We still tend to think of politics in terms of what I shall call the ‘flags of all nations’ model. Politics is of course a much contested concept, but in this model it is perceived as state-centric and every inch of the globe is the territory of some state or another. We know that states differ in size and form, from continental-scale states such as Russia or the USA to small Pacific island statelets like Fiji. Since the 1950s, when as a boy I began collecting the cigarette card set of Flags of All Nations, the number of formally sovereign entities has risen to some 190. Political science has tended to follow folk wisdom in this matter, concentrating on politics within relatively homogeneous state-societies, and leaving politics beyond the state to a related sub-discipline, International Relations.

The widely perceived phenomenon of ‘globalization’ has led many commentators to question the continued salience and viability of the nation state. They see economic and social processes escaping from territorial limitations and becoming truly global. That is, they become supra-national or trans-spatial, taking place either without relevance to borders, or in cyberspace. The consequences for politics are twofold. In the first place, politics disappears as markets and electronic interchanges replace the need for more than basic local regulation. States become like local authorities in a global market system. Markets and the Internet absorb the coordinative function of states. In the second place, politics becomes redefined as a cosmopolitan planetary system based on supra-national entities, like the UN, and orchestrated by new global political forces, such as NGOs. In the first case, politics is an irrelevance: it just gets in the way of more efficient, non-territorial forms of social organisation and resource allocation. In the second case, it becomes a cosmopolis, a world political community, but one which must rely, if it is in fact possible at all, on political processes quite different from those of the nation state.

Globalization is a highly contested concept and the scale and scope of the phenomena grouped under it can be strongly questioned, as we shall see later. Yet the concerns about the eclipse of the nation state stemming from this currently fashionable concept do help to remind us that the nation state is a highly specific historical form. It developed between the sixteenth and seventeenth centuries in Europe. The rise of the modern state not only changed the territorial landscape of Europe: it also transformed our imaginative landscapes. Since the seventeenth century, we have come to see political power as inherently territorial. Politics takes place within the state as the exclusive governor of a definite territory. We also identify political territory with social space, perceiving countries as ‘state-societies’. This is not, as some modern social theorists believe, a conceptual error: it is a fact of political transformation, as sovereign territorial states reinvented themselves as nation states. Rulers and ruled became alike, and together they shared a distinct culture and institutions defined at least in part in opposition to those of other nations. Such differences were shaped in and by the conflict of states that has been inherent in the states system since its formation.

Territory and Political Power Before the Sixteenth Century

Clearly, politics has not always been identified with a power that claims to be the exclusive ruler of a given territory. Rulers and ruled have often been ethnically, culturally and linguistically different. Part of our difficulty in thinking about politics, power and territory is that our political ideas have been shaped by what we might call a ‘double territorialization’. The construction of exclusive territorial rule has been reinforced by a reinterpretation of the past as if it too conformed to modern models of political rule. The Renaissance thinkers who tried to conceive of the modern state, like Niccolo Machiavelli or Jean Bodin, tended to return to Greek and Roman models. They thought of post-feudal politics as a single political community with a single source of legitimate government and in which those who participated in politics shared a common culture. Thinkers like Machiavelli tended to identify the contemporary Italian city-states with the republics of Antiquity or, like Bodin, to compare the power of the French monarchy with Roman Imperium. This identification was made unexceptional by the general Renaissance practice of recovery of classical cultural models.

But Antiquity was actually very different in the relationship of space and politics. The Greek polis created an enduring model of the self-governing political community, a group of people who govern and are governed in turn and who shape every aspect of their lives through common institutions. The Greek city-state was, indeed, territorial and it was defined by a common culture. Yet the city-state was the exclusive ruler of a small territory, defined by the ability of the members of the governing class to meet in common and, therefore, by the distance that could be walked in a day. The culture of the city-state was doubly exclusive, of those in other cities, and of the large proportion of the population of the city who were not political participants: women, slaves, and resident aliens. City-states were limited by a definite and symbolically significant territory, which they neither could nor wanted to greatly extend. Their military power was restricted by their relatively small, free adult male population. States could either grow by founding sister cities in the form of colonies, as the Greek cities did in Sicily and Asia Minor, or they could subordinate others in the form of tribute empires, as Athens did the cities of the Aegean. The empire depended on the Athenian fleet, its maximum size of about 300 triremes determined by the number of rowers in the free population.

