This Research Bulletin has been published in J Friedman and C Chase-Dunn (eds) (2005) Hegemonic Declines: Past and Present Boulder, Colorado: Paradigm Publishers, 117-134.
The title of this paper may appear odd to many readers. Whereas it makes sense to discuss recent American hegemony in conjunction with contemporary globalization, what has the seventeenth century Dutch 'golden age' to do with global-scale economic practices in the twenty first century? The simple answer is that this paper is an attempt to situate globalization in a world-systems analysis, to define its longue duree context. But why specifically pick on the early Dutch contribution to modern world-system development? This will be justified in the first section below but the basic point is that I am approaching this world-systems analysis from a city-centric perspective. Hence my concern for the 'city-rich Dutch' as a pointer for understanding what the world cities literature identifies as a 'city-rich globalization'. In effect I am trying to put some meat on the skeleton I proposed in my suggested recasting of world-systems analysis to include city networks (Taylor, 2002). As with the latter paper, I ask that this argument be treated in the spirit of an exploratory excursion into a world being 'turned upside down' yet again.
In addition I am taking this opportunity to experiment with merging Jacobs' (1984) city economy analysis with Wallerstein's world-systems analysis. Both arguments begin with the same starting point: a fundamental critique of the very existence of 'national economies'. However, they differ in the direction that they then take their respective analyses: Wallerstein incorporates 'national economies' into a single world-economy, Jacobs divides 'national economies' into their constituent city economies. The paper is used as a vehicle to see whether these two sets of heretical economic thoughts can be brought together. This requires a re-evaluation of the so-called 'Dutch Republic' that emerged after 1579. The basic method used is to confront ideas on the Dutch 'golden age' with the basic historiographic question: which present produced this past?
INTRODUCTION: THE IMPORTANCE OF THE DUTCH TO THE MODERN WORLD-SYSTEM
In world-systems analysis the 'Dutch Republic' is normally identified as one of three hegemonic states that have defined the basic trajectory of the modern world-system. However, compared with the British in the nineteenth century and the USA in the twentieth century, the seventeenth century Dutch appear to be a pale shadow of what a 'world hegemon' should be. A very small state both territorially and demographically, it hardly seems feasible that this still new polity could set the path along which the modern world-system embarked to eliminate all rival systems. Of course, if we accept the latter formulation then, far from being inferior, as the initiator of the trajectory the Dutch are the most important of the hegemons. And this is because it is not overt power that defines a hegemon but its infra-structural power: the Dutch developed a social formula, which we have come to call modern capitalism, that proved to be transferable and ultimately deadly to all other social formulations.
It is not usually appreciated just how important this simple fact of being historically first is for taking the Dutch especially seriously. In discussions of the rise phase of the modern world-system much of the argument is taken up with a search for the origins of the system. But there is a real problem with the 'origin' fetish in geohistorical analysis. Asking the question as to when a world-system began is fraught with difficulties not least surrounding the question: origin of what? Furthermore, however we choose to define modern capitalism, its distinctive features will always be traceable to other world-systems: price-setting markets, capital accumulation, spatial divisions of labour, and free labour processes can all be found in times and places outside the modern world-system. And this is to be expected in the rich social tapestry of human endeavour. To use a biological analogy there will be many mutants of social organization within all world-systems and some will prosper a while just as others will wither almost immediately. The important social mutants are those that prosper for much more than a while and ultimately provide the basis of a new world-system. From this analogy it is not the choosing, in hindsight, of one or more origins from a multiplicity of possibilities that is the key, but rather it is the identification of circumstances for sustaining one of these possibilities to create a fully-fledged historical system that matters. And this is the critical role of the Dutch polity as primary hegemon: to continue the biological analogy, it is where a particularly virulent mutant - modern capitalism - finally succeeded.
THE DUTCH AND THEIR CITIES
In his seminal text on the origins of the modern world-system, Wallerstein (1974) covers all bases in his identification of the 'long sixteenth century' that nearly covers two centuries and overlaps with the early decades of Dutch hegemony. Furthermore the latter is the pivot of volume two where Wallerstein (1980) describes the consolidation of the world-system. Thus in this story he certainly gives due weight to Dutch economic and political achievements, although 'consolidation' is not quite as strong an affirmation of the importance of the Dutch as that presented above. However I do not wish to dwell on this difference here. Rather there is another feature of Wallerstein's treatment of the Dutch that particularly contrasts with the argument I develop in this paper. Although self-evidently a land of many cities, these places do not feature prominently in Wallerstein's text. To be sure Amsterdam is frequently referred to but the remainder of the bustling Dutch cities are wholly or relatively neglected.
