Documente Academic
Documente Profesional
Documente Cultură
WDE
G
Cognitive Linguistics Research
17
Editors
Rene Dirven
Ronald W. Langacker
John R. Taylor
Mouton de Gruyter
Berlin New York
Emotions
in Crosslinguistic Perspective
Edited by
Jean Harkins
Anna Wierzbicka
w
DE
G
Mouton de Gruyter
Berlin-New York 2001
Mouton de Gruyter (formerly Mouton, The Hague)
is a Division of Walter de Gruyter GmbH & Co. KG, Berlin
Copyright 2001 by Walter de Gruyter GmbH & Co. KG, D-10785 Berlin
All rights reserved, including those of translation into foreign languages. No part of this book
may be reproduced or transmitted in any form or by any means, electronic or mechanical,
including photocopy, recording, or any information storage and retrieval system, without
permission in writing from the publisher.
Printing: WB-Druck, Rieden/Allgau
Binding: Lderitz & Bauer, Berlin
Printed in Germany
Contents
Introduction 1
Anna Wierzbicka and Jean Harkins
The study of human emotions needs input from the study of languages,
now more than ever before. Advances in the science of brain physiology
are identifying in ever greater detail the specific wave patterns and lo-
cations of brain activity that correspond to different affective or emotional
states. Studies of this kind require high technology, and most of them are
conducted in predominantly English speaking research environments. It is
expected that their findings will apply to human brains generally, not just
those of a particular language or cultural group (cf. e.g. Davidson and
Ekman, eds. 1994; Ekman and Rosenberg, eds. 1997). Further research
may confirm or challenge this expectation, but the research itself and the
interpretation of its results hinge at certain crucial points upon questions of
language. These questions have to be approached seriously, for the validity
of scientific studies of emotional response may depend upon whether the
issue of language is addressed in an informed way or by default.
Along with increased technical understanding of the workings of human
brains, there have also been major advances in the study of emotions from
the perspectives of cognitive and crosscultural psychology, psychological
anthropology, and sociology. Some of this work has recognised the role of
language as central to the study of emotions, particularly when examining
instances where the cultural life of one group seems to focus attention on
emotional states for which other groups don't even have names. Detailed
descriptions of the meanings and manifestations of emotional states in dif-
ferent linguistic and cultural groups have added much to the understanding
of emotions in cultural context.
The purpose of the crosslinguistic studies presented in this volume is to
demonstrate how the tools of linguistic analysis can be applied to produce
more accurate descriptions of the meanings of emotion words and, more
generally, ways of speaking about emotions in different languages. Such
analyses of linguistic meaning not only complement findings from other
approaches to the study of emotions, but help to resolve methodological
problems that arise when these other approaches have to deal with data
2 Anna Wierzbicka and Jean Harkins
pany it, it is only through language (if at all) that we can know that what is
experienced is anger: that is, if the experience! says so, or says other things
by which we know that the person feels anger. Even if an identical physio-
logical pattern is observed on two occasions, only language enables us to
connect these two events with the emotion of anger. Most, though of course
not all, of what we know about people's inner feelings comes to us via lan-
guage, and it is only language that enables people to use the same word,
anger or angry, to designate two different instances of an emotional experi-
ence. While theories of word meaning and use differ, the point here is that
it is language that provides a conceptual connection between two disparate
emotion experiences, by giving them the same label.
The same label, however, may not apply across languages. Bilingual dic-
tionaries usually equate English anger with French colere, for example,
and deal in varying ways (if at all) with the fact that they cannot correctly
be substituted for each other in every situation. The further observation that
the adjective angry is often, though not invariably, better translated as/w-
rieux (which therefore in turn cannot be regarded as an exact equivalent of
furious) is only part of the story (cf. Wierzbicka 1988). So, even if we think
we can understand what anger means for two people who use this English
word to label their experience, how are we to understand the supposed
equivalent or equivalents in another language and culture?
Even in the case of these two fairly closely related European languages,
it is widely perceived that the differences in usage of emotion words are
connected in some way with cultural attitudes and cultural identity. The
different "feel" of the words angry, furious, furieux has much to do with the
kinds of things English and French speakers do, how they look, sound and
behave, when they feel these emotions. Studies of societies in other parts of
the world have explored how certain patterns of emotional response seem
to be centrally involved in the social behaviour of a cultural group. For
example, Briggs (1970) examined an Eskimo group where very little atten-
tion was devoted to feelings comparable to those described in English as
angry, while more socially cohesive emotions received greater cultural em-
phasis. The passionate feeling of liget described by Rosaldo (1980) in the
Philippines had some similarities to anger, but was seen by Ilongot speak-
ers in a more positive light, as a constructive force motivating much worth-
while human activity.
Links between cultural identity and emotions identified by particular
words, often held to be untranslatable, have long been recognised. They
4 Anna Wierzbicka and Jean Harkins
lexical and grammatical devices from interjections to reflexives that are de-
ployed in various languages to express emotional meanings. In analysing
the language data, we show that it is possible to identify what elements of
meaning are specific to the language under examination, and what el-
ements are shared with similar words and expressions from other lan-
guages.
While this book is written in English, we explicitly recognise the limi-
tations imposed by the language of inquiry. We distinguish in a principled
way between words belonging to particular languages, and elements of
meaning that may be applicable across languages and cultures. Throughout
the volume we indicate typographically whether we are speaking of a lan-
guage-specific word or concept, such as English anger or Malay marah (in
italics), an English gloss such as 'anger' (in single quotation marks), or el-
ements of meaning involved in such a concept, such as "feeling something
bad". If a more general reference is made without typographic marking, for
example to an angry expression or an anger-like feeling, it is recognised
that these are English words and cannot be assumed to apply across lan-
guages and cultures.
ordinary English words we unwittingly fall prey to the "naive picture" that
is reflected in them.
Nor can we transcend this "naive picture" by adopting the conventional
technical language of traditional psychology, because this conventional
language has developed on the basis of ordinary language and is coloured
by the naive picture embedded in it. The reliance of many psychologists on
English folk-distinctions such as that between emotion and mood is a good
case in point (cf. Mandler 1975).
Despite insights from, inter alia, psychologists like Mandler, the faith in
technical English as a suitable metalanguage for describing human emo-
tions across languages and cultures is still deeply entrenched in emotion re-
search in general, and the psychological literature on emotion in particular.
For example, Van Geert (1995: 260) writes (with reference to one of the
present authors): "Anna Wierzbicka argues (correctly, I think) that a theory
of emotion universals cannot be stated in terms such as 'sadness is a uni-
versal emotion' because the word 'sadness' has a culture- and language-
specific meaning, and can therefore not be used to refer to universal mean-
ings". But the solution that Van Geert proposes is a retreat into a technical
language (that is, in practice, technical English):
It is very possible that the number of [emotion-related] physiological patterns is lim-
ited and universal, but that there is no universality in the corresponding subjective ex-
perience. That is, the universals are of an entirely technical nature, they require par-
ticular technical instruments and theories and do not correspond with intuitive ex-
periential categories accessible to the layperson. [... ] The emotions are experiential
universals, that is, all people are in principle able to entertain a set of similar subjec-
tive experiences called the universal emotions. But the only way to refer to such
shared emotions is by means of a technical language, (pp. 261-262)
people's subjective experience and their own way of thinking about it.
Here, technical English will help us even less than ordinary English. In
ordinary English, we can single out some words (and grammatical con-
structions) which can be rendered in the language of the speakers whose
experience we are trying to understand - words like "good" and "bad",
"feel", "want" and "think", "people", "body", "part", "inside" and so on.
Technical English, on the other hand, is totally unrelated to the experi-
ence of people in other cultures - or, for that matter, that of most people
in English-speaking cultures. "Technical universals" of emotion (what-
ever they are) cannot throw light on ordinary human experience and con-
ceptualisation of experience. To understand this ordinary experience, and
the way "ordinary" people think about it, we must pay attention to the
way "ordinary" people talk; and to try to understand their talk we must
rely on concepts "they" (the "laypeople") and "we" (the "experts") share.
In this task, simple and universal words such as "good" and "bad", "feel",
"want" and "think", "people", "body", "part" and "inside" will be our
most reliable tools. (For further discussion, cf. Enfield and Wierzbicka, In
press.)
The approach to "emotions" adopted in most chapters of this book (and
in earlier publications by the editors and colleagues) seeks to break the de-
pendence on any one natural language as the source of "common sense in-
sights" by anchoring the analysis in universal human concepts and their
"universal grammar", according to the Natural Semantic Metalanguage
methods developed by Wierzbicka and colleagues. Both the universal con-
cepts and their universal rules of combination (i.e. their "grammar") have
been arrived at by empirical crosslinguistic investigations carried out by
several linguists over many years, working with typologically diverse and
genetically unrelated languages, including Chinese (Chappell 1986, 1991,
1994), Japanese (Hasada 1997, 1998; Onishi 1994, 1997; Travis 1997),
Malay (Goddard 1995 a, 1996), Lao (Enfield In press), Mbula (Bugen-
hagen 1994), Ewe (Ameka 1990 a, 1990b, 1994), French (Peeters 1994,
1997), several Australian Aboriginal languages (Goddard 1991b; Harkins
1995, 1996a; Harkins and Wilkins 1994; Wilkins 1986), among others.
Most words in any language are specific to this particular language or to
a group of languages, and are not universal. For example, neither English
nor Spanish nor Malay has a word with a meaning corresponding exactly to
the meaning of the German word Angst (cf. Wierzbicka 1999). At the same
time, there is evidence that all languages have words with meanings corre-
12 Anna Wierzbicka and Jean Harkins
spending exactly to the meanings of the English words good and bad, or
know and want. This suggests that the concepts of "good" and "bad",
"know" and "want" are universal, and can therefore be used as elements of
a culture-independent semantic metalanguage.
Furthermore, the use of conceptual primitives allows us to explore
human emotions (or any other conceptual domain) from a universal, lan-
guage-independent perspective. Since every language imposes its own
classification upon human emotional experience, English words such as
anger or sadness are cultural artefacts of the English language, not culture-
free analytical tools. On the other hand, conceptual primitives such as
'good' and 'bad', or 'want', 'know', 'say' and 'think' are not cultural arte-
facts of the English language but belong to the universal "alphabet of
human thoughts"; and they do appear to have their semantic equivalents in
all languages of the world. Basing our analysis on lexical universals we can
free ourselves from the bias of our own language and reach a universal, cul-
ture-independent perspective on human cognition in general and on human
emotions in particular.
The latest empirical evidence (cf. Goddard and Wierzbicka (eds.) 1994,
In press; Wierzbicka 1996; Goddard 1998) suggests that the full set of hy-
pothetical conceptual primitives, lexicalised in all languages of the world,
looks something like this:
This is quoted from the translation by Louise and Aylmer Maude; for more
detailed discussion of this literary practice, see Wierzbicka (1973).
The same mode of description is also often used in everyday discourse,
as well as in popular songs and other similar texts. A simple example is
seen in the blues song, "Sometimes I feel like a motherless child". Much
could of course be written about what it means to "feel like a motherless
child", but the expression "I feel like..." itself cannot be defined or ex-
plained any further: it is as simple and clear as anything can be. There is no
point in trying to define or explain the meaning of "I", "feel", "like", or the
combination "I feel like (this)". The understanding of the whole line dep-
ends not only on the assumption that one knows (or can imagine) how
"motherless children" feel, but also that the meaning of the expression "I
feel like this" is intuitively clear.
Although "feeling" cannot be defined, "ordinary people" generally as-
sume that the way one feels can be described and that one can tell other
people how one feels. There are many ways of describing to other people
how one feels, but most of them can be reduced to three basic modes:
(1) one can tell other people that one "feels good" or that one "feels bad";
and (2) one can tell other people that one feels like a person feels in a cer-
tain situation and then identify, in one way or another, that prototypical
situation, (3) one can tell people what seems to be happening inside one's
body. If I tell someone that I "feel wonderful" or that I "feel awful", I am
following the first mode of describing feelings. If I tell them that I "feel like
a motherless child", or that I "feel lost", or that I "feel abandoned", I am
following the second mode. If I tell them that "my heart is breaking" or that
"my heart is heavy", I am following the third.
For example, we can say that to "feel pleased" means, roughly, "to feel
something good, like a person does who thinks: something good happened,
I wanted this to happen"; that to "feel delighted" means, roughly, "to feel
something very good, like a person does who thinks: I know now that
something very good happened, I didn't know before that this would
happen"; that to "feel hope" means, roughly, "to feel something good, like
a person does who thinks: I don't know what will happen, some good
things can happen, I want these things to happen"; and that "to feel fear"
means, roughly, "to feel something bad, like a person does who thinks: I
don't know what will happen, some bad things can happen, I don't want
these things to happen". Using a standardised mode of semantic descrip-
tion we can represent these meanings as follows:
Introduction 15
In these semantic formulae, the symbol "X" stands quite simply for "this
person". This symbol is used for convenience, because in reading the for-
mulae quickly, confusion might arise between the "this person" of the ut-
terance scenario (this person (X) was pleased) and the "this person" of the
prototype scenario (sometimes a person thinks ...) The two are in fact quite
distinct on careful reading, as the prototype scenario is a self-contained
subset of the utterance meaning. The composition of semantic represen-
tations of emotion words and expressions is more fully explained in
Wierzbicka (1999).
Examples of this kind provide an answer for those who, like Paul Harris
(1995), ask "whether we can draw any conclusions - other than lexical
conclusions - about the emotional universe of a culture by examining its
emotion lexicon." The answer is that, by examining the meaning and the
use of words like anger and angry in contemporary English, we can indeed
learn a great deal about the "emotional universe" of the speakers of con-
temporary English. "Emotion words" such as anger reflect, and pass on,
certain cultural models; and these models, in turn, reflect and pass on valu-
es, preoccupations and frames of reference of the society (or speech com-
munity) within which they have evolved. They reflect its "habits of the
heart" (Bellah et al. 1985) and the concomitant "habits of the mind". The
English anger and angry (with their current range of meanings and use)
both reflect and reinforce what Bellah et al. (1985) call the "therapeutic
culture" in modern Anglo society; and the shift from the Shakespearian
wrath to modern anger both reflects, and constitutes an aspect of, the de-
mocratisation of society and the passing of the feudal order (cf. de Tocque-
ville 1953 [1835-40]; Stearns and Stearns 1986; Logan 1998).
Naturally, it is not only the lexicon that provides clues to the "emotional
universe of a culture". Grammar does too, as do phraseology, discourse
structure, gestures, intonation, interjections, swearwords, forms of address,
culture-specific facial expressions and bodily postures, and so on. For
example, it is clearly significant that "active" verbs like rejoice have all but
disappeared from modern English usage, giving way to "passive" adjec-
tives like happy or pleased; and that those "emotion verbs" which remain
tend to have pejorative or humorous connotations (cf. e.g. fume, fret, sulk,
pine, enthuse, rage, and so on; for detailed discussion see Wierzbicka
1988, 1995; cf. also Hbler 1998, Dirven 1997, Osmond 1997 on various
emotion-related grammatical patterns). Similarly, the Russian "emotional
universe" is reflected in Russian expressive derivation, including notably
the numerous "diminutive" suffixes with different emotional shadings en-
coded in each of them (for detailed discussion see Wierzbicka 1992b; see
also Friedrich 1997). It is also reflected in the culture-specific phraseology
centred on the human body and expressive bodily (and facial) behaviour
(cf. lordanskaya and Paperno 1996; Wierzbicka 1999).
The semantic explorations in this volume go far beyond the lexical se-
mantics of emotion words and their equivalence or otherwise in different
languages. As several of these authors point out, the "emotional world" or
"emotional universe" of the speakers of Chinese, Russian, Mbula or Malay
18 Anna Wierzbicka and Jean Harkins
is much richer than just their set of lexical labels for particular feelings.
"Emotion talk" or talk about people's feelings - one's own or other
people's feelings or both - may receive greater or lesser emphasis in each
cultural and social setting, but in every language examined thus far, people
don't just name feelings: they have recourse to a whole realm of discourse
about human feelings and their manifestations. (Cf. also Palmer 1996;
Palmer, Bennett and Stacey 1999, among others.)
The domains of emotion discourse and phraseology in each language
and culture are represented in these studies in a wealth of different ways.
Semantic analysis of particular emotion concepts has to pay close attention
to evidence from their distinctive collocations and associated phraseology.
To give just one example from this volume, an understanding of the nature
of the Malay hau or seat of emotion relies substantially on the discourse
concerning it: the hau can "speak" to the person, though it apparently
doesn't "move" (unlike the seat of emotion in many other languages); it
can be sakit 'sick',/7os 'content', kecil 'small' and many more - and each
of these feeling-states has its own semantic content that cannot be derived
by simply adding together the meanings of the individual words. Such
phraseology can involve both metaphorical and non-metaphorical cultural
models (cf. Emanatian 1999; Gibbs 1999; Yu 1998), which in turn pose
challenges for how to represent these meanings in semantic explication.
Even the non-verbal facial and bodily signals of emotional feeling are in
turn given linguistic representations - and these too have their distinctive
interpretations in each language and culture, so that the Mbula mata- sigsig
(lit. 'eye red') hasn't anything to do with crying, but rather with eagerness;
and the Japanese hiya-hiya 'shivery feeling' isn't so much a "frisson" as a
helpless apprehensiveness. It is this wealth of associated discourse that
makes the domain of feelings such a rich field for linguistic, anthropologi-
cal, psychological and social science investigation. Some evidence is seen
in these studies for possible universals of emotion phraseology, for
example, description of emotional feelings via bodily events, sensations,
images (Wierzbicka 1999), and culture-specific syntactic and pragmatic
patterns (Goddard Forthcoming a; Enfield In press).
All these aspects of both verbal and non-verbal communication deserve
careful attention and analysis both within their specific linguistic and cul-
tural contexts, and across cultures. None of them, however, can be studied
with precision and without an ethnocentric bias if attention is not given to
the vital matter of a suitable metalanguage. Whether one wishes to focus
Introduction 19
Just as it is easy to fall into "either-or" thinking in relation to verbal and diagram-
matic representation, so too with the contrast between symbolic (conceptual) mean-
ing and its experiential (pre-conceptual) underpinnings. However, one may very well
grant that embodied, pre-conceptual experiential Schemas (kinaesthetic image-sche-
mas) underlie, constrain and support the emergence of conceptual meaning, without
accepting that conceptual meaning is reducible to experiential Schemas.... [B]oth
conceptualist and experientialist perspectives are needed if we are to get anything ap-
proaching a full picture of language as a human phenomenon (just as we need both
universalism and relativism...)
into the structures of the languages under investigation. These papers dem-
onstrate a greater convergence than previously found between different re-
searchers' NSM formulations, but important variations in explicatory style
and practice remain. Alternations in phrasing between, for example,
"thinks something like:..." vs "thinks something like this:...", or "I didn't
think before now: this will happen" vs "I didn't think before that this would
happen", or "sometimes a person thinks" vs "sometimes people think" are
more than trivial stylistic variations, in that they represent different posi-
tions with regard to the semantic valency (or combinatorial properties) of
the primes "like" and "think", or the association of prototypical cognitive
scenarios with individuals or groups. Such theoretical issues, which are
only resolved through ongoing empirical work of the kind seen here, are
not generally pursued in this volume, but are explored in Goddard and
Wierzbicka (In press).
Researchers should not be deterred from using this method to describe
cultural meanings of interest to them, by feeling that they must master a
rigidly formalised system of representation. Although the emerging formal
properties of the metalanguage are indeed worthy of close examination
(see Goddard and Wierzbicka In press), it can readily be employed as a
practical tool for investigating meaning. Statements of meaning can be pro-
posed, tested against more data, and tested for intelligibility and translat-
ability as shown in several of these studies. It takes many successive revi-
sions and much analytical work to arrive at optimally clear and accurate
explications, but many researchers are now finding this process immensely
rewarding in terms of the cultural insight gained through it, in addition to
the usefulness of the outcomes for language teaching and intercultural
communication.
In the first paper, Mengistu Amberber demonstrates that the meanings of
both lexical and grammaticised expressions of emotion in Amharic can be
fully described in terms of semantic universale, allowing far greater accu-
racy than approaches based on partial translation equivalents. The poly-
semy of the verb (t3)s9tnma 'feel; hear' is resolved on formal, language-
internal criteria; and it is found that the Amharic data support the main pro-
posals that have been put forward about universal trends in the linguistic
expression of cognitively-based feelings. The anger-like verb tak'ot't'a
reveals an interesting link between emotion and speech act, containing a
semantic element of wanting to rebuke the person who aroused the angry
feeling.
Introduction 23
The Mbula language of Papua New Guinea is one of many in its region
that make extensive use of body image expressions, as the most productive
and semantically rich means of talking about emotional experiences. Ro-
bert Bugenhagen's detailed explications of expressions based on mata-
'eye', lele- 'insides', kete- 'chest, liver', ni- 'being', kuli- 'skin', kopo-
'stomach' show how a language that reflects a view of the human person
markedly different from "Western" concepts (cf. also McVeigh 1996),
draws nonetheless on the same kinds of semantic components (THiNKing
something, FEELing something GOOD/BAD, WANTing to do something) that
other languages do in emotion expressions. These explications occasion-
ally introduce elements from outside the basic NSM set of primes. Al-
though such elements could themselves be reduced to combinations of
primes, it sometimes works better to retain them as "semantic molecules"
to enhance the readability of long definitions, and explicate these mol-
ecules separately elsewhere.
Turning to a European language, a corpus-based analysis by Uwe Durst
of rger, Wut and Zorn in German illustrates how even a very closely re-
lated language does not have an exact match for the supposedly basic Eng-
lish anger, but that the differences in distribution of these words can be ex-
plained in terms of subtle differences in their semantic structure. He also
demonstrates some of the ways in which historical changes in cultural
models, as suggested above, may be reflected in semantic changes and
shifts in usage of a group such as the anger-like emotions.
Several genetically unrelated languages of Asia provide rich sources of
data and analysis. Ways of talking about facial expressions in Lao, studied
by Nick Enfield, reveal not only a culture-specific set of associations be-
tween emotions and facial gestures, but also a distinctive ethnolinguistic
perspective on what are the significant features or components of facial ex-
pression. At the same time, the Lao evidence provides support for propos-
als concerning the universality of "cry" and "smile" as facial expressions
of feeling (cf. Wierzbicka 1999). While this paper does not propose expli-
cations in the same style as most of the others, it nonetheless demonstrates
the same kind of empirical and non-linguocentric approach to the identifi-
cation and discussion of meanings.
Cliff Goddard gives a thorough overview of the Malay cultural keyword
hati, referring to what is seen as the seat of human feelings and a highly
productive source of linguistic expressions of emotion. Semantic analysis
of this word provides a key to Malay ethnopsychology, both in its under-
24 Anna Wierzbicka and Jean Harkins
Notes
1. The work on the language of emotions carried out within the NSM framework in-
cludes Ameka (1990), Bugenhagen (1990), Goddard (1990, 1991, 1995 a, b, 1996,
26 Anna Wierzbicka and Jean Harkins
1997 a, b), Harkins (1990, 1995, 1996), Harkins and Wierzbicka (1997), Hasada
(1997,2000, In press), Kornacki (1995), Mostovaja (1988), Priestley (In press), Tra-
vis (1998), Wierzbicka (1973, 1986, 1990a, b, 1992a, b, c, 1993, 1994a, b, 1995a,
b, 1997, 1998a, b, c, 1999, 2000), Wilkins (1986), and Ye (2000, In press).
References
Ameka, Felix
1990 The grammatical packaging of experiences in Ewe: a study in the se-
mantics of syntax. Australian Journal of Linguistics 10: 139-182.
1994 Ewe. In Goddard and Wierzbicka (eds.), 57-86.
Apresjan, Jurij D.
1992 Lexical Semantics. Ann Arbor: Karoma. [first pub. 1974, in Russian].
2000 Systematic Lexicography. Translated by Kevin Windle. Oxford: Ox-
ford University Press.
Athanasiadou, Angeliki and Elzbieta Tabakowska (eds.)
1998 Speaking of Emotions: Conceptualisation and Expression. Berlin:
Mouton de Gruyter.
Bellah, Robert N., Richard Madsen, William M. Sullivan, Ann Swidler, and Steven M.
Tipton (eds.)
1985 Habits of the Heart: Individualism and Commitment in American Life.
Berkeley: University of California Press.
Benedict, Ruth
1946 The Chrysanthemum and the Sword. Boston: Houghton Mifflin.
Briggs, Jean
1970 Never in Anger: Portrait of an Eskimo Family. Cambridge, MA: Har-
vard University Press.
Bugenhagen, Robert D.
1990 Experiential constructions in Mangap-Mbula. Australian Journal of
Linguistics 10(2): 183-215.
1994 The exponents of semantic primitives in Mangap-Mbula. In God-
dard & Wierzbicka (eds.), 87-108.
Chappell, Hilary M.
1986 Formal and colloquial adversity passives in standard Chinese. Lin-
guistics 24: 1025-1052.
1991 Strategies for the assertion of obviousness and disagreement in Man-
darin: a semantic study of the particle me. Australian Journal of Lin-
guistics 11: 39-65.
1994 Mandarin semantic primitives. In Goddard & Wierzbicka (eds.),
109-147.
Introduction 27
Kupperbush, C, Matsumoto, D., Kooken, K., Loewinger, S., Uchida, H., Wilson-Cohn,
C. and Yrizarry, N.
1999 Cultural influences on non-verbal expression of emotion. In: P. Philip-
pot, R. S. Feldman and E. J. Coats (eds), The Social Context of Non-
verbal Behaviour, 17-44. Cambridge: Cambridge University Press.
Langacker, Ronald W.
1999 Grammar and Conceptualization. Berlin: Mouton de Gruyter.
Logan, Sarah
1998 The changing meaning of English emotion terms. BA Honours thesis,
Department of Linguistics, Australian National University.
Lutz, Catherine
1985 Ethnopsychology compared to what? In Geoffrey M. White & John
Kirkpatrick (eds.) Person, Self and Experience, 35-79. Berkeley: Uni-
versity of California Press.
1988 Unnatural Emotions: Everyday Sentiments on a Micronesian Atoll and
their Challenge to Western Theory. Chicago: University of Chicago
Press.
1995 Need, nurturance and the emotions on a Pacific atoll. In J. Marks and
R. T. Ames (eds.) Emotions in Asian Thought: A Dialogue in Com-
parative Philosophy, 235-252. Albany: State University of New York.
Malcolm, Norman
1966 Ludwig Wittgenstein: A Memoir. London: Oxford University Press.
Mandler, George
1975 Man and Emotion. New York: John Wiley & Sons.
1997 Foreword. In: James A. Russell and Jose Miguel Fernndez-Dols, The
Psychology of Facial Expression, xii-x. Cambridge: Cambridge Uni-
versity Press.
McVeigh, Brian
1996 Standing stomachs, clamoring chests and cooling livers: Metaphors in
the psychological lexicon of Japanese. Journal of Pragmatics 26:
25-50.
Moore, C. C., Romney, A. K., Hsia, T. and Rusch, C. D.
1999 "The universality of the semantic structure of emotion terms: methods
for the study of inter- and intra-cultural variability". American Anthro-
pologist 101 (3): 529-546.
Mostovaja, Anna D.
1998 On emotions that one can "immerse into", "fall into" and "come to":
the semantics of a few Russian prepositional constructions. In: Atha-
nasiadou and Tabakowska (eds.), 295-329.
Niemeier, Susanne
1997 Introduction. In: Susanne Niemeier and Rene Dirven (eds.), vii-xvii.
Introduction 31
Russell, James A.
1991 Culture and the categorisation of emotions. Psychological Bul-
letin 110:426-450.
Russell, James A., Jose-Miguel Fernndez-Dols, Antony S. R. Manstead and J. C.
1995 Wellenkamp (eds.), Everyday conceptions of emotion. Dordrecht:
Kluwer.
Shweder, Richard
1985 Menstrual pollution, soul loss, and the comparative study of emotions.
In Arthur Kleinman & Byron Good (eds.) Culture and Depression:
Studies in the Anthropology and Cross-cultural Psychiatry of Affect
and Disorder, 182-215. Berkeley: University of California Press.
Stearns, Carol Z. & Peter N. Stearns
1986 Anger: The Struggle for Emotional Control in America's History. Chi-
cago: University of Chicago Press.
Taylor, Charles
1979 Interpretation and the sciences of man. In Paul Rabinow & William M.
Sullivan, eds. Interpretive Social Science: A Reader, 25-72. Berkeley:
University of California Press. [1971]
Taylor, John and Robert E. MacLaury (eds.)
1995 Language and the Cognitive Construal of the World. Berlin: Mouton
de Gruyter.
Tolstoy, Leo
1970 Anna Karenina. Translated by L. & A. Maude. New York: W. W. Nor-
ton. [1918]
Travis, Catherine
1997 Kind, considerate, thoughtful: a semantic analysis. Lexikos 7:
130-152. (Afrilex Series 7).
1998 Omoiyari as a core Japanese value: Japanese-style empathy? In: Ange-
liki Athanasiadou and Elzbieta Tabakowska. Speaking of Emotions:
Conceptualisation and Expression, 55-82. Berlin: Mouton de Gruyter.
Van Geert, Paul
1995 Green, Red and Happiness: Towards a framework for understanding
emotion in universals. Culture and Psychology 1: 259-268.
Wierzbicka, Anna
1973 The semantic structure of words for emotions. In R. Jakobson, C. H.
van Schooneveld & D. S. Worth, eds. Slavic Poetics: Essays in Hon-
our ofKiril Taranovsky, 499-505. The Hague: Mouton.
1986 Human emotions: universal or culture-specific? American Anthropol-
ogist 88: 584-594.
1988 L'amour, la colere, la joie, 1'ennui: la semantique des emotions dans
une perspective transculturelle. Langages 89: 97-107.
1990a DuSa (~ soul), toska (~ yearning), sud'ba (= fate): Three key con-
cepts in Russian language and Russian culture. In: Zygmunt Saloni
Introduction 33
(ed). Metody formalne w opisie jgzykow stowianskich. [Formal
methods in the description of Slavic languages], 13-32. Bialystok:
Bialystok University Press.
1990b The semantics of emotions: Fear and its relatives in English. Austra-
lian Journal of Linguistics (Special issue on the semantics of emo-
tions) 10(2): 359-375.
1992 a Defining emotion concepts. Journal of Cognitive Science 16:
539-581.
1992b Semantics, Culture, and Cognition. Oxford: Oxford University Press.
1992c Talking about emotions: semantics, culture and cognition. Cognition
and Emotion (Special issue on basic emotions) 6(3/4): 285-319.
1993 Reading human faces: Emotion components and universal semantics.
Pragmatics and Cognition 1(1): 1-23.
1994 a Cognitive domains and the structure of the lexicon: The case of emo-
tions. In Lawrence A. Hirschfeld and Susan A. Gelman (eds.), Map-
ping the Mind: Domain Specificity in Cognition and Culture, 431-452.
Cambridge: Cambridge University Press.
1994b Emotion, language, and "cultural scripts". In Shinobu Kitayama and
Hazel Rose Markus (eds.), Emotion and Culture: Empirical Studies of
Mutual Influence, 130-198. Washington: American Psychological As-
sociation.
1995 a Everyday conceptions of emotion: a semantic perspective. In: J. A.
Russell et al. (eds.), 17-47.
1995 a Emotion and facial expression: a semantic perspective. Culture and
Psychology 1: 227-258.
1997 Understanding Cultures through their Key Words: English, Russian,
Polish, German, Japanese. New York: Oxford University Press.
1998 a 'Sadness' and 'anger' in Russian: the non-universality of the so-called
'basic human emotions'. In: Athanasiadou and Tabakowska (eds.),
3-28.
1998b Angst. Culture and Psychology 4 (2): 161-188.
1998c Russian Emotional Expression. Ethos 26(4): 456-483.
1999 Emotions across Languages and Cultures: Diversity and Universals.
Cambridge: Cambridge University Press.
2000 The semantics of human facial expressions. Pragmatics and Cogni-
tion 8(1): 142-183.
Wilkins, David P.
1986 Particle-clitics for criticism and complaint in Mparntwe Arrernte (Ar-
anda). Journal of Pragmatics 10: 575-596.
Ye, Zhengdao
2000 The language of emotions in Chinese: a study based on Hong Lou
Meng. MA thesis, Department of Linguistics, Australian National
University.
34 Anna Wierzbicka and Jean Harkins
1. Introduction
3. The hypotheses
The Amharic exponent for the conceptual primitive FEEL is found in the
verb (t9-)s9mma. Consider the following examples:
(3) a. minim a- yi- ss^mma- nfii- mm
nothing NEG- 3M.IMP- feel- lo- NEG
feel nothing.'
b. min yi- sssmma- h- al
what 3M.IMP- feel- 2M- IMP
'How do you feel?'
The Amharic word is used to express concepts which are undifferenti-
ated between "emotions" and "sensations", or feelings of "cognition" and
feelings of "the body" respectively (cf. Wierzbicka 1997: 18)3. Thus, the
following sentences can be felicitous replies to the question "how do you
feel"?
(4) a. ddssita yi- ssamma- fin- al
happiness SM.IMP- feel- i- IMP
feel happy.'
b. himam yi- ssdmma- fin- al
pain 3M.IMP- feel- I- IMP
feel pain.'
The Amharic verb for 'feel' cannot be used to translate the English con-
struction such as "I feel the chair" (say as in touching a chair while trying to
find your way in darkness):
38 Mengistu Amber be r
The common key element in (19) and (20) is that the event that is taking
place was rather unexpected or unforeseen.
It is important to note that some of the interjections are used to vocalise
both cognitively-based feelings and bodily feelings. For example, iffoy
((18 i) above) is an exclamation of 'relief either in the psychological or the
physical domain. Hence, one can say iffoy\ after the aversion of some im-
peding disaster in which case the experiencer is vocalising a psychological
relief- the anticlimax of a worrying state, so to speak. But one can also
say iffoy\ when taking a break after a physically exhausting activity.
Amharic has a wide range of terms which refer to positive and negative
emotions. In this section, we will sample some of the common emotion
terms by focusing on the (intuitively) prototypical emotions for each cat-
egory: roughly 'be happy' and 'be sad' respectively.
For ease of exposition the emotion predicates are presented in the (un-
marked) intransitive syntactic frame. Other possible grammatical strat-
egies and syntactic frames for encoding emotions are presented in 3.10.
(22)
X feels something
sometimes a person thinks something like this:
Testing emotional universals in Amharic 45
can only present a brief general commentary. I will examine the relation-
ship between the Amharic expressions which roughly correspond to 'sad-
ness', 'disappointment' and 'sympathy'.
However, these 'anger-like' verbs often occur with the manner adverbial
b9niddet (which consists of b9- 'with' and -niddet 'anger (noun)'):
(45) setiyy9wa b9-niddet arr9r9- cc
woman.DEF with-anger overcook.PF- 3F
'The woman is extremely angry.' (or 'fuming with anger')
These verbs are often used to express "suppressed" anger in a way which
is perhaps comparable to the English 'fuming'. Furthermore, the verbs ex-
press the level of the emotional intensity and their function is similar to that
of intensifiers such as 'very' and 'extremely'. This is supported by the fact
52 Mengistu Amberher
that some of the verbs, most notably te-k'at't'9te (lit. 'burn'), can be used
in the context of other emotions such as 'jealousy' and even 'love' as in
'that man is burning with jealousy' (= i.e., 'that man is extremely jealous of
someone').
The fact that "anger-like" verbs can be related to verbs which express
non-emotional events associated with the implementation of fire is also
common in English, as argued by Lakoff (1987: 383):
The folk theory of physiological effects, especially the part that emphasises HEAT,
forms the basis of the most general metaphor for anger; ANGER is HEAT. There are
two versions of this metaphor, one where the heat is applied to fluids, the other where
it is applied to solids. When it is applied to fluids, we get: ANGER is THE HEAT OF A
FLUID IN A CONTAINER. (...) When ANGER is HEAT is applied to solid, we get the ver-
sion ANGER IS FIRE (...)
The Amharic words comparable in meaning with 'smile' and 'cry' are
fygagateand atek'k '959 respectively. The former is typically expressed in
a complex verb which involves a non-verbal stem and an auxiliary (see
3.10 below for complex predicates).
(57) a. aster fygsg ah- co
A. smile.vn aux.pf.- 3r
'Aster smiled.'
b. aster alsk'k'dsa- co
A. cry.pf.- 3r
'Aster cried.'
56 Mengistu Amberber
Some emotions are described in Amharic through the use of externally ob-
servable bodily images or movements. For example, one can say some-
thing like 'his face became ashen' to express sudden shock or fright:
Testing emotional universals in Amharic 57
In 3.1. it is argued that the Amharic verb for the exponent FEEL has formal
properties which overlap with the verb for HEAR. For example, the follow-
ing interrogative sentence refers to either an auditory perception (HEAR) or
an emotional state (FEEL):
(64) min yi- ssdmma- h- al
what 3M.IMP- hear/feel- 2M- IMP
'What do you hear?'/'How (lit. What) do you feel?'
There is some evidence which suggests that the polysemy between sen-
sation and emotional predicates is not restricted to the auditory domain.
The following examples from the domains of taste and olfaction show
similar polysemy.
The internal parts of the body - especially the heart, the stomach, and the
intestine - can be used to express various types of emotions in Amharic. In
the following example, the heart's natural rhythm, expressed by the ono-
matopoeic tir tir, is manipulated to express something comparable to 'fear'
or 'anxiety' with the cognitive components 'someone thinks like this';
'something bad can happen to me now':
The verb tembbcfc 'abbao 'a is normally used to express the movement of
a liquid substance bounded within a container.
Pattern A Pattern B
SubjExp ObjExp SubjExp ObjExp
W9dd9d9 as-w9dd9d9
'love y' 'make love y' 'worry' 'make worry'
f9rra as-f9ira t9-d9SS9t9
'fear y' 'make fear y' 'be pleased' 'make pleased'
t'slla as-t'alla t9-k'ot't'a as-k'ot't'a
'hate y' 'make hate y' 'be angry' 'make angry'
It is important to note that the ObjExp verbs are formed by the causative
affix as-, and not by the causative affix a-. Thus, compare the sentences in
(78) below :
(78) a. hmma t9- d9ss9t9
L. DT. be.please.PF.3MS
'Lemma is pleased.'
62 Mengistu Amberber
"patient" argument of typical action predicates (e.g. 'hit') because the ex-
periencer has little or no control over the onset of the event.
There is some language-internal evidence which supports this claim. The
two most common emotional verbs which do not allow the experiencer to
occur in the accusative case are waddadd and t'9lla roughly corresponding
to 'love' and 'hate' respectively. This fact can be accounted for by assum-
ing that with emotions like 'love' and 'hate' the experiencer is relatively
more volitional compared to the experiencer with emotions like 'anxiety'
and 'shock' and thus not aligning itself with patient-like arguments.
Some emotion predicates take a prepositional oblique argument to ex-
press the 'target' of emotion. Thus, for example with the intransitive verb
k'9nna which roughly corresponds to 'be jealous', or 'be envious', the tar-
get of the emotion occurs in a prepositional phrase:
(85) aster ba- (i)htwa k'anna-
A. on- sister.poss.3p be.jealous.PF-3ps
'Aster is jealous of her sister.'
But it is also possible to cast the target of emotion as an object with the
accusative case and an obligatory agreement marker on the verb:
(86) aster ihitwa-n k'9nna- cc- ibb- at
A. Sister.poss.3p-Acc be.jealous.PF- 3p- ON- 3ro
'Aster is jealous of her sister.'
To summarise this section, the important point is that the experiencer ar-
gument of certain emotion predicates behaves morphologically as if it is
the object of the clause rather than the subject. This is accounted for by
proposing a semantic motivation: experiencers by virtue of having less
control over the onset of the (emotional) event align themselves with proto-
typical "patient" arguments rather than with prototypical "agent" argu-
ments. However, the alignment is partial, manifesting itself mainly in mor-
phology: the occurrence of an accusative case morpheme which is other-
wise restricted to definite object NPs of agentive verbs. As far as the overall
clausal structure is concerned there is no evidence which suggests that ex-
periencers are grammatical objects.
Testing emotional universals in Amharic 65
4. Conclusion
This study has presented a general overview of the semantics of some emo-
tional (linguistic) expressions in Amharic. The main purpose of the study
has been to test the set of hypotheses proposed by Wierzbicka (1997,1999)
regarding emotional universals. The description and analysis presented in
the study shows that emotional universals are borne out by the Amharic
data.
At the methodological level, the study shows that a thorough description
of emotional concepts can be accomplished only through the use of a
"meta" language rather than via a language-specific vocabulary. Although
English words are used in the study, the translations have been stated only
as rough approximations of the Amharic concepts. When there is a mis-
match between concepts expressed by the individual languages, the meta-
language can be used to clearly articulate in exactly what sense the words
in question are different in the relevant languages.
The study also shows that in investigating the expression of emotion con-
cepts one should pay close attention to the semantics of the grammar as a
whole. Thus, in Amharic the concept which is close to the English 'sym-
pathy' is expressed by grammatical means, i.e. attaching a prepositional
suffix on the verb which corresponds to 'sad/sorry'. It has been demon-
strated that by simply using a different prepositional suffix the meaning
comparable to 'be disappointed with' can be expressed. One of the crucial
differences between the semantic explications of the Amharic expressions
roughly corresponding to 'be sad/sorry for someone' and 'be disappointed
with someone' is the presence of the component 'it happened because that
person did something' in the latter.
The problem of polysemy has figured rather prominently in the present
study. The standard strategy employed in postulating polysemy is based on
the discovery of language-internal formal evidence (cf. Goddard and
Wierzbicka 1994: 446). Thus, it has been shown that the lexical exponent
of the semantic primitive FEEL and that of the mental predicate HEAR in
Amharic occur in a number of distinct morpho-syntactic environments.
Of course, some aspects of the analysis presented in the study are ten-
tative and only a detailed study of a wider range of data will provide a more
in-depth insight into the cognitive domain of emotions.
66 Mengistu Amberber
Notes
1. I would like to thank Anna Wierzbicka for her detailed comments and suggestions
on an earlier draft of this paper.
2. It is also spoken by a sizeable minority in Israel and Egypt (cf. Grimes 1996).
3. The material referred to here from Wierzbicka's (1997) paper is published in her
subsequent book on emotions (Wierzbicka 1999).
4. It is interesting to note that according to Johnson-Laird & Oatley (1989: 120), the
definition of sympathy incorporates the emotion sadness. They analyse sympathy as
a complex emotion expressing "sadness for (and understanding of) someone as a re-
sult of imagining oneself in their situation". In their model, sadness is one of five
basic emotion "modes" in terms of which a family of other complex emotions can be
elaborated.
5. In the semantic-field analysis of Johnson-Laird and Oatley (1989), the two emotion
concepts are treated as containing the basic emotion "sadness". They define "disap-
pointment" as "sadness caused by failure to achieve goal" (Johnson-Laird & Oatley
1989: 112).
6. For Darwin (1872/1965) the important element in "shyness", "shame", and "mod-
esty" is "self-attention". Although Darwin's discussion of these emotions is in the
context of describing the biological properties of "blushing" what he has to say
about the mental states that trigger the feelings in question is relevant to our purpose.
He argues that the expectation of "praise" and "admiration" can trigger blushing in
as much as the expectation of "blame" or "disapprobation".
7. The relationship between "smile" and "laughter" is by no means simple. Smiling
can be an onset of laughing, or as Darwin (1872/1965: 196) put it, "smiling... grad-
uates into laughter". But there are clear differences as well. For example, if you see a
friend you are likely to smile but not necessarily laugh.
References
Darwin, Charles
1872/1965 The Expression of the Emotions in Man and Animals. Chicago: The
University of Chicago Press.
Goddard, Cliff
1996 The "social emotions" of Malay (Bahasa Melayu). Ethos 24(3):
426-464.
Goddard, Cliff and Anna Wierzbicka (eds.)
1994 Semantic and Lexical Universals: Theory and Empirical Findings.
Amsterdam: John Benjamins.
Grimes, Barbara
1996 Ethnologue. 13th edn. Dallas: Summer Institute of Linguistics.
Testing emotional universals in Amharic 67
1. Introduction
To compare meanings one has to be able to state them. To state the meaning of a
word, an expression or a construction, one needs a semantic metalanguage. To com-
pare meanings expressed in different languages and different cultures, one needs a
semantic metalanguage independent, in essence, of any particular language or cul-
ture - and yet accessible and open to interpretation through any language.
(Wierzbickal991:6)
To study different cultures in their culture-specific features, we need a universal per-
spective; 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 ...We need to find shared terms; that is, universal concepts. I
suggest 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 languages. (Wierzbicka 1991: 9-10)
5. A construction in which the inalienable nouns le- and ka- are added
immediately following the verb. Le- can be used for both pleasant and un-
pleasant experiences, whereas ka- is used only for unpleasant ones.
