the assumption of responsibility for the welfare of the world











2011 December 5


An analysis of Michael Mann’s ‘The autonomous power of the state’


A state of the modern, Westphalian era has two key features, to Michael Mann: territoriality and centrality. Mann finds that, taken together, these features explain why the state is conjured into existence in the first place, and why it is kept around, or, following his assessment, basically keeps itself around. We can begin with his two forms of power, despotic and infrastructural. Despotic power is the ability to do as one pleases, essentially; to act without external restraint. Infrastructural power, while presented as a sister to despotic power, is in fact a very different understanding of the term ‘power’. No one is stopping me from flying, for example, but I physically can’t fly nonetheless. Infrastructural power is an element of the physical ability to do something; for a state, Mann’s concern, infrastructural power is the entirety of physical ability.

To oversimplify, perhaps, centrality is an infrastructural feature of the state, and territoriality is, more than not, a despotic feature. When Robert Jackson speaks of “positive sovereignty” and “negative sovereignty”, he is capturing much the same contrast. A state has positive sovereignty when it has control over its territory; it has negative sovereignty when, by diplomatic consent, no other (in-system) force will presume to hinder its exercise of control over its (recognized) territory. The latter is a gift from the system, and allows the juridical state the freedom to develop without competition from stronger forces (better-established states, that is). But, as Jackson and Herbst separately point out, there are many examples, particularly in Africa, where the juridical state has very little beyond negative sovereignty. Herbst has a particular interest in infrastructural power, in his case roads, because without it, the state remains a fiction, an empty space in which what actually takes place is far from what is presumed to be taking place. In any case, the territoriality of the Westphalian state is a prerequisite to negative sovereignty, to that element of despotic power that depends on the non-interference of other (empirical) states. The juridical state is allowed a zone of free action, that zone contained within the recognized borders.

A state is centralized when authority, the ability to make decisions and to see those decisions carried out, is concentrated in one institution, usually in one place. It presumes, in Mann’s accurate telling, that other, smaller institutions, as well as individuals, originally possessed that power. These entities created the central authority because to do so ultimately brought more power or greater benefits. The catch is that the state, once so constituted, became indispensable, and therefore autonomous.

Anyone who has observed a parliamentary body in action, say the US Senate, or even who has lived through an effort among friends to decide on a restaurant, can confirm the value of centralized despotic power. Large groups of independent individuals make decisions only slowly, or perhaps not at all. There are too many interests, too many nuances, at least as many opinions as there are members of the group (and sometimes many more), and in the end no decision is made. By contrast, a despot can act quite decisively. If the restaurant decision is delegated to one member of the group of friends, it is possible that many will not like the choice of restaurant; but they will, at least, get to eat. Interest groups within the state follow the same reasoning. Beyond that, they recognize that, even were they possessed of consensus on major decisions, the implementation of those decisions would still be more efficient and effective with a single authority — both because implementation may require countless smaller decisions, and because implementation may require infrastructural power that no interest group alone possesses, and that can only be acquired by pooling resources.

Mann makes the same mistake that Paul Hirst does, when contemplating the success of the Westphalian state. Each asserts that the Westphalian state succeeded at the expense of other forms of polity — they offer the examples of bishoprics and the Hanse, to name two. They do this only by means of hindsight, and by committing a fallacy of classification. In essence, they observe the historical struggle for power, wait for the outcome, categorize the winners in one group and the losers in another, and then pronounce that the winners succeeded by virtue of belonging to a particular group, rather than, as is the case, belonging to a particular group by virtue of winning. To presume that all Westphalian states are the same — that they share commonalities that the polities of any given pre-Westphalian era did not share — is wrong. Our own era has seen centralized states, federations, actual monarchies, constitutional monarchies, city-states, theocracies, and a union of soviet socialist republics. The counterargument must take these differences as nominal, while insisting upon the real (not nominal) differences among polities past. But the former and latter are both exaggerations. The USSR truly was at one point a collection of independent local revolutionary councils (“soviets”). Many a bishopric of medieval Europe had genuine territorial control.

This critique of one of Mann’s minor points only strengthens his main point, that centralized territoriality is a proven system of government. The historical reality is that an organization that successfully mastered control over the resources — human and natural — of a given territory, and that pooled power and authority in a central body, was much better positioned to succeed than one that did not. However, it is possible to interpret both centrality and territoriality, as well as the supposedly-resultant autonomy, in ways that historical examples defy. For example, the Roman Republic lasted centuries without the despotic power later developed under the principate (“empire”) of Augustus. It possessed an institution, the dictatorship, that allowed for a temporary delegation of decision-making authority to a central figure, and that institution was invoked numerous times without transitioning to a permanent status, as Mann’s autonomy theory would expect. The Swiss Confederation and the United Arab Emirates are treated as single states, but have been for much of their existence sufficiently loose that control (“sovereignty”) should properly be identified with the components, not the system (much like the European Union, for now). Even the United States, centralized in so many ways, leaves powers to the “states” that few other Westphalian entities would allow to their provinces, such that ‘state’ is not, after all, that anachronistic a designation. And yet Switzerland and the Emirates are stable, successfully-defended members of the international system, and the US, of course, is the most powerful political entity in the world.

If territoriality is taken in a negative Jacksonian sense, then it is clearly not a requirement for the power of the state. Indeed, the crucible of war, of struggle against other states who do not concede a right to rule a particular territory, has the effect of strengthening the state, considered vis-à-vis other states, and vis-à-vis its own citizens and interest groups. In this latter sense, the state is strengthened in its autonomy, as Mann would surely agree. But if territoriality is taken in a positive Jacksonian sense, then it is reasonable to ask whether non-territorial government is even possible. My own view is that there is but one example, a nomadic ransom group, so that through nomadism, it does not maintain a fixed territory, and through ransom, say, by holding hostage specific individuals, it manages to control a certain population which values those hostages more than its own freedom. I cannot imagine another way in which an institution can control individuals — serve as their government — without controlling the land on which they live.*


* And most individuals do live a sedentary existence. But if the subjects of this hypothetical government are themselves nomads, there is no way to rule them short of ransom, because they are perfectly free to leave the group, and as nomads quite capable of it as well.


Jeffrey Herbst. ‘States and power in Africa’. Princeton University, 2000.
Paul Hirst. ‘Space and power: politics, war and architecture’. Polity, 2005.
Robert H. Jackson. ‘Quasi-states: sovereignty, international relations and the Third World’. Cambridge University, 1990.
Michael Mann. ‘The autonomous power of the state’, p185-213 in the Archives Européennes de Sociologie (European Journal of Sociology), v25 (1984).



Home of the Stewardship Project
and O.T. Ford