Rekeying Cultural Scripts for Youth Suicide: How Social Networks Facilitate Suicide Diffusion and Suicide Clusters Following Exposure to Suicide

Abstract

Research suggests that suicide can socially diffuse through social relationships and social contexts; however, little is known about the mechanisms that facilitate this diffusion. Using data from an in-depth case study of a cohesive community with an enduring youth suicide problem (N = 118), we examine how, after repeated exposure to suicide, the community’s cultural script for suicide may have been rekeyed such that suicide became a more imaginable option for some community youth. Essentially, we found evidence that a series of sudden, shocking, suicide deaths of high-status youth may have triggered the formation of new locally generalized meanings for suicide that became available, taken-for-granted social facts. The new meanings reinterpreted broadly shared adolescent experiences (exposure to pressure) as a cause of suicide facilitating youth’s ability to imagine suicide as something someone like them could do to escape. We conclude by discussing the implications of our findings for the scientific understanding of (1) suicide and suicide clusters, (2) social diffusion processes, and (3) suicide prevention.

Keywords

suicide adolescents community-based mental health social psychology

Introduction

Since 1999, the youth suicide rate in the United States has increased dramatically. Among youth ages 10 to 14, it has increased 200 percent among girls and 36 percent among boys (Curtin, Warner, and Hedegaard 2016). While the rates for youth ages 15 to 24 have increased less substantially, the increases are still statistically significant, with a 53 percent increase among girls and an 8 percent increase among boys. In addition, for girls ages 15 to 19, the suicide rate is currently at a 40-year high (Centers for Disease Control and Prevention 2017). These alarming trends have increased attention on the need for effective suicide prevention strategies; however, they also demand an increased focus on helping schools and communities cope with suicide loss and the associated vulnerability to suicidality that comes with exposure to the suicide attempt or death of a peer (Abrutyn and Mueller 2014a; Cerel et al. 2016). Indeed, concomitant with these increasing suicide rates among youth is increased documentation of suicide clusters in schools (e.g., Rosin 2015), stimulating enough concern for the Centers for Disease Control and Prevention to deploy epidemiological-aid studies for the first time to communities with active suicide problems (Spies et al. 2014; Garcia-Williams et al. 2016).

Whether suicide truly has the potential to spread—in places or through social ties—has been controversial since Gabriel Tarde (1903) first suggested the possibility. However, there is now substantial evidence from studies leveraging longitudinal data and causal-modeling strategies that suggests this diffusion is not simply due to shared preexisting risk factors between individuals (Baller and Richardson 2009; Abrutyn and Mueller 2014a; Randall, Nickel, and Colman 2015; Fletcher 2017). Instead, the evidence indicates that exposure to suicide really does exacerbate youth’s vulnerability to suicide ideation and attempts. Given this, the next imperative task for researchers is to examine how the spread of suicide works so that we can better understand how to intervene and support youth’s healing after a suicide. To date, we know very little about the mechanisms undergirding suicide diffusion, whether as part of a suicide cluster or through social relationships (Niedzwiedz et al. 2014; Pirkis and Robinson 2014; Arensman and McAuliffe 2015; Mueller and Abrutyn 2015). This lack of understanding is extremely problematic for suicide prevention efforts after a suicide has occurred (which are also called “postvention” efforts).

The most obvious place to start our endeavor to better understand suicide diffusion would be with the large body of literature within social networks research that focuses on how ideas, innovations, and even behaviors flow through social ties (Kadushin 2012). However, one limitation with this research is that it largely focuses on which types of network structures facilitate diffusion and who within a network is best situated to diffuse (or control the diffusion of) new ideas or behaviors. Though considering structural forces is warranted, the diffusion literature is insufficient to explain why something as complex as suicide has the potential to spread through social ties. We argue that a fruitful literature to engage to understand this aspect of diffusion can be found in a diverse array of interrelated literatures including symbolic interactionism and more recent efforts in cultural sociology (Fine 2010; Lizardo and Strand 2010; Abrutyn and Mueller 2014b; Norton 2014). One useful (and robust) insight from this literature is that shared symbolic meanings are at the heart of how humans come to mobilize their behavior, interpret and justify their and others’ behavior, and therefore become capable of adopting novel attitudes or behaviors.

With this study, we develop a theoretical argument for how the diffusion of emotionally fraught behaviors like suicide may work by integrating insights from structural sociology, symbolic interactionism, and cultural sociology. We then follow up our theoretical discussion with an empirical case that illustrates the potential value of our approach. Specifically, our data come from an in-depth qualitative case study of a community (Poplar Grove, United States¹) with a significant and enduring adolescent suicide problem that includes a history of suicide clusters (total N = 118). To preview, briefly, our empirical findings, we find evidence that suggests that suicide is more easily imagined by youth in Poplar Grove because the community collectively constructed a widely shared local cultural script for suicide, which became an available, taken-for-granted social fact. The new suicide script reinterpreted broadly shared adolescent experiences (exposure to pressure) as a cause of suicide, facilitating youth’s ability to imagine suicide as something someone like them could do to escape. Before presenting our theoretical argument and our empirical analysis, we begin by reviewing what is known about suicide diffusion and suicide clusters, the promise network principles and concepts provide, and the reasons both may benefit from insights gleaned from the sociological study of meaning and action.

Suicide Diffusion, Clusters, and Culture

Because the spread of suicide goes by many names (that often track along disciplinary lines), we begin by defining suicide clusters and suicide diffusion before articulating how they overlap. Though there is variation in what scholars have considered a cluster, a recent meta-analysis defined “point” clusters as “suicides or episodes of suicidal behavior localized in both time and geographic space” (Niedzwiedz et al. 2014:569).² While there is some debate about how many suicides are necessary and over what time period to constitute a cluster as well as whether suicide attempts should “count” or not, in a school community, two suicide deaths followed by at least one suicide attempt is generally sufficient to be considered a suicide cluster (Brent et al. 1989; Gould et al. 1990).

If point clusters are temporally and geographically bounded spikes in suicides, then suicide diffusion is a more general social process that occurs when suicide appears to spread through social relationships or through social groups. This process is sometimes referred to as “exposure to suicidal behaviors” (Myfanwy et al. 2017) or suicide contagion (Randall et al. 2015), although we prefer the language suicide diffusion because it is a less alarmist metaphor that still alludes to the interpersonal dynamics of diffusion. Important to note, diffusion can happen even if a discernable cluster does not emerge. In addition, the exposure to suicide that triggers diffusion can be through a personal role model (like a friend or family member’s suicide or suicide attempt) (Mueller and Abrutyn 2015) or through a member of a relatively bounded social milieu (such as when a peer group, school, or small community member dies by suicide) (Haw et al. 2013).

Though the literatures and methodologies used to garner insights into the consequences of exposure to suicide through each of these sources are often disparate, our point is that at their core, there is a similar general process: namely, the spread of suicide through social ties. The sometimes-contradictory findings in each of these literatures also suggest that diffusion is a complex process—or possibly processes—that requires far more attention to the mechanisms undergirding it than it has received to date. We are not the first to leverage this critique. Recent reviews of the literature across disciplines have repeatedly concluded that methodological challenges and data limitations have restricted our ability to understand and prevent diffusion or facilitate healing after a suicide in schools and communities (Haw et al. 2013; Pirkis, Robinson, and O’Connor 2016; Myfanwy et al. 2017). To that critique, we would add that the literature has been quite limited in its theoretical imagination of which mechanisms may undergird the diffusion of suicide.

Theoretical Dilemmas and Promises

Currently, a very limited set of theoretical insights into human behavior have been leveraged to understand suicide diffusion. When Haw and her colleagues (2013:101–104; also Pirkis et al. 2016) reviewed the extant literature, they concluded that most explanations for clustering focused on intrapersonal psychological mechanisms, many of which rely explicitly or implicitly on the idea of social learning. Social learning theory argues that through observation of others, we learn both behaviors and their real or perceived costs and benefits (Akers and Jensen 2006). This theory suggests that youth might choose suicide because they believe it will produce the benefits they attributed to it previously. Oftentimes, this theory is combined with differential association or identification to emphasize the importance of the exposed person’s being able to identify with the role model such that the behavior is more readily accepted as good (Stack 1990; Baller and Richardson 2009). Despite their allusion to the broader sociocultural environment, social learning theories largely take for granted how the broader structural and—important to note—cultural milieu facilitates or constrains social learning. This is particularly problematic for phenomena like suicide clusters and diffusion, which invoke interpersonal structural-cultural dynamics as much as intrapersonal dynamics. In other words, we need to understand not just why an individual might be suicidal after exposure to suicide but also why certain social groups or social relations facilitate that state more readily than others.

If we turn, instead, to the larger social networks literature on diffusion processes in an attempt to understand interpersonal mechanisms of diffusion, structural explanations dominate the research landscape (Kadushin 2012). For example, research has noted that diffusion processes are augmented by social cohesion, as dense social ties amplify the likelihood that any given actor in a network or social context is exposed to the behavior or idea (Friedkin 2004). Past research also has shown that new ideas or opportunities often diffuse through weak ties (e.g., the friends of our friends) rather than strong or direct ties to people we know (Granovetter 1973). In particular, individuals who bridge different groups of socially cohesive individuals can be important proponents of social diffusion (Burt 2004), though some research has shown that for the diffusion of complex social norms exposure through multiple relationships in a social network is required for diffusion to occur (Centola and Macy 2007). Finally, research within this paradigm often notes that high social status role models and role models who are similar to ourselves can be powerful forces for diffusion, as individuals have social-psychological motivations to conform to admired or similar role models (Marsden and Friedkin 1993). While these insights are useful for understanding structural aspects of diffusion, they are inadequate for making sense of the diffusion of something as emotionally fraught or complex as suicide. They go too far in the opposite direction of the largely psychological and epidemiological literature on suicide diffusion and neglect intraindividual motivations and circumstances. We are not the first to leverage critiques of the existing network literatures, particularly those focused on mental health. Bernice Pescosolido (1994, 2006) has argued convincingly that to understand the complex role of social ties in mental health, the context of networks matters as much as the network’s structural characteristics. Considering the cultural scripts and social norms embedded in networks may help us understand why some people may acquire and internalize the idea of suicide as something they themselves could do after exposure in a particular social context.

