Is low self-esteem an inevitable consequence of stigma? An example from women with chronic mental health problems

Abstract

It is often assumed that membership in a stigmatized group has negative consequences for the self-concept. However, this relationship is neither straightforward nor inevitable, and there is evidence suggesting that negative consequences may not necessarily occur (Psychol. Rev. 96(4) (1989) 608). This paper argues that the relationship has not been sufficiently theorized, and that a more detailed analysis is called for in order to understand the relationship between stigma and the self. The paper presents a critical examination of modified labeling theory (Am. Sociol. Rev. 52 (1987) 96), with examples from a study examining perceptions of stigma and their relationship to self-evaluation in women with chronic mental health problems. Open-ended interviews and qualitative analyses were used in preference to global measures of self-esteem. It was found that although the women were aware of society's unfavorable representations of mental illness, and the effects this had on their lives, they did not accept these representations as valid and therefore rejected them as applicable to the self. The participants did not deny their mental health problems, but their acceptance of labels was critical and pragmatic. Labels were rejected when they were perceived as carrying an unrealistic and negative stereotype, or when the women felt that their symptoms did not fit with the diagnostic criteria. The research illustrates the importance of considering people's subjective understandings of stigmatized conditions and societal reactions in order to understand the relation between stigma and the self.

Introduction

This study investigates the understanding of stigma in women with long-term mental health problems, and the relation of this understanding to self-evaluation. Rather than taking global measures of self-esteem, which do not allow detailed investigation of the relationships between ascribed identities and the self-concept, the study takes a qualitative approach, which allows some of the complexities of the relationship between self and stigma to be described. In particular, the study explores participants’ understanding of how others represent people with a mental illness, and the extent to which participants accept that these representations define the self.

Goffman (1963) described stigmatized attributes as those that are deeply discrediting in particular contexts, which tend to become the dominant identities by which the person is perceived. Goffman stressed that stigma should be understood in terms of relationships, since it refers to a process of social rejection, devaluation or discrimination (see also Jones et al., 1984). Mental illness is a stigmatized attribute (Corrigan, 1998; Goffman, 1963; Wahl, 1995; Wahl & Harman, 1989). Negative representations of those with mental illness are found in advertising, films and in insults and jokes in everyday discourse. People with mental illness are often portrayed as fundamentally different, less competent, and violent (Wahl, 1995). Individuals with mental illness report incidents in which they are treated badly because of the label (e.g. Herman, 1993; Leete, 1992; Link, Struening, Rahav, Phelan, & Nuttbrock, 1997; Markowitz, 1998; Miles, 1984; Murphy, 1998; Wahl, 1999), and often expect others to devalue and reject them (Link, 1987; Link, Cullen, Struening, Shrout, & Dohrenwend, 1989; Markowitz, 1998; Nicholson, Sweeney, & Geller, 1998). People may therefore conceal their psychiatric history, disclosing it only to trusted others, whilst others may disclose it in an attempt to re-educate people about the nature of mental illness (Alverson, Becker, & Drake, 1995; Gilmartin, 1997; Herman, 1993; Link et al., 1989; Link et al., 1997; Lundin, 1998; Wahl, 1999).

Symbolic interactionism proposes that either by taking the perspective of particular others, or by viewing ourselves from the standpoint of the community and its values (the generalized other), we come to see ourselves as others do (Cooley, 1902; Mead, 1934; Blumer, 1969). Many theorists suggest that as a result of this process, possession of a stigmatized attribute is likely to have negative effects on self-evaluation. “Given that the stigmatized individual in our society acquires identity standards which he applies to himself in spite of failing to conform to them, it is inevitable that he will feel some ambivalence about his own self” (Goffman, 1963, p. 130). Social psychological writing on membership in supposed low status groups also often makes this assumption (e.g. Allport, 1954; Blanz, Mummendey, Mielke, & Klink, 1998; Tajfel & Turner, 1979; Taylor & McKirnan, 1984; Wright, Taylor, & Moghaddam, 1990). Some writers have suggested, however, that people may not agree with others’ views of themselves, and that the relationship between stigma and self-esteem is not inevitable (Fine & Asch, 1988; Finlay & Lyons, 2000; Jones et al., 1984).

