Proponents of civil resistance have long believed that nonviolent methods rival the effectiveness of armed force as means of resistance. Among recent advocates, Erica Chenoweth and Maria Stephan have, perhaps, been the most prominent.Footnote 1 Their use of quantitative methods to compare contrasting means of bringing about progressive political change has helped carry forward a powerful tradition found, notably, in the ideas and practices of Mahatma Gandhi, Martin Luther King Jr., and Gene Sharp. In his recent book Ethics, Security, and the War-Machine: The True Cost of the Military, Ned Dobos offers an important extension of this view through a critique of the widespread commitment by states to maintaining established military forces.Footnote 2 Building on Sharp's idea of civilian defense systems (CDSs), Dobos argues that nonviolent methods offer a viable substitute for regular armed forces in the face of international aggression. If CDSs are potentially as effective as regular armies trained in the use of violence in defending against external military threats, then he thinks it undermines the common assumption that democratic states need to maintain professional armed forces. In their present form, these forces pose a number of dangers. Their training commonly generates a moral code and culture at odds with the values of civilian life. And it is impossible to eliminate the danger that soldiers will leverage their violent skills and equipment to overthrow civilian authorities in a coup d’état. So, if the methods of a CDS compare favorably to those of the armed forces in resisting foreign occupiers, then democracies ought to consider defunding the army in favor of training their citizens in organized nonviolence. This offers strong support for Dobos's wider claim that ethicists cannot simply assume that the armed forces on which just wars rely are legitimate institutions in the first place.
I offer a twofold critical response to this line of thought. First, I argue that if the civil resistance methods associated with a CDS really are as effective as their proponents claim, then this would mitigate the dangers of military coups d’état by offering citizens effective tools to defend against usurpers. But second, the same fact should also make us more circumspect about nonviolent methods. Organized civil resistance can be employed subversively as well as progressively. If it is a powerful tool in the hands of democratic egalitarians, then it will be no less powerful in the hands of fascists. This is not to say that civil resistance methods cannot be valuable—they certainly can be. But at least some of the methods used by civil resisters are valuable in a more narrowly instrumental way that is akin to the value we place on armed force: both can serve just ends with varying degrees of effectiveness, but both are dangerous, morally and politically; and both, therefore, ought to be used only for certain ends, under limited conditions, and with due restraint. Dobos recognizes that nonviolent “weapons” could be subject to “occasional misuse.”Footnote 3 But I argue that their versatility indicates a much deeper problem.
To make my case, I focus on a pair of dichotomies around which the argument for nonviolent civil resistance is constructed. The more fundamental one juxtaposes violence with nonviolence. The second builds on this binary, contrasting military establishments, their equipment, training, methods, and ethos with the methods, training, and spirit of civil resistance and CDSs. Turning away from a violence-nonviolence binary structured around harmfulness, I draw on Richard B. Gregg and Hannah Arendt for an account of nonviolent power defined by “noncoerciveness.” Whereas nonviolent coercion has the potential in the wrong hands to subvert democratic institutions, just as armed methods can, nonviolent power is a necessary bulwark against both violent and nonviolent forms of subversion. Rather than divesting from an armed force or investing in a CDS, I therefore think democracies ought to focus on sustaining the beliefs and civic virtues that animate the nonviolent power at the heart of their political institutions.
Civilian Defense and Civil Resistance
My first, more superficial argument takes aim at Dobos's claims about the risks associated with rival types of defense institutions compared with their effectiveness. Dobos suggests that domestic military establishments endanger democracies by equipping a subset of their citizens with hardware, training, and an ethos that lend themselves to military takeover. The resulting risk of coup d’état, combined with what Dobos claims is the potential of a CDS to oppose invasion and resist occupation, bolsters the argument that we ought to consider seriously the possibility of abandoning arms altogether. Were there no other way to resist foreign aggression than a military one, then the coup risk might be worth taking. But if a CDS offers a viable alternative, then disarmament (or “transarmament” from armed force to civil resistance) is worth considering.Footnote 4
Dobos argues from two premises. The first is that we have as much (or nearly as much) reason to fear the tyranny of domestic armed forces as we do the tyranny of foreign armed forces. Life after a domestic coup d’état more closely resembles the experience of foreign occupation than we might think.Footnote 5 The second premise is that unarmed civil resistance is as effective, or nearly so, at resisting foreign invasion forces and occupations as organized professional armed forces are. Based on the second premise, Dobos maintains, we are likely to be as well off employing civilian defense as we are employing military defense, so far as foreign invasion is concerned.Footnote 6 And if we add to that the first premise, then we can even argue that we are better off employing civilian defense than employing a professional military. This is because harnessing civilian defense against foreign invaders, instead of using military force, obviates the risk of coup and domestic armed tyranny.
