems and songs] (Moskva: Sovetskij Pisatel'). O'Neill, Eugene 1958 "Welded", in: Three plays by Eugene O'Neill (London: Jonathan Cape). 1960 Anna Christie (Harmondsworth: Penguin). Dring, Elliott 1981 Israeli humor: the content and structure of the 'chizbat' of the Palmah (Albany: SUNY Press). Paduceva, Elena 1985 Vyskazyvanie i ego sootnesennost' s dejstvitel' nost'ju [The utterance and its relation to reality] (Moskva: N auka). Palakomkul, Angkab "A socio-linguistic study of pronominal usage in spoken Bangkok 1975 Thai", International Journal of the Sociology of Language 5: 11-41. Pascal, Fania 1981 "Wittgenstein: a personal memoir", in: Rhees (ed.), 26-62.
478
Bibliography
Pasicki, Adam 1976 "lui, jeszcze, and their English equivalents", Papers and Studies in Contrastive Linguistics (Poznan;) 5: 103-110. Paterson, Katherine 1978 The great Gillie Hopkins (Harmondsworth: Penguin). Pomerantz, Anita 1978 "Compliment responses: notes on the co-operation of multiple constraints", in: Schenkein (ed.), 79-112. Pride, Janet B. (ed.) 1985 Cross-cultural encounters: communication and mis-communication (Melbourne: River Seine). Psathas, George 1979 Everyday language: studies in ethnomethodology (New York: Irvington). Quirk, Randolph - Sidney Greenbaum - Geoffrey Leech - Jan Svartvik 1972 A grammar of contemporary English (London: Longman). Raphael, Frederic 1977 The glittering prizes (Harmondsworth: Penguin). Rath, Rainer "Doch - eine Studie zur Syntax und zur kommunikativen 1975 Funktion einer Partikel", Deutsche Sprache 1975: 222-242. Rayner, Clare 1978 The body book (Ultimo, NSW: Whizzard Publications). Reisman, Karl 1974 "Noise and order", in: Gage (ed.), 50-61. Renwick, George W. 1980 InterAct: guidelines for Australians and North Americans (Yarmouth, Maine: Intercultural Press). Rhees, Rush (ed.) 1981 Ludwig Wittgenstein: personal recollections (Oxford: Blackwell). Rosaldo, Michelle 1980 Knowledge and passion: Ilongot notions of self and social life (Cambridge: Cambridge University Press). 1982 "The things we do with words: Ilongot speech acts and speech act theory in philosophy", Language in Society 11:203-237. Ross, John R. 1970 "On declarative sentences", in: Jacobs - Rosenbaum (eds.), 222272. 1973 "Slifting" [sentence lifting], in: Gross - Halle - Schutzenberger (eds.), 133-169. Rosten, Leo 1968 The joys of Yiddish (New York: McGraw-Hill).
Bibliography
479
Russell, Bertrand 1962 An inquiry into meaning and truth (Harmondsworth: Penguin). Rymkiewicz, Jaroslaw Marek 1984 Rozmowy polskie latem 1983 [Polish conversations, summer 1983] (Paris: Instytut Literacki). Sacks, Harvey - Emmanuel Schegloff - Gail Jefferson 1978 "A simplest systematics for the organization of turn taking for conversation", in: Schenkein (ed.), 7-55. Sadock, Jerrold 1970 "Whimperatives", in: Sadock - Vanek (eds.), 223-238. 1974 Toward a linguistic theory of speech acts (New York: Academic Press). 1981 "Almost", in: Cole (ed.), 257-271. Sadock, Jerrold - Anthony Vanek (eds.) 1970 Studies presented to R.B. Lees by his students (Edmonton: Linguistic Research Inc.). Sansom, Basil 1980 The camp at Wallaby Cross: Aboriginal fringe dwellers in Darwin (Canberra: Australian Institute of Aboriginal Studies). Sapir, Edward 1921 Language: an introduction to the study of speech (New York: Harcourt, Brace). 1949 "Gradation: a study in semantics", in: Selected writings of Edward Sapir in language, culture, and personality, 122-149 (David Mandelbaum, ed.) (Berkeley: University of California Press). Schachter, Josef 1973 Prolegomena to a critical grammar (Dordrecht: Reidel) [First [1935] published 1935, Vienna: Julius Springer]. Schenkein, Jim (ed.) 1978 Studies in the organization of conversational interaction (New York: Academic Press). Schieffelin, Bambi B. - Eleanor Dchs (eds.) 1986 Language socialization across cultures (Cambridge: Cambridge University Press) (Studies in the social and cultural foundations of language, 3). Schiffrin, Deborah 1984 "Jewish argument as sociability", Language in Society 13:311335. Schreiber, Paul 1972 "Style disjuncts and the performative analysis", Linguistic Inquiry 3:321-347.
480
Bibliography
Schrett, Jozef 1984 "Dolina mi~dzy gorami" [A valley between mountains], Kultura 436:5-18. Schuchardt, Hugo "On sound laws: against the Neogrammarians", in Vennemann 1972 Wilbur (eds.), 39-72. [First published 1885.] Searle, John Speech acts: an essay in the philosophy of language (Cambridge: 1969 Cambridge University Press). "Indirect speech acts", in: Cole - Morgan (eds.), 59-82. 1975 Expression and meaning: studies in the theory of speech acts 1979 (Cambridge: Cambridge University Press). Sebeok, Thomas A. (ed.) 1974 Current trends in linguistics XII: Linguistics and adjacent arts and sciences (The Hague: Mouton). Sekiguchi, Isugio 1977 "Was heisst doch?", in Weydt (ed.), 3-9. Sherzer, Joel "Namakke, sunmakke, kormakke: three types of Cuna speech 1974 event", in: Bauman - Sherzer (eds.), 263-282. Shetter, William Z. 1966 "The meaning of German noch", Language 42:42-66. Shopen, Timothy 1974 "Some contributions from grammar to the theory of style", College English 35:775-798. Shweder, Richard - Robert LeVine (eds.) 1984 Culture theory: essays on mind, self, and emotion (Cambridge: Cambridge University Press). Siatkowski, Janusz 1977 Studia j~zykoznawcze, Streszczenia prac doktorskich II [Linguistic studies: abstracts of doctoral dissertations II] (Wroctaw: Ossolineum) (Prace j~zykoznawcze PAN 88). SJP 1958-68 Slownik j~zyka polskiego (Witold Doroszewski, ed.) (Warszawa: Panstwowe Wydawnictwo Naukowe). Skinner, Quentin 1970 "Conventions and the understanding of speech acts", The Philosophical Quarterly 20: 118-138. Skorupka, Stanislaw 1974 Slownik frazeologiczny j~zyka polskiego [Phraseological dictionary of the Polish language] 1-2. (2nd edition) (Warszawa: Wiedza Powszechna).
