A variety of theories have developed in the study of domestic violence; these theories can be categorized into three general perspectives: (1) individualist, (2) interactional, and (3) sociocultural (Miller et al. 1999). Individualist theories focus on characteristics of the perpetrator or victim of domestic violence, but they do not address how the offender and injured party interact to produce a violent relationship. Interactional theories locate the cause of domestic violence within the interaction. Issues that are examined include how attached actors are to one another, what is exchanged, who has power, and the meanings individuals attribute to themselves and others in the situation. Sociocultural theories address how culture and social norms foster or, alternatively, discourage the use of violence to resolve conflict. For example, the ideology of patriarchy, or the belief that men should be dominant in a society, encourages the use of force by husbands, fathers, brothers, and sons when they do not have power or control in a situation.
A focus on identity theory grows out of symbolic interaction, particularly structural symbolic interaction (Stryker 2002), thus it falls within the second theoretical perspective above. In structural symbolic interaction, society is patterned and organized, and this organization can be seen within any one individual and between individuals as they interact. Within any one person, the self is conceptualized as organized into multiple parts or identities, arranged in an overall hierarchy (Stryker 2002). In the early development of identity theory, each identity was seen as the internal component of a role that one occupied in the social structure; thus individuals had role identities. Role identities were the set of meanings attached to the self while in different roles. Later developments in identity theory opened the analysis of identities to group identities (self-meanings while a member of different groups) and person identities (self-meanings that identified the self as distinct and unique from others) (Burke 2004; Stets and Burke 2000). Identities develop in interaction with others. Persons come to see themselves as they believe others see them. They then act toward other persons, and themselves, based on these meanings.
Identity theory provides an important avenue for theoretical development in domestic violence research because all behavior, including aggression, is rooted in issues of self and identity. To understand aggression, we need to understand the meanings individuals attribute to themselves in a situation, that is, their self-definitions or identities. In all interactions, the goal of individuals is to confirm their identities. When their identities are not confirmed, persons may control others in the situation to make them respond differently in order to confirm their identities. If control does not work, aggression may be used as a last resort to obtain control and, in turn, confirmation of identity (Stets and Burke 2005). Thus, identity theory can help explain domestic violence by showing how a lack of identity confirmation at the individual level is tied to the control process and aggression at the interactive level. This will be discussed in greater detail later. First, identity theory will be reviewed.
I. Identity Theory
II. Identity Control Theory
A. Early Developments in ICT
B. Domestic Violence Research from Early Developments in ICT
C. Further Developments in ICT
D. Domestic Violence Research from Recent Developments in ICT
III. Future Research
Identity theory has three slightly different versions which focus on different aspects of self and identity (Stets 2006). One of the earliest versions, developed by McCall and Simmons (1978), addresses how identities get accomplished in an interaction through negotiation with others. Individuals may have different interpretations or meanings of the same identity, and they must work out how to resolve seemingly contradictory meanings so that interaction can proceed smoothly. A second version of identity theory is in the work of Stryker and his associates (Stryker 2002; Stryker and Serpe 1982, 1994). They emphasize how the social structure influences the identity that one invokes in a situation. The assumption is that social actors are tied to diverse social networks in society, and these networks are premised on particular identities being maintained within and across situations. Thoits’s work (1991, 2003) also has this emphasis. The most active program of research is in the third version of identity theory by Burke and colleagues (Burke 1991, 2004; Burke and Reitzes 1991; Cast, Stets and Burke 1999; Stets and Burke 2000, 2005. These researchers focus on the internal dynamics of the self that emerge when an identity is invoked in a situation, as well as the interactional consequences of those dynamics. In particular, a perceptual control system is offered as a theoretical way of understanding the operations of the self when an identity is activated.
