Mental Illness Stigma and the Sociology of Mental Health

Abstract

Scanning the development of research pertaining to labeling and stigma as it pertains to mental illnesses over the past 30 years reveals a deep connection between these fields of inquiry and the development of the American Sociological Association’s Section on the Sociology of Mental Health. As research in this domain progressed, the need for a specific home for that research within the ASA became apparent, and as the section grew and supported sessions specifically on labeling and stigma this domain of research was further enhanced. At this juncture sociological understandings of labeling and stigma stand as one of the major successes of the section on the Sociology of Mental Illness.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Scanning the development of research pertaining to labeling and stigma as it pertains to mental illnesses over the past 30 years reveals a deep connection between these fields of inquiry and the development of the American Sociological Association’s Section on the Sociology of Mental Health. As research in this domain progressed, the need for a specific home for that research within the ASA became apparent, and as the section grew and supported sessions specifically on labeling and stigma this domain of research was further enhanced. At this juncture sociological understandings of labeling and stigma stand as one of the major successes of the section on the Sociology of Mental Illness. The work has linked the section to the broader discipline through papers published in The American Sociological Review and the American Journal of Sociology and has also brought critical sociological perspectives to relevant disciplines of anthropology, psychology, psychiatry and public health. We seek to capture some of these developments and the impact they have had on understanding the social context of mental illness. Our focus, like the focus of the volume, has been on contributions emanating from the sociology of mental health and consequently we pay less attention to the large advances that have been made in other disciplines, especially anthropology (e.g. Parker and Aggleton 2003; Yang et al. 2007) and social psychology (see Major and O’Brien 2005). Specifically, we point to conceptual advances, studies of public conceptions of mental illnesses, and research that pertains to how people are affected by stigma. We begin with some selected observations about the background of research in this area that help set the stage for understanding some of the advances that have been made in recent years.

4.1 Background

Before Goffman’s book, Stigma: Notes on the Management of Spoiled Identity, the term stigma was used in the social sciences to mean something quite close to its current meaning but was used only infrequently. A Google Scholar search for the period 1900–1960 returns numerous scientific articles using the term “stigma” but almost all of these refer to botany (the receptive apex of the pistil of a flower) or other biological phenomena (a small mark, spot, or pore) rather than to social science meanings of the term. A Google Scholar search in the current era reveals something entirely different with the social science meaning of the term ascendant and being applied to a vast array of stigmatizing circumstances. Another indicator of the large increase in interest is the number of published articles with the word “stigma” in the title or abstract. In 1980 the number stood at 19 for Medline and 14 for Psych Info, but rose dramatically by the end of the century to 114 for Medline and 161 for Psych Info in 1999 (Link and Phelan 2001). Incredibly, by 2010 the numbers were more than five times as high as in 1999: 758 for Medline and 851 for Psych Info. Of course not all of these referred to mental illness stigma but many did and there is no doubt that this trend concerning stigma in general has also applied to mental illness stigma in particular.

4.1.1 Origins: Goffman and the Labeling Debate

In the mid 1950s, Erving Goffman was a research fellow at the Laboratory for Social and Environmental Studies at the National Institute of Mental Health. The unit was headed at the time by sociologist John Clausen, and it was during this period that Goffman did his ethnographic work Asylums at Saint Elizabeth’s Hospital (Goffman 1961). Stigma was on the minds of the small but enormously generative group at the Laboratory for Social and Environmental Studies, especially in the context of qualitative studies they were undertaking concerning wives of men who were hospitalized for mental illness. Whereas the term stigma was not in wide use in the social sciences at the time, one exception was a paper from this group authored by Schwartz (1956) entitled “The Stigma of Mental Illness.” She indicated that stigma had “two connotations: first, that in the minds of others the person is set apart—that is, different from the so-called normal person; second that he is set apart by a ‘mark’ which is felt to be ‘disgraceful,’ or even ‘immoral,’ by which he can be judged to be ‘inferior’” (Schwartz 1956, p. 7). Exposed to these ideas and drawing on his ethnography in Saint Elizabeth’s hospital, Goffman (1961) produced his highly influential introduction to the stigma concept. Goffman’s (1963) description was comprehensive, and it is difficult to find any current consideration of the concept that is not foreshadowed in his 1963 treatise. It is in this work that perhaps the most influential definition of the concept was provided: “an attribute that is deeply discrediting” and that reduces the bearer “from a whole and usual person to a tainted, discounted one” (Goffman 1963, p. 3).

Subsequent to its introduction stigma played a central role in the so-called labeling debate that emerged during the 1960s. Scheff (1966) constructed a formal labeling theory of mental illness that located the origin of stable mental illness in societal reactions including stigmatizing reactions. The essence of his theory is captured in the following quote:

In a crisis, when the deviance of an individual becomes a public issue, the traditional stereotype of insanity becomes the guiding imagery for action, both for those reacting to the deviant and, at times, for the deviant himself. When societal agents and persons around the deviant react to him uniformly in terms of the stereotypes of insanity, his amorphous and unstructured rule-breaking tends to crystallize in conformity to these expectations, thus becoming similar to behavior of other deviants classified as mentally ill and stable over time. The process of becoming uniform and stable is completed when the traditional imagery becomes a part of the deviant’s orientation for guiding his own behavior (Scheff 1966, p. 82).

The theory is called “labeling” theory because of the centrality it gave to social definition of deviant behaviors. The debate concerning the role of labeling in mental illness involved both informal labeling processes (e.g. spouses labeling of their partners) and official labeling through treatment contact (e.g. psychiatric hospitalization). In Scheff’s theory, the act of labeling was strongly influenced by the social characteristics of the labelers, the person being labeled, and by the social situation in which their interactions occurred. He asserted that labeling was driven as much by these social factors as it was by anything that might be called the symptoms of mental illness. Moreover, according to Scheff, once a person is labeled, powerful social forces come into play to encourage a stable pattern of “mental illness.” Stigma was a central process in this theory as it “punished” people who sought to shed the identity of mental illness and return to normal social roles, interactions and identities.

Critics of the theory, especially Walter Gove, took sharp issue with Scheff’s characterization of the labeling process. Gove argued that labels are applied far less capriciously and with many fewer untoward consequences than claimed by labeling theorists (Gove 1975). For some period between the late 1970s and early 1980s, professional opinion swayed in favor of the critics of labeling theory. Certainly the dominant view during that time was that stigma associated with mental illness was relatively inconsequential. Gove for example, concluded that “… stigma appears to be transitory and does not appear to pose a severe problem” (Gove 1982, p. 290) and Crocetti and Spiro (1974) concluded that “former patients enjoy nearly total acceptance in all but the most intimate relationships.” Moreover, when a group of expert stigma researchers was summoned to the National Institute of Mental Health in 1980 to review evidence about the issue, the term “stigma” was intentionally omitted from the title of the proceedings. Apparently, the argument that behaviors rather than labels are the prime determinants of social rejection was so forcefully articulated that the editors of the proceedings decided that stigma was not an appropriate designation when “one is referring to negative attitudes induced by manifestations of psychiatric illness” (Rabkin 1984, p. 327). It was within this context that so-called “modified labeling theory” (described in some detail below) emerged in response to the then dominant anti-labeling, stigma-dismissing stance that characterized the field at the time. In the ensuing years major advances in concepts, measures and empirical assessments of mental illness stigma unfolded, many within the context of the Section on the Sociology of Mental Health.

