Peter Gelderloos in his book How Nonviolence Protects the State claims that nonviolence is ineffective, racist, statist, patriarchal, tactically and strategically inferior, and deluded.[1] His attack on nonviolence is fierce and unrelenting.

To assess Gelderloos’ views, I first outline the case for nonviolence and the associated case against violence. This provides a foundation for examining Gelderloos’ arguments. I give special attention to his questionable assumption that violence always triumphs over nonviolence. In my judgement, Gelderloos’ arguments are based on pervasive double standards. In addition, he fails to spell out what levels and types of violence he considers acceptable, an omission that undermines his argument. Finally, I comment on connections between anarchism and violence/nonviolence.

I am a longstanding supporter of nonviolent action, so it is predictable that I am critical of Gelderloos’s arguments. But I also believe critical analysis is valuable. Nonviolent activists can become more effective by subjecting their beliefs to logical scrutiny and empirical testing.

The Case for Nonviolent Action

Through history classes, Hollywood movies and the daily news, most people come to believe two things about violence. One is that groups with a greater capacity for violence — armies, weapons, military industry and ruthlessness — can nearly always win over those with a lesser capacity. This is the assumption behind the question “What would you do to stop the Nazis?” asked rhetorically as a presumed refutation of nonviolence.[2] Second, most people believe that violence is a tool, usually a neutral tool. If it is used by bad guys — the enemy, terrorists or criminals — violence is bad, but when used by good guys — “our side” — then it is good. Most supporters of revolutionary warfare accept these assumptions: they believe revolution is a good cause and hope to use armed struggle to achieve it.

Nonviolent action challenges both these assumptions: the successes of nonviolent action challenge the belief that superior violence always succeeds; the characteristics of nonviolent action make it an especially appropriate tool for helping create a nonviolent society.

Mohandas Gandhi was the key figure in creating awareness of nonviolent action as a distinctive approach to social change, principally through campaigns in South Africa beginning in 1906 and then in India from 1915 through the 1940s. Nonviolent action had been used for centuries before Gandhi. For example, Hungarians who opposed domination by Austria used a range of methods of noncooperation from 1850 to 1867.[3] One of Gandhi’s achievements was to put nonviolent action on the agenda as a strategic approach.

Gandhi’s campaigns had an enormous influence worldwide, leading to the development and diffusion of nonviolent campaigning skills and insights. In many social movements, nonviolent action has become the preferred approach.

Nonviolent action, as a technique of political communication and waging conflict, can be distinguished from conventional action and from violence.

Conventional political action includes voting, lobbying and campaigning — anything that is routine within a society. Conventional economic action includes working, and buying and selling goods and shares. Conventional social action includes meetings of clubs or neighbours, charitable work and much else. Nonviolent action, in contrast, goes beyond routine behaviour, often by challenging conventional practices. Examples include protesters disrupting a government meeting by dressing as clowns, a neighbourhood association setting up an alternative system of social welfare, war resisters refusing to pay taxes, consumer activists blocking service in a bank by opening and closing small accounts, bus drivers refusing to collect fares, office workers sending large files to clog an email system, and communities setting up local currencies.

The boundary between conventional and nonviolent action depends on the circumstances. When government repression is severe, handing out a leaflet might count as nonviolent action, whereas in some places strikes are so common and widely accepted that participating in one might be considered conventional action.

Violence means physical force used against humans, including imprisonment, beatings, shootings, bombings and torture.[4] Nonviolent action excludes these. Sabotage — violence against objects — lies at the boundary between violence and nonviolence.[5]

Nonviolent action thus encompasses a wide range of activities that go beyond conventional, routine action but do not involve physical violence against humans. When people think about nonviolent protests, rallies and sit-ins commonly come to mind, but there are many other sorts, such as workers refusing to tear down an iconic building, judges resigning in protest over political pressure, roads activists digging up streets and planting crops, and office workers misplacing or destroying files on dissidents targeted for surveillance and arrest.

Nonviolent action is action — it doesn’t include passivity or inaction — and it goes beyond conventional methods of political communication and waging conflict, such as discussion, negotiation or lobbying. Nonviolent action is nonviolent on the part of those who use it. Their opponents can and often do use violence, sometimes brutally.

Nonviolent action can be divided into actions against something, such as most strikes and boycotts, and actions for something, such as workers organising to produce socially useful products or doing their jobs without bosses. The against actions typically target injustices; the for actions typically seek to build a better society.

With this picture of nonviolent action, what are the reasons for choosing it rather than conventional action or violence? There are two main traditions, commonly called principled and pragmatic nonviolence.[6] Principled nonviolence is undertaken for moral reasons, namely that it is wrong to use violence. This is the Gandhian tradition. Pragmatic nonviolence is undertaken because it is believed to be more effective than alternatives, in particular more effective than violence.

Principled nonviolent activists refuse to use violence under any circumstances. For example, they refuse to join armies, no matter how worthy the cause. However, many principled nonviolent activists pay close attention to effectiveness: they refuse to use violence, but they choose their tactics carefully.[7] Pragmatic nonviolent activists, on the other hand, often proclaim their commitment to nonviolence, knowing this increases their credibility. So, in practice, there is an overlap of principled and pragmatic rationales.

Here I concentrate on arguments for pragmatic nonviolence because they allow a more direct comparison with Gelderloos’ case. Principled nonviolence has its own arguments and criteria which are important but given less attention here.

Gene Sharp, the most prominent researcher of pragmatic nonviolence, divides the methods of nonviolent action into three types:

  • protest and persuasion, such as marches, taunting of officials, protest disrobings and renunciation of honours;

  • noncooperation, including many types of social, political and economic strikes and boycotts;

  • intervention, including fasts, sit-ins, occupations, land seizures and alternative institutions.[8]

From a pragmatic perspective, what are the reasons for choosing nonviolent action? In other words, what are the benefits of nonviolent action compared to alternatives? Out of many that could be listed, here are four.

