8/7/2019 TURNER Toward a General Sociological Theory of Emotion
1/30
Journal for the Theory of Social Behaviour29:200218308
Toward a General Sociological Theory ofEmotions
JONATHAN H. TURNER
Over the last two decades, the sociology of emotions has become one of the
leading edges of micro-level theorizing. Building upon this theoretical momentum,
this paper draws together key ideas from diverse theories and presents a synthetic
theory of emotional dynamics in encounters of face-to-face interaction. Emphasis
is on overt emotional arousalthat is, on emotional states visible to the parties
involved. Such arousal shapes the mutual signaling and interpreting of gestures
between individuals, especially gestures that can be read visually; and in so
doing, emotionally-laden gesturing determines the flow of interaction. Yet, as
will be emphasized, emotional displays are only surface manifestations of
complicated and covert emotional dynamics, operating through the neurology
of the brain. The theory to be developed pulls together arguments of several
theoretical traditions, including mainstream symbolic interactionism (e.g., Mead,
1934; McCall & Simmons, 1966; Stryker, 1980; Stryker & Serpe, 1982; Burke,
1980, 1991; Heise, 1979; Smith-Lovin & Heise, 1988; Smith-Lovin, 1990),
psychoanalytic variants of symbolic interactionism (Scheff, 1988; Scheff &
Retzinger, 1992; Turner, 1994), interaction ritual theory (Collins, 1988, 1993),
dramaturgical analysis (Goffman, 1959, 1967, 1974; Hochschild, 1979), powerand status theories (Collins, 1975; Kemper, 1978; Kemper & Collins, 1990),
and expectation-states theories (Berger & Zelditch, 1985, 1998; Berger, Webster,
Ridgeway & Rosenholtz, 1986; Wagner & Berger, 1993; Wagner & Turner,
1998). My view is that each of these approaches has captured an important
dynamic of emotions as they influence the flow of face-to-face interaction, but
at the same time, none has delineated the full range of emotional forces operating
on participants during an encounter.
In Figure 1, a very skeletal outline of the argument is presented. For any
given cycle of emotional arousal, the process begins with expectations frommany diverse sources; in turn, these expectations are collated as part of the
definition of a situation into a generalized expectation about what will or should
occur in the encounter. Self is a special type of expectation because when its
The Executive Management Committee/Blackwell Publishers Ltd. 1999. Published by Blackwell Publishers 108 Cowley
Road, Oxford, OX4 1JF UK and 350 Main Street, Malden, MA 02148, USA.
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
2/30
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
3/30
Sociological Theory of Emotions 135
expectation-state reserach has been conducted in task-oriented experimental
groups, although there has been some effort to move to less controlled situations
(e.g., Berger, 1988). In Table 1, some of the forces generating expectations in
natural settings are listed. This listing does not mean, of course, that parties to an
encounter always have well-defined expectations before they interact; obviously,people can enter interactions with few or vague expectations about what is to
occur; and under these conditions, individuals only begin to form expectations
as the interaction unfolds. And even when expectations are well defined before
individuals enter the encounter, expectations may change, although there are
limits and patterns to change as the large literature on expectation sttes
documents. Thus, Table 1 only emphasizes that as expectations formwhether
before, during, or after an episode of face-to-face interactionthey will become
organized along the dimensions listed in the Table. That is, individuals will
develop expectations with respect to demography, structure, culture, and trans-
actional needs; and as will be emphasized shortly, individuals will also bring to
an encounter, and develop within the flow of the interaction, expectations about
their own self as well as the selves of others. For the moment, however, let me
review just the forces enumerated in Table 1.
(1) Demographic forces revolve around the number and nature of individuals in
an encounter, particularly a focused encounter but an unfocused one as well
(Goffman, 1961). The expectations for a situation are influenced by the number
of individuals present, and equally important, the categories of individuals as
friends, intimates, and strangers who exhibit other characteristics like gender,
ethnicity, class, and age. Affect control theory (Heise, 1979; Smith-Lovin &
Heise, 1988; Heise & Smith-Lovin, 1981; Heise & Smith-Lovin, 1995) is correct,
I think, in emphasizing that others are also assessed with respect to their likely
level of activity, their potency, and their relevance as objects of evaluation.
(2) Structural forces include the nature, number and relationships among status
positions that can become implicated in an encounter. As Kemper & Collins
(1990) have stressed, power (authority) and status (prestige) considertions are the
driving force behind the activation of emotions. As Table 1 suggests, power andstatus are only one set of many forces generating expectations about what
individuals can and should do, a conclusion that follows from expectation-states
theory (e.g., Ridgeway & Smith-Lovin, 1994). Moreover, power and status in a
situation exert much of their influence on emotions by making self highly salient;
that is, when encounters place individuals in hierarchies of power or status, they
become more self aware. And as the salience of self increases, so does the
potential for emotional reactions to behaviors that do not affirm the power and
status orders (Ridgeway, 1982, 1994). Thus, it may be that the effect emphasized
by Kemper & Collins (1990) of power and status on many emotions is due tothe increasing salience of self in power/status encounters. In addition to power
and status, the density of connections among positions is important, because
high density network ties create expectations that are very different than those
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
4/30
136 Jonathan H. Turner
Table 1. Sources of Expectations
1 Demographic forces: the number and nature (categories) of others present in anencounter, especially with respect to the imputation of their activity, potency, andevaluation.
2 Structural forces: the nature, number, and relationships among status positionsimplicated in an encounter, particularly with respect to:
a differentiation of power/authority among positions or actorsb differentiation of prestige/honor among positions or actorsc density of connections among positions or actorsd diffuse status characteristics that actors in positions can claim and use
3 Cultural forces: sets of symbols that individuals are cognizant of, or that they intendto invoke in an encounter, especially with respect to:
a the appropriate ways of behaving (norms) and evaluating such behavior (values)b the appropriate type and intensity of sanctions to be employed
c the appropriate type and intensity of sentiments, or feeling rulesd the appropriate type of ritualse the appropriate frames to be keyed and rekeyed
4 Transaction forces: the intensity of, and relative proportions among, need states thatindividuals seek to meet in an encounter, especially with respect to:
a the need for profitable pay-offs of relevant resourcesb the need for predictability in the actions of othersc the need to trust that others will meet obligations and expectationsd the need for intersubjectivity in order to presume that others experience a
situation in similar ways, or if different, that these differences can be understoode the need to feel implicated and included in the ongoing flow of interaction in an
encounter.
for low density networks. At a minimum, as Collins (1975) has emphasized, high
density networks make conformity to normative expectations more likely and
sanctioning more effective. Finally, as expectation-states theory has documented,
authority, prestige, and other social characteristics of actors outside the encounter
are often brought into the interactions as diffuse status characteristics; and
these characteristics will influence the power-status hierarchy and the behaviorsof individuals in this hierarchy.
(3) Cultural forces are those systems of symbols that provide instructions and
standards for behavior among those participating in an encounter. Some are
behavioral or normative standards about how self and others are expected to
behave; others are more evaluative and stress value standards about what self
and others should do in a situation. Additional cultural sets indicate how
demographic, structural, and self forces are to be deployed in an encounter.
Related to both normative and value standards are sets of symbols indicating
the appropriate type of sanctions (both positive and negative) as well as the formand intensity of sanctioning to be employed in a situation. A very special set of
normative and evaluational standards is, as Hochschild (1979, 1983) first
emphasized, the feeling rules or the expectations that individuals have about
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
5/30
Sociological Theory of Emotions 137
the nature and intensity of emotions that can, or should, be displayed in an
encounter. Yet another set of symbols defines the appropriate use of rituals, or
stereotyped sequences of behavior that guide the flow of interaction (Goffman,
1967). Ritualized opening, closing, repairing, forming, tracking, keying, framing,
totemizing and other interpersonal processes are conducted through rituals(Turner, 1988), with normative and value standards circumscribing the range of
rituals to be employed in an encounter. Finally, again following Goffman (1974),
cultural codes dictate frames demarking what is to be part of the interaction and,
conversely, what is to be excluded; and particularly important, I think, are what I
have termed body frames (the relevance of body to the flow of interaction) and
personal information frames (the relevance of biography, intimacy, and self
involvement) to be used to structure the boundaries of what is inside and outside
the flow of interaction in an encounter (Turner, 1996b). All of these cultural forces,
then, create expectations about what is to transpire in a situation.
