Cross-Cultural Pragmatics

Trends in Linguistics
Studies and Monographs 53

Editor
Werner Winter

Mouton de Gruyter
Berlin • New York
Cross-Cultural Pragmatics
The Semantics of Human Interaction

by
Anna Wierzbicka

Mouton de Gruyter
Berlin • New York 1991
Mouton de Gruyter (formerly Mouton, The Hague)
is a Division of Walter de Gruyter & Co., Berlin.

® Printed on acid-free paper which falls

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

Library of Congress Cataloging in Publication Data

Wierzbicka, Anna.
Cross-cultural pragmatics : the semantics of human
interaction / by Anna Wierzbicka.
p. cm. — (Trends in linguistics. Studies and mon-
ographs ; 53)
Includes bibliographical references and indexes.
ISBN 0-89925-699-6 (acid-free paper)
1. Pragmatics. 2. Semantics. 3. Intercultural commu-
nication. 4. Speech acts (Linguistics) I. Title. II. Series.
P99.4.P72W53 1991
306.4.4-dc20 90-22563
CIP

Deutsche Bibliothek Cataloging in Publication Data

Wierzbicka, Anna:
Cross-cultural pragmatics : the semantics of human inter-
action / by Anna Wierzbicka. — Berlin ; New York :
Mouton de Gruyter, 1991
(Trends in linguistics : Studies and monographs ; 53)
ISBN 3-11-012538-2
NE: Trends in linguistics / Studies and monographs

© Copyright 1991 by Walter de Gruyter & Co., D-1000 Berlin 30

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.
Typesetting: Asian Research Service, Hong Kong — Printing: Gerike GmbH, Berlin.
Binding: Lüderitz & Bauer, Berlin. — Printed in Germany.
Acknowledgements

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 Boguslawski, 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; "Dif-
ferent 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

Acknowledgements v

Chapter 1 Introduction: semantics and pragmatics 1

1. Language as a tool of human interaction 1
2. Different cultures and different modes of interaction 2
3. Pragmatics — the study of human interaction 5
4. The natural semantic metalanguage 6
5. The need for a universal perspective on meaning 9
6. The uniqueness of every linguistic system 10
7. The problem of polysemy 11
8. Semantic equivalence vs. pragmatic equivalence 12
9. Universal grammatical patterns 14
10. Semantics versus pragmatics: different approaches 15
10.1. 'Complementarism' 16
10.2. 'Pragmaticism' 17
10.3. 'Semanticism' 18
10.4. A fourth approach: two pragmatics 18
11. Description of contents 20

Chapter 2 Different cultures, different languages,

different speech acts 25
1. Preliminary examples and discussion 27
2. Interpretive hypothesis 30
3. Case studies 31
3.1. Advice 31
3.2. Requests 32
3.3. Tags 37
3.4. Opinions 41
3.5. Exclamations 45
4. Cultural values reflected in speech acts 47
4.1. Lexical evidence 47
4.2. Objectivism as a cultural value 49
4.3. Cordiality as a cultural value 50
viii Contents

4.4. Courtesy as a cultural value 56

5. Theoretical implications 59
6. Practical implications 64

Chapter 3 Cross-cultural pragmatics and different cultural values 67

1. 'Self-assertion' 72
1.1. 'Self-assertion' in Japanese and in English 72
1.2. 'Self-assertion' in black and white American English 78
1.3. Spontaneity, autonomy, and turn-taking:
English vs. Japanese 80
1.4. 'Spontaneous self-assertion' vs. 'regulated
self-assertion':
black English vs. white English vs. Japanese 82
1.5. 'Self-assertion' as personal display:
black English vs. white English 84
1.6. 'Self-assertion' and 'good interpersonal relations' 85
2. 'Directness' 88
2.1. American culture vs. Israeli culture 89
2.2. 'Indirectness' in Japanese 93
2.3. Greek culture and American culture 95
2.4. 'Indirectness' and 'dissimulation' in Javanese 100
3. Further illustrations: same labels, different values 104
3.1. 'Intimacy' 105
3.2. 'Closeness' 108
3.3. 'Informality' 111
3.4. 'Harmony' 113
3.5. 'Sincerity' 115
4. Different attitudes to emotions 121
4.1. Polish culture 121
4.2. Jewish culture 122
4.3. American black culture 123
4.4. Japanese culture 126
4.5. Javanese culture 128
5. Conclusion 129

Chapter 4 Describing conversational routines 131

1. Conversational analysis:
linguistic or non-linguistic pragmatics? 131
 Contents ix

2. 'Compliment response' routines 136

2.1. Upgrades 138
2.2. Contrastive opposites 139
2.3. Scaled-down agreements 140
2.4. Downgrades 140
2.5. Reassignment of praise 142
2.6. Returns 142
3. 'Compliment responses' in different cultures 143
4. Conclusion 147

