Self-Determination and Secession: Why Nations Are Special

Abstract

In this paper, I consider the objection that unilateral secession by a national group (e.g., the Scots) from a legitimate, nonusurping state would wrong minority nationalities within the seceding territory. I show first that most proponents of this objection assume that the ground of the right to national self-determination is the protection of the group’s culture. I show that there are alternative justifications available. I then set out a version of this objection that does not rely on this claim; on this objection, a national minority that seceded and created its own state would commit an expressive wrong against minorities within the territory over which it claimed jurisdiction. I show that this objection is undermotivated: only under a specific set of circumstances would the minorities of the secessionist region be subject to an expressive wrong. Finally, I show that the correct way to think about the claims of minorities in secessionist regions is in terms of a claim to secure access to equal civic status such that they are not at risk of becoming justifiably alienated from their new state. If a secessionist group cannot provide this guarantee to the minority residents of their territory, then their seceding would commit wrongful harm, and the presumption in favour of collective autonomy is defeated. I call this defeater the ‘Alienation Defeater.’

With this in hand, we are now in a position to explain why nations are normatively special. Responding to the objections broached by Allen Buchanan and others, I show that even if other kinds of groups, such as religious groups, have the features in virtue of which nations have a claim to self-determination, this does not entail that those groups also have the right to secede. This is because an account of self-determination needs a list of ‘defeaters’—features in virtue of which a group’s claim to self-determination is defeated. I argue that religious groups are the strongest candidate for having a claim to collective autonomy in virtue of sharing many features with nations. I then argue that religious groups will run afoul of the Alienation Defeater; religious identity is too narrow to be the basis of the dominant collective identity of a state. This does not apply to nationality. This, I explain, is because of qualitative differences between religious groups, qua religious groups, and nations.

Introduction

In this article, I discuss two objections to nationalist accounts of the right to secede. The first objection is that setting up a state that is too closely identified with a particular national group violates the claim to minorities to equal civic status. The second objection is that it is arbitrary to single out nations for rights of self-determination because other types of groups can have the features that ground nations’ claims to self-determination. This objection can be taken as either an attempted reductio of the nationalist theory of self-determination—by showing that the nationalist must be committed to statehood for religious groups, for example—or as trying to show that it is arbitrary for a theory of secession to single out nations.

In section 1, I will set out the first objection, distinguishing between different versions of it. I show first that most proponents of this objection assume that the ground of the right to national self-determination is the protection of the group’s culture. As my account of the right to national self-determination grounds the right to national self-determination in a different way, it is not vulnerable to this objection. I then set out a version of this objection that does not rely on this claim; on this objection, a national minority that seceded and created its own state would commit an expressive wrong against minorities within the territory over which it claimed jurisdiction. I show that this objection is undermotivated: only under a specific set of circumstances would the minorities of the secessionist region be subject to an expressive harm. Finally, I show that the correct way to think about the claims of minorities in secessionist regions is in terms of a claim to secure access to equal civic status such that they are not at risk of becoming justifiably alienated from their new state. If a secessionist group cannot provide this guarantee to the minority residents of their territory, then their claim to secede is defeated.

Following on from this, in section 2, I develop the ‘Arbitrariness Objection.’ I show that even if other kinds of groups, such as religious groups, have the features in virtue of which nations have a claim to self-determination, this does not entail that those groups also have a claim to self-determination. This is because an account of self-determination needs a list of ‘defeaters’—features in virtue of which a group’s claim to self-determination is defeated. I then argue that merely religious groups will run afoul of the Alienation Defeater—religious identity is too narrow to be the basis of the dominant collective identity of a state. This does not apply to nationality. This, I explain, is because of qualitative differences between religious groups, qua nations and religious groups. In summary, setting out the claims that minority groups have on secessionists allows us to see how nations, but not other kinds of ascriptive groups, can satisfy these claims—nations are special. I consider both of these objections to nationalist theories of secession here, therefore, because dealing with the first allows us to deal with the second: by showing how minority groups can be wronged by nationalist secession, we can show that they need not always be. Because of the kind of group that a nation is, national secession need not involve the wronging of minority, whereas secession by, in particular, a nonnational religious group does.

1. Secession, minorities, and equal civic status

1.a Introduction

In this section, I want to consider the possibility that a national minority seceding and setting up a new state would wrongfully harm a specific set of people: the residents of the new state who do not share the nationality of the majority. In this section, I canvass some candidate harms against these people, and set up two objections to what we can call ‘national secession’—that is, unilateral secession of a territory whose population has a preponderant, distinct nationality, and where secession is carried out with the explicit goal of achieving national self-determination.

First, however, in section 1.b, I briefly describe nationalist theories of secession, setting out in broad terms the details of different accounts and contrasting them with other theories of secession.

In section 1.c, I set up the ‘Neutrality Objection’ to national secession. On this objection, using a state as a vehicle for national self-determination is impermissible because it is wrong for the state to promote one cultural identity at the expense of others. The problem with this objection is that national secession need not result in a state that promotes one cultural identity over others; implicit in the objection is the assumption that a claim to national self-determination must be grounded by an interest in the promotion of a national culture. The account of the ground of national self-determination I defend in this thesis, however, makes no reference to this interest.

In section 1.d, I set up a new objection to national self-determination: that it commits an expressive wrong against minorities resident in the seceding territory. In order to defuse this objection, I compare it to a similar argument against symbolic religious establishment (SRE). As in the case of SRE, only in a particular context does national secession commit an expressive wrong against minorities resident in the seceding territory. Once this is cleared up, the focus on expressive wrongs seems misguided; instead, we should look at the background social context of the secessionist society to see whether there is anything objectionable about the position of minority groups.

In section 1.e, I describe the conditions under which national secession wrongs minority residents. I argue that when members of minority groups are put at risk of what I call ‘justified alienation’ from their state, they are wronged by national secession. Justified alienation can result not only from outright persecution or discrimination, but from pervasive inequality of political influence, or an overly narrow or exclusionary public sphere. To conclude this section, I argue that putting minority groups at risk of justified alienation is a defeater for nations’ claim to secede.

1.b Theories of secession

Theories of justified secession are standardly divided into ‘Remedial Right Only’ and ‘Primary Right’ theories. According to Remedial Right Only theories, a group can only have a general right to unilaterally secede if it has been subject to certain injustices (Buchanan 1997a, 34–35). Primary Right theories deny that suffering injustice is a necessary condition for a general right to unilaterally secede. There are two types of Primary Right theory: ‘Nationalist’ theories and ‘Voluntarist’ theories. For Nationalist accounts, certain kinds of groups characterised by ascriptive characteristics—nationality, for example—have the right to secede. For Voluntarist accounts, on the other hand, any group, provided it can govern in a satisfactory way, and provided it meets certain procedural hurdles (such as a majority or super-majority support for independence in a referendum) has the right to secede (Philpott 1995; Wellman 2005).