Bibliography
481
Smith, Hedrick 1976 The Russians (London: Sphere Books). Smith, Robert 1983 Japanese society: tradition, self, and the social order (Cambridge: Cambridge University Press). Sohn, Ho-min 1983 "Intercultural communication in cognitive values: Americans and Koreans", Language and Linguistics (Seoul) 9:93-136. Solomon, Robert 1984 "Getting angry: The Jamesian theory of emotion in anthropology", in: Shweder - LeVine (eds.), 238-254. Sperber, Dan - Deirdre Wilson 1981 "Irony and the use-mention distinction", in: Cole (ed.), 295-318. Spitzer, Leo 1928 Stilstudien 1-2. (MUnchen: Hueber). Spradley, James P. - Michael A. Rynkiewich 1975 The Nacirema: readings on American culture (Boston: Little, Brown). SRJa 1957-61 Slovar' russkogo jazyka [Dictionary of the Russian language] 1-4. (Moskva: Gosudarstvennoe izdatel' stvo inostrannyx i nacional 'nyx slovarej). SSRLJa 1950-65 Slovar' sovremennogo russkogo literaturnogo jazyka [Dictionary of contemporary literary Russian] 1-17. (Moskva: Izdatel' stvo Akademii Nauk SSSR). Stevenson, Burton 1946 Stevenson's book of quotations (London: Cassell). Stevenson's book of proverbs, maxims and familiar phrases 1949 (London: Routledge & Kegan Paul). Sudnow, David (ed.) 1972 Studies in social interaction (New York: Free Press). Sumerkin, Alexander (ed.) 1982 Russica 1981: literaturnyj sbornik [Russica 1981: a literary collection] (New York: Russica). Suslov, 11 'ja 1982 "Inga i Verocka" [Inga and Verochka], in: Sumerkin (ed.), 183190. Suzuki, Takao 1986 "Language and behavior in Japan: the conceptualization of personal relations", in: Lebra - Lebra (eds.), 142-157.
482
Bibliography
Szelburg-Zarembina, Ewa 1970 Idzie niebo ciemn{l noc{l... [The sky is walking on a dark night. .. ] (Warszawa: Nasza Ksi~garnia). Szwed, John F. (ed.) 1970 Black America (New York: Basic Books). Tannen, Deborah 1981a "Indirectness in discourse: ethnicity as conversational style", Discourse Processes 4:221-238. 1981b "New York Jewish conversational style", International Journal of the Sociology of Language 30:133-149. 1984 Conversational style: analyzing talk among friends (Norwood, N.J.: Ablex). That's not what I meant! How conversational style makes or 1986 breaks relationships (New York: Ballantine). Tannen, Deborah - Piale Oztek '" Health to our mouths': formulaic expressions in Turkish and 1977 Greek", Berkeley Linguistics Society, Proceedings 3: 516-534 [Reprinted in Coulmas (ed.), 37-54]. Taylor, Brian A. 1976 "Towards a sociolinguistic analysis of 'swearing' and the language of abuse in Australian English", in: Clyne (ed.), 43-62. Tolstoy, Nikolai 1983
The Tolstoys: twenty-four generations of Russian history 1353-
1983 (London: Harnish Hamilton). Traugott, Elizabeth C. - John Waterhouse 1969 '" Already' and 'yet': a suppletive set of aspect-markers?", Journal of Linguistics 5:287-304.
Triandaphyllidis, Manolis 1975 Petite grammaire du grec moderne (F. Duisit, O. Merlier, trans.) (Thessaloniki). Triandis, Harry C. - Leigh M. Triandis 1960 "Race, social class, religion, and nationality as determinants of social disclosure", Journal of Abnormal and Social Psychology 61: 110-118. Triandis, Harry C. - Vasso Vassiliou 1972 "A comparative analysis of subjective culture", in: Triandis (ed.), 299-335. Triandis, Harry C. (ed.) 1972 The analysis of subjective culture (New York: Wiley). Trilling, Lionel 1972 Sincerity and authenticity (London: Oxford University Press). Ueda, Keiko
Bibliography
1974
483
"Sixteen ways to avoid saying 'no' in Japan", in: Condon - Saito (eds.), 184-192. Universite de Paris-7 1986 Les particules enonciatives en russe contemporain (Paris: Departement de Recherches Linguistiques (DRL), Laboratoire de Linguistique Formelle) (ATP Nouvelles recherches sur Ie langage, Collection ERA 642). Urmson, James Opie 1963 "Parenthetical verbs", in: Caton (ed.), 220-240. Vennemann, Theo - Terence H. Wilbur (eds.) 1972 Schuchardt, the Neogrammarians, and the transformational theory of phonological change (Frankfurt: Athenaum). Verschueren, Jef 1985 What people say they do with words: prolegomena to an empirical conceptual approach to linguistic action (Norwood, N.J.: Ablex). Vossler, Karl 1904 Positivism us und Idealismus in der Sprachwissenschaft (Heidelberg). 1925 Geist und Kultur in der Sprache (Munchen). Wannan, Bill 1963 Tell' em I died game (Melbourne: Lansdowne Press). Ward, Russel 1958 The Australian legend (Melbourne: Oxford University Press). Weber, Max 1968 [1930] The Protestant ethic and the spirit of capitalism (Talcott Parsons, trans.) (London: Allen & Unwin). Webster 1973 The international Webster new encyclopaedic dictionary (New York: Tabor House). Webster's new world dictionary (Sydney: Fontana/Collins). 1977 Weydt, Harald 1969 Abtonungspartikel, Die deutschen Modalworter und ihre franzosischen Entsprechungen (Bad Homburg: Gehlen). Weydt, H~rald (ed.) 1977 Aspekte der Modalpartikeln: Studien zur deutschen Abtonung (Ttibingen: Niemeyer). Weydt, Harald - Theo Harden - Elke Hentschel - Dietmar RosIer (eds.) 1983 Kleine deutsche Partikellehre (Stuttgart: Klett). White, Geoffrey M. - John Kirkpatrick (eds.) 1985 Person, self, and experience: exploring Pacific ethnopsychologies (Berkeley: University of California Press).
