David Graeber: Some Remarks on Consensus

As part of our recent series on Occupy and consensus, we are posting this timely piece by David Graeber, originally published at OccupyWallStreet.net

There has been a flurry of discussion around process in OWS of late. This can only be a good thing. Atrophy and complacency are the death of movements. Any viable experiment in freedom is pretty much going to have to constantly re-examine itself, see what's working and what isn't—partly because situations keep changing, partly because we're trying to invent a culture of democracy in a society where almost no one really has any experience in democratic decision-making, and most have been told for most of their lives that it would be impossible, and partly just because it's all an experiment, and it's in the nature of experiments that sometimes they don't work.

A lot of this debate has centered around the role of consensus. This is healthy too, because there seem to be a lot of misconceptions floating around about what consensus is and is supposed to be about. Some of these misconceptions are so basic, though, I must admit I find them a bit startling.

Just one telling example. Justine Tunney recently wrote a piece called "Occupiers: Stop Using Consensus!" that begins by describing it as "the idea that a group must strictly adhere to a protocol where all decisions are unanimous"—and then goes on to claim that OWS used such a process, with disastrous results. This is bizarre. OWS never used absolute consensus. On the very first meeting on August 2, 2011 we established we'd use a form of modified consensus with a fallback to a two-thirds vote. Anyway, the description is wrong even if we had been using absolute consensus (an approach nowadays rarely used in groups of over 20 or 30 people), since consensus is not a system of unanimous voting, it's a system where any participant has the right to veto a proposal which they consider either to violate some fundamental principle, or which they object to so fundamentally that proceeding would cause them to quit the group. If we can have people who have been involved with OWS from the very beginning who still don't know that much, but think consensus is some kind of "strict" unanimous voting system, we've got a major problem. How could anyone have worked with OWS that long and still remained apparently completely unaware of the basic principles under which we were supposed to be operating?

Granted, this seems to be an extreme case. But it reflects a more general confusion. And it exists on both sides of the argument: both some of the consensus' greatest supporters, and its greatest detractors, seem to think "consensus" is a formal set of rules, analogous to Roberts' Rules of Order, which must be strictly observed, or thrown away. This certainly was not what people who first developed formal process thought that they were doing! They saw consensus as a set of principles, a commitment to making decisions in a spirit of problem-solving, mutual respect, and above all, a refusal of coercion. It was an attempt to create processes that could work in a truly free society. None of them, even the most legalistic, were so presumptuous to claim those were the only procedures that could ever work in a free society. That would have been ridiculous.

Let me return to this point in a moment. First,

1) CONSENSUS IS "A WHITE THING" (OR A MIDDLE CLASS WHITE THING, OR AN ELITIST FORM OF OPPRESSION, ETC)

The first thing to be said about this statement is that this idea is a very American thing. Anyone I mention it to who is not from the United States tends to react to the statement with complete confusion. Even in the US, it is a relatively recent idea, and the product of a very particular set of historical circumstances.

The confusion overseas is due to the fact that almost everywhere except the US, the exact opposite is true. In the Americas, Africa, Asia, Oceania, one finds longstanding traditions of making decisions by consensus, and then, histories of white colonialists coming and imposing Roberts Rules of Order, majority voting, elected representatives, and the whole associated package—by force. South Asian panchayat councils did not operate by majority voting and still don't unless there has been a direct colonial influence, or by political parties that learned their idea of democracy in colonial schools and government bodies the colonialists set up. The same is true of communal assemblies in Africa. (In China, village assemblies also operated by consensus until the '50s when the Communist Party imposed majority voting, since Mao felt voting was more "Western" and therefore "modern.") Almost everywhere in the Americas, indigenous communities use consensus and the white or mestizo descendants of colonialists use majority voting (insofar as they made decisions on an equal basis at all, which mostly they didn't), and when you find an indigenous community using majority voting, it is again under the explicit influence of European ideas—almost always, along with elected officials, and formal rules of procedure obviously learned in colonial schools or borrowed from colonial regimes. Insofar as anyone is teaching anyone else to use consensus, it's the other way around: as in the case of the Maya-speaking Zapatista communities who insisted the EZLN adopt consensus over the strong initial objections of Spanish-speaking mestizos like Marcos, or for that matter the white Australian activists I know who told me that student groups in the '80s and '90s had to turn to veterans of the Maoist New People's Army to train them in consensus process—not because Maoists were supposed to believe in consensus, since Mao himself didn't like the idea, but because NPA guerillas were mostly from rural communities in the Philippines that had always used consensus to make decisions and therefore guerilla units had adopted the same techniques spontaneously.