This is in stark contrast to Braudel (1984) for whom cities and their inter-relations are at the centre of his work. The difference between these two social analysts is best seen in their alternative concepts of world-economy. Whereas Wallerstein defines an area with an integrated division of labour, Braudel develops a city-centred concept of world-economy. Although these can be alternative descriptions of the same world-economy - geographically, Wallerstein emphasizing the 'extensive' (the zonal range) and Braudel's emphasizing the 'intensive' (the network centres) - there is a basic difference in how the space-economy is understood. The emphasis on cities provides a spatial infra-structure, an organizational framework that enables a world-economy to come into being and expand. Here cities are not merely trading cities/ports, banking centres or industrial towns, they are integral parts of complex networks of capital circulation. Above all they are where the circulation is organised, where the use of the capital is decided. In other words, it is in cities that the fundamental decisions concerning production, distribution and consumption are made: cities represent capital (Braudel 1984; Tilly 1990)
For Wallerstein (1976), the modern world-system represents the one and only time that a world-economy was able to resist conversion to a world-empire and become a capital expanding system. Previous world-economies were 'fragile' lacking overall political structures with 'life spans probably less than a century', which is all that Wallerstein says about them. For Braudel (1984) there have been many past world-economies defined by their vibrant cities and having various relations to co-existing world-empires. Both agree, however, that the rise of the Dutch polity with Amsterdam at its economic heart created a new enlarged European world-economy, trans-Atlantic for Wallerstein, world-wide for Braudel. Which to choose for understanding the new Dutch polity? A key advantage of taking Braudel's city-centric position is that it allows us to explore marrying Jane Jacobs' (1984) heretical economics to world-systems analysis.
JACOBSEAN ECONOMICS, CITY REIFICATION AND INTER-LOCKING NETWORKS
In Wallerstein's (1980, 45) discussion of Dutch hegemony he uses the high degree of urbanization as a 'confirmation of vitality' of the Dutch. In this formulation 'urbanization' is an outcome of the economic development of a particular state, the 'Dutch Republic'. In contrast, for Jacobs (1984) it is cities and their regions that shape economic life, not vice versa. Thus orthodox economics transfixed by 'the potpourris we call national economies' (p.35) operates in a 'fool's paradise' of economic decision-making. Cities, not 'nations', constitute economies. It is in city economies that economic growth spirals are generated in an input-substitution process that creates waves of 'development' into the city's hinterland and surrounding regions. This geohistorical process is not a property of all cities: many cities are 'passive' wherein economic change is a product of outside influences. However, for Jacobs those special cities that are economically vibrant change not just themselves but transform whole regions. Given that we have argued that seventeenth century Dutch cities changed the world, they must qualify as mega examples of this type of economy!
We must be careful in using such models not to reify the city or city-region. Cities, of course, do not of themselves create economies. Abrams (1979) is particularly useful in reminding us of this and attacks the use of concepts such as 'generative city' and 'parasitic city' as popularised by some development literature. Cities are not agencies of economic change and he takes Braudel to task for sometimes lapsing into this type of thinking (Abrams 1979, 17). The agents of economic change are the holders of capital and it is their decisions that are vital for economic growth or stagnation. The fact that these agents congregate massively in cities, and that the most important of them in relatively few selected cities, does beg the question as to how cities are implicated in this economic life. City economies are the lifeblood of economic growth because they are special places where multifarious information and knowledge are specifically available to seize economic opportunities through innovations in production, distribution or consumption. Successful city economies are learning and reflexive economies (Storper 1997), knowledge-rich matrixes of business, banking, professional and creative agents that cannot be easily duplicated in other places. Smith (1984) has described seventeenth century Amsterdam as just such an 'information exchange'.
One final element has to be added to our interpretation of cities within economic life. There is no such thing as an isolated city; cities exist in relation to other cities as the collective 'crossroads of society'. Cities within a single world-economy can thus be viewed as a city network. But it is an unusual type of network, one that is commonly termed an interlocking network (Taylor 2001). Because we do not reify the city, unlike other networks the inter-nodal relations are not the direct product of the nodes (cities) themselves. Rather nodes are 'interlocked' by sub-nodal agents, in the case of cities by the connections among the business interests that define both city economies and their network. For instance, in Braudel's (1984) discussion of city-centred world-economies in late medieval/early modern Europe, families and firms from a given city have economic agents located in other cities to negotiate, organise and report back intelligence thus providing 'interlocks' between the firm's 'home city' and the other cities. It is this interlocking multiplied manifold that creates a city network as the basic spatial organization of a world-economy.
Incorporating Jacobs' heretical economics into our analysis allows us to revisit two important debates on the seventeenth century Dutch from a specifically city-orientated perspective. These debates concern whether the Dutch created a new territorial state or just another city-state, and whether their foreign policies can be classified as mercantilist.
REVISITING DEBATE I: WERE THE DUTCH A 'STATE'?
Maurice Aymard (1984, 10) asks the question whether the Dutch as the centre of a new world-economy created 'the last city or the first state'? Different authorities can be placed on either side of the debate. The city-position is that what was created in the northern Netherlands was an Amsterdam city-state much in the manner of earlier city-states that dominated their world-economy such as Venice in its prime (e.g. Barbour 1963). Their main argument is based upon the quantitative economic dominance of Amsterdam which spilled over into the political sphere. The state-position is that the northern Netherlands constituted a new territorial state taking its place in the inter-state system emerging at this time. Their main argument is that although the Dutch operated through a very distinctive state apparatus, it still was in fact a modern territorial state (t'Hart 1993). Braudel (1984) although initially favouring the city side eventually comes down in the middle - the Dutch polity as a transition form between city-state and modern territorial state.