Of Meaning and Action

We argue that delving into the cultural meanings surrounding suicide that are negotiated and potentially changed after exposure to suicide is a fruitful direction to pursue. Indeed, while social network theories readily acknowledge that culture is embedded in networks (Pescosolido 2006; Kadushin 2012), they rarely consider how the spread of cultural meanings may facilitate the adoption of behaviors encountered through network ties (Emirbayer and Goodwin 1994; Fuhse 2009). This omission is problematic, especially for understanding suicide diffusion, as a growing body of ethnographic research on suicide reveals that different cultural milieus provide very different meanings about who dies by suicide, why, when, how, and where (cf. Chua 2014; Stevenson 2014). For instance, in Japan, suicide is understood as a rational means of taking responsibility for one’s actions (Kitanaka 2012), which contrasts sharply with the more general view in the United States that suicide is caused by mental illness (Marsh 2010). Even further, a cluster of clinical-psychological studies of bereavement has demonstrated that suicide deaths are followed by an intense and almost obligatory process of meaning making (Currier et al. 2015) wherein individuals try to make sense of both their loss and of suicide itself (Neimeyer, Baldwin, and Gillies 2006).

From these literatures, we take two essential points. First, there are meanings about suicide embedded within cultures that are often broadly known and taken for granted. These meanings clarify why people die by suicide; are very often linked to behavioral repertoires of how one should die by suicide; and ultimately allow the act to be a meaningful performance for the suicidal individual, her intended audience, and even unintended audiences (Stack and Abrutyn 2015; Mueller 2017). Important to note, these meanings can vary across macro-level cultures. Second, though these macro- or societal-level meanings attached to suicide may be known, they are not fixed but rather negotiated by individuals and sometimes by social groups. In particular, we posit that exposure to suicide may shape or even transform the cultural meanings we assign to suicide (Abrutyn and Mueller 2018).

The core of our argument is that meanings are learned and internalized in interaction and employed in the anticipation (or planning) and realization of action through cultural scripts (Canetto 1997; Goddard and Wierzbicka 2004; Harding 2007). Although myriad metaphors for mechanisms linking culture to behavior abound (Wood et al. 2018), we prefer scripts as it reflects both the private (e.g., the idiosyncratic “style” afforded individual interpretation of a role) and the public nature of culture (e.g., the performative expression of intrapersonal meaning structures like schema or frames always follows some well-worn “conventions”) (Patterson 2014; Lizardo 2017). In addition, a literal script provides both the substance for acting (e.g., lines, behavioral repertoires) and key background information (e.g., motivation of actors, cues, subtextual meaning). As such, cultural scripts serve as both guidelines for symbolic performances meant to externally represent meanings that actors believe are shared with their audiences, even if their intended (and, of course, unintended) audience may misinterpret or misunderstand the meanings, and as repositories of central narratives about the performances (why, who, when, where, and so forth). Hence, scripts are the mechanism by which real, imagined, and mythic narrative models of behavior overlap with actual performance (Simon and Gagnon 2003).

Because cultural scripts are replete with cultural meanings that guide social behavior, the central argument posited herein is that exposure to suicide may trigger a rekeying of the cultural scripts surrounding suicide that, in turn, may render suicide a more accessible and applicable option for youth exposed to the suicide and the new script. Thus, diffusion is the process by which behaviors are spread not merely through exposure or contact but also through the acquisition of a role model’s cultural script through social interaction directly with the role model and/or with others who were exposed to the role model. Clusters, then, are not just the escalation of dyadic diffusion but may instead emerge from collective processes by which a community comes to make sense of the initial (and ensuing) suicides, and in so doing, they rekey cultural scripts for suicide that expand for whom suicide is an option.

We argue that this process is most likely to happen (1) when a suicide death contradicts existing shared cultural understandings of suicide and/or (2) when the suicide role model’s actions feel salient or affect a large number of people because the decedent was prominent or admired, was someone many could easily identify with (a “similar other”), or was a member of a cohesive group. Furthermore, the process is most likely to create a new shared set of meanings (3) when the new suicide script is objectively or subjectively generated collectively and (4) comports with a more general stock of knowledge, previously unapplied to understanding suicide, but easy to accept as logical. Under these circumstances, the new suicide script can become institutionalized in what sociologists refer to as idioculture (Fine 2010) or group style (Eliasoph and Lichterman 2003), and through this process become a taken-for-granted social fact. Thus, suicide scripts come to embody an idea of both “who” dies by suicide combined with the meanings (“why” do people die by suicide) and, in the case of localized scripts, or scripts shared by relatively small collectives like small towns, schools, families, or peer groups, how, when, and “where” suicide is “appropriate.” The local community (or school, or family) is important as this process depends on collective sharing and sense making of the suicide loss for the new suicide script to diffuse, and with it vulnerability.

In sum, then, our argument is that people generally have a sense of what suicide means and why someone might die by suicide, but we also argue that this understanding is not static. Instead, it is a product of their own experiences and of the general and local scripts that are readily available, accessible, and applicable, particularly those scripts that are mobilized or created through shared experience of suicide. We now turn to our empirical case study to illustrate the efficacy of our theoretical argument.

Methods

Study Location

The data for this study come from an in-depth case study of Poplar Grove, United States (a pseudonym). Poplar Grove is a small (< 50,000), suburban, majority white (> 90 percent), wealthy (> 90 percent home ownership), socially cohesive community with an enduring adolescent suicide problem—including repeating suicide clusters (for more information, see Mueller and Abrutyn 2016). The suicide problem in Poplar Grove is centered on the one academically excellent public high school in the community. Poplar Grove High School (PGHS) serves approximately 2,000 students at any one time. Since 2005, PGHS has lost (at least) 16 current or recent graduates to suicide. This is significantly higher than what one would expect for a high school this size given national suicide rates for ages 15 to 24 (which is 11.0 per 100,000 youth). A more typical rate of suicide for a high school of 2,000 students would involve one suicide death approximately every 4.5 years. In addition, Poplar Grove and PGHS have experienced at least three suicide clusters wherein we were able to confirm the ties between multiple decedents. Community mental health workers reported that nearly every suicide of a current student is followed by the suicide attempts and serious suicide ideation of multiple peers. Finally, three suicide deaths of community youth occurred during our fieldwork.

Data

For this study, we largely draw on data from semistructured in-depth interviews and focus groups in Poplar Grove (N = 98),³ though the larger project also involved participant observation and a media analysis. Our research was facilitated by community leaders who invited us to conduct research in the community, helped recruit respondents, introduced us to other community leaders, and provided private office space. Recruitment for respondents was largely through community organizations including a mental health counseling center, two religious organizations, and the government-sponsored Suicide Prevention Committee (pseudonym). We also set up tables with flyers at community events where we could informally discuss the research with attendees, posted flyers around the community, and relied on word of mouth. Because of the intense emotions surrounding suicide, we did not contact any respondents directly unless they had given explicit permission for us to do so via a third party.

The first stage of our data collection began with focus groups with youth, parents, mental health workers, and young adults who grew up in the community. In the focus groups, we asked very broad questions about what life was like in the community and about perceptions of suicide in the community. Most groups lasted about two hours.

The second stage of data collection involved conducting in-depth interviews with community members who were bereaved by suicide, including youth and young adults who lost friends or classmates to suicide, parents who lost their children to suicide, and parents whose children lost close friends to suicide. These interviews were centered on the experience of suicide bereavement and how people coped with and explained suicide to themselves and others. In addition, we conducted interviews with teachers, counselors, therapists, doctors, nurses, pastors, suicide prevention activists, public health officials, and crisis responders to get a broad perspective on suicide in the community. During this second stage, we also continued to conduct focus groups with adolescents or young adults who indicated that they felt more comfortable speaking with us surrounded by their friends.

Most interviews took place in person, in the community, though some took place by phone or Skype. We allowed respondents to pick the location of the interview. Many interviews took place in a private office located within a small counseling center that a community leader provided. Interviews lasted between 45 minutes and 4 hours, with most lasting approximately 2 hours. Short interviews were usually with professionals who had limited time. When possible, we triangulated data to get multiple perspectives on the same events or relationships—for example, we interviewed 12 pairs of parents and children. Because of our interest in how individuals made sense of suicide we asked a series of questions designed to get at this issue. We started broadly by asking people to simply tell us about their experience with exposure to suicide. Many people voluntarily commented on the person’s motives at this stage. Toward the end of the interviews, we also asked them to discuss (1) how their views about suicide changed after their experience with suicide, (2) whether suicide was a choice, and (3) whether suicide was ever justified (these last two questions emerged as salient from our early interviews). In focus groups, we asked respondents’ perceptions of the main reasons that youth complete suicide in Poplar Grove. Finally, we include data from one 3,600-word journal written by a suicide decedent during the 12 days before her death (which is used with permission of her parents).