When protracted war stretched Athenian manpower and public financing by the gifts of its wealthy citizens and residents to the limit, as it did during the Peloponnesian War, then the limits of the Greek polis were revealed. Greece above the level of the city state was a cultural entity, not a political one. Greece was doomed once it was confronted by a major power, Macedon, that had absorbed its culture. The common elements of Greek society, the games and the great religious sites, did not provide a lasting basis for political institutions above the level of the city-state. The relationship between politics and territory established in Greece was thus quite unlike that of the modern state. The latter absorbed different self-governing cities and local powers into a common system of rule. Thereby, the new unified states were able to expand outside the core political territory and create overseas empires.

Rome was able to do both of these things. It expanded by incorporating Latin cities within its own political system and then by extending the rights of Roman citizenship to all qualifying free men. The Empire was a network of self-governing subordinate cities, of tributary kingdoms, and of tribes living by their own customary laws. Rome’s empire was, however, conceived as universal. It was without fixed boundaries and saw itself as expanding to include all humanity, as the surrounding kingdoms and tribes were conquered and civilized. Rome recognised no legal or cultural limits to its expansion – no political community had a right to exist except as subordinate to it. It was thus quite unlike the modern state, which recognises other states as part of a common states system governed by certain rules of interaction between sovereign powers. Rome’s only partner in the ancient world was Persia. But Rome’s relations with the Parthian and Sassanian Empires were complex and de facto. Persia was never accepted as a legitimate partner in a stable international system. Rome was thus a state without territorial limits.

Other types of regime had little place within modern political theory, except as that against which it defined itself. The Ottoman Empire was perceived as a form of essentially arbitrary power, against which Western sovereigns could be seen as rulers bound to respect both the laws that they had made and the fundamental constitutional laws of the state. Thus, for Bodin, the French king is not a tyrant because, although he may make and change ordinary laws, he may not tax without the agreement of the Estates General and must abide by the Salic Law, which prescribes succession by the first-born male. Like Rome, the Ottoman state was a universal empire. It was committed to imposing the rule of the Sublime Porte and Islam wherever the force of Turkish armies could carry. It was only in the eighteenth century that the Ottoman state began to make normal treaties with other states and only in the mid- nineteenth century that it was accepted by other European states as a full member of the international system.

Other pre-modern forms of rule have had an even less definite relationship between space and politics. Nomad confederacies, like the Mongol Empire, were non-territorial. Mongol rule recognised neither spatial limits nor the right to existence of subordinate powers, let alone legitimate independent rulers. Mongol rule relied on free movement over the steppes of Eurasia, on a military community fed by its own sheep and horses that moved with it. The Mongols incorporated other nomadic tribes within their confederacy, but their ruling elite was defined by clan membership and by a tribally exclusive shamanistic culture. Whether Mongol rulers were Muslim, Christian or Buddhist, the key nomadic clan rituals defined the exclusive culture of the core of those who controlled the state. Mongol rule was thus both culturally exclusive and non-territorial. The rulers lived in a facsimilie of a nomad camp, even when it was a permanent city like Karakorum with thousands of imported artisans providing goods for the court, or, as in Beijing, where the nomads created a miniature steppe in the form of a park at the heart of the imperial city.