I will question the latter claim later in this essay. But, for now, my immediate objection to Dobos's argument is that if the methods of civil resistance outlined by Gene Sharp are likely to be effective against foreign invaders and occupiers (the greater threat), then presumably they are likely to be effective enough to address domestic putschists, too. The availability of civil resistance methods means that the vulnerability of civilian populations to their own armies is less acute than might at first have appeared, which should reassure those worried by the coup risk posed by a professional military establishment. That being the case, the argument about the potential of CDSs to resist military force blunts rather than supports Dobos's argument about the danger of a coup d’état.
But, of course, this objection would not actually defeat Dobos's argument outright. He only suggests that abolishing the military and concentrating instead on building a CDS is something we ought to consider in light of his premises. And he succeeds in showing that this is a question worth asking. However, what the objection does indicate is that a comparison of the respective defense systems—military and civil—is likely to indicate less dramatic differences in their respective balances of benefits vs. risks. If this is the case, then it is less likely that projects for radical transformation will be proportionate. Shifting from armed to unarmed defense will involve costs and risks of its own (including, one would imagine, the risk of triggering a coup). The morally relevant differences between the alternative systems need to be large enough to outweigh these risks. Whether or not they are large enough depends on detailed estimates and comparisons of the salience of different methods and institutions.
The best arrangements are likely to be those supported by relatively small differences in the degree of effectiveness when we compare alternatives. For instance, if maintaining conventional armed forces is generally a little better than civil resistance at warding off foreign invaders but poses a risk of domestic tyranny through a coup d’état, then the fact that civil resistance methods are at least somewhat effective in resisting militaries (whether domestic or foreign) might reduce the coup risk enough to make a continued commitment to arms worthwhile.
Nonviolence as Unarmed Force
However, there is something more fundamental to say about the juxtaposition of military establishments and CDSs. It concerns an ambiguity in the underlying dichotomy between the violence of the one and the nonviolence of the other. The ambiguity comes into view if we turn to the writings of two thinkers whose ideas closely resemble each other in some respects and yet who are not often brought into dialogue: Richard Gregg, whose Gandhian tract The Power of Nonviolence first appeared in 1934, and Hannah Arendt. Dobos contrasts violence and nonviolence in one way; Gregg and Arendt contrast them in another. In doing so, Gregg and Arendt imply a different binary from Dobos on the basis of which to evaluate various types of tactics. Reviewing the practices of civil resistance associated with CDSs in light of this rival binary indicates a further source of potential danger to democracies, one partially concealed by Dobos's Sharpian analysis.
Let us assume that we can agree on what “violence” is. Perhaps, as Iris Marion Young puts it, it can be defined as “acts by human beings that aim physically to give pain to, wound, or kill other human beings, and/or to damage or destroy animals and things that hold a significant place in the lives of people.”Footnote 7 The question is, what should we understand truly “nonviolent” action to be? In Dobos's account, nonviolence appears chiefly in the guise of various kinds of tactics that Gene Sharp mapped out for challenging dictators and foreign aggressors. In a CDS, for instance, citizens would be trained in “how to organize targeted protests and boycotts, rally support, enlist more people to their cause—everything they would need to do to frustrate the ability of a foreign aggressor to benefit from his aggression.”Footnote 8 The moral superiority of these methods to violence, Dobos indicates, has to do with harming: the “important difference,” he writes, is that unlike a military establishment, “a CDS would not rely on killing and maiming to achieve its objectives.”Footnote 9
I wonder, however, whether this really is the most important moral difference. An alternative suggestion can be found in Arendt's On Violence (1969), where she offers her famous twofold argument that not only must power and violence be sharply distinguished but they also should be considered opposites. Conceptually, she writes,
power corresponds to the human ability not just to act but to act in concert. Power is never the property of an individual; it belongs to a group and remains in existence only so long as the group keeps together. When we say of somebody that he is “in power” we actually refer to his being empowered by a certain number of people to act in their name. The moment the group, from which the power originated to begin with . . . disappears, “his power” also vanishes.Footnote 10
Whereas power, Arendt argues, is an end in itself, violence “is distinguished by its instrumental character. Phenomenologically it is close to strength, since the implements of violence, like all other tools, are designed and used for the purpose of multiplying natural strength.”Footnote 11 Of course, as a form of concerted action, power, too, may be directed instrumentally toward ends. And it can also empower some individuals to wield force on behalf of that power; for instance, to defend it against external threats. But instrumentality does not define power. Power does not come into existence purely to serve ends, even though ends may issue from it and may themselves call instruments into being or into service.
