Abstract
The current study reexamines modified labeling theory with a sample of 15 women incarcerated in two medium-security prisons and who have a diagnosable mental illness. Life history calendars (LHC) were employed to document traumatic histories and a host of risks and needs. Qualitative findings indicate that these women negotiated labels of mental illness and incarceration. The labeling process, particularly self-concept, self-esteem, and stigma, emerged as driving forces for criminal behavior, dysfunctional interpersonal relationships, and decision-making. Participants negotiated the label of mental illness, while making meaning of early and continued trauma. These findings are explained with reference to trauma-informed care (TIC), improved staff training, and the role of language. This study offers evidence that labeling concepts may serve as a language foundation for providing trauma-informed services.
It has long been noted that incarcerated women are diagnosed with higher rates of mental illness when compared with their incarcerated male counterparts (Bronson & Berzofsky, 2017). Of women who are incarcerated in state prison, 66% have been found to have a mental illness compared with 35% of men in state prison, while 68% of women incarcerated in jail have been found to have a mental illness compared with 41% of men in jail (Bronson & Berzofsky, 2017). While these differences have been attributed to social factors such as stereotypes underlying masculinity/femininity, a willingness to self-report symptoms and seek treatment, the externalization or internalization of negative emotions, and the availability of mental health resources, they warrant further scholarly attention. This is a particularly salient issue as correctional systems are now operating as default mental health facilities for a substantial portion of society (Smith, 2022). Previous research suggests that the impact of the symptoms of mental illness affect individuals in a variety of ways, particularly as it relates to stigmatization and self-conceptualization (Kroska & Harkness, 2006, 2008; Link et al., 1989). Regarding gender, it is important to understand how mental illness affects the identity of women before and during incarceration.
The effect of a label on an individual has been described and explained primarily through labeling theory, which posits that when an individual is labeled a deviant or a criminal, they adopt that self-concept and behave in a devious or criminal manner (see Mead, 1934; Rosenberg, 1979). An extension of labeling theory, modified labeling theory, addresses how the “mentally ill” label affects an individual. Specifically, a negative self-concept resulting from mental illness stigmatization can result in poor health and life outcomes (Kroska & Harkness, 2008; Wright et al., 2000). While modified labeling theory stems from labeling theory (which has been largely applied to deviant or criminal populations), modified labeling theory has yet to be fully applied to a female correctional population. Applying modified labeling theory to a population of incarcerated women with mental illness could provide insight into how their mental illness has impacted them and how it has affected their overall identity, decision-making processes, and thus guide policy and therapeutic interventions. Moreover, for women incarcerated in prison, research can often be stifled by life histories that are complex, confusing, and traumatic.
To address these gaps, the current study analyzes the meaning of mental illness to incarcerated women in a sample of 15 women with diagnosable mental illness who are housed in a medium-security prison. By following the central tenets of labeling theory, specifically modified labeling theory, the current study examines the perceptions of these women. To measure the considerable histories of the participants, a Life History Calendar methodology is used, where a schema of key events and developmental categories are established during intensive semi-structured interviews.
Labeling Theory
Behavior that is considered within the dichotomies of normal/deviant or legal/criminal varies across countries and cultures, making labels and the definitions attached to the labels subjective. Becker (1963) argued that those in positions of moral authority, such as legislatures, create the rules and laws defining what is deviant. This is a necessary part of society from a functionalist perspective because that moral authority would exist even in a “society of saints,” where any behavior outside of social norms would be met with punishment designed to preserve social order. The negative reactions to deviance and/or crime from society inform those who commit deviant acts that this behavior is unacceptable, which helps maintain order (Durkheim, 1993/1893). Labeling theory, which was developed from Durkheim’s functionalist approach, recognizes that the purpose of the deviant and criminal label is to maintain societal order; however, it contends that the labeling process results in additional deviant and criminal behavior (see Erikson, 1962).
According to labeling theory, deviance consists of two categories: primary and secondary. Primary deviance is the initial action committed by an individual that can result in being labeled a deviant or receive legal sanctions. Secondary deviance is deviant behavior that is a product of being labeled a deviant and is at the center of labeling theory (Lemert, 1951). Under secondary deviance, when an individual violates the law, they are labeled a criminal even when they have been law-abiding any other time. Therefore, while correcting a person’s behavior may be one of the primary purposes of punishment, the person is instead stigmatized and cast into the category of the “other” (Erikson & Puritans, 1966). Not only is the deviant behavior viewed as “evil,” but the act itself manifests as an identity within the person who committed the deviant act (Tannenbaum, 1938). The adoption of the deviant label has been termed a “self-fulfilling prophesy” (Erikson, 1962, p. 311). As a result, the label serves as a stigma, rather than a means of reforming behavior.
It is important to clarify that labeling theory does not postulate that once an individual is labeled, they are destined to lead a life of deviancy and/or crime, but that they are more likely to commit further acts of deviance or crime. The effects of a label can be mitigated by factors such as whether the labeling process was public or private. The labeling effect in public and/or formal settings will be stronger than when conducted in private, as a public labeling process is viewed as being entrenched and shameful (Paternoster & Iovanni, 1989). Labeling effects can also be mitigated by how the labeled individual is treated by their close friends and family. If close friends and family do not view the labeled individual differently, or treat them differently, because they knew who they were before their actions, then they might be able to resist the label. Furthermore, the individual may disagree with the label, therefore resisting the label, which is considered “deviance disavowal” (Davis, 1961).