Nationalist accounts are the focus of this article. Standardly, such accounts identify some interest in virtue of which the groups they pick out have a claim to statehood. Margalit and Raz and Kai Nielsen, for example, argue that people have an interest in having their culture protected from decline and decay (Margalit and Raz 1990, 449; Nielsen 1998, 110). Simon Caney appeals to the symbolic and expressive value of statehood (Caney 1997, 363). Another way of formulating the nationalist view, however, is to argue that nations have certain features in virtue of which they have a ‘presumptive claim’ to collective autonomy—essentially, that a version of the harm principle applies to their collective decisions. This view is essentially a nationalist version of Christopher Wellman’s voluntarist theory of secession (2005). Wellman’s view argues that groups with the ability and willingness to govern in a satisfactory manner have a claim to collective autonomy. The nationalist version of this view must appeal to certain features that are distinctive of ‘ascriptive groups’—for example, it could appeal to the claim that co-nationality (or co-religiosity, for example) is an intrinsically valuable relationship, and then flesh out the ways in which that this claim entails a presumptive claim to collective autonomy (perhaps by appealing to an analogy with other kinds of relationships, such as familial relationships). ¹

The specific way in which one justifies the nationalist claim to self-determination will have some bearing on how one replies to the objections I canvass below.

1.c Civic equality and secession

Critics of liberal nationalism in political theory, such as Anna Stilz and David Copp, have argued that it is wrong for nationalists to treat a state as the vehicle of self-determination for their nation. On this line of thought, when a national group treats a state as the vehicle of national self-determination, it fails to treat minority residents of the territory it wishes to take—people who don’t share the nationality of the majority—as civic equals. We can refine this line of criticism of liberal nationalism into an objection to secession, along the following lines:

The Civic Equality Objection: For a national minority N in territory T to secede from a liberal-democratic state and set up their own state in T, where that national minority does not have a remedial right to secede, is to fail to treat the residents of T who do not share the N’s nationality as civic equals, and therefore to wrong them.

To give an example, if having secured majority support for independence, Scottish nationalists were to unilaterally secede from the United Kingdom and set up an independent Scottish state encompassing the whole of the Scottish territory, this would wrong residents of the Scottish territory who do not have a Scottish national identity, because it would fail to treat them as civic equals. But as it stands, this objection seems to be merely a placeholder; the crucial notion—that of failure to treat a group of people as civic equals—is not fleshed out. How would setting up an independent Scottish state fail to treat people who don’t share the dominant nationality as civic equals? One account goes like this: setting up an independent Scotland would privilege one identity over others, and thus fail to treat those who do not share that identity as civic equals. I will say a bit more about the notion of civic equality in due course. But first, this account needs some explanation of how, exactly, setting up an independent Scotland would privilege Scottish national identity over others. At this point, the proponent of the objection might think I am being obtuse, but there is an implicit assumption behind this objection—namely that the justification for Scottish statehood is, precisely, to privilege and protect Scottish national identity, and to use the state as a vehicle for the promotion of Scottish culture, both symbolically and otherwise. The objection to national secession is set out by Anna Stilz:

I believe it is objectionable for a dominant national group to organize political institutions in ways that preferentially reflect its own history and culture. Such nation building expresses the view that the state (or territorial subunit) specially belongs to that preferred national group, implicitly devaluing minorities.

By symbolically establishing a particular cultural identity as defining the core of the political community and treating major institutions as instruments for the expression and reproduction of that identity, the state communicates to those who do not share this identity that they have a lower rank in the political community. This violates an important obligation of justice: the state should treat all its citizens as civic equals, adapting and formatting its institutions in ways that express their equal status. (2019, 142) ²

The first thing to say is that this is not an objection to secession per se; it is an objection to nationalist accounts of the right to secede. The second thing to say is that the objection is misguided. Even if one grants that, per Margalit and Raz, the ground of the right to national self-determination is a collective interest in protecting a national culture from decline, it need not follow that you are committed to the permissibility of the kinds of (putative) violation of neutrality and civic equality that Stilz describes. You need not, for example, be committed to the permissibility of certain paradigm instances of a violation of neutrality and civic equality, such as discriminatory language laws, national/ethnic preferences in the public sector, or forbidding certain symbols in public. Nor even should we think that nationalists need to engage in lesser violations of civic equality. Secessionists might want to secede in order to protect their national culture from decay just because they think that only national self-government can secure the conditions under which the culture can survive. To give an example, suppose there is a poor region of a multinational state with an agrarian economy and a distinct language. The population of this region is steadily declining because of the emigration to richer regions of the state, and the future of the region’s culture and language is threatened. Secessionists could argue that statehood would allow the region to pursue policies better suited to its economic development, and, through the creation of administrative and public sector jobs in the region, offer an alternative to emigration. Similarly, if we take Caney’s point about the expressive value of statehood, we need not think that a new state would need to enshrine its ethnocentric or exclusive symbols in order for its people to enjoy this symbolic benefit.

What we can see here, then, is that the nationalist goal of protecting a national culture need not be met by privileging a particular national culture; this is what Stilz has in mind when she talks about a violation of civic equality (2019, 144). A national state may better protect its dominant national culture whilst extending equal treatment to other nationalities. And this seems to be what Stilz means by “civic equality.”

The problem for critics of nationalism like Stilz, then, is that they presume that a nationalist account of the right to secede must be committed to the permissibility of certain breaches of equal treatment with respect to culture. This is not the case. At this point, a critic might ask why, if nationalists need not appeal to the defence of culture, self-determination need take the form of secession in the first place. There are two replies to this challenge: the first is that, as I’ve noted, on some nationalist views, something like the harm principle applies to the nation’s collective decisions. As such, the nationalist need not give a sketch of why secession is necessary to protect national interests; rather, they must give some account of why nations are entitled to such wide latitude with respect to their affairs. Secondly, on some views, as I’ve mentioned, statehood is expressively significant; we can simply appeal to this significance in justifying secession versus other self-determination claims.

A further worry for nationalists is that by setting up a state that is too closely associated with a particular nationality, a national group may wrong minorities within its territory. But many (if not most) states are closely identified with a particular national group; at least to the extent that they contain a dominant nationality, are named after that group (e.g., Denmark), and so on. There doesn’t seem to be anything objectionable about this. If we think that national minorities are under a duty not to create states that are associated with one nationality, then why not think that the peoples of existing states are under a pro tanto duty to their minority members to merge with other states in order to divest their states of their national attachments? It seems implausible that there is such a duty.