7. The expression -ur koipa 'put enemy to' which functions to express
'malice/hate' and the desire to have nothing to do with someone, as well as
simple uncooperativeness, when it collocates with personal nouns. With
event expressions like mbulu sananrjana/pakaamrjana/zigzikyana
'bad/false/destructive behaviour', it implies a negative evaluation of such
behaviour and a refusal to do it. An explication of its meanings is given
below:
sire, life, soul', orayin 'eye' or mach 'wind, breath, spirit'. For example, in
Psalm 73:21 the prosperity of the wicked leads to the following feelings in
the psalmist : "My heart was soured/embittered and my kidneys were
pierced." Earlier in the psalm, in verse 16, he talks about trying to under-
stand this enigma and says it was "burdensome/laborious/oppressive to his
eyes." In New Testament Greek emotions like compassion, longing, and re-
freshment are attributed to a person's splangxnon 'intestines', while lust,
lack of generosity, and thinking and understanding or the lack thereof are
attributed to a person's ophthalmos 'eye'. A Mbula speaker finds it
relatively easy to identify with body image expressions like those in He-
brew and Greek, since his/her own language also abounds in such ex-
pressions.
Of course, the English language, too, has its own collection of body
image expressions like butterflies in the stomach and broken heart. But
much greater reliance is placed upon nouns and adjectives encoding emo-
tions like anxious, anxiety, hurt, offended, offence, happiness, happy, joy,
joyful, peace, and peaceful, which do not require a mediating body part.
One of the very important differences between English speakers and Mbula
speakers is that when Mbula speakers want to talk about emotion, they do
not generally have such lexical items to draw on. There are, relatively
speaking, very few lexical items in the language which, on their own, en-
code emotional or physical experiences.
In the preceding section, most of the listings of lexical items and con-
structions having an inherently experiential meaning are exhaustive. It will
be noted that the range of experiences encoded by these lexical items with
inherently experiential meanings is somewhat limited. For example, there
are no specialised words for notions like 'anger', 'love', 'joy', or 'disap-
pointment' . And yet, living among Mbula speakers over the past seventeen
years, I can vouch for the fact that they certainly seem to experience feel-
ings which strike me, albeit I am a cultural outsider, as bearing a consider-
able resemblance to what I understand such terms to mean. But for them, in
order to talk about these notions, they must use body image expressions
which localise the experience. There is no other option in their language.
So for them, most experiences, whether physical sensations or emotions,
are mediated via and localised in various parts of the body.
When Mbula speakers want to talk about their emotions or the physical
sensations they are experiencing, the body part terms which most fre-
quently occur are: mata- 'eye', lele- 'inside(s), feelings', kete- 'liver', ni-
76 Robert D. Bugenhagen
'genitals, being', kuli- 'skin', and kopo- 'stomach'. Less commonly used
are: kwo- 'mouth', taiga- 'ear', rjgure- 'throat', and zuru- 'lips'. (There are
more specific terms for male and female genitals, but these are largely
taboo words. Instead, the more generic ni- is customarily used.)
With regard to the mapping of arguments and syntax in such construc-
tions, the Experiencer is encoded as the ostensible genitive of the body part
noun, and the experiential stimulus is encoded as the object of the Oblique
preposition pa. However, as Bugenhagen (1995: 256-257) notes, the syn-
tactic structure of such constructions is not:
[NPGENiTiVE Body PartN]suBJECTNP [Verb [pa NP]]
with the body part being syntactically incorporated with the verb into a
kind of complex predicate. This is evidenced by the fact that certain ad-
verbs which normally occur between the Subject and the predicate do not
occur between the body part noun and the verb, but instead occur between
the NP encoding the "possessor" of the body part and the body part noun
itself.
In the remainder of this paper, the principal categories of experience as-
sociated with different body parts will be delineated. In many instances,
there is no clear English equivalent for the emotion expressed, so fuller
semantic explications using the natural semantic metalanguage are given.
Due to limitations of space, examples given below in the various listings
are representative only, and not completely exhaustive.
person', zog mataana men 'the sun is very bright'.) The least extended ex-
periential meanings are observed in expressions like mata- yenrjeene 'eye
sleepy - to be tired and find it difficult to keep one's eyes open', mata- ikam
pak 'eye does opening - to wake up and open one's eyes', mata- iyoyou
'eye hurts - feel pain in one's eye', and in expressions having to do with
seeing), like those below:
(11) Ymata-rikrik
(lit. eye move.back.and forth)
'look around'
(12) mata- ito
(lit. eye follows)
'look at something which is moving away'
(13) mata- ila I imar lisu I ise I imiili
(lit. eye goes/comes/descends /ascends /returns)
'look ahead at something'/'look at the speaker'/'look
down' /' look up' /' look back'
(14) mata-ikam Z
(lit. eye do/get/receive)
'involuntarily happen to see something'
something happens in someone Y's eyes
because of this, sees something
not because wants to see it
(15) mata- mbeij
(lit. eye night)
'to not see something which one should be able to see' (used to
scold someone)
someone does not see something.
I think: if wanted to see this, could see this
because of this, I think does something bad
I say is like someone who cannot see things because it is night3
(16) mata-pisana
(lit. eye-closed.one)
'blind'
78 Robert D. Bugenhagen
The next class of experiential examples with mata- have to do with life
and consciousness, the capacity to think and know.
when these people think like this they feel something (bad)
feels something like this
(There is another sense for this expression, which appears to be essen-
tially synonymous with that ofmata- zugutgut: 'be on the verge of fainting
and becoming unconscious'.)
Unlike mata- , lele- is rarely used to express any sort of physical sen-
sation or experience. The only instance I know of is lele- ibayou 'insides
be.hot - to feel hot (and thirsty)' (because of being out in the sun and ex-
erting oneself. Having a temperature/fever is expressed differently as kuli-
ibayou 'skin be.hot'.
There is also a handful of body image expressions where lele- appears to
have a cognitive function. These are given below:
Apart from the few physical and cognitive uses given above, lele- is used
to express desire and a wide range of emotional states, the more common
of which are listed below.
Lele- isaana and lele- ipata differ in: 1) the intensity of the feeling (with
lele- ipata expressing greater intensity), 2) the Experiencer's assessment of
whether or not someone is culpable for having caused the experience, and
3) whether or not the Experiencer has the possibility of doing anything
about the situation. If someone of lesser status steals a person's areca nuts
or publicly disagrees with him/her, both of which are viewed as personal
slights, the expression lele- isaana would be more favoured. If a relative
dies, lele- ipata would be used, because the situation is something very bad
and not really any particular person's fault, and there is nothing one can do
to undo the situation.
92 Robert D. Bugenhagen
Lele- sapa and ni- sapa appear to be completely synonymous. The kinds
of things done in response to this feeling are: 1) does not look Z directly
in the eye, 2) says as little to Z as possible, and 3) does not smile. Lele-
sapa is the expression used when one sees from another's actions that they
are upset with someone, but does not know the reason why.
This expression is the most generic way of all of expressing that someone
is feeling emotion. It indicates nothing whatsoever about the kind of emo-
tion being experienced.
(71) YiraoZlele-
(lit. be.adequate Z's insides)
'something satisfies someone Z'
(72) a. Y lele- imet (kat) pa Z:event NP
(lit. insides snap.in.two (really) Oblique Z)
b. lele- imet (kat) be Z:S
(lit. insides snap.in.two (really) NAF Z)
'someone becomes determined to do something Z, regard-
less of any possible negative consequences doing it might
have for Y'
sometimes people think like this:
I want to do something
I know: if I do this, something bad can happen to me
I don't want not to do it because of this
when these people think like this, they feel something bad
inside because of this
when Y thinks about , feels something like this
With regard to the component "Y feels something (good/bad) inside", it
seems that the explications should make explicit the localisation of the ex-
perience implied by the use of the noun lele-. Explications for the other
body image expressions will also frequently make use of such a compo-
nent, with "inside" replaced by the corresponding body part term. Such ex-
plications make no claim about the validity according to western science of
associating an emotion with a particular body part. There may or may not
be a scientific basis for such associations. That is not the issue here. What
we are seeking to do is to reflect the Mbula conceptualisation of emotion,
as revealed by the way they talk about it.
The five expressions lele- pa, lele- ambai, lele- isaana, lele- ipata, and
lele- iluumu are certainly among the most frequently occurring body image
emotional expressions in the language and correspond to a wide variety of
emotion terms in English. Of note is the fact that lele- ambai spans a range
of emotions which English sub-divides much more: 'happy, pleased, ap-
proving, grateful, content, at peace, free from worry, care, sorrow, and
Emotions and the nature of persons in Mbula 95
3.3. ^^-'chest/liver'
With these comments about lele-, we now turn our attention to another ex-
tremely productive body-part term: kete- 'liver, chest, flat-shaped part of
something'. (The semantic invariant of the physical referents of kete- is
that they are parts of something which are flat in shape. For example,
nama- keteene is the flat part of one's hand or palm, kumbu- keteene is the
flat part of one's foot or sole, kar keteene is a flat area in the centre of a vil-
lage, lele keteenerjana is a flat plain.) Unlike mata- and lele-, body image
expressions containing kete- never express pure cognition. The term is
96 Robert D. Bugenhagen
used to express some physical states like hunger and different ways of
breathing: kete- bilbil 'chest/liver empty -be very hungry', kete- imaraaza
'chest/liver tear - be so hungry as to feel sharp physical pain in one's stom-
ach', kete- isalaklaaga 'be nauseated', kete- kutkut 'chest/liver be.pulsat-
ing - be breathing rapidly (due to exhaustion, pain, or fear)', kete- ipas
'chest/liver is removed - be out of breath', kete- ikam kug kek 'has ceased
to breathe - dead', kete- nono 'chest/liver be.hunting - heart beats rapidly'.
The preeminent emotional function of kete- is to express rash, impetuous
responses which are not well thought through and strong emotions like
anger.
The expression -pas kete- is used to talk about someone causing some-
one else to become angry and focuses upon the transition from a state of
not being angry to one of being angry.
The term ni- is somewhat difficult to define. It is used to talk about physical
sensations which are not localised in any one particular part of the body, as
in the following examples:
(81) ni- imbeeze being is.smooth 'feeling fit or fresh'
ni- itekteege being is.holding 'shiver with cold'
ni- ambai being is.good 'healthy'
ni- iyoyou being hurts 'physically suffer, ache all over'
ni- isaana being deteriorates 'be exhausted, worn out'
ni- ipeele being opens 'feel fully awake'
(A near equivalent of ni- isaana would be mbura- imap 'strength be fin-
ished'.)
Ni- is not, however, equivalent to body in English. For example, a dead
body is not referred to as someone's ni-. When talking about burying
people, the person being buried is referred to by name, kin-term, or via a
pronoun. A person's ni- is never said to be buried. If one wants to talk about
a person's body going into the ground and decaying, it is their kuli- (also
'skin') which is said to undergo the process of decay. A second sense for
ni- is that it functions as a euphemism for male and female sexual organs.
And a third sense is that of a part of something which is long and narrow:
ndapndap niini 'pestle', yok niini 'watercourse', lolo niini 'lightning bolt',
ndeete niini 'step/rung of a ladder'.
In the expression ni- ikilaalalikanamaala 'being recognises/senses', it
functions to encode perception via some means other than the five senses.
In emotional expressions, it often resembles a negative counterpart to lele-,
functioning to encode dislike. But there is also one very important ex-
100 Robert D. Bugenhagen
pression encoding a positive emotion: ni- ise. Since ni- is specifically used
to indicate feelings that are non-localised, it will be observed that in the fol-
lowing explications the component "feels something in a part of F" is ab-
sent.
(82) F ni- pa Z som
(lit. being Oblique Z not)
does not want Z'
(83) F ni- gesges pa
(lit. being tedious Oblique Z)
finds Z tedious, irksome'
sometimes people think like this:
something happened for a long time/I did something
for a long time
before now, if this thing happened/I did this thing,
I did not feel something bad
now, if this thing happens, I will feel something bad
because of this, I don't want this thing to happen after now
when these people think like this, they feel something bad
when thinks about something , feels something like this
(84) -sapaZ
(lit. being non-referential Oblique Z)
See explication for lele- sa pa.
(85) F ni- sanaana pa Z
(lit. being sin Oblique Z)
= F ni- puliizi pa Z
(lit. being [?] Oblique Z)
'upset with'
sometimes people think like this:
someone did something bad to me
because of this I feel something bad
I want this person to know:
when I think about this person, I feel something bad
when these people think like this, they do something bad
because of this, these people do some things
when thinks about someone , feels something like this
Emotions and the nature of persons in Mbula 101
The kinds of things done are: 1) does not look Z directly in the eye,
2) says as little to Z as possible, 3) does not smile, and 4) when is in
Z's presence, does not come close to Z.
(86) ni- ise (pa Z)
(lit. being ascends Oblique Z)
'be visibly exultant over one's good fortune'
sometimes people think like this:
something good happened to me
this thing did not happen to other people
because of this, I am not like other people now
when these people think like this, they feel something very good
these people do some things
because of this, if other people see these people,
they know that these people feel like this
when thinks about something , feels something like this
The last expression, ni- ise, is a difficult one for which to find an English
equivalent. It occasionally borders on pride, feeling that one is superior to
others. But it is also used when someone just feels happy because of some-
thing good happening to them which has not happened to other people. The
key concept is that the person somehow feels "special" and others can see it
from the way the person behaves, because he smiles, is louder and more
forceful in speech, walks more erectly and generally acts more vigorously
than people normally do. The triggering events can be something as simple
as wearing a new outfit of clothes, or getting a new lawn mower, or being
on a winning soccer team. But the term can also apply to Christians who
feel grateful and happy to have been touched by God's grace.
The next term to be considered is kuli- 'skin, bark, surface'. It is used to ex-
press various physical experiences in expressions like: kuli- isu 'skin de-
scends lose weight, skin becomes wrinkled due to aging', kuli- imalai
'skin wilt - feel tired', kuli- ila muriini 'skin goes.to its.place - feel rested
and refreshed', kuli- ire yoyourjana 'skin sees pain - physically suffer',
kuli- iswidit 'skin be.burnt - be sunburned', and kuli- isarakrak 'skin
102 Robert D. Bugenhagen
shivers - have shivers up one's spine (in response to being cold or en-
countering the supernatural)'. Kuli- is probably the nearest equivalent to
the English concept 'body'. When someone is buried and their physical
body decays, it is their 'skin' which is said to rot. But unlike body, in Mbula
one's skin is not said to be buried; it is the person who is buried. The physi-
cal resurrection which, according to Christian belief, will occur on the last
day is described as tamarjga kulindi munrjaana 'We (INC) get up, our skins
whole'. And food that keeps one healthy is characterised as ikis kulindi 'it
holds our (INC) skin'.
It is possible for the skin to function as an organ of perception, as in the
following example:
kuli- is being used in a wider sense than just 'skin', but 'skin' is included in
the prototype proposed here.
(91) kuli- mburaana pa Z som
(lit. skin its-strength Oblique Z NEG)
feels something in 's skin
because of this, does not want Z
(92) a. Y kuli- imet (kat) pa Z:NP
(lit. skin snaps.in.two (really) Oblique Z)
b. kuli imet (kat) be Z:S
(lit. skin snaps.in.two (really) NAF Z)
1. 'someone becomes boldly determined to do something Z,
regardless of any possible negative consequences doing it
might have for Y'
In this first sense, this and the corresponding lele- imet (kat) appear to be,
for all intents and purposes, synonymous.
2. 'become accustomed to someone and no longer feel uncom-
fortable in their presence'
sometimes people think like this about someone:
before now, I did not know this person
because of this, when I was in the same place as this person,
I felt something bad in my skin
I did not want to be in the same place where this person is
now I know this person
because of this, if I am in the same place as this person now,
I don't feel anything bad in my skin
when these people think like this, they feel something
when Y thinks about someone , feels something like this
when these people think like this, they feel something bad
when thinks about something , feels something like this
In addition to these emotional uses, taiga- occurs in various expressions
indicating the physical state of one's ears, such as taiga- imun 'ear
be.closed - deaf, taiga- iyoyou 'ear hurts - ear ache', taiga- ikam 'ear do/
get - happen to overhear something', taiga- ikam kik 'ear does ring-
ing - ringing in ears that hinders one's hearing'.
(100) taiga- iluumu
(lit. ear be.cool)
'not have to hear something anymore'
sometimes people think like this:
before now, other people said things about something many
times
I don't want to hear these things any more
when these people think like this, they feel something bad when
they hear other people say these things
now these people do not hear other people say these things
because of this, now these people do not feel something bad
feels something like this
(The source of the annoyance could be an unpleasant, irritating noise, or
it could be things people were saying which were unpleasant.)
(101) taiga- ituntun pa
(lit. ear itches Oblique Z)
'always wanting to hear about something'
if people say things about something , wants to hear these
things
(102) ikam ggure-
(lit. does/gets throat)
= izuuru ggure-
(lit. pushes throat)
sometimes if people eat/drink something,
these people feel something good in their throat
because of this, these people want to eat/drink this thing
feels something like this when thinks about
108 Robert D. Bugenhagen
This now completes the survey of the most common body image ex-
pressions in Mbula. The explications for some of them, like kete- malmal
'anger', mata- berber 'covet', -ur mata- pa 'hope', lele- isaana 'feel bad',
lele- imbukmbuk 'anxious', lele- ikam uraata 'be emotionally moved',
ni-llele- sa pa 'be upset with someone' seem to map easily into corre-
sponding English emotional terms. In other cases like ni- ise, as would be
expected, it is difficult to find precise English equivalents, because they in-
dicate a combination of semantic components unlike that of any English
Emotions and the nature of persons in Mbula 109
9. Time of response:
someone does something/wants to do something now/after now
(e.g. mata- putput versus mata- pambaard)
10. Inceptive aspect:
a) before now someone did not want to do something; now this person
wants to do something
b) before now someone did not feel something; now this person feels
something
(e.g. lele- iur pa/be, lele- ipuru, lele- imagga, kete- ise, ipas kete-,
kuli-imet (kat) pal be)
11. Completive aspect:
a) before now someone wanted to do something; now this person does
not want to do this
b) before now someone felt something; now this person does not feel
this
(e.g. lele- iluumu pa, lele- imiili. taiga- iluumu)
12. Differences in intensity:
someone feels something (very) good/bad;
someone does something (very) good/bad
(e.g. mata- putput, kete- ise, kete- malmal, kete- ibayou)
13. Evaluation:
I think this is good/bad
(e.g. mata- pot, mata- sigsig, kete- guruk, kete- pitpit)
The data presented in this paper amply illustrate the importance of the body
in Mbula speakers' conception of emotion. In many cases it is impossible
for them to talk about a particular emotion without making mention of
some part of their body. Sometimes these parts are highly generic like lele-
'insides' and ni- 'being'. Other times, the parts are much more specific like
Emotions and the nature of persons in Mbula 111
mata- 'eye', kete- 'liver/chest', and kopo- 'stomach'. With regard to the de-
bate over whether emotion is physiological or cultural and conceptual, the
Mbula lexicographical evidence would suggest the answer: both.
One interesting question is: What do these kinds of expressions imply
about the Mbula concept of the person? People with a western background
have variously viewed human beings as consisting of: 1) mind plus body,
2) body only (which includes the brain), 3) soul plus body, 4) spirit plus
soul plus body, or 5) mind plus will plus emotions plus body (inter
alia) - but how do Mbula speakers view the human person?
The body image expressions presented in this paper do not support the
neat divisions of people into various discrete components with clearly de-
fined functions as the western traditions seem to do. For example, if mata-
can refer to one's physical eyes which see or experience pain, or experi-
ences arising because of having seen something, or the part of one which
thinks and even feels, what is the basis for a clear division between mind
and body? It would be difficult to identify any one of the body parts de-
scribed in this paper with the English terms mind, will, spirit, brain, or
body. (There is a term in Mbula, rjgora, which corresponds to the physical
brain organ. But it plays virtually no role in expressing thinking, feeling, or
desire.) Along different lines, all of the body parts presented in this paper
can be used to encode purely physical sensations, as well as feelings/de-
sires with associated conceptualisations; i.e. emotions. And a wide range
of body parts are used in body image expressions to indicate some sort of
desire or the lack thereof. With regard to the division of a human being into
a material and non-material part, at death the person is buried and their
'skin' decays, but their kunu- 'shadow, image, picture, ghost' survives and
can appear to people, bringing them assistance or causing them mental or
physical harm. (Kunu- is the form having an inalienable genitive; a related
form kon has an alienable genitive and is restricted to the meaning 'ghost'.)
Those Mbula speakers who are Christians believe that when they die it is
their kunu- that goes to be with God in heaven. It is a striking fact, however,
that unlike 'spirit' in Hebrew, Greek, and English, kunu- plays virtually no
role in expressing emotion.
These body image expressions are, furthermore, used to talk about the
emotions of God, angels, and dead people's spirits, despite their lack of a
physical body. The explications, as formulated here, allow for this, since
they present a prototype of how people feel, and then attribute a feeling
"something like this" to the individual experiencing the emotion. This
112 Robert D. Bugenhagen
Notes
1. The Mbula language is an Austronesian language spoken by some 3500 people liv-
ing on the eastern half of Umboi Island and on Sakar Island in the Morobe Province
of Papua New Guinea. The full classification of the language according to Ross
(1988: 20, 122) is: Austronesian, Malayo-Polynesian, Central-Eastern Malayo-
Polynesian, Eastern Malayo-Polynesian, Oceanic, Western-Oceanic, North New
Guinea Cluster, Ngero-Vitiaz Family, Vitiaz linkage. The language exhibits Sub-
ject-Object-Verb word order and nominative-accusative morphological marking
and is prepositional. Ordering of constituents within the noun phrase is basically:
1) inalienable genitive/topicalised alienable genitive, 2) Head Noun, 3) Attributive
Modifiers, 4) Quantifiers, 5) Alienable Genitives, 6) Relative Clause 7) Demon-
strative. Pronouns and verbal affixation distinguish: 1) first, second, and third per-
son, 2) singular versus plural, and 3) (for first person non-singular) whether or not
the hearer is included. The pronouns also exhibit dual forms, which the verbal mor-
phology doesn't. There is no grammatical category of tense. Use is made of tem-
poral adverbials to fix the temporal deictic centre whenever it changes, but most sen-
tences have no indication whatsoever of the time of occurrence of the situation en-
coded. Most verbs are inflected with a set of subject prefixes. This is indicated in the
paper by a preceding hyphen. A subset of the nouns is obligatorily inflected with a
set of genitive suffixes. These are cited in this paper with a following hyphen. Non-
transparent abbreviations used in this paper are: Accusative, CSB cosubordinating
conjunction, DEMonstrative, EXC speaker excluded, GENitive, INC speaker in-
cluded, LOCative, NMZ nominalisation, NOMinative, OBLique, PASSive, PERFect,
RECX Recipient-Experiencer, REDuplication. I would like to very gratefully ac-
knowledge the help of David Aibike, Lukas Aibul, and many other Mbula-speaking
friends, who have spent many hours opening up for me a small window into their
emotional world.
2. In the ensuing semantic explications, use is largely made of a restricted lexicon con-
taining the following items: I, YOU, SOMEONE, PEOPLE/PERSON, SOMETHING/
THING, WORD, THINK, SAY, KNOW, FEEL, WANT, SEE, HEAR, THIS, THE
SAME, OTHER, ONE, TWO, MANY/MUCH, SOME, ALL, DO, HAPPEN, MOVE,
BE (THERE IS/ARE), LIVE, DIE, GOOD, BAD, BIG, SMALL, NOT, IF, IF...
WOULD, MAYBE, CAN, BECAUSE, VERY, MORE, WHEN/TIME, WHERE/
PLACE, AFTER, BEFORE, A LONG TIME, A SHORT TIME, FOR SOME TIME,
NOW, HERE, BELOW, ABOVE, FAR, NEAR, SIDE, INSIDE, PART OF, KIND OF,
LIKE. However, occasionally some other items are used in order to enhance the
readability of the explications.
Emotions and the nature of persons in Mbula 113
3. The phrasing of this component is not strictly NSM, but it is sufficient for the present
purposes.
4. Again, the phrasing of this component is not strictly NSM, but it is sufficient for the
present purposes.
References
Bugenhagen, Robert D.
1990 Experiential constructions in Mangap-Mbula. Australian Journal of
Linguistics 10(2): 183-215.
1994 The exponents of semantic primitives in Mangap-Mbula. In: Cliff
Goddard and Anna Wierzbicka (eds.), Semantic and Lexical Univers-
als, 87-108. Amsterdam: John Benjamins.
1995 A Grammar of Mangap-Mbula: An Austronesian Language of Papua
New Guinea. (Pacific Linguistics C-101.) Canberra: Australian
National University Press.
In press. The syntax of universal semantic primes in Mbula. In: Cliff Goddard
and Anna Wierzbicka (eds.), Meaning and Universal Grammar. Am-
sterdam: John Benjamins.
Darwin, Charles
1872 The Expression of the Emotions in Man and Animals. London: Murray.
Ekman, Paul, E. R. Sorenson and W. V. Friesen
1969 Pancultural elements in facial displays of emotion. Science 164:
86-88.
Geertz, Clifford
1973 The Interpretation of Cultures: Selected Essays. New York: Basic
Books.
Goddard, Cliff and Anna Wierzbicka (eds.)
1994 Semantic and Lexical Universals. Amsterdam: John Benjamins.
Harre, Rom (ed.)
1986 The Social Construction of Emotions. Oxford: Blackwell.
James, William
1884 What is an emotion? Mind 9: 188-205.
Leavitt, John
1996 Meaning and feeling in the anthropology of emotion. American Eth-
nologist 23(3): 514-539.
Leibniz, Gottfried Wilhelm
1903 Opuscules et fragments inedits de Leibniz. Paris: Louis Couturat. [Re-
printed 1961, Hildesheim: Georg Olms.]
114 Robert D. Bugenhagen
Lutz, Catherine
1982 The domain of emotion words on Ifaluk. American Ethnologist 9(1):
113-128.
1987 Goals, events, and understanding in Ifaluk emotion theory. In: Dorothy
Holland and Naomi Quinn (eds.), Cultural Models in Language and
Thought, 290-312. Cambridge: Cambridge University Press.
McElhanon, Keith E.
1977 Body image idioms in Irianese and Papua New Guinean languages.
Irian 6: 3-27.
Rosaldo, Michelle
1980 Knowledge and Passion: Ilongot Notions of Self and Social Life. Cam-
bridge: Cambridge University Press.
Ross, Malcolm D.
1988 Proto Oceanic and the Austronesian Languages of Western Melanesia.
(Pacific Linguistics C-98.) Canberra: Australian National University
Press.
Wierzbicka, Anna
1980 Lingua Mentalis: The Semantics of Natural Language. New York:
Academic Press.
1990 The semantics of emotions: Fear and its relatives in English. Austra-
lian Journal of Linguistics 10: 359-375.
1991 Cross-Cultural Pragmatics. The Semantics of Human Interaction.
Berlin: Mouton de Gruyter.
Zimbardo, Philip and Floyd L. Ruch
1976 Psychology and Life. Glenview, Illinois: Scott, Foresman and Com-
pany.
Why Germans don't feel "anger 55
Uwe Durst
1. Introduction1
About two thousand years ago, Seneca (1970: 106-109) wrote:
But you have only to behold the aspect of those possessed by anger to know that they
are insane. For as the marks of a madman are unmistakable - a bold and threatening
mien, a gloomy brow, a fierce expression, a hurried step, restless hands, an altered
colour, a quick and more violent breathing - so likewise are the marks of the angry
man; his eyes blaze and sparkle, his whole face is crimson with the blood that surges
from the lowest depths of the heart, his lips quiver, his teeth are clenched, his hair
bristles and stands on end, his breathing is forced and harsh, his joints crack from
writhing, he groans and bellows, bursts out into speech with scarcely intelligible
words, strikes his hands together continually, and stamps the ground with his feet; his
whole body is excited and "performs great angry threats"; ...2
For centuries, if not millennia, people have attempted to explain the phe-
nomenon of anger as well as other emotions, pointing to typical character-
istics of behaviour, causal events, and physical incidents in a more or less
scientific way. It could easily be shown that most of the symptoms de-
scribed by Seneca (and other philosophers, doctors, and psychologists)
would be attributed by people around the world to an angry or choleric per-
son rather than a phlegmatic one. And, given a set of five or six "basic"
emotion concepts, these symptoms would undoubtedly be attached to the
category of "anger" rather than to categories such as "fear", "sadness",
"happiness", "disgust", or "desire".
However, to carry out such an experiment, this set of emotion words
would have to be translated into the respective languages, so this test would
really tell us nothing about the psychological reality of universal emotions
or emotion concepts. It could only show the trivial fact that each of these
languages has at least one word that can be used, among other possible ap-
plications, to name this kind of behaviour. But, provided that we could find
such a word in all languages around the world, how could we be sure of
what it really means?
116 Uwe Durst
(1) Als Sabeth es endlich entdeckte, streckte sie die Zunge heraus; ich
filmte sie mit der gestreckten Zunge, bis sie, zornig ohne Spa,
mich regelrecht anschnauzte. [MK1/LFH.OOOOO (Frisch, Homo
Faber. 104)]
When Sabeth finally noticed, she put her tongue out; I filmed her
with her tongue out, until, really angry, she bawled me out prop-
erly. [Frisch 1974: 90]
(2) Irgendeiner hatte ihn sogar zum Minister machen wollen, aber er
war wtend geworden und hatte gesagt: "Ich bin Lehrer, und ich
mchte wieder Lehrer sein". [MK1/LBC.OOOOO (Bll, Ansichten
eines Clowns: 48)]
Someone had even wanted to make him a Minister, but he had got
very angry and said: "I am a teacher, and I want to be a teacher
again." [Boll 1988: 31]
(3) Ich hatte einmal eine ziemlich lange Nummer "der General" ein-
studiert, lange daran gearbeitet, und als ich sie auffhrte, wurde es
das, was man in unseren Kreisen einen Erfolg nennt: d.h. die rich-
tigen Leute lachten, und die richtigen rgerten sich. [MK1/
LBC.OOOOO (Bll, Ansichten eines Clowns: 264)]
I had once spent a lot of time rehearsing a fairly long number
called "The General," and it turned out to be what is known in our
circles as a success: that is, the right people laughed, and the right
people were angry. [Boll 1988: 215]
(4) Ich machte die Augen auf und sah ihn an. Er war bse.
[MK1/LBC.OOOOO (Bll, Ansichten eines Clowns: 197)]
I opened my eyes and looked at him. He was angry. [Boll 1988:
159]
In (l)-(4) angry was used by the translators to convey four different Ger-
man expressions: the adjectives zornig (derived from Zorn), wtend (re-
lated to Wut), and bse (lit.: 'bad'), and the reflexive verb form sich rgern
(which corresponds to rger). In fact, however, angry is by no means the
only word that can be used to translate one of these four German ex-
pressions into English. Nor is anger the only noun that fits, for example, for
the German word Zorn:
118 Uwe Durst
(5) Ivy starrte mich an, dann schleuderte sie pltzlich ihren Mantel
irgendwohin ins Zimmer, stampfend, auer sich vor Zorn ...
[MK1/LFH.OOOOO (Frisch, Homo Faber. 73)]
Ivy stared at me, then she suddenly threw her coat across the room
and stamped her foot, beside herself with rage. [Frisch 1974: 65]
In (5) Zorn is conveyed by rage, which, in turn, can also be used in a con-
text where the German author uses the noun Wut:
(6) Ich heulte vor Wut, zertrampelte die Schiescheiben und schrie
den Jungen drauen, die mich bewachten, immer wieder zu: "Ihr
Nazischweine". [MK1/LBC.OOOOO (Bll, Ansichten eines Clowns:
32)]
I howled with rage, trampled on the targets, and kept on shouting
at the boys outside who were standing guard over me: "You Nazi
swine!" [Boll 1988: 19]
In sum, there is no German word that perfectly matches the English word
anger, and none of the German words given in the examples above has a
clear counterpart in English. Each of the German words has a meaning
which is somewhat different from each of the English words, and there is
no evidence for the "basicness" of one of these words. To grasp their mean-
ings and to be able to compare them and to define them, we have to submit
each word to detailed semantic analyses.
In what follows, the lexical items rger, Wut, and Zorn, which constitute
the most frequent and most common 'anger' words in German, will be sub-
jected to semantic and comparative investigation. The analysis is given
within the theoretical framework of the NSM approach to semantics, which
has turned out to be a most useful way to gain suitable results for this task
(see Wierzbicka 1996, 1999). But before we enter into details, we should
have a brief look at some methodological problems in the work on German
emotion terms and their semantics.
2. Methodological issues
Russell states that "the meaning of each such [emotion] term, the con-
cept it expresses, is a script" (1991: 39; original emphasis). But if this is
true, what, then, are the meanings of similar words, such as rage, fury, ag-
gression, and the like? Couldn't the five components posited by Russell be
applied to these words, too? Thus, what Russell describes may apply to an
artificial supercategory called "anger" alongside four other "basic" emo-
tions ("fear", "sadness", "happiness", "disgust") but not to the meaning of
the everyday word anger. This is not to say that the positing of such cat-
egories may not be useful for the research on the psychology of emotions.
But it should be kept in view that these categories are artificial constructs,
and that their labels belong to the technical terminology of a particular dis-
cipline - a terminology worthy of linguistic investigation in its own right.
While Russell's "anger script" is too extensive from the viewpoint of ex-
tensional semantics, it is too specific with regard to the intensional seman-
tics of anger. For example, is an emotion called anger normally caused by
an "intentional offense"? Apparently, people often get angry at someone
whom they never met, as in the case of political anger [Stearns and Stearns
1986: 25], and a person who is angry at him/herself certainly did not inten-
tionally offend him/herself in a harmful way before. The "desire for retri-
bution" in line 4 seems to belong to words like revenge or vengeance, and
what is described in line 5 might be a component of the prototype of the
phrase angry outburst rather than of anger.
In the NSM approach to semantics, emotion words have been described
within a prototypical framework from the very beginning (cf. Wierzbicka
1973). At the present stage, the most general frame for an emotion term in-
cludes the following lines (cf. Wierzbicka 1999; for a description of anger
and semantically related English emotion words within this framework,
see Harkins and Wierzbicka 1997: 323-328.):
122 Uwe Durst
3. rger
(8) Ich merkte erst spter, was mich rgerte: - Ich war aus dem Bad
gekommen, Hanna am Telefon, sie hatte das Hospital angerufen,
whrend ich im Badezimmer war sie redete mit Elsbeth.
Only later did I realise what 'arger.PA.ST me'. When I came out of
the bathroom Hanna was on the telephone, she had rung the hospi-
Why Germans don't feel "anger" 123
from which the reader will get some evidence for the cause of the person's
anger. If the cause is still unknown, the absence of a prepositional phrase
would presumably evoke the hearer's reaction in the form of a question for
the cause. This also applies to sentences referring to possible future events.
In (14) the hearer probably knows to whom the affair in question should
not become known, and, if so, why it would have some consequences for
the speaker.
to be put straight below the prototypical thought ((e) and (f)). This can also
explain the fact that sich rgern usually takes the preposition ber
('about'), which can also be used in combination with the noun rger.1
Of course, we have no insight into what a person who feels rger really
thinks. Therefore it is necessary to present the person's thoughts within a
prototypical framework. Thus, the semantic description of X rgert sich
(ber Y) can be put as follows:
Wut (fern.) and Zorn (masc.) differ from rger in that they cannot combine
with the verb haben ('to have') without an article (rger haben, 'to have
rger'; but *Wut/Zorn haben, 'to have Wut/Zorn'). But when Wut or Zorn
are combined with the indefinite article and the verb haben, as e.g. in eine
Wut haben ('to have a Wut') or einen Zorn haben ('to have a Zorn'), these
phrases are not counterparts of rger haben ('to have rger') in the sense
that the latter normally combines with the prepositions mit or in and is
understood as what in English could be called 'trouble'. To describe a per-
son's emotion the nouns Wut and Zorn can be used as well as their corre-
sponding adjectives wtend and zornig. Thus, sentences such as (17) and
(18) can both have the same translation in English:
(17) Er hat eine Wut l einen Zorn.
He has a Wut l a Zorn
'He is angry.'
(18) Er ist wtend/ zornig.
He is wtend/ zornig
'He is angry.'
What makes it difficult to draw a distinction between the meanings of
Wut and Zorn (or wtend and zornig) is the fact that they are interchange-
able in a large variety of contexts constituting grammatical sentences.
Both nouns can be combined with possessive pronouns or genitive at-
tributes referring to the experiencer, and with a prepositional phrase desig-
nating the "object", or the "cause'V'causer" of the emotion. The verb
haben can be interpreted as defining a relation that in other terms would be
called "alienable possession", that is, the emotions are conceptualised as a
sort of "things that people can have", and that are located somewhere be-
tween inside and outside the body. Cf. some very common expressions
like:
(19) den Zorn l die Wut l den rger herauslassen
'to let the Zorn l Wut l rger out'
den Zorn l die Wut l den rger in sich hineinfressen
'to swallow the Zorn l Wut l rger'
128 Uwe Durst
While these metaphors are also well applicable to rger, the following
ones seem to be restricted to Wut and Zorn:
(31) Er zahlte den ersten Scheck einfach auf das Caritaskonto seiner
Pfarre ein, lie sich an der Sparkasse Zweck und Ort des Verrech-
nungsschecks erklren, ging dann zu seinem Pfarrer und bat um
einen Barscheck ber fnfzehn Mark - aber der Pfarrer platzte
fast vor Wut. Er erklrte Heinrich, er knne ihm keinen Barscheck
geben, weil er die Zweckbestimmung erklren msse, und so ein
Caritaskonto sei eine heikle Sache...
The first check he simply paid into the welfare account of his par-
ish, had the bank explain to him the purpose and nature of a cross-
ed check, then he went to his priest and asked him for a cash check
for fifteen marks - but the priest nearly exploded with Wut. He
told Heinrich he couldn't give him a cash check because he would
have to state what it was for, and a welfare account was a ticklish
thing... [MK1/LBC.OOOOO (Boll, Ansichten eines Clowns: 193)]
(32) Ivy starrte mich an, dann schleuderte sie pltzlich ihren Mantel
irgendwohin ins Zimmer, stampfend, auer sich vor Zorn... Ivy
hatte sich eingerichtet, eine Woche in Manhattan zu verbringen,
jetzt gestand sie's, und mein pltzlicher Entschlu, nicht zufliegen
wie blich, sondern morgen schon mit dem Schiff zu reisen, um in
einer Woche auch in Paris zu sein, war ein Strich durch ihre Rech-
nung.
132 Uwe Durst
Ivy stared at me, then she suddenly threw her coat across the room
and stamped her foot, beside herself with Zorn... Ivy had arranged
to spend a week in Manhattan, she now revealed, and my sudden
decision not to fly, but to leave tomorrow by boat so as to reach
Paris in a week's time as planned, upset her calculations.
[MK1/LFH.OOOOO (Frisch, Homo Faber. 73)]
(33) Leo wurde immer wtend, wenn ich ihn traf und ihn fragte: "Weit
du noch, wie wir das Holz miteinander durchgesgt haben?" Er
schreit dann: "Aber wir HABEN das Hh nicht miteinander durch-
gesgt"
Leo always got wtend when I saw him and asked: "Do you re-
member how we sawed through that wood together?" He ex-
claims: "But we DIDN'T saw through any wood together."
[MK1/LBC.OOOOO (Boll, Ansichten eines Clowns: 219)] (original
emphasis in capitals)
(34) Sabine wurde regelrecht hysterisch und nannte mich "Mrder",
Karl schrie mich an: "Du Vagabund - Du Hurenbock", und das
machte mich so wild, da ich ihn "verkrampfter Pauker" nannte,
meinen Mantel nahm und in Zorn davonlief.
Sabina got positively hysterical and called me a "murderer", Karl
shouted at me: "You tramp - you lecher," and that infuriated me so
much that I called him a "frustrated pedagogue, " picked up my
coat and left in Zorn. [MK1/LBC.OOOOO (Boll, Ansichten eines
Clowns: 206)]
(35) Ich stand nur da, blickte auf die Koblenzer Strae und dachte an
Marie. Irgend etwas in meinem Gesicht schien den Mann, neben
dem ich stand, wtend zu machen. Er senkte sogar die Zeitung,
verzichtete aufsein "Strau: mit voller Konsequenz!", schob seine
Brille vorne auf die Nase, sah mich kopfschttelnd an und mur-
melte "Unglaublich"
I just stood there, looking out onto Koblenz Street, and thought
about Marie. Something in my face seemed to make the man next to
me wtend. He even lowered his paper, stopped reading his
"Strauss: With unshakeable determination", pushed his glasses
down onto his nose, looked at me, shook his head, and murmured
"Incredible." [MK1/LBC.OOOOO (Boll, Ansichten eines Clowns:
68)]
Why Germans don t feel "anger" 133
(36) Der Grovater war drr und eigensinnig, er geriet leicht in Zorn
und trank, je mehr die Mitbrger ihn mieden.
The grandfather was withered and stubborn, he'd easily fly into
Zorn and he'd drink the more his fellow citizens avoided him.
[MK1/LJA.OOOOO (Johnson, Das dritte Buch ber Achim: 165)]
Yet there are sentences where it seems to be impossible to replace one of
them by the other one:
(37) Mit dieser Entscheidung aber zog er sich den Zorn der guten
Brger zu,...
But with this decision he "caught" the Zorn of the good citizens,...
[H86/KM 4.12384 (Mannheimer Morgen 21.08.1986: 32)]
(38) Den Eintritt der Sachsen in die Geschichte bezeichnet das
Jahr 286, in dem sie gemeinsam mit den salischen Franken von
See her in Nordfrankreich einfielen und die reichen Hafenstdte
" mit furchtbarer Wut" plnderten.
The Saxons' entrance into history is marked by the year 286, when
they invaded from the sea into the north of France together with the
Salian Franks and plundered the wealthy seaports "with an awful
Wut". [MK1/WPE.OOOOO (Poertner, Die Erben Roms: 339)]
(39) Er konnte immer noch zuschlagen wie er es gelernt hatte: nicht
wtend aber hart und bis der Gegner sah da er auf den Boden
kommen wrde unter diesen ausgerechneten zweckdienlichen
Sten.
He was still good at dealing a blow the way he had learned it: not
wtend, but forcefully until the other guy realised that he'd end up
on the floor under these calculated efficient blows.
[MK1/LJA.OOOOO (Johnson, Das dritte Buch ber Achim: 242)]
The phrase sich jemandes Zorn zuziehen ('to "catch" someone's Zorn'}
in (37) is highly idiomatic and does not allow Wut to replace Zorn. In (38) it
would be possible to replace Wut by Zorn, but it makes a difference
whether people plunder with Wut or with Zorn. In the first case, more un-
controlled and spontaneous action seems to be involved, whereas the sec-
ond case seems to imply that the plundering is the result of a certain atti-
tude against those seaports and their inhabitants, an attitude by which the
plunderers' behaviour is guided and determined. In (39) there is an opposi-
134 Uwe Durst
tion between wtend blows and calculated efficient blows. What is con-
veyed by wtend is the notion of someone beating his opponent in an un-
controlled manner and breaking into pieces everything that is in his way.
This may be the reason why the German psychologist Wilhelm Wundt
(1911: 212) put Zorn in the category of "allmhlich ansteigende Affecte"
("gradually rising affections"), and Wut in the category of "pltzlich here-
inbrechende Affecte" ("suddenly overtaking affections"). But actually, this
does not apply to the generation of the emotions as such because a person
can be suddenly befallen by either of them (see examples (20)-(22) above);
it rather applies to the way the corresponding behaviour is generated. This
is why a regent can Zorn und Milde walten lassen ('rule with Zorn and in-
dulgence') [MK1/WPE.OOOOO (Poertner, Die Erben Roms: 259)], but not
Wut und Milde, and why it is quite possible to talk about edlen Zorn ('noble
Zorn') or heiligen Zorn ('holy Zorn'), while Wut better combines with the
adjective rasend ('raging') and can be used to describe not only human be-
haviour but also the intensity of natural phenomena such as storms or
waves. In this respect, Wut seems to be closer to fury, or Italian rabbia
(cf. Wierzbicka 1995 a), than to anger. It is interesting in this respect that in
contemporary written texts the frequency of wtend is higher than the fre-
quency of zornig, and both are much more common than two other adjec-
tives derived from the same lexemes, wtig and zrnend:
MK LIM HBK
the semantics of the words. The adjectives wtend and zrnend are for-
mally identical with the present participles of the verbs wten ('to rage', 'to
be in a fury') and zrnen (roughly: 'to be angry'). Wtig and zornig are
genuine adjective forms with the highly productive denominal derivational
suffix -ig, comparable to -y in English. Not surprisingly, earlier German
dictionaries offered both of them as seperate lemmata, whereas wtend and
zrnend were still considered as verb forms (cf. Adelung 1808, vol. 4:
1733, 1767; Campe 1969-1970[1811], vol.5: 802, 803, 880, 914).