Crucial to Arendt's differentiation of power from violence is not only instrumentality but also coercion. In contrast to speech, by which citizens may interact and conduct their affairs through “persuasion,” violence imposes one's will “through mute coercion.” Coerciveness is what makes violence so “disgraceful.”Footnote 12 As such, it lends itself to those forms of strength that Western thinkers have often mistaken for political power.Footnote 13 Or, to follow Gregg, we might put it more pluralistically: By focusing too closely on one form of power—James Tully suggests calling this “power-over”—Western political thought has neglected another, perhaps more important form of power, “power-with.” In Tully's paraphrase of Gregg's Gandhian philosophy, “The power of non-violence, or satyagraha, is the intersubjective power of interacting ‘with and for each other’ in cooperative ways in interdependent relationships with oneself (ethics), other humans, all life forms, and the spiritual dimension of existence. It is ‘power-with’: the type of power that animates and sustains all branches of a nonviolent way of life.”Footnote 14 The opposing practice, then, involves coercion and subjection to one's will: “In contrast, violence and domination are the general type of power exercised in violent conflicts and unequal relationships of domination and subordination that are imposed and backed up by force, or the threat of force, and various types of legitimation. It is ‘power-over’ in its many forms.”Footnote 15
So, whereas Dobos contrasts the physical harmfulness of violence with the absence of physical harm when using nonviolence, Arendt and Gregg juxtapose a practice of violence identified closely with wielding coercive power over others with the noncoerciveness of the power-with that arises from acting and interacting with others in cooperation. The significance of this alternative distinction becomes clear when we consider the types of action that it excludes from the category of nonviolence. These will include some of the tactics—and, indeed, the general understanding of resistance—promoted by Sharp.
Gregg has a very specific idea of how the right sort of nonviolent action functions: “We must understand one point thoroughly,” he writes:
The aim of the nonviolent resister is not to injure, or to crush and humiliate his opponent, or to “break his will,” as in a violent fight. The aim is to convert the opponent, to change his understanding and his sense of values so that he will join wholeheartedly with the resister in seeking a settlement truly amicable and truly satisfying to both sides. . . . The function of the nonviolent type of resistance is not to harm the opponent nor impose a solution against his will, but to help both parties into a more secure, creative, happy, and truthful relationship.Footnote 16
This cannot be achieved by any kind of arm twisting; only persuasion can do it. Compare this with Dobos's characterization of the tactics of moral “Jiu-Jitsu” advocated by Sharp as they might be used against a foreign occupier:
Non-violent resistance puts an aggressor in a kind of lose-lose situation. If he tolerates the non-cooperation and obstruction of the natives, it effectively means relinquishing control of the territory and its people. It probably also means losing the ability to extract resources, insofar as this depends on the participation of local labourers, farmers, technicians, transport workers, and the like. This would seem to defeat the very purpose of the aggression, leaving the occupier with little reason to remain. On the other hand, if the aggressor uses violence in an attempt to compel cooperation, this is both expensive and liable to backfire in the ways just described, and again the aggressor cedes power. … If the attacker does not use force he cannot take down the defender. But if the attacker does use force, his own momentum is turned against him, he is thrown off balance, and again the defender is left standing.Footnote 17
Like Gregg, Sharp recognizes that nonviolent action might convert its opponents, but he thinks this is very rare. Instead, he emphasizes “mechanisms” that can be used to “chang[e] the conflict situation and the society so that the opponents simply cannot do as they like.” Centrally, these include what he calls “coercion.”Footnote 18 Coercion encompasses measures by which “the system may be paralyzed by resistance,” while “noncooperation” is used to prevent a regime from performing necessary functions “unless the resisters’ demands are met.”Footnote 19
So while Sharp and Gregg both use the term “Jiu-Jitsu,” they use it to refer to quite different things. Sharp relishes the potential of an imaginative array of trip wires that can be used tactically to cause opponents to stumble and become entrapped. The will of the opponents is broken once they realize that they will be continuously thwarted and that further efforts will be self-defeating. By contrast, Gregg thinks that any attempt to gain such an advantage over one's opponent falls back into the trap of domination and, hence, reenacts the distinctive evil of violence. As Tully puts it, when “resisters mobilize ill-will, fear, anger, and enmity, and engage in strategies to gain power over violent opponents, then both contestants are playing the same power-over game. It is war by other means.”Footnote 20 Instead, Gregg thinks, the resister ought to seek to change the opponents’ minds so that there is no need to coerce them.Footnote 21 This contrast highlights, as Tully says, “the greatest and most important dis-analogy with jiu-jitsu and all violent methods”Footnote 22: truly nonviolent action eschews the coercive power-over entirely that is inseparable from violent methods.