To understand the different versions of identity theory as outlined above, consider how each can be used to explain behavior in a situation. McCall and Simmons would argue that behavior in an interaction is a function of individuals attempting to fit the meanings of their identity with the meanings of the identity of others in the situation. Indeed, every identity in an interaction is understood as it relates to a counter-identity. For example, the identity of ‘father’ depends on the identity of a child (the counter-identity of the father) to engage in play activity. The identity of ‘therapist’ needs the identity of a client in order for the therapist to make a diagnosis. The identity of ‘teacher’ requires the identity of a student so that the teacher may instruct. If conflict emerges when identities interact in a situation, negotiation strategies will be used so that each person’s identity can be confirmed and interaction can proceed smoothly.
For Stryker and his associates, behavior in a situation corresponds to a particular identity being salient for the self in the situation. A salient identity is an identity that is high in one’s overall hierarchy of identities. More salient identities are more likely to be invoked across situations. Thus, behavior that corresponds to a salient identity will be more likely to be observed over time. An important factor that influences the salience of an identity is how committed one is to the identity. Greater commitment results when a person has deep network ties to a large social network premised on an identity.
Finally, Burke and his colleagues argue that behavior in a situation is an outcome of the relationship between how people see themselves in a situation and the identity standard that is invoked in the situation. When there is correspondence between peoples’ views of themselves in a situation and their identity standard view, identity verification exists: The meanings of behavior in the situation match the meanings in their identity standard. When there is a lack of identity verification, people behave differently in the situation to restore correspondence between the self-view and the identity standard view. This third perspective, which has come to be labeled identity control theory, or ICT (Burke 2004; Stets and Burke 2005, Stets and Tsushima 2001), has theoretically informed the domestic area the most.
Identity Control Theory
Early Developments in ICT
ICT began almost thirty years ago with the development of a way to measure people’s identity meanings. Since people choose behaviors whose meanings correspond to those of their identity (Burke and Reitzes 1991), identifying the meanings of an identity for individuals allows for the prediction of the meanings of their behavior. Burke and Tully (1977) developed a method for the measurement of the meanings of people’s gender identities. They gave respondents (boys and girls) a set of bipolar adjectives such as ‘‘weak strong,’’ ‘‘not at all emotional very emotional,’’ and ‘‘not at all competitive very competitive.’’ Respondents rated themselves along these adjective pairs to help locate their identity meanings. Then, through a statistical procedure known as discriminant function analysis, the items were selected that best discriminated between the meanings of different groups in the sample, and in Burke and Tully’s sample, this represented boys and girls. From this, a measure of gender identity was derived. For example, items that distinguished between girls and boys in Burke and Tully’s sample included being soft (versus hard), weak (versus strong), and emotional (versus not emotional).
Since any identity contains multiple meanings, multiple bipolar dimensions can exist for any one identity. In further analyses of gender identity, researchers found that femininity includes the multiple dimensions of noncompetitiveness (‘‘competitive not at all competitive’’), passivity (‘‘very active very passive’’), and sensitivity of feelings (‘‘feelings not easily hurt feelings easily hurt’’) (Burke, Stets, and Pirog-Good 1988; Stets and Burke 1996). Additionally, different people can have different meanings for the same identity. For example, while one woman may see herself in feminine terms as described above, another woman may see herself as less feminine and more masculine, as in being more competitive, active, and less sensitive in her feelings. What is important about the Burke-Tully procedure for measuring identities is that it uses the meanings of the individuals in a particular subpopulation to develop a particular identity measure, rather than using the researcher’s own view as to the meanings of an identity or the views from another subpopulation.
Domestic Violence Research from Early Developments in ICT
The discovery of a measure of gender identity was important because it helped in the investigation of how gender identity relates to physical and sexual aggression (Burke et al. 1988). The long-standing argument in the domestic violence area has been that violence is consistent with the masculine ideal; men are more likely to behave aggressively as a way of demonstrating their masculinity (Toby 1966). However, no research has actually evidenced this (Rosenbaum 1986). Burke and his colleagues tested this argument in a study of college students (Burke et al. 1988).