So where does the idea that consensus is a "white thing" actually come from? Indigenous communities in America all used consensus decision-making instead of voting. Africans brought to the Americas had been kidnapped from communities where consensus was the normal mode of making collective decisions, and violently thrust into a society where "democracy" meant voting (even though they themselves were not allowed to vote.) Meanwhile, the only significant group of white settlers who employed consensus were the Quakers—and even they had developed much of their process under the influence of Native Americans like the Haudenosaunee.

As far as I can make out the ideas comes out of political arguments that surrounded the rise of Black Nationalism in the 1960s. The very first mass movement in the United States that operated by consensus was the SNCC, or Student Non-Violent Coordinating Committee, a primarily African-American group created in 1960 as a horizontal alternative to Martin Luther King's (very vertical) SCLC. SNCC operated in a decentralized fashion and used consensus decision-making. It was SNCC for instance that organized the famous "freedom rides" and most of the direct action campaigns of the early '60s. By 1964, an emerging Black Power faction was looking for an issue with which to isolate and ultimately expel the white members of the group. They seized on consensus as a kind of wedge issue—this made sense, politically, because many of those white allies were Quakers, and it was advantageous, at first, to frame the argument as one of efficiency, rather than being about more fundamental moral and political issues like non-violence. It's important to emphasize though that the objections to consensus as inefficient and culturally alien that were put forward at the time were not put forward in the name of moving to some other form of direct democracy (i.e., majority voting), but ultimately, part of a rejection of the whole package of horizontality, consensus, and non-violence with the ultimate aim of creating top-down organizational structures that could support much greater militancy. It also corresponded to an overt attack on the place of women in the organization—an organization that had in fact been founded by the famous African-American activist Ella Baker on the principle "strong people don't need strong leaders." Stokely Carmichael, the most famous early Black Power advocate in SNCC, notoriously responded to a paper circulated by feminists noting that women seemed to be systematically excluded from positions in the emerging leadership structure by saying as far as he was concerned, "the only position for women in SNCC is prone."

Within a few years SNCC began to splinter; white allies were expelled in 1965; after a brief merger with the Panthers it split again, and dissolved in the '70s.

These tensions—challenges to horizontalism and consensus, macho leadership styles, the marginalization of women—were by no means peculiar to SNCC. Similar battles were going on in predominantly white groups: notably SDS, which ultimately ditched consensus too, and ended up splitting between Maoists and Weathermen. This is one reason the feminist movement of the early '70s, which within the New Left began partly as a reaction to just this kind of macho posturing, embraced consensus as an antidote. (Anarchists only later adopted it from them.) But one point bears emphasizing. It's important. None of those who challenged consensus did so in the name of a different form of direct democracy. In fact, I'm not aware of any example of an activist group that abandoned consensus and then went on to settle on some different, but equally horizontal approach to decision-making. The end result is invariably abandoning direct democracy entirely Sometimes that's because, as here, that is explicitly what those challenging consensus want. But even when it's not, the same thing happens, because moving from consensus sets off a dynamic that inevitably leads in a vertical direction. When consensus is abandoned, some are likely to quit in protest. These are likely to be the most dedicated to horizontal principles. Factions form. Minority factions that consistently lose key votes, and don't have their concerns incorporated in resulting proposals, will often split off. Since they too are likely to consist of more horizontally oriented participants, the group becomes ever more vertical. Before long, those who never liked direct democracy to begin with start saying it's what's really to blame for all these problems, it's inefficient, things would run far more smoothly with clearly defined leadership roles—and it only takes a vote of 51% of the remaining, much more vertical group, to ditch direct democracy entirely.