I want to pose a different question about the Dutch political space: must we choose between city-state and territorial state? Classifying an unusual polity into one of two known political entities may make for simple interpretation but what do we lose in the process? I will deal with the state option first since this is the dominant position, certainly within world-systems analysis, and it is one that I have subscribed to (Taylor 1996).
The work of 'tHart (1989, 1993) carefully delineates the Dutch polity as a 'state of 58 cities' (1989, 666) that emerged in a war against centralization. Thus it is to be expected that the resulting polity would be a decentralized state, opposite to the centralising 'absolutist' states of the times. However because the ideal model of the early modern state has come from the ranks of the latter, France to be precise, it follows that Dutch state credentials are left open to doubt. Her research strives to show the fiscal novelty of the Dutch in an alternative form of modern statemaking. In effect she sees the Dutch Republic as a victim of a particular European historiography that equates state with centralization. But in breaking away from one orthodoxy such arguments remain within even more deeply embedded ways of thinking. Let me explain by way of a parallel argument. In his famous discussion of the Dutch golden age, Schama (1987) has pointed out that the traditional interpretation of the Dutch rebellion as a nationalist revolt is merely a projection of nineteenth century romantic historiography on to seventeenth century Dutch politics. I would argue that the contemporary interpretation of the result of the Dutch rebellion as a modern state is a projection of twentieth century social science statism on to seventeenth century Dutch politics. The concept of the inter-state system leaves no geographical space for other than territorial organization of sovereign states, so its intellectual concomitant embedded statism leaves no theoretical space for other types of polity. Projected backwards, if the Dutch are to contribute to the making of the modern world they could only have done this through possessing a modern state.
The latter logic is very clear in Israel's (1989) explanation for the Dutch inheriting Antwerp's economic network after its fall to Spanish troops in 1585. Initially the economic diaspora from Antwerp was fairly dispersed with 'north-western Germany ' (primarily Hamburg) as the 'most likely candidate' to reap the benefits of Antwerp's demise (p. 33). But very soon the focus switched to the northern Netherlands because of the crucial need for the 'protection and active support of a powerful state' (p. 36). Political fragmented northern Germany simply could not compete. Hence the rise of the Dutch is due to their creation of a powerful state. This standard explanation equates state with defence implying that the former is the only institution that can provide the latter. But if this conflation of state and defence is broken we can suggest that what the Dutch possessed was a highly defensible territory that could be marshalled against attack by a 'league' of geographically concentrated cities. Thus there is no logical reason why we have to continue with the orthodox position: in the right circumstances, a polycentric city region should not be able to organise successfully its own defence without the need to create a modern state.
In my view such orthodox constricted thinking is aided by the posing of one alternative that can be shown to be flawed relatively easily. Quite simply there was no 'Amsterdam city-state' for the fundamental reason that the city of Amsterdam did not rule this new political space either directly or indirectly by some subtle form of subterfuge that is difficult to reveal. All writers agree that the Dutch polity operated through shifting coalitions of cities, provinces and the stadholders. In key questions such as peace versus war, for all its economic importance, Amsterdam could be and was outmanoeuvred by counter coalitions. Venice, a true city-state, could never have lost control of Venetian foreign policy! Hence if the choice of where to place the Dutch polity is between city-state and modern state the latter almost wins by default. But this is only the case if the choice is limited to just these two possibilities.
Let us return to 'tHart's 'state of 58 cities' and remove reference to state which leaves a 'political space of 58 cities'. Such a space is a polycentric city-region. This idea is picked up in some of the literature with references to, for example, a 'league of city-states' (Boogman 1979, 398) and 'a confederation of Venices' (Pocock 1992). It was in fact a unique political creation with no name - according Schama (1987) the name 'United Provinces' was coined by a contemporary Englishman, "Dutch Republic" is a more recent statist appellation. This polycentric city-region polity had a very specific spatial structure: a vibrant core of many cities surrounded by a defensive frontier of fortifications (Taylor 1994). Braudel (1984 202) calls the result a 'fortified island' in the middle of which is a 'high-voltage urban economy' (p. 180). Of course there had been polycentric city polities in the past - city leagues as trading networks of cities with political institutions (e.g the Hanseatic League) and coalitions of city-states (e.g in northern Italy) but none of these formed a city-regional structure including a defensive shell. It is the latter's territorial defensive practices that make Braudel's argument for a halfway house between city and state seem so plausible but I prefer to see this political outcome as a distinctive polity, neither city nor state in the terms of this debate. Locating the Dutch polity in transition eliminates contemporary relevance. Instead of the Dutch polity being a stop on the evolutionary path to modern statehood, it can be viewed as a particular capitalist polity historically superseded by modern states but theoretically still available as a political category for future reference.
