Approximately 90 percent of individuals who die by suicide meet the diagnostic criteria for a psychiatric disorder in the last six months of life. … Cross-nosological depressive symptoms drastically alter the individual’s problem-solving capacity and judgment [and] they may be interpreted as precipitants of suicidal crises. These depressive states associate with molecular changes that are likely to underlie proximal risk factors for suicide. (Turecki Reference Turecki2014a, 11-12)
From antiquity to the present, trauma has been a keyword used by clinicians and scholars to describe certain negative experiences and their aftermath (Ben-Ezra Reference Ben-Ezra2011; Berrios and Porter Reference Berrios and Porter1999; Birmes et al. Reference Birmes, Hatton, Brunet and Schmitt2003). Examined as a distinct object across a wide range of disciplines (Caruth Reference Caruth1996; LaCapra Reference LaCapra2001; Leys Reference Leys2000; Luckhurst Reference Luckhurst2008), trauma has been discussed and debated intensively. In recent years, humanities scholars and social scientists have explored trauma through a diverse set of situations of extremity and associated individual and collective experiences and responses (Bracken Reference Bracken2003; Das Reference Das2003; Fassin and Rechtman Reference Fassin and Richard2009; Mbembe Reference Mbembe2006; Young Reference Young2007). Trauma, in these studies, is considered a multifaceted experience variously associated with illness and suffering, bodily experiences, and the moral politics of victimhood, blame, and accountability (Kirmayer, Lemelson, and Barad Reference Kirmayer, Robert and Mark2007, 4; Hinton and Kirmayer Reference Hinton and Kirmayer2013; Kitanaka Reference Kitanaka, Jerome and Demazeux2016; Fassin Reference Fassin2014; Rechtman Reference Rechtman2000).
Contemporary psychiatric research on trauma has a distinctive history in which the concept has been serially reinterpreted as researchers attempt to characterize the specificity the origins of trauma and its consequences. The multiple historical transformations of the concept are inextricably connected yet equally characterized by intellectual disjunctions. The salient examples and prototypes of trauma within psychiatry have changed over time, as trauma theory has moved from surgical metaphors of injuries and healing in the late 1800s to domain-specific models based on psychological and physiological processes in the 1990s and beyond (Leys Reference Leys2000; Micale and Lerner Reference Micale and Lerner2001; Young Reference Young1995). This history is far from linear. As Judith Herman, a psychiatrist and trauma specialist, acknowledges:
The study of psychological trauma has a curious history – one of episodic amnesia. Periods of active investigation have alternated with periods of oblivion. Repeatedly in the past century, similar lines of inquiry have been taken up and abruptly abandoned, only to be rediscovered much later. (Herman [1992] Reference Herman2015, 7)
Throughout this history, the psychiatric profession has periodically been fascinated by the pathogenic nature of traumatic events, while at other times these professionals have called into question the relevance of personal accounts of trauma in the etiology of health and mental health problems (Van der Kolk Reference Kolk, Friedman, Keane and Resick2007, 19). This article focuses on these shifting narratives of trauma, its effects, and the relationship of both to the development of psychopathology. We examine etiological narratives of trauma and most particularly shifts in the importance accorded to traumatic events at different points of a person’s life and increasing interest in childhood trauma. If each narrative affects the significance accorded to personal accounts of trauma, we suggest that contemporary neuroscientific models of the embodied damage and neurobiological risk thought to follow early trauma consign later life experiences to etiological obscurity. Through attention to narratives of neurobiological risk emerging from psychiatric, neuroscience, and environmental epigenetics research, we explore how suicide has come to be seen as a behavior that has no significant precipitating event, but rather an exceptional precipitating neurochemical state whose origins are identified in distant experiences of early traumatic events.
While our research focuses on environmental epigenetics models of suicide risk emerging from the McGill Group for Suicide Studies (MGSS), these models are consistent with increasingly dominant etiological narratives in contemporary biopsychiatry more generally. These narratives are often seen as specific to the postgenomic era in which life is increasingly seen as characterized by different moments of instability and stability rather than being anchored by the “book of life” (i.e., a presumably stable genome) (Lappé and Landecker Reference Lappé, Hannah and Bria2015). While the narratives we trace are closely linked to reasoning in the postgenomic era (Pentecost and Meloni Reference Pentecost and Maurizio2020), we examine earlier narratives that contributed to contemporary etiological accounts of trauma and its effects. In particular, we draw on two narratives of trauma, each of which concerns questions of time and psychopathology, as narratives that had to be stabilized in order for MGSS models of suicide risk to be posited. We suggest that environmental epigenetics has provided a language and mode of reasoning to bring these two narratives together, drawing questions of time, trauma, and psychopathology into the heart of neuroscientific reasoning. In closing, we orient our attention to the question of suicide to consider how environmental epigenetics explanations of suicidal behavior reframe suicidal acts as “post”: a post-traumatic subjectivity characteristic to contemporary neuroscience and biopsychiatry research.
Narratives of post-traumatic subjectivity
The introductory citation is drawn from an article “The Molecular Bases of the Suicidal Brain” by Gustavo Turecki, a psychiatrist and molecular biologist in Montreal (Turecki Reference Turecki2014a). Turecki is the director of the MGSS, where we have carried out ethnographic research since 2013. Footnote 1 The quotation reflects the MGSS’s model of suicide risk, in which suicide is associated with psychopathology, which is in turn associated with specific, underlying neurobiological profiles. It is these brain-based molecular profiles that are thought to constitute suicide risk and the precipitating factors for suicide. The origins of these at risk molecular profiles are believed to be found in the experience of early childhood adversity (ECA). The lives of “individuals who died by suicide” (Chen et al. Reference Chen, Daniel, Laura and Gustavo2018), from this perspective, are seen as pathological trajectories set in place in the early years through experiences of adversity that enact durable epigenetic changes in the brains of youth. These people are considered susceptible to mental illness, poor decision-making, and a variety of “unhealthy” behaviors (Roth et al. Reference Roth2009; Labonté et al. Reference Labonté2012; Jollant et al. Reference Jollant, Nathalia, Emilie, Owen Gareth and Alain2010). Suicide is seen as the last, and most extreme, step in this trajectory (Turecki et al. Reference Turecki2012; Turecki Reference Turecki2014b).
Turecki’s words are drawn from an article section entitled “Proximal or precipitating factors” of suicide (Turecki Reference Turecki2014a, 11). Subsections address a variety of molecular systems, from serotonergic alterations to inflammatory factors, considered to characterize the proximal or precipitating factors related to suicide. While the role of life events in triggering suicidal crises is acknowledged in this article and in MGSS research more generally, these events are typically described as quite common negative experiences, such as a rupture in a relationship or the loss of a job or status (Fiori and Turecki Reference Fiori and Turecki2016; Turecki Reference Turecki2014b; Turecki et al. Reference Turecki2012). Given the relatively ordinary status of these events, it is neurobiological proximal states that are considered the significant precipitating factors in this model of suicide (Turecki Reference Turecki2014a, 33). This research seeks new ways of characterizing suicide risk, recognizing that decades of intensive efforts in fundamental and clinical research to reduce suicide rates have met with limited success (Turecki and Brent Reference Turecki and Brent2016). While the experience of certain forms of mental illness, such as depression, remain risk factors for suicide, research has equally made it clear that most people who experience depression will never commit suicide (Hawton and van Heeringen Reference Hawton and Kees2009; McGirr et al. Reference McGirr2008a). Research agendas like those at the MGSS attempt to identify who, among those people who share risk factors later in life (e.g., depression), are the ones most likely to develop suicidal ideation and behavior (McGirr and Turecki Reference McGirr and Gustavo2007).
What is striking about Turecki’s argument is not its exceptionality among models of neurobiological risk, but its precise fit with contemporary neuroscience narratives. The MGSS models are consistent with a burgeoning set of research programs in neuroscience and epigenetics that aim to identify how specific life “exposures” (both material and social) lead to a range of mental health outcomes as well as a particular set of subjective experiences, behaviors, and risks. What these narratives share is attention to the ways in which the past is thought to invade the present. We will examine two of these. The first narrative concerns studies of ECA. Beginning with growing interest in the experience of ECA in the 1960s, we examine research on abuse in the last twenty years and the production of specific measures to quantify the experience and study its effects. We focus specifically on the measures of ECA used at the MGSS. Contemporary research on child abuse has demonstrated associations between early traumatic experiences and a wide range of mental health problems later in life (Briere and Elliott Reference Briere and Elliott2003; Heim et al. Reference Heim, Shugart, Craighead and Nemeroff2010; Scher et al. Reference Scher2004). Attention within this research has been oriented to classifying the original experience and identifying its heterogeneous effects rather than on classifying a specific resultant psychiatric diagnosis. Footnote 2 As a consequence, research agendas focusing on ECA ran largely in parallel to the disorder-specific studies of post-traumatic subjectivity that have dominated psychiatric research on trauma since the publication of DSM-III (third edition of the Diagnostic and Statistical Manual of the American Psychiatric Association) in 1980.
