Warren Magnusson

My argument is simple. To comprehend the politics of such a global city , we need new categories, quite different from the ones we have inherited from the social sciences. In my view, the conventional categories tend to displace, repress and conceal the political. As such, they inhibit any understanding of global citizenship. We need to begin again, with a different ontology of the political. As Foucault put it, ‘We need to cut off the king’s head: in political theory it has still to be done.’ (Rabinow, 1984, p. 63.) At present, we are caught within a discourse of sovereignty that leads us on a merry chase for the centres of power, and deludes us into thinking that there are commanding heights to be seized and fortified. If not the castle, why not the king? If not the king, why not parliament? If not parliament, why not the nation? If not the nation, why not the economy? If not the economy, why not culture? If not culture, why not the global city? Surely, the centre of power is somewhere. Is that not the expectation that flows from sovereignty-thinking? And, if we break from this pattern of expectation—if we refuse the temptation to model the political universe in terms of highs and lows, centres and peripheries, interiors and exteriors—do we not also have to de-centre the social sciences, cultural studies, and all the other disciplines that make politics into the great impossible?

The naturalization of the city is an effect of social scientific thinking. It is worth remembering that the social sciences were developed in more or less explicit opposition to Aristotelian political science. The Aristotelian approach was bound up with the ideal of a polis or republic : a self-contained community in which people could be active citizens (Barker, 1962). Although this ideal was long preserved in the Western tradition by thinkers like Cicero, Machiavelli and Rousseau, it was challenged effectively by the new ‘social’ scientists of the late eighteenth and early nineteenth centuries, who showed in various ways that ‘political’ self-government was not enough for genuine self-determination. The implication was that people who seemed to have political independence would still be subject to economic and social forces beyond their control. Moreover, these same people would act within the framework of a culture that largely determined what they would do. Thus, from the perspective of the developing social sciences, merely political activity seemed epiphenomenal. Attention had to be directed at the ‘other’ of politics: civil society . The social might be understood in terms of economic relations, biological necessities, cultural norms, or whatever. The point was that it could not be comprehended by politicians or statesmen, let alone by ordinary citizens. Social science—or its ‘other,’ cultural understanding—always trumped merely political science. Indeed, the only kind of political science that deserved the name was one that could be understood as just another branch of the naturalizing social sciences.

Often forgotten now, in the rush to demonstrate the aristocratic, ethnocentric and patriarchal assumptions of the classical political thinkers, is that the Aristotelian and later republican conception of political science was action-oriented. That is, the premise of this way of knowing was that the people who were fit for it had the responsibility for governing. Such people were rulers and had to decide what to do. The ‘science’ of politics was to help them get oriented and then to make the appropriate decisions. Such a science had to encompass philosophy, sociology, psychology, geography, history and so on: it had to be comprehensive, because everything was at stake in the matters at issue and anything might be relevant to the decisions at hand. In this context, political science had to be the study to which all other sciences contributed. The premise, of course, was that the people who used such a master science could and should take responsibility for the human affairs in which they were engaged. If they disclaimed responsibility—if they said that what was happening was determined by fate, or by nature, or by the requirements of God—they would be refusing to act politically. There have been many ideologies—Christianity and liberalism among them—that have encouraged such political passivity. One of the meritorious features of the critical social sciences and critical social theory is that they have exposed the ideological moves involved in such encouragement and shown how the quiescence of some has entrenched the political power of others. On the other hand, there has been much less clarity about the fact that political responsibility, like adulthood, is ultimately thrust upon every one of us. Either we behave like children politically—which is what many social and cultural theorists seem to recommend—or we take responsibility for ourselves and the world around us. If we opt for the latter, a political science of some sort is essential.

Unfortunately, the only sort of political science we have is disabling, insofar as it is focused on the state and on the political field constituted by the state. ‘Policy studies’ are the nearest modern equivalent to Aristotelian political science. The state-centricity of such studies is apparent. To develop a sociology or an economics or a geography or a cultural analysis of politics is not the appropriate alternative, because a move of this sort is vitiated by the de-politicizing assumptions of the discipline invoked. The terms ‘culture,’ ‘geography,’ ‘economy,’ and ‘society’ were developed in opposition to politics. They were meant to denote a reality that was not the effect of politics, but that instead arose more or less naturally out of ordinary human activities. Having thus imagined human life as it would be without politics, we cannot then return politics to its place—as politically engaged writers would like to do—without making nonsense of the ideas that have underpinned the de-politicized social sciences. We cannot throw out ‘the state’ as a focus of our political attention (as some people would no doubt like to do), without also discarding ‘culture,’ ‘economy,’ ‘region,’ ‘nation,’ ‘society,’ and all the other entities that we have invented as alternatives to the polis. In the end, the political is unavoidable, but it is no accident that we have such difficulty writing or even thinking about it, since we work within an intellectual framework that was designed to put politics to one side.