According to Grimm and Grimm (1854 sqq, vol. 14/2: 2514), the con-
temporary use of the present participle wtend in its attributive, predi-
cative, and adverbial function was developed in the second half of the
18th century. This tallies well with the fact that a notion similar to 'anger'
was usually expressed in Old and Middle High German by zorn (>Zorn),
whereas wuot/wuoti (>Wut) originally was used to designate the symptoms
of persons who were seen as demoniacs, and to describe a "state of being
bodily and mentally beside oneself, deprived of one's controlling volition"
("zustand des der lenkenden willkr entzogenen leiblich-seelischen aus-
zersichseins"; 2476). This quasi-medicinal aspect is still existent in the
words Tollwut ('rabies') and Wutbeere (an older name for Tollkirsche,
'atropa belladonna'), and can be considered as being responsible for the
high productivity of -wut as the second element in other compound nouns,
while the number of compounds on the head -zorn seems to be com-
paratively small (118 word forms on -wut with 37 different lexemes, in-
cluding 46 word forms on -tollwut, versus 8 word forms on -zorn with
4 different lexemes, including 3 word forms of Jhzorn ('irascibility',
'passionateness') in the HBK corpus). To give some examples:
(42) Wut
(a) X feels something bad
(b) sometimes a person thinks:
138 Uwe Durst
(43) Zorn
(a) X feels something bad
(b) sometimes a person thinks:
(c) something bad happened because someone did something
(d) I don't want things like this to happen
(e) because of this I want to do something bad to someone
(f) because of this, this person feels something bad
(g) X feels something like this
(h) X cannot not feel like this
The first two lines within the prototypical thought differ in that the "trig-
gering event" seems to be more present for Wut. This is not to be under-
stood in the sense that it would be happening right now in any case. It is
quite possible to say ich bin immer noch wtend ( am still wtend'), but
this rather indicates that the cause is still present in the speaker's view. On
the other hand, Zorn can be used in the sense of a disposition or attitude,
which is reflected in component (e) which, in turn, presupposes a present
or past event that is presumably caused by a person's action ('someone did
something' in (c)), that is, someone can be blamed for what happened. Of
course, you can get wtend 'because someone did something', but this is
not necessarily so. For example, if your computer fails to work unexpect-
edly in the middle of an important operation, your prospective reaction
would be described as wtend rather than zornig.
There is no reference to the duration of the feeling in the formula. It is
clear that both emotions are not just flashes of split-second feelings, but the
limitation at the other end of the scale is not clear-cut. A person who is
characterised as zornig may have this disposition all his life; and while the
author of the following dialogue could write wtend in all three contexts,
the English translation turns from furious to angry where the duration of
the emotion would be too much extended:
Why Germans don t feel "anger" 139
(44) "Ich hob immer gemeint, du bist wtend auf mich", sage ich,
"wegen damals'.'
Hanna nur verwundert.
"Wieso wtend? Weil wir nicht geheiratet haben?" sagt sie. "Das
wre ein Unglck gewesen -"
Sie lachte mich geradezu aus.
"Im Ernst", sagt sie, "das hast du wirklich gemeint, da ich w-
tendbin, Walter, einundzwanzig Jahre lang?"
[MK1/LFH.OOOOO (Frisch, Homo Faber. 165)]
"I always thought you were furious with me," I said, "because of
what happened."
Hanna was merely surprised.
"Why should I be furious? Because we didn't marry?" she said,
"that would have been a disaster."
She positively laughed at me.
"Seriously," she said, "did you think I was angry with you, Walter,
for twenty-one years?" [Frisch 1974: 142]
Line (e) in Wut refers to the somewhat uncontrolled ( have to do some-
thing') and spontaneous ('now') character of the subsequent action, while
in Zorn it indicates a deliberate decision ( want to do something'), and the
consciousness that the intended action will do harm to someone, even if the
agent/experiencer is not conceived of as doing harm to people in principle
(cf. God's Zorn, and expressions like gerechter Zorn ('justified Zorn"), hei-
liger Zorn ('holy Zorn') etc.). This person, however, remains unspecified
in the semantic formula. In this respect Zorn differs from words like Ha
('hatred') or Rache ('revenge'), which seem to include the notion of an ac-
tion that is directed towards specific persons ( want to do something bad
to this person'). The differences in (42e) vs. (43 e) also account for the fact
that people can talk about the Wut of the waves, or of wtende kisses,
whereas the notion of a Zorn of the waves would be much more figurative
and imply a demonistic conception of the world, while zornige kisses
either does not make sense or implies violence in the sense that want to
do something bad to someone' (43e). The last components (42h) and
(43 h) refer to the fact that Wut and Zorn can take the role of an agent in
(semi-) idiomatic phrases, and that their corresponding adjectives cannot
be combined with the negative imperative.
140 Uwe Durst
7. Conclusion
There are three conclusions that can be drawn from the observations re-
ported above. First, none of the three most frequent 'anger' words in Ger-
man matches the meaning of the English word anger. Second, each of them
has its own semantic structure, determining their ranges of use and their re-
strictions in certain contexts. And finally, the meanings of these words as
analysed here are supported by the evidence from etymology if one as-
sumes a "psychological shift" in their histories, which is responsible for
their semantic change and explains the relation between constancy and
variation in the diachronic process. What has been preserved throughout
the centuries are those components that indicate how a person to whom the
respective emotion is attributed is likely to behave, or how the person
would be affected by his/her own behaviour. It would be interesting, in this
context, to compare the proposed meanings with those of the etymologi-
cally related Dutch words zieh ergeren, woede, and toorn, and to extend the
analysis to further German 'anger' words in order to verify the correctness
of the semantic formulae proposed in this paper.
Notes
1. This paper was written while I was a Visiting Fellow at the Australian National Uni-
versity in Canberra. I thank the members of the Linguistics Department for their
hospitality. I also thank Jean Harkins and Anna Wierzbicka for helpful comments on
an earlier version of this paper, presented at the University of New England in Ar-
midale in March 1998.
2. Ut scias autem non esse sanos quos ira possedit, ipsum illorum habitum intuere;
nam ut furentium certa indicinia sunt audax et minax vultus, tristis frons, torva
fades, citatus gradus, inquietae manus, color versus, crebra et vehementius acta
suspiria, ita irascentium eadem signa sunt: flagrant ac micant oculi, multus ore toto
rubor exaestuante ab imis praecordiis sanguine, labra quatiuntur, denies compri-
muntur, horrent ac surriguntur capilli, spiritus coactus ac stridens, articulorum se
ipsos torquentium sonus, gemitus mugitusque et parum explanatis vocibus sermo
praeruptus et conplosae saepius manus etpulsata humus pedibus et totum concitum
corpus "magnasque irae minas agens,..." (De Ira I.I: 34)
Why Germans don t feel "anger" 143
3. The prepositions von (Of, 'from') and ber ('about', Over', 'above') seem to be
mutually interchangeable within this syntactic frame. Their use in this context can
be compared to that of the English words on and about, although, on a closer look,
some differences will certainly be found.
4. ... etiam si amarum nos acerbumque dicimus, nee minus stomachosum, rabiosum,
clamosum, difficilem, asperum, quae omnia irarum differentiae sunt; (De Ira I.IV:
2)
5. The analysis here is based on an online research of selected text corpora held by the
Institut fr Deutsche Sprache (IDS) in Mannheim/Germany, by means of the Corpus
Storage, Maintenance, and Access System (COSMAS), developed at IDS; along
with some random observations and intuitions by the author as a German native
speaker. The corpora are abbreviated as follows:
GOE: Goethe-Korpus, 1 709663 words from J. W. von Goethe's novels and theor-
etical/scientific writings (assembled at the University of Kyushu/Japan);
GRI: Grimm-Korpus, 518827 words from Grimm and Grimm's collection of sagas
and fairy tales (assembled at the Ryukoku University, Kyoto/Japan);
HBK (H85-H88): Handbuch-Korpus, 13911305 words from newspapers, in par-
ticular from Mannheimer Morgen, Die Zeit, and Stern, 1985-1988 (assembled by
IDS);
LIM: Limas-Korpus, 1348813 words, passages from a large variety of text genres,
1970-1971 (assembled by the Forschungsgruppe LIMAS, Bonn und Regensburg/
Germany);
MK (MK1-MK2): Mannheimer Korpus, 3 291 136 words from a large variety of text
genres, including novels, scientific and popular books, manuals, hand-bills, instruc-
tions for product use, etc., 1949-1973 (assembled by IDS).
Interlinear glosses are given where the argumentation is based on a particular struc-
ture. Abbreviations used in glosses are: ACC = accusative, FEM = feminine,
IMP = imperative, INF = infinitive, MASC = masculine, MP = modal particle,
NOM = nominative, PAST = past tense (Prteritum), PRES = present tense,
REFL = reflexive pronoun, SG = singular. If not indicated otherwise, emphases in
examples are added by the present author.
6. This also applies to studies in other languages. For survey and discussion see
Schmidt-Atzert (1980); Debus (1988), who argues for an orientation along Biihlers
trichotomic model of communication; and Schwarz and Ziegler (1996), who further
try to establish a link between neuropsychology/neurolinguistics and lexicology.
7. ber seems to be used where a deliberative process is involved in the action referred
to by the verb: sprechen ber ('to talk about'), nachdenken ber ('to think about'). It
must be mentioned, however, that there are reflexive emotion words in German
which cannot combine with this preposition, e.g. sich frchten ('to be afraid'). Thus,
the semantics of the preposition must not be confused with the semantics of the re-
flexive verb. The semantic interrelation between (reflexive) verbs and their preposi-
tional complements requires further investigation.
8. Actually, two examples for the phrase in rger geraten ('to get into rger') were
found, one of them in a text from the early 19th century [GOE/AGD.OOOOO (Dich-
tung und Wahrheit, Hamburger Ausgabe, vol. 9: 73)], another one in a contemporary
144 Uwe Durst
novel [MK1/LJA.OOOOO (Johnson, Das dritte Buch ber Achim: 211)]. However,
this use seems to be out of date.
9. Two imperatives of the verb zrnen (corresponding to Zorn) were found in the cor-
pus, one of them in Grimm and Grimm's collection of legends [GRI/SAG.00542
(Lohengrin zu Brabant; Sagen: 542)], the other one in Goethe's Werther [GOE/
AGW.OOOOO (Die Leiden des jungen Werther, Hamburger Ausgabe, vol. 6: 91)]. To
use these words in contemporary German would sound quite obsolete, if not rid-
iculous. As Anna Wierzbicka pointed out to me, it is also quite common to use an-
other 'anger' word, bse (see example (4) above), with the negative imperative: Sei
mir nicht bse (lit.: 'be.IMP me.DAT not bad'; hence: 'don't be angry with me').
The semantics of the phrase jemandem bse sein (lit.: 'someone.DAT bad be.INF';
'to be angry with someone') were not investigated, but I would expect a similar re-
sult as for rger, as far as the notion of an intermediate thought, and thus the possi-
bility of controlling the rise of the feeling, is concerned.
10. 'op, za'am, hemh, hdron, qce$cep, and 'cebrah. For a comparative survey of their
meanings, see Bergman and Johnson (1973), and Myhill (1997).
11. Considering the religious usage of Zorn, Philipp Lersch (1956: 235) stated in his
phenomenological analysis that "When we are zornig, it is always in the name of
what is to happen according to generally binding norms... But we get into Wut only
in the name of our individual self-being." ("Wir sind zornig immer im Namen
dessen, was nach allgemein gltigen Normen geschehen soll... In Wut aber geraten
wir immer nur im Namen der Belange unseres individuellen Selbstseins.") How-
ever, this seems to be too schematic and does not account for examples such as (34)
(for Zorn), where the experiencer apparently feels that he was personally offended,
and for (35) (for Wut), where the man's reaction described by the speaker seems to
rely on a certain idea of how a decent citizen should look.
12. This is not a complete analysis of the German present participle (for example, the
aspect of duration is not considered here), but sufficient to draw a distinction be-
tween the word class meanings of present participles and adjectives (see also
note 14).
13. To avoid misunderstandings: I am not saying that Zorn is a "disposition" in the
sense in which Wittgenstein (1967: 87; number 491) called Liebe ('love') and Ha
('hate') dispositions. In German, neither love nor hate can "seize" someone, nor
can they suddenly "break out", whereas Zorn, as well as Wut, can (see examples
(20)-(22) above).
14. The semantic difference between adjectives on -ig and adjectives on -end is still
transparent in some cases. For example, ein tollwtiger Hund ('a tollwutig.MA.SC.
SG.NOM dog') is a dog that suffers from rabies, which is thought of as an incurable
disease rather than a momentary state. Again, a person can have zittrige Hnde
('shaky hands', adj.), which implies, to a certain extent, general weakness, or old
age etc., whereas zitternde Hnde ('trembling hands', pres. part.) describes a tem-
porary state; consequently, zitternd can combine in such contexts with causative
prepositional phrases indicating a temporary deviation from the normal, e.g. zit-
ternd vor Angst/vor Klte ('trembling with fear'/'shivering with cold').
Why Germans don if eel "anger" 145
References
Myhill, John
1997 What is universal and what is language-specific in emotion words?
Evidence from Biblical Hebrew. Pragmatics and Cognition 5/1:
79-130.
Ortony, Andrew and Terence J. Turner
1990 What's basic about basic emotions? Psychological Review 97/3:
315-331.
Plum, Sabine
1992 Gefhlswrter im Wrterbuch: berlegungen zur lexikographischen
Bedeutungserluterung des emotionalen Wortschatzes. In: Gregor
Meder und Andreas Drner (eds.), Worte, Wrter, Wrterbcher. Lexi-
kographische Beitrge zum Essener Linguistischen Kolloquium,
169-182. (Lexicographica: Series Maior; 42). Tbingen: Niemeyer.
Russell, James A.
1991 In defense of a prototype approach to emotion concepts. Journal of
Personality and Social Psychology 60/1: 37-47.
Schmidt-Atzert, Lothar
1980 Die verbale Kommunikation von Emotionen: Eine Bedingungsanalyse
unter besonderer Bercksichtigung physiologischer Prozesse. Ph. D.
Dissertation, Universitt Gieen.
1987 Zur umgangssprachlichen hnlichkeit von Emotions Wrtern. Psycho-
logische Beitrge 29: 140-163.
Schmidt-Atzert, Lothar and Walter Strhm
1983 Ein Beitrag zur Taxonomie der Emotionswrter. Psychologische Bei-
trge 25: 126-141.
Schreiber, Herbert, Karl-Ernst Sommerfeldt and Gnter Starke
1993 Deutsche Substantive: Wortfelder fr den Sprachunterricht. Leipzig:
Langenscheidt.
Schwarz, Manuela and Wolfram Ziegler
1996 Emotionen in Neurolinguistik und Lexikologie: Ein Forschungsber-
blick. Lexicology 2/1: 34-62.
Seneca, Lucius Annaeus
[1970] Moral Essays. Vol. I. With an English translation by John W. Basore
(Loeb classical library; 214). London: William Heinemann; Cam-
bridge: Harvard University Press.
[1989] Philosophische Schriften. Bd. 1. bersetzt, eingeleitet und mit Anmer-
kungen versehen von Manfred Rosenbach. Darmstadt: Wissenschaft-
liche Buchgesellschaft. 4. Aufl.
Septuaginta
[1962] Edidit Alfred Rahlfs. Stuttgart: Wrttembergische Bibelanstalt. Editio
Septima.
Stearns, Carol Zisowitz and Peter N. Stearns
1986 Anger: The Struggle for Emotional Control in America's History. Chi-
cago: University of Chicago Press.
148 Uwe Durst
In the ongoing debate about emotions and their relationship with facial ex-
pression, James Russell has recently campaigned for "the gathering of new
evidence" (1995: 382) concerning the nature and meaning of facial ex-
pressions across cultures and across languages. In this paper, I present data
on some ways in which Lao people describe facial expression, as well as
some of the ways they attribute inner states to people making particular fa-
cial displays. I first discuss briefly a number of words and expressions
available in Lao for the description of emotions and other inner states. In
the second section, I describe some of the ways in which Lao speakers iso-
late, recognise and describe particular facial expressions, as well as making
some comments on their attribution of inner states (of thought, feeling,
and/or emotion) to people making the expressions in question. One point I
want to make is that "folk" analyses of semiotic phenomena, revealed in
linguistic semantics, are fundamental to a well-informed comparative
science of emotion and nonverbal communication.
1. "Emotions" in Lao
There are no exact equivalents in Lao of the English words sad, angry, dis-
gusted, happy, surprised, and afraid. In some cases, there are obvious
rough translations of these so-called "basic emotions" (Ekman 1992),
while in others the equivalents are not so clear. Let me introduce some of
the problems.
1.1. 'Sad'
There are several possible translations into Lao of the English word sad.
The most likely to correspond to the canonical sense of English sad is sia-
150 N.J.Enfield
caj, literally 'lost-heart'. Sia-caj shares with sad the idea of bad feelings as
a response to some unfortunate event. Thus, one can be sia-caj 'about'
something, or sia-caj 'that' something has happened, apparently restricted
to cases in which something happens to the experiencer. For example, a
speaker says he was sia-caj when, having hurried home from the market to
get money to buy a bargain-priced turkey, returns to find that the creature
has already been sold.
Another common term often translatable as sad is sook-saw, which is
more likely to carry the connotation of a longer term state of unhappiness,
due to some sorry state the subject finds themselves in, rather than some
specific event. A third term which overlaps to some extent with sad is
ngaw, which most typically expresses a bad feeling of loneliness or alone-
ness, especially when a person is too often by themselves in overly quiet
surroundings. The feeling of ngaw is no doubt related to the premium in
Lao culture on muan 'fun', and being with muu 'friends, the group'. Lao
people are most comfortable spending time in larger groups of people with
plenty of activity going on. Feeling ngaw is a response to aloneness in this
cultural context.
1.2. 'Angry'
There are two expressions in Lao which can be routinely translated, in vari-
ous contexts, by the English word angry. The first is caj-haaj Offensive-
heart' (in the sense of 'heart-on-the-offensive'), which expresses the emo-
tional state of a person who is responding to some incident which directly
and negatively affects them. Someone who is caj-haaj is preferably
avoided, since they would be likely to say (or perhaps do) something bad to
anyone who they may consider responsible for whatever it is they are react-
ing to. In this sense, caj-haaj is quite like English angry, but the two differ
in that caj-haaj cannot usually be used to express an idea of being angry
about some situation which does not directly affect the person involved. It
is more immediate. (Note that the component haaj may be used as a
speech-act verb - to 'tell somebody off' - and Lao speakers normally find
it upsetting, as recipient or bystander, when this behaviour occurs.)
The second common 'anger' term is khiat, where the negative feeling is
typically not overtly expressed, similar in some contexts to Australian Eng-
lish pissed off. In contrast with caj-haaj, it is on-expression which char-
A Lao perspective on facial expression of emotion 151
1.3. 'Disgusted'
There is a fairly clear translation into Lao of English disgusted, namely khi-
diat. The prefix khi- (related to khii 'shit') is used widely in the language
for a range of idiomatic expressions to do mainly with different kinds of ef-
fluent material and waste products (khi-muuk [muuk = nose] 'mucus'; khi-
thaw [thaw = 'ash'] 'ashes'; khi-luaj [luaj = 'mill'] 'sawdust'), as well as
negative personal characterisations (khi-khuj [khuj = 'chat'] 'snobby'; khi-
lak [lak= 'steal'] 'thief; khi-qaaj [qaaj= 'shy'] Overly shy'). (The mor-
pheme -dial has no independent meaning outside of the expression khi-diat
'disgusted'.) The usage of khi-diat is more restricted than that of the Eng-
lish term disgusted, since the Lao term is only used to refer to responses
evoked by things which are literally physically disgusting, such as the idea
of eating certain things, or of dealing with physically revolting things such
as rotten food, etc. In Lao, for example, one cannot be disgusted 'at' some-
one, or 'that' something has happened.2
1.4. 'Happy'
The English expression happy may be translated into Lao as dii-caj 'good-
heart', sabaaj(-caj) 'content(-heart)' or mii khuam-suk 'have contented-
152 N.J.Enfield
ness'. Dii-caj has a close affinity with the English expression glad, being a
transient positive emotional response to some specific fortunate incident.
Thus, one may be dii-caj 'that' something has happened, but unlike Eng-
lish happy, dii-caj cannot describe a feeling of general contentedness (as in
John is happy with his new job or Mary seems happy these days). This
latter idea is partly covered by the term sabaaj 'content/comfortable', a
cultural key word for Lao speakers (as for speakers of neighbouring Thai).
Sabaaj itself refers to a feeling rather than an emotion, since it does not en-
tail a cognitive component - it often means 'comfortable', for example
with regard to seating, clothing, or ambience. However, in a compound
with -caj 'heart', the meaning does include the cognitive evaluation lead-
ing to the feeling. This prior cognitive evaluation is the hallmark of emo-
tion (according to many; but cf. e.g. Zajonc 1980, 1984).
Another expression corresponding in certain cases to happy is mii
khuam-suk, literally 'to have happiness'. (Khuam 'sense' acts here as a
nominalising prefix to the verb-like bound element suk 'content/happy'.)
This expression is used in many cases which would also be compatible
with sabaaj, but it does not permit reference to exclusively physical com-
fort which sabaaj 'content' allows. Rather, it is a profound state of well-
being which normally refers to one's general condition of mind and soul.
While one may be dii-caj 'glad' to see an unexpected visitor, it would be
odd to mii khuam-suk 'have happiness' in the same way. On the other hand,
an aim in life could be to mii khuam-suk 'have happiness', but it would
make little sense for one to generally pursue being dii-caj 'glad'.
1.5. 'Surprised'
Two Lao translations are possible for English surprised. These are tok-caj
'fall-heart' and tuun 'awaken'. Both involve a feeling of reaction to the un-
expected, but the second entails an instinctive physical reaction (a 'start'),
and there is no entailment of any thought or evaluation preceding and/or
causing the feeling of 'surprise'.3 (Cf. Goddard 1997 on a similar distinc-
tion in Malay.) Tok-caj, however, can include the idea of a cognitive evalu-
ation leading to the feeling of 'surprise'. Thus, one can be tok-caj, but not
tuun, 'that' something has happened. This element of evaluation leading to
feeling would seem to point to tok-caj as an emotion, and tuun as a mere
feeling or instinctive physical reaction. Even so, the term tok-caj is not pre-
A Lao perspective on facial expression of emotion 153
cisely equivalent to surprise. For example, tok-caj cannot express the kind
of 'surprise' expressed in English I'm surprised you're still here. With re-
spect to tok-caj, some specific event must accompany the evaluation lead-
ing to the feeling of 'surprise'.
1.6. 'Afraid'
The most obvious translation of afraid or scared into Lao is jaan. It is un-
clear to me what difference exists between Lao jaan and English afraid or
scared. It would seem that jaan is closer to afraid in its being less sugges-
tive of immediate and present danger. Jaan is also used in Lao as a com-
plement-taking predicate expressing what the speaker merely suspects
rather than literally fears (cf. English I'm afraid she's not coming), and un-
like English it need not have a negative connotation (cf. Lao jaan phoo
[afraid enough] '(I) suspect (there'll be) enough [e.g., rice to eat]' vs. Eng-
lish ?I'm afraid there'll be enough).
The term caj, which elsewhere refers to the physical 'heart' or 'centre', oc-
curs in most Lao expressions related to emotion, and denotes the primary
'seat' of cognitive and emotional activity for the Lao, the so-called 'heart'
or 'mind'. This word is cognate with Thai cay (Diller and Juntanamalaga
1990), and shares many of the same idiomatic and grammatical properties.
Both the Lao and Thai terms also show extensive semantic and grammati-
cal similarity with Malay hati and comparable terms in other Southeast
Asian languages (cf. Matisoff 1986, Oey 1990, Jaisser 1990 and Goddard
this volume - Enfield 2001 provides a methodological and theoretical dis-
cussion).
There are dozens of expressions in Lao involving caj, of roughly two
main types. A set of terms describing lasting character traits treat caj as a
regular nominal head, with some quasi-adjectival modifier following (e.g.
caj-dam [heart-black] 'ruthless', caj-kuang [heart-broad] 'generous', caj-
kaa [heart-dare] 'daring, courageous'). Most emotion terms, referring to
more transient characterisations, place caj second in the combination (e.g.
nooj-caj [small heart] Offended', naak-caj [difficult-heart] 'undeciding,
154 N.J.Enfield
1.8. Comment
I now describe some ways of speaking about facial expression in Lao, con-
centrating on lexical items and idioms which can be used to describe par-
ticular facial expressions (or features thereof). I will note the kinds of at-
tributions of inner/emotional states which Lao people make, in their own
terms, to people with these facial expressions. I am not concerned here
with such questions as the factual accuracy of attribution of a particular
emotional state to some facial expression, nor am I interested here in cor-
relation, causal or otherwise, of inner states and facial displays. The aim is
to present linguistic evidence for the Lao speaker's conventional analysis
of the face.
A Lao perspective on facial expression of emotion 155
While English has a range of simple words for facial expressions (e.g.
scowl, grin, smile, grimace), Lao notably has few. The only one attested
below is mm 'smile'. Many of the rest are more explicitly descriptive poly-
morphemic expressions, such as 'rotten face', 'smelly face', 'stuck eyes',
and 'clenched mouth'. In between these extremes are a number of 'cran-
berry' expressions (i.e. in which one component does not independently
occur), such as naa-beq 'ready-to-cry face' and taa-loo 'exposed-whites-
above-iris eyes', in which the modifying elements (beq and loo, respect-
ively) have no independent meaning.
Syntactically, these facial expression descriptions involving head nom-
inals such aspaak 'mouth', taa 'eye(s)', or naa 'face' may pattern either as
(a) noun + predicate, in a kind of relative clause or noun + attribute con-
struction ( 2.4.4., below):
taa-khaang
eye-stuck
'stuck eyes'; 'eyes which are stuck';
or (b) a kind of incorporating verb + object construction ( 2.3.4., below):
nom-paak
clench-mouth
'clench-mouthed'.
Note also that being naa-buut 'rotten-faced' does not entail an emotional
state (such as khiat 'angry'), but may simply describe perceived ugliness or
generally poor temperament (i.e. as a description of the kind of person who
never smiles).
The face which Lao speakers call 'smelly face' is essentially the classical
"disgust face", with wrinkled nose, and raised upper lip. The element men
'smelly', however, refers not to a property of the face itself (in contrast to
naa-buut 'rotten face', 2.2.1., above), but to something smelly (i.e.
rotten) which would cause someone to pull such a face. While naa-buut
'rotten face' is typically used to describe the face of someone who is of
poor temperament, or who is khiat, or even caj-haaj ( 1.2., above), the ex-
A Lao perspective on facial expression of emotion 157
The term khum refers to a kind of furious anger or rage, although certainly
less strong than these English terms. To have a khum face is to look like
someone in such a state, typically manifest in classical signs of 'rage' in the
face, such as exposed incisors, square mouth, frowning, and perhaps also
exposed-whites-above-iris eyes (cf. 2.4.1., below).
There are a number of terms in Lao for facial expressions involving the
mouth and parts of the mouth, mostly including the terms paak 'mouth',
sop 'lips', and kheew 'teeth'. It is worth noting, incidentally, that the term
paak is only approximately translated as 'mouth' - it can be used to refer to
the lips and outer part of the mouth as well, as in the expression paak teek
[mouth break] 'broken mouth' (referring to split or chapped lip(s)).
158 N.J.Enfield
2.3.1. mm 'smile'
The term mm 'smile' is apparently equivalent in meaning to English smile.
The crucial idea is that the corners of the mouth are raised (and naturally
this includes associated visible muscle movements around the eyes,
cheeks, and jaw). In attributing an emotional state to a smiling person, Lao
speakers generally describe them as dii-caj 'glad', and occasionally sa-
baaj(-caj) 'content' (if the smile in question is not judged to be a transient
response to some specific happy stimulus).
The expression taa-loo (where loo does not occur independently) refers to
wide-open eyes, where the whites of the eyes above the iris are visible. In-
formants most often describe someone with this expression as jaan
'afraid', especially when other aspects of the facial expression support (or
at least do not contradict) this, for example if the subject is also qaa-paak
Open-mouthed' (2.3.3., above). If a subject combines taa-loo 'exposed-
whites-above-iris eyes' with some incongruous pose, such as a smiling
mouth, Lao informants attribute to the person a state of insanity (baa 'in-
sane'), or else describe them as a malevolent spirit (phii 'spirit, ghost',
phii-baa [spirit-insane] 'lunatic').
160 N.J.Enfield
something more, which accounts for the persistence of the pose over time.
For example, the person who is taa-khaang 'stuck-eyes' may be son-caj
'interested', or the like.
While Lao has a basic transitive verb nok 'to raise something', this verb
cannot be used for raising one's eyebrows, unless the meaning intended is
that the person physically pulls up their own eyebrows with the hands. To
raise one's eyebrows as a facial gesture is referred to as nik-khiw (in the
verb + object pattern, perhaps best literally glossed as 'raise-eyebrowed').
(The verb nik does not occur elsewhere.) Informants have responded in a
number of ways in attributing inner states to subjects with nik-khiw 'raised
eyebrows', and it seems that accompanying features of facial expression
are important factors in choosing between alternatives. Someone who is
merely nik-khiw 'with raised eyebrows', and without specific expression
elsewhere in the face, may be described as son-caj'interested\jaak-thaam
'wanting to question', or boo-sua 'non-believing, doubtful'. The common
theme here seems to be that the subject wants to know more about what is
going on. When nik-khiw 'raised eyebrows' is accompanied by taa-naj 'big
eyes', taa-khaang 'stuck eyes', or especially taa-loo 'exposed-whites-
above-iris eyes', this is associated with being jaan 'afraid', as well as tok-
caj 'surprised'.
2.5. Comment
The evidence presented in 2 has shown that Lao speakers' habitual ways
of talking about facial expression differ from those of English speakers. It
appears that for certain English expressions which seem to describe the
face as a whole - such as frown or grimace - the Lao equivalents explicitly
refer to relevant parts of the face, such as the eyebrows or the lips. Simi-
larly, certain specific aspects of facial expression associated with certain
specific emotions are singled out in Lao, such as for example the eyebrows,
and the exposed-whites-above-iris eyes. Thus, it is not simply expressions
on whole 'faces' that reveal and/or communicate feelings and thoughts, but
rather expressions involving features or components of the face, such as the
eyes, eyebrows, lips, mouth, or some combination of these. Certain de-
scriptions of facial expressions make specific reference to components of
the face, and I suggest that in further research, it would pay to consider fa-
cial expression as having a greater componentiality than is sometimes as-
sumed. (Cf. Wierzbicka 1999, Chapter 4.)
Conclusion
While we may be led to attribute thoughts and/or feelings to a person on the
basis of the expression on their face, it is not the case that a particular set of
facial expressions conveys, either universally or specifically, a particular
set of inner states or emotions. Firstly, there is no universal set of "emo-
tions". What are denoted by words for 'emotions' are not actual phenom-
ena, but conventional and emic descriptions of actual phenomena. Emotion
terms denote concepts, by particular names in particular languages, such as
angry, pissed off, or caj-haaj. When we work on the description of emo-
tion, we cannot escape our reliance on language. Scientific inquiry is
A Lao perspective on facial expression of emotion 163
Notes
1. Transcription of Lao is a simplified system, with tone-marking omitted. Note that
the consonant symbol 'j' in this system (as in the IPA system) is pronounced like
English 'y' in you and boy; 'c' is pronounced in a similar manner to English 'j' in
Jill; 'q' represents a glottal stop. Symbols ' ', 'o', 'e' represent high-back-un-
rounded, low-back, and low-front vowels, respectively. All data are from fieldnotes,
and from field interviews conducted in Sisavat and Saphang Mo villages in Vien-
tiane, Lao PDR, January-May, and July-August 1998. Syban Khoukham's assist-
ance is gratefully acknowledged. Pitsana Vayaphanh also provided generous con-
sultation, and Sotaro Kita gave useful comments on a late version of the text.
2. Disgust is a particularly physical emotion, with a genuine functional/biological as-
pect to it, and one could imagine that certain things - such as, say, rotten
food - would evoke disgust in anyone with a human body, in any cultural sphere. But
it is interesting to note that culture can differently define what disgusts people in this
physical way. Thus, one can easily find everyday practices which are embraced by
one group, yet physically revolting for another. Rural Lao people enthusiastically
snack on insects of different kinds (crickets, grasshoppers, beetles), a practice which
the average urban Australian, say, would find physically disgusting. On the other
hand, the popular practice of oral sex in Anglo culture is regarded similarly as physi-
cally disgusting in many other settings (e.g. rural Burma, Spiro 1977: 232). These
fascinating culture-specific visceral responses, showing the influence that culture
may have on our biological responses, are worth closer and broader research.
3. 'awaken' has an interesting argument structure, being S=A ambitransitive,
where the subject is semantically a patient; thus, cf. khooj tuun hok moong [I awaken
six o'clock] woke up at six o'clock', khooj tuun caw [I awaken you] (got a) start
(from) you'/'You gave me a start'. In the second example, the syntactic object caw
'you' is the stimulus for the event of surprise predicated by tuun 'awaken' of its sub-
ject khooj T.
4. Even so, colour research has resulted in interesting findings, such as the various im-
plicational statements which Berlin and Kay (1969) originally put forward. While
there are serious methodological problems with their research (Lucy 1996), it is
nevertheless interesting to consider the kinds of questions being asked. If a language
has only three "basic colour terms" (as Berlin and Kay define these), can we predict
what will they be? Berlin and Kay argued in the affirmative for LIGHT ('white'),
WARM ('red'), and DARK/COOL ('black/blue/green'). If a language had five terms,
they argued, then 'yellow' would appear, along with a split of DARK/COOL into
'black', and 'green/blue'. Could we follow such a line with respect to facial ex-
pressions? If a language had only one "basic facial expression term", would it be
'smile'? (Cf. Wierzbicka 1999: 275, 282.) But this may be the wrong kind of ques-
tion to ask (cf. Lucy 1996 on colour).
A Lao perspective on facial expression of emotion 165
References
Berlin, Brent, and Paul Kay
1969 Basic Color Terms. Berkeley: University of California Press.
Diller, Anthony V. N., and Preecha Juntanamalaga
1990 'Full hearts' and empty pronominals in Thai. Australian Journal of
Linguistics 10 (2): 231-256.
Ekman, Paul
1973 Darwin and Facial Expression: A Century of Research in Review. New
York: Academic Press.
1992 An argument for basic emotions. Cognition and Emotion. 6 (3/4):
169-200.
Enfield, N. J.
2000 On linguocentrism. In: Martin Ptz and Marjolijn H. Verspoor (eds.),
Explorations in Linguistic Relativity, 125-157. Amsterdam: John Ben-
jamins.
2001 Semantic analysis of body parts in emotion terminology: avoiding the
exoticisms of "obstinate monosemy" and "online extension". Prag-
matics and Cognition 9.2.
Foley, William A.
1997 Anthropological Linguistics. London: Blackwell.
Goddard, Cliff
1997 Contrastive semantics and cultural psychology: 'Surprise' in Malay
and English. Culture and Psychology 3 (2): 153-181.
Jaisser, Annie
1990 DeLIVERing an introduction to psycho-collocations with SIAB in
White Hmong. Linguistics of the Tibeto-Burman area 13(1): 159-177.
Lucy, John
1996 The linguistics of 'color'. In: C. L. Hardin and Luisa Maffi (eds.),
Color Categories in Thought and Language, 320-346. Cambridge:
Cambridge University Press.
Matisoff, James A.
1986 Hearts and minds in Southeast Asian languages and English: an essay
in the comparative lexical semantics of psycho-collocations. Cahiers
de Linguistique Asie Orientale 15(1): 5-57.
Oey, Eric. M.
1990 'Psycho-collocations' in Malay: a Southeast Asian areal feature. Lin-
guistics of the Tibeto-Burman Area 13(1): 141-158.
Russell, James A.
1995 Facial expressions of emotion: What lies beyond minimal universal-
ity? Psychological Bulletin 118 (3): 379-391.
166 N.J.Enfield
Spiro, Melford E.
1977 Kinship and Marriage in Burma. Berkeley: University of California
Press.
Wierzbicka, Anna
1996 Semantics: Primes and Universals. Oxford: Oxford University Press.
1999 Emotions across Languages and Cultures: Diversity and Universals.
Cambridge: Cambridge University Press.
Zajonc, R. B.
1980 Feeling and thinking. Preferences need no inferences. American Psy-
chologist 35 (2): 151-175.
1984 On the primacy of affect. American Psychologist 39(2): 117-123.
Hau: A key word in the Malay vocabulary of
emotion
Cliff Goddard
Introduction1
The word hau is one of the "key words" of Malay culture. By this I mean,
following Wierzbicka (1997), that hati functions as a conceptual focal
point for an entire complex of characteristically Malay values, attitudes,
and expectations; and that by studying the meaning and uses of this one
word we can learn a surprising amount about Malay culture - in particular,
about the conceptualisation of emotion in Malay culture.
Though the nearest English gloss for hati is 'heart' (in its emotional-
moral sense) the two words are not semantically identical, if only because
the Malay hati is significantly more active, and more cognitive, than the
English heart. A more revealing, but still inadequate, gloss for hati would
be 'the sensitive part of a person'. Hati has a high frequency in any dis-
course about human interaction - partly on account of its participation in
dozens (if not scores) of fixed expressions concerning attitudes, moods,
and personal traits, e.g. susah hati 'troubled, worried', hati keras 'deter-
mined', rendah hati 'humble, modest', partly because the hati is the locus
for feelings (especially feelings about other people), and partly because
emotional reactions are often presented in terms of the hati "speaking". As
one might expect, the word occurs frequently in traditional sayings and
poems, and in popular song titles. It is no exaggeration to say that one can-
not approach an understanding of Malay attitudes about human nature and
about social life without understanding this quintessentially Malay con-
cept.
It would be impossible in one short paper to detail the full range of usage
of hati. I attempt here to do the following: first, to outline the range of use
and collocational possibilities of hati, informally comparing and contrast-
ing it with English heart, second, to advance and argue for an explicit sem-
antic explication of hati in its core or central meaning (as in an expression
like hati orang 'a person's hati'); third, to explicate the semantics of five
168 Cliff Goddard
The more active nature of the Malay hati is shown by the existence of ex-
pressions such as those in (2) and (3), in which the hati is depicted as ter-
tanya 'spontaneously asking' an emotionally loaded question, or as bisik
'whispering' an emotional reaction. In general, the hati is capable of "ex-
pressing" itself in ways which can be reported with a wide range of speech-
act verbs, includingpujuk 'console, persuade', rungut 'grumble', and plain
kata 'say', aside from those verbs illustrated below. (Note that in examples
throughout the paper, one sees hati suffixed with pronominal clitics: -ku ,
my', -mu 'you, your', -nya 'he/she, his/her'.)
(4) Roslina tidak dapat berkata apa-apa lagi, hatinya terasa diiris-
iris, pedih, pahit.
'Roslina couldn't say anything more, her hati felt cut up, smarting,
bitter.'
(5) Pada masa yang sama teruskan usaha membahagiakan dan
menggembirakan hati ibu dengan bantuan yang terdaya.
'At the same time carry on your effort to make your mother's hati
happy and pleased with the best help you can give her.'
(6) Tapi Chan tak tau pedihnya hati ni bila mendapati Chan tak ada di
tempat biasa... Tak boleh takjumpa. Rosak hati ni.
'But you (Chan) don't know how sad this hati was on realising that
you (Chan) weren't at the usual place... We have to meet. (If not) it
hurts this hati.'
170 Cliff Goddard
The hau may also be the locus of desire or intention, especially if the de-
sire or intention concerns another person. Not only may one speak of niat
hati 'desire in one's hati', hasrat hau 'hati's desire', kehendak hau 'hati's
wish, desire', but also of the hati actively wanting or intending to do some-
thing. For example:
(9) Jauh di sudut hatinya, kasih mula dipintal, sayang mula dianyam
dan cinta mula dipilin.
'Deep in the recesses of her heart, love began to wring out, care
began to be woven, and passion began to be plaited.'
(10) Namun di dalam hatinya dia merasa bangga kerana dipuji, kerana
ramai yang terpesona pada kejelitaannya.
'But in her heart she felt proud of the compliment, because there
were many who were enchanted by her beauty.'
(11) Perasaan marah kerana ayah berkasar dengan ibu itu memang
biasa dalam hati seorang anak.
'The feeling of annoyance on account of a father's harsh treatment
of a mother is simply normal in the hati of a child.'
Consistent with the private, inacessible nature of the hati is the fact that a
person can say things (di) dalam hati 'inside the hati', meaning to say
things to oneself, as illustrated in (12) and (13) below; cf. also the ex-
pression membaca di dalam hati 'read to oneself, as opposed to reading
aloud.
The same broad range of semantic effects is found among fixed nominal
expressions involving hati. Most of these have a two-part structure, con-
sisting of hati and an adjective. They can be grouped formally according to
whether the adjective follows or precedes the noun hati. Attributive adjec-
tives normally follow their head nouns in Malay, but hati + adjective is the
less frequent of the two patterns. Abdullah's (1990) Kamus Simpulan Ba-
hasa [Dictionary of Fixed Expressions], for example, lists around
15 hati + adjective expressions, but over 50 adjective + hati expressions.
Both these figures are likely to be far short of the full number, as several of
the combinations listed below are not on Abdullah's lists, but the relative
preponderance of the adjective + hati pattern is clear.
While the semantic difference between the meanings expressed via these
two patterns is not perfectly clear to me, it seems that while the hati + ad-
jective pattern encodes an attributive relationship (as one would expect),
the adjective + hati pattern is more like a "locus" relationship, at least when
it designates a feeling, as in the examples listed in (15 a). That is, a pre-
posed adjective does not "describe" the hati, but rather it indicates a feeling
or attitude which is being located, so to speak, in the hati (somewhat simi-
lar in construction to English phrases such as sick at heart). The locus con-
struction with hati also seems somehow to "personalise" the state being de-
scribed. A second group of adjective + hati expressions seem to designate
moral aspects of a person's character. Some examples are listed in (15 b).
These might seem to be straightforwardly attributive, but my impression is
that the meaning structure is more subtle than this: that these expressions
are not describing the person's hati as such, but are describing the person
him or herself- by means of saying something about the hati (perhaps a bit
like English expressions such as pure of heart). More research is obviously
needed.
Examples of several semantic subcategories of the two patterns follow.
Notice that the adjectives involved also fall into several types. There are ex-
plicitly moral terms such as murni 'noble' and murah 'generous', there are
more general evaluative terms such as baik 'good', tulus 'pure', kotor
'dirty,' and busuk 'putrid', there are terms for sensation-states such as
panas 'hot', sejuk 'cold', sakit 'hurting', and puas 'sated', and there are
terms for physical dimensions and states such as besar 'big', kecil 'small',
lembut 'tender', keras 'hard', berat 'heavy', andputus 'broken'.
174 CliffGoddard
To round out this quick picture of the uses ofhati, it is instructive to look
at someperibahasa 'traditional sayings' andpepatah 'maxims'. Tradition-
ally, there were literally hundreds of peribahasa, which to a large extent
encapsulated the distinctive Malay adat 'customs, rules for living'. These
days many of the old peribahasa have fallen out of use, but even so scores
of them survive in everyday use. In (17) the reference to hau carries moral
overtones. The hati is the source of morally correct action, so better one's
eyes be blind than one's hati. (18) reflects the view that people's hati can be
reached by tutur yang manis 'sweet words', which can soften even the har-
dest hati. In example (19), however, the reference to hati is less favourable.
The sentiment is cautionary: we will suffer if we ikut hati 'follow hati'
without careful reflection, another persistent theme of Malay culture. Per-
haps the allusion here is primarily directed towards the role of the hati in
romantic matters. In (20), the expression dekat di hati 'close at hati' refers
to fondness: someone may be physically far away and yet be held in fond
remembrance. The expressions in (21)-(22) employ hati from yet another
angle, referring here to a general alignment of views and attitudes. People
who are sehati One hati' are all of the one mind; those who are hati Iain-
lain 'different hati' cannot get on harmoniously.
2. Meaning of hati
Three things should be evident at this point: (1) that hati is one of the key
ethnopsychological terms of the Malay language and culture; (2) that hati
has certain similarities with the English term heart', (3) that hati also differs
in important respects from English heart. The question now is: Is it poss-
ible, in a principled and rigorous fashion, to state the meanings of words
like hati and heart, so that the singularities and differences can be made be
explicit? The most important scholarly precedent bearing on this question
is Wierzbicka's (1992: 31-63) study of the Russian concept of dusa,
roughly, 'soul', and comparable concepts in several other European lan-
guages, such as Russian serdce, English heart, mind, and soul, and German
Seele.