(4) Transactional forces are fundamental need-states that individuals activate in
all encounters and that they expect to be met during the course of interaction.
Previously, I saw these needs as what motivates individuals (Turner, 1987, 1988,
1994), but more recently, I see motivation as a much more inclusive process,
revolving around the emotional dynamics to be examined in this paper. Yet,
individuals do have what I now term transactional needs, or needs for receiving
payoffs of relevant resources (typically in accordance with cultural standards of
equity and justice), for feeling that they can predict the responses of others, for
trusting others to meet their obligations, for presuming (however inaccurately)
that self and other are experiencing the world in similar ways for the purposes
of the interaction, and for sensing that self is minimally included in the ongoing
flow of the interaction. These transactional needs are listed in Table 1 in their
order of importance to the rhythmic flow of an interaction and to individuals
sense of satisfaction; and like the other forces in Table 1, they generate
expectations about what should transpire in the encounter.
In sum, we can begin to develop a theory of emotions by recognizing the
importance of expectations. As a general proposition, we can say that the greateris the number of expectations that form an individuals definition of a situation
and the higher is the salience of these expectations in an encounter, the greater
is the potential for emotional arousal by an individual and for affective displays
of emotions to others in the encounter. However, as we will see, this general
proposition is qualified by the other forces outlined in Figure 1. For a review of
the theory as a whole, consult the sixteen propositions in the conclusion.
SELF AND IDENTITY
My conceptualization of self follows the general line of argument in most
theories within the symbolic interactionist tradition. Self is a set of images that
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
6/30
138 Jonathan H. Turner
individuals have about themselves with respect to (1) who they are, (2) what
they can do, and (3) how they should be treated. Stated in this way, the
formulation is rather vague and, hence, requires elaboration.
The first issue to be addressed is whether or not self is a cognitive or emotional
construct that individuals hold about themselves. So stated, the implied questionis rhetorical because the obvious answer is that self involves both cognitive and
emotional elements. Indeed, as studies on the process of forming cognitions and
remembering them now document, cognitions cannot be remembred without
corresponding emotional tags (for accessible summaries of the neurological
literature see: Damasio, 1994; Le Doux, 1996; Turner, 2000); and for a highly
salient cognition like self, the cognitive and emotional dimensions are so
neurologically embedded in each other that it is impossible conceptually and
empirically to separate them. Thus for my purposes here, self is the activation
of images about oneself as an object, both in the present and from long-term
memory, variously charged with different types and valences of emotion.
A second issue that needs to be addressed is the nature of these images.
Symbolic interactionists have, at times, tended to emphasize self as a verbal
construct. My view is that linguistic designations are only a special and limited
property of self; and in fact, it would be impossible for self to be only verbal
images because the human brain does not think in terms of conventional
language. If humans did think only in words, they would seem very dim-witted,
since language is an auditory mode of communication and, hence, sequential
and very slow (Turner, 1996a, 1999). Humans are a primate and, by virtue of
this fact, they are visually dominant (Maryanski & Turner, 1992:3368; Turner,
1996a, 1996b, 2000); and I have little doubt that humans think in terms of
patterns that better reflect the wiring of the brain around the visual sense
modality that receives, processes, and stores information as gestalts. Indeed,
regardless of the visual bias to thought, the neurology of consciousness in general
appears to generate gestalts and configurations of sensory information. Sensory
stimulation of visual, auditory, haptic, and olfactory receptors in the present are
integrated in association cortices, such as the interior parietal lobule, and sentto working memory simultaneously, with this simultaneity giving consciousness
the character of configurations of sensations and, if tagged with emotions from
the activation of subcortical limbic systems, configurations of feelings as well
(Geschwind, 1965a, 1965b; Geschwind & Damasio, 1984; Damasio, 1994). If
working memory seeks to activate long-term memories, which have been created
by the hippocampus in assigning emotional valences to past configurations of
sensations and feelings, the process appears to involve stimulating clusters of
neurons containing shorthand codes of an unknown nature that, once stimulated,
organize a reactivation of sensory cortices in a manner that reproduces in anearly but far from exact way the previous inputs that had once entered
working memory. This refiring of the sensory cortices leads to the simultaneous
convergence of sensations and associated feelings in short-term or working
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
7/30
Sociological Theory of Emotions 139
memory, thereby imposing new images from long-term memory on those being
immediately received in the present.
Because primate neuroanatomy is organized in ways that make vision the
dominant sense modality, it is most likely that the first languages among hominids
were visual (e.g., hand, face, and body movements); and to this very day, humansrely upon visual cues as much as linguistic utterances in role taking and role
making. The existence of two discrete areas on the left side of the brain
Brocas area in the frontal lobe for speech production and Wernickes area next
to the inferior parietal lobe for speech comprehensionindicates that the brain
must use translation boxes to convert more gestalt-based brain thinking into
sequential speech, although other areas along the Sylvian Fissure are also
essential to speech comprehension and production (Kimura, 1993). Thus, self is
mostly images of oneself couched in the brains way of thinking, revolving
around coded neurons reactivating sensory cortices to reproduce images about
self; and these images are organized as configurations of cognitive-emotional
states. These images can be translated into linguistic modes, as when people
talk to themselves. Indeed, this capacity to slow the process of thought down
by translating brain thinking into linguistic constructions is important in
understanding emotional dynamicsa point to which I will return.
In this context, let me offer a side comment. One of the reasons that theories
of self have proven so difficult to test empirically is that measurements rely on
languagethat is, on utterances that represent translations of the brains way
of thinking. In an old test like the Twenty Statements Test, for example,
subjects were being asked to translate into words what they feel in the brains
way of thinking, and given the nature of self images as cognitive-emotional
constructs and the operation of defense mechanisms (to say nothing of diverse
interpretations of the tests directions by subject), it is not surprising that TSTs
only grossly measured some of the images that individuals have about themselves.
Other tests of self, such as the use of the semantic differential in affect control
theory, produce better results, but perhaps not for the reason that those who
use the test generally assume (e.g., Heise, 1979; Smith-Lovin and Heise, 1988).In employing the semantic differential, subjects produce patterns of responses
to dichotomous pairs of words which mix cognitions and affect. What makes
this test more successful is the responses better reflect the way that the brain
organizes information. Translation is still occurring but the product of this
translation via Wernickes and Brocas area looks more like the way the brain
configures cognitions (particularly given the way the statistical analyses of the
tests configure individuals scores); and hence, we can have greater confidence
in such an approach, although it is still a by-product of translating visually-
organized images into words and, then via statistical analysis, back into scoresthat better approximate how the brain stores images and thoughts.
A third related issue that has been hotly debated within interactionist circles
is the way in which humans organize self images. One aspect of this issue is
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
8/30
140 Jonathan H. Turner
substantive: what are the dimensions along which cognitive-emotional images
are arrayed? For example, in affect control theory, activity, potency, and
evaluation are the critical dimensions; in other theories, we can find somewhat
different, though overlapping, dimensions for arranging self images (e.g., Gecas,
1989, 1991). For my purposes, however, I will resist the temptations to visualizesubstantive images or self as organized along a few dimensions, although as
Table 1 suggests, humans probably organize expectations for others in this way.