A random sample of college students was gathered. Their gender identity was measured using the Burke-Tully method as described above. Respondents’ physical and sexual experiences in their dating relationships were gathered. The results revealed that men and women with more feminine gender identities, that is, those who described themselves as noncompetitive, passive, and sensitive in their feelings, were more likely to inflict and sustain both physical and sexual aggression in their dating relationships. In explaining these findings, Burke and his associates (1988) argued that compared with those with more masculine gender identities, those with more feminine gender identities were: (1) more emotionally expressive and (2) more oriented to their dating relationships.
In terms of inflicting sexual aggression, the researchers argued that a greater orientation to the relationship may lead to wanting to be more involved with the other. To obtain greater involvement, more feminine individuals may initiate sexual activity, and if others do not desire this, their need to be more involved, coupled with their emotional excitability, may result in forcing the issue of sex. If their partners resist and individuals who are more feminine ignore this resistance, then their actions become sexual aggression. Similarly, emotional excitability can lead to physical aggression, particularly when individuals desire, but lack, control of a situation. The idea that one loses control and strikes out aggressively to get control is consistent with other theoretical and empirical work, which will be discussed below (Stets and Burke 2005).
Finally, Burke and colleagues (1988) point out that since research reveals that violence is reciprocal— that is, individuals ‘‘get what they give’’— individuals who are more feminine who inflict sexual and physical aggression will eventually sustain such aggression as well. What is interesting about this study is that it reveals how identity theory can explain not only male aggression but also female aggression. For both genders, it is a more feminine gender identity that helps researchers understand the violence that takes either a sexual or a physical form and is inflicted or sustained by the actors.
Further Developments in ICT
Further developments in identity theory expand on the idea of understanding people’s identity meanings and their corresponding behavior by conceptualizing the identity process as a perceptual control system, based on the work of Powers (1973). The theoretical development of identity theory, as it became formulated into ICT, argues that individuals are goal oriented; they are motivated to control perceptions of who they are in a situation so that these perceptions match their identity standard (Burke 1991, 2004). A correspondence between self-perceptions in a situation and identity-standard meanings results in identity verification and positive emotion. Noncorrespondence between self-insituation meanings and identity-standard meanings results in identity nonverification and negative emotion. This identity control system is described below.
An identity is a set of meanings that are attached to the self. It serves as the reference or standard for a person in situations. When an identity is activated in a situation, a feedback loop is established. This feedback loop has four components:
- The identity standard, or set of meanings defining who one is in a situation
- Perceptual input, or how one sees oneself, which is based in part on directly observing oneself and in part on feedback from others as to how they see the self in the situation
- The comparator, which compares the perceptual input with the identity standard and registers the degree of discrepancy between the two
- Output, or behavior, which is the result of the comparator. Behavior is modified if the comparator signals noncorrespondence between the input and the identity standard.
What is important about the identity process as outlined above is that instead of seeing behavior as strictly guided by self-meanings (the identity standard) or the meanings of self in the situation (perceptual input), behavior is the result of the relationship between identity-standard and self-insituation meanings. When the comparator registers no discrepancy between perceptions and the standard (a value of zero), this is identity verification, and no change in behavior is needed. As the discrepancy departs from zero and, correspondingly, identity nonverification increases—generally because one’s direct observations or feedback from others about the self in the situation do not match identity standard meanings—behavior changes to counteract the discrepancy. The goal is to realign perceptual input with the identity standard.
To illustrate the above, it is helpful to consider how it applies to one’s gender identity. A man’s self-meanings of masculinity may include being dominant, controlling, and aggressive. If at home he perceives that he is not as dominant, controlling, or aggressive as he feels he should be, or alternatively, his friends give him this feedback, then there is a discrepancy between his self-in-situation meanings and the identity standard. In response to this discrepancy, he may work harder to be more dominant, controlling, and aggressive and may use violence as the ultimate resource to realign perceptual input with identity-standard meanings.
