A defense of the moral and legal right to secede

ABSTRACT

We defend the moral and legal right to secede in accordance with plebiscitary theory. Our paper has three main goals. First, by offering a schematic characterization of plebiscitary theory, the main arguments in its favour (and the main objections to them), we contribute to clarify the structure of this complex debate. Second, we stress the point that, if the moral right to secede is established, the resistance for its inclusion into positive law is unjustified. Finally, by addressing old and new objections to plebiscitary theory, we hope to make a compelling case for a wider recognition of secessionist rights.

KEYWORDS:

• Secession
• national minorities
• minority rights
• associative rights
• multiculturalism

Introduction

The normative status of secession is incredibly controversial. The debate around it has nowadays heated practical instances and longstanding theoretical discrepancies. In the theoretical realm, it is common practice to group the different positions in two main camps: remedial-right theories and plebiscitary theories. It is well-known that remedial-right theories see secession as a very last resort and are thus quite strict on the conditions that normatively justify particular instances.¹ In particular, Buchanan, 1991, 2004, has famously defended that the secession of a given group is justified only when one of the following conditions is met²

1. Violation to human rights: there are severe and sustained violations to minimal standards of justice targeted at the secessionist group.

2. Unjust annexation: the secessionist group has been recently and unjustly annexed by the given state.

3. Violations to autonomy agreements: the state has violated agreements conferring degrees of autonomy and self-determination to the secessionist group.

To these conditions posed by Buchanan, some authors (see Costa 2003; Patten 2002, 2014) have added a further one:

• (4) Lack of recognition: the state has failed to establish institutional provisions that guarantee the equal recognition of the secessionist national minority.

By adding this further condition, both Costa and Patten think that a remedial-right theory can respond to the important claims of equal recognition that national minorities usually make against their host states. For with condition (4) on the table, it becomes clear that the lack of equal institutional standing alongside the larger national group is a justified cause of secession. According to this form of remedial-right theory (for short, RRT), thus, each of these four conditions suffices to normatively justify an instance of secession. Let us schematize this claim as follows:

RRT: (1)-(4) and only (1)-(4) are individually sufficient to justify that a group has a right to secede, even if such right is not recognized by its host state.

Conversely, plebiscitary theories are less restrictive. There is a rich variation on the different forms that these theories have. For the purposes of this paper, we will take the plebiscitary theory to subscribe the following claims:

• (5) Eligibility: the members of the secessionist unit form an eligible group for secession.

• (6) Fair terms: the proposed terms of secession must be fair to the current host state.

• (7) Regional security: the new state will not pose a threat to the security of the current host state.

• (8) Justice: the new state will not violate basic rights and further standards of justice.

• (9) Equal recognition: the new state will offer equal institutional standing to the different minority groups within its territory.

• (10) Democratic determination: the decision to secede from the host state will follow democratic procedures either in collaboration with the host state (if they actually exist) or at least in the territory of the secessionist unit.

Conditions (5) to (10) delineate the commitments that a specifically liberal plebiscitary theory should meet. As it is well known in the literature (and as we will defend here at length), plebiscitary theories are less restrictive than RRT in permitting secessions because of the primary importance they put on the political autonomy and self-determination of the preferred secessionist groups. However, by being committed to a broader liberal view, such values should be coherently balanced with other liberal commitments – such as respect for others and their concerns for a just distribution of resources, their physical integrity, etc. These conditions show, then, that political autonomy is not an absolute right – see Philpott (1995): 181–183. They also show why most authors defending plebiscitary theories think that a right to secede is compatible with a broader liberal view – and, as we will see, that it is also a requirement of it.