Finally, to test our theory that the collective experience in the community facilitates the view of suicide as an option, we also conducted in-depth interviews following the same protocol with a reference group of young adults (ages 18–32) who lost a loved one to suicide but lived outside of Poplar Grove (N = 20). All but two of our respondents had substantially more individual experiences with suicide; their beliefs about their loss were developed either in isolation or in conversation with small, intimate groups like their families. Two reference group respondents did lose someone to suicide who was a member of a bounded social group (in one, a dorm and in another, a high school) where the group then engaged in extensive conversation about the suicide. This reference group helped us identify what was unique about Poplar Grove.

Descriptive statistics about our respondents are presented in Table 1. In Poplar Grove, 14.3 percent of our respondents are youth younger than 18; 20.4 percent are young adults ages 18 and 25, and 6.1 percent are older young adults ages 26 to 32. In addition, we interviewed community parents (40.8 percent of our sample) and mental health workers, broadly defined (18.4 percent of our sample). Of our Poplar Grove respondents, 99 percent are non-Hispanic white (which reflects the fact that the community is majority white). Our reference group sample is more diverse: 75 percent of reference group respondents are white. Both our reference group and our Poplar Grove samples were majority female (at 80 percent and 77 percent, respectively), which may reflect gender difference in comfort in speaking about suicide or participation in qualitative research.

Table 1

Descriptive Statistics.

	%	N
Poplar Grove
Youth (ages 15–17)	14.3	14
Young adults (ages 18–25)	20.4	20
Older young adults (ages 26–32)	6.1	6
Parents	40.8	40
Mental health workers	18.4	18
Female	76.5	75
Non-Hispanic white	99.0	97
Subtotal n	98
Reference group
Young adults (ages 18–25)	65.0	13
Older young adults (ages 26–32)	35.0	7
Female	80.0	16
Non-Hispanic white	75.0	15
Subtotal n	20
Total N	118

Data Analysis

Interviews and focus groups were digitally recorded and transcribed by professional transcribers. Transcripts were reviewed by the authors for accuracy and then analyzed for themes in NVivo 11 software. Themes were found through abductive reasoning, which emphasizes identifying “surprising findings” that emerge from the data (Timmermans and Tavory 2012). For this study, the surprising finding that emerged from fieldwork was the difference in the centrality of mental illness to making sense of suicide between our reference group and our respondents from Poplar Grove. As we developed our analysis, the first two authors read all the transcripts, and the second author conducted a detailed coding of the transcripts to ensure unexpected themes could emerge from the interviews. From this detailed coding, we established our major themes and then progressed to “focused” coding.

As a final step, we grouped respondents into categories about the dominant themes within their narratives of suicide that we will discuss in our Results section. Two authors categorized every respondent independently then compared the categorizations for consistency. We flagged two discrepant cases for discussion and, after a brief discussion, reached agreement. Finally, to protect the privacy of our respondents and the community, all names of people, organizations, and places have been changed, and any identifying details, including dates, have been modified (though we aim to keep modifications as minimal as possible). This research received human subjects’ approval from our universities’ institutional review boards.

Results

Our goal with this study is to examine the mechanisms undergirding suicide diffusion by exploring how exposure to suicide shapes a community’s scripts about suicide and how that process contributes to the emergence and perpetuation of suicide. We begin by exploring the scripts people draw from to make sense of suicide in our reference group respondents before moving on to examine the meanings attributed to suicide in Poplar Grove. We then examine why Poplar Grove may have developed a unique local understanding of suicide through a collective process and the potential consequences this local script has had for suicide risk in the community.

Suicide Scripts in Our Reference Group

In the United States, the idea that mental illness causes suicide is broadly accepted by medical professionals (Marsh 2010) and the public (Lake et al. 2013). For example, the following statistic is frequently repeated in academic and lay books about suicide: “At least 90 percent of all people who died by suicide were suffering from a mental illness at the time, most often depression” (Jamison 2001). Indeed, this idea is so broadly endorsed that it is often taught explicitly as a part of suicide prevention materials in the United States (White and Morris 2010).

It is not surprising, then, that among our reference group respondents, most had a strong understanding that suicide is caused by mental illness. Specifically, 90 percent of respondents (18 out of 20) presented mental illness as a primary cause of suicide. One example of this comes from Susan, an older young adult who lost her brother, Jeremiah, to suicide:

I completely understand that Jeremiah was sick, and not to justify [his suicide], but I get the complexity of depression. … The things that happened in his life, he dealt with differently than myself. I understand that people can only process certain things in their life in a certain way, and at some point he lost control of being able to understand how to function, so his depression took over and he felt like he didn’t have an out. From what I gather, in the way that the brain works, as far as once you get to a point of deciding that you’re going to take your own life. … So I get it. I can sit here and consciously talk about why he did what he did and identify that was my brother. (emphasis added)

Susan’s understanding of Jeremiah’s death as caused by mental illness appears in her comments on depression and in using the language “sick.” It is interesting to note that Susan recognizes mental illness as something biological and “in the brain,” which implies that suicide’s roots are somewhat immutable. Though not everyone in our reference group did this as explicitly as Susan does, many did (see Table 2 for additional examples). In addition, even when social factors were identified as mattering to suicide, our reference group respondents often assigned primacy to mental illness. For example, Katie, a young adult who lost her sister shared the following:

The reason my sister [Jenny] committed suicide … [was that she] was having hallucinations and stuff, and she didn’t ever tell anybody [including her doctor], but I read one of her journal entries, and she would talk about how she’s seen and heard [her father who had recently died]. … She told her teammates about it, and they became anti-Jenny. They were like, “I’m not gonna hang out with you anymore ’cause that’s weird.” Those were the only people she told, and they turned their back on her. I understand they don’t know what to do, but they just literally left her. (emphasis added)

Table 2.

Additional Examples of Mental Illness Scripts for Suicide among Reference Group (Non–Poplar Grove) (n = 20).

Respondent	Ethnicity	Mental Illness Scripts
Young adult (18–25 years old)
Andrea	Latina	“I didn’t expect [my friend’s suicide]; I never thought that he was like terribly like depressed or what have you.”“I’m like, ‘no, I need help. And I need it now.’ Um, and my like depression was really like falling, you know. And it was scary. … I didn’t really have a plan [to suicide]. But it was becoming a thing to where I was just like having [suicidal] ideas.”
Jasmine	White	“I think my grandfather had a lot of mental health and depression issues, but I don’t see that—the suicide—being a mental health issue in that moment because it was time to end [his] suffering [from a physical illness]. Which is different from, I think, like … a young person, or suicide clearly when there’s depression.” And through my therapy, … I’ve realized that I was probably depressed for a long time … [and that’s why I was having suicidal thoughts].”
Older young adult (26–32 years old)
Landry	White	“Depression just oozed out of [my dad].”“Something had misfired or shorted out or something [in his brain], because that is not how it works. You do not commit suicide to see your creator in the world I grew up in.”
Finn	White	“She’s got a diagnosis of borderline personality disorder. … Pretty much everything that you think of with borderline [personality disorder] she was already doing, like eating disorder, self-injury, chaotic relationships. … Ugh. No, I was totally not surprised that [suicide] would happen.”
Starla	White	“He was like textbook bipolar.”
Oliver	White	“My first experience [with suicide] was me. I was 19, and I didn’t know it then, but I was starting to become bipolar … and I was in a mania. And I was dating a guy, and he broke it off. And I was like, ‘I can’t stand this,’ [so I attempted suicide].”“Those, again, who have tried [suicide] or know people who have done it. … They don’t take for granted the fact that they have a healthy brain.”

Note. Emphasis added.

Katie feels that Jenny’s friends’ responses amplified Jenny’s risk of suicide, but Katie still emphasizes mental illness by identifying Jenny’s hallucinations as “the reason” her sister completed suicide. In short, the meanings our reference group easily imposed on their experiences is congruent with previous research on how Americans typically “explain” suicide—for example, mental illness or disorder.

Local Suicide Scripts in Poplar Grove

In Poplar Grove, the societal narrative surrounding mental illness and suicidality was present and important. Specifically, 80 percent of our adolescent and young adult respondents (n = 29/38) mentioned mental health or psychological pain as mattering to suicide in the community. For example, Leah, a youth who lost a close friend (Stella Blue) to suicide, attributed suicide both generally and specifically (in Stella’s case) to depression:

It takes a certain type of person to get to that point [suicide], and a certain type of depression that they’re experiencing. It’s not like their personality wants to [suicide]. It’s not like the friend in them or the daughter in them or the father in them wants to [suicide]. It’s the chemicals in their mind, and it takes a certain type of person to be pushed to the edge, a certain kind of impulsive person, or some other quality that makes them more adept to acting on a rash term. But it’s not the person that you see in them. It’s the disease I think. (emphasis added)

Leah’s interview also reveals how this more general narrative is refracted through a local script for teen suicide. At the end of the interview, when asked what was the most important thing for us to understand about her experience with suicide in Poplar Grove, she emphasized the intense local pressure:

Honestly, I just say the most emphasis should be put on the pressure that people feel around here. And even when I try to explain [the pain of the pressure] to my parents, it’s hard for them to understand, and it’s really hard to explain. … They just don’t really realize that it kind of gets old, and it’s a lot of pressure. … Kids … can be really hard on themselves, especially when they’ve been kind of raised thinking that they’re not good enough. … [The parents] do it out of love, but it’s just … a lot, ’cause even when … you know your parents will encourage you, no matter what you do in life, when you go walking down the street or you go eat dinner at your friend’s house, you’re worrying about their parents and what they think of you. It’s just kind of always there.

Thus, for Leah, mental illness is rooted both in biology (“chemicals in [the] mind”) and in social experiences (“the pressure … around here”). She also attributed Stella Blue’s death in part to Stella’s “[wearing] herself ragged trying to please everybody and trying to do the things that she thought she had to do.” It is essential to note that Leah’s self-reference to pressure in the quote above demonstrates that she suffers under the pressure, just like Stella.