1. Nonviolent action involves withdrawal of support from the system. It is a challenge to the legitimacy of standard behaviours or policies.[9] In contrast, conventional actions, such as voting, implicitly support the system by using its own methods.

2. Nonviolent action usually wins more support than does violence. Nonviolent actions — at least when well chosen — leave open a greater opportunity for communication. Opponents, by not being physically harmed, are accorded a certain respect: implicitly, their health and life are respected. Opponents are less fearful and hence do not have to be as ferocious in defence or attack.

When violence is used against nonviolent protesters, this is widely seen as unjust, and can lead to a major reaction against the violent attackers, a process Sharp calls political jiu-jitsu.[10] This can stimulate protest supporters to become more active, encourage uninvolved third parties to join the side of the protesters and even disgust some opponents of the protesters, causing their loyalty to shift.

3. Nonviolent action allows widespread participation. Women, children, elderly people and people with disabilities can participate in many forms of nonviolent action. This can be a goal in itself.

Much nonviolent action can be organised openly. This allows greater participation than clandestine operations.

Much nonviolent action is empowering for participants, promoting feelings of capability, solidarity and satisfaction.[11] Greater participation means greater empowerment.

4. Nonviolent action as a method is compatible with the goal of a nonviolent society.

Using methods that reflect the goal is called prefiguration: the methods prefigure — in other words, anticipate or replicate in advance — the goal. Prefiguration provides training and experience in what is being sought. It helps create an element of the goal, even if a campaign is unsuccessful. And it helps keep efforts on track.

If the goal is a society without organised violence, nonviolent action has all these prefigurative advantages. It provides experiences in living without using violence; it reduces immediate violence in the here and now, even when campaigns fail; and it ensures that efforts are in a nonviolent direction.

Each of these four points can be applied to violence.

1. Violence involves a withdrawal of consent from the system. In this regard, violence and nonviolence are similar.

2. Violence often alienates potential supporters. Opponents may dig in and resist more strenuously. A psychological perspective called correspondent inference theory helps explain why. People often infer someone else’s motivations by looking at the consequences of their actions. If the actions lead to people dying, the inference is that activists are motivated to kill — not to liberate, which might be their actual motivation. This theory helps explain why terrorists’ motives are so widely misinterpreted.[12]

Violence targets individuals, but harming individuals is not an effective way to challenge systems of oppression. Killing a politician does not undermine the state, because politicians can be replaced, sometimes with ones who are worse. Furthermore, a person who is a politician has other roles, such as parent, friend and musician. Violence, by not discriminating between roles, destroys much that is good, rather than targeting the damaging roles and building on the beneficial ones.[13]

When challengers use violence, this gives greater legitimacy to state violence against them. The jiu-jitsu effect is reduced, even when the state uses far more violence than challengers.

3. Violence restricts participation. Young fit men predominate in both armies and armed liberation movements. The secrecy accompanying armed struggle also limits participation.

Violence can be empowering for those involved, but limited participation means the empowerment is restricted.

4. Violence as a method clashes with the goal of a nonviolent society. Using violence gives training, experience and legitimacy to violence. It causes immediate suffering. And it is easier for campaigns to go off track, down a path towards ongoing violence and associated domination.

Points 2, 3 and 4 constitute the core of the case against violence, from a nonviolence point of view.

Note that these arguments for nonviolent action and against violence are tendencies, not universal truths. For example, nonviolent action usually wins more support than violence, but not always. Some nonviolent methods allow only limited participation whereas some violent movements have many participants. In adopting nonviolent action from a pragmatic point of view, attention needs to be given to the circumstances.

Many nonviolent campaigns are largely spontaneous, without much preparation, planning or training. No one expects armed movements without weapons, training or plans to be very successful. Considering the vast amounts of money and effort put into military operations, it is reasonable to expect that nonviolent action could become far more effective with more resources.

For those with a principled commitment to nonviolence, the circumstances do not matter: they reject violence, even if assassinating a dictator might reduce the suffering of millions. But there is an important link between pragmatic assessments and principled stands. If, pragmatically, nonviolent action is usually a better choice, then it can be (pragmatically) sensible to make a principled commitment, because it reduces the risks of misunderstanding by participants, of being falsely labelled violent by opponents, and of going off track in a violent direction.


Gelderloos is an anarchist. He opposes systems based on hierarchy and supports egalitarian social relationships created and maintained by the people involved in them. He is opposed to the state, capitalism, racism and patriarchy. Being opposed to capitalism puts him in the left generally, but as an anarchist he is opposed to the state, including state socialism whether advocated by reformist socialists seeking state power through electoral means or by Marxist-Leninists who want to seize control of the state, most commonly through armed struggle, in order to crush capitalism. Gelderloos wants instead to destroy the state — and capitalism, racism and patriarchy — so that people can create their own non-hierarchical systems of self-rule.

Gelderloos is an activist and has spent time in prison as a result of protest actions. His passionate commitment to liberation cannot be doubted. But while respecting his vision, dedication and energy, it is possible to criticise his arguments, conclusions and methods.

Rather than take up Gelderloos’ claims about nonviolence one by one, it is more illuminating to understand his perspective as stemming from a few key assumptions. At the core of his thinking is the view that nonviolence cannot be successful against violence.

The state, in the conventional sociological conception, is based on a monopoly over the use of legitimate violence. Gelderloos supports revolutionary change including overthrow and dissolution of the state. He believes that leaders of the state will not acquiesce. Hence, he concludes, physical force must be used.

If nonviolent action cannot succeed against violence, then Gelderloos’ other conclusions follow.

  • Nonviolence is ineffective against oppressive systems backed by violence (pp. 7–22).

  • Nonviolence is racist, because it allows racism to persist: “By preaching nonviolence, and abandoning to state repression those who do not listen obediently, white activists who think they are concerned about racism are actually enacting a paternalistic relationship and fulfilling the useful role of pacifying the oppressed.” (p. 34).