For purposes of comparison, it is useful to consider Wierzbicka's (1992:
48) explication (below) of English heart. It is framed within the "natural
semantic metalanguage" (NSM) approach to semantic description, accord-
ing to which meanings are stated in the form of explanatory paraphrases
composed in a small, standardised and translatable metalanguage based on
natural language. For the purposes of this paper, it is not necessary to de-
scribe this methodology or the theory behind it in any detail
(cf. Wierzbicka 1996, Goddard and Wierzbicka eds, In press). A list of
NSM semantic primes, in English and Malay, is given in the Appendix. All
the explications should be readily translatable into Malay, cf. Goddard (In
press).
(23) heart
(a) a part of a person
(b) one cannot see it
(c) one can imagine that it is a part of the person's body
(d) (in the middle of the upper part of the body
(e) one can hear its movements)
(f) because of this part, a person can feel good things and bad
things
(g) because of this part, a person can feel good things towards
other people
It will be be helpful if we work through this explication line by line.
Lines (a)-(c) state that a heart is a 'part of a person', which we cannot see
but which we can 'imagine' to be part of the person's body. The wording
Hali: A key word in the Malay vocabulary of emotion 177
here is intended to capture the fact that although we recognise that heart (in
its emotional-moral sense) is not part of a person's body it can nonetheless
be thought of as such. Lines (d)-(e) indicate which particular physical part
is implied. The key components of the explication are, however, the final
two lines. Line (f) states that it is the heart which enables a person to 'feel
good things and bad things'. This reflects the fact that the heart is linked
only with "valenced" feelings such as joy, bitterness and sadness, and not
with neutral feelings such as surprise and interest, which make no refer-
ence to the concepts of 'good' and 'bad' (one can 'feel' surprise and inter-
est, but not in one's heart). Line (g) goes on to state more specifically that
the heart enables one to 'feel good things towards other people'. This re-
flects the fact that, so far as feelings directed towards other people are con-
cerned, the heart is specifically linked with the potential for positive feel-
ings (love, pity, admiration, etc.).
Since the above explication was originally proposed, there have been
some significant refinements to the natural semantic metalanguage theory,
so not surprisingly there are several matters of phrasing which now appear
less than optimal. In particular: (a) the term imagine is no longer regarded
as a viable semantic prime, (b) the expression feelings towards is now re-
cognised not to be a universal syntactic possibility for the primitive feel. In
addition, (c) it is preferable to avoid using the term and (as in good things
and bad things), since and is not a linguistic universal either. Rephrasing to
dispense with these features would make the heart explication somewhat
longer, but content-wise it is, in my opinion, substantially correct, and
forms a useful point of comparison for our account of Malay hau.
In view of the discussion of hau in the preceding section, I would pro-
pose the following explication.
Susah hati
Susah hati designates a troubled state of mind. This may be focused either
on oneself or on someone else one cares about. A person who is susah hati
is preoccupied with the idea that something bad and undesirable may
happen to someone. The experiencer feels under pressure to do something
to prevent it, but, unfortunately, he or she cannot think what to do. Hence
overall one is left in an unpleasant state of troubled preoccupation.
(28) Dia susah hati memikirkan anaknya yang pergi ke bandar menca-
ri kerja.
She was worried thinking about her son who'd gone to the city
looking for work.
Second, the range of worries which can induce susah hati is quite varied.
It could be a potential problem of just about any kind, and, as shown in
examples (29)-(31), this potential problem may concern either oneself or
someone else one cares about, e.g. a family member. For this reason, com-
ponent (b) is worded in deliberately general and open-ended fashion. The
explication depicts someone who is susah hati as dwelling on the thought
that 'something bad can happen to someone' (where 'someone' could in-
clude oneself). This is a prospect which, as set out in component (c), the
experiencer wishes not to happen. (Note that in the context of (29), the old
Hati: A key word in the Malay vocabulary of emotion 181
man was definitely ruminating on what was going to happen to him. His ill-
fated life was evidence for him that there was worse to come.)
(29) Lelaki tua semakin susah hau. Dia sentiasa mengeluh mengen-
angkan nasibnya yang malang.
'The old man got more and more troubled. He was forever sighing
as he recalled his misfortune.'
(30) Perkara ini akan say a selidiki nanti. Bersabarlah, dan jangan
susah hati.
'I'll sort this matter out directly. Be patient, and don't worry.'
(31) Bila dengar Sarha cakap macam tu, say a susah hati takut Sarha
akan bunuh din.
'When I heard Sarha talking like that, I was susah hati fearing that
Sarha would kill herself.'
Given the experiencer's awareness of the possibility of a bad and un-
wanted event befalling someone, component (d) spells out the consequent
sense of responsibility for finding a solution or resolution ( have to do
something'). Component (e) states that, for the time being, at least, the ex-
periencer is at a loss about what to do ( can't think what to do'). These
components are not as readily or directly illustrated by means of textual
examples, though they emerge clearly from consultations with native
speakers. The following two examples do bear indirectly on compo-
nents (d) and (e), however. In (32) the protagonist is a village girl who has,
for want of money, fallen behind in her studies at a college in the city. The
sight of her books reminds her both of difficulties and of her obligations.
Example (33) is advice being given to a girl who is having problems cop-
ing with the task of looking after her two mischievous younger siblings.
The author (an advice columnist, Kak Nor) assures the girl that the major
responsibility for disciplining the kids lies with the parents. It is enough for
her, as older sister, to give her younger siblings advice to the best of her
ability. The essence of Kak Nor's counsel is to release the troubled young
182 CliffGoddard
writer from her state of susah hati by releasing her from the pressure of
sole responsibility.
(33) Oleh yang demikian adik jangan susah hati. Memadai dengan
memberi nasihat ala kadar.
'Given that this is the case, you (lit. younger sibling) needn't be too
troubled (susah hati). It's enough to give (them) advice to the best
of (your) ability.'
Senang hati
The expression senang hati, roughly 'contented, having peace of mind',
has been identified by some commentators as designating a traditional
Malay social ideal. Djamour (1965: 145-6) says it would be difficult to
overemphasise its importance: "it permeates all fields of human behaviour.
When a Malay contemplates a change of residence, marriage, divorce, or
the exercise of a profession, the primary consideration is... serenity of
mind". Djamour and others (e.g. Maeda 1975) note that Malays would
often rather disregard grievances or dissatisfactions than disturb their
"peace of mind".
I would advance the following explication, much of which is, in effect,
the converse of susah hati. Whereas someone who is susah hati is con-
cerned about possible misfortune (thinking 'something bad can happen to
someone; I don't want this') and feels under pressure to act (thinking
have to do something'), the person who is senang hati is free from such
concerns and pressure. There is also a positive dimension, stated in com-
ponent (e). The person who is senang hati has a sense of being able to do
what he or she wants. This cognitive state brings about a 'good feeling'.4
(35) Anak rusa berasa senang hati kerana tuannya sentiasa menjag-
anya daripada dibunuh anjing-anjing.
The young deer felt very secure (senang hati) because his master
was always watching over him to protect him from being killed by
dogs (from a children's story).'
(36) Dia senang hati bila mendapat tahu anaknya sudah tiba di rumah
sepupunya dengan selamat.
'He was relieved (senang hati) when he found out his child had ar-
rived at his cousin's house safely.'
As suggested by these examples, and also by the following one, the ex-
pression senang hati is typically found in situations which focus on being
free of particular concerns or troubles.
Notice that the expression tidakltak ('not') senang hau is often used as a
euphemism to refer to emotions and reactions which are presumably more
serious in reality than simply an absence of senang hau. There is nothing
particularly surprising about this (much the same happens with the English
word happy, as the translations below illustrate). Example (39) is address-
ed by a girl to a persistent, unwelcome suitor. In example (40) a divorcee is
relating how people reacted when, after some years of financial hardship,
she was finally able to afford to smarten up her image.
(39) Sungguh aku tak senang hati setiap detik melihat tingkah laku kau
yang menjemukan itu.
'Really I'm not happy every time I see the tiresome way you carry
on.'
(40) Tetapi ada orang tidak senang hati, kalau orang perempuan,
mereka mula memikirkan mungkin saya melaram untuk memer-
angkap suami mereka. Bagi orang lelakipula mereka berpendapat
mungkin saya memang hendak menarik perhatian mereka.
'But there some people who aren't happy (with me), if they're
women, they've begun to think that maybe I am showing off with a
view to snaring their husbands. For the men, they think maybe I am
trying to attract their attentions.'
Sakit hati
The primary meaning of sakit is 'in pain, sick, ill', as in expressions like
sakit kepala 'head ache' (kepala 'head') and sakit kuning 'diabetes' (kun-
ing 'urine'), so on the basis of their English counterparts one might im-
agine that sakit hati is something like English sick at heart. But this would
be mistaken: sakit hati may perhaps imply something like pain, but nothing
like incapacity. If anything, it has a rather active flavour, suggesting the
possibility of some kind of reprisal. An interesting interpretation of sakit
hati comes from an old medical work (Ellis 1893) on the amuk syndrome,
i.e., the insane 'killing frenzy' to which Malays occasionally fall victim.
Ellis (cited in Winzeler 1990: 106) makes the observation that the mental
state of a person preceding an attack was invariably described, by Malays,
as sakit hati, a state which he describes as involving 'depression and brood-
ing over wrongs or supposed wrongs'.
Hati: A key word in the Malay vocabulary of emotion 185
(42) Paling menyakitkan hati dia tuduh kita curi duit dia, macamlah
kita ini pencuri besar.
'What really makes us mad (causes sakit hati) was him accusing us
of stealing his money, as though we were a terrible thief.'
(43) Panas telinga Mira mendengar sindiran Razi. Tetapi, bersabarlah
wahai hati, biarlah dia hendak menyindir atau menyakitkan hati
aku, biarlah, asal dia bahagia.
'Mira's ears burnt listening to Razi's needling. But, be patient hati,
let him if he wants to needle me or pain me (causes sakit hati), let
him, as long as he's happy.'
Still with verbal stimuli, (44) is an interesting example. In this case, what
provokes the protagonist's sakit hati is the way his rival is publicly praised
by the manager. Although the speech-act in this case is not actually about
the experiencer, it still reflects on his performance and on his standing in
the eyes of his superior, the manager (if the manager thought him the equal
of his rival, why not praise him at the same time?).
(44) Menurut pengarah, Hasnul seorang pekerja yang baik dan aman-
ah. Lebih menyakitkan hatinya pula, apabila pengarah tanpa
segan-silu memuji kebaikan Hasnul di hadapannya.
'According to the manager, Hasnul was a good and honest worker.
That only further inflamed his hatred (of Hasnul), especially when
the manager praised Hasnul in front of him.'
The following examples show that sakit hati can be induced by non-ver-
bal events - by a social event such as divorce and its attendant compli-
cations, as in (45), or a physical event such as carelessly bumping into
someone, as in (46).
(45) Lagipun saya rasa lebih baik bercerai waktu muda, hati tak sakit
sangat, kalau dah tua nanti lebih menyakitkan hati.
'Besides I reckon it's better to divorce when you're still young, it
doesn't make (one) so bitter, but if you're already old it makes one
more bitter.'
(46) Mohamad sakit hati melihat Mahmud yang melanggarnya tanpa
sebab. (Zakaria Salleh 1998: 18)
'Mohamad was annoyed seeing that Mahmud had bumped into
him without any reason.'
Hati: A key word in the Malay vocabulary of emotion 187
Puas hati
The expression puas hati designates a good feeling resulting from some-
one's behaviour, when this behaviour is something that the experiencer has
wanted for some time.
(48) ... apa yang say a buat tak kena pada padangannya. Hinggakan
cara saya gosok bajupun dia tak puas hati dan tak kena pada mat-
any a.
'...nothing I did was any good in his eyes. Even to the point where
the way I ironed a shirt didn't please him and wasn't good enough.'
(49) Dia berasa puas hati dengan kerajinan Jali. Orangnya agak lurus
dan raj in.
'She felt pleased with Jali's diligent (work). He seemed an honest
and hard-working sort of person.'
(50) "Cakaplah! Cakap apa sahaja yang kamu semua tidak puas
hati!" sambung Tuan Setiausaha lagi.
'"Speak up! Say whatever, whoever isn't contented!", added the
Chief Secretary.'
(51) Saya tidak puas hati hingga saya menyelesaikan kerja saya.
won't be satisfied until I complete my work.'
Generally speaking, the action or behaviour which gives rise to the puas
hati reaction can itself be described as good - but not always. In
example (52), the speaker is expressing her pleasure at the suffering of her
rival. The explication is worded so as to accommodate such uses; it does
not describe the person's behaviour as, so to speak, good in itself, but
merely as something which the experiencer wants.
188 Cliff Goddard
(52) Biar dia cemburu, biar dia marah, biar dia tak tidur malam...
biarkan. Barupuas hati aku!
'Let her be jealous, let her be angry, let her not sleep at night... so
be it. Finally I'm content.'
Kecil hati
When someone unexpectedly does something (typically, but not necess-
arily, says something) and this makes us realise that this person doesn't ac-
tually think much of us, the unpleasant feeling this induces is the feeling of
kecil (lit. 'small') hati.
(54) Saya rasa begitu kecil hati dan tak sanggup melihat mereka ker-
ana setiap kali berjumpa mereka akan mengeluarkan perkataan
yang bukan-bukan tentu juga kita rasa kecil hati.
feel so hurt and I don't want to see them because every time we
meet they come up with false allegations (against me). Of course
one (lit. we) is hurt.'
Hati: A key word in the Malay vocabulary of emotion 189
(55) "Itulah sebabnya aku tanya kau! Adakah dia bercuti saja atau dah
berhenti kerja. Sekurang-kurangnya kalau dia berhenti, hilanglah
satu tugas aku untuk bertegang urat dengan dia!". Menggelegak
dada Ezza mendengar jawapan Razi. Kalaulah Mira dengar kata-
kata Razi itu, tentu kecil hati Mira. Kesian Mira.
'"That's why I asked you! Has she just gone on a holiday or has
she quit. At least if she's quit, it'll be the end of my job of tussling
with her!" Ezza's blood boiled when she heard Razi's reply. If
Mira had heard Razi say that, she'd have been so hurt. Poor Mira.'
Staying with the wording of component (b), it may be asked why it does
not contain the specification that what was done or said was 'something
bad'. The reason for preferring the more general phrasing is that kecil hati
can result from actions which are not necessarily being seen as bad in
themselves. For example, in (56) what Pak Ngah Russin did to cause the
speaker's mother to feel kecil hati was to turn her out of his house, telling
her that he was not willing to help her (despite being her neighbour and
much better off than she). The same point is apparent from example (57).
The key thing about the triggering stimulus is not that it is bad in itself but
simply that it makes us realise that the speaker does not hold us in high re-
gard after all; hence, component (d) know now that this person doesn't
think good things about me'.
(56) Nol, Pak Ngah Hussin tidak ben kita pijak tangga rumahnya lagi.
Lagipun emak melarang kita ke sana. Kau tahu Nol, emak benar-
benar kecil hati.
'Nol, Pak Ngah Hussin won't let us step foot in his house again.
Plus, mother has forbidden us to go there. You know Nol, mother
was really hurt.'
(57) Ini mesti ada cerita best kat panggung ni, sikit pun tak nak sound
aku... ini yang kecik hati aku ni.
'This (film) must be the best story in the cinema at the moment, but
they didn't even say a word to me (about their plan to go)... that's
what hurt me so.'
4. Concluding discussion
What lessons can be drawn from this brief inquiry into the Malay hati? One
lesson is that apparent translation equivalents must be handled with great
care. In a sense, this is a familiar adage, at least to those working in cross-
linguistic semantics. Numerous studies have shown that seemingly "basic"
English words such happiness, anger, fear, love, and pride do not have pre-
cise equivalents in other languages, and, conversely, that the apparently
"basic" emotion vocabulary of other languages often resists straightfor-
ward translation into English. By and large, however, the studies which
have demonstrated this fact have focused on primary (i.e. monolexemic)
emotion terms. There is still a tendency, at least in some quarters, to as-
sume that secondary emotion terms, especially those involving descriptive
words such as 'sick', 'small', or 'soft', can be understood in a transparent
fashion. The present study has shown that any such view would be mis-
taken. For example, the meaning of the Malay expression sakit hati (lit.
sick/pained heart) cannot be deduced, on the basis of an English speaker's
semantic intuitions, from the combination of sick and heart, e.g. sick at
heart or heartsick. Similarly, it would be incorrect to assume that kecil hati
designates either fear or stinginess, although these are the interpretations
which the "translation" small heart tends to suggest to English speakers.
I do not want to suggest that the meanings of "figurative" expressions
like these vary without limit between languages. No doubt there are some
cross-linguistic tendencies to be discovered; I doubt, for example, that in
any language an expression corresponding to 'small heart' would designate
anything like bravery, generosity, or joy. But to establish these tendencies,
Hati: A key word in the Malay vocabulary of emotion 191
APPENDIX:
Semantic Primes - English and Malay Exponents
Substantives Time
I-AKU, - WHEN/TIME-BILA/MASA(WAKTU)
SOMEOWE-SESEORANG NOW-SEKARANG
SOMETHING-SESi/ArC/ BEFORE-SEBELUM, AFTER-SELEPAS
PEOPLE-ORANG A LONG TIME-LAMA, A SHORT TIME-
SEKEJAP
FOR SOME -BEBERAPA LAMA
192 Cliff Goddard
Notes
1 . For helpful comments and advice I would like to thank Nick Enfield, Hazidi Abdul
Hamid, participants in the Symposium on The Semantics of Emotions held at Aus-
tralian National University in 1999, and the editors and reviewers of this volume.
2. Even though its anatomical sense is 'liver', hati is not altogether unconnected with
the heart. For example, the verb debar 'beating of the heart' can be combined with
hati, in the expression berdebar hati, roughly, 'to thrill with feeling, to have one's
heart throb with feeling'.
3. This is not as straightforward an assumption as it may seem. The complicating fac-
tor is that one may form any number of causative expressions which have hati as ob-
Hati: A key word in the Malay vocabulary of emotion 193
ject, without the need for any comparable fixed expression. For example, one may
form the expressions menggembirakan hati (from gembira 'happy, thrilled') and
membahagiakan hati (from bahagia 'happy, content in life'). These look exactly
parallel with menyenangkan hati but there are no fixed expressions *gembira hati
and *bahagia hati, corresponding to senang hati. Nor (consequently) could you de-
scribe a person as *bergembira hati or *berbahagia hati, though one can describe a
person as bersenang hati. In view of these facts, it is not necessarily the case that ex-
pressions like menyenangkan hati and memuaskan hati (for example) are based di-
rectly on the fixed expressions senang hati andpuas hati. Even so, semantically the
alignment does seem to hold; that is, menyenangkan hati means 'to make someone
senang hati', memuaskan hati means 'to make someone puas hati'.
4. The meaning is different if senang 'at ease, comfortable' is applied attributively; for
example, Hati saya tak berapa senang 'My hati isn't very at-ease'. This usage sug-
gests the transient state of the experiencer's hati, and would be appropriate in repor-
ting a reaction to an event. For example, in (a) below the protagonist is being faced
with unwelcome questioning from an older person (questions which, given Malay
cultural values, could not be ignored). In example (b), the protagonist is a child. The
disturbance to his senang hati is induced by his joining Koran-reading classes for
the first time, and encountering his new strict teacher Tuan Syed.
(58) a. "Siapa? Saya?" tanya Hilmy, memperhatikan muka orang lua itu tepat-
tepat. Hatinya tidak begitu senang ditanya-tanya begitu waktu sedang
sibuk-sibuk bekerja.
' "Who? Me?" retorted Hilmy looking fixedly at the face of the old man. He
didn't feel too comfortable being questioned while he was busy working.'
b. Hati teman tak berapa senang juga melihat Tuan Syed buat kali pertama-
petang tu.
was not very happy to see Tuan Syed for the first time that afternoon.'
References
Abdullah Hussain
1990 Kamus Simpulan Bahasa. Kuala Lumpur: Dewan Bahasa dan Pustaka.
Awang Sudjai Hairul and Yusoff Khan
1990 Kamus Lengkap. Petaling Jaya: Pustaka Zaman.
Coope, A. E.
1991 Malay-English, English-Malay Dictionary. (Revised edition). Kuala
Lumpur: Macmillan.
Diller, Anthony V. N. and Preecha Juntanmalaga
1990 "Full hearts" and empty pronominals in Thai. Australian Journal of
Linguistics 19(2): 231-256.
194 Cliff Goddard
Djamour, Judith
1965 [1959] Malay Kinship and Marriage in Singapore. London: The Athlone
Press.
Goddard, Cliff
1995 'Cognitive mapping' or 'verbal explication'?: Understanding love on
the Malay archipelago. Semiotica 106(3/4): 323-354.
1996 The "social emotions" of Malay (Bahasa Melayu). Ethos 24(3):
426-464.
1997 a Cultural values and "cultural scripts" of Malay (Bahasa Melayu).
Journal of Pragmatics 27(2): 183-201.
1997b Contrastive semantics and cultural psychology: 'Surprise' in Malay
and English. Culture and Pyschology 3(2): 153-181.
In press The syntax of universal semantic primes in Malay (Bahasa Melayu).
In: Cliff Goddard and Anna Wierzbicka (eds.), Meaning and Universal
Grammar: Theory and Empirical Findings. Amsterdam: John Benja-
mins.
Goddard, Cliff and Wierzbicka, Anna (eds.)
In press Meaning and Universal Grammar: Theory and Empirical Findings.
Amsterdam: John Benjamins.
Haji Abdul Rahman bin Yusof
1994 Bahasa Malaysia-English, English-Bahasa Malaysia Dictionary.
(Revised edition.) Glasgow: Harper Collins.
Heider, Karl G.
1991 Landscapes of Emotion. Mapping Three Cultures of Emotion in In-
donesia. Cambridge: Cambridge University Press.
Karim, Wazir Jahan (ed.)
1990 Emotions of Culture. A Malay Perspective. Singapore: Oxford Univer-
sity Press.
Maeda, Narifumi
1975 Family Circle, Community, and Nation in Malaysia. Current Anthro-
pology 16(1): 163-166.
Matisoff, James A.
1986 Hearts and minds in Southeast Asian languages and English: An essay
in the comparative lexical semantics of psycho-collocations. Cahiers
de Linguistique Asie Orientale 15(1): 557.
Oey, Eric M.
1990 'Psycho-collocations' in Malay: A Southeast Asian areal feature. Lin-
guistics of the Tibeto-Burman Area 13(1): 141-158.
Peletz, Michael G.
1996 Reason and Passion: Representations of Gender in a Malay Society.
Berkeley: University of California Press.
Hati: A key word in the Malay vocabulary of emotion 195
Wierzbicka, Anna
1992 Semantics, Culture, and Cognition: Universal human concepts in cul-
ture-specific configurations. New York: Oxford University Press.
1996 Semantics: Primes and Universals. Oxford: Oxford University Press.
1997 Understanding Cultures through their Key Words: English, Russian,
Polish, German, Japanese. New York: Oxford University Press.
1999 Emotions Across Languages and Cultures: Diversity and universals.
Cambridge: Cambridge University Press.
Wilson, Peter J.
1967 A Malay Village and Malaysia. New Haven: Hraf Press.
Winzeler, Robert.
1990 Amuk: Historical, Psychological, and Cultural Perspectives. In Wazir
Jahan Karim (ed.), 96-122.
Zakaria Salleh
1988 Simpulan Bahasa Praktik. Kuala Lumpur: Berita Publishing.
Talking about anger in Central Australia
Jean Harkins
there will always be individuals and groups who have lesser or greater in-
terest in such matters than their fellows.
This exploration of a small group of emotion expressions in the Arrernte
language of Central Australia1 takes the view that it is useful and illuminat-
ing, when investigating emotional or other meanings, to cultivate aware-
ness of the cultural and intercultural dimensions of the enterprise, and to
see both the processes and outcomes of the investigation in this light. Aim-
ing for complete scientific objectivity in a culturally situated study would
be not only foolish, but less informative than taking into account the cul-
tural and linguistic realities that present themselves in the course of such
investigation. However, aiming for interculturally comprehensible ways of
expressing what people mean is both practicable and worthwhile.
the etymology of emotion, from e- Outward' and motion, isn't salient for
most English speakers). Awelheme 'feeling' is intuitively related more to
sensations in the middle part of the body than the ears, despite the attract-
iveness of the image of listening to one's own inward state (cf. also God-
dard 1989: 54). Participants referred to the topic of our discussions as ang-
ketye tyerrtye awelhe-me akerte 'word people feel-NPP PROP', which could
be glossed as 'words for people's feelings', or more formally, 'linguistic
expressions of emotion'.
Awelhe- 'feel' refers to physical sensations such as heat, cold, pain, as
well as to emotional or cognitively-based feelings, but not to intentionally
feeling an object (cf. the transitive verb anpe-me 'touch-Npp'). "Feeling
something good/bad" can be physical or emotional:
rough glosses give some indication of the meanings, but for many of these
words there is no adequate English equivalent.
Each of these categories has many more associated words and phrases,
and each could be the focus of an extended study in cultural semantics. Our
202 Jean Harkins
preliminary analysis considered just a few of the good feelings (2 a), and
anger-like feelings (2 d) which are the focus of the present discussion. Wil-
kins (To appear) has examined atere and apure (2f) in depth, and Harkins
(1995; Harkins and Wilkins 1994: 291) discusses ahentye 'want'.
Here we see the same word ahele-irre- (with negative verbal suffix) trans-
lated using the English words angry and cheeky respectively. This reflects
the Aboriginal English usage, where hostile could be a closer non-Aborig-
inal English equivalent. In Aboriginal English, cheeky often refers not to
casual insolence but to a serious potential for violence or harm (Harkins
1994:155) - a semantic difference that can produce significant intercultural
misunderstanding in situations of actual or potential interpersonal conflict.
This is probably the most frequently occurring anger-like word in Ar-
rernte, with a fairly broad semantic scope. While the base, ahe- 'anger', is
almost certainly etymologically related to 'throat' (as is ahentye 'want',
Harkins 1995: 46-54), synchronically its semantic content doesn't include
any body part reference (nor does the probable derivation of English anger
from Old Norse 'narrow' form part of its current semantic composition).
The multifunctional inflectional suffix -le has ergative, locative, instrumen-
tal and other functions in addition to its function here as a manner adverb
formative (Wilkins 1989: 176). This is followed by the inchoative deri-
vational suffix -irre, whose occurrence with many emotion predicates (in-
cluding ahentye 'want', apure 'shame', at ere 'fear' and others) may signal
a conceptualisation of them as processes; this matter requires some further
study (cf. Harkins 1995: 51, Wilkins To appear).
Along with this word, four other anger-like words were selected for sem-
antic comparison and analysis. These are listed below, with English glosses
from Henderson and Dobson (1994):
(5) a. ahele-irreme
'get angry, be angry at someone'
b. arnkelye
'not talking to people because you're unhappy
about something; sulking, in a sulk'
c. akiwarre
'showing anger; in a tantrum, sulking,
having lost your temper, pissed off,
flying off the handle'
204 Jean Harkins
d. ahakngerre
'aggressive, angry, violent, tough, mean,
savage, vicious, dangerous, touchy. In local
Aboriginal English, "cheeky" has this meaning,
and not its standard English meaning.'
e. atnerte ampeme
' 1. have a burning feeling in the stomach,
usually from hunger. 2. be really angry.'
The English glosses give an indication of the various ways these words
tend to be translated in different contexts; all of them are sometimes trans-
lated as angry, but obviously there are important differences in meaning
among them. The Aboriginal English cheeky is associated here with
ahakngerre, while in (4) above it was used to translate ahele-irreme. In
(5 b) and (5 c), sulk is also used in its Aboriginal English sense (cf. Harkins
1994: 157). Non-Aboriginal English speakers can fail to appreciate the
seriousness of this emotional state, because in non-Aboriginal usage sulk-
ing tends to be viewed as rather petty and self-indulgent, something that a
person can and should fairly easily "get over", while arnkelye is more like a
prolonged state of brooding resentment, fuming over some significant of-
fence, and carries a real danger of erupting into violence.
The usual linguistic method of collecting examples of each word in a
variety of natural utterance contexts was followed, aiming for at least twen-
ty examples of each in order to observe its range of use as broadly as poss-
ible, and to consider what elements of meaning were common to all uses of
the word. Some examples, like (3) and (4) above, are seen in written Ar-
rernte texts, but most of the available writings were factual or event-
oriented narratives, containing less emotion vocabulary than is found in
everyday verbal interaction. Ordinary family and community life, on the
other hand, provided a wealth of examples every day. Often, more than one
of the words under investigation were found in the same utterance, indicat-
ing that each contributes some meaning more specific than a generalised
angry feeling, for example:
atwe-tyenhe-le-athathe re ane-pe-ane-me?
hit-Npc-ss-before 3sS be-FREQ-RDP-NPP
'Is that person building up anger inside, heading for a fight?'
The picture here is of a person who is really "pissed off' (akiwarre) at
someone for something they've done, fuming (arnkelye) continuously over
it and working herself up into a real temper (ahe), likely to be vented in a
fight (i.e. hitting somebody).
As the examples were assembled, it soon became apparent that each one
stemmed from some real life incident, a social context or scenario in which
the emotion was triggered by some event or action. Moreover, the emo-
tional response - whether a person felt akiwarre or arnkelye or ahele-ir-
reme or atnerte ampeme or more than one of these feelings, or whether on
the other hand they didn't get angry or upset over it - depended largely on
how the person "took" the situation, that is, how they interpreted it, or
thought about it. Trying to tease out the fairly subtle differences in meaning
and use between arnkelye, akiwarre and ahele-irreme, in particular, thus
provided a startling degree of support for the view that these emotions are
cognitively based, from the viewpoint of Arrernte social perception as well
as that of NSM semantic theory.
Some examples highlighting the nature of each of these three words may
be helpful here (recalling also those given above). The translations are
those provided by the contributors of the examples, containing some of the
Aboriginal English usages mentioned earlier.
'The two of them are sitting facing away from each other and not
talking to each other. They're sulking because they had a fight.'
(9) Re akiwarre-irre-me arelhe arrpenhe-le
3sS infuriated-iNCH-NPP woman other-ERG
renhe terangke ake-ke-le-iperre.
3sO drunk call-pc-ss-after
'He lost his temper because that woman said he was a drunk.'
(10) Ayenge ngkwenge akiwarre ake-lhe-me-le ilelhe-me!
IsS 2SDAT infuriated call-REFL-Npp-ss admit-NPP
'I'm telling you I'm really pissed off with you!'
(11) Ahe-le-irre-tyele-aye, urreke imp-0-aye!
anger-ADV-iNCH-NegiMP-EMPH later leave-iMP-EMPH
'Don't get angry, leave it for later!'
(12) Artwe ampwe nhakwe ahe-le-irre-me
man old that anger-ADV-iNCH-Npp
renhe arle ampe urreye mape-le
3sO THAT child male PL(g)-ERG
ingwe-le apwerte-le we-ke-nge.
night-LOC stone-iNST throw-pc-ABL
'That old man there is really angry because some young boys
threw stones at him last night.'
As each example and the real life scenario it came from - the ayeye
'story' behind it - were discussed, the participants were able to make some
generalisations about what all these situations had in common. For each
word, these took the form of ayeye akweke angketye nhenhe-ke 'story small
word mis-DAT', that is, a little story, script or mini-narrative for each word.
These ayeye akweke, arrived at by discussion among the speakers, show a
surprising degree of convergence with the kinds of semantic explications
produced through more formal NSM analysis.
Perhaps even more surprisingly, participants wanted to use false initials
to represent the protagonists in each scenario and ayeye akweke, for rea-
sons of anonymity. In a close-knit community, it could be undesirably ob-
vious to any reader of our examples which real individuals, family and
friends, had been involved in each of the disputes that produced the
examples, and we could be accused of gossiping about these people instead
of getting on with our linguistic work. Hence the familiar X and of other
Talking about anger in Central Australia 207
NSM explications came into play, though for reasons that had more to do
with social realities than with issues of indexicality in semantic represen-
tation (cf. Introduction to this volume, p. 20).
Thus, a tentative ayeye akweke for arnkelye-irreme took the following
form:
(13) X arnkelye-irre-me
'X is getting arnkelye'
a. X re awelhe-me-le itirre-ke Y-ke alakenhe
X 3sS feel-NPP-ss think-pc Y-DAT thus
'X is feeling something because X thought like this about Y'
b. Y-le akenge mpware-ke-nge
Y-ERG rotten make/do-pc-os
did something very bad'
c. the akurne ile-tyenhenge renhe ikwere-nge
IsA bad do-SBSQ 3sO SSDAT-ABL
will do something bad to because of this'
d. X alakenhe itirre-me-le, X akurne awelhe-me Y-ke
X thus think-NPP-ss X bad feel- Y-DAT
' when X is thinking like this, X feels something bad towards Y'
An equally tentative ayeye akweke for akiwarre has a similar format and
is also directed at a specific offender (Y), but has a slightly different set of
typical thoughts, indicating that the experiencer is personally affected by
the offending action, is indeed too angry for words, and intends to retaliate,
though not as immediately as in the case of arnkelye:
(14) X akiwarre
'X is akiwarre'
a. X re awelhe-me-le itirre-ke Y-ke alakenhe
X 3sS feel-NPP-ss think-pc Y-DAT thus
'X is feeling something because X thought like this about Y'
b. Y-le akenge mpware-ke-nge atyinhe-ke
Y-ERG rotten make/do-pc-os ISPOSS-DAT
did something very bad to something of mine'
c. ayenge ahentye kwenye angke-tyeke nhenhe ikwere-nge
IsS want NomNEG say-PURp this SSDAT-ABL
don't want to say anything because of this'
d. the urreke akurne ile-tyenhe Y renhe ikwere-nge
IsA later bad do-Npc Y 3sO SSDAT-ABL
will do something bad to Y later on because of this'
e. X alakenhe itirre-me-le, X akurne awelhe-me Y-ke
X thus think-Npp-ss X bad feel- Y-DAT
'when X is thinking like this, X feels something bad towards
Y'
The 'something of mine' in (b) can be one's personal property (e.g.
mutekaye atyinhe 'my car') one's family members (e.g. akngeye atyinhe
'my father'), one's reputation or feelings (examples 9-10 above) - anything
that the experiencer values enough to feel infuriated or severely aggrieved
by some offence against it. This feeling is often "bottled up" (component c,
which should perhaps also include anteme 'now'), and vengeful im-
pulses (d) can persist inwardly or find expression in a show of temper. In-
terestingly, although the evidence from (5 b) and (8) above suggests that
not talking is also associated with arnkelye, these speakers felt that it was
an essential element of akiwarre in a way that it wasn't for arnkelye. This is
an example of the kind of issue that may need to be explored further before
optimal definitions are arrived at. The connotation of uncommunicative-
ness probably also accounts for the use of the English word sulk in relation
to these much less petty emotional states; an explication for the Aboriginal
Talking about anger in Central Australia 209
English sulky might be closer to these Arrernte meanings than to the non-
Aboriginal English definition.
Another example scenario or ay eye 'story' also illustrates the kind of
contrast reflected in the ayeye akweke atherre 'two little stories' above:
(15) Re atherre ilte-rre-ke TV-ke arrkngele.
3S DU argue-RECiP-PC television-DAT reason
Marie re arnkelye arrate-le alhe-ke,
female 3sS resentful go.out-ss go-PC
kele re kethe-le anteme ane-tyeme.
then 3sS outside-LOC now sit-pp
Kele anewe ikwere akiwarre-le
then spouse SSPOSS infuriated-ss
apmere-nge alhe-ke akartne-nge.
home-ABL go-PC area-ABL
'The couple were arguing over the TV. The wife went out in a sulk
and was sitting outside. Then her husband stalked off in a really
bad temper.'
The third ayeye akweke, shown below, indicates that ahele-irreme is also
directed at a target person (a) who has done something that negatively af-
fects something important to the experiencer (b). Alhough the offending
action is not as strongly condemned (akurne 'bad' cf. akenge 'very bad,
rotten'), it is still unwanted (c). There is a desire for retaliation (d), but not
as concrete an intention as suggested in (13 c) and (14d) above.
(16) X ahele-irreme
'X is getting angry'
a. X re awelhe-me-le itirre-ke Y-ke alakenhe
X 3sS feel-NPP-ss think-pc Y-DAT thus
'X is feeling something because X thought like this about Y'
b. Y-le atyinhe akurne ile-ke
Y-ERG ISPOSS bad do-pc
did something bad to (something of) mine'
c. Y-ke ay enge ahentye kwenye alakenhe
Y-DAT IsS want NomNEG thus
mpware-tyeke atyinhe-ke
make/do-PURp ISPOSS-DAT
don't want to do (like) this to something of mine'
210 Jean Harkins
Interestingly, it appears that all three of these anger-like feelings are in-
herently interpersonal, directed at some specific offender. Participants in-
sisted that Y-ke -DAT' was a necessary element in the first and final com-
ponents of each of the three ayeye akweke. A non-personal grievance might
be more likely to be spoken of in terms of akurne awelheme (2 e above),
something like 'regret'. This is also in accord with speakers' intuitive cate-
gorisation of bad feelings towards someone (2 d) in a separate group from
feeling bad about something (2e) or as the result of some event (2c). The
Arrernte components proposed here differ in some important ways from
the anger-like emotions analysed by Goddard (1991, 1998), particularly in
focussing negative judgement on the offenders' actions rather than their
character, although there are equally important similarities, such as the in-
terpersonal emphasis (Y-ku -DAT'), and the desire for retaliation in mirp-
anarinyi but not kuyaringanyi.
Space does not permit a full analysis here of ahakngerre or atnerte am-
peme, but we can observe that ahakngerre, somewhat like the Western De-
sert pikaringanyi (Goddard 1991,1998), seems not to be an emotional feel-
ing but a general propensity to do something bad to someone. The fact that
ahakngerre, like Aboriginal English cheeky, can be predicated of non-hu-
mans and inanimates, such as snakes or poisonous plants, raises an inter-
esting problem of whether or not this should be represented as a metaphor-
like "wanting to do something bad to someone", or in terms of the possi-
bility that the person or thing can or will do something bad to someone.
And the emotional sense of atnerte ampeme, lit. 'stomach burning' (5e
above) raises issues of semantic representation of emotions described via
bodily sensations (Wierzbicka 1999: 295-297). Some of these issues are
discussed further in other papers in this volume.
The interpersonal feelings and transactions reflected in these examples
and partial explications occur within a broader sociocultural context,
Talking about anger in Central Australia 211
4. Practical semantics
This small study has found confirmation for several of the hypotheses
about emotional meanings put forward by Wierzbicka and other practi-
tioners of the NSM approach to semantic analysis, and has raised some
questions about other parts of this theory. But perhaps more importantly, it
has demonstrated something of the practicality of the NSM approach in fa-
cilitating intercultural discussion and understanding of people's emotional
life and behaviour in cultural context, and, furthermore, as a tool for stating
meanings in the language of inquiry. This is a practical tool that field-
212 Jean Harkins
workers and research participants can use in their own language and cul-
tural setting - a far cry from some popular notions of semantics as an ab-
stract, ivory-tower activity.
There was surprising convergence between these Arrernte perceptions
and the NSM picture of emotions as cognitively based feelings, and the
script-like nature of their meanings, with a series of typical thoughts giving
rise to the affective state. These anger-like feelings all contain the impulse
to act, proposed as a universal of anger-like feelings by Wierzbicka (1999:
288). The Arrernte ayeye akweke did not have the full prototype structure
for cognitively based feelings ("sometimes a person thinks...", etc.) pro-
posed by Wierzbicka (1999, see also our Introduction to this volume). A
first attempt to apply such a frame rendered the ayeye akweke unintelligible
in Arrernte, and it is not entirely clear how this problem could be resolved,
or whether it would be appropriate to do so. The papers by Wierzbicka and
by Goddard in this volume employ a similarly simplified frame for practi-
cal definitional purposes, and it may be that the full prototype structure is a
little too abstract for workable natural language definitions.
The outcomes of this modest study were of interest to the participants,
but it was clear that such a process of group discussion might be an unman-
ageably time-consuming way to produce Arrernte language definitions for
more than a handful of Arrernte words. Nonetheless, the intercultural in-
sights gained through the use of this method of analysing meanings, par-
ticularly for important social and cultural concepts, were valued by all con-
cerned. These outcomes are at least as important, indeed more so for most
practical puposes, as the artefacts of semantic formulae that were pro-
duced. The fact that a group of interested speakers could gain some satisfy-
ing insights through practical experimentation with this methodology is
promising, and suggests that the approach has much to offer in linguistic
fieldwork and other intercultural communiation applications.
For semantics to be of use outside of academic circles, it needs to offer
something that fairly ordinary people with an interest in meaning and com-
munication can do, and gain some benefit thereby. And for the academic
study of meaning and communication to make progress, it needs to move
beyond the hermeneutical circle of linguocentric discussion and represen-
tation of meanings. While some aspects of the NSM approach remain con-
troversial, many issues in semantic analysis will only be resolved through
the experiences of people analysing meanings in many different human
languages and cultural environments, and thus discovering what is and
Talking about anger in Central Australia 213
isn't valid, what does and doesn't work in describing linguistic and cultural
meanings. This study has shown just a little of how such investigation can
proceed in a practicable way, and it is hoped that others may also find this
kind of approach a useful and practical tool for learning more about emo-
tional and other meanings within and across languages and cultures.
Notes
1. This study was conducted in Alice Springs (Mparntwe), among speakers of Eastern
and Central (Mparntwe) Arremte (described by Green 1994, Henderson & Dobson
1994, Wilkins 1989), and was funded by the Australian Research Council through
the University of New England. The participants wish to thank the Institute for Abo-
riginal Development, the Arandic Dictionaries Project, and Pamela Ditton for prac-
tical assistance that made this work possible.
2. Abbreviations used in morphemic glosses (mostly following Wilkins' 1989 analy-
sis) are: 1, 2, 3[person]s[ingular]/p[lural]; ABLative; ABSolutive; ADVerb; Associ-
ative; Afgent]; CONTinuous; DATive; DS different subject; DUal; EMPHatic; ER-
Gative; FREQuent; IMPerative; INCHoative; INSTrumental; LOCative;
Neg[ative]IMPerative; NMZ nominaliser; Nom[inal]NEGator; NPC non-past com-
pletive; NPP non-past progressive; O[bject]; PC past completive; PLural(g[roup]);
POSSessive; PP past progressive; PURPosive; P[ast].imm[ediate]; QUOTative;
Q[uestion]; RDP reduplication; RECIProcal; REFLexive; SBSQ subsequent; SS same
subject; S[ubject]; THAT complement marker; V[er]bNEGator.
References
Enfield, Nicholas J.
2000 On linguocentrism. In: Martin Ptz and M. H. Verspoor (eds.), Explo-
rations in Linguistic Relativity, 125-157. Amsterdam: John Benja-
mins.
Fillmore, Charles H. and Beryl T. Atkins
1992 Toward a frame-based lexicon: The semantics of RISK and its neigh-
bours. In: Adrienne Lehrer and E. F. Kittay (eds.), Frames, Fields and
Contrasts. Hillsdale, NJ: Erlbaum.
Frijda, Nico H.
1986 The Emotions. Cambridge: Cambridge University Press.
Goddard, Cliff
1989 Issues in Natural Semantic Metalanguage. Quaderni di Semanti-
ca 10(1): 51-64.
1991 Anger in the Western Desert. Man 26: 602-619.
214 Jean Harkins
1. Introduction1
It has often been noted that the Japanese language is rich in sound-sym-
bolic words, which form a conspicuous group in the Japanese lexicon
(cf. Shibatani 1990: 153). Japanese onomatopoeic words are generally re-
ferred to as giongo-gitaigo, and divided into three classes. The first class is
phonomimes (giseigolgiongo), which imitate sounds. The second class is
phenomimes (gitaigo) which describe appearances, states, conditions of
the external world. The third class is psychomimes (gijoogo) which express
one's inner feelings or mental conditions. Phenomimes and psychomimes
are often called mimetic words, as opposed to phonomimes which are
called (sound-)onomatopoeia (cf. Shibatani 1990: 154; Martin 1964).
Japanese everyday conversation is full of these sound-symbolic words.
In contrast to the Indo-European languages whose sound-symbolic words
are mostly phonomimes (sound-onomatopoeia), Japanese has more mi-
metic words than sound-onomatopoeic words (cf. Tamamura 1982).2 It has
also been noted that Japanese is rich in psychomimes which describe vari-
ous emotion/sensation states.
While non-native speakers of the Japanese language will certainly face
difficulties in acquiring psychomimetic words, they need to master the cor-
rect use of them if they wish to communicate effectively with Japanese
people. Psychomimetic words play a vital role in Japanese everyday lan-
guage life, and thus they are an indispensable key for outsiders to under-
stand the Japanese people and culture.
