j,·''

STUDIES IN EMOTION AND SOCIAL INTERACTION

The psychology of facial
Second Series •
expression
Series Editors

Keith Oatley
Uniuersitu of Toronto

Antony Manstead Edited by

University of Amsterdam James A. Russell
University of British Columbia

José Miguel Fernández-Dols

Universidad Autónoma de Madrid
This series is jointly published by the Cambridge University Press and
the Editions de la Maison des Sciences de l'Homme, as part of the joint
publishing agreement established in 1977 between the Fondation de la
Maison des Sciences de l'Homme and the Syndics of the Cambridge Uni-
versity Press.

Cette collection est publiée co-édition par Cambridge University Press et

les Editions de la Maison des Sciences de l'Homme. Elle s'intègre dans le
programme de co-édition établi en 1977 par la Fondation de la.Maison
des Sciences de l'Homme et les Syndics de Cambridge University Press.

For a list of titles in the First Series of Studies in Emotion and Social In-
teraction, see the page following the index.

w~w CAMBRIDGE
= . - UNIVERSITY PRESS

& Editions de la Maison des Sciences de l'Homme

Paris
 PUBLISHED BY THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDGE
The Pitt Building, Trumpington Street, Cambridge, CB2 lRP, United Kingdom
and
EDITIONS DE LA MAISON DES SCIENCES DE L'HOMME Contents
54 Boulevard Raspail, 75270 Paris Cedex 06, France

CAMBRIDGE UNIVERSITY PRESS

The Edinburgh Building, Cambridge CB2 2RU, United Kingdom
40 West 20th Street, New York, NY 10011-4211, USA
10 Stamford Road, Oakleigh, Melbourne 3166, Australia

© Maison des Sciences de l'Homme and Cambridge University Press 1997

This book is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
Foreword page vii
the written permission of Cambridge University Press. GEORGE MANDLER
Preface xi
First published 1997
Contributors xiii
Printed in the United States of America

Typeset in Palatino Part I Introduction

Library of Congress Cataloging-in-Publication Data 1 What does a facial expression mean? 3
The psychology of facial expression / edited by James A. Russell, José JAMES A. RUSSELL AND JOSÉ MIGUEL FERNÁNDEZ-DOLS
Miguel Fernández-Dols.
p. cm. - (Studies in emotion and social interaction, second series)
2 Methods for the study of facial behavior 31
ISBN 0-521-49667-5. - ISBN 0-521-58796-4 (pbk.) HUGH L. WAGNER
l. Facial expression. 2. Nonverbal communication (Psychology)
I. Russell, James A. II. Fernández-Dols, José Miguel. III. Series.
BF592.F33P78 1997 96-36250
Part II Three broad theoretical frameworks
153.6'9 - DC20 CIP
3 Emotions and facial expressions: A perspective from
A catalog record for this book is available from Differential Emotions Theory 57
the British Library CARROLL E. IZARD
4 Facial expressions as modes of action readiness 78
ISBN O 521 49667 5 hardback
ISBN O 521 58796 4 paperback NICO H. FRIJDA AND ANNA TCHERKASSOF
ISBN 2-7351-0726-4 hardback France only 5 The new ethology of human facial expressions 103
ISBN 2-7351-0727-2 paperback France only ALAN J. FRIDLUND

Part III With a biological and developmental focus

6 Animal sounds and human faces: Do they have anything
in common? 133
PETER MARLER AND CHRISTOPHER EV ANS

V
vi Contents
7 Yawns, laughs, smiles, tickles, and talking: Naturalistic
and laboratory studies of facial action and social
communication 158 Foreword
ROBERT R. PROVINE
8 A neurobehavioral approach to the recognition of facial
expressions in infancy 176
CHARLES A. NELSON AND MICHELLE DE HAAN
9 A dynamic systems approach to infant facial action 205
DANIEL S. MESSINGER, ALAN FOGEL, AND
K. LAURIE DICKSON

Part IV With a psychological and social focus The chapters in this collection represent the best thought on the role and
function of human facial behavior, and many of them address what has
10 A Cornponential Approach to the meaning of facial
become central to most contemporary accounts - the link between facial
expressions 229
expression and emotion. The arguments for and against such a link be-
CRAIG A. SMITH AND HEATHER S. SCOTT
tween emotion and facial behavior are represented in the following
11 Spontaneous facial behavior during intense emotional
pages. It is in part a theme of this book that the belief in such a link was
episodes: Artistic truth and optical truth 255
not always thus in the past and that it need not be thus in the future.
JOSÉ MIGUEL FERNÁNDEZ-DOLS AND MARÍA-ANGELES
The current predominance of the Tomkins-Izard-Ekman account of
RUIZ-BELDA
the meaning of facial expressions and their strong dependence on emo-
12 Is the meaning perceived in facial expression independent
tions started sorne 30 years ago, and 30 years is generally the lifetime of
of its context? 275
regnant psychological theories. In recent years, a new wave of thought
JOSÉ MIGUEL FERNÁNDEZ-DOLS AND JAMES M. CARROLL
has emerged and the debate has been joined between those who see
13 Reading emotions from and into faces: Resurrecting a
facial expression as a necessary accompaniment or precursor of emo-
dimensional-contextual perspective 295
tional experience, and those who see facial expression as communicative,
JAMES A. RUSSELL
expressive, and mimetic devices, possibly remnants of an early preverbal
14 Facing others: A social communicative perspective on
language but not necessarily tied to emotion. I joined the wave sorne 20
facial displays 321
years ago and have been delighted by the momentum and the contrib-
NICOLE CHOVIL
utors it has gathered in the intervening period - most all of them rep-
15 Faces in dialogue 334
resented here.
JANET BEAVIN BAVELAS AND NICOLE CHOVIL
In retrospect, it seems strange that something as palpable and measur-
able as facial expression should be seen to be immersed in something as
Part V Integrative summary
vague and intellectually slippery as emotion. Since I have been in the
16 Faces: An epilogue and reconceptualization 349 past accused of "not believing in emotion," let me briefly expand on this
G. P. GINSBURG theme. In the common natural languages, we all know without much
doubt what feeling emotions and being emotional is all about. It is about
involuntary, strong, sometimes irrational feelings and commitments and
Author index 383
mental and bodily reactions to significant or at least significant-seeming
Subject index 391
situations and people.
But when we corne to the psychologists' and physiologists' emotions,

vii
 Facial expression as action readiness 79
emotional meaning does not necessarily imply attribution of inner feel-
ing. The advance was due to phenomenology (Buytendijk & Plessner,
1925; Sartre, 1939; Gestalt psychology) and Wittgenstein.
4. Facial expressions as modes of We think, therefore, that the category of facial expression (and other

I action readiness
NICO H. FRIJDA AND ANNA TCHERKASSOF
bodily expressions) should be defined as a category of impression. Facial
expression is facial behavior that suggests emotional meaning. This
II
leaves the relations to whatever is expressed and what expressing"
means, a matter of empirical and theoretical analysis. For simplicity, we
continue to use the term expression to refer to the facial behavior con-
cerned.