  • Nonviolence is statist, because it protects the state from the only challenge that Gelderloos believes can overthrow it, namely violence: “Put quite plainly, nonviolence ensures a state monopoly on violence.” (p. 45).

  • Nonviolence is patriarchal, because it prevents women from using a powerful tool — violence — against male domination: “a pacifist practice that forbids the use of any other tactics leaves no option for people who need to protect themselves from violence now.” (p. 67).

  • Nonviolence is “tactically and strategically inferior”: “The struggle against authority will be violent, because authority itself is violent and the inevitable repression is an escalation of that violence.... Lobbying for social change is a waste of scarce resources for radical movements.” (p. 94).

  • Nonviolence is deluded about how revolutionary change can occur: “it would be much easier to end the psychological patterns of violence and domination once we had destroyed the social institutions, political bodies, and economic structures specifically constituted to perpetuate coercive domination. But proponents of nonviolence boldly sound the call to retreat, declaring that we should treat the symptoms while the disease is free to spread itself, defend itself, and vote itself pay raises.” (p. 126).

Some of these arguments sound strange. How, for example, can nonviolence be patriarchal when women have been so prominent in nonviolent action whereas armed groups are almost always dominated by men? But Gelderloos’ argument has an underlying coherence built on his assumption that nonviolence cannot succeed against violence, which leads to his conclusion that violence is needed to overthrow oppressive systems, including patriarchy: “if a movement is not a threat, it cannot change a system based on centralized coercion and violence” (p. 22). Here, “threat” means the potential to use violence. Because nonviolence, in Gelderloos’ eyes, doesn’t pose a threat in this sense, he concludes that it is patriarchal.

Therefore, rather than address in detail Gelderloos’ claims about racism, patriarchy and the like, it is more useful to tackle his central claim that nonviolence is unable to be effective against violence, especially because this assumption is a common one. So how does Gelderloos support this claim?

He certainly doesn’t do it by addressing nonviolence theory: he does not systematically examine it.[14] Gelderloos treats all nonviolence as principled nonviolence, thereby missing pragmatic nonviolence. He mentions Gene Sharp only in passing and does not discuss Sharp’s theory of power or Sharp’s methods and dynamics of nonviolent action. He does not address the key dynamic of political jiu-jitsu, which explains how violence used against nonviolent protesters can be counterproductive.

Nor does Gelderloos examine George Lakey’s strategy for nonviolent revolution.[15] In fact, he assumes that nonviolence cannot be revolutionary, for example referring to “nonviolent and revolutionary activists” (p. 83).

Instead, Gelderloos assesses nonviolence by examining a number of nonviolent campaigns. He dismisses every one as not really constituting a success by using a series of arguments, deployed selectively, often with a double standard in relation to violence.

1. Gelderloos’ first argument against nonviolent campaigns is to say that they weren’t entirely nonviolent. If he can point to evidence of violence in campaigns, he dismisses the contribution of nonviolent action. Referring to 1962 black riots in Georgia and Alabama, Gelderloos concludes “Perhaps the largest of the limited, if not hollow, victories of the civil rights movement came when black people demonstrated they would not remain peaceful forever.” (p. 12)

There is a double standard here. In guerrilla struggles and other campaigns involving violence, there is also a great amount of nonviolent action, for example during the Vietnam war, the Iraq war and the second Palestinian intifada. Why should violence be given all the credit when both violence and nonviolence are used?

2. Gelderloos’ second argument against nonviolent campaigns is to say they didn’t really change anything. They weren’t liberation. They didn’t overthrow the state — just the current rulers — and didn’t overthrow capitalism. “The liberation movement in India failed. The British were not forced to quit India. Rather, they chose to transfer the territory from direct colonial rule to neocolonial rule.” (p. 9)

With this argument, Gelderloos again exhibits a double standard, because he doesn’t assess violent campaigns with the same stringent expectations. He refers approvingly to the Black Panthers in the US in the 1960s and 1970s and the anarchist revolutionaries in the Ukraine in the early 1920s, among others, none of which overthrew capitalism or the state. He lauds these initiatives for standing up to the state, for showing what can be accomplished, for striking fear into the heart of rulers and for empowering participants. That is all well and good, but he doesn’t give nonviolent campaigns credit for equivalent accomplishments.

Gelderloos doesn’t give a single example of an armed struggle leading to the sort of liberated society he espouses. Why not? Undoubtedly because successful armed struggles — such as in China, Cuba, Algeria and Vietnam — have not abolished the state but rather, if anything, strengthened it. Armed struggle encourages militarisation of the movement, making it more hierarchical and authoritarian. These features seldom wither away after revolutionary victories.

3. If a campaign fails, Gelderloos attributes this to the use of nonviolent action and insufficient use of violence. In the 1989 pro-democracy movement in China, “the students who had put themselves in control of the movement refused to arm themselves ...” (pp. 122–123). The double standard here is that Gelderloos does not mention the failure of armed movements in Bolivia, Latvia, Malaya, Philippines, Uruguay and many other countries.

Gelderloos ignores the difference between spontaneous and strategic nonviolent action.[16] Many failed campaigns have relied mainly on spontaneous nonviolent action, without careful planning and training. To dismiss these as failures of nonviolent action as a method would be like dismissing violence as a method because of the failure of spontaneous rioting.

4. In referring to recent campaigns that unseated governments in Serbia, Ukraine and other countries, Gelderloos says they were “orchestrated” by the US government (p. 100). He doesn’t give any evidence for this claim, aside from citing one newspaper story.

