from Brian Martin, Uprooting War (London: Freedom Press, 1984); this is the revised 1990 version.
Go to

Index page for Uprooting War

Brian Martin's publications on peace, war and nonviolence

Brian Martin's publications

Brian Martin's website

To develop a strategy to overcome a social problem, it is useful to have a clear analysis of the present situation and a picture of alternative social arrangements which do not give rise to the problem. A strategy is then formulated to move from the present towards the desirable future.

The formulation of alternative social structures is not a trivial matter, nor can their development be left to an afterthought. Many social struggles depend extensively on opposition: opposition to nuclear weapons, to hazardous chemicals, to sexism, to exploitation of workers. This opposition can achieve a lot. But such opposition can leave the driving forces behind the problems unaltered, and so the same problem may persist or a different problem emerge. If an alternative to the underlying cause of the problem is provided, then campaigns against it will have more credibility and direction and also an increased chance of eventual success.

In the case of the problem of modern war, I would argue that the most important driving forces are political and economic inequality, especially as these are embodied in structures such as the state, bureaucracy and the military. The question then arises, what is an alternative to these structures that does not give rise to war? In this chapter I describe one important alternative, characterised by the term self-management.

First I will list some characteristics of self-management, then discuss the possibility and experience of self-management and how self-management fits into campaigns for social change, and finally mention some problems confronting the self-management alternative.

Characteristics of self-management

Self-management refers to forms of social organisation in which people collectively and individually have a great deal of control over the things which affect their lives. Most people are familiar with self-management in some aspect of their lives. For example, I used to play in an amateur woodwind quintet (flute, oboe, clarinet, bassoon and French horn) which we called 'Windpower.' As five individuals with individual lives and interests, we got together because we enjoyed playing as a group and occasionally performing. No one forced us to play in Windpower. But by participating we made an implicit commitment to each other to practise our parts, to come to rehearsals which we arranged, and generally not to let the others down.

The group was self-managing to the extent that everyone had an opportunity to contribute towards decisions about what music we played, when and how often we rehearsed, who we performed for, who filled a vacancy in the group and so forth. This meant that no single individual or other subset of the group was allowed to determine decisions unilaterally or otherwise dominate the others in the group.

This does not mean that everyone in the group had to be identical in personality or musical ability. Far from it! For example, Mary sometimes had suggestions or strong opinions about interpretation of a piece of music. The rest of us could accept her judgement or disagree with it. Even if, in many cases, we thought Mary's views were based on more experience or sensitivity, and accepted her views for that reason, we did not therefore give her any automatic or permanent control over decisions. The implicit principle for decision-making was consensus based on respect for each other's skills, views, experience and individuality.

By comparison with small amateur chamber music groups, the degree of self-management in many other musical groups is much less. In most professional orchestras, participation in decisions about selection of players, style of rehearsals, choice of music and many other things is more restricted than in Windpower. In orchestras it is common for a small group of leading players, or a management committee, or the conductor alone, to make the most important decisions without much input from others. The same pattern prevails in most government bureaucracies, corporations, political parties and professional bodies. Participation by most members is infrequent, formal, non-interactive and limited in scope. This is just the opposite of a self-managing group.

Self-management should not be considered in isolation from wider structures in society, which may limit or expand the self-determination of small groups. In Windpower, we could choose what to play from chamber music published around the world. But in many countries with authoritarian governments, particular types of music are banned for political reasons, not to mention restrictions on theatre, painting and literature. In Windpower, we could choose which opportunities for performance we wished to request and accept, such as the fund-raising 'Anti-uranium Ball' where we first played publicly. In many countries such association with political activities would lead to harassment.

Our own tastes and abilities restricted the interest that others had in hearing us, but opportunities also were affected by the availability of venues, competition from professional musicians and the attitude of certain influential figures in the local music scene. For us, as an amateur group happy to play for small audiences, these obstacles were not too great. But for individual performers hoping to play in the orchestra or opera, getting offside with influential figures in the amateur or professional music scene can lead to the withholding of opportunities to play.

Life in the non-musical world also affects groups such as Windpower. Our individual lives provided the time and incentive for amateur music. If our jobs or other activities had been all-encompassing, we would not have had time for practising individually or as a group. If we had terribly boring or physically debilitating jobs, we might not feel like playing music. If there were no places to practise without disturbing others, we could not play. If transport were inadequate, we could not get together regularly. If we were too poor we could not afford our instruments or music. Being a self-managing group, or indeed a group at all, depends on many such factors.

Finally there is the role of training, education and professionalism. Learning to play a musical instrument well takes a considerable amount of time and commitment not only from the player but usually also from a teacher. In Australian society both children and adults can take up a musical instrument, and many parents or individuals can afford to pay for instruction. This is not true in many parts of the world. Furthermore, at more advanced levels of playing music, there is a strong bias towards students intending to become professional musicians. Many excellent players must choose between a career as a professional musician and a non-musical career in which, by virtue of the required time and commitment, playing music becomes a minor activity or is dropped altogether. In short, amateur music survives in the interstices of professional music. But compared to many other amateur activities, amateur music is very healthy indeed. In areas such as medicine, science, law and many trades the opportunities for amateur involvement are either very limited or even restricted or prohibited by law, custom or other barriers.