218 Rie Hasada
These seven were chosen, because: (1) they are very commonly used, and
thus deserve to be studied; (2) they are, in various ways, related to each
other semantically and it is interesting and revealing to compare them; and
(3) they reflect important characteristics of the Japanese way of concep-
tualising emotions. The semantic metalanguage developed by Wierzbicka
and colleagues is employed as a tool to define these words. This metalan-
guage enables us to explicate complex concepts such as psychomimes by
means of simple words and grammar which are understandable to anyone,
and thus clarifies the similarities and dissimilarities between the compo-
nents involved in different psychomimes.
In order to comment on the variety of sound-symbolic vocabulary, data
have been drawn from a varied range of sources. Most of the Japanese
examples are from actual materials: newspapers, journals (mainly inter-
views), novels (most of which are written in the first person spoken style),
magazines, advertisements, TV commercials, popular songs, scholarly
papers, etc. I have also taken examples from English texts or books, in order
to show how Japanese sound-symbolic words are translated into English, as
well as what kind of English expressions have been translated into Japanese
sound-symbolic words. Examples are also taken from several dictionaries
in order to supplement my data, particularly from the JapaneseEnglish
Onomatopoeia and Mimesis Translation Dictionary [JEOMTD], whose
examples were all taken from novels or film scripts. The source is indicated
in abbreviation in square brackets at the end of each example. Details of ab-
breviations are given in the list of sources at the end of this paper. Unrefer-
enced Japanese examples are from the author who is a native speaker.
The subtle differences of meaning between psychomimes are often said
to be impossible to translate appropriately into a foreign language. Fur-
thermore, the differences between emotions, for example, between hiya-
hiya and biku-biku, which are usually identically explained as "be afraid"
or "be in fear" in Japanese-English dictionaries, are hard to explain even
for the Japanese. Nevertheless, these words are not synonymous and Jap-
anese people use them distinctively in different situations.
As illustrated above, adequate explanations for the use and nuances of
these words are rarely found in dictionaries. Although it is the purpose of
dictionaries to give a definition which captures the invariable semantic as-
pects of a word, they do not usually succeed in doing this. Many Japanese-
English dictionaries often do not even try to give a definition, but instead
only give a single English word or a list of possibly analogous words. For
220 Rie Hasada
Neither vague analogous words from another language, nor lengthy ex-
planations, nor accumulations of sample examples can show us clearly and
precisely the unique nature or concept of Japanese psychomimes.
If we want to define Japanese psychomimes in a way which is truly free
of ethnocentrism and explanatory, we can do this using explicit semantic
formulas, couched in terms of universal semantic primes (cf. Wierzbicka
1986: 530; 1992a). In this approach, a natural semantic metalanguage
(NSM) is used as a tool for explicating the meanings of words in any lan-
guage. These words are "maximally clear, maximally simple, and maxi-
mally universal" in the sense of having absolute equivalents in all the lan-
guages of the world, and thus are intuitively understandable (non-techni-
cal). By using these words which themselves are not names of emotions in
specific languages, we can also avoid ethnocentric 'definitions' of words.
In this paper, I attempt to explicate the meaning of Japanese psycho-
mimes by using this metalanguage. In my explication of Japanese psycho-
mimes, however, in order to avoid long and unwieldy definitions, I have
used an expanded list, including words which, although not universal,
recur widely in the languages of the world as separate lexical items.
As a framework for the explication of psychomimes, I will employ the
prototype scenario proposed by Wierzbicka (1992c: 540). This explicates
the meaning of emotions by means of the framework showing "to feel a
certain emotion means to feel like a person does who has certain (specifi-
able) thoughts, characteristic of that particular situation (and to undergo
some internal process because of this)" (Wierzbicka 1994: 437). Since the
whole range of emotions which psychomimes can represent is too large to
be analysed in this paper, the discussion is limited to seven representative
psychomimes which are frequently used. They are ordered and grouped in
such a way as to show the differences and similarities of meaning between
loosely related concepts.
Three of them, representing momentary mental activity, will be dis-
cussed in section 2. They are gyoQ, haQ, dokiQ for something like 'sur-
prise' feelings. Although there is a standard verb odoroku or a quasi-sound-
symbolic verb bikkuri-suru for the expressions of surprise in Japanese, the
Japanese language possesses those variant psychomimes such as gyoQ,
dokiQ, haQ, which describe the exact moment of being surprised.
One of the characteristics of Japanese psychomimes is that many of them
represent restless emotions. In section 3,1 will examine four psychomimes
which represent continuous restless emotional activity. They are hara-
222 Rie Hasada
hara, hiya-hiya, biku-biku and odo-odo, for bad feelings about an antici-
pated bad happening.
Hara-hara, hiya-hiya, biku-biku, and odo-odo, are often translated or
glossed as "nervous" in Japanese-English dictionaries. According to Mori-
ta, a particular type of neurosis called shinkeishitsu or shinkeishitsu-shoo,
which is translated as "nervosity" or "nervosis", is relatively common
among Japanese (Lebra 1976: 216). Fujita (in Lebra 1976: 217-8) states
that the shinkeishitsu patient is torn apart by strong anxiety, tenseness, and
fear on the one hand, and by the desire to overcome these and to participate
in human interaction, on the other. Psychomimes such as hara-hara, hiya-
hiya, biku-biku, odo-odo all connote some bad feelings, which are related
to this shinkeishitsu-shoo.
The meaning of these psychomimes cannot be explained by English
emotion terms, since they are indeed Japanese language-specific. Also,
while psychomimes of the same group seem to have almost identical mean-
ings, in fact they do mean different things and function differently in Jap-
anese. When they are translated into the universal semantic metalanguage,
the similarities and the differences between those concepts become appar-
ent, and the meaning of each psychomime becomes comprehensible to out-
siders.
2. Momentary emotions
HaQ, gyoQ, dokiQ are psychomimes denoting something like the feeling of
'surprise' about an unexpected happening. The quality of immediate reac-
tion involved in all of these psychomimes is expressed by the component
"this person feels something for a very short time".
While, for example, Japanese-English dictionaries give the English
word "startled" for the meaning of Japanese psychomime haQ, "startled"
cannot always occur in the context where haQ occurs. For instance, haQ
cannot be replaced by "startled" in the following sentence: Natsukashii
kyoku kiite haQ to shitari... feel haQ, hearing the music of old times'.
Moreover, dictionaries often offer one English emotion word for more than
one Japanese psychomime. For example, the major Japanese-English dic-
tionary KNJED offers "be startled" for all three psychomimes haQ, gyoQ,
and dokiQ; "feel a shock" for both gyoQ and dokiQ; "get a start" for both
haQ and dokiQ. Looking at those lists of English emotion words for each
Meanings of Japanese sound-symbolic emotion words 223
psychomime, the reader just gets confused, wondering how each word is
different from the others.
Izard (1977: 277) says that the feeling of 'surprise' is difficult to describe
because the feeling does not last long. However, the common features as
well as differences of haQ, gyoQ, and dokiQ can be clearly explicated if we
examine the underlying concepts of these psychomimes in terms of intu-
itively understandable words.
For the discussion of each psychomime, examples in context are given
first, then my proposed definition, followed by amplification or justifi-
cation of particular elements in the definition.
2.1. HQ
Examples:
(3) HaQ to shita shunkan no kimi wa tenshi sa...
[From a popular song by T. Tahara]
RH4: 'You look like an angel at the moment I feel haQ.'5
(4) Kanojo no utsukushisa ni mitorete, sono ba ni kugizuke ni natte ita
watashi wa, musume no yobu koe ni haQ to ware ni kaetta. [PG]
B: 'Transfixed, lost in admiration for her beauty, I suddenly \haQ
to] came to myself at the sound of my daughter calling me.'
(5) Futon o mochiageru toki, yoisho to kakegoe shite, haQ to omotta.
Watashi wa, ima madej'ibun ga yoisho nante gebita kotoba o iida-
su onna da to wa omotte inakatta. [HM: 71]
J-E: 'As I lift it (my futon) to put it away, I hear myself go "Oof!"
and I think, what? \haQ to}. I've never thought of myself as the
type of girl who'd come out with something like "oof'.' [RM: 44]
(6) Tsumi to batsu, Dosutoefusukii, chira to sore ga zunoo no katasu-
mi o kasumete toon, haQ to omoimashita. Moshimo ano Dosuto-
shi ga tsumi to batsu o shinonimu to kangaezu, antonimu to shite
narabeta mono to shitara? [NS: 104]
J - E: 'Crime and punishment. Dostoyevsky. These words grazed
over a corner of my mind, startling me [making me feel haQ]. Just
supposing Dostoyevsky ranged 'crime' and 'punishment' side by
side not as synonyms but as antonyms?' [NLH: 147]
224 Rie Hasada
Meaning of haQ
(a) X feels something for a very short time
(b) because X thinks something
(c) sometimes a person thinks something like this:
(d) "something happened now
(e) I didn't think before that this would happen
(f) I know something now because of this"
(g) because of this, this person feels something for a very short time
(h) X feels something like this
(i) because X thinks something like this
According to Morita (1988: 83), the Japanese word odoroku, a non-
sound-symbolic emotion word for a 'surprise' feeling, originally had
milder connotations than the present odoroku. For example, in the Heian
period (794-1192 AD), even the feeling experienced when one is woken
from sleep was described as odoroku. Nowadays, only one psychomime
haQ can be used for such an occasion of awakening from sleep, as seen in
sentence (11). Other psychomimes or emotion words for 'surprise'-like
feelings including odoroku are used for referring to more intense feelings.
something happened now
I didn't think before that this would happen
This feeling occurs suddenly when one reacts to some happening im-
mediately. There is an unexpected feeling in haQ, although its degree is not
as strong as that of gyoQ ( didn't know before that something like this
226 Rie Hasada
could happen') and dokiQ ( didn't think before that this could happen').
The trigger of this haQ feeling can be anything. It does not have to be re-
markable, and often is a very trivial thing, such as someone's calling one's
name as seen in sentence (4). Moreover, the unexpectedness included in
the haQ feeling is not necessarily caused by an external thing. This could be
one's own voice as seen in sentence (5); one's mental activity as seen in
sentence (6); or one's own action of getting off the train as seen in sen-
tence (10).
I know something now because of this
Izard (1977: 277-8) describes the thinking activity at the moment of
'surprise' as follows: "our mind seems to be blank. ... It is as though ordi-
nary thought processes are momentarily stopped. Thus there is very little
thought content associated with surprise, and virtually none with startle".
However, the intensity of unexpectedness of the haQ feeling is not strong
enough to stop one's thinking. For example, one could say that one gets an
idea at the moment of this haQ feeling as follows:
Therefore, the time when one feels haQ is rather the crucial moment to
start the active movement of one's mind. At this moment, we recognise or
know something: for example, we come to reality from an enchanted state
as in sentence (4); we recognise something we didn't recognise before,
such as our own habit as seen in sentence (5) or get a flash of an idea as
seen in sentence (6); we remember something we have forgotten as in sen-
tence (7) and (10); we recognise something anew, such as someone's
beauty as seen in sentence (3) and (14). Therefore, haQ is often used to-
gether with the verbs kizuku 'recognise' in (10), or omou 'think' in (5), (6),
(7), or with phrases such as ware ni kaeru 'come to oneself in (4), or ki
torinaosu 'pull oneself together' in (11).
There are a lot of sound-symbolic words which refer to the state of ab-
sent-mindedness, day-dreaming, or unconsciousness in Japanese: uttori
'be transported', ukkari 'inadvertently', uka-uka 'to dream away one's
time'; booQ, pokan, bokeQ, bosa-bosa, bonyari 'to look vacantly', and so
on. HaQ is the feeling experienced when one comes to oneself from these
absent-minded, delusive, or unrecognised feelings.
Meanings of Japanese sound-symbolic emotion words 227
because of this, this person feels something for a very short time
2.2. GyoQ
Examples:
(16) Watashi wa Tsugumi no ude o tsukami gyoQ to shita. Sono hitomi
wa reisei ni watashi mitsumekaketa ga, ude wa monosugoi atsu-
sa datta. [T: 159]
RH: felt gyoQ when I held Tsugumi's arm. She stared at me
calmly, but her arm was awfully hot.'
(17) Tsugumi mo sukoshi waratte miseta. Soshite, itta.
"Ano na, omae ni dake iu kedo na, atashi, dame kamo shirenai.
Kitto shinu." Watashi wa gyoQ to shita. : 203]
RH: Tsugumi smiled a bit. Then she said, "Well, I will tell only
you. I might be hopeless. I am sure I will die." I felt gyoQ?
(18) Sukoshi zutsu, jigazoo no seisaku ni torikakatte mimashita. Jibun
demo gyoQ to shita hodo. insan na e ga dekiagarimashita. Shika-
shi, kore koso munasoko ni hitakakushi nojibun no shootai na no
da. [NS: 34]
J-E:'... I began to draw a few self-portraits... The pictures I drew
were so heart-rending as to stupefy even myself [make even my-
self feel gyoQ]. Here was the true self I had so desperately hidden.'
[NLH: 55]
(19) (Ani wa) kaoiro mo waruku, gyoQ to suru hodo yasete, kewashii
yooboo ni narimashita. [HM: 214]
J->E: '[My oldest brother's] face had a stern look, and I was
shocked [I felt gyoQ] to see how thin he was.' [DO: 166]
228 Rie Hasada
Meaning of gyoQ
(a) X feels something for a very short time
(b) because X thinks something
(c) sometimes a person thinks something like this:
(d) "something happened now
(e) I know: this is bad
(f) I did not know before that something like this could happen
(g) I can't think now
(h) I can't do anything now because of this"
(i) because of this, this person feels something bad for a very short
time
Meanings of Japanese sound-symbolic emotion words 229
2.3. DokiQ
Examples:
(25) ... to igai na koto osshatta. Watashi wa dokiQ to shite...
[Dazai "Shayoo (setting sun)" in JEOMTD]
J-E: was taken aback Tfelt dokiQ^ by this unexpected remark.'
(26) Kurejitto kaado no seikyuu ni dokiQ to shitari...
[A phrase in the advertisement of Gunze, 4/7/1993]
RH: feel dokiQ to see the demand for payment of my credit
card.'
(27) Tetsuo ga itta. "Chigau, nigeru na'.' dokiri6 to shite kare o mit-
sumeruto... [KY: 158]
RH: Tetsuo said, "No, don't run away (from your trouble)." I felt
dokiri. then I gazed at him.'
Meanings of Japanese sound-symbolic emotion words 231
Meaning of dokiQ
(a) X feels something for a very short time
(b) because X thinks something
(c) sometimes a person thinks something like:
(d) "something happened to me now
(e) I didn't think before now that this could happen
(f) something happened to my heart now because of this
(g) I could hear something like this: dokiQ"
(h) because of this, this person feels something for a very short time
(i) X feels something like this
(j) because X thinks something like this
something happened to me now
I didn't think before that this could happen
232 Rie Hasada
One also could feel dokiQ when the trigger of that feeling cannot be
judged good nor bad, such as the sudden unexpected discovery of some-
one's presence in a place as seen in sentence (31).
Unlike gyoQ, whose cause could be anything as long as the experiencer
thinks it bad and incredibly unexpected, the cause of dokiQ is restricted to
something that will affect the experiencer somehow. People feel dokiQ
when they think about something that concerns them [i.e. "to me"]. For
example, we feel dokiQ upon discovering an extremely high balance due on
our own credit card bill as in sentence (26). We would not feel dokiQ, while
we might feel gyoQ, if that is the bill of our friend's credit card.
It is our mune 'heart' where the physical reaction of this dokiQ feeling
occurs, as seen in mune ga dokiQ to suru 'my heart feels dokiQ' in sen-
Meanings of Japanese sound-symbolic emotion words 233
tence (29). Other parts of the body would not feel dokiQ: *atama/hara/ka-
rada ga dokiQ to sum '*My mind/belly/body feels dokiQ'.
The degree of unexpectedness of dokiQ is strong enough to make the ex-
periencer think he/she had a strong heart beat. The intensity of this unex-
pectedness is stronger than that ofhaQ, where no particular physical mani-
festation is seen except for one's mental activation, but this is weaker than
that of gyoQ, where the unexpected shock is so strong as to stop one's
thinking as well as physical activity.
I could hear something like: dokiQ
The experiencer of a dokiQ feeling who physically felt a strong beat of
his heart thinks as if he/she actually heard the sound of its beating. The
sound of one's repeated heart beat is described as the phonomime doki-
doki in Japanese, as in the following sentence:
3.1. Hara-hara
Examples:
(34) Sore kara futari wa yashizake o nomihajimeta. Kupukupu wa
hara-hara shita. Nazette.futari tomo, taihen na ikioi de gabugabu
nomidashita kara datta. [FKB: 87]
J-E: 'The two began drinking the wine. It made Kupukupu ner-
vous [feel hara-hara] to see how fast they were gulping it down.'
[AKS: 80]
(35) Yakuruto, kotoshi mo hara-hara yakyuu. [AS: 5/4/1993: 21]
RH: 'Yakuruto [baseball team] plays the game making the audi-
ence feel hara-hara this year, too.'
(36) Mozuku daihoosaku: Sanchi hara-hara [AS: 11/6/1993: 17]
RH: 'The people in the production area ofmozuku feel hara-hara
because too much is being harvested [since the price of mozuku
might greatly fall].'
(37) Hajimete Shaarokku Hoomuzu o yonde, hara-hara shite kara to iu
mono wa, yamitsuki ni natte shimatta no de aru. [CS: 71]
Meanings of Japanese sound-symbolic emotion words 235
RH: 'He had been keen on them ever since he read Sherlock
Holmes for the first time and felt hara-hara.'
(38) [Sakki kara uma kara ochite bakari iru] Kare gafutatabi uma ni
noru no hara-hara shinagara. mimamotte imashita. [KKA: 215]
J*-E: ['The knight was falling down from the horse many times.]
[Alice] watched him with some anxiety [feeling hara-hara]^ he
mounted again.' [TLG: 116]
(39) Watashi wa mae kara, anta-tachi no mamagoto asobi mitai na kur-
ashikata hara-hara shinagara mite ita n desu. [Dazai "Shayoo"
in JEOMTD]
(J-E: Tve been watching with my heart in my mouth [feeling
hara-hara] the way you two have been living.')
(40) Watashi wa takai tokoro ni nobotte iru hito o miru to, ochinai ka to
omotte, hara-hara shite shimau. [JF: 60]
RH: 'When I look at a person who is climbing a high mountain, I
feel hara-hara, wondering if he/she might fall down.'
(41) Kabu katta hito wa, ichinichi ichinichi no kabu no neugoki ni
hara-hara shiteiru soo da. [JF: 60]
RH: hear that the people who bought the stocks are feeling hara-
hara about the movement of the price of stock day by day.'
Meaning of hara-hara
(a) X feels something
(b) because X thinks something
(c) sometimes a person thinks something like this:
(d) "I see: something is happening to someone/something (Y)
now
(e) something bad could happen to in a very short time because
of this
(f) I don't want this to happen
(g) because of this, I want to do something
(h) I don't know what I can do"
(i) because of this, this person feels something bad
(j) this person thinks:
(k) "I will see (something/someone) is falling down [hara-hara]
into something bad"
236 Rie Hasada
(42) *Takai tokoro ni noboru to ketsui shita hito mite hara-hara sum.
*'I feel hara-hara seeing a person who decided to climb the high
mountain.'
(43) *Takai tokoro ni nobotta hito mite hara-hara sum.
*'I feel hara-hara seeing a person who climbed the high moun-
tain.
Therefore, for example, in sentence (35), the audience feels hara-hara,
watching their team (Yakuruto) play, assuming they will lose if they keep
playing like that. In sentence (40), the person feels hara-hara at seeing
somebody in the process of climbing a high mountain, anticipating the
possibility of them falling down at any time. In sentence (36), the farmers
feel hara-hara at seeing the on-going too abundant harvest of mozuku (ne-
macystis decipiens), knowing this could cause a great fall in price at any
time. In sentence (39), the person felt hara-hara seeing the young couple's
way of living, because she thought their life was like playing at housekeep-
ing, thus it could collapse at any moment. In sentence (41), the person feels
hara-hara because stock prices are changing everyday, which could make
the price of their own stocks fall.
In all these examples, the experiencers of hara-hara feeling presuppose
the possibility of something bad happening to a particular third person/ob-
ject (Y) because of something in the on-going process. They feel this poss-
ible danger in the immediate context ("now").
I don't want this to happen
Meanings of Japanese sound-symbolic emotion words 237
A person who feels hara-hara for does not want something bad to
happen to Y. A person who does not care if something bad happens to
does not feel hara-hara. For example, in sentence (35), a fan of Yakuruto's
opposition team will not feel hara-hara about Yakuruto's unstable playing
during the baseball game, since they want something bad to happen to Ya-
kuruto so that their favourite team will win.
because of this, I want to do something
I don't know what I can do
The experiencer feels like doing something about the possibility of
something bad happening to Y. However, since the experiencer of the hara-
hara feeling is a passive observer of the on-going situation, often he/she
can do nothing but watch helplessly, just hoping that something bad will
not happen to Y. For example, in sentence (36), the farmers feel hara-hara,
but cannot do anything about the increase of mozuku, which is a natural
phenomenon. In sentence (37), the reader of the Sherlock Holmes story
can do nothing about the development of the plot in the story. Since the ex-
periencer of hara-hara feeling cannot do anything about the on-going situ-
ation, the psychomime hara-hara often collocates with phrases like ki
momu 'to fidget', or mimamoru 'to watch over'.
because of this, this person feels something bad:
The experiencers feel "something bad", since they cannot do anything,
while they are faced with possible danger toward Y, which they do not want
to happen.
this person thinks:
I will see Y (something/someone) is falling down into something bad7
Hara-hara originally describes the manner of something small and light
falling down, which causes the observer to feel sad or melancholy, as seen
in the following sentences.
3.2. Hiya-hiya
Examples:
chatter of Karube.'/'At first I could feel myself going cold all over
[feel hiya-hiya\ as I stood there listening to Karube chatting away.'
(51) Kare ga sono koto o ukkari shaberi wa shinai ka to hiya-hiya
UKNJED]
: was in great fear [felt hiya-hiya} lest he should blurt it out.'
(52) Hiya-hiya shinagara gakeppuchi no michi o susunda. [PG]
B: 'We timidly [feeling hiya-hiya\ made our way along the path on
the edge of the cliff.'
(53) Yaku sennin no yajiuma wa, te ni se o nigitte hiya-hiya. [Yomiuri
(evening) 14/3/1972 in GG]
RH: 'About one thousand curious bystanders feel hiya-hiya in
breathless suspense.'
(54) Itai n ja nai ka to, naishin hiya-hiya datta. [AS: 15/11/1972 in
GG]
RH: was feeling hiya-hiya inwardly, thinking it might be painful
for you.'
Meaning of hiya-hiya
(a) X feels something
(b) because X thinks something
(c) sometimes a person thinks something like:
(d) "something bad can happen
(e) I don't want this
(f) because of this, I want to do something
(g) I don't know what I can do"
(h) because of this, this person feels something bad
(i) this person thinks:
(j) "if I were in a cold place, I would feel like this [hiya-hiya]"
(k) this person feels this for some time
(1) X feels something like this
(m) because X thinks something like this
something bad can happen
Unlike hara-hara, where the possible danger is always directed to a third
person, hiya-hiya has no such restriction. It could be directed to a third per-
240 Rie Hasada
son. For example, in sentence (53) curious bystanders feel hiya-hiya in ob-
serving some event. In sentence (54) the person felt hiya-hiya anticipating
the possibility of something bad happening (pain) to another person. On
the other hand, the patient of the bad happening could also be the experi-
encer. For example, in sentence (46), the family members feeling hiya-hiya
are not concerned about the condition of the third person (the patient) but
about the undesirable possibility of they themselves hearing the patient's
shriek. In sentences (47) and (48) the experiencer feels hiya-hiya about
him/herself for fear of being accused of the deed he/she did.
The experiencer feels hiya-hiya when he/she thinks the anticipated bad
happening is still at the level of possibility ('something bad CAN happen')
and when the possibility of danger is not so imminent or urgent. Therefore,
if the people are sure about the occurrence of a bad happening, hiya-hiya
cannot be used. This point can be illustrated by comparing the following
sentence with the sentence (48) above.
3.3. Biku-biku
Examples:
(56) Kowai mono ni hikizurareru yoo ni, suru-suru to soba e itte suw-
atta. Naishin biku-biku shinagara. "Okaasan wa dooshitemo
dame desu ka?" to itta. [HM: 224]
J - E: 'As though drawn by some fearful force, I went towards him
and sat down. I felt an inward tremor [felt biku-biku inwardly! as I
asked, "Will she (mother) die?'" [DO: 173]
(57) Hidoku shikarareru to omotte, biku-biku mono de kita n desu
kedo, kite yokatta wa. [Inoue "obasute" p. 23/89 in JEOMTD]
J -> : " wondered if I was going to be scolded severely and I was
nervous was in the state of feeling biku-biku] about coming - but
I'm glad I came," she said.'
242 Rie Hasada
Meaning of biku-biku
(a) X feels something
(b) because X thinks something
(c) sometimes a person thinks something like this:
(d) "something bad will happen to me
(e) I don't want this
(f) because of this, I want to do something
(g) I cannot do anything"
(h) because of this, this person feels something very bad
(i) this person thinks:
(j) "I am trembling/shaking [biku-biku]"
(k) this person feels this for some time
(1) X feels something like this
(m) because X thinks something like this
something bad will happen to me
Unlike hara-hara or hiya-hiya, the occurrence of danger of biku-biku is
always specifically directed to the experiencer him/herself, as seen in sen-
tence (56) to (66) ("to me"). It could be directed to someone/something
which one regards as a part of oneself. For example, in sentence (67) the
family feel biku-biku about the possible attack on their baby by the dove.
The experiencer's expectation that something bad will happen to him/her
is much higher for the person feeling biku-biku than it is for the person feel-
ing hiya-hiya or hara-hara ("something bad WILL happen" rather than
"something bad COULD/CAN happen"). One of the big differences between
biku-biku and hiya-hiya is this degree of sense of danger felt by the experi-
encer. When one feels the danger very close and immediate, a person
would say biku-biku, but not hiya-hiya. For example:
244 Rie Hasada
(69) Oyoide iru toki ashi ga tsutte, biku-biku keiren shita koto ga aru.
[PG]
B: Once, while swimming I got a twitch [biku-biku] cramp in my
leg.'
The emotional biku-biku feeling is associated with this image of a shak-
ing movement. People do not necessarily actually tremble when they ex-
perience the emotional biku-biku feeling, like biku-biku in the phrase nai-
shin biku-biku shinagara 'feeling an inward tremor' [the inward biku-biku}
of sentence (56): this trembling is metaphorical, and not a physical move-
ment. However, the experiencers of the biku-biku emotion feel as if they
are trembling, and they sometimes actually physically tremble. Thus, this
image of a 'trembling' movement is reflected in the emotional biku-biku
feeling.
When the person actually trembles because of this biku-biku feeling, that
movement is easily observable outwardly. Therefore, the prohibition form
biku-biku suruna! 'Don't feel/be biku-biku!' is possible for this feeling. By
contrast, hiya-hiya suruna! 'Don't feel/be hiya-hiya!' sounds odd, since
the feeling is manifest only to the experiencer him/herself, and a third per-
son cannot tell whether another person is really feeling hiya-hiya or not
from external appearance. Because of the possible objective observation of
biku-biku, it is often used for describing a third person's feeling manifested
as outward behaviour.
3.4. Odo-odo
Examples:
While the experiencer of the odo-odo feeling definitely does not want
something bad happen to them, and thus wants to take some action about it,
they cannot think of a way out of the "upset" situation.
because of this, this person feels something bad
The resulting feeling is "something bad".
4. Conclusion
The major goal of this paper was to explicate the meaning of selected psy-
chomimes to an extent that can give non-native speakers a greater under-
standing of the meaning and correct use of these words.
The problems with presently available definitions of psychomimes given
in dictionaries were discussed above. It was shown that neither vague anal-
ogous words from other languages, nor lengthy explanations, nor accumu-
lations of sample examinations can show the unique nature or concept of
Japanese psychomimes clearly and precisely. I have proposed the use of
the NSM approach to solve this problem. In sections 2 and 3 I hope to have
demonstrated that the complex Japanese-specific meanings involved in se-
lected psychomimes can be clearly shown and made comprehensible to
outsiders, when they are translated into the Natural Semantic Metalan-
guage based on universal human concepts and represented in the frame-
work of a "prototype scenario".
I believe that I have succeeded in showing that the complex and unique
semantic concepts of Japanese psychomimes, which are usually described
as 'untranslatable', are in fact translatable on the level of semantic expli-
cation with a language-independent semantic metalanguage. The extent of
similarities and dissimilarities in labelling and the conceptualisation en-
coded in different psychomimes becomes apparent with the use of the
Natural Semantic Metalanguage.
It can also be seen from the above examples that many psychomimes are
associated with concrete images such as those of:
Notes
1. I would like to express my gratitude to Prof. Anna Wierzbicka and Dr. Anthony E.
Backhouse for their comments on an earlier draft of this paper. It was partially pre-
pared with the financial assistance of the Toyota Foundation.
2. The direction of transfer is always from phonomimes to mimetic words (phenom-
imes/psychomimes), and not vice versa (Kakehi 1992: 18).
3. The Romanisation used here follows the Hepburn system. The Hepburn system in-
dicates a phonetically close approximation of Japanese pronunciation (and thus
spells shi, chi, tsu, etc.) (for further details see cf. e.g. Backhouse 1993: 60-63).
However, I use /aa, ii, uu, ee, oo/ for long vowels instead of the macron. Also, for the
final mora obstruent in sound symbolic words, /Q/ is used instead of/t/, for example:
jit-to -*jiQ-to; mukat-to -* mukaQ-to
4. All translations are my own (marked RH) except when otherwise indicated. When
translations of original texts were available I have used those translations. In these
cases I have shown which part of the English sentence corresponds to a Japanese
250 Rie Hasada
References
Backhouse, Anthony E.
1993 The Japanese Language: An Introduction. Oxford: Oxford University
Press.
Goddard, Cliff and Anna Wierzbicka
1994 Introducing Lexical Primitives. In: Cliff Goddard and Anna
Wierzbicka (eds.), Semantic Primitives and Lexical Universals, 31-54.
Amsterdam: John Benjamins.
Izard, Carroll E.
1977 Human Emotions. New York: Plenum Press.
Inagaki, Yoshihiko
1982 Gendai no Shingo, Ryuukoogo [New words and words in vogue]. In
Yoshihiko Inagaki (ed.), Gendai no Goi. Tokyo: Meiji Shoin.
Kakehi, Hisao
1986 The function and expressiveness of Japanese onomatopoes. Kobe Dai-
gaku bungakubu kiyoo, 1-12.
Kimizuka, Sumako
1967 A Contrastive study of the Japanese Structural Device Onomato-
poeia + verb' and Certain English Verbs. Journal-Newsletter of the
Association of Teachers of Japanese. Vol. III. No. 3. September:
12-16.
Lebra, Takie Sugiyama
1976 Japanese Patterns of Behaviour. Honolulu: University of Hawaii
Press.
Leibniz, Gottfried Wilhelm
1949 New Essays Concerning Human Understanding. La Salle: Open Court
Publishing Company. [1704]
Meanings of Japanese sound-symbolic emotion words 251
Martin, Samuel E.
1964 Speech labels in Japan and Korea. In: Dell Hymes (ed.), Language in
Culture and Society: A reader in linguistics and anthropology. New
York: Harper and Row.
Mito, Y. and H. Kakehi
1981 Nichi-Ei Taishoo Onomatope Jiten [Japanese-English Comparative
Onomatopoeic Dictionary]. Tokyo: Gaku Shoboo.
Morita, Kiichi
1988 Gitaigo no tokusei ni tsuite [On the characteristics of mimetic words].
Iwate-ikadai-kyooyoo-nempoo 23: 107-118.
Ono, Shuuichi (ed.)
1989 A Practical Guide to Japanese-English Onomatopoeia and Mimesis.
Tokyo: Hokuseidoo.
Ootsubo, Heiji
1989 Giseigo no Kenkyuu [A Study of Onomatopoeia]. Tokyo: Meiji Shoin.
Shibatani, Masayoshi
1990 The Languages of Japan. Cambridge: Cambridge University Press.
Tamamura, Fumio.
1982 Goi oyobi go no koozoo [Vocabulary and word structure]. In: Kooza
Nihongogaku, vol. 12. Gaikokugo to no Taishoo, 1-21. Tokyo: Meiji
Shoin.
Wierzbicka, Anna
1986 Introduction. Journal of Pragmatics 10 (5): 519-534.
1991 Japanese Key Words and Core Cultural Values. Language in So-
ciety 20: 333-385.
1992a Semantics, Culture, and Cognition: Universal Human Concepts in
Culture-Specific Configurations. New York: Oxford University Press.
1992b Talking about Emotions: semantics, culture and cognition". Cognition
and Emotions 6(3/4): 285-319.
1992 Defining Emotion Concepts". Cognitive Science 16(4): 539-582.
1994 Cognitive domains and the structure of the lexicon. In: Lawrence
Hirschfeld and Susan Gelman, eds. Mapping the Mind: Domain Spe-
cificity in Cognition and Culture, 431-452. Cambridge: Cambridge
University Press.
When his mother saw me, her eyes immediately became two fireballs. She jumped at
me and jabbed at my forehead with her finger. "So you are Li Lu! You are Li Lu!" She
grabbed me and hit me again. How I hated her! My face was burning with rage. At
any other time I would have bitten the fat hand that was grabbing my collar. But I re-
membered my teacher's story and suppressed my anger. (Li Lu Moving the moun-
tain, 1990: 42)
1. Introduction
This paper focusses on five Mandarin Chinese words - nu, sheng/qi,
nao(huo),fen, taoyan - as well as their figurative associations and elabor-
ations, which are pertinent to the conceptualisation of the "emotions"
(qingxulqinggan) often rendered with, or comparable to the English words
'angry' or 'anger'. Ever since Darwin's (1872/1979) classic treatment of
emotions, "anger", "something like anger", or "a family of anger concepts/
expressions" have been recurrently proposed by a number of Western psy-
chologists as one of the "fundamental", "universal", "primitive" or "basic"
human "emotions" (cf., e.g., Ekman 1992, 1993, Oatley 1992 for the pro
arguments, and Russell 1991, 1994, van Brakel 1994, Wierzbicka 1992 a
and b, Wierzbicka 1999 for a recent contra).
My aim here will be to examine some of the insights which the examin-
ation of the Chinese lexical data might contribute to this debate.
2. Nu
Tian zi zhi nu, fu s hi
Heaven son of nu lie corpse
bai wan, liu xue qian li
hundred 10 thousand flow blood thousand mile
256 Pawet Kornacki
2.1. Communicating nu
(7) nu ya tu zui
nu tooth protrude mouth
'showing teeth in nu'
Yet another aspect of one's countenance involved in showing nu is hair,
as in (8) and (9):
(11) nu ma bu xiu
nu curse not rest
'cursing nu-\y without stopping, revile'
(12) nu de pao xiao ru lei
nu ASSOC roar roar like thunder
'to roar like thunder with nu'
(13) nu de da fa lei ting
nu ASSOC big send thunder- clap
'to thunder with nu'
However, nu can also be communicated with the eyes only, and the si-
lence which ensues then, heightens the tension of the whole situation:
Concepts of anger in Chinese 259
(14) nu s hi bu yu
nu look not speak
'to give nu looks and say nothing'
(15) nu bu zi sheng
nu not self conquer
'nu beyond self-control'
(16) nu bu ke e
nu not can stop
'so nu that it can't be stopped'
The idea that nu implies lack of control can further be corroborated by
the fact that nu can be ascribed to a number of the natural phenomena
which are characteristically violent and uncontrollable. In such contexts
the difference between nu and 'anger' is especially prominent, since in
translating these expressions one is usually forced to speak of 'fury' or
'rage', rather than simply 'anger'. Thus, for instance, water, wind and fire
can be said to exhibit nu, as in the following stock descriptions:
Nu is like fire - both in the experience of the person who feels it, and in its
effects on the people or things that are affected by it. There are quite a few
expressions which metaphorically link nu with fire and smoke, the most
common of them being probably this:
To this list Zhang (1947/1985: 155-156) adds also fatigue resulting from
menial jobs, quoting the following saying:
Gathering the above clues together, and in order to proceed with further
discussion, I would propose the following explication:
nu
(a) X feels something
(b) X thinks something like this:
(c) something happened
(d) this is bad
(e) I don't want this
(f) because of this, I want to do something very bad to someone
now
(g) I can't not do this
(h) X feels something very bad because of this
(i) because of this, X can't think now
(j) something is happening inside X because of this
(k) something bad can happen to X because of this
Arguably, the prototypical case of 'anger' evokes the idea of "bad ac-
tion" (i.e., "someone did something bad" - see, for example, Wierzbicka
1992b: 569, Goddard 1991: 268). This, however, is not necessarily the case
264 Pawel Kornacki
As this example shows, neither the idea that "someone did something
bad" nor that "something bad happened to me" are necessary for nu - the
mule's illness wasn't, so to speak, somebody's action ("against" the
keeper), neither was it something that actually happened TO HIM (although
he was, of course, affected by it). To see that it is also not necessarily the
case that "something BAD happened" when one becomes nu, let us recon-
sider some of the situations mentioned by Zhang.
Thus, for instance, when a person becomes nu upon seeing someone
'hateful/disgusting' (taoyan) to them, or when he or she catches sight of a
weirdly dressed individual, the focus seems to be on the feelings which
arise in that person because of these events, and which are caused by that
person's "negative assessment" ("this is bad") and "disagreement" ("I
don't want this") in the situation, rather than an inherent feature of it - in
which case nu seems to resemble something closer to 'shocked' rather than
to 'angry' - hence it seems more accurate to "split" an otherwise plausible
component "something bad happened". It is, moreover, not really clear
whether "something" should "happen" necessarily "NOW" - since the
onset of nu could be either an unexpected, singular, and salient event (such
as someone's insult, a quarrel, finding oneself abandoned, noticing some-
thing hateful), or a protracted period of toil.
Nu can be a powerful and uncontrollable emotion. The metaphorical ref-
erences to the natural phenomena beyond human influence (cf. the nu of
the waves, wind, and fire versus the unacceptable *'anger' of the waves,
wind, or fire) on the one hand, and certain fairly unequivocal non-verbal
manifestations (bared teeth, clenched fists) of nu, embedded in a number of
lexicalised expressions, strongly suggest the will of immediate aggression
Concepts of anger in Chinese 265
3. Shenglqi
The phrase points to two important aspects of shenglqi: first, that it can
be naturally seen as extending over a period of time - which makes it dif-
ferent from nu, typically construed as a momentary, or a short-time event,
and second, that it can be plausibly viewed as something that happens IN-
268 Pawel Kornacki
SIDE a person. The relevant word here is bie, translatable as 'suppress, hold
back, suffocate'. According to Bucher (1986: 338), it has strongly "physio-
logical" associations, as it is used first of all about the suppressed natural
needs, as in (39):
sheng/qi
(a) X feels something
(b) X thinks something like this:
(c) something happened
(d) I don't want this
(e) I want to do something bad because of this
(f) X feels something bad for a long time because of this
(g) something is happening inside X because of this
(h) something bad can happen to X because of this
270 Pawel Kornacki
unquestionable will to do something bad but does not prejudge the likeli-
hood of "venting" one's qi (since both restraint and retaliation are possible,
as well as lexicalised, although there seems to be more evidence in favour
of 'restraint', such as, for example, the fact that one speaks of "dying"
rather than "killing" with/from qi).
As suggested before, there seem to be fairly good reasons to think of
shenglqi as an emotion which may extend over an indefinite period of time.
It is also potentially harmful to the experiencer, but again, it seems, in a dif-
ferent way from nu. Shenglqi can be rather seen as something lingering
('smouldering' rather than 'exploding') and hidden ('kept in the stomach'),
something that perhaps happens as much TO a person, as it does INSIDE him
or her.
4. Nao (huo)
Someone refuses to pay attention to the criticism. The words simply 'do
not enter' into his head - although we would like this person to behave in
one way, he keeps doing the opposite. This aspect of the meaning of nao-
huo probably accounts for why this word can sometimes, as Bucher (1986:
274 Pawet Kornacki
5. Fen
The next Chinese concept to be analysed in this paper is fen. Its various
(morphemic) collocations point, first of all, to the strongly moral, evalu-
ative, and, in a sense, "public" quality of this emotion, which seems to be
absent from, or relatively downplayed in, the notions of nu, sheng/qi, or
nao(huo).
Thus, for instance, one can speak of such politically significant senti-
ments as gongfen (lit. 'public-/?'), which Metzger (1981: 19) describes as
"the feelings of 'righteous anger' aroused by authority figures perceived as
morally deficient", yifen (lit. 'righteousness-/?') glossed inHan-Ying Ci-
dian (1988: 821) as 'righteous/moral indignation', and exemplified in the
expression geming yifen (lit. 'revolutionary yifen', i.e., 'revolutionary in-
dignation'). According to one native speaker, people labelled as "traitors"
(maiguozei) or "spies" (tewu) were during the Cultural Revolution often
singled out and persecuted in the name ofminfen (lit. 'people-/?'), and re-
ferred to as 'arousing' (jiqi) or 'facing' (mianzhong) minfen.
Fen is typically stirred in the situations which can be summed up with the
phrase bu gongping (roughly, 'not fair/just'). This association is lexicalised
in another collocation -fenfen bu ping (lit. '/<?/<? not just'), which could
describe the feelings aroused in a situation in which, for example, a crimi-
nal escapes or somehow evades the proper punishment, or a student is un-
deservedly (in the eyes of others) praised or criticised by a teacher.
Concepts of anger in Chinese 275
Wang (1993: 74), in his quite revealing paper on different ways of sub-
duing "bad feelings", speaks about/*? associating it mainly with two kinds
of circumstances, induced by specific 'ways of thinking' (sixiang moshi).
Firstly, fen is associated with what Wang calls "Yinggai" huo "Bu ying-
gai" sixiang ('"Should" or "Should not" thinking'), which is implied in
situations where something that we think 'should' (yinggai) happen in fact
doesn't happen, and/or something that we think 'should not' (bu yinggai)
happen does happen. Wang gives here the following example: "When the
girlfriend with whom you were together for several years breaks off with
you, you can become fenfenran (lit. '/<?/z/<?w-like'): 'This should not be so!
(Zhe tai bu yinggai le - this-too-not-should-) There is 'good feeling'
(ganging) between us, why should we split?'" According to Wang,
thoughts of revenge or of getting even with the partner may be the outcome
of this situation.
The second type of context relevant to fen is called by Wang "Bu gongp-
ing" sixiang (' "Not fair/just" thinking'), where the bad feelings are gener-
ated because a person has evaluated a certain situation as bu gongping. In-
terestingly, Wang provides here not only an example of a relevant situation,
but also the hypothesised thoughts of a person involved in it. The key word
in this passage is the bi-morphemic qifen, the most common and colloquial
of the/? compounds:
For example, one day you go window-shopping, and find something which could be
a beautiful gift for your wife. You immediately want to buy it and give it to her, but
unfortunately you run into an 'abominable' (taidu elie) shop-assistant who not only
'neglects you' (daiman ni) but also shouts abuse at you. Obviously, in such a situ-
ation there appear a lot of thoughts in your mind which can make you extremely qifen
(feichang qifen), so that you wish to hit that person, and when you go back home you
can hardly calm yourself.
fen
(a) X feels something
(b) X thinks something like this:
(c) some people did something bad
(d) it is very bad if people do things like this
(e) when someone does something like this people feel some-
thing bad because of this
(f) I would want to do something because of this if could
(g) X feels something bad because of this
(h) X thinks something bad because of this
(i) sometimes it is good if people feel like this
The basic stimulus to fen is the morally reprehensible ("bad") human ac-
tion. It is, however, not necessarily the case that the agent provoking this
emotion must be a "singular" person ("someONE") - in fact, the "social"
sentiments such as gongfen ('public-/?'), or minfen ('people-/?'), could
easily be provoked by the behaviour of some groups of people, labelled,
e.g., as 'bureaucrats' (guanliao), 'bad elements' (huaifenzi), etc., hence
the reference to "some people" in component (c).
Components (d) and (e) represent the negative moral and social judge-
ments attached to fen. The behaviour causing/? breaks expected social
standards ("it is very bad if people do things like this"), and consequently
hurts the interests and feelings of other people (component e). From the
perspective offen, people like an 'abominable' shop-assistant, 'traitors', an
inconsiderate lover, or a lying politician appear, first of all, to share a dis-
regard for others.