It is true that the US government has provided financial assistance to some nonviolent movements, for example Otpor in Serbia, a key resistance group in triggering the mass movement that brought down president Slobodan Milosevic in 2000. The contribution of US government assistance to these movements has been debated, not least among nonviolent activists, some of whom have argued against accepting assistance because of the risk of being accused of being pawns of the US government. Gelderloos addresses none of the complexities of these situations, simply assuming that because the US government was involved, therefore it orchestrated the whole operation.[17]

The double standard in this case is that Gelderloos does not make a similar claim in relation to violent struggles. During the Vietnam war, the National Liberation Front received considerable assistance from the Soviet government. Does this mean the NLF’s victory was orchestrated by the Soviet government? Of course not. The struggle’s success depended on the massive support and sacrifice of the Vietnamese people. Exactly the same can be said about nonviolent campaigns that receive US government funding: the campaigns would not stand a chance without popular support.

There is also a double standard in Gelderloos’ failure to mention cases in which the US government supported violent resistance. In Afghanistan after the 1979 Soviet invasion, the CIA covertly funded mujahideen opponents. In Kosovo, the US government ignored a decade-long nonviolent struggle[18] and then supported an armed movement, the Kosovo Liberation Army, previously classified as terrorists. By Gelderloos’ logic, these armed struggles were orchestrated by the US government and therefore the role of violence can be discounted.

5. Gelderloos ignores a great number of nonviolent campaigns, thereby avoiding the need to address their challenge to his argument. His claim about US government orchestration falls down entirely for nonviolent campaigns prior to the 1990s, before which there is no evidence of any US government assistance. In one of many examples, in 1944 the dictator of El Salvador — a US client state — was toppled in a popular nonviolent campaign.[19]

Gelderloos claims that the nonviolent strategy of generalised disobedience cannot bring power to the people because the state still controls key resources and the loyalty of the military and police: “in recent decades, the only significant military defections have occurred when the military faced violent resistance and the government seemed to be in its death throes.” (p. 99). To the contrary, there are quite a few cases in which military defections have occurred without there being much violent resistance, including the Philippines in 1986, various Eastern European countries in 1989, the Soviet Union in 1991, and Serbia in 2000. Gelderloos is right that military defections are essential for revolution,[20] but defections can occur as a result of nonviolent methods such as fraternisation.

Comparing Violence and Nonviolence

Gelderloos’ dismissal of the power of nonviolent action is so one-sided and filled with double standards that it would be easy to miss an important point: there has been very little systematic comparison of the effectiveness of violence and nonviolence.

Making comparisons is complex, because the success of a campaign or movement depends on many factors, including belief systems, human and material resources, social cohesion, political alignments and international factors, in addition to the methods used by activists. The choice of violent or nonviolent methods may tip the balance in some circumstances, but the other factors still need to be considered.[21]

In the vast body of research on social movements, there is little on the effectiveness of violence. In one of the few relevant studies, sociologist William Gamson in his book The Strategy of Social Protest analyses 53 US challenging groups between 1800 and 1945. In a chapter titled “The success of the unruly,” he assesses outcomes for groups that used violence compared to those that were recipients of violence without fighting back.[22] In this comparison, groups that used violence were far more likely to be successful, namely to gain acceptance and obtain new advantages. However, Gamson is reluctant to attribute success to violence, arguing instead that “it is not the weakness of the user but the weakness of the target that accounts for violence”: violence is “as much a symptom of success as a cause.”[23]

Gamson does not use the expression “nonviolent action” nor refer to any writings in the area. He does, though, analyse movements’ use of “constraints” including strikes, boycotts and denunciation. He finds that movements that used constraints but not violence were far more likely to be successful than ones that did not use constraints. In effect, he shows that coercive methods of nonviolent action — what Sharp would call methods of noncooperation and intervention — are associated with success.

Gamson’s study, while illuminating, does not directly compare the effectiveness of options for a given movement because, as he well recognises, he is analysing different movements in different circumstances. However, he does give strong backing for the conclusion that a movement being “unruly” is associated with success.[24] Violence and nonviolence are different ways of being unruly.

In 2005, Freedom House published a study of 67 political transitions occurring in countries with authoritarian governments in the period 1973–2000, looking at the level of violence, the source of violence and the forces driving the transitions. Using Freedom House’s pre-existing ratings of freedom in the countries before and after the transitions, the authors were able to assess the comparative roles of violence and nonviolence. Their principal findings:

  • “First, ‘people power’ movements matter, because nonviolent civic forces are a major source of pressure for decisive change in most transitions.”

  • “Second, there is comparatively little positive effect for freedom in ‘top-down’ transitions that were launched and led by elites.”

  • “Third, the presence of strong and cohesive nonviolent civic coalitions is the most important of the factors examined in contributing to freedom.”

  • “Fourth, the data suggests that the prospects for freedom are significantly enhanced when the opposition does not itself use violence.”[25]

These conclusions go directly against Gelderloos’ claims. Though he might dismiss them because of the politics of Freedom House or because none of the political transitions involved revolutionary overthrow of capitalism and the state, nonetheless they undermine his claims about the comparative ineffectiveness of nonviolent action.

In a forthcoming paper, Maria Stephan and Erica Chenoweth analyse data for 323 violent and nonviolent resistance campaigns from 1900 to 2006, concluding that nonviolent campaigns are more likely to achieve strategic objectives. They say, “Nonviolent campaigns are more likely to win legitimacy, attract widespread domestic and international support, neutralize the opponent’s security forces, and compel loyalty shifts amongst erstwhile opponent supporters than armed campaigns, which enjoin the active support of a relatively small number of people, offer the opponent a justification for violent counter-attacks, and are less likely to prompt loyalty shifts and defections.”[26] Additional careful studies of campaigns and outcomes are needed because it is so easy to reach a desired conclusion by the choice of a few selected examples.

Is Challenger Violence Justified?

Gelderloos frequently highlights the violence of the state — “the greatest purveyor of violence” (p. 158) — and other systems of oppression, with the implicit assumption that this justifies violence to destroy these systems. The issue of the legitimacy receives quite a lot of attention in discussions of violence. William T. Vollmann in his mammoth analysis of violence, Rising Up and Rising Down, focuses on justifications for and consequences of violence, and includes a detailed moral calculus.[27]

But just because violence might be justified does not mean it is the best option. If someone at a party swears at you, you might be legally and morally justified in suing for slander, but it is seldom a wise idea to do so: it is likely to be very costly and could harm your reputation even more.[28] Similarly for violence: although justified, it might be counterproductive in legitimising counter-violence, reducing participation, and leading down a path towards more violence. This highlights, once again, the importance of careful comparisons of the effectiveness of violence and nonviolence, taking into account both immediate outcomes and longer-term impacts on morale, solidarity and mobilisation.