The example of Windpower and performance of music is not as unrelated to the wider issues of self-management as it might seem at first. Self-management thrives in situations where bureaucratic or professional controls are weak and also where the people involved each have something to contribute and are reasonably small in number. Amateur music is one area where principles of voluntary association still hold sway. Of course, musicians seldom launch wars. But self-management is possible not only in music but also in agriculture, manufacturing, sport, families and many other areas of social life.

There are several characteristics which are associated with self-management, and it is worth mentioning these.

Lack of hierarchy

In a self-managing group, social interactions are not based on positions in a formal hierarchy of power or privilege. Instead, people are treated according to who they are as individuals.

Formal hierarchies are easy to recognise. They include a large fraction of social relations in government bureaucracies, schools, corporations and military forces. Also important are informal hierarchies, usually based on traditional social relations such as the dominance of men in many families or of gurus within certain religions. Hierarchies are at the root of a great amount of suffering and injustice in the world. They allow the exploitation of workers, the oppression of women and the mobilisation of armies.

One of the primary aims of self-management is to overcome these consequences of hierarchy by giving everyone a chance to participate in designing and maintaining social arrangements. Participatory social systems serve to prevent ruthless people from climbing ladders of ambition. Another aim of self-management is to offer individual and social satisfaction through involvement in social planning and decision-making. This helps reduce the drive towards power by providing less socially destructive satisfactions.


Self-management thrives on and promotes a relatively equal distribution of goods, services and skills. Rough equality allows wider participation. Wide divergences in material wealth, for example, often can be used by the wealthy to influence decisions and increase their political power.

Equality does not imply identity. What is important in self-management is a rough equality in those areas which affect on-going decision-making. People can be quite different in personality, tastes, skills, friendship groups, personal possessions and activities, so long as these differences do not permit any individuals or groups to build positions of dominance over others. The differences most threatening to self-management involve small group control over wealth (including land, money and factories), over instruments of violence and over special skills and knowledge.

Self-management, once established, tends to undermine inequality in dominance-creating areas: wealth is more equitably distributed, monopolies over violence are removed and skills and knowledge spread.


Self-management requires an orientation towards cooperation rather than competition. Chamber musicians, like groups building a house or planning an education programme, achieve a lot more when they cooperate. This involves mutual support, mutual learning and mutual constructive criticism.

If self-managing groups are to operate successfully, participants need to have good interpersonal skills: they need to be able to listen, share, be sensitive to the feelings of others, and be committed to the group's endeavours over personal trifles and selfishness. Of course, not everyone can do this readily. Learning and experience in cooperative group dynamics is needed. And one of the best places to do this is in a supportive, self-managing group!

By contrast, competition thrives in hierarchical situations where people strive separately for a small number of privileged positions in society. Competitive systems are characterised by many losers and consequent apathy, disillusionment, resentment and wastage of talent. Contrary to popular belief, cooperation is almost always a more efficient way of doing things than competition.


Self-management has a much better chance when there is a shared set of experiences, circumstances or interests: in short in a community. Communities can be built around occupations such as rail work or scientific research, around interests such as playing music, around shared surroundings as in geographical neighbourhoods, or around shared lives as in extended families or communes. People in communities share experiences and often share goals. This is a useful basis for developing self-management.

Communities provide a basis for opposing centralised control. For example, Nigel Young has found that individual resistance to conscription is greatest where local communities are strong and support the individual acting in opposition to the state.

Self-management does not come automatically to communities! Patriarchal families or office workers in a state bureaucracy are not self-managing. But in these cases, it is often those subject to the arbitrary power of others that share the greatest sense of community: the weaker members of an extended patriarchal family, or the low-level office workers.

Social conditions strongly influence whether members of a community can effectively organise. For example, workers can organise better against employers in industries that provide easy communication, freedom from supervision, or multiple employers.

Where community is lacking, the opportunities for self-management are slim, as in the case of workers who work individually rather than collectively, in highly competitive situations, in neighbourhoods with a high turnover of residents, or among the home-bound.

Small size

Self-management works better with small groups. Five people can play most chamber music without a conductor; for a 100-piece orchestra a conductor often is a necessity. Similarly, it is easier to cooperate efficiently in small groups in making food or diagnosing illness.


To be self-managing, a group needs a reasonable control over its local situation, such as over land, resources and skills. The essence of self-managing decentralisation is that key facilities are controlled by those who use them. Chamber music playing is decentralised to the extent that players control their own instruments, music skills and access to rehearsal facilities. In a centralised system, musical authorities would strictly control use of instruments, music, recording rooms, and opportunities to rehearse and perform. Amateur music is quite decentralised. By contrast, energy supply in most industrialised societies is centrally controlled by governments and corporations, via large-scale production of oil, coal and electricity. The 'alternative energy movement' has promoted energy systems which can be controlled locally, such as energy-efficient buildings and small-scale solar, wind and biogas systems. Decentralisation does not guarantee self-management, but it helps.

To establish self-managing interactions between dispersed individuals and groups, some form of coordination is needed. One example of how this can be done is provided by the Amateur Chamber Music Players (ACMP), a body based in the United States which prepares a list of people around the world interested in playing amateur music. People simply send in their name, address, instrument, and self-rating of playing standard to the ACMP, which regularly sends out a full list to everyone on the list. (The costs of this come largely from donations.) Players make their own arrangements by contacting others on the list, for example when travelling. The ACMP has no power to force or prevent people from playing. The ACMP thus is a network which fosters self-management in music.


