Hochschild EmotionWorkFeeling 1979

Emotion Work, Feeling Rules, and Social Structure
Author(s): Arlie Russell Hochschild

Source: American Journal of Sociology, Vol. 85, No. 3 (Nov., 1979), pp. 551-575
Published by: The University of Chicago Press
Stable URL: https://www.jstor.org/stable/2778583
Accessed: 17-10-2024 17:03 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
The University of Chicago Press is collaborating with JSTOR to digitize, preserve and
extend access to American Journal of Sociology
This content downloaded from 132.174.250.76 on Thu, 17 Oct 2024 17:03:36 UTC
All use subject to https://about.jstor.org/terms
Emotion Work, Feeling Rules, and Social Structure
Arlie Russell Hochschild
University of California, Berkeley
This essay proposes an emotion-management perspective as a lens

through which to inspect the self, interaction, and structure. Emo-
tion, it is argued, can be and often is subject to acts of management.
The individual often works on inducing or inhibiting feelings so as
to render them "appropriate" to a situation. The emotion-manage-
ment perspective draws on an interactive account of emotion. It dif-
fers from the dramaturgical perspective on the one hand and the
psychoanalytic perspective on the other. It allows us to inspect at
closer range than either of those perspectives the relation among emo-
tive experience, emotion management, feeling rules, and ideology.
Feeling rules are seen as the side of ideology that deals with emotion
and feeling. Emotion management is the type of work it takes to cope
with feeling rules. Meaning-making jobs, more common in the mid-
dle class, put more premium on the individual's capacity to do emo-
tion work. A reexamination of class differences in child rearing sug-
gests that middle-class families prepare their children for emotion
management more and working-class families prepare them less. In
this way each class prepares its children to psychologically reproduce
the class structure.
Social psychology has suffered under the tacit assumption that emotion,
because it seems unbidden and uncontrollable, is not governed by social
rules. Social rules, for their part, are seen as applying to behavior and
thought, but rarely to emotion or feeling. If we reconsider the nature of
emotion2 and the nature of our capacity to try shaping it, we are struck
by the imperial scope of social rules. Significant links emerge among social
structure, feeling rules, emotion management, and emotive experience-links
I try to trace in this essay. The purpose is to suggest an area for inquiry.
I This paper summarizes part of the argument presented in a forthcoming book. This
study has been generously supported by a Guggenheim Fellowship. Although gratitude
in footnotes like this is (as this paper demonstrates) conventionalized, and although
convention makes authenticity hard to decipher, I want anyway to express appreciation
to Harvey Faberman, Todd Gitlin, Adam Hochschild, Robert Jackson, Jerzy Michaelo-
wicz, Caroline Persell, Mike Rogin, Paul Russell, Thomas Scheff, Ann Swidler, Joel
Telles, and the anonymous reviewers for the AIS.
2 I define emotion as bodily cooperation with an image, a thought, a memory-a coop-
eration of which the individual is aware. I will use the terms "emotion" and "feeling"
interchangeably, although the term "emotion" denotes a state of being overcome that
"feeling" does not. The term "emotion management" is used synonymously with "emo-
tion work" and "deep acting."
@) 1979 by The University of Chicago. 0002-9602/80/8503-0003$02.00
AJS Volume 85 Number 3 551
American Journal of Sociology
Why is the emotive experience of normal adults in daily life as orderly

as it is? Why, generally speaking, do people feel gay at parties, sad at
funerals, happy at weddings? This question leads us to examine, not con-
ventions of appearance or outward comportment, but conventions of feeling.
Conventions of feeling become surprising only when we imagine, by con-
trast, what totally unpatterned, unpredictable emotive life might actually
be like at parties, funerals, weddings, and in the family or work life of nor-
mal adults.
Erving Goffman (1961) suggests both the surprise to be explained and
part of the explanation: ". . . We find that participants will hold in check
certain psychological states and attitudes, for after all, the very general
rule that one enter into the prevailing mood in the encounter carries the
understanding that contradictory feelings will be in abeyance.... So gen-
erally, in fact, does one suppress unsuitable affect, that we need to look at
offenses to this rule to be reminded of its usual operation" (Goffman 1961,
p. 23). If we take this passage seriously, as I urge we do, we may be led
back to the classic question of social order from a particular vantage point-
that of emotion management. From this vantage point, rules seem to govern
how people try or try not to feel in ways "appropriate to the situation."
Such a notion suggests bow profoundly the individual is "social," and
"socialized" to try to pay tribute to official definitions of situations, with
no less than their feelings.
Let me pause to point out that there are two possible approaches to the
social ordering of emotive experience. One is to study the social factors
that induce or stimulate primary (i.e., nonreflective, though by definition
conscious) emotions-emotions passively undergone. The other is to study
secondary acts performed upon the ongoing nonreflective stream of primary
emotive experience. The first approach focuses on how social factors affect
what people feel, the second on how social factors affect what people think
and do about what they feel (i.e., acts of assessment and management).
Those who take the first approach might regard those who take the second
as being "overly cognitive." But in fact the two approaches are compatible,
and indeed the second, taken here, relies on some accumulation of knowl-
edge garnered from the first.3
If we take as our object of focus what it is people think or do about feel-
3 For comprehensive, if somewhat dated, reviews of the theoretical approaches to emo-

tion, see Hillman (1964), Carlson et al. (1959), Reymert (1950), and McDougall (1937).
These reviews exclude several more recent theoretical attempts by Tomkins (1962),
Arnold (1968), and Plutchik (1962). For a psychoanalytic account, see Rapaport
(1953). Also see Rabkin (1968), Sprout (1952), Scheff (1973, 1977a, 1977b), Levy
(1973), and Katz (1977). Various social scientists have developed their approaches
via inspection of a particular emotion or feeling, as for example, jealousy (Davis
1936; Clanton and Smith 1977), envy (Foster 1972), embarrassment (Goffman 1956;
Gross and Stone 1964), and love (Goode 1974; Swanson 1965). For a selective review
see Hochschild (1975).
552
Emotion Work
ings, several questions emerge. First, with what assumptions about emotion
and situation do we begin? In other words, (a) how responsive is emotion
to deliberate attempts to suppress or evoke it? (b) What sociological ap-
proach is most fruitful? Second, what are the links among social structure,
ideology, feeling rules, and emotion management? To begin with, (c) are
there feeling rules? (d) How do we know about them? (e) How are these
rules used as baselines in social exchanges? (f) What in the nature of work
and child rearing might account for different ways adults of varying classes
manage their feelings? I shall sketch outlines of possible answers with the
aim, in some measure, of refining the questions.
TWO ACCOUNTS OF EMOTION AND FEELING
In order to address the first question, we might consider two basic accounts
of emotion and feeling found in social psychology: the organismic account
and the interactive account. The two approaches differ in what they imply
about our capacity to manage emotion, and thus in what they imply about
the importance of rules about managing it. I cannot do full justice here to
the question of what emotion is and how it is generated, nor can I offer a
full reaction to the ample literature on that question.
According to the organismic view, the paramount questions concern the
relation of emotion to biologically given "instinct" or "impulse." In large
part, biological factors account for the questions the organismic theorist
poses. The early writings of Sigmund Freud (1911, 1915a, 1915b; see Lof-
gren 1968), Charles Darwin ([1872] 1955), and in some though not all
respects, William James (James and Lange 1922) fit this model.4 The
concept "emotion" refers mainly to strips of experience in which there is
no conflict between one and another aspect of self; the individual "floods
out," is "overcome." The image that comes to mind is that of a sudden,
4 W. McDougall (1948) and, to some extent, S. S. Tomkins (1962) also fit this classi-
fication. Both focus on the relation of emotion to drive or instinct (Tomkins elaborates
a relation between emotion and "drive signals" whereby emotion is said to amplify
drive signals). The central issues on which the two theoretical camps divide are fixity,
reflexivity, and origin. (1) The organismic theorists, unlike their interactive counter-
parts, assume a basic fixity of emotion, based in biological givens. (2) They assume
that social interaction does not basically affect emotions; the social surface remains
what is implied by the term "surface." In the interactive account. this is not the case.
Labeling, management, and expression of feeling (more clearly differentiated by the
interactionists) are processes which can reflexively "act back" on emotion, and indeed
come to constitute what we mean by the term "emotion." (3) Again, the organismic
theorists are more concerned with tracing emotion back to its origins. For Freud and
James the origins were energic or somatic, and for Darwin, phylogenetic. The inter-
active theorists are less concerned with origin than with the interface of a situation
and experience. The focus on origin leads the organismic theorists to focus on com-
monalities between different peoples, and between people and animals. The focus on
social interface leads the interactionists to focus on differences. For recent innovations
in the interactive tradition, see Kemper (1978) and Averill (1976).
553
automatic reflex syndrome-Darwin's instant snarl expression, Freud's ten-