The second narrative relates to the development and expansion of the diagnostic category post-traumatic stress disorder (PTSD). Post-traumatic stress disorder, while included in the DSM since 1980, was an exceptional diagnosis in the manual’s symptom-based approach to mental disorders due to its presumed origins in the memory of a traumatic event (Young Reference Young1995). In comparison with research on ECA, the DSM-III category PTSD situated traumatic events experienced in adulthood – and memories of these events – as playing an etiologically-specific role in the disorder. Over time, however, the etiological significance of traumatic events and their presumed effects shifted in PTSD research. Research on the prevalence, course, and comorbidity of PTSD turned from the exceptional character of traumatic events more proximate to the development of the disorder to predispositions to react pathologically to negative or trauma-level events later in life. The result is an understanding of PTSD in which the person’s response to a stressor is determined not by the stressor itself but by the interaction between the stressor and the person’s premorbid vulnerability (Shephard Reference Shephard and Rosen2004, 54). Early life adversity is among the experiences considered capable of engendering this predisposition (Yehuda, Halligan, and Grossman Reference Yehuda2001).
If historically these two areas of research were carried out with little overlap, with notable exceptions, over time the narratives within PTSD and ECA research have become more consistent with one another most specifically as they concern the neurobiological vulnerability thought to result from negative early experiences. As a case in point, many trauma specialists came to agree that the diagnostic criteria for PTSD, understood as a category tightly associated with the experience of early adversity, captured only a fraction of trauma-related psychopathology in children, invoking the need for an expanded PTSD diagnostic category (van der Kolk et al. Reference Kolk and Roth2005, 390). These “plastic” understandings (Meloni Reference Meloni2019) of the relationships between time and (psycho)pathology have become anchored in a wide range of influential research agendas ranging from studies of the developmental origins of health and disease (DOHaD) to “trauma-informed care.”
Early childhood abuse: An etiological significant event without a diagnostic category
Studies of adverse childhood experiences or ACEs are a part of an increasingly substantial and influential body of research that has emerged in recent decades, alongside growing interest in developmental perspectives of risk. Emblematic of this fast-moving field of knowledge is attention to the putative role of early adversity on health and behavior, as demonstrated by foundational ACEs studies (e.g., Felitti et al. Reference Felitti, Anda, Dale, Williamson, Spitz, Valerie, Koss and Marks1998) and the vast body of literature that followed (Anda et al. Reference Anda, Felitti, Douglas Bremner, Walker and Charles2006; Briere and Elliott Reference Briere and Elliott2003; Dong et al. Reference Dong, Anda, Felitti, Dube and Williamson2004; Dube et al. Reference Dube, Anda, Felitti, Chapman, Williamson and Giles2001; Edwards et al. Reference Edwards, Holden, Felitti and Anda2003). While the history of psychiatry is punctuated by episodic attention to the effects of early adversity (Duché Reference Duché1990; Dorahy, van der Hart and Middleton Reference Dorahy, Onno, Warwick, Ruth, Vermetten and Painé2010; Bensel, Rheinberger, and Radbill Reference Bensel, Rheinberger and Radbill1997), it is only since the 1960s that early adversity gained a stronger foothold in psychiatric research, turning attention toward chronic social problems such as childhood abuse and maltreatment (Dorahy, van der Hart and Middleton Reference Dorahy, Onno, Warwick, Ruth, Vermetten and Painé2010, 7-8; Hacking Reference Hacking1995, 58; Hacking Reference Hacking1991; DeMause [1974] Reference DeMause1995; Hendrick Reference Hendrick, James and Prout1997; Myers Reference Myers2008).
From this point forward, it was generally accepted that a history of severe or prolonged abuse, particularly in childhood, “fosters the development of a prodigious array of psychiatric symptoms,” and appears “to be one of the major factors predisposing a person to become a psychiatric patient” (Herman Reference Herman1992, 379). By the 1990s, there were so many possible effects and clinical manifestations attributed to childhood psychic trauma that psychiatrist Leonor Terr argued, “we must organize our thinking about childhood trauma, however, or we run the risk of never seeing the condition at all” (Terr Reference Terr1991, 10).
In the 2000s, a working committee known as the Complex Trauma Task Force (CTTF) of the National Child Traumatic Stress Network (Cloitre et al. Reference Cloitre, Chris, Julian, Bonnie and Pamela2012) was brought together with the objective of consolidating research and clinical understandings of and responses to child abuse (Cook et al. Reference Cook, Blaustein, Spinazzola and van der Kolk2003; Ford and Cloitre Reference Ford, Cloitre, Chrisine and Julian2009; van der Kolk Reference Kolk2005). They proposed a new diagnosis provisionally called developmental trauma disorder (DTD) (Pynoos et al. Reference Pynoos, Steinberg, Layne, Briggs, Ostrowski and Fairbank2009; van der Kolk et al. Reference Kolk2009; van der Kolk Reference Kolk2005) that aimed to bring attention to the ways in which “multiple exposures to interpersonal trauma, such as abandonment, betrayal, physical or sexual assaults, or witnessing domestic violence, have consistent and predictable consequences that affect many areas of functioning” (van der Kolk Reference Kolk2005, 406). Though proponents of the diagnosis argued for its addition to DSM-5 in 2013 (see Schmid, Petermann, and Fergert Reference Schmid2013; Sar Reference Sar2011), the diagnosis has yet to be included in the diagnostic manual. It has nonetheless drawn increasing attention to ECA/ELA (early life abuse) and to the effects of multiple, cumulative, and/or chronic and prolonged childhood adversities from the early 2000s onward. According to the proponents of DTD, forms of ECA were believed to manifest themselves not only in post-traumatic stress psychological symptomatology, but also in the form of various medical illnesses and conditions which tended to be pervasive and multifaceted (Ackerman et al. Reference Ackerman, Newton, McPherson, Jones and Dykman1998; Putnam Reference Putnam2003; van der Kolk Reference Kolk2000). These symptoms, and the biological reactions or responses to stress thought to underlie them, increasingly came to be understood through the lens of emerging neurobiological theories of trauma, a point that we will return to later in the paper (Cook et al. Reference Cook, Blaustein, Spinazzola and van der Kolk2003; Perry et al. Reference Perry, Pollard, Blakley, Baker and Domenico1995; van der Kolk et al. Reference Kolk2009; Yehuda 2001; Reference Yehuda2000).
Defining and measuring childhood adversity
In the present day, the association between ECA/ELA and mental and physical ill health is widely accepted in fundamental and clinical research (Anda et al. Reference Anda, Felitti, Douglas Bremner, Walker and Charles2006; Dube et al. Reference Dube, Anda, Felitti, Chapman, Williamson and Giles2001; Dube et al. Reference Dube, Anda, Withfield, Brown, Felitti, Maxia and Giles2005; Edwards et al. Reference Edwards, Holden, Felitti and Anda2003; Heim and Binder Reference Heim and Binder2012). Among childhood adversities,Footnote 3 child abuse and maltreatment have been strongly correlated with adult psychopathology, psychiatric morbidity, and suicidality (Afifi et al. Reference Afifi, Boman, Fleisher and Sareen2009; Brezo et al. Reference Brezo, Joel, Frank, Martine, Tremblay and Gustavo2008; Briere and Elliott Reference Briere and Elliott2003; Evans et al. Reference Evans, Keith and Karen2005; Gilbert et al. Reference Gilbert, Cathy Spatz, Kevin, David, Elspeth and Staffan2009; Heim and Nemeroff Reference Heim and Nemeroff2001). These trajectories have been documented in clinical and developmental psychiatry using a range of screening tools and standardized instruments to retrospectively assess childhood adversities among both clinical and nonclinical populations (see Bailhache et al. Reference Bailhache, Leroy, Pillet and Salmi2013; Bethell et al. Reference Bethell, Carle, Hudziak, Gombojav, Powers, Wade and Braveman2017; Milne and Collin-Vézina Reference Milne and Delphine2015; Roy and Perry Reference Roy and Perry2004; Strand, Sarmiento, and Pasquale Reference Strand, Sarmiento and Pasquale2005; Thabrew, de Sylva, and Romans 2012; Tonmyr et al. Reference Tonmyr2011). Researchers have favored descriptive instruments to gather data on childhood adversities (Spinhoven et al. Reference Spinhoven2014, 718), which are then aggregated by, for instance, summing the number of traumatic experiences (ibid.) as in the case of the Adverse Childhood Experience Study (ACEs) (Chapman et al. Reference Chapman, Whitfield, Felitti, Dube, Edwards and Anda2004; Dube et al. Reference Dube, Anda, Felitti, Chapman, Williamson and Giles2001; Felitti et al. Reference Felitti, Anda, Dale, Williamson, Spitz, Valerie, Koss and Marks1998) or classifying specific types of adverse experiences. In the latter case, researchers have sought to characterize types of trauma in terms of levels of psychological distress following a stressful life event (e.g., the Impact of Event Scale – revised version [IES-R] [Weiss et al. Reference Weiss and Charles R.1997]) or the specificity of the traumatic event such as exposure to crime, natural disasters, physical or sexual assault (e.g., the Trauma History Questionnaire [Hooper et al. et al. Reference Hooper, Patricia, Krupnick and Green2011] and the Life Incidence of Traumatic Events [LITE] [Greenwald and Rubins Reference Greenwald and Allen1999]).
