Free & Equal: Tweedy Flanigan’s “Why Riot? An Expressive Theory of the Justification of Rioting”, with a critical précis by Candice Delmas

22 Replies to “Free & Equal: Tweedy Flanigan’s “Why Riot? An Expressive Theory of the Justification of Rioting”, with a critical précis by Candice Delmas”

1. Thank you for this very interesting paper, Tweedy, and thank you, Candice, for the excellent précis. I will make one brief comment on the paper’s negative argument, and then spend more time on the positive account.

   1. Just Causes and Lesser Harms

   Havercroft suggests that rioting may be justified by ends such as preserving freedom, promoting equality, and giving voice to marginalized groups. Flanigan responds by invoking a familiar thought from just war theory: such ends are not the right kinds of ends to be pursued through violence.

   That may be right, so far as lethal violence is concerned. But even if freedom and equality cannot justify killing, it does not follow that they cannot justify lesser harms. And rioting, at least paradigmatically, involves precisely such lesser harms—harms to property in particular. Flanigan writes: “No amount of extra freedom or equality could by itself justify harming others as a means to its attainment” (p. 278). Is this right?

   Suppose I own a large, highly visible house. Suppose further that smashing my windows would generate a media storm that would foreseeably pressure the government into changing unjust policies, thereby significantly promoting freedom and equality. In such a case, it seems at least plausible that smashing my windows would be justified. Indeed, if the benefits are large enough, it is not implausible that I might even have a duty to allow my windows to be smashed.

   Flanigan might reply that real-world riots rarely generate benefits of this magnitude. But that is a different claim. It is an empirical claim about the effects of rioting, not a normative claim about the kinds of ends that can justify harms to property. The just riot framework Flanigan criticizes is compatible with the empirical claim. What Flanigan needs, but doesn’t provide as far as I can see, is an argument for why freedom and equality are the wrongs kinds of ends to justify harms to property.

   2. Flanigan’s Positive Account

   Let me now turn to Flanigan’s positive account. My concern here is twofold. First, given the dialectical burdens Flanigan faces, the positive account looks under argued. Second, I have independent doubts about the kind of fittingness justification he appeals to.

   Flanigan notes at the beginning of Section 4 that non-instrumental justifications are not usually thought to license harm to others. This seems right. I myself enter this debate with the intuition that if rioting is justified at all, it is most likely justified for instrumental reasons rather than non-instrumental ones. Of course, this intuition may be mistaken. But the fact that it is a natural starting point for many in the debate matters. It means that Flanigan’s task is not merely to explain a widespread intuition about the non-instrumental justification of rioting, but to persuade many of us to revise our initial judgments. That raises the bar for the positive argument (more on this in a moment).

   Flanigan’s proposal is that rioting can be a way of adequately saying what it is fitting to say. What is fittingness? Flanigan suggests that it is a “basic normative category” where the relation of fit is one of “metaphysical matching or suiting between two objects” (p.284). From here, he goes on to say that the category of the fitting yields normative claims distinct from claims about value or duty. Finally, he asserts that the fittingness of φ-ing provides defeasible justification for φ-ing—even, crucially, where φ-ing involves harming others.

   The core idea, then, seems to be that the justification of rioting is grounded in the normative force of fittingness. But fittingness, as such, does not seem to have this kind of normative strength. To see why, consider a very ordinary fittingness claim:

   Fitting Fear: A fear response is fitting to its object if and only if that object is fearsome in proportion to the magnitude of the fear.

   If any fittingness claims are true, this one is. Now ask: does the fittingness of fear generate reasons capable of justifying harm? Suppose I am walking with my son, and we encounter an aggressive dog. The dog will attack if it smells fear. Due to a genetic condition, neither my son nor I can spontaneously feel fear. As it happens, I’m carrying a pack of fear pills that would induce exactly the fitting amount of fear. Do I have a reason to take a pill?

   If fittingness grounds reasons, it seems I have a reason to take the fear pill. This already looks doubtful. But it looks even more doubtful once we note that taking the pill would cause harm to both of us. So fittingness, at least in this familiar case, does not generate reasons with harm-justifying force.

   One might say: this only shows that some cases of fittingness lack justificatory force. The relevant cases involve fitting expressions of anger at wrongdoing. Perhaps fitting anger, unlike fitting fear, can justify harming. But notice what this response concedes. It concedes that fittingness as such cannot be doing the normative work Flanigan assigns to it. Something further is needed—a story about why fitting anger is normatively special in this way.

   Flanigan later gestures toward such a story (p.293–294) when he notes that, given the nature of political community, injustice calls for loud and forceful expression. But without an explanation of why the appropriateness of such expression can justify harming, these remarks come close to assuming what is at issue. This would be less troubling if Flanigan were explaining a widely shared intuition that riots are justified non-instrumentally. But he is not. As Flanigan himself notes, he is challenging our starting intuitions. In this dialectical context, passing the justificatory burden to fittingness without explaining its special normative power is not enough.

   Finally, I have independent doubts that fitting expressions of anger can justify harm.

   Suppose I’m being attacked—punched and verbally abused—in a glassware shop. Shouting “Stop!” at my attackers would be a fitting response. But it would bring about no instrumental good. Worse, shouting would topple a beautiful glass object that would cost the shop owner lots of money to replace.
   I have little intuition that the fittingness of my shout gives me a reason to shout. And I have even less intuition that it could justify the economic harm to the shop owner. This suggests to me that non-instrumental considerations of fittingness do not have the normative weight required to justify harming others. Perhaps this judgment is mistaken. But if so, I would want to see a careful and compelling argument explaining why.

2. Thanks so much Tweedy for a terrifically interesting paper, and to Candice for a great response. There is an empirical question about what rioters typically intend, and I don’t know the answer to that question (or any other empirical question). One thing I wondered is whether deterrence features in the minds of rioters, and whether that can contribute to the justification of rioting. This initially seems a more central justification than either self-defence or pure expression/communication. Rioters indicate that if they continue to be treated as they are, this is how they will respond. And there is a global phenomenon of rioting where individual riots contribute to global deterrence. Governments disposed to serious social and economic injustice know that if they go too far, a riot will occur, and that restrains them from such injustice to some degree. The communicative aspect of rioting seems to me better understood as making a contribution to deterrence – through rioting, rioters express anger at the way they have been treated, and riot as a punitive response. If I am right that deterrence is at least somewhat effective. If that is right, there are difficult questions about when deterrent violence is justified, but this seems initially a more plausible justification to me than the idea that rioting is a fitting way of communicating that one has been treated unjustly – this mirrors my view about punishment; that the communicative aims of punishment are better understood as integrated into an instrumentalist theory where deterrence is central rather than providing a stand alone justification of punishment. Am I right that deterrence hasn’t really featured in these debates though? It seems at least worth considering.

3. Thanks so much to the PEA Soup editors for organizing this discussion, and thanks especially to Candice for a great précis and set of comments. Let me offer some replies to the concerns and challenges she raises.

   Two quick things first:

   First, a note on terminology that echoes a point I make in the paper: Some people reasonably prefer more positively valenced terms like ‘uprising’ as a way to describe the kind of protests this paper discusses. I’ll use the term ‘riot’ here, despite that term’s historically negative connotation, because I think it picks out a distinctive social practice that alternative terms don’t. But I acknowledge differences of opinion on this point, and of course anyone commenting here should use the term they think is most appropriate — I take it we’re all talking about the same phenomenon, whatever we call it.