Accordingly, condition (5) specifies a very first filter. It identifies which groups can qualify as secessionist groups. In the literature, there are two positions in this regard. According to Margalit and Raz (1990), Miller (1995), Neran 1998, Kymlicka (2001), Bossacoma (2020), and others, only national minorities (i.e. groups that share a particular cultural identity and whose existence usually precede the formation of their current host state), if any, may have the right to secede. Contrarily, according to Beran (1984), Wellman (1995, 2005) and others, any group with the capacity to form a viable state has this right – regardless of whether they share a cultural identity or not. We will not take a stand on this debate here and, thus, our defence of the plebiscitary theory will cover both positions.³ Notice, however, that these two positions in the plebiscitary theory literature are exhaustive: when taken together, the only group of people that is excluded from the possibility of seceding from a state is the one who could not form a viable state. Since the aim of seceding from a state is to form another viable state, excluding that possible group from this right seems to be justified.

Conditions (6) and (7) try to prevent a negative impact of a given secession on the state that experiences it. On the one hand, condition (6) is usually taken to mean that the separation of a given region should be acceptable considering the economy of the larger state, the division of natural resources, and the like. This ensures that secession will not be too burdensome economically speaking for the old state (see Gauthier 1994 for a radical version of that claim; see also Dietrich 2014; Catala 2017). On the other hand, condition (7) guarantees that the newly formed state will not be a permanent and intractable threat to the old state. This ensures that secession will not create a long-standing security problem in the region (see Philpott 1995, 181). Again, these two conditions show that political autonomy is not an absolute right, and that its correct exercise should respect other values as well, such as non-domination (see Catala 2017, 539–546), equal respect for others, and reciprocity (see Moore 2015: 128-134).

Conditions (8) and (9) regulate the internal organization of the new state that a given secession will bring about. Condition (8) ensures that the new state will properly meet standards of justice for its citizens. Condition (9) ensures that the new state will respect the rights of the minority groups that will now leave under its jurisdiction (see Philpott 1995, 1998, 92). Allocating these rights will always be a complex contextual matter; however, a good normative guide here consists in trying to identify the nature of the group in question (i.e. whether the group is a territorially concentrated national minority, a community of disaggregated nationals, etc.). These two conditions are also of primary importance for liberal plebiscitary theories to be successful. As we mentioned earlier, the main argument in favour of such theories is based on the value of autonomy as such, and thus not respecting such a value and its consequences with regards to liberal justice and minority rights for all the members of the new state would simply be incoherent.

Finally, condition (10) sets a standard regarding how the procedure that might or might not end up in secession should be conducted. In particular, this condition is committed to the idea that if feasible democratic procedures already exist or have been agreed with the host state, these must be employed and followed by the minority seeking to secede. However, and this is why the discussion that follows centres mostly on a unilateral right to secede (as opposed to cases of consensual secessions – see Buchanan 2017), condition (10) states that when no agreement between the host state and the minority is in place, it suffices that secession is sought through democratic procedures at least in the secessionist minority’s territory – if, of course, all previous conditions can be met.

We are now in a position to define the plebiscitary theory more succinctly. For the purpose of this paper, we will take such a theory to state:

PT: A group has a right to secede, even if such a right is not recognized by its host state, if and only if one of (1) to (4) holds or all of (5) to (10) hold.

In this paper, we will defend the PT so defined.⁴ In the next section, we will present what we believe is the most compelling moral argument for PT – which is based on the values of individual autonomy and collective self-determination. We will defend this argument from three common and renewed objections against it: what we call the autonomy objection (i.e. that collective self-determination does not foster individual autonomy), the sovereignty objection (i.e. that the sovereignty of the secessionist group’s territory resides on the whole of the state) and the value of multinational states objection (i.e. that dismantling a just multinational state is wrong). Now, interestingly, even those who are highly sympathetic to plebiscitary theories are sometimes reluctant to recommend the incorporation of this right into positive law (see Altman and Wellman 2009) or remain silent on this issue. Thus, in the last section, we will respond to the most common objections against the legal right to secede. Most of them constitute different versions of what we call the stability objection (i.e. that seceding from a just state creates unjustified instability).