Drawing on the wide ranging accumulated evidence of the effects of childhood trauma on psychopathology, epigenetics, neuroscience, and developmental researchers at the MGSS have sought to identify whether an association can be identified between early abuse and specific profiles leading to risk of suicidal behaviors. In doing so, they identify specific groups and subgroups, attempting to delineate the particular types of suicide risk profiles associated with each of them. McGill Group for Suicide Studies environmental epigenetics researchers, for instance, have been particularly interested in what they refer to as impulsive or aggressive suicide risk, while clinical and developmental researchers at the MGSS have studied how irritability might play a role in this trajectory of neurobiological risk Footnote 4 (McGirr et al. Reference McGirr2008b; Orri et al. Reference Orri2018; Zouk et al. Reference Zouk2007). In these models, the category of ECA is considered to have similar effects on a person’s neurodevelopmental profile as trauma or stress, particularly chronic stress (Isingrini et al. Reference Isingrini, Léa, Quentin, Bénédicte, Elisa, Arnaud and Garance2016; Mahar et al. Reference Mahar2014; Mahar et al. Reference Mahar2017; Turecki and Meaney Reference Turecki and Meaney2016). It takes the foreground as the earliest point in a trajectory whose extreme end point is thought to be suicide. Models of early adversity and its hypothesized biological consequences in the form of epigenetic traits imprinted on children’s brains, enable MGSS researchers to envision a critical link between early-life adversity and development of negative mental health outcomes later in life (e.g., depression, personality disorders, suicide) (Turecki and Brent Reference Turecki and Brent2016).
To establish such correlations, researchers at the MGSS conduct clinically-informed interviews to collect detailed information about the lives of people who are referred to as “individuals who died by suicide” in MGSS and other suicide research (Ahmedani et al. Reference Ahmedani, Peterson, Hu, Rossom, Lynch, Lu and Waitzfelder2017; Chen et al. Reference Chen, Daniel, Laura and Gustavo2018; Zhurov et al. Reference Zhurov2012).Footnote 5 These post-mortem, biographical surveys, also called “psychological autopsies,” are conducted with the deceased’s relatives and loved ones, and collect specific information on traumatic events and episodes, personality traits, history of mental disorders, and a variety of other factors about the lives of people who die by suicide (Brent Reference Brent1989; Cavanagh et al. Reference Cavanagh, Alan, Michael and Stephen2003). These broader retrospective early-life history interviews are complemented by the use of the Childhood Experience of Care and Abuse (CECA), a questionnaire used throughout Canada, the United States, and internationally (e.g., Busso et al. Reference Busso, McLaughlin, Stephanie, Matthew, Gold and Sheridan2017; Falgares et al. Reference Falgares, Marchetti, Manna, Musso, Oasi, Kopala-Sibley, de Santis and Verrocchio2018; Fisher et al. Reference Fisher, Craig, Paul, Morgan and Paola2011; Hart et al. Reference Hart, Lena, Mehta, Charles, Xiaohui, Gerome, Andrew, Kah and Katya2018), that quantifies adversity based on reported experiences of care and abuse (Bifulco et al. Reference Bifulco, Brown and Harris1994). The MGSS uses an adapted, proxy-based version of the CECA, in which people close to the deceased complete the questionnaire.
The CECA is based on categorical questions and a scale of four principal measures (including indifference/neglect, family tension and conflicts, physical abuse, and sexual abuse) covering the person’s childhood experiencesFootnote 6 (Bifulco et al. Reference Bifulco, Brown and Harris1994, 1422). Compared to self-reported measures for assessing childhood trauma (e.g., Bernstein and Fink Reference Bernstein and Fink1998; 2003; Greenwald and Rubin Reference Greenwald and Allen1999; Hooper et al. Reference Hooper, Patricia, Krupnick and Green2011), retrospective interview methods for appraising adverse childhood experiences such as the CECA place particular weight “on clarifying the time order of critical experiences” to encompass the breadth and timing of experiences (Bifulco et al. Reference Bifulco, Brown and Harris1994, 1422). The answers are scored and ranked according to intensity and/or severity providing what is described as “factual” measuresFootnote 7 of abuse and neglect to determine where the participant profile fits in the group’s typology of “suicide with or without abuse” (Lloyd, Filipe, and Larivée Reference Lloyd2018). Intensity and/or severity is rated on a four-point severity scale (“1-marked”, “2-moderate”, “3-some/mild”, “4-little/none”) with a global severity rating (Bifulco et al. Reference Bifulco, Moran, Baines, Bunn and Stanford2002; Brown et al. Reference Brown, Craig, Harris, Handley and Harvey2007; Bifulco and Moran Reference Bifulco and Moran1998; Moran et al. Reference Moran, Antonia, Caroline, Catherine and Kate2002). Severity levels are determined by frequency, pervasiveness and/or intensity of the experience – for sexual abuse, relationship to the perpetrator and coercion/secrecy are equally important (Smith et al. Reference Smith2002, 573). At the MGSS, researchers consider adversity as severe when reports of major physical and/or sexual abuse up to age fifteen include cases with the maximum severity ratings of 1 and 2. Histories of abuse are validated with available reports from medical charts, coroner files, or youth protection services reports (Lutz et al. Reference Lutz2017, 1186).
The abuse measured with the CECA is considered to set people on “negative life trajectories”. The degree to which adverse experiences per se serve as independent determinants of negative outcomes – as compared to the contribution of the dynamic interplay of early childhood adversities and circumstances (Boivin, Hertzman et al. Reference Boivin and Hertzman2012, 24) – remains beyond the scope of this epigenetic and neurodevelopmental research. Without being able to account for dynamic changes over the life,Footnote 8 trauma is instead studied as associated with trajectories, but trajectories with a wide range of possible outcomes. In adopting this approach, researchers focus on correlations rather than causal links. Thus, while ECA/ELA refers to specific measures and definitions of environmental circumstances or events, what these particular experiences lead to is far from clear, both within MGSS research and developmental research more generally. These studies nonetheless suggest that the experience of ECA is likely to require “significant psychological, social, or neurobiological adaptations from children who experience them” (McLaughlin Reference McLaughlin2016, 363-364), adaptations that are considered to make people, for instance, more reactive to stress. Consequently, in the MGSS models, while the effects of abuse might be modulated by events later in life that effectively move people off of negative trajectories, for a subset of the people who experience early abuse, the acquired neurobiological risk is thought to represent the origins of a diathesis that leads, for example, to impulsive or aggressive suicide, with specific neurobiological traits as the key correlating factor (Diaconu and Turecki Reference Diaconu and Gustavo2009; McGirr et al. Reference McGirr2008b). These precipitating (neurobiological) profiles or states take center stage as the origins of a person’s vulnerability to suicide risk.
Post-traumatic stress disorder: A diagnostic category with an exceptional etiological event
When the term ‘psychiatric trauma’ is used today, it is generally in connection with the malady called post-traumatic stress disorder (PTSD). The idea that underlies psychiatric trauma is that disturbing experiences can create painful, persistent, and intrusive memories. (Young Reference Young2001, 661)
Since the 1980s, much of the discussion of trauma in psychiatry has been in connection with PTSD (post-traumatic stress disorder). When PTSD entered the DSM-III in 1980, it sat in discord with the rest of the manual in several ways. One of these is the temporal nature of the disorder: its symptoms are believed to emerge from memories of past traumatic events which enact damage on a person’s present. The experience of trauma, in this perspective, refuses to be represented as part of the past, and is instead perpetually re-experienced in a painful, dissociated, traumatic present (Leys Reference Leys2000, 2). Furthermore, while the extirpation of Freud and the concept of neurosis was a central goal of the manual – which was meant to set etiology aside and focus instead on symptoms – anthropologist Allan Young has argued that an exception was made in the case of the category PTSD. This allowed traumatic neurosis, in which the memory of an event plays an etiologically-specific role in a disorder, to live on in “post-Freudian psychiatry” (Young Reference Young1995, chap. 3; Young Reference Young2001, 663).
As Young explains, this was neither an oversight of the DSM committee nor an accident. Rather, it was a response to the political imperative to construct a category that would provide veterans of the Vietnam war with a legitimate diagnosis for their psychological difficulties and enable them to access services and disability pensions. Therefore, the category had to locate the origins of their problems during their time in service – many years in the past – rather than a pre-existing condition or a post-service experience. This created several challenges for the committee and their solution was to identify the memory of an event that took place during their service years as the origin of veterans’ problems (Young Reference Young2001, 664). The delayed effects of the “etiological experience[s]” were considered to make themselves known in the present through chronic polymorphous adaptations to anxiety. According to historian and trauma specialist Ruth Leys:
the invention of PTSD had been propelled by the widespread desire to obtain formal acknowledgement of the idea that severely traumatic events could have prolonged psychological consequences in anyone, regardless of the individual’s prior history or personality. In this regard, the PTSD diagnosis necessitated a shift away from the previous approach to trauma, which concentrated on the ways in which stressful life events were mediated by the subjective interpretation of the victim. (Leys Reference Leys2006, 138)
For Leys, this meant that the previously, largely psychoanalytic emphasis on the effect of the subject’s prior experiences on the traumatic reaction, especially his or her early libidinal history, was replaced by a focus on the role of the external environment (Leys Reference Leys2006, 138).
