Robert Graham: Anarchy, Hierarchy and Democracy

anarchists assembling in Athens

Inspired by the recent online debate at Center for a Stateless Society (C4SS) on anarchy and democracy, I have been posting some material on anarchy and democracy, to complement earlier posts of material by Errico Malatesta, Luce Fabbri and Murray Bookchin. In 2004, I published an anarchist critique of Bookchin’s theory of “confederal democracy” under the title, “Reinventing Hierarchy: The Political Theory of Social Ecology” (Anarchist Studies, Vol. 12, No. 1). I thought now would be a good time to reproduce some excerpts.


The question which… arises is whether [the] face-to-face [non-mediated] political relationships [advocated by Murray Bookchin] are inherently libertarian and non-hierarchical. Certainly, there are many direct relationships that are neither libertarian nor non-hierarchical, for example master-slave and master-servant relationships, and patriarchal familial relationships. In Bookchin’s proposed community assemblies, it will still be possible for some members of the assembly to engage in domineering and manipulative behaviour. That the members of the assembly will know each other personally is no guarantee against that, as anyone involved in familial relationships can attest.

While manipulative and domineering behaviour may be incapable of elimination from social and political life, Bookchin would argue that the assembly remains non-hierarchical, with each member having equal voice and vote. However, policy decisions will ultimately be made by majority vote. If factions develop, as they invariably do, the very real possibility arises that some people will find themselves in the minority on many issues. Unable to marshal a majority in favour of their policy proposals, and against those of their political opponents, they will find their votes ineffective. This may in turn cause them to cease participating in the assembly or even to rebel against it, due to their lack of real decision-making power.

The majority may very well be placed in the position of having to enforce their decisions against a recalcitrant minority. The minority will have to decide whether to abide by the majority decision or face the consequences of disobedience. In either case, the majority will hold political authority over the minority. Whenever there is a lack of unanimity on a policy decision, or someone later decides the policy was mistaken, a hierarchical relationship will arise. That individual members of the assembly will sometimes be with the majority, sometimes not, does not change the fact that, with respect to the adoption and implementation of majority policy decisions, the majority on a particular issue will be in a position of authority over the minority on that issue. Hierarchical relationships will be created and recreated with every vote.

With respect to the so-called administrative functions to be performed by the various workplace and neighbourhood committees and councils, one of those functions will be the implementation of the majority decisions of the community assembly and, presumably, their enforcement, including the monitoring of compliance by community members with the policies adopted by the assembly. The various committees, councils, boards and tribunals will exercise authority over the individual members, associations and groups comprising the community.

The authority and power relationships between these administrative bodies and the individual members and groups in the community are a kind of hierarchical relationship, even if the alleged legitimacy of the authority and power exercised by these administrative bodies is based on policy-making functions being reserved to the community assembly. The fact remains that these administrative bodies will have the authority and the power to implement and enforce the policies adopted by the assembly, and the individual members and groups in the community will have an obligation to comply with these policies, and to abide by the administrative decisions of the administrative bodies delegated the responsibility of implementing and enforcing them.


Whether administrative bodies can limit their functions to strictly administrative ones, without engaging in any policy-making, is open to question. If administrative bodies must engage, at least to some extent, in policy-making, then one of the central bases for the legitimacy of the authority of the community assembly, namely that all policies are made directly by the members of the community in assembly will be undermined.

John Clark has argued that it is impossible for community assemblies to formulate policies with sufficient specificity `that administrators would have no significant role in shaping policy’ (`Municipal Dreams’, p41). The idea is that in applying general policies to specific cases, administrative bodies are themselves engaging in policy-making by giving general policies specific content. This is similar to arguments that conventional courts, in interpreting and applying the law to specific cases, are in reality creating law, a function that is supposed to be reserved to the legislature.

Clark suggests that administrative power can be kept in check by popular juries and citizens’ committees randomly selected from among the members of the community (`Municipal Dreams’, p42). In contrast, Bookchin has proposed that administrative bodies be kept in check by the community assembly itself (TE; p216).

Even if Clark were right that administrative bodies must engage in policy making at some level, creating yet more administrative bodies to oversee them is not a particularly attractive solution. That will simply create yet another level of political authority with which individual citizens will have to deal. In addition these supervisory bodies will themselves presumably have to be overseen by the community assembly or some other, higher, level of government, in which case the assembly or yet another level of authority will still be faced with what Clark believes to be the impossibly complex task of overseeing all administrative activity (`Municipal Dreams’, p47). Bookchin’s proposal that administrative bodies be overseen directly by the community assembly is at least more democratic.