Arguably, the feelings and thoughts induced by such offending attitudes
are not inherently accompanied by the overt aggression, or a certain vio-
lence peculiar to, e.g., nu. The unfortunate lover, according to Wang
(1993), can 'hate/resent' (yuanhen) his partner, or think of revenge
(baofu). The customer thinks (but - at least in Wang's scenario - only
thinks) of hitting the nasty assistant, and his supposed ideas (I-IV) do not
seem to involve anything as specific and immediate as, for instance, "I
want to do something bad to someone/this person now". What is involved
is rather more like a painful mental recycling of the offending event
(points III and IV), the idea that "someone should do something about
this" (point II), and bad thoughts about the moral qualities of the assistant
278 Paw et Kornacki
6. Taoyan
taoyan
(a) X feels something
(b) X thinks something like this:
(c) this is bad
(d) I feel something bad because of this
(e) I don't want this
(f) X feels something bad because of this
The scope of component (c) is very broad. It must account for quite var-
ied things that can be called taoyan, such as, for instance, people, weather,
or sounds, hence also the simplicity of the judgement "this is bad". Taoyan
things are unwanted things (of which 'nasty' illnesses are a good example).
Obnoxious' behaviour certainly qualifies for being called taoyan. An in-
formant who watched the excessively "sociable" behaviour of a student
who got drunk at a disco said:
7. Conclusions
The two Chinese concepts which would merit an especially detailed com-
parison with the English 'angry/anger', are, arguably, nu and shenglqi - the
first one on account of being actually considered a natural "human feeling"
(ren qing) in the traditional Chinese psychology (see Yan Guocai 1982),
the other on account of its social and experiential salience, confirmed by its
early acquisition and widespread use in everyday language.
As discussed in the Sections 2.1-3 of this paper, nu is involved in a
number of expressions which bring to mind concepts like 'rage', 'fury', or
'aggression', rather than simply 'anger'. It is perhaps interesting to note
here that Darwin, while claiming that "anger, indignation - these states of
mind differ from rage only in degree, and there is no marked distinction in
their characteristic signs" (1872: 246), still decided to analyse them in sep-
arate sections. The bodily symptoms of rage which he enumerated, come,
in fact, remarkably close to the lexicalised Chinese 'formulae' (chengyu)
describing nu, so that, for example:
(a) "the face reddens or becomes purple, with the veins on the fore-
head and neck distended" (1872: 240)
- nu de lian hong bozi CM ('so nu that the face is red, and the neck
bulges')
(b) "the teeth are clenched or ground together" (1872: 241)
- nu de yao ya qie chi ('so nu that one grinds one's teeth')
Concepts of anger in Chinese 283
(c) "such gestures as the raising of the arms, with the fists clenched, as
if to strike the offender, are common" (1872: 241)
- nubi ('-raise arms'); nu de wo quan ton zhao ('so nu that one
clenches the fists and shows the fangs')
(d) "the hair sometimes bristles" (1872: 242)
- nu fa chong guan (One is so nu that the hair raises one's cap')
(e) "the lips, however, are much more commonly retracted, the grin-
ning or clenched teeth being thus exposed" (1872: 243)
- nu ya tu zui ('so nu that the teeth protrude from one's mouth')
(f) "the eyes are always bright, or may, as Homer expresses it, glisten
with fire" (1872: 242)
- nu de muguang ru ju ('so nu that the eyes are blazing like the
torches')
(g) "the paralysed lips then refuse to obey the will, and the voice sticks
in the throat" (1872: 242)
- nu bu neng hui ('so nu that one can't reply')
Ekman, who in one of his papers (1993: 386) speaks about "the anger
family" which "would include variations in intensity stretching from an-
noyance to rage", suggested in an earlier article that "in all members of the
anger family the brows are lowered and drawn together, the upper eyelid is
raised and the muscle in the lips is tightened" (cf. Ekman 1992: 172). In-
terestingly, the Chinese expression which comes closest to Ekman's con-
figuration includes a reference to nu - rather than to some other, less "in-
tense" 'anger'-concept (e.g., nao) - so that one speaks of heng mei nu mu
('horizontal eyebrows, nu eyes'). This might suggest that the universal fa-
cial expression of "anger", as envisaged by Ekman, is more naturally
spoken of as applicable only to some members of the "family". It is rel-
evant to note that qi - when compared with the strikingly graphic "descrip-
tions" of nu - shows in the face to a much less specific degree (so that one
generally describes the face as becoming red or white with qi, rather than,
e.g., baring the teeth, or frowning with qi).
Van Brakel (1994: 2) in his critique of the psychological and anthropo-
logical studies on "emotions" lists twelve different definitions of anger put
forward by various scholars. Consider, however, the following statement,
not included in Van Brakel's list, but taken from Izard's Human Emotions
(1977: 329-30) "a very common stimulus to anger is the feeling of being
284 Pawel Kornacki
Summary of explications
nu
(a) X feels something
(b) X thinks something like this:
(c) something happened
(d) this is bad
(e) I don't want this
(f) because of this, I want to do something very bad to someone
now
(g) I can't not do this
(h) X feels something very bad because of this
(i) because of this, X can't think now
(j) something is happening inside X because of this
(k) something bad can happen to X because of this
shenglqi
(a) X feels something
(b) X thinks something like this:
(c) something happened
(d) I don't want this
(e) I want to do something bad because of this
(f) X feels something bad for a long time because of this
(g) something is happening inside X because of this
(h) something bad can happen to X because of this
naohuo
(a) X feels something
(b) X thinks something like this:
(c) something bad happened to me
(d) I don't want this
(e) X feels something bad because of this
(f) something bad can happen to X because of this
fen
(a) X feels something
(b) X thinks something like this:
286 Pawel Kornacki
taoyan
(a) X feels something
(b) X thinks something like this:
(c) this is bad
(d) I feel something bad because of this
(e) I don't want this
(f) X feels something bad because of this
References
Bond, Michael H.
1991 Beyond the Chinese Face: Insights from Psychology. Hong Kong: Ox-
ford University Press.
Bucher, Urs
1986 Vocabulary of Modern Chinese. Felsberg: Tobun.
Chu, Chieh-fan
1970 A study of Chinese proverbs on health and hygiene. Bulletin of the In-
stitute of Ethnology Academia Sinica Taipei, 30(2): 165-237.
Darwin, Charles
1872 The Expression of Emotion in Man and Animals. London: Murray.
(Reprinted 1979, London: Julian Friedmann.)
Ekman, Paul
1992 An argument for basic emotions. Cognition and Emotion 6: 169-200.
1993 Facial Expression and Emotion. American Psychologist 48 (4):
384-392.
Goddard, Cliff
1991 Anger in the Western Desert: Semantics, culture and emotion. Man 26:
602-619.
Concepts of anger in Chinese 287
Ots, T
1990 The angry liver, the anxious heart, and the melancholy spleen: The
phenomenology of perceptions in Chinese culture. Culture, Medicine,
and Psychiatry 14(1): 21-58.
Porkert, Manfred
1976 Theoretical Foundations of Chinese Medicine. Cambridge, Mass.:
MIT Press.
Renmin Wenxue [People's Literature] (various issues). Beijing: Zuojia Chubanshe.
Roth, H. D.
1996 The inner cultivation tradition of early Daoism. In Donald Lopez (ed.),
Religions of China in Practice, 123-148. Princeton: Princeton Univer-
sity Press.
Russell, J.
1991 Culture and the categorization of emotions. Psychological Bulletin 10:
426-450.
1994 Is there universal recognition of emotion from facial expression? A re-
view of cross-cultural studies. Psychological Bulletin 115(1):
102-141.
Shaver, P., S. Wu, and J. C. Schwartz
1992 Cross-cultural similarities and differences in emotion and its represen-
tation. In P.Clark (ed.), Emotion, 175-211. London: SAGE Publi-
cations.
Shweder R.
1993 The cultural psychology of emotions. In: Michael Lewis and Jeanette
Haviland (eds.), The Handbook of Emotions, 417-431. New York:
Guilford Publications.
Van Brakel, J.
1994 Emotions: a cross-cultural perspective on forms of life. Prepared for
W. M. Wentworth and J. Ryan (eds), The Handbook of Social Psychol-
ogy, vol. 2.
Wang, Ling
1993 Kongzhifennu de xinlixue fangfa [Psychological methods of control-
ling anger]. Fuyin Baokan Ziliao, Xinlixue, B4(l): 73-76.
Wierzbicka, Anna
1992 a Semantics, Culture, and Cognition: Universal Human Concepts in
Culture-Specific Configurations. New York: Oxford University Press.
1992b Defining emotion concepts. Cognitive Science 16: 539-581.
1999 Emotions across Languages and Cultures: Diversity and Universals.
Cambridge: Cambridge University Press.
Xiandai Hanyu Cidian [A Dictionary of Modern Chinese]
1988 Beijing: Shangwu Yinshuguan.
Xiaoshuo Lin [Literary Forest]
1984-87 Ha'erbin: Wenyi Zazhishe (various issues).
Concepts of anger in Chinese 289
1. Introduction1
commas, and the third type in Roman characters. However, due to the syn-
cretism commonly found in real word usage, we can use italics for all three
types of entities without special comment.
As noted by many specialists (Tolstoj 1995: 314; Smelev 1997 a: 481), the
Russian linguistic picture of the world is characterised by the opposition of
the "elevated" and the "earthbound", the "world of the sublime" and the
"terrestrial world", with a marked preference for the former. (This dualism
would seem to be rooted in the peculiar character of Russian Orthodoxy,
which determines the features of Russian culture in general: the polar-
isation of value judgments and the absence of an axiologically neutral
zone; see Lotman and Uspenskij 1994). Many important concepts exist in
Russian in two variants corresponding to these two levels, and are some-
times denoted by different words - cf. the following pairs of words which
are contrasted according to the principle of "high low": istina (roughly,
'truth') andpravda (roughly, 'truth'), dolg 'duty' and objazannost' Obli-
gation', dobro 'good' and blago 'good, welfare'. The pair radost' (roughly,
'joy') and udovol'stvie (roughly, 'pleasure') provides an excellent example
of such a polarisation of the Russian linguistic consciousness.
The words radost' and udovol'stvie have been analysed by Pen'kovskij
(1991). Of the differences noted in that article there are, we suggest, two
main ones which determine all the others. The first is that radost' is a feel-
ing, whereas udovol'stvie is merely a "positive sensual-physiological reac-
tion" (linguistic evidence for this statement can be found in the peculiar-
ities of its metaphorisation and conceptualisation in general). The second,
and in a sense main one, is that radost' belongs to the "elevated", spiritual
world, whereas udovol'stvie belongs to the "low", profane, corporeal.
These fundamental differences involve other, more particular ones
(cf. Pen'kovskij 1991). Thus, the axiological polarisation within the pair
radost' - udovol'stvie is conditioned by the fact that radost' is associated
294 Irina B. Levonna, Anna A. Zalizniak
The grounds for the rapprochement of scast'e with radost' are the fol-
lowing. Firstly, both are "high" feelings. The words radost' and scast'e
often appear together in a text, e.g.:
The cliffs and sea, and slanting rays of the setting sun -1 seemed to see all this still
when I awoke I remember feeling joyful [= rod]. A sensation of happiness
[=scast'e], till then unknown to me, passed through my heart (Dostoevsky, "The
Adolescent").
Both states may lack any concrete reason, as in the foregoing example
and the following one:
Human emotions viewed through the Russian language 297
So death has brushed against me, yet I still cannot lose a sense of constant happiness
[= sast'e]. God only knows where it has come from and how it will end... There
seems to have been so much in my life - and I am happy [=soastliva] with life itself!
(N. Berberova. "The Accompanist").
respond to them and which seem to be universal (see, for example, Izard
1977; Johnson-Laird & Oatley 1989; for the criticism of the notion "basic
emotion" see Wierzbicka 1992a: 119 ff.).
First, the Russian word scast'e is by no means an "everyday word". It
definitely belongs to the "high" category and has a very strong emotional
charge, which produces two opposite tendencies in its usage corresponding
to the two extremes of the "Russian soul". The first one is the tendency to-
wards asceticism, anti-hedonism and a certain modesty (a kind of "pudeur"
that is sometimes ascribed to the ideology of Bolshevism, but has of course
a much longer history), which makes the speakers avoid using "high" and
"strong" words and class them as almost "improper", unmentionable ones.
The other, opposite, tendency corresponds to the Russian urge to talk
"about the main things" and vyvoraivat' dusu naiznanku (lit. 'turn the soul
inside-out' = 'pour one's heart out') which explains why, in spite of the
existence of the first tendency, the word scast'e is fairly frequent and char-
acteristic of Russian discourse (cf. data from the abovementioned article
Ufimceva 1996).
Second, sast'e in Russian is certainly not a "basic emotion", in any
sense of that term. It belongs to the sphere of that which is ideal and unat-
tainable in reality (cf. Pushkin's Na svete sfast'ja net... There is no hap-
piness in the world...') alongside the "meaning of life", "ideal love" and
other fundamental categories of being.
English happiness is evidently closer to dovolen 'content' and udo-
vol'stvie 'pleasure'. According to A. Wierzbicka, in contrast to the "weak-
er" English happiness, the French bonheur, German Glck and Italian fel-
icita express "a common European concept" of an emotion which "fills a
human being to overflowing, leaving no room for any further desires or
wishes" (Wierzbicka 1992 a: 299). The abovementioned words from Euro-
pean languages are indeed closer in meaning and in usage to the Russian
sast'e than English happiness, but there are also some important diver-
gencies here. One of them is the absence of a strong emotional charge and
its consequences (see above).
We must also mention the differences in the usage of French heureux and
Russian sfastliv, which derive from the fact that French has no equivalent
for Russian rad (lit. joyful = 'glad'). The meaning of Russian quasi-adjec-
tive rad in general corresponds to that of radost', but a bit "weakened" by
the fact of its use in etiquette formulas like Ja rad tebja videt' ('I'm glad to
see you'). The Russian word dovolen 'content', cognate with udovol'stvie,
Human emotions viewed through the Russian language 299
The connection of toska with the idea of 'crampedness' is the most per-
sistent (although between toska and tesno there is no etymological link): it
has obvious inner justifications and is also confirmed by a multitude of
semantic parallels. The semantic connection here is that the sensation of
crampedness, compression, is evidently a "physical analogue" for the feel-
ing of toska, e.g.:
...trevozit
Ee~ revnivaja toska,
Kak budto xladnaja ruka
Ej serdce zmet.
(Pushkin, "Evgenij Onegin", 6: III).
...Disturbs
her the ache of jealousy,
as if a cold hand
compressed her heart (Pushkin 1975: 228)
Finally, the connection of toska with tosno is as follows. The word toSno
(a fairly specific one in itself) is a derivative from the impersonal verb tos-
nit' 'to feel sick', but, unlike the verb which has both a literal and figurative
meaning, is used only figuratively, i.e., to indicate a certain state of mind
(soul). The connection of physical sickness with the state of the soul de-
noted in Russian by the word tosno and in general the association of tosno-
ta 'sickness' with dusa (cf. expressions such as s dusi vorotit, dusa ne
prinimaet, i.e. 'it makes me sick', lit.: 'the soul doesn't admit it') would ap-
pear to be explained by the simple fact that a sensation of physical sickness
is located in that part of the body (i.e., in the chest) where, according to
naive anatomy, the soul is. This aforementioned state of mind is close to
toska - when toska shows its least "poetic" and most sombre side. This
semantic similarity, reinforced by the phonetic and etymological ones, ex-
plains why the word tosno takes the place of the impersonal predicative:
i.e., similar to the triple sets of the type obidet'sja - obida - obidno ('to
take offence' - Offence' - 'it's offensive'), bojat'sja - strax - strasno ('to
fear' - 'fear' - 'it's frightening'), skucat' - skuka - skucno ('to be
bored' - 'boredom' - 'it's boring'), etc. we have toskovat' - toska - tosno,
cf:
Ax njanja, njanja, ja toskuju,
Mne tosno, milaja moja
(Pushkin, "Evgenij Onegin", 3: XIX)
Oh, nurse, nurse, I feel dismal,
I'm sick at heart, my dear (Pushkin 1975: 158)
There was an unbearable anguish in his heart [= toSno na duse], and the familiar,
well-read books could not drive this oppressive anguish [= toska] away (M. Gorky).
302 Irina B. Levontina, Anna A. Zalizniak
Of all the Russian writers Dostoevsky has always been regarded as the
most striking exponent of the "Russian soul": the Russian passion for suf-
fering, soul-searching, "clarifying relations", Russian nadryv, rushing into
extremes of humiliating oneself and insulting others, of the most foul self-
abasement and the most unrestrained arrogance, between adoration and
profanation - all this forms the world of Dostoevsky's characters, through
which the inhabitants of Western Europe, beginning from the school
bench, try to penetrate the enigmatic "Russian soul". On the other hand,
many studies have been devoted to Dostoevsky's language, certain features
of which have long since become commonplace in courses of Russian lit-
erature. However, the idea of linking Dostoevsky's language directly with
the Russian linguistic picture of the world did not appear until quite re-
cently and belongs to N. D. Arutjunova (see Arutjunova 1996, 1999).
In this connection we shall examine the typically "Dostoevskian" word
nadryv which expresses a characteristic extreme state of the "Russian soul"
and is quite untranslatable, and also the words obidno and sovestno, which
are less obviously but no less indisputably specifically Russian concepts
(and, incidentally, hold a place of considerable importance in Dostoevsky's
world).
Nadryv
The word nadryv is one of the most capacious, expressive and ingrained
words in Russian culture, which is very hard to translate into English. The
Russian poet Lev Losev enumerates in a poem the repellent signs of the
mad Russian life. Among other things the poem says:
And he also said in excitation:
I don't like those drunken nights,
Penitential frankness of drunkards,
The Dostoevski] nadryv of the informers
It is interesting that this word, similarly to the words avos', toska, leri,
etc. is frequently associated with the epithet "Russian". (On these combi-
nations see Plungjan and Raxilina 1996.)
This word comprises a poignant psychological sketch. Nadryv presup-
poses an emotional state when a person cannot handle his or her feelings.
The person is either overcome by emotions making him or her forget about
limits, manners and even decency, or they come out ponderous and disin-
genuous. In addition, nadryv presupposes a certain masochistic self-reflec-
Human emotions viewed through the Russian language 303
tion and hysterical confession which sounds jarring to the people around.
Cf.:
Bely suffered desperately, driven from humiliated meekness to furious arrogance - he
cried that it was a blasphemy to reject his love.... He spent several months abroad and
came back still suffering and with "Cup of the Snowstorms" - the weakest of his
symphonies because it was written in nadryv (V. XodaseviC).
It is noteworthy that the root -ryv- (-rv-) 'to tear' forms many words de-
scribing the emotional life of a human being: poryv 'impulse', razryv 'rup-
ture', sryv 'collapse' - as well as the modern slang otryv 'enjoying doing
something a lot'. The action of tearing is perceived as something closely
related to an emotional state. Therefore, the gesture of tearing open one's
shirt on one's breast (rvanuf rubaxu na grudi), which is considered char-
acteristically Russian, is very important.
The noun nadryv is mainly connected with the verb nadryvat'sja which
means 'to do something with a great effort'. But more important than the
context of work is that of shouting, crying and, especially, coughing. Con-
sider the expression nadryvat'sja ot kaslja 'to have a whooping cough', ad-
jective nadryvnyj, adverb nadryvno and the usage of the noun nadryv in the
word combination kasljat' s nadryvom 'to cough distressingly'. When we
say that a man nadryvajetsja ot kaslja, kaSljajet nadryvno or s nadryvom,
we mean a specific type of cough. It is not one where the throat aches, but
one that comes out of the depths of one's lungs and shakes the entire body.
This cough is not necessarily loud but hollow, as if a man were trying to
push out of his depths the things that hinder him to breathe and to live. Such
a cough takes a person's energy, and the person still coughs uncontrollably.
Similarly, to say something with nadryv does not mean to cry it out. One
can speak with nadryv in a hardly audible manner but in a way it is clear
that the words come out from the bottom of the soul. It is torturous to say
them but it is impossible not to say them.
Thus, it is clear why the word nadryv describes two opposite situations:
an uncontrolled emotional outburst or an expression of forced and inauth-
entic emotions. In the first case one shows one's deeply hidden intimate
feelings with a frightening frankness, that exposes the things which are
supposed to be most personal (cf. the abovementioned example about An-
drey Bely). In the second case one goes even further. One is so deeply in-
volved in introspection that one finds in one's soul something that does not
exist. That is why with nadryv extremely exaggerated, distorted, or even
imaginary feelings are expressed. In this case nadryv verges on falseness or
304 Irina B. Levontina, Anna A. Zalizniak
tasting it and trying to understand the world through this word. The major-
ity of cases of its usage relate to Katerina Ivanovna and her feelings to-
wards Dmitri. First Dmitri humiliated her in promising to give her money,
in order to save her father, only if she comes alone to take it from him. Then
he gave her the money and demanded nothing in exchange. Soon after that
she wrote him and told him she loved him and wished to be his wife. From
the feeling of insulted human dignity, sacrificial ecstasy, and hatred grow
love and passion (which are also accompanied by a certain desire for sac-
rifice). Katerina Ivanovna says:
"Even if he marries that... brute creature", she began majestically, "...From this day
forth I shall never, never give him up! she uttered in a kind of hysteria [= nadryv in the
Russian text-/.L., A.Z.} of pale, forced ecstasy." (Dostoevsky 1983: 215)
comme ilfaut. The raznoftntsy saw in the culture of the gentry falsity and
superficiality. Therefore, they rejected external lustre. Instead, they sug-
gested the cult of frankness and a deep understanding of things. The raz-
notintsy had a peculiar mode of behavior, which was a thundering blend of
reckless, unrestrained frankness, romantic melodrama, and a propensity to
the depraved. This blend is familiar to readers of Belinsky's letters to Ba-
kunin, Cernystevsky's diaries or almost any personal document from the
people in this social group or psychological type. (On the aesthetics of the
everyday behavior of the raznointsy see Paperno 1996.)
Returning to the theme of coughing, let us remember consumption - the
disease of the raznoZintsy. This consumptive whooping (nadryvnyj) cough
accompanies "the cough of the soul." This is an anguishing frank (nad-
ryvnyj) confession where the shameful and ugly truth about a person is re-
vealed.
Manifest drinking is a strikingly peculiar feature of the nadryv aesthetics
cultivated by the raznocintsy. It is said that What a sober person has on his
mind, a drunk has on his tongue: alcohol intoxication is a perfect moti-
vation for refusal to refrain from expressing one's feelings. This refusal be-
came culturally significant fact, and the abyss of alcoholism a synonym of
spiritual depth. Since then this type of behavior has never had such a high
cultural status. It degenerated into self-parody.
The 20th century gave a new aesthetic dimension to the word nadryv. For
Dostoevsky nadryv was still interesting and aesthetically attractive though
it was fraught with lies. Now it is usually assessed as tastelessness: "Do
you like Andreev? - No, states S. Dovlatov. He is flamboyant and with
nadryv". And Brodsky says about Pushkin in admiration: "Still Pushkin
belongs to the gentry, don't forget it. If you like, he is an Englishman - a
member of the English Club - in his attitude towards reality he is reserved.
He has not got the thing that is called nadryv".
man krnken, cf. also Russian ranit'). The corresponding emotion is ob-
viously peripheral for English and German (at least neither word occurs in
the list of thirty-six common English emotion words in Wierzbicka 1972
and in the list of the forty most frequent German emotion words in Mel'ouk
and Wanner 1996). By contrast, for the Russian language the words obida
and obidet'sja are extremely frequent, and the emotion denoted by them is
one of the most important for the Russian linguistic consciousness (the
word obizat'sja is included in the list of forty emotion words described in
lordanskaja 1970).
The verb obidet'sja denotes a certain emotion which arises as a reaction
to an action by another person. One cannot obidet'sja on oneself or on a
natural phenomenon - although one can, for example, be angry with one-
self or despair of the bad weather.
What exactly is the emotion of obidal Obida consists in feeling pity for
oneself linked with a certain disillusionment in another person, caused by
the fact that this person by his or her words or acts has shown, as it were,
less affection or regard for us than what we had expected from him or her
(the disillusion must be a local one, which has not yet become global: in the
latter case there is no room left for obida). Obida is perceived as experi-
encing something unfair towards yourself (cf. the inclusion of the idea of
'unfairness' in the interpretation of the word obizat'sja in lordanskaja
1970: 19). The conception of unfairness arises from the idea that the other
person owes to oneself something good (good feelings or respect), often
without any definite grounds. So, obida is an aggressive feeling: it means a
reluctance to accept the situation that seems unfair, a demand for compen-
sation. Obida is opposed to a resignation, on the one hand, and to the state
of forgiveness on the other.
and "I don't feel anything bad toward this person". Note that Ja prostil
forgave' differs from Ja ne obidelsja first and foremost by the fact that it
does not include the negation of the component (e), meaning don't feel
anything bad toward this person (even though I think the action was un-
fair)'.
The Russian obida can be compared to the Polish word przykro, de-
scribed by Anna Wierzbicka (this volume), who writes: "It is clear that the
concept of Offended' has a good deal in common with the Polish concept
of przykro. Roughly speaking, what the two share is some perceived 'defi-
ciency of regard'. But first, in offended, this 'deficiency' concerns an ap-
parent absence of 'good thoughts about someone'; second, in offended this
appears to be intentionally displayed, whereas in przykro it can be (and
typically is) betrayed or implied unintentionally; and third, in the case of
offended the experiencer must be different from the agent (I cannot be of-
fended by my own action), whereas in the case of przykro the experiencer
can coincide with the agent: I can feel przykro because of something that I
have done myself (causing a bad feeling for someone else)".
The Russian obida shares with the Polish przykro the first two features
(the third making przykro close to the Russian sovestno, see below), and is
opposed to the English offended more or less in the same way.
So, in accord with the second feature (common to przykro and obida)
obida often arises as a result of an inference made by the "offended" per-
son. Different people are more or less inclined to such inferences. Hence
obida is an egocentric feeling, the source of which is rooted, in the final
analysis, in ourselves. A disposition to obida is the characteristic of the
"offended" person: therefore an obidiivyj person is someone who is in-
clined to take offence (and not to give it - which, from the point of view of
Russian morphology is just as possible, cf. obmanivyj = 'which deceives'
and not 'which is deceived').
The meaning of the impersonal obidno is even more specific. The emo-
tion expressed by the word obidno is somewhat different from that con-
tained in the words obida and obidet'sja. Thus we can say Mne obidno, to
nikto ne pritronulsja k moemy salatu 'I'm offended (lit.: to.me it's.a.pity,
ADV.) that no one has tried my salad', but it would hardly be correct in this
situation to obidet'sja 'take offence' at the guests or to harbour an obida
against them (provided, of course, that there are no grounds for suspecting
deliberate malicious intent in this act). On the other hand, the phrase Mne
obidno, to Ivan ne pozdravil menja s dnem rozdenija am offended that
Human emotions viewed through the Russian language 309
Ivan did not wish me a happy birthday' sounds rather strange. It would be
more natural to say Ja obidelsja na Ivana za to, cto on ne pozdravil menja s
dnem rozdenija or Mne obidno, cto nikto iz moix druzej ne pozdravil menja
s dnem rozdenija am offended that none of my friends wished me a
happy birthday': the former sentence describes my attitude towards Ivan,
the latter, my feeling pity for myself. In the impersonal obidno, the obidik
(the person who causes the offence) disappears: there is no syntactic op-
portunity for mentioning this person. Thus, whereas obidet'sja and obida
describe an attitude of the offended person to the offending one, in obidno
the stress shifts to the offended person himself. We feel obidno when we
perceive, as A. Wierzbicka says, a "message of indifference" from the
outer world.
The feeling expressed by the word obidno arises when something dear to
the person in question is abased, ridiculed or simply underrated. This for-
mula can serve as a generalisation for all cases: the "something dear" is
either the person him/herself, or another person, or thing that is dear to
him; it can also be the sum total of one's own wasted efforts: it is obidno if
they are spent in vain (cf. the abovementioned example with salad).
Sometimes obidno comes close to zalko, cf.:
Other examples: obidno /zalko ujezzaf ran'se sroka 'it's a pity to leave
before the time is up'; obidno I zalko, cto ne udalos vstretit'sja 'it's a pity
we did not manage to meet'. Obidno and zalko in this case describe the
feeling that arises at the thought that circumstances which have taken a
certain turn might have developed differently and that this "differently"
would have been better. Both words express the idea of some positive al-
ternative. It is rooted in the very word zalko (see Zalizniak 1991); in obid-
no it appears by the transformation of the idea of a "just" state of affairs
which is being violated. The difference between impersonal zalko and
obidno result from the fact that the "raison d'etre" of sozalenie (regret) is
"it could be otherwise", whereas the "main idea" of obida is "it must be
otherwise".
310 Irina B. Levontina, Anna A. Zalizniak
So, we can say zalko, ito sejas leto: nel'zja pokatat'sja na lyzax 'What a
pity it is summer now and we can't go skiing', but we cannot say *Obidno,
to sejias leto. Obidno is appropriate only when the positive alternative is a
more "normal" version of the development of events, cf.: Kak obidno, ito
isportilas' pogoda 'What a pity the weather has turned bad'; obidno, to iz-
za kakogo-to pustjaka vs sryvaetsja 'What a pity that it has been called off
because of a trifle'; tak obidno zabolet' v pervyj den' kanikul 'What a pity
to get ill on the first day of the holidays'. In a construction with a subordi-
nate perfective infinitive, obidno is close in meaning to ugorazdilo (see Za-
lizniak and Levontina 1996), in the distribution of the communicative
status of the components. Both zalko and obidno combine with the subor-
dinate infinitive form, a construction specific to the Russian language, and
are without any satisfactory equivalent in European languages.
The peculiarity of the Russian obida, obidno can also be seen by com-
paring it with another Russian word: dosada, dosadno (roughly, 'annoy-
ance', 'annoyed'). Whereas obida is always felt in relation to someone
else, i.e. it is a feeling towards someone, dosada can only be felt about
something, some circumstances which prevent the fulfillment of one's
wishes: this could also be another person, but in dosada this person ap-
pears only as a mechanical obstacle. Therefore one can dosadovat' 'be an-
noyed' with oneself, for example, at one's own stupidity or bad luck. The
source of dosada is a single event from which no consequences are
drawn - unlike obida, where what is important for us is usually not the act
itself, but what "lies behind it" (namely, an attitude towards us which we
do not like). Obida is akin to grief and tears, dosada to anger and aggres-
sion.
To conclude we will say that the inner state described by the Russian
word obidet'sja is fairly close to that presented by the English to be hurt
(see the explanation of the latter in Wierzbicka, this volume), the main dif-
ference between the respective concepts being the degree of their salience
in the linguistic picture of the world. The Russian obida is an elaborated
concept of a high cultural significance, while the same cannot be said about
the English to be hurt. Linguistic evidence of this is seen in the existence of
a substantive and other derivatives (lacking for the English to be hurt}. The
feeling called obida possesses a certain independence in Russian, in the
sense that one can do different things with obida: derzat' 'to hold' or kopit'
'to accumulate\pomnit' 'to remember' or zabyvat' 'to forget' them; obida
itself can rasti v dushe 'to grow up in the soul' orproxodit', uletuiivat'sja
Human emotions viewed through the Russian language 311
Sovestno
The concept of sovest' 'conscience' plays an important role in Russian cul-
ture. Sovest' performs a double function: on the one hand, it enables us to
distinguish between good and evil, and on the other, guides our actions, di-
recting them towards good. As Ju. D. Apresjan puts it, in the Russian lin-
guistic picture of the world sovest' is seen as "a strict internal judge" and as
"a moral restraint which blocks the realisation of amoral desires" (Ju. D.
Apresjan 1995: 353).
The fact that sovest' can gryzt' 'gnaw', terzat' 'tear to pieces', mudt'
'torment' and in general arouse various unpleasant sensations postfactum,
when the act has already been committed (in the case when sovest' has per-
formed its first function and failed in its second) is common to many lan-
guages. The sensations aroused by sovest' are connected usually with the
idea of "gnawing, biting" - cf. Russian ugryzenija sovesti, German Gewis-
sensbisse, French remords, English remorse having similar etymology (no
longer alive in the last two). Specific to Russian is the fact that it has an
idiomatic way of representing the role of sovest' in taking a decision: the
predicative adverb sovestno, whose meaning differs considerably from that
of the substantive sovest'. Let us take a closer look at this.
legedly universal (see Izard 1977) human emotions, whereas the word so-
vestno contains a specifically Russian concept and may be translated into
other European languages only insofar as it is synonymous with stydno
(cf. the dictionary translations offered for Russian sovestno: English to be
ashamed, German sich sch men, French avoir honte). (On the concepts
styd and sovest' see also Arutjunova 1996, 1999.)
The point of intersection for the spheres of sovestno and stydno are bad
actions, i.e. actions condemned by moral norms (cf. Apresjan 1997 b: 417).
However, each of these concepts also has its own spheres. First and fore-
most, unlike sovestno, which characterises actions (i.e., intentional acts
controlled by the subject), the sphere of stydno covers the most varied cir-
cumstances, including some that do not depend on the subject at all (as, for
example, one's origin or physical defect); stydno may be for actions com-
mitted by someone else. Thus, it can be both stydno and sovestno to mis-
lead other people (because a lie is a morally reprehensible, i.e. bad action),
whereas to believe cock-and-bull stories can be only stydno, but not sovest-
no - because the stupidity you thereby reveal doesn't belong to the cat-
egory of "bad actions".
On the other hand, stydno is not necessarily for something bad. It may be
simply something intimate: something which should be private, but has be-
come public. This includes various circumstances concerning the body and
sex, and also, for example, the manifestation of strong feelings or excessive
exposure. Sovestno cannot apply to any of this.
The emotion of shame can be rendered in Russian by a verb: stydit'sja 'to
be ashamed'. But there is no *sovestit'sja, because there is no such emo-
tion except for the construction with a subordinate infinitive where it ap-
pears in the perfective form posovestit'sja: X posovestilsja to do = 'X
hasn't done because X felt sovestno about it' (cf. section 5).
The word neudobno is also specifically Russian. The nearest English
equivalent inconvenient differs from neudobno in frequency and stylistic
register, neudobno being a common and everyday word. The original
meaning of the pair udobnolneudobno is physical comfort. Neudobno re-
flects a sense of the possible violation of some norms of decency or rules of
conduct, which are usually rather complex and subtle and are determined,
inter alia, by relations between the people in question (cf. Mne neudobno
ego ob tom prosit', a ty mozes' eto sdelat' 'It's inconvenient for me to ask
him, but you could'). In certain contexts sovestno seems almost synony-
mous with this neudobno, cf. Mne sovestno/'neudobno, to ty vse vremja
Human emotions viewed through the Russian language 313
stydno
(a) someone else can think something bad about me
(a) because of this I can think something bad about me
(a) because of this I feel something bad
(d) like I do when I know that I have done something bad
sovestno
(a) I have done (want to do) something
(a) someone else can feel something bad because of this
(a) because of this I feel something bad
(d) like I do when I know that I have done something bad
314 Irina B. Levontina, Anna A. Zalizniak
neudobno
(a) I want to do something
(a) someone else can feel something bad because of this
(a) because of this I feel something bad
passing', cf. obnesti [kogo-to ugo$eniem] 'to leave someone out when
serving food', obdelit' 'do someone out of his share', obvesit' 'cheat in
weighing', obojti vnimaniem 'neglect, disregard'. Thus, the Russian
obida has grown out of the idea of obdelit' vzgljadom 'not to look at
someone'.
A similar idea is included in the word prezrenie 'contempt'; prezirat'
[kogo-to] literally means 'to look over someone', i.e., not to see the person
(being somehow above him/her).
In this section we shall consider the words drug 'friend', obs enie 'contact'
and otnosenie 'relation, attitude'. The word drug (and also its synonyms in
Russian and equivalents in other languages, in particular, English friend) is
analysed in detail in Wierzbicka (1997, ch. 2). Moj drug in Russian is a per-
son I want to be with and talk to, and to talk po duSam 'intimately, sin-
cerely' - cf. Wierzbicka's components want to be with this person',
want to talk to this person' and' I can say anything to this person' (1997:63).
All that can be added to Wierzbicka's analysis is perhaps that the Rus-
sian word drug seems about to share the same destiny as English friend.
This word in English, as Wierzbicka points out, has undergone a semantic
evolution consisting of a "weakening" in its meaning: from the meaning,
close to that of the Russian drug, with the main semantic component
want to do GOOD THINGS FOR this person' to the meaning centered round
the idea want to do THINGS WITH this person' (Wierzbicka 1997: 40). In
modern Russian discourse (particularly in the speech of people who have
lived abroad for a long time) the use of the word drug increasingly appears
as a nomination for people with whom the speaker goes on Sundays to a
barbecue.
The words obS enie (roughly, 'contact, communication') and obscat'sja
'to have contact with' have not yet attracted the attention of linguists. Yet
these words are very specific and difficult to translate. There are at least
two features which distinguish them from their translation equivalents. The
first is that the words obScenie and particularly obscat'sja belong to every-
day language and contain the idea of something "warm" and "friendly";
the second is that obscat'sja in the modern language refers to a concrete
process.
316 Irina B. Levontina, Anna A. Zalizniak
Obscat'sja in Russian means 'to talk with someone for a while in order to
preserve dusevnyj kontakt with that person' (dusevnyj kontakt, close con-
tact, literally 'contact between the souls of the interlocutors'). The fact that
the generic seme here is 'to talk' is demonstrated by the existence of the
valency of content: na kakuju temu On what subject'. The talk which forms
the content of obscenie need not necessarily be about "highbrow things". It
can be about mere trifles as well, because the main thing in obSenie is to
strengthen friendly contact. The word obsat'sja also contains the idea of a
certain purposelessness in this activity and the radost' or udovoVstvie ob-
tained from it (the former being due to the abovementioned dusevnyj kon-
takt, the latter, to the intellectual interaction, cf. section 2.1). This meaning
of the word obscat'sja appeared relatively recently: it is a colloquial new
formation and by older and/or more conservative speakers of the Russian
language is regarded as vulgar. Not long ago this verb used to be used only
to denote an "generic state" (using Z. Vendler's terms), like pitat'sja 'to
feed' or pravit' 'to rule', cf. Deti v etom vozraste obscajutsja v osnovnom
so svoimi sverstnikami 'Children of this age mix mainly with their peers'.
It is characteristic that from the verb obSoat'sja in modern colloquial Rus-
sian different manners of verbal action are easily formed: delimitative
(poobscaf sja polcasa 'to do it for half an hour'), perdurative
(proobscat'sja ves' veier 'to do it all evening'), saturative (naobscat'sja
vdovol' 'to do it to one's heart's content') and even semelfactive (pb-
scnut'sja 'to do it within a very short time, but intensively'). The last verb
is regarded as a vulgarism not only by the purists.
In order to do something, including poobsiat'sja s druz'jami, Russians
must do three difficult things: sobrat'sja ('to make up one's mind', see Za-
lizniak and Levontina 1996), vybrat'sja (more or less the same as so-
brat'sja, but when speaking about actions one does outside, e.g. visiting
people) and dobrat'sa (reach the place). All three words (derived from the
same root meaning 'to take') encode language-specific concepts and are
untranslatable: the first two of them reflect properties of the "Russian
soul", and the third, those of the Russian landscape (and mode of life in
general). Indeed, the five hundred years that, according to Pushkin, Russia
needed to improve its roads, are not yet over.
Another purely Russian occupation, which is, in a sense, something quite
opposite to the obScenie, is called vyjasnenie otnosenij (lit. 'clarifying re-
lations'). This language-specific expression means: to express resentment
to each other concerning not actions or statements (i.e., something "tan-
Human emotions viewed through the Russian language 317
Zalost' in the Russian language is both one of the most spontaneous and
one of the most culture-mediated emotions. It is an uncontrollable and im-
mediate response of the soul to the suffering of others. Zalost' may em-
brace, cover, squeeze the heart, etc. Cf.:
An unexplainable zalost' always grew in me when I spoke to him (N. Berberova,
"Italics are mine").
318 Irina B. Levontina, Anna A. Zalizniak
v
During the Soviet epoch, due to the cult of force fostered by official So-
viet ideology, the mere word zalost' became associated with something hu-
miliating - cf. Gorky's expression which became a pithy saying Zalost'
unizaet eloveka 'Pity humiliates man'. Soviet books and movies are full
of similar rhetoric. Even V. Vysotsky sang: "When I see broken wings/I do
not feel any zalost' and there's a reason for it: I hate violence and impo-
tence".
Human emotions viewed through the Russian language 319
When understood this way, zalost' is different from love: one can often
hear it said in Russian something like this: Ne ljubit, a tol'ko zaleet 'He
does not love her, he simply has some pity for her'. But this concept of
zalost' is alien to traditional Russian culture, where "to feel zalost'" means
"to love", and this concept is specific to Russian (cf. Wierzbicka 1992 a:
168-169; Stepanov 1997: 281). In Russian culture zalost' was always
understood as one of the supreme values, as a feeling very close to love and
not offensive at all. Cf.: "Sometimes people experience a genuine and pro-
found feeling on Earth. There is always a trace of zalost' in it. The more we
love, the more we perceive the object of our admiration as a victim" (B.
Pasternak, "Doktor Zivago"). V. V. Rozanov wrote about his sick wife:
"Her face was all in tears. I froze. Both in admiration and zalost'". This
philosopher believed zalost' to be one of the fundamental values of life; he
wrote: "Is there any zalost' in the world? Beauty - yes, sense - yes. But
zalost")... Do stars feel zalost'l The mother does: and let her be above the
stars".
A. Platonov usually names the feeling of maternal love by the word
zalost', cf.:
It turned out that Simon was alive because he felt his mother's zalost' for him, and he
could keep the peace of her mind just remaining safe in this world. (Platonov 1978 a:
299)
In this world, only his sister would live on, but she then would give birth to a child,
and zalost' for it would become stronger than her grief for her dead, crushed down
brother." (Platonov 1978b: 60)
The mother half opened her eyes, they were suspicious, prepared for any kind of mis-
fortune in life, and already whitish with indifference - and she said in her own de-
fense: "I don't feel sorry for you [= nine tebja ne zalko] now and I don't need any-
one -1 have become like stone; please put out the lamp and turn me on my side, I
want to die." (Platonov 1978b: 60)
Rodnoj
There are three ideas that form the basis of terms of endearment. Some of
them directly indicate the respective feeling: ljubimyj, zelannaja, ljubov'
moja, radost' moja, siast'e moe, nenagljadnaja ('beloved', 'wished', 'my
love', 'my joy', 'my happiness', 'the person I can't stop looking at admir-
ingly'), etc. Others express the value and uniqueness of the object of the
feeling: dorogoj, edinstvennaja, bescennyj, zolotoj, sokrovisie moe, zo-
lotce ('dear', 'the only one', 'priceless', 'golden', 'my treasure', 'my
gold'). And lastly, many words indicate the pleasant features or prettiness
of the object of the feeling (and other more sophisticated attributes): xoro-
Saja 'good', sladkij 'sweet' etc.; cf. in English: honey, sweetypie and
sweetheart. This idea is often expressed through comparison with small
and pretty animals - zajka, kiska ('bunny', 'pussy'), etc.: a creature of
small size and young age is seen as always nice.
Among the numerous Russian terms of endearment, however, there is
one which stands apart. It seems to be one of the most important and, no
doubt, the most peculiarly Russian term of address - rodnoj (literally,
'native'). It has a diminutive form rodn'enkij and a number of other de-
rivatives. The obsolete addresses rodimyj and krovinoska 'a small drop of
blood' are close to it in their meaning. The hero of Ju. Malecky comments
on the address rodnaja in the following way: "Rodn'en'kaja. Closer than of
the same [Russian] tribe, belonging to a very peculiar kin of my own rather
than to a masculine or feminine or neutral gender, she is the only one of
just-my-own kind."
The address rodnoj means: I think about you as if you were a (beloved)
blood relative. This type of address is rare even in other Slavic languages. It
differs from others by its special emotional coloring. Rodnoj, rodnaja ex-
presses profound tenderness, confidence, feeling of mutual understanding
and spiritual intimacy rather than romantic love and passion. "When this
adjective is used as a form of address outside the family, and outside the
Human emotions viewed through the Russian language 321
domestic circle, its great emotional strength is also linked with the impli-
cation of belonging, a total lack of distance and the existence of ties felt to
be absolute and unseverable" (Wierzbicka 1997: 84). Although rodnoj is
related to erotica to a lesser extent than other words of love, the degree of
intimacy of that word is higher than of the words milyj (darling) and even
ljubimyj (beloved). One would hardly call a person rodnoj at the beginning
of a love affair, even when the feeling of love has been already expressed
through other words of endearment. Some people do not have the word
rodnoj in their love lexicon at all, as it seems to them shockingly open.
The Russian adjective rodnoj primarily states the existence of blood ties.