A related question about the organization of self images is structural: Do
individuals organize images of themselves into hierarchies of identities (e.g.,
McCall & Simmons, 1966; Stryker, 1980, 1991; Stryker & Serpe, 1982; Serpe
& Stryker, 1987)? I suspect that some images of self are more salient and
emotionally charged than others, but just which self images become salient is
not just a reflection of clear rank ordering within the brain but also a response
to purely situational cues activating particular images from long-term memory
and, via working memory, integrating them with sensory images present in the
immediate situation. Yet another question addresses how self images are organized
and asks whether or not humans carry multiple identities and, relatedly, whether
or not there are core identities (or self) and more peripheral and situational
identities (or selves). My position is that individuals do have core selves or
identities that they seek to confirm in all interaction situations; and these
identities carry with them high-intensity emotional valences. Individuals also
have situation-specific selves, tailored to images about who they are, what they
can do, and what they can expect from others; and these too carry emotional
tags but not to the same intensity as the emotions attached to core self. Moreover,
there is a relatively high degree of consistency between core and peripheral
selves, and if this is not the case, the individual suffers considerable emotional
distress because of the incongruity (Higgins, 1987, 1989). Thus, in terms of how
the cognitive-emotional images of self are organized, the theory to be developed
here emphasizes that self is a series of emotional valences attached to images
about oneself as an object, some of which are trans-situational and fundamental
to a person and others of which are tied to particular types of situations. Ineither case, where others do not confirm self, emotions are aroused and defense
mechanisms are activatedall of which influence the nature and level of covert
and overt emotional arousal evident for a person during interaction (Higgins,
1987, 1989; Heise, 1989; Smith-Lovin, 1990).
A fourth issue is the behavioral responses that individuals make toward
themselves as objects. One very important response to others and self is
subcortical and involves activation of those of areas of the brain, sometimes
loosely termed the limbic systems (Maclean, 1990), that generate unconscious
emotional responses (Damasio, 1994; Le Doux, 1996). Another form of behavioralresponse is dramaturgical, revolving around the orchestration of gestures, both
consciously and unconsciously, so as to elicit desired and expected responses
from others or to achieve a particular goal or purpose (Goffman, 1959). Yet,
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
9/30
Sociological Theory of Emotions 141
another behavioral response is to react to oneself as an object before, during,
and after interaction (Serpe, 1991). As I will argue, self reflection is a very
important aspect of self, as all interactionist theories would emphasize, but it is
important in a sense not always stressed by these theories. Much self reflection
is, I believe, conducted in the brains mode of thinking and is not, therefore,linguistic (Turner, 1996a, 1996b); and in fact, self reflection is typically a blur
of emotionally-charged images and feelings that are not so easily put into
wordswhich is, perhaps, another way to phrase William James famous phrase,
the stream of consciousness. This stream is being activated by various
subcortical and unconscious limbic systems, thereby making emotionally-charged
images about self surface manifestations of complex neurological forces that
operate outside consciousness. The vast majority of self reflections in non-
problematic interactions operates in this way, and for the most part, avoids the
auditory mode of thought which immediately puts thinking into slow motion.
Verbalized self reflection, as an internal conversation is, however, critical to
understanding emotional arousal, especially since so many of the limbic forces
causing the arousal are subcortical and, to an individual, not easily understood
(although those interacting with an individual can often see the bodily changes
occurring as subcortical limbic systems mobilize motor and visceral responses).
Such self talk slows the process of thought and self reflection down so that
individuals ponder why they are experiencing emotions and, if necessary, so that
they can deal with problematic situations. Moreover, self talk gives individuals
the ability to anticipate and re-live the emotions aroused in a situation,
particularly when these emotions carry high valences. Additionally, and this is
important for the theory that I seek to present, self talk can often delay and, if
sufficiently frank and open, forestall the activation of defense mechanisms which
will, if activated, compress and distort emotional responses.
Yet, self talk can also sustain emotions that have been aroused in a situation,
whether positive or negative; and if self talk is sufficiently vocal about negative
experiences, emotions can be raised to a new amplitude, although eventually
fatigue and defense mechanisms will lower the emotional voltage. When selfreflection about a situation allows an individual to re-experience positive
outcomes of an interaction, however, the emotional energy generated can be
sustained and periodically reactivated; and as Collins (1988) has emphasized,
this kind of self reflection can become part of the particularized cultural capital
of an individual and, hence, set up expectations for the next episode of
interaction. The same is true for negative experiences that have been reflected
upon, although in this case, such reflection will generally lower expectations
unless defense mechanisms have so distorted the process of self talk that the
person is no longer in touch with the interpersonal realities behind their ownexperiences. In this latter case, expectations can remain artificially high, thus
setting off another round of negative experiences which only force the escalated
use of defense mechanisms. But, as will be argued later, this use of defense
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
10/30
142 Jonathan H. Turner
mechanisms ultimately depresses emotional energy, along several fronts: repres-
sion will eventually increase the valences for disappointment-sadness; projection
will lead to aversion-fearful responses from others or aggressive-angry negative
sanctioning from these others; and displacement of negative emotions onto
others will elicit aggressive-angry forms of negative sanctioning from the victimsof such displacement.
A final issue in the conceptualization of self is defense mechanisms, a topic to be
examined in greater detail later. Defense mechanisms seek to protect self from
unpleasant emotions, especially those directed at core self but peripheral selves as
well. Repression, displacement, and projection all distort individuals perceptions
of, and self reflections about, themselves in ways that change the overt emotional
reactions and the level of emotional energy displayed in a situation. With some
highly notable exceptions (e.g., Scheff, 1988; Scheff and Retzinger, 1992; Turner,
1994), symbolic interactionists and many other micro theories have not adequately
conceptualized the operation of defense mechanisms; and as we will see, they are
central to a theory seeking to explain peoples level of emotional energy.
Thus, when self is highly salient, the potential for activation of subcortical
areas of the brain generating emotions increases. Just whether or not this
potential is realized depends on the degree to which expectations in general and
those about self in particular have been met during the course of interaction.
To anticipate the theory summarized in the conclusions, when expectations have
been met or exceeded and when others have offered positive sanctions, or both,
people arouse a high level of positive effect towards themselves and others; and
conversely, when expectations have not been met or others have given negative
sanctions, or both, negative emotions toward self and others are activated.
Attribution processes qualify this relationship. When success in meeting expecta-
tions or receiving positive sanctions is attributed to self, self is evaluated positively,
and individuals will sustain relations with others in the encounter; when success
is attributed to internal states of others or categories of others, positive affect
towards these others will be aroused; and when success is attributed to external
causes in the situation, positive emotions will be of lower intensity. Conversely,when individuals do not meet expectations or receive negative sanctions from
others, or both, negative affect towards self, and others ensues; and this process
too is channelled by attributions. When self is blamed, powerful negative
emotions toward self ensue; when internal states of others are seen as the cause
of failure to meet expectations and receive negative sanctions, or both, aggressive
emotions toward others will be evident; and when external causes are seen to
be the cause, negative emotions toward self and others will be muted. Self talk
can increase the intensity and duration of both positive and negative emotional
arousal toward self and others; and when such self talk can target correctly thesources of emotional arousal, individuals will experience more positive emotions
toward self and others in the present encounter and in other encounters as well.
The activation of defense mechanisms generally lowers the modal level of overt
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
11/30
Sociological Theory of Emotions 143
emotional display by individuals, but increases the likelihood of short and
periodic emotional outbursts. Of course, these generalizations, summarized by
the propositions in the conclusions, get us ahead of the argument, but they give
us a sense for the direction of the theory that I seek to develop.
COLLATING EXPECTATIONS INTO A DEFINITION OF THE SITUATION
Depending upon the situation, all or various subsets of the expectations listed
in Table 1 can be operative. Moreover, these expectations are all interrelated.
For example, transactional needs for payoffs in terms of standards or equity and
justice are related to normative and value standards. Thus the listing in Table
1 represents only a rough statement of research topics on expectation states
rather than a refined and mutually exclusive categorization of variables. Yet, for
the purposes of the argument to be developed, the critical consideration is that
expectations exist along many dimensions, and humans can simultaneously
activate all of these sources of expectations, although more typically some are
more salient than others.
Similarly, self as a special kind of expectation is also likely to be operative;
and depending on the extent to which core or peripheral self is salient, the
expectation-states for a situation will vary. Moreover, since the forces listed in
Table 1 can increase the salience of selfas is the case, for example, when
differentiation of power and prestige among interactants is high, when values
dominate expectations, or when payoffs involve intrinsic resources from others.