sion discharge at a given breaking point of tension overload, James and
Lange's notion of an instantaneous unmediated visceral reaction to a per-
ceived stimulus, the perception of which is also unmediated by social
influences.
In this first model, social factors can enter in only in regard to how emo-
tions are stimulated and expressed (and even here Darwin took the uni-
versalist position) (see Ekman 1972, 1973). Social factors are not seen as
an influence on how emotions are actively suppressed or evoked. Indeed,
emotion is characterized by the fixity and universality of a knee-jerk re-
action or a sneeze. In this view, one could as easily manage an emotion as
one could manage a knee jerk or a sneeze. If the organismic theorist were
to be presented with the concept of feeling rules, he or she would be hard
put to elucidate what these rules impinge on, or what capacity of the self
could be called on to try to obey a feeling rule (see Hochschild 1977). Re-
cent attempts to link an organismic notion of emotion to social structure,
such as Randall Collins's (1975) wonderfully bold attempt, suffer from the
problems that were implicit in the organismic account to begin with. By
Collins, as by Darwin on whom he draws, emotions are seen as capacities
(or susceptibilities) within a person, to be automatically triggered, as Collins
develops it, by one or another group in control of the ritual apparatus that
does the "triggering" (1975, p. 59). A wholly different avenue of social
control, that of feeling rules, is bypassed, because the individual's capacity
to try to, or try not to feel-that to which the rule applies-is not suggested
by the organismic model with which Collins begins.
In the interactive account, social influences permeate emotion more in-
sistently, more effectively, and at more theoretically posited iunctures. In
large part, sociopsychological factors account for the questions the inter-
active theorist poses. The writings of Gerth and MIills (1964), Goffman
(1956, 1959, 1961, 1967, 1974), Lazarus (1966), La7arus and Averill
(1972), Schachter and Singer (1962), Schachter (1964), Kemper (1978),
Averill (1976). and aspects of late Freudian and neo-Freudian thought fit
this model. To invoke the Freudian vocabulary, the image here is not that
of a "runaway id," but of an ego and superego, acting upon, shaping, nag-
ging, however ineffectively, temporarily, or consciously, the id.5 Emotion
is sometimes posited as a psychobiological means of adaptation-an ana-
5 The stress on will (included in the concept of ego, but not all that "earo" refers to)
is not a clean divider of the organismic from the interactive theorist. Schachter and
Gerth and Mills, whom I see as members of the interactive camp, lay no particular
stress on volition. Goffman stresses the phenomena that call tacitly for will. He stresses
the patterned results of it, but provides no theoretical account of will itself. He posits
no actor qua emotion manager who might accomplish the acts that, by inference, must
get accomplished to pull off the encounters he describes so well. In my view, we must
reinstitute a self capable of experiencing emotion and of working on it in socially
patterned ways. (On the issue of will, see Piaget in Campbell [1976]; Solomon [1973].)
554
Emotion Work
logue to other adaptive mechanisms, such as shivering when cold or per-

spiring when hot. But emotion differs from these other adaptive mechanisms,
in that thinking, perceiving, and imagining--themselves subject to the in-
fluence of norms and situations-are intrinsically involved.
As in the first model, social factors affect how emotions are elicited and
expressed. However, in addition, social factors guide the microactions of
labeling (Schachter 1964; Schachter and Singer 1962; Katz 1977), in-
terpreting (Gerth and Mills 1964), and managing emotion (Lazarus 1966).
These microactions, in turn, reflect back on that which is labeled, inter-
preted, and managed. They are, finally, intrinsic to what we call "emotion"
(see Schafer 1976). Emotion, in this second school of thought, is seen as
more deeply social. Lazarus's (1966) work in particular lends empirical
weight to the interactive model. It suggests that normal adults, such as
university students on whom he conducted experiments, have a considerable
capacity to control emotion. It is more control than one might expect from
a small child, an insane adult, or an animal, from all of which Freud in his
earlier writings and Darwin drew inspiration. But since it is the emotive
experience of normal adults we seek to understand, we would do well to
begin with the interactive account.6
The Interactive Account of Emotion and Social Psychology
If emotions and feelings can to some degree be managed, how might we get
a conceptual grasp of the managing act from a social perspective? The in-
teractive account of emotion leads us into a conceptual arena "between"
the Goffmanian focus on consciously designed appearances on the one hand
and the Freudian focus on unconscious intrapsychic events on the other.
The focus of Mead (1934) and Blumer (1969) on conscious, active, and
responsive gestures might have been most fruitful had not their focus on
deeds and thought almost entirely obscured the importance of feeling. (See
Shott [1979] for an attempt to consider emotion from a symbolic inter-
actionist perspective.) The self as emotion manager is an idea that borrows
from both sides-Goffman and Freud-but squares completely with neither.
Here I can only sketch out a few basic borrowings and departures, focusing
on the departures.
Erving Goffman
Goffman guides our attention to social patterns in emotive experience. He

catches an irony: moment to moment, the individual is actively negotiating
6 My own account of emotion draws on that proposed by Katz (1977). For Katz emo-
tion is generated by a "schematic discrepancy," that is, a discrepancy between the
individual's schemata and his current perception, memory, or imagining of an event
or object. Also see the interesting work of Lazarus, Kanner, and Folkman (1979).
555
a course of action, but in the long run, all the action seems like passive
acquiescence to social convention. The conserving of convention is not a
passive business. Goffman's approach might simply be extended and deep-
ened by showing that people not only try to conform outwardly, but do so
inwardly as well. "When they issue uniforms, they issue skins" (Goffman
1974) could be extended: "and two inches of flesh."
Yet, ironically, to study why and under what conditions "participants . . .
hold in check certain psychological states . . ." (Goffman 1961, p. 23), we
are forced partly out of the perspective which gave birth to the insight. I
shall try to show why this is so, what the remedies might be, and how the
results could be conceptually related to aspects of the psychoanalytic
tradition.
First, Goffman, for reasons necessary to his purpose, maintains for the
most part a studied disregard for the links between immediate social situa-
tions and macrostructure on the one hand, and individual personality on
the other. If one is interested in drawing links among social structure, feeling
rules, and emotion management, this studied disregard becomes a problem.
Goffman's "situationism" is a brilliant achievement, one that must be
understood as a development in the intellectual history of social psychology.
Earlier in the century a number of classic works linked social structure to
personality, or "dominant institutions" to "typical identities," and thus
also related findings in sociology and anthropology to those in psychology
or psychoanalytic theory. These studies appeared in a number of fields-in
anthropology, Ruth Benedict (1946); in psychoanalysis, Erich Fromm
(1942), Karen Horney (1937), and Erik Erikson (1950); in sociology,
David Riesman (1952, 1960), Swanson and Miller (1966), and Gerth and
Mills (1964).
Possibly in response to this paradigm Goffman proposed an intermediate
level of conceptual elaboration, "between" social structure and personality.
His focus is on situations, episodes, encounters. The situation, the episodi-
cally emergent encounter, is not only nearly divorced from social structure
and from personality; he even seems to intend his situationism as an analytic
substitute for these concepts (see Goffman 1976, p. 77). Structure, he
seems to say, can be not only transposed but reduced "in and down," while
personality can be reduced "up and out" to the study of here-now, gone-then
interactional moments.
Each interactional episode takes on the character of a minigovernment.
A card game, a party, a greeting on the street exacts from us certain "taxes"
in the form of appearances which we "pay" for the sake of sustaining the
encounter. We are repaid in the currency of safety from disrepute. (Thanks
to Harvey Farberman for discussion on this point.)
This model of the situation qua minigovernment, while useful for Goff-
man's purposes, leads us away from social structure and personality-two
556
Emotion Work
concepts with which any study of feeling rules and emotion management
would be wise to deal.7 To study why and under what conditions "partici-
pants ... hold in check certain psychological states . . ." (Goffman 1961,
p. 23), we are forced out of the here-now, gone-then situationism and back,
in part at least, to the social structure and personality model. We are led
to appreciate the importance of Goffman's work, as it seems he does not,
as the critical set of conceptual connecting tissues by which structure and
personality, real in their own right, are more precisely joined.
Specifically, if we are to understand the origin and causes of change in
"feeling rules"-this underside of ideology-we are forced back out of a
study of the immediate situations in which they show up, to a study of such
things as changing relations between classes or the sexes.
If we are to investigate the ways people try to manage feeling, we
shall have to posit an actor capable of feeling, capable of assessing when
a feeling is "inappropriate," and capable of trying to manage feeling. The
problem is that the actor Goffman proposes does not seem to feel much, is
not attuned to, does not monitor closely or assess, does not actively evoke,
inhibit, shape-in a word, work on feelings in a way an actor would have
to do to accomplish what Goffman says is, in fact, accomplished in one
encounter after another. We are left knowing about "suppressive work" as
a final result, but knowing nothing of the process or techniques by which
it is achieved. If we are to argue that social factors influence how we try
to manage feelings, if we are to carry the social that far, we shall have to
carry our analytic focus beyond the "black box" to which Goffman ulti-
mately refers us.
Goffman's actors actively manage outer impressions, but they do not
actively manage inner feelings. For example, a typical Goffmanian actor,
Preedy at the beach (Goffman 1959), is exquisitely attuned to outward
appearance, but his glances inward at subjective feeling are fleeting
and blurred. The very topic, sociology of emotion, presupposes a hu-
man capacity for, if not the actual habit of, reflecting on and shaping
inner feelings, a habit itself distributed variously across time, age, class,
and locale. This variation would drop from sight were we to adopt an ex-
7Time: to link the momentary act of emotion work with the concept of personality,
we must alter our perspective on time. An emotive episode and the attempt to shape
it is, after all, a brief strip of time. Situations such as Goffman studies are also short.
The focus is on the act, and the act ends, so to speak, when the theater closes and
starts again when it reopens. If we extend Goffman's analysis, by speaking now of
"deep" acting, we, like him, are focusing on short episodes, on "stills" from which
long movies are composed. The notion of personality implies a fairly durable, trans-
situational pattern. The Casper Milquetoast personality may lead an anxiety avoidant
life of 73 years. Not momentary stills, but many decades are at issue. Again, we
must shift our situationist focus at the structural end when we come to speak of
institutions, which live even longer than people do.
557
clusive focus on the actor's attentiveness to behavioral facade and assume