   Second, a shameless plug! If you’re interested in the view developed and applied here to rioting, I invite you to take a look at two other papers, “Futile Resistance as Protest” (in Mind) which concerns interpersonal cases of futile resistance, and “Justifying Futile Climate Resistance” (forthcoming in PPR), co-authored with Ten-Herng Lai, which develops this framework with reference to the climate crisis.

   # Rioters’ characteristic aims

   The idea for this paper came originally from reflections on some histories of major riots. This led me to spend a fair amount of space in the paper thinking about how best to characterize the typical aims of riot participants, based on what I understand to be the best empirical evidence we have. (In earlier drafts of the paper, this section was much longer.) This is for several reasons. For one, as philosophers studying conditions of oppression, it’s worth taking care to try to understand the phenomenology of those conditions for those facing them – simply as a way of respecting and doing justice to the subject matter and the people who make it up. For another, the contours of complex social phenomena are sometimes drawn by the way people engage in them, which in turn is shaped by people’s aims and intentions in their engagement. And finally, I take it that the moral justification of acts can partly depend on an actor’s aims and intentions.

   So, what is the best evidence we have about the aims of riot participants? It’s worth stating up front that the best evidence we have is still very limited. Rioting is an empirically understudied phenomenon, and among those who do study it, disciplinary concerns tend to focus research on rioting’s causes and effects. So there’s even less evidence about the aims and intentions of riot participants. There’s also a selection issue: to the extent that we have this kind of evidence, it’s been collected after, and about, specific riots that were considered notable, which is to say that the evidence we do have may not be representative of rioting as a whole. But with these limitations noted, there is some fairly robust evidence on rioters’ aims and motivations, and not all of it is anecdotal. An analysis of interviews with 270 participants in the 2011 UK Riots concluded that “a consistent theme emerging from interviews with the rioters across England was that they harboured a range of grievances and it was anger and frustration that was being expressed out on the streets.” Similar judgments were made by the authors of the Kerner Commission Report, the “People Beyond 12th Street” report, and the McCone Commission report, which were based on interviews with residents of neighborhoods in which major riots took place in the late-1960s US. It’s a theme of David Carter’s landmark history of the Stonewall Riots. These point to a widespread set of closely related motivations that are, I argue, about the hopelessness and futility of political action toward change, rather than some means-ends orientation whereby the act of rioting is supposed to bring about change.

   (This is notably not true of many 18th century English and French food riots: villagers – mostly women – “rioted” to “state the price of corn” (or wheat, etc.), often following an established script, sometimes having organized themselves in advance. So it’s worth being clear that I’m focusing on a particular slice of riots in the long history of the practice. I can’t recommend EP Thompson’s classic “The Moral Economy of the English Crowd in the Eighteenth Century” enough to anyone interested.)

   Now, it’s important to emphasize that rioters are typically only a loosely connected entity, and so, in agreement with Candice, I would be shocked if rioters ever completely, univocally shared an aim – their motivations will instead always be multiple and mixed. Some people may be there to “loot”, some to see what’s going on, some because their friends are there, etc. The question however is about whether we can make some kind of overall generalization. The evidence I mentioned above suggests that in at least some cases we can. The broader point though is that these things do matter to the moral assessment of rioting, and we should do our best to bring evidence to bear on the question, and to make our theories responsive to that evidence. To the extent that we just stipulate an answer to the question, “What are rioters’ aims?”, or stipulate that there is no answer, I think we’re making a mistake. There’s a point about the methodology of normative political theory here: when we write about social practices, and particularly high-stakes practices of resistance, we should give special weight to the expressed views of those who participate in them. And that demands that we do our best to bring evidence to bear on what those views, taken generally, are. And that demands robust engagement with social science.

   Lastly on this point, Candice points out that I characterize the ultimate shared expressive *content* of what I take rioters, through their rioting, to be saying in very general terms: “listen!”, “this is wrong!”, “enough!”, “no more!”. She worries that these are hopelessly vague, and that they don’t accurately represent the much more specific things individual rioters may wish to express. I agree. I don’t mean to suggest that individual riot participants have only these general messages in mind. On the contrary, their grievances will surely often be more specific, and will therefore vary substantially across riot participants as a group. The idea rather is that the underlying message that unites these more specific, varied messages will indeed often (but not always!) be general and something basic like “stop!”, “no more!”, etc. So I think that, typically, while there will be many specific messages expressed by riot participants, there will be something like an underlying consonance that yields a (relatively) univocal expressive content. (On the question of when and whether riots are relatively “univocal” or “multivocal” in their expressed grievances is something else on which I have empirical social scientific data to appeal to, but this didn’t make it into the paper: I can share it with anyone who’s interested.) Jose Medina makes a similar point in *The Epistemology of Protest*; his notion of “polyphony” is helpful here.

   # Justifying violence

   ## Violence and adequate expression

   The paper’s most controversial claim is that the violence involved in rioting may sometimes be the best or only way, under the circumstances, for the oppressed to adequately protest their oppression. This is why, I claim, violence can be called for; and that is a major part, on my view, of its justification in rioting.

   To this claim, seeming counterexamples and objections may come flooding to mind. Do those who march peacefully do something inadequate and therefore wrong? Are victims morally required to harm their attackers when the attacks are severe enough? Surely not.

   The basic form of my answer to these kinds of challenges is to claim that those who act in these ways do not act wrongly, but they do (often — this is nuanced) act expressively inadequately. But this fact is unsurprising — I think there are often lots of good reasons why someone might not act expressively adequately, some of which I’ll mention in a moment. And in light of these many good reasons, I think that failing to express oneself adequately is often not wrong. In this way, fittingness is generally speaking *permissive* of violent expression in response to severe wrongs rather than *requiring*.

   Why might someone respond to a wrong in a way that is not expressively adequate? Most obviously, it might be unwise to do so. Those who enact serious wrongs are often powerful, and correctly responding expressively to those wrongs might only invite further wrongs upon the victim. In other circumstances, victims may choose to forego an expressively fitting response as an act of grace. (In draft work, I’m developing an account of principled nonviolence that has this shape.) People who choose not to blame the blameworthy are often, I think, making a choice of this kind. In general, victims have a prerogative to do less than would adequately protest the wrongs they suffer; it’s in the nature of prerogatives that they can be exercised largely for the holder’s own reasons.

   For the same kinds of reasons, when organized movements are available that stand a good chance of bringing about positive change with respect to an injustice, I am inclined to think that the instrumental value of pursuing those options will often trump the importance of engaging in a fitting expressive response. If you can stand up to a bully *or* prevent him from bullying, who wouldn’t choose the latter?

   There is also the matter of protest on another’s behalf. I don’t have any detailed knowledge of the Silent Parade protests Candice mentions, but I do notice that they took place in New York in response to the horrors of racial lynching, which was a predominantly (but not exclusively) Southern phenomenon. I suspect that the magnitude of fitting protest on someone else’s behalf is often or always smaller than fitting protest carried out by the victim.

   Finally, I want to allow emphatically that there are ways of adequately responding to serious wrongs that do not involve violence. Art can be one such way. Emma Sulkowicz’s *Mattress Performance*, for example, might have been more expressively adequate than any verbal or physical confrontation with her attacker could have been. The famous “I AM A MAN” signs that were part of the 1968 Memphis sanitation worker’s strike are another candidate. But I don’t think we can demand of ordinary people facing oppression that they become artists just to avoid violence. We might ask them to join in dignified protest movements instead of rioting *when those movements exist* — but often, in the circumstances that characteristically give rise to rioting, they do not. Rioting is special in this regard. It’s an eruption of discontent; other expressively adequate channels are ipso facto unlikely to exist.