Leah was far from alone in her assessment of how central social pressure was to understanding depression and, thereby, suicide in Poplar Grove. Beth, a young adult who lost her friend Michelle to suicide, attributed Michelle’s suicide partially to depression and partially to “culture” and “pressure”:

Anna:

So why do you think that Michelle died by suicide? And … I’m asking more because I’m interested in your perception, not in the capital T “Truth”… if that makes any sense.

Beth:

Well, I don’t know. The combination of a lot of things … and I mean, I know she had … depression. I think, I know for sure that she had it for a couple of years. So I think that … I don’t know, it could be genetic, biological … but also … kind of like a culture, like socialization stuff. So I guess that’s what put her in that suicidal mindset. That’s all I know.

Anna:

Fair enough. Do you have any thoughts on what caused her depression? Other than the biological or is that what you attribute it to?

Beth:

I think just the pressure, I guess. … I think that she was under a lot of pressure, and I think that seeing that externally, around her. … I think that internally, she didn’t have the capability to deal with it. … I think it’s understandable when you are … constantly surrounded by people who are achieving certain things that you feel like you have to be like them.

The “youth under pressure” (to meet social expectations) narrative was broadly recognized in the community as playing a role in the local suicide problem (see Table 3 for additional examples). Specifically, 87 percent (n = 33/38) of young community members (ages 15–32) referenced pressure as a major cause of local youth suicide, suggesting pressure as salient to understanding suicide in the community as mental illness. Indeed, eight young respondents did not mention mental illness at all and only mentioned the pressure to meet expectations as the cause of suicide (compared to five youth who did not mention pressure).

Table 3.

Additional Examples of Scripts about the Causes of Suicide among Poplar Grove Youth and Young Adult Respondents (n = 38).

	Mental Illness Script	Pressure Script
Youth (15–17 years old)
Hannah	“She was about to go back into treatment, I think. And she just wanted to get her finals over with, is what somebody told me. She wanted to get her finals in. She just wanted to take care of her grades before she went back into treatment, and she just couldn’t make … get that.”	“I know there is a lot of pressure to like, fit in … school is a big thing, and sports is a big thing. But also just like fitting in like having straight, long hair, and like skinny jeans, and UGGs or whatever is in style that particular season. I know I’ve heard guys it’s hard too, but for girls there’s a lot of stuff. I don’t think I have a low self-esteem, but even though in the back of my mind, I know I don’t have to fit those standards, it’s hard to like, push it away. So I think that kind of makes kids feel like they’re not necessarily fitting in if they don’t look the right way.”
Scott	None	“I think people really don’t see the light at the end of the tunnel. That this is a small portion that I know. … I’ve heard it said that mediocre here is extraordinary everywhere else. Which, I’m going to assume is true … , but I think the mindset here is weird. … You have to play a sport, you have to have at least four or five APs, and we are pressured into this, which only adds more pressure and some people can’t handle the pressure. … I get this feeling that a lot of the suicides are because … after three or four years of the pressure, they just couldn’t take it. … It makes me feel better about us if I think we’re the only place where people are so stressed that they commit suicide. Because there is clearly something wrong if it’s like this everywhere.”
Young adult (18–25 years old)
Vanessa	“I mean like, a couple of my friends last semester had depression, and they got medication and didn’t say anything. And I was just like, they eventually did, but now I like worry about them constantly. And they’re like, I’m better now, but I still like, unconsciously worried about them [dying by suicide] all the time. I’m like, are you really? Are you guys just saying that? Like, I don’t know.”	“I think [suicide] has a lot to do with just the pressure here. That’s all ever—people ever talk about here, is just how much pressure there is, like academics and sports, and I just think they just can’t take it anymore.”
Chloe	“I mean, like I said it first, it was like, ‘Why wasn’t it me?’ Like, I had all these fucked-up issues [including a diagnosed mental illness]. … Like, [my friend Kennedy who died by suicide] had the entire future for her. … Like, it should’ve been me [who died]. … And actually after she did commit suicide, I stopped taking all of my antidepressants. I was 18 at the time. I went to my psychiatrist. I told him, ‘I’m done.’”“I know she was on Paxil when she did do it. I think she was too smart, personally. I don’t think she knew how to cope. … She was going through a lot of mental things that she didn’t know or want to communicate with people. And then you give someone a drug to cover them up and repress it. It’s just going to build up, you know?”	“I’m blessed that I got to experience her adolescent/childhood side. You know? Before the pressures really got to us. Because she was such an awesome person from the get-go, really. And honestly I don’t know if she, you know, if she. … It was all the pressures, or if she really just wanted to be with her boyfriend [who died by suicide before her]. You know? It could totally be a Romeo and Juliet syndrome.”“Definitely a social aspect to this area, I think, with the pressure and the number of suicides. This is an abnormal amount of suicides in this area.”
Molly	“I’m not sure if she ever said the word ‘depression.’ Maybe I heard that from [our mutual friend] or something. I just knew that she struggled with herself and her body image and being—I use the word perfect because I feel like that is an accurate term.”	“I just feel like she never—well, I feel like she never felt like she could be perfect in her own eyes, and she had explained this to me just a week and a half before she died. She was telling me, ‘No matter how many people tell me that I’m skinny or pretty or whatever, it doesn’t matter because it matters what I think.’ I feel like she didn’t think that things were going to get better and that she was ever going to be happy with herself. Even though on the outside she’s like super athletic, varsity athlete, stellar grades … and pretty and skinny and all of those things, but to her, she wasn’t who she wanted to be. Also, I guess with her personality, she felt like she wasn’t doing everything right, and she told me all these things, and obviously, I tried to talk to her about it, but I didn’t know how serious that really was. That was … not very long before [her suicide]. The way I’ve always thought about it is that she wanted to be perfect and needed to see herself as perfect, but didn’t and couldn’t, and no one could convince her that she was perfect except herself. She couldn’t do that.”
Sloane	“Brian had battled with depression for a long time. And then my other friend, his name was Trevor, … he had battled with depression. … He had also battled with addiction.”“He’s like manic depressive, and he’ll just go crazy, so it was more for that than anything else, but I know he was not dealing with that well.”	“I think that’s a huge issue around here, and I think that’s what a lot of people struggle with—pressure to succeed and then drugs. That’s a big thing around here because it’s easy to get and everybody has money. … I just think there’s so many expectations of people, especially here, you know. … Like, everything around here is competitive—everything. And it’s horrible. …I just think there’s so much pressure on kids to be … to grow up and buy their own [Poplar Grove] house, and maybe they don’t want to do that.”“I think his last year of high school, he felt a lot of pressure to conform and be a certain way, … and, you know, ‘Everyone has to like me,’ and I was like, ‘No. Like, just do your own thing. Surround yourself with people that have your best intentions, and you will feel a lot better.’ And I think that was a huge struggle for him. And I think with Brian, a huge struggle for him was to fit in, because he was … I mean, he was like laid-back, jokester, like he wasn’t really preppy or anything. And I think he felt a lot of pressure to be like …‘I have to find a sport I’m good at. I have to do this. I had to make everybody happy.’ And I think he just couldn’t handle it.”

Note. Emphasis added.

Perhaps most interesting, this explanation for suicide appears to have become locally diffuse, such that youth with less direct experience with suicide (e.g., losing a classmate versus a friend to suicide) framed local suicide as resulting from the intense pressure Poplar Grove imposed on youth. Lily, a youth in Poplar Grove who lost a classmate, illustrates this pattern after being asked in a focus group why youth die by suicide:

From the beginning of middle school, we start thinking about a career path, and I think that’s really hard … especially for me, personally. I have no idea what I want to do when I grow up, and from that time, most of your classes and teachers are thinking, “Oh, what are you going to be? What are you going to do? How are you going to make any money? How are you going to support your family? You are going to have a family, right? Blah, blah, blah, blah, blah.” I feel like a lot of kids feel like they don’t know where they’re going, so they’re not going anywhere. And then, they realize that there is so much pressure around them that maybe they’ll never get out of the pressure, or the town, and they’ll never get a real job, or a real life outside of their teenage nightmare.

It is interesting to note that Lily, like Leah and Beth, presented a clear notion that pressure matters to suicide, but she never directly referenced mental illness or even emotional distress during her interview. Instead, it was implied, with pressure foregrounded in her script.

This emphasis on pressure was substantially different from the emphasis in our reference group, where respondents more frequently saw mental illness as rooted in biological factors or as a fixed, diagnosable trait. In addition, in our reference group, when mental illness was not immediately visible as a cause of suicide, the suicide itself was generally taken as evidence of mental illness. For example, Tiffany (young adult, reference group), in trying to make sense of her friend’s suicide, shared, “I don’t know his mental makeup. I don’t know if he had anxiety and depression,” implying that where there is suicide, there is also a problem with a person’s “mental makeup.” This was rarely the case in Poplar Grove, where pressure seemed sufficient to understanding suicide.

The Limits of Mental Illness

The emergent centrality of “pressure” and the growing inadequacy of “mental illness” to understanding suicide in Poplar Grove is not entirely surprising. First, many suicide decedents in Poplar Grove did not appear publicly “mentally ill” or even emotionally distressed prior to their deaths. Quite a few decedents were, in fact, broadly perceived as “bubbly,” gregarious, popular types who seemingly “had it all.” Subsequently, their suicides triggered widespread confusion and a compelling need to make sense of why they died by suicide. For example, Michelle’s death was particularly confusing for many of her peers, like Shannon (a youth who was a few grades behind Michelle in school): “From the outside [Michelle] was this picture-perfect girl—the sports star, and super smart, and definitely going places, and dating this really awesome guy. And [her suicide] just seemed a little bit—[it] was confusing; there’s not really another word to use for it.”

