Why Riot? An Expressive Theory of the Justification of Rioting
Edmund Tweedy Flanigan
It must be massive, militant, nonviolence, or riots. The discontent is so deep, the anger so ingrained, the despair, the restlessness so wide, that something has to be brought into being to serve as a channel through which these deep emotional feelings, these deep angry feelings, can be funneled.
Martin Luther King, Jr.1
I. INTRODUCTION
Political rioting—roughly, political protest events characterized principally by, or substantially including, moments of decentralized violence—is a durable and significant social phenomenon.2 Indeed, it is a recurrent feature of societies across wide differences in time, space, and political structure.3 It is also highly morally contentious. Leaders across the political spectrum regularly condemn the practice, which much of the general public appears likewise to stand against.4 Many others regard rioting as neither altogether condemnable nor morally justifiable, but rather as an understandable, perhaps inevitable, perhaps excusable reaction to social injustice.5 Still others take the view that some cases of rioting may be, at least in principle, morally justifiable.
Among those who hold the latter position, the principal burden lies in offering a satisfactory account of how the violent harms characteristically involved in rioting could be justified. The dominant approach to this question, represented in recent literature by Avia Pasternak and Jonathan Havercroft, has been to regard rioting as akin to an act of collective defense and as governed, morally, by the principles of war and self-defense.6
I too believe that political rioting can be morally justified—or rather, as I prefer to say, that rioters can be justified in their rioting.7 However, I believe that the just war tradition and its interpersonal counterpart, the framework of defensive ethics, are not tools well suited to this task. This is not because rioting is completely unlike cases of war or self-defense—in some respects the two are similar, and in other respects they are not. Indeed, the conditions for justified rioting that I shall propose bear important similarities to the conditions set out by these frameworks. Rather, it is because rioting is often not, in Pasternak’s words, “a form of defensive harm,”8 nor is it otherwise amenable to, in Havercroft’s words, “a systematic theory of a just riot.”9 My first aim is to establish this conclusion.
If we are to think productively about the justification of rioting, we would do best, I believe, to look elsewhere. Drawing on the tradition of expressive theories of the justification of harm in political and interpersonal contexts,10 I outline a theory according to which rioting—including the violent harms it involves—is justified when and because it is a form of fitting expression. Rioting is not only, in King’s memorable formulation, “the language of the unheard”;11 it may also be, in the very same dimension, their righteous sword.
II. CHARACTERIZING RIOTS
The phenomenon of rioting is notoriously difficult to characterize.12 At the same time, however, the normative dynamics of rioting depend, as do many social practices, on central characteristics of the practice itself: its nature, aims, and so on. It is therefore important to attempt to accurately characterize it.13 Indeed, as I’ll argue, differences in characterization turn out to have important implications for the normative analysis of the practice.
Avia Pasternak offers a careful, empirically-informed characterization, which makes for a good starting point. For her, a political riot is:
a public disorder in which a large group of actors, acting spontaneously and without formal organization, engages in acts of lawlessness and open confrontation with law enforcement agencies…[leading typically to] damage to public property…damage to private property…and harm to persons…[as] a response of a disadvantaged group to shared experiences of “subjective deprivation, social exclusion, political powerlessness, and moral outrage.”14
It is also, she emphasizes, “a form of political protest”:
The political riot is a communicative episode—its participants intend[] to deliver a message to the police, the government, and to fellow citizens…[of] anger…condemnation of [] injustice…demand for change…[and] defiance.15
As for rioters’ aims, Pasternak characterizes them as threefold. The first is “to bring about a change of public policy that will eradicate, or in the least ameliorate, the substantive violations of justice they experience at the hands of the state”;16 the second is “to resist…political marginalization”;17 and the third is to “communicate anger toward the state and defiance of its political authority.”18
This characterization is (in my estimation) largely accurate, but not entirely so, and some of the inaccuracies bear directly on the question of rioting’s justification. So let me now mark a few points of difference.
First, several ameliorative suggestions. For one thing, referring to riots as “spontaneous,” while common, is not entirely apt. Although riots are usually unannounced and unplanned and their occurrence difficult to precisely predict, they also “do not,” as King said, “develop out of thin air.”19 They are, rather, moments of inflection in much broader and deeper social and political currents. For this reason, I think it is better to call them “eruptive.”20 This stands in sharp contrast with other forms of social action, which are often the product of movements built slowly and painstakingly over time.
Pasternak’s claim that riots are characteristically “without formal organization” also bears emphasis and further articulation. The key point, in my view, is that rioters are only loosely coordinated, and they lack directive leadership.21 Each participant’s activity is at their own discretion and direction, even though it may be locally closely coordinated with others (“let’s overturn this car”) or globally loosely coordinated (“people are gathering near City Hall”). Rioting represents, in this way, an uncommon type of joint activity: rioters clearly act together, but they do so in an especially diffuse way—one that most plausibly falls short of joint agency. A consequence of this is that individual participants, rather than the group of rioters writ large, represent the primary site of moral decision-making, and so too, therefore, of moral justification.
This is important, because justification attaches to agentive acts, and if rioters do not constitute a joint agent, then only in a loose or summative sense could we ask whether some riot was justified. What we can clearly ask, by contrast, is whether individual riot participants are justified in their rioting. This is important because a theory of the morality of rioting should not merely morally assess riots, it should provide moral guidance to prospective rioters weighing whether and how to participate.
To be clear, this is not to say that we cannot morally assess riots as events, nor that the justification of individual rioters’ acts cannot depend on facts about the joint act as a whole. Indeed, I’ll argue that whether a riot participant is justified in their rioting depends on facts about the overall expressive mood of the rioters. Moreover, it can sometimes make sense to summarize many shared facts about the individual justifications of riot participants by describing a riot itself as justified or unjustified. For example, if rioting in response to sports outcomes is never justifiable, then we can sensibly claim that sports riots are unjustifiable, since no participant could have been justified in their rioting. The point is rather to identify the locus of moral agency in rioting—and so the locus of moral justification—at the level of the individual.
For another, while many riots involve confrontation with law enforcement, not all do, and this feature strikes me as incidental, rather than central, to rioting as a practice. When rioters engage with the police, this is often as a response to law enforcement activity in the midst of a riot; only sometimes are the police also specific targets of rioters’ anger and action.22 The underlying point is that rioters often target those they take to be responsible for the wrongs to which they are responding (or the agents of those responsible).
Now for the point of substantial disagreement. Pasternak claims that rioters act “in order to bring an end, or at least to ameliorate, their on-going unjust treatment at the hands of their state” and to bring about a change in policy.23 This is, I believe, typically false. But, as I shall argue in what follows, it matters significantly what we take the characteristic aims of rioters to be, and particularly whether we take their aims to be oriented towards the instrumental achievement of social change.
Very rarely do riot participants express the view that their activity is aimed at improving the conditions against which they are rioting. On the contrary, rioters commonly describe their activity as taking place against a background feeling of hopelessness or inability to expect, much less effect, the political change that would ameliorate their condition. As one representative participant in the 2011 U.K. riots said:
Fuck knows, don’t know, don’t really care about [change] no more. I’ve gone past caring. Just think there’s no point in me wishing, wanting things to happen.24
Similarly, Michael Fader, a participant in the Stonewall riots, remarked that “of course you couldn’t” achieve the liberatory ends hoped for (“go after the police and free the Stonewall or whatever you want to call it”), but that the events of the riot “were symbolic gestures, and they were enough.”25 These sentiments are not merely anecdotes; they reflect large-scale research into the attitudes and aims of riot participants.26
Insofar as rioters share an aim, that aim seems better characterized as principally and distinctively expressive or communicative. Riots are, of course, moments of intense political affect, and a desire to express anger or frustration, among other similar emotions and attitudes, is a commonly cited aim. Also cited is the aim of making others “take notice” against a pervasive feeling of social or political “invisibility.”27 In the words of one riot participant, “it’s not like your voice is heard; they don’t care about you because you’re poor.”28 Similarly, at a community meeting with Martin Luther King following the 1965 Watts riots, Connie Griffin, a riot participant, said:
You couldn’t talk to anybody because there was nobody to talk to. … The only way—the only way, it seems—that we can ever get anybody at any time to listen to us is to start a riot.29
Riot participants, in other words, characteristically understand their activity as a form of protest and as a claim to be heard, which is to say as a principally expressive or communicative activity, rather than as instrumental action oriented towards the achievement of concrete political ends.
As will become clear later, it matters to the justifiability of rioting whether we take it to be an expressive or communicative act. We can ask already, then, which of these best describes rioters’ characteristic acts. No doubt individual rioters vary in this dimension as they do in others, and no doubt the aims of some are mixed. For some, pure expression, i.e., the venting of anger or grievances, may be the aim, using the targets of violence as one might use a punching bag. More commonly stated, however, is the aim not just of speaking but of “being heard,” which suggests the aim of communication.
A difficulty for the latter analysis, however, is that rioting is not an activity well suited to communication—a fact that riot participants are well aware of. That is, while rioting often makes noise, and is often noticed (“heard”) in this respect, it also often fails to be understood. It is blunt and antagonistic, frequently leading to misinterpretation and backlash. If communication is the aim, there would seem to be better avenues available.
This is a difficulty because, as Elizabeth Anderson and Richard Pildes write, “the constitutive aim of…communication is that [it] be publicly acknowledged as governing the interactions of the speaker and the addressee.”30 For a practice to have a constitutive aim is (at least) for it to be practically oriented towards the accomplishment of that aim. But for the reasons just mentioned, it is doubtful that rioting can be construed as constitutively aiming at successful communication of its attendant attitudes. If it were, it would (or at least should) limit and reform itself so that its message would be likely to be understood and, ultimately, accepted. Clearly rioting is not like this.
It might be answered that since rioting is often claimed to be the only way to be heard, its communicative defects may yet be intelligible as part of an act that constitutively aims at communication. Perhaps; but then it is not clear why, from a communicative standpoint, violence that is likely to be heard but misunderstood (like rioting) is to be preferred to a clearer, non-destructive form of expression (like traditional forms of protest) that is likely to be easily understood but ignored. Both seem equally defective as far as communication is concerned. If this is the goal, why opt for the more destructive, harmful method?
My view is that the characteristic aims of rioters in this respect are best described as something between (mere) expression and (robust) communication, which we can call directed expression. They are claims to be heard, directed at an addressee in response to a perceived wrong, but, crucially, without the primary aim of being understood or accepted. In this way, they are members of a familiar species of righteous moral claims—ones whose primary aim is simply the making of the claim, and which only secondarily aim to be understood or accepted. This is like the way we might shout “STOP!” at an attacker we are sure will not heed our demand; or like the way we might participate in ordinary political protest regardless of the likelihood of anyone powerful taking notice, simply because “it is important to speak up”; or even like the way a parent might insist to their teenage child that they must do this or mustn’t do that, despite knowing that the child won’t listen. Sometimes the point lies in the importance of making the claim more than in the importance of its making a difference.
In summary, then, I suggest that we understand riots as (characteristically) large, eruptive public disorders in which participants act at their own discretion and direction in order to violently express a family of emotions and attitudes—anger, frustration, disaffection, despair—as a way of protesting common (subjective) experiences of social injustice, directed towards those taken to bear responsibility for that injustice.