484
Bibliography
Whorf, Benjamin Lee 1956 Language, thought, and reality (John Carroll, ed.) (New York: Wiley). Wierzbicka, Anna 1971 "The deep or semantic structure of the comparative", Linguistische Berichte 16:39-45. 1972 Semantic primitives (Frankfurt: Athenaum) (Linguistische Forschungen, 22). "The semantic structure of words for emotions", in: Jakobson 1973 van Schooneveld - Worth (eds), Slavic poetics, 499-505 (The Hague: Mouton). 1974 "The semantics of direct and indirect discourse", Papers in Linguistics 7:267-307. 1976 "Particles and linguistic relativity", International Review of Slavic Linguistics 1:327-367. 1977 "The ignorative: the semantics of speech acts", International Review of Slavic Linguistics 2:251-312. Lingua mentalis: The semantics of natural language (Sydney/ 1980 New York: Academic Press). 1984 "Diminutives and depreciatives: semantic representation for derivational categories", Quaderni di Semantica 5: 123-130. 1985a "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: 1985b Polish vs. English", Journal of Pragmatics 9:145-178. Lexicography and conceptual analysis (Ann Arbor: Karoma). 1985c 1986a "A semantic metalanguage for the description and comparison of illocutionary meanings", Journal of Pragmatics 10:67-107. "Does language reflect culture? Evidence from Australian 1986b English", Language in Society 15:349-373. "Human emotions: universal or culture-specific?", American 1986c Anthropologist 88:584-594. English speech act verbs: A semantic dictionary (Sydney: 1987 Academic Press). The semantics of grammar (Amsterdam: John Benjamins). 1988 "Semantic primitives and lexical universals", Quaderni di 1989a Semantica 10.1:103-121 (Round table on semantic primitives, 1). "Semantic primitives: the expanding set", Quaderni di Semantica 1989b 10.2 (Round table on semantic primitives, 2). "Antitotalitarian language in Poland: some mechanisms of linguis1990 tic self-defence", Language in Society 19: 1-59.
Bibliography
485
in press a "Lexical universals and universals of grammar", in Michel Kefer - lohan van der Auwera (eds.) Meaning and grammar. (Berlin: de Gruyter). in press b "Semantic complexity", to appear in Theoretical Linguistics 17, 1990 [special issue on "Complexity"]. to appear Semantics, culture and cognition: universal human concepts in culture-specific configurations. Wierzbicka, Anna (ed.) 1972 Semantyka i slownik [Semantics and lexicon] (Wroclaw: Ossolineum). Wilkes, Gerald A. 1978 A dictionary of Australian colloquialisms (Sydney: Fontanal Collins). Wilkins, David P. "Nominal reduplication in Mparntwe Arrernte", Language in 1984 Central Australia 1: 16-22. 1986 "Particle/clitics for criticism and complaint in Mparntwe Arrernte (Aranda)", Journal of Pragmatics 10:575-596. Williams, Tennessee 1961 The night of the iguana (London: Secker & Warburg). Williamson, David Don's party (Collected plays, 1) (Sydney: Currency Press). 1973 1974 The coming of Stork, Jugglers three, What if you died tomorrow: three plays (Sydney: Currency Press). Wittgenstein, Ludwig 1953 Philosophical investigations (G.E.M. Anscombe, trans.) (Oxford: Blackwell). Zapolska, Gabriela 1978 Moralnose pani Dulskiej [Mrs. Dulska's morality] (Wroc}aw: Ossolineum). Zasorina, L.N. (ed.) 1977 Castotnyj slovar' russkogo jazyka [Frequency dictionary of the Russian language] (Moskva: Russkij Jazyk). Zipf, George K. 1949 Human behavior and the principle of least effort: an introduction to human ecology (Facsimile edition, 1965) (New York: Hafner).
Subject and name index
Abby, Dear 143-144 Abelard, Peter 198, 253 Aboriginal (Australian) culture 63, 102-103, 159-60 Abrahams, Roger 63, 69, 103, 150, 183-184,277 abuse, verbal 35, 40, 232-235, 249-251 address forms 48,51,56-59,86, 106107, 111-112, 122 advice 31-32 agreement 140 Ali, Mohammed 68 Ameka, Felix 69, 131,342,344,459 American culture, see English, American AND (Australian National Dictionary) 165-168,170-178,181-183 Anglo-American culture, see English, American Anglo-Saxon culture 26, 47-60, 64-65, 69,72-77, 103-104,276-277,279, 446-448, 454 animals, calls to 292-293 apologies 86, 126, 152, 156-157 Arensberg, Conrad 91 argument 68-69 Aristotle 161 Arnold, Matthew 116-117 Ascham, Roger 104 assertiveness, see self-assertion Atlas, Jay 396 augmentatives 1 Austin, John L. 197, 240 Australian culture, see English, Australian; Aboriginal culture autonomy 37,49,52,76, 80-82, 86,91, 95, 99, 276
Bach, Kent 198,200-201,209,253, 449 Bacon, Roger 198, 253 Baker, Sidney 3, 177 Bakhtin, see Baxtin Bally, Charles 393,445,451 Bankowski, Andrzej 367 baptism 151-152 Bamlund, Dean 12, 53, 82, 86, 93, 103, 106, 110, 113-114, 129 Bartlett, John 401, 415 Barzini, Luigi 280-282 Bates, Elizabeth 77 Baxtin, Mixail 149, 197,457 Beeman, William 62, 77 Benedict, Ruth 57,71,129,176 Black (American) culture, see English, Black blessings 4, 122, 150 Blum-Kulka, Shoshana 77, 89-92, 96-98, 138, 141 Blyton, Enid 415 boasting 84-85, 119, 198 Bogardus, Emory 108 Boguslavskij, Igor' 342 Bogus.fawski, Andrzej 7-8, 164,217, 344, 397, 412 Bohn's Library 438 Bolinger, Dwight 248, 268-269, 271 Bowles, Colin 3 Braun, Friederike 457 Brown, Penelope 30, 67, 108-109, 279,444-445,454 Brown, Roger 47, 58 Brzechwa, Jan 306-307 BUhler, Karl 288 Butwin, Frances 123