Questions for research on facial expression

Many questions one could ask about facial expressions seem to be fixed
by the use of the very word expression. Facial expression, first, refers to Analysis of facial expression has to address the following major ques-
facial behavior that suggests emotional meaning to an outside observer. tions:
Second, the term carries the implication that that facial behavior has the II
l. What does facial expression express"? That is, what information do

i
function or purpose of conveying such meaning. Third, it suggests that
there exists something (say, an inner feeling) independently of that be- observers perceive or infer from facial expressions (the receiver ques-
tion), and what psychological states or processes produce them (the

¡
havior to which the behavior called expression is added as an extra.
These aspects are not necessarily all true of the same behaviors. Facial sender question)? What does a smile convey to an observer, and what
behaviors may suggest emotional meanings to observers, but that may state or process actually underlies that smile?
not be their function or purpose. Receiving the epithet "expressive" in 2. Why do certain psychological states or processes lead to particular
fact says nothing about the nature of the behavior concerned. "Hasty" expressions? That is, what is the functional explanation of facial ex-
or "greedy" behaviors, for instance, are made to arrive as fast as one can pression? Why do we smile when happy (if we smile when happy
at the object of desire, and not to inform others about one's state of mind. and if we smile when happy)?
Also, nonbehavior may on occasion be highly expressive, such as un- 3. Why is that content so expressed? Why do facial expressions look the
deracting in the theater and Jesus's remaining silent under accusation. way they do and have the temporal properties that they have? Why
And there are phenomena that are expressive by suggesting emotional do we smile when happy (if indeed happiness is what makes us
meanings in which no inner feelings of whatever produced the phenom- smile)? Why its temporal and topographic variations?
ena are involved, such as joyful bird songs, angry bursts of wind, sad 4. When is that content expressed? When do expressions occur? Do facial
music, nervous lines, and solemn penguins. expressions appear whenever the expressed state occurs, or, if not,
Most past and current theorizing on facial expression starts from the what are the conditions?
assumption that it expresses emotional feelings and exists for the sake Traditional theory has simple answers to all four questions:
of doing so. The study of expression pretty much originated in the phil-
osophical problem of the knowledge of other minds. It sought to solve l. Facial expressions express emotional states, particularly feelings. Dif-
the riddle of how it is possible to obtain knowledge of others' unspoken ferent emotions or feelings correspond to different facial expressions
inner feelings (e.g., Bain, 1859; Berkeley, 1709; Lipps, 1905). That expres- (at some level of categorization of "different emotions").
sions manifest inner feelings probably is dogma to many researchers; at 2. Facial expressions exist for the sake of communicating emotions to
least it was when the first author started his work. Advance has been others.
possible only by abandoning this dogma, recognizing that perception of 3. Presumably, Darwin's (1872) three principles provide a satisfactory

78
80 N. H. Frijda & A. Tcherkassof Facial expression as action readiness 81
account of the nature and origin of expressions. However, most con- sistently assigned to six or so major emotion categories when judges are
temporary theorists are silent on why expressions are as they are. forced to choose from among those six or so (Ekman, 1982; Izard, 1977).
Other explanatory principles (notably, those from Bühler, 1934; Du- Such assignments are not usually made with 100% consistency, and ma-
mas, 1933a, 1947; Piderit, 1867) are largely ignored, except in some jor confusions occur repeatedly (Russell, 1994); still, agreement generally if
ethological work.
4. Expression occurs whenever the corresponding emotional state oc-
is way above chance and tends to occur across cultures (Ekman, 1982;
Russell, 1994). The affinity also appears from indications that the para- I
~
curs, except insofar as it is suppressed by control processes. digmatic expressions actually tend to occur under the expected emo-
tional conditions, and again cross-culturally so. For laughter, smiling,
and crying, there need hardly be discussion on this point. Subtler indi-
The basic facts of facial expression cations came, for instance, from Darwin's (1872) observations and from
This chapter is concerned mainly with the first question: What does facial the questionnaires he sent to about threescore missionaries. For instance,
expression express? Our answer follows from what we consider to be the facial expression of surprise, as described earlier, was made by the
the basic data on facial expressions. inhabitants of Tierra del Fuego "when the Beagle fired some rockets"
'j
(Darwin, 1872). I
l. There is a clear and distinct affinity between particular facial expres-
Each of these three sources of evidence is open to valid criticisms (Fri-
sions and particular categories of emotion. This affinity exists cross- dlund, 1994; Russell, 1994). However, together they constitute a solid
culturally and probably universally. body of indications, sufficient to establish an emotion-expression affinity
2. Emotion categories and facial expressions do not possess more than as a basic fact that expression theory has to explain.
an affinity, however. A given kind of emotion may give rise to dif-
Yet there is no more than an affinity. The second basic fact of facial
ferent facial expressions or to no facial expression at all. expression is that the expressions actually shown in emotional conditions
3. A given facial expression may be common to different kinds of emo- often differ drastically from the paradigmatic ones. Crying in happiness
tion as well as to psychological processes that are not distinctly emo- and nervous giggling provide obvious examples. It is true that such cry-
tional. ing or giggling is usually taken as evidence that the system is upset, but
"Affinity" between expressions and emotions, the first basic fact, deviations from what theory expects are also common under normal
means first of all that certain expressions form the preferred or paradig- conditions. For instance, in a study of the ecology of facial expression
matic representations of certain emotions, such as crying being typically (Frijda, 1953), a large number of emotional and other reactions were pro-
linked to sadness or grief, laughter to joy, and wide open eyes, lifted voked in two women in a conversational setting. Their facial expressions
brows, and dropped jaw to amazement or surprise (Ekman & Friesen, were filmed, their introspections were obtained after each segment, and
1975). This affinity forms the substance of traditional analyses of facial detailed recordings made of each eliciting event. Very many expressions
expression, throughout history and across cultures. These analyses are differed from those one would theoretically expect, given the eliciting
surprisingly consistent, from at least Lebrun (1667), Engel (.1785), and event. For instance, a self-reported incident of deep happiness was ac-
Camper (1792) onward.1 They also tend to be consistent with the 2,000- companied only by concentrated staring into space, and one of anger led
year-old mudras, coded facial expressions from the Indian baratha na- only to looking away. Wagner, MacDonald, and Manstead (1986), too,
tyam dance repertoire (see, for instance, Gopal, 1951), and their found a high proportion of nonparadigmatic expressions under natural
equivalents in kathakali dancing. They are said to follow the rules of conditions. Fernández-Dols and Ruiz-Belda (chapter 11, this volume)
"rasabhinava," which means "communication of emotional states by ex- present observations along the same line.
pressions of the face" (Bonneau-Le Breton, 1994, p. 175). Analysis by Emotions may be accompanied by no facial expression at all, or not
FACS (Facial Action Coding System) could and should be made to verify by characteristic ones (Fridlund, 1994; Frijda, 1986; Izard, 1977). This ap-
the similarities. plies even to very strong emotions, and not only because of self-control.
"Affinity" also refers to the fact that these same expressions are con- The expression researcher Dumas (1933a) reproduced photographs of
82 N. H. Frijda & A. Tcherkassof Facial expression as action readiness 83