This version of PTSD was narrow, conceived of to acknowledge how trauma – the specific atrocities soldiers experienced during the war – affected them in the present. According to Young, this is how Freud’s traumatic neurosis gained an exceptional status in DSM-III. At this time in the disorder’s history, it became:
simply taken for granted that time and causality move from the traumatic event to the other criterial features and that the event inscribes itself on the symptoms. Because the traumatic event is the cause of the syndromal feelings and behaviours, it is logical to say that it precedes them…. The discovered memory is now the explanation (post hoc) for the onset of his symptoms. (Young Reference Young1995, 115-116, emphasis in original)Footnote 9
In this “traumatic neurosis,” the specific nature of these events and the memories associated with them were important: the events were considered sufficient to cause “significant distress to anyone, anywhere” (Young Reference Young2001, 668). It was these events and their related intrusive memories that both bound together and rendered distinct the broad range of symptoms of PTSD, which overlapped substantially with other disorders such as major depression and generalized anxiety disorder (Young Reference Young2001, 677).
For the purposes of our argument, we are most concerned with several specific elements of the diagnosis PTSD and its changes over time. We will devote the remainder of this section to these elements. We will first focus on the diagnosis in terms of the depiction of the relationship between traumatic events and psychopathology in early descriptions of the category. We will then examine changing understandings of the disorder since, with particular attention to the ways in which PTSD research has come to frame the disorder as the reproduction of the stress process in which it is thought to originate. In effect, that the stressful or traumatic events – and the psychological and biological effects of these events – at the base of PTSD, are considered to produce a certain kind of person: someone whose subsequent reactivity to stress is seen as the key pathological element of the disorder. We are particularly interested in how the temporal particularity of etiologically-significant traumatic event has shifted over the years.
In the original formulation of the disorder, the occurrence of the significant traumatic experience during war experiences was essential to its diagnosis. Yet, like many psychiatric classifications, PTSD – and understandings of trauma as the etiological event responsible for the disorder – is a moving target (Hacking Reference Hacking2007). After its inclusion in DSM-III in 1980, PTSD shifted quickly both in terms of its patient profile (with corresponding changes in its diagnostic criteria) and the trauma associated with the disorder. Allan Young noted, in Reference Young2001, that “in less than a decade [after its inclusion in DSM-III], PTSD had acquired a contagious quality, manifested variously in ‘transgenerational PTSD’ … ‘vicarious PTSD,’ [or] ‘compassion fatigue’” (Figley 1995; Motta et al. Reference Bensel, Rheinberger and Radbill1997; Solomon et al. 1988; Yehuda, Beirer, et al. Reference Yehuda2000, cited in Young Reference Young2001, 662). Likewise, he suggested there had been “a corresponding extension in the effects attributed to PTSD” (Young Reference Young2001, 662). Symptoms expanded from dreams, intrusive images, and phobias, to a variety of physical ailments including lung disease, obesity, and diabetes. PTSD was often diagnosed as comorbid with a variety of other mental disorders ranging from personality disorders to schizophrenia (Pynoos et al. Reference Johnson, Patricia, Jocelyn, Elizabeth and Bernstein1999, cited in Young Reference Young2001, 662).
At the same time, the definition of the traumatic event at the heart of PTSD began to shift (Young Reference Young2001, 668). Whereas in DSM-III (1980) it was defined as an event that would cause “significant distress to anyone, anywhere” (Young Reference Young2001, 668), by the time DSM-IIIR was published in 1987, the traumatic events considered significant to PTSD shifted from being “out-of-the-ordinary” to “quite ordinary (but awful) experiences” (ibid.). While these changes are substantial, Young notes that they “simply codified what had become a convention by the end of the 1980s” (ibid.) in terms of the application of the diagnostic category. Nonetheless, the textual changes to PTSD significantly and formally broadened the range of people who could be diagnosed. PTSD diagnosis has since expanded to encompass an increasingly wide range of populations and presentations, regardless of background, history, or personal sensitivity (Summerfield Reference Summerfield2001). This potentially includes “anyone who presents a psychological affliction following an unusual event, whether he is the victim or the perpetrator” (Rechtman Reference Rechtman2002, 790, our translation).
Other post-traumatic syndromes conceptualized prior to PTSD such as the “rape trauma syndrome” (Burgess and Holstrom Reference Burgess and Holmstrom1974) or the “battered women’s syndrome” (Walker Reference Walker1984) were subsumed under the new description of the diagnosis in DSM-IIIR, despite the fact that the different syndromes had originally been described with considerable variations from the eventual definition of PTSD (Van der Kolk et al. Reference Kolk2005, 389). These studies of physically and sexually abused children, as well as of women who were exposed to prolonged interpersonal violence, played a major role in establishing an integrated, post-Vietnam War-specific approach to trauma (Herman Reference Herman1992, cited in Leys Reference Leys2000, 5). In the process, forms of trauma were conceptually merged. Concomitantly, epidemiological studies of the era suggested that exposure of the American population to “trauma-level events” might be as high as 90 percent (Breslau 1993, cited in Young Reference Young2001). Trauma and responses to trauma needed to be reconsidered in light of this, and they were. Consequently, at this point in its history, PTSD became not about a normal response to an abnormal experience, but the opposite. Research was reoriented to identifying why, in response to relatively common (if awful) experiences, some people developed PTSD while others did not (Young Reference Young2001, 672). Molecular and genetic research was seen as a potential means of answering this question through the identification of a pathogenesis distinct to PTSD, setting subjectivity, memories, and differential diagnoses aside.
Diatheses and vulnerabilities: Environmental epigenetics and the unfurling of the past in the present
It is at this point in history that narratives of child abuse and suicide, and trauma and PTSD begin to more consistently share a number of common pathways that fit into parallel transformations in other research agendas and funding interests as they concern the Developmental Origins of Health and Disease (DOHaD) (Ngo and Sheppard Reference Ngo and Allan2015; Penkler et al. 2019) as well as the later phases of ACEs research (e.g., Sun et al. Reference Sun2017; Thompson et al., Reference Thomson2015). Central to the shared reasoning in these narratives is a distinct etiological association between early trauma and a wide variety of specific forms of (psycho)pathology and (neuro)biological risk later in life.
PTSD, cortisol dysregulation, and the origins of vulnerability
Throughout the 1980s, a growing body of research focused on the molecular biology of PTSD and attempted to ascertain the specific impact of trauma on the body as a means of explaining its enduring effects on certain people (Mason 1986, cited in Young Reference Young2001, 273). Research emerging from these studies suggested that traumatic stress leaves its mark on the brain in the form of literal “eidetic” or iconic “imprints” (van der Kolk et al. 1985, in Leys Reference Leys2000, 250) and would be expected to exert profound changes over a wide range of biological structures and functions in a way that is theorized as standing “outside all ordinary stress response[s]” (van der Kolk Reference Kolk2000, 13). Cortisol emerged as a potentially significant hormone involved in the structural and functional profiles associated with the disorder and substantial research was dedicated to situating PTSD in relation to this hormone and its regulation via the hypothalamic-pituitary-adrenal (HPA) axis (Young Reference Young2001, 671-672; Young Reference Young and Rosen2004, 14). Complicated findings resulted, without clear associations emerging even as to whether PTSD was associated with hypo- or hypercortisol excretion (Young Reference Young2001, 672, 677).