a uniform passivity vis-a-vis feelings.
This skew in the theoretical actor is related to what from my viewpoint
is another problem: Goffman's concept of acting. Goffman suggests that we
spend a good deal of effort managing impressions-that is, acting. He posits
only one sort of acting-the direct management of behavioral expression.
His illustrations, though, actually point to two types of acting-the direct
management of behavioral expression (e.g., the given-off sigh, the shoulder
shrug), and the management of feeling from which expression can follow
(e.g., the thought of some hopeless project). An actor playing the part
of King Lear might go about his task in two ways. One actor, following
the English school of acting, might focus on outward demeanor, the con-
stellation of minute expressions that correspond to Lear's sense of fear
and impotent outrage. This is the sort of acting Goffman theorizes about.
Another actor, adhering to the American or Stanislavsky school of acting,
might guide his memories and feelings in such a way as to elicit the cor-
responding expressions. The first technique we might call "surface acting,"
the second "deep acting." Goffman fails to distinguish the first from the
second, and he obscures the importance of "deep acting." Obscuring this,
we are left with the impression that social factors pervade only the "social
skin," the tried-for outer appearances of the individual. We are left under-
estimating the power of the social.
In sum, if we are to accept the interactive account of emotion and to
study the self as emotion manager, we can learn from Goffman about the
link between social rule and feeling. But to elaborate this insight we might
well selectively relax the theoretical strictures Goffman has stoically im-
posed against a focus on social structure and on personality.
Freud
The need to replace Goffman's "black-box psychology" with some theory

of self, in the full sense of the term, might seem to lead to Freudian or neo-
Freudian theory. Yet, here, as with Goffman, only some aspects of the
Freudian model seem useful to my understanding of conscious, deliberate
efforts to suppress or evoke feeling. I shall briefly discuss psychoanalytic
theory to show some points of departure.
Freud, of course, dealt with emotions, but for him they were always
secondary to drive. He proposed a general theory of sexual and aggressive
drives. Anxiety, as a derivative of aggressive and sexual drives, was of
paramount importance, while a wide range of other emotions, including
joy, jealousy, depression, were given relatively little attention. He devel-
oped, and many others have since elaborated, the concept of ego defenses
as generally unconscious, involuntary means of avoiding painful or un-
558
Emotion Work
pleasant affect. Finally the notion of "inappropriate affect" is used to point

to aspects of the individual's ego functioning and not used to point to the
social rules according to which a feeling is or is not deemed appropriate to
a situation.
The emotion-management perspective is indebted to Freud for the general
notion of what resources individuals of different sorts possess for accom-
plishing the task of emotion work (as I have defined it) and for the notion
of unconscious involuntary emotion management. The emotion-management
perspective differs from the Freudian model in its focus on the full range of
emotions and feelings and its focus on conscious and deliberate efforts to
shape feeling. From this perspective, we note too that "inappropriate emo-
tion" has a clearly important social as well as intrapsychic side.
Let me briefly illustrate the differences between the two perspectives. In
Shapiro's well-known work on "neurotic style," he gives an example:
An obsessive-compulsive patient-a sober, technically minded and active

man-was usually conspicuouisly lacking in enthusiasm or excitement in
circumstances that might seem to warrant them. On one occasion, as he
talked about a certain prospect of his, namely, the good chance of an im-
portant success in his work, his sober expression was momentarily inter-
rupted by a smile. After a few more minutes of talking, during which he
maintained his soberness only with difficulty, he began quite hesitantly to
speak of certain hopes that he had only alluded to earlier. Then he broke
into a grin. Almost immediately, however, he regained his usual somewhat
worried expression. As he did this, he said, "Of course, the outcome is by
no means certain," and he said this in a tone that, if anything, would sug-
gest the outcome was almost certain to be a failure. After ticking off
several of the specific possibilities for a hitch, he finally seemed to be
himself again, so to speak. [Shapiro 1965, p. 192, emphasis mine]
What seems interesting about this example differs according to whether

one takes the psychiatric perspective or the emotion-management per-
spective. First, to the psychiatrist in the case above, what circumstances
warrant what degree and type of feeling seems relatively unproblematic.
A doctor "knows" what inappropriate affect is; the main problem is not
so much to discern misfits of feeling to situation but to explain them
and to cure the patient of them. From the emotion-management per-
spective, on the other hand, the warranting function of circumstances
is problematic. Further, the means used to assess this warranting function
may well be the same for a psychiatrist as for a salesclerk or school
disciplinarian. For, in a sense we all act as lay psychiatrists using un-
examined means of arriving at a determination about just "what" circum-
stances warrant "that much" feeling of "that sort."
What the psychiatrist, the salesclerk, and the school disciplinarian may
share is a habit of comparing, situation (e.g., high opportunity, associated
with an accomplishment at work) with role (e.g., hopes, aspirations, ex-
559
pectations typical of, and expected from, those enacting the role). Social
factors can enter in, to alter how we expect a role to be held, or played.
If, for example, the patient were a "sober, technically minded and active"
woman and if the observer (rightly or wrongly) assumed or expected her
to value family and personal ties over worldly success, less enthusiasm at
the prospect of advance might seem perfectly "appropriate." Lack of en-
thusiasm would have a warrant of that social sort. Again, if the patient
was an antinuclear activist and his discovery had implications for nuclear
energy, that would alter the hopes and aspirations he might be expected
to have at work and might warrant dismay, not enthusiasm.
We assess the "appropriateness" of a feeling by making a comparison
between feeling and situation, not by examining the feeling in abstracto.
This comparison lends the assessor a "normal" yardstick-a socially
normal one-from which to factor ouit the personal meaning systems
which may lead a worker to distort his view of "the" situation and feel
inappropriately with regard to it. The psychiatrist holds constant the
socially normal benchmark and focuses on what we have just factored out.
The student of emotion management holds constant what is factored out
and studies the socially normal benchmark, especially variations in it.
There is a second difference in what, from the two perspectives, seems
interesting in the above example. From the emotion-management perspec-
tive, what is interesting is the character and direction of volition and con-
sciousness. From the psychiatric perspective, what is of more interest is
pre-will and nonconsciousness. The man above is not doing emotion work,
that is, making a conscious, intended try at altering feeling. Instead he is
controlling his enthusiasm by "being himself," by holding, in Schutz's
term, a "natural attitude." He "no longer needs to struggle not to grin;
he is not in a grinning mood" (Shapiro 1965, p. 164). In order to avoid
affective deviance, some individuals may face a harder task than do others,
the task of consciously working on feelings in order to make up for "a nat-
ural attitude"-explanable in psychoanalytic terms-that gets them in
social trouble. The hysteric working in a bureaucratic setting may face the
necessity for more emotion work than the obsessive compulsive who fits in
more naturally.
In sum, the emotion-management perspective fosters attention to how
people try to feel, not, as for Goffman, how people try to appear to feel.
It leads us to attend to how people consciously feel and not, as for Freud,
how people feel unconsciously. The interactive account of emotion points
to alternate theoretical junctures-between consciousness of feeling and
consciousness of feeling rules, between feeling rules and emotion work,
between feeling rules and social structure. In the remainder of this essay,
it is these junctures we shall explore.
560
Emotion Work
EMOTION WORK
By "emotion work" I refer to the act of trying to change in degree or quality