victims of the Chinese torture called "fragmentation into a thousand Interestingly, the "errors" in interpretation usually appear quite rea-
fragments." Victims' faces were bland, or showed merely gasping. And sonable. Although some of them appeared to be due to neglect of some
ever since antiquity, philosophers have discussed the story of the Egyp- subtle feature of the stimulus pattern, in most cases the expressions
tian king Psammenitus, related by Herodotus (Histories, III, 14). When might very well have come from the states to which the subjects attrib-
taken prisoner by Cambyses of Persia, Psammenitus was forced to watch uted them. Startle, distress, and deep thought may well on occasion pres-
his daughter passing by dressed as a slave, and his son on his way to ent the same frown, bent head, and forcefully closed eyes. We can
execution. The king was observed to keep a bland, unmoved face. How- conclude that the same or similar facial expressions can occur with dif-
ever, he burst out weeping when noticing a friend reduced to a beggar. ferent states, emotional as well as nonemotional.
When questioned by Cambyses, Psammenitus answered that some griefs Certain facial expressions occur under conversational rather than emo-
are too great for tears, an interpretation accepted by Cambyses and by tional conditions. They are the more voluntarily produced conversational
later philosophers. signals called emblems by Ekman and Friesen (1969) and mimiques by
The third basic fact is that expressions tend to be common to several Dumas (1933b, 1947). Distinctions, between voluntary and less voluntar-
states, emotional as well as nonemotional. This fact is suggested by the ily produced expressions are perhaps hard to make on a neurological
range of emotion interpretations given to almost every expression in rec- basis (Fridlund, 1994); in actual practice, they seem to be made smoothly.
ognition experiments, and by the range of conditions under which any We leave emblems or mimiques out of our discussion.
given expression actually occurs.
It is rare to find an expression in a recognition experiment that is What is inferred from facial expressions?
interpreted by the subjects in one sole way (Russell, 1994). For instance,
the expressions described by Ekman and Friesen (1975) as expressions The three basic facts discussed lead us to a paradox. On the one hand,
of surprise were considered to be fear expressions by a nonnegligible there are reasons to doubt a strict correlation between particular emo-
proportion of the subjects in several studies. Also, providing the subjects tions and particular expressions. On the other hand, there is a general
with more labels to choose from than the seven emotion names increases tendency to attribute emotions to other people on the basis of their facial
the range of interpretations (Russell, 1994). Yet the choices do not become expressions.
random. These data have led to the conclusion that facial expressions do But is that tendency indeed so general? Do people indeed attribute
not correspond to particular emotions, but to regions in a two- or three- emotions to other people whenever such others show facial expressions?
dimensional emotion plot (Russell, 1980; Woodworth & Schlosberg, 1954) Is "attribution of emotion" the best way to characterize what people do
or to relatively broad emotion classes (Ekman, 1993; Russell & Bullock, under such circumstances? Data on the process whereby people assess
1986; Woodworth, 1938). In experiments that leave the subject entirely the meaning of expressions suggest otherwise.
free in what interpretations to make, ranges are still larger. In one such Not many studies have looked at process. Standard expression rec-
experiment (Frijda, 1953), most interpretations of a given expression ognition experiments do not allow us to do so because they force the
could indeed be viewed as falling within an emotion region or class that subjects to select an emotion label; subjects cannot respond in other ways.
also contained the emotion label considered "correct" (that is, the one Some information does come from experiments that allow subjects to
given by the filmed subject or the filming experimenter). However, a respond freely, as in the experiment by Frijda (1953). In that experiment,
sizable number of the interpretations could not easily be seen as belong- subjects were presented with the film segments of spontaneous expres-
ing to such a region or class. They did not refer to emotions but to sions described earlier, and with slides taken from those films. They were
cognitive or instrumental responses, such as "deep thought" for a startle asked to describe "what might be going on in the person shown or what
response, and "water is splashed in her face" for the reaction to a grue- might have happened to her." Responses were recorded verbatim, or
some story. The reverse also occurred: Expressions of physical effort or nearly so. Various conclusions could be drawn.
concentration were often interpreted as expressions of emotion (e.g., First, quite often the subjects mentioned no emotion label. Fitting an
pulling a rope as aversion and deep thought as distress). emotion label to a perceived expression is clearly not an ubiquitous el-
84 N. H. Frijda & A. Tcherkassof Facial expression as action readiness 85

