In 1891, Henry Sidgwick defined territory as a political concept, inextricably connected with the idea of political jurisdiction. ‘The connection between a political society and its territory’, he wrote, ‘is so close that the two notions almost blend’ (1891: 213). On this reading, territory represents the geographical domain of political society and must be distinguished both from land (any physical portion of the earth's surface not covered by water) and from property (which as real estate is land owned and used by individuals or associations, typically under the auspices of, and regulated by, political authorities). This understanding of the concept was hardly surprising: for several centuries prior to Sidgwick's time, the territorial state had been the basic building block of the international order, and its exclusive right to control everything within its territorial borders was central to that order.
In the modern world, there is much talk of globalization and deterritorialization, but the basic organizational structure of the international order remains stubbornly territorial: indeed, the entire usable earth is divided into political units, each claiming jurisdictional authority over a distinct geographical area. States are the paradigm case, but jurisdiction can be held at both higher and lower levels: suprastate entities (such as the European Union) can be said to have territories, and so can substate entities (such as Quebec, Bavaria or Maine). Indeed, any politically organized group capable of exercising jurisdiction is a potential bearer of territorial rights; this includes indigenous peoples who claim as territory the land that they regard as their homeland and over which they claim specific (jurisdictional) rights, such as fishing rights or hunting rights, or other kinds of jurisdictional authority.
The concept of territory is complex in much the same way as is the concept of property. In each case we discover a bundle of different rights – claim rights, liberties, powers and immunities – that when held together with respect to some material thing add up to the fullest form of normative control over that thing (in the case of property, to what Honoré, 1961, in a famous analysis, called ‘full liberal ownership’) but that are also capable of being parcelled out and held by different agents. Each right within the bundle may call for a separate justification. Central to the idea of territory is the right to jurisdiction, which we understand to mean not only the right to create and enforce laws within the domain in question, but also the power to make changes to the territory's status, for example by incorporating it into some larger entity such as the EU, or by creating subjurisdictions in a federal system.1 However territorial rights are normally understood to include other rights beyond jurisdiction: rights to control resources within the geographical area, rights to control borders and regulate the flow of people and goods across them and rights to defend the territory against outside aggression (Simmons, 2001; Miller, 2012). These are typically assumed to be inherent in the very idea of a sovereign state, but we can envisage entities that have some rights of jurisdiction but lack other elements in the territorial bundle. This is true of most substate authorities: there are significant jurisdictional powers at the US state level, but no border checks between Massachusetts and New Hampshire, for example; and there can be rights to control resources (which in the Canadian federation are controlled at the substate level, by provinces) in the absence of other dimensions of territory, such as collective rights of self-defence or rights to control the migration of people or flow of goods across provincial boundaries.
Although the territorial state has been central to our international order for several centuries, it faces both practical and philosophical challenges. The most important practical challenges arise from rival claims to the same territory. This can take the form of a secessionist movement, as in Scotland, Quebec, Catalonia, the Basque area of Spain, East Timor, the Kurdish areas of Iraq, Syria and Turkey; military annexation, such as China's occupation of Tibet, Israel's occupation of the West Bank and Gaza, and, for cases of occupation combined with irredentism, Turkey's occupation of the northern part of Cyprus and the Russian occupation of Crimea and the eastern part of Ukraine; boundary disputes between neighbouring states, such as between India and Pakistan over Kashmir or Morocco and Algeria over the Western Sahara; disputed claims to offshore islands, such as between Argentina and the UK over the Malvinas/Falkland islands and between China and Japan over the Diaoyu/Senkaku islands; and, finally, rival claims to unoccupied territory, in the Arctic, the Antarctic and the seabed.
There are also philosophical challenges to territory, which mainly take the form of unpacking the bundle of territorial rights, and criticizing one or more of its elements. The territorial state's claim to monopoly control over the resources found within its domain is challenged by the idea that the world belongs to humanity as a whole, and its bounty should be used to benefit everyone who lives in it; no particular group can make a claim of entitlement or desert to exclusive control of natural resources. The right to control borders is also challenged, on the grounds that the state's right to exclude violates the human rights (to free movement) of those people who are prevented from migrating, and also serves to perpetuate the massive inequalities that currently exist between the inhabitants of rich and poor territories.