4.2 Conceptualizing Stigma

4.2.1 What Is Stigma?

In the literature on stigma, the term has been used to describe what seem to be quite different concepts. It has been used to refer to the “mark” or “label” that is used as a social designation, to the linking of the label to negative stereotypes, or to the propensity to exclude or otherwise discriminate against the designated person. Even Goffman’s (1963) famous essay includes several somewhat different, albeit very instructive, definitions. As a consequence of this variability, there has been confusion as to what the term means. Additionally, an intense dissatisfaction with the concept emerged in some circles for at least two reasons. First, it was argued that the stigma concept identifies an “attribute” or a “mark” as residing in the person—something the person possesses. The objection to this conceptualization was that it took for granted the process of affixing labels and did not interrogate the social processes that led to such labeling (Fine and Asche 1988). In particular, far too little attention had been focused on the selection of a single characteristic for social salience from a vast range of possible characteristics that might have been identified instead. Second, it was argued that too much emphasis had been placed on cognitive processes of category formation and stereotyping and too little on the broad and very prominent fact of discrimination and the influence that such discrimination has on the distribution of life chances (Oliver 1992).

In light of this confusion and controversy, Link and Phelan (2001) put forward a definition of stigma that recognized the overlap in meaning between concepts like stigma, labeling, stereotyping and discrimination. This conceptualization defined stigma in the relationship between interrelated components. The idea to do this followed an insight from Goffman who at one point indicated that the essence of stigma lay in the relationship between an attribute and a stereotype. As described below, this conceptualization defines stigma in the co-occurrence of interrelated components of labeling, stereotyping, separating, emotional reactions, status loss and discrimination. The approach also responds to the criticism that the stigma concept locates the “mark” or “attribute” in the person by making it clear that such “marks” (or “labels” as designated by Link and Phelan) are selected for social salience from among many possible human traits that might have been selected. This approach also responds to prior criticisms by making the social selection of designations a prominent feature, by incorporating discrimination into the concept, and by focusing on the importance of social, economic and political power in the production of stigma. Link and Phelan describe their conceptualization as follows:

In our conceptualization, stigma exists when the following interrelated components converge. In the first component, people distinguish and label human differences. In the second, dominant cultural beliefs link labeled persons to undesirable characteristics—to negative stereotypes. In the third, labeled persons are placed in distinct categories so as to accomplish some degree of separation of “us” from “them.” In the fourth, labeled persons experience status loss and discrimination that lead to unequal outcomes. Stigmatization is entirely contingent on access to social, economic and political power that allows the identification of differentness, the construction of stereotypes, the separation of labeled persons into distinct categories and the full execution of disapproval, rejection, exclusion and discrimination. Thus we apply the term stigma when elements of labeling, stereotyping, separation, status loss and discrimination co-occur in a power situation that allows them to unfold (Link and Phelan 2001, p. 367).

A detailed exposition of each of these components is available elsewhere (Link and Phelan 2001, 2012). Here we provide a brief description of each component, connecting each component to the stigma associated with mental illness (although they are intended to be applicable to other stigmatized circumstances as well).

Distinguishing and labeling differences. The vast majority of human differences, e.g., eye color, favorite ice cream or ear lobe width, are not considered to be socially relevant bases for constructing sharp group boundaries. However, some differences, such as skin color and sexual preferences, are currently awarded a high degree of social salience. Both the selection of salient characteristics and the creation of labels for them are social achievements that must be present for stigma to exist. In the area of mental illnesses The Diagnostic and Statistical Manual of Mental Disorders (DSM) of the American Psychiatric Association represents an attempt by professionals to decide which human differences should be selected for designation as mental illnesses and which should not. This social selection of human differences and social production of designations is particularly apparent when the criteria are contested, as they were when homosexuality was removed from the DSM, and as they are now with respect to whether or not normal human emotional states such as sadness are being pathologized (Horwitz and Wakefield 2007).

Associating differences with negative attributes. In this component, the labeled difference is linked to negative stereotypes. For example, one common stereotype is that a person who has been hospitalized for mental illness is likely to be unpredictable. Other powerful stereotypes associated with mental illnesses involve inferences about competence, dangerousness, cleanliness and trustworthiness. In the Link and Phelan conceptualization, stereotypes like these must be present and operative for stigma to exist—there must be some linking of a label to a stereotype.

Separating “us” from “them”. Central to early and nearly all definitions of stigma (e.g. Jones et al. 1984; Schwartz 1956) a third aspect of the stigma process occurs when labels connote a separation of “us” from “them.” Examples can be found with respect to certain ethnic or national groups (Morone 1997), people with mental illness, or people with a different sexual orientation who may be considered fundamentally different kinds of people from “us.” In the area of mental illness such a separation is sometimes embedded in the language we use to describe people. For example a person has heart disease, cancer or an infection but a person who develops schizophrenia “is” a “schizophrenic”—a different sort of person than the rest of “us.”

Emotional Responses. The Link and Phelan conceptualization of stigma subsequently was expanded to include emotional responses. Link et al. (2004) noted that from the vantage point of a stigmatizer, emotions of anger, irritation, anxiety, pity and fear are likely. From the vantage point of the person who is stigmatized, emotions of embarrassment, shame, fear, alienation or anger are possible.

Status loss and discrimination. When people are labeled, set apart and linked to undesirable characteristics, a rationale is constructed for devaluing, rejecting and excluding them. When devaluation, discrimination and exclusion are widespread, a persistent pattern of unequal social relationships arises that creates social structures of disadvantage. Once in place these structural arrangements (segregation, exclusion, downward occupational placement) feedback to reinvigorate the labels, stereotypes, setting apart and emotional reactions that disadvantage stigmatized groups.

The dependence of stigma on power. A unique feature of Link and Phelan’s (2001) definition is the idea that stigma is entirely dependent on social, cultural, economic and political power. Lower-power groups (e.g., mental health consumers) may label, stereotype and separate themselves from higher-power groups (e.g., psychiatrists) by perhaps labeling the psychiatrists “pill pushers,” stereotyping them as “cold” “haughty,” and “clueless” and seeing them as a separate group that is distinct from “us.” But in this case, stigma as we define it does not exist, because the potentially stigmatizing group (the mental health consumers) do not have the social, cultural, economic and political power to imbue their cognitions (labels and stereotypes) with serious discriminatory consequences. The psychiatrists are not severely damaged materially by the consumers’ stereotypes about them. Stigma is dependent on power.

