Researching nonviolent action: past themes and future possibilities

Published in Peace & Change, Vol. 30, No. 2, April 2005, pp. 247-270. The published version includes slight sub-editorial changes from the version here.
pdf of published article

Brian Martin

Go to

Brian Martin's publications on nonviolence

Brian Martin's publications

Brian Martin's website

Research into nonviolent action has developed in the past century from nothing to a small but thriving enterprise. Past nonviolence research can be usefully assessed in terms of its relevance to three principal audiences: nonviolence researchers, activists and policy makers. Nonviolence research has often served to inspire and inform researchers and activists but rarely has had an impact on policy. Future nonviolence research can do more of what has gone before, especially case studies and theory. As well, there is a potential to open up new areas of investigation including the role of technology, the absence as well as the presence of action, and new action arenas including cyberspace and organisational struggles, as well as to pioneer participatory methods of research. The future development of nonviolence research is likely to be symbiotic with the growing use of nonviolent action.

Part 1: Past themes

People have been engaging in nonviolent action for centuries, and today there are millions participating in rallies, boycotts, strikes and a host of other actions. Some participants think long and hard about how to make their actions more effective; some try to develop sets of ideas for understanding these actions, while others gather evidence. Somewhere along the line it's reasonable to say that this thinking and evidence-gathering should be called research. Although researchers are often thought of as separate observers, participants can be researchers too, and it may be wise not to draw too strong a distinction between research and other activities.

Whatever definition is used, it is fair to say that research into nonviolent action made great strides during the 1900s, starting from nothing and growing to encompass a small but dedicated network of researchers who collectively have made many valuable contributions. How can this research be assessed? This question immediately raises the issue of the audience of research. Three obvious audiences are scholars, nonviolent activists and policy makers.

The advantage of orienting research to other scholars - both nonviolence researchers and others - is that quality is maintained by the familiar processes of peer review. Intellectual advance can occur by building on others' findings. But there is a major risk in too strong a scholarly orientation: the research enterprise can become cut off from practice, being uninteresting and unintelligible to activists and policy makers. This is a familiar scenario in the social sciences. A large amount of research in disciplines such as sociology and political science, and even in interdisciplinary fields such as environmental or women's studies, is of no interest to anyone but scholars.

An alternative is to orient nonviolence research to activists, with the great advantage of connecting theory and practice, providing a regular testing ground for theory and offering theoretical insights to activists. Nonviolence research can be relevant both to activists committed to nonviolence on principle and to activists using nonviolent methods for pragmatic reasons. The risk in this orientation is that research is tailored to what activists like to hear, with potentially uncomfortable directions not pursued and uncomfortable findings submerged. Another risk is that research findings published in magazines for activists may not get to a very large audience.

A third option is for nonviolence research to be oriented to policy makers, especially governments. This has the potential for gaining significant institutional support for nonviolence, including funding for research. One risk in this orientation is that nonviolence may be coopted by policy makers, being given nominal support without substantive change in policies. Another risk is that nonviolence research is steered away from directions useful to activists.

Each of these three orientations for nonviolence research thus has advantages and disadvantages. There is no neutral point for judging these orientations, since they depend on value judgements about the best way to achieve a better world and better knowledge about it.

My plan here is to look at the work of small number of nonviolence researchers with an eye to their key contributions, taking note of the audiences for their work. My selection is not intended to suggest that these are the most important researchers. That would be impossible given that English is my only language, and in any case is not a useful exercise. Rather, I've picked works with which I have some familiarity, with the aim of highlighting some important themes.

My friend Tom Weber, a leading Gandhian scholar, pointed out that I have focussed on the collective end of the nonviolence spectrum - organised group actions, including nonviolence defence and social transformation - and neglected the individual end that includes interpersonal conflict resolution and spiritual development of the individual. This latter end of the spectrum is important - I see it going hand-in-hand with collective action - but I only mention it in passing here.[1]

My discussion here focuses on the realm of ideas. I do not address funding for research, professional associations, teaching, career opportunities and the like. These matters are vitally important but must be left for another occasion.

Mohandas K. Gandhi was not a researcher in the style of today's scholars and he never sought to publish his work in scholarly journals. Nevertheless, it is worth looking at Gandhi's contributions through the lens of research.

Gandhi made several lasting contributions to nonviolence research. His personal example and his leadership of the Indian independence movement provided inspiration for activists and intellectuals alike, a process of inspiration that continues today. Also vitally important was Gandhi's conceptualisation of nonviolent action and its deployment as a planned method for social change. People had used methods of nonviolent action long before Gandhi. He, more than anyone else, combined theory and practice. Gandhi was committed to nonviolence on ethical grounds, an approach now commonly called principled nonviolence, though he had a astute eye for what would work in practice.

Gandhi subtitled his autobiography "The story of my experiments with truth,"[2] and there is definitely a sense in which he was an experimentalist. Gandhi produced a vast amount of writing, but did not systematise his ideas. Therefore, it seems reasonable to say that Gandhi's work was primarily oriented to activists. He was not interested in academic research.

Gandhi declined the opportunity to be the political leader of newly independent India, but he groomed his key followers to make government policy along nonviolence lines. Aside from President Rajendra Prasad, they dropped any commitment to pursuing Gandhian policies as soon as they were in power.[3] It could be said that Gandhi had hopes of influencing policy but had not developed tools for doing so.

Richard Gregg was one of Gandhi's many followers and admirers. Spending some time working with Gandhi, he took on the task of expounding Gandhi's ideas and practice for other audiences. Gregg's book The Power of Nonviolence, first published in 1934, is an impressive exposition and interpretation of Gandhi's methods, aimed at Western audiences, augmented by Gregg's own insights.[4] He covers examples of nonviolent action, the effectiveness of mass nonviolent action, nonviolence as a substitute for war, nonviolence and the state, and nonviolence training.