We take our paper to have three main strengths. First, by setting up the main views, arguments, and objections, we contribute to clarify the structure of this complex debate. Second, in an effort to connect these two issues, we defend in a single paper both the moral and legal rights to secession. Finally, we address recent and recalcitrant objections to plebiscitary theories. All in all, we hope these arguments make a compelling case for a wider recognition of secessionist rights.

The moral right to secede

Why is it that a group satisfying (5)-(10) has a moral right secede? The argument we will present follows a line of thought in the literature that begins with the next uncontroversial premise

(P1) The individual autonomy of a group’s members should be respected.

(P1) simply asserts the importance of respecting the individual autonomy of persons; as such, it is a statement of one of the most important premises in which liberal democracies are built. As it is well known, different and prominent defences of such a premise can be found in Mill, Rawls (1999), Dworkin (2000), Raz (1986) and Kymlicka (1989) – to mention a few. It is common ground of these authors that persons must be able to build their own lives, decide for themselves which plan they want to pursue (and change it if so they wish), being able to partake on equal footing in the political organization of their society, etc.

In turn, some authors have tried to ground the value of group self-determination on (P1). According to Philpott (1998), for instance, a constraint on the right for self-determination would arbitrarily limit citizen’s autonomy and should be rejected for this reason. More recently, Stilz (2015, 2016) has defended that citizens have an interest in being the authors or ‘makers’ of their own life. This interest requires being able to ‘affirm’ their own political institutions, and that is what ultimately grounds the value of a group’s self-determination. Consequently, and ideal political unit is one in which not only the value of justice is perfectly realized but also one that is ‘self-determining – i.e., it would rest on the shared will of its people’ (Stilz 2016: 125). She also notes that only by assuming this understanding of self-determination we can explain the wrong of a putative benevolent colonialism – i.e., taking control of another country to impose a perfectly just political regime without its consent (Stilz 2015: 8). Despite the fact that Philpott and Stilz’ arguments differ in significant respects,⁵ they point to the very same idea: that collective self-determination is grounded in a specification of the value of individual autonomy:

(P2) Respecting a group’s moral right to self-determination is a necessary condition for respecting the individual autonomy of the group’s members.

From (P1) and (P2) we can infer

(P3) Thus, a group’s moral right to self-determination should be respected (from P1 and P2).

We think this is a compelling reasoning for the recognition of a group’s right to self-determination. As we saw, (P2) is supported by important reasons that explain why benevolent forms of colonialism are wrong. Furthermore, note that (P2) is compatible with reductive and non-reductive views of group rights on the rights individuals. In any event, for our purposes we only need to vindicate (P3). Accordingly, if there are further reasons for vindicating a group’s right to self-determination (for instance, if one thinks groups have properties that ground this right independently of its contribution to individual autonomy – see Altman and Wellman 2009: 4-5 – or that granting self-determination to a group is required to treat its members as equals – see Copp 1997: 291-292) this is perfectly congenial with our main strategy.⁶

Once respect for the group's self-determination is established, the argument in favour of secession seems to follow. Again, with important differences and caveats, Philpott, Altman & Wellman, and others defend a distinct version of this claim. In our case, the premise takes the following form:

(P4) The moral right to secede of a group that satisfies (5)-(10) is a justified exercise of a group’s right to self-determination.

With (P4) on the table, we can finally

(C) Thus, a group that satisfies (5)-(10) has the moral right to secede (from P3 and P4).

This is the main moral argument that defends the most controversial part of PT as we have defined it.⁷ The main two elements of it are, first, vindicating the idea that groups have a right to self-determination and, secondly, stating that when a group satisfies conditions (5)-(10) its right to secede is a justified instance of self-determination.

How could a critic of PT challenge this conclusion? Given that (P1)-(P3) seems hard to deny, the literature against PT usually focuses on objections to (P4). Here we will revise what we consider the three most salient and frequent of them: we call them, respectively, the autonomy objection, the sovereignty objection and the value of multinational states objection.