1.d The expressive wrong of national secession

There might still be something to this general line of criticism regarding civic exclusion. Regardless of how the Scots justify their claim to secede or what they intend to do with independence, doesn’t the bare act of creating a Scottish state—that is, one where boundaries of the state and citizenship are defined by Scottish nationality—express to people in Scotland who don’t share that nationality that they are not regarded as equal citizens? If the state is the vehicle of Scottish self-determination, then can non-Scots regard it as equally their own? To sharpen this point, let’s put the Civic Equality Objection in revised form:

The Expressive Citizenship Objection: For a national minority N in territory T to secede from a liberal-democratic state and set up their own state in T, where that national minority does not have a remedial right to secede, is to express to (inter-alia) the residents of T who are not members of N that (i) the new T-state belongs to N and (ii) nonmembers of N are not civic equals in T-state. For the N’s to express this message to non-N members of T would be to wrong them.

Now this claim is similar to claims made by opponents of symbolic religious establishment (SRE). Martha Nussbaum, for example, has objected to symbolic religious establishment on the basis that it sends a message that nonadherents of the established religion are second-class citizens:

By throwing its support behind an orthodoxy, government makes a statement: this is the official doctrine of our nation. Such a statement, as Madison saw, suggests that non-adherents are not fully equal members of the political community, and they don’t enter the public square “on equal conditions.” Even if they are not coerced, the implication is that they exist at the sufferance of the dominant group, not as citizens of equal worth in their own right (2008, 247). ³

The objection to symbolic religious establishment, at least in this form, does not straightforwardly apply to secession by a national group, however. Nussbaum’s argument here relies on two claims: (i) by establishing a religion, government expresses the view that that religion is the official doctrine of the nation, and (ii) such a statement undermines the civic equality of nonadherents of that religion. The problem for transposing this objection to the secession case is that there is no clear analogue of (i) in the case of national secession. How would Scotland or Quebec declaring independence express the attitude that the Scots or the Quebecois are its Staatsvolk? One explanation goes like this: “the Scots or the Quebecois expressly set up a state on the basis that they are a separate nationality—that clearly sends a message that the new state belongs to them.” The problem with this explanation, however, is that movements for national independence almost never appeal to the bare fact of nationality itself to explain why they should exercise their right to secede. Although on my account, and in the view of nationalists in general, the fact of their being a separate nationality explains why a group has the (presumptive) right to secede, it does not explain why they should or should not exercise it. Typically, national separatists appeal to, for example, economic and political subordination, differing political values, or a desire to escape a malfunctioning state to justify their claims that they should exercise their right to secede. There is a good example of this from Quebec: Manon Masse, the leader of the left-wing Quebec nationalist party Quebec Solidaire, often says that the province should break free of the “petro-state” Canada—an appeal to environmentalist values in the service of a nationalist cause. ⁴ This sort of justification for seeking independence is not inherently exclusionary towards people who don’t share the dominant nationality. Their interests and values are, or can be, factored into the economic and political arguments for seeking independence. ⁵ The fact that separate nationality grounds the presumptive claim to secede says nothing about the political dynamics of secessionist movements, or the justifications they offer for exercising the right to secede. There might be a few Scottish nationalists who appeal to the Declaration of Arbroath to explain why Scotland should go it alone, but most appeal to Scotland’s purported divergence in political values from England, or to the supposed economic benefits of independence. ⁶ The setting up of an independent national state might be an expression of ethnic chauvinism, but it also might be the expression of a collective desire to have a more robust (or a leaner) welfare state, a greener energy policy (or more direct control over oil and gas royalties), or simply to be rid of a self-serving political class. That is not to say, however, that when a national group does justify secession by its desire to become a Staatsvolk that it does not wrong minorities. The precise form that this wrong takes is perhaps best left to the following subsection, where I explain what it is that national secessionists owe to minorities.

One worry here is that in my response to this objection, I have relied on a particular (false) view of what it is to treat someone as a civic equal—namely, one treats someone as a civic equal when one justifies policies or constitutional changes on the basis of reasons that are not exclusionary, or which do not presuppose the truth of a particular conception of the good, or the value of some particular identity. This view has a structural similarity with the neutrality of justification or the neutrality of aim interpretation of the putative normative ideal of liberal neutrality. And these interpretations have come in for strong criticism by proponents of liberal neutrality. First, however, we should try to get a sharper grip on the idea of neutrality of justification and neutrality of aim. Allen Patten characterizes them as follows:

Neutrality of Aim: The state violates this requirement when it adopts any policy with the aim of making some particular conception of the good more or less successful.

Neutrality of Justification: The state violates this requirement when its fundamental reason for adopting some policy involves a judgement about the value of a particular conception of the good. (2014)

As Patten and others (such as Matthew Kramer), have pointed out, however, both of these interpretations are vulnerable to the objection that they “count as neutral policies that seem, intuitively, to be nonneutral” (Kramer 2017, chap. 1). Patten gives the example of a state that establishes a particular religion, not on the basis that the religion is the true religion, but on the basis of the desirable social consequences of doing so (2014, 113). What Patten and like-minded critics suggest instead, therefore, is something like the following principle:

Neutrality of Treatment: The state violates this requirement when, relative to an appropriate baseline, its policies are more accommodating to some conceptions of the good than they are of others (115).

What matters for whether a secessionist policy violates this principle, however, is how it conducts itself after it secedes. It’s not obvious that the mere fact of setting up a state that is, let’s say, closely associated with a particular national group is to be more accommodating to some conceptions of the good than others, at least “relative to an appropriate baseline.” If having a state that is closely associated with a particular national group—say, by having a dominant nationality, or being named after that group—is a violation of neutrality of treatment, then existing states that are associated with particular national groups have pro tanto reasons, grounded in neutrality of treatment, to merge with other states in order to eliminate this association. This seems, as we’ve seen, implausible. So the appropriate baseline cannot be defined by a lack of association between a particular national group and a state.

What really matters for whether the national minority expresses the attitude that the newly independent state “belongs to them,” in a suitably objectionable way, is their behaviour towards minorities, both before and after independence. I will return to this near the end of the section.