   ## Proportionality

   I argue that fitting expression must be proportionate to what it responds to. Unlike the proportionality condition in defense, as it is usually construed, which weighs the expected (relevant) achievements of a defensive act against the harm it imposes, proportional response just asks whether the magnitude of a response is proportionate to the magnitude of what’s being responded to. Candice suggests that proportionate response should likewise take account of the response’s effects, given the many potentially negative consequences of rioting.

   I’m not sure. Consider a parallel: suppose you wrong me, and I blame you for it. Blame is similarly governed to conditions of fit; fitting blame must be proportionate. But it should be proportionate to the wrong, not to (or including accounting for) the effects of blame. The effects might be something to take account of in your all-things-considered judgment about whether to blame, but I don’t think they would affect what counts as proportionate blame. I’m inclined to say the same thing about proportionate fitting expression.

   ## Democracies and Non-democracies

   Social invisibility and voicelessness are commonly cited motivating grievances among riot participants. As Candice notes, the demand for voice has different significance in democracies versus non-democracies. She writes, “it should arguably make a difference in one’s chosen means of expression whether the society in question professes or not to treat all citizens as political equals and to protect their basic liberties.” An interesting question is whether the expressive form of rioting, when it is correctly responding to voicelessness, should take a different shape in democracies versus non-democracies. Certainly the precise nature of the grievance differs: in democracies, which are often thought to be committed to something like everyone having a voice, it is distinctively wrong to deny such a voice to only some subjects. On the other hand, if (in some non-democracies) *no one* has a voice, but everyone ought to, that too seems distinctively wrong in a different way. So the wrongs are different — but I’m not sure whether the fitting response necessarily is, especially when we’re talking about a blunt form of expression like rioting. I’d want to hear more about what (reasonably available) forms of expression would be differentially appropriate.

   ## Who is this for?

   Finally, Candice calls attention to my claims that a moral theory of rioting should “provide moral guidance to prospective rioters weighing whether and how to participate.” There is of course something odd about offering “moral guidance” to riot participants in the pages of an academic journal. I don’t take myself to be doing that, although I see that how I’ve written that sentence lends itself to that interpretation. The point I mean to be making in that paragraph is just that a theory of the justification of rioting needs to *apply* at the level of individual decisions about whether and how to participate, rather than to the event as a whole.

   —

   Thanks again, Candice, for your engagement with the piece!

4. Thanks Tweedy for a very thought provoking paper, and Candice for the equally thought-provoking precis!

   My question picks up on Faron’s. It concerns the claim that ‘no amount of freedom and equality can justify (instrumentally) harming others (unless the freedom and equality in question concerns our most basic rights)’. Faron rightly questions whether this is true, once we move from killing to lesser harms.

   But I found Tweedy’s endorsement of this claim odd for two reasons.
   1. Tweedy says that most just war/self-defence theorists accept this claim, and then seems to accept it himself. Why not just say they’re wrong about that, but the ‘just war/self defence/instrumentalist’ *framework* stands (and therefore can in theory justify rioting, even if this will not justify many actual riots)? Tweedy may be dissatisfied with this, since he wants to justify a much wider range of riots, including those that don’t end injustice or aim at ending injustice. But then why not say riots can (in theory) be justified instrumentally *and* as a form of expression. In other words, Tweedy’s paper sometimes seems to assume that there is a single way to justify rioting — so it’s the just riot account, or his. But why? Why not say that in principle the achievement of justice/moving toward it can justify rioting, but most riots will be justified expressively (this also picks up on Candice’s point)?
   2. It seems especially odd to deny that the ACHIEVEMENT of freedom or equality could never justify rioting, but expressing anger about lack of freedom and equality can. Surely if expressing frustration about an injustice justifies rioting, ending that injustice could, in principle, justify rioting. If so, then for any circumstance x that justifies an expressive riot, ending x would (in principle) justify an instrumental/just war/Pasternak-style riot.

5. @Victor: Thanks – that is an interesting idea; I don’t think I’ve come across work in any discipline that focuses on rioting as deterrence. (Separately, I’ve thought a little bit about the implications of my kind of view for punishment, especially mixed expressive-instrumental views like yours and Scanlons, but I haven’t come to any conclusions.) As you note, this raises the thorny question of the deterrent effects of rioting. Political science verdicts on the medium-run political effects of major riots is fairly mixed: some find positive effects, others find backlash effects. So that’s one provisional obstacle to the strategy you’re suggesting. This also raises the issue I raise for defensive justifications: rioting is characteristically undertaken by the powerless; so, do we really think the justifiability of rioting should vary along how accommodating the powerful are? If rioting could ever be an appropriate response to conditions that by hypothesis admit of no deterrent possibilities – and I’m inclined to think it could – then I wouldn’t want to center the account on deterrence.

   @Faron: Thanks; these are tough challenges! On lesser harms: others will know better than I about this, but I suspect that while property violence is lesser in kind, it also matters that in riots, the property harms are typically extreme. If a country attacks another with an army (so, with lethal violence) or by causing its victim’s economy to collapse (so, with something like property violence) doesn’t seem to me like it should dramatically affect what kinds of causes would count as a just cause. But I realize this is a live debate in other parts of the just war/defensive ethics space.

   On the normative and moral force of fittingness: I don’t think that fitting response necessarily carries any moral force. If it is fitting to believe P in light of the evidence, that doesn’t (I would say) yield any moral reasons whatsoever. If it’s fitting to admire what’s beautiful, you don’t gain any moral reasons for admiration. If it’s fitting to fear the bear, you aren’t thereby given any moral reasons whatsoever to fear the bear. Normative categories have their attendant upshots: the deontic gives you requiredness/permission/prohibition, the evaluative gives you (non-moral) value, and the fitting gives you fittingness. But they only give you moral requiredness, goodness, fittingness (etc.) when they operate in a moral domain. I think just talking about the reasons that follow from normative categories obscures this. My claim is about fitting reactive attitudes, which following Strawson I take to exist within (or even be the foundation of) the moral domain. That’s why I think that if the attitudes expressed by rioters are fitting, then we can say those attitudes are morally called for – and thus can interact naturally with all the other moral considerations that apply here.

   On the glass shop: The question here is whether responding fittingly can be outweighed or overridden by the effects my response will have on third parties. Sometimes, it’s clear that it can. Imagine I intend to give you fitting thanks for your kindness, but if I do so a grump will punch a bystander. Arguably I should withhold my thanks. I don’t think this shows there’s no significant normative force to the demand of thanks; it’s just been overriden by the disvalue of the foreseeable consequences on the poor bystander. Could the demand for thanks ever *not* be overriden by practical consequences? Sure. If you’ve done me a great kindness, but my thanking you will annoy those who envy you, causing them to be rude to some bystanders, I should probably thank you anyway. In the case of rioting, we’re talking about especially severe and pervasive wrongs, so I think the importance of carrying out the fitting response will carry substantial weight.

6. @Patrick: Thank you! On rioting being multiply justifiable: I’m inclined to think that insofar as the main obstacles to a defensive justification of rioting have to do with the facts and evidence about rioting’s effects, if those facts and evidence were otherwise, or *when* they’re otherwise (in a particular case), then rioting can be multiply justifiable, i.e. as expressive or as instrumental harm. I say something to this effect in footnote 44. (There are some tricky questions about how these two forms of justification interact, which I’ve ignored because a) they’re very tricky and b) I don’t think these cases are basically ever likely to arise.)