Feudal forms of government that had developed after the decline of Roman rule in the West were non-territorial. They were based on personal ties between lord and vassal. The fief was a gift of land in return for obligation and service. It was contingent where it was located; territorial rights derived from personal obligations and could be changed. The unfree population was tied to the land by labour service and had no part in governance. Feudal elites served with their lords as dynastic acquisitions, conquest, or crusade dictated. Norman nobles, for example, could be found ruling in England, in Sicily, in the Holy Land, and on the Baltic coast. Typical of Medieval feudal states were borderlands where rule and title were ill defined. The frontier was not a fixed and determined line, but an area of disputed marchlands. Medieval Europe steadily expanded across such shifting frontiers - raiding, settling and conquering as the occasion arose. In the Baltic and in Spain Christian power steadily expanded after the year 1000 against pagan tribes and Muslim rulers respectively. In the Balkans, by contrast, the Ottomans expanded by similar processes against the Byzantine Empire and the various kingdoms of Serbia, Hungary etc. Such moving frontiers existed within Europe before they became typical of the neo-Europes of North America, Argentina or Australia. Note 1

Medieval states lacked not only exclusive control of a given territory by a single ruler but also a clear and coherent division of labour in governance. Powers competed to control the same spaces, claiming forms of territorial or functional rule that were ill defined in their scope and rights. Different entities would rule in the same space, often making contradictory claims upon the ruled. The Church claimed not only functional rule over religious matters and over clergy, the right to raise revenue, to have its own law, but it also disputed temporal power. The Pope claimed universal dominion over Christendom, as did the Holy Roman Emperor. Popes claimed the right to invest secular rulers in office, with the threat of veto, and hence to exercise hierarchical control. Kings in turn demanded the right to appoint bishops, for example, the French and the English kings. Not only was the Pope a secular ruler in the Romagna, but Prince Bishops elsewhere ruled their dominions and had armed forces.

Cities enjoyed extensive powers of self-government, either, granted as particular liberties or privileges by monarchs in return for money or military aid, or appropriated de facto, as with the Italian city states in respect of the Empire. Cities raised taxes, possessed armed forces, their own system of justice, and made treaties with rulers. Leagues of cities, like the Hanseatic League of northern trading cities acted as powers in their own right. The Hanseatic League’s cities coined money, donated armed forces to the common purposes of the League, as defined by its ruling institutions, made commercial laws, enforced trading privileges, and dealt diplomatically with rulers. The cities were drawn from territories that would later be part of Germany, Poland, Sweden, etc. The Hanse obtained exclusive trading privileges from rulers and the right to establish its own trading factories, with extra-territorial rights for its members. The League was a quasi-polity with common decision-making institutions and rules, its members being the participating cities. Within the League certain cities like Lübeck were its vital sources of financial and naval power.

Within Medieval cities the Guilds enjoyed functionally specific governance of their particular trade and its practitioners. Guilds determined entry to the trade, training, the quality of work, and prices. Thus regulating the production and sale of goods and the number and character of the labour force. The guilds were typically controlled by the leading guild masters. Cities were ruled by an urban patriarchate made up of guild masters and leading merchants. For the ordinary journeyman such functional governance was more important in ruling their lives than any other form of governance.

Within the Ottoman Empire the millets system of self-regulating religious communities allowed the subordinate peoples of the Empire a measure of control of their own affairs, according to their own religious laws. Christians and Jews had to pay additional taxes and were subject to other liabilities, but as religions of the Bible they were legitimate communities and had some rights to self-government. Ottoman cities were typically divided into closed quarters where the different religious laws prevailed. Throughout the medieval world, Muslim and Christian alike, laws thus depended on status. Priests, serfs, guild members, and so on, had different rights and were subject to different laws. Rulers and the ruled might be unlike in culture and rights. Nobilities and peasantries often had little in common. These are examples of the parallel governance by different groups within the same territory, but also of the functional governance of people across spaces.

