“Violent Possible”: The Stochasticity of Institutional Violence

1 Introduction

“The very substance of violent action is ruled by the means-end category,

whose chief characteristic, if applied to human affairs,

has always been that the end is in danger of being overwhelmed

by the means which it justifies…

Since the end of human action can never be reliably predicted,

the means used to achieve political goals

are more often than not of greater relevance to the future world…”

Hannah Arendt.

The close-up on the rusted imprint of a bullet that pierced, to fulfill its homicidal purpose, a blue door – evocative of the sunniest skies of Bogotá or Málaga – resembling those of a 1960s Ford abandoned outdoors, constitutes the powerful metonymy of the cover of Bernstein’s (2013) book. This image revives the emotional jolt of discovering, in an old wound on the skin of the upper part of the shoulder blade of a young Nukak-Makú woman from the Colombian Guaviare jungles, the vestige of a bullet likely impacting her as a child, which over the years scarred as a protrusion similar to the largest lesions caused by measles. Now, in these turbulent times, rife with violences that once again the bells toll, both shots reveal the ultimate outcome of civilized institutions: their immense capacity to exert control over everything (people, animals, and things) using any type of violence, subtle or forceful, always armed, whether with bullets or signs.

Reading Bernstein’s preface to his book (2013) obviates further elaboration, as it already highlights the institutional capacity of states – tested by the COVID-19 pandemic – to wield their intimidating force – a generator of legitimacy and institutionalizer of “violent possibles” – among whose probable outcomes is the activation of global war (against a global enemy like the pandemic, or among numerous localized groups, independent of states; although they serve it, its remains unimaginable). This war already exhibits its stochasticity. Hence, the author invokes a phrase attributed to Hannah Arendt, “thinking without banisters,” as a guiding principle for the governability of the violent.

As a corollary, this article is structured as follows: this introduction and a general conceptualization of stochasticity and institutional violence; three sections focusing on violence, the violent, stochasticity, the stochastic, and their interconnections; and concluding remarks on the stochasticity of institutional violence. Thus, the stochastic quality of violent behavior – stochasticity – opens the possibility of dynamically understanding institutional violence, while simultaneously outlining a critique of the notion of structural violence.

2 General Conceptualization

“[…] I saw millions of delightful and horrible acts;

none amazed me so much as the fact that they all occupied the same point,

without superposition and without transparency.

What my eyes saw was simultaneous; what I shall write is successive,

because language is successive.”

Borges (1999)

“A stochastic process is the mathematical abstraction of an empirical process whose development is governed by probabilistic laws” (Doob, 1990, p. v), given that an empirical process can be understood as the successive convergence of observable or verifiable events over time. Stochasticity would be the evolutionary, random, and irreversible quality of concrete events within the process, regardless of whether they are mathematically abstracted or modeled. This idea recalls Heraclitus’s metaphor regarding the impossibility of bathing twice in the same river. Thus, stochasticity would reflect the inherent uncertainty and unpredictability of many real-world events that exhibit randomness and are susceptible to being treated – in addition to mathematically – ethnographically.

Since this work addresses concrete institutional violences, not only those of the state but in the broad sense afforded by anthropology (Hodgson, 2006), the central point is the possibility of ethnographically describing the empirical evolution, the stochasticity, present in the subtle behaviors of institutional reality, so that violence can be predicted or contained.

Stochasticity has been employed in biological anthropology and evolutionary archaeology. In the latter, not without reluctance, for “most archaeologists shy away from explaining…cultural changes as random”; researchers such as Binford, Braun, Cavalli-Sforza, Dunnell, and Feldman admit, however, the importance of stochastic processes for paleontologically explaining the evolution of human institutions (Nicholson & Sibani, 2015; Powers et al., 2016). Their work has stimulated the boldness of these prolegomena on stochasticity (Billiard & Alvergne, 2018; Sahay, 2024; Wolpert et al., 2024), and their critical possibilities regarding the linguistic violence of contemporary “stochastic parrots” (Bender et al., 2021), as some authors call chatbots.

The study of violences – in the plural – should consider the stochasticity of violent phenomena, as it is an objective quality of each and every one of them, because violences – regardless of content (magnitude) and form (organization) – are usually the effect of a succession of probable, interconnected, and random events that at a precise moment converge and erupt (with or without control) in an aggressive manner (Wolpert et al., 2024). The drift from the accumulation of tensions in triggering episodes towards aggressive actions is presented here as the – unexpected – transition from “violent possibles” to “violent acts”; therefore, with the aim of supporting the unpredictability of the violent, this perspective challenges deterministic ideas that seek a single cause or a fixed set of them as inevitable generators of violences.