Self-managing systems tend to be more flexible than large, hierarchical, centralised systems. We in Windpower can change our rehearsal schedules to suit our individual and group needs. This is not always easy, but there is a lot more flexibility than with a large orchestra. Similarly, self-managing groups of students can try out special methods for individual learners, and self-managing groups of workers can alter their schedules and work arrangements for a member who has a special engagement, illness or handicap. By contrast, bureaucracy, the epitome of a non-self-managing system, is notorious for being inflexible to the individual requirements of workers and clients.

Purposefully designed

The possibility of self-management does not depend on the innate goodness of people or natural compatibility within groups. Anthropological studies of numerous non-industrial societies show that there is scope for a wide range of cultural arrangements, from non-hierarchical cooperative societies to violent competitive systems. There seem to be no genetic or other intrinsic barriers to self-management.

Furthermore, why be bound by the models from past or existing societies? Humans have the capacity to become aware of their own social arrangements and to design social systems with desired properties, just as technology can be designed with certain mechanical and social goals in mind.

For example, the methods used in nonviolent action training, including techniques for consensus decision-making, clarification of individual and group goals, and analysis and modification of group dynamics, can be used to mould individual and group behaviour in ways desired by the participating individuals. To be viable alternatives, self-managing political and economic systems do not have to be immediate, spontaneous or fully elaborated in advance. Instead, they can be developed as part of a process of self-conscious individual and social transformation.

With this last point in mind, it is worth considering some experiences with self-management with an eye toward their relevance to eradicating the causes of war.

Lessons from history

The most dramatic examples of large-scale self-management have occurred in revolutionary situations.

In these and other cases, major social, political and economic activities have been organised by the workers or population concerned, without the need of bureaucracies, managers or political elites to tell people what to do. Workers have taken over and run factories. Self-organised groups have handled distribution of food, goods and health care. Distribution of land has been made to those who farm it, by agreement of all involved. Education, media and welfare have been organised by workers and community.

There is much to learn from such revolutionary episodes, and indeed there is a thriving literature on the Spanish events, for example. One vital lesson is that widespread self-management is not only quite possible, but that its potential normally lies hidden under the weight of oppressive and debilitating social structures, only to be revealed in conditions of enormous social upheaval.

The Paris Commune was destroyed by the French military. The Russian soviets were used by the Bolsheviks to gain power, but later destroyed by them. The Spanish collectives were opposed both militarily by Franco's fascist forces and also by elements within the Republican coalition, especially the communists. In each of these three cases, the self-managing initiatives originated and were destroyed in the context of a major military struggle: the Prussian defeat of France in 1870, World War One and the war by the Bolsheviks against numerous opponents and invaders in 1918-1920, and the Spanish Civil War.

Self-management provides an alternative to the hierarchy and inequality which characterises war-based structures. But when self-management develops spontaneously in the midst of major military and social upheaval, the chances of its survival seem to be low. On the one hand, military forces mobilised on the occasion may be used to destroy the social revolution, as when the French army, defeated by the Prussian army, was used to smash the Paris Commune. On the other hand, the self-managing bodies may initiate or join a military resistance, as in the case of the Spanish Civil War, with a corrosive effect on self-management even if the military resistance is organised democratically and is successful. In the case of the Russian Revolution, both mobilisation for the civil war and the developing centralisation of Bolshevik power served to destroy the soviets.

Social revolutions often seem larger than life. It couldn't happen here, could it? Possibly not at the moment, since such upheavals depend on a particular set of social conditions, which seldom can be consciously planned. Nevertheless, even seemingly stable societies may suddenly pulse with the demand for more participatory social structures, as in the case of France in the events of May and June 1968. The opportunities opened by such events are often wasted. Spontaneity is seldom enough. Prior planning, preparations and working out of strategies may not be enough either, but they may provide politically fruitful channels for the spontaneous energy unleashed in revolutionary situations.

On a much less grandiose scale, but still highly important, are experiences with self-management in particular industries. In July 1971 the British government announced that two of the four yards of the Upper Clyde Shipbuilders (UCS) were to be closed. For about six weeks the workers had been discussing and planning what to do when this closure occurred. What they did was organise a 'work-in.' They took control of the yards and continued construction on the projects currently in the yards, organising jobs, pay, insurance and publicity themselves. The work-in continued until October 1972, when all the yards were made viable through government grants.

The UCS and other similar experiences show that workers can organise production by themselves quite adequately. The work-in as a form of social action is only suitable under certain conditions, in particular when (as in the case of shipbuilding) there are not constant problems of obtaining supplies and selling products. Employers and governments invariably oppose work-ins and other strong self-management initiatives, and take various measures to induce these alternatives to fail. Because of this opposition and because of the risks generally, workers usually contemplate work-ins, sit-ins and other forms of direct action only when other avenues have failed. The UCS work-in was catalysed by the threat of massive layoffs.