III. JUST RIOT THEORY
Pasternak and Havercroft have advanced views according to which the morality of rioting is governed, respectively, by the principles of defensive ethics or by an analogue of the principles governing the morality of war. On some views, these are two ways of describing the same thing. Many now take defense to be the primary, or even only, cause that could justify resort to war; and similarly, many “reductivists” now take the permissible imposition of harm in war to be only a special case of the permissible imposition of defensive harm in general. For those holding such views, there would be little difference between a view of rioting regulated by the principles of defensive ethics and a view of rioting regulated by analogues to the principles of just war.
According to an older tradition, however, a variety of causes can justify resort to war, among which defense is only one. Analogously, rioting might be permissible when and because it is a means to achieving some other just cause within the constraints set out by the principles of the ethics of war.
I’ll now argue that either approach faces substantial challenges when applied to characteristic cases of rioting.
III.A. Defense
While Pasternak recognizes various valuable goals realizable through rioting, she takes the justifiability of rioting—and in particular, the justifiability of the harms to persons and property that it characteristically involves—to turn on its functioning as a form of defense:
The most serious challenge rioters face…is that although their political goals may be worthy, the resort to violence (rather than peaceful protest) in order to attain them violates [the] natural duty [not to inflict unjustified harm on others]. In order to determine whether there exists a form of political rioting that can avoid this charge, we need to turn to existing standards of permissible defensive harm. For…political rioting has clear defensive goals, as its participants aim to undo, or in the least ameliorate, the conditions of injustice they experience, and to resist their political marginalization. It follows, then, that the permissibility of their actions should be assessed in light of the constraints offered by ethicists of self-defense and war—success, necessity, and proportionality.31
Pasternak’s claims are (i) that rioting is a form of defense; (ii) that this is true because rioters aim to ameliorate (perceived) social injustice and resist their political marginalization; (iii) that rioters in fact intend to achieve these aims through rioting; (iv) and thus, that rioting is permitted (indeed justified) when it meets the conditions on permissible defensive force, i.e., when it stands a sufficiently good chance of being successful at achieving its aims, is necessary to doing so, and inflicts only proportionate harm.
This argument faces objections at each step.
To begin with, we can ask whether rioting is best understood as a form of self- or other- defense. This is a question, first, about whether the things rioting putatively defends are proper objects of defensive action, and if so, also a question about whether rioting as a mode of action can be properly said to defend them. Pace Pasternak, fighting political injustice and marginalization are not, by themselves, “clear defensive goals” that can be properly pursued with force. For example, persistent unfairness in the tax code, while a political injustice, is not something that should be corrected with violence; and the coordinated exclusion of far-right parties from German ruling coalitions, while plausibly a case of political marginalization, is not clearly wrong at all, much less something that far-right partisans may exercise defensive force to change. Rather, the objects of rioting are most plausibly defensive when what rioters face are infringements of their most basic political rights and entitlements, or equally, wrongful threats to their physical safety; for it is clear that in non-rioting contexts (such as war), people do permissibly use force to defend their basic political rights and entitlements, as well as their bodies and lives. Clearly, rioting is sometimes about these things, but even in these cases, we must then also ask whether rioting defends them. For an act to count as a defense of d, it must be instrumentally oriented towards the establishment or preservation of d in the face of a wrongful threat against it.32 But as I shall argue below, rioting is characteristically not instrumentally oriented towards rectifying injustice. If I am right about this, then even when rioting is about things that are proper objects of defense, and so could be defended, it is nevertheless not, in most cases, practically oriented toward defending them, and so not a form of defense.
Second, it matters whether rioting is taken to try to “undo” or “ameliorate” injustice and marginalization, i.e., whether rioting secures (by establishing or preventing the loss of) justice or else simply promotes justice. For even if securing justice is a proper object of defense, it is not clear that its promotion would be. For example, while many believe defensive force to be justifiable in the establishment or preservation of national self-determination, few believe force to be justifiable against acts that merely undermine it (e.g., economic or diplomatic coercion) or in service of its promotion (e.g., to establish a better voting system). It is not just that these things might be disproportionate to armed resistance, it is that they seem to be the wrong types of things to be pursued violently.
It might be replied that merely contributing to the achievement of defense, or a just cause, sometimes licenses harm, as when individual soldiers (who cannot alone win or lose a battle) fight to contribute to victory. Perhaps rioting is like this—something that contributes to the long-term fight for justice. But rioters are in this respect unlike soldiers. “Taking that hill” and other potentially lethal offensive acts that contribute to victory within just defensive wars can often only be justified when coordinated by officers and generals who judge that they are necessary to winning the battle, or the war. Without such coordination, killing in order to “take that hill” might simply be murder. But as I have argued, rioters lack precisely this kind of coordinative leadership; and as I will argue below, rioters also individually lack the information necessary to make these judgments themselves.
Third, even if we regard the amelioration of injustice and resistance to political marginalization as proper objects of defense, it is doubtful whether, as Pasternak claims, riot participants characteristically aim to achieve these ends. Indeed, as I argued above, we have good reason to think that riot participants characteristically do not riot in order to end or ameliorate conditions of injustice. Rather, rioting is often a response to the perception that change remains too far beyond the horizon and that acting for change is futile. Rioters may of course hope that unjust conditions will end, and some may even hope that their acts will hasten the arrival of more just conditions. But acting with hope for some end is not the same as acting in order to bring that end about.
An immediate reply is that whether the aims of rioters are defensive may be irrelevant, so long as their activity is in fact oriented toward the achievement of defensive ends. For, surely defense is permissible in ordinary cases even when (say) a victim, in a state of shock, cannot be said to aim at anything specific at all; and, in the context of war, surely soldiers may permissibly fight in just defensive wars, even if some of them aim only (say) to fulfill the legal requirements of service and return home, or to earn a salary for fighting.
This may be true.33 But notice that in the above cases, the actual aims of defenders and the defensive aims that help justify their acts are at least compatible: one could fight to defend one’s country and to serve it, or to earn a salary, for example (whether or not one actually holds the defensive aim). By contrast, rioters are often motivated by the perceived futility of instrumental action towards the improvement of political conditions. Their aims, that is, are at odds with the defensive aims that would help justify their acts. This forces us to adopt a “Government House” moral theory of rioting, according to which the true justification of the practice is incongruent with the way its participants understand it. But rioters often have a robust sense of their aims and exercise complex moral judgment about their participation, as is common among those who engage in high-stakes political resistance.34 Absent good reason to think that rioters misunderstand these aspects of the nature of rioting, this counts as a strike against the view of rioting as defense.
Pasternak’s final claim is that, in light of the valid defensive aims she takes rioters to (sometimes) justifiably pursue, the moral permissibility of rioting turns on whether it meets the conditions of success, necessity, and proportionality.35 Despite her arguments to the contrary, I doubt that rioting characteristically meets these conditions.
According to the necessity condition, a justified act of defense must be the least harmful effective means of achieving a defensive aim. According to the closely related success condition, a permissible defensive act must have a sufficiently good prospect of achieving that aim.36 As Pasternak notes, theorists often read this condition according to an evidentiary standard: a person is justified in some defensive act only if they reasonably believe, based correctly on the (sufficient) evidence available to them, that the act has a sufficiently good prospect of being successful, and that it is the least harmful available means of achieving this success. Pasternak claims that, because we have reason to believe that some riots are effective in ameliorating injustice, and because those who riot often have no more effective means available to them, rioters—if aware, or made aware, of this evidence—could reasonably believe that by rioting, they are engaged in defensive action that meets both the necessity and success conditions.
But this conclusion does not, I think, plausibly follow. The moral question facing would-be rioters is not “Shall we riot?” because rioters do not engage in collective deliberation of this kind. Rather, the moral questions faced by rioters are faced by individual participants: whether to break some particular window or overturn some particular car, whether to retreat or fight the police on some particular street, and, most generally, whether to participate in a particular riot. But there is great distance between these questions and the question of whether riots in general are necessary and effective means to political progress, and then whether this can be said prospectively of (or during) any particular riot, and whether an individual’s contributory acts can be justified by these judgments.
Imagine that you are a riot participant considering whether to throw a brick through a window. You must ask yourself, on this view, “Does breaking this window stand a good chance of helping to bring about an end to the injustice I am protesting, and is it necessary to that end?” By itself, it is hard to imagine that breaking any window could realistically end, or even ameliorate, serious political injustice. The thought must instead be that by breaking this window alongside many others performing similar acts of destructive resistance, we together might achieve this effect. That is, the justification of an individual rioter’s participatory acts must depend on judgments about the potential effectiveness of the riot as a whole.
Could an ordinary citizen living under conditions of oppression, knowing what they know about daily life, reasonably judge that in their case, rioting stands a good (and the best) chance of improving things? It is hard to see how they could. The best evidence on this question is not dispositive; in fact, it is mixed. As Pasternak notes, some riots have indeed resulted in progress on matters of social justice. On the other hand, many others have failed, and moreover have led to backlash that only worsened conditions.37 Still others have fizzled without making any significant political impact at all.
Of course, what riot participants need is not access to the best evidence, but to sufficient evidence available in their epistemic positions. But the potential effectiveness and necessity of a riot is an abstract and complex question about the direction of future policy. Unlike many other questions, whose answer may simply look different from different evidential perspectives, this is a question to which it’s hard to find answers from any perspective. It implicates further questions about (for instance) the likely acts of many—how many?—uncoordinated others, as well as the effects of those acts, and the likely response to all of this by still others. It is hard to see how being embedded in the daily conditions of oppression could make answering these questions any easier.
Moreover, this is a question not only about which direction the evidence points but also about what standard of evidence is sufficient to support such a judgment given the stakes of the answer. The stakes in this case are not just the possibility of unnecessarily harming a liable oppressor or their property. Rioting also risks backwards progress on the very issues that, on this view, it must stand a good chance of making forwards progress on in order to be justified. Given these stakes, one’s evidence should be very good. But in most cases, no one possesses very good evidence on this question.38
It may be objected that rioters commonly see rioting as the only political tool available to them. Perhaps when one has just one tool with which to defend oneself, one needs less evidence about its likely effects than might otherwise be required to justify wielding it. But though they may have just one tool, rioters do not have just one option available to them. They can also choose not to riot. If rioting stands a significant chance of producing substantial backlash, and its justification depends on its chances of improving social conditions, the option of not rioting becomes an important one—one that, again, makes it difficult to imagine making reasonable evidentiary judgments about from any perspective.
These judgments are complicated by the fact, mentioned already, that the acts in need of justification are participatory acts, not riots taken as a whole, and each participatory act should be assessed according to its marginal contribution toward a riot’s overall success. Thus, even if a rioter reasonably judges that rioting is sufficiently likely to achieve political progress, they must still ask whether breaking this window is necessary to achieve it. But in an uncoordinated joint activity like rioting, it is hard to know whether any marginal participatory act is necessary, and thus justifiable by the standards of defensive action. Solutions to collective action problems like this one often appeal to coordinative agents who can assure sufficient participation to achieve the activity’s ends. But such coordinative activity is arguably just what is absent in the chaotic circumstances of rioting.39
Finally, note that if we make the justification of political rioting turn on the ends it achieves, we make the permissibility of rioting as a response to oppression directly dependent on the attitudes of the oppressor.40 In ordinary defensive cases, such a restriction makes sense: if I justifiably punch you to defend myself, doing so must have a reasonable chance of averting your attack. But in characteristic circumstances of rioting, non-accommodation, and indeed failure or refusal to even hear rioters’ grievances, is a major part of what is at issue. To tell those who would riot that they are permitted to do so only if their acts succeed in appealing to the goodwill of their oppressors may seem to miss the point.
