Cultural nationalist accounts

Philosophical liberal nationalists have been at the philosophical forefront of examining people’s relationship to territory, and have advanced a number of arguments to justify the nation’s rights over a particular land. A nation, Miller (Reference Miller1995, 27) writes, is 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’. This definition combines objective, cultural features with subjective features, such as shared beliefs and mutual commitments. Miller’s account is based on a complex argument about the inter-relationship between group culture, land, and state territorial rights. Although criticisms of his argument have focused on different aspects of the nexus, much of the focus has been on the way, in which the culture–land argument does not seem to justify clearly territorial right as we know it.

The relationship between group culture and land is established through an argument that emphasizes the symbiotic relationship between them. In Miller’s view, the people who inhabit a certain territory shape the land that they occupy; their culture is mixed with the physical characteristics of the land, and the physical characteristics shape the culture that they develop. Through custom and practice, as well as by explicit political decision, they create laws, establish individual or collective property rights, engage in public works, and shape the physical appearance of the territory. Over time this takes on symbolic significance as they bury their dead in certain places, establish shrines or secular monuments and so forth. There is a sense in which the cultural national community can come to have ‘rights’ to land by mixing its culture with it (Meisels Reference Meisels2005, 95; Miller Reference Miller2007, 217–18). For the group, the land has, in addition to universal value, particular value: it is important to the culture and identity of the group in question.

While this argument suggests that particular land is important for particular groups, this does not yet constitute a justification of territorial right. This brings us to the third element of the culture–land–territorial right nexus. Miller (Reference Miller2012, 263) argues: it is ‘not difficult to justify rights of jurisdiction on the basis of what has been said. Rights of private property alone will not do the job of protecting this added cultural value] because a) such rights are always susceptible to being redrawn by whoever holds rights of jurisdiction and b) much of the embodied value that the group has created is likely to be located in public space – in public architecture, landscapes of historic significance, and so forth. The group needs to maintain overall control over the territory in order to secure that value over time, and for that it needs rights of jurisdiction such as those normally exercised by a state’.

There are three basic criticisms of this cultural nationalist account: the agent-indeterminacy criticism;Footnote ¹⁸ the territorial under-inclusive criticism;Footnote ¹⁹ and the group over-inclusive criticism. All three have been articulated most forcefully by Anna Stilz, who has questioned the extent to which the cultural nationalist account, which she associates primarily with Miller’s work, grounds territorial (jurisdictional) rights, as we know it. I believe – though do not argue the point here – that Miller could address the first criticism, that the second one misses the mark, but that the third one, which focuses on the cultural component of the nationalist definition of the nation, is the most telling. Cultural nationalism, she complains, ‘seems to grant territorial jurisdiction to any group that constructs a culturally marked infrastructure, including immigrant groups’ and cites the case of Little Havana in Miami (Stilz Reference Stilz2011a, 577). At this point, Miller might object that he distinguishes national groups from immigrant groups in his account of the five components that are necessary to qualify as a ‘nation’. However, since his account of the relationship of culture to land (cultural infusion) and the relationship of that to territorial rights (or at least rights of jurisdiction) is doing most of the philosophical work ‘attaching’ people to land, it is not clear why only nations, as distinct from other cultural groups with the right sort of identity and land-transforming capacity, are not also entitled to jurisdictional rights.

Although I think that Miller’s argument could address this criticism, it is only by modifying his account in the direction that I suggest in part three of this article. As it stands, there is a gap between Miller’s argument that ‘attaches’ national groups to particular pieces of territory and the claim that this grounds territorial rights.

Legitimate state theory

If the cultural nationalist account of territory offers a plausible theory of attachment, but a weaker link between territory and jurisdictional authority, legitimate state theory has an opposite, and, I would argue, more serious problem. This account of territorial rights, associated with the work of Kant, and recently defended, in a sophisticated and nuanced form, by both Lea Ypi (Reference Ypi2012) and Anna Stilz (Reference Stilz2011a), has difficulty articulating principles of attachment and has counter-intuitive implications in a number of cases. Indeed, it avoids these counter-intuitive implications only by incorporating non-Kantian elements into the theory, which brings the account much closer to the self-determination account that I defend in part three.