Equally important, multiple expectations are combined, as expectation-states
research suggests (e.g., Wagner and Turner, 1998). Just how this combining occurs
is not fully understood, but at some level, it represents a mechanism for simplifying
actors definitions of the situation. In combining expectations, some become a
more significant subset than others; and in terms of arousing emotional reactions,
Kemper and Collins (1990) are probably correct in seeing authority and status as
highly salient, but the effect of these structural forces operates, to some degree,through self and identity forces. Thus, authority/status and self would be more
likely, I believe, to dominate the formation of a subset of expectations and, thereby,
to circumscribe the inclusion or exclusion into the subset of other forces listed in
Table 1. Still, this point should not be pushed too far; and for the theory presented
here, self-expectations are the most influential on an individuals level of emotional
arousal, use of defense mechanisms, and emotional energy.
EXPECTATIONS AND EXPERIENCE
Several theoretical traditions document that incongruities between what indi-
viduals expect in a situation and what they experience generate emotional
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
12/30
144 Jonathan H. Turner
responses. For example, B.F. Skinner observed that pigeons under a reinforcement
schedule appeared to become angry when not reinforced, a finding that
Homans (1961) originally elaborated into a law of distributive justice (i.e.,
rewards to individuals should be proportional to their investments relative to
the investments of others in a situation) and later (Homans, 1974) an aggression-approval proposition (i.e., anger and punishment are likely responses to not
receiving an expected reward, satisfaction-approval for receiving a reward, and
happiness-approval for receiving more than expected). Molm (1991) in her
ongoing research program on punishment in exchanges has reported that
satisfaction and affect in exchanges of resources are mediated by expectations
and that units of punishment have greater effects on actors satisfaction because
rewards are more common than punishments, giving punishment a kind of
shock (or surprise) value. This effect, however, is reduced as punishments
become as frequent as rewards, although the basic relationship among expecta-
tions, experiences, and emotions remains. Within the expectation states theoretical
research program, it is consistently reported that emotions escalate when
individuals do not honor the expectations associated with the distribution of
status (Ridgeway, 1982, 1994; Ridgeway and Johnson, 1990; Berger, 1988).
Among the symbolic interactionist theories, the affect control approach developed
by David Heise and Lynn Smith-Lovin (Heise, 1979, 1989; Smith-Lovin and
Heise, 1988; Smith-Lovin, 1990) is built around the level of deflection between
fundamental sentiments and transient sentiments which escalates emotional
arousal and motivates actors to reduce the deflection. Within a somewhat
different wing of symbolic interactionism, Sheldon Stryker (1980) and collabor-
ators (Stryker and Serpe, 1982) have argued that emotions are a marker or sign
that a particular identity, which an individual expects to confirm, has not
received such confirming responses. Similarly, other interpersonal theories
emphasize congruence between self and responses of others (e.g., Backman,
1985, 1988; Swann, 1983, 1990; Higgins, 1987, 1989). And, within the power-
models of Kemper and Collins (1990), the activation of either negative or
positive emotions is related to whether or not anticipated confirmations ofrelative status or power have been realized.
Thus, it is clear that emotional arousal is related to the relationship between
expectations and experience. I will simply term this force the degree of congruity
or incongruity between expectations and experiences in a situation. If there is
an incongruity below expectations for reinforcing outcomes or associative
experiences, then the three primary emotions of assertion-anger, aversion-fear,
and disappointment-sadness are activated in varying degrees of intensity and in
different combinations and ratios, as will be discussed shortly (see Tables 2 and
3). If there is an incongruity that exceeds expectations for either positive ornegative outcomes, a variant of the primary emotion of satisfaction-happiness is
stimulated. And if expectations and experiences are congruent, then comparat-
ively low levels of emotion are aroused: if expectations were for positive
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
13/30
Sociological Theory of Emotions 145
Table 2. Variants of Primary Emotions
Satisfaction-Happiness
Low intensity Moderate intensity High intensity
content cheerful joysanguine buoyant blissserenity friendly rapturegratified amiable jubilant
enjoyment gaiety
Aversion-Fear
Low intensity Moderate intensity High intensity
concern misgivings terror
hesitant trepidation horrorreluctance anxiety high anxiety
Assertion-Anger
Low intensity Moderate intensity High intensity
annoyed displeased dislikeagitated frustrated loathingirritated belligerent disgust
vexed contentious hate
perturbed hostility despisenettled ire desestrankled animosity hatred
seethingwrath
Disappointment-Sadness
Low intensity Moderate intensity High intensity
discouraged dismayed sorrowdowncast disheartened heartsickdispirited glum despondent
resigned anguishedgloomywoefulpained
Startlement-Surprise
Low intensity Moderate intensity High intensity
sudden-awareness sudden-alertness amazed
astonishedastoundedincredulous
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
14/30
146 Jonathan H. Turner
Table 3. First-order combinations of primary emotions
Satisfaction-Happiness
Satisfaction-Happiness aversion-fear wonder, hopeful, relief, gratitude, pride,
reverenceSatisfaction-Happiness assertion-anger vengeance, appeased, calmed, soothed,
relish, triumphant, bemusedSatisfaction-Happiness disappointment-sadness nostalgia, yearning, hopefulnessSatisfaction-Happiness startlement-surprise elation, exhilaration, thrilled
Aversion-Fear
Aversion-Fear satisfaction-happiness awe, reverence, venerationAversion-Fear assertion-anger revulsed, repulsed, antagonism, dislike,
envyAversion-Fear disappointment-sadness dread, warinessAversion-Fear startlement-surprise panic, consternation, alarmed, scared
Assertion-Anger
Assertion-Anger satisfaction-happiness snubbing, mollified, rudeness, placated,righteousness
Assertion-Anger aversion-fear abhorrence, jealousy, suspiciousnessAssertion-Anger disappointment-sadness bitterness, depression, betrayed
Assertion-Anger startlement-surprise offended, incited, perturbed, piqued,outraged, irate, insulted, appalled,provoked, indignant, furious, inflamed,incensed
Disappointment-Sadness
Disappointment-Sadness satisfaction-happiness acceptance, moroseness, solace,melancholy
Disappointment-Sadness aversion-fear regret, forlornness, remorseful, miseryDisappointment-Sadness assertion-anger aggrieved, discontent, dissatisfied,
unfulfilled, boredom, grief, envy,sullenness
Disappointment-Sadness startlement-surprise crestfallen, shaken, dejected, stunned,upset
Startlement-Surprise
Startlement-Surprise satisfaction-happiness delightStartlement-Surprise aversion-fear shock, alarmed, unnerved, unsettled,
agitated, vigilantStartlement-Surprise assertion-anger disgust, upset, abhorrence, irritatedStartlement-Surprise disappointment-sadness dejection, incredulity, dismay
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
15/30
Sociological Theory of Emotions 147
experiences, then satisfaction is activated; and if expectations for negative
experiences are met, then relatively low level variants of disappointment-sadness
are likely to ensue.
Aside from this gross relationship between the level of emotional arousal
and the congruity/incongruity between expectations and experiences, anotherprimary emotionstartlement-surprisecan increase the intensity of the emo-
tions aroused. Of course, incongruity between expectations and experience is,
per se, a type of surprise; and so surprise is always involved in emotional
reactions. Yet, there can be varying degrees of surprise, and the more one is
surprised by an incongruity the greater is the intensity of the emotions
experienced, at least in the short term. Part of the reason for this enhanced
intensity is that surprise activates motor and visceral system responses in ways
that feed back and feed forward to stimulate subcortical emotion systems as well
as cortical attention to stimuli (Le Doux, 1996; Damasio, 1994); another part is
that surprise highlights the incongruity between expectations and experiences;
and most significantly, surprise makes individuals more self aware, and as a
consequence, surprise increases the salience of self. And, escalating the visibility
of self in a situation raises the emotional stakes. Even when self is not originally
salient, surprise as it activates other emotions pulls at least situational selves into
the emotional response; and if more core cognitions-emotions about self become
salient, then the emotional intensity increases considerably. Some have argued,
it should be noted, that surprise may not be a primary emotion; yet, whatever
its status in this respect, surprise as a primary or learned response intensifies
emotional experiences.