Empirical Support
When someone adopts a deviant label, they are viewing themselves and basing their behavior on the way that others see them, which is termed a self-concept (Mead, 1934; Rosenberg, 1979). A deviant self-concept—viewing oneself as deviant—is inherently linked to risk of committing future deviant and/or criminal acts. The self-concept in this section will refer to the deviant and/or criminal label, although the self-concept will be addressed later regarding the “mental patient” label to which this definition also applies.
In one early test of labeling theory, Farrington (1977) found that additional acts of deviance (secondary deviance) were linked to the labeling process, but he was unable to determine the effect of a deviant self-concept that may emerge during primary deviance. Scholars, however, have continued to highlight the potential role of one’s self-esteem or self-concept (Cechaviciute & Kenny, 2007; Matsueda, 1992; Ray & Downs, 1986; Restivo & Lanier, 2015). Ray and Downs (1986) hypothesized that an individual with high self-esteem might be less likely to accept a deviant label compared with an individual with low self-esteem. To evaluate this, Ray and Downs (1986) included the Rosenberg Self-Esteem Scale in an assessment of the Adolescent Drug and Alcohol Intervention and Referral (ADAIR) program. While self-esteem was not found to be statistically significant, the self-label was a significant predictor of future drug use for males, but not for females whose drug use remained constant. Ray and Downs (1986) suggested that this was due to gender differences in social desirability, with male participants being less influenced by the perceptions of others when compared with female participants.
Matsueda (1992) studied the notion of self-esteem, which he termed “reflected appraisals,” of youth and their future delinquency by measuring youths’ rule violations with parents’ appraisals of them. In support of labeling theory, the way these youth viewed themselves was significantly impacted by how their parents viewed them, while also influencing their participation in delinquent acts (Matsueda, 1992).
In addition, assessing the impact of adjudication on juveniles, Murray et al. (2017) found that when a juvenile is adjudicated for delinquent acts, they are at an increased risk for criminality in adulthood. Liberman et al. (2014) examined the impact that arrest has on future delinquency and arrest among juveniles finding that those who had been arrested were significantly more likely to offend among all of the offending categories including violence, property crime, and selling drugs (Liberman et al., 2014). Motz et al. (2020) examined the impact of criminal justice contact on subsequent delinquency in a sample of twins. For their study, one of the twins did not make criminal justice contact while one did. Results indicate that the twin who was labeled via criminal justice contact continued to offend (Motz et al., 2020). In a large sample of 95,919 adjudicated adults, Chiricos et al. (2007) found that those who were convicted of a felony were more likely to recidivate within two years when compared to people whose conviction/s were withheld. Labeling effects were stronger for women, whites, and people under the age of 30 (Chiricos et al., 2007).
During the 1960s and 1970s, there was an abundance of research on labeling theory, which focused primarily on deviant or criminogenic concerns often excluding other deviant behaviors that co-occur, such behavior resulting from mental illness. This is a noteworthy gap in research as mental illness greatly increases the likelihood that an individual will come into contact with the criminal justice system (Smith, 2021b).
Modified Labeling Theory
The emergence of scholarly work that examines the labeling processes of mental illness within a criminal justice context has largely occurred within modified labeling theory. An extension of labeling theory, modified labeling theory, addresses how the “mentally ill” label affects an individual. Using data that were collected by employing vignettes for psychiatric patients, Link (1982) examined the effects of both the symptoms of mental illness and being labeled mentally ill (through treatment) on employment and income. He found that, controlling for other variables (marital status, education, age, and occupation), patient status had as much of an effect on income and employment as did the symptoms. This study led Link (1982) to conclude that labeling theory was useful in exploring how a label impacts other areas of an individual’s life, which led to the development of modified labeling theory.
According to modified labeling theory, the stigma of a mental illness label impacts an individual’s life course by affecting their peer and social support networks, educational and employment opportunities, and access to other social resources (Link et al., 1989, 1991). Link and colleagues (1991) posited that these networks and opportunities are impacted by the labeled individual adapting to or coping with the negative label, concealing their diagnosis, aversion to socializing with others, or attempts to prevent stigmatization by educating others. This results in the labeled individual avoiding people that are outside of their trusted social group, thus limiting educational and job opportunities (Link et al., 1989, 1991).