Before elaborating on the precise account offered by Kant and contemporary Kantians, it is worth explaining the central contrast between this account and the cultural nationalist account. In both Miller’s and Meisel’s cultural nationalist theory, there is a pre-political conception of the people (the cultural nation) that has a connection to land, and which is the fundamental holder of territorial right. The state is seen as a necessary instrument of ‘the people’ but not as the fundamental holder of territorial right. In legitimate state theory, put forward by Kant and developed in relation to territory by Ypi and Stilz, the state is the fundamental holder of territorial right. It gains these rights by being legitimate, but it is the state that makes ‘the people’ into one collective body through being subject to state institutions and participating in shaping these institutions (Stilz Reference Stilz2011a, 579).

Kant’s main contribution to our understanding of the relationship between territory and the state is through putting forward an account of the territorial state as required by justice. According to Kant, individuals have a permissive claim to be in occupancy of landFootnote ²⁰ and to obtain objects, and both activities are justified by the exercise of their freedom. However, these acts are permitted only as long as individuals also endeavour to leave a state of nature, that is, a state with no political authority and therefore lacking in justice, and submit to a common jurisdictional authority. This obligation is initiated by the possession of things, which is justified in one sense as an extension of individual freedom but which also contradicts (other people’s) freedom, since removing things from common use prevents other people from enjoying the object in question.Footnote ²¹ Since the exercise of my freedom seems to unavoidably involve restrictions on the exercise of someone else’s freedom, the dilemma can only be resolved by reciprocal recognition of everyone’s obligation to respect the fundamental principles around the acquisition, transfer and use of objects in the external world, which we normally think of as ‘rights of property’. This reciprocal recognition involves creating a civil condition, which ensures that my freedom will be respected and that others’ freedom will be respected, an assurance that cannot be met in the state of nature, and which transforms the use of these external things from the coercive will of one person to the general will of a jurisdictional authority, because it provides a framework of general rules, which is the foundation of civil society.

The main achievement of the Kantian account, therefore, is in explaining why the state is necessary for justice and why individuals are obligated to submit to a state rather than remain in the state of nature. This is a powerful argument for the coercive state, and for territorial rights as necessary for state functioning. However, since Kant’s argument is directed, in the first instance, to demonstrating why in general states are necessary, and so justifying in general territorial rights, it does not show how particular states can have authority over particular territories. Moreover, since the defence of state jurisdiction over territory emphasizes its importance in realizing justice, it has, I will argue, some counter-intuitive implications in cases of annexation, injustice, and where the state is failed. In cases where a state is annexed by another state but the other state is more effective in securing justice, it would seem to license the aggressor in thereby gaining territorial right. In cases where the state is unjust, it would seem that another state can gain control of the territory and implement a more just order, and thereby gain territorial right. Moreover, in cases where a state is failed, it would seem that an effective state, which is able to implement rules of justice, can thereby gain rights over the territory. These results are counter-intuitive, as I will argue below; and they also do not correspond to international law, which is clear that aggression should not be rewarded; that intervention in an emergency, such as when the state is failed, can only be justified for the minimum amount of time necessary to return a functioning order to the people living on the territory; and that, if we viewed justice as the grounding principle of territorial rights, many states would not have the rights over their territory (as argued by Ayelet Banai in this issue).