488
Subject and name index
Buzo, Alexander 27,29,35,39,4244,46, 171 calling 292-293, 300-301 Cattell, Ray 226 change, language 169-170, 173, 177, 189-190 Chekhov, Anton 300-301 Chinese 144, 423-428, 435, 454 Chomsky, Avram Noam 14, 164, 283 circularity 6, 161 Clancy, Patricia 113-114 Clark, Herbert 25 closeness, see intimacy codes 67 cognitive style 282-283 Cole, Peter 16, 392 Colquhoun, Archibald 438 commands 30, 39, 61, 150-151, 199, 203, 261, 295; see also orders; imperatives communicative style 454-455 complaints 60-61, 64-65, 181 complementarism 16-17 complements 221, 238-239 compliments, responses to 136-147 Comrie, Bernard 77,202-205 concessive tautologies 423-426 Confucius 427 conjunctions 240-243, 381 conversation 18, 21, 59-64, 80-82, 116,118,120,131-148,149-151, 202,219-235,253-254,449,451 Conway, Ronald 183 Cooke, Joseph 12 cordiality 27, 50-56, 70, 86 Coulmas, Florian 86, 156, 163 courtesy, see politeness Cowie, Anthony P. 398,401,440 Crawford, Raymond M. 177 criticism 213-214, 230
cultural differences 2-4, 25-65; see also values cultural kinds 17-18 cultural style 276-284; see also values; identity Cuna 150 curses 4, 122-123,213,249-251, 338; see also swearing Dakota 302 Dal', Vladimir 305 Danet, Brenda 77,89-90, 138, 141 Danish 307,313-314,338 Darwin, Charles 312-313, 315-316 Davies, Norman 58, 112, 192-193 Davison, Alice 207 Dawes, Arthur 166 de Tocqueville, Alexis 90 Deakin, Greg 163,247-248 deference, see politeness; honorifics; address forms definitions, see metalanguage DeVos, George 154-155 diminutives 1, 51-53, 55-56, 255, 280 Dineen, Anne 307 directives, see orders; commands; requests; imperatives directness 44, 60, 63-64, 70-71, 88104, 186-187,454 discourse 258-260 disgust 302-317, 338 dissimulation 63-64, 100-104 distance 37, 47-48, 108-111; see also calling; intimacy Dixon, Robert M.W. 273 Djilas, Milovan 193 Doherty, Monika 367 Doi, Takeo 74, 76, 113, 155 Doroszewski, Witold 49 Dostoevsky, Fedor 280, 312 Dovlatov, Sergej 327
Subject and name index
downgrade 140-141 Doyle, Arthur Conan 419 Dragunskij, Viktor 327, 336 Drazdauskiene, Maria-Liudvika 43, 45, 116 Dumas, Alexandre 285 Eades, Diana 63,69, 102-103 emotions 53-57,87-88,119,121129,244-245, 266-269, 271,275, 279-280, 302-326, 331, 339 emphasis 258; see also intensification; intonation English, see specific subject headings; Black (American) 2-3, 21, 26,63,68,71,73,78-79,82-87, 97,119,123-126,150,165,183185,277,454; American 2, 26, 42, 48, 67, 71, 72-82, 85-87, 9899,103-108,110,113-121,124126, 128, 136, 144, 248, 339, 371, 454-455; Australian 2-3, 2022, 26, 42, 48, 55-56, 91, 111112, 116, 136, 165-183, 185, 187188,249,277,303-304,454,459460; Scottish 180 equivalence, semantic 12-14, 152, 286 Ervin-Tripp, Susan 30, 58, 60 ethnocentrism 21, 25-26, 67-68, 71, 131, 150,454 ethnography, of speaking 67, 85, 122, 126, 149-151 exclamations 31, 45-46, 222, 232-235, 249-251; see also interjections explication, see metalanguage face 67, 454 Facey, Albert 357 fatalism 431-433 Field, Deborah 396, 428-430
489
Fillmore, Charles 220 Finnish 293 first person, in definitions 162-165 fixed expressions 56, 245-247 Fochi, Franco 271 folk labels 150-160 Folsom, Franklin 334-335 formality 48,57, 104, 111-113,358; see also politeness; distance Forster, E.M. 395 Francis, W. Nelson 339 Fraser, Bruce 26, 152 French 1, 13-14,71, 103,270,392394,445 future 236, 431-433 fuzziness, see indeterminacy Garton Ash, Timothy 189-190, 192 Gazdar, Gerald 451 Geertz, Clifford 63,87,92, 100-101, 128-129 gender, grammatical 1; see also sex differences generative semantics 15-16, 18, 197198, 253 genres, speech 149-165, 183-196 Gerber, Eleanor 339 German 13-14,60-61, 103, 286, 302, 304,310,313-315,338,342,367, 370, 375-376, 392-393, 445 gesture 446 Gherson, Rimona 77, 89-90, 138, 141 Gibbs, Raymond 25 Gilman, Albert 47, 58 Givan, Talmy 203-205, 397 Goddard, Cliff 7-8,69,243,293, 344, 346 Goffman, Erving 150, 289-290 Goldstein, Bernice 120-121, 127-128 Goljavkin, Viktor 311 Goody,E~her 103,276
490
Subject and name index
Gordon, David 32, 59, 62, 213 grammaticality 209-210, 232, 253, 449-452 Grandgent, Charles 255 Greek 95-99, 128,240,315-316, 457-458 Green, Georgia 32-33, 211 greetings 3, 132, 246 Grice, H. Paul 21,23,59,62,67, 366, 391-392, 396-397, 398-400, 403, 449-452, 454 Grochowski, Maciej 344, 367, 377 Grodzienska, Wanda 290, 320 Grossman, Vasily 394 Gumperz, John 62-63, 150 Haiman, John 288, 302, 304 Halliday, Michael 240, 242 Hamlet 118 Harasawa, Itsuo 396 Hardy, Frank 166 Harkins, Jean 69,267,277,307,344 harmony 92, 104, 109-110, 113-115, 118-119 Harnish, Robert 198, 200-201, 209, 253, 449 Harris, Max 169 Harris, Stephen 103, 158 Hasan, Ruqaiya 240, 242 Hawke, Robert 2 Hayakawa, Haruko 270 Hebrew 21, 71, 89-93, 119, 165, 185-188,454 hedges 43-44 Henschke, Sharon 135 Hibberd, Jack 27-29,35,213 Higa, Masanori 30 Hijirida, Kyoko 48,69, 105-107, 111 Hill, Deborah 103, 459 Hirszowicz, Maria 192 Hoffer, Bates 68, 93, 127, 147 Hollos, Marida 62, 77