There is another way to put the Expressive Citizenship Objection. On this version, the analogue of Nussbaum’s (i) is something like: making salient ethnic divisions, where this is at the expense of minorities with socially vulnerable identities. For an explication of this line of criticism, we can look again at critics of symbolic religious establishment, specifically the case of what Cecile Laborde and Sune Laegaard call “neo-establishment”—that is, when a political community decides to symbolically reaffirm the position of the majority or “historically dominant” religion (Laborde and Laegaard 2019, 184). The example they give of neo-establishment is the decision by the government of Bavaria to have the crucifix displayed in all classrooms. Neo-establishment is contrasted with “vestigial establishment,” where the prominence of the historically dominant religion in various areas of public life is a historical holdover. They give the examples of the practice of putting crosses beside dead authors’ names, or the ubiquity of church spires and Christian religious symbols in medieval university–towns such as Oxford and Cambridge. Neo-establishment, unlike (at least typically) the vestigial kind, aims at the affirmation of a particular religious identity. This unavoidably sends a message of exclusion to minorities (184–85). The point of similarity, on this objection, between national secession and neo-establishment is that both make ethnic or religious more salient and exacerbate the social vulnerability of minorities.

There are two problems for this analogy between national secession and neo-establishment, however. The first is that unlike neo-establishment, national secession may not be a move from relative neutrality in terms of state support for certain identities towards nonneutrality; it may involve merely a change in which identities are promoted. We cannot assume a background situation where the state does not promote a particular national identity at the expense of the secessionist group. A critic of national secession might have the following background picture in mind: most liberal democratic states, even if they at one time actively promoted the symbols and culture on one dominant ethnic group, are gradually having their symbols and the collective identity of their citizenries drained of any ethnic content. We might illustrate this process with the example of Canada. At its foundation, Canada was dominated by two ethnic groups: British-Canadians and what were then referred to as French-Canadians. Canada’s symbols and institutions were mostly reflective of its British heritage: its original flag bore the Union Jack, its head of state was the British monarch, and God Save the Queen was sung at official functions. Gradually, however, the British element in Canadian identity receded in importance. A new national flag, the Maple Leaf, was adopted in 1965; the title of “Dominion” fell out of official usage; and, starting in the 1960s, large-scale immigration meant that Anglophone Canada no longer had a dominant ethnic group. But most states are not like this. Even multinational states like Britain typically have a preponderant national group, and the state’s symbols and civic identity are infused with the heritage of this group. In other words, there is usually no background of neutrality from which national secession departs. The second problem is that the examples of neo-establishment described by Laborde and Laegaard—the Bavarian crucifixes and the Swiss minaret ban—were both motivated by anti-Muslim sentiment. In their telling, neo-establishment “[…] is not a benign cultivation of vestigial traditions and symbols but, rather, a distinctively reactive movement motivated by fear of migration, globalization, and Muslims” (2019, 184). But national secession might not take place in a context like this. The profiles of separatist movements are quite different in this respect. We can’t assume that every movement for national independence has ethnic chauvinism festering under its skin, and so the suggestion that national secession will always have an expressive meaning of exclusion to minorities is unfounded.

Let’s now step back and see if we can get to the root of the Expressive Citizenship Objection. Continuing with the analogy between SRE and national secession, let’s try to see if we can find a more general statement of the problem with SRE, and see if it carries over to the case of national secession. In Liberalism’s Religion, Cecile Laborde argues that what is wrong with SRE is not the failure of neutrality towards competing conceptions of the good, but the constitution and perpetuation of “social relations of hierarchy, subordination and domination” (Laborde 2017, 126). On her view, “[…] symbolic establishment is wrong if religious identity independently functions as a marker of social vulnerability and domination in the society in question, or if such establishment can reliably be predicted to increase the social salience of religious identity” (126). We might be tempted to reason straightforwardly from this argument against SRE to an argument against national secession. But again, there are two crucial disanalogies. The first is that in the SRE case, the state actively promotes one religious identity over others. As I’ve argued, this needn’t be true of national secession. Secondly, the social vulnerability of members of minority ethnic groups might have nothing to do with movements for national independence, and the increased salience of say, Scottish versus British identity mightn’t have consequences for ethnic minorities on either side of the divide. There are no good grounds for thinking that the position of Scots of Asian ethnic background, for example, would be put in worse position by Scottish independence. If certain national-cum-ethnic distinctions are made more salient by secession or by secessionist movements, it does not follow that the social vulnerability of a given ethnic or national minorities will be exacerbated where the divisions are between people whose primary political identity is that of the larger state and those whose primary political identity is that of the secessionist region. As David Miller points out (in his criticism of Laborde’s objection to SRE), “it is not enough to point to the fact that having a religious identity may be a source of vulnerability: it must also be shown that the presence of an established church magnifies that effect” (2019, 88).

Collecting the various strands of the analogy between objections to SRE and the Equal Citizenship Objection, it seems clear that the objection to national secession is dependent, first, on the salience of certain ethnic or national distinctions, and secondly, on some causal connection between national independence, or moves towards it, and the exacerbation of the vulnerable position of minorities. Let’s say then that national secession might be taken to express a wrongful attitude towards internal minorities when the following condition is in place:

The Alienation Condition: There is a high degree of mistrust and friction between nonmembers of the majority group and members of the majority or there is a recent history of persecution, discrimination, or ethnic conflict or negative attitudes towards internal nonmembers of the majority national group play a major part in mobilising support for independence or are publicly and frequently expressed by leading supporters of independence.

As with SRE, then, national secession need not involve the expression the objectionable attitudes towards nonmembers of the majority national group. If we want to find out whether an instance of national secession does involve the expression of this attitude towards minorities, we will have to look at the social context in which that secession takes place. ⁷

1.e The Alienation Defeater

Having specified the two conditions above, notice that the Expressive Citizenship Objection, even as directed against cases where both conditions are satisfied, seems, if not misdirected, then misplaced. The primary problem from the point of view of ethnic minorities with the new secessionist state doesn’t seem to be with what it expresses symbolically. Rather, it seems to be that a large proportion of people in their state are actively hostile to them or disregard their interests and preferences. Even if they are not likely to be subject to active persecution or discrimination, it is unlikely that they will be treated as full members of the political community: their concerns and interests will not be taken seriously in public debate, and the state will not make much effort to ensure that they are represented or included in public institutions. Subjecting one’s fellow citizens to this kind of insecurity seems to be the most important objection in this case. There will still be a further expressive wrong attaching to national secession in cases where the two conditions are satisfied, just as there is an expressive wrong when Fred assaults Johns and steals his wallet. But John will be more concerned with his broken bones and his missing debit card than the expressive meaning of Fred’s actions.