This is not to say that likelihood of accommodation should be irrelevant to the moral calculations of rioters. If the consequences of rioting would clearly be very bad, or very good, then naturally this should affect what rioters ought, all-things-considered, to do. But these are best thought of as external considerations that could defeat or bolster a prima facie justification for rioting. By contrast, the view of rioting as defense makes them a direct and central justificatory concern: as the likelihood of accommodation falls, so does the scope of rioting’s justifiable harms, and vice versa. But this variance runs, it seems to me, in the wrong direction. In the face of serious oppression, as the prospects for accommodation fade, the prima facie case for rioting only grows.
III.B. Another Just Cause
In light of these difficulties with construing rioting and its justification as defensive, we might ask whether some other valuable end achievable by means of rioting could justify its characteristic harms instead. As mentioned already, just war theory traditionally permits such a move: wars may be undertaken as a means to achieving a variety of just causes, among which defense is only one. If rioting can be successfully argued to pursue some other just cause, it might yet be justifiable within the framework set out by ethicists of war and defense.
Havercroft argues along these lines. In developing the outlines of “a theory of a just riot,” he identifies eleven “just riot criteria” corresponding to what is required of legitimate action outside of established venues for political contestation. Several constitute constraints on the use of violence: rioting must harm only “legitimate targets,” must be “proportionate to the injustice” protested,41 and must be a method of “last resort.” These correspond roughly to the constraints of liability, proportionality, and necessity. The other criteria offer candidates for the ends that might be taken as rioting’s just cause: rioting should “preserve freedom,” “promote equality,” and “give voice to the grievances of marginalized communities”; and it should only follow “systematic…violat[tions] [of] basic rights” and failures of “genuine conditions of reciprocity [for the] most disadvantaged.”42 Together, these suggest a just riot framework according to which rioting’s characteristic violence is permitted when it is a means to these ends, as long as it meets the constraints set out by the other conditions.
So, could the establishment or preservation of freedom, promotion of equality, giving voice to the voiceless, or responding to serious political wrongs furnish an alternative just cause for rioting? Despite the significance of these ends, I think not—at least not within the framework of just war or a parallel framework of just rioting.
To see why, note first that “just cause” is a technical term in just war theory: it is not simply an end whose attainment would promote justice, it is an end that may be properly pursued using war, or, more generally, violence as a means. Just as defensive ethics permits only harms that are instrumental to the achievement of defensive ends, just war theory permits only violence that is instrumental to the achievement of a just cause. But not all valuable ends may be pursued violently. Indeed, just war theorists typically agree that a mere benefit, no matter how large, cannot furnish a just cause for war; this is why the economic benefits of war, even if they would be very great, could never justify one. A more plausible view, as Jeff McMahan argues, is that a just cause for war must be something of the right type to be pursued violently, in addition to being a weighty enough instance of that type.43
When we ask whether some other end could be just cause for rioting, then, we are asking in the first instance whether it is an end of the right type to justify violence. Freedom and equality, however, are political values too capacious to serve as ends of the right type. No amount of extra freedom or equality could by itself justify harming others as a means to its attainment, nor could any mere reduction in either. Only when violations of freedom and equality are so severe that they infringe basic rights could they be such a cause—but then it would be the protection of these rights, not the value of the goods promoted, that would be the justifying cause.
The same might be true, depending on how we interpret it, of the value of giving voice to the voiceless. If we understand this as a way of promoting valuable political speech, then violence seems the wrong tool to use: we may not harm those whose political speech drowns out our own just for this reason, even if our speech is very important. On the other hand, if we take voicelessness to be indicative of a more basic political wrong—when some people are systematically “unheard,” then the state is failing in its basic duty to be responsive to all those it governs—then it would be the protection of this more basic entitlement, not giving voice per se, that is the plausible candidate for a just cause.
This leaves the protection of basic political rights and entitlements as the most plausible candidates for a just cause for rioting. I grant that these could be causes of the right type. Indeed, such things, in sufficiently serious cases, have been thought to be just cause for armed revolution, so they might also justify the lesser violence involved in rioting. But although this thought is plausible, it has the consequence of leading us back to problems directly analogous to those faced by the view of rioting as defense.
To protect one’s basic rights and entitlements is to act so as to secure them, which is possible only if one’s acts in fact stand a good chance of helping to do so. But it is not clear that rioting does so, given the significant risk of not only failing to improve conditions but of worsening them, nor is it clear that individual rioters are characteristically in a position to make judgments about this issue—at all, but also of the kind that could justify their participatory acts. And like that case, this view also runs the risk of forcing us to adopt a Government House theory of the justification of rioting, of making its justification depend on the accommodative attitudes of oppressors, and of making its justification vary incorrectly along this dimension.
The underlying problem, as I see it, for any approach that seeks to justify the characteristic harms of rioting by appeal to the framework of war or defense, as just riot theory seeks to do, is that these frameworks permit us to pursue certain ends violently only insofar as violence is a necessary and effective instrument to achieving those ends. But as I have argued, the ends realizable through rioting are either of the wrong type to be pursued via instrumental violence, or else are unlikely to be effectively pursued by wielding violence as a tool.
This is not to say, however, that there aren’t important ends achievable through rioting that might justify the use of violence as a non-instrumental part of their achievement via some other justificatory framework. Indeed, we can agree with Pasternak and Havercroft about the wrongs rioters seek to address through rioting without adopting the framework of defensive ethics, or just riot theory, as our mode of moral justification. In what follows, I shall defend such a view. Namely, I’ll argue that the moral dynamics of protest itself can sometimes justify harms to others.
This offers an alternative way of understanding the appropriate and justifiable response to the wrongs rioting characteristically responds to, including especially, I’ll argue, the problem of voicelessness. In other words, we can agree with Pasternak and Havercroft that being “unheard” is a central issue for rioters, and one on which rioting’s justification can turn, while turning to a different framework for understanding how and when it is justified. My claim will be that violence is sometimes the only available way of protesting the wrongs of oppression and marginalization loudly enough, and that making such a protest adequately is sometimes both permitted and demanded by the circumstances of oppression. But this is not because violence is a means to achieving political voice, much less concrete political ends; it is because violence is, in extreme circumstances, how one speaks.44
IV. RIOTING AS FITTING EXPRESSION
Among those who reject the view that rioting is mere “lawlessness” or “criminality,”45 the phenomenon is widely held to be a form of political protest. Taking this view is, even by itself, theoretically fruitful. From this frame, it has been argued, for example, that riots qua protest can be ways of constituting an audience—vocative speech acts, like yelling “Hey, listen up!” in a noisy room—when a group is otherwise unheard;46 of disrupting the “motivated ignorance” of privileged political groups;47 of enacting expressive “forms of democratic repair”;48 and of expressing “symbolic impure dissent” to political oppression in ways that realize intrinsic value, support self-respect, and perform similar non-instrumentally valuable functions.49 As Pasternak notes, however, and in line with the discussion of the previous section, the non-instrumental ends realizable by riots are not of the kind typically thought to justify the imposition of harms on others. The primary justificatory burden facing riot participants remains, then, the justification of the characteristic harms that rioting involves.
My view is that the expressive aims—and the corresponding expressive demands—of rioting, understood as a form of protest, can themselves sometimes directly license many of rioting’s characteristic harms. Rioting, when it is justified, is a way of engaging in fitting political expression, a way of giving proper voice to the family of interrelated attitudes and emotions that often give rise to it. Justified rioting, I shall argue, is a way of adequately saying what it is fitting to say—a way of meeting the gravity of the moment, given the expressive avenues available.
As we shall see, this view yields standards that are in several respects like those yielded by the view of rioting as defense—though only, I shall argue, at surface level. In other respects, they are importantly different. First, they locate the primary site of application of the normative principles governing rioting at what I take to be the correct site: individual decisions about whether and how to participate in rioting. Second, they tie the justification of rioting closely to its central aims and nature as an expressive practice, including the dimensions along which the justification of rioting varies. And third, these principles permit (and indeed demand) more riotous activity than the principles of defense permit, in ways that avoid the objections to which the view of rioting as defense is vulnerable.
IV.A. Expression
When we call rioting a form of protest, the implication is that (at least some of) the harmful and destructive acts characteristic of the practice are themselves expressions of dissent. For it is not as though riots involve marches, slogans, and signs on one hand and the destruction of cars and buildings on another. To participate in a riot is to engage in the latter sort of activity. Burning a car in protest, on this understanding, is like the act of resigning in protest: the message is expressed by, not in addition to, the act.
Unlike protest via signs and slogans, however, the message expressed by the burning of cars is typically not spelled out for all to see; and unlike resignations-in-protest, the breaking of shop windows is typically not accompanied by a formal objection to clarify its meaning. If riots are expressive episodes of protest, then what, more precisely, do riots say?
It should be acknowledged that this question may not always yield a determinate answer. Riots vary in the extent to which their expressive content is unified versus disunified (or more aptly, polyphonous versus cacophonous) precisely because what is collectively expressed is a product of many individual, self-directed, uncoordinated expressive acts taking place in the context of a chaotic form of joint action. Whether and to what extent a particular riot can be said to say something determinate will depend on where it falls on the spectrum between these two extremes.
Evidence on this question supports the view that rioters can be relatively unified in both their basic grievances and in what they collectively express. Researchers who studied the attitudes of participants in the 2011 U.K. riots, for example, wrote that in response to “a range of grievances,” it was:
anger and frustration that was being expressed out on the streets…They expressed it in different ways, but at heart what the rioters talked about was a pervasive sense of injustice.50
Similarly, the Kerner Commission, which interviewed several hundred residents of riot-affected neighborhoods following the Detroit and Newark riots of 1967, concluded that those events were:
linked in the minds of many…with a reservoir of underlying grievances…[resulting from] discrimination, prejudice, and powerlessness…[and] severely disadvantaged social and economic conditions.51
These large-scale research findings match individual accounts of other riots. Writing just after the Stonewall riots in the East Village Other, for example, one participant wrote that “we have reached the bottom of the oppressed minority barrell [sic]” and described the “predominant theme” of the first three nights of unrest as, simply, “this shit has got to stop!”52 Likewise, Connie Griffin, a participant in the Watts Riots of 1965 said:
Nobody wants to lay down and get kicked in the head. As long as we lay down, we know we’re gonna get kicked. We’ve got enough sense to know it. This is what the noise is about.53
We can, I think, draw some generalizations from these observations. Note first that rioting often involves a claim to be heard by people who take themselves to be, in the relevant respects, ignored—that is, wrongly socially or politically marginalized. This is supported by King’s famous dictum that riots are “the language of the unheard”54 as well as by recent work on the linguistic function of rioting.55
As I argued earlier, these claims to be heard are best understood as directed expression: claims to someone who (it is thought) ought to listen, which can be appropriate to make even when one does not expect that they will be heeded. Who this someone is will vary with social and political circumstances but will typically be a regime, government, or its agents, as well as, in democratic societies, one’s fellow citizens. Recall that directed expression is, I have argued, neither a shout into the void (mere expression), nor is it a direct attempt to gain an understanding or accommodating response from its target (robust communication).