An essential part of the cortisol story changed in the mid-1990s, and this is the part of the story on which we will focus. Specifically, PTSD was reconceptualized as the consequence of a diathesis, similar to models of diathesis-stress interactions involving life stressors and predispositional factors thought to be involved in the development of schizophrenia and depression (Monroe and Simmons Reference Monroe and Simons1991, 406-407). While understandings of the effects of trauma as part of a diathesis had been in circulation since the days of Charcot, it was only at this time that they became a part of the PTSD research narrative (Young Reference Young2001, 678), with eventual implications for the interpretation of a wide range of other forms of psychopathology and neurobiological risk. Cortisol remained central to the story, but it became a story of “dysregulation” rather than “down regulation.” The diathesis model, Young argues, “shifts PTSD’s centre of gravity from the traumatic event to the traumatized patient” (Young Reference Young2001, 677). The origins of his/her/their current problems, in this model, relate to an interaction between a predispositional vulnerability and a stress caused by life experiences. The “vulnerability” or “predisposition” of the veterans could have its origins in, for instance, negative early life events or specific genetic profiles. While “pre-existing” conditions were seen as a potential challenge to the legitimacy of veterans’ PTSD in 1980, by 1995 the disorder had become sufficiently anchored in American psychiatry and beyond that the diathesis model was no longer considered to undermine the credibility of veterans’ suffering (ibid., 676-7). Considerations of predisposition and vulnerability to PTSD multiplied during this period and drew on studies of pre-war weakness before and after WWI (Scott Reference Scott1993; Shephard Reference Shephard2001; Young Reference Young1999), WWII-era predictions of which soldiers would “break down” in combat based on their psychiatric histories (Grinker and Spiegel Reference Grinker and Spiegel1945; Kardiner and Spiegel Reference Kardiner and Herbert1947), and multicausal models of PTSD supported by post-Vietnam research regarding acute stress reactions and their possible origins in a wide variety of relatively ordinary yet severe stressful events (Blank 1985, in Shephard Reference Shephard and Rosen2004, 53-54). Overall, studies of prevalence, course, and comorbidity of PTSD raised questions concerning the role and magnitude of proximate stressors as primary etiological factors (Breslau and Davis Reference Breslau and Davis1987), challenging the idea that PTSD was a typical and inevitable outcome of exposure to traumatic stress (Yehuda and McFarlane Reference Yehuda and McFarlane1995, see Shephard Reference Shephard and Rosen2004, 54–55 for details).Footnote 10
Biomarkers specific to this diathesis model of PTSD vulnerability were sought in both gene-environment interaction (GEI) and epigenetic studies (Comings et al. Reference Caruth1996, 368–369, cited in Young Reference Young2001, 678). The glucocorticoid receptor, or GR, and its regulation of cortisol became central to these models of vulnerability. In the epigenetic models in particular, based on studies of rodent model organisms, early stress was seen to induce vulnerability and reactivity to future experiences of stress. “In other words, the epigenetic marks reproduce the stress process in which they originate” (Young Reference Young2012, emphasis in original). Yet despite an increasing number of molecular studies of PTSD, across broad and varied datasets, these studies have yet to offer evidence of a molecular pathogenesis distinct to pure PTSD (in its common form it is presented with comorbidities), which would enable researchers to distinguish it from other disorders with similar symptoms (Young and Breslau Reference Young, Naomi, Devon and Good2016). In the end, or at least to date, it is only an indelible or “iconic” memory that distinguishes a diagnosis of PTSD from a diagnosis with other conditions with which PTSD shares a large number of symptoms and potentially neurobiological processes (ibid., 150; see also McNally Reference McNally2009; and Weathers et al. Reference Weathers, Marx, Friedman and Schnurr2014). Consequently, diathesis-stress models of PTSD, rather than identifying the molecular specificity of PTSD, instead contributed to a portrait of neurobiological vulnerability more broadly. Specifically, these models describe processes that have become core features of contemporary psychiatric models of risk and vulnerability in which early stress or trauma, via epigenetic changes of the GR,Footnote 11 are correlated with a wide variety of forms of psychopathology (Radtke et al. Reference Radtke, Martina, Gunter, Katalin and Maggie2011; Turecki and Meaney Reference Turecki and Meaney2016; Tyrka et al. Reference Tyrka, Price, Carmen, Walters and Carpenter2012; Binder et al. Reference Binder, Bradley, Liu, Epstein and Deveau2008; Katz and Yehuda Reference Katz, Yehuda and Schein2006; Heim et al. Reference Heim, Shugart, Craighead and Nemeroff2010; Heim and Binder Reference Heim and Binder2012; Roth et al. Reference Roth2009).
To a certain extent, these emerging understandings of mental illness recall earlier critiques of the limits of classificatory systems based in categorical approaches to psychopathology, such as those found in the work of Hempel (Reference Hempel1965) and Eysenck (Reference Eysenk1964 and Reference Eysenk1967) (see Demazeux Reference Demazeux2008 for details). At the same time, they situate mental illness in ways that are consistent with orientations of the National Institutes of Mental Health’s (US) Research Domain Criteria (RDoC) initiative, in which the origins of mental illness are thought to be found in a variety of dimensional traits (domains), from cognitive to arousal/modulatory systems, that cross-cut categories in current diagnostic systems (Sanislow et al. Reference Sanislow, Michele, Jennifer, Rudorfer and Morris2019). In these studies, while genetic vulnerability remains a key etiological consideration, researchers now look for additional sources of vulnerability given that decades of research which sought to identify the “gene for” a number of common forms of mental illness met with inconclusive results. Genetic studies identified the potential and partial role of many different genes and, in some cases, the role of a very wide range of common genes carried by the majority of the population (Kendler and Donovan Reference Kendler and Donovan2014; Border et al. Reference Border, Johnson, Evans, Smolen and Berley2019; Banerjee, Morrison, and Ressler Reference Barnerjee, Morrison and Ressler2017). Continued efforts to identify origins of vulnerability have been described as the search for “missing heritability” “in which heritability estimates are significantly higher than the variance explained by known genetic variants” (Mayhew and Meyre Reference Mayhew and Meyre2017). The search has led researchers to increasingly examine the role of the environment in the development of vulnerability in the form of gene-environment interactions, but especially epigenetic mechanisms. This is the case not only for research focusing on neurobiological risk for psychiatric conditions, but also for a wide range of research agendas in the postgenomic era that focus on the plasticity of human bodies and in which nurture is increasingly seen as indissociable from nature (Meloni Reference Meloni2019; Fox-Keller Reference Fox-Keller2010). As an environmentally-informed understanding of human development has begun to eclipse pre-genomic models, attention has turned to the moments at which people are considered to be most “plastic” or susceptible to environmental influence. Youth, childhood, and in utero experiences are now seen as not only relevant but crucial targets for research. In a context where scientists must explain why someone would respond more negatively than someone else to an “ordinary” adverse experience in adulthood, the source of exceptionality (i.e., vulnerability) is increasingly sought in adversity in the early years. Attention to time and trauma shifts in these accounts, with increasingly limited consideration of the etiological significance of the iconic memories historically associated with PTSD. In the process, it is the origin of the diathesis that has become the focus of research in molecular biology including environmental epigenetics.
Diathesis and/or Complex PTSD
In parallel with studies of diatheses, the search for biomarkers specific to pure PTSD, and studies of potential molecular pathways for a wide range of psychiatric conditions thought to be associated with the experience of early adversity, another history was unfolding, that of complex PTSD, or CPTSD. First proposed by Judith Herman in Reference Herman1992, the disorder was conceived of to recognize more chronic and coercive forms of trauma. The diagnosis PTSD, Herman argues:
derives primarily from observations of survivors of relatively circumscribed traumatic events: combat, disaster, and rape. It has been suggested that this formulation fails to capture the protean sequelae of prolonged, repeated trauma. In contrast to the circumscribed traumatic event, prolonged, repeated trauma can occur only where the victim is in a state of captivity, unable to flee, and under the control of the perpetrator. … The psychological impact of subordination to coercive control may have many common features, whether that subordination occurs within the public sphere of politics or within the supposedly private (but equally political) sphere of sexual and domestic relations. (Herman Reference Herman1992, 377-8)
Complex PTSD, as formulated by Herman and later by others, was meant to account for those types of trauma that did not fit with the existing criteria for PTSD. In particular, the disorder sought to describe the pervasive effects of prolonged trauma on a person, including symptoms, character and personality, and the “survivor’s vulnerability to repeated harm, both self-inflicted and at the hands of others” (Herman Reference Herman1992, 379; van der Kolk Reference Kolk and Yehuda2001).
While research on the diathesis model of PTSD turned attention to the potential for early life factors to render someone vulnerable to mental illness in response to trauma later in life, CPTSD framed the effects of post-traumatic stress more explicitly in a developmental perspective (Roth et al. Reference Roth1997). Specifically, post-traumatic stress was described as resulting in an “impairment in developmental processes related to the growth of emotion regulation and associated skills in effective interpersonal behaviours” (Cloitre et al. Reference Cloitre, Stolbach, Herman, Bessel, Robert and Eva2009, 400). It is the experience of child abuse in particular that is considered by some researchers to be the key factor that distinguishes CPTSD from PTSD (Cloitre et al. Reference Cloitre, Garvet, Brewin, Bryant and Andreas2013, 8; van der Kolk et al. Reference Kolk2005). This explicit association between early abuse and (C)PTSD, one that had circulated in terms of possible etiological explanations of PTSD but that had not been expressly put forward as a developmental process, produced a narrative continuum between early trauma and traumatic experiences later in life. Researchers further argued that the distinction between PTSD and CPTSD provided a “conceptual coherence to the multiple, diffuse, and apparently contradictory symptoms of complex PTSD” (Cloitre et al. Reference Cloitre, Stolbach, Herman, Bessel, Robert and Eva2009, 400) through attention to vulnerability caused by and the heterogeneity of adaptation to trauma (Resick et al. Reference Resick, Boivin, Calloway, Dick and King2012). Complex PTSD was elaborated as a diagnosis alongside discussions of DTD. Complex PTSD and DTD share many features and both correlate early abuse with a wide range of psychiatric conditions as well as an array of “cognitive, language, motor, and socialization skills.” (Van der Kolk Reference Kolk2005, 404-5) Van der Kolk and others consider early abuse and its after effects to be pervasive in society, arguing that they are leading to a “silent epidemic of neurodevelopmental injuries” (Kaffman 2009, cited in Ford et al. Reference Cloitre, Garvet, Brewin, Bryant and Andreas2013). The results of these injuries, according to emerging narratives in developmental research, would include CPTSD.