Self is always hightly salient if individuals expect reinforcing and confirming
responses from others. Thus, while self can be pulled into the emotional
dynamics, thereby increasing the emotional valences involved, interactions that
begin with a high sense of self as an object in relation to others will produce
more intense emotional reactions when incongruity occurs. When self is not
confirmed, individuals will experience higher levels of assertion-anger, aversion-
fear and disappointment-sadness; when self is viewed negatively but actuallyreinforced positively, the sense of relief (and other mixtures of fear, satisfaction,
and sadness) will be more acutely experienced (see Tables 2 and 3); when more
positive views of self are confirmed, the sense of satisfaction will move to the
happiness side of this primary emotion.
As anticipated earlier, a key factor in increasing the salience of self and the
intensity of the emotions experienced is the level of direct sanctioning by others.
Often individuals simply do not realize their expectations or goals with no
sanctioning from others directly involved, but with active sanctioning by others,
whether on the positive or negative side, the emotional stakes are raised becausesanctions focus the responses of others on self, making self even more of an
object to itself. Thus, when congruities and incongruities between expectations
and experiences follow from active sanctioning by others, the emotions experi-
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
16/30
148 Jonathan H. Turner
enced will be more intense. If others praise self, satisfaction-happiness is more
acutely felt; and if such praise is a surprise and exceeds expectations, then the
emotional reactions are escalated further. If others react negatively to behaviors
of self, then more intense mixes of assertion-anger, aversion-fear, and disappoint-
ment-sadness ensue; and if surprise is involved, then anger or fear is even morestrongly felt, often drowning out the sense of disappointment-sadness that also
accompanies negative sanctioning.
As noted earlier attribution processes are also involved in the relationship
among expectations, experiences, and emotions. If incongruity occurs, and the
incongruity is attributed by the individual to self, then the emotions experienced
are intensified, whether positively or negatively. If the incongruity is attributed
to the behaviors and motives of others (as a kind of internal attribution), and
the experience has been negative, then more heightened variants of assertion-
anger are aroused, often mixed with aversion-fear if the other is in a position
to apply negative sanctions. If the experience is positive, however, then other
will be viewed, and responded to, in terms of combinations and variants of
satisfaction-happiness. If the incongruity is attributed to situational forces without
reference to others or, if others are involved, to forces beyond their control,
then the emotions experienced will be lower key. Under these conditions of
external attribution, incongruities on the negative side will be more likely to
produce reactions on the disappointment end of this primay emotion and its
variants, and if anger and fear are activated, these too will have lower valances;
and incongruities on the positive side under conditions of external attribution
will qualify and mitigate happiness (see Tables 2 and 3).
To sum up, the emotional arousal stimulated by congruity or incongruity
between expectations and experiences is influenced by a number of variables:
the nature of expectations, the extent to which the incongruity exceeds or fails
to meet expectations, the extent to which surprise has been activated, the degree
to which self is salient and implicated in the experience of congruity or
incongruity, the extent to which congruity or incongruity is related to direct
sanctioning by others, and the nature of the attributions made for the cause ofthe incongruity or congruity. Yet, these variables interact in complex ways in
generating emotions, per se, and in activating defense mechanisms that distort
emotional experiences. Moreover, there are complex feedback and feed forward
relations with subcortical systems involved in the production of emotional
responses. At this point, then, we can only see the generalizations presented
above as first approximations of a very complicated process.
EMOTIONAL AROUSAL
Emotional arousal always involves activation of at least one, and typically several,
primary emotions. This arousal begins subcortically, as sensory inputs move to
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
17/30
Sociological Theory of Emotions 149
the thalamus which then activates the hypothalamus as well as various limbic
systems, such as the amygdala, septum, and cingulate gyrus; but as working
memory experiences these subcortical inputs, while activating coded neurons to
restimulate sensory cortices to reproduce images from long-term memory,
individuals experience feelings organized in terms of variants and combinationsof primary emotions. Following Kempers (1987) review of the large literature
on the primary emotions, or what can be reasonably seen as biologically-
based emotions in humans, five such emotions appear to be hard-wired into
humans neuroanatomy and, indirectly, into the endocrine system (Ekman, 1982).
Table 2 is my own adaptation of the literature on primary emotions (see
Turner, 2000, for a review). As is evident I see each primary emotion as
potentially moving from a low to a higher intensity state: satisfaction to happiness,
aversion to fear, assertion to anger, disappointment to sadness, and startlement
to surprise. This is not an issue of mere labels, because the low intensity variant
(the left side of each pair in Table 2) is, I believe, the more prevalent biological
state for most mammals; in contrast, humans and other higher mammals have
the neurological capacity to elaborate this lower intensity state to a much higher
level of amplitude. Thus, for humans, satisfaction can move from contentment
through buoyancy and cheerfulness to joyfulness and, as we will see in Table 3,
to many first-order combinations of primary emotions, such as pride; aversion
moves from hesitancy through anxiety and trepidation to terror and many first-
order combinations as well as becoming implicated in complex second-order
emotions like shame and guilt as is delineated in Table 4; assertiveness goes
through such moderate states as displeasure, frustration, and hostility to high
intensity states like loathing, disgust, and hate, while becoming implicated in the
many first-and second-order combinations listed in Tables 3 and 4; disappoint-
ment moves from a sense of being discouraged through pain and dismay to
sorrow and anguish as well as many first- and second-order combinations;
startlement goes from sudden awareness through sudden alertness to amazement
and incredulity, while being implicated in more complex combinations of
emotions and, as emphisized earlier, in the intensification of all emotions.As a basic hypothesis, I would argue that when expectations and experiences
are congruent, whether activating positive or negative emotions, the low intensity
state is the most likely node of activationthat is, satisfaction, mild aversion, mild
assertiveness, disappointment, and startlement. Conversely, as the incongruence
between expectations and experiences increases, especially as self becomes ever
more salient, the high intensity end of these emotions is activated, as are the
various first- and second-order combinations portrayed in Tables 3 and 4.
As is evident in Table 3, I see not only variants but first-order combinations
of primary emotions as crucial. Following Kempers (1987) lead, emotions arecombined in first-order combinations revolving around the mixing of two
primary emotions in varying proportions or ratios (Turner, 1996a, 1996b, 1997a,
1997b, 1999, 2000). Thus, the emotions that humans use in their role taking
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
18/30
150 Jonathan H. Turner
and role making efforts are variants of primary emotions and combinations of
these basic emotions; it is the capacity to combine primary emotions that makes
humans emotional repertoire so large compared to other animals. Table 3
presents my current candidates for some first-order combinations.
In Table 3, each of the primary emotions is combined with another primaryemotion to form a first-order combination. This process is very much like the
mixing of colors, as Plutchik (1980, 1993) has made explicit in his theory of
basic and derived emotions, but I do not think that the mixing is quite as orderly
and systematic as implied by Plutchiks emotions wheel. Rather, the neural
mechanisms that perform the mixing are, themselves, complex systems, revolving
at a minimum around feed forward and feed back relations among (1) thalamus
and hypothalamus, (2) cortical and subcortical motor areas, (3) subcortical visceral
responses and neurotransmitters like dopamine, norepinephrine, serotonin, and
acetylcholine, (4) subcortical emotion centers such as the septum, cingulate
gyrus, and amygdala, (5) basal forebrain, (6) hippocampus, and (7) neocortex.
Thus, the neurological capacity to generate and mix emotions is vast, far greater
than the actual numbers and range of emotions that structure human interaction.
The neuro substrate of emotions is an important topic itself, but for my purposes
here, the critical point is that humans emotional repertoire revolves around a
complex process of activating interconnected systems in the brain and the body
in ways that generate mixes among primary emotions.
Table 3 lists combinations that involve one dominant primary emotion mixed
with lesser but varying proportions of a second primary emotion. For example,
when the more intense variant of satisfaction-happiness is dominant, but mixed
with varying levels of aversion-fear, emotions such as wonder, hopefulness, relief,
gratitude, reverence and, most importantly, pride are generated. As a first-order
combination of mostly happiness (toward self as an object), tinged with a small
amount of fear (about potential failure), pride needs to be emphasized because,
following Cooley (1916) and Scheff (1988, 1990), its activation will be very
important in understanding self and the relation of self to other emotional states.