For a closer look, let me focus on just one element in Gregg's book: the concept of moral jiu-jitsu. When a nonviolent activist - a satyagrahi in Gandhi's terms - comes under attack, for example being beaten by police, and maintains nonviolent discipline, this can generate support for the nonviolent activist. According to Gregg, nonviolence shows respect for the opponent's integrity, thereby putting the attacker to shame and impressing onlookers, who may be won over. By analogy to the sport of jiu-jitsu, in which the opponent's strength is used to destabilise them, Gregg says that nonviolence causes the attacker to lose moral balance while the defender maintains it, thus producing what he terms moral jiu-jitsu.

Gregg's analysis is along the lines of Gandhi's: both of them see nonviolence working through psychological processes. Gregg added a coherent explanation, references to the psychological literature and a descriptive name. What is lacking is any evidence of the psychological effects of nonviolent action. Years later, Thomas Weber re-examined the jiu-jitsu process in Gandhi's 1930 salt march.[5] When satyagrahis came forward to be beaten by lathi-wielding police, this apparently did not lead to a psychological transformation in the police. Some of them became angry at the lack of resistance by the protesters, redoubling their attacks. Weber says that the attacks triggered outrage by third parties around the world who were informed through eloquent news reports by journalist Webb Miller. Contrary to Gregg's idea that nonviolence works primarily through direct psychological effects on attackers, Weber found that the main effect, in this instance at least, was through influence on third parties.[6] Dennis Dalton, another Gandhian scholar, documented the effects of Gandhi's campaign on British officials.[7] According to Dalton, the salt march brilliantly exploited Brish ambivalence by appealing to individuals' higher sentiments (such as Lord Irwin's religious beliefs), by presenting the cause sympathetically to moderates and by putting the British in a lose-lose situation: either toleration, with the independence movement making advances, or repression leading to a backlash. These more recent works show both the limitations of Gregg's analysis and the value of his framework as a foundation for further investigation.

Gregg's book is relevant to both activists and scholars. For activists, it presents Gandhi's method in a more systematic form than Gandhi ever did. For scholars, it lays out a conceptual framework for nonviolent action. Being firmly in the Gandhian tradition of principled nonviolence, Gregg's approach has little appeal for policy makers.

Stephen King-Hall was a British naval officer, intelligence officer, playwright, member of parliament and iconoclastic commentator on British foreign policy. Just before the outbreak of World War II, he advocated sending messages directly to the German people in an attempt to undermine support for Hitler. Gaining no support from the British government, he financed a private effort along these lines which, at least according to King-Hall, had considerable impact.[8]

In 1958, King-Hall's book Defence in the Nuclear Age was published.[9] He argued that the rise of nuclear weapons had made conventional military defence obsolete and instead argued for British nuclear disarmament - unilateral if necessary - and nonviolent resistance should the country be occupied. Unlike pioneering peace researchers arguing for nonviolent defence[10] - all of whom explicitly derived ideas from the nonviolence tradition - King-Hall was primarily concerned about the communist threat and derived his radical conclusions from a pragmatic analysis of the implications of nuclear weapons. He was quite uncritical of the "western way of life" which he saw as the central thing to be defended. One of the core elements of this way of life, as he saw it, was British parliamentary democracy.

King-Hall's unconventional approach to nonviolent defence led him to make some innovative proposals. He suggested, for example, that Khrushchev could have been invited to appear on television with the British prime minister, who would then invite 100,000 Soviet citizens to live with British families in their homes for two weeks, seeing the western way of life, at British expense. If Khrushchev had refused this generous public offer, then this rejection could have been publicised, especially in the Soviet Union. If he had accepted, then the visitors' commitment to communism would have been undermined, or so King-Hall assumed.[11] This is a type of international diplomacy with fraternisation as the underlying method.

King-Hall's book was aimed squarely at policy makers. Though generating some discussion, it essentially represented a cry in the wilderness given the low receptivity of western governments to a recommendation to renounce nuclear weapons and rely on nonviolent defence. King-Hall did not orient his recommendations to activists, and academics were even further from his concerns, so it is perhaps not surprising that his book did not become the foundation for future research. It did show, though, that nonviolence need not be the province only of activists: as nonviolent defence, it could be treated as a serious policy option.

Gene Sharp is widely recognised as the world's leading nonviolence researcher. Among his many contributions are the documentation and classification of hundreds of different methods of nonviolent action. He also spelled out a framework for the dynamics of nonviolent action, with a series of typical stages: laying the groundwork for nonviolent action; making challenges, which usually brings on repression; maintaining solidarity and discipline to oppose repression; political jiu-jitsu; achieving success through conversion, accommodation or nonviolent coercion; and redistributing power.

Whereas Gandhi was unsystematic in his observations and analysis, Sharp is relentlessly thorough, most distinctively so in his epic book The Politics of Nonviolent Action.[12] Sharp's classification of methods of nonviolent action uses three main categories - symbolic action, noncooperation, and intervention - various subcategories, such as strikes and boycotts as types of noncooperation, and then numerous specific methods such as farm workers' strikes and traders' boycotts. For each method, Sharp provides brief historical examples plus references.

Another important contribution by Sharp is his use of the consent theory of power as the theoretical foundation for nonviolent action. In essence, this theory proposes that the key to the power of rulers is the consent or acquiescence of their subjects: if that consent or acquiescence is withdrawn, the ruler's power dissolves. Although others had argued along the same general lines previously,[13] Sharp gave it a more prominent and practical role.

Though significantly influenced by Gandhi, Sharp broke with him in a major way. Whereas Gandhi advocated nonviolence as a moral imperative, Sharp advocates nonviolence as more effective than violence. These positions are commonly called principled nonviolence and pragmatic nonviolence respectively.[14] Sharp's pragmatic approach can be considered an adaptation to western culture, where the basis for widespread principled commitment to nonviolence seems to be lacking.