An important assumption in ICT is that individuals seek situations in which their identities will be verified. They may choose to interact with others who they know will confirm their identities and avoid those who they know will not confirm their identities. They may even ‘‘act the part’’ by dressing a particular way or using a certain style of speech so that others recognize who they are and confirm their identities. When individuals act to verify not only their own identity, but also that of others in the situation, a ‘‘mutual verification context’’ exists (Burke and Stets 1999). Two or more individuals may act to mutually support each other’s identities. Disturbances in these situations are countered in order to protect and preserve the identities of the actors, the relationship in which the identities are embedded, and, by extension, the social structure in which the identity ultimately belongs. For example, a married couple often develops a mutual verification context in which each partner not only verifies his or her own spousal identity, but also acts to support and maintain the spouse’s identity. In turn, the relationship—and the institution of marriage more broadly—is maintained.
At the individual level, self-verification sustains the belief that one’s world is controllable (Swann 1990). As Pinel and Swann (2000, p. 133) have remarked, ‘‘Self-verifying evaluations are what the purr of the automobile is to the driver or the roar of the jet engine is to the pilot: a signal that all’s as it should be.’’ When others see the self in a verifying manner, it also provides an emotional anchor that leaves one less vulnerable to the slings and arrows of life events. Individuals know who they are, others come to know and support that view, and this keeps individuals on an even keel. These feelings get reverberated at the interactive level. Mutual verification contexts often produce very stable relationships and result in positive emotions and feelings of trust and commitment among individuals (Burke and Stets 1999).
Domestic Violence Research from Recent Developments in ICT
Identity nonverification threatens the maintenance of one’s own and others’ identities in a situation. It also threatens people’s sense of control over their environment. In response to identity nonverification, individuals may withdraw physically or psychologically from the interaction, may selectively dismiss nonverifying information or selectively recall verifying information, or may work harder in the interaction to counteract disturbances and seek a match between self-perceptions and their identity standard. If, in working harder, they are still unable to obtain identity verification, their sense of control over the environment will diminish further. In turn, they may increase their control over others in the situation so that others will respond in a way that verifies their identity and they regain the perception of control over the environment. Stets and Burke (2005) revealed how control provides the very seeds of domestic violence. They argued that when increased control in the situation fails to achieve identity verification, that is, one cannot get another to verify the self, aggression may be used as a last resort to regain control over the environment and obtain identity verification.
Individuals control others to compensate for a loss of control over the environment (Stets 1993, 1995). Since a major theme in the domestic violence literature during the 1990s has been the relationship between control and aggression (Johnson and Ferraro 2000), Stets and Burke’s research attempts to theoretically develop this relationship by showing how it is importantly influenced by the underlying process of identity verification.
Stets and Burke examined couples in the first two years of their marriage in two mid-sized communities in Washington State. They obtained an identity verification measure of each partner’s spousal identity by calculating the amount of agreement between an individual’s self-rating of how he or she should behave with respect to a series of spousal role activities (the spousal identity standard) and how the partner expected the person to behave with respect to these spousal role activities. Thus, each partner in the marriage had a self-rating and a rating of how the partner expected the person to behave in terms of the spousal identity. The greater the agreement between the self-rating and the rating of the other, the more there was identity verification for that individual. They also obtained measures of how much control each partner felt that they had over the environment, how much they controlled their partner in the marriage, and how much each was physically aggressive toward the other.
The results provided support for the finding that nonverification of a spousal identity reduced the self’s perceived control over the environment. In turn, the self increased acts of control over the partner. And, heightened control over the partner was associated with acts of aggression toward the partner. Unfortunately, Stets and Burke found that using aggression in an attempt to regain control was disruptive to later self-verification. Specifically, the use of aggression in one year significantly reduced verification of the spousal identity in the following year. Further, aggression ultimately led to a spiral of more aggression, since aggression in one year significantly influenced aggression in subsequent years. In this way, identity disruptions at the individual level threaten established relationships at the interactive level by influencing controlling and aggressive behavior, and such controlling and aggressive behavior jeopardizes stable social structural arrangements such as the institution of marriage.