Apparently, they emerge not because of human ineffability, but because they are a possibility within a spectrum of probabilities resulting from the interaction between different levels of reality (local, national, or global; material, symbolic, or institutional, etc.). All violence is exerted with force to obtain some change in the regimes of coexistence, and all change can affect some more than others, even if they are supporters of the same interest.

Each social event contains its own “violent possibles,” probabilities of concretion, and its own stochasticity and potential interactions with other factors present in reality, either independently or correlatively; therefore, it is possible to consider them as variables in research and to be treated ethnographically (Wolpert et al., 2024). Such an idea, the stochasticity of violence as a describable social fact, would be useful for better observing and understanding institutional violences that, if manifested, would be the structuring effect of their own capacity and organizational strength. On the other hand, such ethnographic meticulousness would also help to prevent violence because it is a correlative action, decidedly opitulational, understood within the definition of “violent possibles” such as internal actions of control, regulation, and governability, typical of institutions.

The “violent possible” events and “violent acts” are different; the former are based on possibility, and the latter are concrete violent actions. Both are probable and inherent to the exercise of institutions, each with its specific stochasticity. That is, it is necessary to conceive them according to the probabilities of evolution, considering three possible times: birth, maintenance, and expiration.

If violences are the effect of everyday imponderables and circumstantial dynamics that provide them with the characteristics of indeterminacy and stochasticity, it would be difficult to find structural violence because, being so varied and so integrated, the factors of violence are affected by the functional conditions of violence. Once identified, their specific structural values must be determined. Furthermore, following the same argument, given that violences are derivations of social relations that result from violent possibles and potential interactions, violences are not consequences of state failure, and even if they were, not all violences would be (although the state is always a part), because that is its more or less standard mode of operation; which does not mean that it is structural. It is still functional and circumstantial that the conditions of violence become aggressive; hence, it is interesting to rethink the idea of institutional inaction. By themselves, they are generators of cycles of violence. Institutional action and inaction may or may not be part of a “violent possible,” because the inaction and action of an institution are two sides of the same coin (Dye, 2013).

After the questions about the nature of violence or its origin, the question of the dynamics of violent behavior and the way to observe it arises. What is violence, and what is the violent? How are they produced, manifested, and behaved? These questions should be framed within the everyday normality of violence and the violent, perhaps taking into account how they are naturalized, sedimented, or institutionalized. A whole challenge posed by Herzfeld (1992), yet to be addressed, on the production of indifference in the face of circles of inoperability, bureaucratic indolence, the basis of our concerns, about the symbolic and real roots of bureaucratic violence.

In the following section, the analysis will address the distinction between violence and the violent; it can be understood, roughly, that to understand violence, one must access it as a phenomenology of the practices of the violent. And, to the violent, in turn, by identifying the expression of force in each event. However, they are only that, accesses, if the social production of violence is not taken into account, whose main quality is stochasticity.^[1]

3 Violence

Violence is ubiquitous and condemned, almost without exception, universally (Bourdieu, 1998; Curtin & Litke, 2022). However, despite this condemnation, there is no hesitation in engaging with it through television, video game consoles, books, or sensationalist news; it overflows into daily life. Perhaps for this reason, there is also no hesitation in realizing that it is an evident, normal, habitual, sedimented, and daily characteristic of social institutions such as the family, school, church, political party, health center, police office, etc., all organized around the “do not do,” “you are not” and sanctions. Moreover, it is not that latent violence is absent; rather, processes exist that escalate this apparent normality and trigger specific episodes (violence is synecdoche, the part representing the whole, as when “domestic violence” is used to describe a territorial disagreement between siblings over the use of space in a shared room, which does not involve parents or other family members).

Violence, therefore, is produced by the convergence of a plurality of probabilities with intentions to control and seize (maintain or reinforce) a position of dominance in contentious matters – motivated emotionally or rationally – which coincide in a single event, positioning, organizing, unifying, and confronting individuals, families, groups, communities, or nations; it behaves relationally (even with a phantom) and manifests diversely, at a particular moment, erupting into aggression. It is not an isolated event, not arising from nothing, but rather the product of a chain of events that, individually, may seem normal or even innocuous.

The “violent potential” – not to be confused with potential violence – manifests through three interdependent probabilistic movements: preparation, disposition, and materialization. Each generates particular dynamics of participation, derivation, delimitation, and culmination, whose development, though uncertain, is directly or indirectly detectable until the eruption of the violent act. Preparation involves the interaction of concrete factors that engender “violent potential”; disposition configures the propensity for violent action; and materialization corresponds to the triggering of violent acts, which cause violence to emerge as a discrete (though probabilistic) and singular event.

