This Research Bulletin has been published in ProtoSociology, 20, (2004), 30-45.
Please refer to the published version when quoting the paper.
Social science was conceived and created within modernity and has continued, by and large, to be constituted of discourses that attempt to explain, interpret or understand the modern experience. It is a modern knowledge par excellence: its origin-myth is that the methods of the natural sciences can be used to understand and improve society. And two key trajectories of this modern society, not found in other societies, have been the extraordinary increasing powers of states, and the ever-increasing rise in urbanization that has recently made city dwellers a global demographic majority. Quite simply, modern states and modern cities are unprecedented in their political capacity and economic production respectively.
In this paper I explore how cities and states relate to each other as social collectivities. It might be thought that this question would be a well-worn one within social science but surprisingly this is not the case. There are, of course, very large literatures on each collectivity but their relationship is under-researched. It seems that, post-founding fathers, social-science specialization into disciplines left cities and states located in separate intellectual camps: cities came under the purview of sociologists and geographer-planners, while states were the special subject of political scientists and economists. It has taken the current exigencies of globalization to bring my question to the fore again: today it is seriously asked whether cities, viz global city-regions, are replacing states as the prime organizational nexus of social change (Scott 2001).
I do not think that cities are replacing states but I do believe that the evidence for fundamental changes in the relationships between the two has become overwhelming. This paper will attempt to justify this claim. To achieve this aim I take a particular perspective on both of my subjects by focussing on their spatialities. The concept of spatiality derives from the rejection of two traditional views on the role of 'space' in social change: first, that it is merely an inert platform upon which social events take place, and second, that it is a simple context for events, an 'active' locale that 'influences' social action but is still divided from it. Spatiality is the concept of the alternative idea that the spatial and the social cannot be treated separately: we make our own geographies as well as our own histories. Thus spatiality, like historicity, is inherent in all social activities; all social institutions have their spaces and times, and therefore cannot be 'abstracted' as if space and time are simply the metrics we use to keep track of the where and when of life. Spatiality can be defined as the materially-constructed, socio-spatial orders of life that are routinely produced and reproduced by people carrying out their everyday pursuits. It is because cities and states exhibit quite distinctive, indeed contrasting, spatialities that this perspective is chosen as a lever to understand their changing relationship.
The paper is divided into two related arguments: development of a spatiality perspective on cities and states through identifying types of material spaces, and application of this perspective to evaluate contemporary and near-future world-spatialities. The arguments are separated by an interlude; the latter acts as an intellectual bridge by summarising my revisionist world-systems position that has been presented more fully elsewhere.
1. TYPES OF MATERIAL SPACE AND SOCIAL SCIENCE
Socio-spatial formations are constituted as two types of space. These are spaces of places and spaces of flows (Castells 1996). Thus social space is comprised both of processes operating in a specific place to make it distinctive and, perhaps, provide it will an identity, and by processes that operate across places to create a coherent social formation of connections. These two sets of processes are not exclusive to each other; they operate simultaneously to produce their spatial outcomes in complex social packages. Nevertheless it is analytically useful to distinguish between them.
Spaces are created by social practices. Thus they are not universal, they are always historically specific. Spaces become socially important when they are constituted by myriad social practices. In this form they define a material spatiality of life. All spatialities are products of social agents: activities by people operating as the makers of spaces. There are makers of spaces of places who create worlds of 'local' identities, and there are makers of spaces of flows who create worlds of connections. Through these social practices a spatiality is formed in which spaces of places and spaces of flows mutually co-exist but where one may be dominant. Thus we can contrast the dominance of flows over places in the current Kondratieff B-phase (neo-liberal globalization) with the dominance of places over flows in the previous B-phase (1930s nationalist protectionalism). These represent two distinctive world-spatialities half a cycle apart.
1.1 Making Spaces of Places in the Modern World-system
Making a place is to give a particular segment of space a social meaning. Here I focus on places that provide an encompassing identity to their inhabitants. I use the concept of home to describe the result of combining place with identity. Home implies a place where a person feels comfortable, literally feels 'at home': disruption of home is therefore socially catastrophic. I treat home as a construction that exists at different geographical scales: with no suffix it refers to the locale of modern households, otherwise it can be 'home neighborhood', 'home village', 'hometown', 'home province' (home state in the USA e.g. Californian or Texan), homeland (referring to the territory of one's state or country), and culminating with contemporary environmentalists who try and persuade us of Earth as the (only) home of humanity, our 'home planet' (Taylor 1999). In this discussion I will focus on one key scale - the makers of spaces that are the places we call homelands because they are the ultimate reproducers of nation-states.