Note too that while the specific grievances expressed by individual riot participants likewise vary, they are often united in kind. Rioters characteristically express frustration (at their oppression, economic position, or other aspects of felt injustice) and anger (at these conditions and at the agents taken to be responsible for them) as well as a family of related attitudes and emotions: despair, disappointment, hopelessness, and so on. Beyond these specific attitudes, moreover, is the sense that these conditions are not merely wrong but intolerable. Riots are ways of saying “no” or “no more,” that “this shit has got to stop!”—often despite (and indeed because of) having no way to make it stop. These are, at bottom, simple messages—“Listen!” “This is wrong!” “No more!”—whose simplicity makes them amenable to expression via blunt means.56
IV.B. Fit
In many of the conditions of oppression that characteristically give rise to rioting, the messages expressed are ones that are worthy of being expressed. This is true of protest generally. In some circumstances, however, to express these messages adequately—in a way that meets the gravity of the circumstances—it may not be enough to paint these words onto signs and carry them out into public. Rather, forceful expressive acts, like overturning a car, or burning a building, or destroying parts of a ghetto, may be what is called for. These bits of normative vocabulary—what is “worthy” of being said or done, what is “adequate” to the circumstances, what is “called for”—are pieces of the normative vocabulary of the fitting.57
I take the fitting to be a basic normative category,58 and I take the relation of fit to be one of metaphysical matching or suiting between two objects. More precisely, fit is the relation that stands between an object and a response to that object when, as Christopher Howard puts it, “the object merits—or is worthy of—that response.”59 As I use the term, whenever we say that something is the “fitting” response, we can equivalently say that it is “correct,” appropriate,” “called for,” “demanded,” and other synonyms. For example, gratitude is the correct response to kindness, and blame is the appropriate response to what is blameworthy. Likewise, I claim that rioting can be the correct response to the conditions that give rise to it.
The category of the fitting yields normative claims, but they are distinct from familiar normative claims about value or duty. When we say that something is worthy of being done or said in this sense, we do not mean that doing so would be valuable, nor that it is what duty requires. Rather, we mean that it is the appropriate response in the circumstances. Correspondingly, when we say that it would be fitting to do or say something, we do not mean simply that it would be permissible to do or say it, but neither do we mean, typically, that it must be done or said. In this way, fitting responses appear to bear a distinctive, intermediate deontic status.60 When we say that some fitting response is called for or demanded, we make this kind of intermediate normative claim. This status can justify acts. To say that something is worthy of being done or said is one way of saying it should be said or done, and thus also one way of (defeasibly)61 justifying doing or saying it.
This kind of justification is, like others, subject to being overridden by all-things-considered considerations. Thus, if it is fitting to say something, but saying it would cause great harm to innocents, it may be that I ought, all things considered, not to say it. As a demand, it is also subject to prerogatives to do otherwise: I might permissibly opt not to say it, despite its being called for, if doing so would cause me distress that I would prefer not to endure.
Fit has, like other normative standards, necessary conditions for application. I’ll discuss four here: type-appropriateness, proportionality, correct-directionality, and adequacy. I’ll also discuss the sense in which a necessity condition may be said to apply to rioting. Clearly, these are facially similar to the conditions governing war and defense; I explain below what makes them different. In the case of rioting, the application of these conditions is complicated by the fact that that what is expressed by a particular act of rioting will be multiform, both in that its expressive targets may be multiple (the proprietor of this shop whose window is being smashed, say, but also the wider democratic community) and in that it is at the same time part of a joint expressive act whose content depends on its many constituent acts.
IV.B.1. Type-Appropriateness
To say that a fitting response must be of the appropriate type is just to say that one’s response must be the right kind of response to the thing responded to. Gratitude is a fitting type of response to kindness, and blame is a fitting type of response to wrongs—and this is true whether or not any particular instance of gratitude or blame is fitting. In the case of fitting expression, this means that what is said must be the fitting type of thing to say in the circumstances. I’ll say more in Section IV.C. about what makes rioting an appropriate type of expressive response to some social injustices. Here, I want to note the practical constraints this condition places on riot participants.
First, the individual acts of riot participants should themselves be of an appropriate type: they should express fitting types of attitudes in response to the injustices they protest. This may seem an obvious point to make. However, it is at the core of some moral disagreement about rioting, including whether “looting” is ever appropriate—a question about whether taking goods legally owned by another could be a type-appropriate response to conditions of social injustice. When individual acts in riots have multiple expressive targets, what is expressed to each should likewise be type-appropriate.
Second, the joint act of rioting should be expressively type-appropriate. However, since no one directs the joint act, no one can make it the case that the joint act is type-appropriate, and so individual rioters must judge for themselves whether this is (likely to be) the case and direct their own participation accordingly. Since the “mood” of many joint activities is discernible by participants—think not only of riots but of concerts, large sporting events, and so on—I believe these judgments are ones rioters can reasonably make. Those joining in the Stonewall Riots, for example, could judge well enough that the riots were about sexuality-based oppression, that the kinds of attitudes and emotions being expressed constituted defiance of this oppression, and whether the latter were fitting types of responses to the former. If, however, a riot has no determinate expressive content, as in the case of what I earlier termed cacophonous riots, it could not be expressively type-appropriate; and if a riot’s mood cannot be discerned by prospective participants, they could not justifiably participate on this basis.
Third, individuals’ acts should be consonant with what they judge to be the character of the joint act. That is, participants should aim for their individual expressive acts to contribute appropriately to what is jointly expressed. There is something clearly defective, for example, about the justification of someone who participates in an expressively fitting riot only to seek, say, the social esteem of their friends.
IV.B.2. Proportionality
Fitting responses must also be proportionate in their relative magnitude. For example, it would not be fitting to react furiously to some minor slight, even in cases when some measure of anger or resentment is called for, because severe anger or resentment is a disproportionate response to minor wrongs. Similarly, participants in riots should avoid acts that would be disproportionate to the wrongs or injustices to which they respond.
Proportionality is also, of course, a condition on justifiable war and defense. It is important to note, however, that the comparands in those cases are very different. Proportionality in war and defense involves a forward-looking comparison between harms imposed and defensive aims likely to be thereby achieved. By contrast, proportionality in fit is responsive: fitting gratitude responds proportionately to the magnitude of a kindness, fitting blame responds proportionately to the magnitude of wrongdoing, and fitting rioting (I propose) responds proportionately to the magnitude of conditions of injustice.
Individual acts should be proportionate with respect to each of their expressive targets. This is important in the case of riots because the expressive harms of rioting often fall only indirectly on their primary expressive targets, or because these targets are groups made up of many individuals. It might be clear, for example, that breaking a shop window is expressively proportionate in response to conditions of deep social injustice, while remaining unclear whether it is proportionate with respect to what is expressed to the individual shopkeeper. Insofar as the shopkeeper is an appropriate target of an expressive harm (more on which below) whose wider aim is to say something to a group of which the shopkeeper is only one part, the harm must be expressively proportionate in how it addresses each of its expressive targets.
Judgments of proportionality also apply to the joint act of rioting. Since, again, no one directs the joint act, individuals must judge that the overall expressive force of the riot is not disproportionate to the injustice to which it responds. Since rioting often responds to very serious injustice, it will often be the case that rioting could not realistically be forceful enough to become disproportionate to the injustice it protests. For example, no actual riot, I believe, could have been expressively disproportionate to the magnitude of racial injustice faced by Black Americans in the mid-20th century.
Judgment is more difficult when we consider proportionality with respect to the various other targets who may bear rioting’s characteristic harms. If enough rioters impose disproportionate harm on local businesses, for example, that may be sufficient to make the joint act disproportionate in this dimension. Judgments about such issues, while sometimes difficult, can nevertheless, I believe, often be made. For, unlike such judgments in defense, where the comparand of proportionality is some future policy outcome, rioters engaging in expressive protest must judge whether ongoing collectively imposed harms have exceeded what proportionality permits. Besides a robust sense of their collective grievances, riot participants will often be in a position to judge how widespread and how violent the group of which they are a part is, as well as how unified the group is in its expressive aims. When an individual rioter cannot reasonably make such judgments, however, their participation could not be justified on these grounds.
IV.B.3. Correct-Directionality
Fitting responses must also be correctly directed. When gratitude is appropriate in response to a kindness, for example, it is gratitude to the kind person, not gratitude to anyone or gratitude in general, that is called for. Similarly, it was important that the participants in Detroit’s 1967 riot targeted property in Detroit rather than property just across the Detroit River in Windsor, Ontario. Conversely, a natural thought is that it is the culpability for wrongs of some—the misconduct of the police, say, or the unjust indifference of the majority—that makes them appropriate targets of the kinds of messages rioting characteristically expresses. This is like the principle of liability in defensive ethics, which requires that defensive acts harm only those who are proper targets of such harms.
Liability and correct-directionality are not, however, entirely analogous, and two distinctive features of the correct-directionality requirement are worth noting. First, it is helpful to distinguish between the addressee and the audience of directed expression, which often come apart in cases of protest (generally) and rioting (specifically). For example, the addressee of a protest might be an absent political leader—“hey, hey, LBJ, how many kids did you kill today?”—while its audience might be other assembled citizens. Similarly, a riot participant might appropriately write “stop killing black people” on the wall of a private citizen’s house, despite the apparent mismatch between the message’s addressee (the police) and the citizen that bears its cost (one member of its audience).62 Standards of fit apply both to expression’s addressee and audience, and thus, each aspect of such a message may or may not be correctly directed. Second, while both the correct-directionality requirement and the liability principle can explain why harms should target those responsible for wrongs, the correct-directionality condition can also explain why, in some cases, it is appropriate for rioters to target the very ghettoes in which they reside (which are of course not liable to be destroyed)—including when doing so causes more harm than good. To put the point simplistically, if one is wrongfully confined, it can be expressively appropriate to destroy one’s prison, even if it is one’s only shelter.63
Whether and to what extent someone or their property is an appropriate target of expressive harms in the context of rioting will often depend on complex and difficult questions, particularly in democratic societies, about democratic complicity and individual responsibility for collective injustice. Ultimately, these questions demand substantive answers not decidable here, and one’s particular view about the extent of permissible rioting will simply (partly) depend on them. However, it is clear that expression that targets responsible collectives more directly will permit proportionately more than expression that targets only collectives’ individual members. This offers one explanation for the common judgment that in riots, greater harm to government property or to large corporate interests is more clearly permissible than similar harm to private property and persons, as well as the common judgment that rioting targeting property is permissible to a greater extent than rioting that targets persons.64
Things are more complex, and depend more fully on one’s stance on further questions both moral and empirical, when we consider the appropriateness of harms to individual co-citizens and their property, or to small businesses. If rioters’ co-citizens are complicit in the wrongs rioters protest, and if this complicity makes them culpable for these wrongs, then they would be appropriate targets of riotous protest. A more common view is that co-citizens are usually not culpable for wrongs committed in their names but are nonetheless at least partially responsible for them. (Pasternak elsewhere defends such a view.65) In such cases, harms to co-citizens’ property might be appropriate if they are necessary to adequately protest the wrongs of the collective. Similar remarks apply to small businesses, which may not be culpable for economic injustices, but which may nevertheless bear some responsibility for them. (This is arguably the claim made by the film Do the Right Thing.66) Moreover, even when co-citizens are not responsible, solidarity might demand of them that they accept some costs, such as damage to their property, if it is necessary to adequately protesting injustice. These considerations apply most clearly in democratic societies. In non-democracies, the harms of rioting should more clearly be confined to government and other public targets.
