This post is by Alex Oaten, the Lincoln based Research Associate on the Suicide in/as Politics project. You can find out more about all the team on our People page!
Our reading group recently met to discuss a 2020 article authored by Heidi Hjelmeland and Birthe Loa Knizek entitled ‘The emperor’s new clothes? A critical look at the interpersonal theory of suicide’. In the article, the authors are highly critical of the work of leading suicidologist Thomas Joiner and his Interpersonal Theory of Suicide, arguing that Joiner’s predictive theory is “reductionist and decontextual” and raising serious concerns about the way in which it has been “so uncritically … embraced by a scientific audience” (2020, p.176).
This criticism of Joiner’s theory served to draw my attention to the role that predictive knowledge plays within contemporary suicidology and I wanted to think about the wider contextual landscape in which Joiner and his colleagues operates within. In order to do this, I turned to the work of French philosopher Michel Foucault who developed the concept of governmentality, which he understood as “techniques and procedures for directing human behavior. Government of children, government of souls and consciences, government of a household, of a state, or of oneself” (Foucault 1997, p.82). Governmentality is the efficient management of individual and group conduct.
Developed out of the eighteenth-century desire to govern the collective subjects of a particular territory governmentality “put the notion of population and the mechanisms for ensuring its regulation at the center of its concerns” (Foucault cited in Tierney, 2010, p.361). Governmentality draws our attention to the complex of actions and practices that are aimed at regulating and directing individuals and populations, and this regulation requires knowledge, because knowledge produces the ‘truths’ about that which we seek to govern.
Governmentality: The management of Suicide
To focus on the governmentality of suicide is to consider how the problem of suicide is addressed through technologies and regimes of knowledge and how these are formulated in order to manage and govern the problem of suicide within the population. Suicidology as a discipline, with its academic research and expert analyses, forms a central part of the governmentality of suicide.
Foucault recognised that suicide presented a disconcerting challenge to the modern administration of the life of the population with suicide becoming one of the first phenomenon to be studied by the new discipline of sociology. Durkheim published his classic text Le Suicide in 1897. Durkheim’s sociological intervention on suicide was an attempt to establish sociology’s status as a discipline capable of generating useful knowledge that would contribute to the effective administration of life and enable more efficient population management (Tierney, 2010, p.377). This pressure to generate knowledge of suicide which is ‘useful’ for the governmental administration of the life of the population has not diminished since Durkheim was writing. Such knowledge is now produced within a specific ‘psy-centred’ knowledge regime which, as Ian Marsh notes, understands suicide “as primarily an individual act of pathology… [with] the social and political context of its arising…taken to be of secondary importance” (2010, p.221).
By focusing on the individual and individual psychology much modern suicidology research aims at the prediction of suicide. However, such a focus results in the deletion of the contextualised individual, who is replaced with abstracted and decontextualised risk factors and behaviours. Or, as Hjelmeland and Knizek point out in their criticism of Joiner’s theory, individuals become “stripped of their contextual biography” (2020, p.169) in the service of a universal predictability which is seen as essential for the effective management of suicide rates within the population.
Joiner’s model: The Popularity of Prediction
Joiner’s model seeks to predict and explain the processes which lead individuals to engage in suicidal behaviour; stating that “an individual will engage in serious suicidal behavior if he or she has both the desire to die and the capability to act on that desire” (Ribeiro and Joiner, 2009, p.1291). Despite Joiner himself acknowledging that “the proposition may seem somewhat obvious at first” (Ribeiro and Joiner, 2009, p.1291), it has become one of the leading models within mainstream suicidology, with its predictive powers considered a key strength.
Joiner’s theoretical model has met with widespread acclaim because the governmentality of suicide requires that suicide be constructed as regular, patterned and predictable. We want to try and prevent suicides and, in order to feel empowered to do so, we want to think that there is some patterned regularity or certainty to suicide. Without certainty we feel much less able to manage and control a problem.
Hjelmeland and Knizek strongly criticise Joiner’s conviction that “suicide is a phenomenon with a universal explanation” existing independent of contextual factors (2020, p.169). I agree with their criticism, but we also need to explicitly recognise that the reason that Joiner is so focused on universal explanation and less interested in contextual specificity and complexity is because prediction is seen as essential to our collective ability to govern and manage populations. So, we like universal explanations that enable predictions because they make us feel in control; such knowledge enables strategies, plans and techniques to be developed and implemented.
Prediction is comforting, whereas an explicit acknowledgement of the contextually complex, inherently uncertain, and often contradictory nature of suicide would undermine our confidence in our ability to really ‘know’ suicide as a phenomenon, and thus our prowess to successfully control and reduce it.
Diminishing Complexity
Joiner’s theoretical model:
In order to make this prediction testable and thus demonstrate its usefulness, Joiner and his colleagues created the Interpersonal Needs Questionnaire (INQ). This is a questionnaire based upon 15 statements scored on a scale of 1 (‘not true at all for me’) to 7 (‘very true for me’). The contextual complexity of suicide is thus reduced to a set of universal statements. Hjelmeland and Knizek argue that the INQ test is far too unspecific to have value. Whatever we may think about the INQ, and opinions in our reading group were divided, I think that we need to understand that the value that the INQ and Joiner’s model have is that they make suicide seem knowable, explainable and thus predictable.
In contrast to Joiner’s theory, Smedslund has argued that “the openness of the person makes it virtually impossible to conceive of any given behavioural measure as being an invariant function of only a limited number of determinants” (2009, p.785). Smedslund’s position, which would encourage us to recognise the inherent contextual complexity of suicide, would not assist the efficient governmentality of the phenomenon. Joiner’s success is that he tries to make suicide predictable. By doing so he may simplify and decontextualise, but that is the only way that he can claim universal prediction and as I have noted above, it is that kind of knowledge which is deemed most useful for the efficient governmentality of suicide. Joiner is comforting, he makes us feel like we can know suicide and that we have the tools to control and manage it.
One negative consequence of this pursuit of prediction is that, as Hjelmeland and Knizek argue, Joiner’s model results in the contextual biographical individual being marginalised, complexity lost in the service of a model that seeks to maintain our grip on suicide as knowable, predictable and manageable. But this is the landscape in which Joiner and his colleagues are operating within, a landscape that prioritises certain types of knowledge as more or less useful.
Just as Durkheim sought to study suicide in order to demonstrate the usefulness of sociology, so contemporary models like Joiner’s primarily focus on demonstrating usefulness rather than recognising the complexity of suicide. The risk here is that the model becomes more important than the phenomenon being studied. Such a model centric approach leads to “the telescope becom[ing] greater than the sky” and individuals becoming detached from their social structures.