The autonomy objection

Some critics object (P4) by stating that secession is not a justified instance of self-determination precisely because it does not foster individual autonomy in the best way (see Buchanan 1998 17–18; Patten 2002, 574-580). For when a group of people decides to secede, those who disagree with the majority would see their autonomy violated; as soon as there are dissenters in the seceding territory, some people would be deciding for others. We call this the autonomy objection.

This is a fair point, but note that this happens with any policy implemented democratically: i.e. it is always the case that some citizens do not see their preferences realized. Accordingly, in this respect, a democratic secession mandated by PT is no different from any other legitimate democratic decision: it respects the decision that is in accordance with the preference of the majority of those living in the secessionist unit – despite being unable to meet all relevant preferences.

The same idea can be expressed with the notion of a ‘weak right’. In a different context, Gauthier (1994): 360 claims that each person has a weak right to associate with whom she wishes, where a ‘weak right’ is one whose exercise must be coordinated with that of others in such a way that as many persons as possible will find themselves in mutually desirable association. Thus, self-determination seems to support autonomy understood as a weak right; conceding to a group the right to self-determination approaches the ideal in which as many people as possible end up in their desired association.

However, Patten has challenged this idea by noting that secessionist claims are always advanced in the context of identity pluralism. Because of this, in secessionist territories there are at least three types of persons with specific preferences: those who identify exclusively with the nationals struggling for the creation of the new state, those who identify as nationals of the larger state and those who have a dual identity (see Patten 2002: 579). In this context, a democratic secession would only foster the autonomy of the individuals of the first group and because of that Patten thinks it would not be the best result – even if such a group were the largest one:

This kind of result does not come very close to the ideal in which everyone finds themselves in their preferred association. Although it is true that the majority are able to put themselves in exactly the association they most prefer, the minority are excluded from their preferred association altogether. In general, however, a distribution in which all get some of what they want should be regarded as superior to one in which a majority gets all of what they want while the minority get none. Patten (2002): 579.

According to Patten, assuming that a multinational state already guarantees equal institutional recognition to the nationals of the secessionist minority, the best way to maximize the preferences of all individuals living in the secessionist territory is to uphold the existing multinational state. For the individuals of the three mentioned groups get ‘some of what they want’: the first group is ensured the institutional recognition of its identity (albeit within the existing state), the second group is granted its belonging to the state they recognize as theirs, and the third group maintains its dual identity.

This is an interesting reasoning that opens serious questions in regard to the aptness of the majority rule. However, there are at least three complications with it. First, once it is discarded that a simple or qualified democratic majority is the strategy that mostly respects the autonomy of group members, Patten owes us an account of how many people on groups two and three are needed in order for their preferences to tip the balance against secession. For instance, is he recommending a general principle according to which if some people want A and not B, others B and not A and a third group prefer A and B, then A and B should always be implemented, irrespective of majorities? That looks like a very implausible principle to us. For one thing: in any secessionist dispute some people are attached to the larger state and some other to the secessionist group so that, if majorities were not taken into account, all cases of secession would be illegitimate. This result not only makes resistance to secession too cheap (one just needs to find some people with allegiance to the two groups) but in certain cases would lead to very counterintuitive results. For instance, if 90% of the population in the territory identifies as nationals wanting to secede, 5% as nationals of the current state, and 5% as both, why is it that forming a new state is not the best way to respect everyone’s individual autonomy, given that all had the same opportunity for political influence by means of their vote? Surely, it would be important to try to meet the preferences of these last two groups; however, it seems that the best result in this case is to comply with the decision of the vast majority. This is especially clear once we remember that, according to PT’s condition (9), full institutional recognition and rights would be granted to minorities in the state formed after secession.

