The contemporary political order is still first and foremost a state system. One of its elements is an array of political centers. The future course of the state system is in dispute (Hanagan & Tilly 1999, McGrew 2000). It is a historical system, preceded by different political orders and highly variable over the course of its lifetime. A state system is necessarily more than a collection of separate states. It is somehow held together. In this paper I concentrate on two features (not the only ones) that have always played a role in cementing the elements of the state system: interstate cooperation and transnational group formation. I am interested in their evolution during the history of the state system and in their current status. More particularly I am interested in what way interstate cooperation and transnational group formation contribute to the distribution of political centres that the state system has produced.
The state system encompasses a series of political centers, as a first approximation the collection of capital cities. They show a locational pattern and signs of a hierarchical order based on differences in state power and network connectivities. Do interstate cooperation and transnational group formation result in additional political centers, do they reproduce and strengthen the distribution of political centers that resulted from other formative factors, or do they have locational patterns with different accents? In short, how do they affect the distribution of political centrality? For practical reasons I will now only consider these questions in the context of the European state system but I will take into consideration that the state system has transgressed its originally European character.
The state system is just one out of a series of possible political orders. Political orders are systems of rules and systems of rule. Systems of rules and systems of rule are mutually dependent. A system of rules can be thoroughly homogenized and universally applicable or differentiated and specific, though somehow dictated by shared first principles. A system of rule can be more or less decentralized. If pronouncedly decentralized it needs a coordinating mechanism to maintain the integrity of the system. If pronouncedly centralized it needs organizational efficiency to make its presence felt throughout the system. The degree of political centralization in a system of rule translates in the locational pattern of political authority and administrative capacity and results in a system of political central places. Their locational pattern is not only, not even primarily governed by Christaller-like considerations of consumer demand and transport efficiency. Their locations are conditioned by the results of power struggles of different kinds (religious, economic and military), spatially differentiated societal development and the patina of history. They are nonetheless central because it is from them that flows of authority emanate and towards them that flows of loyalty and recognition of submission are directed.
In a state system a large degree of decentralization is given. Otherwise its defining characteristic of state sovereignty would be void. At the same time considerable differences between state systems as to the degree of centralization based on variations in distinction between larger and smaller powers or a variable degree of hegemony can still be perceived (Wight, 1977). In any case a state system needs some overarching coordination mechanisms to maintain its integrity as a political order. These rules of the game may be completely informal without any institutionalization or they may be incorporated in permanent organization. They may be based on purely utilitarian principles or they may become part of a legal system that lacks the unequivocal powers of domestic legal provisions but that through its historical status and reference to principles of justice may gain legitimacy. However this may be, these rules provoke and guide political action that, I submit, in most cases is dominated from certain places. Actions are directed at sorting out problems that potentially undermine the system, e.g. by war, or that contribute to a superior functioning of the different individual states (e.g. functional cooperation). Congresses are held and organizations and their secretariats instituted. These actions are therefore instances of political center-formation and it is my aim to find out how this type of center formation adds to the existing centralities.
In the largely decentralized state system, as we know it, there have been long-standing and powerful processes in many places to make state and society largely co-incident. Society for all practical purposes became state society in those cases. However, as an historic system the state system had to put aside earlier social formations, locally bound or transgressing national boundaries. During the course of its lifetime the state system may have successfully moulded societies as state societies, but the alternatives have not disappeared altogether. As the state system becomes more disputed, the importance of alternative bases for societal formation may once again grow.
Transnational group formation in particular plays an important role in the way in which the state system is integrated. If completely successful, the state-society is a sharply delimited entity that can only coordinate its actions with the outside world through the state apparatus that supposedly speaks and acts on behalf of society. This is not completely impossible but it is risky, as disputes and misunderstandings may easily give rise to all out conflict. Transnational group formation, possibly a sign of emergent transnational society, enables additional links and the possibility of cross-cutting loyalties that supposedly help dampen serious conflict and thus contribute to integration of the state system. Structurally this repeats the debate between pluralistic and consociational democracy in domestic politics (Lijphart 1968).
Societies articulate their political ideals, fears and claims in different ways, but in most instances some institutional framework will be in place that generates central positions. These materialize in most cases in locations giving rise to yet another kind of political center. Local societies have their influentials who often gather in the bars along the public square in front of the town hall. State societies have their plethora of national associations, movements and political parties with their secretariats often close to the national government. And transnational society? How does it organize itself? Where does it congregate? It is my aim to indicate how this type of society generates its own centers and how this center-formation adds to the existing patterns.
In section 2 I rely mainly on a few historical sources to do a thumbnail sketch of earlier cases of interstate cooperation and earlier instances of politically relevant transnational group formation plus their consequences for political centrality and central places. Before permanent institutions slowly started to grow from the 19th century (O'Loughlin & Van der Wusten 1993), 'congress diplomacy' was the established manner of international multilateral cooperation between states. 'Congress diplomacy' in fact continued until the present time (e.g. Helsinki Conference 1975). Although there is a sizable literature on 'congress diplomacy' where the studies of political history and international law overlap, it is pretty difficult to obtain a clear empirically grounded overview over the longer run as many case studies are very specialized without due regard to the other instances of possibly the same phenomenon and specialists of international law tend to concentrate on the texts and to withdraw from the empirical implications (Lesaffer, 1999 is a recent effort to diminish the gap). As to the early manifestations of transnational group formation, cosmopolitan society and 'Weltbürgertum' particularly during the later stages of the 'ancien régime' and in the early 19th century are well recognized categories in the literature. However, they have not given rise to a clear empirical demarcation at different moments in time in different parts of Europe.