Modified labeling theory (MLT; Link, 1987; Link, Cullen, Struening, Shrout, & Dohrenwend (1989), Link, Struening, Rahav, Phelan, & Nuttbrock (1997); Rosenfield, 1997) suggests that people with a mental illness are likely to have internalized negative representations of mental illness before they became labeled. In anticipation of devaluation and discrimination, they may adopt strategies such as withdrawal, secrecy or educating others. A variety of negative consequences occur as a result of both discrimination and the effects of the coping strategies themselves (e.g. in social interaction and employment). Low self-esteem is assumed to arise as a result of these processes, and also because individuals apply unfavorable social conceptions of mental illness to themselves (Link, 1987; Link & Cullen, 1990; Link, Mirotznik, & Cullen, 1991). It should be noted, however, that Link et al. (1989) accepted that some people with mental illness believe that others’ attitudes are rather benign (see also Pinel, 1999).

The research examining MLT has been broadly supportive. Studies have shown that people's expectations of devaluing and discriminating reactions are positively related to depression (Link et al., 1997), demoralization and unemployment (Link, 1987), and negatively related to perceived quality of life (Rosenfield, 1997), income and social support (Link et al., 1989). However, researchers, using mostly multiple regression procedures, stress the difficulties in making interpretations of causality (Link, 1987; Rosenfield, 1997). Relationships are likely to be bi-directional and affected by unmeasured factors such as actual experiences of discrimination. In addition, the hypothesized link between internalized negative attitudes towards mental illness and self-esteem is not tested in these studies; there are no measures of the person's own representations of mental illness, and self-esteem as an independent outcome has rarely been assessed. Link's original study (Link, 1987), widely cited as showing evidence of negative effects on self-esteem, did not have a measure of self-esteem, but rather a measure of demoralization, a composite of self-esteem, hopelessness, pessimism, dysphoria, thought confusion and sadness. Rosenfield (1997) found a significant effect in a model using self-esteem as a mediating variable between perceived stigma and quality of life, but pointed out that causality might be in either direction (see also Markowitz, 1998). One recent longitudinal study found that actual experiences of rejection affected self-deprecation (Wright, Gronfein, & Owens, 2000). There is also some evidence that mental illness leads to lower self-esteem from retrospective personal and ethnographic accounts (Gilmartin, 1997; Herman, 1993; Wahl, 1999). However, it is difficult to separate the effects of stigma from other aspects of the illness such as loss of roles and independence in these studies. Lowered self-esteem may be an antecedent or symptom of the illness itself (Brown, Andrews, Bifulco, & Veiel, 1990) or may be the result of grief associated with the loss of cognitive or social functioning (Appelo, Sloof, Woonings, Carson, & Louwerens, 1993). In addition, the use of coping strategies or other temporal factors may alter self-evaluations (Breakwell, 1986), making it difficult to demonstrate lowered self-esteem as a separate consequence of stigma.

Despite a widespread assumption that the stigma of mental illness creates problems of self-esteem, which people attempt to overcome through a variety of strategies (e.g. Herman, 1993; Link et al., 1989; Sayre, 2000; Wahl, 1995), there is little empirical support for this proposition. This is hardly surprising considering that a number of differing predictions might be made from stigma theories, each of which specifies different mediating variables. It seems too crude to suggest that merely being in a stigmatized group leads to low self-esteem. Lower self-esteem might be found only in those for whom the label is central to their self-concept, who perceive a more negative attitude in their community, whose own attitudes to mental illness are more unfavorable, or who have had more direct experience of rejection. Self-esteem might be affected in a variety of ways: by internalizing negative representations of mental illness, by being affected by the response of others in the immediate environment, by unfavorable social or temporal comparisons, by the loss of valued roles and relationships (Nicholson et al., 1998; Wright et al., 2000), or by lowered perceptions of personal control (Fife & Wright, 2000).