What I suggest, then, is that we change the focus of the worry about the equal citizenship from symbolic equality to something like (justified) alienation. The thought is that minority members of a secessionist region are wronged when it is very likely they could reasonably become justifiably alienated from their new state. What do I mean by justified alienation? As I mean it here, it has both a cognitive and an affective component. First, minority members of the secessionist group come to believe (reasonably) that their state does not represent them. Second, they come to feel attitudes of resentment and distrust for their state and their compatriots, and these attitudes are fitting. But what are the circumstances under which they could reasonably form this belief, and which attitudes would be fitting? The most straightforward examples are cases of outright persecution, where either the state is the persecutor or deliberately or negligently fails to protect the group from persecution. Serious discrimination—for example, formal or informal dispreferment in jobs, housing, education, and various public services—is another example. There is a more difficult kind of case, however: cases characterised not by persecution or discrimination, but by a justified sense by members of minorities that they are not considered full members of the public sphere. Although there may be no pattern of discrimination in essential services or employment, the members of the group are not well-represented in politics or the media, do not have their concerns or interests taken seriously, or are treated with suspicion or resentment by compatriots. Let me flesh this out with the following example:

Devoutia: The overwhelming majority of Devoutia’s citizens are observant members of theologically and socially conservative denominations of Christianity. There are small minority groups of liberal Christians, atheists, and non-Christians. Devoutia’s politics are dominated by theologically conservative Christians; it has conservative laws on bioethics and matters of public morality, Christian religious symbols are prominent in public places and at state occasions, and non-Christian religious or secularist views are not given much public discussion. Although there is no overt persecution or discrimination against them, liberal Christians, the nonreligious, and members of other religions encounter mild hostility when they express views that depart from the societal consensus, and few public representatives are willing to align themselves with them or to attempt to defend their policy preferences. They participate in politics at much lower rates and generally keep to themselves as much as possible. They have a sense that they should “keep their heads down.” ⁸

In this case, the elements of persecution and discrimination are missing. Furthermore, there is no complaint about the legitimacy of Devoutia’s conservative policies on bioethics and other matters of social morality. ⁹ These are publicly justified, even if religiously motivated. The problem in this case is not that any particular measures or elements of public life in Devoutia are objectionable at the bar of liberal political morality, but that, in the round, the members of minority religious groups have inferior civic status. Now it would be useful here to describe their condition in a more general way. One way to characterise an aspect of their condition is to say that public life in Devoutia is carried out on terms that are, to use Patti Lenard’s term “inaccessible” to religious minorities (2019, 160). The values that are appealed to in public debates and the modes and norms of public engagement cannot be adopted by these people without serious cost to their own group identities. A Devoutian Muslim, for example, could not expect a sympathetic or receptive hearing in a public debate over abortion, for example, unless they were to express their arguments in terms acceptable to theologically conservative Christians. Part of the cost of participating in public life is demonstrating one’s fidelity to important group attachments, and the religious minorities of Devoutia will not be able to do so without serious cost. Another aspect of the condition of Devoutia’s minorities is that they are a “sticky” minority; there is, as Philip Pettit puts it, “[…] ex ante reason, associated with their independently fixed identity, to think that certain individuals will be in the minority on given issues: their identity pre-commits them, as we might say, on those issues” (2012, 213). They are in a position of unequal influence, as Pettit puts it, with respect to democratic decisions.

Can we put a name on their general condition? Should we think of them as being dominated, for example? I will remain neutral about this. I think it’s enough for my purposes to just say that religious minorities are, with good reason, alienated from Devoutia, without specifying what it is in the most general terms that makes their alienation justified. I want to leave my account such that others can fill in the details with respect to what exactly is objectionable about the religious minorities of Devoutia’s situation. It’s for this reason that I’ve chosen to focus on the justified alienation—which some might say is a by-product of a wrong-making feature such as domination.

I will now try to sharpen this account of justified alienation. Let’s say that alienation from a state S is justified for a member, m, of a minority group N if: ¹⁰

1. (i) S persecutes, or culpably fails to prevent, serious systematic rights violations against members of N; or

2. (ii) S formally discriminates against members of N in matters of education, housing, employment, or access to public services; or S culpably fails to prevent informal systematic discrimination against members of N; or

3. (iii) m has the correct belief that his membership of N makes him a “second-class citizen” of S, as evidenced by the fact that:

   1. (a) there is an ex ante reason to think that because m is a member of N, he will be in the minority of some important political questions (Inequality of Influence); ¹¹ and

   2. (b) public life in S is dominated by norms, values, and political practices that are not, by virtue of n’s membership of N, accessible to n;

      or

   3. (c) there is widespread hostility towards members of N, and this hostility is either encouraged or tolerated by S. ¹²

      or

4. (iv) The content of the N’s collective identity is not itself such as to make identification with N impermissible (e.g, the N’s are neo-Nazis).

The religious minorities in Devoutia have a claim against their compatriots to be put on more equal terms of citizenship. But now suppose that Devoutia were not an independent state, but on the verge of independence from a more pluralist liberal democracy. What can we say about the religious minorities of Devoutia in that case? They might have no complaint against their current state, but they know that their position will change once Devoutia achieves independence. Their condition on the verge of Devoutian independence is this: if Devoutia becomes an independent state, they will be at high risk of becoming justifiably alienated. Plausibly, people have a claim against being exposed to this risk. They will have formed their life plans and attachment to the places in which they live in reasonable expectation of continuing to enjoy equal civic status. Once their status is threatened, they may wish to substantially revise their life plans, or simply to pack up and leave.

Another important point: a group may be at risk of becoming alienated without there being an expressive wrong at all. The religious minorities in Devoutia might never be the target of hostility, and their social position may be completely irrelevant to the move for independence in Devoutia. Yet they are still at risk of becoming justifiably alienated. If the content of the dominant collective identity in Devoutia is such that members of religious minorities cannot adopt elements of it without serious cost, then they are unlikely to be able to participate in the public sphere on equal terms. They will be second class citizens, even without being the target of active persecution.

One final point: if this discussion has been focused on what secessionists owe to minorities within their claimed territory, it also has implications for existing states and their legitimacy. It might be thought that this discussion could be incorporated into a ‘remedial-right’ account of secession—an account that sets out the kinds of injustices that can give a group a claim to secede. That is to say, one might think that being alienated in this way could give a group a claim to secede from a state. Of course, this would be a remedial claim to secede, and thus would not be explained by a primary right theory of secession (although it would be compatible with such a theory).

Conclusion

My argument in this section has been that (i) the neutrality objection to national secession is misguided, and that (ii) in order for the Expressive Citizenship Objection to go through, it must be shown that the group claiming to be disrespected by national secession has, or is at risk of having, inferior civic status. Expressive wrongs of this sort that feature in the Expressive Citizenship Objection are typically dependent on some objectionable features of the broader social context. From this, I argue that the focus on expressive wrongs is misguided: if there is an objection to national secession that is connected to the equal civic status of minorities, it is that they are at risk of what I have described as justified alienation from the state in which they live.