Work-ins are a fundamental challenge to managerial control, and also give workers an experience in running their own affairs. In contrast, strikes seldom challenge hierarchical social relations themselves, but are used to achieve more benefits for the strikers within existing social relations. The work-in, in various forms, would seem to have a lot of potential in other contexts, especially in service areas. Instead of going on strike, bus drivers can work as usual but refuse to collect fares. Students can organise their own learning.

Self-management in social revolutions and in work-ins are examples of what can be achieved in the course of an urgent social struggle. But in neither case has a long-term achievement of self-managing structures been easy. What are the prospects for self-management as a more gradually introduced alternative?

There are quite a number of otherwise conventional enterprises around the world which incorporate features of self-management. One example is the Scott Bader Company in Britain, a substantial chemical company which is organised to allow workers to be involved in policy formulation and also to share in profits. Many large companies, as in West Germany and Sweden, have worker representatives on company boards, and also promote various types of job rotation and worker control over conditions at the shop floor level.

The best known example of worker self-management on a large scale is Yugoslavian enterprises, about which much has been written. As a matter of state policy, formal structures for industrial democracy are found in all substantial enterprises. Policy by enterprises is decided upon by representatives from all parts of the workforce. The Yugoslav experience has been mixed. Certainly there has been much more worker participation in industry than in other countries, capitalist or state socialist. But the balance, degree and enthusiasm of participation has not reached expectations. One problem is that skilled and better educated workers tend to become representatives, while others remain apathetic. Another limitation is that overall economic policy is decided at the state level, and hence the scope for self-managing initiative is not as great as it might appear on paper.

Worker representation at management levels and job enrichment can be steps in the direction of worker self-management, but they are severely limited. Worker representatives often become coopted into management perspectives, and job enrichment and limited autonomy at the shop floor level may be used to reduce discontent and head off demands for more fundamental changes.

It is important to distinguish between worker participation and worker self-management. Worker participation is sometimes promoted by management, who use it to consult with workers about decisions, thereby reducing discontent and gaining some useful ideas while still maintaining ultimate control over decisions. Worker participation can be a means of mobilising a degree of support or tolerance for management policy. Worker self-management, by contrast, implies a dissolving of the management-worker distinction.

What can be learned from the various official forms of self-management in industry in relation to the problem of war? First, measures of self-management do not automatically lead to antimilitaristic orientations. The Spanish anarchists organised themselves to fight in the Civil War in a self-managing way, but this approach has little potential for undermining the roots of war. Most of the companies with features of self-management manufacture the same range of goods as other companies, such as the chemicals produced by the Scott Bader Company. Indeed, arms production by workers' co-ops is not uncommon. Even the UCS workers began their work-in to continue making ships, not to challenge the orientation of industrial development. This is more or less inevitable given that these companies and workers must survive within the wider economic system, which itself is tightly linked with the state and war.

There is an interesting connection between Yugoslav self-management and Yugoslav national defence. Yugoslav defence planning is based on the principle of 'a nation in arms': rather than relying entirely on professional military forces, the Yugoslav government supplements its conventional forces with plans and preparation for arming the people and conducting partisan warfare. There are some geographical and political reasons for this, namely Yugoslavia's rugged terrain and its position outside the major military alliances in Europe. But defence by arming the people also goes hand in hand with self-management in industry. In both cases a much greater reliance on and trust in the people is apparent. Guerrilla warfare is not a solution to the problem of war, but it does not hold the same potential for mass destruction or of extreme violence controlled by a very small political or military elite. [Events since the breakup of Yugoslavia suggest that this assessment was too optimistic. An alternative conclusion is that training in violence is inherently risky.]

The two other countries in Europe which have adopted the 'nation in arms' approach to the greatest extent are Sweden and Switzerland. Their governments have supplemented their conventional military forces with militias, civil defence preparations and various other avenues for participation in national defence. It so happens that the Swedish and Swiss political systems also allow a somewhat higher degree of democratic participation than most other European countries. So, while recognising the many limitations of the social systems in Yugoslavia, Sweden and Switzerland, the evidence suggests some degree of general correlation between greater self-management in political and economic systems and in the mode of defence. This gives some hope that campaigns towards more extensive and deeper self-management in political and economic spheres will support and be supported by campaigns for wider self-management in defence, and social defence in particular.

Self-management within existing structures

Self-management is sometimes associated with dramatic social confrontations, such as social revolutions or work-ins. But there are also elements of self-management in many aspects of daily life, so commonplace as to be unremarked, as in the case of Windpower. It is worth becoming aware of these elements of self-management, so that they can be protected, deepened and extended.

Public parks and public libraries contain elements of self-management. Although they are usually managed by professionals or bureaucrats, their use by the public depends on widespread acceptance of their value as a public resource of potential benefit to all. If even a tiny minority set out to cut down trees in parks or steal books from libraries, these public resources would quickly degenerate. It is not often realised today that parks and libraries were only established after significant social struggles in the mid 1800s. The ruling elites at that time did not believe that the common people were sufficiently responsible to care for parks and libraries. The struggle to protect these public resources has not ended, and indeed public parks have been the focus of many environmental battles between citizens and government or corporate developers.

Public facilities such as parks and libraries show that communal resources are both possible and beneficial. The extension of such resources to include films, computers, printing equipment, videos, machine tools, bicycles and many other items would be a valuable focus for social action campaigns. At the moment, public parks, libraries, telephones and swimming pools (not to mention schools and hospitals) are managed by professionals or government bureaucrats. Another important focus for social action is to increase local community control over such public resources.

