The Democracy Project

by David Graeber

  Similarly, declaring members of an informal leadership clique to be members of a “coordinating committee,” but allowing everyone else to decide whether to reappoint them every six months or so, does not make them “more accountable,” as is often suggested (contrary to all experience); it clearly makes them less. One might well ask why anyone would imagine otherwise.‡

  Q: I’ll allow that consensus works well enough in a small group or neighborhood or community where everyone knows one another, but how can it work in a large group of strangers where there’s no initial foundation of trust?

  A: We shouldn’t romanticize community. True, people who have lived together all their lives in, say, a rural village are more likely to share perspectives than those who live in a large, impersonal metropolis, but they are also more likely to be bitter enemies. The fact that they can nonetheless come to consensus is a testimony to humans’ ability to overcome hatred for the sake of the common good.

  As for meetings between strangers: if one just assembled a random group of people off the street and forced them to attend a meeting against their will, probably they would be unable to find much common ground (other than in forming a plot to escape). But no one comes to a meeting of their own free will unless they want to get something out of it, a common goal everyone is there to achieve. If they don’t get sidetracked and constantly bear in mind what they came for, they can, generally speaking, overcome their differences.

  Q: If you have a fallback on a 66 percent, or 75 percent, or even 90 percent vote in larger meetings, why call this “modified consensus”? Isn’t that just a supermajority voting system? Why not just be honest and call it that?

  A: It’s not actually the same thing. What’s crucial to consensus is the process of synthesis, of reworking proposals to the point where the largest possible percentage of participants likes it, and the smallest percentage objects. Sometimes in larger groups you will find that despite this someone will block, and there will be fundamental disagreements about whether that block is a genuine expression of the group’s basic principles. In that case you have the option of going to a vote. But as anyone who has actually sat through a meeting based on, say, two-thirds voting can attest, if you just go to a vote immediately, the whole dynamic will be different because there is never the presumption that everyone’s perspective is equally valuable. Anyone whose views would appear to represent less than a third of the people of the meeting can simply be ignored.

  Q: What to do if people abuse the system?

  A: There are people who are, for whatever reasons, too damaged or disturbed to take part in a democratic assembly. There are others who can be accommodated, but who are so disruptive and difficult, who demand such constant attention, that indulging them would mean devoting so much more time to their thoughts and feelings than those of everyone else in the group that it undermines the principle that everyone’s thoughts and feelings should have equal weight. If a person is continually disruptive, there should be a way to ask that person to leave. If they refuse, the next step is generally to reach out to their friends or allies to help convince them. If that’s not possible, the best approach is to make a collective decision to systematically ignore them.

  Q: Isn’t the insistence on consensus stifling of creativity and individuality? Doesn’t it promote a kind of bland conformity?

  A: Yes, if done badly. Anything can be done badly. Consensus process is often done very badly. But this is mostly because so many of us are new to it. We’re effectively inventing a democratic culture from scratch. When done right, there’s no other process so supportive of individualism and creativity, because it is based on the principle that one should not even try to convert others entirely to one’s point of view, that our differences are a common resource to be respected, rather than an impediment to pursuing common goals.

  The real problem here is when consensus is a decision-making process by groups that are already based on sharp inequalities of power (either recognized or not) or that already have a culture of conformism—to take an extreme example, the way consensus is practiced within a Japanese corporation, or even an American one like Harley-Davidson. In cases like this, there’s no doubt that demanding “consensus” can make all this even worse. But in cases like this we’re not really talking about consensus at all, in the terms being laid out here, but rather, forced unanimity. There is no more effective way to destroy the radical potential of such democratic procedures than to force people to pretend to use them when actually they’re not.

  Q: Is it reasonable to expect people to constantly attend fourteen-hour meetings?

  A: No, it is completely unreasonable to expect that. Obviously no one should be forced—even by moral pressure—to attend meetings they don’t want to. But neither do we want to divide into one class of leaders who have time to attend long meetings, and another class of followers who never get to weigh in on key decisions. In traditional societies that have been practicing consensus for centuries, the usual solution is to make meetings fun: introduce humor, music, poetry, so that people actually enjoy watching the subtle rhetorical games and attendant dramas. (Here again, Madagascar provides my favorite example. The kind of rhetoric deployed in meetings is so appreciated there that I have seen particularly skilled orators come out and perform it as a form of entertainment between sets by rock bands at music festivals.) But of course these are societies where most people have a lot more time on their hands (not to mention don’t have TV or social media to distract them). In a contemporary urban context, the best solution, when one is not at a moment of initial ferment when everyone is thrilled to be taking part at all, is simply not to have fourteen-hour meetings. Be assiduous with time limits: allocate ten minutes for this item of discussion, five for that, no more than thirty seconds for each speaker. Constantly remind speakers there’s no need to repeat what someone else has said. But most important, do not bring proposals before a larger group unless there is a compelling reason. This is absolutely essential. In fact it’s so important I will give it an entire section of its own.