Because of the complexity of the relationship between stigma and self-conceptions, a more fine-grained analysis is called for, one that examines how people's understanding of mental illness, stigma, and their identities are related. Although it is acknowledged that stigma might affect self-evaluations through many routes (see above), the following section questions the suggestion that stigmatized people see themselves as others see them and struggle with low self-esteem as a result. Specifically, we address two assumptions, namely that the individual not only accepts that mental illness is central to their own identity but also shares the unfavorable representations of mental illness found in the wider community.

Thoits (1985) criticizes the assumption that when people accept that they have a mental illness, it becomes central to their self-concept (see also Taylor & Perkins, 1991). The importance that mental illness assumes as an aspect of the self-concept varies among individuals and over time (e.g. Appelo et al., 1993; Estroff, Lachicotte, Illington, & Johnston, 1991; Lundin, 1998; Taylor & Perkins, 1991). This is found in studies of insight in schizophrenia (David, 1990), and people's explanations for their symptoms and hospitalizations (e.g. Estroff et al., 1991; Sayre, 2000). Accounts of people with mental illness often stress that people should see beyond the label, to other aspects of the person (e.g. Herman, 1993; Wahl, 1999). Deegan (1997) talked about the importance of resisting seeing oneself the way others see you when you are labeled as having schizophrenia, where otherwise unremarkable behaviors are attributed by others to the mental illness: ‘No, I am not an illness. I am first and foremost a human being’ (p. 19).

If self-esteem is affected by self-stigmatization, we might expect that those who accept the label, or for whom the identity is more central, would be more likely to have lowered self-esteem. However, because people have varied understandings of mental illness, evidence concerning the relationship between self-labeling and self-esteem is contradictory (Birchwood, Mason, MacMillan, & Healy, 1993; Mechanic, McAlpine, Rosenfield, & Davis, 1994). Estroff et al. (1991) recommended that research on identification explore the meanings of mental illness for the respondents (see Finlay & Lyons, 1998, concerning people with mental retardation). In their study, self-labeling was related to people's varied explanations of their illness (see also Sayre, 2000). For example, those who gave medical or emotional/developmental accounts were more likely to say they had a mental illness than those who gave situational or religious/spiritual accounts. Different accounts of mental illness would be expected to have different implications for the self-concept.

People with a mental illness are often aware of the negative impression others have of them. Birchwood et al. (1993), with a sample of people with schizophrenia and bipolar disorder, found that those who had accepted the negative stereotypes of mental illness were more depressed than those who had not (for a similar result see Link, 1987). This implies that individuals vary in the degree to which they accept these negative representations. It is more likely that a person would share negative representations of mental illness early in the course of the illness, since over time an increased understanding of the condition and the varied reactions of others to it is likely to develop (see Deegan, 1997; Gilmartin, 1997). Reports of people with mental illness disagreeing with popular representations are common in the literature, and people who find themselves stigmatized often try to educate the public into a more realistic understanding (e.g. Gilmartin, 1997; Herman, 1993; Wahl, 1999). Negative reactions or evaluations by others would be expected to have less effect on self-esteem when they are attributed to prejudice (Crocker & Major, 1994). Even people with a mental illness who do share the stereotype may not apply it to themselves. For example, O’Mahony (1982) found that psychiatric hospital patients shared negative stereotypes of ‘the mentally ill’ with hospital staff, but did not represent themselves in terms of this stereotype. They were not, however, engaged in a global denial of their problems.

While many studies describe the type of strategies people might adopt to manage hostile reactions from others (e.g. Goffman, 1963; Herman, 1993; Link, Cullen, Struening, Shrout, & Dohrenwend (1989), Link, Mirotznik, & Cullen (1991); Wahl, 1999), there has been little attention to date on the phenomenological aspects of stigma for the self. This study therefore examined in detail, using semi-structured interviews and qualitative analyses, the ways in which understandings of stigma, mental health problems, and self-concept were related in a small sample of women with chronic mental health problems.

The major research questions were as follows:

• How do the women represent other people's understandings of mental illness?

• To what extent do they agree with these understandings?

• In what ways are aspects of mental illness presented as part of the women's identities?

• Is there evidence of negative self-evaluations related to this stigmatized identity?