By looking down the right side of Table 3 under stiuations where one primaryemotion is dominant, the varieties of first-order combinations of emotions can
be seen. I am not sure that the combinations presented in Table 3 are exactly
right, but the general point is that humans have the neurological capacity in
their covert thoughts, in their self presentations and role making, in their self
talk, and in their role taking to draw upon a very large repertoire of first-order
combinations of primary emotions, thereby giving human interaction, thought,
decision making, sanctioning, and moral coding a subtle and fine-tuned quality.
But one critical difference with Kempers and my earlier work with first-order
combinations is that some of the most important emotions in human interactionare what I term second-order combinations of three primary emotions mixed in
different ratios. As Table 4 emphasizes, the emotions of shame and guilt are the
most important combinations of three primary emotions. Shame is an emotion
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
19/30
Sociological Theory of Emotions 151
Table 4. The Structure of Shame and Guilt
Rank-order of primary emotions Second-order emotionsShame
1 Disappointment-sadness (at self)2 Assertion-Anger (at self)3 Aversion-fear (about consequences to self)
Guilt
1 Disappointment-sadness (at self)2 Aversion-fear (about consequences to self)3 Assertion-anger (at self)
in which self is highly salient; the most prominent emotion in shame is
disappointment-sadness at self for behaving inadequately in the eyes of others
and in ones own self-assessments, coupled with anger at self and some fear
about the consequences to self for having engaged in behaviors are seen as
inadequate and, perhaps, incompetent. Guilt activates the same three primary
emotions but in a slightly different combination: the dominant emotion is still
disappointment-sadness at self for having failed to meet expectations, but the
second most important emotion is now fear about the consequences of this
failure, coupled with some anger at self for having so failed. Many have argued
(e.g., Lewis, 1971; Scheff, 1988) that guilt is simply extreme shame, but the
subjective feelings for shame and guilt are quite different, especially in extreme
form. Others have suggested different ways to combine or distinguish shame,
guilt, and related emotions like embrarrassment (e.g., Sabini & Silver, 1997,
1998; Tangney, et al., 1996). I may be wrong in this conclusion, but the general
point remains the same: shame and guilt are at least a second-order emotion,
mixing three primary emotions; and the different rankings among the primary
emotions give shame and guilt distinctive guidelines and make their experience
by an individual distinctive. Moreover, the causes of the two emotions vary:guilt is activated with failure to meet moral expectations, while shame is
generated when one has behaved incompetently. Because shame and guilt are
such unpleasant emotions, they are very likely to mobilize defense mechanisms.
Since disappointment-sadness is the dominant emotion in shame and guilt,
the depression that often accompanies shamed and guilty individuals may not
reflect the use of repression and other defense mechanisms; rather, the depression
may simply be an artifact of the dominance of disappointment-sadness in both
shame and guilt. Yet, as I will argue below, the activation of defense mechanisms
accelerates and strengthens this depressive effect. For, it takes a great deal ofemotional energy to repress; and when one uses other defense mechanisms, such
as attributions to others and situations, projections onto others of ones own
emotions, and displacement of the anger and fear contained in shame and guilt
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
20/30
152 Jonathan H. Turner
onto others, one also tends to receive back negative sanctions, making self feel
even more disappointed and angry as well as fearful about the consequences of
ones actions. In the end, the effort to control these unpleasant feelings and
responses of others lowers the overt emotional energy in interaction.
ACTIVATION OF DEFENSE MECHANISMS
Subcortical systems generating emotions are, to a degree, controlled by the
centers for rational and purposive thinking in the neocortex, especially the
prefrontal cortex; and this was probably so long before Homo sapiens sapiens
emerged on the hominid line. And it may well be that the hominid neocortex
became ever larger over the last few million years not so much for language-
production and comprehension as for cortical control of hominids expanding
emotional capacities.
Such control enables humans to present self and fronts in ways that do not
disrupt the flow of interaction; and in fact, cortical control of emotions as well
as the inadvertent leaking of emotions from such control facilitates interpersonal
attunement and sanctioning. Yet, as hominids ability for seeing self as an object
(an ability that is probably an unselected by-product of an expanding of the
neocortex, especially the prefrontal neocortex), cortical control could also be
used to defend self from particularly unpleasant emotions, especially those
building upon and combining fear, anger, and sadness. If surprise was also
involved, neocortical efforts to protect self from unpleasant emotions would
simply intensify. Emotional variants of happiness and first-order combinations
such as pride did not generate such protective reactions, and so the cortical
system of defense that evolved was designed to protect individuals from highly-
charged elaborations and combinations of fear, anger, and sadness. As a relevant
aside, I think that humans can read in others anger and fear as well as all of its
variants far more readily and subtly than they can read satisfaction-happiness
because these are the emotions that bring pain. Such emotions are activated bynegative sanctions from others as well as failures to meet expectations; and as a
result, these emotions activate powerful subcortical limbic responses (principally,
the systems revolving around the amygdala) which, in turn, can activate efforts
at cortical control.
The most commonly discussed defense mechanismsrepression, projection,
and displacementare simply ways to protect self from experiencing negative
emotions, particularly second-order emotions such as shame and guilt (Freud,
1900 [1960]; Horney, 1950; Sullivan, 1953). Ironically, these are also the very
emotions that are essential for social control, because if individuals do not feelshame at themselves for disassociative and inadequate behaviors or for failing
to meet order-sustaining expectations as well as guilt for such behaviors, social
organization would not be possible. The world would be filled with sociopaths
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
21/30
Sociological Theory of Emotions 153
who would soon become extinct. Thus, defense mechanisms are maladaptive to
the social order only if they keep individuals from experiencing those very
emotions that are essential to the smooth flow of interaction, while being
maladaptive to the individual only if they depress emotional energy and lower
the capacity to role take and role make effectively. Although the causes of shameand guilt as well as the subjective experience of these emotions differ, they are
both subject to the same defense mechanisms. Let me offer some general and,
I grant, fairly obvious generalizations as they relate to understanding defense
mechanisms and emotional energy in human interaction.
Activation of defense mechanisms such as projection and displacement in the
face of either guilt and shame tends to be counterproductive in the long run
because use of these mechanisms invites negative sanctions from others. If an
individual is seen as imputing his or her own negative emotions on others or as
venting on others negative emotions, especially versions of anger, people will
respond in ways that increase an individuals sense of shame and guilt, or
alternatively, force a person to repress to an even greater degree. Repression is
often necessary before other defense mechanisms can be activated, but it also
operates by itself. Highly repressed individuals often have episodes of emotional
outburst, typically anger but anxiety as well, which once again invite negative
responses from others, thereby heightening the shame for engaging in such
outbursts and the guilt over having done so.
As Lewis (1971) and Scheff (1988) have emphasized, repression operates at a
number of levels, from simply bypassing shame and not acknowledging it to
burying negative feelings about self from conscious reflection. If individuals
habitually bypass shame, and I would add, guilt, as well, repressive responses
must become ever more mobilized. The end result is that persons are drained
of emotional energy, except for disproportionate and typically inappropriate
outburst of emotion, especially anger but fear as well, that escape the cortical
censors. These emotional outbursts only feed back and require the censors to
increase their pressure which, in turn, lowers peoples emotional energy in ways
that can disrupt interaction. A biography of such cycles will, in the end, decreasethe modal levels of emotional energy generated by individuals, even when they
are in situations, such as those giving them power and authority, that normally
produce positive emotional arousal.
Self thought and self talk are important elements in these dynamics. If second-
order variants of primary emotions causing shame and guilt are only experienced
in the brains mode of processing information as images and gestalts, there is a
greater likelihood that the emotional arousal will go unacknowledged or that
more active repression will occur. The key to dealing with negative emotions is
self talk, where the blur of cognitive-emotional images can be slowed down formore careful inspection and introspection. Of course, the goal of most therapy
is to get individuals to engage in such self talk, but the difficulties that therapists
have are really built into the very nature of human thinking.