   On the second point: There’s some ambiguity when we talk about the values of freedom and equality. In one sense, those are foundational political values that correspond to basic entitlements of political subjects. These are traditionally thought of (and I agree) as the kinds of things wars can be fought for. In another sense, they might mean something much less important: I’m freer if (say) there are more places I can go without needing papers, and I’m more equal if (say) income inequality is narrower rather than wider. These are not the kinds of things we fight wars for. Similarly, serious violations of basic entitlements of freedom and equality are worth rioting over, but that doesn’t extend to all instances of those values. (I’m not sure if this fully responds to your worry – I’m sorry if not!)

7. Thanks a million, Tweedy, for your brilliant paper and thanks to Candice for the excellent précis. I have a couple of questions. One is about implications for war and international injustice; the other is about your empirical assumptions.
   First, on the implications of your account of rioting as a form of permissibly fitting expressive violence and international warfare: If we ought to accept that sometimes it is permissible for individuals to join a riot that will cause harm both to those who are liable and to others as well as to property and built environments as a fitting way of expressing outrage, then should we accept that the same is true for states facing international injustices? Or for non-state actors or individuals within those states?
   Say, for instance, a state or a people is impoverished and oppressed as a result of imperial legacies and contemporary global power relations and has no obvious pathway to a remedy: might your argument imply that expressive violence could be permitted as a result? Sometimes, for instance, non-state actors engaged in ‘terrorist’ attacks have presented themselves as fighting against this sort of injustice. Is there a way of permitting riots along the lines you suggest without opening the door to some of these other forms of violence? Or, similarly, if a state faced annexation of part of its territory by an aggressor but had no chance of defending itself by means of instrumental violence, might it be permitted (or even obliged) to launch some sort of expressive warfare against its attacker?
   Second, the paper assumes that riots are generally ineffective, i.e. they can’t be expected to achieve benefits of the right kind instrumentally. But it seems to me that sometimes riots do achieve things that could justify violence, so it’s hard to believe that rioters should generally be entirely without hope of material success. For instance, after violent protests following the murder of George Floyd, prosecutions were pursued against officers in a way that arguably helped establish law enforcement against the police more generally. Insofar as this might have helped diminish the arbitrary power of police officers to inflict harm, it maps onto legitimate defensive objectives: the riots arguably aimed to achieve a reduction in the unchecked power to harm on the parts of the police and achieved some success.
   If that’s right and if riots are sometimes directed against pervasive threats from security forces, then why couldn’t we interpret riots (at least sometimes) as being attempts to achieve goals of the sort that can justify violence instrumentally? If so, then in the least promising cases, rioters might still view the expressive value of violence as a second best justification compared to securing their, as it were, material goals.

8. Hi Tweedy. Thanks for your responses. In terms of freedom and equality, the examples you give seem to point toward very small improvements, but you say (provided basic rights are not implicated) NO AMOUNT of freedom can justify ANY VIOLENCE. Much is going to depend on what we take ‘basic rights’ to incorporate.

   But let me just push you on the formal version of what I said:

   ‘If expressing dissatisfaction with X justifies a riot, so would ending X’.

   Is there anything where expressing dissatisfaction would justify a riot, but seriously improving the situation would not?

9. Thank you Tweedy for your excellent paper and everyone else for their insightful comments. I would like to make a few points following on from the above.

   # Rioters’ aims
   I find Tweedy’s account of riots more convincing than the alternative just war/social justice approaches. As Tweedy notes, rioting involves cases where one has gone ‘past caring’. Thinking of the 1992 Los Angeles riots after Rodney King brutal beating (or other riots following the all-too-frequent police shootings of unarmed African American in the US), I don’t think those who burnt down the city centre had any desire to bring about change. The timing of their action is crucial (and is the main reason I find Tweedy’s account more convincing than Avia’s): riots tend to occur shortly (even just a few hours) after the event. They are a quick reaction to an egregious injustice, an injustice that signals, at least in the rioters’ mind, that there is no hope for change. That is why some people decide to go out and smash everything, instead of e.g., organising a peaceful event to raise the issue. Other citizens will probably do that, later on: in fact, even the rioters themselves may realise, after having let some of the steam out in the riot that, if they want to bring about lasting change, they have to adopt a less ‘riotous’ approach. But the point remains, in my view, that at the moment those citizens decide to riot, they do it to express rage.

   # Instrumental vs non-instrumental
   This is just a thought, but I wonder whether the riot’s expressive role may also have instrumental value (not sure if Tweedy considers that possibility). By engaging in collective vandalism against property (not people), citizens may release some uncontrollable anger, which would allow them to the better contribute to more organised (and less violent) action. I am thinking of Willaim Smith’s account of civil disobedience as playing an instrumental role as a ‘safety valve’. In the case of riots, one may argue that the expression of rage would prevent those citizens from further escalating their actions (e.g., engaging in acts of terrorism).

   # Last resort
   I wonder whether we need this requirement. As with civil disobedience (who is traditionally deemed justifiable only as a last resort), one may reply that the rioters always have alternative ways to express their rage. In civil disobedience, for example, the idea is that citizens should first exhaust all legal avenues – except, the legal avenues never end! Should the rioters (e.g., in the aftermath of King’s beating) first use legal avenues to protest (a petition?), then try civil disobedience, then progressively escalate to the riot? I am not sure this just is how riots tend to work in practice.

   # Violence vs non-violence
   In his response to Candice, Tweedy writes that “when organized movements are available that stand a good chance of bringing about positive change with respect to an injustice, I am inclined to think that the instrumental value of pursuing those options will often trump the importance of engaging in a fitting expressive response. If you can stand up to a bully *or* prevent him from bullying, who wouldn’t choose the latter?”
   On this point, it might be worth mentioning the idea of so-called ‘Radical Flank effect’ (I think Candice discussed this in her excellent 2018 book). This refers to the fact that we should not think of social movements as EITHER ‘violent’ OR ‘nonviolent’. Most social movements have both a peaceful and a radical fringe.
   For example, the official reading of the Civil Rights Movement is that, by choosing peaceful rather than violent protest, the movement led by Reverend King managed to reach its goal. This reading, however, ignores the contributions of e.g., the Black Panther Party or the Nation of Islam, which were less keen on nonviolence. The contention is that, without these radical groups, King’s movement would not have looked so moderate, and therefore may not have gained support from White liberals. This suggests the possibility of a division of labour – where different forms of expression can contribute to the final result. Perhaps some people should stand up to the bully, while some other should seek to prevent him from bullying.

10. @Chris: Thanks, these are great questions. On the view’s extension to state/collective action: I certainly think the view is in principle applicable to state/collective action. Insofar as states are just complex collections of people, if people want to stand up for themselves against a much stronger adversary, I think that could very well be an appropriate (i.e. fitting) course of action. You might read the Warsaw Ghetto Uprising in this way, for instance (though the actor there is not a state). But important complications and obstacles arise as soon as you consider collectives whose individuals are not strongly unified in their attitudes and willingness to bear costs. Wars are usually things decided upon by political and military leaders, whereas the costs are born mainly by individual soldiers and civilians. Engaging in futile resistance is costly; when the agent who acts and those who bear the costs aren’t the same, we can’t appeal straightforwardly to prerogatives to bear costs. In other words, leaders will have to judge for the collective that “standing up for what’s right” is worth the costs that others will bear to do so. That’s a complex issue in all sorts of ways – not having specifically to do with the framework for expressive harms, but certainly applying to it. Something similar applies when the actors are informal representatives of those they act on behalf of.