However, rodnoj is not quite the same as rodstvennik (relative); cf.:
Nastja, her daughter and the only rodnoj person, lived far away, in Leningrad (K.
Paustovskij).
If in this phrase we replace the word combination "the only rodnoj per-
son" by the combination "the only relative", the meaning will change. The
original message - human loneliness will be replaced by data from a
questionnaire. The word rodnoj has some special warmth - as if blood re-
lations were not only an objective fact, but something one feels in one's
heart:
Suddenly Nadja felt in her heart that aunt Frosja was there her only rodnoj person.
She saw the familiar shape of her fingers similar to that of her mother's and the fam-
iliar slightly high cheekbones. And she felt a sudden tenderness for aunt Frosja as
never before (Yu. Trifonov).
Just because rodnoj is more than simply a relative, this word freely com-
bines with such words as such and forms degrees of comparison (rodnee,
samyj rodnoj):
"Mother! How could it happen? I have no one else in my life who is or will be rod-
nee" (K. Paustovski).
Since the word rodnoj means primarily not "of the same blood", but
"someone organically linked to the speaker", it is widely used to describe
the relationship of people who are not blood relatives. In Russian it is poss-
ible to "become rodnoj", cf.:
Once Katja even burst into tears - though she never cried - and suddenly those tears
made her extremely rodnoj to him, made him feel acute pity and a sort of guilt with
respect to her" (I. Bunin).
322 Irina B. Levonna, Anna A. Zalizniak
Blood relationship is a great value in all cultures, being a part of the basic
opposition "own vs alien". But for the Russian culture the feeling of kin-
ship is especially important. Moreover, in Russian the power of love can
transform a stranger into a sort of blood relative: rodnoj and even rodneje
rodnogo (closer than rodnoj}. The feeling of unity with the beloved person
can be so strong that this unity seems to be as indissoluble as blood ties.
Naturally, this feeling can be expressed, one way or another, in any other
language, but the Russian language has canonised it through providing a
special word for it and putting this word into the mouths of people speaking
of love.
The metaphor of blood relations is widely used in Russian colloquial and
common forms of addressing strangers. In other than Slavic languages one
can hardly find such an abundance of this type of address forms: otec, pa-
pasa, mat', mamasa, synok, doka, sestrenka, brat, bratok, bratcy, tetka,
djadja, ded, babuSka, babulja, vnudka ('father', 'dad', 'mother', 'mum',
'son', 'daughter', 'little sister', 'littlebrother', 'brother', 'brothers', 'aunt',
'uncle', 'grandfather', 'grandmother', 'granny', 'my dear granny', 'grand-
daughter') (cf. Lixaoev 1987; Kronhaus 1996). The form of address is de-
termined by the speaker who correlates his or her own age with that of the
addressee and figures out what relative that person could be for him or her.
It should be noted that religious addresses brat (brother) and sestra (sister)
are different in principle: they have nothing personal in them and just indi-
cate equality of all people before God.
So, for the Russian culture kinship is a strong value that has a strong
emotional coloring to it. Love for one's own brethren however does not
imply indifference or hostility to others. On the contrary, the warmth of
familial or congenial relations serves as an example of a kind attitude to
people in general. The linguistic facts that have been discussed here seem
to confirm the traditional conception of "breadth" (= Sirota) of the "Rus-
sian soul", its ability to embrace the entire world with a kind of parental,
fraternal or filial love.
The word soskuiit'sja implies that the emotion felt becomes ever
stronger, the longer the separation; it presupposes a certain "accumulation"
of feeling. It is correct to say "In a fortnight he soskuiilsja even more." On
the contrary, the sentence *Snaala on soskuailsja, no ierez dve nedeli os-
voilsja i perestal skudat' (At first he soskuftlsja, but in a fortnight he got
accustomed to the situation and this feeling disappeared) sounds incor-
rect - although in real life such a situation would be quite normal. For in-
stance, if a young man has left for another city to study there it will be cor-
rect to say: "At first he was missing home a lot, but then he becomes com-
pletely carried away with student life, and he absolutely forgot his home
and mother". But in this case we must use the verb skuat', and not sosku-
Zit'sja because soskuit'sja doesn't allow for the possibility of reverse
development of the feeling: it presupposes that the longing will persist till
the desired resumption of the contact.
324 Irina B. Levontina, Anna A. Zalizniak
So, razluka is a particular state that has both internal and external as-
pects. To be in razluka means to be separated physically but remain to-
gether emotionally. This idea distinguishes the Russian razluka from ras-
stavanie as well as of the English separation, French separation, German
Trennung, Italian separazione etc. All these words can designate both
physical and emotional distancing in various combination, but they lack
that sense of contradiction between the internal and external aspects which
is inherent in the Russian word razluka. Moreover, the word separation
(and its equivalents in other West European languages) is used as a legal
term designating a particular marital status: a state normally preceding di-
vorce, when the spouses live separately. This kind of separation can take
place even within the same flat. In this case the word separation (German
Trennung etc.) becomes an opposite to Russian razluka, because it desig-
nates a situation when two persons are physically together while emotion-
ally separated from each other.
Normally, razluka is a temporary absence of contact. This word wouldn't
be used in the situation where no reunion is planned. It would be strange to
say about a person who is leaving for good (without the possibility of re-
turn, as it was e.g. for Soviet emigration in the '70s), that he is facing raz-
luka from his relatives. It would be as strange as to speak of making a stop
when one has no intention to travel onward. When lovers think of their sep-
aration in terms of razluka, they hope that their feelings will withstand the
razluka, that the beloved person will not stop loving and waiting. In the
Russian lexicon of love there is a special term for it: dozdat'sjalne
dozdat'sja ('not stop/stop loving', literally, a resultative mode of action
from the verb meaning 'to wait'), cf.:
Razluka bystro proneslas'
Ona menja ne dozdalas',
No ej proSoaju
(V. Vysotskij. "Toj no'ju ja ne pil ne el")
Human emotions viewed through the Russian language 327
The words razluka, razlu it' derive from the root luk and are etymologi-
cally connected with the word luk 'bow' (see Fasmer 1996, II: 537);
cf. words with other prefixes: otlucit' (ot cerkvi) 'excommuniate', otlu-
328 Irina B. Levontina, Anna A. Zalizniak
All that geometry and physics were invented by Brodsky when he lived
in Russia. Then his beloved was an American. But when Brodsky crossed
the ocean, he retained this happily found lyrical construction and started
addressing it to a Russian love. Having changed his address, he changed
also the addressee, so that both could be, as before, separated by the ocean.
His life became a realisation of his own metaphor.
5. Conclusion
The Russian language creates language-specific ways of viewing emo-
tional states first of all owing to the words which contain unique (i.e. absent
in other languages in a ready-made form) conceptual configurations, the
analysis of which was the subject of the present article. However, a no less
important part of the picture of the world seen through the Russian lan-
guage can be found in the unique conceptual configurations contained in
the meaning of syntactic constructions.
The Russian language possesses a formal device which makes it possible
to express idiomatically the connection between the inner states of subjects
Human emotions viewed through the Russian language 329
Notes
1. This work was supported by the Research Support Scheme of the Open Society In-
stitute, grant # 797/1997. We express gratitude to those who have assisted with this
paper: Konstantin Bogatyrev, Jean Harkins, Steven Hassman, Alexej Smelev, Eliot
M. Tretter, Anna Wierzbicka, Andrej Zalizniak.
2. Some interesting conclusions concerning the present state of the Russian linguistic
consciousness have been obtained by processing data from the new Associative Dic-
tionary of Russian (see Ufimceva 1996). On the basis of the frequency with which a
word occurs as a response to a given verbal stimulus the author comes to a con-
clusion about the relative significance of this or that concept for the Russian lin-
guistic consciousness. The results obtained by N. V. Ufimceva confirm both certain
common ideas about the "Russian soul" and also the results of linguistic analysis by
other authors. In particular, words with the highest rate (and higher than the corre-
sponding words in English) in modern Russian include: radost' 'joy', stast'e 'hap-
piness', drug 'friend' and razgovor 'conversation'.
3. Cf. the explanation given in Dictionnaire dufrangais contemporain, by J. Dubois et
al. (Paris: Larousse, 1971): "Seditd'unetreanimedontlesdesirs, les gouts sontsat-
isfait, qui a ce qui lui plait". And the difference between rad and dovolen in general
corresponds to the opposition radost' - udovoljstvie, which we have considered
here, both in their syntactic properties and the presence of a mental component
(cf. Wierzbicka 1988: 312, Zalizniak 1992: 41).
4. Etymologically both words go back to a root meaning 'empty'. It is noteworthy that
the Russian v/ordpustoj 'empty' reproduces the same semantic derivation 'empty'
-'vain', as in: pustye ugrozy 'empty threats',pustaja trata vremeni 'time-wasting'
vpustuju 'in vain'.
5. Lexical function, a notion introduced by I. Mel'Cuk (e.g. Mel'cSuk 1995: 26), associ-
ates with a given lexical item a lexeme or phrase which expresses a certain general
meaning, such as 'very' or 'great' (LF Magn), 'begin' (LF Incep) etc.
6. The differences between obida and oskorblenie (note that the latter doesn't describe
an emotional feeling) consist in the fact that oskorblenie is caused when socially de-
Human emotions viewed through the Russian language 331
termined norms are violated, whereas obida arises on individual, personal grounds:
the set ofoskorbitel'nye (causing oskorblenie) utterances is a part of the cultural text
of the given social and cultural community, whereas the set of obidnye (causing
obida) utterances cannot be determined. One can say also that oskorblenie arises
when honour is involved, in obida - feelings. The Russian language reacts more
sharply to the latter (cf. Ufimceva 1996 on the relatively slight significance of the
concept of cest' 'honour' for the Russian linguistic consciousness).
Cf. the famous translation by Catullus (Carmen LI):
Hie mi par esse deo videtur
llle, si fas est, superare divos
Qui sedens adversus identidem te
Spectat et audit
There is a third item here - the word skuCat' mentioned above. We do not treat it in
detail here, as it is less specific than soskuiit'sja and razluka (cf. English to miss,
French s'ennuyer [de quelquun]. This word is related to the word soskuiit'sja, but
its tone renders it fairly compatible with the word razluka. Expression skuat' v raz-
luke 'to miss somebody in separation' is a very frequent word combination in Rus-
sian. These two words could be similarly explained through each other: one can say
that razluka is such a type of separation when people skuajut 'miss each other',
and, vice versa, skuat' means to have such a feeling which one normally experi-
ences in razluka. If a person does not skucaet v razluke, that means there is no raz-
luka at all.
References
Apresjan, Jurij D.
1995 Obraz Celoveka po dannym jazyka [The model of a person based on
linguistic evidence]. In Jurij D. Apresjan, Izbrannye trudy, vol. II,
348-388. Moscow: Jazyki russkoj kul'tury.
1997 a Radovat'sja, likovat', torzestvovat' [dictionary entries for Russian
verbs comparable to rejoice]. Novyj ob"jasnitel'nyj slovar' sinonimov
russkogo jazyka, 312-315. Moscow: Skola Jazyki Russkoj Kul'tury.
1997 b Stydit'sja, smu$at'sja, stesnjat'sja, konfuzit'sja [dictionary entries for
Russian words comparable to shame]. Novyj ob"jasnitel'nyj slovar'
sinonimov russkogo jazyka, 417-423. Moscow: Skola Jazyki Russkoj
Kul'tury.
Apresjan, Valentina
1995 Emotions: Linguistic representation and cultural conceptualization.
PhD dissertation. Los Angeles, USC.
Apresjan, Valentina and Jurij D. Apresjan
1995 Metafora v semantieskom predstavlenii emocij [Metaphor in the
semantic representation of emotions]. In: Jurij D. Apresjan, Izbrannye
trudy, vol. II, 453-465. Moscow: Jazyki russkoj kul'tury.
332 Irma B. Levonna, Anna A. Zalizniak
Arutjunova, N. D.
1996 Stil' Dostoevskogo v ramke russkoj kartiny mira [The style of Dos-
toevsky and the Russian model of the world]. Poetika, Stilistika, Jazyk
i kul'tura. Pamjati T. G. Vinokur, 61-90.
1976 Predloienie i ego smysl [The sentence and its meaning]. Moscow:
Nauka.
1997 sty de i stuzhe [ On the relatedness of the Russian words for shame
and cold]. Voprosy jazykoznanija 2: 59-70.
1999 Ponjatie sty da i sovesti v tekstax Dostoevskogo [The concepts of
shame and conscience in the works of Dostoevsky]. In N. D. Arutju-
nova and I. B. Levontina (eds.) Logieskij analizjazyka. Obraz ielove-
ka vjazyke i kul'ture, 320-345. Moscow: Indrik.
Beyer, Th. R., Jr.
1995 Marina Tsvetaeva and Andrej Belyj: Razluka and Posle razluki.
Wiener zionistischer Almanack 35: 97-132.
Baranczak, Stanislaw
1990 Breathing under water and other East European Essays. Cambridge,
MA: Harvard University Press.
Cernyx, P. Ja.
1994 Istoriko-etimologiieskij slovar' sovremennogo russkogo jazyka [His-
torico-etymological Dictionary of Contemporary Russian], vols. 1-2.
Moscow: Russkij jazyk.
Chekhov, Anton
1984 The Russian Master. Oxford: Oxford University Press.
Dostoevsky, Feodor M.
1983 The Brothers Karamazov Translated by David McDuff. Harmonds-
worth: Penguin.
Fasmer, M.
1996 Etimologioeskij slovar' russkogo jazyka [Etymological Dictionary of
the Russian Language], vols. 1-4. St. Petersburg: Azbuka.
lordanskaja, Lidija
1970 Popytka leksikografieskogo tolkovanija gruppy russkix slov so
znaceniem Cuvstva [An attempt at a lexicographic analysis of a group
of Russian emotion terms]. MaSinnyjperevod iprikladnaja lingvistika
13: 3-26.
1972 Leksikografioeskoe opisanie russkix vyrazenij, oboznaajuScix fizie-
skie simptomy cuvstv [A lexicographic description of Russian ex-
pressions for physical symptoms of emotion terms]. Masinnyjperevod
iprikladnaja lingvistika 16: 3-30.
1984 Slovarnye stat'i bojat'sja, vostorg, vosxiSoat', gnev, strax. Tolkovo-
kombinatornyj slovar' russkogo jazyka [Dictionary entries for Russian
words comparable to fear, delight and anger]. Wiener Slavistischer Al-
manach 14.
Human emotions viewed through the Russian language 333
P: I don't think so. I am sure that people in those other cultures also ex-
perience fear and anger, even if they don't have words for them.
L: Maybe they do experience fear and anger, but their categorisation
of emotions is different from that reflected in the English lexicon.
Why should the English taxonomy of emotions be a better guide to
emotional universale than that embodied in some other language?
P: Let's not exaggerate the importance of language.
But, by neglecting the semantic differences between words from differ-
ent languages, scholars who take this position end up doing precisely what
they wished to avoid, that is, "deifying" some words from their own native
language and reifying culture-specific concepts which are encapsulated in
them. Thus, unwittingly, they illustrate once again how powerful the grip
of our native language on our thinking habits can be.
As William James noted a hundred years ago, we know from introspec-
tion that, on the one hand, we are capable of a great variety of feelings, and
on the other, that these different feelings are not clearly separated from one
another and can not be counted. Furthermore, as James also noted, every
language imposes its own interpretive grid upon this largely nebulous
world of feelings.
... if one should seek to name each particular one of [the emotions] of which the
human heart is the seat, it is plain that the limit to their number would lie in the in-
trospective vocabulary of the seeker, each race of men having found names for some
shade of feeling which other races have left undiscriminated. If we should seek to
break the emotions, thus enumerated, into groups, according to their affinities, it is
again plain that all sorts of groupings would be possible, according as we chose this
character or that as a basis, and that all groupings would be equally real and true.
(James 1890:485)
Thus, the way people interpret their own emotions depends, to some ex-
tent at least, on the lexical grid provided by their native language. Trying to
develop these ideas in an earlier study I wrote:
... the set of emotion terms available in any given language is unique and reflects a
culture's unique perspective on people's ways of feeling. It also reflects the links be-
tween feelings, cognition, moral norms, and social interaction.
To draw on my personal experience (as a bilingual and bicultural person), I would
say that I tend to perceive my daily emotions in terms of lexical categories provided
by Polish (my native language), such as those illustrated in the following sentences:
A culturally salient Polish emotion: Przykro 339
This paper is devoted to the first and perhaps the most salient of all these
culture-specific emotion concepts.
To give the reader some idea of how salient the concept of przykro is in
Polish, we can compare the frequency of przykro with that ofsmutno 'sad'
given in a Polish frequency dictionary (Kurcz et al. 1990) based on a cor-
pus of 500,000 words. POT przykro the figure is 15, and for smutno, 8. If we
include the figures for the nouns przykrofo and smutek (8 and 11), and for
the adjectives przykry and smutny (23 and 27), the joint frequencies of the
two groups will be the same (46). Thus, although "sadness" is supposedly a
"basic human emotion", innate and universal, the concept of przykro ap-
pears to be equally, if not more salient in Polish culture. Figures of this
kind, based on a relatively small corpus of half a million words, should not
be pushed too far, nonetheless they do give us some idea of the salience of
the emotion concept przykro in Polish culture.
2. Przykro
The adverb przykro is one of the most important Polish emotion words. It is
normally used in a dative construction, as in the following sentence quoted
in the Dictionary of the Polish Language (Slownik J^zyka Polskiego
1958-69):
340 Anna Wierzbicka
But przykro differs also from sorry, insofar as it refers necessarily to in-
terpersonal relations (like hurts). For example, I can be sorry when I hear
that someone else lost their keys. Przykro, however, could not be used in
such a context:
For example:
(9) Byio mi bardzo przykro i glupio z powodu tego, co sie_ stalo. (Pie-
tak)
had a painful and embarrassed feeling because of what had hap-
pened.'
Here, the sentence (adduced by Stownik Jgzyka Polskiego without any
context) does not refer to any preceding human action. Yet the word
przykro (as well as glupio, to be discussed below) makes it quite clear that
the event referred to must have involved some human action: one could feel
A culturally salient Polish emotion: Przykro 343
the meaning of the word przykro guarantees that the sentence refers to
some interpersonal act: either X felt something bad (przykro) because
someone else (Y) had done something that suggested coldness, indiffer-
ence, or lack of regard for X, or X felt something bad (przykro) because X
thought that someone else (Y) may have perceived (construed) some action
of X as suggesting coldness, indifference, or lack of regard towards that
other person (Y).
The painful effect of someone's perceived "coldness" or indifference is
often experienced as extremely strong, so much so that the word przykro
often occurs in Polish in combination with the intensifier strasznie 'ter-
ribly':
(16) Strasznie mi przykro.
terribly to.me painfully
am terribly hurt/sorry.'
(17) Byio mu strasznie przykro.
it.was to.him terribly painfully
'He was terribly hurt/sorry.'
This frequent co-occurrence with the intensifier strasznie highlights the
depth of the hurt: even when there is no implication that anyone has done
"anything bad" to another person, or that the two persons in question have a
close emotional relationship, the "slight" of "coldness" or "indifference"
can be perceived as very painful.
4.1. Offended
What matters here is not the action as such but its apparent meaning: I
feel "offended" by your action if this action suggests to me that you "don't
think good things about me", and that you want me to know it. Schemati-
cally, this can be represented as follows:
4.2. Hurt
But the range of use of the two words (hurt and offended) is clearly not
the same, and clearly hurt has more to do with feelings than offended; it is
not an accident that while one can either hurt or offend a person, one can
only "hurt someone's feelings" (not "offend feelings"). A few examples
(from Bernard Shaw's plays):
(24) a. / can' t fall in love and I can' t hurt a woman 's feel ings by tell-
ing her so.
b. Mrs. Warren winces, deeply hurt by the implied indifference
towards her affectionate intention.
c. You can marry her, if you hurt her feelings by refusing...
d. How deep it would hurt her to think it was because I didn't
care for her.
e. Oh, do stay, Rejjy, I shall really be hurt if you desert us.
What all these examples suggest is that one person "feels something
bad" as a result of something that another person does, and that the "hurt-
ful" action is unexpected and apparently inconsistent with the affected per-
son's expectations. Schematically, this can be represented as follows:
350 Anna Wierzbicka
4.3. Sorry
(30) Sorry
(a) I feel something bad (now)
(b) because I think:
(c) "I did something
(d) it was bad for you"
Of the three uses of sorry considered here and linked with three different
grammatical frames (X was sorry for Y, X was sorry that Z, and Sorry) it is
the second use which comes closest to the Polish przykro and could some-
times be used as its approximate translation equivalent. For example:
The phrase feel bad looks so simple and straightforward that it is seldom
recognised for what it really is: a language-specific English idiom, with a
meaning different, and not predictable, from the meanings of its two parts.
For example, a question like "How are you feeling today?" is more likely to
be answered with "I feel good", or "I feel wonderful", or "I feel awful",
than with "I feel bad". In contrast to the other three phrases, "I feel bad"
(often, "I feel bad about it") suggests a cognitive scenario which is irrel-
evant to a question about a person's well-being.
Broadly speaking, "I feel bad" can be compared to "I feel guilty", al-
though the two are by no means identical in meaning. First of all, "I feel
bad" refers to not just any "bad feeling (or feelings)" that I may have, but to
a bad feeling resulting from something that I have done or that I have failed
to do. Secondly, "I feel bad" refers, in an oblique way, to "other people". If
my conscience is telling me that I have done something bad which doesn't,
however, involve other people, I could readily say that "I feel guilty", but
not that "I feel bad". A person who is on a diet and who has gone on a
binge, could say, half in jest, that they "feel guilty", but probably not that
they "feel bad about it". At the same time, if something really terrible
happens as a result of my actions, I could say that "I feel guilty", even "ter-
ribly guilty", but not that "I feel bad about it". In fact, saying that "I feel
bad about it" is almost like saying that "I didn't really do anything
bad" - while at the same time admitting that there was something about my
behaviour that was not quite right, and that was bad for someone else.
"Feeling bad" seems to be particularly appropriate for situations where I
could have done something good for someone else and didn't. I don't have
to "feel guilty" (I tell myself), for I don't think that I have done "anything
bad"; at the same time I can't deny to myself that what I have done was bad
for someone else. An example of such a scenario:
354 Anna Wierzbicka
Polish teachers don't like cheating any better than the Americans. One Polish in-
structor, who is infuriated by cheating, caught a university student red-handed and
sent her out of the exam room. "I felt really bad", he said, "I was wet with sweat, she
was crying". He thinks that unclear policies on cheating and a reluctance among
teachers to act as executioners result in some degree of tolerance. (Klos Sokol 1997:
194)
5. Conclusion
The English words and expressions discussed are probably closer to
przykro than any others. It can be safely concluded, therefore, that English
doesn't have a word or expression corresponding exactly to this Polish con-
cept.
This is not to say that speakers of English never experience the emotion
associated in Polish with the word przykro; only that they do not think ha-
bitually about their experiences in these terms. The situation conceptual-
ised by a Polish speaker in terms of przykro would be likely to be concep-
tualised in English in a different way - for example, as a situation linked
with the experience of being "hurt", "offended", "sorry", or "feeling bad".
Each language provides its speakers with a different set of interpretive
categories and encourages them to conceptualise their emotional experi-
ence in terms of a different set of scenarios. It is likely that the set of scen-
arios linked with frequently used language-specific categories (lexical and
lexico-grammatical), points to a culture's central concerns and values. The
Polish word przykro and the grammatical construction associated with it is
a good case in point.
In an earlier study of the norms relating to emotion in Polish culture
(Wierzbicka 1994) I argued that Polish culture places a great deal of em-
phasis on the expression of interpersonal "good feelings", noting that "the
Polish cultural emphasis on warmth or affection must be distinguished
from the Anglo emphasis on consideration and tact", and that "Polish cul-
ture encourages the showing of good feelings toward the addressee rather
than efforts not to hurt or offend the other person" (1994: 163; for fuller
and more recent discussion, cf. Wierzbicka 1999). I also noted that in
Polish culture, behaviour that displays emotion is seen as the norm, not as a
departure from the norm, and that "perhaps no feeling is more valued and
more expected in Polish discourse than a good feeling towards the ad-
dressee". In support of these observations, I wrote, inter alia:
The Polish cultural value of expressing good feelings toward the addressee is... re-
flected in its rich store of terms of endearment, especially metaphorical endearments
(usually used in the vocative case), such as ptaszkul (bird-DlM), sioneczko\ (sun-
DIM), kotku\ (kitten-DiM) robaczku\ (worm-DiM), zabko\ (frog-DiM), skarbiel (treas-
ure), or zlotko! (gold-DiM). Words of this kind may be used not only as parenthetical
terms of address (like honey or sugar in American English), but also as exclamative
utterances in their own right (like darling\ or sweetheart*, in English). Frequently,
356 Anna Wierzbicka
terms of this kind are used with the possessive pronoun moj ('my'), which under-
scores the personal character of the relationship: moje ztoto (my gold), moj skar-
bie (my treasure), or moje sioneczko (my dear-little-sun; cf. English My darlingl
vs. *My honeyl). Metaphorical endearments are particularly common in address-
ing children, but are not restricted to adult-child interaction. (Wierzbicka 1994:
161)
Notes
1. This and subsequent examples from leading Polish authors are as cited in Stow-
nik Jezyka Polskiego (1958-69).
2. The format used in this explication and in the other explications of emotion con-
cepts included in this paper is a simplified one. A full explication of an emotion
concept should, I believe, be based on a prototype, and should attempt to de-
scribe not only the cause but also the quality of the emotion. This can be done'as
follows (cf. Wierzbicka 1973, 1992, 1999):
I felt joy
(a) I felt something good
(b) because I thought something
(c) sometimes a person feels something good
(d) because this person thinks:
(e) something very good is happening now
(f) I want this to happen now
(h) I felt something like this
(i) because I thought something like this
In the case of przykro a full explication of this kind would read as follows:
Byto mi przykro
(a) I felt something bad
(b) because I thought something
(c) sometimes a person feels something bad
(d) because this person thinks:
(e) someone did something
(f) because of this someone else could think:
(g) "this person doesn't feel anything good toward me"
(h) I felt something like this
(i) because I thought something like this
A culturally salient Polish emotion: Przykro 357
References
Collins English Dictionary and Thesaurus
1993 John M. Sinclair, general consultant. London: Harper Collins.
James, William
1890 The Principles of Psychology. London: Macmillan.
Klos Sokol, Laura
1997 Shortcuts to Poland. Warsaw: Wydawnictwo IPS.
Kurcz, Ida, Andrzej Lewicki, Jadwiga Sambor, Krysztof Szafran and Jerzy Woronczak
1990 Stownik Frekwencyjny Polszczyzny Wspotczesnej [A Frequency Dic-
tionary of Contemporary Polish] Krakow: Polska Akademia Nauk.
Longman Dictionary of the English Language
1984 London: Longman.
The Oxford English Dictionary
1933 12 vols. Oxford: Clarendon Press.
Sapir, Edward
1949 Selected Writings of Edward Sapir in Language, Culture and Person-
ality. David Mandelbaum (ed.). Berkeley: University of California
Press.
Stownik Jgzyka Polskiego [Dictionary of the Polish Language]
1958-69 Witold Doroszewski (ed.). 11 vols. Warsaw: Panstwowe Wydawnict-
wo Naukowe.
Smith, Charles John. (1970 [1903]). Synonyms Discriminated: A Dictionary of
Synonymous Words in the English Language. H. Percy Smith (ed.).
Detroit: Gale Research Co.
Wierzbicka, Anna
1973 The semantic structure of words for emotions. In: Roman Jakobson,
C. H. van Schooneveld and Dean S. Worth (eds.), Slavic Poetics: Es-
says in Honor ofKiril Taranovsky, 499-505. The Hague: Mouton.
1992 Defining emotion concepts. Cognitive Science 16: 539-581.
1994 Emotion, language, and "cultural scripts". In: Shinobu Kitayama and
Hazel Rose Markus (eds.), Emotion and Culture: Empirical Studies of
Mutual Influence, 130-198. Washington: American Psychological As-
sociation.
1997 Understanding Cultures through their Key Words: English, Russian,
Polish, German, Japanese. New York: Oxford University Press.
1999 Emotions Across Languages and Cultures: Diversity and Universals.
Cambridge: Cambridge University Press.
An inquiry into "sadness" in Chinese1
Zhengdao Ye
What we have to translate, in effect, is not a word but a whole culture, to see how an
emotion and the name for that emotion fits into the systematic worldview, lan-
guage, and the way of life of the society. (Robert C. Solomon 1995: 256)
No inquiry into the emotions (affections) can afford to overlook poetry and narrative,
since the writers of such works are more perceptive than anyone else on these
matters. (Zhongshu Qian 1998: 84)
carries an assumption that the world above the "well" of the Anglo lan-
guage is the whole world outside the well. They forget that the high wall of
the well is built up by the bricks of Anglo values and judgments. It is not
that all researchers of emotions are unaware of the problem of absolutising
their own values. Lutz (cf. 1985, 1987, 1988, 1995), for example, has re-
peatedly emphasised semantic differences between cultures as a methodo-
logical problem in her study of the indigenous psychology on Ifaluk Atoll
(Micronesia). She stated that "in the present study, the question of the
meaning and translation of those terms has been taken as the primary and
logically prior question" (Lutz, 1985: 39). She emphasised this point a dec-
ade later by warning herself to guard against her own preconceptions: "My
sense of what people were saying - requiring an effort to disentangle my
own native emotional understandings from theirs" (Lutz, 1995: 253). Rus-
sell made the same point from a more global perspective: "I take it for
granted that psychologists are interested in the emotions of all people, not
just those who speak English. We must therefore understand the vocabu-
laries of all people" (Russell 1991: 428).
The same problem comes up in the study of Chinese emotions (as well as
in the study of Chinese indigenous concepts in general, cf. Bond 1996;
Yang 1988). Russell and Yik (1996), in emphasising that analysis of the
Chinese lexicon can help reveal the concepts available to Chinese speakers
through which they categorise their emotional experiences, also raised the
fundamental question of what. In their comprehensive and revealing re-
view of the research on Chinese emotion, Russell and Yik tabulated
95 Chinese emotion-describing words that were translated to or from Eng-
lish from 18 studies (Table 12.2: Translation of Emotion Terms, in Russell
and Yik 1996: 177-179). They were listed with their "English equivalents".
This list is illuminating as well as alarming. Russell and Yik pointed out
that among the aforementioned 95 words, 66 were found in only one study
(where the translation has not been replicated). Among the remaining
29 cases, which allowed researchers to examine the replicability of trans-
lation across studies, the one-to-one correspondence between Chinese and
English was replicated in just five translations. The researchers supposedly
investigated the same concepts, but in fact they did not.
Some researchers do recognise the disparity and inconsistency between
concepts across cultures (cf. e.g. Schwartz, White, and Lutz 1992;
Shweder and LeVine 1984). The question is how to make explicit the
meaning of emotion concepts without imposing researchers' own cultural
An inquiry into "sadness" in Chinese 361
The three emotion-describing terms bei, ai, and chou, are often glossed in-
discriminately in various studies and in dictionaries as either sadness, sor-
row or grief. Bei and ai appear in different versions of "basic emotions" in
classical Chinese texts. Russell and Yik (1996) labeled bei and ai as sorrow
in discussing five versions of the "basic emotions" in Chinese. When 50
educated Hong Kong Chinese adults were asked to list the seven basic
emotions, as in the proverb ren you qiqing liuyu (lit. 'people have seven
emotions and six desires'), bei, ai, and chou were mentioned, with an oc-
currence of 66 percent for ai, 14 percent for bei, and eight percent for
chou.5 The English translations were as follows: ai - 'sorrow/grief,
bei 'sorrow/sadness', and chou 'sadness'. This seems to suggest that ai
is perceived as one of the more basic emotions; and that these three Chi-
nese emotion terms share the same conceptualisation with each other, and
also with the English sadness. Is this the case? The discussion will start
with bei.
An inquiry into "sadness" in Chinese 363
Alone at night, sitting in an empty hall, the poet Wang Wei (701-761 AD)
reflects on his state of isolation in the darkness of night, and on his ageing
marked by the universal, unavoidable life change - black hair turning
white. As the temporal emptiness progresses towards the darkest part of the
night (ergeng), the poet can sense the decline of his life. He is conscious
that what awaits him is death. Ageing, the cycle of life, are inevitable and
irreversible: his white hair will never again turn black and no elixir will
ever bring back his youth. He has to submit to the "law of nature". This
"naturalness" of the cycle of life is heard through the thud of autumn fruit
falling from the tree in the silence of the night mountain in the rain, and
through the wailing of the insects, whose imminent death approaches with
every passage of autumn.
364 Zhengdao Ye
When black hair starts to whiten, this indicates that a person is entering
shengming zhi qiu ('the autumn of life'), in other words something bad is
happening (to somebody). White hair is a sign of the decline of life, with
the worst yet to come - death. What makes one bei is not just a bad hap-
pening at that moment, but an uncontrollable force pulling towards a sub-
sequent "tragic" happening determined by nature. Furthermore, bei lies in
the consciousness of the experiencer, who is aware that the expected course
of events is irreversible. One can do nothing to "beat the odds" and reverse
the happening from worse to worst. Thus, the meaning of bei can be expli-
cated along the following lines:8
bei
(a) X felt something because X thought something
(b) sometimes a person thinks:
(c) "something bad happened now
(d) I know that after this good things will not happen any more
(e) I do not want things like this to happen
(f) I cannot do anything
(g) because I know no one can do anything when things like this
happen"
(h) when this person thinks like this this person feels something very
bad
(i) X felt something like this
(j) because X thought like this
Components (a), (b), (i) and (j) set up a prototypical cognitive scenario
referring to a standard situation in which X feels like a person does when
they have certain specifiable thoughts. Facing the same antecedents, emo-
tional stimuli or situational antecedents, different people may feel differ-
ently because of the different thoughts they have, but it is the prototypical
thoughts that single out a particular situation (cf. Fehr and Russell 1984;
Shaver et al. 1987; Wierzbicka 1972).
Component (c) ("something bad happened now") suggests that the
thoughts of bei are a response to a recent event, whose effects are still evi-
dent. The event may happen to somebody, or it may just happen as a natural
process like autumn. In any case, this event triggers a further pessimistic
view of the situation. Component (d) points to the experiencer's conscious-
An inquiry into "sadness" in Chinese 365
ness of the on-going deteriorating situation: the very source of bei. The no-
tion that the future happening lies towards the bad end of the scale ("I know
after this good things will not happen any more") is reflected in such com-
monly used compounds as (2 a) and (b):
(2) a. bei ju
bei play/drama
'tragedy'
b. bei guan
bei view
'pessimistic'
"Something good will not happen any more" proposed in component (d)
indicates thoughts about the absence of a good outcome. Compared with a
component like "something bad will happen", which refers to the presence
of a bad result, "something good will not happen" brings out a pessimistic,
rather than a "fearful" view. This pessimism is one of the most important
aspects of the meaning oibei. The following examples provide further evi-
dence for this key conceptual component. In example (3), a reading of the
thoughts that go through Jia Zheng's mind as he listens to all the riddles
from the young members of the family sheds light on the pessimistic char-
acteristics of bei:
(3) Jia Zheng xinnei chensi dao: 'niangniang suo zuo baozu, ci nai yi
xiang er san zhi wu. Ying-chun suo zuo suanpan, shi dadong luan
ru ma. Tan-chun suo zuo fengzheng, nai piaopiao fudang zhi wu.
Xi-chun suo zuo haideng, yifa qingjing gudu. Jin nai shangyuan
jiajie, ruhe jie zuo ci buxiang zhi wu wei xi ye?' Xinnei yu si yu
meng, yin zai Jia mu zhi qian, bugan xing yu se, zhide mianqiang
kan xiaqu... Jia Zheng kanwan, xinnei zi cun dao: 'ci wu dao hai
youxian. Zhishi xiaoxiao zhiren zuo ci ciju, gengjue buxiang, jie fei
yongyuan fushou zhi bei'. Xiangdao cichu, da you bei qi zhi
zhuang. (HLM: v. 1:304)
Her Royal highness wrote about a firecracker which disintegrates
after a single explosion. Ying-chun's subject, the abacus, is in con-
stant commotion; Tan-chun's kite is something which drifts away
with the wind; Xi-chun's temple lamp is even more lonely and ne-
glected. What ill-omened subjects for all of them to choose so soon
after the New Year!' The more he reflected, the deeper his dismay.
366 Zhengdao Ye
(4) (Bao-yu yangwang xingzi bushe. You xiangqi Xing Xiu-yan yi zele
fuxu yishi.) Sui shuo shi nann da shi, bu ke bu xiang, dan weimian
you shaole yige hao emu. Buguo liangnian, bian ye yao 'lu ye
cheng yin zi man zhi' le. Zai guoji ri, zhe xing shu zi luo zhi kong,
zaiji nian, Xiu-yan weimian wufa ru yin, hong yan si gao le, yinci
bu mian shangxin, zhi guan dui xing liulei tanxi. Zheng beitan
shi... (HLM:v. 1:800)
Though marriage was something everyone must go through with,
this would mean one good girl the less; in just a couple of years she
would be burdened with children as this tree was with fruit; and
just as the apricots would soon be gone, leaving the branches bare,
in a few more years Xiu-yan's hair would be turning silver and she
would lose her beauty too. He could not help shedding tears as he
gazed at the tree. But as he was sighing... (DRM: v. 2:295)9
It is interesting to note that the translator was unable to render the mean-
ing of bei. Nonetheless, tan ('to sigh') conveys, to a certain extent, the
powerlessness and resignation on the part of the experiencer, which is an
integral part of the meaning of bei. Although the experiencer rejects the
outcome (component (e) "I don't want this bad thing to be happening"),
they cannot prevent it from happening. They merely subscribe to the ac-
ceptance of, and the subsequent resignation before the irreversible out-
come of the event. Component (f) ("I cannot do anything") shows one's in-
ability to reverse the course of the event. This is because of the fatalistic
realisation that the event happens independently of one's will. Compo-
nent (g), "because I know that no one can do anything when things like this
An inquiry into "sadness" in Chinese 367
happen" reflects this resignation. This powerless feeling bears some resem-
blance to sorrow, where "I cannot do anything" is part of the meaning.
The powerlessness of the individual before the "law of nature" causes a
poignant bitterness, pain and sombreness, almost to the point of "chill and
cruelty" as reflected in the component "this person feels something very
bad", supported by the following compounds:
(9) (Xiang-yun) you xiangdao ziji mingku, gang peile yige caimao
shuangquan de nanren, xingqing you hao pianpiande dele yuannie
An inquiry into "sadness" in Chinese 369
(10) Ta he ta de ganging youshi name hao. Ran' er xianzai fuqin que gei
ta xuanqu le ling yige ta bu renshi de guniang, bingqie haijueding
jiuzai niannei jiehun, tade shengxue de xiwang chengle paoying,
er ta shuo yao qu de you bushi suo zhongyi de nage 'ta'. Duiyu ta,
zhe shizai shi yige dade daji. Tade qiancheng duansongle. Tade
meimiao de huanmeng pomie le.... Ta bufankang, ye xiangbudao
fankang, ta zhi bei dao ziji de mingy un. Tan ta renshou le. Ta shun-
congle fuqin de yizhi, meiyou yidian yuanyan, keshi zaixinli ta que
weizhe ziji tongku, weizhe ta shouhou de nage shaon tongku.
(J: 28)
He was deeply in love with Mei, but now his father had chosen an-
other, a girl he had never seen, and said that he must marry within
the year. What's more, his hopes of continuing his studies had
burst like a bubble. It was a terrible shock to Juexin. His future was
finished, his beautiful dreams shattered.... He did not fight back,
he never thought of resisting. He only bemoaned [bei and mourn]
his fate. But he accepted it. (SWBJ: 28)
When Jue-xin finds out about the arranged marriage ("something bad is
happening to me"), he knows that his future is totally destroyed ("I know
after now good things will not happen"). Juexin does not want this arrange-
ment ("I don't want things like this to happen"), for he is in love with Mei,
yet he knows, being the eldest son in such a large and feudal family, that he
cannot do anything but fulfil his filial responsibility ("I know I cannot do
anything", "because when things like this happen, no one can do any-
thing"). Marriage, in such a feudal family, is not dependent on his own will.
The examples cited above, particularly the opening poem, indicate that
bei involves a process of "transfer" and momentary "stirring" (therefore
370 Zhengdao Ye
only "feel very bad" in (h) without reference to a period of time). The ex-
periencer, stirred by some external stimulus, reacts to it. In this regard, bei
is very different from the long-term feeling of sorrow. The following sen-
tence, which serves as a title for a chapter in HLM, typically reflects a natu-
ral change that is intrinsically conducive to bei, or the kind of external
stimulus that the emotion of bei is typically associated with:
(11) Gan qiu shen fu qin bei wang shi.
stir autumn deep touch zither bei past matters
'Moved by autumn, a zither musician mourns the past.'
The music of the zither is desolate and heavy, like the thud of autumn
fruits falling on the ground. It represents the "explosive" tune of bei, which
seems to be symbolised by bilabial plosives /p/. It is interesting to point out
that many scholars have mentioned that bei is a key emotional tone in tradi-
tional Chinese music (cf. Qian 1998).
From ancient times until now, the emotion of bei has typically been
"stirred" by the season of qiu ('autumn'). Bei qiu (lit. 'bei autumn'), which
is a lexicalised item, has been one of the most important and enduring
themes in Chinese literature (e.g. Xin 1993). Song Yu, from the Kingdom
of Chu (around 2nd century BC), believed to have been the first person to re-
flect this folk view of human feelings in Chinese literature, started the lit-
erary tradition of beiqiu:
the start of Yin energy ('negative energy'), which initiates the decline of
the process of Nature until it meets Yang energy ('positive energy') again
to generate new life. Bleak and desolate autumn scenes are transplanted
into a kind of reflexive awareness in the individual, who naturally becomes
aware of unavoidable ageing and death, under the Chinese cosmic belief of
tian ren he yi ('the unity of Heaven and Man') (cf. Fang 1957; Feng 1953).
The processes of mutual generation and mutual restraint in nature are in-
dependent of human will.
To some extent, bei crystallises the process from stirring to reflection,
which is construed on the basis of the particular Chinese folk view of the
universe. The explication of the cognitive content of bei seems to suggest a
plausible explanation as to why bei, not ai, is posited as one of the basic
emotions in the Daoist doctrines, such as the Huangdi Nei Jing ('The Yel-
low Emperor's Canon of Medicine', 5th-3rd century BC) and the Dao De
Jing, which are closely associated with Yin-Yang cosmology.
Reflected in beiqiu is a unique aspect of Chinese emotions, that is, the
fact that seasons play an important role in Chinese emotional experiences.
This is well documented by Liu Xie, a sixth century literary critic, in one of
the most important Chinese treatises on literature Wen Xin Diao Long (lit.
'literature heart carve dragon'):
Springs and autumns follow on in succession, with the brooding gloom of dark Yin
and the easeful brightness of Yang. And as the bright countenances of physical things
are impelled in their cycles, so the affective capability of mind (hsin) too is shaken....
And when autumn's skies are high and the animating air takes on a chill clarity, our
thoughts, sunken in the darkness of Yin, touch on far things. The year has its physical
things, and these things have their countenance; by these things our emotions are
shifted, and from emotions language comes. The fall of a solitary leaf finds its place
in our understanding [and we know that autumn is coming], (in Owen 1985: 22; em-
phasis mine).
For Chinese people, to think about autumn is to think about its correlativ-
es in the cycle of human life, in the dynastic cycle, and in all domains of
life (Owen 1985: 22; Xin 1993; Zhang 1989). In the Chinese cultural con-
sciousness, time, major (seasonal) changes in nature, human life, and the
movement of history are seen as unified, operating in a cyclic pattern.
372 Zhengdao Ye
Both bei and ai are considered as being among the cardinal emotions in
traditional Chinese texts. However they are very different in their proto-
typical cognitive structures. Ai is closely associated with someone's death,
as reflected in a cluster of death-related phrases (13). This can be regarded
as the first major difference between bei and ai.
(13) a. ai yue
ai music
'funeral music'
b. ai ge
ai song
'an elegy, a dirge'
c. ai dao
ai mourn
'to mourn for the deceased'
d. mo ai
silent ai
'stand in silent tribute (to someone's death)'
e. ai zi
ai son
'a male bereaved of this father'
f. ai si
ai thought
'sad memories (of the deceased)'
It is interesting to note that in HLM, one of the most popular Chinese
classics (which I have examined in detail in Ye, 2000 a), ai occurs in death-
related situations only. For example, when her mother dies, the character
Dai-yu is said to be shou sang jin ai (14a), and ai tong guo shang (14b):
The close link between ai and death seems to suggest grief, which is
prototypically connected with death. However, phrases like ai si (lit. 'ai
thoughts', 'sad memories of the deceased') and ai ge (lit. 'ai song', 'an
elegy') strongly suggest that thoughts of ai are focused not on the loss
event, but on the other person. Thus, ai is very different from grief. Long,
deep and sympathetic thoughts are key elements in the meaning of ai.