  DO NOT SUBMIT A PROPOSAL FOR CONSENSUS UNLESS THERE IS A COMPELLING REASON TO DO SO

  Consensus process only works if it is combined with a principle of radical decentralization.

  I really can’t stress this enough. If there is any silver lining to the cumbersome nature of formal consensus process, it’s precisely this: that it discourages people bringing proposals before a General Assembly or Spokescouncil or other large group unless they really have to. It’s always better, if possible, to make decisions in smaller groups: working groups, affinity groups, collectives. Initiative should rise from below. One should not feel one needs authorization from anyone, even the General Assembly (which is everyone), unless it would be in some way harmful to proceed without.

  Let me give an example.

  At one point when we were still meeting in Tompkins Square Park, before the actual occupation began, the Outreach group came very close to quitting en masse when they proposed a two-line description of the nature and purposes of the Occupy Wall Street group to be used in flyers, only to see it blocked at General Assembly. The woman who was point person for Outreach at the time could barely disguise her exasperation, and finally sought me out—as the presumed process maven—to see if there was some way to mediate. I thought about it for a moment and asked: “Well, why did you bring the text to the group at all?”

  “Because I figured it would be better to have everyone approve the way they were being described to others. But it looks like whatever language we come up with, no matter how minimal, someone will object to it. I mean, that was a really unobjectionable statement we came up with!”

  “Are you sure they’re not objecting to the fact that you brought it before the group at all?”

  “Why would they do that?”

  “Well, okay, let’s think about it this way. You’re the Outreach group. You’re a working group that has been empowered by the General Assembly to do outreach. Well, I
guess you could argue, then, if you are empowered to do outreach, you have therefore also been empowered to do those things it is necessary to do in order to do outreach. Like, say, coming up with some way to describe the group. So I don’t think there’s any real reason you should have to ask the group’s approval unless you think there’s something sufficiently controversial in there that you want to check. I mean, I wasn’t there: was it controversial?”

  “No. I thought if there was a problem it was because it was actually pretty bland.”

  Which is what happens if you think you need approval for anything.

  After this conversation I hunted down the person who had originally blocked the language and discovered that he completely agreed with my assessment. He blocked because he wanted to establish that working groups should decide such matters for themselves. (The main problem, then, wasn’t a difference about process at all; it was that the blocker hadn’t properly explained it.)

  As a general rule of thumb: decisions should be made on the smallest scale, the lowest level, possible.§ Do not ask for higher approval unless there’s a pressing need to. But when does a need become pressing? What are the criteria for deciding who really ought to have the opportunity to weigh in on a question, and who doesn’t?

  It you look at it this way, much of the history of radical thought—and particularly radical democratic thought—turns on exactly this question. Who gets to make the decisions and why? It has largely taken the form of a debate between two principles: one is often referred to as workers’ self-organization, or just workers’ control; the other we can just call direct democracy.

  In the past the concept of workers’ control has most often been applied, as its name suggests, to the organization of workplaces, but as a basic principle, it can be applied anywhere. The basic principle boils down to the idea that anyone actively engaged in a certain project of action should be able to have an equal say in how that project is carried out. This is the principle, for instance, that lies behind theorist Michael Albert’s proposed system of participatory economics (or parecon), which tried to answer the question: what kind of work organization allows for a genuinely democratic workplace? His answer was “balanced job complexes”—organizations in which everyone would have to do a certain share of physical, mental, and administrative labor. The basic idea of workers’ control is that if you’re involved in a project, you should have an equal voice in how it’s executed.

  The second principle, that of direct democracy, is that everyone affected by a project of action should have a say in how it is conducted. Obviously the implications here are quite different. If such a notion were formalized, it would lead to some form of democratic communal assemblies in order to gather together the opinions of everyone who had a stake in the project. But things need not become so formalized. In many circumstances it’s probably critical that they aren’t. In Madagascar, where people have been operating by consensus for a very long time, there is what is called the “fokon’olona” principle, which is hard to translate, since sometimes it’s translated as “public assembly” and sometimes just as “everybody.” French colonists tended to assume that fokon’olona were local political institutions that could be turned into extensions of their administration; later Malagasy governments often attempted to make them the grassroots cells for local democracy. It never really works, and it’s largely because these aren’t formal bodies at all, but assemblies brought together around a particular problem—resolving a dispute, distributing irrigation water, deciding whether to build a road—uniting anyone whose lives are likely to be affected by the decision made.

  While there are some who have tried to present these two principles—direct democracy and workers’ control—as a stark choice, a truly democratic society would likely have to rely on a combination of both. If there were a paper mill in a small town, there’s no reason everyone in the town whose life was somehow affected by the mill would need or want to weigh in on its vacation policies; there’s every reason they might wish to be consulted on what the mill was pouring into the local river.