Gelderloos’ arguments against nonviolent action in part miss the mark because he misconstrues nonviolence and makes claims without evidence. For example:

  • “... many pacifists in the US today also believe that if you are rocking the boat or causing conflict, you are doing something wrong” (p. 20). The only evidence given for this sweeping claim is one email. Note that Gelderloos’ persistent use of the term pacifist is misleading. Historically, many pacifists do not engage in nonviolent action and many nonviolent activists are not pacifists.

  • “But pacifists seem not to have given the violence of patriarchy its due consideration.” (p. 66). Actually, many nonviolent activists are feminists and have worked against male violence.[29]

  • “... nonviolence ... glorifies passivity” (p. 67). This is a common misconception.[30] The terms satyagraha and nonviolent action were developed to supersede the misleading expression “passive resistance.”

  • “Nonviolence focuses on changing hearts and minds, but it underestimates the culture industry and thought control by the media.” (p. 85). Persuasion is only one of the ways that nonviolence works. Sharp lists dozens of methods of noncooperation and intervention.[31]

  • Pacifists “have the option of forswearing confrontation with state power and pretending they are engaged in some process of magically transforming the state through the ‘power of love,’ or their ‘nonviolent witness,’ or by disseminating heartwrenching images of cardboard puppets through the media, or some other swill.” (p. 108). Actually, nonviolent activists have repeatedly confronted state power.[32]

  • “nonviolence is deluded in repeating that means determine ends...” (p. 130). The usual claim is that means influence ends.[33]

  • “The pacifist vision of struggle, based on a polar dichotomy between violence and nonviolence, is unrealistic and self-defeating.” (p. 139). Nonviolent activists are well aware of degrees of violence. It is possible to draw a line between violence and nonviolence in theory and practice while recognising different types of actions within each category.

There are many other examples of contentious statements by Gelderloos for which he gives no evidence or cites an email or personal comment but draws conclusions apparently intended to apply to all nonviolent activists. Gelderloos may be correct that some nonviolent activists are — perhaps unconsciously — racist, paternalist or too timid. But this does not mean that nonviolent action, as a method of struggle, has the same characteristics, any more than the racism or other features of violent activists mean that violence is racist.

Gelderloos often uses sources in a selective, misleading fashion. For example, he criticises an article by Carol Flinders about women and nonviolence, incorrectly portraying it as saying women are inherently nonviolent.[34] He cites Martha McCaughey’s book Real Knockouts, an analysis of the women’s self-defence movement, in support of his argument for women’s violence against patriarchy, missing the complexity and sophistication of McCaughey’s argument.[35]

Gelderloos approvingly quotes (pp. 114–115) Martin Oppenheimer’s book The Urban Guerrilla concerning shortcomings of nonviolence but omits any mention of Oppenheimer’s trenchant criticisms of violence, such as:

  • “For the organization of violence, for whatever reason, is basically subversive of democratic values and institutions, and the habit of solving political issues through violent means, far from liberating, imprisons persons and personalities so that truly democratic participation in decision-making become nearly impossible.”

  • “... the kind of people who become active in insurrections and survive it tend not to be the kind of people who will create a positive, humanistic order ... the kind of organization seemingly required to conduct a violent effort is inherently subversive of such an order.”

  • “... both in terms of personality and organization, violence, far from being therapeutic, endangers when it does not utterly destroy the humanistic component of a social movement.”[36]

Violence: What Sort and How Much?

How Nonviolence Protects the State is curiously coy about the actual role violence might play in liberation. Gelderloos explicitly rejects presenting a definition of violence: “... one of the critical arguments of this book is that violence cannot be clearly defined.” (p. 3). He says that many activists consider everyday activities, such as “buying clothes made in a sweatshop”, to be violent and says the concept of violence isn’t useful when “no two people can really agree on what it means” (pp. 124–125). However, even though activists may have different conceptions about terms, it’s still possible for analysts to agree on meanings.

Gelderloos, by leaving violence ill-defined, is able to avoid spelling out what he sees as the appropriate or inappropriate use of violence in liberation. He prefers to focus on hierarchy as the key to oppression and to say that all means of challenging hierarchy should be considered, without bothering about the difference between violence and nonviolence. He advocates a diversity of tactics, assuming that the more tactics are available to be used, the more effective a movement can be. Because a commitment to nonviolence means ruling out some tactics, Gelderloos concludes that nonviolence is bound to be less effective than a broader diversity of tactics.

There are a few clues in the text about what sorts of actions Gelderloos is thinking about:

  • “fighting cops or engaging in clandestine acts of sabotage” (p. 4);

  • “violent protests, bombings, and property destruction” (p. 15)

  • “hits a cop or throws a brick through a window” (p. 58)

  • “Killing a cop who rapes homeless transgender people and prostitutes, burning down the office of a magazine that consciously markets a beauty standard that leads to anorexia and bulimia, kidnapping the president of a company that conducts women-trafficking” (p. 67)

  • “blow up a dioxin-emitting factory that is making your breast milk toxic”; “kill the general who sends out the soldiers who rape women in a war zone” (p. 69)

  • “expropriate money to fund and greatly increase the capacities of grassroots media outlets” (p. 90).

The question arises: are there any methods that Gelderloos rejects? Does he reject use of machine guns? Does he reject missiles? Does he reject biological weapons? Does he reject nuclear weapons? Does he reject torture?[37] If Gelderloos rejects any of these methods, perhaps because they are inhumane or counterproductive, then he is drawing a line, accepting that not all methods are acceptable in a diversity of tactics.

