One of the strongest arguments for the use of violence on the left is that it is tactically necessary in order to defeat structural or systemic violence, or to dismantle a violent regime, such as apartheid, dictatorship, or totalitarianism.10 That may well be right, and I don’t dispute it. But for that argument to work, we would need to know what distinguishes the violence of the regime from the violence that seeks to take it down. Is it always possible to make that distinction? Is it sometimes necessary to suffer the fact that the distinction between the one violence and the other can collapse? In other words, does violence care about that distinction, or for that matter, any of our typologies? Does the use of violence reduplicate violence, and in directions that cannot always be restrained in advance?
Sometimes the argument in favor of violence is that it is only a means to achieve another goal. So one question is: Can violence remain a mere instrument or means for taking down violence—its structures, its regime—without becoming an end in itself? The instrumentalist defense of violence depends quite crucially on being able to show that violence can be restricted to the status of a tool, a means, without becoming an end itself. The use of the tool to realize such purposes presupposes that the tool is guided by a clear intention and remains so guided throughout the course of the action. It also depends on knowing when the course of a violent action will come to an end. What happens if violence gets out of hand, if it is used for purposes for which it was never intended, exceeding and defying its governing intention? What if violence is precisely the kind of phenomenon that is constantly “getting out of hand”? Lastly, what if the use of violence as a means to achieve a goal licenses, implicitly or effectively, the use of violence more broadly, thereby bringing more violence into the world? Does that not lead to the possibility of a situation in which others with contrary intentions rely upon that revitalized license in order to realize their own intentions, to pursue destructive aims that are contrary to the ends constrained by its instrumental use—aims that may not be governed by any clear intention at all, or may prove to be destructive, unfocused, and unintentional?11
We can see that at the outset of any discussion about violence and nonviolence, we are caught up in another set of issues. First, the fact that “violence” is used strategically to describe situations that are interpreted very differently suggests that violence is always interpreted. That thesis does not mean that violence is nothing but an interpretation, where interpretation is conceived as a subjective and arbitrary mode of designation. Rather, violence is interpreted in the sense that it appears within frameworks that are sometimes incommensurable or conflicting, and so it appears differently—or altogether fails to appear—depending on how it is worked over by the framework(s) at issue. Stabilizing a definition of violence depends less on an enumeration of its instances than on a conceptualization that can take account of its oscillations within conflicting political frameworks. Indeed, the construction of a new framework tasked with such a purpose is one of the aims of this project.
Second, nonviolence is very often understood to be a moral position, a matter of individual conscience or of the reasons given for an individual choice not to engage in a violent way. It may be, however, that the most persuasive reasons for the practice of nonviolence directly imply a critique of individualism and require that we rethink the social bonds that constitute us as living creatures. It is not simply that an individual abrogates his or her conscience or deeply held principles in acting violently, but that certain “ties” required for social life, that is, the life of a social creature, are imperiled by violence. Similarly, the argument that justifies violence on the basis of self-defense appears to know in advance what that “self” is, who has the right to have one, and where its boundaries lie. If the “self” is conceived as relational, however, then the defenders of self-defense must give a good account of what bounds that self. If one self is vitally connected to a set of others and cannot be conceived without them, then when and where does that singular self start and end? The argument against violence, then, not only implies a critique of individualism, but an elaboration of those social bonds or relations that require nonviolence. Nonviolence as a matter of individual morality thus gives way to a social philosophy of living and sustainable bonds.
Moreover, the account of requisite social bonds has to be thought in relation to the socially unequal ways that “selves” worth defending are articulated within a political field.12 The description of social bonds without which life is imperiled takes place at the level of a social ontology, to be understood more as a social imaginary than as a metaphysics of the social. In other words, we can assert in a general way that social interdependency characterizes life, and then proceed to account for violence as an attack on that interdependency, an attack on persons, yes; but perhaps most fundamentally, it is an attack on “bonds.” And yet, interdependency, though accounting for differentials of independence and dependence, implies social equality: each is dependent, or formed and sustained in relations of depending upon, and being depended upon. What each depends upon, and what depends upon each one, is varied, since it is not just other human lives, but other sensate creatures, environments, and infrastructures: we depend upon them, and they depend on us, in turn, to sustain a livable world. To refer to equality in such a context is not to speak of an equality among all persons, if by “person” we mean a singular and distinct individual, gaining its definition by its boundary. Singularity and distinctness exist, as do boundaries, but they constitute differentiating characteristics of beings who are defined and sustained by virtue of their interrelationality. Without that overarching sense of the interrelational, we take the bodily boundary to be the end rather than the threshold of the person, the site of passage and porosity, the evidence of an openness to alterity that is definitional of the body itself. The threshold of the body, the body as threshold, undermines the idea of the body as a unit. Thus equality cannot be reduced to a calculus that accords each abstract person the same value, since the equality of persons has now to be thought precisely in terms of social interdependency. So, though it is true that each person should be treated equally, equal treatment is not possible outside of a social organization of life in which material resources, food distribution, housing, work, and infrastructure seek to achieve equal conditions of livability. Reference to such equal conditions of livability is therefore essential to the determination of “equality” in any substantive sense of the term.
Further, when we ask whose lives count as “selves” worth defending, that is, eligible for self-defense, the question only makes sense if we recognize pervasive forms of inequality that establish some lives as disproportionately more livable and grievable than others. They establish this inequality within a particular framework, but this inequality is historical and contested by competing frameworks. It says nothing about the intrinsic value of any life. Further, as we think about the prevailing and differential ways that populations are valued and disvalued, protected and abandoned, we come up against forms of power that establish the unequal worth of lives by establishing their unequal grievability. And here, I do not mean to treat “populations” as a sociological given, since they are to some degree produced by their common exposure to injury and destruction, the differential ways they are regarded as grievable (and worth sustaining) and ungrievable (already lost and, hence, easy to destroy or to expose to forces of destruction).
The discussion of social bonds and the demographics of unequal grievability may seem unrelated to the opening discussion of the arguments used to justify violence or to defend nonviolence. The point, however, is that all these arguments presuppose ideas about what counts as violence, since violence is always interpreted in such discussion. They presuppose as well views on individualism and on social relationality, interdependency, demographics, and equality. If we ask what violence destroys, or what grounds we have for naming and opposing violence in the name of nonviolence, then we have to situate violent practices (as well as institutions, structures, and systems) in light of the conditions of life that they destroy. Without an understanding of the conditions of life and livability, and their relative difference, we can know neither what violence destroys nor why we should care.
Third, as Walter Benjamin made clear in his 1920 essay “Critique of Violence,” an instrumentalist logic has governed the prevailing ways in which violence has been justified.13 One of the first questions he poses in that complex essay is: Why has the instrumentalist framework been accepted as the necessary one for thinking about violence?