4.2.1.1 Why Do People Stigmatize?

Whereas the Link and Phelan (2001) conceptualization provided concepts that help us to understand what stigma is, the scheme does not tell us why people stigmatize nor why mental illnesses might be stigmatized. Filling this gap Phelan et al. (2008) provide a conceptual scheme that addresses the issue of why people stigmatize. The essence of the answer they provide is that stigmatizing helps people attain ends they desire. They propose three generic ends that people can attain by stigmatizing others: (1) exploitation/domination or keeping people down, (2) enforcement of social norms or keeping people in, and (3) avoidance of disease or keeping people away. We briefly review these reasons for stigmatization and then use them to consider why mental illnesses might be stigmatized.

Exploitation and domination. Wealth, power, and high social status can be attained when one group dominates or exploits another. Ideologies involving stigmatization develop to legitimate and perpetuate these inequalities with the group designated as the one to be kept down being deemed to be inferior in terms of intelligence, character, competence and the basic human qualities of worthiness, and value (Phelan et al. 2008). Classic examples are the racial stigmatization of African Americans beginning in the era of slavery, the Europeans’ colonization of countries around the globe, and U.S. whites’ expropriation of the lands of American Indians (Feagin 2009).

Enforcement of social norms. People construct a labyrinth of written and unwritten rules governing everything from how nation states should wage war to how a New Yorker should make it into a subway car. Some degree of investment in norms like these develops; people come to count on them and to be outraged or annoyed when they are violated. Failure to comply with these norms is often cast in terms of the flawed morality or character of the transgressor (Goffman 1963; Morone 1997), and stigma processes are deployed as a corrective mechanism. One way that stigma is useful, then, is that it imparts a stiff cost—a strong social disapproval—that can make subsequent transgressions less likely. In this use of stigma, people are kept in by influencing the behavior of the norm violator. A related use is that the people around the norm violator are kept in by learning the boundaries of acceptable behavior and by observing the stern example of what happens to someone who goes beyond those bounds (Erikson 1966).

Avoidance of disease. Many illnesses and disabilities (e.g. HIV, facial disfigurement, limb loss) are probably not stigmatized in order to exploit or dominate or in order to directly control behavior and enforce norms. Kurzban and Leary (2001) provide another explanation for stigma in these circumstances by arguing that there are evolutionary pressures to avoid members of one’s species who may spread disease. They focus on parasites noting that infection can lead to ‘‘deviations from the organism’s normal (healthy) phenotype’’ (Kurzban and Leary 2001, p. 197) such as asymmetry, marks, lesions and discoloration; coughing, sneezing and excretion of fluids; and behavioral anomalies due to damage to muscle-control systems. They argue that the advantage of avoiding disease might have led to a more general aversion to deviations from any local standard for the way humans are supposed to look or carry themselves (Kurzban and Leary 2001). Thus a broad band of deviations might lead to a visceral response of disgust and a strong desire to keep the person with the deviation away.

Why do people stigmatize mental illnesses? In keeping with the strong emphasis in sociological thinking about “residual rule breaking” (Scheff 1966) and the extension of that thinking through the sociology of emotions to “feeling rules” (Thoits 1985), we believe that the major reason for the stigmatization of people with mental illnesses is an attempt to keep people in. Initial reactions to symptoms are generally common-sense attempts to rein in the rule-breaking behavior by strongly disapproving of odd beliefs expressed by people with psychosis, admonishing a person with depression to “snap out of it,” or passing favorite foods into the sight lines of a person with anorexia. At the same time, the bizarre behavior of psychosis; the weight loss, enervation, and anhedonia of depression; or the extreme underweight associated with anorexia could stimulate a desire for “disease avoidance.” Although there is little reason to suppose that mental illnesses are stigmatized so that those who suffer from them can be exploited or dominated for pecuniary gain when efforts to keep people in fail, keeping people away can be substituted as a strategy to avoid non-normative behavior. And to the extent that keeping people away is more easily achieved when people are relatively powerless, we might expect that keeping people down would also be prominent in the case of serious mental illnesses. Thus we expect a strong initial motivation to stigmatize mental illnesses resides in efforts to keep people in, but when symptomatic behaviors endure and efforts to keep people in fail, motivations to keep people down and away are also evident.

4.2.2 Stigma Power: People’s Use of Stigma to Achieve Desired Ends

A novel feature of the Link and Phelan definition of stigma (described above) is the incorporation of “power” in that definition. Successful stigmatization requires power; it requires the ability to construct stereotypes that are broadly endorsed and deeply held; and it requires control over jobs, housing, and education to enact discriminatory behavior that has teeth. Thus, it “takes power” to stigmatize.

However, in light of Phelan et al.’s (2008) consideration of the reasons people stigmatize, we now realize that people achieve outcomes they desire when they stigmatize others. Whether the motive is to keep people down, keep people in or keep people away, stigma is a useful instrument to accomplish ends that are congenial to the stigmatizer’s interests. Conceived in this way, stigma is a source of power that helps the stigmatizer control the stigmatized person and thereby keep them down, in or away. Thus, we now take the idea that it “takes power” to stigmatize and add the notion that stigmatization confers power—“stigma power.”

The concept of stigma power can be thought of as one form of what Bourdieu called “symbolic power.” For Bourdieu (1987), symbolic power is the capacity to impose on others a legitimatized vision of the social world and of the cleavages within that world. Bourdieu developed and used the concept mainly to understand class and class reproduction, adding a cultural element to the understanding of those phenomena. But three aspects of Bourdieu’s concept are extremely useful with regard to understanding stigma and the power it confers. First, cultural distinctions of value and worth are the critically important mechanisms through which power is exercised. Stigma is in many respects a statement about value and worth made by a stigmatizer about those he or she might stigmatize and, thus, one form of symbolic power in Bourdieu’s terms. Second, those who are disadvantaged by the exercise of power are often persuaded, sometimes without realizing it, to accept as valid the cultural evaluations that harm them. Finally, the exercise of symbolic power is often buried in taken-for-granted aspects of culture and thereby hidden, or “misrecognized” as Bourdieu (1990) put it, by both the people causing the harm and by those being harmed.

To explore the utility of the stigma power concept with respect to mental illnesses, Link and Phelan (2014) examined ways in which the goals of stigmatizers are achieved but hidden in the stigma coping efforts of people with mental illnesses. Capitalizing on new measures from a small study of stigma and psychosis, Link and Phelan found that people with psychosis are aware of the cultural assessment of their lower social standing, show a high degree of concern about staying within normative boundaries, an inclination to stay away from others to avoid rejection and a feeling of being downwardly placed in terms of the experience of low self-esteem. In keeping with the concept of stigma power, results are consistent with the possibility that a cascade of circumstances in which stigmatized people, in seeking to avoid rejection by others, accomplish what those others want—keeping them in, down and away. Because it is new, the usefulness of the stigma power concept remains for future research to further assess. At the same time the idea that people’s interests underlie their inclination to stigmatize and that these interests are often achieved in hidden misrecognizable ways is an idea that could help us understand why stigma has been so difficult to address.