This threefold movement introduces an objective quality of violences, hitherto untreated, which is their stochasticity; one might say it introduces diachrony. That is, the historical as probability and singularity within diversity. Stochasticity, due to its probabilistic basis, contributes to improving the perception of the singularity of violences and to better projecting prevention methods, transcending deterministic approaches and acknowledging the inherent randomness of transitions from emergencies to violent acts. Perhaps the heterogeneity of violent manifestations can be better understood without discriminating against the potential importance of individual “violent possibles” (often overlooked in other perspectives) to large-scale collectives, recognizing the stochastic influence on their respective developments. Causal, historical, and anthropological mechanisms of violence are not, therefore, discarded, but rather integrated into this framework, where uncertainty, instead of predictability, determines the transition between the potentiality and materialization of the “violent act.”

The concatenation of these events, their interconnection, is what creates the potential for violence. Randomness plays an influential role because chance and the unforeseen affect the unfolding of events leading to violence and seem to operate systemically. They are not loose pieces in this complex device; thus, recalling Bertalanffy (1968), a small change in the system generates proportional changes within it; therefore, varying the sequence can alter the trajectory of the violent eruption (p. 56).

4 “The Violent”^[5]

The violent is unpredictable; its indeterminate nature prevents certain predictions of its form or occurrence. Furthermore, its expression through a vast diversity of forms complicates its definition and the capacity to predict its specific manifestations. Due to its variability, exhaustively defining violence requires more time than has been allotted thus far and must also consider its probabilistic nature. It is difficult to determine with certainty when or how it will occur, even when it is before our eyes or when its steps disturb tranquility. However, it is possible to deeply describe and study, and evaluate the probabilities of different “violent possibles.”

“The violent” arises as an attribute of the convergence of multiple possible factors interwoven into a unity, enabling a deeper and more nuanced understanding of the violent event. Consequently, this surpasses the reductionist view of the violent as a physical, direct, and intentional act. It manifests according to context, situation, intention, and impact. Therefore, considering its temporal dimensions – providing access to its stochastic nature – leads to a more exhaustive, rich, precise, and complete understanding of any social event. The violent is, thus, the inherent, diverse, and objective attribute of violences (plural), helping to avoid the simplification that violence rests solely on its physical, direct, and intentional particularity, reinforcing the idea of violence’s multiple manifestations.

5 Stochastic, the Stochastic, and Stochasticity^[6]

“Stochastic” is a probabilistic mathematical method that allows for identifying the behavior of plausible variables to model the evolution of systems where indeterminacy plays a significant role. “The Stochastic” would be the set of empirical manifestations of events studied using stochastic methods. And “Stochasticity” is the empirical quality of any process or event understood as a chrono-situated, open, and dynamic social construct, capable of evolution. General Systems Theory (Bertalanffy, 1968) can encompass them, and to deal with them, of opitulation^[7] (Zambrano, 2024). Institutional violence, particularly in its diversity, is a prime subject for analysis using these theories.

Both systemic and opitulational theories offer a complementary holistic interpretive framework, interconnected and in constant interaction with its environment, for understanding institutional violence as systemic violence and its social intervention. This possibility can contribute to reinforcing the explanatory power of both systemic and opitulational theories, because they can foreground the influence of probability and uncertainty within systemic dynamics. Since an open system, by definition, constantly exchanges matter, energy, and information with its environment, it creates a complex scenario where probability, rather than certainty, governs the diverse, differential behavior and unfolding of institutional violence, which can be described under the concept of stochasticity.

Because each theory possesses its own stochastic specificities (continuous changes, adaptations, etc.), an apparently minor change in a variable or the emergence of an event can trigger, through the intricate network of interrelations within the system, considerably larger and unpredictable effects. This recalls the non-linear nature of many open systems, where small causes can produce large effects – a fundamental principle that stochasticity helps explain and ethnography helps describe (Bertalanffy, 1968, p. 56).

6 Concluding Remarks

This work has attempted to contribute to the study of institutional violences by introducing a non-deterministic perspective. The fact that violence is unpredictable does not mean that it is impossible to determine and prevent it with some accuracy. While certainty regarding the time, place, mode, and magnitude of a specific violent act is unattainable, it is possible to identify contexts and situations of higher or lower risk and to develop prevention strategies focused on reducing the probabilities of tensions escalating into aggressive episodes. Deterministic theories relating institutions and violence based on single, predictable causes, binary constructs, and polar categories – useful as points of reference – fall short in the face of the complexity, diversity, and relativity of institutional violence. This work aimed to advance beyond such limitations.