ement when interpreting expressions. When an emotion label was men- erroneous to regard the free descriptions as the results of recalling sit-
tioned, it often was not the first and most direct step in the response uations that had been linked to the expressions in the past, as Fridlund
process. It usually followed and was an inference from other kinds of (1994, pp. 276-277) seems to suggest. The facial expressions did not re-
mind the subjects of situations in which they had seen them before. They
response.
Second, the most direct response was often to imagine and describe invented or imagined the situations, as several subjects made clear (see
an emotionally charged situation that appeared to fit the perceived ex- also Ruckmick, 1921, for similar observations). They creatively imagined
pression. For instance, one film clip showed the target person waiting situations that fit the expressions' apparent emotional implications.
for an electric shock with manifest tenseness, anxious attention, and try-
ing to brace herself. One subject described it as follows: "As if she is
looking at something with fixed attention, a game or something tense, Recognizing expressive information
two cars which almost get into collision, but nothing happens" (Frijda, The fact that subjects imagined situations that they felt fit the expressions
1953, p. 314). One of the slides showed the target person dreamily think- has an important implication. It means that the expressions contain in-
ing of her work (painting). It was given the interpretation: "She looks formation. Terms like attribution or inference obscure the fact that expres-
the way you look at a small child playing." A film clip of the target's sions possess a meaning that the subject perceives or deciphers in them
head when pulling a rope: "Just like she sees something very nasty." rather than adds to them. Indeed, that meanings are inherent in expres-
Sometimes an emotion label was added that was felt to fit the target's sive phenomena has been the accepted view in earlier discussions of
response to the imagined situation. "Something of contempt in it, a bit perceiving such phenomena; Gestalt psychologists coined the term phys-
tense" was the label given for the response to the almost-collision situ- iognomic perception (Koffka, 1935; Köhler, 1929; Werner, 1926/1948).
ation (Frijda, 1953, p. 312). Expressions dictate a Gibsonian view. There is information contained
Third, the expressions were perceived as parts of the target person's in them that can be picked up by an observer before interpretations or
interaction with her environment. The slides and films showed only the attributions are made. This raises the question of what that information
target person's head and shoulders; yet she was perceived in a situation. is. The preceding has shown that it is not "emotions." Attributing emo-
She was perceived not as "displaying a facial expression" that signaled tions goes beyond the information that the expressions contain because
some emotional state "within" her but as a person actively responding emotion labels may vary with one particular expression. We think that
to an event in her environment or her thoughts, attending to something the information contained in expression can be identified by examining
or as explicitly not attending to something . .She was seen as shielding what in fact is conveyed by them in daily interactions.
herself from something, withdrawing from something, accepting or not In such a context, it rarely is some verbal or abstract categorization,
accepting something, opening up to it or closing off from it, or some as in expression recognition experiments. People do not usually mutter
combination of these. In brief, the target persons were seen as interacting something like "Lo, anger!" to themselves. Under many conditions cat-
with their environment, and the facial expressions were seen as behav- egorization would not even be possible - for instance, when confronted
iors that were part of the interaction. The behavioral and interactive na- with the dynamic expressions found in ballet, mime, or music. But in
ture of the expressions was particularly evident in the film clips, where daily interactions, people give evidence that another person's expression
it contributed to the accureçy of interpretations (accuracy scores for the is recognized or understood, be it correctly or incorrectly, in at least three
films were 50% higher than for the slides; Frijda, 1953, p. 306). nonverbal, nonabstract ways. All three can be traced in the protocols
Fourth, it should be strongly emphasized that what the expressions from the Frijda (1953) experiment.
convey is essentially emotional, even if the information grasped from
them is not best represented by emotion labels. The situations were imag- Environmental expectation. Another person's facial expression may
ined for their emotional significance: as suspenseful, frightening, nasty, direct one's attention to a particular object and evoke an expectation
or endearing situations, to fit the expressions. An expression evidently about that object that fits the expression. Recognizing a fearful ex-
suggests a situation with a particular emotional content. It would be pression may just mean looking around for a threat or perceiving
 Facial expression as action readiness 87
86 N. H. Frijda & A. Tcherkassof
The data discussed lead to the hypothesis that expressions correspond
a given object as a threat. Understanding the meaning of someone's
to something more general than "emotions," or at a different plane of
crying may just mean that one wonders what loss occurred.
analysis - namely, the subject's "positionality" (Frijda, 1953) or "rela-
Affective response and behavioral expectation. Another person's facial tional activity" (Frijda, 1986). Facial expressions represent the manner in
expression may modify one's affective state and evoke expectations
which the individual at that particular moment relates (or does not re-
about how the other person will respond in further interactions that
late) to the environment. They represent the position taken: accepting or
fit the expression. Understanding someone's angry expression often
refusing, moving toward (in the sense of proximity seeking), moving
just means becoming frightened and expecting further hostile be-
away, or moving against. They represent the activity or lack of activity
havior. "Understanding" of this nature has been observed even in in taking position: highly or weakly active, or inactive as in apathy or
babies under 6 months (Schwartz, Izard, & Ansul, 1985). rest. They also represent the manner of that activity: whether it is being
Empathic identification response. Another person's facial expression deployed freely or under restraint, inhibited as in anxiety paralysis, or
may engender an empathic response: imitatory movement, or just lacking in direction as in nervousness.
the sense that one can identify the perceived expression in terms Because of their temporal dynamics, among other things, facial ex-
of one's own expressive repertoire. Expression empathy is best con- pressions usually point to the motivational states that engender the re-
sidered a nonverbal categorization or identification act. Recogniz- lational activity. We call them states of action readiness. Facial expressions
ing an angry face sometimes just means sensing the contraction and express states of action readiness, which we specify as states of readiness
nasty, pushing forward movement implied, or even involuntarily to establish, maintain, or change a particular kind of relationship with
producing them by motor mimicry (see Frijda, 1953, 1956, for ob- some object in the environment or in thought, or with the environment
servations and discussion). as a whole (Frijda, 1986, chapter 2). States of action readiness vary ac-
None of these three kinds of recognition response implies attribution cording to their aim (obtaining proximity, avoiding contact, neutralizing
of a feeling state to the perceived person nor any form of explicit cate- obstruction, etc.) and their degree and manner of activation (hyperactì-
vation, hypoactivation, tenseness). Major modes of action readiness cor-
gorization. Attribution and categorization, by contrast, would seem to
depend upon prior nonverbal recognition of expressive meaning and to respond with major modes of subject-environment interaction or major
interactional goals (cf. Roseman, Wiest, & Swartz, 1994). Theoretically,
involve going beyond that meaning.
only a limited number of modes of subject-environment interaction can
be distinguished, and these can meaningfully be called basic, as all states
What does facial expression express? The notions of relational of action readiness represent one or more of these modes.
activity and action readiness State of action readiness implies a tendency to control behavior. That is,
states of action readiness have the property of" control precedence" (Fri-
Identifying the information contained in expressions is of double interest.
jda, 1986). They are likely to lead to action and to interfere with ongoing
It clarifies the process of understanding expressions by an observer and
actions. They are involuntary and "impulsive" or unplanned in nature.
of his or her making emotion attributions on that basis. And it identifies
Because they are motivational states or goals, each state of action read-
what in fact is expressed in expression - that is, what the states or pro-
iness may get expressed through a variety of behaviors, including mere
cesses are in a sender that most directly cause the expressions.
How do we characterize that which is expressed? The information con- mental actions (wishes, plans, fantasies). Therefore, although facial ex-
pressions point to states of action readiness, the reverse is not always
tained in facial expressions is, we think, that which is common in raising
true. They may or they may not appear in expression or in other behav-
appraisal expectations, evoking affect and behavior expectations in in-
ior. States of action readiness can remain impulses or states of readiness
teractions, and empathic responses. It is, in addition, what is common to
and nothing more.
the various conditions under which a given expression arises and to the
Different modes of action readiness correspond to the 6 to 10 dimen-
various emotional and nonemotional states that may elicit a given ex-
sions derived from theory (Frijda, 1986) and from questionnaire research
pression.
 Facial expression as action readiness 89
88 N. H. Frijda & A. Tcherkassof
(Davitz, 1969; Frijda, Kuipers, & Terschure, 1989). The notion of states Table 4.1. Action readiness ratings: Percentages of "somewhat applicable"
of action readiness thus allows for more differentiation than the two or and "very much applicable" ratings (entries of 60% or over)
three dimensions of expression posited by Schlosberg (1954) and Russell
Facial expression group
(1980). This is as it should be. The information contained in facial ex- Action readiness
pressions is richer than variation along two or three dimensions. This mode Anger Sadness Disgust Fear Contempt Surprise Joy
appears from several studies. In one study, ratings were made on 22 Approach 81
bipolar scales of 30 posed facial expressions. Factor analysis of the ratings Being with 76
yielded four orthogonal factors when average ratings per photograph Protect oneself 74 77
Avoid 75 81
were used (Frijda & Philipszoon, 1963), and six when the individual Attending 81 69 72 90 77 95 76
ratings were used: pleasantness, activation, attentional activity, sponta- Keep distance 70 83 81
neity-reactivity, surprise, and simple-complex (Frijda, 1969). Almost the Reject 95 92 81
Boil inwardly 89
same factors appeared in a second study with posed photographs of a Agonistic 92 90.5. 71.5
different target person (Frijda, 1969). In a third study, subjects were Reactant 88
asked to check which of 110 emotion adjectives applied to each of 62 Interrupt 71.5 90.5
In command 72.5 82
posed expressions of an actress and 68 posed expressions of an actor. Submitting 69
Factor analysis of the frequencies of co-occurrence of the adjectives Helplessness 92 93 84.5
yielded 17 or 18 unipolar factors with over 1 % contribution to variance; Tensely contracted" 93 83 94 100
the factors from both sets were rather similar. Other subjects rated the Open 65.5 64 96
Crying 80
same photographs on 40 bipolar 7-point scales, which resulted in seven Laughing 92
bipolar factors. Cluster analyses yielded similar differentiation (Frijda,
1970, 1973). ªThe phrase "tensely contracted" translates the French "crispation." The latter
Most of these factors, if not all, do not just represent the semantics of has a defensive, inhibitory overtone that the English designation does not have.
emotion words. In Frijda (1969), significant correlations were obtained
between the factor scores and ratings of various simple and complex
facial feature measures (e.g., smiling, frowning, approach-avoidance emotion items representing the Matsumoto and Ekman (1989) emotion
labels.
score, tenseness) for both target persons.
Table 4.1 gives the percentages of subjects who checked the action
readiness items for each group of slides meant to represent a given emo-
Expressions and action readiness: Empirical support tion category. (Scores for the four slides in a group are combined, as are
If facial expressions correspond to states of action readiness, subjects ratings of "somewhat applicable" and "very much applicable"; the "very
should have little trouble associating particular states of action readiness much" scores alone give almost the same picture. To save space, only
with particular facial expressions. Support for this prediction was ob- the items showing variation over the seven groups are included in the
table.)
tained in a preliminary experiment (Tcherkassof, in preparation). Sub-
jects were presented with 28 facial expression slides from Matsumoto The subjects clearly found the task meaningful. A large majority
and Ekman' s (1989) series, four for each of the seven emotion categories. agreed upon at least one of the action readiness items for each slide
They rated each expression on 34 action readiness items. The items came group; for all but the contempt slides, at least one item was checked by
from the questionnaire used by Frijda et al. (1989), with slight adapta- over 90% of the subjects for the four slides in the group together. The
tions, and were presented as 3-point scales (not applicable, somewhat ap- subjects showed unanimity on at least one item for 21 of the 28 individ-
plicable, and very much applicable). Items for crying and laughing were ual slides. All but two of the slides (both from the contempt group)
included. In a second round, subjects rated each slide on seven 3-point showed at least one action readiness item with 90% agreement or over.
90 N. H. Frijda & A. Tcherkassof Facial expression as action readiness 91