Honna, Nobuyoki 68, 93, 127, 147 honorifics 12; see also address forms; politeness Hornby, Albert 161 Horne, Donald 167, 181-182 House, Juliane 60 HUbler, Axel 276 Hudson, Joyce 159-160 Hughes, Geoffrey 103 Humboldt, Wilhelm von 282 Hungarian 77 Hymes, Dell 67, 149-150, 165,258, 283, 341 identity, cultural 129-130 illocutionary force 17, 22, 60-61, 150, 158, 161-165, 197-254,288289, 342 illocutionary grammar 258-260, 276280 imperatives 27, 30-32, 36-37, 51-53, 60,76-78, 89, 122, 204-205, 227229,231,243,258-260,263,451 implicature 207, 228, 255, 261, 391, 397-403; see also logic indefinables, see metalanguage indeterminacy 197-210,212,224, 253-254, 351, 389 indirectness, see directness infinitive 36-37 informality, see formality; intimacy insults 151, 183-185, 187,232-233; see also abuse intensification 168-170, 256-258, 264, 268-270, 278 intentions 164, 198 intercultural understanding 9, 21, 24, 64-65, 69-70, 129-130, 144, 147148, 210, 455 interjections 23, 45-46, 243-245, 285-339; cognitive 326-336; emotive 302-326; to animals 292293; volitive 292-301
Subject and name index
interrogative, see questions intimacy 47-48, 70, 86, 104-111; see also distance intonation 247-248 invitations 29, 132, 174, 211-212, 215-216, 228 Israeli Hebrew, see Hebrew Italian 22, 51, 71, 77, 103,255-284, 435-436,438-439,446,454 Japanese 17, 21,48, 68, 70-71, 145147,270,454; culture 2,53-54, 57,72-83,85-88,93-95, 103, 105-107, 112-114, 120-121, 126-128,129-130,144,176; imperatives 30-31, 88, 258-259; pronouns 12-14; speech acts 152158; tautologies 396, 428-431 Javanese 48, 57, 63-64, 92, 100-104, 128-129 Jefferson, Gail 254 Jespersen, Otto 277, 279,401 Jewish culture 2-4, 68-69, 71, 119, 122-123, 144, 252,446-448; see also Hebrew; Yiddish Johnson, Christina 407 Johnson, Ken 86 jokes 188-192 Kaczynski, Mieczysiaw 271 Kageyama, Taro 78 Kalish, Donald 391 Karcevski, Serge 285, 293, 302, 304-305,310 Kasper, Gabriele 60 Katriel, Tamar 67,69,91, 185-187 Keen, Ian 103 Kemme, Hans-Martin 342 Keneally, Thomas 181 Kerry, Tristram 442-443 Kinchin-Smith, F. 316 King, Jonathan 183
491
kinship 158-160 Kipling, Rudyard 395, 415 Kirby, Michael 419 Kochman, Thomas 3, 68-70, 73, 7879,82-85,119,124-125,277 Kolankiewicz, George 192 Konig, Erich 367 Korean 48,105,107,112,396,454 Kucera, Henry 339 Kurokawa, Shozo 13 Labov, William 183-184 Lakoff, George 18, 30, 32, 59, 62, 213 laments 338 Lanham, Betty 80, 86, 126 Latin 7,255-256,270,458,460 Lawson, Henry 167 Lebra, Takie 57, 70, 72-73, 75, 85, 87,91,103,127,153-155,176 Lee, David 350-353 Leech, Geoffrey 15,17,19,21,68, 132, 198-200, 209-210, 277-279, 282 Leibniz, Gottfried Wilhelm 10, 343 Lepschy, Anna and Giulio 257, 265 Levenston, Edward 92 Levinson 30, 67, 108-109,207-209, 254,279,391,397,399-401,408, 432,444-445,454 Levy, Robert 339 Lewis, Clive Staples 414 Lewis, Paul 192 Lithuanian 43, 45 Locke, John 196,341-342 logic 60, 62-63, 207, 391, 397,399400,454 Longman (English dictionary) 6, 286-287,289,295,303,317,330 Luo 423, 426-428, 436 Lutz, Catherine 9,54-55, 103, 121, 339
492
Subject and name index
Malamud, Bernard 3-4, 446-448 Manifold, John 182 Manzoni, Alessandro 256-257, 259, 261-262, 266,438-439 Matisoff, James 4, 97, 122-123,249251,338,447 Matsumoto, Yoshiko 30, 67, 77-78 McGregor, Craig 170 Mediterranean cultures 47,50,65, 279 Mel'cuk, Igor 270,280 Melluish, T.W. 316 mental verbs 238-240 metalanguage 6-8, 71-72, 130-131, 134, 137, 147-149, 151, 156, 161165, 184, 199-202, 212, 218-219, 253, 283,288,291, 314, 337-338, 343-344,403-404,451,453-455 Mezzrow, Milton 183-184 Mickiewicz, Adam 50, 297-298, 312 Miller, Roy 145-146 Minami, Hiroshi 154 Mitchell-Kernan, Claudia 277 Mittwoch, Anita 208-209 Mizutani, Osamu and Nobuko 68, 69,74-75,81,86,93-95, 126, 145-147 modal verbs 235-238 Moiseev, Aleksandr 367 Monahan, Barbara 47, 53 Montague, Richard 391 Montgomery, Lucy 396 Moravcsik, Edith 270 Morgan, Jerry 16,451 Morimoto, Junko 153-155 Morris, Charles W. 15, 16 Morris, Desmond 446 Myerhoff, Barbara 92 Nabokov, Vladimir 285 Nakane, Chie 93, 112, 153 names, see address forms
natural kinds 17-18 natural semantic metalanguage, see metalanguage Nesbitt, Edith 408 Nevile, Ann 152-153 Newmeyer, Frederick 18 Niehoff, Arthur 91 Nikolaeva, Tat'jana 342 Nuchelmans, Gabriel 240, 458 O'Grady, John 177-178 O'Neail, Emmie 228 O'Neill, Eugene 459 objectivity 49-50 obligation 157-160,175-176,419423 Ochs (Keenan), Elinor 69 OED (Oxford English Dictionary) 177, 180, 360 offers 27-30, 211-212, 215-216, 228 Okud~ava, Bulat 393, 435 Olshtain, Elite 77 opinions 31, 41-44, 219 opposite, contrastive 139 orders 30, 39, 151-152, 158-160, 199-202, 205, 215-216; see also commands; imperatives Oring, Elliott 185-186 other minds 162-165 Oztek, Piale 97 Paduceva, Elena 17, 252, 363 Palakornkul, Angkab 12 particles 23, 240-243, 341-389; approximative 354-366, 384-389; quantitative 345-366, 379-389; temporal 367-378 Pascal, Fania 405 Pasicki, Adam 367, 371 past 434-439 paternalism 154 Paterson, Katherine 459
Subject and name index