Ultimately, a sociology of politics can only offer a naturalized account, which leaves the action-questions—the political questions—to be decided later. Other disciplinary (and interdisciplinary) analyses have the same effect: specifically political analysis is subject to infinite deferral. Taking a stance, proclaiming an allegiance, or offering an ironic comment is not the same as developing a political analysis. And yet, these are the interventions we pass off as political, in response to the naturalizing accounts that dominate academic (and popular) understandings of human possibility. This is not good enough, especially in an age which is supposed to be one of increasing democracy . Whatever else democracy means, it entails a broadening of political responsibility, in the sense that people who were once conceived as the innocent subjects of government now appear as agents of their own destiny. As political agents, they need ‘to think what they are doing’ (to borrow a phrase from Hannah Arendt) and thus to take responsibility for their own actions. If we have learned anything from two centuries of social science, it is that the most innocent-seeming, de-politicized activities—caring for our children, disposing of our household wastes, purchasing the things we need, arranging for our own security—can have widespread political effects. Thus, to be responsible political agents, we need to think about our lives as a whole, relate what we are doing to the actions of others and consider our individual and collective responsibilities. It is not easy to do any of this well and it will be doubly difficult if we continue to substitute cultural commentary, pious moralizing, and pseudo-scientific description for political analysis.

These issues are extremely complex and cannot easily be analyzed here. However, I want to suggest that the question of the global city brings these issues into view in a particularly helpful way. This is because the globalization of an urban way of life gives a discernible form to the politics of the world as a whole—a form which is analytically familiar in various respects. Although the existing disciplines obstruct the effort to come to terms with this form, we do have resources in urban analysis that could be tapped in an effort to work out a new political science.

The concept of the global city has been most fully developed within the allied disciplines of urban geography and urban sociology (Friedmann and Wolff, 1982, Knox and Taylor, 1995). Geography has long been the least respected of the social sciences, because of the spatial determinism that seems implicit in its concept of human affairs. Urban sociology has been a bastard study, because most of what is important in modern societies is urban and the focus on that feature seems to beg the question of analytical significance. In this context, the concept of the global city is of some importance to both disciplines, because it suggests that there is a spatial form of the urban that bursts the bounds of particular societies. By analyzing the hierarchy of cities in a globalized urban system and giving particular attention to the peak command centres—especially London , New York and Tokyo (Sassen, 1991) —it seems possible to analyze patterns of development and forms of social power in new ways. This elevates the status of the spatial sciences. On the other hand, sociologists and historians who want to put the urban back in its place prefer the concept of ‘globalization ’—a looser formulation that enables them to look at cultural and economic relations, without relating them to patterns of urban development. Sociology has been somewhat embarrassed by its lack of global theory, which has put this discipline at a disadvantage in relation to political science and economics (Robertson, 1992; Waters, 1995). The latter disciplines have the state system and the market respectively to account for world order. Modernity—or, in more radical versions, capitalism—has been the closest analogue to a sociological concept of the global, but it is more useful for analyzing the transition from then to now, than for specifying the dynamics of the present itself. This opens the way for a sort of revenge of the spatial sciences, in the form of a theory of the global city.

