By George

This was written during one of my ‘collections’ (non-assessed exams sat each term at the University of Oxford). It was written in one hour, under exam conditions (closed book). I have gone back over it to correct typos, but that is all.

Is violence a form of power, or something fundamentally different?

In this essay, I will argue that violence is a form of power, and can in fact be one of the most effective forms. I will begin by defining power, distinguishing the concepts of power-to, power-over, and power-with, before discussing how violence might fit into each of these categories. Against Arendt, I will conclude that violence is closest to a form of power-with, since it most closely resembles other forms of power when it is being used strategically and discreetly, creating a culture where each additional act of violence, or even its threat, constitutes an action modifying action.

To begin with, I will clarify the concept of power. Power has been called an ‘essentially contested concept’ by Lukes (2005), and it is true that definitions often seem contradictory. The idea of ‘power-to’ finds its origin in Hobbes. For Hobbes, power was a person’s capacity to effect change in the world around them. Therefore money, friends, and of course the capacity for violence, were all forms of power. If somebody has many friends, then they can be leveraged in order to achieve that person’s desired ends. Similarly, the capacity for violence, and violence itself, can be used to force a particular outcome. This is exemplified in war, torture, military coups – in all cases, violence is obviously power-enhancing. A problem with this view is that it seems to be more concerned with the social basis of power, rather than power itself. The statement ‘I have weapons’ is not semantically identical to the statement ‘I have power’ – it depends on the context. If weapons are not what is needed to achieve your ends, then you do not have power. Thus, it is perfectly possible to have all of the things which Hobbes considers as constituting power, and be quite powerless in the real, personal sense. An improved definition might describe power as a capacity to effect one’s desired ends, but this also encounters problems. It does not seem as though somebody is powerful if they have nothing, but also desire nothing. What is needed is an account of power-to which does not conflate it with material conditions, but also does not leave it meaningless as a political concept.

My suggestion here is to define power-to more negatively – rather than the capacity to achieve your own desires, it is the capacity to get in the way of other people’s desires without fundamentally harming your own. Thus, the person who desires nothing is not powerful, but the dictator who arbitrarily wages war is, even if what they really desire is admiration, happiness, et cetera.

Violence is clearly a way in which you can disrupt other people’s desires, and hence is a form of power-to. However, it is unclear whether this is interesting, normatively speaking. Power as a capacity does not mean that it is being used, and hence does not seem particularly problematic. Of greater interest in political theory are two particular denominations of power-to – power-over and power-with.

Power-over, unlike power-to, is especially concerned with human relationships. Foucault conceptualised power-over as a struggle which takes place between free individuals. The individual with power attempts to modify the actions of another person, and that person resists. He writes that every relationship of power dreams of domination, and every strategy of resistance dreams that, if it acts correctly, it will become the winning strategy. In this sense, power-over is distinct insofar as it is ‘zero-sum’ – if I have the ability to dominate you, you necessarily do not have the ability to dominate me.

Is violence a form of power-over? Arendt would say no – violence between individuals is force, or strength, rather than power. The meaning of this statement is that violence is based in natural, physical potential, and that such potential is not correctly termed power. It is unclear to me why this should necessarily be the case. Some might bring in Foucault, arguing that violence destroys a subject’s capacity to act, so counts as domination rather than power. From a Foucauldian perspective, power only operates where there remains space for resistance, whereas domination forecloses that space. However, this response should be rejected for two reasons.