Stigmatization
According to Goffman (1961), a stigma manifests when an “attribute that is deeply discrediting” (p. 3) is placed upon an individual and the concomitant reaction. A more contemporary description has included the concepts of labeling, separating, status loss, and discrimination (Link & Phelan, 2001). As these definitions suggest, stigma is not a ubiquitous nor standardized concept. The stigma attached to a voluntary action such as a criminal behavior may differ from involuntary occurrences such as childhood trauma, adult victimization, or mental illness. As such, the devaluation-discrimination scale has been a crucial tool in gauging stigmatizing views of people with mental illnesses (Link et al., 1989). The scale consists of 12 items measuring the extent to which a person believes that most people will devalue or discriminate against someone with a mental illness (Brohan et al., 2011). Using this measurement approach, there is evidence that self-stigma negatively impacts people with mental illness, particularly via the emergence of alienation, stereotype endorsement, social withdrawal, and discrimination experiences (Brohan et al., 2011). Link et al. (1989) explained that self-stigma is inherently connected to treatment, with early interactions with mental health professionals often leading to new conceptualizations of self, as labeling becomes more salient. This was supported through Oexle et al.’s (2017) study on the impact stigma has on suicidal ideation between those labeled mentally ill (N = 429) compared with those not labeled (N = 252). Participants were included in the labeled group based on their use of mental health services. Results indicated significant association between perceived stigma and suicidal ideation among the labeled participants and not the unlabeled participants (Oexle et al., 2017).
Dohrenwend et al. (1985) hypothesized that individuals with mental illness hold beliefs, developed through socialization, of how others will view and treat them based on the notion that mental illness is generally perceived negatively by society. This hypothesis was supported with a sample of patients who had been diagnosed with major depression and schizophrenia. Likewise, Rosenfield (1997) found that 65% of individuals who were receiving treatment for mental health expressed perceptions that society would not view them as trustworthy or as intelligent as people who do not have mental health issues. Further support was found in a sample of mental health patients with 72% agreeing that they are consistently “devalued and discriminated against” (Markowitz, 1998, p. 339). Together this has guided additional research on the labeling of mental illness, with significant research finding evidence of stigmatization (Kroska & Harkness, 2006; Link & Phelan, 2001; Martin et al., 2000) and negative self-concept (Kroska & Harkness, 2006, 2008; Markowitz, 1998; Markowitz et al., 2011; Martin et al., 2000; Rosenfield, 1997; Wright et al., 2000).
Of note, respondents are often more likely to support interacting with an individual who has mental illness when the cause is presented as biological (i.e., something that the individual did not have control over; Martin et al., 2000). In contrast, individuals perceive being less likely to interact with a person who had a mental illness if the cause is presented as the product of socialization during childhood (Martin et al., 2000).
Self-Concept and “Mental Patient” Label
The stigmatization that mentally ill individuals perceive from others has been found to negatively affect their self-concept (Kroska & Harkness, 2006, 2008; Markowitz et al., 2011; Markowitz, 1998; Rosenfield, 1997; Wright et al., 2000). In studies of the self-concept among individuals with mental illness, the term self-concept is not always used, instead, “self-concept” is often replaced with “self-identity” or “self-appraisal.” Measuring self-concept can relate to the individual’s self-esteem or how they view their quality of life (Rosenfield, 1997). Other scholars have combined self-identity with reflected appraisals to represent “self-meaning.” Reflected appraisals refer to how the individuals view themselves based on how others see them. Therefore, self-meaning is the combination of their own identity and how they believe others see them (Kroska & Harkness, 2006). Furthermore, Markowitz and colleagues (2011) added the measure “significant other appraisals” to not only understand how persons with mental illness view themselves and how they view themselves based on the opinions of others, but also how mothers of people with mental illness view their own children.
Kroska and Harkness (2006) presented what they call “stigma sentiment,” to capture stigma more accurately as it relates to the cultural concepts of people with mental illness. Stigma sentiments are the evaluation, potency, and activity of that cultural concept. Concerning self-identity, stigma sentiments were hypothesized to be unrelated to the self-identity of mental health patients and nonpatients, which was supported. Concerning reflected appraisals, stigma sentiments were hypothesized to be positively related to the reflected appraisals of mental health patients, but unrelated to nonpatients, which was supported (Kroska & Harkness, 2006). They expounded on this study with the addition of diagnostic categories and questioned if stigma sentiments were related not just to mental illness in general, but the type of mental illness (adjustment, affective, or schizophrenic) and found that the type of diagnosis had a mitigating effect on their self-identity (Kroska & Harkness, 2008). This negative self-concept was adversely affecting the individual’s life satisfaction, for patients newly diagnosed and those who have been receiving treatment for some time (Markowitz, 1998; Rosenfield, 1997; Wright et al., 2000).
Current Study
Traditional labeling theory has focused on how the deviant or criminal label affects the individual. Specifically, that the individual is stigmatized with the label and then accepts the label, which results in further deviance (Erikson, 1962; Lemert, 1951). While labeling theory has focused specifically on the label of “deviant” or “criminal,” modified labeling theory was developed to explore the label of “mentally ill” (Link, 1982; Link et al., 1989). Modified labeling theory has been applied in research regarding the stigma that being labeled mentally ill has on an individual (Kroska & Harkness, 2006; Link & Phelan, 2001; Markowitz et al., 2011; Martin et al., 2000), and to understand the development of a self-concept on an individual with mental illness (Kroska & Harkness, 2008; Markowitz et al., 2011; Restivo & Lanier, 2015). Yet, to date, there has been a lack of studies that examine modified labeling theory with incarcerated samples who have a diagnosable mental illness.