Concepts described above pertaining to what stigma is, why people stigmatize, and how people use stigma to gain desired ends have grown alongside empirical studies of public conceptions or what some have called public stigma (Corrigan et al. 2004). We turn next to studies of public conceptions focusing attention on their importance as sociological phenomena and especially on the importance of assessing changes in such conceptions over time.

4.3 Public Conceptions of Mental Illnesses

One way to think about domains of public conceptions, and the most common way to do so, is to ask what determines individual differences in such conceptions and what consequences such differences might have for individual behaviors. Construed in this way, research about public conceptions is sometimes challenged when it either does not assess individual behaviors at all or predicts such behaviors with less than ideal accuracy. The lack of correspondence between attitudes and behaviors is brought to the fore and the utility of research focused on public conceptions is sharply questioned.

However, another way to reason about public conceptions that is informed a by a sociological perspective is to view them at the collective level—as indicators of cultural context. Specifically, if we could obtain an accurate and comprehensive assessment of the public conception concerning what mental illness is, what people with mental illness are like, what causes mental illnesses, what kinds of emotional reactions mental illnesses evoke, what should be done when a person develops a mental illness, how much social distance should be kept from someone with a mental illness, and what are preferred policies to address the problem of mental illness, we would have a portrait of the cultural conception of mental illness in a given place and at a particular time. It would tell us how people think and feel about mental illnesses and how such illnesses should be managed. As a context, this cultural conception becomes an external reality—something that individuals must take into account when they make decisions and enact behaviors. The idea is that individuals (e.g. people with mental illnesses, care givers, policy makers) know about cultural conceptions and shape their behaviors to some significant degree to take account of them no matter what their own knowledge, attitudes and beliefs (KABs) happen to be (see description of modified labeling theory below). In this way, cultural conceptions can have an important impact on things that matter for people with mental illnesses through mechanisms that do not involve individual attitudes influencing individual behaviors toward people with a mental illness. Consider just four ways in which cultural conceptions can affect the structural circumstances that people with mental illness encounter. First, cultural conceptions are the ways we have of thinking about the issue, about what mental illness is, what people who have mental illnesses are like, what people need, and how we should manage people who develop mental illnesses. This will influence the kind of policies and practices we conjure to address the problem, putting bounds on what we think makes sense and what we think is possible. Second, to the extent that there is a societal downward placement of mental illness in a hierarchy of importance or worthiness, structural stigma is likely to be a consequence. When studies of the prestige of medical specialties and of specific illnesses have been undertaken, psychiatry and psychiatric illness are rated low (Norredam and Album 2007). Moreover, coverage of mental illness in high-impact prestigious journals like Lancet is dramatically lower than one would expect given the burden of disease that mental illnesses impose (Griffiths 2010). Lower prestige and reduced coverage in major journals means less prominence that could lead to lower funding, fewer scientific discoveries and less money for good patient care. Third, cultural conceptions impose constraints on many policies and practices. For example, as treatment providers and policy makers make decisions about where to locate a new board and care home for people with serious mental illness, they are likely to include in their considerations the expected response of the neighborhoods they imagine placing the facility in. Processes like these have resulted in a clustering of board and care facilities in neighborhoods that do not have the clout to exclude such facilities, thereby creating what have been called “psychiatric ghettos” (Arboleda-Florez 2006). Fourth and finally, structural stigma is not only induced by cultural conceptions, it is also sustained by such conceptions. Imagine for example that, unlike all other illnesses, insurance policies for heart disease (instead of mental illnesses) were capped at lower levels of compensation and included higher co-pays. Despite the fact that a rationale could be conjured for such a policy, given that heart disease is influenced by behaviors people can control, such as sedentarism, fatty diets and smoking, an insurance policy disadvantaging heart disease is unimaginable at the current time. Similarly, if cultural conceptions deemed some of the disadvantages that people with mental illnesses currently endure as simply outrageous and unthinkable, pressure would be brought to bear to change them, and they would not exist.

To the extent that cultural conceptions are important, it becomes critical to understand them. Members of the Mental Health Section have taken great strides in deepening that understanding in what Pescosolido has deemed a “resurgence” of interest in this area (Pescosolido 2013). Signature accomplishments were the fielding of three mental health modules in the General Social Survey, one focused on adults views of children’s mental-health disorders (Pescosolido et al. 2007) and two focused on adults’ mental-health conditions that were fielded in 1996 and 2006 (Link et al. 1999; Pescosolido et al. 2010). As Pescosolido (2013) points out, these surveys have taught us a great deal and have reenergized a sociological perspective on public conceptions. Among all these accomplishments, we focus on the comparisons over time because, in our view, these are particularly revealing concerning the role of social factors over time. The reason a long-term perspective is required is that cultural conceptions change relatively slowly and it is impossible to observe such change in cross-sectional studies. Cross-sectional (or other studies with a short time frame) direct attention to the variation such studies can capture—individual differences in cultural conceptions and their associations with outcomes variables of interest. Important as such studies can be, they leave us blind to the impact of changes in levels of KABs over time and to the powerful implications such changes have for the context in which people experience mental illnesses. In keeping with this rationale, we present evidence from two studies that have enacted the same or similar methods over multiple time periods. We begin by considering a study conducted by Phelan et al. (2000) that provided evidence on changes in the stereotype of dangerousness associated with mental illness in the United States.

The Dangerousness Stereotype in the United States from 1950 to 1996. Over the period from the 1950s to the 1990s two perspectives on public attitudes were in play. On the one hand the “optimists” (Crocetti and Spiro 1974) heralded a new era in which the public was more knowledgeable and much more tolerant than it had been in the 1950s. “Pessimists” took a decidedly different and much more skeptical view, attributing any apparent positive trends to surface-level changes in knowledge about the socially desirable response to survey items (Link and Cullen 1983). There was little if any consideration of the possibility that things might have gotten worse. It was in this context that Phelan et al. undertook a study that could reflect on such changes over long periods of time.

In 1996, teams of investigators at Columbia and Indiana Universities constructed the MacArthur module of the General Social Survey (Pescosolido et al. 2010). Interested in trends over time, the team directed attention to the first nationwide U.S. study of public attitudes conducted in 1950 by Shirley Star. Unfortunately, the questions in the original study generally used language that had become dated by 1996. However, the following open-ended question was available and could be repeated, “Of course, everyone hears a good deal about physical illness and disease, but now, what about the ones we call mental or nervous illness … When you hear someone say that a person is “mentally-ill,” what does that mean to you?” In both the Star study and the MacArthur module of the 1996 survey, answers were recorded verbatim. Fortunately, every tenth interview of the original Star survey had been saved by the librarian at the National Opinion Research Center in Chicago where both studies were conducted. This allowed trained coders to reliably rate the 1950 and 1996 responses to this question with respect to whether the respondent spontaneously referred to violent behavior in describing a person with mental illness. Thereby the study allowed a rare glimpse at trends in one key stereotype in the stigma associated with mental illnesses involving psychosis. Remarkably, the analysis revealed that despite massive efforts to educate the public about mental illness and enormous advances in treatment, respondents whose descriptions indicated a person with psychosis were nearly two and a half times as likely to mention violent behavior in 1996 (31.0 %) as in 1950 (12.7 %) (Phelan et al. 2000). Whatever the reasons for this change, at the very least, it represents a discomforting fact for people with a psychotic illness seeking broader social acceptance.