    On instrumental justification of rioting: I agree that rioting sometimes brings about positive outcomes. (See e.g. Enos, Kaufman, and Sands in APSR on the 1992 LA Riot.) But these are case-by-case empirical issues, and largely (I think) knowable only in retrospect. So I’m skeptical that we could judge in advance of a riot that it stands a good (enough) chance of realizing positive outcomes to yield an instrumental justification for participation. Then there’s the further question of how/whether individual acts of rioting contribute to, and are then justified by, the attempt to achieve such outcomes. My skepticism about this seems to outstrip that of most of the people I talk to about this! On the other hand, I feel it pretty firmly, and I think it’s based on a good reading of the empirics, which is why I keep insisting on it. But as I said to Patrick above, it’s certainly possible in principle that rioting could be justifiable both by its expressive fit and by its instrumental effects.

    @Patrick Thanks for the follow-up. I worry we may still be misunderstanding one another. The part of the paper you reference is trying to draw a contrast between graded contributions to some value and violations of that value. Surely enough graded contributions to injustice will amount to a violation, but I think it’s the violation that’s licensing a harmful instrumental response, and only indirectly the graded contribution. (I’m imagining ‘violation’ here as taking some kind of sufficiency analysis.) That’s why I’ve written “no amount of *extra* freedom or equality could *by itself* justify harming others…” – although I see how you could read that as strictly false.

    On the formal version of the challenge: I hesitate to make a “for all X” claim about a substantive question, but I *do* think that the kinds of wrongs that would justify rioting, i.e. wrongs violent protest against which would be fitting, are the same kinds of wrongs that would justify an instrumental response, i.e. defensive action. The challenges I raise in the paper for the defensive account of the justification of rioting are principally about whether the conditions on justifiable defense can be met in the typical case of rioting, not whether the grievances to which rioting characteristically responds are the kinds of things one might justifiably defend against.

    Taking a step back though, I think I see how I might face a bind based on the argument I’ve brought against Havercroft, and I take it this is what you’re pushing me on. I suggest that grievances and corresponding goals of the “freedom preserving” or “equality promoting” kind are either, depending on how we fill them out, insufficient to license harm or else *are* sufficiently severe but then collapse into violations that we already think may be violently defended against. Couldn’t there be cases that fall somewhere in the middle? Serious wrongs that wouldn’t license revolution, say, but which might license rioting? I’ll have to think more carefully about this, but it seems like such cases must exist. Then the question is whether in those cases, violence could be justified instrumentally. In principle, I think yes. But in practice, we’d need good evidence that sufficient improvement is likely to be brought about by rioting, and then also by an individual rioter’s contributory acts. For reasons I bring up in the paper and that I’ve been returning to in this discussion, I’m doubtful those conditions will be met.

    @Piero: Thanks for the kind words! I agree – I think the eruptive nature of riots, including specifically their timing, is very important to what kind of response they typically constitute, and I think that’s important to their moral justification. I also agree that fitting expression *may* have instrumental value, but for the reasons I emphasize in the paper, I would hesitate to make those a central part of any justification for participating in rioting. On last resorts: for reasons you bring up, I think the term is misleading. The main issue is that you shouldn’t use violence when you can accomplish the same thing just as well without it, and you should use less violence rather than more given the option. The way this cashes out in real cases is clearly very complex, and the “last resort” language tends to obscure that. And yes – I take your point certainly about radical flank effects, although I can’t say with any authority what social scientists have found about when/to what extent that effect operates.

11. Thanks, Tweedy, for the excellent paper, and Candice for the précis! This has been fun to read. I want to pick up on a few points raised in the précis before raising some worries about Strawson and fittingness.

    Starting with the last question from the précis—what should we want from the theory of rioting?—it’s worth highlighting that one’s answer to this question and the shape of the theory of rioting one ends up offering are connected. In an earlier comment, Tweedy points out that the aim is not to offer guidance to rioters necessarily, but rather to offer a theory applies at the level of individual decision-making. This is a reasonable response, but I’m not sure that it gets to the root of Delmas’s point. Since the view focuses on individual actions, it seems like it is best-equipped to speak to the individual considering whether to riot. It’s less clear to me how the view might answer other moral questions about protest, such as policy questions or questions about the justification of riots considered as events.

    I also find myself puzzled by the claim that rioting does not aim at some instrumental end. Delmas points out that this question might not be settled by the empirics, but even beyond that point, the claim is slippery. Flanigan supports the claim by arguing that rioters typically feel hopeless, and that rioting is a response to conditions of futility, even though rioters might, of course, hope that their actions will have some instrumental effects. But we can distinguish between how rioters feel, what they aim at, and what their activities aim at. And it does not seem farfetched to me that rioters can feel hopeless, but nevertheless engage in activities aimed at creating instrumental change. Flanigan rejects this move on the grounds that, even if the aims of rioters and the aims of their activities can come apart, the two must at least be compatible; and, he claims, it would be incompatible to engage in an activity aimed at creating change when one is motivated by the futility of one’s situation (273). But is that latter claim true? For example, couldn’t rioters attempt to defend themselves even when they know it to be futile, precisely as a way of expressing the futility of their situation? If so, then one can maintain that the aim of rioting is defensive, even if rioters also often have expressive aims.

    Moving on from the précis, I had some worries about the operative notion of fittingness. I think there is a tension between Flanigan’s account of the expressive content of riots and some of his remarks about what makes riots fitting as a form of expression. On the one hand, Flanigan writes that riots are a way of saying something simple like “no,” “no more,” or “this shit has got to stop!” This thinness of expressive content is important to the argument: the thinner the content, the more likely it is that it can be expressed through the blunt means of rioting, and the less demanding it will be for prospective rioters to engage in fitting political expression. On the other hand, Flanigan argues that what makes these forms of expression fitting is that they are part of the sorts of moral and political claims we exchange with other participants in the moral community, similar to what Strawson says about the reactive attitudes.

    This strikes me as an odd move for a few reasons. To start, it seems to me that the sorts of claims Strawson has in mind are at least partly instrumental. We blame others to express our resentment, but also to engage others and demand that they treat us with a certain degree of goodwill. But this seems inconsistent with Flanigan’s claim that rioting is not motivated by an attempt to achieve some instrumental end.

    Second, it also seems to me that the sort of claims Strawson has in mind often have substantially more content than “no” or “no more.” Blame is a typical example here, so we might think about someone who blames their spouse for cheating on them. This person is saying “no,” but they are also plausibly saying something about how the action has wronged them and demanding a particular form of regard from their spouse. Now, one could fill in the account so that rioters are also saying something more through their actions, but I think more would have to be said to substantiate this sort of claim, and I worry that doing so would make the account either more demanding or less universally applicable.

    Third, the reactive attitudes are occasioned by participants in the moral community failing to satisfy the moral demands made of them. But it seems to me that, in the cases described in the paper, something more fundamental is going wrong. That is, rioters are not treated as participants in the moral community in the first place—instead, others seem to adopt the objective stance towards them. It seems to me that, under these circumstances, a more radical and fundamental claim for recognition might indeed be called for than the sort that people make of one another in the moral community. To put it another way, the claim that someone ought to take up the participant stance towards you is different from the claim that, within the participant stance, they ought to manifest a greater degree of goodwill towards you. And it seems more plausible to me that rioters are making the first claim than that they are making the second.

12. This has been an interesting discussion to follow so far. Thanks Tweedy, Candice, and everyone else for your contributions. I wanted to raise a note of curious skepticism about the idea that a theory of the justifiability of riots should be focussed on — as Tweedy puts it above — “individual decisions about whether and how to participate, rather than to the event as a whole.”