Second, there were social incentives in Poplar Grove to come up with an explanation for suicide that did not involve mental illness. Mental health stigma was prevalent in the community (Mueller and Abrutyn 2016); thus, there was a desire to preserve the reputation of youth who had died by suicide (and their families) by not mentioning anything “bad” about them, including mental illness. As Lily (a youth) noted bluntly, “no one trash talks the dead kid.”⁴

Finally, because of the repeated nature of suicide in the community, it was insufficient to simply explain the suicide of an individual youth; the community also needed to understand “what’s going on here, in Poplar Grove.” Indeed, mental illness—at least if perceived as a biologically driven phenomenon—works less well to understand why a place is plagued by suicide. Thus, it was understandable that the community turned to social explanations for why suicide was such a problem.

The Role of the Collective

Though the evidence we present above draws on individual youth’s scripts, these meaning-making projects in Poplar Grove were far from individual. The fact (1) that the community was very socially cohesive (Mueller and Abrutyn 2016); (2) that many of the decedents were highly visible, well-known youth; and (3) that many of the deaths occurred in public places (sometimes with witnesses) meant that the community, as a whole, was deeply affected by the majority of the deaths. Indeed, the drive to understand suicide in Poplar Grove was because the community was subject to a (recurring) collective trauma—specific in its content (suicide) but generic in its form and impact (cf. Erikson 1978; Alexander 2004). Collective trauma is unique insofar as it is does not simply challenge a given individual’s reality, but also challenges the very meanings that provide the hooks upon which group members craft their social identity. Because of this, shocks, like collective traumas, are known to provide opportunities for culture creation (Fine 2010). And with each recurring suicide and foment of a new cluster, these newly available local narratives became more and more accessible as subsequent suicides continued to fit the new script, objectively or intersubjectively.

Thus, it is not surprising that we see the emergence and spread of an ostensibly “new” shared suicide script in which “perfect” youth also could be understood as youth who are vulnerable to suicide and where a need to escape social pressures and stress became a plausible motive for suicide. In this way, our Poplar Grove respondents’ experiences were quite different from those of our reference group respondents, who often mourned more individually or at most within a family group or close circle of friends.

Though we lack the space to delve too far into the role of the collective in rekeying suicide scripts, it is worth mentioning a clear way that this happened: People in positions of authority in Poplar Grove tended to reaffirm the emergent “pressure” narrative that students were using, thus facilitating the rekeying of local suicide scripts. For instance, the local newspapers disproportionately published high-profile articles featuring “experts” who testified to the toxicity of pressure and buried articles that focused on mental illness (Mueller 2017). In addition, other community adults—like teachers—further reified pressure as a valid cause of suicide. “I have this one teacher,” Natalie, a Poplar Grove youth, shared,

and he’s always like, “You know what? I get it, you all are [Poplar Grovians]. You are being pushed so hard, and if my class ever stresses you out that much, just let me know because I don’t need you freaking out and losing it, because my class added enough stress to your lives.”

This reassurance meant all the more to Natalie because, as she put it, “he’s one of the teachers of the girl that killed herself.” Indeed, many youth reported that empathic adults like Natalie’s teacher were quite comforting because they recognized the daily struggle of life in “the Grove.” Krista, a young adult Poplar Grovian, shared the following about after her friend Michelle’s suicide:

I remember talking about the pressure of [Poplar Grove] [with adults—not at school], and I think it was helpful because then you understand that like, other people feel this way too. And that you aren’t alone feeling that way; it’s okay you don’t have to be perfect. … So it was helpful.

While these conversations reflected a real source of distress for youth, they also affirmed youth’s understanding of pressure as a “risk factor” for suicide.

Finally, it is worth noting that while the localized suicide script was being reinforced by authoritative community members, the school as a whole had a policy of not talking about suicide with students out of fear of generating “a ripple effect.” This was according to parents, local public health officials, people who work for the school, and our own observations of postvention strategies following three deaths during fieldwork. Hence, no suicide prevention programs that might formally teach youth about risk factors for suicide (which often include information about mental illness) were never systematically offered to youth and never were offered at school.

The Consequences for Suicide Risk

Because exposure renders youth more vulnerable to suicidality, we now examine how youth reacted to the local script for suicide and present our evidence for the script’s contribution to vulnerability to suicide. Of course, due to the rare nature of suicide, we are largely limited to examining factors that may render a youth probabilistically more vulnerable to suicide or suicidality (Roth and Mehta 2002), like attitudes toward suicide (Gould et al. 2004; Lake et al. 2013; Pitman et al. 2017), rather than suicide itself, or to evaluating post hoc evidence of a decedent’s motives. Indeed, the local belief in Poplar Grove that pressure “causes” suicide is concerning because this suicide narrative invokes a broadly shared experience (pressure) that plays a role in why many youth in the community are miserable (Mueller and Abrutyn 2016). Despite these data challenges, we did indeed find some evidence that youth could turn the perceived motives for the suicides of others embedded within local scripts into something applicable to themselves. This was particularly true among youth who were suffering under the local pressure and who wished they could do a better job living up to expectations. That is, some youth appeared more vulnerable to linking their own personal experiences with pressure and distress to the broader scripts as justification for or explanation of their own suicidality. Our first example of this comes from Becca (a youth), who perceives suicide as caused by pressure and who has lost several classmates to suicide.

Anna:

Do you always worry that one of [your friends] is going to kill themselves?

Becca:

Every day. Every day I walk in the school, and I’m like where’s Madison, where’s [their mutual friend] Nicole? …

Anna:

I’m curious. … Why suicide? Why is that so present in your mind?

Becca:

For me it’s because I had my parents yelling at me every day, every, every, every day because my grades weren’t good enough to live up to my sister, Cindy. She took almost every AP exam and got a five on it. … If I don’t do all of that I’m screwed because my parents are going to kill me. … [She sighs deeply.] … I definitely try to work on my problems and try to make myself feel better about myself. I’ve wanted to get away from having all these problems. Seeing all these other people go through all these problems, their answer is suicide, so why can’t my answer be suicide? (emphasis added)

Becca’s quote illustrates how easy it is for some teens in Poplar Grove to see suicide as a possible solution to their problems. Her comments also reveal the dualistic character of the local script Poplar Grove has adopted for understanding teen suicide: While Becca is eager to reject suicide for others—as she constantly worries about preventing suicide in her friends—she is at the same time sympathetic to the suicidal impulse and the need to escape. Notice how even though the interviewer’s question (and the larger flow of the conversation) was about Becca’s concern for her friends’ vulnerability to suicide, Becca answers the question “why suicide” by referencing her own distress, revealing how intimately she understands her friends’ motives. But Becca does not stop at identifying with escape from pressure as a motive for suicide; she also identifies with the person who died by suicide and sees herself as the same type of person and as facing the same type of ordeals.

Becca was not alone in her ability to identify with and claim others’ suicide motives as her own. Becca’s friend Madison (a teen), who lost her close friend Mark to suicide, also saw suicide as an “answer” to her own need to escape local pressures:

Like four years [after Mark’s death] … I would think like, “[Mark] had so many issues of his own, and now he doesn’t have them anymore. Like, how great would that be? To not have to like go through your life thinking about every little thing that you do. …” So it’s just kind of like, What a great idea. Like, you don’t have to like deal with any of your problems anymore. Like, you could just be. (emphasis added)

While thinking about suicide is not the same as completing suicide, internalizing suicide as a “great idea” could exacerbate risk for suicide should circumstances ever get sufficiently psychologically painful. Indeed, research identifies prosuicide attitudes as a risk factor for future suicide deaths (Renberg, Hjelmeland, and Koposov 2008; Phillips and Luth 2018).

Finally, we can provide some suggestive post hoc evidence that the local cultural script for suicide played a role in actual suicide deaths in the community. Since 2000, at least 18 PGHS youth or young adults have died by suicide; of those, we have detailed information about the circumstances surrounding the deaths of 9 decedents (based on interviews with multiple individuals close to the decedents). Respondents close to 5 of the 9 decedents explicitly noted that a need to escape the community pressure played a partial role in their significant others’ suicides. Leah and Beth provided examples of this pattern (for suicide decedents Stella Blue and Michelle, respectively), and their perspective was echoed by others close to Stella and Michelle. In addition, Bonnie, a youth who died by suicide and who left behind a 3,600-word journal documenting her decision to suicide, referenced her desire to escape the local pressure several times as a partial reason for her suicide. For example, in a table detailing her perception of the pros and cons for suicide, the pros all centered on academic pressure: “no school; no work; no college; no more stress” or social pressure: “no more bitches; no one getting mad at me; no more haters.” Though it is not possible to know where Bonnie came up with the idea of suicide as an escape, her close friends reported knowing about “three or four” suicide decedents, including Michelle, whose suicide was a “really really big deal” to them.

Discussion

Given the substantial increases in youth suicide rates since 2007 (Curtin et al. 2016) and the increased documentation of suicide clusters in school (Garcia-Williams et al. 2016), understanding why these events occur and, if possible, how to disrupt them is a pressing public health problem. With this study, we examined the case of a community that has experienced repeated suicide clusters to offer a more nuanced understanding of how and why suicide clusters may form and persist within a relatively bounded social unit like a tight-knit community. Though several decades’ worth of research has demonstrated that suicide, like other social behaviors, can spread through social relationships (Abrutyn and Mueller 2014a), rarely have these studies homed in on the specific processes or mechanisms by which exposure generates diffusion (cf. Mueller and Abrutyn 2015). Thus, we lack satisfying answers about why some places are sites of suicide clusters while others that appear similar are not.