Rioting also sometimes involves serious physical harm to individuals. Even when co-citizens are complicit in serious injustice, this seems unlikely to make them appropriate targets of serious physical harm, since serious physical harm would often be disproportionate to any individual’s contribution to injustice. In some other scenarios, however, serious harms to individuals strikes me as possibly appropriate. If one’s fellow citizens are not merely complicit, but are active agents of extreme injustice, such harms to them might be expressively appropriate. (Think of members of the Klu Klux Klan in the American South, for example.) To a lesser extent, riot police, when they act as agents of injustice, also strike me as appropriate targets of some measure of expressive harm.67
This condition is complex, and may seem to demand more discernment than the chaotic realities of rioting characteristically allow. To the contrary, however, it is worth emphasizing that riot participants do regularly make substantive judgments of this kind, distinguishing different targets according to their perceived responsibility for, or complicity in, the wider wrongs protested.68 These constraints are thus in line with what many, including riot participants, take to be the proper moral constraints on rioting.
IV.B.4. Adequacy
The most distinctive condition on fitting response is the adequacy condition, which demands that the magnitude of a fitting response be adequate to what it responds to. For example, to give fitting thanks requires doing what would adequately convey one’s gratitude, and to express fitting regret, an adequate apology is what is called for. These examples are not about what would suffice for the relevant expression, but about what kind and magnitude of expression adequately meets the expressive demands of the circumstances. This means that more than what is strictly necessary to accomplish expression is often demanded of fitting expression. In this way, the adequacy condition exerts upward pressure on the magnitude of fitting responses.
Note that the adequacy condition on fitting response describes a relation between a response and what it responds to. This distinguishes it from other ways of describing something as “adequate” that describe other relations. For example, we might call a disproportionate defensive response “inadequate” in the sense that it does not achieve its defensive aim “well enough” to justify the harm it imposes. But this sense describes—more specifically, negatively assesses—a relation between costs (harms) and benefits (defensive aims). Whether some response is adequate in the fittingness sense is not about costs or benefits, only whether the response meets the magnitude of what it is a response to. Fitting anger is anger adequate to the wrong it responds to; fitting thanks is an expression of gratitude adequate to the kindness it responds to; and so on. Likewise, a fitting expression of protest, including fitting rioting, is expressively adequate to the wrong protested, independent of any further considerations of the costs or benefits of the protest.
This is crucially important with respect to the moral assessment of rioting. The view proposed here demands that acts of rioting adequately protest conditions of injustice by saying what is worthy of being said, and by saying it loudly and forcefully enough. The question facing prospective rioters is thus, “What will say what is worthy of being said—‘Listen!’ ‘This is wrong!’ ‘No more!’—well enough?” In this way, it is concerned not with achieving justice per se but rather with adequately protesting injustice. If this contributes to social change, so much the better; but even if it does not—even if one has “gone past caring” about achieving change—these messages remain worthy of being said.
Now, if adequacy demands responses of a certain magnitude, are responses which fall short of this wrongful? This may seem counterintuitive, for surely the oppressed do not act impermissibly whenever they choose not to protest their oppression. However, while the demand of adequacy calls for certain kinds of protest, it does not require them. Moreover, victims of oppression have a prerogative to do less than would adequately protest their oppression, and considerations of prudence may affect what the oppressed have most all-things-considered reason to do.
Like the others, this condition applies both to individual acts and to the joint act of rioting. Practically speaking, this means that in addition to acting individually so as to adequately expressively respond to the wrongs rioting protests, riot participants should aim for their individual acts to contribute to a collectively adequate expression of protest. This imposes only a weak practical requirement on individuals, however, since for reasons just noted, adequacy is primarily normatively permissive rather than restrictive. Indeed, considering grave wrongs to which we might think no expression of protest could ever fully adequately respond, it seems clear that some expression of protest remains not only permitted but worthy of expression.
The idea that, in order to say what is fitting to say, an act may be called for, and moreover that this might justify harmful acts, may seem extreme. After all, we often sharply distinguish between what one may say to someone and what one may do to them. But philosophers have long recognized that speech may constitute acts and that acts may carry principally expressive dimensions; and moreover, there is a robust history, particularly in the philosophy of law, which recognizes that expressive acts may themselves be objects of fit.69 For example, Anderson and Pildes, defending what they call an “expressive theory of law,” write that:
There are some things we can express only with deeds because words alone cannot adequately convey our attitudes.70
And Feinberg claims that expressive demands are what justify the harms involved in punishment, and that this is (partly) because:
when [the state] speaks by punishing, its message is loud, and sure of getting across.71
This idea remains controversial, of course, in legal philosophy. How, it is asked, could the expressive value of condemning criminal acts justify the harsh treatment characteristic of punishment, given the power of the state and its many attendant alternatives to such acts?72 But the thought that there must be some important connection between the expressive function of punishment and the harms it imposes is common even among those who are critical of purely expressive theories.73 Moreover, these concerns lose much of their force when the expressive agent is not the state but rioters—the oppressed, the marginalized, and the powerless. When rioting really is the last resort of the powerless, we cannot reasonably point to alternative means of expression and argue that those should be pursued instead,74 nor can we argue that rioters should invent some alternative, less harmful means of expression that can adequately say what it is fitting for them to say. Rioters, it is plausible to think, really do express with deeds what cannot be adequately said with words, and do so in the only way that is, in their positions, loud enough, and so able (perhaps) to get their message across.
IV.B.5. Necessity
It is worth asking, finally, whether fitting responses might also be subject to a necessity condition. In earlier work, I suggested that they are not.75 For, whereas necessity in defense imposes a strict upper limit on permissible force—roughly, only what suffices to successfully defend oneself—adequacy often seems to demand more than a successful response necessitates. For example, to successfully express rejection of a wrong, one need only say (or even whisper) “no.” But an adequate expression of rejection might demand much more than this.
While I stand by these observations about the demands of adequacy, I now believe I was mistaken about necessity. Necessity limits the harm permissibly inflicted in defense in light of the aim of that practice: successful defense. The aim of the practice of fitting expression is not, as suggested by the train of thought above, successful expression—it is, rather, successful fitting expression. Put generally, the aim of fitting response is not successful response but successful fitting response. But there is no conflict between what is demanded of an adequate response and what is necessary to a fitting response, since adequacy is part of what makes a response fitting. Put simply, we might claim that only what is necessary to a fitting (and hence adequate) response is permitted.
Should we make this claim? At least when fitting responses impose harms, we should. It is plausible to think that the underlying motivation for the necessity condition in war and defense is the avoidance of gratuitous harm. We should be concerned with avoiding gratuitous harm everywhere, not only in war and defense. If two expressive acts fittingly (and hence adequately) respond to some injustice, but one is more harmful than the other, then clearly the less harmful act should be chosen instead. This consideration is important when we consider rioting, since it supports the widespread view that rioting should be a measure of last resort.
***
I have argued that fitting rioting must be a type-appropriate, proportionate, correctly directed, and adequate response to what it protests. In addition, it should avoid inflicting gratuitous harm, beyond what is expressively adequate. Importantly, these principles are all properly applied at the level of, and are generally epistemically available to, individual riot participants. Riot participants characteristically possess a robust sense of the grievances that motivate their protest, as well as a sense of whom their protest is properly directed towards. This, together with moral judgment, can yield correct judgments regarding both the proportionality and adequacy of decisions to participate in rioting.
Since the expressive justifiability of rioting depends not only on the fittingness of individual acts but also on the fittingness of rioting as a whole, there remain epistemic barriers to justifiable participation similar to those discussed critically in Section III. In some cases, these barriers may be great enough to defeat justifiable participation in rioting. But unlike the view of rioting as defense, the present view does not require participants to make reasonable judgments about the future policy effects of a riot, only that they make reasonable judgments about a riot’s ongoing mood and character. While these judgments may sometimes be difficult to make, they are an order of magnitude easier than judgments about rioting’s eventual effects, and I believe that in characteristic cases of rioting, they can typically be made. (Compare: it is one thing to judge the mood of a crowd at a sports game, and another to judge how that will help or hinder the home team’s chances of victory.)
IV.C. Rioting
When rioting is justified, it is a response to real and serious grievances—deeply objectionable defects in the social or political order, which are, moreover, often unacknowledged by those in power or those who keep them there (such as a democratic majority). Serious wrongdoing calls for protest, and the deeply objectionable calls for objection, both in interpersonal relations and in political society. Facing such circumstances, it is fitting to demand that one’s grievances be heard and to demand that ongoing wrongs cease. This is so not primarily as a matter of duty, nor because of any prospects for righting wrongs. It is so because it is fitting. It is fitting because to respond to those who do wrong by making claims that they stop just is how one responds appropriately to the way one is treated by others, just as (in other circumstances) one might respond appropriately to kindness with gratitude, wrongdoing with blame and resentment, and so on.
These are substantive claims about what is fitting to say and do. They are supported, I believe, by intuition, but I have said little about the deeper question of what makes them true. This deeper question—whose answer requires appeal to a substantive normative theory of fit—is important but impossible to fully answer here. I can, however, make some remarks in this direction.
Strawson claimed that engaging in interpersonal responsiveness is what it is to be a “participant” in interpersonal morality, to bear the “strains of involvement,” and to be “humanly connected” to one another, which, he says, is rooted “in our human nature and our membership of human communities.”76 In other words, to respond to one another with claims and reactive attitudes is how we live together as members of a moral community. In a parallel vein, engaging in the exchange of distinctively social and political claims and reactive attitudes is, I believe, a core part of how we live together in political communities. When we are denied regard as proper sources of political claims and proper objects of political concern, it seems clear that the appropriate response is to insist, loudly if necessary, that this is seriously wrong.
Moreover, to insist upon this loudly enough may demand action. When I consider, for example, the de facto and de jure apartheid, ghettoization, and racial terror to which so many Black Americans were subject in the mid-20th century, it seems clear that adequate protest of these very deep injustices required more than words alone could have achieved: dramatic, forceful, and indeed violent acts of protest were called for instead.