In early modern Europe states were largely dynastic in character and therefore were identified with a 'royal house' that supplied its rulers. This was a territorial solution to the gross political instability consequent upon the Reformation and Counter-Reformation. Codified in the Treaty of Westphalia (1648), state subjects were required to adhere to the religion of their sovereign. This created a mosaic of politico-religious spaces, sovereign territories enclosing jurisdictional places. These were states created by 'high politics' (through dynastic marriages and succession wars) that were, once military security was established, only of marginal relevance to the everyday lives of ordinary people. But these remote dynastic states, the first state form of the modern world-system, did not survive the exigencies of ever-changing modernities. As socio-territorial containers, modern states developed functions beyond the initial politico-security imperative to encompass, first, economic functions (mercantilism), second, cultural identities (nationalism), and third, social needs (welfare state) (Taylor 1995). These state developments cumulatively produced what Jessop (2002) has termed the 'Keynsian Welfare National State' of the twentieth century. In other words the original 'high politics' had to cede ground to a new 'low politics' that addressed the lives of ordinary people. This is the basis of the common conflation of nation (which is a cultural grouping) with state (which is a political apparatus): the harnessing of the concepts together as nation-state confirms their indelible combination in the contemporary modern state. It is with the nation-state that the world-spatiality of modernity becomes dominated by spaces of places.
The high politics of states produced the containers in which places were nationally constructed as homelands. Through a variety of state institutions from nineteenth century 'national education' through to twentieth century 'national media', there is a powerful socialization process that nationalizes people into the state. Thereafter their behaviors - working (providing income taxes for states), shopping (providing sales taxes for states), TV watching (receiving news about states), driving (licensed and regulated by states), voting (providing leaders for states), etc. - create a 'national world' in which person and state are entwined. This is consolidated by cultural celebrations that 'bring the nation together'. The state-as-nation becomes the dominant lens through which the social world is viewed and constructed: it is a world mosaic, as depicted by that most familiar cartographic product, the world political map. Nation is imbued so deeply that this map has come to be seen as 'natural': how else would humanity be divided? The fact that this mosaic is a relatively recent modern artefact is lost in the nationalist myths of multiple special, sometimes 'chosen', peoples. Thus, Anderson's (1983) famous dictum, that, in later modernity, 'nation' has replaced early modern religion as the prime identity for people locating their place in the Cosmos. The result is a mosaic world-spatiality.
1.2 Making Spaces of Flows in the Modern World-system
The making of spaces of flows involves the intersection of two well-known stories: an economic one about the spread of markets, and a technological one about the 'shrinking' of distance. Thus the cultural and political focus of the last story is replaced by the doings of economic and financial agents and the consumption of their goods within a changing context of transport and communication.
It is the business of merchants to create spaces of flows. They make their profit by buying cheap and selling dear and this requires 'movement' of commodities in either time or space. The former involves holding stock to 'play' the market, the latter involves transporting stock to create commodity chains. Commodity chains typically define a myriad space of flows organised through cities.
Cities are critical to economic spaces of flows in a second, more opaque, way. According to Jacobs' (1984) classic theory of dynamic cities, it is cities that are the crucibles of economic innovations that generate economic growth. Such innovation feeds directly into trade through import replacement. For instance, an entrepôt is a mere distribution center that does not of itself create economic growth. However, it becomes a dynamic city when it begins to manufacture goods it previously imported. It can then begin to consume and export its own products. This need not be a problem for the city it previously imported from; as a dynamic economic entity the latter can switch to new commodities of higher value such as the machine tools that the former entrepôt now requires. This process emphasizes the interdependencies of cities in a virtual spiral of economic growth. The unit of economic change is thus city networks in which cities are more than trading nodes, they are the basic locales of economic growth (Jacobs 1984). It is just such a space of flows - a world city network - that is undermining the modern mosaic spatiality of nation-states (Taylor 2004a).