By examining a community with an enduring adolescent suicide problem and comparing it to a reference group of suicide-bereaved individuals outside the community, we aimed to push the study of suicide more generally and suicide diffusion and clustering specifically in a new direction, namely, one that considers cultural sociology and social psychology—in particular, symbolic interactionism—in addition to the more standard structural, psychological, and Durkheimian insights. This move is necessary if we are to fully capture the complexity of how human social behaviors, including suicide, spread. To guide future research (and to facilitate summarizing a complicated process), we offer a formalized theoretical model that was informed by our empirical case for explaining why some places are more vulnerable to clusters; see Figures 1 and 2.

Figure 1.

Typical meaning-making process.

Figure 2.

Cognitive dissonance and reframing process.

To begin, Figure 1 visualizes the “typical” process by which an event (in our case, suicide) occurs and individuals make sense of it using existing cultural scripts. As long as the interpretation of the recent event is congruent with existing cultural scripts, meaning they “make sense,” the event easily can be defined and understood. Our reference group respondents illustrated this pattern quite consistently. Though suicide was often traumatic and painful, broad meanings about who dies by suicide were not challenged for this group of individuals. They generally interpreted the cause of suicide as mental illness, which matches broad, documented cultural scripts for suicide in the United States (Marsh 2010).

However, when we turn to our data from Poplar Grove—home to repeated adolescent suicide clusters—we see that the process depicted in Figure 1 can become a much more complex locally shared project that may alter the content of local meanings of suicide, a process we theorize in Figure 2. In Poplar Grove, our respondents also acknowledged the conventional set of meanings for why people die by suicide (like our reference group respondents: “mental illness”). Beginning from left to right, our theory suggests that suicide clusters emerge when cultural scripts for suicide are rekeyed such that suicide is a more imaginable option for a larger proportion of youth. This process begins with suicide deaths that do not publicly fit the existing scripts for understanding suicide (A1 in Figure 2). The cognitive dissonance generated by experiencing deaths that don’t “make sense” (A2) triggers both an individual and collective processes of attempting to make sense of why an individual died by suicide. As Fine’s (2010) work, among others, teaches us, shocking events that do not fit existing cultural scripts can put pressure on groups to rekey said scripts and, in so doing, can create new scripts that then any individual in the collective or, potentially, the community can draw on (see A3 → A4 in Figure 2). In Poplar Grove, many of the suicide deaths were quite shocking. Many deaths involved high-status, ideal typical girls—for example, high-GPA, beautiful, and athletic “stars”—and appeared to trigger a collective rekeying of the cultural scripts surrounding suicide in the local community (see also Mueller 2017).

Over time, these newly available local narratives became more and more accessible as subsequent suicides continued to fit the new cultural script, objectively or intersubjectively (A5). Dissonance gradually dissipated and was replaced with a well-worn story of why kids in Poplar Grove died by suicide (pressure). In addition, when the mental illness script was drawn upon, we found that it was often conceptualized as “caused” by the pressure. Ultimately, this scenario heightened vulnerability to suicide in the community, particularly among kids who both valued the community ideals and struggled to meet them, because the script made so much sense to them. Suicide was essentially something that youth like them did to escape the pressure.

This rekeying process ultimately offers a different but complementary take on structural explanations of diffusion while also offering new sociological insights into practical questions related to both the prevention of diffusion or clustering and postvention following a suicide or successive suicides. The remainder of the article addresses these two implications and, necessarily, the limitations of this study.

Implications for the Sociology of Diffusion

Arguably, by considering the role cultural mechanisms, like scripts, have in patterning suicide in this particular community, sociology can begin to more clearly understand why suicides cluster in some places and not others. To be sure, structural dimensions still matter, as the social cohesion in this community greatly facilitated the formation and diffusion of new cultural scripts and the salience of the collective in the interpretation of traumatic experiences. For example, as we showed, authoritative community members reified new cultural scripts for suicide, while other authorities (e.g., the local newspapers and the school) failed to discuss or suppressed other explanations for suicide (Mueller 2017). But our case also suggests that structural explanations will be lacking, particularly for complex diffusions, if diffusion theories do not also consider the role of cultural processes, particularly those surrounding local cultural meanings, in the translation of something new into an available, accessible, and applicable option. To be sure, we are not the first to suggest that meaning matters to understanding network processes. As Bernice Pescosolido (2006; see also Pescosolido 1994) has argued, the context of social networks matters as much as their structural factors. What we add, however, is an emphasis on not only exposure to influential people’s behaviors but a mechanism (cultural scripts) that transmit the meanings assigned to these behaviors.

In making this argument, we are joining the ranks of individuals calling for renewed effort at bridging social network theories and symbolic interactionism and cultural sociology (Fine and Kleinman 1983; Pescosolido 1992; Emirbayer and Goodwin 1994; Lizardo 2006; Fuhse 2009). Notably, symbolic interactionism teaches us that meanings are central to how people label, interpret, and express emotions, attitudes, and behaviors (McCall 2018), and thus they are central to cultural diffusion. Of course, as we illustrated with our empirical case, how meanings come to be relevant to the self for action also involves identity. Thus, another fruitful direction for future research involves examining more clearly how identity is related to suicide. Identities carry internalized meanings for the roles we play (e.g., an “ideal” youth), the status characteristics we possess (e.g., gender), and groups with which we affiliate (e.g., PGHS). Important to note, identities present social scientists with a link between intrapersonal processes (like social learning), the structural aspects of social networks (e.g., density of social ties), and the cultural milieu in which identity meanings are acquired and internalized (Abrutyn and Mueller 2016). Furthermore, previous research has repeatedly demonstrated that our identities matter to our physical, mental, and emotional health (Thoits 1983) and that “threats” to them produce intense negative affect (Burke and Stets 2009). As repositories of scripts, the role identity plays in protecting or making someone vulnerable to suicidality, as well as facilitating or constraining diffusion, seems like a crucial direction for suicide research.

Implications for Suicide Prevention

In addition, our study has implications for suicide prevention. This study, along with a small body of research, makes it clear that the stories we tell about suicide matter to whether suicide is seen as an option (Canetto 1997; Chua 2014; Stevenson 2014; Abrutyn 2017; Mueller 2017). Ideally, we would encourage communities to promote narratives about suicide that constrain suicide diffusion or prevent future clusters, but currently it is hard to know what these stories should look like. Prior research does suggest that exposure to salient role models who faced psychological pain and even contemplated suicide, but never attempted it and instead found another way to cope, can evince decreases in local suicide rates (Niederkrotenthaler et al. 2010). To the extent that communities can promote scripts about other ways to cope, told by role models who kids admire and with whom they identify, this may be a promising strategy.

However, there are some important cautions to consider before we promote controlling scripts as a suicide prevention strategy. First, rekeying suicide scripts as a mental health problem is not necessarily “safer” than other ways of understanding suicide. While the mental illness frame illustrated by our reference group respondents may have provided some respondents with some protective distance from identifying with suicide motives, this distance rests in part on mental health stigma. The line between mental “illness” (which many of our respondents felt they did not have) and psychological pain (which many were experiencing) is at least somewhat arbitrary. Thus, “othering” suicide as something that happens to mentally ill people, an unfortunately stigmatized category, may be self-defeating as it may discourage people from identifying their own suffering and seeking help (Pescosolido 2013).

A second caution is that controlling scripts in larger social groups can be extremely challenging. In part, this is because agents like newspapers act somewhat autonomously and often at cross-purposes with carefully crafted plans laid out by school officials, community organizations, or parents (Mueller 2017). In addition, youth have their own communication channels that adults are not always privy too, and thus, they may develop their own narratives regardless of adult interventions. One strategy that prior intervention research shows would be helpful is that communities should provide factual and clear information about suicide that is not romanticized or distorted.⁵ In Poplar Grove, youth’s distress was amplified by their confusion about why youth were dying and the denial, particularly by the school, that there was a problem. In addition to making suicide less of an option, youth may have had an easier time coping with the repeated tragedies if the community (and particularly the school) had agreed to talk about suicide rather than denying the problem.

Limitations

Despite our contributions, our study does have some limitations worth noting. Our fieldwork began after the pressure explanation was already in existence; thus, we had to rely on document analysis and interviews with older community members to re-create its emergence (Mueller and Abrutyn 2016; Mueller 2017). The community did have a well-documented “ground zero” suicide cluster around 2000 that most community members (even the very youngest) remembered or knew about that helped us determine the timeline (Mueller 2017), but we were not able to observe the emergence of the pressure script as it unfolded. In addition, suicide not only is stigmatized but also triggers intense emotions. While we tried our best to interview a wide range of respondents, we interviewed only people who volunteered, and these respondents represent only a small portion of Poplar Grove. We did not solicit interviews from individuals (except for one community leader we emailed directly so we could inform the person about the study); individuals had to express an interest before we contacted them. Hence, some people’s views are missing, some of which, undoubtedly, were those experiencing the strongest emotions or, perhaps, sense of stigma. Furthermore, some adults did not want us to talk to their children about suicide, and the school was not very cooperative—that is, while we did speak with current and former students and school personnel, we did not have the opportunity to observe within the school itself.

Finally, there is one wrinkle to the suicide problem in Poplar Grove that was beyond the scope of this study to explore: The majority of youth who died by suicide while at PGHS are young women. This is particularly surprising since suicide is generally more prevalent among boys than girls in the United States (Baca-Garcia et al. 2008). In addition, girls who died by suicide were more likely to be high-status role models and triggered cognitive dissonance, reframing, and in several cases suicide clusters. There is almost certainly something gendered about this process; however, particularly with our overrepresentation of female respondents (something we tried unsuccessfully to fix during data collection) it was challenging to determine the nuances of how gender mattered. Future research should endeavor to better understand the gendered nature of suicide and whether gender renders some suicides more surprising than others or whether there is something uniquely distressing for girls in this particular context or even more generally during this historical moment.