Although ai is prototypically linked with death, it can also refer to other
situations where the person is suffering as a result of some disastrous
events. For example, in his famous Ai Jiangnan Fu (On Grieving Jiang-
nan'), Yu Xin (513-581 AD), exiled after the demise of the Liang dynasty,
expresses his ai over the fall of the Liang dynasty not in terms of personal
'loss', but of the suffering of the nation (ZLWZX: v. 2 (1): 224). In exile,
and thinking of his people who are suffering, Qu Yuan (3rd century) utters
the following lines in his famous poem "Li Sao":
In (17), what the experiencer feels ai about is the misery of the soldiers
who were sent on expeditionary duties and who were treated very badly.
On the one hand, "death" seems to be regarded as the prototypical situ-
ation when ai is aroused; on the other hand, it can be extended to situations
where an unfortunate event happens to people or where a nation (in this
sense, personified) is suffering. How is this going to be reflected in the
meaning of ail Using the prototype approach and relying on the semantic
prime LIKE, I would explicate the meaning of ai as follows:
ai
(a) X felt something because X thought something
(b) sometimes a person thinks about another person like this:
(c) "this person died
(d) this is very bad for this person
(e) I don't want things like this to happen to this person
(f) I want to do something good for this person because of this
(g) I want to think about this person for a long time"
(h) when this person thinks like this this person feels something very
bad for a long time
(i) X felt something like this
(j) because X thought like this
(k) when something very bad happens to another person
(1) a person can feel something like this
(m) because a person can think something like this
cause of this, ai can not only be used for people one knows, but also be ex-
tended to one's compatriots.
With regard to component (f) ("I want to do something good for this per-
son"), bei and ai diverge fundamentally - the fatalistic view of bei
contrasts with altruistic thoughts in ai. This marks the second main differ-
ence between the two. Ai encompasses 'sympathy', 'compassion' or even
'love', revealing a wish for bad things not to happen to the other person,
and a desire to do something good for the unfortunate. In this respect, ai is
much more active than grief in that the experiencer of grief does not go
beyond dwelling on the loss ("I want to think about this; I can't think about
other things"), whereas the experiencer in ai not only actively remembers
the other party but wants to do something beneficial for them. Compo-
nent (g) - "I want to think about this person for a long time" - shows this
altruistic aspect of ai. The volition in (g) makes ai highly compatible with
lian ('loving dearly/pity'), as in ai Han:
(18) ai lian
ai pity/commiserate
'to pity; to commiserate; to have sympathy or compassion for'
(ZSHC: 151)
(? bei lian *chou lian)
The difference between ai lian and bei lian (if bei lian is used at all) is
that in the former the emotion of lian arises from the realisation that dis-
aster befalls the other, while, in the latter, it arises from the realisation of
the powerlessness of those affected and the unavoidability of the event.
When a person dies, perhaps the good wish of doing good things for this
person is best carried out by remembrance - want to think about this per-
son for a long time' (component (h)). Words such as ai si (lit. 'ai thoughts',
'sad memories of the deceased') clearly point in this direction. Ai si is a
long and lingering thought that always occupies people's minds.10 People
may commemorate a person by various means such as songs and elegies.
Component (h) does not have any suggestion of resignation, which is pres-
ent in bei. Its 'sympathy' is realised through active thoughts about the other
person.
The ethical aspect of ai, which is totally altruistic, seems to shed some
light on the positing of ai (not bei) as one of the basic emotions in Confu-
cian texts, such as Li Ji (Book of Rites) and Zuo Zhuan (A Commentary of
Zuo). Ai is valued by Confucius and his disciples. This is consistent with
376 Zhengdao Ye
two of the core values that Confucius advocated: ren ('benevolence') and
ai ('love', a different character from the ai in question). The following
examples show when ai was mentioned in Confucius' Lun Yu (The Ana-
lects).
(20) a. ai ji
ai sacrifice
'mourning rites' (ABC: 2)
b. ju ai
lift/uphold ai
'wail in mourning; go into mourning' (ABC: 322)
An overt display of ai is wailing. This is probably why ai often collocates
with ku ('cry') or da ku ('big cry'), and why ai seems to be often linked
with sound and voices.
The third difference between bei and ai lies in the long-term intense feel-
ing of the experiencer of ai, and the explosive, momentarily intense stir of
bei. Bei erupts like a volcano. It is easier for the experiencer to get over it,
where ai is a lingering emotion, haunting, and penetrating, which is in line
with tunes found in funeral music. It also reminds one of the wailing of
dying birds in an echoing valley, or the residue of the sound persistently
penetrating one's ears. Ai is also cumulative to the point of ai hui gu li:
(21) ai hui gu U
ai destroy bone stand
'emaciated with grief (ABC: 2)
Generally speaking, ai differs from grief in three main aspects: "person-
orientation" vs. "event-orientation" (the loss); "altruistic" ("another per-
son", "for this person") vs. "personal" ("to me"); and lingering feeling
("feel very bad for a long time") vs. not long-term feeling.
In a sense, ai is closer in meaning to mourning, in that both are related to
death and the thoughts are focussed on the other person. However, there are
some major differences. Firstly, ai, though closely related to death situ-
ations, is not exclusively limited to them. It can be extended to other situ-
ations where the targeted person is suffering greatly. Secondly, ai is a much
more intense emotion that can cause harm to the experiencer. Thirdly, the
moral connotation of ai is missing in mourning. Ai is an emotion that is
highly valued in Chinese culture. Its status as a cardinal human emotion
and the elaboration of /-related phrases are strong manifestations of its
importance in Chinese culture.
378 Zhengdao Ye
Both bei and ai can be found as early as the Shi Jing ('The Book of the
Songs', ll lh -6 th century BC) and pre-Qin texts (before 3 century BC),
where chou was an emotion concept hardly mentioned (it did not appear in
The Analects at all). However, it is chou that has become the predominant
emotional message and theme in Chinese poetry from the Qin dynasty until
the present. The very reason behind this, I believe, is meaning-motivated.
Unlike bei and ai, where the experiencer is more or less "stirred" by or di-
rected towards the third party, chou is fundamentally personal. The concept
"TO ME" is so pervasive that almost all Chinese poets have touched upon
and elaborated this theme. Poems have long been regarded as expressing
the true inner self. It comes as no surprise that chou holds such a special
position in the hearts of poets.
Gale (1992, in Myers 1996) treats chou as a key word in Chinese culture
and glosses it as 'sorrow, loneliness, and homesickness'. The most widely
used bilingual dictionary for students studying the Chinese language,
Jingxuan Yinghan H any ing Cidian (JYHC), gives the English translation
of 'worry; be anxious' under the chou entry, yet the same dictionary in its
English-Chinese section makes no mention of chou under the worry entry.
Apparently there is no equivalent or near equivalent in English for chou.
Chou is specific to the Chinese emotional experience, just as fago is to the
people of Ifaluk Atoll (cf. Lutz 1985). So what exactly is choul What is its
conceptual structure? A reading of the last four lines of a famous Chinese
poem by the lauded "immortal poet" Li Bai, "Xuanzhou Xietiaolou Jianbie
Xiaoshu Shuxu" ("At Xie Tao's High Mansion in Xuanzhou: a Parting
Banquet for the Collator Shu-yun"), would be a good starting point for an
inquiry into the meaning of chou.
hardship. This poem was written in his later years, when a parting triggered
mixed feelings of loneliness, sadness, depression, confusion, and anguish
(choii). To Li Bai, this chou was concurrent with his existence and unquen-
chable, as again expressed in the last line of another of his poems "Jiang Jin
Jiu" ("Invitation to Wine"):
A typical scene or life experience that would evoke the emotion of chou
is usually cast in the setting of parting, separation, or wandering on foreign
soil without any immediate prospect of return. It is small wonder that Gale
glosses chou as 'sadness, loneliness, and homesickness'. A person living in
a foreign country or in exile is likely to have a kind of 'nostalgia' called
xiang chou (lit. 'hometown chou').12 The following lines from "Huang He
Lou" ("Yellow Crane Tower") and "Feng Qiao Ye Bo" ("Mooring at Night
by Maple Bridge") respectively, written by different poets when they saw
friends off or were away from their home, reflect these typical situations
when chou is experienced:
In parting from a friend, chou is diffusive like a white cloud in the vast
sky (24). In drifting on a boat far away from home during the long night,
chou is pervasive like the endless night itself. One can imagine the toss and
turn of the poet in the long, dark night. In both situations, chou is marked
by loneliness and preoccupied thoughts: "I cannot not think of my best
friend and my hometown".
The numerous writers who have touched on the emotion of chou include
an emperor, Li Yu (937-978 AD), the last monarch of the South Tang dyn-
asty. Let us first look at Li Yu's chou in some of his most famous poems, es-
pecially the last one (25), which has been set to a popular song:
Here Li Yu shows us his most private and troubled feeling: chou. Li Yu's
life went through a series of drastic changes. In an earlier period, he lived
an extremely luxurious court life filled with love and joy. Later, he was a
political prisoner after losing his kingdom, left to live a life of melancholy
and desolation (cf. ZLWZX). Beautiful things like spring flowers and the
autumn moon could only arouse his deep sorrow. The turbulent eastbound
water is an extension of on-going chou, in time and space. Unbearable nos-
talgia for his lost kingdom, rejection of his present situation, and the diffi-
culties of extricating himself from that undesirable situation left Li Yu fad-
ing away in ill health- 'My complexion, once ruddy, has become pale'.
Based on the examples and discussion, the meaning of chou can be ex-
plicated as follows:
chou
(a) X felt something because X thought something
(b) sometimes a person thinks:
(c) "something very bad is happening to me
(d) before this, I did not think this would happen
(e) I don't want things like this to happen to me
(f) because of this, I want to do something if I can
(g) I don't know what to do
(h) I cannot not think about this all the time"
(i) when this person thinks this, this person feels something bad for a
long time
(j) X felt something like this
(k) because X thought something like this
Component (c) ("something very bad is happening to me") reflects the
personal characteristic of chou and the miserable situation that the experi-
encer encounters. The happening is closely related to the experiencer's per-
sonal benefit. This difficult situation is not just momentary, but continuing
and co-existent with the experiencer of chou. It is like an obstacle. Compo-
nent (d) ("before this, I did not think this would happen") indicates that the
event is unexpected, leaving the experiencer unprepared. While rejecting
the adverse situation (component (e) - "I don't want things like this to
happen to me"), the experiencer wants to do something to change the s'tu-
ation (f) ("because of this, I want to do something if I can"), yet he or she is
caught up in confusion, not knowing what can be done (component (g) - "I
384 Zhengdao Ye
don't know what to do"). The struggle between wanting to find a solution
and not finding one is clearly exemplified in the poems cited above,
through, for example, the interplay of thoughts and actions while drinking
wine, tossing through the chilly night and lamenting in a beautiful season.
Maybe the only thing a person can do is to try to forget chou. Li Bai made
the effort, yet the feeling of chou was so haunting that it could not even be
dissolved by wine. Jie jiujiao chou (lit. 'to quench chou with wine') has
become a typical action when people experience chou. Unlike bei, which is
easy for the experiencer to get over, chou stays with the experiencer. How
can a poet forget the honour of the older days; how can one forget one's
hometown; how can an emperor forget his lost kingdom? The persistence
and unforgettableness of chou is embodied in component (h) - "I cannot
not think about this all the time". "Feel something bad for a long time" in
component (i) accounts for the prevailing preoccupied feeling, and the on-
going misery, like eastbound water spreading over time and space.
The meaning of chou is crystallised in the poems of the most outstanding
and influential Chinese poetess, Li Qingzhao (1084-1151 A.D.). Chou ap-
pears in almost every one of her poems composed in later years. Among
them, "Sheng Sheng Man" ("Every Sound Lentamente"), is a condensed
outpouring of chou. The famous opening lines epitomise the thought "I
don't know what to do" (i). The poem goes:
z he n g s hang xin,
PROG hurt heart
que shi jiu shi xiang shi.
but be old time acquaintance
Man di huang hua dui ji,
fill ground yellow flower pile accumulate
qiaocui sun,
haggard/thin/pallid fade away/damage
rujin you she kan zhai!
these days there be who bear pick
Shou zhe chuangr,
guard DUR window
du zi zen shengde hei?
one self how bear black
wutong geng jian xi yu,
wutong more with fine rain
dao huanghun,
till dusk,
dian dian di di.
dot dot drip drip
Zhe cidi,
this situation
zeng yi ge chou zi liaode!
how one CL chou character PART (say all)
I've a sense of something missing I must seek.
Everything about me looks dismal and bleak.
Nothing that gives me pleasure, I can find.
Even the weather has proved most unkind.
It is warm, but abruptly it turns cold again.
An unbroken rest - most difficult to obtain.
Three cups of thin wine would utterly fail -
To cope with the rising evening gale.
Myself, into woe, a flight of wild geese has thrown.
But with them, very familiar I have grown.
About the ground, chrysanthemums are bestrewn.
Gathering into heaps - bruised - withering soon.
With myself in utter misery and gloom,
Who cared to save them from their approaching doom?
386 Zhengdao Ye
c. chou ku le
chou cry PFV
'to cry from chou'
d. chou bai le toufa
chou white PFV headrhair
'to grow white-haired from chou'
e. chou si le
chou die PFV
'to be dying from chou'
Compared to bei and ai, chou is inextricably linked with a personal pre-
dicament that is forced on one by circumstances. It is active both in "want-
ing to do something" and in preoccupied thoughts. What sets it apart from
both bei and ai is its future-oriented thoughts. In this regard, chou is quite
far away from "sadness-like" emotions. What allows it to retain its link
with "sadness-like" emotions is that chou responds to a concurrent event. It
is not surprising, then, that the meaning of chou has been confused, its
glosses ranging from 'sadness', 'sorrow', and 'melancholy' to 'worry' and
'anxiety'.
4. Sadness - Universal?
Sadness has been proposed as one of the basic emotions by different psy-
chologists in the study of emotions (e.g. Ekman 1992; see Plutchik 1994
and Van Brakel 1993 for different lists of basic emotions). In this chapter,
the precise analysis of the cognitive structure of Chinese "sadness-like"
emotion terms, which are considered to be basic emotions in Chinese cul-
ture, provides convincing empirical evidence that what Chinese people
consider to be basic emotions are very different from those proposed in the
English language. If the postulation of basic emotions were based on the
Chinese language, it could include bei ('fatalist/tragic sadness') and ai
('ethical/altruistic grief/mourning'). Neither of these emotion terms, as
discussed in this paper, have exact English equivalents. There is no reason
to assume that bei and ai are not as basic as sadness (cf. Wierzbicka 1992;
1999).
The "basic emotions" proposed by Chinese people single out what the
Chinese people consider to be important in Chinese culture. The basic
390 Zhengdao Ye
emotions of the Chinese people are shaped by Chinese culture. The whole
concept of bei, which highlights a sense of powerlessness due to the laws
of nature and which is coloured by Chinese people's cyclic view of life
and change, may be absent in other cultures. Bei is more tragic and fatal-
ist, involving a momentary transference. Its association with nature and
seasonal changes is characteristic of Chinese people's emotional experi-
ences. Ai, though involving 'loss', is ethical in nature, arousing altruistic
compassion and sympathy. Although chou does not appear on the lists of
"basic emotions" proposed in traditional Chinese texts, its frequency con-
firms its salience in Chinese emotional experience. It is personal, intro-
spective, mundane and ongoing. English does not share the same Gestalt
as the Chinese "sadness" experience. The claim that sadness is a universal
emotion category is biased towards the English language. It is also inter-
esting to note that there is a strong connection between "sadness-like" and
"worry-like" (as well as "love-like") emotions in Chinese (cf. Shaver, Wu
and Schwartz 1992; Ye 2000 a). Intuitively, and as is often the case in clas-
sifying dictionaries, chou is grouped with you ('loving concern'), another
emotion term bordering on "sadness-like" and "worry-like" emotions,
forming a separate category from both bei and ai. To simply gloss chou as
sadness would distort the true picture of the Chinese people's categoriz-
ation of emotion domains. Solomon has succinctly pointed out this pre-
sumption:
The problem, among others, is that we take the category of emotions to be something
settled, and then wonder whether or to what extent the Chinese category corresponds
to our own. But, of course, it does not, not only because we do not have any such
settled category ourselves but because there is no reason whatever to think that the
Chinese concept - which seems to have more to do with what we might call "auth-
enticity" than with a particular realm of "the mind" - bears any interesting relation-
ship to the kind of classificatory category that 'emotions' is supposed to be in Eng-
lish. (Solomon 1995: 179)
This points to a broader issue of whether or not there are universal emo-
tions - a question that has been one of the central concerns in the study of
emotions among psychologists (cf. Ekman and Davidson 1994). Indeed,
there is no reason to think that the question of whether there are basic emo-
tions is valid. As Shweder rightly argues: ' "Are there basic emotions?" is
such a deceptively simple question that it deserves to be deconstructed'"
(Shweder 1994: 32).
An inquiry into "sadness" in Chinese 391
To totally reject the quest for the universality of human nature is not a
constructive attitude. The analyses in this paper, on the other hand, show
that the quest for a common core among human beings is not a question of
all or nothing. One can extract the common structure once the cognitive
scenario is explicated in universal human concepts. On the one hand, bei,
ai, chou and sadness have very different cognitive structures, on the other,
they share a semantic invariant core of "something bad happened/is hap-
pening", which may cross-linguistically link these Chinese emotion terms
with other emotion concepts that are constructed upon this cognitive
basis. What is basic is not the emotion terms, but the shared common "re-
sponse elements" (Scherer 1994: 25) that can be couched in lexical uni-
versals. It is the cognitive element found in emotions that is fundamental
to the universality of emotion. The configuration of these elements is
shaped by culture. They are artefacts of the Chinese culture, shaped by
uniquely Chinese experiences and views of life and the universe.
Substantives:
wo ni shei shenme renmen~ren shenti
I YOU SOMEONE SOMETHING PEOPLE-PERSON BODY
Determiners:
zhei(ge) tong ling(wai)
THIS THIS:SAME OTHER
Quantifiers:
yi Hang (ge) dou duo you de
ONE TWO ALL MANY/MUCH SOME
Evaluators:
hao huai ~ bu hao
GOOD BAD
Descriptors:
da xiao
BIG SMALL
Mental Predicates:
xiang zhidao ganjue-gandao yao kan dao ting dao
THINK KNOW FEEL WANT SEE HEAR
Speech:
shuo zi zhen
SAY WORD TRUE
Actions, Events and Movement:
zuo fasheng dong
DO HAPPEN MOVE
Existence and Possession:
(zai PLACE) you X X you
THERE IS/THERE ARE HAVE
Life and Death:
huo si
LIVE DIE
Time:
shihou xianzai yiqian yihou henjiu
TIME/WHEN NOW BEFORE AFTER A LONG TIME
duanqi you yi duan shijian
A SHORT TIME FOR SOME TIME
An inquiry into "sadness" in Chinese 393
Space:
shenme difang ~ nar shang xia yuan jin
WHERE/PLACE ABOVE UNDER FAR NEAR
pang(bian) limian zher
SIDE INSIDE HERE
Logical Concepts:
bu keneng hui yinwei ruguo
NOT MAYBE CAN BECAUSE IF
Intensifier and Augmentor:
hen zai... duo
VERY MORE
Metonymy and Taxonomy:
X you ~ bufen zhong
IS A PART OF X A KIND OF
(double subject construction)
Similarity:
xiang... zheyang ~yiyang
zheyang Verb
LIKE
The parallel explications of bei, ai, and chou in English and Chinese primi-
tives look as follows.15
bei
(a) X felt something because X thought something
(b) sometimes a person thinks:
(c) "something bad happened now
(d) I know that after this good things will not happen any more
(e) I do not want things like this to happen
(f) I know I cannot do anything
(g) because I know no one can do anything when things like this
happen"
(h) when this person thinks like this this person feels something very
bad
(i) X felt something like this
(j) because X thought like this
394 Zhengdao Ye
bei
(a) X ganjuedao le shenme, shi yinwei X xiangdao le xie shenme
(b) youshihou, yi ge ren hui zheyang xiang:
(c) "buxin de shiqing fasheng le
(d) wo zhidao zhi yihou, haoshi buhui zai you
(e) wo buyao fasheng zheyang de shiqing^
(f) wo (dui zhejian shi) zuobuliao shenme
(g) yinwei, wo zhidao zai zhezhong shi fasheng de shihou,
meiyou ren neng zuo xie shenme"
(h) zhe ge ren zheyang xiang de shihou, zhe ge ren xinli ganjue hen
buhao
(i) X you xiang zheyang de ganjue
(j) shi yinwei X you zheyang de xiangfa
ai
(a) X felt something because X thought something
(b) sometimes a person thinks about another person like this:
(c) "this person died
(d) this is very bad for this person
(e) I don't want things like this to happen to this person
(f) I want to do something good for this person because of this
(g) I want to think about this person for a long time"
(h) when this person thinks like this this person feels something very
bad for a long time
(i) X felt something like this
(j) because X thought like this
(k) when something very bad happens to another person
(1) a person can feel something like this
(m) because a person can think something like this
ai
(a) X ganjuedao le shenme, shi yinwei X xiangdao le xie shenme
(b) youshihou, yi ge ren hui zheyang xiang lingwai yi ge ren:
(c) "zhe ge ren si le
(d) dui zhe ge ren lai shuo shi henda de buxing
An inquiry into "sadness" in Chinese 395
chou
(a) X felt something, because X thought something
(b) sometimes a person thinks:
(c) "something very bad is happening to me
(d) before this, I did not think this would happen
(e) I don't want things like this to happen to me
(f) because of this, I want to do something if I can
(g) I don't know what to do
(h) I cannot not think about this all the time"
(i) when this person thinks this, this person feels something bad for a
long time
(j) X felt something like this
(k) because X thought something like this
chou
(a) X ganjuedao le shenme, shi yinwei X xiangdao le xie shenme
(b) youshihou, yi ge ren hui zheyang xiang:
(c) "henbuxin de shiqing zheng fasheng zai wo shenshang
(d) zai zhe yiqian, wo meiyou xiangdao hui fasheng zheyang de
shiqing
(e) wo buyao zheyang de shiqing fasheng zai wo shenshang
(f) suoyi, wo yao jin wo suoneng de zuo dian shenme
(g) wo bu zhidao wo neng zuo shenme
396 Zhengdao Ye
These Chinese explications via the Chinese primes are, generally speak-
ing, quite natural. This "feasibility study" of the NSM framework in the
Chinese language lends support to the universal applicability of NSM. The
idea of introducing a value-free interpretive and comparative tool into
"ethno" studies in general is not just necessary, but possible.
Blowers (1996) points out that there have been more recent attempts to
study indigenous concepts from indigenous viewpoints (undertaken by
Chinese researchers), but that there has been little enthusiasm for reporting
them to the English-speaking world. Blowers does not mention the reason
for this lack of enthusiasm. Presumably it was because of mismatches be-
tween concepts in other languages. There are indeed no exact equivalents
in English (or in other languages) for Chinese indigenous concepts. Re-
searchers do not wish an incorrect or inaccurate label to be attached to in-
digenous expressions. However, as this paper has demonstrated, the prob-
lem is not insoluble. With the Mandarin Chinese semantic primitive set and
its syntax, Chinese researchers can explicate meanings of indigenous con-
cepts accurately and neutrally and, equally importantly, they can explain
indigenous concepts from an insider's point of view. The explications in
NSM can be safely translated into semantic primitives identified in other
languages. This could facilitate the two-way information flow among re-
searchers across cultures, and contribute to a better understanding of in-
digenous concepts among researchers across different disciplines.
6. Conclusions
This paper has undertaken a textual based semantic analysis of three emo-
tions terms - bei, ai, and chou - that are often translated as equivalents of
sadness, grief or sorrow. Bei and ai are considered basic emotions in Chi-
nese. Using the semantic analytical framework of NSM, the paper has
shown that these three "sadness-like" emotions do not have exact semantic
equivalents in English. The claim that sadness is a universal emotion is
An inquiry into "sadness" in Chinese 397
called into question. The case of chou also shows that simply labelling
chou with other "sadness-like" terms hinders a true understanding of the
Chinese categorization of the emotion domain.
As this paper has illustrated, framing definitions in NSM has three main
advantages in the study of emotion. Firstly, it makes possible the expla-
nation of meanings from an insider's perspective. Secondly, the exact dif-
ferences and connections between concepts, within a culture and across cul-
tures, can be clearly identified. This point is most clearly illustrated by com-
paring, on the one hand, ai and bei, and, on the other, ai and grief. Thirdly, it
allows definitions to be translated into different languages while retaining
neutrality. Emotion terms are like the tip of an iceberg with a deep and co-
lossal block of meaning underneath. The NSM approach allows us to ex-
plore this. This paper has also shown that lexical items, conventionalized
expressions and literary examples are all valuable linguistic sources that
should not be overlooked. The lexicalisation of words such as bei qiu (lit.
'bei autumn') bears witness to the Chinese view of life and the universe, and
to the role that seasons play in shaping Chinese emotional experiences.
Once the meaning is unpacked, the "cultural logic" is readily untangled. The
elucidation of meanings for bei and ai sheds light on the "logic" of their dif-
ferent status in Daoist and Confucian versions of "basic Chinese emotions".
Linguistic theory can contribute to the study of emotions, and the study of
emotions can, and will, benefit from an interdisciplinary perspective.
Notes
1. I am deeply indebted to Professor Anna Wierzbicka for the perspective on cross-
cultural semantics that her work has opened for me, and for her encouragement and
support. I also wish to express my sincere thanks to Dr. Jean Harkins for the many
stimulating talks on various issues of emotions and for her valuable suggestions and
continuous support. My thanks also go to Professor Wenguo Pan, Dr. Xiaonong Zhu,
and Ms Changxin Cong for spending time with me discussing the Chinese emotion
universe and the Chinese translations of the explications.
2. This idiom, jingdizhiwa, originated in the writing of the Chinese Taoist philosopher
Chuang Tzu (cf. HCDC: 265; Feng, 1953).
3. Chinese, in this paper, refers to Mandarin Chinese. The romanisation used through-
out that paper is the Pinyin system. For ease of notation, tones are not marked.
4. I use zi not just to refer to the basic graphic or writing unit, but to an integral unit
constructed of characters where sound, form and meaning converge (cf. Xu 1997,
1998).
398 Zhengdao Ye
5. The Chinese frequency dictionary (XHZTB) shows that that the number of occur-
rences of bei, ai, and chou from a computation of 11,873,029 randomly sampled
characters in a corpus of Social and Natural Sciences are 835,328, and 314 respect-
ively, ranking 1525, 2040, and 2199 among 7754 characters. This means that they
are of relatively high frequency.
6. The provision of these closest (albeit idiosyncratic) English glosses is the task of a
separate project headed by Professor Paolo Santangelo of Istituto Universitario
Orientale of Naples. This project aims at compiling a glossary of expressions and
terms for emotions and mental states used in Chinese texts of the late imperial peri-
od (personal communication).
7. Most linguistic examples follow a tripartite notation, with first line romanisation,
second line morpheme to morpheme notation, followed by a literal translation.
Names in translations (except those that are conventionalised names such as
Chuang Tzu), if in the Wade-Giles system, are changed to thepinyin system for the
purpose of consistency. Where the texts are quite lengthy, only English translations
are given. Where sources are unspecified, the translation is literal. The abbrevi-
ations used for morpheme-to-morpheme notation are as follows: CL = classifier;
DUR = durative; PART = particle; PFV = perfective aspect; PROG = progressive.
8. An explication is a definition formulated through a configuration of semantic primi-
tives. It takes the form of a componential analysis, which is in line with the general
view of psychologists that emotions are "multicomponential phenomena" (Scherer
1994; cf. Frijda, Sato, and Wiers 1995). The semantic framework for explicating
emotion concepts adopted here has been developed by Wierzbicka (e.g., 1972,
1986, 1992, 1999). Further detailed explanations of the rationale behind the NSM
framework for emotion words, and how it reflects and concurs with latest develop-
ments and findings in psychological and anthropological studies of emotions, can
be found in Harkins (1990), Harkins and Wierzbicka (1997), and Ye (2000a).
9. The association between apricots and Xiu-yan is based on the homonymic relation-
ship between xingzi ('apricot') and the surname of Xing Xiu-yan.
10. Si ('thoughts/thinking/loving') plays an important role in Chinese emotions. It is
considered one of the "basic Chinese emotions" in traditional texts, and forms its
own distinct category. Liang (1996: 29) points out that, compared to Plutchik's
(1994) multidimensional model of the emotions, the Chinese classification of basic
emotions has an emphatic si category that cannot be contained within Plutchik's
model. The presence of this emotion category seems to indicate that 'thinking' has
an important status in shaping Chinese emotional experiences.
11. The English translations in the following poems are all from ZGWG. All the poems
cited are well-known among Chinese people and some have been set to popular
songs.
12. Xiang chou is a very important theme among Chinese people living abroad. Articles
on xiang chou are common in newspapers such as the overseas version of the
People's Daily.
13. High pitch is usually associated with 'happy' feelings in English (cf. Hinton, Ni-
chols, and Ohala 1994). The Chinese lines here seem to suggest a different inter-
pretation. This matter requires further investigation.
An inquiry into "sadness" in Chinese 399
14. Wierzbicka (1999: 201) explains the meaning of brows drawn together as "I am
thinking now; I want to do something now; I am not doing it now". The expression
chou mei supports this interpretation. This suggests that what is universal is the
iconic meaning of some universal facial gestures. But the interpretation of a par-
ticular facial expression may vary from culture to culture.
15. The syntactic aspect of the combinatorial behaviour of Chinese primitives cannot
be discussed here. It requires a separate study to fully investigate various issues
such as valency options (see Chappell 1994 and In press for a detailed treatment of
this issue).
16. For Mandarin speakers, yao ('want') sounds less natural than xiang when it is ne-
gated by bu. For speakers of other dialects (for example Wu speakers), yao sounds
much more natural.
References
Blowers, Geoffrey H.
1996 The prospects for a Chinese psychology. In: Bond (ed.), 1-14.
Bond, Michael Harris (ed.)
1996 The Handbook of Chinese psychology. Hong Kong: Oxford University
Press.
Chai, Ch'u - Winberg Chai (eds.)
1967 Book of Rites: an encyclopedia of ancient ceremonial usages, religious
creeds, and social institutions. New York: University Books.
Chao, Yuan Ren
1968 A Grammar of Spoken Chinese. Berkeley: University of California
Press.
Chappell, Hilary
1994 Mandarin semantic primitives. In: Goddard and Wierzbicka (eds.),
109-47.
In press Universal syntax and Mandarin Chinese grammar. In: Goddard and
Wierzbicka (eds.).
Chen, Lifu
1977 An Introduction to Chinese Culture. Taipei: The Council of Chinese
Cultural Renaissance.
Ekman, Paul
1992 An argument for basic emotions. Cognition and Emotion 6: 169-200.
Ekman, Paul and Richard J. Davidson (eds.)
1994 The Nature of Emotion: Fundamental Questions. Oxford: Oxford Uni-
versity Press.
Fang, Tung-mei
1957 The Chinese View of Life: The Philosophy of Comprehensive Har-
mony. Hong Kong: Union Press.
400 Zhengdao Ye
Izard 5, 29, 223, 226, 250, 283-4, 287, Levontina 24, 291, 310, 316-7, 327, 333
298, 312, 333 Lexicalisation 268,271,292,338
Li Bai 378, 379-80, 384, 386
Jger 146 Li, C.N. 362, 400
Jaisser 153, 165 LiLu 255,287
Jakobson 330,333 LiQingzhao 384-6
James 16, 29, 69, 113, 338, 357 LiYu 386-7,381-3
Japanese 4, 11, 18, 24, 217-53 Liang 367,398,400
Jealousy 64,72, 83-4,170, 201 Lin 262,268,280,287
Johnson 133,144-5 Lips 76, 108-9, 157-8, 162
Johnson-Laird 8, 29, 66-7, 298, 333 Literature 13-14, 24-5, 116-8, 302-4,
Joy 5, 75, 101, 177, 291-9, 356, 367-8 361,370-1,378
Juntanamalaga 153, 165, 168, 191 Liu 271-2,287
Liver 23,75,95-9,111-12
Kakehi 220,249-51 Lixacev 322,333
Karim 171, 172, 178, 194 Logan 16-17,30
Kay 164-5 Loneliness 150, 201, 323-8
Keywords 23, 167, 176 Longing 8-9, 72, 108-9, 170, 291-2,
Khan 193 299-302, 323-8, 331, 380-1, 398
Kimizuka 220,250 Losev 302, 304
Kinship 320-2 Lotman 293,301,305,333
Kirkpatrick 359,400,402 Love 61, 64, 75, 88-9, 136, 170-1, 190,
Kita 164 314,319-22,326-7,390,396
Klappenbach 122, 140, 146 Lu 362,401
KlosSokol 354,357 Lucy 163-4, 165
Kondo 4,29 Luther 136, 146
Korean 229 Lutz 4, 30, 60, 114, 278, 287, 360, 378,
Kornacki 24, 26, 29, 255, 388, 400 401-2
Koukham 164
Kvecses 19, 30, 140, 146, 211, 214 MacLaury 32, 148
Kronhaus 322,333 Madsen 26
Kupperbusch 10, 30 Maeda 182, 194
Kurtz 339,357 Malay 6, 7, 11, 13, 17-18, 23-4, 152-3,
167-95
Lakoff 19,52,67,211,214 Malayo-Polynesian 112
Langacker 19, 134, 146 Malcolm, I.G. 211,214
Lao 11,23,149-66 Malcolm, N. 8, 30
Latin 116, 136, 141,300 Malecky 320
Leavitt 69,70, 113 Mandler 10,30
Lebra 222,250 Marks 361,401
Leibniz 71,113,220,250 Martin 217,251
Lerner 116 Marx 119, 146
Lersch 144,146 Matisoff 153, 165, 168, 194
Leslau 42,67 Matsumoto 30
LeVine 360,402 Matthews 256
Subject and name index 409
186, 190; se~ (of one mind) 175; irrareltheme (depressing, Ar) 201
sejuk ~ (calm) 174; senang ~ (relaxed) -irre (inchoative, Ar) 203
168, 174, 179,182, 183-4, 193; suci ~
(pure) 174; suka ~ (please oneself) jaak-thaam (questioning, Lao) 161
172; susah- (troubled) 167-8, 174, jaan (afraid, Lao) 153, 159, 161
179, 180, 181-2; tabah ~ (determined) jaga hati (protect feelings, Ma) 172
174; tawar ~ (lost interest) 174; ter- jantung (heart, Ma) 168
pendam di ~ (buried in ~) 170; tidak jatuh hati (fall in love, Ma) 171
sampai ~ (can't stand) 172; tinggi ~ jingdiziwa (frog in the well, Ch) 359,397
(proud) 174; tulus ~ (sincere) 174; jay 75, 177,291,356
ubah ~ (change ~) 172
hear 38-41,58,65 ka- (recipient, Mb) 73
heart 167-8, 170, 173, 176, 177-9, 190; kaipa (gloat, Mb) 71
broken ~ 75, 168; pure of~ 173; sick kak by (as if, Ru) 329-30
at- 173, 184, 190 -kam (do/get, Mb) 72
het khiw sonlcoot-kan (frown, Lao) 161-2 kanamaala (sense, Mb) 72
het taa-naj (make big-eyes, Lao) 160-1 kar kopoono isoloorjgo (collective anxie-
heureux (happy, Fr) 298-9 ty, Mb) 105, 109
hiya-hiya (shivery, Ja) 18, 218-9, 222, kasih (care, Ma) 190
234,238-8,239,240-1,243-5,249-50 kecil hati (feel hurt, Ma) 168, 173-4,
hod-e tambboc'abbacd (stomach lurched, 179,188, 189-90
Am) 59-60 k' anna (jealous, Am) 64
hope 15 k^nakkana (uneasy, Am) 58
hostile 203 kete- (liver, Mb) 23, 75, 95-9, 111-12;
hurt 24,169,306-7,310-11,340,346-7, ipas ~ (enraged) 98, 99, 110; ~ guruk
349-50, 350, 355 (impulsive) 96, 109-10; ~ ibayoulibe-
leu (furious) 98, 109-10; ~ ikam kegl
iffoy (interj., Am) 44 we k (startled) 97; ~ imap (astonished)
-ig (adj. suffix, Ge) 135,144 99; ~ ipakpak (furious) 98; ~ ise (en-
ikarrut zuru- (biting lips in anger, Mb) thusiastic) 97, 110; ~ kembeita lai
108-9 (furious) 98; ~ malmal (angry) 97-8,
ikut hati (follow heart, Ma) 175 108-10; -pitpit (impetuous) 96-7,
ilkelheme (enjoy, Ar) 201 109-10
ilkelhetyakenhe (uncomfortable, Ar) 201 kheew (teeth, Lao) 157; kat (biting
imagine 177 teeth) 159
implication 8 khiat (angry, Lao) 150-1, 156
m (in, Ge) 123 khi-diat (disgusted, Lao) 151
indignation 5 khit (think, Lao) 161
ingkerte-ingkerte (jealous, Ar) 201 khiw (eyebrows, Lao) 161-2; het ~ son/
ingketye (desirable, Ar) 201 coot-kan (frown) 161-2; nik-~ (raise
interest 111 eyebrows) 161
ipas kete- (enraged, Mb) 98, 99, 110 ki o torinaosu (pull oneself together, Ja)
ira (anger, La) 116,136,141 226
irao lele- (satisfied, Mb) 94 -kilaala (recognise, Mb) 72
irrare (lonely, Ar) 201 kilyowt (kidneys, He) 74-5
Words and phrases index 417
-mbur (jealous, Mb) 72, 84 ni- (being, Mb) 23, 75-6, 99-101, 110;
-meete(die, Mb) 72 ~ ambai (healthy) 99; ~ gesges (bored)
-mel (fall, Mb) 72 100,109; ~ ikilaalalikanamaala (sense)
men (smelly, Lao) 156 99; ~ imbeeze (fit) 99; ~ ipeele (awake)
menmeen (happy, Mb) 71, 95 99; ~ isaana (exhausted) 99; ~ ise (jubi-
marrara (bitter, Am) 58 lant) 101, 108-9; ~ itekteege (shiver)
miar) (shame, Mb) 102 99; ~ iyouyou (ache) 99; ~ pa som (not
mii khuam-suk (happy, Lao) 151-2 want) 100; -puliizi (upset with) 100,
milyj (darling, Ru) 321 109; ~ sapa (resentful) 92,100,108-9;
mind 111, 176,294 ~ sanaana (upset with) 100, 101, 109
mmg(fate, Ch) 368-9 nik-khiw (raise-eyebrows, Lao) 161
miri (thirsty, Mb) 71 Him (smile, Lao) 155, 158
mirpanarinyi (angry, Ya) 6, 210 nom-paak (clench-mouthed, Lao) 155,
miss 331 159
m/Y(with, Ge) 123 nu (rage, Ch) 24, 255-65, 263, 267-70,
mood 10 274,276-8,282-4,285; bao~ (explode
-morsop (startled, Mb) 72 in rage); fa ~ (get angry) 261; kuang-
-moto, motogana (fear, Mb) 72 (mad rage) 260; ~bi (raise hands) 258;
mourn 377 ~ma (spirited horse) 260; xi~ (calm
muan (fun, Lao) 150 one's anger) 262, 268
mune (heart, Ja) 232-3 Ogora (brain, Mb) 111
murah (generous, Ma) 173-4 Qgumtur (embarrassment, Mb) 72
murni (noble, Ma) 173-4 -Ogug (cough, Mb) 72
-murur (surprised, Mb) 72 ggure- (throat, Mb) 76, 107-8; ikaml
muu (friends, Lao) 150 izuuru ~ (savour) 107, 108; ~ bulbul
mwarre awelheme (feeling good, Ar) 201 (disgusted) 108; ~ imamaaza (thirsty)
108; ~ imun (breathless from crying)
-n (accusative, Am) 37 108; ~ isaana (crave) 108,109; kumbu-
naa (face, Lao) 155-7; beqlnuk (ready- ~(calf) 108; nama- ~ (wrist) 108
to-cry face) 155,157-8; buut (scowl) -gu (offended, Mb) 72
156; khum (furious-face) 157; ~-men
(smelly-face) 156-7; pii ~ (next year) obida, obidet' (offence, offend, Ru) 291,
155 301-2, 306-7, 307, 308-11, 314-5,
nadryv (hysteria, Ru) 291,302-6 329-31
nama- ggure- (wrist, Mb) 108 obSfenie, obSoat'sja (communication,
-namut (sweet, Mb) 72 communicate, Ru) 315-6
nao(huo) (annoyed, Ch) 24, 255, 269, odo-odo (timid, Ja) 222, 234, 245-7,
271-3, 273, 274, 276, 283, 285 247, 248
naslazdenie (enjoyment, Ru) 295-9 odoroku (surprise, Ja) 220, 225, 229
ndddada (burn, Am) 51 offence, offended 6, 24, 291, 306-8,
nakka (touch, Am) 38 346-8, 348, 349, 355
nenavist' (hatred, Ru) 314 ogorenie (grief, Ru) 297
neudobno (inconvenient, Ru) 312-4, 314 omou (think, Ja) 226
naw (be, Am) 45 open-mouthed 158
ngaw (lonely, Lao) 150 opthalmos (eye, Gr) 75
Words and phrases index 419
orge (fury, Gr) 136 radost' (joy, Ru) 291, 293-9, 316, 330
otnoSenie (attitude, Ru) 315-7 rage 17, 118, 121
otryv (revel, Ru) 303 rasa (feel, Ma) 170
otvraSienie (revulsion, Ru) 314 razgovor (conversation, Ru) 330
outrage 5 razluka (separation, Ru) 291, 323-8,331
raznofintsy (intellectuals, Ru) 305-6
pa (prep., Mb) 72 -re (see, Mb) 72
paak (mouth, Lao) 155, 157-9; nom-~ rebuke 52
(clench-mouthed) 155, 159; qaa rejoice 17
(open-mouthed) 158 remorse 311
pepatah (maxims, Ma) 175 ren (benevolence, Ch) 376
pepe (don't, Mb) 71 ren qing (human feelings, Ch) 256, 282
perasaan (feelings, Ma) 170 rendah hati (humble, Ma) 167
peribahasa (trad, sayings, Ma) 175 revenge 121
petel (hungry, Mb) 71 rindu (longing, Ma) 170
phii (spirit, Lao) 159, baa (lunatic) rodnoj (beloved, Ru) 291, 320-2
159 rodstvennik (relative, Ru) 321
pikaringanyi (angry, Ya) 6-8, 210
pine 17 sabaaj (content, Lao) 151-2,158
pissed off 150, 162 sad, sadness 10, 13, 46-7, 66, 121,
pity, 319 149-50, 154, 169, 177, 359-99
pleased 15, 17 sak' 3 (laugh, Am) 56
-po se mata- (memorise, Mb) 80 sakit hati (annoyed, Ma) 168, 173-4,
-pok (uneasy, Mb) 72 179, 184,185, 186,190
poryv (impulse, Ru) 303, 305 samodovol'stvo (self-content, Ru) 295
prezrenie (contempt, Ru) 315 sbr (break, Am) 36
pride 190 scared 153
progolodats'ja (hunger, Ru) 324 sdast'e (happiness, Ru) 291, 295-9, 320,
przykro (hurt, Po) 25, 308, 337-43, 343, 330
344-56, 356 scastliv (happy, Ru) 298-9
puas hati (satisfied, Ma) 168, 173-4, scowl 155-5
179, 187, 188, 193 sdelat' (do, Ru) 9
pustoj (empty, Ru) 330 sedih (sorrow, Ma) 170
Seele (soul, Ge) 176
qaa-paak (open-mouthed, Lao) 158 sehati (of one mind, Ma) 175
qi (energy, Ch) 265, 268, 270-1, 283; samma (feel, Am) 22, 37-40
bie-zhe ~ (resentful) 268; chu~ (vent senang hati (relaxed, Ma) 168, 179,182,
anger) 270-1; -fen (infuriated) 275-6; 183-4, 193
~ren (irritate) 266; ~si (half-dead with separation 291, 326
anger) 268 serdce (heart, Ru) 196
qiu (autumn, Ch) 370-1 shame 4, 8, 66, 102, 198
shenglqi (anger, Ch) 24, 255, 265-9,
rabbia (fury, It) 134 269, 270-4, 276, 282, 284, 285
Rache (revenge, Ge) 139 shinkeishitsu (hyperanxiety, Ja) 222
rad (glad, Ru) 298 shy 55, 66,102
420 Words and phrases index