The space of places that is the world of nation-states exults homogeneity; it is what nationalism, in its many guises, is ultimately about. Such politics defines who is and who is not a citizen (or potential citizen when the state-making process is still ongoing). In contrast spaces of flows organized through cities are about heterogeneity: trading only makes sense across spatial differences in production. Thus cities are traditionally full of merchants from different places: the more successful a city, the wider and denser its inward flow of 'foreigners'. It follows that cities are inherently cosmopolitan, thriving on their heterogeneity in spaces of flows.
Technological innovations that have made transport and communications faster and cheaper over the last two centuries have profoundly affected potentials for the scope and scale of economic activities. While trading has been a world-wide activity since the 'long sixteenth century' (c.1450-1650), it remained largely coastal and riverine until the nineteenth century. Communication was subsumed into these trading patterns. However railways and improved road transport, and the telegraph/telephone and new media such as the radio, altered fundamentally the space of flows. The initial effect was that the new spatial possibilities were harnessed to 'nation-building' by states through development of national railway and highway networks, and national telephone and national media as communication separated out from transport. In other words localisms were eroded as states promoted the new technologies to homogenize their territories. This stimulus for a mosaic state world spatiality lasted about a century from the mid-nineteenth to the mid-twentieth century. However, in the last half-century technologies have made possible increases the scale of spaces of flows, containerisation of land and sea transport, and air transport have created new global commodity spaces of flows while telecommunications have enabled global organization to become a reality. In particular the combination of communication and computing technologies have provided the means of world-wide command and control for 'global corporations' since the 1970s. Subsequently media has become less national and more global. Collectively these technological advances have provided the infrastructure of the space of flows that is globalization.
1.3 The Unexamined Spatiality in Macro-Social Science
For Jacobs (1984) there is no such thing as a 'national economy': it is a myth to imply state territories are coherent economic units. She calls them 'grab-bags' (p. 32) of different economies - aggregations of city economies - that create wholly artificial economic units. She asks why there is an assumption that economic processes should conform to political units, states, created by military and political processes. Jacobs traces the problem back to mercantilism and its state-centric growth theory and development policy: economic success was measured by hoarding bullion within a state. Thus economic study came to be about the state and this easily elided into the assumption that states define economies. This 'mercantilist tautology' (p.31) continues in contemporary macro-economics of all political hues. I accept Jacobs ' argument, which I call an unexamined spatiality in economics knowledge. I take the argument a stage further by finding the same unexamined spatiality at the heart of mainstream social science knowledge.
The mercantilist tautology has been consolidated in economic knowledge in the last century through reform of state 'treasuries' to encompass fiscal/economic regulation, management and planning roles for the state. This reform of economic practice in the state has proved to be important for place-making due to its insistence on the importance of state boundaries as containers. Simultaneously, the broad intellectual tradition of 'political economy' was converted into a specialist discipline of 'economics' to provide the relevant knowledge for the new expanded state. Thus, following Jacobs, the result was an embedded statism within the new discipline. Other social knowledge was similarly afflicted. The late nineteenth century and early twentieth century was the era both of reforms and discipline-making that went beyond state economics. The discipline of 'sociology' was a product of state social reforms that emerged from a more general, positivist 'social science'. And political reforms led to more specific knowledge than traditional 'political philosophy' that came to be called the discipline of 'political science'. It is this specialist trilogy of disciplines that came to be recognised as the 'social sciences' in the mid-twentieth century when this form of knowledge became widely legitimised through its institutionalisation as chairs/departments in universities (Taylor 1996).
Among the many effects of this embedded statism, one stands out as particularly relevant here: the strange assumption of spatial congruence in people's collective behavior. The study of economic processes, social processes and political processes are all circumscribed by state boundaries in such a way that economy, society, and politics are nationalized. Whether theoretical or empirical, macro studies use identical spatial units (state territories) to delimit the scope of their arguments. In the case of empirical research this is further enforced by the fact that macro-level data is produced by states for states so that statistics (state-istics) are state-delimited. But theoretical treatments are equally susceptible to this unexamined spatiality. Wallerstein (1984) doubts the validity of the concept of society since it always implies a national entity; he suggests abandoning the term. Recent concerns for replacing 'government' by 'governance' are based on a similar revelation that not all relevant politics are national. In other words, for most of their short history the social science trinity has been a vehicle for reproducing a world space of places; they are best termed mosaic social sciences. Such a knowledge form has had a deep effect on our understanding of cities (Taylor 2004a).