The territorial state has been much criticized, on these and other grounds, but increasingly in terms of its failure to grapple with many of the problems that arise in the world today, which require international or global coordinated responses. In Leviathan, Hobbes remarked on the fact that the territorial state system was effectively a state of nature: ‘Kings, and Persons of Soveraigne authority, because of their Independency, are in continuall jealousies, and in the state and posture of Gladiators; having their weapons pointing, and their eyes fixed on one another’ (1985/1651: 187). Nevertheless, he argued that the sovereign authority was able to provide its citizens with effective security, thus avoiding the miserable condition endured by individual men in the state of nature. We live, however, in an era of much greater interdependence, including global markets, global trade, global flows of ideas and migration. These in turn produce pathologies that require global responses: countries need to act in concert to combat international terrorism and crime, and to agree on effective ways of limiting greenhouse gas emissions in order to reduce harms to the environment, to future generations and to low-lying areas that will be devastated by climate change. Poverty too is exacerbated in many cases by global capitalism, where countries adopt inadequate safety standards, have poor or non-existent environmental regulation and permit child labour and other forms of exploitation, principally because they fear that they are in a competitive ‘race to the bottom’, as multinational corporations threaten to exit to cheaper, less regulated locations. None of these problems can be tackled by states acting independently. The competitive – indeed gladiatorial – posture of most territorial states is a serious obstacle to arriving at a better, more just global order.
What, then, can be said in favour of the territorial state? Its central feature is that it applies uniform rules and common policies to people living in proximity to one another. It makes sense to apply a single framework of law to people living close to one another, since they will unavoidably have more intense contact, and therefore more occasion for conflict. Such a uniform body of law both provides predictability and makes possible coordinated decision-making. It is superior to law organized on the personal principle, where each individual or each group in a geographical area may be subject to a different legal regime, since by applying laws equally to everyone on the territory, regardless of their status or identity, it advances values such as equal freedom and equality of opportunity. The benefits of a territorial form of organization, in terms of coordination, predictability, equality before the law and the achievement of social justice, are therefore considerable.
As we have seen, however, these benefits come accompanied by significant disadvantages at the global level. So a more complete justification would have to be comparative. What alternatives to a territorially organized state might we envisage? There are at least three.
We could imagine a global free market, in which the protective and other services that people need would be supplied on a contractual basis by organizations that had no territorial limits – so two people might be neighbours but yet buy healthcare, legal protection and personal security service from two entirely different suppliers. All of the goods that states now provide would instead be supplied by competing market-based associations. Some elements of this broadly libertarian vision were articulated by Robert Nozick (1974), though Nozick's own conclusion was that competition between what he called ‘protective associations’ would lead to one such association becoming the dominant provider of protective services in any given geographical area – so free-market anarchy would eventually culminate in the emergence of an ‘ultraminimal state’. Nozick's underlying assumption, however, was that the only grounds on which coercion could be exercised, whether by states or private associations, were to protect individual rights and ensure that contracts and other market transactions are free of force and fraud. So many of the functions that states now perform – whether providing welfare to those unable to work, ensuring equal opportunities in education and the job market, or providing public goods such as roads, parks and museums – would be dispensed with. Thus, there are significant costs attached to abandoning the territorial state, even assuming that the global market in protective services turned out to work efficiently in other respects
A second alternative is a global state. Obviously, a global political order would avoid the static and essentializing effects associated with hard boundaries between territorially distinct political communities, and would allow coordination to solve global problems. However, it is less clear that the global state would be either functional or just. As Rawls argued in The Law of Peoples, a world state (which he regards as a precondition for a global scheme of distributive justice) would lead either to tyranny or to lawlessness. Specifically, Rawls writes:
I follow Kant's lead in Perpetual Peace (1795) in thinking that a world government – by which I mean a unified political regime with the legal powers normally exercised by central governments – would either be a global despotism or else would rule over a fragile empire torn by frequent civil strife as various regions and peoples tried to gain their political freedom and autonomy.
One reason why we might think a global state would degenerate into tyranny is that states are not simply territorial entities that offer coordinating rules and dispense justice: their boundaries also define the demos as part of which the people can make decisions over their collective lives, thus enabling the members of the political community to be collectively self-determining. Citizen alienation is more likely when individuals cannot see their role in decisions that are made by distant central organs. It is one thing to be one voice in a hundred or a thousand, but quite another to be one vote in seven billion. One may also argue that an important precondition for unalienated citizenship is identification with co-members in a political project, which is unlikely to reach across the globe, but tends to occur in more local settings among people with commonalities of language, values and a shared historical experience of political self-rule.