Stilz’s Kantian-inspired argument aims to address these potential problems with Kantian theory by modifying it. She argues for two further conditions that help to specify why particular states have a right to a particular territory. The conditions are (1) the legitimate occupancy condition and (2) the non-usurpation condition. The legitimate occupancy condition is based on people’s right to live in a place and pursue plans and goals. Stilz argues that states have territorial rights over the geographical space that is occupied by the people they govern. This helps Stilz solve one aspect of the ‘attachment problem’: a state ‘has no claim to territory that its people have no prior right to occupy’ (Stilz Reference Stilz2011a, 579). This right to occupancy is based on a more fundamental interest in autonomy, in securing the background conditions for the exercise of choice. The non-usurpation condition specifies that the state has rights to territory if the state is not a usurper, that is, as long as it did not violate justice by displacing another justice-enforcing state. These conditions are at the heart of Stilz’s attempt to provide a response, broadly in line with Kant’s account of territory, both to the ‘attachment problem’ and to the counter-intuitive implications of pressing a tight link between justice, the state, and territorial rights.

There are – as will become clearer below – two problems with Stilz’s analysis. First, I will argue that these conditions cannot be simply added to a Kantian analysis because they suggest that the appropriate right-holder is not the state but the people. Second, in her account, the value of territorial rights is centred on their role in realizing justice, whereas, I will argue that it is primarily by realizing forms of collective self-determination. Although Stilz occasionally gestures in the direction of recognizing the value of collective self-determination, these gestures are not sufficient: as I will show in the final section of this paper, these do not represent a modification of her basic statist account but suggest a wholesale rejection of it. It suggests that the people are the fundamental territorial right-holder, and that they do so through exercising their moral rights of occupancy and self-determination. In what follows, I will explicate an argument for viewing the people as the appropriate holder of territorial right in a way that coherently explains the relationship between state, territory, and people, and confers fundamental status on the people in a way that cannot be reconciled with statism but does not collapse into cultural nationalism.

A non-statist self-determination account

In view of these challenges, I propose the following principle for demarcating territory. A ‘people’ has jurisdictional rights (and the related liberties, claims, and entitlements that generally flow from this) over land in which members of the group reside, if they legitimately reside on the land.Footnote ²² There are three distinct elements embedded in this principle. The first pillar of this account is a moral right of residence, which attaches to individuals. Since, on this account, the group – a people – must be comprised of individuals who are legitimately resident on the land, we are confronted with the question of how to identify a people and what distinguishes it from other kinds of groups, which are not entitled to territorial rights (even though they might be composed of the very same individuals). The second element in the account is then an account of the group: I argue that a ‘people’ has rights to jurisdictional authority over the territory on which its members are legitimately residing if and only if (a) there is a shared political commitment to establish rules and practices of self-determination on the part of a wide majority of members, (b) they have the political capacity to establish and sustain institutions of political self-determination, and (c) they possess an objective history of political cooperation together, through, for example, state or sub-state institutions, or in a resistance movement. The idea is that the people constitute a collective agent in the sense that the decisions and actions of the group are different from, and not reducible to, the decisions and actions of any individual member of the group: they can be explained through the complex ways in which individuals operate in relationships with others. In this political relationship, they participate in and benefit from these relationship goods; and the main good is that of creating a common life together, producing or maintaining institutions and practices that help to shape the collective circumstances of people’s existence. Third, this argument is grounded in terms of a people’s moral right of self-determination, that is, its right to shape the collective conditions of its existence and make decisions over its collective life.