an emotion or feeling. To "work on" an emotion or feeling is, for our pur-
poses, the same as "to manage" an emotion or to do "deep acting." Note
that "emotion work" refers to the effort-the act of trying- and not to the
outcome, which may or may not be successful. Failed acts of management
still indicate what ideal formulations guide the effort, and on that account
are no less interesting than emotion management that works.
The very notion of an attempt suggests an active stance vis-'a-vis feeling.
In my exploratory study respondents characterized their emotion work by
a variety of active verb forms; "I psyched myself up.... I squashed my
anger down. . . . I tried hard not to feel disappointed. . . . I made myself
have a good time. . . . I tried to feel grateful. . . . I killed the hope I had
burning." There was also the actively passive form, as in, "I let myself
finally feel sad."
Emotion work differs from emotion "control" or "suppression." The
latter two terms suggest an effort merely to stifle or prevent feeling. "Emo-
tion work" refers more broadly to the act of evoking or shaping, as well as
suppressing, feeling in onself. I avoid the term "manipulate" because it
suggests a shallowness I do not mean to imply. We can speak, then, of two
broad types of emotion work: evocation, in which the cognitive focus is on
a desired feeling which is initially absent, and suppression, in which the
cognitive focus is on an undesired feeling which is initially present. One
respondent, going out with a priest 20 years her senior, exemplifies the
problems of evocative emotion work: "Anyway, I started to try and make
myself like him. I made myself focus on the way he talked, certain things
he'd done in the past.... When I was with him I did like him but I would
go home and write in my journal how much I couldn't stand him. I kept
changing my feeling and actually thought I really liked him wlhile I was
with him but a couple of hours after he was gone, I reverted back to different
feelings...."8 Another respondent exemplifies the work, not of working
feeling up, but of working feeling down:
Last summer I was going with a guy often, and I began to feel very
strongly about him. I knew though, that he had just broken up with a
8 The illustrations of emotion work come from a content analysis of 261 protocols
given to students in two classes at the University of California, Berkeley, in 1974.
Many of the illustrations come from answers to the question, "Describe as fully a,nd
concretely as possible a real situation, important to you, in which you experienced
either changing a real situation to fit your feelings or changing your feelings to fit a
situation. What did it mean to you?" Three coders coded the protocols. The findings
will be reported in a later study. I will only mention here that 13% of the men but
32% of the women were coded as "changing feeling" instead of changing situation,
and of those who changed feelings, far more women reported doing so agentially
rather than passively.
561
girl a year ago because she had gotten too serious about him, so I was
afraid to show any emotion. I also was afraid of being hurt, so I attempted
to change my feelings. I talked myself into not caring about Mike . . .
but I must admit it didn't work for long. To sustain this feeling I had to
almost invent bad things about him and concentrate on them or continue
to tell myself he didn't care. It was a hardening of emotions, I'd say. It
took- a lot of work and was unpleasant, because I had to concentrate on
anything I could find that was irritating about him.
Often emotion work is aided by setting up an emotion-work system, for

example, telling friends of all the worst faults of the person one wanted to
fall out of love with, and then going to those friends for reinforcement of
this view of the ex-beloved. This suggests another point: emotion work can
be done by the self upon the self, by the self upon others, and by others
upon oneself.
In each case the individual is conscious of a moment of "pinch," or dis-
crepancy, between what one does feel and what one wants to feel (which
is, in turn, affected by what one thinks one ought to feel in such a situation).
In response, the individual may try to eliminate the pinch by working on
feeling. Both the sense of discrepancy and the response to it can vary in
time. The managing act, for example, can be a five-minute stopgap measure,
or it can be a more long-range gradual effort suggested by the term "working
through."
There are various techniques of emotion work. One is cognitive: the at-
tempt to change images, ideas; or thoughts in the service of changing the
feelings associated with them.9 A second is bodily: the attempt to change
somatic- or other physical symptoms of emotion (e.g., trying to breathe slow-
er, trying not to shake). Third, there is expressive emotion work: trying
to change expressive gestures in the service of changing inner feeling (e.g.,
trying to smile, or to cry). This differs from simple display in that it is
directed toward change in feeling. It differs from bodily emotion work in
that the individual tries to alter or shape one or another of the classic public
channels for the expression of feeling.
These three techniques are distinct theoretically, but they often, of course,
go together in practice. For example:
9 There may be various types of cognitive emotion work. All can be described as
attempts to recodify a situation. By recodification I mean reclassification of a situation
into what are previously established mental categories of situations. As in an initial,
more automatic codification of a situation, deliberate recodification means asking onself
(a) What category in my classification schema of situations fits this new situation?
(the schema may include blame-in situations, blame-out situations, credit-in situations,
credit-out situations, etc.), and (b) What category in my classification schema of emo-
tions fits the emotion I'm feeling right now? (i.e., is it anger, general anxiety, disap-
pointment?). In deliberate recodifications, one tries to change the classification of
outward and inward reality. (To translate this idea into Lazarus's framework, we
might speak of the individual trying consciously to alter his or her appraisal of a
situation so as to change the coping process [Lazarus 1966].)
562
Emotion Work
I was a star halfback in high school. Before games I didn't feel the up-
surge of adrenalin-in a word I wasn't "psyched up." (This was due to
emotional difficulties I was experiencing and still experience-I was also
an A student whose grades were dropping.) Having been in the past a
fanatical, emotional, intense player, a "hitter" recognized by coaches as a
very hard worker and a player with "desire," this was very upsetting. I
did everything I could to get myself "uip." I would try to be outwardly
"rah rah" or get myself scared of my opponents-anything to get the
adrenalin flowing. I tried to look nervous and intense before games, so at
least the coaches wouldn't catch on.... When actually I was mostly bored,
or in any event, not "up." I recall before one game wishing I was in the
stands watching my cousin play for his school, rather than "out here."
Emotion work becomes an object of awareness most often, perhaps, when

the individual's feelings do not fit the situation, that is, when the latter
does not account for or legitimate feelings in the situation. A situation (such
as a funeral) often carries with it a proper definition of itself ("this is a
time of facing loss"). This official frame carries with it a sense of what it is
fitting to feel (sadness). It is when this tripartite consistency among situa-
tion, conventional frame, and feeling is somehow ruptured, as when the
bereaved feels an irrepressible desire to laugh delightedly at the thought
of an inheritance, that rule and management come into focus. It is then that
the more normal flow of deep convention-the more normal fusion of situa-
tion, frame, and feeling-seems like an accomplishment.
The smoothly warm airline hostess, the ever-cheerful secretary, the un-
irritated complaint clerk, the undisgusted proctologist, the teacher who
likes every student equally, and Goffman's unflappable poker player may
all have to engage in deep acting, an acting that goes well beyond the mere
ordering of display. Work to make feeling and frame consistent with situa-
tion is work in which individuals continually and privately engage. But they
do so in obeisance to rules not completely of their own making.
FEELING RULES
We feel. We try to feel. We want to try to feel. The social guidelines that
direct how we want to try to feel may be describable as a set of socially
shared, albeit often latent (not thought about unless probed at), rules. In
what way, we may ask, are these rules themselves known and how are they
developed ?10
10 That we can single out such a thing as "feeling rules" is itself a commentary on
the ironic posture of the self legitimated in modern culture. Modern urban cultures
foster much more distance (the stance of the observing ego) from feeling than do
traditional cultures. Jerzy Michaelowicz, a graduate student at the University of
California, San Diego, observed that traditional and tight-knit subcultures put people
directly inside the framework of feeling rules and remove ironic distance or sense
of choice about them. He reported on some research in which one Hassidic rabbi was
asked, "Did you feel happy at the Passover ceremony?" "Of course!" came the in-
credulous reply.
563
To begin with, let us consider several common forms of evidence for