In this spirit let us consider the primary political dimensions of a polycentric city-regional polity. Such a polity will combine co-operation with competition among its constituents. In the case of the Dutch polity, co-operation is basic to the defensive function as written into the founding treaty (e.g. joint responsibility for the fortifications - see original texts in Kossmann and Mellick (1974, 167-9)). However it occurs also in the economic division of labour that developed. According to Braudel (1984, 182) Amsterdam was the apex of a pyramid of cities and co-operated with the following cast:
'industry propered in Leyden, Haarlem and Delft; shipbuilding in Brill and Rotterdam; Dordrecht made a living from the heavy flow of traffic along the Rhine; Enkhuisen and Rotterdam controlled the fisheries of the North Sea; Rotterdam, again, the most important city after Amsterdam, handled the lion's share of trade with France and England; the Hague, the political capital ... (Braudel 1984 184)
But this division of labour was also a source of conflict within the polity. For instance, industry and trade interests differed with respect to war and peace with Haarlem and Leiden leading the 'war party' that prevented the 'peace party', led by Amsterdam and Rotterdam, from concluding a truce in 1630 (Israel 1990, 56). Of course, all polities are a matrix of conflict and compromise but what makes the Dutch polycentric city-region polity so unusual is that it had no institution to act as political umpire to resolve differences, a unanimity rule ruled. In modern states the 'umpire' role is the function of the executive constrained by the legislature and judiciary, backed up by the coercive arm of the state. The nearest the Dutch got to an executive with coercive capability was the stadholder position that was periodically powerful but never attained sovereign status. This further suggests that we should not confuse the Dutch polity of the seventeenth century as a modern state. The Dutch political space was filled by a unique political entity, a polycentric city-region polity.
REVISITING DEBATE II: WERE THE DUTCH 'MERCANTILIST'?
The Dutch hegemonic cycle coincides with the 'age of mercantilism'. Since contemporaries were forever complaining that the Dutch were 'ruled by merchants' it might be thought that the Dutch polity of the seventeenth century would be the prime example of mercantilism. Not so. The classic 'mercantilists' were the English and the French. So what about the Dutch? Traditionally not identified as mercantilist, latterly they have been brought to the very centre of the 'mercantile system'. As in the previous debate, this debate has an important historiographical dimension.
The term 'mercantile system' was defined by Adam Smith in the eighteenth century to describe the restrictive economic policies of states he was keen to critique. These were policies of partnership between economic elites and the monarch on the basis of the equation: more wealth equals more power in and for the state (Wilson 1958). However in the late nineteenth century this pragmatic policy became reinterpreted as a policy of economic nationalism to justify state protectionism for new states (Coleman 1969). In this way, older states, England, and especially, France, became the model cases of mercantilism. The common missing link in these arguments was the target of these mercantilist policies. It was the Dutch who stimulated the need to devise restrictive policies: England and France were trying to counter the economic successes of the Dutch cities. Hence, the so-called 'merchant state' was not mercantilist as defined by Smith since it was in its general interest, as hegemon, to have as few economic restrictions on trade as feasible (Wallerstein 1980). Of course, Smith's definition of mercantilism, by omitting the Dutch, describes only half the world-economic process as it existed in the seventeenth century. If we devise a broader definition of mercantilism as political policy for the benefit of merchants then the Dutch become the super-mercantilists (Rowen 1978, 189). Without the need for partnership with the 'private interests' of a monarch, there was little or no dilution of 'public' economic policy except in emergencies when the Stadholders played their military role.
As the discussion above shows, it is difficult to develop ideas on mercantilism without consideration of the state again. The 'super-mercantilist' conclusion merges with the Boogman's (1978) thesis that the Dutch created a new political-economic raison d'état that contrasted with the standard political-military raison d'état of the times. Although very tempting - again I have used it in my work (Taylor 1996) - this idea contradicts our previous conclusion not to interpret the Dutch polity as a state; no state equals no raison d'état. As Braudel (1984, 205) tells it: 'for the Dutch, commercial interests effectively replaced raison d'état''. But this leaves us with no obvious way of characterising the Dutch polity's pro-economic policies. To unravel this conundrum we have to return to Jacobs' heretical economics.
At the core of Jacobsean economics is the revealing of a taken-for-granted assumption within macro-economics. This is the geographical assumption that political sovereign territories are spatial economic entities. Jacobs traces this fallacy back to mercantilism, all orthodox economics from Smith through various political economies to communist economic planning, Keynsianism and monetarism, for all their many differences, share this 'mercantilist tautology' (Jacobs 1984, 32). All of these schools of economic thought have as their purpose the construction of policy levers to 'steer' the 'national economy'. In this sense they can all be termed 'political economy', they are developments of economic ideas to serve the state. Thus statist mercantilism created a political economy discourse that has survived through to the embedded statism of the contemporary social science discipline of economics. The point is, of course, that this statist mercantilism is only the 'half-system' component of the seventeenth century world-economy. It misses out the Dutch, the target, the hegemon.