Although both male and female incarcerated populations could provide valuable insight regarding stigmatization, women routinely display higher rates of mental illness than men in both prison and jail populations, and there is evidence that gendered pathways toward mental illness, crime, and incarceration are distinct. For example, incarcerated women have unique needs such as those related to their medical needs (see Binswanger et al., 2011; Pickett et al., 2018), their role as mothers (see Glaze & Marushak, 2010), and their interpersonal relationships. Employing modified labeling theory in a sample of incarcerated women with mental illness could shed some light on their meanings of being labeled mentally ill and how it relates to their risk and needs. This work can be guided by previous empirical studies that employ the Life History Calendar as a means of facilitating rich data (Axinn et al., 2020; King & Smith, 2023; Lynch et al., 2012; Smith & Power, 2015). The goal being to provide policy implications for a vulnerable group, one that often has complex and challenging needs.
Method
Sample
The current study features a purposive sample of 15 women from two medium-security prisons in the Southeast. At the time of the study, Prison 1 had a population capacity of 706 with 556 women housed, whereas Prison 2 had a population capacity of 844 with 610 women housed. Prison 1 is the medical facility for women with physical or mental disabilities, debilitating conditions that necessitate around-the-clock medical care, and women who are pregnant. These are the only female prisons for the entire state, and they include a range of housing and programmatic settings, including mental health dorms, major special management units, and female death row.
Sampling was purposive to include the following: (a) females currently incarcerated in prison, (b) a diagnosable mental health disorder and/or condition, and (c) not currently placed on disciplinary or lockdown status. Previous research on females with mental illness in this prison system (King & Smith, 2023) found that a subset of these incarcerated persons received contradictory and conflicting diagnoses throughout their life course (i.e., school, foster care, hospitalizations in the community, during incarceration, etc.). For example, one participant had received the diagnoses of bipolar disorder, major depressive disorder, and schizoaffective disorder at different stages of her life course. As such, the focus of the study was on a sample of incarcerated women who had been classified by the prison system as having a mental illness, rather than a strict clinical approach using the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5; American Psychiatric Association, 2013). The prison system where these participants were housed has long documented rates of 66% of incarcerated females having a diagnosable mental illness.
Procedure
Nineteen participants were recruited through mental health professional staff at two female prisons resulting in a 79% participation rate. A member of the research team met with a total of nine mental health personnel from both prisons and detailed the study goals, methods, and procedures. These staff were asked to assist in recruiting women who had regular meetings with the staff member related to their diagnosable mental illness. Participants were recruited at the discretion of the mental health personnel. Due to the staff member having an established working relationship with the women, it was decided not to provide a script, but to rely on the expertise of the staff in these interactions. The final sample characteristics can be seen in Appendix A: Participants’ Information. Once participants were recruited, the mental health professional scheduled a day and time for the interview to take place with the researchers.
All interviews were conducted in a private office with the door closed, whereby participants received an explanation of the study purposes, the informed consent form, details of confidentiality protocols, and contact information of the researcher. Interviews were recorded and later transcribed by one of the authors for analysis. No data were collected from the women who decided not to participate in the study. The interviews took place immediately before the participant’s regularly scheduled meeting with a mental health professional. This approach was designed to limit any negative emotional reactions to the study, as participants were able to meet with certified professionals. To ensure that participation was entirely voluntary in this vulnerable population, participants were not compensated for participating in the study. An institutional review board (IRB) application was submitted and approved by both the University of South Carolina IRB and the research division of the prison system under study. All participants were provided contact information for the University IRB and committee.
Analysis
Due to the complex and often confusing life course of these female participants, a Life History Calendar (herein LHC) approach was used as a means of gathering data. In a sample where there is significant early childhood abuse, trauma, criminal behavior, adult victimization, violence, and incarceration experience, the LHC method was deemed suitable. As Yoshihama et al. (2002) explained, when using an extensive timeframe, using shorter time intervals such as “months” or “years” make recalling events more difficult for the participant when compared to recalling events of their life. A blank example of an LHC calendar can be seen in Appendix B: Life History Calendar. In practice, the LHC relies on a large paper sheet consisting of an X- and Y-axis. The X-axis consisted of four stages: childhood, adolescence, adulthood, and incarceration. These categories were based on previous research of incarcerated people using the LHC (King & Smith, 2023; Lynch et al., 2012; Smith & Power, 2015; also see Axinn et al., 2020), and with recognition that the traditional category of desistance from crime in a study focused on mental illness was not suitable. The Y-axis consisted of life event domains, which served as anchors or memory cues (Bellair & Sutton, 2018; Roberts & Horney, 2010; Yoshihama et al., 2002). The life event domains consisted of topics related to school, residence, cohabitation, intimate relationships, childbirth, drug/alcohol use, victimization, witnessing violence, family problems, mental health treatment, support, and turning points (Lynch et al., 2012). The interview, accompanied with the LHC, included questions specifically related to symptoms of mental illness or the diagnosis of mental illness during childhood, adolescence, adulthood, and incarceration.