In section 3 I construct the current state of the proposed additions to the pattern of political centrality following the same two lines of inquiry, on the basis of the latest edition of the Yearbook of International Organizations (2000). Since the early years of the 20th century the Union of International Associations, that has always been in Brussels Belgium, has at regular intervals published this invaluable source book that tries to capture the degree to which states and state-societies interweave as seen by the evolution of the body of intergovernmental and international non-governmental organizations. This is a formidable task as these organizations come in very different guises. In the latest edition (nr. 38) a renewed effort has been made to register them as precisely as possible, to draw their profiles, to give some ideas about their activities and where they occur, and to indicate the centers from where they are run.
THE EMERGENCE OF INTERGOVERNMENTAL AND TRANSNATIONAL POLITICAL CENTRALITY IN THE WESTPHALIAN STATE SYSTEM
If it has a firm beginning the state system was born in 1648. It was brought into being by a series of international conferences in Westphalia that marked the end of a long period of warfare across Europe. By the agreements amongst contending parties it was determined who would now be the actors that made up the resulting political order (Caporaso 2000). Not that this was all that clear at the time. The German lands formed the Holy Roman Empire. They were nominally under the leadership of the emperor who as a rule governed from Vienna. But most of these lands belonged to different states, the more important rulers of these states behaved as sovereign powers on an equal footing with other European powers. Several cities had farreaching forms of autonomy under the distant reign of the emperor (Duchhardt 1990). The Republic of the Seven United Provinces (the Netherlands for short) had delegations of all the provinces present at the conference in Münster and it could not be taken for granted that they spoke with one voice (Prak 2002, 50-51). There were also people in the conference representing territorial powerholders, who were not a subject of these authorities.
Already some time before 1648 bilateral treaties had been made between parties that looked very much like states, but in 1648 most relevant states were present. From now on most of the time they recognized each other's existence if not necessarily the precise size and shape of each individual state and they agreed to the notion of mutual sovereignty based on territorial authority. Treaties involving subdivisions of states became increasingly exceptional. In this way the decentralized state system came into being containing a multitude of more or less sovereign units and their capital cities.
Europe's state system has functioned as a political order that knew a long series of wars. These were so to speak the unavoidable collisions of a set of actors who were first of all keeping as firm a grip on their territories as was technically possible while at the same time looking for opportunities for extension if the risks were not too large or withstanding pressure from 'colleagues' who thought that the risks of agression were not too large. There were always challengers, at least potential challengers, and defenders of the status quo. They operated in a nearly uninterrupted diplomatic dialogue between relevant partners and adversaries (Watson 1982), occasionally complemented or replaced by warfare. The art of diplomacy got its modern appearance from early in this period on the basis of the experiences among the city-states of Italy. Alliances were on the whole very temporary and lacked any organizational machinery. There were of course stronger and weaker powers and the fate of some powers changed dramatically over time (e.g. Prussia's, Russia's).
After episodes of warfare international congresses were often organized in order to establish a just peace and the new equilibrium position of the state system. Some congresses were also held with the express purpose to diminish tension, to prevent war or to rule out types of behaviour in warfare. "Diplomatic congresses were not regarded in the seventeenth century and subsequently as isolated events. They were climaxes of the dialogue between the independent states of the European res publica which went on all the time" (Watson 1982, 103). These multilateral negotiations, highlights of interstate cooperation, often resulted in one or several treaties that confirmed the latest status quo. The most important ones among them set precedents about the conduct of negotiations (e.g. in a neutralized environment, language used, treaty formulas) and about the principles to respect in the outcomes (e.g. balance of power) that acted as guidelines on later occasions.
It is not generally agreed when these climaxes or highlights of interstate cooperation occurred. Westphalia (1648) stands out as the most important initial occasion in this age and there is no question that Vienna 1814/5 and Paris and environment 1919/20 also qualify. In addition, there is a pretty general consensus on the importance of some of the other congresses. Apart from that, there is a long list of other gatherings that might or might not be called congresses. In some cases the participation is far less than general, in some cases real negotiations took place elsewhere, in some cases the congress was called but hardly met, in some cases the meeting is called a conference while the distinction with a congress is unclear. Phillipson (1916, 120-122) who has made an extended analysis of most peace agreements in the state system, comes down to the view that the distinction between congresses and conferences is no longer relevant in the 20th century. In earlier times congresses have tended to be more final, more dominated by frontline politicians, more solemn and less technical.
Places where congresses were held acted for a short while as political centers. Let me give two illustrations. "Nimwegen was the place where the dramatic decisions were made. In 1678/9 it was in fact the political focus of Europe, as Münster and Osnabrück had been thirty years earlier " (Roelofsen, 1999, 115, my translation). Friedrich von der Gentz, the famous secretary of the Congress in Vienna 1814/5 wrote about the period of the congress: "Vienna is enclosed in itself. We do not know anything about the exterior nor do we want to know what happens in the other countries. For everything that is powerful, interesting and enviable, is in this place. A journey of three months through the whole of Europe would not equal the value of what somebody in a favourable position within 24 hours can observe, learn and rejoice in Vienna at the present time."(Haüsler 1999, 140, my translation).
In the 17th and 18th century congresses could easily take months and sometimes years but even then their role was meant to be and remained temporary. Under the 'ancien régime' the selection of a congress city was a long drawnout process in which several considerations were important: religious services for all creeds possible, well connected from a transportation point of view, diplomatic immunity assured, city and its surroundings should enable private meetings, not too attractive in order not to encourage a prolonged stay of diplomats (Duchhardt 1999, IX) (although in a further chapter it turns out that Aix-la-Chapelle was selected in 1748 for its qualities as the best spa of the moment, Kraus 1999, 119). From the 19th century onwards there was a pronounced preference for places with high symbolic value (to underline victory, defeat, the damnation of a protagonist etc.).