If the Dutch did not constitute a state then their pro-merchant policies cannot be described as political economy. Rather we might turn the term around and refer to Dutch policy as 'economic politics'. The latter are, of course, policies that independent cities pursued before the rise of the modern state. In effect, what I am doing here is dividing Arrighi's (1994, 10) 'political capitalism' into two types, political economy and economic politics. The Dutch polity as a polycentric city-region continued city-centric economic politics while others created the political economy of statist mercantilism. The clearest indication of this difference can be seen in the work of Misra and Boswell (1997) where seventeenth century Dutch leadership, as measured by ships, in trade is very much greater than in military seapower (from Modelski and Thompson 1988). This contrasts markedly with their great rival England where seapower was consistently much the more important than trade. In other words, the Dutch were much more of an economic power than a military one (i.e. economic politics) with England much more of a military power than an economic one (i.e. political economy).
Generally economic politics can be seen in a concern for networks rather than territory. The Dutch polity of the seventeenth century was famously unconcerned with territorial expansion: as long as the frontier operated effectively as a defensive shield no extra land was deemed necessary. But this did not mean isolationism, far from it for the Dutch were 'intimately concerned with the maintenance of favourable conditions in the world beyond their boundaries' (Wilson 1957, 6). And this meant developing and maintaining networks of trade and finance. Thus, the Dutch were concerned to keep open the Sound to the Baltic and intervened in Swedish-Danish relations on numerous occasions to keep the seaway open. In stark contrast the Dutch polity was adamant that the Scheldt should not be opened. Closing off Hapsburg Antwerp from the sea in war, the Dutch were just as keen to keep it closed in peacetime to stop this erstwhile world-city regaining any economic clout to rival Dutch cities. Hence economic politics do not translate simply into 'free trade', rather it is 'pro-trade' policy for the particular city or cities. This simply means the city or city-region attending to its networks, which was the basis of Dutch economic politics.
Such economic politics on a world-economy scale ended in 1672 with the demise of civilian leadership of the Dutch polity in the wake of the French invasion. It is from this point that we can identify the Dutch converted into a state pursuing vigorous military and economic wars against its rivals. As a state the Dutch were a failure but more important was the closing off of the city-region polity option. Subsequent hegemons were explicitly statist however liberal their policies: both the British Manchester liberals and Americans proclaiming 'the business of America is business' can only be described as political economists. Thus the seventeenth century Dutch polity of cities is to be intellectually treasured as a unique alternative to all-pervasive modern statism, especially given that is may have particular resonance for contemporary globalization.
STATES DESTROYING CITIES
It has been pointed out on many occasions that while the rest of Europe was suffering its 'crisis of the seventeenth century', the Dutch bucked this trend with their 'golden age'. In short the Dutch are interpreted as the only successful state of their era. However, taking a city-centric position we come to a different position. Certainly the cities of the Dutch were very successful in the seventeenth century and this economic vibrancy was reflected in their rapidly growing populations. But they were not the only cities to experience rapid demographic growth: Madrid, Vienna, Paris and London were also cities that prospered in the seventeenth century. Of course, the latter quartet grew for different reasons than the Dutch cities, they were the beneficiaries of political restructuring that produced the more centralised state-form that the Dutch rebelled against. Therefore their growth can be said to reflect a political vibrancy whose benefits were harnessed in just a few capital cities. These are successful early modern 'political economy cities' that should be contrasted with the Dutch cities and their economic politics.
The effects of political economy cities on other cities can be starkly seen with the decline of the vibrant urban system of central Castile in the early seventeenth century (Ringrose 1989: Albaladejo 1994)). In the sixteenth century this system had a hierarchical structure with Toledo at the top that lasted until Madrid expanded to outgrow Toledo and all the other local cities put together. The effect of Madrid's growth was catastrophic, instead of representing new economic investment in the region it acted as a capital-sink politically distorting local markets and thus destroying city economies. The pertinent figures are shown in Table 1 where the two leading cities of central Castile are compared to the two largest cities of central Holland. Note that initially Madrid actually outgrows Amsterdam. However the real contrast is between the fate of the two 'second cities': unlike Toledo, Leiden keeps up with its illustrious local rival. This stark difference between Toledo and Leiden represents a key dissimilarity between political economy and economic politics. While the rise of Madrid as a great Hapsburg capital city was destroying an adjacent urban economic structure, the rise of Amsterdam was accompanied by the rise of other Dutch cities in a mutually-reinforcing polycentric city-region.
Of course, the elimination of Toledo as a major city-economy is a very extreme political economy example chosen to make a point. But, nevertheless, demotion of city-economies by states is by no means uncommon. This relates to Jacobs' (1984, 32) identification of so-called 'national economies' as 'collections or grab bags of very different economies'. The geohistory of European states is awash with examples of contrasting city-economy needs where political elites have to choose policies that favour some cities over others. Often these are generalised into 'maritime-trade versus continental land' disputes. Classic examples are the 'two Frances' - Paris versus Atlantic France', the 'two Germanies' - Prussia/Berlin versus Rhinelands, and the 'two Britains' - 'rentier south' versus 'industrial north'. But the point is that these are not regional problems, they are city-economy conflicts and must be addressed in that way.