feeling rules. In common parlance, we often talk about our feelings or those
of others as if rights and duties applied directly to them. For example, we
often speak of "having the right" to feel angry at someone. Or we say we
"should feel more grateful" to a benefactor. We chide ourselves that a
friend's misfortune, a relative's death, "should have hit us harder," or that
another's good luck, or our own, should have inspired more joy. We know
feeling rules, too, from how others react to what they infer from our emotive
display. Another may say to us, "You shouldn't feel so guilty; it wasn't
your fault," or "You don't have a right to feel jealous, given our agree-
ment." Another may simply declare an opinion as to the fit of feeling to
situation, or may cast a claim upon our managerial stance, presupposing
this opinion. Others may question or call for an account of a particular
feeling in a situation, whereas they do not ask for an accounting of some
other situated feeling (Lyman and Scott 1970). Claims and callings for an
account can be seen as rule reminders. At other times, a person may, in
addition, chide, tease, cajole, scold, shun-in a word, sanction us for "mis-
feeling." Such sanctions are a clue to the rules they are meant to enforce.
Rights and duties set out the proprieties as to the extent (one can feel
"too" angry or "not angry enough"), the direction (one can feel sad when
one should feel happy), and the duration of a feeling, given the situation
against which it is set. These rights and duties of feeling are a clue to the
depth of social convention, to one final reach of social control.
There is a distinction, in theory at least, between a feeling rule as it is
known by our sense of what we can expect to feel in a given situation, and
a rule as it is known by our sense of what we should feel in that situation.
For example, one may realistically expect (knowing oneself and one's neigh-
bor's parties) to feel bored at a large New Year's Eve party and at the
same time acknowledge that it would be more fitting to feel exuberant. How-
ever, "expect to feel" and "should ideally feel" often coincide, as below:
Marriage, chaos, unreal, completely different in many ways than I imag-

ined. Unfortunately we rehearsed the morning of our wedding at eight
o'clock. The wedding was to be at eleven o'clock. It wasn't like I thought
(everyone would know what to do). They didn't. That made me nervous.
My sister didn't help me get dressed or flatter me (nor did anyone in the
dressing room until I asked them). I was depressed. I wanted to be so
happy on our wedding day. I never dreamed how anyone would cry at their
wedding. A wedding is "the happy day" of one's life. I couldn't believe
that some of my best friends couldn't make it to my wedding and that
added to a lot of little things. So I started out to the church and all these
things that I always thought would not happen at my wedding went through
my mind. I broke down-I cried going down. "Be happy" I told myself.
Think of the friends, and relatives that are present. (But I finally said to
myself, "Hey people aren't getting married, you are. It's for Rich [my
husband] and you.") From down the pretty long aisle we looked at each
564
Emotion Work
other's eyes. His love for me changed my whole being. From that point
on we joined arms. I was relieved and the tension was gone. In one sense
it meant misery-but in the true sense of two people in love and wanting
to share life-it meant the world to me. It was beautiful. It was inde-
scribable.
In any given situation, we often invest what we expect to feel with

idealization. To a remarkable extent these idealizations vary socially. If
the "old-fashioned bride" above anticipates a "right" to feel jealous at any
possible future infidelity, the young "flower child" below rejects just this
right.
. . .when I was living down south, I was involved with a group of people,
friends. We used to spend most evenings after work or school together.
We used to do a lot of drugs, acid, coke or just smoke dope and we had
this philosophy that we were very communal and did our best to share
everything-clothes, money, food, and so on. I was involved with this one
man-and thought I was "in love" with him. He in turn had told me that
I was very important to him. Anyway, this one woman who was a very
good friend of mine at one time and this man started having a sexual
relationship, supposedly without my knowledge. I knew though and had a
lot of mixed feelings about it. I thought, intellectually, that I had no claim
to the man, and believed in fact that no one should ever try to own an-
other person. I believed also that it was none of my business and I had no
reason to worry about their relationship together, for it had nothing really
to do with my friendship with either of them. I also believed in sharing.
But I was horribly hurt, alone and lonely, depressed and I couldn't shake
the depression and on top of those feelings I felt guilty for having those
possessively jealous feelings. And so I would continue going out with these
people every night, and try to suppress my feelings. My ego was shattered.
I got to the point where I couldn't even laugh around them. So finally I
confronted my friends and left for the summer and traveled with a new
friend. I realized later what a heavy situation it was, and it took me a
long time to get myself together and feel whole again.
Whether the convention calls for trying joyfully to possess, or trying casual-
ly not to, the individual compares and measures experience against an
expectation often idealized. It is left for motivation ("what I want to feel")
to mediate between feeling rule ("what I should feel") and emotion work
("what I try to feel"). Some of the time many of us can live with a certain
dissonance between "ought" and "want," or between "want" and "try to."
But the attempts to reduce emotive dissonance are our periodic clues to
rules of feeling.
A feeling rule shares some formal properties with other sorts of rules,
such as rules of etiquette, rules of bodily comportment, and those of social
interaction in general (Goffman 1961). A feeling rule is like these other
kinds of rules in the following ways: It delineates a zone within which one
has permission to be free of worry, guilt, or shame with regard to the sit-
uated feeling. Such zoning ordinances describe a metaphoric floor and ceil-
565
ing, there being room for motion and play between the two. Like other rules,
feeling rules can be obeyed halfheartedly or boldly broken, the latter at
varying costs. A feeling rule can be in varying proportions external or in-
ternal. Feeling rules differ curiously from other types of rules in that they
do not apply to action but to what is often taken as a precursor to action.
Therefore they tend to be latent and resistant to formal codification.
Feeling rules reflect patterns of social membership. Some rules may be
nearly universal, such as the rule that one should not enjoy killing or wit-
nessing the killing of a human being, including oneself.1" Other rules are
unique to particular social groups and can be used to distinguish among
them as alternate governments or colonizers of individual internal events.
FRAMING RULES AND FEELING RULES: ISSUES IN IDEOLOGY
Rules for managing feeling are implicit in any ideological stance; they are
the "bottom side" of ideology. Ideology has often been construed as a flatly
cognitive framework, lacking systematic implications for how we manage
feelings, or, indeed, for how we feel. Yet, drawing on Durkheim (1961),12
Geertz (1964), and in part on Goffman (1974), we can think of ideology as
an interpretive framework that can be described in terms of framing rules
and feeling rules. By "framing rules" I refer to the rules according to which
we ascribe definitions or meanings to situations. For example, an individual
can define the situation of getting fired as yet another instance of capitalists'
abuse of workers or as yet another result of personal failure. In each case,
the frame may reflect a more general rule about assigning blame. By "feeling
rules" I refer to guidelines for the assessment of fits and misfits between
feeling and situation. For example, according to one feeling rule, one can
be legitimately angry at the boss or company; according to another, one
cannot. Framing and feeling rules are back to back and mutually imply each
other.
11 But this, too. seems to be culturally variable. Erving Goffman points out that hang-
ings in the 16th century were a social event that the participant was "supposed to
enjoy," a rule that has since disappeared in civilian society.
12 Durkheim, in The Elementary Forms of Religious Life, conveys just this under-
standing of the relation of world view to feeling rules: "When the Christian, during
the ceremonies commemorating the Passion, and the Jew, on the anniversary of the
fall of Jerusalem, fast and mortify themselves, it is not in giving way to a sadness
which they feel spontaneously. Under these circumstances, the internal state of the
believer is out of all proportion to the severe abstinences to which they submit them-
selves. If he is sad, it is primarily because he consents to being sad. And he consents to
it in order to affirm his faith" (Durkheim 1961, p. 224). Again, "An individual . . . if
he is strongly attached to the society of which he is a member, feels that he is morally
held to participating in its sorrows and joys; not to be interested in them would be
equivalent to breaking the bonds uniting him to the group; it would be renouncing
all desire for it and contradicting himself" (1961, p. 446, emphases mine).
566
Emotion Work
It follows that when an individual changes an ideological stance, he or

she drops old rules and assumes new ones for reacting to situations, cogni-
tively and emotively. A sense of rights and duties applied to feelings in
situations is also changed. One uses emotion sanctions differently and ac-
cepts different sanctioning from others. For example, feeling rules in Ameri-
can society have differed for men and women because of the assumption
that their natures differ basically. The feminist movement brings with it a
new set of rules for framing the work and family life of men and women:
the same balance of priorities in work and family now ideally applies to
men as to women. This carries with it implications for feeling. A woman
can now as legitimately (as a man) become angry (rather than simply upset
or disappointed) over abuses at work, since her heart is supposed to be in
that work and she has the right to hope, as much as a man would, for ad-
vancement. Or, a man has the right to feel angry at the loss of custody if he
has shown himself the fitter parent. "Old-fashioned" feelings are now as
subject to new chidings and cajolings as are "old-fashioned" perspectives
on the same array of situations.
One can defy an ideological stance not simply by maintaining an alterna-
tive frame on a situation but by maintaining an alternative set of feeling
rights and obligations. One can defy an ideological stance by inappropriate
affect and by refusing to perform the emotion management necessary to feel
what, according to the official frame, it would seem fitting to feel. Deep
acting or emotion work, then, can be a form of obeisance to a given ideologi-
cal stance, lax emotion management a clue to an ideology lapsed or rejected.
As some ideologies gain acceptance and others dwindle, contending sets
of feeling rules rise and fall.13 Sets of feeling rules contend for a place in
people's minds as a governing standard with which to compare the actual
lived experience of, say, the senior prom, the abortion, the wedding, the
birth, the first job, the first layoff, the divorce. What we call "the changing
climate of opinion" partly involves a changed framing of the "same" sorts
of events. For example, each of two mothers may feel guilty about leaving
her small child at day care while working all day. One mother, a feminist,
may feel that she should not feel as guilty as she does. The second, a tradi-
tionalist, may feel that she should feel more guilty than, in fact, she does
feel.
Part of what we refer to as the psychological effects of "rapid social