Sharp has oriented some of his work squarely at policy makers, especially his writings on nonviolent defence or, to use his preferred expression, civilian-based defence.[15] His argument that nonviolence is pragmatically superior can be interpreted as an appeal to policy makers - especially in the US, his primary target audience - to rationally consider options and to choose the more effective option. Though there has been some polite attention and the occasional enthusiastic support, by and large Sharp's ideas seem to have had minimal impact on US defence policy. One explanation for this is that for US policy elites, maintaining the military and state apparatus is a higher priority than, or a precondition for, considering effectiveness.

Sharp has also oriented his writings to scholars. This is apparent, for example, in the care he takes in documenting historical sources and in listing prior theoretical work relevant to the consent theory of power. Although Sharp's works are widely known to and regularly cited by nonviolence scholars, he has been virtually ignored in mainstream disciplines such as political science. It is still possible to read accounts of nonviolent struggles without so much as a mention of Sharp or any other nonviolence scholar.[16] Part of the reason may be that Sharp did not often publish in mainstream disciplinary journals outside peace research. More fundamentally, his approach is at odds with dominant frameworks in academic social science.

Sharp has personally inspired and advised nonviolent activists in numerous countries around the world, but his substantial writings are not explicitly oriented to activists.[17] Especially in his policy-relevant work on civilian-based defence, he does not want this alternative to be seen as the special agenda of social movements, such as socialism and feminism, since this might taint it in the eyes of policy makers.[18] Ironically, though, it is among activists that Sharp has had by far his greatest impact; years ago I suggested that Sharp has had more influence on social activists than any other living theorist.[19] This is due to the very practical nature of his classificatory framework, its inspirational value deriving from so many historical examples, and the consent theory of power that offers activists an explicit warrant for bringing about change.

Alex P. Schmid in 1985 produced a comprehensive analysis of Social Defence and Soviet Military Power, in which he concluded that "the Soviet military power instrument cannot be balanced by economic noncooperation and cultural persuasion alone as the USSR is economically invulnerable and culturally impenetrable."[20] In other words, social defence - another term for nonviolent defence or civilian-based defence - would not work against a Soviet military invasion. The research was funded by the Dutch government, the only project that proceeded out of a whole series of projects originally planned. Social Defence and Soviet Military Power contains a short survey of concepts of nonviolent action and social defence, an examination of Soviet military interventions and nuclear threats since 1945, four case studies of East European resistance to Soviet domination, and an assessment of social defence as a component of a national defence system.

Schmid's analysis is important as one of the few sustained investigations undertaken without an obvious sympathy for nonviolence, thereby allowing articulation of insights that would be uncomfortable for scholars such as Gregg or Sharp. One important point made by Schmid is that in many struggles, the outcome depends primarily on the international configuration of power, with the method used - violent or nonviolent - being of lesser significance. The Lithuanian partisan resistance against Soviet re-occupation, from 1944 to about 1952, is a case in point: without Western intervention, the partisans had little hope of success.

This cool-headed examination stands in contrast to the more familiar approach by nonviolence scholars of highlighting success stories - such as toppling the 1920 Kapp Putsch in Germany, resistance to the Nazi occupation in Norway and the Netherlands, the US civil rights movement and the Czechoslovak resistance to the 1968 Soviet invasion - and ignoring or downplaying both the problematical features of these examples as well as other examples where nonviolence had only limited success.

In order to develop the case for social defence, it can help to scrutinise the arguments of critics, so Social Defence and Soviet Military Power has a special value given that few careful critiques exist. Four assumptions made by Schmid are open to question: that social defence is necessarily national defence (rather than both a local and a transnational process), that social defence has no offensive capacity, that social defence must substitute for all the strengths of military defence, and that social defence would be introduced without any other significant changes in society. Given the collapse of the Soviet Union and its satellite regimes, it is easy now to dismiss Schmid's assessment,   but it would be to throw away all potential insights just because some assessments were wrong.

Schmid's treatment was aimed at policy makers, in particular in the Netherlands, though it may be that the book's negative assessment of social defence was redundant and that even a glowing endorsement would not have led to serious government attention. The book is scholarly in construction but seems to have received little attention from nonviolence or other scholars. Finally, the book is definitely not aimed at activists and they have shown no interest in it.

Robert Burrowes would be considered by many to have been Australia's leading nonviolent activist in the 1990s. He was an inspiring presence in the Melbourne-based Rainforest Action Group, a shrewd strategist and a key networker at a national level. He took a highly principled position and promoted it astutely. For example, when refusing to pay the portion of his income tax that would go to the military, he instead delivered shovels - symbols of constructive work - to the taxation office in a symbolically potent act of resistance.

Burrowes read widely in the nonviolence literature and undertook a PhD. His thesis, in revised form, was published in 1996 as The Strategy of Nonviolent Defense: A Gandhian Approach.[21] The title indicates his distinctive contribution: a Gandhian approach that deals with strategy, in particular for defence. Among the innovations in the book is a reassessment of the political purpose and strategic aims of nonviolent defence. Richard Gregg may have been the first to apply Clausewitz's ideas about strategy to a nonviolent struggle; this was done much more thoroughly by Anders Boserup and Andrew Mack in their book War Without Weapons. Boserup and Mack concluded that the strategic aim is maintaining the unity of the resistance. Later, Gene Keyes decided instead that the strategic aim should be maintaining the morale of the resistance.[22] For Burrowes, though, the strategic aim of the resistance is "to consolidate the power and will of the defending population to resist the aggression" and the strategic aim of the counteroffensive is "to alter the will of the opponent elite to conduct the aggression, and to undermine their power to do so."[23] This combination of will and power might be taken to reflect a combination of Gandhi's emphasis on moral persuasion and Sharp's emphasis on nonviolent coercion.

Although Burrowes talks about nonviolent defence, his perspective is more akin to nonviolent revolution. He strongly criticises non-Gandhian approaches, for example criticising Sharp's model of civilian-based defence as not focussing on satisfying human needs, as relying on a conception of society oriented to elites, and as being based on a faulty strategic theory.