Bodin’s various marks of sovereignty – to give orders but not to receive them, to make laws, to administer justice, to coin money, to tax, to raise armies, to deal with other rulers – were complexly distributed across territory before the sixteenth century. Various agencies could do these things – including raise armed forces and enter into relations with other rulers. The Hanseatic League, the monastic military orders like the Teutonic Knights or the Hospitallers, mercenary forces like the Catalan Company, city-states, bishopics- all acted much as later ‘sovereign’ states would claim exclusively to do and often across the same territory. Late medieval society was complex and political power was distributed differentially across it, creating multifaceted relations between space and politics.

The Sovereign Territorial State

From the sixteenth century onwards - starting with the centralising late medieval monarchies of England, France and Spain - states across Europe struggled and eventually succeeded in becoming the dominant powers within a definite territory to which they laid claim. Note 2 This is not the place to rehearse the reasons for the rise of the modern state or to look too closely into the temporality of the process with its numerous challenges, crises and conflicts. Suffice it to say that it has two defining characteristics:

I. Exclusion: all entities that are not exclusively sovereign are de-legitimised and eventually expelled from the international system. Thus the Hanseatic League, the monastic military orders, the Church as a pan-European institution, and the city-state are all either eliminated or marginalized. Both the Papacy and the Hanseatic League, having previously been major powers in the politics of the Germanic lands, therefore have no effective part in the Treaty of Westphalia of 1648 which ends the 30 Years War. Likewise, the Hospitallers, having played a vital role in resisting Ottoman advance in the Mediterranean at the siege of Malta in 1565, become an increasingly honorific irrelevance in the European system, until removed by Napoleon. The Holy Roman Empire, itself a ghost after Westphalia, lingers on ever more marginal in relation to the member states until abolished by Napoleon in 1806. Exclusion is thus a process that takes place at the international level, as well as in terms of redirecting the capacities of local and functional powers within centralising states.

II. Mutual Recognition: states acquire powers over their societies to a substantial degree because they recognise each other as exclusive rulers of a definite territory. A central aspect of such recognition is non-interference, states refraining from acting directly within the territory of another state. During the religious wars of the sixteenth and seventeenth centuries, such intervention had become normal practice, with states aiding religious dissidents in other countries either for reasons of ideological affinity and/or for reasons of balance of power. France and the German lands were torn apart by religious civil wars, fostered in part by external powers. The Treaty of Westphalia established the principle of non-interference in domestic religious conflicts, thus enabling states to begin to assert control over their societies. Until they did so, confessional loyalty tended to take priority over loyalty to monarch or state. Once a stable relationship between religion and territory was established, with states being recognised as either Reformed or Catholic, the process of building a form of political or the territorial loyalty of population, incorporating religion as part of state identity, can begin. The control of internal violence allows states to turn the aggression of their peoples outward, towards other states. Hence the succession of wars from the late seventeenth century into the early nineteenth century, ending with the peace of Vienna in 1815. States use wars to build forms of identification between rulers and ruled. Where this is successful, as in Britain, state, regime and nation come to be defined together in one form of legitimacy. Where this is less successful, as in France, state and nation are built against the existing regime, as after 1789.

Without territorialisation and exclusivity, there could not be political nationalism. Why should rulers and ruled be alike if political institutions are distributed by function and status, and different institutions compete to rule the same territory? How can subjects or citizens identify with a territory if it does not have definite symbolic spatial features? Nationalism typically claims not just an ethnos/national group, but also a territory that this group should inhabit as its homeland as of right. Without the prior existence of the sovereign state claiming a definite territory it is difficult to see how nationalism might arise. Typically, nationalism begins as the project of rulers seeking to harness people to states. Then rival projects begin, demanding unification or secession.

Without territorial nationalism, it is difficult to see how there could be democracy. Representative government can exist in heterogeneous polities: feudal estates or the councils of the Hanseatic League are ‘representative’ in that delegates speak for statuses or cities. People expect such delegates to be different one from another – to speak for their estate or the interests of their city. Democracy – where the government is chosen by the majority vote of the people – is quite distinct from this traditional representation. People have to feel sufficiently like their neighbours for a simple majority decision to be acceptable. Nationalism provides a basic cultural-social homogeneity that enables people to trust majorities.