Importantly, this justified alienation objection won’t be an objection to national secession in all circumstances. All the objection says that is that members of minority groups have a strong claim against being put at risk of being justifiably alienated from the state in which they live. For their compatriots to put them in this position is pro tanto to wrong them. We can now give a clearer statement of a candidate defeater for the right to secede:

The Alienation Defeater: If a national minority N in territory T secedes from a liberal-democratic state to set up their own state in T, and if non-N residents of T have a reasonable prospect of being justifiably alienated from the new state in T, then the people of N pro tanto wrong the non-N members of T.

This formulation best expresses the thought behind the civic equality objection to national secession. We can see that it isn’t a blanket objection; the failure to secure the equal civic status of minority groups is a possible defeater of the right to secede for nations. Now I have chosen to offer a relatively minimal account of justified alienation; as I use the term here it is effectively a placeholder. But I think it’s clear that on my view, the bare fact of national secession doesn’t undermine the civic status of minorities; the worry about symbolic equality can only get a grip against a background of more tangible inequalities of status. In connection with this, I have argued that we need to look at the broader social context, rather than whether individual policies run afoul of specific principles of liberal political morality, to see whether national secession will undermine the equal civic status of minorities.

2. The Arbitrariness Objection

2.a Introduction

In section 1, I hope to have defused the objection that national secession always infringes on the equal civic status of minorities. I argued that national secession only infringes on the equal civic status of minorities in certain social and political contexts. This response, however, opens me up to two related objections. The first objection is to my general approach to the right to secede—that is, that nations are normatively special in relation to secession. This is the objection that it is arbitrary to single out nations for the right to secede, when another kind of group—religious groups—has many of the features in virtue of which nations have, according to nationalists, the right to secede. I call this the ‘Arbitrariness Objection.’ The second objection is that my account of the ground of national self-determination, in conjunction with my reply to the Civic Equality Objection, entails that religious groups have a presumptive claim to unilaterally secede. This is intended to be a reductio of my account of the right to secede.

In this section, I show that it is not arbitrary to single out nations for the right to self-determination; there are important differences between nationality and religious identity. Following on from my discussion of the Alienation Defeater in the last section, I argue that the multistranded nature of nationality means that national secession is less likely to run afoul of the Alienation Defeater than secession by a religious group. Religion is too narrow in content to be acceptable as a basis for collective identity, I argue.

2.b The Arbitrariness Objection

One common criticism of the “first-wave” of nationalist accounts of the right to secede is that these accounts arbitrarily single out nations for the right to self-determination. Other kinds of groups, these critics point out, can have the features on the basis of which it is claimed that nations have the right to self-determination (Lee 2017, 78; Buchanan 1997b, 297). Allen Buchanan puts the objection this way:

Suppose we grant Margalit and Raz’s premise that the best protection for those interests of individuals that are served by belonging to a nation is for the nation to have its own state. That might be true as well for a number of other types of groups—religious, political-ideological, ethnic, and so on. (1997b, 302)

There are two ways to take this objection. One way is to take it as a reductio of nationalist theories of self-determination in general: “nationalist theories entail that not only nations, but also religious groups and political-ideological groups have the right to set up their own state. ¹³ But then nationalist theories can’t be right, because those groups certainly don’t have the right to set up their own state.” The other way to read it is as an objection to the claim that nations are special: “Fair enough, nations have the right to set up their own state, but then so do religious and political-ideological groups.” We can take both versions together.

To start, it’s not clear at all that political-ideological groups have the interest that, on the nationalist theory, grounds the right to secede. Members of political-ideological groups typically want (or at least claim to want) to gain political power and steer the state in their direction. If they have a distinct culture, it is usually neither as all-encompassing nor as important as national culture, and it is not typically valued except insofar as it contributes to the political objectives of the group. I will say more about this in the next subsection.

It is true, however, that nationalist theories entail that some religious groups have many of the features in virtue of nationalist theorists argue that nations have the right to self-determination: members of a territorially concentrated religious minority, we can imagine, might have a strong interest in ensuring the continued viability of their collective way of life, or in enjoying the symbolic benefits of statehood. So by their own lights, then, shouldn’t nationalists regard religious groups as having a claim to secede? Again, this objection can be taken either as a reductio of the nationalist account of self-determination, or as a criticism of its specifically nationalist bent. If the account entails that some religious groups also have the right to secede, then how is the theory a nationalist theory of secession?

But it doesn’t follow that someone who is committed to the account of the value of collective self-determination nationalists defend has to say that religious or political-ideological groups have a right to set up their own state. A complete theory of self-determination will also have a list of defeaters—that is, conditions such that, when a group satisfies them, their claim to self-determination is defeated. We can bring the Alienation Defeater to bear here again.

2.c Religious groups and the Alienation Defeater

In which cases will the Alienation Defeater apply? Compare the following cases:

Divinitia with Symbolic Establishment: Divinitia secedes from Secularia and re-constitutes itself as the “Apostolic State of Divinitia.” The new state gives expression to its Catholic identity through its symbols, holidays, etc. However, the state does not enforce Catholic social teaching—divorce, artificial contraception and blasphemy are all permitted, but the state recognises in law “the special position of the Catholic Church.”

Divinitia without Symbolic Establishment: Divinitia secedes from Secularia and constitutes itself as the “Republic of Divinitia.” The new state does not symbolically establish its majority religion and protects the religious and political liberty of religious minorities. ¹⁴

Now even granting that Divinitia with Symbolic Establishment runs afoul of the Alienation Defeater, what grounds do we have for thinking that it applies to Divinitia without Symbolic Establishment? Let’s distinguish between two further cases:

Pious Divinitia: The Catholics of Secularia are intensely religious. Their collective identity is entirely defined by their religious identity. They have no attachment to any particular region of Secularia; they are scattered throughout the country. At some point, there is a movement among them to move to sparsely populated region of Secularia called Divinitia, where they will form a majority. They do this, and then vote to secede from Secularia. They will settle for some form of autonomy if Secularia refuses to allow them to secede. They do not intend to symbolically establish their religion, but they are seceding with the explicit aim of creating a majority Catholic state.

Lax Divinitia: The Catholics of Secularia are concentrated in and make up a majority of the population of one region, Divinitia. They are from the same ethnic group as the rest of the Secularia and share a language and many of the same traditions. The majority of them practice their faith, but their collective identity also has a cultural dimension, and some non-Catholics also identify with the broader culture of Divinitia. Divinitia votes to secede. They do not intend to symbolically establish their religion.