In this paper I do not discuss at length the moral right of residence, in part because there is convergence by all the competing theories of territory on the idea that rights attach to individuals who live on and are settled on land (Walzer Reference Walzer1983, 43; Moore Reference Moore2001; Stilz Reference Stilz2011a; Reference LefkowitzLefkowitz unpublished). On my account, a moral right of residence attaches to individuals and has three components: a liberty right to settle in an unoccupied area; a right of non-dispossession, by which I mean a right to remain, at liberty, in their homes and communities and not be removed from the place of one’s projects, aims, and relationship; and a right of return, when an individual has been unjustly dispossessed of the land on which s/he has a right to reside. The moral right of residence, or some form of occupancy right, is normally justified either by arguing for the importance of the particular place that we live in for an autonomous life; or by invoking welfarist considerations that people need security of place to live well and function effectively, whether or not people are autonomous in the liberal sense. It is important to the second kind of justification that there is strong empirical evidence that many of the projects and relationships that give meaning to one’s life can only be pursued in a particular location, with a particular institutional structure and geography (Wilson Reference Wilson1982; Brice-Bennet Reference Brice-Bennett1994; Moore Reference Moore2013). While the right is an individual right, in the senses that it attaches to individuals and protects individual interests, individuals are not isolated and atomistic but operate within a structure of relationships that give meaning to their lives. This individual right, then, is connected to a different sort of place-related right that attaches to groups that individuals are members of: a group right of occupancy. Although I do not have space to argue this here (but see Moore Reference Moore2013), the central idea is that individuals have collective identities that are also integral to their sense of who they are, and their collective aspirations as members of these groups are an important part of their lives. Individuals have a right to a place not only because of a connection forged independently by individuals living in a place, but by individuals as members of a collective, whose members share a geographical location with one another. On this conception, the locus of group occupancy rights is defined by the projects and plans central to the relationship; and the central relationship for territorial rights, I will argue below, is that of peoplehood.

I will also not defend specifically my view that territorial rights are primarily rights of jurisdiction, rather than conceived of as something like property. This point is accepted by the two leading types of theories discussed in this paper: nationalist theories, such as Miller’s, and justice theories, such as Stilz’s. Both of these, like this theory, regard territory primarily as the geographical domain of the exercise of jurisdictional authority. Although it would require a complete theory of territory to demonstrate the inter-relationship between jurisdictional authority and other territorial rights – such as the right to control and make rules about natural resources and the right to control borders for the passage of goods and peoples – here I simply assume without a full or complete argument that the exercise of jurisdictional authority is the fundamental element of territorial right. I also assume that there are good reasons, connected to equal treatment of members within the territorial unit, administrative convenience for dispute resolutions, and to solve collective action problems, why jurisdictional authority takes a territorial form. This is not a controversial element in the debate between my theory, legitimate state (justice) theorists, and nationalists.

The ‘people’ as territorial right-holder

In the previous two sections of this paper, I argued that we need to avoid the pitfalls of both the nationalist reliance on culture and statism. In this section, I argue that the territorial right-holder is a collective agent, the people, which should not be defined in either statist or cultural terms. It is necessary to articulate a clear picture of ‘the people’ who hold territorial right beyond the point made in the previous section that it is comprised of individuals who possess moral residency rights. This is a challenging task. It is not helpful to appeal to international law and practice on this topic, for it is incoherent: it promises rights of self-determination to ‘peoples’ but then limits who counts as ‘a people’ in relatively arbitrary ways. Statist views, such as put forward by Stilz, have a plausible social ontology, because the state is a stable, territorially limited set of institutions, which defines who counts as ‘the people’. It is much harder to define the collective, ‘the people’ without reference to a state or set of institutions that can give them corporate identity. Miller’s cultural nationalist account avoids relying on an institutional structure to give the group corporate identity; instead, he appeals to objective cultural markers, which link both people’s identities and attachments, and land, but it is precisely this move which I argued gives rise to the group ‘over-inclusive’ criticism.

In this section I argue in favour of a conception of ‘a people’ that meets the following three conditions, all of which have a political component. First, they must share a conception of themselves as a group – they subjectively identify with co-members, in terms of either being engaged, or desiring to be engaged, in a common political project. Second, they must have the capacity to establish and maintain political institutions, through which they can exercise self-determination. Third, the people have a history of political cooperation together; we can identify objective and historically rooted bonds of solidarity, forged by political practices. These do not necessarily presuppose that the people live within state or state-like institutions; rather, as an extension of my criticism of Stilz, I argue here that many different forms of political cooperation can be important to creating a ‘people’ with potential for self-determination.Footnote ²³ Each of these elements will be discussed in turn.