The final chapters of The Strategy of Nonviolent Defense cover planning and organising nonviolent defence (including strategic assessment, organisation, leadership, internal communications, advance preparations, constructive programme and evaluation) and the strategy and tactics of both the defence and the counteroffensive. These chapters are rich with practical insight drawn from Burrowes' activist experience combined with his vision of strategy.

Burrowes wanted his book to reach activists. For example, he arranged for information about it to be sent to dozens of activist groups around the world, but few of them responded. Part of the problem may be the packaging for his ideas, namely a thesis with many preliminary chapters on various conceptions of military strategy, society and power. The book has been taken up more by scholars, aided by its publication by a major university press, though it remains to be seen whether the book's contents will receive serious scholarly attention or whether the book will be simply cited as a work on nonviolence. The Strategy of Nonviolent Defense is definitely not aimed at policy makers and they have taken no apparent notice of it.


There is the vast amount of nonviolence research that I have not even mentioned, including works by famous names in the field. My choice of six individuals omits any women[24]; it omits researchers who have not written extensively in English[25]; and it reflects a bias toward a western style of research. In exculpation I can only comment that this brief exposition is not intended as a review of the literature and even less as a ranking of contributors, but rather as a means of highlighting a few points. The works of these six individuals point to many of the key areas addressed by twentieth-century nonviolence research:

The linking of names to themes is somewhat arbitrary. With a comprehensive survey of the field, many more names could be added to each theme. The key thing here is that the themes represent crucial developments in nonviolence research, even though there is disagreement about the significance or value of some of them. Let me now turn to a few aspects of nonviolence research not yet addressed.

Nonviolence research often serves to inspire others. Many activists have been inspired by writings by Gandhi and Sharp, for example. Though there are other ways to inspire people besides research writings - for example television documentaries such as A Force More Powerful, for which Ackerman and DuVall's book of the same name serves as a valuable aid[26]  - some nonviolence research can inspire activists. It can also inspire current and potential researchers. Indeed, without awareness of other work, researching nonviolence would be a lonely occupation. There is also the possibility of inspiring policy makers, though so far this seems to have been limited in effect.

Another value of nonviolence research is to provide useful ideas or frameworks. If activists can obtain ideas for action or a way of better understanding what they are doing and trying to achieve, that is of tremendous value, potentially making the difference between successful and unsuccessful campaigns. For scholars, ideas and frameworks are grist for the mill of future research. For policy makers, ideas and frameworks are potential tools for introducing and implementing policy.

So far, the success of nonviolence researchers in providing useful ideas and frameworks has been mixed. Few activists have the time or inclination to study scholarly work, with many of them relying on popularisations or on a few individuals to interpret research ideas. (Robert Burrowes was one of those who effectively bridged the research and activist cultures.) Part of the problem is that few scholarly works are written in a form oriented to activists, given that scholarly credibility is often reduced by adoption of more accessible writing styles. Researchers need to learn how to couch their results so that they can be effectively taken up by activists. Within scholarly forums, there is little discussion of how this can best happen.

But there is a deeper problem too: the capacity of the nonviolence movement to learn. If research confirms standard activist ideas about nonviolence, then it can serve as validation and inspiration. Arguably, though, a more important function of research is to come up with results that are not obvious.[27] Researchers who reach conclusions that are unwelcome to activists - perhaps because they suggest that certain familiar techniques are not effective or because they suggest activities that would shift the balance of influence within activist groups - are less likely to receive encouragement.

For researchers communicating to other researchers, there is plenty of inspiration and plenty of ideas. So far, there is nothing close to agreement about the priorities for the field or key issues to be resolved. There is actually a vast scope for nonviolence research but relatively few individuals who pursue it. No doubt more could be done to link nonviolence researchers with each other, but the wider obstacle is the credibility of nonviolence in scholarly arenas and the wider society. Until that changes, it is seldom a good career move to study nonviolence. This has the advantage of discouraging careerists but it also limits the opportunities for intellectual advance and greater credibility.

The linkage of nonviolence research to policy makers is weak at best and often non-existent. Some activists and scholars would say this is a good thing, given the risk that nonviolence could be coopted by the state, having its radical potential defanged. Another perspective, though, is that policy makers need to be at least aware of thinking about nonviolence, even if they do not actively support it, so that they can respond more appropriately when nonviolent struggles occur. The failure of governments to support the decade-long nonviolent struggle in Kosovo reflects a lack of understanding of the potential of nonviolence, if not hostility to it.[28]

Finally, it is worth mentioning that nonviolence research does not always have the effects intended by its authors. Sharp aimed his studies more at policy makers and scholars but has had his greatest effect on activists, whereas Burrowes was more interested in activism but his book is more noticed by scholars.

What I have not indicated here is the complexity and contingency of interactions between activism, scholarship and policy, not to mention other arenas such as media and personal behaviour. There are many personal as well as formal networks in the nonviolence field, now facilitated by the internet. Contingencies, such as someone happening to talk to a particular individual or read a particular article, can have a major impact on actions and research directions. This should not be a cause for concern. If the nonviolence field were too well organised, it would also be more at risk of manipulation and subversion. That may be a risk in the future, but for now the greater challenge is to encourage greater participation in both research and action.

Part 2: Future possibilities

There are many possible directions for future nonviolence research and many possible ways of discussing them. Rather than predict what will happen or recommend what should happen, I'll outline a range of things that could happen, though this inevitably will overlap with what I think will or should happen.

The most obvious path for future research is more of what has gone before. This means more case studies of nonviolent struggles. Especially since the 1990s there have been quite a number of excellent collections of case studies covering different geographical regions and cultures,[29] but there are many more cases to be studied. Observing the vast quantity of writing on military history available in large bookshops, it's apparent that the historical study of nonviolence has barely begun.

As well as doing additional case studies, more can be done with the case studies that are produced: more on the political, economic and social context; more on psychological dimensions; more on nonviolence strategy; more on strengths and weaknesses within the nonviolence camp; more on long-term consequences.