Changes in Public Conceptions in the United States from 1996 to 2006. Another study of change in public conceptions used the general social survey in the United States and employed the exact same methods and measures ten years apart (Pescosolido et al. 2010). With benchmarks like the 1999 U.S. Surgeon General’s report, efforts designed to reduce stigma were often predicated on assumptions that neuroscience offered the most effective tool to lower prejudice and discrimination. For example the National Alliance for the Mentally Ill’s (NAMI’s) “Campaign to End Discrimination” sought to improve public understanding of neurobiological bases of mental illnesses and substance abuse, facilitating treatment-seeking and lessening stigma. Over the decade between 1996 and 2006, the American public was taught about the symptoms of mental illnesses, educated about biological theories of etiology, and informed about the value of treatment to thereby underscore the basic argument that mental illnesses are diseases, no different from others. The National Stigma Study—Replication (NSS-R) which used modules from the 1996 and 2006 General Social Surveys in the United States provides evidence on the consequences of these efforts for public conceptions.

The NSS-R is a nationally representative study conducted under the umbrella of the General Social Survey that is implemented on an ongoing basis by the National Opinion Research Corporation. Individuals (N = 1,956) were interviewed face-to-face and presented with vignettes describing cases meeting DSM-IV criteria for schizophrenia, major depression or alcohol dependence. Approximately half of the respondents were interviewed in 1996, the other half in 2006, and all were randomly assigned to one vignette and then asked questions about the described person.

Results show that there were widespread increases in public acceptance of neurobiological theories about the causes of mental illnesses and significant changes in public support for treatment but no reduction in social distance or perceptions of dangerousness to self or others. Further, in both years and across all vignette conditions, holding a neurobiological conception was either unrelated to social distance and perceptions of dangerousness or was actually associated with an increase in these indicators of stigmatizing reactions (Pescosolido et al. 2010).

Again, these striking findings could not have been revealed without the trend over time, and we could not have known with any precision that the portrait of current KABs is the outcome of changes from an earlier era. Interestingly, the data provide rationales for both optimism and pessimism. On the one hand the data tell us that at least certain aspects of public conceptions are subject to change—tremendous effort was exerted to realize change and change did in fact occur in some dimensions. The public is much more likely to see mental illnesses as medical conditions and to believe that seeking help from medical and psychiatric sources is appropriate for such conditions. This change in the cultural context has been accompanied by a dramatic increase in the number of people with disorders seeking appropriate help for those disorders (Wang et al. 2005). The rational for pessimism resides in the relatively high and enduring levels of social distance and beliefs in dangerousness. There are three reasons for pessimism associated with this finding. First, a highly touted approach—advancing a neurobiological basis for mental illnesses—that was believed to be a potent factor in reducing negative attitudes failed to reduce core aspects of stigma. Second is the possibility that social distance and perceptions of danger are simply more deeply ingrained in history and the human condition and, as a consequence, much more difficult to change. Third is the possibility that social distance and stereotypes are not core concerns of the most powerful groups shaping public opinion—the medical, psychiatric and psychotherapeutic professions and pharmaceutical companies. The changes that have occurred are consistent with the interests of these groups: to be recognized on a par with other medical conditions for treating a “real” illness, to underscore the legitimacy of approaches to treatment that these groups deliver and, for the industry, an expansion of the market. Changing stereotypes and social distance are not as essential to the aims of these interest groups. People with mental illnesses who are often deeply harmed by stereotypes and rejection have great interest in these domains but less power to shape messages directed at what matters most to them.

4.3.1 Geneticization and Stigma

It certainly seems that, in the wake of the Human Genome Project and the sequencing of the human genome, genetic explanations for illness, personality, behavior, and every imaginable human characteristic are on the rise. As described above, the General Social Surveys conducted in 1996 and 2006 confirm this impression for alcohol dependence, schizophrenia, and major depression. Genetic attributions increased for all three disorders (from 61 to 71 % for schizophrenia, 51 to 64 % for major depression and 58 to 68 % for alcohol dependence) (Pescosolido et al. 2010). But what are the likely consequences of such a rise?

4.3.1.1 Theories Relating to Genetic Causal Attributions and Stigma

Attribution theory (Weiner 1986, 1995) states that the attributions people make about the cause of a personal outcome influence emotions, expectancies, and behavior toward the individual affected by the outcome. One important application of the theory has been to stigmatized behaviors (Corrigan 2000; Weiner et al. 1988). According to the theory, attribution of low causal responsibility for a stigmatized characteristic (e.g., brain dysfunction due to accidental injury rather than illicit drug abuse) is associated with less blame and more positive emotions, that is, pity rather than anger, which in turn lead to an inclination to help the person and a disinclination to punish (Corrigan et al. 2000; Rush 1998; Weiner et al. 1988).

Because one cannot be considered to have caused one’s own genetic makeup, attribution theory suggests that genetic attributions should reduce the perceived causal responsibility and, consequently, the negative emotions and behaviors associated with a stigmatized characteristic. Less sanguine predictions concerning the effect of genetic causal attributions on stigma are generated by the concept of essentialism (Rothbart and Taylor 1992) and the idea (“genetic essentialism”) (Lippman 1992; Nelkin and Lindee 1995) that genetic understandings of human behavior and other characteristics increases the belief that those characteristics are an essential part of the person. In a genetic essentialist view, genes form the basis of our human and individual identities (i.e., “we are our genes”) and are strongly deterministic of behavior, so that if one has genes associated with some behavior, that behavior will definitely occur and “is fixed and unchangeable” (Alper and Beckwith 1993, p. 511). A genetic essentialist viewpoint suggests that genetic characteristics are irrevocably, or at least very firmly, attached to an individual and by extension to those with whom the person shares genes. Consideration of positively valued characteristics such as beauty or intelligence make it clear that genetic essentialism is not inherently stigmatizing. However, when applied to negatively valued qualities, genetic essentialism should exacerbate stigma via its influence on several perceptions: (1) that the person is fundamentally different from others, (2) that the problem is persistent and serious, and (3) that the problem is likely to occur in other family members. These perceptions in turn should increase negative behavioral orientations such as the endorsement of reproductive restrictions and social distance, particularly “associative” (Mehta and Farina 1988) or “courtesy” stigma (Goffman 1963), in which social distance is desired from the biological relatives of the stigmatized individual.

In sum, attribution theory predicts that geneticization will reduce stigma, while the idea of genetic essentialism predicts that stigma will be exacerbated. However, different outcomes are implied by the two theories. Attribution theory predicts stigma reduction via reduced blame, anger, and punishment and increased sympathy and helping. Genetic essentialism predicts stigma magnification via increased perceptions of differentness, and—indirectly through increased perceptions of seriousness, persistence, and risk to family members—via increased social distance and reproductive restriction. Thus, it is possible that both theories are correct and operate simultaneously.