One of Gelderloos’ chief complaints is that nonviolent activists are unwilling to support activists who use violence. Is Gelderloos willing to support any activist, even ones who use land mines and chemical weapons? If not, then his strictures against nonviolent activists, who draw a line at a different place, reflect a double standard in his argument.

Anarchism and Violence

Because Gelderloos articulately describes himself as an anarchist, some readers might gain the mistaken impression that he speaks for anarchists generally. Actually, anarchists have long debated and disagreed about the use of violence in bringing about social change. Some anarchists believe violence is warranted and necessary. The most famous armed struggle by anarchists was during the Spanish revolution and civil war from 1936–1939, when workers ran farms and factories and anarchist militias defended the revolution both from Franco’s fascists (backed by Hitler) and from communists. However, since then anarchist armed struggle has not played a prominent role, though a few anarchists have advocated using guerrilla methods designed for an industrialised country.[38]

There is also a parallel strand within anarchism that opposes use of violence.[39] Furthermore, some anarchists are openly committed to nonviolence.[40]

Pragmatically, the prospects for armed liberation within industrialised countries have been minimal for decades. In this context, many anarchists believe that using violence would be futile.[41] Prominent anarchist Murray Bookchin wrote,

The state power we face is too formidable, its armamentarium is too destructive, and, if its structure is still intact, its efficiency is too compelling to be removed by a contest in which weaponry is the determining factor. The system must fall, not fight; and it will fall only when its institutions have been so hollowed out by the new Enlightenment, and its power so undermined physically and morally, that an insurrectionary confrontation will be more symbolic than real.[42]

Another rationale arises out of the anarchist belief in prefiguration, namely that means should reflect ends. If the goal is a world without organised violence, this means avoiding violence in the struggle for such a world. However, anarchists have long debated the application of this principle to the use of violence.[43]

Marxists, in contrast, seldom subscribe to prefiguration. They instead assume that the ends justify the means, most notoriously in capturing state power to smash capitalism, after which the state is supposed to wither away: the ultimate Marxist goal is communism, in which there is no state, the same goal as anarchists. (This is the original meaning of communism, as distinct from the reality of states ruled by Communist parties.) Anarchists in the 1800s argued against Marxists, prophetically warning that capturing state power was a prescription for dictatorial rule.

Gandhi can be considered an anarchist.[44] He opposed the state, proposing instead village democracy. He refused an offer to lead newly independent India, unlike those anarchists who joined the government of Republican Spain in the 1930s.

The goal of anarchists is a society built around non-hierarchical structures organised by the people involved in them — an approach known as self-management — and hence anarchists strive to create non-hierarchical structures in their own organisations and campaigns. They have promoted cooperatives and workers’ control as alternatives to capitalist enterprise. They have promoted egalitarian relations between men and women.

Another point of debate within anarchism is the issue of revolution. Some, like Gelderloos, believe in destruction of capitalism and the state in a mass uprising to bring about self-managed systems. Others point to examples of self-management within today’s societies, believing that anarchist practices can grow up in the interstices of existing institutions, eventually supplanting them.[45] If social structures are likened to a forest, then revolution is like cutting down the trees and planting new species, whereas in an evolutionary model, new species grow up between the old, eventually becoming dominant.[46]

Gelderloos gives no hint of debates over violence and revolution within the anarchist movement: for him, anarchism seems to mean his own particular set of beliefs. Nor does he acknowledge the anarchist sensibility that is widespread in nonviolent movements.[47] Many activists are sceptical of the state and other dominant institutions and favour non-hierarchical forms of organisation, even though they are unfamiliar with anarchist writings. They also favour nonviolent action, which may explain why Gelderloos does not give credit to their anarchist sensibility.


Since the days of Gandhi’s campaigns, nonviolent activists have been criticised by supporters of armed struggle.[48] In recent years, the most comprehensive critiques of nonviolence have been by Howard Ryan,[49] Ward Churchill[50] and Gelderloos. Unfortunately, many critiques suffer through inadequate understanding of nonviolence, often due to a failure to engage with writings in the area.[51]

Gelderloos has shown enormous commitment as an activist and great energy in compiling a comprehensive critique of nonviolence. Unfortunately, he has missed his main target: in essence, he attacks principled nonviolence from a perspective in which the ends justify the means. He dismisses nonviolent action campaigns using a set of arguments that display systematic double standards.

Underlying Gelderloos’ argument is the assumption that violence is more effective than nonviolence. This is certainly a common assumption, but if a critique of nonviolence is to have any real teeth, the assumption needs to be justified and counterexamples addressed.

Gelderloos shows almost no awareness of the pragmatic tradition in nonviolent action. He misrepresents nonviolent action as consisting solely of protest and persuasion, missing the more coercive methods of noncooperation and intervention. Furthermore, he ignores a large number of major nonviolent struggles, successful and unsuccessful.

A key omission in Gelderloos’ argument is a discussion of limits in a diversity of tactics: he does not say whether any methods should be ruled out. Almost any activist will agree that some methods should not be used, whether it is assassination, land mines or biological weapons. The question then becomes where to draw the line.

The limitations of Gelderloos’ argument point to some ways for nonviolent activists to improve the presentation of their own views.

  • Challenge the assumption that violence is always more effective than nonviolence by using simple examples, for example of when violence is counterproductive for the state or protesters.

  • Expose double standards in discussions of violence and nonviolence, for example by pointing out failures of violence.

  • Become more familiar with case studies of nonviolent campaigns.

  • Be familiar with both the principled and pragmatic traditions in nonviolence, and the interaction between principles and effectiveness.


I thank Andy Chan, Howard Clark, Jørgen Johansen, Colin Salter, Maria Stephan, Ralph Summy, Tom Weber and Steve Wright for valuable comments.

[1] Peter Gelderloos, How Nonviolence Protects the State (Cambridge, MA: South End Press, 2007).