In parallel with more case studies, another obvious direction is more research into nonviolence theory. Given the array of theory already available - without widespread agreement - there is plenty to work with. For example, Sharp's consent theory of power has been subject to critiques[30] but no one has come up with an alternative theory of power that is more productive for activists. Similarly, Sharp's model of the dynamics of nonviolent action deserves elaboration and testing.[31] Another area where theory lacks resolution concerns application of Clausewitz's strategic theory to nonviolent defence. Yet another area worth exploring is Ackerman and Kruegler's principles of strategic nonviolent conflict, which essentially are recommendations for what is required for effective use of nonviolent action.[32] These are only a few samples of theoretical areas deserving further investigation. There are many others worthy of attention, such as premier peace researcher Johan Galtung's vast array of provocative theoretical explorations.[33]

Another avenue is to take nonviolence research into disciplinary fields. This has the virtue of bringing the ideas to new audiences and of reinvigorating nonviolence theory itself through engagement with different schools of thought. Exemplary efforts along these lines are Roland Bleiker's book Popular Dissent, Human Agency and Global Politics, which engages international relations scholars, and Glenn Paige's book Nonkilling Global Political Science.[34]

For theoretical work, one road is development of new theory; another is testing and revision of previous theory. For both roads, case studies are immensely useful for stimulation and testing of ideas. One risk in developing theory is to become too detached from practice. Regular revisiting of case studies helps keep the theory "relevant." On the other hand, if theory is too tied to case materials, then it is more difficult to see the big picture and grasp general principles.

Theory can be a useful guide when doing case studies. Rather than simply describing an event involving nonviolent action - which itself depends on implicit assumptions about what is worth describing - deployment of a theory helps to highlight particular features of the event. For example, the first of Ackerman and Kruegler's principles of strategic nonviolent conflict, "formulate functional objectives," draws attention to a particular aspect of the conflict, and similarly for each of the remaining 11 principles. Theory thus can provide a window through which to perceive particular features of a conflict.

As well as doing more of the same sort of nonviolence research, another option is to tackle new arenas. Many people have an implicit image of nonviolent action in their minds that shapes the way they think about all nonviolent action. What is the implicit image of nonviolent action? No doubt this needs investigation (more research!). My guess is that for many people it is the rally.[35] Antiglobalisation protests testify to the continuing salience of the rally. The key elements of the standard image of protest are people taking action by putting their bodies on the line in a public space. This image does not capture all methods of nonviolent action, as a brief examination of Sharp's catalogue reveals. For example, in boycotts, no bodies need be in public spaces.

It is worth looking closely at each of the elements of the usual image:

The first element is that nonviolent action is undertaken by people. What is the alternative? Non-human animals perhaps? That's a possibility, but there is more scope in looking at technology. In the exercise of violence, technology has been an almost inevitable partner, with weapons including everything from machetes and rifles to aircraft carriers and nuclear explosives. Some weapons are becoming semi-autonomous, such as guided missiles and pilotless aircraft. The heavy reliance on technology by the military and police contrasts with a relative neglect of technology by nonviolent activists. My own assessment is that technology can be an important support for nonviolent struggle, most obviously in the case of network communication technologies such as the telephone and internet but also in the case of resilient systems for transport, energy, agriculture, health, water supply and housing, making it far easier for a population to survive and resist aggression, a military coup or terrorist attack.[36] That the military spends tens of billions of dollars each year on researching and developing new weapons suggests the potential scale of technology-oriented research and development for nonviolent struggle.

The second element of the standard image of nonviolent action is that action is involved. After all, "action" is part of the standard expression "nonviolent action." Researchers have been attracted by all sorts of actions but especially by the largest and most successful ones. If hundreds of thousands of people protest in the streets, helping to topple a repressive regime - as in the Philippines in 1986 or in East Germany in 1989 - this is seen as an example of nonviolent action that is especially worthy of investigation. That is reasonable enough. But what is neglected is the converse situation: when nonviolent action is warranted but little or none occurs, for example in the Philippines in 1985 or East Germany in 1988, or any number of other situations where repression, war or genocide occur but there is little or no resistance. This includes lack of resistance by vulnerable people plus lack of resistance by people in the rest of the world. The point of this is not to lay blame but rather to point to areas worthy of research: situations where nonviolent action is warranted and needed but little or none occurs.

Expanding nonviolence research to look at "non-action" opens up a vast arena of human behaviour for investigation. Consider two contrasting cases in Indonesia: the popular protests in 1998 that led to the resignation of dictatorial ruler Suharto, in a scenario fitting the standard image of nonviolent action; and the 1965-1966 massacre of hundreds of thousands of Indonesians as part of a brutal anticommunist takeover, when Suharto came to power, during which time there was relatively little resistance either within Indonesia or from other countries. Indeed, some world leaders welcomed the killings as a favourable political development. An analysis shows that the conditions for encouraging political jiu-jitsu were present in 1998 but not in 1965.[37] The point here is not to recommend this particular analysis but to illustrate the potential value of studying non-action or relative lack of action.

The third element in the usual image of nonviolent action is that bodies are put on the line. Activists on the "front line" certainly receive the bulk of media attention, and most researchers followed suit. Yet important things can be happening elsewhere. In conventional warfare, each front-line soldier may be supported by up to ten others carrying out tasks such as supply, maintenance, intelligence, training, accounts and cooking. Also essential to modern military forces are research, development and production of weapons. A full understanding of military systems requires addressing all these dimensions. Similarly, in most nonviolent actions there are people working behind the scenes doing tasks such as publicity, community organising and legal support - perhaps research should fit in here somewhere! Support systems for nonviolent action have not been given much attention by researchers.