The first condition for constituting a ‘peoples’ specifies that it must be a collective group, with collective agency, which acts together in organized ways, either through informal methods of coordination or through more formal methods (Copps Reference Copp2006; Pettit Reference Pettit2007; Stilz Reference Stilz2011b). It is not an aggregate group that happens to share a certain feature, such as a group of blue-eyed people. It must be united by what we might call, following Searle, shared ‘we’-intentions. According to Searle (Reference Searle1998, 143–58), we can distinguish social activity from individual activity in virtue of the presence of ‘we’-intentions, which, in Searle’s work, refers to collective, or shared, intentional states. As an adaptation of Searle, and moving away from the language of ‘intentions’, we could say that action is collective when individuals conceive of themselves as participating in a plural subject, which may not have a formal group structure (Gilbert Reference Gilbert1989; May Reference May2005). Similarly, Raimo Tuomela and Kaarlo Miller (Reference Tuomela and Miller1988, 367–89) argue that groups consist of collections of persons with a shared conception of themselves as ‘we’ and that they act by way of a special kind of group-intention. The definition of ‘peoples’ then draws on this account of social group, and it defines ‘peoplehood’, as distinct from other social groups, in terms of a shared ‘we’, and especially a shared aspiration to exercise collective self-government.

One of the difficulties with the generic analysis in terms of shared intentional states or cooperative activity on behalf of a plural subject is that it is difficult, in the absence of an institutional decision-making structure, to be sure when there are these shared aspirations and shared recognition of membership. Elites acting on behalf of a group, or claiming to be representative of the group, will claim that there is this shared sentiment underlying the group, which they represent. This is a serious problem. For this reason, it is important that the first condition be strengthened: this means requiring that there be strong, persistent, empirically verifiable evidence that they either exercise and value collective self-government, or they aspire to collective self-government. The aspiration can be subject to empirical verification, either through clear evidence that the group has mobilized to achieve self-government, or there have been free and fair elections in which parties are elected that have a platform of collective self-determination for the group.Footnote ²⁴ Although the evidence for we-intentions might involve examining the way in which people vote, this account is not equivalent to Wellman’s plebiscitary method, which is an aggregate account in the sense that it aggregates people in accordance with their voting patterns (Wellman Reference Wellman2005). Rather, voting patterns here simply supply evidence to support the claim that there is a social group, which acts in concerted ways to achieve self-government. It is meant to challenge or support elite claims, but only represents one kind of evidence of the existence of a social group.

Second, the groups must have the capacity for political self-determination. The right of self-determination is contingent on the possibility that it can be exercised, and the form that collective self-determination takes has to be consistent not only with the aspirations of the group but the capacities of the group to exercise it. One important condition is that the group has to be territorially concentrated in an area that they legitimately occupy. Although, of course, there is some possibility of non-territorial self-determination for dispersed groups, most significant kinds of self-determination, and especially the kind of self-determination that carries with it territorial right, involves having jurisdictional authority over a geographical area, and the area is coterminous with the area occupied by the group. This means that diaspora people or people that are not territorially concentrated on an area of sufficient size and resources to enable them to fulfil the functions of a state are not entitled to be collectively self-determining in a state, but they can exercise less robust non-statist forms of self-determination, either in sub-state units or through non-territorial mechanisms.

Finally, there must be historical evidence that the people do indeed constitute a single people rather than a collection of individuals, which can be discovered through examination of the existence of political relationships to constitute the group. A people entitled to exercise collective self-determination (over a territory) is not a collection of individuals, but a group whose shared identity has been forged and maintained by actual historical relations – such as a history of sharing a sub-state unit, or a history of political mobilization against an oppressive colonial state. The territorial right-holder should be characterized by group solidarities that spring from a valuable history of shared political association, which will be important in creating the moral bonds that is necessary for the exercise of collective self-determination.