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
22/30
154 Jonathan H. Turner
The key to using self talk effectively is seeing the sources of emotional reactions
and recognizing the differences between guilt and shame. Ultimately, these
sources are the expectations listed in Table 1 as well as the degree to which
experiences, especially those which make self highly salient, correspond to
expectations. If self talk can be reflectively analytical, seeing incongruities forwhat they are and making selfattributions about failures to meet expectations,
then the emotional reactions can be culled from the brains mode of organizing
cognitions in terms of fields and gestalts. In turn, if open and honest, such
reflection can stave off the activation of defense mechanisms, with the result
that the negative emotions experienced, particularly anxiety, shame and guilt,
can be managed and become a part of an individuals cultural capital and
used to more effectively present self and role make in future interactions. Such
analytical reflection, as it creates cultural capital, raises emotional energy in
subsequent interactions. One way in which analytical self reflection works is to
lower or alter expectations, and as a result, the emotional stakes are lessened
because incongruities between expectations and experiences, even those where
self is salient, will be either eliminated or, at the very least, greatly reduced.
Self talk can also focus on more positive emotions, especially those variants
and combinations of primary emotions that generate pride. Pride not only
enhances peoples sense of well being, it charges emotional energy because pride,
like its more negative counterparts, becomes a part of the cultural capital of
individuals that can be invested in interactions (Collins, 1988). Indeed, high
levels of pride are often necessary for individuals to engage in self talk about
shame or guilt, and under ideal circumstances, frank self talk about shame or
guilt can become yet another source of pride. For this to occur, however, a
persons cultural capital must have an existing reservoir of prideful cognitions
accumulated in past interactions of a given type or, more generally, in his or
her cumulative biography. Highly depressed individuals, who have cultural
capital filled with shame and guilt, generally have such lowered pride that they
cannot engage in effective self talk; and indeed, their self talk can often turn
into self-recriminations which, in the end, will reactivate defense mechanisms.Moreover, self talk can also be used for self deception; and as a result,
individuals can simply employ talk to impose a defensive regime that protects
self. When confronted with shame and guilt, especially a biography of bypassed
and repressed experiences, talk can make external or internal attributions of
emotions, projecting them onto particular people, categories of people, and
situations; self talk can justify displacement, once again through internal and
external attributions to others and situations; and self talk can jabber around
guilt and shame in order to drown out awareness of those emotions that have
been pushed below conscious reflection (such individuals are often compulsivetalkers not only to themselves but to others as well).
If humans did not have the capacity to bypass minor moments of shame and
guilt, however, they would be overwhelmed by emotions. Any given moment
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
23/30
Sociological Theory of Emotions 155
of emotionally-charged interaction or any legacy of accumulated emotional
experiences in an individuals biography would make a person too emotionally
charged to function effectively in interactions. Thus, defense mechanisms evolved,
no doubt, as part of the neocortical package to keep emotions from getting out
of control and for enabling individuals to use them in associative ways. Moreover,most emotions are read visually, bypassing the translation boxes for speech
comprehension and production, and feeding directly into the brains mode of
organizing memories and images, thereby keeping emotions from becoming
ready topics of slow self talk. Again, this is a useful way to avoid being
overwhelmed by emotions. It is, therefore, a matter of how active defense
mechanisms are. As a rule of thumb, if the activation of defense mechanisms
and the natural operation of the brain in terms of imaging prevent individuals
from retrieving painful cognitions and making them available for self talk, a
decrease in modal levels of emotional energy will ensue.
CONCLUSIONS: STATING THE THEORY FORMALLY
I have delayed to the end presenting the theory because the dynamics of
emotions discussed thus far are interrelated in direct, indirect, and reverse causal
patterns, thereby requiring that I first describe the operative processes behind
the labels presented in Figure 1 before pulling these together into a more formal
and parsimonious theory.
The propositions enumerated below formalize the basic lines of argument
stated discursively thus far. Given the highly speculative nature of the argument,
however, these propositions should only be considered tentative hypotheses,
although they do summarize various empirical findings from well-known
theoretical traditions. As speculative extensions of existing theories and research
findings on these theories, the proposition do not represent a refined theory.
Rather, the propositions are only a first step in trying to develop a more general
sociological theory of emotions in interpersonal encounters. When stated aspropositions, however, the argument can more readily be assessed and criticized
in light of other theoretical positions or empirical findings, thereby paving the
way for correction and increased refinement. I offer these sixteen propositions
in the hopes of stimulating critique and theoretical cumulation.
1. The greater is the number of expectations that form an individuals
definition of a situation and the higher is the salience of these expectations
in an encounter, the greater is the potential for emotional arousal by an
individual and for affective displays of emotions to others in the encounter.2. If individuals have received positive sanctions from others and have met
or exceeded expectations, or both, in an encounter, they will experience at
least satisfaction and more typically happiness; and, as a consequence, they
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
24/30
156 Jonathan H. Turner
will display associative affect towards others and will receive similar affect
from others in the encounter.
2a. If self is salient in an encounter and if attributions for success in
meeting expectations and/or for receiving positive sanctions are placed
on self, then these individuals will experience pride; and, as a consequence,they will seek to develop and sustain associative relations with others in
this encounter.
2b. If attributions for success in meeting expectations and for receiving
positive sanctions are attributed in an encounter to the internal states of
others or categories of others, then these individuals will experience
variants of satisfaction-happiness; and, as a consequence, they will display
towards and receive from others these positive emotions, and, thereby,
develop and sustain associative relations with others in this encounter.
2c. If attributions for success in meeting expectations and for receiving
positive sanctions are attributed in an encounter to external causes in the
situation, then the level of satisfaction-happiness expressed by individuals
towards others will be lower, but nonetheless, individuals will still
seek to interact and to create associative relations with others in the
encounter.
3. Both hapiness and pride under conditions of positive attribution increase
the level of associative affect emitted by an individual towards others in an
encounter who, reciprocally, will tend to increase their display of affect to
the individual in an escalating cycle of ever-more associative and emotionally-
charged relations until fatigue and satiation set in, or the encounter is
terminated.
4. Self talk about interactions in which variants and combinations of
satisfaction, happiness, and pride have been experienced will sustain the
positive flow of emotions previously displayed and become a part of
individuals cultural capital which, in turn, will set expectations for the
subsequent interaction, making individuals more inclined to pursue associat-
ive interaction with particular others in encounters now and in the future;and, as a consequence, the level of associative affect displayed towards and
received from others will increase in these encounters.
5. Such self talk of variants and combinations of satisfaction-happiness will
often be generalized to similar others or similar encounters, thereby raising
the expectations for positive emotions and making individuals more likely
to seek associative relations with others or to enter encounters like those
which, in the pasts, were rewarding to the individual; and, as a consequence,
the level of associative affect displayed towards others and received from
others will increase in such encounters.6. If, however, individuals receive negative sanctions from others or are
unsuccessful in meeting expectations, or both, in an encounter, they will
experience varying combinations of assertion-anger, aversion-fear, and
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
25/30
Sociological Theory of Emotions 157
disappointment-sadness, which, in turn, will activate the converse of processes
(2) through (5) above.
6a. If self is salient in an encounter and if attributions by individuals for
failure to behave competently and for receiving associated negative
sanctions are placed on self, then these individuals will experience shamewhich will eventually lower their modal level of emotional arousal; and,
as a consequence, their modal level of affective display will decline and,
thereby, dampen the associative affect in relations with others in the
encounter.
6b. If self is salient in an encounter and if attributions by individuals for
failure to meet normative expectations and for receiving associated
negative sanctions are placed on self, then these individuals will experience
guilt; and as a consequence their modal level of emotional arousal will
decline and, thereby, dampen the associative affect in relations with others.
6c. If attributions for failure to meet expectations and for receiving
negative sanctions in an encounter are attributed by individuals to the
internal states of others or categories of others, then variants of assertion-
anger toward these others or categories of others will be aroused and
displayed by these individuals up to the point of fatigue, or escalated
negative sanctions by others, or termination of the encounter.
6d. If attributions for failure to meet expectations and for receiving
sanctions in an encounter are attributed by individuals to external causes
in the situation, then variants of assertion-anger will be less intense and
the level of negative emotional arousal and display by these individuals
will be relatively low; and, as a consequence, they will sustain relations
with lessened associative affect with others in the encounter.
7. All of the effects described in (2) through (6) will be magnified under
conditions of increasing surprise; and the greater is the level of surprise
experienced by individuals, the higher wil be their levels of emotional
arousal and display in an encounter.