As well as looking at bodies behind the scenes, another potential arena for nonviolent action is where no bodies are involved at all. A prime example is cyberspace. Not only is the internet used to support activism, as an additional communication channel, but it can be the principal arena for some types of activism. For example, dissidents in repressive countries such as China have used email and websites to promote freedom and democracy. Others have defaced websites of government agencies or launched denial-of-service attacks. There are online struggles over surveillance, domain names, spamming, intellectual property, encryption and other issues. Just as militaries have added information warfare to their arsenals, so cyberspace has become a new arena for activism. Nonviolence researchers have a new area for study. But can cyberactivism be considered nonviolent action? After all, no physical violence is involved online, even if a Trojan horse or computer virus can do lots of damage to information. In order to study cyberactivism as nonviolent action, a generalisation of nonviolence theory is required.

The fourth element in the standard image of nonviolent action is that it takes place in public places. Media coverage of action reinforces this image. But it is quite possible for nonviolent action to occur out of the public eye. One key location is within organisations. This doesn't just mean strikes and boycotts, which usually challenge organisational elites by pulling things away - with workers withdrawing labour or consumers withdrawing their custom - but as an intimate challenge to the routine workings of organisations. The work-to-rule tactic is an example, as are some sorts of sabotage at work, though sabotage can stretch the boundaries of nonviolence. But there are also many subtle types of symbolic action and noncooperation by workers that are not easily captured by usual categories that focus on public action.

A useful beginning for analysis of organisational struggles is Deena Weinstein's picture of bureaucracies as analogous to authoritarian states.[38] Workers in bureaucracies do not have rights to free speech or assembly or election of their leaders. Those who openly oppose the hierarchy can be fiercely attacked, as usually happens to whistleblowers. The main way that bureaucracies differ from authoritarian states is that bureaucratic elites do not have violence at their direct disposal (with some exceptions such as the military), which may help explain why nonviolence theory has not been brought to bear.

Gene Sharp's theory of power applies most readily to authoritarian systems, where there is clear division between rulers and subjects. In general, it is easier to see how to proceed with nonviolence theory and practice when the opponent is clearly defined. Therefore, using Weinstein's conceptualisation of bureaucracies, it should be straightforward to apply nonviolence theory inside organisations. But only a tiny beginning has been made along these lines.[39] Again, this involves a generalisation of nonviolence theory to apply to situations where exploitation occurs but physical violence is seldom used.

One reason for the neglect of organisational struggles by nonviolence researchers may be an assumption that the key arena of struggle is the public sphere, where what is conventionally called "politics" takes place. If changes are made in government policies then, it is assumed, change elsewhere will follow, as in military policy, economic policy and organisational governance. So it is still far more common to be aware of public rallies against government war-making than of behind-the-scenes organising of nonviolent resistance within the military, and more common to be aware of public rallies against global economic planning meetings than of resistance organised by employees of multinational corporations. Part of this imbalance of awareness is that there is a lot more attention, especially in the media, to action in the public sphere. Nonviolence researchers tend to go along with this, giving more of their attention to public actions. Also, being in public, they are far easier to study. Gaining access to a secretive organisation can be a major challenge.

Some nonviolence researchers assume that nonviolence is a functional alternative for violence, doing the same job as violence but more effectively or with fewer negative consequences. This assumption is made by those advocates of nonviolent defence, such as Sharp, who envisage government leaders replacing military defence by a civilian-based alternative because they are convinced nonviolence will work better. With this assumption, there is little thought to changing other parts of society such as the economic system or organisational structures. This may help explain why there has been so little attention to nonviolent action as a challenge and alternative to capitalism[40] and as well as the neglect of democratising organisations through nonviolent action. There are many activists who are working toward economic, organisational and personal change, but nonviolence researchers have not paid much attention to them.

In recent years, the movement against corporate globalisation has used nonviolent action extensively. Many writers have analysed or glorified the movement,[41] but contributions by nonviolence theorists have been few.

Researchers and audiences

Most case-study-based nonviolence research is carried out by observers or observer-participants. The researcher collects information about the events or struggle of interest - from media accounts, documents, statements and reports by participants, interviews and sometimes personal participation - and then, using this information and drawing on concepts from nonviolence theory, writes an article or book. This is familiar and can be highly useful both intellectually and practically. The audience may be primarily scholars and students but sometimes is activists, for example if findings are circulated to participants in a timely fashion.

Most nonviolence research is non-interventionist: it relies on study of events that would happen whether or not they were being researched. This is the way most social research is carried out, but there are several alternatives. One is action research: carrying out research that is itself a form of social action.[42] For example, interviews might be carried out with workers to find out how they could nonviolently resist an invasion or coup.[43] The process of doing the interviews helps to increase awareness about the potential of nonviolent action. Action research could be oriented to policy makers, for example by interviewing government officials about how they could support a popular nonviolent struggle against a coup or other threat.

Some research in the natural sciences, such as astronomy, is carried out purely by observation, but in most fields experimentation is of crucial importance. In the social sciences, experimentation is far more common in psychology than in sociology or political science. The most well-known large-scale nonviolence experiment took place on Grindstone Island in Canada in 1965. It was a simulation of nonviolent resistance to a military takeover and resulted in many insights, especially about the need to rethink assumptions about how nonviolence would work.[44]

This sort of social experimentation has much to offer. It allows careful testing of hypotheses in a way not possible in "real-life" events. It can provide a stimulating experience for participants. And it can help bridge the gap between researchers and subjects.

In typical psychology experiments, the researcher sets up the situation and the subjects - often university students - are not involved in planning or assessment. Sometimes the subjects are dupes, not being told the real purpose of the experiment. This model might have something to offer to nonviolence researchers, but a more participatory model has greater potential: researchers and participants can jointly design, run and analyse the experiment. In a perfectly egalitarian model, everyone involved would be an equal researcher-participant, though in practice it is likely that some individuals will contribute more in certain areas. Even so, social experimentation could break down the some of the barriers between researchers and activists.

Social experimentation could focus on particular methods of nonviolent action, such as how to make a sit-in more effective. This overlaps with nonviolence training, for example to maintain nonviolent discipline in the face of police violence.[45] Studies of intuition have great potential relevance to understanding and improving nonviolent action, but of the vast amount of relevant research virtually none has been undertaken with nonviolence in mind.[46] Another direction for social experimentation is to examine alternative social arrangements, such as whether cohousing facilitates resistance to aggression and, if so, what features of the built or social environment make the difference. There are untold opportunities for such experimentation. Again, the audience for the research is partly determined by the participants in the experiments.