Paul of Venice 240 performatives 37, 60, 163, 197-198, 201-202, 208-209, 236, 253-254 Pitjantjatjara 293 Polish 26,71, 103,279,454; advice 31-32; address forms 47-49,51, 107, 112; culture 47-60, 105, 114115, 121-122; diminutives 51-53, 122; exclamations 45-46; imperatives 30-31, 77, 88, 122; interjections 23, 286, 290, 292-308, 310-315,318-334,336-337,315; offers 27-29; opinions 41-44; particles 23, 344, 362, 367, 370389; requests 32-37; speech acts 20-21,27-60, 165; speech genres 188-196; tags 37-41; tautologies 394-395, 402-403, 413, 422-423, 425-426,432,434,436-437,445 politeness 13, 27, 30, 34-36, 40, 4344, 52, 56-64, 86, 90, 100, 205, 213; see also address forms; honorifics polysemy 11, 350 Pomerantz, Anita 21, 131, 133, 136144, 148 practical applications, see intercultural understanding pragmaticism 17-18 pragmatics, non-linguistic 6, 131136 pragmatics, vs. semantics 1-24, 366, 449-455 praise 68-69, 136, 141-142, 146-147, 232-235 Pride, Janet 283 primitives, semantic see metalanguage; universals Pritchard, Katherine 167 promises 151-152, 198 pronouns 12-14 Psathas, George 254
493
questions 5-6, 27-30, 32-35, 45-46, 52-53,61-63,75-8,89, 102, 132, 150-151, 199-201,203-204,209217,220,224-232,240-241,451; see also tags Quirk, Randolph 274 Raphael, Frederic 459 Rath, Rainer 393 Rayner, Clare 286 realism 404-405 reciprocity, see obligation reduplication 22, 255-284 Reisman, Karl 85 relevance 220 Renwick, George 3, 136, 169, 173, 183 repetition, clausal 258-263, 268-270 reprimands 223-224 requests 25, 31-37, 39, 52, 60-62, 76-78, 138, 151-152, 159, 193195,201-203,215-217, 228 respect, see politeness; formality restraint 45, 53, 74-76, 79, 83, 8587, 99, 128 return, of compliment 142-143 Rintell, Ellen 26, 152 Rosaldo, Michelle 152, 339 Ross, John R. 197, 238 Rosten, Leo 285, 294,308-310,317, 371, 373 Russell, Bertrand 14 Russian 7, 17, 30, 71, 279-280, 454; address forms 57-59; culture 2, 105, 110, 114, 129-130; imperatives 77, 88; interjections 23, 285, 293,296,300-302,310-315,322323, 326-330, 334-336, 338-339; particles 342, 367, 393; requests 204; tautologies 392-394, 425426,435,445 Rymkiewicz, Jaroslaw 413
494
Subject and name index
Sabra (Israeli) 185-186 Sacks, Harvey 254 Sadock, Jerrold 23,200-201,214, 229, 356, 363-364, 366 Sakurai, Kaoru 157 Sansom, Basil 63, 103 Sapir, Edward 279,282, 365 sarcasm 38 Schachter, Josef 197 Schegloff, Emmanuel 254 Schenkein, Jim 148, 254 Schiffrin, Deborah 68,69,92,119 Schreiber, Paul 207 Schrett, J6zef 59 Schuchardt, Hugo 19 Schunk, Dale 25 Searle, John 59-62,67, 151,201202, 214, 454 Sekiguchi, Isugio 393 self-assertion 72-88 self-restraint, see restraint semantic metalanguage, see metalanguage semanticism 18 Serbo-Croatian 30, 435 sex differences 58, 136, 145-146, 166, 168, 172, 175, 183 Shakespeare, William 118,313 shame 306-307 Sherzer, Joel 150 Shetter, William 367 Shin, Gi-Hyun 396 Shopen, Timothy 217, 220 silence, calls for 293-296 Simpson, Jane 64, 269, 279 sincerity 70-71, 104, 115-121, 133, 135, 186; see also directness SJP (Polish dictionary) 193, 294, 297-300, 306, 312, 318-321, 323327, 331-333, 376, 380-382, 385, 389 Skinner, Quentin 165 Skorupka, Stanislaw 193
Slavic languages 47, 50-51, 56, 107,114,118,279 Smith, Gerda 375 Smith, Hedrick 53, 86 Smith, Robert 57,74,91, 153 sociology 254 Sohn, Ho-min 48, 69-70, 105107, 111 Solomon, Robert 339 sound symbolism 294,315-317, 337-338 Spanish 30, 51, 402-403, 431-434 speech acts 17, 21-22, 25-65, 149-196, 197-254 Sperber, Dan 449 spitting 310-313 Spitzer, Leo 282 spontaneity 57, 70-71, 80-83, 86 Sprachgeist 282 SRJa (Russian dictionary) 300, 327, 329, 335 SSRLJa (Russian dictionary) 296, 300,305,311-312,322,329, 335 Stein, Gertrude 442 Stevenson, Burton 104,401,411 Stoics 197, 240 Sudnow, David 254 suggestions 27, 94, 198, 211-212, 214-216, 219 superlatives 270-276 Suslov, 11'ja 304 Suzuki, Takao 12,72-73,85 swearing 3, 35, 169, 187 Swedish 307,313-314 syntax 197-198, 209-210, 245246, 253, 403-404 Szelburg-Zarembina, Ewa 307, 320-321 tags 31, 37-42, 224-232 Tamil 444-445 Tamori, Ikuhiro 78
Subject and name index
Tamura, Kyoko 120-121, 127-128 Tannen, Deborah 1,67,69,79, 97-99 tautologies 23, 391-455; nominal 391-431; semantic invariants 439-444, verbal 431-444 taxonomy, folk 150 Taylor, Brian 169 Thai 12, 48, 88 thanks 126, 158 threatening 197-198 tolerance 398, 405-410, 421 Tolstoy, Alexei 296 Tolstoy, Leo 329 Tolstoy, Nikolai 443 Traugott, Elizabeth 367 Triandaphyllidis, Manolis 258 Triandis, Harry 47,96,99, 108 Trilling, Lionel 115-119, 120 truth 103-104, 124, 186, 397, 448 Turkish 12 tum-taking 80-82 Veda, Keiko 145 understatement 44-45, 135, 276-280 ungrammaticality, see grammaticality universals 9-15, 59-62, 67-69, 71-72, 270,286.289,302,392-397,403, 432, 444, 446, 448, 453-455 Vniversite de Paris-7 342 upgrade 138-139 urging 298-299 Vrmson, James O. 238 values, cultural 21,37,47-65,67-130 Vassiliou, Vasso 96, 99 Verschueren, Jef 152 Volksgeist 282 Volosinov (pseudonym of Baxtin) 457 Vossler, Karl 282
495
Wajda, Andrzej 36 Waley, Arthur 427 Walmatjari 21, 158-160,454 Walters, Joel 26, 152 Wannan, Bill 182 Ward, Russel 177, 182 warmth, see cordiality warning 152, 197-198 Waterhouse, John 367 Weber, Max 104 Webster's (English dictionary) 317, 360 Weydt, Harald 342 whimperatives 2, 77-78, 200-207, 209-218 Whorf, Benjamin Lee 282 Wierzbicka, Anna 3, 7-8, 15, 17, 32, 44,49,51,55,57,67,69,86,91, 105-107, 121-122, 128, 133, 136, 153, 155, 157, 160, 162, 165, 172,181,193,195,200-202,217, 221, 243, 268, 272, 277, 287-289, 338-339, 342-344, 432, 459 Wilkes, Gerald 166-167, 178, 183 Wilkins, David 270, 344 Williams, Tennessee 241, 459 Williamson, David 35, 39-40, 50, 178,205,460 Wilson, Deirdre 450 Winter, Werner 460 Wittgenstein, Ludwig 15, 150, 197 Wolfe, Bernard 183-184 Yiddish 4,21,23,71, 122,249-252, 285-286,294,308-310,315-318, 338,370,371,447,454 Yolngu 158 Zapolska, Gabriela 36-37 Zasorina, L.N. 339 Zipf, George 11