As we have seen, cities are organizational centers of spaces of flows. They are inter-locked by economic (and other activities) activities that create networks in which the cities act as nodes. However for states, their cities are administrative sub-units, mini-territories that require 'local government'. This bureaucratic perspective creates an administrative urban hierarchy with the capital city (national government) at the top and other cities largely ordered by size in different layers of hierarchy. Of course, major cities continued to engage in important 'cross-border trade' but this has been treated as largely irrelevant to what was viewed as the national urban hierarchy. And this position was bolstered by social scientists who produced a large literature on 'national urban systems' wherein were found urban hierarchies (Taylor 2004a). Purporting to be a tool for state economic management, this 'systems - thinking' intellectually legitimised the imposition of the state's self-image as a hierarchy on to knowledge of cities. The irony of this unexamined spatiality is that the most networked of social collectives, cities, were territorialized; a mosaic urban studies was created wherein a strange world of multiple and separate national urban hierarchies was conceived. This has been disastrous for understanding cities within spaces of flows (Taylor 2004a).
The lesson I learn from this section is that mosaic social science is particularly unsuitable as a vehicle for understanding globalization with its transnational processes. We need a 'network social science', social knowledge not mesmerised by state and nation. World-systems analysis is such knowledge, depicting as it does a world of connections within and without the nation-state. But this alternative spatiality embedded in world-systems analysis has never been fully exploited as a radical departure from conventional social science. Such a revisionist world-systems analysis is offered here.
2. INTERLUDE: REVISIONIST WORLD-SYSTEMS ANALYSIS
My revision is a geohistorical one. As well as using the spatiality perspective previously developed, I focus upon a particular time period: the 'long sixteenth century' that culminates in the establishment of a modern world-system (Wallerstein 1974). This timeframe is chosen because it shows the operation of a nascent modern world-system before both the consolidation of the inter-state system and the rise of nationalism. The expectation is that a glimpse of a less state-centric early modernity can help us transcend embedded statism for understanding late modernity, viz globalization. The result is a city-centric variation on Wallerstein's (1979) world-systems analysis. Most of the ideas have been presented in more detail elsewhere (Taylor 2004a, 2004b, 2004c), here I provide a summary listing of the argument.
The French invasion of 1672 is usually taken as the marker for the end of Dutch hegemony. Although (unlike the invasions of northern Italy in the 1490s) this attempted political take-over failed, the United Provinces were irretrievably weakened. In this revised world-systems analysis this year also marks the confirmation of the modern world-system as a world political-economy. The post-hegemonic Dutch polity begins to behave just like any other mercantilist state. Its specificity as a multi-nodal urban polity rapidly wanes in an interstate system that leaves no room for autonomous city polities. It is not until the rise of contemporary globalization that cities are again considered seriously as major players in the modern world-system (Taylor 1995).
3. MATERIAL SPACES OF THE TWENTY-FIRST CENTURY
The revisionist interpretation of the modern world-system as a world political-economy has profound implications for interpreting contemporary globalization. Following Wallerstein's argument for the decline of US hegemony, the recent worldwide promotion of neo-liberal economic agendas represents a major global shift of power from political elites to economic elites. Thus, accepting that we are in the demise phase of the modern world-system, interpreting the latter as a world political-economy means that we are finally on a possible transition to capitalism, to a capitalist world-economy. This is to argue for a world-system that is about to continue where the Dutch left off in the seventeenth century. Here I present two world-spatialities, one that interprets the present and one that extrapolates to a generation hence. I focus upon concrete outcomes in which national spaces of places are interwoven with global spaces of flows.
3.1 World-Spatiality Today: Post-hegemonic Superpower
Despite the rise of world cities, the most impressive feature of the contemporary world is the unprecedented political power of one state, the USA. From a standard world-systems position this is interpreted as a post-hegemonic effect: a political-military attempt to compensate for relative economic decline. The classic case is the expansion of the British Empire after British hegemony in the late nineteenth century. But the American situation today is different. The British may have built the largest empire in history but their power within the interstate system was far inferior to the current projection of US power across the world. This reflects one simple fact, hegemons have got larger over time: hegemonic United Provinces were a small polity, hegemonic Britain was a medium-sized state, hegemonic America was a continental state. Political compensation in the latter case has produced a unique superpower.