Stets and Burke found that the link between identity verification, control, and aggression was more likely to predict minor aggression than severe aggression. Since minor aggression is less likely to cause serious injury compared with severe aggression, individuals may be more likely to use minor aggression as a strategy to regain control when it is lost. Severe aggression may be more likely to be interpreted as deviant, if not criminal, and its use may lead to the irretrievable breakdown of a relationship. Given the lesser costs associated with minor aggression, individuals may be more inclined to use it as a last resort.
In general, the findings of Stets and Burke reveal that if interaction is to continue smoothly, each person must act to verify not only his or her own identity, but the identity of the other in the situation. When identity verification is not forthcoming and a person engages in maladaptive behaviors such as controlling or behaving aggressively toward the partner in order to coerce identity verification, the person will find it even more disruptive and costly to the relationship both in the short and in the long run.
ICT is a rapidly developing area of study (Stets and Burke 2005), which includes application to domestic violence research. More research is needed to investigate whether the identity process can help one understand not only dating aggression and marital aggression but also child abuse, sibling violence, parental abuse, and elder abuse. Perhaps it might even help in the understanding of violence between strangers. To the extent that another person does not confirm who one is, one will work harder to obtain that confirmation, although the person may work harder for confirmation from someone with whom he or she is close, whose opinion matters, and with whom he or she will likely interact in the future.
If a lack of identity verification fosters increasing control over another and, in turn, leads to aggression, intervention strategies may need to be developed that teach individuals alternative ways of responding to nonverification. Since part of the identity processes within ICT are psychological, having to do with the self, including self-perceptions and others’ perceptions of the self, intervention may involve therapeutic efforts to understand the inner workings of the self. For example, one may be misinterpreting others’ feedback. Others may be verifying one’s identity but the self sees it as nonverifying, perhaps as a result of low self-esteem. Thus, these misinterpretations would need to be identified. Alternatively, others may be ignoring the self’s display of behaviors that correspond to his or her identity standard such that the self may have to point out the consistency to his or her audience in a clear way. Finally, nonverification feedback from others may imply that the self needs to change its identity standards. This change would likely occur at a very slow rate, if the person were open to it.
In general, ICT is a coherent, cumulative, and everdeveloping theory in social psychology that shows promise in explaining domestic violence. By focusing on the internal dynamics of the self as an identity control system, and the relation between those dynamics and the interactional dynamics in interpersonal settings, one can study how disrupted identities in situations produce aggression in interaction. The key is identifying the mechanisms that disrupt the identity process and finding ways to minimize the effects of these disruptions so that individuals do not resort to aggression.
- Burke, Peter J. ‘‘Identity Processes and Social Stress.’’ American Sociological Review 56 (1991): 836–849.
- ———. ‘‘Identities and Social Structure: The 2003 Cooley- Mead Award Address.’’ Social Psychology Quarterly 67 (2004): 5–15.
- Burke, Peter J., and Donald C. Reitzes. ‘‘The Link between Identity and Role Performance.’’ Social Psychology Quarterly 44 (1981): 83–92.
- ———. ‘‘An Identity Theory Approach to Commitment.’’ Social Psychology Quarterly 54 (1991): 239–251.
- Burke, Peter J., and Jan E. Stets. ‘‘Trust and Commitment through Self-Verification.’’ Social Psychology Quarterly 62 (1999): 347–366.
- Burke, Peter J., Jan E. Stets, and Maureen Pirog-Good. ‘‘Gender Identity, Self-Esteem, and Physical and Sexual Abuse in Dating Relationships.’’ Social Psychology Quarterly 51 (1988): 272–285.
- Burke, Peter J., and Judy Tully. ‘‘The Measurement of Role/Identity.’’ Social Forces 55 (1977): 881–897.
- Cast, Alicia D., Jan E. Stets, and Peter J. Burke. ‘‘Does the Self Conform to the Views of Others?’’ Social Psychology Quarterly 62 (1999): 68–82.