Complex PTSD was not included in DSM-5 in 2013. Arguments against its inclusion involved its frequent comorbidity with PTSD and Borderline Personality Disorder (BPD) (Resick et al. Reference Resick, Boivin, Calloway, Dick and King2012).Footnote 12 However, given that that CPTSD was included in ICD-11 in 2018 (the World Health Organization’s International Statistical Classification of Diseases and Related Health Problems, 11th edition), there is reason to believe it will be included in future editions of the DSM.
Through the development and formalization of the classification CPTSD, we see the emergence of a model of trauma and its effects that is highly consistent with environmental epigenetics models of suicide wherein trauma becomes a part of a person’s future. Depending on the person and his/her/their later life experiences, this trauma might be seen as entering, shaping, and never leaving the body.Footnote 13 Moreover, this body of research frames childhood as a period in which specific types of adversities, such as neglect and peer victimization (Geoffroy Reference Geoffroy, Michel, Louise, Johanne, Perret and Gustavo2018; idem 2016), under-resourced health services (Renaud et al. Reference Renaud2014), and poor nurturing environments (Bethell et al. Reference Bethell, Carle, Hudziak, Gombojav, Powers, Wade and Braveman2017) appear endemic, placing children and youth at risk of negative developmental outcomes. The weight of claims emerging from molecular models of trauma have already resulted in shifts in the focus of ACEs research in recent years from the study of health outcomes in adult life to the study of neurobiological and psychosocial effects of abuse believed to already be tangible during childhood and adolescence (Sun et al. Reference Sun2017; Thompson et al. Reference Thomson2015). In this context, there have been calls for health policies and services (e.g., psychiatric, social work) that focus on “trauma-informed care” (Edwards, Gillies, and White Reference Edwards, Val and Sue2019) to promote “resilience” in children through interventions that aim to mitigate the potentially durable, negative effects of ECA and prevent them from entering stable, long-term negative trajectories (Davidson and Carlin Reference Davidson and Eric2019; Larkin, Felitti, and Anda Reference Larkin2014; Logan-Greene et al. Reference Logan-Greene2014; Lang et al. Reference Lang2019; Bethell et al. Reference Bethell, Gombojav, Solloway and Wissow2016; Ranjbar and Erb Reference Ranjbar and Matt2019).
Constructing the etiology of suicide
The environment epigenetics models of suicide risk at the MGSS emerged from the same diathesis-stress models that were at the base of PTSD, and later CPTSD, research. The origins of the MGSS epigenetics research program, for instance, can be found in the same rodent models of stress and epigenetic modifications of the GR that informed PTSD diathesis models, models that have had an immense impact on contemporary environmental epigenetics research. Researchers at the MGSS were the first scientists to translate these exceptionally influential studies on the effects of early adversity on stress reactivity later in life, carried out on model organisms (Meaney Reference Meaney2001; McGowan et al. Reference McGowan2009; Weaver et al. Reference Weaver, Nadia, Champagne, D’Alessio and Shakti2004), into human populations in the form of studies of people who experienced early abuse and who died by suicide later in life.Footnote 14
Researchers at the MGSS argue that ECA leads to specific epigenetic profiles, at least some of which are sufficiently stable to be found in the brains of people who die by suicide (Labonté et al. Reference Labonté2012; Lutz and Turecki Reference Lutz and Gustavo2014; Lutz et al. Reference Lutz2015). They consider these epigenetic traits, identified in postmortem brain tissue, to constitute a molecular basis for suicide risk. In the words of Gustavo Turecki, ECA “lead[s] to the development of these maladaptive trajectories. … So, suicide risk is, perhaps the most severe negative end point of those psychopathological conditions that are in turn predicted by these negative trajectories.”Footnote 15
While their highly influential research is oriented toward the identification of a model for suicide risk, we suggest that their primary contribution to contemporary narratives of neurobiological risk is different. The core of their research program focuses on ascertaining the biological nature of the origins of the diathesis and identifying the biological changes during childhood that are thought to occur in response to early abuse (e.g., Brezo et al. Reference Brezo, Alexandre, Chantal, Valérie, Barker, Frank, Martine, René, Tremblay and Gustavo2010; Diaconu and Turecki Reference Diaconu and Gustavo2009; McGirr and Turecki Reference McGirr and Gustavo2007; Turecki Reference Turecki2014a). So, if PTSD researchers reoriented attention from the traumatic event itself – initially seen as abnormal – toward a person’s abnormal reaction to a normal traumatic event by recourse to a diathesis-stress model, the MGSS researchers are reorienting attention toward characterizing the origins of this vulnerability.
Within this research program, a specific story of the lives of people who die by suicide is beginning to emerge, in which negative early life events set people on pathological life trajectories whose extreme end point is suicide. This conceptualization of suicide as the product of a diathesis is stabilized through recourse to temporal arguments about the presumed indelibility of early experiences on the body. Suicide in this view is bookended by two precipitating events: one in the past and (probably) “abnormal,” the other more recent and “normal.”
The biological embedding of time
Epigenetic traits are generally described as unstable precisely because they are considered responsive to environmental cues. As Turecki and the highly influential epigenetics researcher Michael Meaney acknowledge, recent evidence in epigenetic research supports the hypothesis that “epigenetic plasticity is sustained in the brain throughout adulthood, potentially as a mechanism to cope with the evolving demands of the environment, yet there are clear moments during development when plasticity is heightened, and these may be more strongly associated with the establishment of life-long epigenetic modifications” (Auger and Auger 2013, in Turecki and Meaney Reference Turecki and Meaney2016, 7). Researchers at the MGSS explain the enduring effects of ECA in terms of, first, “critical windows of neuroplasticity” and, second, severity of abuse, as measured with the CECA (Lutz et al. Reference Lutz2017; Turecki Reference Turecki, Spengler and Binder2016, 168; Fiori and Turecki Reference Fiori and Turecki2016). In the first case, events that occur earlier in life are considered to have more significant consequences for subsequent development than events that occur later in life (Turecki Reference Turecki2014b, 144; Turecki et al. Reference Turecki2012).Footnote 16 Furthermore, the early abuse is considered to correspond to the timeframe of the stabilization of a specific epigenetic trait, CH methylation, in the first fifteen years of life. This form of methylation is thought to be present in higher proportions in the brain than other parts of the body (Barnett-Burns et al. Reference Barnett-Burns, Szyszkowicz, Luheshi, Lutz and Turecki2018; Lutz et al. Reference Lutz2018; Lutz et al. Reference Lutz2017, 1186). Thus, MGSS researchers expect that the biological consequences of an experience during the first fifteen years of life, if significant enough to affect CH methylation, might be of greater significance and potentially more durable given that it stabilizes relatively early in life, as compared to other epigenetic traits (Lutz et al. Reference Lutz2018). Second, the severity of the abuse is considered sufficient to unleash biological consequences for anyone (Brezo and al. 2008). This is how MGSS researchers describe the origin of the diathesis, or developmental processes, underlying suicide risk – how one moment in time, a traumatic experience, becomes indelibly impressed on a person with enduring consequences for the rest of that person’s life.
A life of various psychopathologies and personality traits (e.g., impulsivity, bad decision-making) is considered to follow early abuse and methylation: in these models, epigenetic marks reproduce the stress process in which they are thought to originate, as in the PTSD research Young studied (Young Reference Young2012). In MGSS models, the remainder of a person’s life and the events preceding his/her/their later death by suicide appears to carry less weight than early traumatic experiences in terms of the suicide risk they confer. MGSS researchers consider the negative events proximate to suicidal behavior to be common problems (e.g., loss of a job, interpersonal conflict leading to rejection and isolation, stressful life events) (Jollant et al. Reference Jollant, Nathalia, Emilie, Sébastien and Philippe2011, 320; Turecki Reference Turecki2014a, 9; Turecki Reference Turecki2014b, S145; Zouk et al. Reference Zouk2006, 200, 202Footnote 17). In this vision of time, trauma, and psychopathology, produced in the afterlife of a person who died by suicide, the events surrounding his or her death are considered almost an afterthought when compared to early life events.
Many questions remain. In the epigenetics studies of suicide, ultimately researchers do not know when these epigenetic traits or vulnerabilities were established – the epigenetic profiles identified in post-mortem brains could reflect a change near the time of death – but this is not the narrative that has been stabilized. In the PTSD studies, if researchers initially had to justify the persistence of a memory against research that showed memories to be highly malleable processes rather than “timeless photos,” epigenetics researchers have to explain the apparent formation and persistence of what might otherwise be a highly changeable biological trait. These parts of the puzzle have yet to be answered.
The correlation of ECA/trauma first to an epigenetic trait and, second, to a variety of “negative mental health outcomes” could be described as the reduction of the complex or multiple to the simple or singular. Yet what we’re interested in, beyond reductionisms or neoreductionismsFootnote 18 (Lock Reference Lock2015), is what is elaborated in these studies: the stabilization of a particular narrative of time, trauma, and psychopathology. This is a story of recurrence, of the re-enactment of an original stress. It is about the nature of what stays and affects the rest of a person’s life. It is a story about which experiences in our lives are considered significant.