Here I think that the Pious Divinitians run afoul of the Alienation Defeater for the following reason. Even though the Pious Divinitans might have no intention of even symbolically establishing their religion, it is clear that (a) a separate religious identity is the whole rationale for independent statehood, and that (b) the political significance of religion and religious identity is likely to be high in this new state. Whatever collective political identity emerges in Pious Divinitia will draw heavily from the religious identity of the majority. And this collective identity will be too narrow in content to sustain a genuinely pluralistic society. ¹⁵

But it’s not clear that the Lax Divinitians run afoul of the Alienation Defeater. They haven’t gone out of their way to exclude non-Catholics, and their collective identity has enough cultural content to be accessible to nonbelievers. Does this mean that we have to abandon the nationalist theory of self-determination? My response is to note that Lax Divinitia looks suspiciously like a nation. It is a territorially concentrated group whose members share an ascriptive characteristic—religion. Remember that one of the defining characteristics of nations is that they have a public culture. Religion can be one of the elements of a public culture. Indeed, in some cases it may be the most salient difference between two nationalities. Think, for example, of the Bosniaks. Until quite recently they were not regarded as a separate nationality at all, but merely as Muslim Serbs or Muslim Croats. In Yugoslav-era censuses, there was no category of “Bosniak,” but only of “Muslim.” But we can perfectly well think of the Bosniaks as a nation; certainly, they now think of themselves this way. Having a different religious identity to your neighbours makes you different in lots of ways. For one, it colours a group’s sense of its history—who it regards as hero or a villain, which battles it counts as victories or defeats, and so on. A shared sense of history is another defining characteristic of nationhood. The Bosniaks, to continue the example, have a strong sense of continuity with the people of Bosnia before the conversion to Islam; there is a strongly held myth that the Bosniaks converted to Islam partly in rejection of the Catholic and Orthodox Churches, who had persecuted the Bosnians for their adherence to the sect of Bogomilism. ¹⁶ We can make a distinction between a merely territorially concentrated religious groups, (e.g., Nonconformist Protestants in the North of England) and groups where religion is perhaps the dominant feature of religious identity, but there is enough distinctive nonreligious content to the collective identity that we can regard them as a nation as well. And it is precisely because we can regard groups like this—like the Lax Divinitians—as a nation, that they can avoid running afoul of the Alienation Defeater.

2.d National identity and the Alienation Defeater

What exactly does the fact of nationality have to do with whether or not a group can avoid the Alienation Defeater? Nations have, I will argue, certain qualitative features such that they are less likely than religious groups to run afoul of the Alienation Defeater. ¹⁷

The first of those features is territoriality. Nations are typically geographically bounded communities. There are two germane implications of this. The first is that a national identity will typically accumulate over time features of the identities of those who live in the national territory, even if they are not initially members of the national group itself. Many European national identities, for example, have elements drawn from the cultures of their immigrant-descended populations. In addition, even though there might be a formerly ethnic core to a national identity, that identity will also include elements drawn from the national minorities living in the territory. The second implication is that connection to a particular region is usually one of the core components of national identity. Although political theorists who write about nationalism have noticed this connection between nationality and territory, they have failed to distinguish between two modes of territorial attachment. ¹⁸ The first mode is where the territory is seen as either the ethnic homeland or the ethnic birth-right of a nation—that is, where the national territory is seen as belonging to a particular ethnic group, either because they are believed to originate there or because they have some special entitlement to it by virtue of conquest or divine favour. ¹⁹ We might call this mode ‘ethnic attachment.’ The second mode, however, we might call ‘affective attachment.’ On this mode, there will be either no sense of ethnic attachment to a particular territory, or it will not be prominent. Instead, nationals will have certain affective ties to their national territory and will regard it as the object of pride and will value things because they come from that territory. This mode of attachment is similar to the attitudes that people sometimes feel towards their towns or cities. For example, a Londoner or a New Yorker might feel strongly attached to their city and wish to learn about its history and distinctive cultural traits despite having no ethnic attachment to their city—indeed, they might not even have been born there. Many nations—particularly multi-ethnic nations or nations formed out of former settler colonies—have affective, but not ethnic attachment to their territory. ²⁰ The point of distinguishing between these two modes of territorial attachment is to illustrate that the bedrock of a national identity may simply be a sense of attachment to a particular place, its character, and history. Just as people may freely adopt and discard the local “patriotisms” of the cities or the towns in which they live, so, in many cases, do people with national identity, without serious cost to the other aspects of their identity.

The second feature is that national identities in liberal-democratic societies are subject to deliberation and contestation. Features of national identity that are not accessible to certain groups, or which are outright insulting to them, may be discarded after public debate (Miller 2012, 176; 1995, 96). One common example of this sort of thing is debates over which historical figures the nation ought to celebrate. For example, consider the public debate in Canada regarding John A. MacDonald, the country’s first prime minister. MacDonald was once celebrated as one of the country’s founders. In recent years, however, there has been greater public attention to his governments’ treatment of Canada’s indigenous First Nations peoples and his introduction of racist legislation against Chinese migrants. MacDonald has become a figure of increasing controversy and is less often celebrated as a positive figure in Canadian history. This feature of national identity—its openness to contestation with respect to content—can be contrasted with religious identity, where the terms of contestation are narrowed by the sharper boundaries between members and nonmembers, and the frequently hierarchical character of deliberation within religious groups.

The third feature is that national identities are often nested—e.g., Scottish and British, or Canadian and Quebecois. Having one national identity does not preclude having another. The national identity of a seceding group will often contain much content that is shared with the people of the remainder state. To make a slightly different point, national identities can share the same content with other forms of identity, such as regional identity. For some, their Quebec or Scottish identity is a regional identity. For others, it is a national identity. It is often the case that minorities of secessionist states who identify with their former state regard the national identity of the secessionists as their own regional identity. Often the national identity has content that the regional identity doesn’t have. But there is usually a good deal of shared ground. In the case of religion, however, there is typically a clearer boundary between the identity of a religious group and other forms of identity, such as regional attachment. The sharper boundary—indeed incompatibility—between membership of different religious groups is paradigmatically explained by the fact that religious belonging requires affirming certain propositions, and that some of these propositions are contradictory to the propositions affirmed by other religions. Think for example, of the incompatibility of Catholic and Lutheran Eucharistic doctrine. Of course, however, not all religious practices or religious identities can be characterised in this way (Laborde 2017, 21–24). ²¹

The point of elaborating these three features of national identity is to argue that nationality need not be too narrow and exclusionary a basis of majority collective identity to sustain a pluralistic society—one in which members of minority groups have equal standing and have access to the public sphere. My argument is, in summary, that to secure substantive civic equality, minorities will have to adopt aspects of the dominant collective identity; this will be too costly for them if the dominant collective identity is primarily religious in character, but this need not be true of national identity for the reasons I have outlined.