Both Clark’s and Bookchin’s schemes entail a hierarchical structure of authority. In implementing and enforcing the policies adopted by the assembly, the firsl level administrative bodies endorsed by Bookchin exercise authority over individual community members. In supervising the exercise of this authority, the popular juries and citizens’ committees proposed by Clark exercise authority over the first-level administrative bodies and, indirectly, over the individual community member. In both cases the highest authority, at least at the community level, remains the assembly of all community members based on majority vote.

If individual members of the community are also members of the governing authority, then how can it be said that there is a hierarchy of authority? Bookchin goes so far as to say that `the self that finds expression in the assembly and community is literally, the assembly and community that has found self-expression – a complete congruence of form and content’ (PSA, p 167, fn.). Yet this would only be the case if the assembly always spoke in one voice. However, when decisions are made by majority vote, this often may not be the case. The minority on an issue will be subject to the authority of the majority and to the derivative authority of the administrative and supervisory bodies charged with implementing, interpreting, applying and enforcing the policies adopted by the assembly by majority vote.


The question that naturally arises is whether or not any properly political relationship can be non-hierarchical. It may be that Bakunin was right when he wrote, `whoever talks of political power talks of domination’ (The Anarchist Reader, p109). How is it possible to create political relationships that are truly non-hierarchical? Can there be such a thing as non-hierarchical political authority?

These are questions to which Bookchin has never provided satisfactory answers. To critics of majoritarian direct democracy, Bookchin has responded that the majority `could hardly “dictate” to anyone. The minority would have every opportunity to dissent, to work to reverse that decision through unimpaired discussion and advocacy’ (AMFL, p147). This response ignores the fact that unless and until the minority is able to reverse the decision (thereby creating yet another dissenting minority, unless unanimous agreement is reached), it remains subject to the decision, and the authority, of the majority.

The feminist political theorist, Carole Pateman, has proposed a model of direct, participatory democracy that is non-hierarchical and anti-authoritarian. To give substantive recognition to the freedom and equality of all citizens, Pateman argues, one must give practical recognition to `the right of minorities to refuse or withdraw consent, or where necessary, to disobey’ majority decisions (PPO, p162). Political relationships remain non-hierarchical, because the majority does not exercise institutional power over the minority. The minority is free to decide `whether or not they ought to consent to, or comply with’, majority decisions (PPO, p137). Direct democracy conceived in these terms is compatible with a social ecological and anarchist conception of non-dominating, non-hierarchical community.

Bookchin does not consider this alternative, but appears to believe that the only real alternative to majority rule is decision-making based on consensus, or unanimous agreement. The important difference between consensus-based decision-making and the kind of direct democracy advocated by Pateman, is that only in the former can a `minority of one’ prevent the rest of the community from adopting a policy or deciding on some collective action (Bookchin, AMFL, p147). This does give the dissenters their own kind of de facto authority over the majority because their refusal to consent to a proposal governs the outcome of the decision making process. However, under Pateman’s proposal, the majority can adopt policy and act on it despite minority dissent, although they may decide not to in the face of such dissent. What the majority cannot do is force the minority to obey its decisions, which is different from a minority being able to force the majority act in accordance with its wishes. This kind of political `authority’ does not legitimize the exercise of `power over others’ but rather gives `citizens collective power to, or the ability to, act for themselves’ (PPO, p136).

Bookchin himself proposed a kind of non-dominating authority as a means of undermining the authority of existing, statist political institutions in The Rise of Urbanization and the Decline of Citizenship (RUDC). Neighbourhood assemblies are to elect mandated, recallable delegates to municipal and state councils assembly delegates, creating a parallel moral authority to oversee and influence the legal, civic and state governments (pp271-273). Although these municipal statewide councils of neighbourhood assembly delegates would not exercise and official political power, they would `function as the popular voice of the citizenry articulated into communities rather than anonymous voters’ (p273). Through this process, `governance by legislative command, with its panoply of penalties an coercion, would begin to yield to governance by moral suasion, with its evocation of public responsibility and individual probity’ (p274).