Analysis was based on a constant comparative analytic approach to develop thematic outputs. Once transcribed, participant responses were entered into a database and analyzed using Excel and ATLAS/ti. Software (ATLAS.ti). The thematic analysis started with open coding, a microanalysis of the data where the goal is to generate initial code categories. Thematic analysis begins with a microanalysis of the data, where open coding schemas are used to generate categories in responses. Then axial coding is used to further establish categories and relationships within the data. The final step centers on selective coding, where these categories are merged and refined to meet the theoretical structure. This process continues until no new themes emerged, and thematic saturation is achieved (Braun & Clarke, 2006; Green & Thorogood, 2004). The thematic analysis occurred in line with the central theoretical tenants of modified labeling theory, with the central focus being the perceptions of incarcerated women with a diagnosable mental illness.
Results
Analysis of the data produces several themes in this sample of incarcerated women, including coming to terms with the mental illness label, labeling mental illness and criminogenic behavior, and labeling and stigma.
Coming to Terms With the Mental Illness Label
Participants expressed a negotiation approach to internalizing the label of “being mentally ill” with some participants explaining that they are close to “accepting” their mental health diagnosis while others used language to reconceptualize mental illness. Six of the participants expressed acceptance of the label by acknowledging that they have a lifelong illness, highlighting that they need medication, stating that they finally have answers to unresolved issues, perceiving that their mental illness is not their fault, or articulating that their mental illness was controllable though not fixable. Acceptance of the label of mental illness was evident in the following statements: I think medicine is better in me . . . and I think for the long run I think I’m [going to] have to be on this for the rest of my life because it settles me. (R15) It’s helpful for me because I know, like I said, now that I know what’s wrong, I know how to handle it, how to deal with it . . . so if anything it, it helps knowing than me not knowing, so I’m okay with it. (R13) I hate it. I hate it but I can’t change it. (R10)
For other participants, the labeling process of mental illness was influenced by a reconceptualization of abnormal psychology being normal, particularly in the context of a prison setting. This can be seen here: I just feel like it’s normal . . . It’s a normal part of life . . . I have not felt like I was normal until I came here at this institution and have been placed with people that experience the same thing. (R4) I feel like it’s normal . . . I think everybody’s got some kind of crazy in [them]. (R9)
Normality in mental illness was also extended to include reflections on early childhood abuses. It is here that labeling during adulthood served as a relief that could explain symptoms occurring during childhood and adolescence, as one participant states: I remember being in like maybe third grade and having an anxiety attack and didn’t know what it was. I thought something was wrong with me. (R13)
Labeling Mental Illness and Criminogenic Behavior
Participants described ways by which the labeling process of mental illness was linked to their engagement in crime. This included a negative influence on their own behavior and the behavior of others. Several participants noted that the mental illness label was linked to coping responses and to feelings of anger, which can be seen in the following accounts: My mental status has a lot to do with my anger and rage . . . I’m an assaultive inmate because I’ve been in so many fights. (R6) It gets me in trouble. Gets me in a lot of trouble . . . I just get mad and my anger and I just start thinking about everything that’s going on with me and I just lash out and just take it out. (R3)
One-third of the participants felt that the label of mental illness had a direct connection to the crime they were incarcerated for, and each described how their mental illness and offense were intricately related. For one participant, the connection was from a failure to comply with antipsychotic medication. For other participants, their mental illness impacted their decision-making. The impact of the mental illness label on their crime can be seen in the following accounts: Because of what I’ve done this time . . . I was having paranoia schizophrenia. (R8) I know and I think that if I had medication when I was on the street . . . my actions would have been different. I wouldn’t have been quick to do things that I was just doing out of boredom or felt like I had to do . . . And I probably would’ve coped with whatever I had going on more better if I was on medicine . . . I probably wouldn’t have been here. (R15) In a way I probably wouldn’t have made the decision with them people that night, if . . . I wasn’t dealing with mental illness. (R11)
Participants were also apt to assign the label of mental illness to their own victimization. This occurred even in recollections where there was an absence of a mental health diagnosis for the participant. One participant explained that people with mental illness “don’t think about others,” because she did not care about the feelings of others before she was incarcerated. Likewise, another participant perceived that mental illness, specifically bipolar disorder, was the cause of her mother’s aggressive demeanor and violent actions: She had . . . a lot of some anger, like violence . . . She had an extremely bad temper. I put it that way. (R6)
According to another participant, labeling her brothers as mentally ill was a technique used to explain the sexual abuse she experienced during her childhood and adolescence: I would assume that it was an issue of mental health for my brothers to start molesting me. And that doesn’t seem like it would be like normal thinking. (R2)
Labeling and Stigmatization