Second, Patten’s argument is committed to another deep counterintuitive consequence: let us suppose that we can divide the whole state, and not only the secessionist territory, into the same three groups of citizens (those who identify exclusively with the nationals struggling for the creation of a new state, those who identify as nationals of the larger state, and those who have a dual identity). Now, suppose that 80% of all citizens in this state vote for the secession of the given territory. This would be an example of a non-unilateral secession in which a majority in the entire state thinks the best thing is for a region to secede. However, if Patten’s argument is correct, it follows that it would be preferable to maintain the already existing multinational state since this arrangement gives all citizens some of what they want in regard to their associative preference.⁸

Finally, Patten’s argument seems to commit a common pitfall of restrictive accounts of secession: they confuse reasons against secession for reasons against the right to secession (for more on this see footnote 11 below). Even if one thinks that in a certain case secession is not the best option, this should not be used as an argument against the right of a certain group to decide which is the option they prefer.

The sovereignty objection

Another quite frequent objection holds that (P4) is incorrect because the sovereignty of the secessionist territory resides on the host state as a whole (see Buchanan 2004: 374; Miller 1998: 68). According to this view, a country’s sovereignty is not divided into parcels which should be ruled by those who happen to live in a certain region, but it is the whole body of citizens that belong to a certain state who have a legitimate claim to exercise the authority over the state’s territory and should have a say in any decision that might affect it. For it is a distinctive feature of secessionist claims that they presuppose that a group can legitimately take a part of a territory in order to create a new state, and this territorial dimension of secession is one of the central features distinguishing it from other forms of self-determination (see Brilmayer 1991; Moore 1998). Drawing on this essential but often neglected aspect, some have tried to argue that the claim for a part of the state’s territory is illegitimate. This idea has sometimes been presented as a challenge: the secessionist should justify why a group of people is entitled to take away a part of the state’s territory (see Copp 1997: 281; Patten 2002). We call this the sovereignty objection.

To this objection we have two responses. First, the analysis of territorial rights is a complex and unresolved issue (Ypi 2013; Moore 2020). Furthermore, extant approaches fail to analyse in detail the connection between territorial rights and theories of secession (Buchanan 2017). Some accounts (e.g. Miller 1995; Kolers 2009; Moore 2015) might be able to justify PT’s assumption that a group other than the whole state has a territorial right over the seceding territory, whereas this claim might be harder to defend in other frameworks (e.g. Buchanan 2004). Still, the connection between the two debates is far from straightforward. Buchanan (2004), for instance, endorses a legitimacy-based account, which grounds territorial rights on the state’s ability to protect human rights and to guarantee the rule of law. However, an objection against this type of account is precisely that ‘all that the legitimate-state argument can show is how boundaries could be protected, not how they ought to be drawn’ (Ypi 2013, 250). For, within the debate regarding territorial rights, it is hard for the legitimate-based account to explain why it is not compatible with putative benevolent forms of colonialism – see Moore (2015): 107-108 for this and further problems.

Consequently, although we certainly agree that this point needs to be further developed by theories of secession, the sovereignty objection simply assumes a particular take on the complex debate of territorial rights that is incompatible with PT. Thus, we think the objection as such begs the question against PT. Considering this, PT defenders can legitimately place the same challenge over PT deniers: i.e., that they should justify why it is that only the state as a whole is entitled to allow separation from its territory.

Second, if the sovereignty objection is correct, then again counterintuitive consequences follow: suppose that 49% of the population of the entire host state votes in favour of a given secession, including a 100% of the citizens living in the secessionist territory. This would be an example in which citizens in the secessionist region unanimously want to secede and pretty much half of the citizens in the entire host country approves this result. Still, if Buchanan’s view of popular sovereignty over territory is correct, this secession will be unjustified because the majority of the host state is not willing to give up part of its territory. Although intuitions might differ in some respects, we think rejecting the legitimacy of secession in these situations is a cost that non-plebiscitary theories have to pay.⁹

The value of multinational states objection

The third and final objection we would like to consider against (P4) holds that it is wrong to secede from a just, or sufficiently just, multinational state. A recent and succinct version of this argument is found in Patten (2014). Like Costa (2003), Patten argues that the existence of national minorities imposes certain requirements on just states. In particular, the state needs to introduce meaningful constitutional arrangements that recognize the distinct national identity of the members of national minorities. Failing to adopt an appropriate institutional response to a situation of identity pluralism undermines the principle of equal recognition and, as a result, gives rise to a valid claim for secession – claim that is captured in condition (4) above.