Yet this is not a story that is specific to suicide or even the broader narratives of neurobiological risk at the MGSS. It is one of many incremental contributions that have led to current understandings of neurobiological risk and risk of ill health more generally. While much attention has been given to narratives of plasticity and vulnerability as characteristic of the postgenomic era, we suggest that these narratives of vulnerability can be traced back further in time. One precursor to current understandings of plasticity can be found in narratives of PTSD and, particularly, shifts in attention from the traumatized soldier to the vulnerable soldier. This narrative shift can now be seen as coherent with more recent research trends including, for example, the RDoC initiative, which in many respects represents a return to etiology in biopsychiatry after it had been banished from official psychiatric nosology only forty years ago. Moreover, this is a return to specific etiological narrative, one that traces significant etiological events to youth, reconfiguring how we account for their effects and people’s subsequent subjective experiences.
The MGSS environmental epigenetics research on suicide risk and recent research on PTSD focusing on the origins of the diathesis are producing new narratives of vulnerability. In current narratives, with a focus on the GR and other biomarkers, a new timeline – a new sense of time – is created: a biologically embedded time. What, in other models, might have been considered indelible memories of early childhood abuse (Janet Reference Janet1889; Breuer and Freud Reference Breuer and Sigmund1893) become past traumas stamped onto post-mitotic cells in a child’s brain. This biologically embedded time is thought to live on in a person, recreating the stress out of which it was born, explaining why a person might respond “abnormally” later in life to a “normal” traumatic event. New etiologically significant events are created. And along with them, universal vulnerability – given a particular set of early negative environmental factors – to these conditions and behaviors.
It is in this way that, as in the case of PTSD, models of suicide provide insights into psychiatric reasoning about “the past invading the present” in the form of past abuse, diatheses, and development (Young Reference Young1995, 7; Young Reference Young2006). Narratives of the people who die by suicide allow molecular traits to be associated with specific moments in time, and further, allow them to be seen as indelible when they might otherwise be seen as fleeting. Ultimately, both research programs are ever-changing stories of time, trauma, and how we come to understand trajectories, futures, through people’s pasts, even if, in the case of suicide, these are constructed in their afterlives.
When suicide no longer has a significant precipitating event
In these models, the scope of vulnerability changes. Researchers have moved beyond an interest in the exceptionality of proximate traumatic or stressful events (and their memories) as the cause of present disorders or behaviors, and (largely) beyond gene-environment interactions as a source of vulnerability in which case it would only be a certain, stable, and limited subpopulation that would be at risk (i.e., people born with a genetic predisposition). The meaning of wartime traumatic events involved in the original formulation of PTSD shift from exceptional to unexceptional, as predispositions and vulnerability established in youth become the key etiological factors that allow normal negative events later in life to be seen as sufficient to trigger pathological stress reactions. In environmental epigenetics models of suicide risk, potential vulnerability becomes universal: given the right (which is to say wrong) environment or traumatic experience early in life, anyone can be placed on an “unhealthy trajectory.”
The temporal nature of risk and vulnerability in these models changes as well. The important event is not something later in life or closer to death, but early in childhood. As a result, perceived windows of therapeutic opportunity, origins of vulnerability, and evaluations of risk change. In some areas of research, these essential windows roll back even further than childhood to perinatal or in utero experiences and the preconception period. In such a context, the answer to the question “When is the beginning?” (Pentecost and Meloni Reference Pentecost and Maurizio2020, 8) reveals an increasing schism between biographic and biological perspectives (Lloyd and Larivée Reference Lloyd and Alexandreforthcoming). Pathology comes to be seen as the result of not only experiences during one’s own lifespan, but also inter- and transgenerational life experiences, conceived of as able to shape biological traits and personal trajectories.
For the remainder of this paper we will focus on the ways in which shifting understandings of vulnerability and risk have resulted in historically specific neurobiological models of suicide. Through these models, we have come to a point where suicide can be studied as an action without a significant precipitating event, only a significant precipitating biochemical state as a result of early-acquired neurobiological risk. This neurobiological state is considered to be the key precipitating factor for suicide. The common negative experiences that often precede suicidal acts are reconceived as triggers (Jollant et al. Reference Jollant, Nathalia, Emilie, Sébastien and Philippe2011; Turecki 2014a, idem Reference Turecki2014b; Zouk et al. Reference Zouk2006). According to MGSS neuroscience researcher and psychiatrist Fabrice Jollant:
The triggering of the suicidal crisis often has external causes. Environmental events are frequently put forward in notes left by people who die by suicide or reported by suicide attempters themselves (Heikkinen et al. Reference Heikkinen, Hillevi and Jouko1994). However, these reasons are in most cases very common problems including marital difficulties and separation or job problems and loss (interestingly, these reasons may be universal as shown by anthropological studies in pre-industrialized populations [MacDonald Reference Macdonald2006]). In addition, suicidal acts may sometimes take place without any clear external triggers. Therefore, these environmental events are apparently not sufficient to totally explain the occurrence of a suicidal crisis. As external triggers of suicidal acts are often social by nature – they either involve other people (e.g., difficulty with one’s partner) or one’s place in society (e.g., loss of social status) – individual differences in the way people experience social relationships, perceive oneself in society and respond to the social environment should be central to understanding the suicidal process. (Jollant et al. Reference Jollant, Nathalia, Emilie, Sébastien and Philippe2011, 320, emphasis added)
Thus, while devoid of exceptional qualities, triggers are considered sufficient to set in motion affective responses and suicidal behavior among individuals who are vulnerable to pathological stress reactions.
The absence of a significant precipitating event in these understandings of suicide, replaced by triggers and affective states, raises questions about the cognitive and emotional aspects of people’s suicidal behaviors. Ruth Leys is among a number of social scientists and humanities researchers who have critically analyzed neuroscientific explanations of affect as a neurobiological state (e.g., Blackman and Venn Reference Blackman and Couze2010; Connolly Reference Connolly2002; Hemmings Reference Hemmings2005; Massumi Reference Massumi2002; Papoulias and Callard Reference Papoulias and Felicity2010; Thrift Reference Thrift2004). She argues that this framing of affect has significant implications for the ways in which people’s actions are understood:
there is a gap [in these explanations] between the subject’s affects and its cognition or appraisal of the affective situation or object, such that cognition or thinking comes “too late” for reasons, beliefs, intentions, and meanings to play the role in action and behaviour usually accorded to them. The result is that action and behavior are held to be determined by affective dispositions that are independent of consciousness and the mind’s control. (Leys Reference Leys2011a, 443)
Leys suggests that these models have the effect of replacing “the idea of one’s intentionFootnote 19 with regard to objects or of the meanings those objects might have” with the idea of affect as triggered neurobiological states “operating outside the domain of consciousness and intentional action” (ibid., 465). These models, she writes, “posit a constitutive disjunction between our emotions on the one hand and our knowledge of what causes and maintains them on the other, because according to them affect and cognition are two separate systems” (ibid., 437). In other words, affects are only contingently related to objects in the world, defined in non-intentional terms: “they operate blindly,” Leys adds, “because they have no inherent knowledge of, or relation to, the objects or situations that trigger them” (ibid.). This understanding of affect within the neurosciences is thus often aligned with models of self-organizing networks and bodily systems as processual and dynamic entities that attribute the origins of action to “the neural infrastructure of consciousness and not to our experience of the lived present” (Papoulias and Callard Reference Papoulias and Felicity2010, 47, emphasis in the original).
Debates over the nature and limits of intentionality and non-intentionality of affect thus speak to questions of whether affect requires (or not) a subject in order to understand or produce its effects (Blackman Reference Blackman2012, 12). This has a direct bearing on the extent to which personal experiences across time, social contexts, and relationships, are seen as tightly bound-up with human meaning-making (Dragojlovic Reference Dragojlovic2015; Walkerdine Reference Walkerdine2010) and associated with neurobiological processes (Blackman Reference Blackman2010; Clough Reference Clough2009).
The intersection between these debates and contemporary reasoning about a wide range of psychopathologies – including PTSD and the forms of mental illness considered associated with suicidal behavior, but extending well beyond them – are held together through narratives of diatheses and early trauma. Within these models, once an early neurobiological profile has been established, subsequent decision-making is seen as relatively impaired, with greater reactivity toward normal negative events, which are now conceived of as triggers. The content of these events is seen as relatively unimportant. In the case of suicide, this conceptualization of triggers speaks directly to the role of precipitating events, reasons for suicidal behavior, and the meaning of suicidal acts.