Obviously, the widespread perception of consensus process as white isn't just be a hold-over from events that took place forty years ago. A lot of the problem is that, since the '70s, consensus process has largely been developed among direct-action oriented groups, and, while there are certainly African-American-based groups operating in what might be called the Ella Baker tradition, most of those groups have been largely white. The reasons are pretty obvious. Those lacking white privilege face much higher levels of state repression, and (unlike, in say, Mexico, or India, where those who face the most repression are generally speaking already organized in semi-autonomous communities that operate at least partly by consensus), in the US, this limits the degree to which it's possible to engage in creating experimental spaces outside the system. Communities face immediate such practical concerns so pressing many feel working outside the system would be irresponsible. Those who don't often feel they have no choice but to adopt either strict, rigorous, MLK-style non-violence, or adopt revolutionary militarism like the panthers—both of which tend to lead to top down forms of organization. As a result, the culture of consensus, the style in which it's conducted, the sensibilities surrounding it, inevitably comes to reflect the white middle-class background of so many of those who have created and shaped it, and the result is that those who do not share these sensibilities feel alienated and excluded. Obviously this is something that urgently needs to be addressed. But the problem here is not with the principles underlying consensus (that all voices have equal weight, that no one be compelled to act against their will), but with the way it's being done—and the fact that the way it's being done have the effect of undermining those very principles.

2) RULES VERSUS PRINCIPLES

I think the real problem here is a misunderstanding about what we're basically arguing about. A lot of people on both sides of the debate seem to think "consensus" is a set of rules. If you follow the rules, you're doing consensus. If you break the rules, or even do them in the wrong order it's somehow not. I've seen people show up to meetings armed with elaborate diagrams or flow-charts for some kind of formal process downloaded from some web page and insist that only this is the really real thing. So it's hardly surprising that other people put off by all this, or who see that particular form of process hit some kind of loggerhead, say "well consensus doesn't work. Let's try something else."

As far as I'm concerned both sides completely miss the point.

I'll say it again. Consensus is not a set of rules. It's a set of principles. Actually I'd even go so far to say that if you really boil it down, it ultimately comes down to just two principles: everyone should have equal say (call this "equality"), and nobody should be compelled to do anything they really don't want to do (call this, "freedom.")

Basically, that's it. The rules are just a way to try to come to decisions in the spirit of those principles. "Formal consensus process," in is various manifestations, is just one technique people have made up, over the years, to try to come to group decisions that solve practical problems in a way that ensures no one's perspective is ignored, and no one is forced to do anything or comply with rules they find truly obnoxious. That's it. It's a way to find consensus. It's not itself "consensus." Formal process as it exists today has been proved to work pretty well for some kinds of people, under some circumstances. It is obviously completely inappropriate in others. To take an obvious example: most small groups of friends don't need formal process at all. Other groups might, over time, develop a completely different approach that suits their own dynamics, relations, situation, culture, sensibilities. And there's absolutely no reason any group can't improvise an entirely new one if that's what they want to do. As long as they are trying to create a process that embodies those basic principles, one that gives everyone equal say and doesn't force anyone to go along with a decision they find fundamentally objectionable, then what they come up with is a form of consensus process—no matter how it operates. After all, it a group of people all decide they want to be bound by a majority decision, well, who exactly is going to stop them? But if they all decide to be bound by a majority decision, then they have reached a consensus (in fact, an absolute consensus) that they want to operate that way. The same would be true if they all decided they wanted to be bound by the decisions of a Ouija Board, or appointed one member of the group Il Duce. Who's going to stop them? However, for the exact same reason, the moment the majority (or Ouija board, or Il Duce) comes up with a decision to do something that some people think is absolutely outrageous and refuse to do, how exactly is anyone going to force them to go along? Threaten to shoot them? Basically, it could only happen if the majority is somehow in control of some key resource—money, space, connections, a name—and others aren't. That is, if there is some means of coercion, subtle or otherwise. In the absence of a way to compel people to do things they do not wish to do, you're ultimately stuck with some kind of consensus whether you like it or not.