Conclusion

Since the time of the Greeks, anecdotal evidence of suicide diffusion and clustering has been reported. Five decades of systematic research on diffusion has strengthened suicidology’s confidence that exposure to suicide makes some populations, like adolescents, more vulnerable to suicidality. However, our understanding about why and how exposure leads to diffusion and in some places clustering remains an unresolved puzzle. Leveraging an in-depth case study of a community with an enduring suicide problem, including repeating clusters, we advance a theory of suicide diffusion that demonstrates (1) the pressing need for the inclusion and investigation of cultural factors in the social scientific study of suicide and (2) the consequences exposure has for potentially rekeying a community, school, family, or peer group’s suicide script and therefore the meanings attributed to why people die by suicide and who is most at risk of suicide. Of course, this study is just a first step. We hope, however, that our findings and our theoretical model encourage new waves of sociological research that bridge structural and cultural insights into social behavior. By drawing on the full theoretical and methodological tool kit of sociology, we believe sociology can offer compelling insights into pressing public health problems like youth suicide and important process like social diffusion.

Footnotes

Acknowledgements

This project was supported by a grant from the Social Science Research Center at the University of Chicago (Anna Mueller, principal investigator). We would like to thank Basia Ellis, Ken Frank, John Levi Martin, and April Sutton for their detailed comments and suggestions on previous drafts as well as the reviewers for their insightful comments and suggestions. We also would like to express deep gratitude to the people of “Poplar Grove” for sharing their lives with us and facilitating the research.

Authors’ Note

A previous version of this article was presented at the 2017 annual meetings of the American Sociological Association.

Notes

References

Abrutyn

Seth

. 2017. “What Hindu Sati Can Teach Us about the Social Psychological Dynamics of Suicide.” Journal for the Theory of Social Behaviour 47(4):522–39.

Abrutyn

Seth

Mueller

Anna S.

2014a. “Are Suicidal Behaviors Contagious? Using Longitudinal Data to Examine Suicide Suggestion.” American Sociological Review 79(2):211–27.

Abrutyn

Seth

Mueller

Anna S.

2014b. “Reconsidering Durkheim’s Assessment of Tarde: Formalizing a Tardian Theory of Imitation, Contagion, and Suicide Suggestion.” Sociological Forum 29(3):698–719.

Abrutyn

Seth

Mueller

Anna S.

2016. “When Too Much Integration and Regulation Hurt: Re-envisioning Durkheim’s Altruistic Suicide.” Society and Mental Health 6(1):56–71.

Abrutyn

Seth

Mueller

Anna S.

2018. “Towards a Cultural-structural Theory of Suicide: Examining Excessive Regulation and Its Discontents.” Sociological Theory 36(1):48–66.

Akers

Ronald L.

Jensen

Gary F.

2006. “The Empirical Status of Social Learning Theory of Crime and Deviance: The Past, Present, and Future.” Pp. 37–76 in Taking Stock: The Status of Criminological Theory, edited by Cullen

F. T.

Wright

J. P.

Blevins

K. R.

New Brunswick, NJ: Transaction.

Alexander

Jeffrey C.

2004. “Toward a Theory of Cultural Trauma.” Pp. 1–30 in Cultural Trauma and Collective Identity, edited by Alexander

J. C.

Eyerman

Giesen

Smelser

N. J.

Sztompka

Berkeley: University of California Press.

Arensman

Ella

McAuliffe

Carmel

. 2015. “Clustering and Contagion of Suicidal Behavior.” Pp. 99–105 in Suicidal Behavior: Underlying Dynamics, edited by Kumar

London, England: Routledge.

Baca-Garcia

Enrique, M.

Perez-Rodriguez

Mercedes

Mann

J. John

Oquendo

Maria A.

2008. “Suicidal Behavior in Young Women.” Psychiatric Clinics of North America 31:317–31.

10.

Baller

Robert D.

Richardson

Kelly K.

2009. “The ‘Dark Side’ of the Strength of Weak Ties: The Diffusion of Suicidal Thoughts.” Journal of Health and Social Behavior 50(3):261–76.

11.

Brent

David A.

Kerr

Mary M.

Goldstein

Charles

Bozigar

James

Wartella

Marty

Allan

Marjorie J.

1989. “An Outbreak of Suicide and Suicidal Behavior in a High School.” American Academy of Child and Adolescent Psychiatry 28(6):918–24.

12.

Burke

Peter J.

Stets

Jan E.

2009. Identity Theory. New York: Oxford University Press.

13.

Burt

Ronald S.

2004. “Structural Holes and Good Ideas.” American Journal of Sociology 110(2):349–99.

14.

Canetto

Silvia Sara

. 1997. “Meanings of Gender and Suicidal Behavior during Adolescence.” Suicide and Life-threatening Behavior 27(4):339–51.

15.

Centers for Disease Control and Prevention. 2017. “Suicide Rates for Teens Aged 15–19 Years, by Sex—United States, 1975–2015.” Morbidity and Mortality Weekly Report 66(30):816. Retrieved March 5, 2019 (https://dx-doi-org.web.bisu.edu.cn/10.15585/mmwr.mm6630a6).

16.

Centola

Damon

Macy

Michael

. 2007. “Complex Contagions and the Weakness of Long Ties.” American Journal of Sociology 113(3):702–34.

17.

Cerel

Julie

Brown

Margaret M.

Maple

Myfanwy

Singleton

Michael

van de Venne

Judy

Moore

Melinda

Flaherty

Chris

. 2018. “How Many People Are Exposed to Suicide? Not Six.” Suicide and Life-threatening Behavior. Retrieved March 7, 2019 (https://doi.org/10.1111/sltb.12450).

18.

Cerel

Julie

Maple

Myfanwy

van de Venne

Judy

Moore

Melinda

Flaherty

Chris

Brown

Margaret

. 2016. “Exposure to Suicide in the Community: Prevalence and Correlates in One U.S. State.” Public Health Reports 131(1):100–107.

19.

Chua

Jocelyn Lim

. 2014. In Pursuit of the Good Life: Aspiration and Suicide in Globalizing South India. Berkeley: University of California Press.

20.

Currier

Joseph M.

Irish

Jennifer E. F.

Neimeyer

Robert A.

Foster

Joshua D.

2015. “Attachment, Continuing Bonds, and Complicated Grief Following Violent Loss: Testing a Moderated Model.” Death Studies 39:201–10.

21.

Curtin

S. C.

Warner

Hedegaard

2016. Increase in Suicide in the United States, 1999–2014. NCHS Data Brief No. 241. Hyattsville, MD: U.S. Department of Health and Human Services, CDC, National Center for Health Statistics. Retrieved March 5, 2019 (https://www.cdc.gov/nchs/data/databriefs/db241.pdf).

22.

Eliasoph

Nina

Lichterman

Paul

. 2003. “Culture in Interaction.” American Journal of Sociology 108(4):735–94.

23.

Emirbayer

Mustafa

Goodwin

Jeff

. 1994. “Network Analysis, Culture, and the Problem of Agency.” American Journal of Sociology 99:1141–54.

24.

Erikson

Kai T.

1978. Everything in Its Path: Destruction of Community in the Buffalo Creek Flood. New York: Simon and Schuster.

25.

Fine

Gary Alan

. 2010. “The Sociology of the Local: Action and Its Publics.” Sociological Theory 28(4):355–76.

26.

Fine

Gary Alan

Kleinman

Sherryl

. 1983. “Network and Meaning: An Interactionist Approach to Structure.” Symbolic Interaction 6(1):97–110.

27.

Fletcher

Jason M.

2017. “Gender-specific Pathways of Peer Influence on Adolescent Suicidal Behaviors.” Socius. Retrieved March 7, 2019 (https://doi.org/10.1177/2378023117729952).

28.

Friedkin

Noah E.

2004. “Social Cohesion.” Annual Review of Sociology 30:409–25. doi:10.2307/29737700.

29.

Fuhse

Jan A.

2009. “The Meaning Structure of Social Networks.” Sociological Theory 27(1):51–73.

30.

Garcia-Williams

Amanda

O’Donnell

Julie

Spies

Erica

Zhang

Xinjian

Young

Randall

Azofeifa

Alejandro

Vagi

Kevin

. 2016. Epi-aid 2016-018: Undetermined Risk Factors for Suicide among Youth, Ages 10–24—Santa Clara County, Ca, 2016.” Retrieved March 7, 2019 (https://www.sccgov.org/sites/phd/hi/hd/epi-aid/Documents/epi-aid-report.pdf).

31.

Goddard

Cliff

Wierzbicka

Anna

. 2004. “Cultural Scripts: What Are They and What Are They Good For?” Intercultural Pragmatics 1/2:153–66.

32.

Gould

Madelyn S.

Wallenstein

Sylvan

Kleinman

Marjorie

O’Carroll

Patrick

Mercy

1990. “Time-space Clustering of Teenage Suicide.” American Journal of Epidemiology 131(1):71–78.

33.

Gould

Madelyn S.

Velting

Drew

Kleinman

Marjorie

Lucas

Christopher

Thomas

John Graham

Chung

Michelle

. 2004. “Teenagers’ Attitudes about Coping Strategies and Help-seeking Behavior for Suicidality.” Journal of the American Academy of Child and Adolescent Psychiatry 43(9):1124–33.

34.

Granovetter

Mark S.

1973. “The Strength of Weak Ties.” American Journal of Sociology 78(6):1360–80.

35.

Harding

David J.

2007. “Cultural Context, Sexual Behavior, and Romantic Relationships in Disadvantaged Neighborhoods.” American Sociological Review 72(3):341–64.