An important element of self-management within the legal system is the jury. The jury system limits the control of professionals and bureaucrats. The selection of jurors by lot means that no individual can serve on and dominate a succession of juries. Selection by lot also minimises inequalities in participation by class, age, sex and race. Of course juries are manipulated and circumvented in various ways by politicians, judges, police, lawyers and others, for example through the design of the law and through selective arrest and trial. But as an element of self-management, the jury system should be defended and the principles of its operation extended to other fields.

Self-management and social action

Some of the most systematic attempts to develop and promote self-managing political and economic structures have been taken by social action groups and social movements, such as the feminist movement, the workers' control movement and the alternative technology movement. There are two major and interlinked focusses: self-management as a goal and self-management as a method.

As a goal, self-management can be seen as an implicit or explicit guiding principle in many campaigns against systems of unequal power. Feminists oppose the laws, regulations, habits, practices and attitudes which perpetuate the domination of men over women. Environmentalists oppose nuclear power partly because of its intimate connection with centralised political and economic power.

As a method, self-management is the preferred organisational principle for many social action groups. Self-managing social action groups try to promote participation by all members, share both routine tasks and exciting opportunities, use consensus or other democratic forms of decision-making, encourage members to develop a range of skills, and provide a social environment for emotional satisfaction and sharing as well as accomplishing tasks.

While some self-managing social action groups mainly focus on challenging oppressive structures, others attempt to build alternative structures. A standard form for a self-managing structure is the cooperative, whether it is a food co-op, media co-op, women's refuge co-op, health co-op or manufacturing co-op. Cooperatives, as the name implies, are an alternative to the bureaucratised provision of goods and services. Co-ops are based on sharing work equitably, encouraging participation, reducing exploitation of workers, and sharing the benefits of the cooperative endeavour.

There are quite a number of action groups, co-ops and other groups around the world which aim at self-management as a method and a goal. So far their impact on the dominant hierarchical structures of corporations, state bureaucracies and professions has been limited. There are several problems facing self-managing social action groups.

First, it is not easy for a group of people to operate cooperatively and non-hierarchically when most of its members grew up in and live in a hierarchical society: in families, schools, bureaucracies and the like. As most participation in social action groups is a marginal and part-time activity, there is a continuing struggle to overcome the conditioning and experience that is so at variance with self-management.

Second, for most people there is a shortage of time which can be committed to self-managing groups. This is a straightforward practical problem in the face of job, family and other commitments.

Third, there is a continual conflict between achieving internal participation and agreement, and running campaigns or doing a job. The poles of danger are either becoming an introverted support group with no external impact, or running an externally successful campaign or service at the expense of internal self-management.

Fourth, there is little sense of overall coordination or strategy guiding the action of self-managing groups. Many groups focus on their own little area without a picture of where these efforts fit into a programme for social change. Although many activist groups keep in informal contact with each other, there is surprisingly little effort to jointly assess goals and methods and coordinate campaigns.

Fifth, self-managing groups are relatively few and weak compared with the dominant bureaucratic organisations and infrastructures, and the risks of being destroyed, being coopted or just giving up are quite high. Being small and weak is self-perpetuating, since few people will join a movement which seems ineffectual and lacking a convincing programme.


How can the present large, hierarchical and centralised political structures be transformed into or replaced by self-managing structures? This is a key question for all individuals and groups which adopt self-management as a goal or method. It is also a key question for those who see present structures as a root cause of war. Sadly, there is relatively little discussion of this question even in self-managing social action groups, much less any widely accepted answers. The reason for this is that since most self-managing groups are small and weak, it takes most of their energy just to survive and to run a few campaigns. In addition, the general problem of major social transformation seems so enormous and remote as to be not worth considering yet.

Actually, it is unfair to say that most self-managing social action groups have not thought about how social change towards self-management will occur, since they are in the process of promoting such change themselves. Once general principles and goals are decided upon, such as self-management, nonviolence, sexual equality, respect for nature and organising at the grassroots, then it is not hard to orient campaigns in these directions.

For example, if maximising participation is a goal, then an unemployment action group will favour signature drives and door-to-door canvassing over running advertisements, and a food co-op will avoid dependence on paid staff and large corporate suppliers. In my experience, the implicit strategy underlying the actions of most social action groups is that the important thing now is to move in the correct direction. This requires both deciding on basic principles and promoting changes either internal or external to the group.

There is a lot to be said for this implicit strategy. It allows useful action to be carried out without waiting for a comprehensive programme. It is flexible and does not prejudge the possibilities for change. And it is not exclusive, but rather encourages individuals and groups to contribute in their own ways.

There are also some severe disadvantages to the implicit strategy of moving in the correct general direction without a more fully developed programme. It is easy to become sidetracked into superficial issues or actions, for example focussing solely on opposing particular weapons rather than focussing also on the systems which promote the creation and use of weapons. It is also easy to only use methods which make people feel they are doing something, such as rallies and letter writing, rather than ones which are actually effective in helping change structures.