The majority of participants (11 out of 15) identified a stigma with being diagnosed with a mental illness. The participants expressed feeling different from others, being subject to labeling either directly or vicariously, and attempts at rejecting the label. Shame of having a mental illness was also acknowledged. Awareness of the stigmatization of having a mental illness was specifically addressed by seven of the participants. The stigma was something that they were aware of from childhood and these reflections included negative associations to their prescribed medication regime. This often carried the label of being “an outcast” (R4). Experiences related to mental health stigma can be seen here: I wouldn’t get on medicine because in my mind, only the kids at school were on medicine were kids that were crazy, so I didn’t want to be seen as a crazy kid. (R13) They even say “even though she’s bipolar and crazy . . . deep down she’s, she’s good people, she’s sweet, she’ll do anything for you.”(R12) I’m mentally ill not stupid . . . I’m mentally, I have a mental illness. I’m not stupid. I’m smart, I am mentally ill, but I can say I’m smart. (R8)
The stigma of mental illness also occurred during incarceration, with six of the participants noting that they were labeled as well as other incarcerated women being labeled with negative terms. This “name calling” included terms such as “slow,” “sick,” “crazy,” plus phrases such as “you aren’t all there” and “something is wrong with you.” One participant recalled stigma when a mental health counselor ascribed a label to her following a suicide attempt during adolescence, which is described in the following account: She would say . . . “do you realize you tried to kill yourself and only crazy people kill yourself, try to kill they self . . . what’s wrong with you?” (R14)
Labeling was also contextual within a prison setting with two participants stating that they experienced individuals close to them labeling them, though they differentiated between serious versus joking expressions. One of the participants acknowledged being called “crazy” in a “playful way.” The other participant stated, I mean, when you’re, not meaning, it’s one thing if we’re joke . . . laughing and talking, but it’s different ways of saying. Your inflection on how you’re saying it. (R1)
Five of the participants expressed, either explicitly or implicitly, that the labeling process produced shame. After stating a variety of diagnoses to the researcher, one participant said that saying them made her sounds like “a hot mess” (R1). Similar responses include: Sometimes I feel ashamed, I feel like I’m not normal. (R8) I think a time before on medication I was embarrassed to say that I needed it because I felt that I was a strong person that I could do it on my own. But after talking to my mental health counselor, she let me know that . . . it might be a slight chemical imbalance . . . “it’s not your fault. It’s nothing to be ashamed of.” Once she told me that, I felt okay about it. I’m okay with it, you know. It’s nothing that I’ve done. (R14)
Four of the participants described instances where they had rejected the negative labels that were ascribed to them or others. For one participant, rejecting the label of mental illness was a long-term course of action and she reflected, “I don’t let it define me as much as I used to” (R8). Some participants expressed overt rejection of labels by other incarcerated women and correctional officers, which is evident in the following accounts: You get a 20-year sentence, you might be on medication, too! (R14) I’m not crazy, I may be mentally ill, but I’m not crazy by any stretch of the imagination. (R1)
Discussion
The current study highlights several key themes in a sample of women with serious and persistent mental illness incarcerated in a medium-security prison. While insightful it is not without limitations. Generalizability was restricted by a sample of 15 female participants from two medium-security prisons located in one southeastern state (though this constituted the entirety of female prisons for the entire state). Moreover, while the sample was purposive to include participants who had been diagnosed with a mental health disorder, there is recognition that intake data, clinical data, and institutional records were not included. In line with the study methodology, the perceptions of these participants were recorded and analyzed, though no additional documentation was merged to create a triangulation of data. Future studies that use modified labeling theory to examine mental illness occurring in corrections could also benefit from the inclusion and comparison of different perspectives, such as incarcerated populations and the correctional officers who have contact with them (see Smith, 2021a).
With these limitations in mind, several key themes did emerge from the analysis. The first theme, coming to terms with the mental illness label, consisted of accepting the diagnosis of having a mental illness or by viewing it as normal. Scholars have questioned if individuals with high self-esteem are better equipped to reject the label that comes with delinquency or deviancy with the findings being mixed (see Cechaviciute & Kenny, 2007; Matsueda, 1992; Ray & Downs, 1986; Restivo & Lanier, 2015). According to Link (1987), being diagnosed with a mental illness is going to impact one’s self-esteem, which, in turn, impacts one’s self-concept. How much it impacts one’s self-concept might be mitigated by their level of self-esteem prior to diagnosis as well as the type of diagnosis they receive (see Kroska & Harkness, 2008).
This labeling process was not produced by a single event (i.e., a diagnosis, incarceration, etc.) but rather a series of discrete events over the life course. Labeling was more the product of repeated phrases by people these women interacted with over the life course. In line with modified labeling theory, these incarcerated women were aware of the negative perceptions of mental illness in society, and the impact of these perceptions on their own actions (see Dohrenwend et al., 1985; Link, 1987; Rosenfield, 1997). There was clear linkage between the label of mental illness and antisocial behavior, as the sample highlighted negative interactions with others, engagement in criminal behavior, and using mental illness to explain the behaviors of their abusers. As such, the label of mental illness was internalized (as self-concept) and externalized (as a form of meaning making to explain their own traumatic experiences, as well as the unpredictable, aggressive, and/or abnormal behaviors of other incarcerated women). The context of prison was paradoxical, as being incarcerated with a majority of other women who had a mental illness led to the destigmatization of mental illness. The high prevalence of mental illness allowed for perceptions that abnormal behavior was in fact normal. Conflict arose when these perceptions moved toward the labeling of symptoms of mental illness indicating a lack of intelligence, a notion that these incarcerated women strongly rejected.