Where such homogeneity does not exist, the introduction of democratic procedures simply exposes the conflicts of the wider society, as in the former Yugoslavia. ‘Nations’ are seldom homogeneous enough to begin with to suffice for political purposes; they are made up of ethnoi, the local customs and patois of various pays. Only homogeneous languages, customs and institutions will suffice for real political homogeneity, imposed by national schools, the army and universal military service, ‘national’ cultural institutions, and patriotic organisations in civil society, like the Boy Scouts. Given such a network of ‘nationalising’ cultural and social institutions, state and society merge. The continuation of such politics of nationalisation and the wars that are its inevitable outcome, as nations define their differences and establish their territories in conflict with others, leads to the unity of politics, culture and territory.

Such a unity gives borders a special salience. Borders existed before the modern state. They were either marchlands or zones of control, such as the border regions of the Roman Empire – Hadrian’s Wall, the Rhine frontier, and the limes on the Danube. Such systems controlled the movement of barbarian peoples on the frontiers of the Empire in much the same way as the Chinese frontier with the desert and steppe. Where borders mattered otherwise, this was mostly in terms of property rather than politics. The borders of manors were typically well defined, whereas those of feudal states were not. With the modern state, the frontier becomes not a disputed region or a zone of control, but a line. Then we expect things to be socially and culturally different on one side as against the other. The frontier is marked and policed; that is a secondary effect of exclusive governance of territory and of peoples being defined culturally. Within those EU states that are signatories of the Schengen Agreement, borders now matter less and less, in some cases less than borders between US states. Once such frontiers were closely guarded: now they are often imperceptible. The reason is not merely because the states in question are no longer in conflict, but because it is increasingly difficult to tell the peoples of adjacent states apart.

Applying the classic definition of Max Weber and amplifying it, we may define the characteristic features of the modern state as follows:

1. Hierarchy: the state is a superior political agency that decides on the role and powers of all subsidiary governments. Its rules – laws – are the primary rules.
2. It possesses or claims a definite territory with clear boundaries and defines who may or may not reside in it.
3. It has exclusive control of territory – ‘sovereignty’ means that no other agency has a legitimate or viable claim to rule.
4. It possesses a monopoly of the means of violence within the territory – the state alone controls and sanctions the political use of armed force.
5. Exclusive control by the state of the use of external violence – only the state can make war on other states.
6. Mutual recognition by states of each other and their territories.
7. A system of uniform and continuous administration throughout the territory of the state.
8. The separation of the affairs of state from the private affairs of both rulers and ruled.

As we have seen, the formation of exclusive control within the state’s territory owed a great deal to the fact that states were members of an international system. The modern state evolved not merely pari passu with the states system, but to some considerable degree as an effect of it. War was also a central fact in producing identification with the state, creating the territorialisation of populations that served as the foundation of nationalism. From the beginning, however, states did not just compete; they also cooperated, and from the earliest times their interactions were governed by norms. This normatively governed system created a radical disjuncture between the space governed by the states system and its rulers, and that beyond, the civilised and the uncivilised.

As Carl Schmitt realised, international law created the conditions for the control of the ‘non-civilised’: "A nation that was not civilised….could not be a member of the community of states. It would be regarded not as a subject but as an object of that community’s civil laws. In other words, it was part of the possessions of one or another of the civilised nations, as a colony or colonial protectorate." (Carl Schmitt, Land and Sea, p. 39). By the late nineteenth century, European states had extended the states system to the whole of the earth. They had converted most non-European territory into colonies and had reduced China to a series of spheres of influence. Colonial control created new spatial patterns in Africa, the Americas and Asia, divisions based on European conquest, and political entities that were artefacts of conquest. Most of the states created by de-colonisation are thus marked in their very territory by the effects of European conquest.