Another big problem is that whole areas of vital importance to social change towards self-management are overlooked since groups stick only with their particular issue. Direct effort towards conversion of military establishments, for example, is not even on the agenda of most peace movements. Finally, the implicit strategy of moving in the correct direction avoids the difficult problems confronting social movements as they become larger and more powerful. How are they to be coordinated? How are they to be institutionalised? Many a social movement which has suddenly grown large, as the peace movement in the early 1980s, finds itself without ideas or methods for utilising and channelling this upsurge of support in an effective way.

Another strategy for replacing hierarchical structures by self-managing ones is the strategy of dual power. The basic idea is to start creating self-managing structures, even if only on a small scale, in all areas and at all levels: co-ops, alternative justice systems, alternative media and communications, etc. The hope is that more and more people will transfer their allegiance to the self-managing structures and that the hierarchical structures will collapse or wither away, leaving the self-managing structures as the dominant organisational form.

The strategy of dual power has been promoted especially by sections of the anarchist movement. Certainly it is a clear application of the anarchist principle that the ends should be incorporated in the means. Dual power consists of making the desired end, self-management throughout society, as the basis for present campaigns.

In spite of its attractiveness as a principled method for social change, the strategy of dual power has several shortcomings. The main one in my view is that the organs of dual power (the co-ops, alternative health services and so forth) by themselves are unlikely to cause the established structures to decay or collapse. Indeed, so long as hierarchical structures remain strong, they can suppress, tolerate or coopt small self-managing alternatives. The continuing difficulties confronting co-ops, initiatives for local democracy and so forth testify not only to overt obstacles but also to the pervasive role of socialisation in bureaucratic organisations. The building of self-management at the grassroots needs to be supplemented by campaigns to challenge and undermine large-scale hierarchical structures.

Problems and limitations

There are numerous obstacles, traps and unknown areas to be encountered in working towards self-managing political and economic structures. Here I outline a few of the key problems and limitations.

Personal skills for self-management

People need to develop skills and attitudes conducive to self-management. For most people who have grown up and lived in hierarchical families, schools, corporations, state bureaucracies and professions, working in a non-hierarchical group takes a fair bit of adjustment and learning. This requires tolerance and support from all concerned. It also benefits from practical tools and exercises for encouraging and equalising participation, handling disagreement and conflict, sharing interpersonal skills, and evaluating activities and feelings. Self-management is not an instant utopia, and unrealistic expectations for it should not be fostered. But it can be satisfying and challenging given sufficient commitment and favourable external conditions.

Limits of consensus

Since the late 1960s, there has been a great expansion in the use of methods for consensus decision-making in social action groups. This has been a fundamental part of the promotion of self-management in these groups. Rather than use a decision-making procedure such as voting which assumes conflict and creates losers, the aim has been to look for solutions acceptable to everyone. If a strong objection to a proposal is made by a single member, no action is taken until an agreed-upon solution is reached. Associated with use of consensus is the principle of equal respect for each person and the practice of meeting face-to-face. Opportunity and usually encouragement is given to each person to participate fully.

There are many advantages to use of the consensus approach. It helps to orient thinking to common interests, to avoid prematurely settling on a less than adequate solution, to incorporate insights from all people, and especially to unify the group behind the decision taken. Most important in terms of self-management, consensus techniques are effective in preventing the rise of traditional power elites.

While consensus has many advantages for social action groups, it is not the solution to all problems. Yet use of consensus methods has become a virtual dogma in some activist circles. At the same time, consensus has largely been ignored or rejected in other areas such as trade unions and political parties. What are the limits of consensus? A major contribution towards answering this question has been made by Jane J. Mansbridge in Beyond Adversary Democracy. Mansbridge distinguishes between unitary and adversary democracy. Unitary democracies are like friendships: they are based on a high degree of common interest, and tend to be based on consensus. Adversary democracies assume the existence of strong conflicts of interest, and typically use majority rule, the secret ballot and other means for equal protection of interests (rather than equal respect).

Mansbridge notes that liberal democratic theory has focussed almost entirely on adversary situations. In practice, many unitary elements enter in, as in the Vermont town meeting government which she studied. The promotion of consensus methods by social action groups has redressed this imbalance and drawn attention to importance of seeking out common interests.

Mansbridge has made important observations about the limits of consensus, based on part on her detailed study of a successful participatory workplace. The most difficult problem for the consensus approach is dealing with fundamental conflicts of interest. In face-to-face meetings, genuine conflict tends to be suppressed. Those who are less confident or experienced tend to feel intimidated. Those with more experience in the organisation tend to have more contacts, inside knowledge and social networks, all of which give them more power. By contrast, adversary methods are well suited for dealing with conflicts.

In situations of conflict, settlements based on taking turns, majority rule or arranging a set of outcomes proportional to those people desiring them are often best. But with consensus methods it is hard to reach these outcomes because the adversary methods of bargaining and compromise are discouraged. For example, consider the problem of a peace group deciding whether to criticise Soviet militarism. With majority rule, an adversary procedure, a vote would be taken. But if consensus is the procedure, then one or two pro-Soviet members can prevent action. When there is a strong disagreement over Soviet militarism (or United States militarism for that matter), the effect of relying solely on consensus is to suppress debate and end up with a pro-peace stance with little substance or scope for organising action. Using consensus, it would be difficult to reach a solution based on separate efforts, namely forming different groups or allowing independent actions by subgroups.