The women in this study were double-stigmatized, being both diagnosed with a mental illness and being incarcerated. Stigmatization consisted of being aware of the negative label of having a mental illness, including the manifestation of symptoms, feeling shame in having a mental illness, and attempts to reject or transform the label. There was universal support for the notion that having, or being perceived as having, a mental illness carries stigma. This supports the hypothesis of Dohrenwend et al. (1985) that people with a mental illness are socialized knowing how others will see and treat them differently from people without a mental illness. Specifically, negative stereotypes regarding mental illness within this sample may lend support to Rosenfield’s (1997) study where individuals with a mental illness expected others not to trust them or see them as intelligent. This has obvious consequences in prison, where perceptions of dishonesty have risks of punitive responses from correctional staff, and perceptions of a lack of intelligence may reduce motivation and opportunities for participating in programs.
Implications for Theory and Practice
The current study has implications for labeling theory, specifically modified labeling theory. Labeling theory has focused on the deviant or criminal label; therefore, it was helpful to use tenets of labeling theory with a sample of incarcerated women. In addition, because modified labeling theory concerns the stigma of being diagnosed with a mental illness (Kroska & Harkness, 2006; Link & Phelan, 2001; Markowitz et al., 2011; Martin et al., 2000), it was beneficial to apply tenets of modified labeling theory to a sample of incarcerated women who have been diagnosed with a mental health disorder and/or condition. Scholars have suggested that the label of mental illness affects how an individual views themselves and the way they think others view them (Dohrenwend et al., 1985; Link, 1982). The current study indicates that the label of “mentally ill” affects how the participants view themselves during incarceration as well as how they think others view them. While labeling theory has its roots in the social movements of the 1960s and 1970s, there is applicability to the modern-day correctional system that operates as a default mental health system. Specifically, the dual stigmatization of being diagnosed with a mental illness and being incarcerated has yet to be fully explored, and there is room for further development of modified labeling theory in this context.
The negotiation between the label and their self-concept might be different in the correctional setting versus the outside world or society. In the correctional setting, there is a higher population of individuals who have a mental illness compared with the general population. It would be important to understand how being in an environment with this higher population of individuals with mental illness impacts the negotiation between the label and the self-concept. For one of the participants, being incarcerated with other women who had a mental illness made her feel more normal and less stigmatized. For another participant, being around other women with mental illness did not mitigate the negative self-concept that she had to overcome. She was adamant that although she was diagnosed with schizophrenia, she was smart. She was aware that others viewed her as unintelligent due to her mental health diagnosis.
Future research is needed to replicate the study in a sample of incarcerated men who have been diagnosed with a mental illness. While there are higher rates of mental illness among incarcerated women than incarcerated men, there are higher rates of mental illness among incarcerated men than in the general population. Thirty-five percent of men in state prison and 41% of men in jail reported having a mental illness compared with 15.8% of men in the general population (Bronson & Berzofsky, 2017; Substance Abuse and Mental Health Services Administration, 2020).
There is a lack of training on mental health and trauma for correctional officers (DeHart & Iachini, 2019). The majority of mental health training for correctional staff involves crisis intervention followed by general mental health education (Kois et al., 2020). DeHart and Iachini (2019) created a promising and accessible training program for correctional staff to educate them on mental illness, trauma, and appropriate responses to incarcerated individuals in crisis. Having all correctional staff receive mental health training could help in minimizing the labeling that the incarcerated individuals experience, which could help reduce the stigma of mental illness in the prison environment. In addition, several women brought up that they are targeted and stigmatized in the institution because when it is time to take their medication each day, they have to line up in the “pill line” with other incarcerated persons to get their cup of medication from the prison staff. Participants stated that even if they do not disclose their mental health status to other incarcerated persons, others suspect that they have a mental illness and will label them as “crazy” or make fun of them for taking medication. Correctional institutions might be able to take better precautions to minimize the labeling and stigmatization experienced by incarcerated individuals with mental illness by not distributing medication through a pill line that everyone can see, and taking similiar steps in other correctional procedures.
In terms of practice, the current study lends considerable support to the need for trauma-informed care (TIC). TIC centers on topic areas that often drive the risks and needs of incarcerated women with mental illness, specifically outreach and engagement, screening and assessment, resource coordination and advocacy, crisis intervention, mental health and substance abuse services, trauma-specific services, parenting support, and health care (Elliot et al., 2005). As Elliot and colleagues (2005) stated, “histories of trauma and violence are integrally intertwined with histories of substance abuse and/or mental health disorders among women” (p. 473). This suggests that programs for these incarcerated women should recognize patterns of substance abuse and mental illness (including self-injury and suicidal behaviors) as they reflect attempts to cope with trauma. Correctional policies and practices should be guided by TIC, to ensure that they are safe, appealing, and relevant to survivors of trauma (Harris & Fallot, 2001). Moreover, correctional staff trained in TIC can be empowered as they can recognize maladaptive coping strategies and act to refer at-risk women to appropriate care.
One unexpected finding that can guide practice centers on the labeling of mental illness by the female sample. Here, labeling was connected to an umbrella of behaviors, including, though not limited to, reflections of unstable childhood homes, engagement in deviancy and criminal activity, continued dysfunctional interpersonal relationships, and even to perceptions of abusers. While more work is needed, there is evidence that renegotiating the label of mental illness while incarcerated offered some utility toward empowerment. The lives of these women were so complex, traumatic, and disorganized that having access to a mental health staff member in an incarcerated setting provided them a respite to consider the mental illness label and possible treatment options.