For the cities themselves congresses were important occasions because they drew a lot of attention. While an opportunity for private conversations was important, a congress was also a highly public affair. The local environment was very much involved in most of them. The Paris congress of 1856 that ended the Crimean War was one of a series of events that underlined France's temporarily regained high status. In a city of Paris that was under Hausmannian reconstruction, there had been a successful Exposition Universelle and an official visit of Queen Victoria in the preceding year. Now in the French Ministry of Foreign Affairs at Quai d'Orsay an important new peace agreement was crafted. 'Quai d'Orsay' as the selfevident reference to French foreign affairs dates from this occasion. The negotiations were accompanied by massively reported opera concerts and grands dîners. As the treaty was signed hundredthousands of people celebrated in the streets of Paris (Wunsch 1999, 173-180). Only in modern times the publicity could be reserved to the mass media and the local environment be completely disengaged from the congress itself. An extreme example is the Potsdam conference (this may also well be called a congress) in 1945 (Angelow 1999). On all earlier occasions congresses stimulated tourism. Under the 'ancien régime' city governments made considerable investments to beautify the town hall (on several occasions the official meeting place), the churches, the theatre, to decorate the town, to improve personal security and public order, and to ensure that the necessary extra food would be available. In Aix-la-Chapelle public expenditure in 1748, the year of the congress was 15-17% higher than in 1745 and 1749 for which such figures are also available (Kraus 1999, 132).
In one of the 18th century treatises on international relations it was prescribed that the location of a congress should be "le Temple de la Paix et de la sûreté publique" (G. Réal de Curban, La science du gouvernement, quoted in Schilling 1999, 83). Slowly a series of conventions had developed concerning the order within the congress city. As regards relations with the outside world these were differently regulated depending on circumstances. When this seemed appropriate (because of uncertain or nontrustworthy overall state authority) congress cities were on several occasions officially neutralized. A perimeter was then agreed and demarcated by delegates. Armed forces were prohibited to enter. The highest authority inside the city or the perimeter for the duration of the congress was the collectivity of official representatives who conducted the negotiations. They nominated somebody as their representative and he was primarily responsible to maintain public order and punish crime.
To answer the question where congresses have been held more precisely, one should first of all enumerate the occasions that qualify. I will answer this question for the 1648-1920 period. I take 1648 to be the beginning of the state system. I take it that a new period starts in 1920. From that moment onwards interstate cooperation, not least in the European state system, has increasingly been managed through permanent institutions as exemplified at the level of the system as a whole by the League of Nations and then the United Nations and by the institutions of the European Union and its predecessors.
As mentioned earlier there is no question about Westphalia (1643/8), Vienna (1814/5), and Paris and environment (1919/20). In addition there is pretty general agreement about Nimwegen (1676/9), Rijswijk (1696/7), Utrecht (1712/5), Aix-la-Chapelle 1748, Paris 1763, Paris 1856, Berlin 1878. In sum, following these two lists congresses were held in and around the United Provinces outside the major cities during the first century of the state system. Only Rijswijk could possibly qualify as an existing political central place as it was the site of the Stadtholder's major palace just outside the political center of the United Provinces in The Hague. During most of this time the United Provinces were certainly a major power, perhaps for a while even a hegemonial power. Later on capitals of major powers acted as the gathering place of the most important congresses, particularly Paris.
What if we widen the scope and look at a more exhaustive list of congresses? To construct an acceptable list I relied to a large extent on Phillipson's Index of Treaties, Preliminaries of Peace, Conferences, and Congresses (Phillipson 1916, 473-486); this list is to a large extent replicated in Vol. I, Appendix XXX , table 31-42, 636 a.f. of Wright's monumental study of war (1942). I excluded places outside Europe (a considerable number of European congresses dealt to a considerable extent with outer European business, but European states were heavily involved). I ignored the armistice agreements in Phillipsons list, collapsed several cases into a single case when several places in each other's vicinity were simultaneously used for a congress (e.g. Westphalia: Münster and Osnabrück) or when a number of agreements were made in one place in a short period that were apparently related (e.g.various conferences held in Frankfurt in 1871 about the ending of the Franco-German War). I added several cases that are known as congresses but did not result in peace agreements of any sort for different reasons (e.g. Soissons and Cambrai in the 1720s, Laibach, Trappau, Verona in the 1820s, the Hague conferences in 1899 and 1907). All in all I ended with 133 cases. I was particularly interested how the frequency of congresses developed over time and to what extent these congresses were held in the political centers of the state system (in most cases the capital cities) and to what extent in the capitals of major powers (I used Levy's (1983, table 2.1, 48) classification of major powers as the basis for this distinction). The results are summarized in table 1.
Table 1: Number of congresses, those held in capital cities, and those held in capital cities of major powers per period 1648-1920
Over the whole period of the 'ancien régime' a congress was held on average every four to five years. The pace intensified tremendously during the Revolution and the Napoleonic period to a congress nearly every six months. In the years to the Crimean War the number of congresses fell to one in about every four years. Then the numbers increased again to one in every sixteen months or so. This recaptures our general impression of periods of lower and higher tension in the interstate system. One might perhaps have expected higher numbers during the early life of the state system given its still volatile nature but apparently the capacity to organize common understanding was still limited. It should however be added that in the short list of pretty generally agreed congresses mentioned earlier, numbers under the 'ancien régime' are relatively higher.
Over time the congresses were increasingly held in capital cities. The proportion grows monotonously over time. But congresses were not limited to capitals of major powers although these were important. Their relative importance did not grow over time. While the Concert of Europe period in the first part of the 19th century indeed shows up in these figures in a pronounced concentration of congresses in major power capitals, there appears to be some decline in the following period.
All in all these figures show that during the whole period and increasingly so, capital cities have acted as important temporary centers for interstate cooperation in addition to their regular role as the permanent nodes in the network of interstate relations. Major power capitals have always been particularly important in this respect. There have been additional temporary centers particularly in the early stages of the system in and around the United Provinces. The decentralized nature of this peculiar state that was at the same time a major power during this period may have played a role here. A wide variety of other cities has occasionally hosted nearly half of all congresses.