  In the case of an activist group, when we ask this question, we are really asking about the role of working groups. Every Occupy General Assembly had these; by November 2011, the New York City General Assembly already had over thirty of them. Some were permanent and structural: Media, Facilitation, Housing, Accounting, Direct Action. Some were permanent and thematic: Alternative Banking, Ecology, Transgender Issues. Some were organized around specific actions or campaigns and might therefore be either permanent or temporary: Occupy Foreclosed Homes and Oakland Solidarity March would be examples. Action working groups will themselves tend to have their own structural working groups: media, outreach, transportation, and so forth.

  Working groups are created by the General Assembly or larger group, in order to fulfill a specific task or carry out some kind of work: research, education, whatever it may be. Sometimes this happens because there’s a generally recognized need (“Is anyone willing to take responsibility for sanitation issues for the camp?”), sometimes because a group of people has an idea (“Some of us want to create a group to think about how sanitation systems would work in an egalitarian society”). The New York City GA operates by the principle that anyone wishing to create a working group needs to assemble at least five initial members and submit the request to the group. Some requests have been blocked.

  Anyone is free to meet in a room and discuss anything they like, of course: what the GA is doing when it approves a working group is empowering it to act in the name of the GA. It’s basically a form of delegation. It doesn’t create vertical hierarchies because working groups are open to anyone. In fact, when a General Assembly or action planning meeting breaks out into working groups during the course of a meeting, this is actually a way of ensuring no one takes on too much influence, since it’s physically impossible to take part in more than one at the same time. In principle, even point people, who volunteer to be the person to contact if you want to reach members of the working group, ought to be rotated. At Spokescouncils, where only one “spoke” from each working group can take part in the formal discussion (other members are encouraged to attend and whisper in his or her ear or consult discreetly), no one can speak for the same group twice in a row. Still, once the work has been divvied up, or once an existing group has been authorized to pursue some project, there comes to question of how often one needs to check back for approval. The general rule really ought to be: only when it’s obvious it would be wrong to do otherwise. If there’s any reason to doubt you have to check, just go ahead and do it, you probably don’t.

  DIRECT ACTION, CIVIL DISOBEDIENCE, AND CAMPING

  The original inspiration of Occupy Wall Street was the tradition not just of direct democracy, but of direct action. From an anarchist perspective, direct democracy and direct action are—or ought to be—two aspects of the same thing: the idea that the form of our action should itself offer a model, or at the very least a glimpse of how free people might organize themselves, and therefore what a free society could be like. In the early twentieth century it was called “building the new society in the shell of the old,” in the 1980s and 1990s it came to be known as “prefigurative politics.” But when Greek anarchists declare “we are a message from the future,” or American ones claim to be creating an “insurgent civilization,” these are really just ways of saying the same thing. We are speaking of that sphere in which action itself becomes a prophecy.

  The original conception of OWS reflected this anarchist sensibility in several different ways. Most obviously, the refusal to make demands was, quite self-consciously, a refusal to recognize the legitimacy of the existing political order of which such demands would have to be made. Anarchists often note that this is the difference between protest and direct action: protest, however militant, is an appeal to the authorities to behave differently; direct action, whether it’s a matter of a community setting up an alternative education system or making salt i
n defiance of the law (an example from Gandhi’s famous salt march), trying to shut down a meeting or occupy a factory, is a matter of proceeding as one would if the existing structure of power did not exist. Direct action is, ultimately, the defiant insistence on acting as if one is already free.

  (Everyone is perfectly well aware the power structure does exist. But acting this way denies any moral authority to their inevitable, usually violent, response.)

  The refusal to ask for permits was in the same spirit. As we kept pointing out to one another in the early days of meeting in Tompkins Square Park, New York codes are so restrictive that any unpermitted assembly of more than twelve people in a public park is technically illegal (it’s one of those laws that are never actually enforced except against political activists): therefore, even our meetings were at the very least a form of civil disobedience.

  This raises another important distinction: between civil disobedience and direct action, which is often, mistakenly, thought to be simply a difference of militancy (civil disobedience is assumed to be a matter of blockading things, direct action, of blowing them up). Civil disobedience means refusal to comply with an unjust law, or legally valid, but unjust, order. As such, an act of civil disobedience can also be a direct action: as when one, say, burns one’s draft card, on the principle that one would not have draft cards in a free society, or insists on one’s right to be served at a segregated lunch counter. But an act of civil disobedience does not have to be a direct action, and ordinarily acts of civil disobedience do not question the legal order itself: only specific laws or policies. In fact they often explicitly aim to work within that legal system. This is why those engaged in civil disobedience so often welcome arrest: it allows them a platform to challenge the law or policy either legally or in the court of public opinion.