The wider problem, only touched upon by Mansbridge, is how to organise systems incorporating the strengths of both unitary and adversary democracy. How can groups switch from consensus to voting and back again depending on the issue and level of agreement? And how can large-scale coordination of self-managing groups be organised without losing the benefits of small-scale consensus? This latter question leads to the next topic, coordination.


How can and how should different self-managing organisations coordinate their activities? Consider a group of co-ops for food production and distribution, mining, manufacture and distribution of goods and so forth. How should they coordinate their activities? One method that avoids hierarchy is to demand a high degree of local self-sufficiency, so that little overt coordination is required. But such a high degree of self-sufficiency may not be possible or desirable even in the long term. Certainly in a transition to greater self-management, the existing interdependencies must be dealt with. The concepts of the market and of centralised planning are the standard coordinating principles, but they each generate strong pressures for the creation of elites.

The most common idea for coordination of self-managing groups is through some sort of federation, in which member bodies would be represented in some way. How would representation be made? The problem with elected representatives is that those elected soon develop special interests of their own, and use their position to entrench their power and that of others like themselves. One way to tackle this is to have delegates rather than elected representatives. The delegate from a group would be expected by the group to present the group's views, within a predecided range of personal initiative. Delegates could be changed at any time: they would be revocable.

The idea of a revocable delegate implies that the bodies to be represented are relatively small: from perhaps ten up to several hundred people. For larger scale coordination, several layers of representative bodies based on delegates could be envisaged.

The delegate method has worked reasonably well in consensus decision-making involving hundreds and even thousands of people, especially within the coalitions against nuclear power. Affinity groups of perhaps a dozen people send delegates to a central meeting. If consensus cannot be reached immediately, delegates report back to their affinity groups and reconsideration of views or new proposals is sought.

In spite of the successes of this consensus procedure, and in spite of the scope for learning and improvement in consensus methods, there are likely to be serious obstacles to coordination of self-managing groups for which the delegate method will have difficulty. There is the danger of strong personalities building positions of considerable informal power. There is the problem of reaching agreement when quite contrary interests and viewpoints are involved. And there is the problem of the experts (those who know more about a specialised subject, such as health, farming or consensus methods themselves) who may find themselves thrust into positions of power without the incentive to spread knowledge and decision-making in these areas.

One possible method for overcoming some of these problems of coordination is use of the lot or jury system. The lot system was used in ancient Athens to select public officials. (The exclusion of women, slaves, children and resident aliens from consideration in the lottery indicates its incomplete application.) This system, as mentioned earlier in regard to the jury system, contains many features supportive of self-management: participation is equalised and the consolidation of elite power is minimised. The lot system could readily be used to select gradually rotating committees, for example to coordinate co-op production and distribution or to make suggestions on major projects such as selecting transport routes.

The lot system would not disenfranchise people who were not selected, since after all they could still communicate their views in various ways and if necessary use nonviolent action to press their claims. What the lot system so nicely achieves is the prevention of people obtaining formal positions of influence simply because of connections, personality or skills. It is revealing to hear objections to this system, which often come from people who would normally expect to gain positions of formal or informal influence.

John Burnheim, in his penetrating book Is Democracy Possible?, has developed the idea of the lot system into a full alternative to the state and bureaucracy. Burnheim's alternative, called 'demarchy,' involves a dense network of local decision-making groups chosen by lot from volunteers. Each group would deal directly with a functional area, such as roads, schools or garbage collection. There would be higher-order groups (composed of people from the regular groups) to adjudicate procedural issues, but there would be no central decision-making body.

The great advantage of demarchy is that it allows a large degree of direct citizen control without requiring everyone to know everything about every topic. Personally, I believe it is the most promising prospect for dealing with problems of coordination in a self-managing society.

The lot system has hardly been tried by self-managing groups. One reason is that such groups are seldom large enough to require the coordinating role facilitated by the lot system. Another reason is that influential people in social movements have not been on the lookout for ways to counteract their own power. More often than not, large meetings of social activists are dominated by experienced 'heavies' from local areas and especially by full-time activists, often in paid positions. The lot system provides a threat to those who seek power and influence in the 'alternative' movements.

Because the lot system can be used in small ways immediately, it is an alternative that is readily turned into a campaign. But so far this process has hardly begun.


Another big problem is maintaining self-managing features in the face of pressures for bureaucratisation: cooption in the bureaucratic system. Consider for example a consumer movement (though the process applies also to political parties, trade unions, peace groups and sundry others). In the early stages there are many small independent grassroots initiatives. To coordinate these a national body is set up. As more people join the movement, several full-time staff are hired to handle the national administration. The government, media and other influential groups take notice of the movement, and consult the movement leaders and staff. Perhaps government funds are provided, or employers agree to administer a salary deduction scheme to fund the movement. Pressures mount to increase the paid staff and to formalise the administration. The national office, because of its dependence on funding or its contacts with national elites, supports a more moderate line on consumer issues than most of its constituents. The process of bureaucratisation has begun.