Within the existing discipline of political science, the concept of the global city has relatively little purchase, because there is already a powerful theory of the global, in the form of the ‘realist’ account of international relations (Walker , 1993). This account is deeply invested in the old spatial imaginary. It suggests that states are ultimately the dominant actors in global affairs, by virtue of their monopoly over the means of extreme violence, their command over popular loyalties and their legal supremacy in relation to economic, cultural, religious, social and other political institutions. One might well suggest that the dominance of states is not what it appears to be (and indeed this is a major theme in contemporary commentary on international affairs: see, for instance, Ruggie, 1993, or Sassen, 1996). On the other hand, if municipalities are the political organizations on which the global city (or global cities) must rely, it is not clear how they are to match the power of states. Municipalities are conventionally understoods as miniature states, stripped of sovereign authority. How can such stripped-down, miniaturized states be effective, when the organizations on which they are modelled seem to have lost control? An increasingly popular idea is that there is an emergent ‘civil society’ that brings people from different countries together. (See, for example, Lipschutz with Mayer, 1996.) However, to conceptualize world politics on the model of civil society is simply to project a familiar form onto a different scale. (Compare Shaw, 1994, and Walker, 1994.) Something similar happens in current speculations about ‘global governance". (See, for instance, Held, 1995, and Linklater, 1998.) To think differently about world politics is extraordinarily difficult, because we are so used to the idea that there is a centre from which ‘government’ occurs and to which ‘politics’ must relate.

The point of focusing on the urban is to move away from social scientific categories, in the hope of recovering a sense of the political. Used in the way I suggest, the urban is both a spatial and a temporal category—that is, it is both geographical and historical. In its original sense, it denotes a difference between the urban on one side and the rural or simply natural on the other. At any moment, this can be mapped in two or three dimensions. On the other hand, the urban is an historical movement, which can be traced to its origins in the ancient civilizations and followed in its development into a global system which actually encompasses the rural and the natural. The predominance of urbanism as a way of life thus can be shown geographically and accounted for historically. It is important to emphasize that urbanism is not just any way of life, but is a particular form of it, in which people gradually free themselves from their immediate dependence on the natural environment and create for themselves environments of their own making. As we know well, the latter environments are not always pleasant and their making is constrained by the parameters of the natural world. Nonetheless, it is obvious that the world in which we live is increasingly something of our own creation and it is this feature, more than the presence of dense clumps of buildings, that makes urbanism as a way of life so different from what went before.

To think of the urban in this way is to bring us nearer to a political conception of it, for the emphasis is on the process whereby we make the world in which we live. There is an Aristotelian echo here, for what Aristotle noted in The Politics was the fact that the ultimate political questions—the ‘constitutional’ questions that interested him most—related to the problem of producing and maintaining the polis in the form most conducive to human aspirations. What was at stake for the founders of a polis was the creation of a new way of life that could be self-sustaining. Genuinely political issues always related to these problems of founding and sustenance. Significantly, the polis was conceived as a city: that is, as an organization of urbanism as a way of life. As urbanism became self-conscious, it became political in the fullest sense. To organize a city politically was to make a declaration of independence from agrarian, pastoral, or hunter-gatherer life: it was to say that people could create and maintain their own artificial environment , which would be more or less adequate to their own purposes. We may believe, as some ecologists and religious thinkers do, that such an aspiration is a sign of hubris and that people are liable to destroy themselves if they push urbanism to its logical conclusion. However, it is hard to deny the fact that urbanism has become global and that the rural and the natural are quickly being reduced to the status of urban parks.

The nineteenth century social sciences encourage us to think of urbanism as an unintended consequence of things we do for other reasons. Urbanism thus becomes susceptible to naturalistic explanation. A politicized understanding of the urban leads us to refuse this evasion of responsibility. It is true that no one planned the world to be exactly as it is—any more than anyone planned Athens to be exactly as Aristotle found it. Nonetheless, the world in which we live is largely of our own making and we sustain it as such in our daily routines. To take responsibility for that world—as the Athenians took responsibility for Athens—is to take responsibility for urbanism as a way of life. This means attempting to understand the global city —urbanism as a way of life—as a political order that people have created and that they continue to sustain. To appreciate what it is, we have to give up the idea that political order necessarily conforms to the sovereignty principle. The latter idea encourages us to think that there can be no real political order if there is no sovereignty present. Since urbanism does not conform to the sovereignty principle, it appears to be apolitical. But to mistake this appearance for reality is to make a grave theoretical error. To the extent that sovereignty exists, it is always localized and its existence is always in the form of a political claim (Magnusson , 2000). That claim is never uncontested and thus paradoxically sovereignty is always limited. In the wider politics of the global city, sovereignty-claims appear beside other sorts of political claims—rights-claims, property-claims, knowledge-claims, identity-claims and so on—which are not necessarily less important. What is contested politically is actually quite open, for it is always possible for someone to identify a new problem with the existing way of life and to demand fundamental change. That is what ‘new’ social movements are always about.