Is there a third option, neither market nor global state? It is not difficult to find authors who, with the benefit of hindsight, are able to demonstrate the historical contingency of the territorial state, drawing attention to the other forms of political order that it managed to displace (see, for example Ruggie, 1993; Agnew, 1994; Spruyt, 1994). The suggestion here is that if we recognize this contingency, we will be better able to understand how forms of political coordination that lack a territorial basis may emerge – and indeed are beginning to appear in the contemporary world. The problem, however, is that many of these coordinating bodies are either established by states, or depend upon states for their stability (for example, as providers of legal systems that can be used to resolve disputes in the event that these cannot be settled internally). Outlining a fully fledged alternative to the territorial state has proved difficult. The most likely contender would be a functional theory that disaggregates the various tasks that states now discharge and assigns them to different institutions, some local, some regional and some global – so the territorial principle would not be entirely abandoned, but instead different spatial boundaries would be treated as relevant for different functions.2 By way of illustration, we might consider the concept of ‘ecological space’ favoured by a number of environmental theorists (see Hayward, 2014), which might be used to mark out the contours of ecosystems that need to be managed collectively by all those whose life plans rely upon them, regardless of whether they exist within or cut across existing territorial boundaries; thus Avery Kolers speaks of ‘interstitial settlements’ where groups with different ‘ethnogeographies’ overlap in space, and argues that in such cases individuals can be subject to multiple jurisdictions: ‘individuals can be governed under multiple jurisdictions concurrently, with the relevant legal system determined not by location but by activity, or even which computer program they are using’ (2012: 341).3 However, Kolers does not discuss how this understanding of territory can be implemented, in particular how government services can be provided, how legal relationships between different jurisdictions can be understood and conflicts between them sorted out. Others have proposed the partial unbundling of territorial rights: thus Nine (2014) uses the example of rivers whose waters flow between two or more states to argue against exclusive control of territory and in favour of a shared or joint sovereignty arrangement in such areas, and Armstrong (2015) argues against sovereign control of natural resources on the grounds that there are no good reasons to link jurisdiction over land to ownership of the usable or extractable resources that it contains. But in neither case do we find a head-on challenge to territorially based authority as such. At present, then, there is no clearly articulated third alternative, which means that the territorial state remains the principal – indeed, hegemonic – political form in the modern world, both practically and philosophically.
Showing that the territorial state has certain advantages that its would-be competitors lack does not, however, give us a full theory of territorial rights. Most notably, it does not explain how such a state can legitimately claim to exercise authority over any particular portion of the earth's surface. We need to solve this ‘particularity problem’ if we are to address the many cases referred to earlier in which territory is contested between rival states. Here we examine five influential theories of territory: utilitarian, Kantian, Lockean, nationalist and self-determination. As we shall see, the last three theories address the particularity question directly, while the first two do not.
We take Sidgwick as our representative of a utilitarian theory of territory that perhaps comes closest to the way territorial rights are currently understood in international law. Its central thesis is that over the area in which a state achieves an effective social order, it has a prima face justified claim to territorial rights. In his view:
the main justification for the appropriation of land to the exclusive use of individuals or groups is that its full advantages as an instrument of production cannot otherwise be utilised; the main justification for the appropriation of territory to governments is that the prevention of mutual mischief among the human beings using it cannot otherwise be adequately secured.
(1891: 239)
But what if two governments each claim that they are able to perform this function in a given area: should the territory be awarded to whichever is likely to create more happiness overall, as the utilitarian standard might suggest? In fact, Sidgwick sets the bar low, in such a way that territorial rights are awarded to any state so long as it is ‘exercising a tolerably effective and continuous governmental control’ over the territory in question (1891: 242). Thus, when he discusses secessionist claims, he argues that such claims have force only when the secessionists can demonstrate ‘some unjust sacrifice or grossly incompetent management of their interests, or some persistent and harsh opposition to their legitimate desires’ (1891: 217). Moreover, the preferences of the secessionist group have to be set against ‘the dislike of the community from which secession is proposed to lose territory that has once belonged to it, and to which it has a claim recognized by foreigners’ (1891: 219). Given the difficulty in weighing up the welfare gains and losses experienced by the two sides, and bearing in mind the disruption that a secession will cause, Sidgwick's position supports the claim of nearly all established states to their territories. At the same time, a utilitarian can raise no objection in principle to a successful conquest, and Sidgwick himself was perfectly ready to defend colonialism so long as the material interests of the colonized peoples were adequately protected (1891: ch. 18, §§7–8).4
Moving beyond Sidgwick, we can ask how a utilitarian might try to resolve the problem of indeterminacy just identified. Here we can distinguish an act-utilitarian approach that resolves specific territorial disputes by a direct application of the greatest happiness principle, and a rule-utilitarian approach that asks which rule for assigning territorial rights will maximize human welfare in the long run. An act-utilitarian will in particular want to investigate which claimant group will make the best use of the disputed land, for example, farming it or developing it economically in the most efficient way. But this might lead to instability, as over time different contenders can make the strongest claim; and it is also likely to be biased against indigenous groups with a close attachment to the land who might wish to use it sustainably but not ‘efficiently’ in the usual sense (for contrasting views about the appeal to efficiency as a ground for territorial rights, see Moore, 1998; Meisels, 2009). A rule-utilitarian, on the other hand, may want to follow Sidgwick's lead in defending the claims of the current holders, whoever they are, on the grounds that stability of possession gives political communities the strongest incentive to use land responsibly, knowing that it will pass in due course to their descendants. This, of course, provides no guidance in the case of territory that has yet to be occupied by anyone.