The modern states system created the principle of no territory without a state, and this principle was vigorously applied, squeezing out marginal areas and political entities that did not conform to the norms of modern statehood, like the pirate enclave of the Dey of Algiers. The sea, however, was another matter. States did try to claim the sea. The Treaty of Tordesillas, concluded by the Pope between Spain and Portugal in 1494 imagined that the globe, including the sea, could be divided like a land barrier. Neither country in fact could exclude the French, British and Dutch from raiding and trading.

Control of the high seas depended not on borders, but on the effective capacity to trade and on the dominance of one fleet over others. As the American strategist A.T. Mahan noted in The Influence of Sea Power on History, naval conflict and wars on land followed a different logic – one could not hold positions in the sea. In he early seventeenth century, the Dutch theorist, Hugo Grotius, developed the notion of the seas as commons open to all to trade and use. Law was based on fact not fiat, the seas could not be monopolised, nor did the passage of one ship diminish the utility of the sea to another.

The principal political projects that emerged from modern territorial statehood -dynastic absolutism, nationalism and democracy- all depended on a certain spatial order, of a claim to a given form of rule over a place. Dynasticism by right of inheritance, nationalism by the established co-residence of a people with distinct attributes, democracy as the will of the nation, link power and place together. The inter-state order, however, was never limited in this way. From medieval times, states had sought to foster and control commerce; trade bringing money and thus the sinews of war in train.

In the mercantilist period in the seventeenth and eighteenth centuries, states sought to annex and monopolise trade, in efforts to concentrate long-distance trade in their ports and in their ships. They sought to control access to trade through monopoly companies like the Dutch VOC or the English Hudson’s Bay Company. During the eighteenth century, these projects of monopolisation faltered as private traders sought to enter these markets. Adam Smith provided a rationale for free trade in The Wealth of Nations, and the origin of a new political principle, commercial liberalism.

In the nineteenth century, under British hegemony, a new international order was erected, based on the freedom to trade and the right of citizens freely to engage in private actions across the borders of their states. Commercial liberalism was the only doctrine that could build a genuine international order. Liberal sovereignty was founded on the freedom of private individuals to trade and, therefore, on the redefinition of the role of the state in facilitating such commerce. The liberal state was thus committed to definite international norms and thus used its power to enforce freedom to trade.

Hence the forced opening of China and Japan that refused such freedoms. Hence the stigmatisation of states like the Russian Empire that prevent the free movement of people and goods as backward countries. The ‘long’ nineteenth century (1815 – 1914) created a world free trading system based on growing levels of international trade, investment and migration. In 1914, Britain, France and Germany had attained trade to GDP ratios comparable to those of today, levels of capital export to GDP not exceeded today, and levels of mass migration that dwarf those of today. Between 1800 and 1930 40 million Europeans migrated permanently overseas. The world created by commercial liberalism was unlike the world of today, which is controlled by borders, passports and visas that in the nineteenth century were regarded as devices of barbarous regimes like the Tsars.

The nineteenth century was the period in which an open global economy was erected, one based on exchanges across the borders of sovereign states. These states, far from being threatened by such developments, saw it as their role to promote and protect free commerce. Liberal sovereignty implies a world of other states bound by rules of international civility and a system of free exchange by private individuals across the borders of states. The liberal state is thus inherently internationally oriented; its exclusive control of its territory is designed to promote an open commercial system within and without. Liberal sovereignty is state power applied to promote commerce, whether in the form of compelling reluctant ‘partners’ to trade, as in the Opium Wars the British fought in China, or in the domestic social policies designed to force workers to participate in the capitalist economy, as catalogued in Karl Polanyi’s The Great Transformation.