This is supported by the thought, common to many traditions in political philosophy, that responsiveness—to the subjective or objective interests of subjects—is a primary virtue of good government.77 This is a pillar of the democratic tradition, which makes proper embodiment of this kind of responsiveness the core of its institutional design and moral justification, but it is also present even in non-democratic theories, including (for example) Plato’s78 and those of the social contract tradition. Rioters often describe their motivations as having to do with a feeling of collective invisibility or voicelessness in political society. The kind of invisibility and voicelessness in question is, very plausibly, incompatible with any suitably articulated norm of responsiveness. Clearly, in non-democratic societies, insisting upon visibility and voice may be especially urgent. In democratic societies, there is a different but also very deep significance to claims that one’s interests are not taken account of, and are indeed ignored, in public affairs. When one’s urgent social and political claims fail to be heard or taken account of, it seems that, in any political system, insistence on the opposite is the correct response.
But even if it is fitting to say the things that rioters characteristically say through their rioting, is it also fitting to say so in the way that rioters characteristically say them? Is violent protest the appropriate response to failures of responsiveness? For, while destruction and confrontation may be one way of speaking loudly, there may be other, less harmful responses that are also adequate to the circumstances and also capable of attracting attention. This was King’s argument against early Black nationalist calls for violent resistance, which he saw as posing a false dilemma between being “cringing and submissive” on one hand and taking up arms on the other. Are there not, as he put it, “other meaningful alternatives”?79
It is important to emphasize that nonviolence movements, like Gandhi’s and King’s, can indeed succeed in adequately insisting upon and demonstrating the dignity and righteous complaints of the oppressed. However, the presence of these examples, powerful as they are, does not license the claim that would-be rioters must opt for nonviolent means of protest instead. This is for several reasons. First, as the bloody histories of Indian independence and the American civil rights movement demonstrate, nonviolence movements are often extremely costly to those who participate in them—they are “nonviolent” only in the sense that nonviolent activists commit not to impose violence on their oppressors. People have no duty in such cases to take the more dangerous path. Second, such movements’ prospects for success are uncertain and depend on a wide range of contextual factors. Even considering only expected results, then, it is not clear that nonviolence is always to be preferred. Third, part of the power in nonviolence, I believe, lies in the way in which it responds to wrongdoing with an act of grace rather than with a fitting response to the wrong. (To “turn the other cheek” is not to engage but rather to rise above.) If this is correct, then it cannot also be claimed that nonviolence is the more fitting response; it must be preferable on other grounds. Finally, and most decisively to my mind, nonviolence movements are just that: social movements, which demand deep coordination, commitment, and leadership, and which take time, effort, and good fortune to successfully build. But in the circumstances in which rioting characteristically “erupts,” organized social movements that could adequately protest the wrongs in question are often unavailable. We cannot claim that people must opt for an alternative that does not meaningfully exist.
Similar remarks apply to other still forceful but less harmful means of protest: boycotts, strikes, and organized “uncivil” disobedience, for example. When movements do exist that can coordinate these forms of protest, and when these forms of protest would adequately answer the conditions of oppression being protested, then people ought to participate in them rather than resort to rioting. For, in such cases, resorting to rioting, and thereby inflicting its characteristic harms, would be gratuitous. Moreover, even when such forms of protest would not be adequate, if they nevertheless seem likely to actually improve conditions, perhaps they should be opted for anyway. This would be like the way that, on some views, practical reasons to forgive can win out even when forgiveness does not seem apt.
But it is important not to overstate these limitations. One might be tempted, for instance, to argue that participants in the American race riots of the 1960s ought instead to have joined King’s nonviolence movement. King’s movement, however, addressed issues that were relevant largely only in the South: ending racial terrorism and securing meaningful formal political rights for Black citizens. Rioting, by contrast, took place largely in the North and West, where formal political rights were more robust and lynching was much rarer, but where substantive exclusion, abuse, inequality, and other forms of oppression remained widespread. While this simplifies a complex historical case, the point remains that rioting voiced distinctive grievances that participation in King’s movement could plausibly have been judged to neither effectively nor adequately address. Rioting is, I claimed earlier, characteristically an “eruptive” reaction to moments of inflection in wider struggles in which a reservoir of underlying grievances becomes too much to tolerate. These intense moments of political affect often naturally arise because alternative modes of pursuing change seem fruitless or unavailable.
V. CONCLUSION
Rioting is often a form of protest. When things are bad enough—when the wrongs one suffers are intolerable, and when change seems out of reach—it is worth protesting these facts loudly; worth insisting, in the words of one participant in the Stonewall Riots, that “this shit has got to stop!”80 I have argued that this fact can itself license many of the harms that rioting characteristically involves. The violence that distinguishes rioting as a form of protest is how it speaks loudly—how it says what is worthy of being said, and says it in a manner befitting the seriousness of the circumstances.
The benefits of this approach are several. For one thing, it ties the justification of rioting to its distinguishing characteristics as a social practice. When rioting is justified, it is justified by its distinctive expressive power and force, not in spite of it. For another, it makes this justification available to individuals, and indeed to individuals possessing the limited information they characteristically possess. It is, moreover, responsive to moral reflection engaged in by riot participants themselves about which particular acts of rioting are justifiable. Finally, the normative verdicts yielded by this approach align with common intuitions about which episodes of rioting could be justifiably participated in, and along which dimensions this justification varies. Rioting is most clearly justified when the wrongs it protests are most clearly worth protesting, rather than when it is most likely to receive sympathetic or accommodating responses, or otherwise to achieve liberatory aims.
Rioting is also often the last resort of the powerless—a way of responding to wrongs that cannot be forestalled or overcome but, equally, cannot be tolerated. But even then, it is not a way of redressing those wrongs. It is, rather, a way of insisting loudly that they are wrong, and that they should cease. These are things worthy of being said, and that, I think, may sometimes be justification enough.
Notes
- Martin Luther King, Jr., “Showdown for Nonviolence,” Look 32 (April 1968): 23–25. ⮭
- I offer a more complete characterization of rioting in Section II. I confine my discussion here to political rioting, insofar as it may be thought to be a distinct subclass of a more general phenomenon. Some prefer the terms “uprising” or “revolt” as a way of avoiding or countering the historically negative connotations of the word “riot.” However, because rioting is, in my view, a distinctive mode of social protest that these more general terms fail to capture, I believe it is worth using its distinctive name. ⮭
- An early recorded example is the Nika riots of 532 AD, which reportedly saw half of Constantinople burned, led to 30,000 deaths, and nearly resulted in the fall of Emperor Justinian; see Geoffrey Greatrex, “The Nika Riot: A Reappraisal,” The Journal of Hellenic Studies 8, no. 117 (1997): 60–86, https://doi.org/10.2307/632550. On other historical cases, see especially George Rudé, The Crowd in History: A Study of Popular Disturbances in France and England, 1730-1848 (New York: Wiley & Sons, 1964) and E. P. Thompson, “The Moral Economy of the English Crowd in the Eighteenth Century,” Past & Present 50, no. 1 (1971): 76–136, https://doi.org/10.1093/past/50.1.76. On the distribution of riot events globally and by regime type, see the ACLED Database (Clionadh Raleigh, Andrew Linke, and Håvard Hegre, “Introducing ACLED-Armed Conflict Location and Event Data,” Journal of Peace Research 47, no. 5 (2010): 651–50, https://doi.org/10.1177/0022343310378914) and the Mass Mobilization Project (David Clark and Patrick Regan, “Mass Mobilization Protest Data” (Harvard Dataverse, 2016), https://doi.org/10.7910/dvn/httwyl). ⮭
- The list includes, among many others, Barack Obama, Donald Trump, Recep Erdoğan, and Carrie Lam. On the attitudes of the public, see YouGov, “Yahoo News/YouGov Presidential Election Poll 8/27–28/2020: Toplines,” https://docs.cdn.yougov.com/7oai6ahvoj/20200828_yahoo_election_toplines.pdf, 2020. ⮭
- This is the view taken most notably by King (“Showdown for Nonviolence”) and by many academics and policymakers; see, e.g., McCone Commission, “Violence in the City: An End or a Beginning? A Report of the Governor’s Commission on the Los Angeles Riots” (California State Archives, 1965); Philip Meyer, “The People Beyond 12th Street: A Survey of Attitudes of Detroit Negroes After the Riot of 1967” (Detroit Urban League and Detroit Free Press, 1967); Kerner Commission, “The Report of the National Advisory Commission on Civil Disorders” (US Government Printing Office, 1968); Robert M. Fogelson, Violence as Protest: A Study of Riots and Ghettos (Garden City: Doubleday, 1971). ⮭