During the 19th century the first examples of functional interstate cooperation resulting in permanent organizations with secretariats also took shape. One of the very first was the Rhine Commission providing a forum for consultation between the states sharing the Rhine River. The fear of infectuous diseases spreading across countries and the intensification of international postal traffic resulted in other intergovernmental organizations. The number of such organizations gradually increased and several of them were subsumed under the League of Nations after 1919. The League meant a new step in the provision of intergovernmental machinery that was meant to be global in range and all encompassing in scope. For various reasons it was still very much orientated to Europe. Based in Geneva the League was dominated by France and Britain. After World War II much of what the League embodied was moved to the headquarters of the UN organization in New York (though some important institutions remained headquartered in Geneva) and some of it went into the construction of European cooperation for the most part based in Brussels. Functional organizations emerged in all sizes and shapes. Only a few of them are really global in scope. By far most of them are much smaller and much more regional, a sizable proportion geared to or headquartered in Europe. Over the course of the 20th century numbers have grown from about ten to the current 243 IGO's according to a restrictive definition (O'Loughlin & Van der Wusten 1993). It is intriguing that the two principal locations for the establishment of these permanent forms of intergovernmental cooperation (Geneva and New York) were not located in capital cities and the most recent one was outside Europe. While congresses were increasingly limited to capital cities, other options were used for these permanent intergovernmental machineries. In the next section we come back to the question how these different options are reflected in the current pattern of locations for the establishment of permanent interstate cooperation in Europe.
As the European state system emerged and provided a modicum of political order across Europe, there also developed an elite society stretching from the Atlantic to Russia, although it could be thin on the ground in places (Darnton 2002). Its members led a life that was pan European in its orientations and that was at the time called cosmopolitan. It was sharply distinguished from all those whose life experience did not exceed the range of the local community and that was in contradistinction to the cosmopolitan way of life called l'esprit de clocher, campanilismo or Kirchturmschattenbeschränktheit to underline the everpresent local churchtower. The civilized language among cosmopolitans changed from Latin to French, the Gelehrtenrepublik of an earlier period was replaced by a somewhat more broadly based Republic of Letters (that also encompassed other arts).
This transnational society subscribed in the 18th century increasingly to some version of the enlightenment stressing good manners, tolerance and the acceptability of the innerworld pursuit of happiness. In the later part of the 18th century Voltaire's estate in Ferney on the border between France and Geneva became one of its most important central places where cosmopolitans stopped and were invited to share a meal and listen to the master who started to command public opinion on a European scale. In European cities developed a set of institutions that scaffolded the society of cosmopolitans: coffeehouses, Masonic lodges, salons and academies. Close by newspapers were printed, again sometimes in French and aimed at a European audience, e.g. La Gazette de Leyde (published in the Dutch university town of Leiden). Darnton (2002) says in conclusion: "Print, talk and coffee combined to create a powerful new force everywhere in Europe: public opinion, and public opinion took a radical turn in all the great cities". The power of public opinion was directly transmitted to those in charge of the political order. E.g. Voltaire wrote 187 letters to Catherine the Great, that have been preserved.
Among these cosmopolitans was a lively interest in the improvement of the 'European republic of states'. Proposals for a permanent peace started to be published even before 1648, but from then on there was a sustained and growing series of widely discussed plans aimed at this laudable objective. Twentytwo of these peace projects printed between 1680 and 1800 have been discussed in the literature. They always contained a supra-national institution that in practice remained unattainable, although the contents of such proposals were well known among professional diplomats and politicians. "These private peace designs were one of the most cherished kinds of popular writing in the age of the enlightenment" (Duchhardt 1976, 28-34).
After the French revolution and the Napoleonic era this high culture society did not altogether subside, but it was for a long time increasingly covered by and mingled with a different process: the nationalizing of elites and masses. In hindsight we should call this elite European society transnational, but it only could become transnational after national state societies had become the most prominent features of societal construction. In the course of the 19th century cosmopolitanism continued as transnationalism. Obviously it had to adapt to the new circumstances. It had to function in the context of ever more powerful and imposing states that garnered everlarger amounts of infrastructural power. And at the same time new social categories became involved in transcending the limits not only of the church tower but also of the national boundaries. The internationalization of the working class is a case in point as is the increase in international mail traffic during the second half of the 19th century and the beginnings of international tourism for social layers beyond the upperclass.
These pointers to transnational group formation became increasingly organized in voluntary associations. E. g. the literary interest in the subject of peace during the enlightenment broadened and deepened in the 19th century in a transnational movement of peace associations with international meetings since the late 1840s. Around 1900 there were several plans to erect a new world capital that would somehow be the new political central place of the world (they were always somewhat curious blends of science, politics and religion, sometimes also sports). Candidate locations for one of the most discussed projects (started by a Norwegian sculptor) were Brussels, Bern, Constantinople, The Hague, New Jersey, Paris, Riviera, Rome. The emphasis is still clearly on Europe and on capital cities, or in any case administrative centers. But Patrick Geddes, the urban planner, proposed a world union of provincial towns in order to bypass the warmongering national capitals (Bosma 1993, 42-70). The Yearbook of International Organizations, that will be the basis for our portrait of the current state of international cooperation in the next section, comes from the same background and started in 1908. Among other things its various editions portray the emergence of transnational associational life during the 20th century. It has recently captured this notion by adding 'Guide to global and civil society networks' as its subtitle. Numbers of associations have since 1908 increased from a handful to at least 6357 in 2000.