    We can evaluate individuals for their actions, of course, and we each face a choice as individuals about whether to join a riot and (if so) how. A normative theory of action should have something to say about these issues. But I find it somewhat puzzling to think about how well individual choices like these /fit/ as expressive responses to the sorts of injustice that typically spark political riots.

    Naively, I guess I’m inclined to think the only thing that could really be interpreted as an appropriate expressive response to things like systematically racist police violence, the hegemony of transnational capital, or elite manipulation of grain prices is going to be something similarly big, structural, and collective. Of course, an individual can be outraged by such things and express that outrage (although I’d guess individual expressive acts are more often about particular symptoms or manifestations of these structural things). But is a riot really legible — morally and politically — as nothing more than a summation of such individual expressions? I’m kind of skeptical.

    As an analogy, consider spoiling a ballot in protest of the slate of candidates running for election. Although there might be a kind of self-satisfied purity in this minor expression of frustration, there seems to me to be something qualitatively different in a movement to do this which succeeds at a larger scale. When the returns show that a few people spoiled their ballot, it has virtually zero effect on how the election is interpreted, whereas 20% off spoiled ballots has a kind of political significance that doesn’t seem to reduce to the mere sum the minor expression of frustration in individual spoiled ballots.

    I don’t know if this really matters for the larger debate between Tweedy and his targets, but I’m wondering if the full moral “fit” of rioting as an expressive vehicle is something that can really be evaluated at the individual level.

13. Thanks Tweedy for the excellent paper and Candice for the thought provoking responses.

    You say that we can assess both individual acts of rioting, and the joint act of rioting, based on whether they satisfy certain conditions of fit (e.g. type-appropriateness, proportionality, etc). Take type-appropriateness, for instance. Whether the joint act of rioting is type-appropriate depends on the character of the riot. As you say, if might not be so, if it is cacophonous and without determinate expressive content. Suppose that large numbers of people joining an ongoing riot that is (up till that point) type-appropriate, but whose individual actions fail to be type-appropriate. Perhaps they don’t express fitting types of attitude in response to the injustice at hand, or maybe they’re not about the injustice at all. Suppose that such participation renders the joint act of rioting type-inappropriate. How should we now evaluate the actions of individual rioters who joined the riot prior to this shift in its character (at the joint act level)? I think the same point could be made about the other conditions.

14. Thanks to all! I want to pick up on some remarks from Candace regarding the justification of violence (Comment B). The first is this: “Justifying violence in rioting cannot rest on a simple comparison with the scale of the injustice protested.” The second comes a paragraph later, but I think is immediately connected: “What expressive rules or conventions, if any, make smashing shop windows, setting fire to state offices, and overturning cars the fitting vehicles for rioters’ worthy message of defiance?” Putting the two points together, I ask: what standards do we use to evaluate the fittingness of an expression, both in general and specifically in the case of rioting?

    Tweedy’s discussion of fit in IV.B. suggests an answer. For an expression to be fitting, it must satisfy the four criteria discussed there: it must be type-appropriate, proportional, correctly directed, and adequate. But for each of these, I ask the same question: what standards do we use to evaluate whether the criterion has been satisfied? Tweedy illuminates each criterion with examples, and while this may help clarify what is meant by each and why each plausibly has a place in an account of fit, it does not help one understand what is required to satisfy any of the criteria.

    My worry here is that we are supposed to have some “intuitive” grasp of what is needed to satisfy the criteria, which in turn makes me worry that the standards that are doing the account’s ultimate justificatory work have been uncritically assumed. Consider the standards that might apply to acts of gratitude. Tweedy explicitly discusses gratitude to illuminate type-appropriateness: “gratitude is a fitting type of response to kindness.” He might also have mentioned it in the discussions of proportion and adequacy. In many circumstances, if you buy me a beer and I respond with a garish display of gratitude, it is out of proportion; in many circumstances, if a dean expresses gratitude to an overworked chair with a five-dollar gift card for coffee, it is inadequate. The caveat “in many circumstances” matters in these and all cases, however, because norms of gratitude operate against a set of background social conditions. If we live, not in a liberal democracy, but in some hierarchically structured community, and I am your master and you are my servant, it may *never* be appropriate for me to express gratitude to you, no matter what you do, because gratitude presupposes a sort of status-equality that does not obtain between us. Similarly, there may be no limit to the gratitude that it suits you to express to me, should I do something kind for you, so it may be impossible for you to express disproportionately great—or perhaps, even adequate—gratitude to me.

    Another example: suppose a child screams at her mother, “I hate you! You’re the worst mom ever!” Is it fitting for the mother to respond, “Oh yeah? Well you’re the worst child ever!” One answer is, It *is* fitting, but there is a whole host of other reasons a parent should never say such things to their child. In line with the previous example, one might instead say, It would only be appropriate on the assumption of some equality between parent and child, but obviously that equality does not obtain. In that case, what is a fitting response? It wasn’t long ago that many places, a spanking was in order; in many places now, spankings are never in order. So my question, again: what are the norms that govern what is a type-appropriate, proportionate, and adequate response?

    When we turn this line of thought to riots, one might object that property destruction is simply never a type-appropriate response to injustice, full stop, or that it may be type-appropriate but can never be proportionate, full stop. What standards does one turn to to push back against such claims, besides appeals to “intuitions”, which, again, may amount to little more than uncritical assumptions?

15. Post script: my apologies, Candice, for misspelling your name above.

16. Thanks, Tweedy, for the fascinating paper, and thanks to Candice and everyone else for all the stimulating and perceptive commentary.

    I hope I won’t seem to be changing the subject if I propose here a broadening in the way we think about the justification of rioting. I think it could be helpful to widen our focus of concern to consider not only the ethical quality of the acts of rioting but also the character and dispositions of those who riot.

    If we consider the anti-ICE militancy that has unfolded in Minneapolis in the past week or two, I think that at least some of it could be called “rioting.” (This depends on how one defines rioting, obviously, but surely much of it is at least rioting-adjacent.) I think it is clear that many of us — millions of Americans, for instance — admire the anti-ICE rebels, who defiantly attempt to disrupt the actions of the ICE agents, sometimes chasing them out of neighborhoods, insulting and condemning them, attempting to intimidate them and to force the ICE agents to retreat. Can we justify our admiration — our ascription of civic virtue to the rebels — in terms of the “fitting expression” view of what’s best about rioting? Or do we need to find a different basis for our admiration?

    If we admire and ascribe civic virtue to the anti-ICE rioters, it seems not to be because we think their actions are permissible, although that might be a background assumption without which our admiration would seem paradoxical. We (have reason to) admire them because we grasp that people like that, people willing to riot, are indispensable in a political community that aspires or should aspire to promote justice and democracy. In the face of intransigent elites and unresponsive systems of power, we need some people who are willing to go beyond the requirements of duty, willing to take risks and pour uncommon energy into confronting a manifest evil that — because it may be perpetrated by the authorities, or may be perpetuated over time by the authorities, and so is not likely to be stopped by the authorities — forces us to look outside of and away from the police, the courts, or legislatures for the kind of forceful response we need.

    What we admire, it seems, is that the rioters rush to the scene to join the fight, in spite of various risks, and in this way they meet an urgent civic need, rising to the occasion of a civic emergency. What, if anything, the rioters are likely to achieve — an important theme in this discussion — can’t be the main thing that justifies our admiration, since we admire their rush to the scene of rioting immediately, as soon as we see it, not only after we form an opinion about whether or not it might accomplish something important. In this respect, it seems to operate like our admiration for a person who, seeing that someone is drowning, jumps into the water to attempt a rescue. The rescue-attempter’s admirableness is mostly, although perhaps not completely, independent of their prospects for success. We admire them because we sense that these are exactly the kind of people we need in a situation like this.