Instead of viewing hegemony at the state-territorial level, the revisionist position is that it is in cities that the wealth-generation to create hegemony occurs. And the cities are not evenly spread across the territories identified as hegemonic: the key Dutch cities were in or near Holland (Amsterdam, Rotterdam, Leyden, Utrecht, Haarlem, Delft), key British cities were in the north (Manchester, Birmingham, Glasgow, Liverpool, Leeds, Newcastle), and key US cities were in the 'manufacturing belt' (New York, Chicago, Boston, Detroit, Pittsburgh, Cleveland). Notice that the capital cities are not featured: The Hague, London and Washington, DC. were not central to hegemony-making. Also noteworthy is the fact that beyond these cities there were lands that looked anything but hegemonic: the Dutch outer provinces, Ireland in the UK, and the American 'Deep South'. In other words, it is practices in clusters of dynamic cities that initially created each hegemonic interval.
This is particularly pertinent to understanding US hegemonic decline. The peak of US hegemony in the 1960s coincided with the final domestic confrontation with the racist ex-Confederate South. The resulting rise of the 'New South' coincides with the hegemonic decline of the US. If you divide the US into a continental wide north-south division we get the two world faces of the country: producers and traders in the north (San Francisco - Silicon Valley, Seattle, Chicago, New York, Boston), and circuses and soldiers in the south (Los Angeles - Orange County, San Diego, Las Vegas, Texas, Florida). And in our consumer society it is the latter that have dominated politically (Kennedy in 1960 is the last time a candidate from the north won a Presidential election). Northern US cities created hegemony, the southern USA is using it up. This is the source of the US post-hegemonic political-military compensation but in the unique circumstances of creating a superstate there is an additional interpretation. In past hegemonic declines there has been rivalry between two states, one dominated by economic elites, the other by political elites: after Dutch hegemony England/Britain versus France, after British hegemony USA versus Germany. Today, although there is no hegemonic (economic) successor to the USA, it does seem to be the case that the political-military challenge has been internalised. In this argument the southern-dominated USA is the successor of Bourbon/Napoleonic France and Imperial/Nazi Germany.
The current balance between national spaces of places and global spaces of flows is that a neo-conservative USA - the 'last sovereign state' (Martin and Schumann 1997, 216) - is operating alongside a neo-liberal world city network (Taylor 2004a). This contrast in contemporary spaces is made particularly intriguing because US cities are relatively underrepresented within the world city network: only 7 of the top 50 world cities, as measured by commercial connectivity, are located in the USA (Taylor and Lang 2004a).
3.2 World-Spatiality in the Near Future (c. 2025): A Bi-polar States / World Cities Nexus
Looking ahead is speculation; we cannot know the future but we can make informed guesses. From a spatiality perspective I conjecture that the current spatiality will evolve as a reversal of the early modern world-system. That is to say, the earlier erosion of city power by states will turn full circle so that in the late modern world-system we will find erosion of state power by cities.
Unlike Petrella's (1995) famous scenario that depicts a city dystopia in which states have been eliminated, in my scenario states and cities continue to operate through their two distinctive spaces. State authorities are eroded by cities which become the main sources of connections across the world. In the erstwhile periphery and semi-periphery zones of the world political-economy, major world cities develop (Sao Paulo, Mexico City, Johannesburg, Moscow, Mumbai, Bangkok) and where there are failing states, cities substitute (e.g. Kinshasa for the Congo). In the European Union (EU) the territorial-sovereign conflict between the nation-states and the EU-center is elided as the trans-border connections of Europe's cities makes this 'old' politics less and less salient. In 2025 most of the world is clearly being run as network relations between cities with states as places of secondary importance.
But not all the world is city-dominated. The declining importance of states is hardly operating in two political poles where strong states continue to be powerful. In the western pole, the power of the USA of today remains remarkably resilient cemented domestically by a strong civic nationalism. In the eastern pole, the rise of the Chinese state creates a political balance with the USA, its ethnic nationalism connecting it to a wider Chinese diaspora. Both poles will have important cities linked to the world city network (for the USA: New York, Washington, Atlanta, Houston, Miami, Los Angeles; for China: Hong Kong, Beijing, Shanghai, Singapore, Taipei, Guangzhou) but they will continue to operate through a neo-conservative territorialist logic: to accumulate wealth within a place in the manner of traditional mercantilism. Economically and demographically these two states provide a severe challenge to the maturing world city network. Too large to dislodge, the modern world-system has come to an impasse between political elites in their two continental territories and economic elites in their global networks.