One would predict that those action readiness items in particular will Table 4.2. Emotion ratings: percentages of "somewhat applicable" and "very
be checked that figure prominently in the semantics of the emotion label much applicable" ratings (entries of 60% or over)
of the slide group concerned. Specific predictions were derived from
questionnaire data on action readiness ratings for emotion incidents (Frij- Facial expression group
da et al., 1989). For contempt, no predictions could be derived. The pre- Checked emotion Anger Sadness Disgust Fear Contempt Surprise Joy
dicted entries have been underlined in Table 4.1. All of them were
Anger 87
checked by 77% of the subjects or more. A number of nonpredicted high Sadness 97 62
scores also emerged, however. Disgust 62 98.5 60
We hypothesized that emotion attributions are inferences from assess- Fear (57) 97
Contempt 84 68
ments of states of action readiness. This implies that expressions labeled Surprise 92 98.5
differently probably differ in terms of action readiness. The present ex- Joy 93.5
periment gives clear cues that this might indeed be the case. Discriminant
analysis (using all 34 items), with the emotion group labels as the cri-
terion, yields 74.8% correct assignments. Our hypothesis further implies and other object-related actions with a similar relational sense. One
that agreement on action readiness assignments to expressions should would expect a frightened face made when confronting a material event
be at least as high as on emotion attributions (Table 4.2), provided that to be at least accompanied by withdrawal movements of head and shoul-
all expressions given the same emotion label contain the same mode of ders, if it is not accompanied by crouching or flight. During anger, the
action readiness. This latter condition was not fulfilled, however. For facial expression is often accompanied by general muscular tensing, fist
instance, two of the sadness photographs were rated as distinctly agonis- clenching, and forward bending or stiffly erect posture. A relaxed smile
tic, while the other two were not. The prediction could also not be rig- tends to be accompanied by a slowing down of respiration (Dumas,
orously tested because the number of action readiness items was about 1948). In fact, respiration changes accompany facial expressions that are
five times that of emotion items. Yet it is interesting to compare the two adopted voluntarily, and they are probably responsible for the auto-
types of ratings. Two results are worthy of note. First, the highest per- nomic changes observed by Ekman, Levenson, and Friesen (1983)(Boiten,
centages in comparable columns of Tables 4.1 and 4.2 do not differ much, 1996).
and as already mentioned, almost all individual slides showed such a Although little or no research exists on this issue, there are scattered
high percentage on at least one action readiness item. Second, similar descriptions, mostly of emotional expressions of mentally ill people ( e.g.,
confusion patterns occur for both rating types. Expressions meant to de- Darwin, 1872; Dumas, 1933a) and of children (Bonneau-Le Breton, 1994),
pict contempt do less well than those from other groups in both rating that support the suppositions of behavioral context with a similar sense,
types, while the similarity in action readiness patterns of the sadness, fear, and so does available work on posture (e.g., De Meijer, 1991). Recogni-
and disgust groups in Table 4.1 (and as evident in the discriminant anal- tion studies show that the information conveyed by posture and gross
ysis) is reflected in the off-diagonal elements in Table 4.2. body movement primarily concerns action readiness in the most literal
sense, approach-address versus withdrawal-avoidance and dominance-
submission, and activation control (tenseness) (Riskind, 1984; see De Mei-
The behavioral context jer, 1991, for review and empirical data).
To view expressions as forms of relational activity is strengthened by
examining the behavioral context in which facial expressions usually ap- What kinds of behavior are facial expressions?
pear. Facial expression literature curiously neglects that context. The ex-
pressions are generally treated as if they stand on their own. This neglect "Expressive behavior" is not a substantive category but an impression
most probably is a mistake. Facial expressions tend to appear in a context notion, as we said earlier. One cannot, therefore, expect facial expressions
of head and body orientations, gross body movements, posture changes, to be all of a kind. States of relational action readiness are indeed re-
92 N. H. Frijda & A. Tcherkassof Facial expression as action readiness 93
fleeted by different kinds of behavior. Even instrumental behavior like of intimidating the obstructor, who has to "cooperate" by understanding
eating is expressive (of greed, for instance) when its dynamic properties the threat and responding to it (Fridlund, 1994; Frijda, 1982, 1986).
show the intensity of readiness to get to the goal. 3. Activation and deactivation manifestations. Activation is defined as
Facial behavior that carries expressive information can be subsumed "tonic readiness to act" (Pribrarn & McGuiness, 1975). Activation and
under four headings (1) relational activity proper; (2) social signals; (3) activation loss are equivalent to Darwin's third principle of" direct action
activation and deactivation manifestations; and (4) inhibition manifes- of the nervous system" as sources of expressive movement. Activation
tations. variation was considered the major explanatory principle for expression
l. Relational activity proper is behavior that directly modifies the indi- by both Spencer and Dumas (1933a). There are different modes of acti-
vidual's relationship to its environment. It is the kind of behavior from vation: that involved in relational or instrumental actions, that corning
which Darwin derived his first principle, but that was discussed more from physical effort such as the face made when cracking a nut, "su-
elaborately by Engel (1785), Piderit (1867), and Wundt (1902). Sorne facial perfluous" activation such as characterizes much joyful behavior, and
expressions are actions to decrease or increase sensory intake (e.g., open- restrained activation that goes by the name of "tenseness." Deactivation
ing or closing the eyes or nostrils, head aversion, mouth movements in is illustrated by the drooping features of relaxation, fatigue, and sadness.
response to aversive tasting substances; Chiva, 1985). Other expression 4. Inhibition involves response blocking under activating conditions.
features are parts of approach and withdrawal movements, such as winc- Freezing in rodents is usually quoted as exemplifying inhibitory process,
ing, and probably raising the eyebrows in surprise (Fridlund, 1994; Fri- and so is generalized response blocking in anxiety (Gray, 1982); both can
jda, 1986). The surprise expression as a whole is an orienting reaction. be understood as cautionary responses to unpredictable or aversive con-
Frowning is a curious case of relational behavior. It is relational in that ditions. Facial expression features that are best explained as inhibitory
it corresponds with the effort of maintaining focus upon one's goal under features are the sagging jaw in amazement, and the rigid, drooping and
difficulty (Schänzle, 1939; Smith, 1989); it seems to be effective in that bland features in anxiety and certain anger reactions (Darwin, 1872; Du-
function (Frijda, 1986, p. 21), although it is obscure how. mas, 1933a).
Many expressive movements are intention movements, the initial
stages of relational actions proper (Handlungsinitien, action onsets, Facial expression is here given a functional analysis. Expressions are
Bühler, 1934). Readying one's jaw or fist for aggressive approach in anger the way they are because the relational activity, social influencing, and
is an example. Intention movements are evident states of readiness. activation implement the aims of the state of action readiness at hand.
2. Certain facial expressions are social signals meant to influence the Facial expression at any given instant of time can be understood from
behavior of others. They are nonverbal requests or commands (Frijda, the functions of the composing elements in protection, orientation of at-
1982, 1986), and are extensively discussed by Fridlund (1994, and chapter tention, motor realization of activation, signaling affiliative intent, and
5, this volume). A threat display discourages approach or persistence in the like. This analysis clearly leads to a cornponential view of facial ex-
whatever elicits it. Certain smiles signal willingness to establish contact pressions, such as proposed by Scherer (1992) and Smith (1989, and chap-
without aggressive intent. Crying tends to induce succorance and prob- ter 10, with Scott, this volume). Individual components of expression
ably is meant to serve that purpose. each have their functions, either as independent bits of relational activity
Relational activity proper and social signals are functionally not so or as modifiers or intensifiers of other components. Scherer (1992) and
different. Relational activity proper serves the prevailing state of action Smith and Scott (chapter 10, this volume) primarily link the components
readiness directly: It directly helps to maintain, achieve, or modify a to their appraisal antecedents. We link them to their action readiness
given type of relationship. Facial social signals likewise serve that pre- sources and relational functions.
vailing state, but they do so indirectly by invoking the activity of an Components belonging to different modes of relational activity may
interactant. The aim of aggression is to end an obstruction by neutral- coexist. For instance, protective eye closure coexists with keeping the
izing the obstructor; the aim of threat is to obtain the same end by way eyes open to maintain visual contact in sorne fearful expressions. Each
94 N. H. Frijda & A. Tcherkassof Facial expression as action readiness 95