Understanding that much institutional violence is systemic and affects society is very important; this systematicity is characterized by the naturalization of oppressive practices that are transferred to society as innocuous, necessary, or reliable. This makes identifying and understanding them extremely difficult. One does not know how much violence one is capable of enduring, nor how much one is capable of inflicting, Bourdieu (1998) and Lorente Molina and de Torres (2014) observed.

This analysis avoids explaining institutional violence by labeling institutions as violent or non-violent, as not all institutions can be categorized in this way, and not all institutions with the potential for violence exercise it (furthermore, there are numerous instances of institutions without inherent potential for violence that suddenly escalate to violence). Therefore, the aim has been to understand the systemic processes within which institutional violence is normalized, making it difficult to identify, as acting, sedimented violence. By reducing institutional violence to common, verifiable aspects such as its state-based nature, hegemonic management capacity, or the cultural origins of its functions and officials, the difficulty of observation is also reduced. This is further simplified by studying the intricate network of interactions that fuel violence contextually and situationally, whether symbolic, political, or economic. All these elements are essential and must be considered, along with their interactions and syntheses, to improve the perception of “violent possibles” and institutional violence.

The presented approach may be more fruitful by understanding institutional violence according to its stochasticity; that is, according to the set of probable events in which contingency and the interaction of multiple factors and intentions play a decisive role in shaping patterns or trends of violent institutional acts. Therefore, the impossibility of predicting with certainty when and how the violent potential of an event will manifest is recognized. Stochasticity necessitates considering the multiplicity of factors that contribute to the emergence of a violent act (e.g., inequalities, individual, group, or community experiences, social constructs, media impacts, etc.) to seek conditions that make certain forms of violence more probable than others.

Recognizing the stochasticity of violence is the first step toward developing more effective prevention strategies, rather than seeking magical solutions. The stochastic perspective necessitates addressing both risk and safety factors within the same event (situated and contextualized), as these interact to reduce or escalate the probability of violence manifesting, suggesting an ecological rather than a purely statistical mode of configuration. Hence, the proposal of a conception of stochasticity that views institutions as complex, systemic ecosystems, where violence is not an intrinsic characteristic but the product of the dynamic interaction between their various components and their environment. Internal segmentation, contextual heterogeneity, and the temporal relativity of institutions are key to understanding how violence materializes and transforms within these settings.

Stochasticity acknowledges the existence of a ubiquitous “potential for violence” in any social structure; this potential, far from being inherent to the institution itself, manifests as a probable event, dependent on the complex interaction of multiple internal and external factors. Institutional violence is not an inevitable outcome but a conglomeration of “violent possibles,” whose probability of occurrence depends on the confluence of specific circumstances. This foundation does not negate the responsibility of institutions in generating violence, but rather emphasizes understanding the conditions that increase or decrease the probability of its manifestation. The stochastic model, therefore, offers the possibility of a more precise analytical tool for identifying risk factors and designing more effective intervention strategies aimed at mitigating the probability of the “potential for violence” materializing into concrete actions.

Institutions – be they “social, cultural, economic, or political; altruistic or charitable; public or private” – sediment, at a probable moment, what we have termed “violent possibles.” This represents a paradigm shift from the search for culprits to the construction of complex and contextualized solutions capable of navigating the inherent uncertainty of this institutional phenomenon. Future research should delve deeper into this area, developing methodologies appropriate for analyzing the complexity and uncertainty inherent in the relationship between institutions and violence.

Stochasticity is a fundamental component for understanding the complexity of the open and dynamic systems described by Bertalanffy’s General Systems Theory. Incorporating probability and uncertainty into systemic analysis allows for the development of more realistic models and more robust predictions, recognizing the intrinsically random nature of many real-world phenomena. By embracing stochasticity, systems theory is strengthened as a tool for analyzing and understanding the intricate network of interactions governing the behavior of complex systems.

Stochastic studies are crucial for a complete understanding of violence, particularly in its historical and social manifestations. While research has explored deterministic factors such as poverty, inequality, or lack of opportunity, the influence of chance and contingency on the genesis and propagation of violence has been systematically neglected. By integrating stochasticity into models, we can better explain the variability in the intensity and form of violence over time and across different contexts, overcoming the limitations of purely deterministic explanations. This stochastic perspective, applied to the past, allows for extrapolation to the future, modeling the probability of outbreaks of violence considering the contingency inherent in social processes. Since the present is situated between the past and the future, a stochastic approach allows not only to understand the origin of current violence patterns but also to predict, with greater precision, their possible future trajectories, recognizing that random forces, such as unexpected events or fluctuations in social dynamics, play a fundamental role in their configuration. The need for stochastic models specific to violence data, rather than simple adaptations of models from other disciplines, becomes evident for achieving significant progress in understanding violence in its entirety.