Index of words and phrases
This index lists words and phrases explicated or discussed in the book; explications are indicated by bold page numbers. Languages are abbreviated as follows: Da(nish), Ch(inese), Fr(ench), Ge(rman), Gr(eek), He(brew), It(alian), Jp(Japanese), Jv(Javanese), Ko(rean), La(tin), Pi(tjantjatjara), Po(lish), Ru(ssian), Se(rbo-Croat), Sp(anish), Sw(edish), Wa(lmatjari), Yi(ddish). All items are listed in the order of the English alphabet. a joke is a joke 410, 422 a kiss is just a kiss 415, 418, 441 a man is a man 391, 411,412-413, 424,439 a party is a party 403, 408-409, 417, 418-419, 441 a picnic is a picnic 408, 409 a point is a point 415, 418, 441 a promise is a promise 23, 391,419420,421,441 a rose is a rose is a rose 442 a steak is a steak 403,416, 419 aah 325 about 23, 344, 355-357,358,361364, 366 ach (aah, Po) 323-326,324 adagio adagio (very slowly, It) 255, 264-265 aga (aha, Ru) 327-328,329-330 agreement 137; scaled-down 140 ah 286-287, 289, 317-318, 325, 339 aha 285,330 aha (aha, Po) 326-332,327, 337 ahoj (ahoy, Ru) 293 aizuchi (turn-taking, Jp) 81 ale co S to S (what has been has been, Po) 436-437 almost 23, 344, 361-365,366,387, 388 already 1,23, 344,367-368, 369-373
amae (love, Jp) 155 announce 164 apologise 156, 453 application 194, 195-6 appreciation 145 approximately 23, 359, 358-360, 362, 364-365 around 23, 344, 355-364, 358, 366 as many as 345 as much as 345,381 ask 5-6, 32, 60, 159, 160-161, 195, 202, 213 at least 23, 354 at most 354 atashi (I, Jp) 13 au (yoo-hoo, Ru) 300 ax (aah, Ru) 326, 339 az (as much as, Po) 380,381
bark 163 bastard 1, 3, 119, 169, 187 b~c (fall, Po) 290 because 208 believe 42-43, 238-239 bella bella (very beautiful, It) 22, 256-258, 265, 278 blisko (nearly, Po) 386-387, 388-389 bloody 3, 55, 169 blue 1 boast 163-164
498
Index of words and phrases
boku (I, Jp) 13 boys will be boys 391-394, 398-403, 407,408,410,424,440-441,444445, 449-450 bugger 2-3, 169 bullshit 119, 169 business is business 391, 398,404, 450 bygones, see let caffe caffe (real coffee, It) 265, 267268 can you 25-26,61,89,204 can't you 35, 38,207, 227-228,229 cat 17 chat 171, 172 chiack 165-170,168,173,175,177, 187 cii (shh, Po) 293-294, 295 cip-cip-cip (here chickens, Po) 292 closeness 108-111, 109 co b~dzie to b~dzie (what will be will be, Po) 402-403,432 co by/o to by/o (what has been has been, Po) 434-436 co X to X (an X is an X, Po) 395, 413 come in, come in! 258, 260, 262, 263, 269 command 34, 150-151, 199 complain 164, 181,453 compromise 48-49 conclude 241-242 confess 161-162 cooee 301 cordiality 122 could you 38, 53,88-89,96,201, 204, 213 country 49 criticise 213 cto bylo to bylo (what has been has been, Ru) 435
cup 17 curses, Jewish 123 dakara (of course, Jp) 429 damn you 249,250-251 darling 1 dear sir 118 demo (something, Jp) 94-95 did you know 221-222 ding shi ding (ding is ding, Ch) 426427 disagreement 137 do it 96,205 do you know 220-222, 221 dob 177-180,179 dobrze? (okay?, Po) 38 doch (indeed, Ge) 393 doggie 1, 50, 55 don't tell me 222, 223 dopiero (only, Po) 376-378,377, 380 dowc~ Uoke,Po) 188 downgrade 137, 141-142 dozens 165, 183-185, 184 du (you, Ge) 13 dugri (straight, He) 165, 185-188, 187
east is east 395, 404, 413, 414, 426427 either he will or he won't 400-402, 433,447 emotional 53-55, 121 empathy 127 enough is enough 410, 439, 442, 450 enryo (self-restraint, Jp) 74-78, 76, 95, 99 -est 272-276 est' (be, Ru) 393-394 e{ok-e{ok (dissimulation, Jv) 100101, 102 exactly 359
Index of words and phrases
excuse me 89, 108 fe (yuk, Po) 23, 306-307,308,314 fear (I fear) 239 feh (yuk, Yi) 308-309, 310, 314, 316 feu (phew, Gr) 315-316 for God's sake! 143-144 forbid 152 frankly 207-208,210 friendliness 87 fu (yuk, Po) 23, 286, 302-306, 303, 313-314,316 fu (yuk, Ru) 302, 304-306,305,313314, 338-339 fy (yuk, Sw/Da) 307-308, 313-314, 338
gather 241-242, 238, 239-240 gee 244, 286-287, 290, 330 go jump in the lake 246, 247 good morning 246 good night 246 good-bye 1, 246 grandstanding 68, 79, 84 guess 43 hallo 293 hardly 382 harmony 114-115 have a nice day 86-87 hej (hey, Po) 293, 298, 299 hejta (gee, Po) 292 hejie (come on, Po) 293, 299 hell! 290-291 hito wa hito da (a person is a person, Jp) 430-431 hop hop (ahoy, Po) 293,300, 301 hop, hopla (jump, Po) 290, 293,298 how about 27,62,215-218,216,245 how are you 116, 118, 132, 133-136; replies 134-135 how dare you 246, 247 how do you do 62
499
how many times have I told you? 223, 224 how nice 45-46, 287 huzza (sic-em, Po) 292 hyc (jump, Po) 290 I 12-14 ie (house, Jp) 1 -ie 55-56 inflexible 49 inform 177-179 informality 111-113, 112 intimacy 105~ 106-108 invite 211-212, 215 is it 226, 227 isn't it 45-46, 225, 226 -issimo/a (-est, It) 270-271, 272, 276, 278-279 -ja (certainly, Ko) 396 japirlyung (request, Wa) 159, 160 jeszcze (still, Po) 371-375, 372 jinjinyung (order, Wa) 160 joke 188-191,192 just 23, 344, 349-353, 354 jui: (already, Po) 371-376, 373 kansha suru (thank, Ja) 157 kawaJ (joke, Po) 165, 188-192, 191 kici-kici-kici (here kitty, Po) 292 kids are kids 391, 405-406, 407, 411,440-441,450 kimi (you, Jp) 13, 48 kompromis (compromise, Po) 48-49 kraj (country, Po) 49 kudasai (please, Jp) 78 kure (please, Jp) 78