8. All of the effects listed in (1) through (7) will feed back and changeexpectations; and the more intense are the emotions experienced by
individuals, whether positive or negative, the greater will be the change in
their expectations for subsequent interactions in similar encounters.
9. The more expectations escalate among individuals in an encounter,
whether positive or negative, the more likely are incongruities to emerge in
subsequent encounters, thereby raising the level of emotional arousal and
display towards others in subsequent interactions in similar encounters.
10. To the degree that defense mechanisms have been used by individuals
in an encounter to repress disappointment-sadness, aversion-fear, andassertion-anger, especially as these combine to produce shame and guilt,
then the modal level of overt emotional energy arousal by these individuals
will be lowered in this encounter; and, as a consequence, the level of
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
26/30
158 Jonathan H. Turner
associative affect displayed towards others and received from others will be
correspondingly dampened in the encounter.
11. To the degree that defense mechanisms have been used habitually and
over a long period in an individuals biography, then the modal level of
emotional displays by this individual will be low; and, as a consequence,the individuals level of associative affect displayed towards others and
received from others will be dampened in all encounters.
12. To the extent that attributions for disappointment-sadness, aversion-
fear, and assertion-anger are directed by individuals toward others and
categories of others in an encounter, their modal level of overt emotional
arousal and display will increase, especially episodes of anxiety and anger,
up to the point where fatigue, shame, or guilt about such responses reduce
the level of emotional display; and the more negative sanctions from others
lead these individuals to experience shame and guilt, the more powerful
will be the effect of shame and guilt on reducing their emotional displays
toward others in the encounter.
13. To the extent that disappointment-sadness, aversion-fear, and assertion-
aggression are projected or displaced by individuals on others, the more
likely are such individuals to experience negative sanctions from these
others; and, as a consequence, the more likely are the dynamics in (6a6d)
and (12) above are to be operative.
14. To the degree that the processes described in (12) and (13) have occurred
over a long period of time and typify the biography of an individual, the
less will be this individuals modal level of emotional display in an encounter
when not engaged in periodic outbursts of assertive displays revolving
around attribution, projection, and displacement; and, as a consequence,
this individuals modal level of associative affect displayed towards and
received from others will be dampened.
15. To the extent that individuals can engage in analytical self talk that
correctly targets the sources for their emotional experiences in encounters
whether the target be self, others, or situationsthe greater will be theirsense of satisfaction and, hence, the more this sense of satisfaction will
mitigate the dampening affects of negative emotions and, thereby, raise
these individuals modal level of positive emotional arousal in encounters;
and, as a consequence, they will display toward others and receive from
others associative affect in all encounters.
16. If self talk by individuals is effective in reducing negative emotional
arousal, especially those negative emotions that have accumulated over
longer periods of time and that revolve around anxiety, guilt, and shame,
the more likely will these individuals experience pride in encounters, andhence, the greater will be these individuals modal level of positive emotional
arousal; and, as a consequence, they will display toward others and receive
from others associative affect in all encounters.
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
27/30
Sociological Theory of Emotions 159
Jonathan A. Turner
Department of Sociology
University of California, Riverside
Riverside, CA 92521-0719
USA
REFERENCES
BACKMAN, C.W. (1985). Identity, Self-presentation, and the Resolution of Moral Dilemmas:Toward A Social Psychological Theory of Moral Behavior. In The Self and Social Life,ed. B.R. Schlenker. New York: McGraw-Hill, pp. 23656.
BACKMAN, C.W. (1988). The Self: A Dialectical Approach. Advances in Experimental Social
Psychology, 21, 22960.BERGER, J. (1988). Direction in Expectation States Research. In Status Generalization: New
Theory and Research, eds. M. Webster and M. Foschi. Stanford, CA: Stanford UniversityPress, pp. 45074.
BERGER, J . & ZELDITCH M. Jr. (1985). Status Rewards and Influence. San Francisco: Jossey-Bass.BERGER, J . & ZELDITCH M. Jr. (1988). Status, Power, and Legitimacy. New Brunswick:
Transaction Books.BERGER, J., WEBSTER, M.A. Jr., Ridgeway, C. & ROSENHOLTZ, S.J. (1986). Status Cues,
Expectations, and Behavior. Advances in Group Processes, 3, 122.BURKE, P.J. (1980). The Self: Measurement Implications From a Symbolic Interactionist
Perspective. Social Psychology Quarterly, 43, 1829.
BURKE, P.J. (1991). Identity Processes and Social Stress. American Sociological Review, 56,83649.
COLLINS, R. (1975). Conflict Sociology: Toward An Explanatory Social Science. New York: Academic Press.
COLLINS, R. (1988). Theoretical Sociology. San Diego: Harcourt Brace Jovanovich.COLLINS, R. (1993). Emotional Energy and the Common Denominator of Rational
Action. Rationality and Society 5, 20330.COOLEY, C.H. (1916). Social Organization: A Study of the Larger Mind. New York: Scribners.DAMASIO, A. (1994). Descartes Error: Emotion, Reason, and the Human Brain. New York:
G.P. Putnam.EKMAN, P. (1982). Emotions in the Human Face. Cambridge, UK: Cambridge University Press.
FREUD, Sigmund. (1990) [1960]. The Interpretation of Dreams. New York: Wiley.GECAS, V. (1989). The Social Psychology of Self-efficacy. Annual Review of Sociology, 8, 133.GECAS, V. (1991). The Self Concept as a Basis for A Theory of Motivation. In The Self-
Society Dynamic: Cognition, Emotion and Action, eds. J.A. Howard and P.L. Callero.Cambridge, UK: Cambridge University Press, pp. 17188.
GESCHWIND, N. (1965a). Disconnection Syndromes in Animals and Man, Part I. Brain,88, 23794.
GESCHWIND, N. (1965b). Disconnection Syndromes in Animals and Man, Part II. Brain,88, 585644.
GESCHWIND, N. & DAMASIO, A. (1984). The Neural Basis of Language. Annual Review of Neuroscience, 7, 12747.
GOFFMAN, E. (1959). The Presentation of Self in Everyday Life. Garden City, NY: Anchor Books.GOFFMAN, E. (1961). Encounters. Indianapolis, IN: Bobbs-Merrill.GOFFMAN, E. (1967). Interaction Ritual: Essays on Face-to-Face Behavior. Garden City, NY:
Anchor Books.
The Executive Management Committee/Blackwell Publishers Ltd. 1999
8/7/2019 TURNER Toward a General Sociological Theory of Emotion
28/30
160 Jonathan H. Turner
GOFFMAN, E. (1974). Frame Analysis: An Essay on the Organization of Experience. New York:Harper and Row.
HEISE, D. (1979). Understanding Events. New York: Cambridge University Press.HEISE, D. (1989). Effects of Emotion Displays on Social Identification. Social Psychology
Quarterly, 53, 1021.
HEISE, D . & SMITH-LOVIN, L. (1981). Impressions of Goodness, Powerfulness, andLiveliness From Discerned Social Events. Social Psychology Quarterly, 44, 93106.
HIGGINS, E.T. (1987). Selfdiscrepancy: A Theory Relating Self and Affect. PsychologicalReview, 94, 31940.
HIGGINS, E.T. (1989). Self-discrepancy Theory: What Patterns of Self-beliefs CausePeople to Suffer? Advances in Experimental Social Psychology, 22, 93136.
HOCHSCHILD, A.R. (1979). Emotion Work, Feeling Rules and Social Structure. AmericanJournal of Sociology, 85, 55175.
HOCHSCHILD, A.R. (1983). The Managed Heart: Commercialization of Human Feeling. Berkeley:University of California Press.
HOMANS, G.C. (1961). Social Behavior: Its Elementary Forms. New York: Harcourt Brace.HOMANS, G.C. (1974). Social Behavior: Its Elementary Forms, revised edition. New York:
Harcourt Brace Jovanovich.HORNEY, K. (1950). Neurosis and Human Growth: The Struggle Toward Self Realization. New
York: Norton.KEMPER, T.D. (1978). An Interactional Theory of Emotions. New York: Wiley.KEMPER, T.D. (1987). How Many Emotions Are There? Wedding The Socia