Firstly, it is not only violence that can destroy a subject’s capacity to act – psychological coercion can too. The phenomenon of ‘social death’ seems relevant here. Social death takes place when acts of violence or degradation result in the breakdown of the social being, resorting to a non-social self which is totally isolated and unaware of its own reality. Slavery and the Holocaust are paramount examples. During the Holocaust, Primo Levi wrote that the only way he maintained the belief that he was in fact a person was by participating in small acts of autonomy – washing himself, shaving, and so on. Viktor Frankl shows in Man’s Search for Meaning how this was not a level of self-worth maintained by many victims of the Holocaust, who became hollow and passive figures. This point is echoed by Harriet Tubman, who wrote regarding emancipation that she ‘would have freed a thousand more, had they only known they were slaves’. Of course, all these examples included horrendous acts of violence. But it does not seem to me that they relied on brute violence. Imagine a Holocaust where guards did not assault prisoners, but verbally degraded them, isolated them from others, and ensured the ever-present threat of the gas chambers loomed over them (noting that the threat of violence is not equivalent to violence itself). The effects would surely have been similarly paralysing. Haugaard calls this ontological power – the power to destroy someone’s identity and remould it at your will. Such power can be solely psychological, yet tantamount to the most extreme acts of physical violence. I do not necessarily suggest that social death is possible; I am merely pointing out that even if we accept the premise that violence destroys subjectivity, it does not follow that we should normatively separate violence from other forms of coercion or psychological control.

Secondly, we might doubt the notion that subjectivity can be ‘destroyed’ at all, since this has the perverse implication that power ceases to be power when it becomes too powerful. That is, when a power strategy becomes so successful that it becomes total domination, it becomes normatively distinct. A concept should be able to capture its archetypal illustrations, which include extreme forms of oppression as well as subtle instances of social discipline, so I reject this categorical separation. Said and Fanon go further, arguing that domination is not the total destruction of subjectivity, but rather the formation of a particular ‘colonial subjectivity’ which still finds ways to resist. In other words, domination is a particular kind of subjectivity, not its annihilation, and this is reflected in colonial discourse. Accordingly, Arendt’s idea that violence and power are fundamentally distinct – physical violence on one side, and free collective action on the other – falters when we see how the two techniques can be interwoven to create unique forms of power, such as a culture of silence.

I have delayed addressing power-with because it seems to me that it is not clearly distinct from power-over. Arendt defines power-with as the human ability to act in concert. The reason it appears distinct from power-over is because power-over is zero-sum, whereas she defines power-with as an end in itself, which spontaneously arises and dissipates when collectives rise together. Thus, power-with would not seem to come ‘at the cost’ of anyone else.

I think this is mistaken. The power of a group may be temporary, but a group is not powerful unless it is more powerful than another group. In this regard, power-with seems like a collective form of power-over. Individuals can combine their individual power into a greater whole, but they cannot change it into something fundamentally different. Hence, my reasoning for why power-with is not fundamentally different from violence is similar to why power-over is not: in both cases, violence can augment the potential of its possessor. Arendt rejects this conclusion because she views violence as instrumental, whereas she defines power as an end in itself. However, she fails to consider how the power of a collective could be used by an unjust majority in order to enforce oppression. For example, patriarchy might be termed a version of power-with since it involves the (general, and often tacit) consensus of men. When men commit acts of violence against women, it does not signal the faltering of their power, but is rather further evidence of it. Not only this, it reinforces the power (and is thus a form of it) by promoting a culture of silence and fear. This is why I argued in my introduction that power-with is perhaps the form of power where violence is most immanent. Violence committed by a powerful group creates social norms, which construct realities. It contributes to cultural hegemony and ‘common sense’, found in casual phrases concerning women’s clothing, ethnic identities, and accepted social categories. In this sense, violence ‘ends action modifying action’, but precisely by ending it in the short term, it can create and sustain a culture in which certain groups retain their political domination.

Thus, I conclude by repeating that violence is not fundamentally different from power on any account. Violence is both a tool that enhances power and an example of it, and its frequent normative separation relies on a desire for conceptual neatness that is not reflected in reality. While Arendt, Lukes, and Foucault each have reasons for distinguishing violence from power (based on their focus on legitimacy, subject-formation, or collective freedom), their distinctions sometimes overlook how violence can be embedded in social systems in such a way that it produces, shapes, and even solidifies power. The fact that violence can end action modifying action does not mean it eliminates power entirely; rather, it can shift and channel power in ways that perpetuate domination and subjectivity simultaneously.