In the first half of the twentieth century, the system based on liberal sovereignty nearly foundered as the defeated and frustrated powers of World War I sought through neo-mercantilist economics and authoritarian nationalist politics to create an alternative order based on the state control of large economic areas (Haushofer’s Grossraumwirtschaft). The liberal powers, Britain and the USA, proved more resilient – aided by the USSR. After 1945 the USA rebuilt the liberal international economy, much as Britain had built it in the first place after 1815. After World War 2 two projects for the international order competed: American hegemony based on commercial liberalism and Soviet hegemony based on administered trade and satellite states. As in the nineteenth century, commercial liberalism proved far more capable of creating an international order in which states had a measure of autonomy and citizens were free to trade. The Soviet order crumbled precisely because it was not international; it relied on domination and the forcible control of movement of citizens.

Globalization

What most people mean by ‘globalization’ is the continued development of the international system of commercial liberalism. This system remains inter-national, not truly global, because it involves high levels of trade and investment between distinct national economies centred on major states. As Grahame Thompson and I have shown in Globalization in Question, a global economy based on supra-national market forces has not developed. Instead, most major companies continue to sell about 2/3rds of their products and keep the bulk of their assets in their home country/region. Despite the integration of short-term financial dealing, capital markets remain stubbornly local, with about 90% of investment sourced locally in the advanced countries, and migration is more highly controlled than ever. Borders function now not to exclude invading armies, but to keep economic migrants out of welfare states. Even the vaunted de-localisation of information flows must be set in perspective. The telegraph transformed the world in the nineteenth century; IT makes running that new industrial world a whole lot easier. The clicks of e-commerce depend crucially on the spatially localised and bricks and mortar based necessities of fulfilment.

If we look at globalisation in historical perspective, we can argue that after the massive contraction of international trade in the 1929 Great Crash, and after the effects of two World Wars in the twentieth century, the international economy is returning to something like its late nineteenth century heyday. Liberal states in the nineteenth century were limited governments but they were not weak rather, they used their power to promote commerce abroad and to clear the way for private industry at home. In some ways current policies in the USA, UK and Australia mirror this, promoting global free trade and domestic de-regulation in a ‘sound money’ economy.

It would be foolish to pursue this analogy too far. The modern world is not like that of the Pax Britannica. Equally, it would be foolish to judge the capacities of the modern state by the excesses of regulation and control prompted by the World Wars and their aftermath. The essential differences between the pre-1914 world and now are twofold:

First, a complex division of labour in governance has emerged in which states share power and governance capacity with supra-national agencies like the WTO, IMF and World Bank, and also with functionally-specific public, quasi-public and private bodies that control a plethora of things from the global radio wave spectrum to the insurance classification of merchant ships.

Second, states have chosen to associate into trading blocs, like NAFTA, the EU and MERCOSUR, in which they accept free movement of goods and – in the case of the latter two – people. States are increasingly embedded in larger entities to which they have ceded certain sovereign powers; for example, members of the WTO accept its adjudication over a wide range of trade-related matters that impinge on the scope of national policy, and the EU member states under the Single European Act accept EU legislation in matters facilitating the single market as superior to those of their national legislatures and enforcible as such in their courts.

States are not only sharing power ‘upwards’ within supra-national bodies, but also ‘downwards’, with bodies that, whatever the constitutional formalities, are no longer subordinate agencies that exist by the state’s fiat alone. Note 3 Economic regulation has increasingly devolved to regional governments and to functional agencies that possess more effective local knowledge in a volatile, flexible and rapidly changing economic system. Examples are the German Länder, the new Scottish parliament, or the English Regional Development Agencies.

States that fail to effectively participate in this division of labour, shunning international agencies and failing effectively to devolve power are not, however, strong, but weak. They seek to preserve political control at the price of sacrificing effective governance. For the paradox is that by sharing power with other institutions and agencies, states can stabilise their external environment and maximise their domestic economic performance, thus creating a less volatile environment in which to govern and enhancing the resources that feed into governance capacity. It is a mistake to view sovereignty in zero-sum terms, as if the state must become a weaker and less effective governor if it shares the tasks of governance with others.