4.3.1.2 Empirical Evidence on Genetic Attributions and Stigma

We have already seen that genetic attributions increased over this period for alcohol dependence, schizophrenia and major depression but that there was no evidence that stereotypes of violence and incompetence, or the willingness to interact with people with mental illnesses changed for the better over the period between the studies (Pescosolido et al. 2010). This evidence challenges the idea that stigma will dissipate when the public is moved toward more medical and genetic views of mental illnesses. A review of additional evidence, much of it experimental, was conducted by Phelan and Link (2012). They located 17 studies that examined the association between genetic attributions and stigma-related outcomes for a variety of stigmatized characteristics (Angermeyer et al. 2003; Bennett et al. 2008; Boyle et al. 2009; Dietrich et al. 2004; Eker 1985; Feldman and Crandall 2007; Jorm and Griffiths 2008; Magliano et al. 2004; Martin et al. 2000; Menec and Perry 1998; Phelan 2002, 2005; Phelan et al. 2006; Piskur and Degelman 1992; Schnittker 2008; Teachman et al. 2003). These focused primarily on mental illnesses (N = 12) and usually specifically on schizophrenia (N = 11) and/or major depression (N = 9). The most commonly studied stigma-related outcome was social distance (N = 11). Other outcomes included blame, perceived dangerousness, unpredictability and incompetence, emotions of anger and sympathy, and intentions to help, punish or restrict reproduction. Nine of the studies employed experimental designs, randomly assigning research participants to be exposed to different causal statements. The eight non-experimental studies assessed participants’ causal beliefs and stigmatizing attitudes and measured the relationship between the two. The studies more often indicate a stigmatizing effect of genetic attributions than a de-stigmatizing effect, but the findings are far from consistent. Nineteen significant positive associations (P < 0.05) between genetic explanations and stigma, eight significant negative associations (P < 0.05), and twenty-eight non-significant associations were reported in these 17 studies. Importantly, these stigmatizing effects varied by outcome in ways that speak to the two theories (attribution and essentialism) that have been applied to understand them.

First, when outcomes specified by attribution theory—blame, anger, punishment, pity and willingness to help—were examined, 13 associations were found. Four were significant in a direction consistent with attribution theory whereas in the other nine no significant associations were uncovered. Second, when outcomes specified by the theory of genetic essentialism—seriousness, persistence, differentness and the spread of stigma to genetically related individuals—12 were found. Six associations concerned perceived seriousness and persistence and four of these were significant in the direction predicted by essentialist predictions. Genetic attribution was not significantly related to persistence in another instance and was not associated with perceived differentness in the one case where differentness was examined. Four other associations relevant to essentialist predictions concerned associative stigma measured in terms of desired social distance from the relative of a person who was described as having a genetically caused disorder. In all four instances significant associations were found.

In sum, the preponderance of results suggested stigmatizing rather than de-stigmatizing effects of genetic attributions, but this pattern was by no means strong or dramatic. There were a considerable number of non-significant associations as well as a number of instances in which genetic attribution was associated with less stigma. Focusing on outcomes related to the two theoretical perspectives with specific relevance to genetic explanations for stigmatized characteristics, the number of relevant findings is small, but somewhat more consistent. Although we still find non-significant associations, all statistically significant findings are in the predicted direction. Clearly, the number of findings reviewed by Phelan and Link is too small to draw firm conclusions; nevertheless they are suggestive that genetic attributions may have opposite effects on stigma through the dual pathways of reducing blame and increasing genetic essentialism.

More generally, and in keeping with the GSS results concerning population change, these results tell us that there is no easy fix to public stigma. We cannot just convince the public that mental illnesses are genetically based biological illnesses and expect issues of stigma to be resolved.

4.4 Personal Consequences of Stigma for People with Mental Illnesses

Interest in how people with mental illnesses are affected by stigma and how they seek to cope with or resist such stigmatization has been a central concern for members of the Section on the Sociology of Mental Health. We begin our consideration of these issues by focusing attention on modified labeling theory (Link 1982, 1987; Link et al. 1989) and evidence that has accumulated about that theory. It is an auspicious place to start because it links public conceptions (the focus of the last section) with personal consequences (the focus of this section).

Modified Labeling Theory. In the 1980s Link and his colleagues developed a “modified” labeling theory that derived insights from the original labeling theory, but stepped away from the claim that labeling is a direct cause of mental illness (Link 1982, 1987; Link et al. 1989). Instead the theory postulated a process through which labeling and stigma jeopardize the life circumstances of people with mental illnesses, harming their employment chances, social networks, and self-esteem. By creating disadvantage in these domains and others like them, people who have experienced mental illness labels are put at greater risk of the prolongation or reoccurrence of mental illness. The modified labeling theory also provided an explanation as to how labeling and stigma might produce these effects and how key concepts and measures could be used in testing the explanation with empirical evidence.

The theory begins with the observation that people develop conceptions of mental illness early in life as part of socialization (Angermeyer and Matschinger 1996; Scheff 1966; Wahl 1995). Once in place, people’s conceptions become a lay theory about what it means to have a mental illness (Angermeyer and Matschinger 1994; Furnham and Bower 1992). People form expectations as to whether most people will reject an individual with mental illness as a friend, employee, neighbor, or intimate partner and whether most people will devalue a person with mental illness as less trustworthy, intelligent, and competent. These beliefs have an especially poignant relevance for a person who develops a serious mental illness, because the possibility of devaluation and discrimination becomes personally relevant. If one believes that others will devalue and reject people with mental illness, one must now fear that this rejection will apply personally. The person may wonder, “Will others look down on me, reject me, simply because I have been identified as having a mental illness?” Then, to the extent that it becomes a part of a person’s worldview, that perception can have serious negative consequences that affect self-confidence, social relationships, employment and other life-domains.

To test this explanation, Link (1987) constructed a 12-item scale measuring the extent to which a person believes that people who have been labeled by treatment contact will be devalued and discriminated against. Respondents are asked the extent to which they agree or disagree with statements indicating that most people devalue current or former psychiatric patients by seeing them as failures, as less intelligent than other persons, or as individuals whose opinions need not be taken seriously. The scale also includes items that assess perceived discrimination by most people in jobs, friendships, and romantic relationships. The scale was administered to people with mental illnesses and to community residents from the same general area of New York City in a case-control study of major depression and schizophrenia. Link (1987) showed that the degree to which a person expects to be rejected is associated with demoralization, income loss, and unemployment. This association occurs in individuals labeled mentally ill but not in unlabeled individuals, thereby supporting the idea that labeling activates beliefs that lead to negative consequences.