[2] For a discussion, see Michael C. Stratford, “Can nonviolent defence be effective if the opponent is ruthless?: the Nazi case,” Social Alternatives, Vol. 6, No. 2, April 1987, pp. 49–57; Brian Martin, “The Nazis and nonviolence,” Social Alternatives, Vol. 6, No. 3, August 1987, pp. 47–49; Jacques Semelin, Unarmed Against Hitler: Civilian Resistance in Europe 1939–1943 (Westport, CT: Praeger, 1993); Ralph Summy, “Nonviolence and the case of the extremely ruthless opponent,” Pacifica Review, Vol. 6, No. 1, May-June 1994, pp. 1–29.

[3] Tamás Csapody and Thomas Weber, “Hungarian nonviolent resistance against Austria and its place in the history of nonviolence,” Peace & Change, Vol. 32, No. 4, October 2007, pp. 499–519.

[4] The damage done to people through oppressive systems, such as exploitation, poverty and preventable disease, is commonly called structural violence.

[5] Some forms of sabotage, for example workers damaging equipment to interrupt production, such as in Nazi weapons factories, are commonly seen as nonviolent action. Others, such as blowing up large dams, are not, especially when the risk of hurting people is significant. Disagreements and disputes about sabotage are a recurring feature of discussions about nonviolent action.

[6] Judith Stiehm, “Nonviolence is two,” Sociological Inquiry, Vol. 38, Winter 1968, pp. 23–30.

[7] Gene Sharp, Gandhi as a Political Strategist (Boston: Porter Sargent, 1979).

[8] Gene Sharp, The Politics of Nonviolent Action (Boston: Porter Sargent, 1973), Part Two. On Sharp’s relation to Gandhi, see Thomas Weber, “Nonviolence is who? Gene Sharp and Gandhi,” Peace & Change, Vol. 28, April 2003, pp. 250–270.

[9] Sharp, Politics of Nonviolent Action, pp. 7–62; Gene Sharp, Social Power and Political Freedom (Boston: Porter Sargent, 1980), pp. 21–67, 309–378.

[10] Sharp, Politics of Nonviolent Action, pp. 657–703.

[11] Ibid., pp. 777–799.

[12] Max Abrahms, “Why terrorism does not work,” International Security, Vol. 31, No. 2, Fall 2006, pp. 42–78.

[13] I thank Jørgen Johansen for this point.

[14] Kurt Schock, “Nonviolent action and its misconceptions: insights for social scientists,” PS: Political Science and Politics, Vol. 36, No. 4, October 2003, pp. 705–712, lists 18 common misconceptions about nonviolent action. Gelderloos holds quite a few of them.

[15] George Lakey, Strategy for a Living Revolution (New York: Grossman, 1973). On nonviolent revolution see also Bart de Ligt, The Conquest of Violence: An Essay on War and Revolution (London: George Routledge & Sons, 1937); Dave Dellinger, Revolutionary Nonviolence: Essays (Indianapolis: Bobbs-Merrill, 1970); Narayan Desai, Towards a Non-violent Revolution (Rajghat, Varanasi: Sarva Seva Sangh Prakashan, 1972); Geoffrey Ostergaard, Nonviolent Revolution in India (New Delhi: Gandhi Peace Foundation, 1985). On anarchist views, see Andy Chan, “Violence, nonviolence, and the concept of revolution in anarchist thought,” Anarchist Studies, Vol. 12, No. 2, 2004, pp. 103–123.

[16] On the latter, see Peter Ackerman and Christopher Kruegler, Strategic Nonviolent Conflict: The Dynamics of People Power in the Twentieth Century (Westport, CT: Praeger, 1994); Robert L. Helvey, On Strategic Nonviolent Conflict: Thinking about the Fundamentals (Boston: Albert Einstein Institution, 2004).

[17] Valerie Bunce and Sharon Wolchik, “Bringing down dictators: American democracy promotion and electoral revolutions in postcommunist Eurasia,” Mario Einaudi Center for International Studies, Cornell University, Working Paper 5–07, July 2007, p. 15: “To reduce the electoral revolutions to the machinations of the U.S., however, is a serious mistake.” See also Andrew Wilson, Ukraine’s Orange Revolution (New Haven, CT: Yale University Press, 2005), pp. 183–189. I thank Howard Clark for recommending these references.

[18] Howard Clark, Civil Resistance in Kosovo (London: Pluto, 2000).

[19] Patricia Parkman, Nonviolent Insurrection in El Salvador: The Fall of Maximiliano Hernández Martínez (Tucson: University of Arizona Press, 1988). For other cases and analysis see Peter Ackerman and Jack DuVall, A Force More Powerful: A Century of Nonviolent Conflict (New York: St. Martin’s Press, 2000); Ralph E. Crow, Philip Grant, and Saad E. Ibrahim (eds.), Arab Nonviolent Political Struggle in the Middle East (Boulder: Lynne Rienner, 1990); Philip McManus and Gerald Schlabach (eds.), Relentless Persistence: Nonviolent Action in Latin America (Philadelphia: New Society Press, 1991); Patricia Parkman, Insurrectionary Civic Strikes in Latin America 1931–1961 (Cambridge, MA: Albert Einstein Institution, 1990); Kurt Schock, Unarmed Insurrections: People Power Movements in Nondemocracies (Minneapolis, MN: University of Minnesota Press, 2005); Stephen Zunes, “Unarmed insurrections against authoritarian governments in the Third World: a new kind of revolution,” Third World Quarterly, Vol. 15, No. 3, 1994, pp. 403–426; Stephen Zunes, Lester R. Kurtz, and Sarah Beth Asher (eds.), Nonviolent Social Movements: A Geographical Perspective (Oxford: Blackwell, 1999).

[20] Katherine Chorley, Armies and the Art of Revolution (London: Faber & Faber, 1943).

[21] I thank Howard Clark for emphasising this point.