Jacobs (1984, chapter 11) provides a simple political economy model of how city-economies are treated within a state. If the 'national economy' is in reality an amalgam of city-economies it follows that those pulling the policy levers cannot satisfy the economic needs of all their economic components. In a situation where one city begins as the largest city-economy, usually the capital city, political economy policy will tend to favour that city because its economic circumstances are more reflected in the aggregate 'national economic' statistics. Thus national policy, for instance in interest rates to control currency levels, will respond to the requirements of a state's largest city-economy at the expense of the rest. Over time such a political economy process will produce a situation where one city increasingly dominates: the aggregate 'national economy' gradually evolves towards a city economy with 'surplus' territorial appendages. This seems to have happened in many European states where one city-economy dominates: Stockholm/Sweden, Copenhagen/Denmark, Oslo/Norway, Helsinki/Finland, Vienna/Austria, Athens/ Greece, Lisbon/Portugal, and, of course, Paris/France and London/UK. For instance, in the latter case, successive governments have pursued policies favouring the rich south-east - London's city region - at the expense of what remains of the great Victorian cities of northern Britain. This process does not occur where there is a more decentralised state (Switzerland), and where the state is large enough to accommodate numerous large city-economies such as in the USA and Germany. In the latter cases, political economy policies will impact neutrally on the various city-economies over time so that no one city will grow to dominate. But still there will be no economic policy customised to the needs of any one city-economy. This is important for cities that have to restructure their economies (not simply 'reinvent' themselves via a city boosterism) and need the helping hand of an economic policy e.g. without its own policies the Detroit city economy will be just as bad off as the Manchester city economy. Both these cities desperately need an economic politics.
FORWARD TO CONTEMPORARY GLOBALIZATION
There is a massive literature on contemporary globalization which inevitably means a vast array of different positions invariably backed up by different definitions. In many ways the best text that tries make sense of all this is Held et al.'s (1999) Global Transformations. They have provided what has become a standard way of classifying the globalization literature. Three 'broad schools of thought' are identified: the hyperglobalists, the sceptics and the transformationists (p. 2). The first group define a new post-state era of transnational processes, the second group reject such novelty and equate globalization with already-existing international processes, and the third group believe globalization is historically unprecedented but that states, far from being transcended, are adapting to the new circumstances.
Notice that these 'schools' actually lie along a single axis that measures the degree to which the state is assumed to be under threat from globalization. We might call this the state-erosion scale with the hyperglobalists at one end, the sceptics at the other, and the transformalists in between. From their title it is apparent that the authors belong to the latter group but this classification tells us more than authors locating themselves in a safe middle position in a debate. The key point is that the scaling is actually about the state, not globalization per se. In fact, their whole treatise is actually about the state and its future as if this is all that globalization is about. It represents an indication of how the embedded statism in the social sciences is able to encompass globalization within its thinking (Taylor 2000). In some ways this is a quite bizarre position to take for understanding globalization: the book would be better entitled 'State transformations'. At the very least studying globalization should enable us to transcend the taken-for-granted mosaic world of the states and point us towards a much more network view of how the modern world-system operates. This is not a matter of choosing between places and flows but rather of taking the opportunity to readjust our social scientific thinking that has been so much dominated by places, especially nation-states.
Cities are back. A world cities literature that does not necessary belong to any of Held et al.'s 'globalization schools' has defined a new spatial structure of nodes and connections in a global space of flows. And this is why the seventeenth century Dutch are so particularly interesting for analysing contemporary globalization and why it is important to consider whether their polity was a state or not. Answering yes puts the 'Dutch Republic' at the beginning of the political economy tradition of thought and practice. I have argued that the Dutch political space encompassed a different thought and practice, an alternative to territorial political economy. If the latter is indeed under threat from globalization, then we need to begin looking for historical alternatives. For instance, one popular approach has been to search outside the modern world and proclaim a complex 'new medievalism' - I agree with Bromley (1996, 5) that this is a rather 'fanciful assertion'. In contrast the seventeenth century Dutch are inherently modern and they provided a political space, both geographically and metaphorically, for a city-centred economic politics that may have relevance for understanding contemporary globalization organised through world cities.