Link et al. (1989) extended the forgoing reasoning in two ways. First they brought into the analysis empirical measures of coping orientations of secrecy (concealing a history of treatment), withdrawal (avoiding potentially threatening situations), and education (attempting to teach others in order to forestall the negative effects of stereotypes). Consistent with the idea that the stigma associated with mental illness creates expectations of rejection, they showed that people with mental illnesses tend to endorse these strategies as a means of protecting themselves. Second, the researchers extended the analysis to a consideration of the effects of these processes on social network ties. They found that people who fear rejection most and who endorse the strategy of withdrawal have insular support networks consisting mainly of household members.

Aspects of the theory have since been tested with a broader range of outcomes, in different samples, by different investigators, and often using longitudinal data. These studies generally showed that the perceived devaluation-discrimination measure is associated with outcome variables including quality of life (Rosenfield 1997), self-esteem (Link et al. 2001, 2008; Livingston and Boyd 2010; Wright et al. 2000), social networks (Link et al. 1989; Perlick et al. 2001) depressive symptoms (Link et al. 1997; Perlick et al. 2007), treatment adherence (Sirey et al. 2001) and treatment discontinuation (Sirey et al. 2001). But one particular aspect of the theory, the idea that cultural conceptions (perceived devaluation and discrimination) have effects on outcome variables in labeled persons but not in unlabeled ones, has not been as thoroughly tested in subsequent research. One study that did do so was undertaken by Kroska and Harkness (2006) using samples of psychiatric hospital patients and community residents in Indianapolis. This study operationalized cultural conceptions in a completely different way than Link and his colleagues did. Patients and community residents were asked to evaluate the concept “mentally ill person” using a seven-point semantic differential scale with opposing adjectives such as “good/bad,” “useful/useless,” and “powerless/powerful.” Respondents also rated the concept “myself as I really am” and “myself as others see me” with the same adjective pairings. The researchers’ modified labeling theory hypothesis was that associations between ratings of the concept “mentally ill person” and “myself as I really am” and “myself as others see me” concepts would be stronger in the labeled group (where a mental illness label is potentially personally relevant) than in the unlabeled group (where it is not personally relevant). Their results were generally consistent with this prediction, providing additional support for this key prediction of modified labeling theory.

Labeling as a “Package Deal.” Evidence from modified labeling theory and other approaches to labeling, stereotyping, and rejection strongly suggest that negative consequences associated with labeling are experienced by many people. At the same time, evidence from a voluminous body of research indicates that a variety of psychotherapies and drug therapies can be helpful in treating many mental illnesses. Given this, existing data simply do not justify a continued debate concerning whether the effects of labeling are positive or negative—clearly they are both. Rosenfield (1997) was the first to bring this point to light in a single study. She examined the effects of both treatment services and stigma in the context of a model program for people with severe mental illnesses. She showed that both the receipt of services (specific interventions that some people in the program receive and others do not) and stigma (Link’s 1987 measure of perceived devaluation and discrimination) are related—in opposite directions—to multiple dimensions of a quality of life measure. Receipt of services had positive effects on dimensions of quality of life, such as living arrangements, family relations, financial situation, safety, and health, whereas stigma had equally strong negative effects on such dimensions.

A second study that explores the idea of joint effects in opposite directions is one by Link et al. (1997). In a longitudinal study, men who were dually diagnosed with mental disorder and substance abuse were followed from entry into treatment (when they were highly symptomatic and addicted to substances) to a follow-up point one year later (when they were far less symptomatic and largely drug- and alcohol-free). Despite these dramatic benefits, the results also showed that perceptions of devaluation and discrimination and reported experiences of discrimination continued to affect the men’s level of depressive symptoms. Similar results, showing evidence of improvement in symptoms with treatment but enduring effects of stigma on self-esteem, were reported in a recent study by Link et al. (2008). The effects of stigma and discrimination endure and are apparently unaffected by any benefits of treatment.

Thus the evidence indicates a bundling of labeling effects that are currently joined in a kind of “package deal.” People seeking mental health treatment navigate this deal in one way or another. Sometimes they do so in ways that mental health professionals think they should not, such as avoiding treatment, denying their illness, or ending treatment earlier than their treatment team thinks they should. A kind of collective finger-wagging ensues that at times shifts from admonitions and warnings to using the “leverage” of housing or financial benefits to ensure treatment compliance (Monahan et al. 2001). If leverage fails, more direct forms of coercion are also possible such as involuntary inpatient commitment or the ascendant “outpatient commitment“(Hiday 2003). Of course, there is an intense debate about the utility and effectiveness of leverage and coercion, with some believing that these practices are necessary (Torrey and Zdanowicz 2001) and others seeing them as counterproductive (Pollack 2004). What a sociological perspective adds to this debate is evidence that there is indeed a package deal and that people face real choices and real dilemmas as they navigate its parameters. It also suggests that the ingenuity invested in constructing strategies to leverage compliance or to coerce it needs to be complemented or replaced by efforts that really change the balance of the package deal to one that delivers more benefit and less stigma. When that happens, more people will choose treatment, and less leveraging and coercion will be required. Mental health sociologists can help by continuing to unpack the package deal so that its existence is more widely acknowledged and our understanding of the mechanisms that undergird it more complete. Recent research by mental health sociologists is engaged in precisely these issues. For example, Markowitz and Angell (2011) elaborated on the modified labeling theory to probe more deeply into the mechanisms involved by including the reflected appraisals of family members in their empirical analysis. Lucas and Phelan (2012) conducted experimental work integrating paradigms from the expectation-states tradition in sociology (Berger et al. 1972) with work in psychology on the sources of stigma in interaction processes to investigate whether and to what extent a mental illness label reduces influence in interactions and engenders behavioral social distance. Identifying sources, probing mechanisms, and evaluating novel attempts to respond to stigma will deepen our understanding and enable us to address the stigma processes that affect people with mental illnesses.

4.4.1 Stigma Coping and Stigma Resistance

Individually-Based Stigma Coping Responses. The idea that people who are stigmatized actively respond to their situation has been a central element of theories about stigma ever since the concept emerged as a critical social scientific idea in the 1960s. It is, for example, a key component of classic labeling theory’s concept of “secondary deviance” as something brought on by “defense, attack, or adaptation” to the overt or covert problems produced by societal reactions to “primary deviance” (Lemert 1967, p. 17). And of course Goffman’s (1963) essay is all about the active “management” of stigma both by those who are the object of stigma and by those who do the stigmatizing. This active response to stigma is carried forward by Link et al. (1989, 1991, 2002) in the empirical elaboration of modified labeling theory through the conceptualization and measurement of stigma coping orientations. In earlier work, coping orientations of “secrecy” (concealing labeling information) “education” (providing information to counter stereotypes) and “withdrawal” (avoiding potentially rejecting situations) were measured and assessed (Link et al. 1989, 1991) followed by the addition of coping orientations of “challenging” and “distancing” (Link et al. 2002). Challenging is the active confrontation of stigmatizing behavior. For example, one might challenge by pointing out stigmatizing behavior when it occurs and by indicating that one disagrees with the content of stigmatizing statements or disapproves of stigmatizing behaviors. Distancing is a cognitive separation of the potentially stigmatized person from the stigmatized group. In distancing, one dodges the stereotype that others might apply or that one might apply to oneself by essentially saying—“I am not like them!” “Your stereotypes of them are misapplied to me.”