The question then is what kind of decision making process is most likely to lead to decisions that no one will object to so fundamentally that they will march off in frustration or simply refuse to cooperate? Sometimes that will be some sort of formal consensus process. In other circumstances that's the last thing one should try. Still, there's a reason that 51/49% majority voting is so rarely employed in such circumstances: usually, it is the method least likely to come up with such decisions.

Think of it this way.

Imagine the city is about to destroy some cherished landmark and someone puts up posters calling for people to meet in a nearby square to organize against it. Fifty people show up. Someone says, okay, "I propose we all lay down in front of the bulldozers. Let's hold a vote." So 30 people raised their hands yes, and 20 people raise their hands no. Well, what possible reason is there that the 20 people who said no would somehow feel obliged to now go and lay in front of the bulldozers? These were just 50 strangers gathered in a square. Why should the opinions of a majority of a group of strangers oblige the minority to do anything—let alone something which will expose them to personal danger?

The example might seem absurd—who would hold such a vote?—but I experienced something almost exactly like it a few years ago, at an "all-anarchist" meeting called in London before a mass mobilization against the G8. About 200 people showed up at the RampArts Social Center. The facilitator, a syndicalist who disliked consensus, explained that another group had proposed a march, followed by some kind of direct action, and immediately proceeded to hold a vote on whether we, as a group, wanted to join as. Oddly, it did not seem to occur to him that, since we were not in fact a group, but just a bunch of people who had showed up at a meeting, there was no reason to think that those who did not want to join such an action would be swayed by the result. In fact he wasn't taking a vote at all. He was taking a poll: "how many people are thinking of joining the march?" Now, there's nothing wrong with polls; arguably, the most helpful thing he could have done under the circumstance was to ask for a show of hands so everyone could see what other people were thinking. The results might even have changed some people's minds—"well, it looks like a lot of people are going to that march, maybe I will too" (though in this case, in fact, it didn't.) But the facilitator thought he was actually conducting a vote on what to do, as if they were somehow bound by the decision.

How could he have been so oblivious? Well, he was a syndicalist; unions use majority vote; that's why he preferred it. But of course, unions are membership-based groups. If you join a union, you are, by the very act of doing so, agreeing to abide by its rules, which includes, accepting majority vote decisions. Those who do not follow the group's rules can be sanctioned, or even expelled. It simply didn't occur to him that most unions' voting system depended on the prior existence of membership rolls, dues, charters, and usually, legal standing—which in effect meant that either everyone who had voluntarily joined the unions was in effect consenting to the rules, or else, if membership was obligatory in a certain shop or industry owing to some prior government-enforced agreement, was ultimately enforced by the power of the state. To act the same way when people had not consented to be bound by such a decision, and then expect them to follow the dictates of the majority anyway, is just going to annoy people and make them less, not more, likely to do so.

So let's go back to Justine's first example,

the first time I saw a block used at Occupy was at one of the first general assemblies in August 2011. There were about a hundred people that day and in the middle of the meeting a proposal was made to join Verizon workers on the picket line as a gesture of solidarity in the hope that they might also support us in return. People loved the idea and there was quite a bit of positive energy until one woman in the crowd, busy tweeting on her phone, casually raised her hand and said, "I block that". The moderator, quite flabbergasted asked why she blocked and she explained that showing solidarity with workers would alienate the phantasm of our right-wing supporters. Discussion then abruptly ended and the meeting went on. The truth was irrelevant, popular opinion didn't matter, and solidarity—the most important of all leftist values—was thrown to the wind based on the whims of just one individual. Occupy had to find a new way to do outreach.

Now, I was at this meeting, and I remember the event quite vividly because at the time I was one of the participants who was more than a little bit annoyed by the block. But I also know that this is simply not what happened.