facial expression can be read in terms of composing modes of action to generate expectations about environmental events and about possibly
readiness, as in this example. Traces of relational activity can be subtle. forthcoming behavior by the observed person, and to identify the per-
A tinge of reserve may occur in an expression that is otherwise open ceived readiness in terms of one's own felt readiness repertoire. Knowl-
and receptive; that is, some slight narrowing of the eyes, or the head edge and imagination then help out, given the time and the inclination,
being held somewhat stiffly, may occur with directed attention and for- to construct hypothetical representations of the viewed person's emotion
ward stretching of the head. Systematic variation on facial components or other state, and to find a fitting label for it.
allows analysis of how the information content of expressions varies cor-
respondingly. A nice example is found in a study by Garotti, Caterina, Facial expressions and emotions
Brighetti, Giberti, & Ricci-Bitti, (1993). They systematically varied specific
features in schematic drawings of smiles (e.g., eye narrowing, opening We claim, then, that facial expressions are relational activities, social sig-
the mouth) and, submitting the drawings to subjects in a recognition nals, activation manifestations, and inhibitions, all of which flow from a
experiment, obtained clearly different emotion attribution patterns for state of readiness to maintain or change the relationship with the envi-
ronment. Does this mean that facial expressions have nothing to do with
each variation.
emotions? Not at all. On the contrary. Emotions and facial expressions
are intrinsically related for the simple reason that emotions are states of
The process of expression understanding action readiness. More precisely, emotions are best viewed as action dis-
How is it possible to understand facial expressions, the old philosophical positions (Lang, 1995) or states of action readiness elicited by antecedent
question with which expression study began? Understanding facial ex- events as appraised and manifesting some degree of control precedence
pressions becomes rather transparent from the present perspective, (see Arnold, 1960; Frijda, 1986; Lang, 1995; Scherer, 1984, for convergent
which may be taken as an argument in its favor (Frijda, 1956). views). Also, a change in state of action readiness is perhaps the major
Usually, one of the three traditional explanations - associative learning aspect of behavior or experience that leads to use of the word "emotion,"
(Berkeley, 1709), "reasoning by analogy" (Bain, 1859), and "empathy" both to denote one's own state and that observed in others. The word
(Lipps, 1907) - is taken for granted. None provides a satisfactory expla- emotion is rarely used for .mere feelings that do not involve some claim
nation. Associative learning does not because it renders novel expres- on change in action readiness.
sions (like those in ballet dancing) unintelligible. Reasoning by analogy The emotion-expression relationship is greatly clarified by the com-
does not because it appears too sophisticated for, for instance, animal ponential approach to emotions (e.g., Frijda, 1986; Lang, 1995; Lazarus,
recognition of expressions. Empathy does not because it itself is largely 1991; Mandler, 1984; Scherer, 1984). According to that approach, emo-
tions are structures of moderately correlated components. Affect, ap-
an interpretative response (Frijda, 1956).
An ability to grasp the sense of relational activity and activation, how- praisal, action disposition, and physiological response are the major
ever, is not a great mystery. It merely requires that movements be components. Emotional feelings are considered as one's awareness of one
viewed as behavior - that is, as purposive, as movements related to the or more of these components. Different emotions can be viewed as struc-
organism's environment and as guided by aims in relation to that en- tures that differ in one or more of these components. Emotion words
vironment. It requires, in short, that movements are perceived from an specify such structures at a given level of specification, though not nec-
intentional stance (Dennett, 1978), which presumably implies a process essarily with regard to all components together.
of the same elementary nature as perceiving causality (Michotte, 1950). Many emotion words specify a particular state of action readiness. In
Recognizing action readiness in behavior requires no great step either. several languages there exist close links between major emotion catego-
It just amounts to picking up the cues that tell that things may come ries and modes of action readiness (Davitz, 1969; Frijda et al., 1989;
from the things that are. From here, the three recognition modes of daily Frijda, Markam, Sato, & Wiers, 1995; Roseman et al., 1994). The links
life are not so difficult to understand. Once relational activity is grasped between major emotion categories and modes of action readiness parallel
as involving relational action readiness, the material is there from which those between major emotion categories and facial expressions. For in-
96 N. H. Frijda & A. Tcherkassof Facial expression as action readiness 97

stance, hostile or threatening impulse is one of the major semantic com- overt behavior, including facial expression. As we mentioned, states of
ponents of "anger" and its nearest equivalents in many other languages, action readiness are not always expressed in overt behavior, and they
and a prototypical angry expression translates a hostile or threatening may lead to widely divergent overt behaviors. In addition, control pro-
action readiness. This explains Ekman's (1982, 1993; Ekman & Friesen, cesses, including social display rules, obviously work to widen the split.
1975) findings. It accounts for the affinity between particular emotions Moreover, there is no simple relation between the impact of an emo-
and particular expressions, the first basic fact mentioned. tional event and action readiness. The event aspect that controls action
Yet those links between emotions and expressions are neither neces- readiness, or which mode of action readiness controls behavior, is not
sary nor exclusive, as follows from multicomponential emotion theory. always the central aspect of the event's impact. We earlier gave the ex-
First, emotion words are generally used in fuzzy, nondeterministic fash- ample of a target person in the Frijda (1953) experiment who was think-
ion, to denote now this, then that of the moderately correlated compo- ing with feelings of deep happiness about some work she had been
nents (Fehr & Russell, 1984; Russell, 1991; Shaver, Schwartz, Kirson, doing. Her face mostly showed her concentrated attention, and not so
& O'Connor, 1987). They may primarily indicate particular appraisals much the happiness. Similar observations are made by Kraut (1982) and
rather than modes of action readiness. "Anger" is often used to denote Fernández-Dols and Ruiz-Belda (chapter 11, this volume). They found
the response to an event appraised as blameworthy, regardless of the that events causing important satisfaction or happiness frequently do not
nature of the resulting state of action readiness. That may be an impulse lead to happy expressions; the subjects smile only when facing the pub-
to break off contact rather than hostile impulse, or drastic loss of action lic. Part of the explanation may be that concentration and exhaustion
readiness, as when paralyzed and almost fainting from anger. Other after effortful feats overlay or delay actual feelings of joy or triumph.
emotion words do not specify the state of action readiness at all. "Being Then both action readiness and facial expression may vanish when
upset" suggests that action readiness is somehow interfered with - action action appears meaningless. Brehm (1994) advances the hypothesis that
is interrupted - but no more. The links between emotion categories and the relationship between event intensity and emotional response has the
modes of action readiness, while sometimes clear and intrinsic, thus are shape of a sawtooth rather than being monotonic. In maximally intense
in no way fixed or absolute. emotions, response magnitude drops steeply. The hypothesis is corrob-
Second, not all states of action readiness (and thus, not all emotions) orated by the Chinese torture report and the story of Psammenitus. These
give rise to facial expression or to a very specific facial expression. The considerations, too, account for aspects of the second basic fact.
emotion of desire, understood as the impulse to get closer in order to Fourth, relational actions, social signals, and activation manifestations
possess, is an example. Facial expression, moreover, is only one among need not originate in a state of action readiness. They may come just
many kinds of action that a given state of action readiness may com- from physical exercise or represent hypotonie boredom reactions that
mand. Each kind of action has its own determinants in addition to the look like dumb amazement. They may also come from voluntary intent
state of action readiness. One additional determinant for facial expres- to signal a particular event appraisal to others ("how awful is what you
sion is physical context: A fearfully contorted face is more likely in front tell me") or to suggest a state of action readiness that is not really there.
of a threatening physical object than when facing failing an examination. These are the conditions for mimiques or emblems (Dumas, 1933b; Ekman
Another determinant is prevailing activation mode. For instance, the & Friesen, 1969).
nonsocial smile may represent "active rest" or "sense of mastery" rather Attributing emotions necessarily involves going beyond the informa-
than just generally pleasure or joy (Buytendijk, 1947; Frijda, 1986; Sroufe tion given by expression. One has to add that the relational action, social
& Waters, 1976); and Ellgring (1989) notes that facial expressions in de- signal, or activation state was elicited by an appraised event, and that
pressed patients depend more upon their initial behavioral level than these actions involve a state of action readiness with control precedence.
upon the prevailing emotion or mood. Other determinants again are so- One frequently has to add hypotheses on the nature of the eliciting event
cial influence and expected effectiveness, as extensively examined by Fri- and appraisal - hence, the confusions in recognition experiments. This
dlund (1994). All this accounts for the second basic fact: the loose and accounts for the third basic fact, the looseness of the emotion-expression
variable relation between particular emotions and particular expressions. relationship.
Third, there is no simple relationship between action readiness and As indicated before, the major forms of action readiness can be con-
98 N. H. Frijda & A. Tcherkassof Facial expression as action readiness 99