change," or "unrest," is a change in the relation of feeling rule to feeling
and a lack of clarity about what the rule actually is, owing to conflicts and
13 Collins suggests that ideology functions as a weapon in the conflict between contend-
ing elites. Groups contend not only for access to the means of economic production
or the means of violence but also for access to the means of "emotion production"
(1975, p. 59). Rituals are seen as useful tools for forging emotional solidarity (that
can be used against others) and for setting up status hierarchies (that can dominate
those who find that the new ideals have denigrating effects on themselves).
567
contradictions between contending sets of rules. Feelings and frames are

deconventionalized, but not yet reconventionalized. We may, like the mar-
ginal man, say, "I don't know how I should feel."
It remains to note that ideologies can function, as Randall Collins rightly
notes ( 1975), as weapons in the conflict between contending elites and social
strata. Collins suggests that elites try to gain access to the emotive life of
adherents by gaining legitimate access to ritual, which for him is a form of
emotive technology. Developing his view, we can add that elites, and indeed
social groups in general, struggle to assert the legitimacy of their framing
rules and their feeling rules. Not simply the evocation of emotion but
laws governing it can become, in varying degrees, the arena of political
struggle.
FEELING RULES AND SOCIAL EXCHANGE
The seemingly static links among ideology, feeling rules, and emotion man-
agement come alive in the process of social exchange. Students of social
interaction have meant two things by the term "social exchange." Some
have referred to the exchange of goods and services between people (Blau
1964; Simpson 1972; Singelmann 1972). Others (G. H. Mead) have re-
ferred to an exchange of gestures, without the cost-benefit accounting re-
ferred to in the first usage. Yet acts of display, too, may be considered "ex-
changed" in the limited sense that the individual very often feels that a
gesture is owed to oneself or another. I refer, then, to exchange of acts of
display based on a prior, shared understanding of patterned entitlement.
Any gesture-a cool greeting, an appreciative laugh, the apology for an
outburst-is measured against a prior sense of what is reasonably owed
another, given the sort of bond involved. Against this background mea-
sure, some gestures will seem more than ample, others less.
The exchange of gestures has, in turn, two aspects; it is an exchange of
display acts (Goffman 1969, 1967)-that is, of surface acting-and also an
exchange of emotion work-that is, of deep acting. In either case, rules
(display rules or feeling rules), once agreed upon, establish the worth of a
gesture and are thus used in social exchange as a medium of exchange. Feel-
ing rules establish the basis of worth to be ascribed to a range of gestures,
including emotion work. Emotion work is a gesture in a social exchange;
it has a function there and is not to be understood merely as a facet of
personality.
There seem to be two ways in which feeling rules come into play in social
exchange. In the first, the individual takes the "owed" feeling to heart,
takes it seriously. For example, a young woman on the eve of her college
graduation felt anxious and depressed but thought that she "ought to feel
happy," and that she "owed this happiness" to her parents for making
568
Emotion Work
her graduation possible. The parents felt entitled to a series of gestures

indicating her pleasure. The young graduate could "pay" her parents in
emotive display, a surface acting dissociated from her "real" definition of
the situation. Going one step further, she could pay them with a gesture
of deep acting-of trying to feel. The most generous gesture of all is the
act of successful self-persuasion, of genuine feeling and frame change, a
deep acting that jells, that works, that in the end is not phony (since it is
what the emotion is) though it is none the less not a "natural" gift.
The second way feeling rules come into play in exchange is shown when
the individual does not take the affective convention seriously; he or she
plays with it. For example, an airport observation: There are two airline
ticket agents, one experienced, one new on the job. The new agent is faced
with the task of rewriting a complex ticket (involving change of date, lower
fare, and credit of the difference between the previous and present fare to
be made toward an air travel card, etc.). The new ticket agent looks for the
"old hand," who is gone, while the customers in line shift postures and stare
intently at the new agent. The "old hand" finally reappears after 10 min-
utes, and the following conversation takes place: "I was looking for you.
You're supposed to be my instructor." Old hand: "Gee," with an ironic
smile, "I am really sorry. I feel so bad I wasn't here to help out" (they
both laugh). The inappropriate feeling (lack of guilt, or sympathy) can
be played upon in a way that says, "Don't take my nonpayment in emotion
work, or display work personally. I don't want to work here. You can
understand that." The laughter at an ironic distance from the affective con-
vention suggests also an intimacy; we do not need these conventions to hold
us together. What we share is the defiance of them.
COMMODITIZATION OF FEELING
In the beginning we asked how feeling rules might vary in salience across
social classes. One possible approach to this question is via the connections
among social exchange, commoditization of feeling, and the premium, in
many middle-class jobs, on the capacity to manage meanings.
Conventionalized feeling may come to assume the properties of a com-
modity. When deep gestures of exchange enter the market sector and are
bought and sold as an aspect of labor power, feelings are commoditized.
When the manager gives the company his enthusiastic faith, when the air-
line stewardess gives her passengers her psyched-up but quasi-genuine re-
assuring warmth, what is sold as an aspect of labor power is deep acting.
But commoditization of feeling may not have equal salience for all social
classes. It may have more salience for the middle class than for the working
class. The way each class socializes its children may, furthermore, prepare
them for future demands for the skill of emotion management.
569
When I speak of social class, it is not strictly speaking to income, educa-

tion, or occupational status that I refer, but to something roughly correlated
to these-the on-the-job task of creating and sustaining appropriate mean-
ings. The bank manager, the IBM executive, for example, may be required,
in part, to sustain a definition of self, office, and organization as "up and
coming," or "on the go," "caring," or "reliable," meanings most effectively
sustained through acts upon feeling. Feeling rules are of utmost salience
in jobs such as these; rule reminders and sanctions are more in play.
It is not, as Erich Fromm suggests, that the modern middle-class man "sells
his personality," but that, more precisely, many jobs call for an apprecia-
tion of display rules, feeling rules, and a capacity for deep acting.
Working-class jobs more often call for the individual's external be-
havior and the products of it-a car part assembled, a truck delivered 500
miles away, a road repaired. The creation and sustaining of meanings goes
on, but it is not such an important aspect of work. Physical labor is more
commoditized, meaning-making and feeling, less. Surely, too, there are
working- or lower-class jobs that do require the capacity to sustain meanings
and to do so, when necessary, by emotion work; the jobs of prostitute (El-
mer Pascua, work in progress) and personal servant require feeling manage-
ment. But to the extent that meaning-making work tends to be middle-class
work, feeling rules are more salient in the middle class.
There are jobs, like that of secretary or airline stewardess, with relatively
low financial rewards and little authority, which nonetheless require a high
degree of emotion and display management. Such jobs are often filled by
women, many of whom come from the middle class. Such workers are espe-
cially important as a source of insight about emotion management. Being
less rewarded for their work than their superiors, they are more likely to
feel detched from, and be perceptive about, the rules governing deep act-
ing. Deep acting is less likely to be experienced as part of the self and
more likely to be experienced as part of the job. Just as we can learn more
about "appropriate situation-feeling fits" by studying misfits, so too we
can probably understand commoditized feeling better from those for whom
it is a salient form of alienation (see Kanter's [1977] excellent chapters on
secretaries).
CLASS, CHILD REARING, AND EMOTION WORK
Middle- and working-class parents tend to control their children in different

ways (Kohn 1963, 1969; Bernstein 1971). Given the general pattern of
class inheritance, each class tends to prepare its children with the skills
necessary to "its" type of work environment and to pass on class-appro-
priate ways.
Middle-class parents tend to control via appeals to feeling, and the con-
570
Emotion Work
trol is more of feeling.14 The working-class parent, by contrast, tends to