Kantian arguments for territorial rights also proceed at the general level, by justifying the state as necessary to the pursuit of justice. By the same token, they have difficulty in explaining which bits of territory each state ought to have, and, in the same way as Sidgwick's, reject a maximalist interpretation of the principle of justice, opting instead to define a threshold above which a state's territorial claims must be regarded as legitimate.
In Kant's own case, the argument proceeds in three steps.5 He first defends the right to occupy and appropriate objects as implicit in, and justified by, the exercise of individual freedom. Next, he points out that removing things from common use prevents other people from enjoying the object in question. Since the exercise of my freedom seems unavoidably to involve restrictions on someone else's freedom, the third step involves multilateral recognition of the obligation to respect the acquisition, transfer and use of objects in the external world, which we normally think of as ‘rights of property’. In consequence, according to Kant, people who live in close proximity to one another, and therefore cannot avoid interacting, are morally obliged to enter the civil condition and acknowledge a political authority whose coercive law can guarantee their property rights. There is no choice about this, so anyone meeting the proximity condition can be compelled by others to join the state if he refuses. The state's territorial rights correspond to the area over which it exercises jurisdiction. The justification for its exercise of those rights is simply that it thereby replaces ‘a state devoid of justice’ with ‘a rightful condition’ (Kant, 1996/1797: 90).
This theory fixes territorial rights only in the sense that the principle of proximity marks out a domain in which people are unavoidably interacting with one another, wherever that happens to be. It does not, however, tell us which territorial unit we should join in cases where we stand midway between one group of proximate people who are in the process of creating state A, and a second group who are creating state B; indeed it appears that both the As and the Bs might be justified in compelling us to join their political community, since otherwise we will be left in a lawless condition. Kant's argument cannot address this dilemma. Indeed, he avoids discussion of such a case by appealing to considerations that fall outside the Kantian theory itself: he noted that people tend, as an empirical fact, to be naturally grouped into linguistic or religious collectivities, and that membership in these cultural groups helps to define particular jurisdictional domains (1971/1795: 113–14). This empirical assumption also helps Kant to escape what might appear to be the cosmopolitan logic of his account: without appeal to that purely contingent (and, in his terms, unjustified) fact of people's preference to associate politically with others who speak the same tongue or belong to the same confession, there is no reason internal to the theory why jurisdiction should not be universal, why we would not end up in a global state.
Neo-Kantian theories of territory, such as those of Lea Ypi, Allen Buchanan and Anna Stilz, have had to grapple with at least two problems bequeathed to them by Kant. One is the problem of particularity just identified (what gives this state rights over that territory, with its specific boundaries?). The other is how to set and defend any given threshold of justice in order to define what counts as a legitimate state capable of holding territorial rights (why not assign territory to whichever state will do the best job, not merely an adequate job, regardless of the wishes of the current inhabitants?). Neo-Kantians have responded to the latter question by building some democratic requirements into justice. Thus Ypi lists as ‘essential criteria’ for legitimacy ‘the ability to guarantee the rule of law; to protect basic human rights; and to provide sufficient opportunities guaranteeing citizens' democratic participation’ (2014: 300). Buchanan's position is more nuanced: his minimal view is that any wielder of political power over a territory is legitimate ‘if and only if it (1) does a credible job of protecting at least the most basic human rights of all those over whom it wields power and (2) provides this protection through processes, policies, and actions that themselves respect the most basic human rights’ (2004: 247). But he adds to this that where ‘institutional resources allow for democratic authorisation’, full legitimacy then requires ‘all persons to participate as equals in the public processes for determining who shall wield political power’, that is, some form of democracy (2004: 256). Stilz's position is nuanced in a different way. Having initially stated that a state's claim to territory requires a system of law that ‘ “rules in the name of the people”, by protecting basic rights and granting the people a voice in defining them’ (2011: 578) – which seems to imply democratic government – she later concedes that there may be ‘forms of political participation other than democracy that suffice to render a state legitimate’. Thus:
I would support extending provisional territorial rights to nondemocratic regimes that (a) protect basic rights and institute the rule of law, (b) provide meaningful nondemocratic forms of political consultation and contestation, and (c) are reformist regimes, i.e. they aim at reforming the political culture in the long term, in a manner that is supportive of democracy.