But if both classic and modified labeling theories have emphasized the active response of the stigmatized, what are the consequences of these efforts according to these research traditions? Are individually-based efforts to cope or resist effective? The concept of secondary deviance suggests not—at least not always. The responses of “defense, attack or adaptation” by the stigmatized induce additional “secondary” deviance that further sets the person apart (Lemert 1967). And, when Link et al. (1991) assessed the coping orientations of secrecy, education and withdrawal, they found no evidence that these approaches buffered people with mental illnesses from untoward consequences, but did find some evidence that these orientations actually exacerbated these consequences. They conclude that individual coping orientations are unlikely to be effective because they do not deal with the fundamental problem of deeply embedded cultural conceptions and stereotypes. According to Link et al. (1991), the best solutions are ones that change societal conceptions or involve the collective action of people with mental illnesses that change power differentials.

More recently, Thoits (2011) developed new concepts and theory suggesting the possibility of “stigma resistance” at the individual level particularly as it might protect the self-esteem of people with mental illnesses. Thoits points to what she calls a moderate association between perceived or experienced stigma and self-esteem and infers that a less than perfect association means that some people effectively counteract the effects of stigma on self-esteem. Thoits identifies two forms of resistance: “deflecting, impeding or refusing to yield to the penetration of a harmful force or influence” and “challenging, confronting, or fighting a harmful force or influence” (Thoits 2011, p. 11). In “deflecting,” a person responds to mental illness and associated stereotypes by concluding “that’s not me,” “that is only a small unimportant part of me,” or that the designation “mentally ill” does not apply to me because my problems are something different than mental illness. According to Thoits (2011, p. 14), deflecting strategies offer the possibility to “dramatically reduce, if not eliminate, potential threats to self-regard.” “Challenging” as described by Thoits (2011) differs from deflecting in that it involves an effort to change other people’s beliefs or behaviors. A person can challenge by (1) behaving in ways that contradict stereotypes, (2) educating others to move them away from stereotyped views, (3) confronting people who express prejudicial sentiments and behave in discriminatory ways, or (4) engaging in advocacy and activism.

Whether, to what extent and under what conditions stigma resistance can protect self regard or other potential consequences of stigma is an empirical question that has not been fully resolved. Our conjecture is that individually based efforts will generally fail. We base this in part on Link et al.’s (1991) study suggesting that at least three individually-based coping approaches (secrecy, withdrawal and education) were not effective in reducing distress or counteracting negative consequences for employment for people with mental illnesses. Additionally, although there is something alluring about the idea that the stigmatized can fight back or cognitively manipulate their orientation to stigmatizing circumstances, one must keep in mind that stigmatizers are actively pursuing their own interests at the same time. To the extent that stigmatizers have an interest in keeping people, down, in or away we can expect them to counter the efforts of stigmatized groups to resist with the exercise of power. Foucault’s famous aphorism “Where there is power there is resistance” can be turned around to read “where there is resistance there is power.” Agency is operative for both the stigmatized and the stigmatizer and it is likely that the ultimate outcome will depend on the relative power of the two groups. This leads to our pessimism about individually-based coping or resistance—the actions of single individuals are very unlikely to change the power difference between stigmatized and stigmatizing groups.

Group-based Resistance—Social Movements. We are much more optimistic about the long-term effectiveness of group-based resistance. One reason is that we can point to some social movements that have been at least partially successful such as the civil rights movement and the gay and lesbian liberation movement. Another reason is that sustained collective action over long periods of time affects a mechanism we believe is critical to the successful production of stigma—it alters the balance of power between stigmatizing and stigmatized groups thereby altering the capacity of the stigmatizing group to exert their desire to keep people down, in or away. In the long run it may even change the stigmatizers’ inclination to keep people down, in or away. This is not to say that collective social action proceeds in a linear fashion toward success. Instead collective action proceeds in fits and starts, sometimes gaining ground, sometimes losing it and sometimes failing altogether. But social movements usually aim to directly resist the power of the stigmatizing groups, thereby seeking changes that can be sustained over time. Interestingly, research has shown just how important sociological processes are in such social movements (Jasper 2011). For example, social movements of stigmatized groups often seek a shift in identity from shame to pride, set in place interaction rituals that sustain commitment, manipulate “moral shocks” to recruit new members and keep old ones engaged by highlighting or otherwise enhancing the emotional impact of events (Jasper 2011). In sum, sociological processes are critical to understanding stigma coping and resistance, and both individually-based and group-based resistance should be studied from a sociological perspective. What we expect is that individually-based efforts will be less effective in resisting stigma than group-based social movements and that this will be especially true if one adopts a long-term perspective.

4.4.2 The Future

This chapter provided examples of sociologically inspired efforts to conceptualize and understand mental illness stigma. A definition of stigma that includes multiple components and that is attentive to the power differences between stigmatizers and the stigmatized was one such example. Another was the sociologically informed inquiry into why people stigmatize with its focus on the interests stigmatizers have in keeping people down, in or away. Viewed from this vantage point stigma is form of power, a way that people achieve desired outcomes. The chapter also provided examples of research concerning public conceptions, how such conceptions are distinctly sociological and how much progress has been made in recent years in understanding them. Finally, the chapter provided an example in the form of modified labeling theory of a sociological approach to understanding how stigma harms people with mental illnesses and also evidence about how people seek to side step or resist the impact of stigma on their lives. In light of this work it seems fair to say that substantial progress has been made in the sociological understanding of mental illness stigma. At the same time, it is also true that large gaps in understanding remain. In particular, there is much to be done in terms of filling out the promise of some of the concepts and theories that have already been introduced and, of course, new concepts and understandings are likely to emerge in the time ahead. But we end with one recommendation for the future and that is the development of a sociologically informed approach to stigma change.

Sociologists are trained to be adept observers who use their sociological imagination to understand social processes. Of course this kind of understanding carries the capacity to create change, especially to the extent that it reveals processes that might otherwise be hidden. Whereas 50 years ago the word stigma was rarely used, today it is widely employed to bring to light the predicaments faced by people experiencing a wide variety of circumstances. It is difficult to quantify the effect of such a transformation but to the extent that people can identify the processes that trouble them we presume they will be more able to challenge and perhaps change them. Still, if stigma processes are deeply set in sociological phenomena, it would seem that sociologists could play an active role in conjuring ways to change stigma processes. This is rare. Elaine Cummings, a sociologist, collaborated with her husband John Cummings, a psychiatrist, in an effort to change attitudes in a town in Saskatchewan, Canada in the early 1950s. The failed attempt was published in the now classic and extremely instructive book, Closed Ranks (1957). But since that undertaking, sociologists have only rarely engaged in efforts to change mental illness stigma. We believe that such an undertaking represents a strong challenge for the future mental health sociology—to lay out a sociologically informed strategy and to suggest lynch-pin intervention approaches that could move that strategy forward.