A number of explanations of suicide are incompatible with these models of vulnerability. Suicides of resolve, in which a person makes a reasoned decision to commit suicide, as documented by anthropologist Junko Kitanaka do not fit (Kitanaka Reference Kitanaka2011, see chap. 7). Kitanaka describes a process in which suicide came to be interpreted, if not as a result of specific neurobiological risk, almost exclusively as a result of mental illness, the result of impaired reasoning. Suicide as “way of belonging,” when, in regions characterized by high rates of suicide, all you have left is your connection to the dead is equally incompatible with emerging models of vulnerability that focus on early events and predispositions rather than temporally proximate reasons for suicidal behavior (Niezen Reference Niezen2009). Suicide as a shared experience under the sway of personal ties, regimes of desires, imagination and remembrance, and daily struggle with life and death, as documented by anthropologist Lisa Stevenson based on her experiences listening to suicidal youth in Nunavut, is an action that is not easily understood as set off only by a trigger. Suicide, in this instance, is seen as being deeply a “part of their everyday fabric of life” (Stevenson Reference Stevenson2014, 9). Suicide as a protest to political or economic upheaval (Widger Reference Widger2015; Imberton Reference Imberton2012) sits uneasily with environmental epigenetics explanations of suicidal behavior, in which precipitating chemical states replace precipitating social, economic, and political circumstances. Suicide bombers, as examples of late twentieth century reframings of suicidal acts as “weapons” (Hacking Reference Hacking2008), further complicate and distribute intentionality into the realms of social and political violence and imagination in which notions of class, gender, and ethnicity become intertwined (Brunner Reference Brunner2007; Jaworski Reference Jaworski2010; Victor Reference Victor2003). These contextualized experiences are pushed to the side when suicide is no longer seen as having a significant precipitating event or being the result of a rational decision. When neurobiological risk established in youth takes the foreground, other factors considered associated with distress and suicidal acts later in life recede to the background. Viewed from this perspective, the MGSS account of suicidal behavior and the wide-ranging neuroscientific literature that draws on and develops similar narratives of neurobiological vulnerability appear to sideline many of the considerations that are central to other accounts of suicide.
As sociologist of the life sciences Maurizio Meloni argues contemporary bioscientists’ task “is not so much to map whether a disease is the product of gene or environment, or an interaction of the two, but how environmental exposures in critical windows of plasticity are biologically embedded and reproduced through a mixture of bodies-biohabits-culture” (Warin et al. 2015, cited in Meloni Reference Meloni2018, 6). Currently, in environmental epigenetics research on suicide, some of these windows receive more attention than others. Understandings of youth as a period of elevated vulnerability to the impact of negative events is stabilized in MGSS research through recourse to psychological tests and neural markers that act as proxies for proof of childhood trauma. Identification of this trauma is the basis of an explanatory model in which biomarkers identified in molecular research are understood as embodied evidence of abuse in the brains of people who have died by suicide or who are considered at risk of suicide. These preliminary hypotheses, which can also be seen as proxies, about how and under what circumstances negative experiences become part of one’s cerebral constitution guide scientific models and, increasingly, priorities for intervention (White and Wastell Reference White and David2017; Gillies, Edwards, and Horsley Reference Gillies, Rosalind and Nicola2016). While proxies might be understood as interim solutions that “carry us through a period of ignorance until we find the underlying causes” (Duster Reference Duster2005, 1050), as sociologist Troy Duster points out, these proxies can have powerful consequences. While new scientific findings and narratives will ultimately emerge and amend or displace prevailing hypotheses, Duster warns that a proxy can be “such a dominant category in the cognitive field that the ‘interim solution’ can leave its own indelible mark once given even the temporary imprimatur of scientific legitimacy by molecular genetics” (ibid.). In this sense, proxies play powerful roles in scientific narratives and must be considered more than simply place-holders. While details of the MGSS model shift over time, perhaps the durable impact of this narrative of neurobiological risk, along with those widely circulating in other areas of research, might most specifically concern time. When the diagnosis PTSD was initially described in 1980, it was “simply taken for granted that time and causality move from the traumatic event to the other criterial features and that the event inscribes itself on the symptoms. Because the traumatic event is the cause of the syndromal feelings and behaviours, it is logical to say that it precedes them…. The discovered memory is now the explanation (post hoc) for the onset of his symptoms” (Young Reference Young1995, 115-116). If many elements of the diagnosis have changed since 1980, this conceptualization of time, trauma, and causality have not. While the relevant time at which the pertinent trauma is thought to have occurred has shifted from adulthood to childhood, time and causality are considered to move just as surely from the traumatic event to the symptoms. This is how trauma at one point in time has come to be seen as having specific etiological consequences. Through the logic of narratives of plasticity and critical windows of neuroplasticity, this notion of how trauma leads to symptoms has taken a place at the heart of neuroscientific and biomedical reasoning more broadly.
Media and cultural studies specialist Lisa Blackman (Reference Blackman2012) has argued that what lays in the background of concerns about neuroscientific models of psychopathology is a history in which, for the most part, psychologists and neuroscientists consigned threshold experiences and “immaterial process” to the category of pathology or viewed them as irrational (Blackman Reference Blackman2012, see chaps. 7 and 8). For Blackman, though, this does not necessarily reflect an attempt to dismiss the meaning or intentionality associated with these experiences. Indeed, iterative reinterpretations of experience have allowed assumptions about volition, consciousness, and intention to be actualized and transformed in different ways (Blackman Reference Blackman2014). The neuroscience and epigenetics research we discuss in this article, then, might be perceived not as necessarily implying the absence of a relationship between “the objects or situations that trigger them” and psychological subjects “that elicit them” (Frank and Wilson Reference Frank and Wilson2012, 874-75), but rather as providing an historically particular way of understanding the relations between affects and life. From such a perspective, these etiological explanations might be seen as an addition to rather than an erasure of preexisting and complementary understandings of experiences and their effects.
Indeed, at other research units and in parallel research activities at the MGSS, researchers draw attention to the effects of personal experiences across the lifespan (Nemeroff Reference Nemeroff2016; Provençal and Binder Reference Provençal and Binder2015; Roberts et al. Reference Roberts2018).Footnote 20 These models speak to flexible and complex timelines of the development of specific forms of subjectivity that might lead to suicide risk. This research intersects with broader neuroscientific visions of life concerned with the ways in which personal experiences articulate with events and traces of the past, while also being a part of a person’s future and ongoing day-to-day life (Blackman Reference Blackman2014, 7; Lloyd and Larivée Reference Lloyd and AlexandreForthcoming). In these models, clear boundaries between “pre” and “post” significant etiological experiences, for example, become difficult to identify and render unproductive attempts to either limit or rescue intentionality from one of these two poles. From this perspective, an approach that only focuses on “pre” or “post-abuse” would be a limited conception of life experiences, reproducing the idea that there might be a developmental point before which biology is uncompromised or brought under the sway of affects. Consequently, rather than trying to identify a point at which people – materially and experientially – might be considered to have a pure and unadulterated form of intentionality, this approach instead considers all subjectivities ‘in the post’ and as possible sites of entanglements (Wilson Reference Wilson2015, 39, 164). Interest in characterizing and understanding life as post, in which life is seen as shaped by early traumatic events, is, then, not only a concern of contemporary neuroscientists, but also of social science and humanities researchers who attempt to understand and document the experiences of people in conditions ranging from conflict or post-conflict environments to natural or human instigated disasters or persistent toxic exposures and the effects of these experiences on their lives and those of their families or descendants (Lester Reference Lester2013; Kirmayer et al. Reference Kirmayer, Hanna, Abel Hamid, Duncan, Morgan and Bhugra2010; Moghnieh Reference Moghnieh2017; Roberts Reference Roberts2017; Yamaguchi Reference Yamaguchi2018). These accounts depict intentionality or volition as fluid states always anchored in personal trajectories within specific environments eschewing explanations that permit experiences at one moment in life (i.e., childhood) casting all others into the etiological and experiential shadows. This leaves a fertile space within which one might be able to consider suicide as both an act with a significant precipitating biochemical state and a significant precipitating event.
Acknowledgments
This research was Canadian Institutes of Health Research (CIHR) Project Grant (166036) “Situating Suicide Risk: An inquiry into the production of the lives and afterlives of neurobiological vulnerability”.
Stephanie Lloyd is an Associate Professor in the Department of Anthropology at Université Laval whose research sits at the intersection of medical anthropology and science and technology studies. Through two main research axes, her work explores (1) narratives emerging from neuroscience and epigenetics research on trajectories of neurobiological risk whose extreme end point is considered suicide, and (2) perceptions and experiences of hearing through cochlear implants (CIs), from the perspective of neuroscience and audiology researchers and CI users. She has carried out research at the McGill Group for Suicide Studies, where she is a principal investigator, since 2013, including the CIHR-funded project “Situating Suicide Risk”. Her research has been funded by the Social Sciences and Humanities Research Council and the Canadian Institutes of Health Research, among other funding bodies. She is co-editor of The Palgrave Handbook of Biology and Society.
Alexandre Larivée is a doctoral candidate in the Department of Anthropology at Université Laval and a graduate research assistant in CIHR-funded project “Situating Suicide Risk.” His doctoral research is the basis of one axis of the larger project and explores the ways in which explanations of depressive disorder and suicide risk are enrolled at the intersection of neuroscience and epigenetics and clinical and psychiatric care.