(2011: 589)
Such nuancing seems unavoidable if neo-Kantians are to avoid the conclusion that most contemporary states have no legitimate claim to the territories they govern. Yet even the basic human rights condition cited by all three authors seems to disqualify many of them (as Stilz admits, citing Zimbabwe, Iran, Sudan, China and North Korea as examples). Thus neo-Kantians have to explain why justice-promoting states should be prevented from annexing or colonizing justice-deficient ones. Might they do this by addressing the particularity problem – the problem of identifying the specific territory that any state, no matter how legitimate, can claim?
Ypi (2014) tackles this issue by laying heavy stress on the de facto possession of territory, whatever its origins, but then coupling this with the proviso that full territorial rights can only be established by states subjecting themselves to an international political authority that will adjudicate all territorial claims. This solution draws inspiration from Kant, but appears to extend the powers of the authority more widely than Kant did in his ‘league of free states’.6 In other words, her solution to the particularity problem is a procedural one: states are (conclusively as opposed to provisionally) entitled to whatever the global association says they are. Buchanan does not address the particularity problem directly, but appears to believe that the territorial rights legitimate states currently hold are justified unless the territory in question has been unjustly annexed. His main concern is to attack what he calls ‘primary right’ theories of secession which would permit territorially concentrated groups within established states to demand independence without having to show that their basic rights were being systematically violated under the status quo (see Buchanan, 2004: ch. 8; also see Buchanan, 2003). Stilz appeals to the fact of occupation to define the group and the territory over which a state may exercise its authority, so long as the occupants in question are there through no fault of their own – that is, they have not unjustly displaced some other group. On this view a state cannot be legitimate if it usurps an already legitimate state, and neo-Kantians rely upon this principle to explain the wrongness of annexation. But they will find it harder to show why a just state could not gain territorial rights from a much less just state if they annex its territory through victory in a defensive war. This of course was precisely the situation of the Allied Powers who occupied Germany in 1945, a case addressed by Stilz (2011: 590-1; for a critical appraisal, see Moore, 2014). To resist the conclusion that the Allies could legitimately incorporate the divided territory of Germany within their various political projects, thereby gaining rights of jurisdiction, rights to resources and so on, she has to invoke the idea of a German ‘people’ who hold meta-jurisdictional rights over the territory in question – but it's not clear that a Kantian-derived theory, which justifies territorial rights in terms of the enactment of justice, has the conceptual resources to do so.7
As we have seen, both Kantian and utilitarian theories of territory are better able to explain what justifies territorial rights in general than to explain how states come to have such rights over specific areas of land. Either they end up heavily biased towards the status quo – awarding rights to any political authority that is able to establish a threshold level of social order or rights-protection – or, in the case of some versions of utilitarianism, they recommend awarding rights to groups that appear not to deserve them (such as invading colonizers who can claim that they will increase the productivity of the land they are about to occupy). For a theory that begins with landholding and treats territorial rights as derivative from this, we turn to John Locke and contemporary neo-Lockeans.8 Their common starting point is the Lockean theory of property in land, which claims that, antecedently to any political authority, individual persons may acquire such property by ‘mixing their labour’ with the land and thereby improving it (Locke, 1988/1689: Second Treatise, ch. 5). Locke then asks us to imagine persons, either individuals, or loosely associated persons (e.g., families), many of whom have property in land, combining together to create a state. Individuals in the state of nature, Locke argued, would consent to majority rule, to obedience to, and support for, law (within the limits of natural law) and, importantly, for this account of territory, consent to incorporate their rightful landholdings into that territory over which the society will have jurisdiction (see Simmons, 2001: 313). The crucial passage in Locke linking the creation of authority over persons and authority over territory is the following:
By the same Act therefore, whereby any one unite his Person, which was before free, to any Commonwealth; by the same he unites his Possessions, which were before free, to it also; and they become, both of them, Person and Possession, subject to the Government and Dominion of the Commonwealth, as long as it hath a being.