First of all, as I remarked, OWS from the beginning did not have a system where just one person could block a proposal; in the event of a block, we had the option to fall back on a 2/3 majority vote. So if everyone had really loved the proposal, the block could have been simply brushed aside. While many felt the woman in question was being ridiculous (most of us suspected the "national movement" she claimed to represent didn't really exist), the facilitator, when she asked if anyone felt the same way, was surprised to discover a significant contingent–some, but not all, insurrectionist anarchists–did in fact object to holding the next meeting at a picket line, since they didn't want to immediately identify the movement with the institutional left. Once it became clear it was not just one crazy person, but a significant chunk of the meeting—probably not quite a third, but close (there weren't really a hundred people there, incidentally; more like sixty)—she asked if anyone felt strongly that we should move to a vote, and no one insisted. Was this a terrible failure of process? I must admit at the time I found it exasperating. But in retrospect I realize that had we forced a vote, the results might well have been catastrophic. Because at that point we, too were just a bunch of people who'd all showed up in a park. We weren't a "group" at all. Nobody had committed to anything; certainly, no one had committed to going along with a majority decision.

A block is not a "no" vote. It's a veto. Or maybe a better way to put it is that giving everyone the power to block is like giving the power to take on the role of the Supreme Court, and stop a piece of legislation that they feel to be unconstitutional, to anyone who has the courage to stand up in front of the entire group and use it. When you block you are saying a proposal violates one of the group's agreed-on common principles. Of course, in this case we didn't have any agreed-on common principles. In cases like that, the usual rule of thumb is that you should only block if you feel so strongly about an issue that you'd actually leave the group. In this sense I suspect the initial blocker was indeed being irresponsible (she wouldn't have really left; and many wouldn't have mourned her if she had.) However, others felt strongly. Had we held a vote and decided to hold our next meeting at a picket line over their objections, many of them would likely not have shown up. The anti-authoritarian contingent would have been weakened. Had that happened, there was a real chance later decisions, much more important ones, might have gone the other way. I am thinking here in particular of the crucial decision, made some weeks later, not to appoint official marshals and police liaisons for September 17. Judging by the experience of other camps, had that happened, everything might have gone differently and the entire occupation failed. In retrospect, the loss of one early opportunity to create ties with striking unionists now seems a small price to pay for heading off on a road that might have led to that. Especially since we had no trouble establishing strong ties with unions later—precisely because we had succeeded in creating a real occupation in the park.

There are a lot of other issues that one could discuss. Above all, we desperately need to have a conversation about decentralization. Another point of confusion about consensus is the idea that it's crucial to get approval from everyone about everything, which is again stifling and absurd. Consensus only works if working groups or collectives don't feel they need to seek constant approval from the larger group, if initiative arises from below, and people only check upwards if there's a genuinely compelling reason not to go ahead with some initiative without clearing it with everyone else. In a weird way, the very unwieldiness of consensus meetings is helpful here, since it can discourage people from taking trivial issues to a larger group, and thus potentially waste hours of everyone's time.

But all this will no doubt will be hashed out in the discussions that are going on (another good rule of thumb for consensus meetings: you don't need to say everything you can think to say if you're pretty sure someone else will make a lot of the same points anyway). Mainly what I want to say is this:

Our power is in our principles. The power of Occupy has always been that it is an experiment in human freedom. That's what inspired so many to join us. That's what terrified the banks and politicians, who scrambled to do everything in their power—infiltration, disruption, propaganda, terror, violence—to be able to tell the word we'd failed, that they had proved a genuinely free society is impossible, that it would necessarily collapse into chaos, squalor, antagonism, violence, and dysfunction. We cannot allow them such a victory. The only way to fight back is to renew our absolute commitment to those principles. We will never compromise on equality and freedom. We will always base our relations to each other on those principles. We will not fall back on top-down structures and forms of decision making premised on the power of coercion. But as long as we do that, and if we really believe in those principles, that necessarily means being as open and flexible as we can about pretty much everything else.