Chapter 5 Speech acts and speech genres across

languages and cultures 149
1. A framework for analysing a culture's 'forms of talk' 149
1.1. The importance of folk labels 150
1.2. Two approaches 150
1.3. Some examples: English vs. Japanese 152
1.4. Another example: English vs. Walmatjari 158
1.5. The elimination of vicious circles 161
1.6. Evidence for the proposed formulae 161
1.7. The first-person format 162
1.8. The problem of other minds 164
2. Some Australian speech-act verbs 165
2.1. Chiack (chyack) 165
2.2. Yarn 170
2.3. Shout 173
2.4. Dob 111
2.5. Whinge 180
3. Some examples of complex speech genres 183
3.1. The black English dozens 183
3.2. The Hebrew 'dugri talk' 185
3.3. The Polish kawal 188
3.4. The Polish podanie 192
4. Conclusion 196

Chapter 6 The semantics of illocutionary forces 197

1. Are illocutionary forces indeterminate? 197
1.1. Illocutionary forces as bundles of components 199
1.2. Illustration: the discrete and determinate
character of 'whimperatives' 202
x Contents

1.3. Syntax and illocutionary force 207

2. More whimperative constructions 211
2.1. Why don't you do X (tomorrow)? 211
2.2. Why do X? 213
2.3. How about X? 215
3. Additional remarks on the explication of
illocutionary forces 218
4. Selected conversational strategies 219
4.1. Tell you what, S! 219
4.2. Do you know, S? 220
4.3. Don't tell me S! 222
4.4. How many times have I told you (not) to do X! 223
4.5. Who's talking about doing X? 224
5. Tag questions 224
5.1. Tags with declarative sentences 224
5.2. Tags with imperative sentences 227
5.3. Why can't you (do X)! 229
5.4. OK? 231
6. Personal abuse or praise: You X! 232
7. Illocutionary forces of grammatical and
other categories 235
7.1. Modal verbs 235
7.2. Mental verbs 238
7.3. Particles and conjunctions 240
7.4. Interjections 243
7.5. Fixed expressions 245
7.6. Intonation 247
8. Comparing illocutionary forces across languages 248
9. Conclusion 252

Chapter 7 Italian reduplication: its meaning and

its cultural significance 255
1. Italian reduplication: preliminary discussion 255
2. Discourse and illocutionary grammar 258
3. The illocutionary force of clausal repetition 260
4. The illocutionary force of Italian reduplication 263
5. Clausal repetition as a means of 'intensification' 268
6. The absolute superlative in Italian and in English 270
7. Illocutionary grammar and cultural style 276
8. Conclusion 282
 Contents xi

Chapter 8 Interjections across cultures 285

1. Preliminary discussion 285
1.1. Interjections: physis and thésis
('nature' and 'convention') 285
1.2. Defining the concept of 'interjection' 290
1.3. Types of interjections 291
2. Volitive interjections 292
2.1. Interjections directed at animals 292
2.2. Interjections directed at people 293
2.2.1. The 'I want silence' group 293
2.2.2. The 'I don't want you in this place' group 296
2.2.3. The 'I want you to jump' group 298
2.2.4. The 'urging' group 298
2.2.5. The 'communication over distance' group 300
3. Emotive interjections 302
3.1. Interjections of 'disgust' and similar feelings 302
3.1.1. The Polish fu and the English yuk 302
3.1.2. The Russian/« 304
3.1.3. The Polish fe 306
3.1.4. The Yiddish feh 308
3.1.5. The Polish tfu and the Russian t'fu 310
3.1.6. 'Disgust' and bodily gestures 313
3.1.7. 'Disgust' and sound symbolism 315
3.2. 'General purpose' interjections 317
3.2.1. The Polish oj 318
3.2.2. The Russian oj 322
3.2.3. Ochs and achs 323
4. Cognitive interjections 326
4.1. The Polish aha and Russian aga 326
4.2. The Polish oho 331
4.3. The Polish o 333
4.4. The English oh-oh 334
4.5. The Russian ogo 334
5. Conclusion 337

Chapter 9 Particles and illocutionary meanings 341

1. English quantitative particles 345
1.1. Non-approximative particles:
only, merely and just 346
xii Contents

1.1.1. Only 346

1.1.2. Merely 348
1.1.3. Just 350
1.2. English approximative particles 354
1.2.1. Around and about 355
1.2.2. Approximately 358
1.2.3. Roughly 360
1.2.4. Almost and nearly 361
2. English temporal particles 367
3. Polish temporal particles 371
3.1. Juz and jeszcze 371
3.2. Dopiero 376
4. Polish quantitative particles 379
4.1. Non-approximative particles 379
4.1.1. Tylko 379
4.1.2. Az 380
4.1.3. Zaledwie 381
4.1.4. Ledwie 382
4.2. Polish approximative particles 384
4.2.1. O malo nie 384
4.2.2. Niemal and prawie 385
4.2.3. Blisko 388
5. Conclusion 389