control via appeals to behavior, and the control is more of behavior and its
consequences (Kohn 1963; Bernstein 1971). That is, the middle-class child
is more likely to be punished for "feeling the wrong way, or seeing things
in the wrong light," or having the "wrong intention," whereas the working-
class child is more likely to be punished for wrong behavior and its con-
sequences. The class difference in socialization amounts to different degrees
of training for the commoditization of feeling. This is yet another way the
class structure reproduces itself. (Thanks to Caroline Persell for this point.)
It may well be that, especially among the middle class, a corresponding
value is now placed on "authenticity," on things as they "truly are" or
"once were." Authenticity, which Lionel Trilling has described as the "new
moral virtue," when it refers to unworked-over feeling, may be rendered
scarce for those in the meaning-making sector. For this sector, the pattern
may be that of conventionalizing feeling, putting it on the market, and
looking for "authenticity" (see Trilling 1972).15
SUMMARY AND CONCLUSION
Why, we asked at the beginning, do we feel in ways appropriate to the

situation as much of the time as we do? One suggested answer is: because
we actively try to manage what we feel in accordance with latent rules. In
order to elaborate this suggestion we considered first the responsiveness of
emotion to acts of management as it is treated in the organismic and inter-
active account of emotion. According to the interactive account, we are not
14 As Kohn comments, ". . . middle class mothers are far more likely to punish their
son physically for what they call loss of temper than for behavior defined as wild
play. They appear to find the child's loss of temper, but not his wild play, particularly
intolerable" (1963, p. 308). Again, ". . . The interview reports indicate that the dis-
tinction between wild play and loss of temper was most often made in terms of the
child's presumed intent, as judged by his preceding actions . . . if his actions seemed
to stem from the frustration of not having his own way, they were judged to indicate
loss of temper" (Kohn 1963).
15 According to Trilling, the place of "sincerity," as a moral virtue, has been taken
(Trilling's verb is "usurped") by "authenticity." Sincerity refers to the relation between
inner feeling and outward display. Trilling offers many definitions of authenticity, but
one seems to refer to the relation between the self as emotion manager and the inner
feelings so managed. He cites Wordsworth's poem "Michael" about a very old shepherd:
"When Michael, after having lost his son Luke to the corruption of the city, continues
to build the sheepfold which he and the boy had ceremonially begun together, his
neighbors report of him that sometimes he sat the whole day, 'and never lifted up a
single stone"' (1972, p. 93). There is no act of self upon feeling; he is not psyching
himself "up" or "down," he is not "letting himself" feel grief, or deliberately "getting
into" his grief. It is on this account called "authentic," and deemed nowadays valuable.
What now ironically undermines authenticity, as a virtue, is the cultural belief in the
mutability of inner feeling and the individual's capacity, with therapeutic guidance
or otherwise, via "emotion work" or otherwise, to change fundamentally what is not
implacably ascribed after all.
571
always passive vis-a-vis an uncontrollable flood of feeling, and those occa-

sional efforts to actively shape our feeling can sometimes be effective. Tak-
ing this account I began to articulate the emotion-management perspective
first by distinguishing it from the dramaturgical perspective on the one
hand and the psychoanalytic perspective on the other. I then suggested
some links among emotive experience, deep acting, and feeling rules. In
turn, feeling rules were seen as the bottom side of ideology and therefore
subject to the same pressures for change as are ideologies.
Conventions of feeling (i.e., what one is supposed to feel) are used in
social exchange between individuals. Individuals operate their exchanges
according to a prior sense of what is owed and owing. Individuals see
themselves as being owed and as owing gestures of emotion work, and they
exchange such gestures. People bond, in the emotive sense, either by ful-
filling the emotive requirements situations call forth (e.g., the graduate
trying to feel happy) or by holding just these requirements to one side
(the ironic ticket agent acknowledging the sincere feeling "due" the other
even while playing with that notion of debt).
Just as gestures of emotion work can be exchanged in private, so they
can be exchanged in the marketplace, as an aspect of what is sold and
bought as labor power. In such a case we can speak of the "commoditiza-
tion" of emotion work. This prevails more for workers whose job it is to
make and sustain meanings (e.g., "this is an up-and-coming company";
"this is a pleasant, safe airplane")-jobs found more in the middle class.
Commoditization is less salient for those in physical labor or nonsocial
mental labor, more common in the working class. A reexamination of class
differences in child rearing suggests that middle-class families prepare their
children for emotion management more, working-class families less. Each,
in this way, prepares its children to psychologically reproduce the class
structure.
The emotion-management perspective can be applied to any number of
areas. We know little about how feeling rules vary in content from one
occupation to another. The funeral parlor director, the doctor, the com-
plaints clerk, the day-care worker all apply a sense of "should" to the
situated feelings that emerge in the course of a week. How do these
"shoulds" differ? Crosscutting occupational and class differences, they are
likely to exhibit cultural differences associated with gender and ethnicity.
Indeed, a good place to study change in feeling rules would be the strata of
persons for whom the right of men to cry, or feel fearful, is extended over
a greater range of situations, and for whom the right of women to open
anger is extended over a larger, sanction-free, zone. How has this set of
feeling rules, as the underside of feminist ideology, altered the understanding
between men and women as to what feelings are latently "owed" and
"owing"? We need to ask how different sexes, classes, and ethnic and
572
Emotion Work
religious groups differ in the sense of what one "ought to" or "has the
right to" feel in a situation. How different is the burden of hidden work
trying to obey latent laws? Finally, in whose interest are these feeling
rules? Some managing of feeling promotes the social good. Some does not.
Surely the flight attendant's sense that she "should feel cheery" does more
to promote profit for United than to enhance her own inner well-being.
REFERENCES
Arnold, Magda B. 1968. The Nature of Emotion. Baltimore: Penguin.

Averill, James R. 1976. "Emotion and Anxiety: Sociocultural, Biological, and Psycho-
logical Determinants." Pp. 87-130 in Emotion and Anxiety: New Concepts, Methods
and Applications, edited by M. Zuckerman and C. D. Spielberger. New York: Wiley.
Benedict, Ruth. 1946. Patterns of Culture. New York: Penguin.
Bernstein, Basil. 1971. Class, Codes and Control. New York: Schocken.
Blau, Peter. 1964. Exchange and Power in Social Life. New York: Wiley.
Blumer, Herbert. 1969. Symbolic Interactionism: Perspective and Method. Englewood
Cliffs, N.J.: Prentice-Hall.
Campbell, Sarah, ed. 1976. The Piaget Sampler. New York: Wiley.
Carlson, Anton J., et al. 1959. Feelings and Emotions: The Mooseheart Symposium.
New York: McGraw-Hill.
Clanton, Gordon, and Lynn G. Smith. 1977. Jealousy. Englewood Cliffs, N.J.: Prentice-
Hall.
Collins, Randall. 1975. Conflict Sociology. New York: Academic Press.
Darwin, C. (1872) 1955. The Expression of the Emotions in Man and Animals. New
York: Philosophical Library.
Davis, Kingsley. 1936. "Jealousy and Sexual Property." Social Forces 14:395-410.
Durkheim, Emile. 1961. The Elementary Forms of the Religious Life. New York: Collier.
Ekman, P. 1972. "Universals and Cultural Differences in Facial Expressions of Emo-
tion." Pp. 207-83 in Nebraska Symposium on Motivation 1971, edited by J. K. Cole.
Lincoln: University of Nebraska Press.
, ed. 1973. Darwin and Facial Expression. New York: Academic Press.
Erikson, Eric. 1950. Childhood and Society. New York: Norton.
Foster, George M. 1972. "The Anatomy of Envy: A Study in Symbolic Behavior."
Current Anthropology 13:165-202.
Freud, Sigmund. 1911. "Formulations regarding the Two Principles in Mental Func-
tion." Pp. 13-21 in Collected Papers. Vol. 4. New York: Basic.
. 1915a. "Repression." Pp. 146-58 in Standard Edition. Vlol. 14. London: Hogarth
Press.
. 1915b. "The Unconscious." Pp. 177-79 in Standard Edition. Vol. 14. London:
Hogarth Press.
Fromm, Erich. 1942. Escape from Freedom. New York: Farras & Rinehard.
Geertz, Clifford. 1964. "Ideology as a Cultural System." Pp. 49-76 in The Interpreta-
tion of Cultures. London: Hutchinson.
Gerth, Hans, and C. Wright Mills. 1964. Character and Social Structure: The Psy-
chology of Social Institutions. New York: Harcourt, Brace & World.
Goffman, Erving. 1956. "Embarrassment and Social Organization." American Journal
of Sociology 62 (November): 264-71.
. 1959. The Presentation of Self in Everyday Life. Garden City, N.Y.: Dou-
bleday.
1961. "Fun in Games." Pp. 17-84 in Encounters. Indianapolis: Bobbs-Merrill.
1967. Interaction Ritual Garden City, N.Y.: Doubleday.
1969. Strategic Interaction. Philadelphia: University of Pennsylvania Press.
1974. Frame Analysis. New York: Harper & Row.
573
. 1976. "Gender Advertisements." Studies in the Anthropology of Visual Com-