If urbanism denotes the process whereby we come to create and sustain our own environment —a place in the natural world in which we can live in accordance with human purposes—then the politics of urbanism is the activity by which this creative and sustaining process is put into question. To the extent that we assume responsibility for what we are doing, we become engaged politically and in its broadest sense this engagement is with the phenomenon of urbanism.

It is worth pausing to consider the ‘ism’ in urbanism. The term, ‘urbanism as a way of life,’ comes from Louis Wirth (1938). He contrasted urbanism with industrialism and capitalism and appealed for a sociological understanding of the phenomenon. However, he was a social scientist, looking for social not political reasons. This led him toward the naturalism for which he was ultimately criticized by neo-Marxists like Manuel Castells (1977; see also Smith, 1979, pp. 1–48). Castells attempted to re-think urbanism as an effect of the class relations implicit in capitalism. This led him to the idea of ‘urban social movements’ as modes of resistance to capitalism. His thinking has evolved since, as has the critical analysis of urban political economy, an analysis that he and David Harvey did so much to inspire. In a way, Castells and others have been trying to radicalize and politicize Wirth’s original analysis. In so doing, however, they have pushed Wirth’s original insight into the background. To retrieve that insight, we have to link it not only to the ancient Greek conception of politics as an ongoing activity (Arendt, 1961), but also to the modern notion of a social movement as a challenge to the existing order (Tarrow, 1994).

In a sense, a social movement is the modern equivalent of a polis . If the polis offered to its ancient inhabitants a sense of identity and purpose and constituted their collective existence in a form that enabled them to act together to carry this identity and purpose forward, then the contemporary social movement does something rather similar. There is a crucial difference, however, and this is that a movement is not necessarily tied to a particular place . A movement moves, in every sense of that word (Magnusson , 1997). In this respect, it is akin to urbanism as a way of life. Particular cities, habitations, modes of transportation, forms of cultural expression and patterns of economic activity may be established as this way of life develops, but these fixtures are always temporary. Urbanism is characterized by movement, flux, restlessness. The global city is never finished, because new possibilities are always being discovered. Thus, the politics of urbanism is a politics of movement, akin in this respect to the politics of any other social movement. So, does it make sense to conceive of urbanism as a sort of social movement?

To affix an ‘ism’ to the ‘urban’ as Wirth did is to suggest such a possibility. Then as now, an ‘ism’ could be conceived as an ideological construct which ordered people’s political activity in a movement toward a particular goal. The goal of urbanism is not easy to specify, but if the analysis above is correct we can identify urbanism with the effort to make human life self-dependent—that is, to free human beings from the limitations of their natural existence. There are many aspects to this aspiration and it is subject to constant revision, but we can well see that urbanism in this sense is associated with a broad political agenda. The space of urbanism is the space of the global city , but not in the sense in which the geographers and sociologists have tended to use that term. The global city is not just a particular site like New York or Tokyo . Rather, it is urbanism as a ‘way of life’. The argument here is that we must learn to think of this way of life as a way of organizing ourselves politically. From this perspective, urbanism appears as an over-arching social movement, which constitutes the political hyperspace within which other, more familiar social movements appear.

Every important social movement creates its own political space . It defines ‘us and them’, ‘here and there’, ‘then and now’. It establishes an identity and a set of goals for the people involved and thus sets a direction for their activity. This direction is both spatial and temporal: we are here and want to be there, we were this and want to be that. The political space that the movement defines is the terrain in which it must act if it is to be successful. Its enemies appear in particular locations on that terrain, with particular powers at their disposal. Its friends and potential supporters are in other locations, with powers to be developed and deployed in accordance with the strategic objectives and tactical capacities of the movement. The space of action must be defined if decisions are to be made, but it is impossible to know that space—and in particular to appreciate how it might change—without understanding how it was produced historically. Thus, the political space available to a movement must be interpreted temporally if it is to serve as a space for action. In the end, what a movement shapes out of what it can discover is a space and time for its own activity: a where and when for its own efforts. Movements like socialism, feminism and environmentalism —like liberalism , secular humanism and scientific rationalism before them—create worlds of their own in which the friends and enemies, pasts and futures, homes and aways, opportunities and threats are quite distinctive. It would be a mistake to suppose that each of these movements conforms to the same model.