If councils of neighbourhood assembly delegates can, through moral suasion, influence the exercise of political power by existing institutions, then one would think they would be able to exert an even more powerful influence over the individual members of the community for whom the councils would be providing a voice, without resorting to the `panoply of penalties and coercion’ upon which existing political institutions and governments depend. If majority rule is ultimately upheld by the use of coercive sanctions, the focus of political activity will be on mobilizing majority support instead of achieving mutual understanding, cooperation and agreement by rational persuasion. Bookchin’s `vision of community life as an ethical compact’ will be seriously, if not fatally, undermined if the community assembly must ultimately resort to coercive measures in order to maintain its authority (RUDC, p274)…

Robert Graham (2004)

CrimethInc: From Democracy to Freedom

vote for nobody

Last week, I posted a brief section on “community assemblies” from the “Anarchist Current,” the Afterword to Volume Three of my anthology of anarchist writings, Anarchism: A Documentary History of Libertarian Ideas. I raised some concerns regarding proposals for direct democracy that to my mind create structures that are too rigid and will result in a return to political parties and power politics as people coalesce into groups with sometimes conflicting interests (a critique I have more fully developed in my article, “Reinventing Hierarchy: The Political Theory of Social Ecology,”[6] in Anarchist Studies, Volume 12, No. 4 (2004)). Previously, I posted some selections from Malatesta, Luce Fabbri and Murray Bookchin setting forth different views about anarchy and democracy. Coincidentally, CrimethInc. has been running a serious of articles providing an anarchist critique of even directly democratic forms of government. Here, I present some excerpts from the section on democracy and freedom.

democracy means police

Anarchist critiques of democracy

Democracy is the most universal political ideal of our day. George Bush invoked it to justify invading Iraq; Obama congratulated the rebels of Tahrir Square for bringing it to Egypt; Occupy Wall Street claimed to have distilled its pure form. From the Democratic People’s Republic of North Korea to the autonomous region of Rojava, practically every government and popular movement calls itself democratic.

And what’s the cure for the problems with democracy? Everyone agrees: more democracy. Since the turn of the century, we’ve seen a spate of new movements promising to deliver real democracy, in contrast to ostensibly democratic institutions that they describe as exclusive, coercive, and alienating.

Is there a common thread that links all these different kinds of democracy? Which of them is the real one? Can any of them deliver the inclusivity and freedom we associate with the word?

Impelled by our own experiences in directly democratic movements, we’ve returned to these questions. Our conclusion is that the dramatic imbalances in economic and political power that have driven people into the streets from New York City to Sarajevo are not incidental defects in specific democracies, but structural features dating back to the origins of democracy itself; they appear in practically every example of democratic government through the ages. Representative democracy preserved all the bureaucratic apparatus that was originally invented to serve kings; direct democracy tends to recreate it on a smaller scale, even outside the formal structures of the state. Democracy is not the same as self-determination.

To be sure, many good things are regularly described as democratic. This is not an argument against discussions, collectives, assemblies, networks, federations, or working with people you don’t always agree with. The argument, rather, is that when we engage in those practices, if we understand what we are doing as democracy—as a form of participatory government rather than a collective practice of freedom—then sooner or later, we will recreate all the problems associated with less democratic forms of government. This goes for representative democracy and direct democracy alike, and even for consensus process.

Rather than championing democratic procedures as an end in themselves, then, let’s return to the values that drew us to democracy in the first place: egalitarianism, inclusivity, the idea that each person should control her own destiny. If democracy is not the most effective way to actualize these, what is?

As fiercer and fiercer struggles rock today’s democracies, the stakes of this discussion keep getting higher. If we go on trying to replace the prevailing order with a more participatory version of the same thing, we’ll keep ending up right back where we started, and others who share our disillusionment will gravitate towards more authoritarian alternatives. We need a framework that can fulfill the promises democracy has betrayed…


Creating Spaces of Encounter

In place of formal sites of centralized decision-making, we propose a variety of spaces of encounter where people may open themselves to each other’s influence and find others who share their priorities. Encounter means mutual transformation: establishing common points of reference, common concerns. The space of encounter is not a representative body vested with the authority to make decisions for others, nor a governing body employing majority rule or consensus. It is an opportunity for people to experiment with acting in different configurations on a voluntary basis.

The spokescouncil immediately preceding the demonstrations against the 2001 Free Trade Area of the Americas summit in Quebec City was a classic space of encounter. This meeting brought together a wide range of autonomous groups that had converged from around the world to protest the FTAA. Rather than attempting to make binding decisions, the participants introduced the initiatives that their groups had prepared and coordinated for mutual benefit wherever possible.