sidered basic forms of subject-environment interaction. They form dis- other factors, in addition to feeling, event appraisal, and action readiness.
crete categories, very much as the basic emotions in Izard's (1977) theory. These factors dilute the link between emotions and expressions. They do,
Since certain emotions paradigmatically embody these forms of action however, in no way sever it. Although the link is neither exclusive nor
readiness, and the corresponding states may paradigmatically be mani- necessary, it is an intrinsic one.
fest in certain facial expressions, it makes sense to consider these emo-
tions as basic emotions, and certain facial expressions to represent them. Acknowledgment
However, recognizing basic emotions in this sense in no way implies the This manuscript was prepared with the assistance of the members of the CERE
lore around that notion: unitary biological predispositions, felt qualia, group, and the support of the Maison des Sciences de l'Homme, Paris, and the
Université de Paris-X Nanterre.
solid links between the various components, a stiff emotions hierarchy.
That lore, in fact, we firmly reject.
Note
1 Lebrun was a painter, Engel a theorist of the stage, and Camper a Dutch anat-
Conclusions omist (in fact, the founder of comparative anatomy).
11
Facial expressions, we propose, express" the individual's state of rela-
tional action readiness or unreadiness. State of action readiness is the References
proper content of facial expressions, and that which observers infer from Arnold, M. B. (1960). Emotion and personality (Vols. 1 and 2). New York: Columbia
II
them in the first place. Expressions express" it, in the sense that an University Press.
Bain, A. (1859). Emotions and the will. London: Longmans and Green.
action expresses its underlying intention. They implement or effectuate Berkeley, G. (1709). An essay towards a new theory of vision. London.
action readiness. That is, they do so unless they come from a different Boiten, F. A. (1996). Autonomic response patterns during voluntary facial action.
source, such as social habit, deceit, or voluntary intent. Psychophysiology, 33, 123-131.
II Bonneau-Le Breton, C. (1994). Expression faciale des émotions: comparaison des mi-
Because the core of what one calls emotions" consists, by and large, miques évoquées et provoquées, dans une perspective developpementale. Ph.D. the-
of variations in the individual's state of action readiness, facial expres- sis, Université de Paris-X Nanterre.
sions can be said to express" emotions. Facial expressions, when they
II
Brehm, J. (1994). The intensity of emotions. Unpublished manuscript.
Bühler, K. (1934). Ausdruckstheorie. Jena, Germany: Fisher.
are contingent upon states of action readiness, are intimately bound to
Buytendijk, F. J. J. (1947). De eerste glim/ach van het kind [The infant's first smile].
emotions. Emotions are expressed in facial expressions and lead to such Nijmegen: Dekker en van der Vegt.
expressions when additional conditions are appropriate for state of action Buytendijk, F. J. J., & Plessner, H. (1925). Die Deutung des mimischen Ausdrucks.
readiness to do so. Facial expressions thereby are often fairly diagnostic Philosophische Anzeiger, 1, 26-48.
Camper, P. (1792). Redencoeringen over de Wyze om de onerscheiden Haristochien op
for the prevailing states of action readiness and, with sufficient cues (be- onze Wezens te Verbeelden [Discourses on the manner to depict the various
havior context, time course, information that there is some eliciting passions on our creatures]. Utrecht.
Chiva, M. (1985). Le doux et l'amer [The sweet and the bitter]. Paris: Presses Univ-
event) for emotion.
ersitaires de France.
At the same time, from the point of view of the observer, the issue is Darwin, C. (1872). The expression of emotions in man and animals. London: John
not simple. Facial expressions represent relational activities, activation Murray.
modes, social signals, and inhibitory states. Correctly recognizing emo- Davitz, J. R. (1969). The language of emotion. New York: Academic Press.
II De Meijer, M. (1991). Emotional meaning in large body movements. Tilburg, Neth-
tion needs those additional sufficient cues" just mentioned, which allow
erlands: Tilburg University Press.
going from relational activity to action readiness to correctly assessing Dennett, D. C. (1978). Brainstorms: Philosophical essays on mind and psychology.
whether the action readiness is the response to an appraised event. Montgomery: Bradford Books.
The relationship between facial expressions and emotions, as felt or as Dumas, G. (1933a). L'expression des émotions. In G. Dumas (Ed.), Nouveau traité
de psychologie. Tome 3 (pp. 39-292). Paris: Alean.
defined by eliciting event, type of appraisal, or action readiness, is var- Dumas, G. (1933b). Les mimiques. In G. Dumas (Ed.), Nouveau traité de psychol-
iable. Manifest state of action readiness may be peripheral to what is ogie. Tome 3 (pp. 293-360). Paris: Alean.
central to experience. Occurrence of facial expression depends upon Dumas, G. (1947). La vie affective. Paris: Presses Universitaires de France.
100 N. H. Frijda & A. Tcherkassof Facial expression as action readiness 101
Ekman, P. (1993). Facial expression and emotion. American Psychologist, 48, 384- Kaffka, K. (1935). Principles of Gestalt psychology. New York: Harcourt Brace.
392. Köhler, W. (1929). Gestalt psychology. New York: Liveright.
Ekman, P. (Ed). (1982). Emotion in the human face (2nd ed.). New York: Cambridge Kraut, R. E. (1982). Social presence, social feedback, and emotion. Journal of Per-
University Press. sonality and Social Psychology, 42, 853-863.
Ekman, P., & Friesen, W. V. (1969). The repertoire of nonverbal behavior: Cate- Lang, P. J. (1995). The emotion probe. American Psychologist, 50, 372--385.
gories, origins, usage and coding. Semiotica, 1, 49-98. Lazarus, R. S. (1991). Emotion and adaptation. New York: Oxford University Press.
Ekman, P., & Friesen, W. V. (1975). Unmasking the face. Englewood Cliffs, NJ: Lebrun, C. (1677). Conférences sur l'expression des différents charactères des passions
Prentice-Hall. [Lectures on the expression of the different passions]. Paris.
Ekman, P., Levenson, R. W., & Friesen, W. V. (1983). Autonomic nervous system Lipps, T. (1905). Die Erkenntnis from fremden Ichen [The knowledge of other
activity distinguishing among emotions. Science, 221, 1208-1210. minds]. Psychologische Untersuchungen, 1, 1-53.
Ellgring, H. (1989). Nonverbal communication in depression. Cambridge: Cambridge Lipps, T. (1907). Zur Einfühlung [On empathy]. Psychologische Untersuchungen II,
University Press. 73-102.
Engel, J. J. (1785). Ideen zu einer Mimik. Berlin: Mylin. Mandler, G. (1984). Mind and body: The psychology of emotion and stress. New York:
Fehr, B., & Russell, J. A. (1984). Concept of emotion viewed from a prototype Norton.
perspective. Journal of Experimental Psychology: General 113, 464-486. Matsumoto, D., & Ekman, P. (1989). American-Japanese cultural differences in
Fridlund, A. J. (1994). Human facial expression: An evolutionary view. New York: intensity ratings of facial expressions of emotion. Motivation and Emotion, 13,
Academic Press. 143-157.
Frijda, N. H. (1953). The understanding of facial expression of emotion. Acta Psy- Michette, A. E. (1950). The emotions as functional connections. In M. Reymert