Urbanism seems to be distinctive in that it is an architectonic movement, which occurs in a more complex political space . The urban is associated with contrary ‘affects’: on the one hand, the excitement of human achievement, human expression and human contact; on the other hand, dismay at the effects of human activity in relation both to individual and social life and to the natural environment . Thus, the urban is rarely posed as an unambiguous ideal. It is rather the form that our way of life has taken in consequence of our efforts to make the world over in accordance with our purposes. To the extent that different purposes are expressed in different social movements , urbanism is the ensemble of these purposes and more generally of the activities that give effect to them. From this perspective, the key social movements are not the ones that bring smiles to the faces of critical social theorists. The really powerful social movements are ones like capitalism and statism, which we can trace to the beginnings of the modern era and which have obviously reshaped the space and time for human activity in fundamental ways. Capitalism and statism—to which one might add the ideological ensemble of secular humanism, scientific rationalism and political liberalism —appear to have been the most vital movements within urbanism as a way of life and to have given contemporary urbanism its most significant characteristics. Most of the other social movements can be understood in terms of their reaction to these hegemonic movements.

Elsewhere I have suggested that we might invoke the concept of ‘hyperspace’ to make sense of the relation between urbanism and other social movements (Magnusson , 1996). I take this term from contemporary physics (Kaku, 1994), in which the currently dominant theory is that space actually has ten, rather than four dimensions. According to this theory, the four-dimensional world in which we live day-to-day is in a sort of giant, expanding bubble on the surface of a six-dimensional space curled up in a tiny ball. The only evidence that the other six dimensions are there is that, without hypothesizing them, we cannot make sense of the relations between the four fundamental forces of the universe. It would be wrong to suppose that the physicists’ concept of an n-dimensional hyperspace could be applied directly in an analysis of our social existence—what would it mean?—but two important ideas can be derived from this physical theory. The first is that there can be separate spatial domains, in which what is outside remains largely unnoticeable. And the second is that there are features of the interior of any domain that are unintelligible except with reference to what is outside. We can apply these ideas to the analysis of social movements. On the one hand, every major social movement seems to form its own political space —positing an identity (workers, women, whatever), developing an appropriate history, situating itself geographically. As a result, no two movements share exactly the same political space and so none can simply read off its future or pull down its lessons from the experience of another. On the other hand, there are parameters for each movement that are given by the character of the exterior space. A movement cannot simply give its political space the character it wants: that depends on what is outside. If the analysis above is correct, the ‘outside’ can best be understood as the ‘inside’ of a ‘global city ’—the hyperspace of urbanism as a way of life.

Arguably, the concept of urbanism that I am advancing is a functional equivalent of older theories of modernity. After all, it was by means of the distinction between the ancient and the modern or, later, between the traditional and the modern, that Western thinkers began to make sense of what distinguished post-Renaissance Europe from what went before and what was outside. The modern was actually a four-dimensional, spatio-temporal concept which was heated up for explanatory purposes at points and times of transition. For good or ill, the concept is now burdened with the assumptions of Western imperialism, which are ethnocentric and presentist. Much of the recent fascination with the concept of the post-modern derives from the frustration of intellectuals who are attempting to think beyond the tradition of modernism that they have inherited. However, the concept of the post-modern simply confirms the spatio-temporal assumptions it was supposed to challenge. (‘We were modern then, but now we are post-modern’.) If we are to break out of this frame, we need to think of our lives in a way that does not burden us again with the same history and geography. The concept of urbanism as a way of life is not to be identified simply with Europe and its effects since 1492. On the contrary, we can see that this way of life has taken a variety of forms and has developed over thousands of years in different parts of the world. What is unique about the present era is that the various urbanisms have been integrated with one another in the context of a developing global city .

It is not hard to see that some elements of the contemporary global city —long-distance trading routes and universal religions, for example—were already in place two thousand years ago (Times Atlas, 1993, pp. 70–3). However, the world was still strongly separated into distinct regions and some regions—the Americas, southern Africa, Australia , the Arctic—were largely isolated from the rest of the world. In any case, the predominance of urbanism in relation to other modes of life was far from complete. Only with the long outburst of European expansionism in the so-called modern era did the various regions of the world become firmly integrated with one another and urbanism become everywhere predominant. In fact, it is only in the last few decades that most countries in Africa, Asia and Latin America have become preponderantly urban. To the extent that this process is as yet incomplete, there is every indication that it will continue until it is complete. People do not like starving in the countryside and they will make what efforts they can to join in the urban prosperity of which they are now aware.