Leadership is an important issue that links research, action and policy.[47] Leadership is needed in each of these areas; for those committed to egalitarian group dynamics, models of shared or rotating leadership are available. The inspirational role of activist leaders is attested by the examples of Gandhi, Martin Luther King Jr., and Aung San Suu Kyi. There are also behind-the-scenes "heroes" of nonviolent action who contribute through political analysis and rational argument.[48] Similarly, in both research and policy on nonviolent action there is a need for both inspirational public figures and for those with a lower profile who get the job done. At a meta level, mentors are needed to encourage people to take leadership roles and to advise them through the challenges they inevitably face.

I've suggested some possible directions for future nonviolence research, but there are many others, such as nonviolence in everyday life, feminist struggles and global institutions. There is far too much to do even if the number of nonviolence researchers were suddenly many times larger. After all, nonviolence research has just as wide an ambit as military research but thousands of times less money. There is a surplus of ideas for nonviolence research; the challenge is to decide what is worth doing given limited resources.


Tom Weber provided many valuable comments on a first draft of this paper. The material here was presented at the Symposium on Nonviolent Research, Tromso, Norway, on 28-29 November 2002. I thank participants for many insightful comments. Thanks also to an anonymous referee for very helpful suggestions.

1 See, for example, Thomas Weber, "Gandhian philosophy, conflict resolution theory and practical approaches to negotiation," Journal of Peace Research 38 (2001): 493-513. In this context Weber notes especially the work of Johan Galtung (e.g. The Way is the Goal: Gandhi Today (Ahmedabad: Gujarat Vidyapith Peace Research Centre, 1992)) and Arne Naess, Gandhi and Group Conflict: An Exploration of Satyagraha (Oslo: Universitetsforlaget, 1974).

2 M. K. Gandhi, An Autobiography or the Story of My Experiments with Truth (Ahmedabad: Navajivan, 1940, 2nd ed.).

3 Tom Weber, personal communication, 12 November 2002.

4 Richard B. Gregg, The Power of Nonviolence (New York: Schocken, 1966).

5 Thomas Weber, "'The marchers simply walked forward until struck down': nonviolent suffering and conversion', Peace & Change 18 (1993): 267-289.

6 This sort of influence through intermediaries has been called the great chain of nonviolence by Johan Galtung, "Principles of nonviolent action: the great chain of nonviolence hypothesis," in Nonviolence and Israel/Palestine (Honolulu: University of Hawaii Institute for Peace, 1989), pp. 13-33.

7 Dennis Dalton, Mahatma Gandhi: Nonviolent Power in Action (New York: Columbia University Press, 1993).

8 Stephen King-Hall, Total Victory (London: Faber and Faber, 1943).

9 Stephen King-Hall, Defence in the Nuclear Age (London: Victor Gollancz, 1959).

10 Theodor Ebert, Gewaltfreier Aufstand: Alternative zum Bürgerkrieg [Nonviolent Insurrection: Alternative to Civil War] (Freiburg: Rombach, 1968); Bradford Lyttle, National Defense Thru Nonviolent Resistance (Chicago, IL: Shahn-ti Sena, 1958); Adam Roberts (editor), The Strategy of Civilian Defence: Non-violent Resistance to Aggression (London: Faber and Faber, 1967).

11 King-Hall, 1958, op. cit., pp. 128-129.

12 Gene Sharp, The Politics of Nonviolent Action (Boston: Porter Sargent, 1973).

13 For example, Etienne de La Boetie, Anti-Dictator (New York: Columbia University Press, [1548] 1942); Antonio Gramsci, Selections from the Prison Notebooks (London: Lawrence and Wishart, 1971). See Roland Bleiker, Popular Dissent, Human Agency and Global Politics (Cambridge: Cambridge University Press, 2000), for an examination of the trajectory of this idea.

14 Judith Stiehm, "Nonviolence is two," Sociological Inquiry 38 (Winter 1968): 23-30. On Sharp's relation to Gandhi, see Thomas Weber, "Nonviolence is who? Gene Sharp and Gandhi," Peace & Change 28 (April 2003): 250-270.

15 Gene Sharp, Making Europe Unconquerable: The Potential of Civilian-based Deterrence and Defense (Cambridge, MA: Ballinger, 1985); Gene Sharp with the assistance of Bruce Jenkins, Civilian-Based Defense: A Post-Military Weapons System (Princeton, NJ: Princeton University Press, 1990).

16 For example, James C. Scott is a high-profile and insightful scholar whose work - such as Domination and the Arts of Resistance: Hidden Transcripts (New Haven, CT: Yale University Press, 1990) - has obvious affinities to nonviolence but who does not forge connections with nonviolence perspectives.

17 For example, in the preface to The Politics of Nonviolent Action (op. cit., vi), Sharp states that his study is "a tool for increasing our understanding and knowledge; its propositions, classifications, analyses and hypotheses should be subjected to further examination, research and critical analysis."

18 Sharp, Civilian-Based Defense, op. cit., 124: "Very importantly, no peace or pacifist group or radical political organization should identify itself as the prime advocate of civilian-based defense. Neither should the new policy be presented in ways that might alienate conservatives and members of the existing defense establishment or the independent social groups and institutions that would bear responsibility for carrying out the future policy."

19 Brian Martin, "Gene Sharp's theory of power," Journal of Peace Research 26 (1989): 213-222, at 219.

20 Alex P. Schmid, with Ellen Berends and Luuk Zonneveld, Social Defence and Soviet Military Power: An Inquiry into the Relevance of an Alternative Defence Concept (Leiden: Center for the Study of Social Conflict, State University of Leiden, 1985), 209.

21 Robert J. Burrowes, The Strategy of Nonviolence Defense: A Gandhian Approach (Albany, NY: State University of New York Press, 1996).