One objection to this line of argument is that religious groups can have the features of territoriality, deliberativeness, and nestedness. Many religious groups, after all, have a strong attachment to a particular place: think of Jews and Israel, for example. ²² Others have elements of deliberation and contestation: Quakers are a good example of this. And some religious identities can be nested: Shintoism and Buddhism come to mind here. I have two replies to make: the first is to note that I am not saying that religious groups cannot have these features taken in isolation. But they rarely have all three. Most importantly, when a religious group does have all three, it is hard to explain why it does not count as a nation. This is particularly true I would note, for religious identities that can be characterised as more “practice-based,” and which do not appear to be incompatible with other religious identities.

A final objection is that my view of nationality is too rosy. A critic could claim that most nations’ claims to self-determination wouldn’t survive the Alienation Defeater. I am perfectly happy to accept this claim. My goal in this article has been to point out that nationality can have certain features that make it acceptable from the point of view of liberal political morality as a basis for majority collective identity. If a given national identity doesn’t have those features—if it is narrowly ethnic in content and closed to outsiders—then it isn’t acceptable from the point of view of liberal political morality, and so doesn’t have the right to unilaterally secede.

2.e Political identities and nationality

Recall that, up to this point, I have been arguing that nationality is normatively “special” when it comes to the right to secede. A critic of this line of argument could respond that the account of nationality defended here is too thin to distinguish nations from mere regional groups with distinctive political values. Take the people of Alberta, for example. The Albertans are not ordinarily reckoned to have a distinct national identity. They do, however, seem to have different political values from the people of the rest of Canada. These political values are subject to deliberation and contestation. Moreover, the Albertan identity seems to be a nested identity—it is compatible with a broader Canadian identity. And Alberta is a territorial entity. So it seems, prima facie, that there is little to distinguish nations as such from groups that happen to have distinctive political values.

The problem with this objection is that it moves from the claim that (1) shared political values can be an important element of nationality to the claim that (2) shared political values are (in conjunction with territorial concentration) sufficient to make a group a nation. Although this move is clearly illicit, we should take some time to distinguish between what we might think of as purely “political identities” and national identities in order to see why political identities are not good candidates for self-determination.

First, what kind of relationship in general do people who merely share political values—as cashed out in ideological similarities, voting patterns, party membership, and so on—have to one another qua sharing political values? They are not likely to see their relationship with those who share their values as anything more than instrumentally valuable—instrumental for the sake of the promotion of those values. They might attempt to foster certain bonds of solidarity between political comrades through songs, associations, and so on. But this is usually seen as instrumental to their shared political goals.

Co-nationals, in contrast, tend to see their relationship with one another as intrinsically valuable—as mattering for its own sake, regardless of any further goals or goods it may facilitate. That is not to say, of course, that the co-national relationship is not (typically) of benefit to parties of it. In fact, the fact that co-nationality is typically beneficial is often used as evidence for the claim that the relationship is intrinsically valuable! As Seth Lazar points out, this is the typical strategy theorists take when trying to defend the claim that some relationship is of special significance (2016, 30–31). What are some of the benefits that people derive from their relationship with their co-nationals? Here are some good candidates:

1. (i) The development of valuable capacities and dispositions: regarding yourself as belong to a nation is a way for you to develop valuable capacities, such as your capacity for reciprocity and fellow-feeling. In addition, it may lead you to develop appreciation for, for example, pieces of music, landscape and works of art, by virtue of regarding them as part of your national heritage, where you would not have appreciated them otherwise. ²³

2. (ii) Providing a secure source of self-respect and esteem: nationality, as mentioned above, is usually a matter of belonging, not of choice. The achievements and positive features of nations are a source of self-respect and esteem for their members. Because membership is a matter of belonging, national membership, provided things are going well for the nation itself, is a highly secure source of self-respect and esteem. ²⁴

3. (iii) Interpersonal projects and transgenerational value: a person’s relationship to their co-nationals is a way for them to put their lives in the context of a larger project, one that extends across generations. (Miller 2005, 68–69; Gans 2003, 52–54)

4. (iv) Cultural goods: nationality provides people with cultural goods such as a historical narrative, music, national literature, folk customs, etc. Part of the benefit that people derive from having these goods as elements of their national culture, as opposed to the common heritage of the world, is dependent on the fact of their relationships with their co-nationals. (McMahan 1997, 130)

Of course, some of these goods or benefits are benefits that may also be generated by forming purely political associations with others. But the range and depth of the benefits that can be produced by the co-national relationship clearly distinguishes it from purely political association.

What is the upshot of this as far as self-determination is concerned? The upshot is that nations are far stronger candidates for the right of self-determination than people in purely political association because their relationship with one another is far more important and worthy of protection by political means. The interest that co-nationals have in protecting both their national identity and in setting the terms of their relationship with one another is of greater significance. Now of course, this claim is in part dependent on what account of self-determination one prefers—if one thinks that the ground of collective self-determination is something like political autonomy, or some Rousseauvian conception of individual self-determination, then one might be inclined to say that purely political associations are more worthy of protection. But of course, the nationalist theorist of self-determination is not likely to think along these lines, and our central task here is to defend the nationalist theorist from charges of arbitrariness, not to defend his or her entire account of self-determination.

At this stage, a critic might object that I am re-injecting nationality with too much nonpolitical content in a way that makes it vulnerable to the Alienation Defeater. In response, I have to ask why nonpolitical cultural content need be necessarily exclusive in a way that political values need not be? The cultural content need not be the thick sort of cultural content characterised by ethnic folk practices, national dress, etc. It needs to merely be the sort of thing that comes about in virtue of people living a national life in an age of mass communications together—for example, having a shared memory of particular sporting events, shared jokes and cultural references, and so on.

Conclusion

I have argued that it is not arbitrary for a theory of secession to single out nations. Unlike religious identity, nationality can serve as a basis for collective political identity without being impermissibly exclusionary towards minorities. My main goal in this article overall has been to outline the conditions under which the setting up of a new state is compatible with substantive equality of citizenship for people who do not share the dominant identity of the new state, whether that be national or religious. My argument has been that a state where the dominant identity is purely religious in character will not produce social and political conditions that are compatible with this substantive equality, whereas states where the state’s dominant identity is a national identity (including national identities that have religious membership as a substantial component), may well be satisfactory from the point of view of minority citizens. I have outlined some features of nationality that explain why this is the case. My conclusion, therefore, is two-pronged: (i) national secession can be compatible with the claims of minority groups and (ii) nationality is special in this regard. With this, I hope to have answered Allen Buchanan’s question: “What’s so special about nations?”