least, at 354 ledwie (just, Po) 382-385, 384 let bygones be bygones 434, 441 letter 194 list (letter, Po) 194
500
Index of words and phrases
many, as 345 materit'sja (mother-swear, Ru) 17 merely 23,348-351,349 molto (very, It) 257, 264-266, 268, 278 more 372 more or less 345 most (Adv.) 22,272-275,276,278279 most, at 354 mostly 344 mozzies 55, 56 much, as 345 mug 17 must 236-238, 237 na (take it, Po) 293 nanren shi nanren (men are men, Ch) 423, 424 naplevat' (spit, Ru) 310-312 narod (nation, Po) 49-50 nearly 365, 366 nedaru (ask for, Jp) 155 neri neri Get black, It) 256, 267, 269, 278 nezumi (rat/mouse, Jp) 17 niemal (almost, Po) 385-387, 386 nieomal (almost, Po) 385 nieugi~ty (inflexible, Po) 49 niezlomny (unbreakable, Po) 49 no 92-93 no less 355 no more 355 noch (yet, Ge) 375-376 nu (interj, Yi) 285-286 nuie (come on, Po) 293, 298, 299 o (oh, Po) 326, 333, 334 o (oh, Ru) 339 o malo nie (only just, Po) 362, 384, 385, 387 och (ooh, Po) 323-326,324, 337 offer 211-212, 215
ogo (oho, Ru) 334, 335-336 oh 325,339 oh my God 223, 243, 244-245 oh-oh 332, 334 oho (oho, Po) 326-327, 331-334, 332, 337 oj (oy, Po) 318-322,323,325,327 oj (oh dear, Ru) 322, 323, 339 ojej (oh no, Po) 320,321 OK 38, 97, 231,232 omal (almost, Po) 385 omoiyari (empathy, Jp) 87 only 23,344,346,347-351,355, 370, 378-381 ooh 325 oops 287, 289, 290 opposite, contrastive 139 order 17, 34, 151-154, 158-159, 160, 164-165, 198-201, 202 ore (I, Jp) 13 otaku (esteemed house, Jp) 1 ow/ouch 286-289 ox (ooh, Ru) 326, 339 oy 285,317-318 oy vay (oh no, Yi) 338 pan/pani (Mr/Mrs, Po) 48, 56-59, 107 pfui 286,313-315,338 phew 286-287, 304, 313-315, 339 pismo (official letter, Po) 194 please 201, 203-204 pledge 17 podanie (application, Po) 165, 192196, 195 pooh 286,315,339 praise 453 prawda (true, Po) 38-40, 42 prawie (almost, Po) 362, 385-388, 386-387 prawo prawem (the law is the law, Po) 422-423, 425~426 precz (go away, Po) 293, 296, 297
Index of words and phrases
privacy 47 problem6n/-azo (big problem, Sp) 1 promise 17, 149, 151, 164,453; see also a promise is a promise prosz~ (please, Po) 28, 34, 60, 195 prr (whoa, Po) 292 psst 295-296, 337 pst (psst, Po) 293-296, 295 pst (pah, Ru) 296 que sera sera (what will be will be, Sp) 402, 431-432, 433-434 question 5-6, 149-151, 199 radz~
ei (I advise you, Po) 31-32, 60 rapping 3, 68, 79, 150 rather 43-44, 276-277 reckon 42-43 referent shift 137, 142 reprimand 154, 453 request 17,32,149,151,161,194195, 198-202 return 143 reveal 161-162 rieeo rieeo (very rich, It) 267 roughly 23, 360, 361 rubbish 119, 187 sake (rice whisky, Jp) 17 same here 50 satosu (warn, Jp) 153, 155-156 say 11 scab 170 schon (already, Ge) 375-376 sensei (teacher, Jp) 48 shout 170, 173-174, 175, 176-177 showboating 68, 79, 84 sincerity 71, 115-121, 117, 121 sio (shoo, Po) 293 sit down! 27 so 240-242, 241 sook 182 sort of 44
501
sounding 150, 183 spin a yarn 171 still 23, 344, 367-368, 369-372 sto je bi/o bi/o je (what has been has been, Se) 435 stylin' out 79, 84 subito subito (at once, It) 257-258, 266,267,271 suggest 202, 211-212, 215 sumimasen (indebted, Jp) 157 super/ativo (superlative, It) 270-271 sza (hush, Po) 293, 294-295 tanomu (ask, Jp) 155 tas-tas-tas (here rabbits, Po) 292 tell 34, 108-110 tell on 179 tell you what 219, 220 t'fu (spit, Ru) 23, 310-313, 311, 314, 339 tfu (spit, Po) 23, 310-313, 312, 314 thank 156, 157, 262, 453 thanks 1 that's that 425, 439-441 the law is the law 404, 419-420,421 then 240-242 think 41-43 threaten 153, 156, 163, 198, 201, 453 thunder 163-164 tju-tjuu (here doggie, Pi) 293 -to (too, Ko) 396 tpru (whoa, Ru) 293 truth 103-104, 125 tu (you, Fr) 13, 48 ty (you, Po) 48, 56-59 ty (you, Ru) 58 tylko (only, Po) 379, 380-381 uczuciowy (emotional, Po) 54 upgrade 137, 138 understand 238 ux (oof, Ru) 335
502
Index of words and phrases
ver (become, Yi) 250 very 256-257, 263-269 vow 17 vy (you, Ru) 58
war is war 23, 394, 400-401, 403404,405,420,422,440-441,444445 wara (keep away, Po) 293, 296-299, 298 warmth 87 warn 149, 152-154, 153, 156, 163165, 198,201,453 watakushi (I, Jp) 13 watashi (I, Jp) 13 what's done is done 437 whatever will be 440-441, 431-433 whinge 180, 181-182 who's talking about 224 why can't you 38, 229, 230-231 why do X 213, 214-215 why don't you 32-35, 38, 60, 62, 210-213, 212, 215-216, 218, 245, 449, 452 why not 214, 215 widz~ (I see, Po) 39 wii (whoa, Pi) 293 will 235-237, 236
will you 27, 32, 35, 38-39, 60, 204, 206, 207, 227-229, 228 wio (gee-up, Po) 292 wista (haa, Po) 292 won (get out, Po) 293, 296,297 won't you 27,38-39,227-228,229 worries, no 56 would you 2,27,29,38,51,53,62, 88-89,96, 203-204, 206, 207, 449, 452 wow 286-287, 289, 290, 330,459 wy (you, Po) 58-59 yarn 170-1 71, 172-173, 175, 177 yellow 1 yet 23, 344, 367-369,368,371 you 11, 12-14,47-48,56-59 you beauty! 234,235 you X! 233-234, 235 yuk 1,23,288,302-303,304,313315,339,459-460 zaledwie (only, Po) 381-383,382 zart Uoke, Po) 188-189 zarty zartami Uokes are jokes, Po) 422 ze (indeed, Ru) 393 zol (let/should, Yi) 251