36.

Haw

Camilla

Hawton

Keith

Niedzwiedz

Claire

Platt

Steve

. 2013. “Suicide Clusters: A Review of Risk Factors.” Suicide and Life-threatening Behavior 43(1):97–108.

37.

Jamison

Kay Redfield

. 2001. Night Falls Fast: Understanding Suicide. New York: Vintage.

38.

Kadushin

Charles

. 2012. Understanding Social Networks: Theories, Concepts, and Findings. New York: Oxford University Press.

39.

Kitanaka

Junko

. 2012. Depression in Japan: Psychiatric Cures for a Society in Distress. Princeton, NJ: Princeton University Press.

40.

Lake

Alison M.

Kandasamy

Suganthi

Kleinman

Marjorie

Gould

Madelyn S.

2013. “Adolescents’ Attitudes about the Role of Mental Illness in Suicide, and Their Association with Suicide Risk.” Suicide and Life-threatening Behavior 43(6):692–703.

41.

Lizardo

Omar

. 2006. “How Cultural Tastes Shape Personal Networks.” American Sociological Review 71(5):778–807.

42.

Lizardo

Omar

. 2017. “Improving Cultural Analysis: Considering Personal Culture in Its Declarative and Nondeclarative Modes.” American Sociological Review 82(1):88–115.

43.

Lizardo

Omar

Strand

Michael

. 2010. “Skills, Toolkits, Contexts and Institutions: Clarifying the Relationship between Different Approaches to Cognition in Cultural Sociology.” Poetics 38(2):205–28.

44.

Marsden

Peter V.

Friedkin

Noah E.

1993. “Network Studies of Social Influence.” Sociological Methods & Research 22(1):127–51.

45.

Marsh

Ian

. 2010. Suicide: Foucault, History and Truth. Cambridge, UK: Cambridge University Press.

46.

McCall

George J.

2018. “Symbolic Interaction.” Pp. 5–28 in Contemporary Social Psychological Theories, edited by Burke

P. J.

Stanford, CA: Stanford University Press.

47.

Mueller

Anna S.

2017. “Does the Media Matter to Suicide? Examining the Social Dynamics Surrounding Media Reporting on Suicide in a Suicide-prone Community.” Social Science and Medicine 180:152–59.

48.

Mueller

Anna S.

Abrutyn

Seth

. 2015. “Suicidal Disclosures among Friends: Using Social Network Data to Understand Suicide Contagion.” Journal of Health and Social Behavior 56(1):131–48.

49.

Mueller

Anna S.

Abrutyn

Seth

. 2016. “Adolescents under Pressure: A New Durkheimian Framework for Understanding Adolescent Suicide in a Cohesive Community.” American Sociological Review 81(5):877–99.

50.

Myfanwy

Maple

Cerel

Julie

Sanford

Rebecca

Pearce

Tania

Jordan

Jack

. 2017. “Is Exposure to Suicide Beyond Kin Associated with Risk for Suicidal Behavior? A Systematic Review of the Evidence.” Suicide and Life-threatening Behavior 47(4):461–74.

51.

Neimeyer

Robert A.

Baldwin

Scott A.

Gillies

James

. 2006. “Continuing Bonds and Reconstructing Meaning: Mitigating Complications in Bereavement.” Death Studies 30:715–38.

52.

Niederkrotenthaler

Thomas

Voracek

Martin

Herberth

Arno

Till

Benedikt

Strauss

Markus

Etzersdorfer

Elmar

Eisenwort

Brigette

Sonneck

Gernot

. 2010. “Role of Media Reports in Completed and Prevented Suicide: Werther V. Papageno Effects.” British Journal of Psychiatry 197(3):234–43.

53.

Niedzwiedz

Claire

Haw

Camilla

Hawton

Keith

Platt

Stephan

. 2014. “The Definition and Epidemiology of Clusters of Suicidal Behavior: A Systematic Review.” Suicide and Life-threatening Behavior 44(5):569–81.

54.

Norton

Matthew

. 2014. “Mechanisms and Meaning Structures.” Sociological Theory 32(2):162–87.

55.

Patterson

Orlando

. 2014. “Making Sense of Culture.” Annual Review of Sociology 40:1–30.

56.

Pescosolido

Bernice A.

1992. “Beyond Rational Choice: The Social Dynamics of How People Seek Help.” American Journal of Sociology 97(4):1096–1138. doi:10.1086/229863.

57.

Pescosolido

Bernice A.

1994. “Bringing Durkheim into the Twenty-first Century: A Network Approach to Unresolved Issues in the Sociology of Suicide.” Pp. 264–93 in Emile Durkheim: Le Suicide 100 Years Later, edited by Lester

Philadelphia: Charles Press.

58.

Pescosolido

Bernice A.

2006. “Of Pride and Prejudice: The Role of Sociology and Social Networks in Integrating the Health Sciences.” Journal of Health and Social Behavior 47(3):189–208.

59.

Pescosolido

Bernice A.

2013. “The Public Stigma of Mental Illness: What Do We Think; What Do We Know; What Can We Prove?” Journal of Health and Social Behavior 54(1):1–21. doi:10.1177/0022146512471197.

60.

Phillips

Julie A.

Luth

Elizabeth A.

2018. “Beliefs about Suicide Acceptability in the United States: How Do They Affect Suicide Mortality?” Journals of Gerontology: Series B. Retrieved March 7, 2019 (https://www.ncbi.nlm.nih.gov/pubmed/29378018).

61.

Pirkis

Jane

Robinson

. 2014. “Improving Our Understanding of Youth Suicide Clusters.” Lancet Psychiatry 1(June):5–6.

62.

Pirkis

Jane

Robinson

O’Connor

Rory

. 2016. “Suicide Clusters.” Pp. 758–74 in The International Handbook of Suicide Prevention, edited by O’Connor

Pirkis

Malden, MA: Wiley Blackwell.

63.

Pitman

Alexandra

Nesse

Hedvig

Morant

Nicola

Azorina

Valeriya

Stevenson

Fiona

King

Michael

Osborn

David

. 2017. “Attitudes to Suicide Following the Suicide of a Friend or Relative: A Qualitative Study of the Views of 429 Young Bereaved Adults in the UK.” BMC Psychiatry. Retrieved March 7, 2019 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5729247/).

64.

Randall

Jason R.

Nickel

Nathan C.

Colman

Ian

. 2015. “Contagion from Peer Suicidal Behavior in a Representative Sample of American Adolescents.” Journal of Affective Disorders 186:219–25.

65.

Renberg

Ellinor Salander

Hjelmeland

Heidi

Koposov

Roman

. 2008. “Building Models for the Relationship between Attitudes toward Suicide and Suicidal Behavior: Based on Data from General Population Surveys in Sweden, Norway, and Russia.” Suicide and Life-threatening Behavior 38(6):661–75.

66.

Rosin

Hanna

. 2015. “The Silicon Valley Suicides: Why Are So Many Kids with Bright Prospects Killing Themselves in Palo Alto?” The Atlantic. Retrieved March 7, 2019 (https://www.theatlantic.com/magazine/archive/2015/12/the-silicon-valley-suicides/413140/).

67.

Roth

Wendy D.

Mehta

Jal D.

2002. “The Rashomon Effect: Combining Positivist and Interpretivist Approaches in the Analysis of Contested Events.” Sociological Methods and Research 31:131–72.

68.

Simon

William

Gagnon

John H.

2003. “Sexual Scripts; Origins, Influences, and Changes.” Qualitative Sociology 26(4):491–97.

69.

Spies

Erica

Ivey-Stephenson

Asha

VanderEnde

Kristin

Lynch

Sean

Dean

David

Gleason

Brigette

. 2014. Epi-aid 2015-003: Undetermined Risk Factors for Suicide among Youth, Ages 10–24—Fairfax County, Va, 2014. Retrieved March 7, 2019 (https://www.fairfaxcounty.gov/health/sites/health/files/assets/images/suicide-epi-aid-questions-answers.pdf).

70.

Stack

Steven

. 1990. “Divorce, Suicide, and the Mass Media: An Analysis of Differential Identification, 1948–1980.” Journal of Marriage and Family 52(2):553–60.

71.

Stack

Steven

. 2005. “Suicide in the Media: A Quantitative Review of Studies Based on Nonfictional Stories.” Suicide and Life-threatening Behavior 35(2):121–33.

72.

Stack

Steven

Abrutyn

Seth

. 2015. “Suicide by Fire: Ethnicity as a Predictor of Self Immolation in the U.S.” Pp. 193–210 in Suicide as Performance, edited by Lester

Stack

New York: Transaction.

73.

Stevenson

Lisa

. 2014. Life Beside Itself: Imagining Care in the Canadian Arctic. Berkeley: University of California Press.

74.

Tarde

Gabriel

. 1903. The Laws of Imitation. New York: Henry Holt.

75.

Thoits

Peggy A.

1983. “Multiple Identities and Psychological Well-being: A Reformulation and Test of the Social Isolation Hypothesis.” American Sociological Review 48(2):174–87.

76.

Timmermans

Stefan

Tavory

Iddo

. 2012. “Theory Construction in Qualitative Research: From Grounded Theory to Abductive Analysis.” Sociological Theory 30(3):167–86.

77.

White

Jennifer

Morris

Jonathan

. 2010. “Precarious Spaces: Risk, Responsibility and Uncertainty in School-based Suicide Prevention Programs.” Social Science and Medicine 71(12):2187–94.

78.

Wood

Michael

Stoltz

Dustin S.

Van Ness

Justin

Taylor

Marshall

. 2018. “Schemas and Frames.” Sociological Theory 36(3):244–61.