The way in which these transnational groups had to accommodate with national state societies was perhaps most problematic where national state societies were less the fruit of longstanding social processes of modernization than of violent upheaval and enduring social conflict. Prominent examples of the painful constitution of national state societies are Germany and Russia. In Germany a very influential book on the topic was published in 1907 by the famous historian Friedrich Meinecke entitled 'Weltbürgertum und Nationalstaat'. It had 7 printings until 1928 and was republished in 1962. The book has two parts: 'Nation, state and world citizenship in the development of the German national state' (which covers the 18th and 19th century to 1871); and 'The Prussian national state and the German national state' (which is mainly concerned with Bismarck's political construction). Meinecke is at the same time acclaiming the German circle of world citizens (cosmopolitans, transnationals) out of whos midst developed the national idea on the basis of an ideal of freedom, and the iron hand of Bismarck who constructed a state that he considered a happy mixture of 'Macht und Geist' (might and mind). At the same time he was well aware of the tensions between these two driving forces and he wavered repeatedly about the success of the synthesis. In 1944 he wrote as an old man to a friend: "When we were young we wanted to unite organically in ourselves the heritage of the age of Goethe and of Bismarck - we imagined, that we should be able to succeed" (quoted in Meinecke 1962, XVIII)
A good Russian example of a similar tension between world citizenship and national state loyalty is in the career of the painter Ilja Repin. Born in humble circumstances in the province of Charkov in Ukraine in 1844, he made it into the Academy of Arts in St. Petersburg and traveled 'Europe' in 1873-1876 for the first time with a grant. In Paris he saw the first signs of the Impressionists (and met with and painted Turgenjew). He moved back to Moscow and then to St. Petersburg where he became a very well respected artist and close observer of the turmoils during the 1880s. At the same time he made four long journeys to 'Europe' in 1883, 1887, 1889 and 1894. In between he traveled extensively in Russia itself particularly to observe the countryside. Repin sold to the Russian elite, but he exhibited his paintings also in 'Europe'. He was influenced by midcentury French and Dutch realist painters and also by Rembrandt (whom he knew from the Hermitage, he disliked the Nightwatch when he saw it in Amsterdam). In a recent approving critical essay on his work, the art historian Van Os stresses the unusual number of sharp shifts and ruptures in style and content in his work that seem to result from constant changes in the relative impact of cosmopolitan and more narrowly focused nationalist circles that partly overlapped and in which he was active. They pulled him in different directions. During his career debates were raging in Russia about the proper orientation of the country as one other member of a well acquainted European family of nations or as a distinct nation that should nurture its own roots quite separate from the other Europeans. Repin could never find a permanently satisfactory position. He withdrew in 1903 to an estate in Finland and ended up outside Russia after the Revolution. He did not go back but was used as one of the predecessors of the infamous social realism that dominated Russian art in the 1930s (Van Os & Scheijen 2001).
In sum, interstate cooperation has slowly institutionalized during the lifetime of the state system. In earlier periods occasional congress diplomacy created ephemeral extra political centrality (but to a considerable extent and ever more prominently in capital cities) in a system of decentralized political centrality that only knew pretty stable configurations of international power distribution concentrated in the set of capital cities. As interstate cooperation became more formally organized new central places where secretariats resided emerged. Prominent examples are outside capital cities but it remains to be seen if this fits the general pattern.
A European wide elite civil society inspired by cosmopolitanism emerged as the state order came of age. It was organized around a narrow set of institutions for the most part located in the larger cities (most often capital cities) and sometimes in university towns, possibly continuing an even earlier tradition of networks of scholars. It also had the occasional temporary central place resulting from its function as the residence of a prominent member. As societies became more national and state power increased, the cosmopolitan way of life became transnational, it started to encapsulate broader social layers and it became more formally organized. During the 20th century it increased tremendously in numbers of associations and in number of people involved. It remains to be seen where its organizational centers were located and if we can see any change in the direction of new political centrality in this regard.
CONTEMPORARY POLITICAL CENTRALITY IN EUROPE
The following overview of the current pattern of political centers as regards interstate cooperation and transnational group formation for Europe is based on data from the last edition of the Yearbook of International Organizations. I use the data in the Yearbook mainly for two purposes. First of all, the intergovernmental organizations (IGO's) refer to cooperation between state machineries. They have secretariats, that coordinate and instigate the main activities. The question here is to what extent the places where these secretariats are established might be considered as emerging new political centers possibly challenging the current set of political centers. Secondly, the nongovernmental organizations (INGO's) are the organizational manifestations of a possibly emerging transnational society replacing, complementing or perhaps overlaying the existing set of national state societies. Again, the places where the secretariats of these organizations are established, might be considered as the focal points of this transnational group formation. Such focal points in my view also demonstrate political centrality. Finally I look briefly at venues of international meetings.
It is important to stress that international organizations in their turn are mutually interwoven and in this way form their own network. This is demonstrated in the Yearbook by means of a citations analysis of their statutory documents. A count has been made of the references that every single organization in these documents makes to other international organizations.
Table 2: Frequency of references to IGO's and INGO's in statutory documents in percentages compared to proportions of organizations (fig. 0.3.8 in Yearbook)
References to IGO's, the instances of interstate cooperation, turn out to be much more frequent than those to INGO's, the approximation of transnational society formation. Consequently certain initiatives in the field of interstate cooperation apparently provide most of the basic framework for all international organization efforts. The preceding data refer to the widest possible collection of international organizations. The smaller number of international organizations of the conventional type that are central in the remainder of this analysis attract much more attention than others in terms of numbers of citations (according to fig. 0.3.1 in the Yearbook). In the virtual space of meaning and significance (some of) the conventional intergovernmental organizations are the backbone of the whole network.
A few times I have indicated that the number of international organizations, be they governmental (IGO's) or nongovernmental (INGO's), is not a selfevident piece of information. The Yearbook has fifteen different types of organization for IGO's as well as INGO's. Four categories define well established, conventional types of organizations. In 2000 there are 243 conventional IGO's registered in these four categories and 6357 conventional INGO's. If we include all those registered as non-conventional we arrive at hugely different figures: 6743 IGO's and 47098 INGO's. But out of these 22% of the IGO's is currently inactive and for INGO's this is even 38%. In the remainder of the paper I use as often as possible the set of conventional organizations, thus the limited numbers, appreciating full well that in this way possibly interesting but still ephemeral developments may get lost.