Arendt's and Gregg's reflections call into question the simple violence-nonviolence binary on which Sharp's and Dobos's analyses are constructed. Let's say we continue to identify “violence” in the ordinary sense of the word, more or less as Young defines it. We now have two types of unarmed action, each distinguished from violence in a different way: on the one hand, there is what we can call the “unarmed force” of Sharp's jiu-jitsu; on the other, there is the persuasive power-with of Gregg's pure nonviolence: let us call this “nonviolent power.”Footnote 23 So instead of one dividing line, we now have two. Moreover, each construal of the violence-nonviolence binary establishes a different moral hierarchy. What distinguishes violence from both unarmed force and nonviolent power and renders it morally dubious by comparison is its intentional harmfulness. But what distinguishes nonviolent power from both violence and unarmed force is the absence of any attempt to use nonviolent power to dominate and coerce.Footnote 24
Dobos's analysis builds on the first binary, contrasting harmfully destructive acts (violence) with actions that do not instrumentalize harmful destruction but that, at least some of the time, coerce (unarmed force plus, perhaps, nonviolent power)—tactics (“non-violent ‘weapons’”) such as boycotts, strikes, ostracizing those who cooperate with the opponent, blockades, and sit-ins.Footnote 25 My suggestion, by contrast, is that the second binary is at least as important. Instead of emphasizing destructive harm (violence alone), we ought to focus on coercion (not only violence but also unarmed force). To see why, we need to turn to the dangers that coercive forms of nonviolence can pose. Like organized arms, organized nonviolence, too, can be turned to subversive, anti-democratic ends.
Dangerous Nonviolence
Dobos's argument for a Sharpian CDS is not based solely on tactical grounds but also on the grounds that it promotes civic virtue. Citizens trained in CDS methods learn new political skills and are inspired to effect change through action outside the formal constitutional and institutional channels of the state:
If they previously thought that non-electoral political action couldn't achieve anything tangible, they will have been disabused of this notion. And if they previously lacked the practical know-how necessary to effectively take such action themselves, they will have gained some valuable experience—and perhaps even a taste for it. . . . [T]hese men and women would henceforth be better democratic citizens, in the sense that they would be more inclined to participate in politics beyond casting their votes in periodic elections. A post-military defence system . . . would double as a kind of participatory citizenship training.Footnote 26
Characterized in this way, what Sharp and Dobos propose sounds like it could deepen nonviolent power in a way that speaks particularly to Arendt's civic humanist orientation. But recent events point to some dangers too.
These dangers come into view if we compare different types of coups d’état. One such possibility is exemplified by the storming of the U.S. Capitol on January 6, 2021. Robert O. Paxton, a historian of fascism, thought that Donald Trump's “open encouragement of civic violence to overturn an election crosse[d] a red line” that finally made it “not just acceptable but necessary” to describe Trumpism as “fascist.” He compared the event to similar attempts to thwart the outcome of democratic elections by armed force in Europe in the 1930s.Footnote 27 Many commentators have described the riot at the Capitol in 2021 specifically as an “insurrection” and a “failed coup.”Footnote 28 But perhaps when we picture a coup, we might imagine something more orderly than this. More typical is a case like Chile in 1973 or Egypt in 2013, in which some of the generals stepped in to snuff out civilian power, arrest civilian leaders, and deploy troops to take control of state media. If Trump was engaged in an attempted coup d’état, he acted without this sort of military backing. But there is also a third possibility. It is more than imaginable that a coup that chiefly employs nonviolent tactics might occur.