Of course, prison can be retraumatizing, though for these women the label of “being mentally ill” developed over time, and often centered on language (rather than tangible events like incarceration). This raises the potential for TIC to address and correct labels, as language training is a core component of this programming. For example, TIC replacing the labeling of clients as sick, resistant, or uncooperative with that of the client being affected by an injury or condition (Substance Abuse and Mental Health Services Administration, 2014). This shifts the conversation from “What is wrong with you?” to “What has happened to you?” which was a key consideration in the methodology of the current study (Klinic Community Health Centre, 2023). These services can also be used to develop cognitive reframing that can be positive, healthy, and prosocial. TIC aims to maximize the individual’s power in personal and interpersonal interactions which would be suitable for the population in this study (Gutierrez et al., 1998).
Labels in treatment, even with good intentions, can lead to negative outcomes such as a negative self-concept (see Willis, 2018). Therefore, TIC removes labels from the treatment while helping the individuals renegotiate the labels that they have already internalized (Bolton et al., 2013; Levenson & Willis, 2019). For example, using person-first language such as “people with mental illness” instead of “the mentally ill” (Willis, 2018, p. 732). This renegotiation of labels is accomplished through treatment focused on dignity, compassion, safety, respect, and trust (Bolton et al., 2013; Levenson & Willis, 2019). TIC has the potential to increase this knowledge of self and other, enhance self-esteem, and develop plans toward achieving personal goals (Miller & Stiver, 1997).
Related to these practical implications is the role of labeling of both mental illness and incarcerated person as a means of establishing a self-concept. These findings suggest that an examination of the labeling process may in fact offer utility as a language tool that can address trauma (and recovery). Trauma-informed services recognize the impact of violence and victimization on development and coping strategies, maximizing choices and control, while aiming to avoid retraumatization (Butler et al., 2011). This requires an assessment of elements that are inherently contained in labeling theory, namely self-esteem, self-concept, and stigma. Moreover, the connection between labels and further engagement in crime and/or changes in mental health status are foundations to TIC. These language strategies can be based on the five principles that guide trauma-informed practice: safety, trustworthiness, choice, collaboration, and empowerment (Fallot & Harris, 2001). This may serve as a conduit to improve treatment for incarcerated women with mental illness, as they are at risk of experiencing repeated, chronic, and multiple traumas that emerge in symptoms of mental illness, substance abuse, and crime/incarceration. Understanding the labeling processes that drive these pathways can facilitate more informed interventions.
Conclusion
The goal of this research was to reexplore modified labeling theory in the context of mental illness occurring in a sample of females incarcerated in two state prisons. These women present complex challenges for prison staff and administration as they have a constellation of risks and needs. The current study highlights commonalities with these women coming to terms with early labels of mental illness, including social stigmatization. This labeling process is linked to acts of criminality, dysfunctional interpersonal relationships, and making meaning of early trauma (including perceptions of abusers). These finding offer utility as modified labeling theory can be applied to better understand key driving factors, such as self-concept, self-esteem, and stigma. This represents a conduit to TIC, improved staff training, and the implementation of trauma-informed services.
Footnotes
Appendix A
Participants’ Information
| Race | Age | Education | Crime | ||||
|---|---|---|---|---|---|---|---|
| Black | 10 | 18–29 | 5 | Less than HS | 4 | Murder | 3 |
| White | 5 | 30–39 | 6 | HS/GED | 7 | Attempted murder | 3 |
| 40–49 | 3 | Some college | 3 | Armed robbery | 2 | ||
| 50–59 | 1 | College degree | 1 | Manslaughter | 3 | ||
| Drugs | 2 | ||||||
| Arson | 1 | ||||||
| Assault/battery | 1 |
Note. HS = High School; GED = General Educational Development.
Appendix B
Life History Calendar
| Life Events | Childhood | Adolescence | Adulthood | Incarceration |
|---|---|---|---|---|
| Family dynamics | ||||
| Substance use | ||||
| Mental health diagnosis/symptoms | ||||
| Accident/illness/death/incarceration | ||||
| Parents/caregivers/authority figures | ||||
| Parental figures | ||||
| Relationship with parents | ||||
| Education | ||||
| School/education achievements | ||||
| School/education problems | ||||
| Interpersonal relationships | ||||
| Marriage | ||||
| Romantic partners | ||||
| Children | ||||
| Friends | ||||
| Trauma | ||||
| Physical trauma | ||||
| Emotional trauma | ||||
| Sexual trauma | ||||
| Mental health | ||||
| Symptoms | ||||
| Diagnosis | ||||
| Treatment | ||||
| Substance abuse | ||||
| Drugs | ||||
| Alcohol | ||||
| Prescription drugs/other | ||||
| Juvenile delinquency/adult criminality | ||||
| Misconduct/criminal offense | ||||
| Punishment/sanction/sentence |
Note. Appendix B is condensed for purposes of publication. When used in practice each category was more fully extended and placed visually on a large sheet of paper that participants could view. This allowed for the holistic development of the life history calendar, with the researcher and participant working in a collaborative manner.
AUTHORS’ NOTE:
The authors declare that they have no relevant or material financial interests that relate to the research described in this article. The authors also have no conflict of interests regarding this research.