Europe has for a long time been the core of the state system. By leaps and bounds other parts of the world have been incorporated in Europe's political order without acceding fully to the state system. In those other parts of the world European rule was imposed on older forms of political order sometimes resembling European forms, sometimes not. With the end of the colonial period the political order in the world as a whole has taken the form of a state system while the precolonial forms of rule and the colonial experience have produced state forms that considerably add to the variety of states that has always existed in the European realm. Europe's unique position in current networks of intergovernmental and transnational cooperation in terms of political centrality is indicated by the following results.
Table 3: Secretariats of IGO's and INGO's in Europe and European capital cities as a proportion of total numbers of organizations and of total numbers with secretariats in Europe
Europe is generally a very important location for international secretariats, but relatively the most important for INGO's. Within Europe capital cities are generally prefered locations for the establishment of international secretariats but this is relatively much more the case for IGO's than for INGO's. To put these figures in perspective it is useful to note that the US has 7 IGO secretariats and 622 INGO secretariats, out of which 4 resp. 38 are in Washington. Compared to Europe as a whole the numbers of secretariats are low. The US is in this respect better compared to individual, well endowed European countries. The proportion of IGO secretariats in the US hosted by the capital city is in line with the European average. The US proportion of INGO secretariats hosted in the capital is very low suggesting a dispersed pattern.
To what extent do these figures refer to organizations that are limited to and preoccupied with Europe? This is difficult to know but we can make an estimate. First of all, it is important to remember that most organizations are regional in scope (as regards aims and/or membership): this accounts for 71,2% of all IGO's and 75,2% of all INGO's.
At the same time 41% of all IGO's and 31% of all INGO's lack any European membership (table 2.1.4., 49 in Yearbook). In general these will be regional organizations outside Europe. In other words, according to this yardstick, 59% of all IGO's and 69% of all INGO's are either European or larger (often global, but including Europe). From another table (fig. 2.1.8, 108 in Yearbook) we learn that according to the aims of the different organizations 26% of all IGO's and 71% of all INGO's are preoccupied with Europe. In combination this suggests that in the case of the IGO's less than half of all organizations with European membership are specifically European, whereas in the case of INGO's organizations with European membership are in actual practice European organizations. With respect to the secretariats this apparently means that the strong presence of IGO-secretariats in Europe is partly produced by a sizable share of European organizations but to perhaps nearly the same extent (26 versus 21%) is it the result of wider-ranging organizations that have their center in Europe. In the case of the INGO's the preponderance of secretariats in Europe is the direct result of the large proportion of organizations with a European orientation.
In what places do we find the largest concentrations of secretariats? The following table ranks the ten most important locations hosting IGO and INGO secretariats in Europe and adds the cases of Washington and New York for comparison.
Table 4: Rank order of cities hosting IGO and INGO secretariats in Europe plus Washington and New York
The 10 first ranks have 64% of all IGO secretariats and 39% of all INGO secretariats in Europe. INGO secretariats have a more dispersed pattern while at the same time Brussels really is the primate INGO city (the production site of the Yearbook has been selected with foresight, and/or some bias in its coverage might be present). The two lists overlap to a very large extent. The only exceptions are Moscow and Helsinki in the lower part of IGO secretariat concentrations and Amsterdam and Rome in the lower part of INGO secretariat concentrations. Brussels, London and Paris are in a class of their own. Genève is the only non-capital city in both top ranks. In both ranks the emphasis is clearly on Western Europe, with Moscow (but only on the IGO list) as the major exception. Nordic interstate cooperation and Nordic transnational group formation show up in these figures as is the case with Spains role in the Hispanic world (but the attribution of many INGO secretariats to this function would be anomalous in light of earlier findings). In the US Washington exceeds New York in both rankings, while its small preponderance in INGO secretariats in US totals compared to the European average reiterates the more dispersed pattern of these institutions in the US. Both in terms of IGO and INGO the numbers of secretariats in Washington and New York are low compared to the European top rankings. But it is undoubtedly true that the weight of organizations has been neglected in this analysis and this may well counterbalance lower numbers to an extent.
What cities host international secretariats in Europe apart from a number of capital cities?
The information on IGO secretariats is deficient. In a non-exhaustive list numbering 81 out of 113 secretariats (fig. 2.1.8,94/5 in Yearbook) we find apart from the cities mentioned (capitals plus Genève) Strasbourg, Monaco and The Hague with a few IGO secretariats. The Hague is in fact the political center of the Netherlands, Monaco is a capital city of sorts and Strasbourg has an international vocation similar to Genève but of a more recent origin and more oriented to Europe. Information on INGO secretariats is more complete. It turns out that different countries have different proportions of secretariats outside the capital city. Table 5 provides an overview.
Table 5: European countries according to the proportion of INGO secretariats in the capital city (figs. 2.1.8c, 2.1.5c. in Yearbook)
There is a pretty clear divide between Eastern and Western Europe. In Western Europe secretariats are more dispersed, with France and Belgium as the most important exceptions. This apparently reflects the French centralizing tradition although it must be added that concentration in Paris is only just above the 50% threshold. Brussels in a number of ways emulated Parisian attitudes for a long time, but Brussels' emergence as European capital undoubtedly provides an extra impetus for the concentration of secretariats in the city.
A second suggestion for the explanation of the establishment of INGO secretariats outside capital cities refers to the individual attributes of cities that host them. One outstanding feature is the large number of university cities. Louvain-la-Neuve, the offshoot of the classic University of Louvain established just beyond the language border in Wallonia in 1970, is a pure university town, there is hardly anything else: it has 12 INGO secretariats.
The network of medieval and renaissance university towns hosts an impressive array of INGO secretariats: Louvain (23), Freiburg (5), Göttingen (2), Tübingen (3), Bologna (5), Leyden (21), Krakow (1), Lund (13), Uppsala (16), Basel (19), Cambridge (10), Oxford (10), to cite only the most obvious examples (neglecting the cases where other explanations are feasible as in the case of the Sorbonne, Paris).