(1988/1689: 348)
Legitimate territorial right is established, on this account, through the subjection, by free consent, of persons and their land to state authority. This has the advantage of linking people and land together prior to the creation of political authority. It also explains the domain of jurisdictional authority: the boundaries of the state are coextensive with the set of properties incorporated either by individual owners or the holders of common land (such as parishes) when political society was created. However, it encounters the problem that the domain created in this way might not be continuous. Because individual consent is required to make political authority legitimate, there may be dissenters whose land is interspersed with the properties of those who are willing to sign the social contract. If we think of the territorial state as an institution with fixed and regular boundaries whose laws are applied consistently across the whole of the area within those boundaries, the Lockean theory provides no guarantee that such a state can be (legitimately) established in any given area.9 Moreover, it might seem that those who initially give their consent might later choose to withdraw it, leaving the state's territory with internal gaps occupied by property-owning dissidents.
Contemporary followers of Locke disagree over whether this should be seen as a failing. We referred earlier to Nozick, whose ‘ultraminimal state’ tolerates ‘independents’ living within the area in which one protective agency has achieved dominance. Hillel Steiner defends the right to dismember territory: ‘precisely because a nation's territory is legitimately composed of the real estate of its members, the decision of any of them to resign that membership and, as it were, to take their real estate with them is a decision that must be respected’ (1996: 144). Others, however, follow Locke himself in arguing that the transfer of land to a Commonwealth, once undertaken, is irrevocable and binds those who might later inherit the property. For instance Simmons claims that, in Locke's theory: ‘when people consent to make or join a political society … their consent should normally be understood as consent to whatever arrangements are necessary for a peaceful, stable society’ (2001: 313). This is what justifies Locke's insistence that property once incorporated cannot be detached from the Commonwealth and that ‘subsequent holders of that land will also be bound by the obligations of membership, including subjection of the land to state jurisdiction’ (Simmons, 2001: 313). Another reading of Locke draws attention to his remark that once political communities have been formed, their territorial boundaries are settled by ‘compact’ between them, thereby severing the connection between individuals' property rights and states' territorial rights (Van der Vossen, 2015). This, however, would bring Lockean theory into much closer alignment with the Kantian theory discussed above, and deprive it of its distinctive method of determining the legitimate territorial boundaries of states.
Contemporary nationalist conceptions of territory begin, like Locke's, by treating the link between people and land as more primitive than the state's. In this case, however, people are understood collectively as a nation – ‘a community (1) constituted by shared beliefs and mutual commitments, (2) extended in history, (3) active in character, (4) connected to a particular territory, and (5) marked off from other communities by its distinct public culture’ (Miller, 1995: 27). This builds the idea of a national homeland into the very definition of nationality. The connection between people and land is forged in two main ways. First, by interacting with the land over a long historical period and shaping it so that it better serves the needs and interests of its members as defined in part by the national culture, the nation has a strong claim to enjoy the fruits of its collective labour, which it can only do if it retains possession of the land itself. Second, the particular tract of land that the people have occupied over time acquires a special significance for them, as they bury their dead in certain places, build monuments to significant historic achievements or losses, and so on (see Miller, 2007: 214–21; 2012: 257–62; Meisels, 2009: esp. chs 3 and 7). In light of this intimate connection between the nation and its territory, a state that effectively represents the nation can claim territorial rights in that area so long as it also satisfies general criteria of legitimacy such as those discussed earlier. The nationalist argument is that only a state organically linked in this way to the nation – ideally but not necessarily a democratic state – can ensure that the material and cultural value that the land holds for the people is safeguarded.
This argument has to confront a number of challenges. The most radical asserts that nations are largely fictitious entities, at least if we are thinking about most contemporary states, which are predominantly multicultural and/or multinational (Brilmayer, 1995; Barry, 1999: 42–3; Watner, 2010: 252–3). Thus there is no culturally homogenous group of the kind that is presupposed when claims are made about land being physically shaped and imbued with cultural significance by the nation. Or, if such a group can be identified, it will turn out to be an ethnic ‘core’ within the state, with the worrying implication that these are the ‘true’ owners of the land. A second challenge points out that those who have been responsible for adding material value to the land are not necessarily the same group as those for whom the land has cultural significance. According to Stilz, ‘much of the improvement of land that goes on within a national territory – the construction of houses, churches and the like – is not carried out by the nation but by individuals or private associations. “The nation” does not mix its labor with these objects in any sense except metaphorically’ (2011: 577). A third challenge points to cases in which rival nations make claims to the same piece of territory, and are each able to advance arguments about physical improvement and cultural significance in support – the city of Jerusalem, for instance. What use is the nationalist theory in these circumstances? In reply, nationalists argue that their approach will still provide guidance, by allowing us to evaluate the relative strength of the competing claims, and that in cases such as this a more complex political arrangement, such as consociationalism or joint sovereignty, may be the solution that the theory prescribes (see Miller, 2014).