To say this is not to re-invoke a theory of historical inevitability. Rather, it is to call attention to the political choices that are shaping the world in which we live. Elitists that we are, we like to assign a political character to the decisions made by national governments, international agencies and great corporations. We have a little more trouble thinking of a peasant’s decision to move into the city as political. But what else is such decision? It is by no means simply an economic move, since it is vested with social and cultural aspirations. Moreover, it involves an implicit repudiation of one sort of political community in favour of another. The sum of these innumerable political choices is of enormous consequence for the world as a whole. On the other hand, it is difficult if not impossible to call the people concerned to account for the choices they have made, because those choices seem to have been impelled by dire necessity. What we can see from a distance is that the global city is organized in a way that offers only bad choices to the majority of people and gives others enormous power to organize global life in accordance with their own desires. The global city may be de-centred, but it is by no means democratic in its mode of political organization. Indeed, the claim that the world is becoming ‘more’ democratic seems to depend on an extraordinarily narrow conception of democracy , which depends on the illusion that states normally govern human affairs.

If we want to understand the political organization of the contemporary world, we are well advised to look at the cities in which we live. These places are but nodes in the global city , but in their local and regional organization they seem to replicate most features of the global order. Anyone who has studied urban government carefully will tell you that sovereignty is illusory in urban affairs. Of course, there are states that claim authority over cities and intervene regularly in urban affairs, but what becomes apparent on the ground is that states have a very limited capacity to order things as they would wish. In fact, state authority seems to dissipate at the local level, as it spreads through a bewildering array of agencies that overlap the familiar analytical boundaries. Consider a state-funded charitable agency, with a board of directors representative of local elites, a mandate to raise funds from the community, a core of volunteers and professional workers and a set of elaborate procedures for consulting client groups and affected members of the wider community. Is such an agency within the state or in civil society ? Is it economic, social, political, or cultural? Or is it all these things simultaneously? The ambiguity we encounter in this instance is typical: nothing on the ground where political decisions are being made in the local community seems to conform to the state-centric categories of the social sciences. Concrete analysis pushes us towards categorical invention.

One such invention is ‘the metropolis’. The idea of the metropolis was adopted about a century ago to make sense of the fact that there was a difference between the city proper—that is, the municipality with the name concerned or the densely built-up core—and the surrounding suburbs. As the distinction between suburban and rural areas blurred, people began to talk about metropolitan regions to denote the wider area of urban influence. From the beginning it was noticed that, in densely populated areas, one metropolitan region overlapped another, so that rather arbitrary distinctions had to be made for analytical or administrative purposes. Gradually, it was recognized that wider urban systems were national and international and had been so for some time. Thus, the obvious fact about the metropolis is that it is both centripetal and centrifugal: it draws activity towards itself and spins it outward, so that every metropolis interacts with other metropolises in extremely complicated ways. The relations between one metropolis and another cannot be understood as if the two entities were self-contained and things like national boundaries are of only relative significance in determining the pattern of interaction. Every metropolis is characterized by an internal dispersal of political authority—a dispersal which is in part geographic and in part functional, but in the end much more complex. This dispersal does not conform to neat distinctions between state and society, public and private, national and local.

A similar ambiguity is implicit in the older idea of the municipality . Municipalities date from the medieval era and hence from a time when sovereignty had yet to congeal as a principle of personal and political identity. A municipality is a sort of polis , but it is not autonomous. It has come to be regarded as an organ of the state and yet it is always also in civil society . Not surprisingly, it has the legal status of a corporation: an artificial person without sovereignty. Although analysts sometimes treat municipalities as if they were miniature states, such an analytic strategy never works, because municipalities are not self-contained. The municipality is a liminal space. As such, it may be paradigmatic. Modern (or post-modern ) identities are increasingly porous, complicated and ambiguous. If the personal is the political and the political is the personal, then the municipality is emblematic of our condition. It is limited, but never self-contained: always beyond itself, but nevertheless quite particular. Significantly, the organizations we call states are coming to resemble municipalities in their relation to the global economy and to replicate the patterns of municipal politics on a global scale.