There is still another way of looking at political centrality as produced by international organizations. This is the distribution of meetings across cities. This refers in its own context of transnational group formation to a similar phenomenon in the realm of interstate cooperation that we encountered in the preceding section, the European states' peace congresses. International meetings in the Yearbook are meetings sponsored by the international nongovernmental organizations mentioned in its tables plus national meetings of appropriate organizations with over 300 participants and at least 40% foreign participation during at least 3 days. It should be stressed that in this case the organizations are all those in the Yearbook, that is to say a large number that we earlier excluded. Table 6 presents the most important venues in Europe plus Washington and New York for comparative purposes.
Table 6: Most important international meeting places (according to number of meetings) in Europe plus Washington and New York in 2000
The ten most popular venues accomodate 31% of all international meetings held in Europe. This distribution is appreciably more dispersed than that of the two types of secretariats. Although we should expect this distribution to vary across the years, for the year 2000 the overlap with the location of the secretariats is considerable. The three top ranking cities are still the same, there are only two newcomers: Berlin and, barely, Barcelona. Genève and Barcelona are the only cities that are not (national) capital cities. The two US cities are in terms of meetings somewhat more preponderant than in the case of the secretariats. More in general the US (all cities together) is by far the most important hosting country for international meetings (13,8 % of all international meetings compared to France, 6,7%, UK 6,5%, Germany 6,3%).
In brief, in interstate and particularly in transnational organizations Europe is much more involved than other continents. A considerable part of the interstate organizations that have their center in Europe are more wideranging if not truly global, this is hardly the case for transnational organizations that for the most part have a European vocation if their secretariat is established in Europe. Europe's capital cities still count for a disproportionate share of interstate central functions in particular, but also of transnational central functions. This applies to the hosting of secretariats but also to the venues for meetings. Brussels, Paris and London are in a class of their own, Genève is the only non-capital city that shows up in the top of all the rankings. Transnational organizations show clear signs of dispersal of central functions. This is particularly the case in Western Europe and university cities play an important role in this respect. Meetings are still somewhat more dispersed. Compared to Europe the US hosts central functions in the order of the larger single European countries, but it is preponderant in terms of meetings. Center formation in transnational organizations in the US shows an even more dispersed pattern than is the case in most West European countries. Meetings in the US are also relatively dispersed.
CONCLUSIONS AND IMPLICATIONS
Interstate and transnational organizations are an important feature of international politics and have become more prominent, at least more numerous, in recent years. In Europe they have grown out of earlier, more informal and less permanent ways to keep the dialogue across state borders going by means of congresses of statesmen and the cosmopolitan networks of an elite society. Contemporaneous organizations not only communicate with other political actors, but they also communicate with each other and refer in many instances to other international organizations in their statutory documents. Congress texts refered to earlier precedents and in this way built up important political conventions, while social standards were set by example in the international society of (European) world citizens. In this way they all can be considered as networks, part of the tapestry of relations that knit the social world together at scales surpassing the units of the state system.
This part of the tapestry is not equally strong in all parts of the world. There are several options for macro-regions like Europe. Macro-regions may be highly connected internally and also richly endowed with links to other parts of the world. They may be internally closeknit but less well endowed globally. And there may be those who lack a dense internal network, but are linked into the global system and finally those that are short of links either within their own region or externally (Nierop 1994). Europe has at least since the inception of the state system been of the first type. This does not exclude that its history, also internally, may have been unusually violent at the same time.
Organizations and their international predecessors have centers. In this instance I have looked at the location of congresses, at nodes in the network of international elite interaction, at the secretariats of organizations and at venues for international meetings. Do all these different actions give rise to other, additional patterns of political centrality apart from the wellknown distribution of capital cities or do eventual concentrations just join the existing distribution of central places that embody the political order?
Europe, the subject of my enquiries, is a very important part of the tapestry of intergovernmental relations. A long series of peace congresses has contributed to the integration of its state system from very volatile beginnings and extremely violent eruptions over time. Nowadays, Europe is well endowed with forms of interstate cooperation geared to European matters, countries in Europe are deeply involved in interstate organizations that are globally active and at a European scale, Western Europe in particular hosts very many of the secretariats of both types of organizations. The tapestry of international nongovernmental organizations, expressions of transnational group formation, apparently has a different form. Historically it continues an extensive network of transnational elite communication that has assisted in softening the hard edges of national state societies, although by no means sufficiently to guarantee permanent peaceful integration. Europe's countries are now extremely well connected among each other in terms of transnational group formation. There are apparently relatively few global associations. Other parts of the world are much less well connected internally. European countries host a huge proportion of all secretariats because a huge proportion of all organizations have a European vocation.
Europe's capital cities have become ever more prominent in hosting peace congresses and they have always played a major role as nodes in the transnational networks of elite communication. They are also prominent in hosting secretariats. In this way the nodes in the tapestry of connections produced by the center-formation of international organizations first of all add to the pattern of political centrality that is already present. This refers in particular to intergovernmental organizations as expressions of interstate cooperation and somewhat less to international nongovernmental organizations as expressions of transnational group formation. Whereas these types of political center formation particularly underline the existing centers of Paris, London and Brussels within Western Europe, they also gives rise to center formation outside the capital cities particularly again in Western Europe. The best example is Genève but there is an intriguing role of university cities primarily as hosts of international nongovernmental organizations. This might reiterate a very traditional function that universities have fulfilled since the renaissance in Europe and that was also present during the onset of civil Europe for cosmopolitans in the 18th century. It is now possibly continued in the center formation that is part of the emergence of transnational society.