Chapter 10 Boys will be boys: even 'truisms' are

culture- specific 391
1. The meaning of tautologies 391
1.1. Gricean maxims: universal or language-specific? 392
1.2. Problems in interpreting implicatures 397
1.3. Context as an excuse for analytical failure 400
2. English nominal tautologies: semantic representations 403
2.1. 'Realism' in human affairs 404
2.2. Tolerance for human nature 405
2.3. Tolerance at 'special times' 408
2.4. The limits of tolerance 410
2.5. Seeing through superficial differences 411
2.6. Recognising an irreducible difference 413
2.7. Tautologies of value 414
2.8. Tautologies of obligation 419
 Contents xiii

3. Some comparisons from Chinese and Japanese 423

3.1. Chinese concessive tautologies 423
3.2. 'Irreducible difference', Chinese style 426
3.3. Chinese tautologies of unreserved praise 427
3.4. Japanese tautologies of 'a matter of course' 429
3.5. Japanese tautologies of irrelevance 430
4. Verbal tautologies 431
4.1. Future events 431
4.2. The immutability of the past 434
5. Is there a semantic invariant? 439
6. The deceptive form of English tautological
constructions 444
7. The culture-specific content of tautological patterns 446
8. Conclusion 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 pragmatics

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 interac-

tion. 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, ani-
mals, 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 grammati-
cal 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 fille 'girl' vs. le
garqon '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

imperative (a) and a so-called 'whimperative' (b):
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 parlia-
mentary 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', 'anti-
sentimentality', '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
exchanges (Bowles 1986:37; cf. Chapter 4):
G'day, mate, owyagowin?
Nobbad. Owsyerself? (Or: Cam 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 example of 'rap-
ping' 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 non-
standard 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 discus-
sion, 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 — the study of human interaction 5

3. Pragmatics — 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, grammati-
cal 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 unrival-
led access to contemporary English and the way it is used", offers us
the following definitions:
question - 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) respond - to write or speak in reply
(to) reply - to respond in words or writing
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 diction-
aries (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', 'inti-
macy', '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, English-
specific), and which provide no language-independent, universal per-
spective 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 seman-
tic 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

I propose for this purpose a 'natural semantic metalanguage', based on

a hypothetical system of universal semantic primitives, which my
colleagues and I 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 I will try to state the meanings under consideration
in terms of simple and intuitively understandable sentences in natural
language. This, I believe, will ensure that the proposed semantic explica-
tions 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-independ-
ent (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 I believe to be the
universal core of natural languages. For example, if I say in an
explication:'I want', I mean something that could be just as easily repre-
sented 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, I think,
to call it a 'natural semantic metalanguage', (cf. Goddard 1989a,b), be-
cause 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

The lexicon of the 'natural semantic metalanguage' is based on the

current version of the hypothetical set of universal semantic primitives
(see Wierzbicka 1989a,b; Goddard 1989a,b; Boguslawski 1981b, 1989).
The first version of this set, posited in Wierzbicka (1972), included just
fourteen elements. The current, considerably expanded, version of this
set, contains over two dozen elements (some of them regarded as more
hypothetical than others). They are:
Pronouns Determiners Classifiers Adjectiv
I this kind of good
you the same part of bad
someone two
something all
Verbs Modals Place/Time Linkers
want can place like
don't want if/imagine time because
say after (before)
think above (under)
know
do
happen
It is not clear at this stage to what extent the items listed above can be
viewed as true lexical universals. It is impossible to determine without
painstaking semantic analysis whether or not a given language has
separate words embodying concepts such as, for example, 'part', 'know',
or 'like'. (It is certainly not sufficient to consult dictionaries, native
speakers, or standard linguistic descriptions of the languages in ques-
tion.) It seems certain, however, that a semantic metalanguage based on
the twenty-odd hypothetical primitives listed above can be matched to a
considerable degree across different languages of the world. In addition,
the metalexicon employed in the present work includes a limited number
of other concepts, which are regarded as neither indefinable nor univer-
sal or near-universal, but which are still relatively very simple and which
recur widely in the languages of the world as separate lexical items. This
larger set, whose items can be defined in terms of the basic set of
primitives, includes concepts such as 'feel', 'small', 'much', 'a little',
'more', 'less', 'different', and so on.
 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 inevi-
tably 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 idiosyn-
cratic 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 (i.e.
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 univer-
sal 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, how-
ever, the presence or absence of a word for a given concept cannot be
established by any mechanical, checklist method. The search is empiri-
cal, but it also has necessarily an analytical dimension. Above all, there
is the problem of polysemy. For example, I 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 SG/PL'. Yet one doesn't have to
look further than modern 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 connaître, 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 yousc and youPL
(cf. youSG yourself vs. youPL 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, how-
ever, 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. Barnlund 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', self-
deprecating 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;
Palakornkul 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 in-
comparably 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 soci-
ety 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 commensu-
rable sentences. This means that we must look not only for commensu-
rable 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, there-
fore, 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 universal-
ity, 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 rela-