A final challenge to the nationalist theory is that it appears to grant nations territorial claims to places that have symbolic meaning to them, whether or not the places in question are currently occupied by co-nationals, and whether or not the territory formed by ‘joining the dots’ would be continuous or not. Thus, Serbian claims to the territory of Kosovo centre on the site of the battlefield where in 1389 the Serbian Prince Lazar fought against the invading Ottoman army – an event that now marks the most important day in the Serbian national calendar. But this site is now deep in territory occupied by Albanian Kosovans, so in order to reunite the Gazimestan monument that commemorates the battle to Serbia itself, a large chunk of Kosovo would need to be transferred, along with its unwilling population. What this reveals is that the two components of the nationalist argument – the investment of collective labour in transforming the land, and the cultural significance of place – can in some circumstances pull in opposite directions. One way to resolve the conflict is to say that claims of cultural significance only have force in the case of land that either now or in the recent past has been occupied by the nation in question. Otherwise the claims can only be claims of access to monuments or sacred sites, not claims to exercise territorial control.
Self-determination theories of territory have the same overall shape as nationalist theories, but they seek to avoid the difficulties of assuming culturally homogenous nations by ascribing territorial rights to a ‘people’ understood as a political rather than a cultural unit. Territory is here conceived as the site of collective self-determination. On Moore's account, states (or substate units) hold territorial rights by acting as a vehicle of self-determination for some group G (Moore, 2015: esp. ch. 3). However not just any group can qualify as a potential holder of territorial rights. Three specific conditions are required: political identity, political capacity and political history. The political identity condition requires that the group is united by a shared aspiration for wide-ranging powers of jurisdictional authority or political control over the territory. The political capacity requirement refers to a predicted and/or demonstrated ability to exercise self-determination and maintain effective forms of governance. The third condition requires that the members of the group have a history of shared practices and mobilization in terms of political projects. Unless these conditions are fulfilled, a group will not be equipped to exercise jurisdictional authority over an extensive territory. The territory itself over which jurisdiction is to be held is determined by the fact of occupancy. Occupancy is more than just physical presence: it requires a stronger connection to land. The group must occupy the land legitimately (meaning that it has not displaced some other group), and it must be rooted in that geographical space by the individual life plans and collective and political projects of the group's members. In short, when group G (which meets the three conditions above) legitimately occupies territory T (understood as geographical space), it can be said to hold territorial rights over T.
There are three important challenges to self-determination theories. First, connection to land, in terms of group occupancy, can identify heartlands of groups, but does not offer precise guidelines for boundary-drawing because, at the fringes, more than one group can claim the same land. It does not therefore provide an easy solution to border disputes. Second, the emphasis on collective self-determination as the value that territorial rights realize seems to apply most straightforwardly to groups that engage in democratic self-determination, because then we can be confident that the state is the vehicle of the group's self-determination. This, however, may seem too restrictive a condition. Third, such a theory, with its emphasis on capacity, cannot assign territorial rights to people who inhabit failed or conquered states.
Self-determination accounts could respond to the last criticism by pointing out that, in contrast to theories that make territorial rights dependent on the state's achieving a certain standard of justice or social order, these theories vest them directly in groups with the requisite political capacity to be self-determining, regardless of whether they are currently able to exercise it. In the case of a failed state, for example, the normative implication is not that external actors are entitled to claim jurisdiction, but that they may have an obligation to help the rights-holding people create the conditions under which they can be self-governing. In reply to the charge that only groups with democratic institutions can be awarded territorial rights, Moore (2015) defends a precautionary principle for institutional design, which entails that where a state governs on behalf of a group, we should assume that it acts as a vehicle of self-determination for that group unless we have good reasons to think otherwise. Even though democratic institutions are usually optimal for achieving self-determination, they are not always necessary.