The value of collective self-determination of peoples

In this section I ask and answer the question: what values are served by conferring territorial rights (and specifically the right to jurisdictional authority) on peoples? The question of the values served by a right of collective-self-determination is important, because a theory of attachment has to link the account of territorial right-holder with the justificatory argument that underlies territorial right in the first place. This was accomplished in the cultural nationalist account by linking jurisdictional authority over land with the preservation of value in the land. It was accomplished in the statist account by linking the state, which has authority over territory, with the establishment of justice (combined with some fairly straightforward reasons why the exercise of justice should take a territorial form). I suggested above that while this very tight normative connection gives us a strong reason for conferring rights on states, it is also part of the problem, since many states do not realize justice (nor do they function effectively). This leads proponents of legitimate state theory to offer a more modest account of the requirements of the state – that it be legitimate rather than fully just. The account offered here in reply to this question does not emphasize justice nor the preservation of value in land, but the moral value of self-determination.

The right of self-determination of peoples is enshrined in Article 1, par. 2 and Article 55 of the United Nations Charter. I assume in this paper that this is also a moral (not merely legal) right to collective self-determination. This moral right follows from the idea that political communities are valuable in part because they are spaces in which members co-create their own political project and together implement their own conception of justice. Institutions of political self-determination gives expression to the communities in which people live; they express people’s identities; and they are an important forum in which collective autonomy can be expressed, and people can shape the context in which they live, and realize their political aspirations, free of external domination.

The moral value of self-determination does not inhere in the fact that through it people create objective justice; rather, the process of making the rules that govern a people’s collective existence is itself morally valuable. People who exercise collective self-government have the institutional mechanisms to shape the conditions of their existence, and their future together, and are thereby more autonomous – or experience a different (collective) dimension of autonomy – compared with the strict individual (private sphere) protection of autonomy model.

The right of self-determination, conferred on different similarly situated peoples, also ensures that the international order is structured in such a way that it supports the conditions of (inter-peoples) non-domination. A people can be subject to domination, as is the slave to the master, even if the aggressive power or master does not actually interfere with the actions of the dominated. It inheres in the structure of relations, and especially the all-present threat of interference that affects the choices and actions of dominated agents.Footnote ²⁵ The institutional mechanisms – the legal and moral international rights – by which a people realize collective self-determination gives them protection from the threat of interference, and thereby ensures that inter-people relations are on a footing of relational non-domination, rather than structured by domination and subordination. At this point, however, it is necessary to say something about the conceptual connection between collective self-determination and non-domination. They are kin concepts and express very similar notions, so one might think that one may be superfluous, and that a clearer analysis could be achieved by expanding only one of them (Bachvarova Reference Bachvarova2013). Collective self-determination explains the moral value of territorial rights for each set of peoples, from the perspective of each people, while non-domination is a relational concept, which considers the power relations among different peoples in the international system. There is an interesting dynamic relationship between them. Recognition of a right of self-determination at the international level is supportive of relations of non-domination, because it facilitates people in making decisions free from the threat of interference; and non-domination is an institutional precondition for the realization of collective self-determination. The two concepts are related and mutually reinforcing, but not identical.

Applications

The theory presented so far explains most of our intuitions about the appropriate relations between people and state in the inter-state order. It can account for our intuitions in cases of (a) a failed state; (b) a conquered state; and (c) an unjust state.Footnote ²⁶