Capital cities are apparently well suited to act as hosts of international organizations. Jean Gottmann (in Gottmann & Harper 1990) appreciated the 'hosting environment' of capital cities as one of their main attractions thus providing room for the idea that in the competition for central place positions capital cities - and even former capital cities - are favourable locations per se. The results of this paper underline that observation. Obviously cities should not be looked at as passive sceneries showing their strong points as one passes by. City governments and national governments are strongly competing for the location of secretariats of international organizations. Brussels has done this occasionally in a very roundabout way since the 1950s but with astounding success despite a lot of intra-urban struggles (Lagrou 2000). Vienna has constructed an UNO-city as part of a major expansion and reconstruction plan for the city during the 1970s (Lichtenberger, 1993 ). The Hague has softly been competing on and off for an international role since the early part of the 20th century. After the early establishment of the International Court of Justice (the Peace Palace, constructed in 1913), it has for a long time been unsuccessful in attracting new institutions of any renown. During the last years this has changed and the city now has a very active policy resulting in the location of various organizations on a city axis that originally had been reserved for this purpose just after 1945 (Van der Wusten 1998, 2002). In university cities, on the contrary, the gown apparently is reason enough to establish secretariats in town. They may be less spectacular, but the network apparently has deep roots.
This is a thoroughly revised and extended version of a paper entitled 'Political centrality in the state system: the case of Europe', that was first presented at the AAG Annual Conference in Los Angeles (19-22 March 2002) and then posted on the GaWC website as research bulletin 91. I thank Professor C.G. Roelofsen, Utrecht University, for valuable advice on the literature about diplomatic peace congresses.
Angelow, J. (1999) Die Stadt Potsdam und die Potsdamer Konferenz von 1945, in: H. Duchhardt (ed.) (1999) 197-212
Bosma, J. (1993) Ruimte voor een nieuwe tijd: vormgeving van de Nederlandse regio 1900-1945. Nederlands Architectuurinstituut Rotterdam
Caporaso, J. (2000) Changes in the Westphalian order: territory, public authority, and sovereignty. International studies review . Summer 2,2, 1-28 plus other features in special issue
Darnton, R. (2002) An Euro state of mind. New York Review of Books, 28 February, 30-32
Duchhardt, H. (1976) Gleichgewicht der Kräfte, Convenance, europäisches Konzert: Friedenskongresse und Friedensschlüsse vom Zeitalter Ludwig XIV bis zum Wiener Kongress. Wissenschaftliche Buchgesellschaft Darmstadt.
Duchhardt, H. (1990) Altes Reich und europäische Staatenwelt 1648-1806. Oldenbourg Verlag München
Duchhardt, H. (ed.) (1999) Städte und Friedenskongresse. Böhlau Verlag Köln, Vorwort VIII-XI
Gottmann, J. & R.A.Harper (eds.) (1990) Since Megalopolis. The urban writings of Jean Gottmann. Johns Hopkins University Press Baltimore
Hanagan, M. & Ch.Tilly (eds.) (1999) Extending citizenship, reconfiguring states. Rowman & Littlefield Lanham MD
Haüsler, W (1999) 'Europa bin ich - nicht mehr eine Stadt'. Die Haupt- und Residenzstadt Wien als Schauplatz des Kongresses 1814/15, in: H. Duchhardt (ed.) (1999) 135-158
Kraus, Th. R. (1999) Aachen und der Aachener Friede von 1748, in: H. Duchhardt (ed.)(1999), 117-134
Lagrou, E. (2000) Brussels. Five capitals in search of a place. The citizens, the planners and the functions. GeoJournal 51,1-2, 99-112
Lesaffer, R. (1999) Europa. Een zoektocht naar vrede? Universitaire Pers Leuven
Levy, J.S. (1983) War in the modern great power system, 1495-1975. University Press of Kentucky Lexington
Lichtenberger, E. (1993) Vienna. Bridge between cultures. Belhaven London
Lijphart, A. (1968) The politics of accommodation:pluralism and democracy in the Netherlands. University of California Press Berkeley
McGrew, A. (2000) Power shift: from national government to global governance? In D. Held (ed.) A globalizing world? Culture, economics, politics, 127-168.
Meinecke, F. (1962, or. 1907) Weltbürgertum und Nationalstaat. Oldenbourg München
Nierop, T. (1994) Systems and regions in global politics. An empirical study of diplomacy, international organization and trade 1950-1991. Wiley Chichester 1994
Os, H. van & J. Scheijen (2001) Ilja Repin. Het geheim van Rusland (Exhibition catalogue). Groninger Museum Groningen
Phillipson, C. (1916) Termination of war and treaties of peace. Fisher Unwin London
Prak, M. (2002) Gouden eeuw. Het raadsel van de Republiek. Sun Nijmegen 2002.
Roelofsen, C.G. (1999) Von Nimwegen (1676-79) bis Utrecht (1712-13). Die 'niederländische Epoche' in der Geschichte des europäischen Kongresswesens, in : H. Duchhardt (ed.) (1999), 109-116
Schilling, L (1999) Zur rechtklichen Situation frühneuzeitlicher Kongresstädte, in: H. Duchhardt (ed.) (1999), 83-108
O'Loughlin, J. &. H. van der Wusten (1993) The geography of war and peace ,in: P.J. Taylor (ed.) Political geography of the twentieth century. Belhaven London, 63-106.
Watson, A. (1982) Diplomacy. The dialogue between states. Methuen London
Wight, M. (1977) Systems of states. Leicester University Press Leicester
Wright, Q (1942) A study of war. 2 vols. University of Chicago Press Chicago
Wunsch, S. (1999) Paris und der Pariser Friede 1856, in: H. Duchhardt (ed.) (1999), 159-184
Wusten, H. van der (1998) The state political geography is in. Tijdschrift voor Economische en Sociale Geografie 89, 1, 82-89
Wusten, H. van der (2002) New law in fresh courts. Progress in Human Geography, 36,2, 151-153
Yearbook of International Organizations (2000). Brussels 5 volumes. Edition 38.
* Herman van der Wusten, Department of Geography & Planning, University of Amsterdam.
Edited and posted on the web on 24th June 2002; last update 20th January 2003
Note: This Research Bulletin has been published in Political Geography, 23 (6), (2004), 677-700