If so, some will wonder why global analysis ought not be proceed under the rubric of ‘capitalism’. The so-called triumph of the liberal-democracies is obviously a triumph of capitalism and if there is a world order it is evidently now a capitalist world order. The global economy we now have is not just any economy: it is the one that emerged within the womb of European feudalism and that was established on a world-scale in the context of European imperialism. It might be argued that, once we strip Marxism of its dross, what we have—especially in Marx ’s own writings—is a brilliant historical-geographic analysis of the way the contemporary world-order came to be (Harvey, 1982, 1996). Moreover, we have in Marxist theory an account of the organizational principles of capitalism—the core system of the world order—which helps us to understand (at least in retrospect) why and how the world evolves as it does. What better analysis of the collapse of the Soviet Union could there be than the one we could derive from Marx? Was the Soviet planning system not one that had developed the economy in Russia to the point where the integument of the existing system of production had to be broken? If we were to reinvigorate Marxist analysis by applying it ruthlessly to the utopian socialists of the twentieth century—among whom Lenin, Attlee, and Blair might all be numbered—then it might be possible to produce a convincing account of our world that put the phenomenon of capitalism at its centre.

Capital is a masterpiece of political analysis, in that it exposes as political a set of relationships that had been labelled as natural (or at least as apolitical). What Marx does so brilliantly is to explain how capitalism works as a political system. All subsequent analysis has to begin where he left off. On the other hand, we must be conscious now that an analysis of capitalism as a political system tells us only some of the things we need to know to make sense of the global politics in which we are engaged. A political science that centres itself on the phenomenon of capitalism may be superior in some respects to a state-centric political science, but it replicates many of the unfortunate features of sovereignty-thinking. It misleads us into thinking that our world has a central determinant principle, one that can be comprehended naturalistically and that will somehow give us our political precepts directly. These were the presumptions of the political economy to which Marx was responding and they are presumptions from which he himself never fully escaped.

To conceive of capitalism as instead one of an ensemble of modern social movements is to situate it in a more complex political space . This space is not uniquely defined by the logic of capital or by labour’s response to it. What is incommensurable with that logic—be it religious understanding, ethnic hatreds, feminist aspirations, or eco-centric concerns—is neither simply external nor simply subordinate. To understand this is to recognize that capitalism, like statism, produces a show of sovereignty to which we must respond, but respond in a way that does not reproduce the show’s illusions.

With respect to citizenship, the state is the show we encounter first. It constitutes the political in a particular way, by attributing natural sovereignty to the individual. As Hobbes argued, such natural sovereignty can lead only to the worst of all possible worlds. So, the artificial sovereignty of the state seems necessary to produce social order. However, the global order thus effected turns out to be a place of mutually exclusive artificial sovereignties: hence, it reproduces the problems of violence and disorder on a different scale. According to the received view, both citizenship proper and politics proper are attributes of statehood (Linklater, 1990). States determine who may be citizens and constitute spaces within which citizenship can be practiced. Citizenship thus defines a relation that absolves individuals of responsibility for what happens beyond the borders of their particular states. Political responsibility is state-centred and as such is distinguished from the personal responsibility that each person has for his or her own life. What cannot be assigned either to the personal or to the political is somehow beyond anyone’s responsibility.

The de-politicizing social sciences enable us to analyze human affairs naturalistically and so to evade responsibilities that might otherwise come with citizenship. In deferring to such intellectual practices, we accept the shows of sovereignty as reality. To politicize the global city is to refuse these shows and to accept our responsibility for matters that we can never understand or control completely. Even more than a particular metropolis, the global city is decentred. It is not subject to a single, sovereign authority. There are multiple systems of power at work within it, some of which can be connected with capitalism, others with statism, still others with tribal nationalism, cultural imperialism, religious fundamentalism and so on. If there is a logic to the development of the global city, it can more easily be appreciated in terms of self-organizing systems or chaos theory than otherwise. However, to adopt such theories as a basis for analysis is to retreat once again into naturalism . A political analysis ought instead to return us constantly to the recognition that the city we have created is our own and that what happens in it is our own responsibility. The global city is both the venue and the product of our own struggles to become what we would like to be, and in the end there is no alternative but to take responsibility for what we have created.