Consider first the case of a failed state, by which I mean a state that fails to exercise authority throughout its territory, where large swathes of land are in the control of lawless bands of criminals. One of the classic examples of a failed state is Somalia. On the self-determination account advanced above, there is an empirical question that is important to any assessment of the rights that attach to failed states. One of the conditions for counting as ‘a people’ who are entitled to territorial rights is that they must be capable of exercising collective self-government. In a case where a people (e.g. the Somalis) have conferred on them by the international community jurisdictional authority over their territory, but they are unable to exercise that authority effectively, the important question is: why? If they are incapable of doing so, then the exercise of territorial rights should fall to some other agent. But we should not infer from the empirical fact of a failed state that the Somalians are incapable of creating a viable political order. They have many of the conditions necessary to exercise territorial self-government: they are geographically concentrated on territory that they occupy not unjustly; they do not seek to be part of another state or ruled by an external power, so we can assume that they wish to exercise collective self-government as Somalians. There may be many reasons why Somalia is a failed state – including menacing external neighbours or lack of effective state infrastructure – none of which speak to the innate ability of Somalians to govern themselves as a people; and if this is so, then the right conclusion is to assist the Somalians in their endeavour to create a functioning self-governing community, both through respecting principles of non-intervention, so that they can exercise control over their own affairs, and by facilitating the creation of what Rawls called ‘favourable conditions’ to ensure that they can be collectively self-governing.

This is superior to the statist account of territorial rights, which reasons from the fact of no effective state and therefore no justice to the conclusion that they ought not to have territorial right. On the statist view, if Ethiopia or some other adjacent state, or even some distant power (the US) were able to impose order and protect human rights in Somalia, they would thereby gain territorial rights (rights to jurisdictional authority, to regulate, sell, and extract natural resources, and the right to control entry, and exit). This does not accord with our own sense of justice nor with accepted international norms. It is counter-intuitive because it denies the people on the land the possibility of exercising their own forms of government, and to be collectively self-determining over that territory and it is contrary to international norms, too – for while it is sometimes thought that humanitarian intervention is justified in certain conditions, it is also thought that the intervention should be as minimalist or as temporary as possible, and that the ultimate aim is to restore jurisdictional authority to the people living on the land.

What about the case of a conquered state? Who has territorial rights when a state has been conquered? In international law, the answer is clear. The international order does not permit aggression, and so does not confer rights on a state that takes over the territory of another by force; that would simply create perverse incentives. This consequentialist line of reasoning can be applied to all theories of territorial right-holders and so would seem to suggest that there is agreement that acquiring territorial right as a result of conquest is disallowed. The theory advanced here – emphasizing the moral and legal right of self-determination of peoples – does better than this, however. It gives a non-consequentialist explanation of the wrong of conquest. The problem with conquest is that it violates peoples’ right to be self-determining. It is not about the stability of rules of the international order, nor imposing a non-usurpation condition on the exercise of territorial right to avoid perverse incentives, although all this is true, but that conquest violates self-determination. Moreover, on the theory advanced above, since the territorial right-holder is the people, it is not possible to acquire the moral right to exercise authority over territory from the people as a result of conquest of the state, because the people are the territorial right-holder and as such have the moral rights to reconstitute themselves in a political community.

The self-determination account is also superior to statist theories of territorial rights, which tend to invoke a non-usurpation condition, and emphasize the violation of justice. Legitimate state theories, such as Stilz’s, are not satisfied with consequentialist reasons for denying territorial rights to usurping states. They justify the non-usurpation condition by arguing that states have rights to territory if the state did not violate justice by displacing another justice-enforcing state. This line of argument is in accord with the Kantian story about justice, but raises another kind of problem, namely, that it might be that the conquered state is in fact an unjust state and/or the aggressor state (both of which apply in the case of the Allied defeat of Nazi Germany, and will be discussed in relation to the third problem, of injustice, below).

This brings us to the case of an unjust state. Who has territorial right in cases where the state fails to live up to standards of justice (or, more minimally, legitimacy)? In the case of an illegitimate state that has been conquered, who, or what, would have territorial right? Does it mean that another state that met the minimalist threshold of justice or legitimacy could thereby gain territorial right (on legitimate state theory), simply by acting in the interests of the people who are occupants of the territory? This is a parallel question to the one raised above, where I asked who has territorial right in the context of a failed state? These questions are pressing for statist accounts that link territorial right with justice, because it seems to follow that injustice would nullify the rights over territory.