Much of the decision-making occurred afterwards in informal intergroup discussions. By this means, thousands of people were able to synchronize their actions without need of central leadership, without giving the police much insight into the wide array of plans that were to unfold. Had the spokescouncil employed an organizational model intended to produce unity and centralization, the participants could have spent the entire night fruitlessly arguing about goals, strategy, and which tactics to allow.

Most of the social movements of the past two decades have been hybrid models juxtaposing spaces of encounter with some form of democracy. In Occupy, for example, the encampments served as open-ended spaces of encounter, while the general assemblies were formally intended to function as directly democratic decision-making bodies. Most of those movements achieved their greatest effects because the encounters they facilitated opened up opportunities for autonomous action, not because they centralized group activity through direct democracy.16

Many of the decisions that gave Occupy Oakland a greater impact than other Occupy encampments, including the refusal to negotiate with the city government and the militant reaction to the first eviction, were the result of autonomous initiatives, not consensus process. Meanwhile, some occupiers interpreted consensus process as a sort of decentralized legal framework in which any action undertaken by any participant in the occupation should require the consent of every other participant.

As one participant recalls, “One of the first times the police tried to enter the camp at Occupy Oakland, they were immediately surrounded and shouted at by a group of about twenty people. Some other people weren’t happy about this. The most vocal of these pacifists placed himself in front of those confronting the police, crossed his forearms in the X that symbolizes strong disagreement in the sign language of consensus process, and said ‘You can’t do this! I block you!’ For him, consensus was a tool of horizontal control, giving everyone the right to suppress whichever of others’ actions they found disagreeable.” If we approach the encounter as the driving force of these movements, rather than as a raw material to be shaped through democratic process, it might help us to prioritize what we do best.

Anarchists frustrated by the contradictions of democratic discourse have sometimes withdrawn to organize themselves according to preexisting affinity alone. Yet segregation breeds stagnation and fractiousness. It is better to organize on the basis of our conditions and needs so we come into contact with all the others who share them. Only when we understand ourselves as nodes within dynamic collectivities, rather than discrete entities possessed of static interests, can we make sense of the rapid metamorphoses that people undergo in the course of experiences like the Occupy movement—and the tremendous power of the encounter to transform us if we open ourselves to it.

democracy autonomy

Community Assemblies

A communal assembly in Rojava

A communal assembly in Rojava

Here is a very short excerpt from the “Anarchist Current,” the Afterword to Volume Three of my anthology of anarchist writings, Anarchism: A Documentary History of Libertarian Ideas, in which I discuss Murray Bookchin’s idea of community assemblies, an idea which appears to have been taken up by people in Rojava, under constant threat from both ISIS and Turkish armed forces. The issue for me is whether these assemblies form voluntary federations or whether they become communal or cantonal systems of government.

bookchin next revolution large

Community Assemblies

The contractarian ideal seeks to reduce all relationships to contractual relationships, ultimately eliminating the need for any public political process. Murray Bookchin has argued to the contrary that there is, or should be, a genuine public sphere in which all members of a community are free to participate and able to collectively make decisions regarding the policies that are to be followed by that community. Community assemblies, in contrast to factory councils, provide everyone with a voice in collective decision making, not just those directly involved in the production process (Volume Two, Selection 62). Such assemblies would function much like the anarchist “collectives” in the Spanish Revolution documented by Gaston Leval (Volume One, Selection 126).

Questions arise however regarding the relationship between community assemblies and other forms of organization, whether workers’ councils, trade unions, community assemblies in other areas, or voluntary associations in general. In addition to rejecting simple majority rule, anarchists have historically supported not only the right of individuals and groups to associate, network and federate with other individuals and groups but to secede or disassociate from them. One cannot have voluntary associations based on compulsory membership (Ward: Volume Two, Selection 63).

Disregarding the difficulties in determining the “will” of an assembly (whether by simple majority vote of those present, as Bookchin advocated, or by some more sophisticated means), except in rare cases of unanimity one would expect genuine and sincere disagreements over public policy decisions to continue to arise even after the abolition of class interests. The enforcement of assembly decisions would not only exacerbate conflict, it would encourage factionalism, with people sharing particular views or interests uniting to ensure that their views predominate. In such circumstances, “positive altruism and voluntary cooperative behaviour” tend to atrophy (Taylor, Volume Two, Selection 65), as the focus of collective action through the assemblies becomes achieving coercive legal support for one’s own views rather than eliciting the cooperation of others (Graham, 2004).

Robert Graham

voting fair vote