The territorial state will not disappear. Indeed, this new division of labour in governance makes it ever more necessary. It becomes the key locus that ties the different levels and forms of governance together. States, because they are territorial, and if they are liberal democracies, are able to speak legitimately for their populations and to make international commitments that successor governments accept as binding acts of sovereignty. As such they provide the legitimacy for supra-national bodies, a derived or indirect legitimacy stemming from the will of the people.

Equally, territorial states remain our primary source of accountability and democracy in such a complex system. Their representatives remain, in theory at least, subject to domestic political pressure. Cosmopolitan forms of democratic governance are unlikely to develop because we still operate in a world shaped by nationalism. Citizens still identify with nation state. They are the largest bodies that can claim any sort of primary legitimacy.

International bodies are the preserve of elites, and the international technocracy needs the check of politicians directly answerable to national politics. Accountability of international agencies through national publics is at best indirect and weak, but strong supra-national democracy is just impossible. Democracy implies homogeneity; the world is just too unequal economically and too different culturally for the rich to submit to the decisions of the poor, or for one established culture to accept the internationalisation of the norms of another. Hence the unwillingness of the G7 states to give a greater say to developing nations in the core institutions of supra-national governance. Hence also the widespread resistance by other major cultures to international human rights norms that come in a box marked ‘made in the USA’.

We thus live in a world constituted out of apparently contradictory components: territorial sovereignty and commercial liberalism; nationalism and international accountability; supra-national institutions and the continued viability of nation states. The territorial states will remain a central component of the new division of labour in governance, even if it no longer has quite the salience it had when it appropriated political power from the complex division of labour in governance of the later Middle Ages. Politics is no longer exclusively territorial; on the other hand, it cannot hold together unless it is rooted in the democratic political will of territorial states that practice liberal policies, that are internationally oriented and that submit to supra-national norms.

1 On the structure of feudal society see Bloch 1965, on the frontier in Medieval Europe Bartlett and MacKay 1989, and on the nomads and the Chinese empire see Franks and Twitchett 1994.

2 On the rise of the modern state see Ertman 1997, Spruyt 1994 and Van Creveld 1999. On the formation of sovereignty see Krasner 1988, and on territoriality Ruggie 1993.

3 For a further discussion of this role of the state as the suture that binds different levels of governance together see Hirst and Thompson 1999 ch.9.

R. Bartlett and A. MacKay (eds.) Medieval frontier Societies Oxford, Clarendon Press 1989.

M. Bloch Feudal Society 2vols 2nd edition London, Routledge and Kegan Paul 1965.

T. Ertman Birth of the Leviathan Cambridge, Cambridge University Press 1997

H. Franks and D. Twitchett Eds. The Cambridge History of China, Vol.6, Alien Regimes and Border States 907-1368 Cambridge, Cambridge University Press 1994.

P. Hirst and G. Thompson Globalization in Question 2^nd. Ed. Cambridge Polity Press 1999.

S.D. Krasner "Sovereignty: an institutional perspective" Comparative Political Studies Vol21 No.1 1988 pp.66-94

A.T. Mahan The Influence of Sea Power upon History 1660-1783 London, Methuen 1965.

K. Polanyi, The Great Transformation Boston MA, Beacon Press 1957.

J.G. Ruggie " Territoriality and Beyond: problematizing modernity in international relations" International Organization Vol.47 No.1 1993 pp.134-72.

C. Schmitt Land and Sea Washington DC, The Plutarch Press 1997

A. Smith The Wealth of Nations 1776, Chicago IL, University of Chicago Press 1976.

H. Spryut The Sovereign State and Its Competitors Princeton NJ, Princeton University Press 1994.

M. Van Creveld The Rise and Decline of the State Cambridge, Cambridge University Press 1999.