    So, perhaps fittingness is indeed the right concept, except that it is not a matter of the rioting as fitting expression of indignation (or ‘stop!,’ ‘enough!,’ etc.), in response to a grievous wrong, but fitting character in a situation where an urgent and demanding type of intervention is needed and many people are unlikely to provide it. We admire the rioters for being just the kind of people that we need in a situation like this, when our elites, institutions, and systems of power, are unresponsive or intransigent, outrageously insensitive to our most pressing grievances. In such circumstances, the uncommon willingness of some to riot, the fittingness of their character to what is demanded of us in a critical situation, is worthy of admiration, imitation, and cultivation. And this is a way of justifying it as a fitting response, albeit in a way that doesn’t quite conform to the ‘expressive’ view of what is morally important in rioting.

17. Good afternoon from Europe! Thanks so much to everyone who engaged with this discussion overnight here.

    @Henry: These are subtle challenges, so thank you. On riots considered as events: I certainly think we can assess riots as events – as things that are good or bad, for instance. What I’m doubtful of is that we can ask whether a joint activity that is only loosely coordinated like this is (non-derivatively) justified/unjustified. I *do* think that we can make derivative generalizations: if we take most of the riot participants to have been justified, it makes sense to talk about the riot as justified. This could be highly relevant to policy discussions. But the derivative/non-derivative distinction matters a lot to what kinds of frameworks are appropriate to apply to the phenomenon.

    On mixed motivations, and hoping for change while protesting its impossibility: That’s an interesting possibility that I should think more carefully about. My flat-footed response to the particular example you raise is that you can’t aim to defend yourself as a way of showing that you can’t defend yourself — then you’re actually just aiming at demonstrating the futility of defense, rather than aiming at defending yourself. Some work is being done here by the ambiguity of ‘defense’/’defend’, but for this purpose we need to read it strictly as aiming at success.

    Likewise, the set of challenges you raise to my appeal to Strawson involve subtle issues I can’t do justice to here, and which I’ll have to think carefully about. I’m inclined to think that basic reactive attitudes like blame, while they may have complex specific content, also have a thin expressive ‘core’ — in that case, something like ‘that was wrong!’. I also want to resist the construal of the practice of engaging with others via reactive attitudes as instrumental (though we’re reaching the boundaries of usefulness of the instrumental/non-instrumental distinction here). I don’t blame you *in order to* get you to (say) apologize, even though I might want and expect that you do that. We can see this by considering the appropriateness of blame even when we’re sure the wrongdoer won’t repent. And on your final point: I agree, treating someone as outside the moral community is different from treating someone wrongly but as a member of the community. However, I think the notion that insisting that one is owed concern (in this sense) is a fitting response to denials of proper concern is consistent with the spirit of the Strawsonian idea. (I make this point more explicitly in the discussion of Strawson in “Futile Resistance as Protest.”)

    @Matthew: Let me say first, since I only touch on it in passing in the paper, that I really admire yours and Graham’s work on the subject, and I’ve learned a lot from it! On the point you raise: There is something interesting and complex about the *kind* of joint action that rioting is. Because on the one hand, riot participants are often out there doing things basically at their own direction and discretion. But on the other hand, they’re doing those things in part because they understand them to be part of something that all the other participants are also contributing to. This is stretched, but it might be a bit like joining in on a spontaneous public singing of some song. When we think about the thing as a whole, we think about what The Riot expresses, or what The Song sounded like. But when we ask questions about the justifiability of the thing, for reasons I emphasize in the paper, I think we have to focus on individual acts. I agree that this doesn’t seem like a simple summation, and I agree that it’s interesting and a bit mysterious. But it doesn’t strike me as implausible that we could care about the character of the whole event while focusing our moral attention on individual participatory acts.

    @Chong-Ming: As you point out, I link the justifiability of decisions to participate in rioting to judgments about the overall mood of the event. Those judgments might turn out to be wrong, and they might also be harder to make at different times in the event. If you participate in a riot that later turns out to be mostly about something odious (because it turns out most participants were there to protest something else), you might regret your participation. Depending on how well-founded your initial judgment was, you might either have been a) justified but mistaken, or else b) simply unjustified in your decision to participate. Rioting is chaotic, so it doesn’t strike me as problematic that it should have features like this.

    @Graham: Let me reiterate how much I appreciate yours and Matthew’s work on this topic! On your question: I completely agree that there’s a big question here of the form, Why are *these* acts fitting responses to *those* wrongs, which is a question not only about the particular form of appropriate response but also about the social context that partly determines the question of appropriateness. I only gesture at answers to these questions in the paper. But I’m not *too* worried about that for the viability of the argument. Saying ‘no’ is basically how we respond to wrongful treatment. Insisting on that, including shouting, is how we respond to repeated failures to heed the initial ‘no.’ For beings like us, in societies like ours, violence is the most extreme form of shouting ‘no’ — one that is rarely called for, but sometimes nevertheless is. So I don’t think this kind of response is highly conventional or socially contextual in a way that would open it up to mere insistence on the opposite.

    On telling your child they’re the worst: That children aren’t fully responsible members of the moral community is one (main?) reason adults shouldn’t do this. Not only can children not engage in wronging in the same ways that adults can – which is why their insults, while hurtful, can often ultimately be disregarded – the right response to the wronging they *can* do is dominated by the importance of moral education rather than fitting interpersonal response. But imagine the same example with a spouse or friend. Telling your partner that no, they’re the worst! may seriously damage the relationship. So for its sake, you might not respond that way, even if it would be fitting. Or you might choose to rise above the hurt they’re doing you for their sake, as an act of grace. I take it we all do this, and are given this in turn, by friends and partners over the long course of relationships. I think these kinds of alternative responses can be worth pursuing, and I think the best way to read commitment to principled nonviolence is basically like this.

    @Steve: I’ve also learned a lot from reading your work on this topic, so thanks so much for your engagement! Admiration is certainly the kind of response that falls under the category of the fitting, and brave people who resist oppression certainly merit our admiration. I’m not sure, but I’m inclined to think of admiration – including moral admiration – as a form of appreciation, and so as a form of aesthetic (rather than moral) response. So then the connection to justifiable harm would be more difficult to draw. But the thought is interesting: maybe one should act in ways that would be fitting for others to morally admire? This also suggests an interesting connection to virtue ethics and the normativity objection that people often raise against it. I’ll have to keep thinking about that.

18. I would also like to very briefly say something about the good point Graham (if I may) raised above, about how to evaluate whether a particular instance satisfies each of the four conditions Tweedy proposed. I had the fortune to have some good discussions with Tweedy, and this is the “methodology” we used in a joint paper:

    We first find one or two cases where it is intuitive that one condition is met or not met. We then try to say something like “in the case we want to justify, we think it is less severe than the paradigm case.” So for example, we think that some cases of property damage in climate protest may be justifiable as fitting protest, because a) the magnitude of harm caused by some corporation is more severe than a paradigm case, and b) property damage is less severe than acts of interpersonal violence in some paradigm cases, so on and so forth.”

    We understand that if people don’t share the intuition of some paradigm cases, the argument won’t work. But what we try to do is to draw from the growing literature of futile resistance, and start with the common assumption that some paradigm cases of futile resistance are justified. The argument then goes as “if we think the story of *why* this paradigm case is justified works, then by the same story, certain acts of resistance (in our case climate resistance), should be justifiable by the same story.”