- Avia Pasternak, “Political Rioting: A Moral Assessment,” Philosophy & Public Affairs 46, no. 4 (2019): 384–418, https://doi.org/10.1111/papa.12132; Jonathan Havercroft, “Why Is There No Just Riot Theory?” British Journal of Political Science 51, no. 3 (2021): 1–15, https://doi.org/10.1017/S000712342000085X. Pasternak proposes that in making a “moral assessment” of political rioting, which she calls “a form of defensive harm,” we should “examine [rioters’] actions in light of the various constraints offered by just war theorists” (386; see also 398); and Havercroft claims that to offer a “normative defense of rioting” would be to offer “a theory of a just riot” (3), which his paper aims to present. Pasternak and Havercroft, whose views I discuss at length here, offer the only sustained philosophical discussions of the moral dimension of political rioting. Friendly amendments to Pasternak’s view are proposed by Ricky Mouser, “How to Read a Riot,” Journal of Ethics & Social Philosophy 26, no. 3 (2024): 445–68, https://doi.org/10.26556/jesp.v26i3.2316 and Thomas Carnes, “Privileged Citizens and the Right to Riot: A Reply to Pasternak,” Journal of Ethics and Social Philosophy 26, no. 3 (2024): 633–640, https://doi.org/10.26556/jesp.v26i3.3086. Partial discussions of the phenomenon are also offered by Stephen D’Arcy, Languages of the Unheard: Why Militant Protest Is Good for Democracy (London: Zed Books, 2014); Candice Delmas, A Duty to Resist: When Disobedience Should Be Uncivil (Oxford: Oxford University Press, 2018); Juliet Hooker, “Black Lives Matter and the Paradoxes of U.S. Black Politics: From Democratic Sacrifice to Democratic Repair,” Political Theory 44, no. 4 (2016): 448–69, https://doi.org/10.1177/0090591716640314; and Tommie Shelby, Dark Ghettos: Injustice, Dissent, and Reform (Cambridge, MA: Harvard University Press, 2018). Matthew Chrisman and Graham Hubbs, using tools from the philosophy of language, address rioting’s communicative aspect but officially set aside the matter of moral justification (“‘The Language of the Unheard’: Rioting as a Speech Act,” Philosophy & Public Affairs 49, no. 4 (2021): 379–401, https://doi.org/10.1111/papa.12201). Similarly, Alexis Bibeau-Gagnon argues for the “democratic potential” of riots as a means of political representation (“Can Riots Represent? A Democratic Theory,” American Journal of Political Science, forthcoming, https://doi.org/10.1111/ajps.12989). Rioting is discussed but dismissed by Hannah Arendt, who claims that violence is “instrumental by nature” and that riots “can only promote violence in the end” (“Reflections on Violence,” New York Review of Books, February 1969); by John Rawls, who suggests that resistance should be either nonviolent or else revolutionary (A Theory of Justice, rev. ed. (Cambridge MA: Harvard University Press, 1999), sec. 55); and by Michael Walzer, who (unfathomably) calls riots “spontaneous terror” and rioters “spontaneous terrorists” (Obligations: Essays on Disobedience, War and Citizenship (Cambridge: Harvard University Press, 1982), 65–66). Outside of political philosophy, E.P. Thompson offers an important historical account of what he calls the “moral economy of the crowd” in 18th-century England (“The Moral Economy of the English Crowd”) and Fabien Truong discusses some moral aspects of rioting from the standpoint of sociology (“Total Rioting: From Metaphysics to Politics,” The Sociological Review 65, no. 4 (2017): 563–77, https://doi.org/10.1111/1467-954X.12436). ⮭
- Why I prefer this way of speaking will become clear later on. ⮭
- Pasternak, “Political Rioting,” 386. ⮭
- Havercroft, “Why Is There No Just Riot Theory?,” 2. ⮭
- E.g. Joel Feinberg, “The Expressive Function of Punishment,” The Monist 49, no. 3 (1965): 397–423, https://doi.org/10.5840/monist196549326; Elizabeth Anderson and Richard Pildes, “Expressive Theories of Law: A General Restatement,” University of Pennsylvania Law Review 148, no. 5 (2000): 1503–75, https://doi.org/10.2307/3312748; Edmund Tweedy Flanigan, “Futile Resistance as Protest,” Mind 132, no. 527 (2023): 631–58, https://doi.org/10.1093/mind/fzad015. ⮭
- Martin Luther King, Jr., “The Other America” (speech, Stanford University, 1967), Civil Rights Movement Archive, https://www.crmvet.org/docs/otheram.htm. ⮭
- Many make this point. For example, Daniel Myers writes that “although most social scientists have an intuitive sense of what constitutes a riot, the edges of the definition are fuzzy and it can be difficult to determine whether or not some events are actually ‘riots’” (“Riots, Race,” in The Wiley Blackwell Encyclopedia of Race, Ethnicity, and Nationalism, eds. John Stone, Rutledge M. Dennis, Polly S. Rizova, Anthony D. Smith, and Xiaoshuo Hou (Oxford: John Wiley & Sons, 2015), 1, https://doi.org/10.1002/9781118663202); Thompson writes that “it has been suggested that the term ‘riot’ is a blunt tool of analysis for so many particular grievances and occasions…it is also an imprecise term for describing popular actions” (“Moral Economy,” 107); and Eric Hobsbawm writes that “the mob” (as he calls it) “is a particularly difficult phenomenon to analyse in lucid terms” (Primitive Rebels: Studies in Archaic Forms of Social Movement in the 19th and 20th Centuries (New York: W.W. Norton, 1965), 7). ⮭
- I use the language of “characterization” rather than “definition” because rioting is not amenable to precise definition. It may be more of a family of action types than a single clearly definable practice. On this methodological point, see Kimberley Brownlee, “Features of a Paradigm Case of Civil Disobedience,” Res Publica 10, no. 4 (2004): 337–51, https://doi.org/10.1007/s11158-004-2326-6. ⮭
- Pasternak, “Political Rioting,” 388, quoting David Waddington, “The Madness of the Mob? Explaining the ‘Irrationality’ and Destructiveness of Crowd Violence,” Sociology Compass 2, no. 2 (2008): 675–87, https://doi.org/10.1111/j.1751-9020.2007.00079.x. ⮭
- Pasternak, “Political Rioting,” 391. ⮭
- Ibid., 392. ⮭
- Ibid., 393. ⮭
- Ibid., 394. ⮭
- King, “The Other America”; see also Thompson, who notes that 18th century food riots in England and Wales became, over time, highly ritualized affairs, which were sometimes even announced in advance (“The Moral Economy of the English Crowd in the Eighteenth Century”). Of course, the more these events are led, coordinated, outcome-directed affairs, the more we may doubt they are to be analyzed as the same kind of phenomenon as other riots. ⮭
- I borrow the term from Kwame Ture (writing as Stokely Carmichael) and Charles V. Hamilton, Black Power: The Politics of Liberation in America (New York: Vintage Books, 1992), who describe riots as “eruptions” (48) and as “explosions of frustration, despair, and hopelessness” (160). ⮭
- Riots may have leaders in the sense that they have vanguards—the Stonewall riots were begun by a small group of drag queens, even though the mass of participants were more traditionally gendered gay men—or in the sense that leadership may arise against the background of a riot—Rodney King famously pleaded during the 1992 Los Angeles riots for people to “all just get along.” By directive leadership, in contrast, I mean leaders in a position to substantially coordinate and direct riotous activity. ⮭
- Police are often but not always, and also often not the only, targets of rioting. The 1992 riots in Los Angeles, for example, also (in)famously targeted Korean American grocers, and the Stonewall riots were equally directed at the Mafia, which ran the Stonewall Inn and was widely regarded as exploitative of the gay community. See David Carter, Stonewall: The Riots That Sparked the Gay Revolution (New York: St. Martin’s Press, 2013); Ronnie Di Brienza, “Stonewall Incident,” East Village Other 4, no. 32 (July 1969): 2. ⮭
- Pasternak, “Political Rioting,” 386. ⮭
- Quoted in Tim Newburn, Paul Lewis, Esther Addley, and Matthew Taylor, “David Cameron, the Queen and the Rioters’ Sense of Injustice,” The Guardian, December 5, 2011, http://www.theguardian.com/uk/2011/dec/05/cameron-queen-injustice-english-rioters. ⮭
- Quoted in Carter, Stonewall, 166. ⮭
- See especially Paul Lewis, Tim Newburn, Matthew Taylor, Catriona Mcgillivray, Aster Greenhill, Harold Frayman, and Rob Proctor, “Reading the Riots: Investigating England’s Summer of Disorder” (London: London School of Economics and The Guardian, 2011), http://eprints.lse.ac.uk/id/eprint/46297, who interviewed 270 participants in the 2011 U.K. riots. See also again Kerner Commission, “The Report of the National Advisory Commission on Civil Disorders”; Meyer, “The People Beyond 12th Street”; and McCone Commission, “Violence in the City.” ⮭
- This sentiment is reflected in large-scale data and in other first-personal accounts of other riots. Participants in the Stonewall riots described that event as involving “all kinds of people, all different reasons, but mostly…total outrage, anger, sorrow” (Michael Fader, quoted in Carter, Stonewall, 161) and as “a slight lancing of the festering wound of anger at this kind of unfair harassment and prejudice” (Morty Manford, quoted in Carter, Stonewall, 161). Similarly, King described rioting as centrally involving “moments of anger” and also “moments of deep bitterness” (“The Other America”). Related to the aim of making others “take notice” is the aim of asserting one’s agency. King described rioters in Watts as joyful because “they had asserted themselves against a system which was quietly crushing them into oblivion and they were now ‘somebody.’” The Autobiography of Martin Luther King, Jr., ed. Clayborne Carson (New York: Grand Central Publishing, 1998), 293. Also worth mentioning is the aim of small-scale economic opportunism, often called “looting,” which is a significant feature of some riots. It is important to note, however, that not all riots involve looting, and those that do often do so in a way that is governed by moral norms reflecting substantive views of distributive justice, e.g., by targeting certain classes of shops or avoiding damaging others. The targeting of Asian-owned shops in Koreatown in Los Angeles in 1992, for example, reflected a feeling of exploitation of the Black community by the Korean American community. See Michael Omi and Howard Winant, “The L.A. Race Riot and U.S. Politics,” in Reading Rodney King: Reading Urban Uprising, ed. Robert Gooding-Williams (New York: Routledge, 1993), 104–105; and Ice Cube, “Black Korea,” Death Certificate (Lench Mob Records and Priority Records, 1991). The aim of looting is argued by some to be in service of the other aims discussed here rather than independently motivating. See, e.g., Martin Luther King Jr, “The Role of the Behavioral Scientist in the Civil Rights Movement,” American Psychologist 23, no. 3 (1968): 181. ⮭
- Quoted in Newburn, et al., “David Cameron, the Queen and the Rioters’ Sense of Injustice.” ⮭
- Hezakya News & Films, “MLK Goes to Watts After Riots,” August 1965, https://archive.org/details/hez-news-films-archive/. ⮭
- Anderson and Pildes, “Expressive Theories of Law,” 1530. ⮭
- Pasternak, “Political Rioting,” 398. ⮭
- Helen Frowe calls this condition “instrumentality.” Helen Frowe, “The Role of Necessity in Liability to Defensive Harm,” in The Ethics of Self-Defense, eds. Christian Coons and Michael Weber (Oxford: Oxford University Press, 2016), 156, https://doi.org/10.1093/acprof:oso/9780190206086.003.0008). ⮭
- Intending a just cause is traditionally taken to be a necessary condition on a justifiable war. If there is a corresponding “right intention” condition on defense, that too would be a challenge for Pasternak’s and Havercroft’s analyses, since (I claim) rioters rarely intend rioting to ameliorate social wrongs. I take no stand on this issue, however. (The objection raised here is distinct.) ⮭
- See Paul Lewis, Tim Newburn, Matthew Taylor, and James Ball, “Rioters Say Anger with Police Fuelled Summer Unrest,” The Guardian, December 4, 2011, http://www.theguardian.com/uk/2011/dec/05/anger-police-fuelled-riots-study; Fogelson, Violence as Protest, 17; S. D. Reicher, “The St. Pauls’ Riot: An Explanation of the Limits of Crowd Action in Terms of a Social Identity Model,” European Journal of Social Psychology 14, no. 1 (1984): 1–21, https://doi.org/10.1002/ejsp.2420140102. ⮭
- Though she does not list it among her set of conditions, Pasternak also takes the justifiability of rioting to turn on questions of liability vis-à-vis those it harms. ⮭
- This statement of the principles oversimplifies, though it suffices for present purposes. (It is also in line with how Pasternak treats necessity.) On the complexities, see Seth Lazar, “Necessity in Self-Defense and War,” Philosophy & Public Affairs 40, no. 1 (2012): 3–44, https://doi.org/10.1111/j.1088-4963.2012.01214.x; Jeff McMahan, “The Limits of Self Defense,” in The Ethics of Self-Defense, eds. Christian Coons and Michael Weber (Oxford: Oxford University Press, 2016), 185–210, https://doi.org/10.1093/acprof:oso/9780190206086.003.0010; Kieran Oberman, “Killing and Rescuing: Why Necessity Must Be Rethought,” The Philosophical Review 129, no. 3 (2020): 433–63, https://doi.org/10.1215/00318108-8311248. Note too that many take the success condition to be entailed by the other conditions rather than a freestanding principle. ⮭