We can see what this would look like by peering behind the growing violence of pro-Trump activists early in 2021. The Capitol riot was the final denouement in a longer sustained attempt to mobilize popular and elite resistance to the democratic transfer of power—through nonviolent means classically associated with civil resistance. Organizers used social media to expand and galvanize support for the defeated incumbent and coordinate action. They deployed armies of volunteers to engage in different sorts of protest. Some posted themselves as “observers” at election counts, exerting pressure on those counting the votes with the ostensible aim of deterring them from disregarding pro-Republican votes. Others chanted “Stop the count” outside voting stations that were likely to contribute to the growing tally of Democrat votes. Various kinds of public protest were staged to pressure state officials. Some were more targeted, such as those staged at officials’ offices or, as in the case of the Georgia governor, at their residences. Other large, urban protests and rallies demonstrated public dissatisfaction more generally with the emerging electoral picture. Protesters’ immediate aims varied, but the overarching political strategy seems to have converged in an effort to coerce some officials and representatives and embolden others within the U.S. electoral institutions into overturning the election result. Protesters hoped that these officials might do so by boycotting or thwarting procedures; for instance, by refusing to validate electoral counts or by disregarding the state-level results when selecting electors to the Electoral College. If enough officials and representatives had been cajoled into coordinated action, then their withdrawal of support from legitimate electoral procedures—it seems to have been hoped—could have facilitated a bloodless coup (or what Secretary of State Mike Pompeo called, “a smooth transition to a second Trump administration.”)Footnote 29
Threats of violence often accompanied intimidation tactics, of course—sometimes they were implicit, as when protesters openly carried arms, and sometimes explicit, in threatening text and voice messages. Such threats fall well outside the Sharp playbook. But many of the tactics deployed on behalf of Trump can quite readily be characterized in terms redolent of civil resistance and CDSs. At the very least, they show how powerful these methods can be when used for anti-democratic purposes. And this puts the civic skills they cultivate in quite a different light. Trump-supporting citizens, too, put their faith in “non-electoral political action,” hoping it could “achieve [something] tangible.” They showed a growing “inclin[ation] to participate in politics beyond casting their votes in periodic elections.” And, following the lead of social media influencers and networks, they learned how they might “effectively take such action themselves, [gaining] some valuable experience—and perhaps even a taste for it.”Footnote 30
If Trump's attempted coup employed some of the methods of nonviolent civil resistance, it was thwarted by a different kind of nonviolent—but also, crucially, non-coercive—action. What ultimately defeated the Trumpists was the continued active power-with invested in institutions.Footnote 31 Even when their sympathies were with the Republicans, most officials, representatives, and leaders were too deeply invested in the principles and practices constitutive of the U.S. political system to succumb to external pressure. This institutional investment and the coordinated action through which it is expressed is precisely what Arendt calls (nonviolent) power. By contrast, if coercive violence had any significance in those events, it was chiefly as a means available to the state (especially in the form of the National Guard) to contain the chaotic destruction threatened by frustrated pro-Trump activists once it had become clear that their attempt at civil subversion had failed.
Conclusion
Sharp declares that “due to their nonviolent nature, the weapons of civilian-based defense cannot generally be used for the purposes of repression.”Footnote 32 This might be true, but it does not mean that these weapons cannot be used to bring about more repressive political outcomes. What is most threatening to the institutions of democracy is not the ability to inflict violent harm as such, but the widespread availability of the means of coercion. And this is precisely what the forms of civic action Sharp and Dobos envisage are designed to supply. The ability to inflict physical harm can, of course, amplify one's ability to coerce. But it is not the only means of doing so. The aim of CDSs is to train citizens in a wide array of different techniques, many of which serve the purposes of resistance by coercing opponents. From the point of view of Gregg or Arendt, this is little different from training them in the use of arms.
From an Arendtian theoretical point of view, the sort of power that can truly provide hope of withstanding the coup risk—whether posed by the armed forces or by a nonviolent army of CDS practitioners—is fundamentally different from the coercive power-over that many nonviolent tactics are designed to secure. Unless this power already exists within the institutions of the state, turning to nonviolent means of resistance as an alternative to military force merely trades one form of subversion for another.
The use of coercion does, of course, potentially have a place in democratic political activism. In nonviolent forms (and perhaps, in some cases, even in violent forms), it has played a role in the establishment, defense, and deepening of democratic rights and institutions.Footnote 33 It can also help defend democratic power and its institutions from actions aimed at overturning them. But like all coercive means, their employment for such purposes must be constrained within an ethical framework that carefully defines the appropriate occasions for using them, the limited ends for which they might be used, and the restraints that activists must observe when using them. In other words, like violence itself, nonviolent methods of coercion need to be constrained and guided by an ethics akin to just war theory.Footnote 34
Open access