My purpose here is not to 'learn lessons' from the golden-age Dutch by drawing simple analogies across the centuries. (Although I must admit that equating the Bush family with the Orange dynasty and having the current president as 'global stadholder' brought back to check calamity (9/11 as the new 1672) does appeal!) More fundamentally, city-based thinking beyond political economy provides for a network description and analysis of globalization. At GaWC we have been trying to define a world city network for the contemporary world-economy. This is specified as an interlocking network with global service firms in the role of 'interlockers'. In effect we are defining a network of service centres for global capital, places where today's necessary inputs to capital accumulation processes take place. The global service firms provide 'advanced producer services' such as accountancy, advertising, banking/finance, insurance, law, and management consultancy. All of these services have generated their own global corporations with offices in cities across the world to provide the 'seamless service' their corporate clients demand. Thus there is a network of nodes (clusters of these services in cities) with connections between them (intra-firm flows of instruction, information, knowledge, ideas, plans, etc.). This world city network transcends states in some ways (e.g. most of London's advanced producer services are provided for clients operating outside the UK) but also operates within regulatory frameworks that are state-based (e.g. law). Thus the world city network and the nation-state mosaic are both separate and intertwined within contemporary globalization. Hence while cities have not 'escaped' from states - this is not an era of city-states - they are implicated in important trans-state processes the intensity of which makes contemporary globalization different from previous world-economy eras. It is this 'city manoeuvrability' within the world-economy that marks out the present.
Although we would not expect political economy to disappear with globalization, we can ask if there are indications of a new economic politics appearing. Political economy is reflected globally in an international relations (IR) that is essentially 'realist' in nature, a thought and practice that is about competition. There is an idealist counter to this dominant school of IR but in practice co-operation across states always fails when perceived threats to 'national interest' are involved. However, in contrast to this mosaic world of boundaries, in a network world it is co-operation that is inherent in the structure. In general, mutuality is at the heart of all networks, without coincidence of interests networks simply collapse. In the case of the world city network the mutuality exists through the service firms and their global location strategies. Investing in expensive offices across many cities means that these firms have a vested interest in all the cities they operate through. Hence although the traditional city development literature emphasises competition, a sort of 'lower-level IR', the reality is a symbiosis among world cities as a network.
This conclusion provides a basis for searching out economic politics in contemporary globalization. At GaWC we have just completed a study comparing London and Frankfurt as world cities (Beaverstock et al. 2001). This is a particularly interesting pair of world cities because the location of the European Central Bank in Frankfurt followed by the launch of the euro currency led to much financial press speculation that this city was 'catching up' London. A review of the press in both cities found this to be by far the dominant discourse. However a very different story emerged when interviewing practitioners in both cities. The idea that London and Frankfurt were in competition was an alien notion to their economic practices. Working with colleagues in their firm's office in the other city, and visiting that office in the other city, meant that the two cities were seen as part of a single practice. There was a division of labour - London dealing with more global packages, Frankfurt with European packages - but the dominating process was one of co-operation. This idea was even found in the interviews with urban officials working for city institutions in both places. Quite simply, the view was that 'what's good for London is good for Frankfurt and vice versa'. Growth in spheres of work attended to in Frankfurt had potential for spilling over into London's global practice, and the latter provided opportunities for developing Frankfurt's European work. In short, London-Frankfurt is an important European dyad in a world city network where co-operative relations interlock the cities.
One swallow doesn't make a season and we must be careful about reading too much into the recent growth of a world city network of service centres. It is clear, however, that the logic of the processes we are describing is for city authorities to attend to much more than just their place, they need to devote as much political energy to their networks (Beaverstock et al. 2002), other places beyond their specific jurisdictions. Herein lies a reason why we might predict that big city mayors will begin creating a global economic politics for the twenty first century.
CONCLUSION: SUCCESS TO DEMISE
All this might sound like another argument for a triumphant globalization, but I follow the dictum that it is through its successes that the modern world-system approaches its demise. There are two ways this can be argued from the above thesis.
My emphasis on cities to counter the centrality of political economy in our thinking, right through to the embedded statism of the social sciences, has implications for radical theories of the state that conclude the capitalist state to be essential for capitalist reproduction. The basic message of these theories is that without a restraining political hand, short-term capitalist calculation will destroy the means of production. Hence the degree to which the economic politics of cities replaces the political economy of states, the asymptotic drivers to the demise of this world-system will be accentuated.
Alternatively we can view the idea of 'capitalist state' to be a based upon a non sequitur. It is a classical political economy concept that misses both the urban and world basis of economic life. This is not to denigrate the importance of modern states over the last few centuries but, as I have previously argued (Taylor 2002), this requires us to rethink the nature of the modern world-system. If it is a world political economy (i.e. shared rule by economic elites (capitalist executives) and political elites (state executives)) rather than simply a capitalist world-economy, then Dutch hegemony and, potentially, contemporary globalization, represent glimpses of a capitalist world-economy process. In these contexts it is not states per se that are attacked but their impinging on profit-taking: the Dutch regents freed themselves from the Spanish state above and urban guilds below; economic globalization's prime target is the redistribution we call the welfare state. Contemporary profit-taking is dependent on ever-growing mass consumption and the spatial organization of the world appears to be bifurcating with states handling 'old-fashioned' welfare, and cities being centres for the latest consumptions. Again, the rise of the latter means accentuating the system's asymptotes. In this case the system begins and ends with episodes of economic politics encompassing a three centuries of a political economy world.