tionship 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

semantics semantics I (semantics) I

V W W W
i 1
I (pragmatics) I pragmatics 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 'illocu-
tionary' 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 con-
sists in human interpretation of the world. It is subjective, it is anthropo-
centric, 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 Padu-
ceva 1985).
Since the meanings conveyed in natural language are inherently
subjective and anthropocentric, they cannot be neatly divided into 'refer-
ential' 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 'semanti-
cism' 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 convey-
ing 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 'radi-
cally 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 com-
pare 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 impover-
ished 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 com-
petence, 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 belong-
ing 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 seman-
tics. 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

Non -linguistic
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 univer-
sal 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 con-
versational 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 speech-
act 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 com-
ponents, 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 theo-
retically 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 signifi-
cance', 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 de-
vices', 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 distin-
guished 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 charac-
teristic 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 yuk! is
compared and contrasted with its nearest Polish and Russian counterparts
fu!, fe!, tfu! and t'fu! 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 possi-
bility 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 'approxima-
tive' 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 'approxi-
matives' (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 pragmat-
ics', 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', culture-
specific, 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 im-
peratives like Tell me the time, which are direct requests, in preference
for questions like Can you tell me the time? or assertions like I'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 re-
quests 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' deter-
mined 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 f r o m 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 any one 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
d i f f e r e n c e s b e t w e e n A u s t r a l i a n E n g l i s h and A m e r i c a n E n g l i s h ,
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 com-
piled 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 21

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 con-
text 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 indis-
tinguishable from that of genuine questions (except of course for the
intonation):
28 Different cultures, different languages, different speech acts

Moze sif czegos 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 inap-
propriate. 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 con-
cerned 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:
Mialbys ochotp 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 troszkg!
Ale juz nie mog$!
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 ochotg ... ?
'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:
Mozebysmy poszli do kina?
'Perhaps we would go to the cinema?' (implied: if I asked you)
rather than:
Czy mialabys ochotg 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, zebys miala ochote zjgsc lunch
ze mng, 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 system of
devices reflecting a characteristically Anglo-Saxon cultural tradition:
a tradition which places special emphasis on the rights and on the auton-
omy of every individual, which abhors interference in other people's
affairs (It's none of my business), which is tolerant of individual idiosyn-
crasies 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 impera-
tive. 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 o nomimasho! (Literally, 'We will drink Coca
Cola!') rather than the imperative Coca Cola o nomel 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 instruc-

tions 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:
Ja ci radz% powiedz mu prawdq.
'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 radzic used
performatively could be so reported:
32 Different cultures, different languages, different speech acts

Radzg ci, zebys mu powiedzial prawdg.

'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 radzg 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 follow-
ing 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 combina-
tion 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 chusteczkq?
'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 proszq '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 possibili-
ties 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 moglbys ... ?
'Could you ... ?'
Czy bylbys tak dobry, zeby ... ?
'Would you be so good as to ... ?'
Czy byl(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 interroga-
tive 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? (William-
son 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? (Wil-
liamson 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 re-
spect 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 inter-
rogative 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 extra-
polite 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-so-
polite 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 feel-
ings, 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$ tutaj!
'Not to show oneself to me here!' (i.e. 'You are not to come here.')
Wynosic siq stqd!
'To get away from here!' (i.e. 'Get away from here!')
Zabierac si% stqd!
'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 sig) 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 'non-
existence'): 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:
Made parasol? Isc prosto - nie oglpdac si§. Pamigtac:
skromnosc - skarb dziewczgcia. (Zapolska 1978:30)
'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 combina-
tion with a performatively used verb:
Proszg sig do tego nie mieszac. (Zapolska 1978:108)
'I ask not to interfere.'
Proszg - proszg powiedziec, proszg siq 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 the door, will you?
Close the door, won't you?
Close the door, could you?
Close the door, can't you?
Close the door, why don't you?
Close the door, why can't you?
Close the door, 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.