3. But it is unlikely that world revolution can be achieved in one single blow. This means that the only way in which we can conceive of revolution is as interstitial revolution, as a revolution that takes place in the interstices of capitalism, a revolution that occupies spaces in the world while capitalism still exists. The question is how we conceive of these interstices, whether we think of them as states or in other ways.

4. In thinking about this, we have to start from where we are, from the many rebellions and insubordinations that have brought us to Porto Alegre. The world is full of such rebellions, of people saying NO to capitalism: NO, we shall not live our lives according to the dictates of capitalism, we shall do what we consider necessary or desirable and not what capital tells us to do. Sometimes we just see capitalism as an all-encompassing system of domination and forget that such rebellions exist everywhere. At times they are so small that even those involved do not perceive them as refusals, but often they are collective projects searching for an alternative way forward and sometimes they are as big as the lacandon jungle or the argentinazo of three years ago or the revolt in Bolivia just over a year ago. All of these insubordinations are characterised by a drive towards self-determination, an impulse that says "No, you will not tell us what to do, we shall decide for ourselves what we must do."

These refusals can be seen as fissures, as cracks in the system of capitalist domination. Capitalism is not (in the first place) an economic system, but a system of command. Capitalists, through money, command us, telling us what to do. To refuse to obey is to break the command of capital. The question for us, then, is how do we multiply and expand these refusals, these cracks in the texture of domination.

5. There are two ways of thinking about this.

a) The first says that these movements, these many insubordinations, lack maturity and effectiveness unless they are focussed, unless they are channelled towards a goal. For them to be effective, they must be channelled towards the conquest of state power - either through elections or through the overthrowing of the existing state and the establishment of a new, revolutionary state. The organisational form for channelling all these insubordinations towards that aim is the party.

The question of taking state power is not so much a question of future intentions as of present organisation. How should we organise ourselves in the present? Should we join a party, an organisational form that focuses our discontent on the winning of state power? Or should we organise in some other way?

b) The second way of thinking about the expansion and multiplication of insubordinations is to say "No, they should not be all harnessed together in the form of a party, they should flourish freely, go whatever way the struggle takes them." This does not mean that there should be no coordination, but it should be a much looser coordination. Above all, the principal point of reference is not the state but the society that we want to create.

6. The principal argument against the first conception is that it leads us in the wrong direction.

The state is not a thing, it is not a neutral object: it is a form of social relations, a form of organisation, a way of doing things which has been developed over several centuries for the purpose of maintaining or developing the rule of capital. If we focus our struggles on the state, or if we take the state as our principal point of reference, we have to understand that the state pulls us in a certain direction. Above all, it seeks to impose upon us a separation of our struggles from society, to convert our struggle into a struggle on behalf of, in the name of. It separates leaders from the masses, the representatives from the represented, it draws us into a different way of talking, a different way of thinking. It pulls us into a process of reconciliation with reality, and that reality is the reality of capitalism, a form of social organisation that is based on exploitation and injustice, on killing and destruction. It also draws us into a spatial definition of how we do things, a spatial definition which makes a clear distinction between the state's territory and the world outside, and a clear distinction between citizens and foreigners. It draws us into a spatial definition of struggle that has no hope of matching the global movement of capital.

There is one key concept in the history of the state-centred left, and that concept is betrayal. Time and time again, the leaders have betrayed the movement, and not necessarily because they are bad people, but just because the state as a form of organisation separates the leaders from the movement and draws them into a process of reconciliation with capital. Betrayal is already given in the state as an organisational form.

Can we resist this? Yes, of course we can, and it is something that happens all the time. We can refuse to let the state identify leaders or permanent representatives of the movement, we can refuse to let delegates negotiate in secret with the representatives of the state. But this means understanding that our forms of organisation are very different from those of the state, that there is no symmetry between them. The state is an organisation on behalf of, what we want is the organisation of self-determination, a form of organisation that allows us to articulate what we want, what we decide, what we consider necessary or desirable. What we want, in other words, is a form of organisation that does not have the state as its principal point of reference.

7. The argument against taking the state as the principal point of reference is clear, but what of the other concept? The state-oriented argument can be seen as a pivoted conception of the development of struggle. Struggle is conceived as having a central pivot, the taking of state power. First we concentrate all our efforts on winning the state, we organise for that, then, once we have achieved that, we can think of other forms of organisation, we can think of revolutionising society. First we move in one direction, in order to be able to move in another: the problem is that the dynamic acquired during the first phase is difficult or impossible to dismantle in the second phase.

The other concept focuses directly on the sort of society we want to create, without passing through the state. There is no pivot: organisation is directly pre-figurative, directly linked to the social relations we want to create. Where the first concept sees the radical transformation of society as taking place after the seizure of power, the second insists that it must begin now. Revolution not when the time is right but revolution here and now.

This prefiguration, this revolution here-and-now is above all the drive to self-determination. Self-determination cannot exist in a capitalist society. What can and does exist is the drive towards social self-determination: the moving against alien determination, determination by others. Such a moving against determination by others is necessarily experimental, but three things are clear:

a) The drive towards self-determination is necessarily a drive against allowing others to decide on our behalf. It is therefore a movement against representative democracy and for the creation of some form of direct democracy.

b) The drive towards self-determination is incompatible with the state, which is a form of organisation which decides on our behalf and thereby excludes us.

c) The drive towards self-determination makes no sense unless it includes as its central point the self-determination of our work, our activity. It is necessarily directed against the capitalist organisation of work. We are talking, therefore, not just of democracy but of communism, not just of rebellion but of revolution.

8. For me, it is this second conception of revolution that we have to concentrate on. The fact that we reject the state-centred conception does not mean obviously that the non-state centred conception does not have its problems. I see three principal problems, none of which is an argument for reverting to the idea of taking state power:

a) The first issue is how to deal with state repression. I do not think the answer is to arm ourselves so that we can defeat the state in open confrontation: we would be unlikely to win and anyway it would involve reproducing precisely the authoritarian social relations we are fighting against. Nor do I think that the answer is to take control of the state so that we can control the army and the police forces: the use of the army and police on behalf of the people obviously comes into conflict with the struggles of those who do not want anyone to act on their behalf. This leaves us with trying to find other ways of dissuading the state from exercising violence against us: this may have to involve some degree of armed resistance (as in the case of the Zapatistas), but must surely rely above all on the strength of the integration of the rebellion into the community.

b) The second issue is whether we can develop alternative doings (alternative productive activity) within capitalism, and to what extent we can create an alternative social nexus between activities, other than value. There are many experiments that point in the direction of some sort of solution (the fábricas recuperadas in Argentina, for example) and the possibilities will obviously depend on the scale of the movement itself, but this remains a major problem. How do we think of a social determination of production and distribution that moves from the bottom up (from the interstitial revolts) rather than from a central planning body.

c) The third issue is the organisation of social self-determination. How do we organise a system of direct democracy on a scale that goes beyond the local level in a complex society? The classic answer is the idea of councils linked by a council of councils to which the councils send instantly recallable delegates. This seems basically correct, but it is clear that even in small groups the operation of democracy is always problematic, so that the only way in which direct democracy can be conceived is as a constant process of experimentation and self-education.

9. Can we change the world without taking power? The only way to find out is to do it.