munication, vol. 3, no. 2 (Fall).
Goode, William. 1974. "The Theoretical Importance of Love." Pp. 143-56 in The
Family, Its Structure and Functions, edited by Rose Coser. New York: St. Martin's.
Gross, F., and G. P. Stone. 1964. "Embarrassment and the Analysis of Role Require-
ments." American Journal of Sociology 70 (July): 1-15.
Hillman, James. 1964. Emotion: A Comprehensive Phenomenology of Theories and
Their Meanings for Therapy. Evanston, Ill.: Northwestern University Press.
Hochschild, Arlie Russell. 1975. "The Sociology of Feeling and Emotion: Selected Possi-
bilities." Pp. 280-307 in Another Voice, edited by Marcia Millman and Rosabeth
Moss Kanter. New York: Anchor.
. 1977. "Reply to Scheff." Current Anthropology 18, no. 3 (September): 494-95.
Horney, Karen. 1937. The Neurotic Personality of Our Time. New York: Norton.
James, William, and Carl B. Lange. 1922. T he Emotions. Baltimore: Williams & Wilkins.
Kanter, Rosabeth. 1977. Men and Women of the Corporation. New York: Basic.
Katz, Judith. 1977. "Discrepancy, Arousal and Labelling: Towards a Psycho-Social
Theory of Emotion." Mimeographed. Toronto: York University.
Kemper, Thomas D. 1978. "Toward a Sociological Theory of Emotion: Some Problems
and Some Solutions." American Sociologist 13 (February): 30-41.
Kohn, Melvin. 1963. "Social Class and the Exercise of Parental Authority." Pp. 297-313
in Personality and Social Systems, edited by Neil Smelser and William Smelser. New
York: Wiley.
-. 1969. Class and Conformity. Homewood, Ill.: Dorsey.
Lazarus, Richard S. 1966. Psychological Stress and the Coping Process. New York:
McGraw-Hill.
Lazarus, Richard S., and James R. Averill. 1972. "Emotion and Cognition: With Special
Reference to Anxiety." Pp. 242-83 in Anxiety: Current Trends in Theory and Re-
search, edited by D. C. Spielberger. Vol 2. New York: Academic Press.
Lazarus, Richard S., Allen D. Kanner, and Susan Folkman. 1979. "Emotions: A Cogni-
tive-phenomenological Analysis." In Theories of Emotion, edited by R. Plutchik and
H. Kellerman. New York: Academic Press, in press.
Levy, Robert I. 1973. Tahitians, Mind and Experience in the Society Islands. Chicago:
University of Chicago Press.
Lofgren, L. Borge. 1968. "Psychoanalytic Theory of Affects." Journal of the American
Psychoanalytic Association 16 (July): 638-50.
Lyman, S., and Marvin Scott. 1970. Sociology of the Absurd. New York: Appleton-
Century-Crofts.
McDougall, W. 1937. "Organization of the Affective Life: A Critical Survey." Acta
Psychologica 2:233-46.
. 1948. An Outline of Psychology. 12th ed. London: Methuen.
Mead, A. H. 1934. Mind, Self and Society, edited by Charles Morris. Chicago: Univer-
sity of Chicago Press.
Plutchik, Robert. 1962. The Emotions: Facts, Theories and a New Model. New York:
Random House.
Rabkin, Richard. 1968. "Affect as a Social Process." American Journal of Psychiatry
125 (December): 772-79.
Rapaport, D. 1953. "On the Psychoanalytic Theory of Affects." International Journal
of Psycho-Analysis 34:177-98.
Reymert, Martin, ed. 1950. Feelings and Emotions: The Mooseheart Symposium. New
York: McGraw-Hill.
Riesman, David. 1952. Faces in the Crowd: Individual Studies in Character and Politics.
New Haven, Conn.: Yale University Press.
. 1960. The Lonely Crowd: A Study of the Changing American Character. New
Haven, Conn.: Yale University Press.
Schachter, Stanley. 1964. "The Interaction of Cognitive and Physiological Determinants
of Emotion States." Pp. 138-73 in Psychobiological Approaches to Social Behavior,
edited by P. H. Leiderman and D. Shapiro. Stanford, Calif.: Stanford University
Press.
574
Emotion Work
Schachter, S., and J. Singer. 1962. "Cognitive, Social and Physiological Determinants
of Emotional State." Psychological Review 69:379-99.
Schafer, Roy. 1976. A New Language for Psychoanalysis. New Haven, Conn.: Yale
University Press.
Scheff, Thomas J. 1973. "Intersubjectivity and Emotion." American Behavioral Scientist
16 (March-April): 501-22.
. 1977a. "The Distancing of Emotion in Ritual." Current Anthropology 18, no. 3
(September): 483-91.
. 1977b. "Toward a Sociology of Emotion." Mimeographed. Santa Barbara:
University of California.
Shapiro, David. 1965. Neurotic Styles. New York: Basic.
Shott, Susan. 1979. "Emotion and Social Life: A Symbolic Interactionist Analysis."
American Journal of Sociology 84 (May 1979): 1317-34.
Simpson, Richard. 1972. Theories of Social Exchange. Morristown, N.J.: General Learn-
ing Press.
Singelmann, Peter. 1972. "Exchange as Symbolic Interaction: Convergences between
Two Theoretical Perspectives." American Sociological Review 37 (4): 414-24.
Solomon, Robert. 1973. "Emotions and Choice." Review of Metaphysics 27 (7): 20-41.
Sprout, W. J. H. 1952. Social Psychology. London: Methuen.
Swanson, Guy E. 1965. "The Routinization of Love: Structure and Process in Primary
Relations." Pp. 160-209 in The Quest for Self-Control: Philosophies and Scientific
Research, edited by Samuel Z. Klausner. New York: Free Press.
Swanson, Guy E., and Daniel Miller. 1966. Inner Conflict and Defense. New York:
Schocken.
Tomkins, S. S. 1962. Affect, Imagery, Consciousness. 2 vols. New York: Springer.
Trilling, Lionel. 1972. Sincerity and Authenticity. Cambridge, Mass.: Harvard Univer-
sity Press.
575

Hochschild EmotionWorkFeeling 1979

Uploaded by

Copyright:

Available Formats

Hochschild EmotionWorkFeeling 1979

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hochschild EmotionWorkFeeling 1979

Uploaded by

Copyright:

Available Formats

Emotion Work, Feeling Rules, and Social Structure

Author(s): Arlie Russell Hochschild

This essay proposes an emotion-management perspective as a lens

@) 1979 by The University of Chicago. 0002-9602/80/8503-0003$02.00

AJS Volume 85 Number 3 551

Why is the emotive experience of normal adults in daily life as orderly

3 For comprehensive, if somewhat dated, reviews of the theoretical approaches to emo-

TWO ACCOUNTS OF EMOTION AND FEELING

automatic reflex syndrome-Darwin's instant snarl expression, Freud's ten-

logue to other adaptive mechanisms, such as shivering when cold or per-

The Interactive Account of Emotion and Social Psychology

Goffman guides our attention to social patterns in emotive experience. He

clusive focus on the actor's attentiveness to behavioral facade and assume

The need to replace Goffman's "black-box psychology" with some theory

pleasant affect. Finally the notion of "inappropriate affect" is used to point

An obsessive-compulsive patient-a sober, technically minded and active

What seems interesting about this example differs according to whether

By "emotion work" I refer to the act of trying to change in degree or quality

Often emotion work is aided by setting up an emotion-work system, for

Emotion work becomes an object of awareness most often, perhaps, when

To begin with, let us consider several common forms of evidence for

Marriage, chaos, unreal, completely different in many ways than I imag-

In any given situation, we often invest what we expect to feel with

FRAMING RULES AND FEELING RULES: ISSUES IN IDEOLOGY

It follows that when an individual changes an ideological stance, he or

Part of what we refer to as the psychological effects of "rapid social

contradictions between contending sets of rules. Feelings and frames are

FEELING RULES AND SOCIAL EXCHANGE

her graduation possible. The parents felt entitled to a series of gestures

When I speak of social class, it is not strictly speaking to income, educa-

CLASS, CHILD REARING, AND EMOTION WORK

Middle- and working-class parents tend to control their children in different

trol is more of feeling.14 The working-class parent, by contrast, tends to

SUMMARY AND CONCLUSION

Why, we asked at the beginning, do we feel in ways appropriate to the

always passive vis-a-vis an uncontrollable flood of feeling, and those occa-

Arnold, Magda B. 1968. The Nature of Emotion. Baltimore: Penguin.

. 1976. "Gender Advertisements." Studies in the Anthropology of Visual Com-

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.