This is not a problem, however, for self-determination accounts, in which territorial rights are justified by reference to the moral value of ensuring that members of the political community can have control over the collective conditions of its existence.Footnote ²⁷ Injustice is a terrible thing, and it should be met by resistance, rebellion, or revolution, but it would be counter-intuitive to suggest that people who suffered injustice at the hands of their governing regime should thereby also be denied the possibility of exercising collective self-government at all because some other justice-producing entity can thereby be entitled to govern the territory. This result is deeply counter-intuitive and suggests a very passive view of the relationship between the government and its subjects.

Significantly, contemporary attempts to modify Kant’s account have recognized the problematic nature of the justice account in cases such as this. Consider, for example, Stilz’s (Reference Stilz2011a) discussion of Nazi Germany, which is clearly an illegitimate state, and therefore not entitled to territorial right. In 1945, it was defeated and occupied by the Allied powers: it cannot be claimed that the Allied powers violated justice by defeating Germany in war. The legitimate state account of territorial right seems to justify the Allied (occupying) powers in gaining territorial right by imposing a system of just laws and rules over the territory and incorporating the territory into their own. Yet, this seems counter-intuitive for the same reason that the statist treatment of the Somalia example seems counter-intuitive. We might agree that the Nazi state is illegitimate, but we do not thereby think that the German people (or the Somalian people) should lose their previous capacity to make rules to govern their collective life, and be thereby collectively self-determining. (Of course, on legitimate state theory of territorial right, the German people do not lose this right, since it is held by the state, not the people.)

In order to respond to this objection, Stilz modifies the non-usurpation condition so that it does not simply refer to the non-usurpation of a legitimate state. There is, she notes, a ‘widely shared intuition that for such an annexation [the US annexing Germany in 1945], there must be some expression of collective consent on the part of the annexed group’ (Stilz Reference Stilz2011a, 590). She notes that this is a difficulty for her theory since her theory denies that there is a pre-political nation, independent of the state, and it denies that consent is necessary for state legitimacy. Nevertheless, she argues that ‘unilateral annexation [of Nazi Germany by the United States,] would infringe the residual right of the German people in their territory. This gives the German people a claim against annexing states, even when their own states fails or becomes illegitimate’ (Stilz Reference Stilz2011a, 538). How can a statist analysis possibly explain these residual rights to territory? Stilz answers that she does not rely on a pre-political ‘nation’; indeed, if she did, it would mean that her argument has just collapsed into the cultural nationalist one. Rather, she writes: ‘Recall that on the legitimate state theory the people are made into one body by being subject to common institutions and by participating together in shaping these institutions’ (Stilz Reference Stilz2011a, 592). On her view, while a people can only be created by being united into a state, once a people have been so created, through the coordinated social behaviour that marks a state, they develop moral bonds that persist even when the state fails or becomes unjust.

This is the right sort of conclusion, but it is the wrong kind of argument. It appears that Stilz relies on an implausible empirical claim to maintain a purely statist account, where she claims that ‘only a history of sharing a state demonstrates the existence of the moral bonds that support political authority’ (Stilz Reference Stilz2011a, 593). If this is intended to be an empirical claim, about what kinds of ties are necessary to forge a ‘people’, it is implausible, since there is ample evidence of state failure to ‘nation-building’, to build a single people, and some evidence of state success in units that did not have prior institutional recognition. Here, Israel is a paradigmatic example, but many countries were carved out of the defeated (Austro-Hungarian, Ottoman, and German) empires after World War 1 and went on to be successful states, even though the state had no prior history as a state. It is possible – conceptually possible and empirically true too – that a history of cooperation against a state or against a usurper or a settler group, might also forge the moral ties that could sustain institutions of justice. If, on the other hand, Stilz’s argument is intended to be a normative claim, we need to be given reasons to exclude forms of political mobilization that are not institutionalized in a state but which express strong ties and cooperative practices.