Is this process inevitable, as social theorist Robert Michels and his 'iron law of oligarchy' claims? No: some groups largely avoid the process, such as Friends of the Earth in Australia. But the pressures in this direction are strong, especially the pressures arising from the bureaucratic organisation of the state, including political parties. To work through the state bureaucracies or the major political parties, it is often more effective in the short term to adopt a similar bureaucratic form. To challenge the fundamental interests of the state bureaucracies is another matter. Without a firm commitment to a grassroots strategy for challenging structures, it is all too easy to fall into the process of bureaucratisation.


While a prime danger of working within established structures is bureaucratisation, a danger in staying too far away is marginalisation, or in other words being permanently on the fringe. Many of those who set up rural communes in the 'back to the land' movement which burgeoned in the 1960s wanted to avoid being compromised by the 'system,' including schools, corporations, and the consumer society. So they left the scene physically. One problem with this is that a gradual re-creation of many conventional structures is likely unless collective efforts are made to prevent this. After all, it is one thing to move physically and another to change mentally, emotionally and organisationally.

Another problem with many rural communes and other attempts to break significantly with established structures is that a connection with on-going struggles to challenge those structures may be lost. The temptation is to become engrossed in surviving independently and enjoying apparent freedom and so to opt out of social struggle. Unfortunately it is not enough simply to live the alternative. Some governments use repressive measures against those adopting 'alternative life styles,' while others more tolerantly allow them to pursue their own way as long as they don't create a fuss. Living the alternative will not lead, by itself, to the downfall of systems of hierarchy and centralised power.

The challenge to those in rural communes and others in 'alternative' structures is to maintain a connection with struggles to change mainstream structures. This is not easy. Some governments may tolerate a minority of 'alternative lifestylers' on the fringe, but they are less likely to turn a blind eye to political action from such groups.

Self-management for elites

Consider a team of doctors, all highly qualified and experienced specialists. They might organise their practices in a non-hierarchical, mutually supportive way, in most ways quite in tune with the principles of self-management. But what about the nurses, technicians, receptionists, accountants, cleaners and spouses? They might be left in the familiar hierarchies and division of labour. Indeed, in many oligarchies, the rulers practise self-management: they organise their own lives, and everyone else's too.

The possibility that privileged groups may practise self-management without including others points to a problem in promoting self-management in hierarchical organisations, or in occupations such as medicine based on professional dominance. Should self-management be promoted initially at different levels of the hierarchy, or should the full division of labour be confronted from the outset? This is an important problem. It is not too difficult to find support for self-management if it can be restricted to a particular stratum or occupational specialty: people often know and sometimes respect each other, and there is little loss of status involved in fostering participation.

In many professional groups, a considerable degree of professional equality and autonomy is taken for granted. In academia, scholarly contributions are supposed to be treated according to merit, and autonomy is justified on the basis of the benefits of academic freedom. But such groups are strongly antagonistic to sharing their 'privileges' or involving others. Academics are usually very hostile to students sharing in academic decision-making such as deciding on curriculum, hiring of staff or even serving on committees. The implication is that campaigns for self-management cannot uncritically accept occupational or organisational structures. Promoting self-management among particular groups can be a useful contribution, but not if it strengthens the position of those groups over others.

A related problem is the promotion of self-management among those doing useless or harmful things. The self-managing team of doctors might be cosmetic surgeons for the rich, or specialists in heart transplants. Given the pressing health needs of the poor and the diversion of resources and attention away from preventive health measures towards high technology curative medicine, these are examples of useless or harmful medicine. Other examples can be found in the growing of tobacco, the building of planned obsolescent cars, the designing of tax loopholes and the manufacture of chemical weapons. Is it really worth promoting self-management among tobacco growers or tax avoidance lawyers? The answer is not simple. Self-management can help rescue or streamline failing enterprises, and thus prop up harmful activities. But as suggested by the case of the Lucas Aerospace workers, participation in decision-making about one's work can lead to questioning about the products as well as the work organisation. The challenge is to develop campaigns which encourage this sort of questioning.

The goal of workers' control in industry illustrates both these problems. Workers' control can be used to entrench the position of current workers, and so keep out women, racial minorities and even political dissidents. And workers' control can be introduced in undesirable or marginally beneficial industries.

Preventing counter-revolution

In social revolutions, the greatest immediate threat is counter-revolution: the smashing of people's self-managing organs, often by military attack. To oppose the counter-revolution, there is a strong temptation to rely on a revolutionary elite, as in the case of the Russian Revolution and the ensuing civil war. How can counter-revolution be prevented without creating a new permanent elite? There is also the problem of the conservatism of the elites in supposedly 'progressive' groups, as illustrated by the lack of support by the French Communist Party in May-June 1968 for the student and worker initiatives. How can the opposition of 'progressive' elites to self-managing initiatives be overcome?

A general answer can be given. The idea and practice of self-management must be spread at all levels, laying the groundwork for action in a crisis. The trust in elites of any variety must be countered by spreading leadership skills and creating networks for decision-making. Strategies must be analysed and discussed widely. Earlier experiences must be studied and digested. Preparations for resisting counter-revolution, such as social defence for resisting aggression, need to be made. And plans for circumventing 'progressive' elites need to be formulated.

Is this approach sufficient to prevent counter-revolution? Can it work? Where should efforts begin? What does it mean in detail? These are hard questions, since so little has been done to build systematically towards self-managing social transformation.