Whatever demise scenario is preferred, the basic implication of this paper is to separate the political from the state so as to address networks as well as mosaics in contemporary politics.
Abrams, P (1978) 'Towns and economic growth: some theories and problems' in P brams and E A Wrigley (eds) Towns in Society. Cambridge: Cambridge University Press
Albaladejo, P F (1994) 'Cities and the state in Spain' in C Tilly and W P Blockmans (eds) Cities and the Rise of States in Europe, AD 1000 to 1800. Boulder, CO: Westview
Arrighi, G (1994) The Long Twentieth Century. London: Verso
Aymard, M (1982) 'Introduction' in M Aymard (ed) Dutch Capitalism and World Capitalism. Cambridge: Cambridge University Press
Barbour, V (1963) Capitalism in Amsterdam in the Seventeenth Century. Ann Arbor: University of Michigan Press
Beaverstock, J V, Hoyler, M, Pain, K and Taylor P J (2001) Comparing London and Frankfurt as World Cities London: Anglo-German Foundation
Beaverstock, J V, Doel, M A, Hubbard, P J and Taylor P J (2002) ''Attending to the World: Competition/Co-operation and Co-efficiency in the World City Network' Global Networks
Boogman, J C (1978) 'The raison d'etat politician Johan de Witt' The Low Countries History Yearbook 1978, 55-78
Boogman, J C (1979) 'the Union of Utrecht: its genesis and consequencies' BMDGN 94, 277-407
Braudel, F (1984) The Perspective of the World. London: Collins
Bromley, S (1996) 'Globalization?' Radical Philosophy 80, 2-5
Coleman, D C (1969) Revisions in Mercantilsm. London: Methuen
Held, D, McGrew, A, Goldblatt, D and Perraton, J (1999) Global Transformations. Cambridge: Polity
Israel, J I (1989) Dutch Primacy in World trade, 1585-1740. Oxford: Clarendon
Israel, J I (1990) Empires and Entrepots. London Hambledon
Israel, J I (1995) The Dutch Republic. Oxford: Clarendon
Jacobs, J (1984) Cities and the Wealth of Nations. New York: Random House
Kossmann, E H and Mellick, F (1974) Texts concerning the Revolt of the Netherlands. Cambridge: Cambridge University Press
Misra, J and Boswell, T (1997) 'Dutch hegemony during the age of mercantilism' Acta Politica 32, 174-209
Modelki, G and Thompson, W (1988) Sea Power in Global politics, 1494-1993. Seattle: University of Washington Press
Pocock, J G A (1992) 'The Dutch republican tradition' in Jacon, M C and Mijnhardt (eds) The Dutch Republic in the Eighteen Century. Itahaca, NY: Cornell University Press
Ringrose, D (1989) 'Towns, transport and crown: geography and the decline of Spain' in E D Genovese and L Hochberg (eds) Geographic Perspectives in History. Oxford: Blackwell
Rowen, H H (1978) The Princes of Orange. Cambridge: Cambridge University Press
Schama, S (1987) The Embarrassment of Riches. London: Collins
Smith, W D (1984) 'The function of commercial centres in the modernization of European capitalism: Amsterdam as information exchange in the seventeenth century' Journal of Economic History 46, 985-1004
Storper, M (1997) The Regional World. New York: Guilford
Taylor, P J (1994) 'Ten years that shook the world? The United Provinces as the first hegemonic state' Sociological Perspectives 37, 25-46
Taylor, P J (1996) The Way the Modern World Works. London: Wiley
Taylor, P J (2000) 'Embedded statism and the social sciences 2: geographies (and metageographies) in globalization' Environment and Planning A 32, 1105-14
Taylor, P J (2001) 'Specification of the world city network' Geographical Analysis 33, 181-94
Taylor, P J (2002) 'Recasting world-systems analysis: city networks for nation-states?' in W Dunaway (ed) New Theoretical Directions for the Twenty-First Century World-System. Westport, CN: Greenwood
t'Hart M C (1989) 'Cities and statemaking in the Dutch Republic, 1580-1680' Theory and Society 18, 663-87
t'Hart M C (1993) The Making of a Bourgeois State. Manchester: Manchester University Press
Tilly, C (1990) Coersion, Capital and European States AD 990-1990. Oxford: Blackwell
Wallerstein, I (1974) The Modern World-System. New York: Academic
Wallerstein, I (1976) 'A world-system perspective on the social sciences' British Journal of Sociology 27, 345-54
Wallerstein, I (1980) The Modern World-System II. New York: Academic
Wilson, C (1957) Profit and Power. London: Longman
Wilson, C (1958) Mercantilism. London: Historical Association
Table 1: Populations (000s) in selected Castilian and Holland Cities 1570-1670
Sources: Ringrose (1989) and Israel (1995)
Edited and posted on the web on 10th May 2002
Note: This Research Bulletin has been published in J Friedman and C Chase-Dunn (eds) (2005) Hegemonic Declines: Past and Present Boulder, Colorado: Paradigm Publishers, 117-134