[22] William A. Gamson, The Strategy of Social Protest (Homewood, IL: Dorsey Press, 1975), pp. 72–88. I thank Doug McAdam for recommending this reference.

[23] Ibid., p. 82.

[24] A similar argument is made by Frances Fox Piven and Richard A. Cloward, Poor People’s Movements: Why They Succeed, How They Fail (New York Vintage, 1979). See also Doug McAdam and Yang Su, “The war at home: antiwar protests and Congressional voting, 1965 to 1973,” American Sociological Review, Vol. 67, October 2002, pp. 696–721.

[25] Adrian Karatnycky and Peter Ackerman, How Freedom is Won: From Civic Resistance to Durable Democracy (New York: Freedom House, 2005), pp. 6–8.

[26] Maria J. Stephan and Erica Chenoweth, “Why civil resistance works: the strategic logic of nonviolent political conflict,” International Security, Vol. 33, No. 1, Summer 2008, pp. 7–44.

[27] William T. Vollman, Rising Up and Rising Down, 7 volumes (San Francisco: McSweeney’s Books, 2003). Vollman makes derogatory comments about Gandhi’s views. However, he neither addresses pragmatic nonviolence nor systematically compares nonviolence and violence.

[28] Truda Gray and Brian Martin, “Defamation and the art of backfire,” Deakin Law Review, Vol. 11, No. 2, 2006, pp. 115–136.

[29] Aruna Gnanadason, Musimbi Kanyoro and Lucia Ann McSpadden (eds.), Women, Violence and Nonviolent Change (Geneva: WCC Publications, 1996); Pam McAllister, The River of Courage: Generations of Women’s Resistance and Action (Philadelphia: New Society Press, 1991).

[30] This is the first misconception listed by Schock, “Nonviolent action and its misconceptions.”

[31] Sharp, Politics of Nonviolent Action, Part Two.

[32] See the sources cited earlier on toppling regimes.

[33] See the quote from Martin Oppenheimer in the text below.

[34] Carol Flinders, “Nonviolence: does gender matter?” PeacePower: Journal of Nonviolence & Conflict Resolution, Vol. 2, Issue 2, Summer 2006. Gelderloos (p. 160, note 31) says that Flinders praised the “innate pacifism of ‘devout Hindu wives’” whereas Flinders actually just referred to the behaviour of “devout Hindu wives” without praising it or attributing it to innate pacifism.

[35] Martha McCaughey, Real Knockouts: The Physical Feminism of Women’s Self-Defense (New York: New York University Press, 1997).

[36] Martin Oppenheimer, The Urban Guerrilla (Chicago: Quadrangle Books, 1969), pp. 50, 57 and 64. Emphasis in the original.

[37] Producing effective fighters requires special training to break down instinctive reluctance to kill: see Dave Grossman, On Killing: The Psychological Cost of Learning to Kill in War and Society (Boston: Little, Brown, 1995). Gelderloos does not say whether he endorses this sort of training for activists.

[38] International Revolutionary Solidarity Movement, First of May Group, Towards a Citizens’ Militia: Anarchist Alternatives to NATO and the Warsaw Pact (Over the Water, Sanday, Orkney: Cienfuegos Press, 1980).

[39] You Can’t Blow up a Social Relationship: The Anarchist Case against Terrorism (San Francisco: See Sharp Press, 1990).

[40] For example, Anarchists Against the Wall is “a direct action group” of Israeli activists that “works in cooperation with Palestinians in a joint non violent struggle against the occupation”: [][]]. I thank Maria Stephan for this example.

[41] Andy Chan, “Anarchists, violence and social change: perspectives from today’s grassroots,” Anarchist Studies, Vol. 3, No. 1, Spring 1995, pp. 45–68.

[42] Murray Bookchin, Toward an Ecological Society (Montreal: Black Rose Books, 1980), p. 260.

[43] Benjamin Franks, Rebel Alliances: The Means and Ends of Contemporary British Anarchisms (Edinburgh: AK Press, 2006), pp. 139–152; Vernon Richards (ed.), Violence & Anarchism: A Polemic (London: Freedom Press, 1993).

[44] Geoffrey Ostergaard and Melville Curle, The Gentle Anarchists: A Study of the Leaders of the Sarvodaya Movement for Non-violent Revolution in India (Oxford: Clarendon Press, 1971).

[45] Colin Ward, Anarchy in Action (London: George Allen and Unwin, 1973).

[46] I owe this metaphor to Val Plumwood and Richard Sylvan.

[47] Brian Martin, “Eliminating state crime by abolishing the state” in Jeffrey Ian Ross (ed.), Controlling State Crime: An Introduction (New York: Garland, 1995), pp. 389–417, at pp. 400–401.

[48] Gandhi’s approach has been criticised from a variety of perspectives: Thomas Weber, “Gandhian nonviolence and its critics,” Gandhi Marg, Vol. 28, No. 3, October-December 2006, pp. 269–283.

[49] Howard Ryan’s book Critique of Nonviolent Politics: From Mahatma Gandhi to the Anti-Nuclear Movement — a sympathetic critique that engages with nonviolence writings in a well-informed way — is no longer available online. For a detailed review see Brian Martin, “Critique of violent rationales,” Pacifica Review, Vol. 9, No. 1, 1997, pp. 83–91. [Update, 2010: Howard Ryan’s book is now available. The review gives links.]

[50] Ward Churchill, Pacifism as Pathology: Reflections on the Role of Armed Struggle in North America (Edinburgh: AK Press, 2007). Churchill’s analysis has many of the same assumptions, omissions and double standards as Gelderloos’.

[51] William Meyers’ Nonviolence and Its Violent Consequences (Gualala, CA: III Publishing, 2000) shows little awareness of writings on nonviolence. Steven Best and Anthony J. Nocella II (eds.), Terrorists or Freedom Fighters: Reflections on the Liberation of Animals (New York: Lantern Books, 2004) has many contributions discussing nonviolence but almost no recognition of pragmatic nonviolence.