- See, e.g., Omar Wasow, who finds that violent protests following the assassination of Martin Luther King “likely caused a 1.5–7.9% shift among whites toward Republicans” and tipped the 1968 U.S. presidential election in favor of Richard Nixon. Omar Wasow, “Agenda Seeding: How 1960s Black Protests Moved Elites, Public Opinion and Voting,” American Political Science Review 114, no. 3 (2020): 638–59, https://doi.org/10.1017/S000305542000009X. In the other direction, Ryan Enos, Aaron Kaufman, and Melissa Sands find that the 1992 Los Angeles Riots led to “a marked liberal shift in policy support at the polls.” Ryan Enos, Aaron Kaufman, and Melissa Sands, “Can Violent Protest Change Local Policy Support? Evidence from the Aftermath of the 1992 Los Angeles Riot,” American Political Science Review 113, no. 4 (2019): 1012, https://doi.org/10.1017/S0003055419000340). ⮭
- If one is considering participating in an ongoing riot that has already, for example, achieved concessions from political leaders, the instrumental evidence regarding the effects of one’s participation may be clearer, and so may license participation on those grounds. Most cases, however, are clearly not like this. ⮭
- Similar challenges to those raised here arise in light of the proportionality condition, which requires that permissible defensive acts be proportionate in the seriousness of the harms they involve to that of the defensive goals they aim to achieve. Reasonable judgments about this, while difficult, can more plausibly be made, since information about the destructiveness of rioting is more generally available than evidence about rioting’s effectiveness. See also my further remarks about proportionality and rioting in Section IV.B.2. ⮭
- Pasternak recognizes this: she writes that in order for a riot to be permissible, “the means that rioters deploy [must] have a reasonable prospect of generating an accommodating response from state authorities.” Pasternak, “Political Rioting,” (404). ⮭
- Pace Havercroft, if rioting is justified by the ends it pursues (its “just cause”), then it must be proportionate to the value of those ends, not proportionate to the magnitude of the wrong it responds to. Only sometimes are these the same, viz., when the end pursued is the prevention of the wrong responded to. But this cannot be the case for rioting, whose progress towards justice is almost only ever (if even) incremental. ⮭
- Havercroft, “Why Is There No Just Riot Theory?,” 12. ⮭
- Jeff McMahan, “Just Cause for War,” Ethics & International Affairs 19, no. 3 (2005): 1–21, https://doi.org/10.1111/j.1747-7093.2005.tb00551.x; see also Thomas Hurka, “Proportionality in the Morality of War,” Philosophy & Public Affairs 33, no. 1 (2005): 34–66, https://doi.org/10.1111/j.1088-4963.2005.00024.x. ⮭
- The view I propose is in principle compatible with the defensive and just riot theory approaches to the justification of rioting. If a version of those views is developed that is not subject to the objections raised here, or in cases of rioting in which those objections do not apply, rioting could then be multiply justifiable. Note too that although I shall develop a view on which rioting is justifiable, I resist calling my view a species of just riot theory, as that name is too closely tied to the instrumental justification of violence that, for characteristic cases of rioting, I reject. ⮭
- These are the words of Joe Biden, quoted in Eric Bradner, “Biden Condemns Violence and Asks If Americans ‘Really Feel Safe Under Donald Trump,’” CNN, August 31, 2020, https://edition.cnn.com/2020/08/31/politics/joe-biden-pittsburgh-violence-speech/index.html, and David Cameron, “Riots: Cameron Statement in Full,” BBC News, August 11, 2011, https://www.bbc.com/news/uk-politics-14492789. ⮭
- Chrisman and Hubbs, “Rioting as a Speech Act”; see also Quill Kukla (writing as Rebecca Kukla) and Mark Lance, “Yo!” And “Lo!”: The Pragmatic Topography of the Space of Reasons (Cambridge, MA: Harvard University Press, 2009), chs. 6 and 7. ⮭
- Clarissa Rile Hayward, “Disruption: What Is It Good For?,” The Journal of Politics 82, no. 2 (2020): 448–59, https://doi.org/10.1086/706766. ⮭
- Hooker, “Black Lives Matter and the Paradoxes of U.S. Black Politics,” 450. ⮭
- Shelby, Dark Ghettos, 272–73. ⮭
- Lewis et al., “Rioters Say Anger with Police Fuelled Summer Unrest,” 24. ⮭
- Kerner Commission, “The Report of the National Advisory Commission on Civil Disorders,” 64; see also Meyer, “The People Beyond 12th Street”; McCone Commission, “Violence in the City.” ⮭
- Di Brienza, “Stonewall Incident.” ⮭
- Hezakya News & Films, “MLK Goes to Watts,” my emphasis. ⮭
- King, “The Other America.” ⮭
- Chrisman and Hubbs, “Rioting as a Speech Act.” ⮭
- The idea that the active resistance of the oppressed is a way of saying “no” and “no more” is also a central theme of theorists of resistance such as Ture and Hamilton (Black Power); Frederick Douglass (My Bondage and My Freedom (New York; Auburn: Miller, Orton & Mulligan, 1855)); and Frantz Fanon (Les Damnés de La Terre (Paris: François Maspero, 1961)). José Medina also argues that similar messages—“‘No,’ ‘Enough!,’ ‘Stop!,’ or ‘We will not take it anymore’”—are characteristic of what he calls “polyphonous” dissent movements. José Medina, The Epistemology of Protest: Silencing, Epistemic Activism, and the Communicative Life of Resistance (Oxford: Oxford University Press, 2023), 98. ⮭
- This section further develops conditions outlined in Flanigan, “Futile Resistance as Protest.” ⮭
- See Selim Berker, “The Deontic, the Evaluative, and the Fitting,” in Fittingness: Essays in the Philosophy of Normativity, eds. Christopher Howard and Rach Cosker-Rowland (Oxford: Oxford University Press, 2022), 23–57, https://doi.org/10.1093/oso/9780192895882.003.0002; Christopher Howard, “The Fundamentality of Fit,” in Oxford Studies in Metaethics, ed. Russ Shafer-Landau, 14 (2019): 216–34, https://doi.org/10.1093/oso/9780198841449.003.0010. ⮭
- Christopher Howard, “Fittingness,” Philosophy Compass 13, no. 11 (2018): 1–14, https://doi.org/10.1111/phc3.12542. ⮭
- For more on this topic, see again Berker, “The Deontic, the Evaluative, and the Fitting,” especially sec. 4. ⮭
- I call this justification “defeasible” rather than “pro tanto” to avoid the implication that justification comes in degrees. Like Ross’s “prima facie duties” (which might similarly be better called “defeasible duties”), the point is that this a robust kind of justification that can, nonetheless, be outweighed or overridden. ⮭
- See Chrisman and Hubbs, “Rioting,” 387. ⮭
- Rioting that substantially destroys the neighborhoods of the oppressed may remain inappropriate if it destroys the property or collective goods of the fellow-oppressed who opt not to riot, since “destroying one’s own home” is a prerogative that weakens, of course, if the home is shared. (Though considerations of solidarity may partly push back in the other direction.) The point, in any case, is that the symbols and sites of oppression could be an appropriate target of protest in a way that distinguishes this consideration from the notion of liability used in the ethics of war and defense. ⮭
- See also Ten-Herng Lai, “Political Vandalism as Counter speech: A Defense of Defacing and Destroying Tainted Monuments,” European Journal of Philosophy 28, no. 3 (2020): 602–16, https://doi.org/10.1111/ejop.12573, esp. sec. 3.3. ⮭
- Avia Pasternak, Responsible Citizens, Irresponsible States: Should Citizens Pay for Their States’ Wrongdoings? (Oxford: Oxford University Press, 2021), https://doi.org/10.1093/oso/9780197541036.001.0001. ⮭
- Spike Lee, Do the Right Thing (Universal Pictures, 1989). ⮭
- It may remain hard to believe that expression that is carried out without hope of bringing about positive change could really justify physical harm. I take up this question separately in Flanigan, “Futile Resistance as Protest.” ⮭
- On rioters making distinctions between targets that follow substantive judgments of responsibility or symbolic importance, see Michael Omi and Howard Winant, “The L.A. Race Riot and U.S. Politics,” 104–5; see Carter, Stonewall, and Di Brienza, “Stonewall Incident” on the mafia’s involvement in running the Stonewall Inn; and see again Thompson, “The Moral Economy of the English Crowd in the Eighteenth Century.” ⮭
- Outside of legal philosophy, see also Christopher Bennett’s similar view regarding expressive reasons for action. Christopher Bennett, “Expressive Actions,” in Emotional Expression: Philosophical, Psychological and Legal Perspectives, eds. Catharine Abell and Joel Smith (Cambridge: Cambridge University Press, 2016), 73–94, https://doi.org/10.1017/CBO9781316275672.004). ⮭
- Anderson and Pildes, “Expressive Theories of Law,” 1503. ⮭
- Feinberg, “The Expressive Function of Punishment,” 408. ⮭
- Many raise objections along these lines. See especially Victor Tadros, The Ends of Harm: The Moral Foundations of Criminal Law (Oxford: Oxford University Press, 2013), ch. 5, though for a detailed reply, see Joshua Glasgow, “The Expressivist Theory of Punishment Defended,” Law and Philosophy 34, no. 6 (2015): 601–31, https://doi.org/10.1007/s10982-015-9235-z. ⮭
- See, e.g., T. M. Scanlon, “Punishment and the Rule of Law,” in T.M. Scanlon, The Difficulty of Tolerance: Essays in Political Philosophy (Cambridge: Cambridge University Press, 2003), 219–33. ⮭
- Some do make such claims of rioters. In Section IV.C., I consider whether rioters ought to pursue nonviolent protest or other still forceful but less harmful forms of disobedience. ⮭
- Edmund Tweedy Flanigan, “From Self-Defense to Violent Protest,” Critical Review of International Social and Political Philosophy 26, no. 7 (2021): 1094–1118, https://doi.org/10.1080/13698230.2020.1870859; Flanigan, “Futile Resistance as Protest.” ⮭
- P. F. Strawson, “Freedom and Resentment,” Proceedings of the British Academy 48 (1962): 187–211. ⮭
- Hanna Fenichel Pitkin, The Concept of Representation (Berkeley: University of California Press, 1967). ⮭
- See Rachana Kamtekar, “Social Justice and Happiness in the Republic: Plato’s Two Principles,” History of Political Thought 22, no. 2 (2001): 189–220; Donald Morrison, “The Happiness of the City and the Happiness of the Individual in Plato’s Republic,” Ancient Philosophy 21, no. 1 (2001): 1–24, https://doi.org/10.5840/ancientphil20012116. ⮭
- Martin Luther King, Jr., “The Social Organization of Non-Violence,” Liberation, October 1959, 5–6. ⮭
- Di Brienza, “Stonewall Incident.” ⮭
Acknowledgements
For their generous and helpful comments on various versions of the ideas presented here, I am grateful to Eric Beerbohm, James Brandt, Susanne Burri, David Clark, Thomas Fossen, Jonathan Havercroft, Nik Kirby, Ten-Herng Lai, Avia Pasternak, Tommie Shelby, Laura Valentini, and two reviewers for this journal; as well as to audiences at the 2021 Association for Political Theory Annual Conference, the conference “The Ethics of Defending Yourself and What’s Yours Under Incomplete Information” at the University of Zurich, the LMU Munich Social and Political Philosophy Colloquium, the 2023 Joint Session of the Aristotelian Society and Mind Society, the University of Bamberg Political Theory Research Seminar, and a workshop on defensive ethics at the University of Konstanz. For their help with the empirics of violent protest, I am grateful to Holly Dykstra, Anna-Lena Hönig, and Roman Krtsch.
Competing Interests
The author has no competing interests to declare.