19. Thank you very much Tweedy for your paper. I enjoyed reading it!

    While reading the argument that treats rioting as a fitting form of expression, I began to wonder how this view should handle cases in which riots stem from false beliefs.

    Consider riots that are motivated by anger or frustration grounded not in actual social injustice, but in misinformation or fabricated claims. The attack on the U.S. Capitol on 6 January 2021 is a clear example. The participants expressed anger, resentment, and a sense of grievance, but these attitudes were largely based on demonstrably false beliefs about electoral fraud. In such cases, there appears to be a gap between the expressive content of the riot and the reality of the social world it purports to respond to.

    This raises a question about the conditions under which expressive aims can genuinely justify the harms of rioting. It seems plausible that expressive justification requires more than the mere presence of anger or frustration. Rather, there must be a real injustice or wrongful social condition that makes such expression fitting in the first place. If there is no such injustice—if the grievance is epistemically defective—then the riot risks becoming normatively empty or unjustified, even if it is sincerely expressive of participants’ emotions.

    This suggests that there is a closer connection between the permissibility of rioting and the epistemic conditions under which it arises. Expressive anger may be morally relevant only if it is appropriately responsive to the facts. Where the underlying beliefs are false, distorted, or culpably maintained, the expressive aims of rioting lose their justificatory force. In that sense, the riot is not merely unjustified because of its harms, but because its expression fails to track any genuine moral wrong.

    The concern, then, is that an account that focuses solely on expressive aims risks overlooking the epistemic dimension of protest. Without an account of when anger, resentment, or outrage are warranted, it becomes unclear how to distinguish justified riots from cases of collective wrongdoing driven by misinformation. If rioting is to be justified as fitting expression, the theory must say something about the epistemic standards that grievances must meet. Otherwise, it cannot adequately explain why riots grounded in false beliefs—such as the Capitol attack—fail to qualify as legitimate forms of protest.

20. Hi Tweedy. Thank you for this excellent piece. And thanks also to Candice for her thorough and provoking précis. The ongoing discussion in the comments is really enlightening, and it’s pushing the stakes in important theoretical discussions.
    Reading thought the paper, the précis, as well as the various comments posted since yesterday, I find myself wondering mostly about two points.

    # Property destruction

    I think that some pressure might be put on the correct-directionality of riots in relation to property destruction.

    On this point, Tweedy builds an expressive analog to the “right target” of just war, and he is right, in my view, in pointing out that the addressee and the audience are not the same, and that targeting the actual addressee would be normatively stronger than targeting the audience.

    Of course, one issue is that this is difficult in practice. Property destruction during riots very often does not target the addressee but seemingly random targets (eg. local shops, convenience stores), which may not even be part of the target audience beyond simply being part of some “general public” or collective.

    For Tweedy, such cases of property destruction can be justified, as “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.” (p. 288)

    In my view, this position is too permissive. I read Candice Delmas as making a similar point in her précis above when she argues that justifying violence and property destruction requires evaluating how it “setbacks […] affected people’s life plans.” Expressive damage should probably take into account the direct and indirect harms that it has on targets, especially when these targets’ connection with the underlying motive of the riot is at best ambiguous and at worst inexistent.

    But it seems to me that we might even legitimately ask the following question: Even when expressive damage is *correctly* directed, should property destruction face a much higher bar when it predictably harms someone’s basic means of subsistence and livelihood—things like housing, basic financial security, basic local infrastructure?

    I’m curious about how concerns about threats to the basic livelihood of those targeted during riots can be balanced against the greater expressive aim of riots.

    # Democratic implication?

    Another point I wish to make concerns Tweedy’s move from communication to what he calls “directed expression.”

    On his view, rioters’ claims are “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.” (p. 282) I think that this is basically right—insofar as riots’ audience generally rejects rioters’ claim because of the (destructive, violent, disruptive, transgressive) way in which such claim is articulated.

    My concern is whether the fact that rioters’ claim is not at least oriented toward uptake by some public audience undermines the political—and perhaps even the democratic?—potential of riots. It seems to me that riots are quite clearly communicative: They are a directed communication of grievances at target audiences. (Although, as the discussion so far shows, there is disagreement about the empirical evidence concerning the “commonality” of the grievances and of the claim made by rioters as a whole, and of their intended audience.) However, as Tweedy correctly points out, most of these grievance-based communicative claims do not end up being “taken up” by their target audience and “often fails to be understood.” (p. 269)

    Now, I wonder about the democratic dimension that might come up here. In my view, the normative issue that follows from the particular rioters-audience dynamic is that, for riots to have some political weight, they have to somehow “translate” into the public sphere—meaning that while people might not be responsive, rioters at least have to attempt at making some sort of contribution to the overall communicative realm on which democracy operates. However, my concern is that Tweedy’s “directed expression” abandons that basic attempt at communication by simply falling back on the non-instrumental value of expression by rioters.

    Perhaps this point falls outside the purview of the sort of justification Tweedy entertains. Nonetheless, it seems that, if we accept the non-instrumentality of “directed expression,” this has some intriguing democratic implications that might be worth exploring. In any case, this is perhaps more an invitation to Tweedy about what he considers the democratic implications of his approach.

21. @Ten: Thanks for popping in 🙂 And thanks for the clarification! The joint work Ten mentions, about futile resistance and climate change, is up on PhilArchive now, and I encourage anyone interested to take a look.

    @Ignacio: You’re right, it very much matters to the justifiability of rioting whether the grievances expressed by rioters reflect reality. This is similar to paradigm reactive attitudes: my gratitude is fitting only if you’ve in fact done me a kindness, not if I mistakenly believe you to have; likewise with blame, and so on. A consequence of this is that, in order to be justified, rioters need to be right, or at least epistemically justified, about what their grievances are. This is a substantial burden — but then again, rioting is an extreme form of political dissent, so that’s the kind of standard that seems appropriate to me. And you’re also right: this is exactly what I’d say about the January 6 Riot — insofar as it was a riot (some people disagree about this), its central failure was in being about a core grievance that was simply false.

    @Alexis: About property destruction: I take all of your points, and I guess I suspect judgment will come down to the particulars of particular cases. That may be unsatisfying, but given the chaotic and varied nature of rioting, I’m not sure there is anything less schematic and more substantive to say. Regarding harms to co-citizens’ property: my general view is that robust solidarity or allyship requires a willingness to bear actual costs, and I think that informs my stance here. I also think that yes, one shouldn’t destroy another’s basic means of subsistence or livelihood, although I don’t know whether you have in mind something like Sal’s Pizzeria (from Do the Right Thing) or something like the home of the fellow-oppressed. I don’t think I’d want to put those in the same bucket.

    About directed expression: I think this is an interesting category — it seems clear to me that sometimes we do make moral claims on those we don’t expect to listen. But I don’t think we have to be doing it for our own sakes, for instance, for the practice to make sense. We’re saying something *to* someone who *should* listen, and you’d like them to listen and respond. There’s just not a means-ends orientation to it; you aren’t saying it, or saying it in any particular way, in order to best get the addressee to listen and respond. This sounds like it could count as what you call a “contribution to the overall communicative realm”? Separately, I’m sympathetic to what you write in “Can riots represent?” about the democratic potential of riots; and of course I think it’s good when rioting contributes positively to democracy, social liberation, and so on. My point here is just that I think rioting’s justification is best understood as not dependent on those things.

22. Let me just say thank you once more to Candice for a wonderful critical précis, to everyone for their thoughtful discussion, and to Susanne, Kartik and all the PEA Soup folks for making this possible!

    I’d love to continue talking about any of this going forward, so please don’t hesitate to write to me.

Comments are closed.