22 Anders Boserup and Andrew Mack, War Without Weapons: Non-violence in National Defence (London: Frances Pinter, 1974); Gene Keyes, "Strategic non-violent defense: the construct of an option," Journal of Strategic Studies, 4 (June 1981): 125-151.

23 Burrowes, op. cit., 209; emphasis in the original.

24 Partly this reflects the predominance of men among nonviolence researchers: in terms of prominence and impact, there are no female equivalents to Gandhi or Sharp, for example. It would have been possible, though, to include women such as Joan V. Bondurant, known especially for her book Conquest of Violence: The Gandhian Philosophy of Conflict (Princeton: Princeton University Press, 1958/1988), who presents a Gandhian picture; Elise Boulding, Quaker, activist and peace researcher; or Barbara Deming, whose essays - for example, Revolution and Equilibrium (New York: Grossman, 1971) and We Cannot Live without our Lives (New York: Grossman, 1974) - might be classified more as insightful inspiration than scholarly analysis.

25 For example, leading social defence researchers who have written primarily in languages other than English include Theodor Ebert (German), Jean-Marie Muller (French), Johan Niezing (Dutch)

26 Peter Ackerman and Jack DuVall, A Force More Powerful: A Century of Nonviolent Conflict (New York: St. Martin's Press, 2000).

27 Randall Collins, Sociological Insight: An Introduction to Nonobvious Sociology (New York: Oxford University Press, 1982).

28 Howard Clark, Civil Resistance in Kosovo (London: Pluto, 2000).

29 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); Yeshua Moser-Puangsuwan and Thomas Weber (eds.), Nonviolent Intervention across Borders: A Recurrent Vision (Honolulu, HI: Spark M. Matsunaga Institute for Peace, University of Hawai'i, 2000); Roger S. Powers and William B. Vogele (eds.), Protest, Power, and Change: An Encyclopedia of Nonviolent Action from ACT-UP to Women's Suffrage (New York: Garland, 1997); Paul Wehr, Heidi Burgess, and Guy Burgess (eds.), Justice Without Violence (Boulder, CO: Lynne Rienner, 1994); Stephen Zunes, Lester R. Kurtz, and Sarah Beth Asher (eds.), Nonviolent Social Movements: A Geographical Perspective (Oxford: Blackwell, 1999). Ronald M. McCarthy and Gene Sharp, Nonviolent Action: A Research Guide (New York: Garland, 1997) is an annotated bibliography, mostly of books, covering cases of nonviolent struggle around the world as well as methods and dynamics of nonviolent action and theoretical works on power, conflict, and violence.

30 Sharp spells out his consent theory of power most explicitly in Sharp (1973), 7-62 and Gene Sharp, Social Power and Political Freedom (Boston: Porter Sargent, 1980), 21-67, 309-378. For critiques see Martin, 1989, op. cit. and Kate McGuinness, "Gene Sharp's theory of power: a feminist critique of consent," Journal of Peace Research, 30 (1993): 101-115.

31 Sharp (1973), op. cit., 447-814.

32 Peter Ackerman and Christopher Kruegler, Strategic Nonviolent Conflict: The Dynamics of People Power in the Twentieth Century (Westport, CT: Praeger, 1994).

33 Galtung's many contributions to nonviolence research are not collected in a single place. For an overview of his ideas see Johan Galtung, The True Worlds: A Transnational Perspective (New York: Free Press, 1980).

34 Bleiker, op. cit.; Glenn D. Paige, Nonkilling Global Political Science (US: Xlibris, 2002).

35 Tim Jordan, Activism! Direct Action, Hactivism and the Future of Society (London: Reaktion Books, 2002), 8, comments on the familiarity of the image of rallies. Another possibility for an implicit image is the sit-in.

36 Brian Martin, Technology for Nonviolent Struggle (London: War Resisters' International, 2001).

37 Brian Martin, Wendy Varney and Adrian Vickers, "Political jiu-jitsu against Indonesian repression: studying lower-profile nonviolent resistance," Pacifica Review 13 (2001): 143-156. See also Brian Martin and Wendy Varney, Nonviolence Speaks: Communicating against Repression (Cresskill, NJ: Hampton Press, 2003).

38 Deena Weinstein, Bureaucratic Opposition: Challenging Abuses at the Workplace (New York: Pergamon, 1979).

39 Brian Martin, Sharon Callaghan and Chris Fox, with Rosie Wells and Mary Cawte, Challenging Bureaucratic Elites (Wollongong: Schweik Action Wollongong, 1997). See also Brian Martin, "Whistleblowing and nonviolence," Peace & Change, 24 (1999): 15-28.

40 Brian Martin, Nonviolence Versus Capitalism (London: War Resisters' International, 2001).

41 One of the most perceptive accounts is Naomi Klein, No Logo (New York: Picador, 1999).

42 William Foote Whyte (ed.), Participatory Action Research (Newbury Park, CA: Sage, 1991).

43 Jacki Quilty et al., Capital Defence: Social Defence for Canberra (Canberra: Canberra Peacemakers, 1986).

44 Theodore Olson and Gordon Christiansen, Thirty-One Hours (Toronto: Canadian Friends Service Committee, 1966).

45 Howard Clark, Sheryl Crown, Angela McKee and Hugh MacPherson, Preparing for Nonviolent Direct Action (Nottingham: Peace News/CND, 1984); Virginia Coover, Ellen Deacon, Charles Esser and Christopher Moore, Resource Manual for a Living Revolution (Philadelphia: New Society Publishers, 1981); Per Herngren, Path of Resistance: The Practice of Civil Disobedience (Philadelphia: New Society Publishers, 1993).

46 For a readable summary, see David G. Myers, Intuition: Its Powers and Perils (New Haven, CT: Yale University Press, 2002).

47 I thank an anonymous referee for raising this point.

48 Ralph Summy, "Ordinary mortals of nonviolence," Gandhi Marg, 22 (October-December 2000): 279-290.