In this final section, we ask how theories of territorial rights can respond to the practical and philosophical challenges we identified in our introduction. The main practical challenge concerned contested territories, where the solution will depend on which theory we favour. Take secessionist demands as an example. Utilitarian and Kantian theories, as standardly formulated, are hostile to secession unless it can be shown that the state that the seceders are trying to leave is manifestly failing to protect their rights and legitimate interests. Some neo-Kantians would make an exception for groups that have been incorporated into the state against their will and without just cause. Notice, however, that the status quo bias implicit in these theories has the paradoxical consequence that a secessionist group will lack territorial rights up to the moment at which it succeeds in its struggle for independence, at which point, and providing it creates a political entity that meets the threshold these theories set, it gains full entitlement over the territory it commands. Locke's theory, in its original formulation, is directly hostile to secession, since the initial transfer of jurisdiction over property to the state is irrevocable. In contrast, those such as Steiner who offer a libertarian reading of Lockean theory leave the door wide open to secession by any group able to command a majority in a compact territory (for what this might mean in practice, see Beran, 1988). For nationalist and self-determination theories, the issue is whether the seceding group qualifies as the kind of agent entitled to claim and exercise territorial rights. For nationalists, this will often be complicated by the fact of multilevel identities: if the secessionist region also contains many individuals who identify with the larger nation of which it forms a part, then the case for outright secession (as opposed to some form of devolved government) may be weak (see Miller, 1998). Self-determination theories too may award competent subgroups some forms of non-exclusive jurisdiction as an alternative to full secession.
In responding to the philosophical challenges, we should begin by observing that all the theories we have examined justify territorial rights by appeal to the interests, broadly conceived, of the inhabitants of the territory in question. Thus they are vulnerable in principle to cosmopolitan critics who point out that the interests of those who live outside the territory are ignored. Sidgwick, for example, candidly admitted that his treatment of territorial rights, and associated questions such as the right to immigrate, assumed what he called ‘the national ideal of political organization’. If one were to switch instead to the cosmopolitan ideal (which ‘is perhaps the ideal of the future’) then the state's task would be ‘to maintain order over the particular territory that historical causes have appropriated to it, but not in any way to determine who is to inhabit this territory, or to restrict the enjoyment of its natural advantages to any particular portion of the human race’ (1891: 308). For Kant, too, a full theory of territorial rights had to recognize the principle of ‘cosmopolitan right’, deriving from the fact that ‘all nations stand originally in a community of land’ – in other words, that the earth was originally held by the human race in common (1996/1797: 121). Kant drew the relatively modest conclusion that each person had the right to travel abroad for purposes of engaging in commerce, and when he did so ‘he must not be treated with hostility, so long as he behaves in a peaceable manner in the place he happens to be in’ (1971/1795: 106).
Their underlying cosmopolitanism notwithstanding, neither Sidgwick nor Kant directly challenged the right of a legitimate state to determine who should be allowed to settle permanently within its borders. But others have asked why territorial rights should include the right to prevent strangers from entering. Once again the answer will depend on the precise theory of territory that we invoke, but broadly speaking it will turn on the conditions that are necessary for a territorial state to function effectively. Immigration changes both the set of people who fall within the jurisdiction of the law and, in democratic states, the citizen body that is responsible for making it. If, like Kant, we imagine state representatives deliberating together to establish rules to govern their intercourse, it seems very likely that the right to control borders would be accepted as one such rule, since many internal policy decisions depend on the state's ability to exercise it. Against that might be set the human rights of immigrants who would be denied protection if barred from entering. But this would only establish a limited right on the part of refugees to enter some state that could provide them with shelter.10
The final challenge we shall consider concerns the natural resources that lie within the boundaries of each territorial state. In some intuitive sense it seems that different groups of citizens are arbitrarily advantaged or disadvantaged by the resources that lie unearned beneath their feet. Yet it is hard to see how jurisdiction can be exercised effectively if it does not include the right to legislate on (and thereby control) the use of resources that fall within the territory. A possible solution to this dilemma is to tax each state for the resources that it either possesses or uses, the funds raised in this way to be distributed either on an equal per capita basis to all of the world's inhabitants (see Steiner, 1994: ch. 8; 1996), or to people living in countries whose per capita income takes them below the global poverty line (see Pogge, 1998; 2002b). But the problem here is to find a way of valuing unimproved natural resources that is impartial in the sense of being independent of any particular political community's beliefs about the appropriate use of a particular resource: what members of one society may regard as a holy site, others may regard as merely a lucrative development opportunity (for an elaboration of this point, see Miller, 2007: ch. 3; Moore, 2012). If territorial rights can be defended, therefore, on one or other of the grounds canvassed in this essay, some degree of arbitrariness in access to resources may be the unavoidable price.