chologica, 9, 294-362. (Ed.), Feelings and emotions (pp. 114-126). New York: McGraw-Hill.
Frijda, N. H. (1956). De betekenis van gelaatsexpressies [The meaning of facial ex- Piderit, T. (1867). Mimik und Physiognomik. Detmoldt: Meyer.
pressions of emotion]. Amsterdam: Van Oorschot. Pribram, K. H., & McGuiness, D. (1975). Arousal, activation, and effort in the
Frijda, N. H. (1969). Recognition of emotion. In L. Berkowitz (Ed.), Advances in control of attention. Psychological Review, 82, 116-149.
experimental social psychology (Vol. 4, pp. 167-223). New York: Academic Riskind, J. H. (1984). They stoop to conquer: Guiding and self-regulatory func-
Press. tions of physical posture after success and failure. Journal of Personality and
Frijda, N. H. (1970). Emotion and recognition of emotion. In M. B. Arnold (Ed.), Social Psychology, 47, 479-493.
Feelings and emotions: The Loyola symposium (pp. 241-250). New York: Aca- Roseman, l. J., Wiest, C., & Swartz, T. S. (1994). Phenomenology, behaviors, and
demic Press. goals differentiate discrete emotions. Journal of Personality and Social Psychol-
Frijda, N. H. (1973). The relation between emotion and expression. In M. von ogy, 67, 206-221.
Cranach & l. Vine (Eds.), Social communication and movement (pp. 325-340). Ruckmick, C. A. (1921). A preliminary study of the emotions. Psychological Mon-
New York: Academic Press. ographs, 30.
Frijda, N. H. (1982). The meanings of facial expression. ln M. Ritchie-Kye (Ed.), Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social
Nonverbal communication today: Current research (pp. 103-120). The Hague: Psychology, 39, 1161-1178.
Mouton. Russell, J. A. (1991). ln defense of a prototype approach to emotion concepts.
Frijda, N. H. (1986). The emotions. Cambridge, England: Cambridge University Journal of Personality and Social Psychology, 60, 425-438.
Press. Russell, J. A. (1994). Is there universal recognition of emotion from facial expres-
Frijda, N. H., Kuipers, P., & Terschure, E. (1989). Relations between emotion, sion? A review of the cross-cultural studies. Psychological Bulletin, 115, 102--
appraisal, and emotional action readiness. Journal of Personality and Social 141.
Psychology, 57, 212--228. Russell, J. A., & Bullock, M. (1986). Fuzzy concepts and the perception of emo-
Frijda, N. H., Markam, S., Sato, K., & Wiers, R. (1995). Emotion and emotion tions in facial expression. Social Cognition, 4, 309-341.
words. ln J. A. Russell, J-M. Fernández-Dols, A. S. R. Manstead, & J. Wellen- Sartre, J. P. (1939). Esquisse d'une theorie phenomenologique des emotions. Paris: Her-
kamp (Eds.), Everyday conceptions of emotion (pp. 121-144). Dordrecht: Klu- mann (The Emotions, New York, Philosophical Library, 1948).
wer. Schänzle, J. (1939). Der mimische Ausdruck des Denkens [The facial expression of
Frijda, N. H., & Philipszoon, E. (1963). Dimensions of recognition of emotion. thinking]. Berlin: Grafe.
Journal of Abnormal and Social Psychology, 66, 45-51. Scherer, K. R. (1984). On the nature and function of emotion: A component pro-
Garotti, P. L., Caterina, R., Brighetti, G., Giberti, E., & Ricci-Bitti, P. E. (1993). Il cess approach. In K. R. Scherer & P. Ekman (Eds.). Approaches to emotion (pp.
ruolo del sorriso come indicatore di stati emotivi differenziati [The role of 293-317). Hillsdale, NJ: Erlbaum.
the smile as indicating differentiated emotional states]. Rassegna di Psicologia, Scherer, K. R. (1992). What does facial expression express? In K. T. Strongman
10, 29-53. (Ed.), International review of studies of emotion (Vol. 2, pp. 139-165). New York:
Copal, R. (1951). Indian dancing. London: Phoenix House. Wiley.
Gray, J. A. (1982). The neuropsychology of anxiety: An enquiry into the functions of Schlosberg, H. (1954). Three dimensions of emotion. Psychological Review, 61, 81-
the septo-hippocampal system. Oxford: Oxford University Press. 88.
Izard, C. E. (1977). Human emotions. New York: Plenum. Schwartz, G. M., Izard, C. E., & Ansul, S. E. (1985). The 5-month-old's ability to
 102 N. H. Frijda & A. Tcherkassof
discriminate facial expressions of emotion. Infant Behavior and Development,
8, 65-77 .
Shaver, P., Schwartz, J., Kirson, D., & O'Connor, C. (1987). Emotion knowledge:
Further exploration of a prototype approach. Journal of Personality and Social 5. The new ethology of human
Behavior, 52, 1061-1086.
Smith, C. A. (1989). Dimensions of appraisal and physiological response to emo- facial expressions
tion. Journal of Personality and Social Psychology, 56, 339-353.
Sroufe, L. A., & Waters, E. (1976). The ontogenesis of smiling and laughter: A
perspective on the organization of development in infancy. Psychological Re- ALAN J. FRIDLUND
view, 83, 173-189.
Tcherkassof, A. (in preparation). La perception des expressions émotionelles faciales.
Unpublished doctoral dissertation, Université de Paris-X Nanterre.
Wagner, H. L., MacDonald, C. J., & Manstead, A. S. R. (1986). Communication of
individual emotions by spontaneous facial expression. Journal of Personality
and Social Psychology, 50, 737-743. Historically, researchers on facial expressions have tried to forge links
Werner, H. (1948). Comparative psychology of mental development. New York: Sci-
ence Editions. (Original work published 1926) between movements of the face and those ineffable states called "emo-
Woodworth, R. S. (1938). Experimental psychology. New York: Holt. tions" (e.g., Mandler, 1984). The advocates of this "Emotions View" are
Woodworth, R. S., & Schlosberg, H. (1954). Experimental psychology (rev. ed.). not homogeneous in all their axioms and precepts, but they share the
New York: Holt.
Wundt, W. (1902). Grundzüge der pysiologischen Psychologie (Vol. 3). Leipzig: En- belief in the centrality of emotion in explaining facial movements. I my-
gelmann, 5th. Ausgabe. self worked within this tradition for many years (e.g., Ekman & Fridlund,
1987; Fridlund, Ekman, & Oster, 1987; Fridlund & Izard, 1983; Fridlund,
Schwartz, & Fowler, 1984; Matsumoto, Ekman, & Fridlund, 1990) but
began to be troubled by certain insurmountable problems with the ap-
proach. My apostacy led to a search for a better way to understand our
facial expressions.
I have proposed an alternative (e.g., Fridlund, 1991a, 1994), termed the
Behavioral Ecology View, because it derives from modern accounts of the ev-
olution - both genetic and cultural - of signaling behavior.1 This account,
based on work by biologists like Maynard Smith, Hinde, Smith, Krebs, Da-
vies, and Marler, contrasts with the Emotions View of faces (see Izard; Fri-
jda & Tcherkassof; Smith & Scott; chapters 3, 4, and 10, respectively, this
volume) in its view of how facial expressions evolved, what they signify,
and how they function in our everyday lives. This chapter presents the
fundamentals of the Behavioral Ecology View, followed by the reasons
why it may afford the better understanding of human facial expressions.

., The Behavioral Ecology View of faces

Most theorists within the Emotions View essentially espouse a two-factor
model, depicted in Figure 5.1, that posits two basic kinds of faces. First
are the innate reflex-like faces that read out ongoing emotion; these are
"facial expressions of emotion." Second are learned, instrumental faces

lín