4. CONCLUSION: SYSTEMIC BIFURCATION
I have attempted to demonstrate in this paper that spatiality is a valuable analytic concept for studying world-systems. My revisionist world-systems analysis has ended with an impasse and I will conclude this demonstration by showing how this relates to Wallerstein's (2003) argument that the demise of the modern world-system will create a systemic bifurcation.
The prediction is that the transition to a capitalist world-economy through the world-city network will be hampered by the two champions of political economy, the USA and China. This is a contest between the alternative ruling elites - political and economic - now increasingly divided as the modern world-system declines. The hiatus will finally be broken when the cities in the USA and China begin to feel the economic costs of neo-conservative use of boundaries to divert flows. It is only with the re-emergence of an incipient capitalist world-economy that the critical bifurcation appears. The world city network can operate in two ways. One builds upon existing power differentials to create strong hierarchical tendencies within the network, with new monopolies to enable enhanced capital accumulation. This route consolidates a capitalist world-economy. The second route involves a mobilization of citizens to attack existing power differentials and ensure the world city network lives up to its potential for social mutuality. We might call this the anarchist moment. Their previously political flowering coincided with national consolidation of states (a maximal inhospitable social environment for anarchists), but with states gone or tamed the systemic bifurcation provides a practical opening for creating a non-hierarchical, egalitarian world-system of cities (Taylor 2004b).
Braudel, F. 1984. The Perspective of the World. London: collins
Castells, M.1996. The Rise of Network Society. Oxford: Blackwell
Israel. J. I. 1995: The Dutch Republic. Oxford: Clarendon Press
Jacobs, J. 1984. Cities and the Wealth of Nations. New York: Random House
Jessop, B. 2002. The Future of the Capitalist State. Cambridge: Polity
Martin, H-P., and H. Schumann. 1997. The Global Trap. London: Zed
Petrella, R. 1995. A global agora vs, gated city-regions. New Perspectives Winter: 21-22
Schama, S. 1987. The Embarrassment of Riches. London: Collins
Scott, A. J. 2001. Global City-Regions. New York: Oxford University Press
Spufford, P. 2003. Power and Profit. London: Thames and Hudson
Taylor, P. J. 1994. The state as container. Progress in Human Geography 18: 151-62
Taylor, P. J. 1995. World cities and territorial states: the rise and fall of their mutuality. In World Cities in a World-System, edited by P. L. Knox and P. J. Taylor, 48-62. Cambridge, UK: Cambridge University Press
Taylor, P. J. 1996. Embedded statism and the social sciences. Environment and Planning A 28: 1910-16
Taylor, P. J. 1999. Places, spaces and Macy's. Progress in Human Geography 23: 7-26
Taylor, P. J. 2004a. World City Network. London: Routledge
Taylor, P. J. 2004b. Homo Geographicus: a geohistorical manifesto for cities. Review (Fernand Braudel Center) 27: ???-???
Taylor, P. J. 2004c. Dutch hegemony and contemporary globalization. In Hegemonic Declines: Present and Past, edited by J. Friedman and C. Chase-Dunn, ???-??? . Boulder, CO: Paradigm Press
Taylor, P. J., and R. L. Lang 2004. Us Cities in the World City Network. Washington, DC: Brookings Institution
T'Hart, M. 1993. The Making of a Bourgeois State. Manchester, UK: Manchester University Press
Wallerstein, I. 1974. The Modern World-System. New York: Academic
Wallerstein, I. 1979. The Capitalist World-Economy. Cambridge, UK: Cambridge University Press
Wallerstein, I. 1980. The Modern World-System II. New York: Academic
Wallerstein, I. 1984. The Politics of the World-Economy. Cambridge, UK: Cambridge University Press
Wallerstein, I. 2003. The Decline of American Power. New York: The New Press
Edited and posted on the web on 10th February 2004
Note: This Research Bulletin has been published in ProtoSociology, 20, (2004), 30-45