- Johnson, Michael P., and Kathleen J. Ferraro. ‘‘Research on Domestic Violence in the 1990s: Making Distinctions.’’ Journal of Marriage and the Family 62 (2000): 948–963.
- McCall, George J., and J. L. Simmons. Identities and Interactions. New York: Free Press, 1978.
- Miller, JoAnn Langley, Dean D. Knudsen, and Stacey Copenhaver. ‘‘Family Abuse and Violence.’’ In Handbook of Marriage and the Family, edited by Marvin B. Sussman, Suzanne K. Steinmetz, and Gary W. Peterson. New York: Plenum Press, 1999, 705–741.
- Pinel, Elizabeth C., and William B. Swann, Jr. ‘‘Finding the Self through Others: Self-Verification and SocialMovement Participation.’’ In Self, Identity, and Social Movements, edited by Sheldon Stryker, Timothy J. Owens, and Robert W. White. Minneapolis: University of Minnesota Press, 2000, 132–152.
- Powers, William T. Behavior: The Control of Perception. Chicago: Aldine Publishing, 1973.
- Rosenbaum, Alan. ‘‘Of Men, Macho, and Marital Violence.’’ Journal of Family Violence 1 (1986): 121–129.
- Stets, Jan E. ‘‘Control in Dating Relationships.’’ Journal of Marriage and the Family 55 (1993): 673–685.
- ———. ‘‘Job Autonomy and Control over One’s Spouse: A Compensatory Process.’’ Journal of Health and Social Behavior 36 (1995): 244–258.
- ———. ‘‘Identity Theory.’’ In Contemporary Social Psychological Theories, edited by Peter J. Burke. Stanford, CA: Stanford University Press, 2006, chapter 5. Stets, Jan E., and Peter J. Burke. ‘‘Gender, Control, and Interaction.’’ Social Psychology Quarterly 59 (1996): 193–220.
- ———. ‘‘Identity Theory and Social Identity Theory.’’ Social Psychology Quarterly 63 (2000): 224–237.
- ———. ‘‘Identity Verification, Control, and Aggression in Marriage.’’ Social Psychology Quarterly 68 (2005a): 160–178.
- ———. ‘‘New Directions in Identity Control Theory.’’ Advances in Group Processes 22 (2005b): 43–64.
- Stets, Jan E., and Teresa Tsushima. ‘‘Negative Emotion and Coping Responses within Identity Control Theory.’’ Social Psychology Quarterly 64 (2001): 283–295.
- Stryker, Sheldon. Symbolic Interactionism: A Social Structural Version. Caldwell, NJ: Blackburn Press, 2002.
- Stryker, Sheldon, and Richard T. Serpe. ‘‘Commitment, Identity Salience, and Role Behavior: A Theory and Research Example.’’ In Personality, Roles, and Social Behavior, edited by William Ickes and Eric S. Knowles. New York: Springer-Verlag, 1982, 199–218.
- ———. ‘‘Identity Salience and Psychological Centrality: Equivalent, Overlapping, or Complementary Concepts?’’ Social Psychology Quarterly 57 (1994): 16–35.
- Swann, William B., Jr. ‘‘To Be Adored or to Be Known? The Interplay of Self-Enhancement and Self- Verification.’’ In Handbook of Motivation and Cognition, edited by E. Tory Higgins and Richard M. Sorrentino. New York: Guilford, 1990, 408–450.
- Thoits, Peggy A. ‘‘On Merging Identity Theory and Stress Research.’’ Social Psychology Quarterly 54 (1991): 101–112.
- ———. ‘‘Personal Agency in the Accumulation of Multiple Role-Identities.’’ In Advances in Identity Theory and Research, edited by Peter J. Burke, Timothy J. Owens, Richard T. Serpe, and Peggy A. Thoits. New York: Kluwer Academic/Plenum, 2003, 179–194.
- Toby, Jackson. ‘‘Violence and the Masculine Ideal: Some Qualitative Data.’’ Annual American Academic Political Social Science 364 (1966): 623–631.