Now, Patten argues that plebiscitary theories fail to respect arrangements that establish an equal recognition for the different national minorities in a certain community. To illustrate the problem, he asks us to consider a state S in which two main national identities can be found: a national identity focused on the whole state S and another identity focused on a substate unit N. Additionally, we are to suppose that S is a just state in the sense that it contains meaningful constitutional arrangements that recognize these different national identities (and, of course, one that complies with all further requirements of justice – such as respect for human rights and sufficient standards of distributive justice). Patten rightly suggests that PT would allow the national minority in N to secede from S and argues that this consequence would be unacceptable:

In the case being considered, this theory gives the go-ahead to a secession that dismantles arrangements providing for equality in the recognition of the different national identities prevalent in the secessionist region and replaces them with a new set of arrangements that exclusively recognize the substate identity. Prior to the secession, a multinational federation was in place that provided institutional space for the expression of both the substate and statewide identities. By putting together a majority in a referendum on secession, however, those bearing the substate identity are able to obtain a monopoly on the good of recognition and shut the minority statewide identity out completely. (Patten 2014: 252).

In a nutshell, Patten’s objection is that N’s secession not only would break a whole set of constitutional arrangements that comply with the condition of equal recognition but would also substitute them for a political unit that fails to satisfy this requirement. He convincingly argues that this is a situation in which secession should not be permitted (for a similar argument, see Miller 1998).

Yet this objection assumes that the seceding state would fail to include the meaningful arrangements necessary to fulfil the equal recognition requirement. As Patten admits, ‘a well-founded expectation that a new set of confederal arrangements would be established in the aftermath of secession would diminish, and perhaps nullify, the recognition-based objection to plebiscitary secession’ (Patten 2014, 263). But why should we assume that the new state would be unable to comply with this desideratum? Note that the whole force of Patten’s objection depends on justifying the claim that the new state will fail to appropriately recognize its own national minorities. This assumption, however, needs to be backed up. In its support, Patten mentions the difficulty of creating successful institutions that are able to guarantee the equal recognition of all national groups in a given state since to get such institutions up and running requires a degree of genius at institutional design, together with a high level of trust and cooperative commitment on the part of the major parties:

There is no guarantee that an intention, however sincere, on the part of one or more of the parties concerned to build institutions providing for the recognition of different national identities will meet with success. (Patten 2014: 263)

Unfortunately, this reason seems too weak to support the value of multinational states objection. Although we certainly agree that it might be insufficient to merely assert the intention to create respectful multinational arrangements in the new state, there are different strategies for making this intention more credible – such as a legal commitment or allowing some supranational organization to supervise the process. On the other hand, whether the new state can be trusted to appropriately recognize its national minorities seems to be an empirical question that should be assessed on a case-by-case basis. For this reason, the importance of ensuring a just treatment to minorities should not be used as a general argument against PT but as a qualification that, when it is not properly met, offers sufficient ground to invalidate a given claim to secede. PT contemplates this qualification as we defend it, for it includes the requirement that a newly formed state produced by a given secession should respect and confer equal institutional standing to national minorities (i. e., condition (9)).

A PT opponent could insist that a secessionist unit is wrong in dismantling a just multinational state even if this is for the purpose of creating another fully functioning just multinational state. After all, a just arrangement is already in place. By arguing in this way, this PT opponent would seem to commit herself to a particular view on the debate regarding the justification of political obligation: i.e. the natural duty account, according to which citizens have a natural duty to uphold and preserve just institutional arrangements (see Waldron 1993; Wellman 2005; Stilz 2013; for different defences of this view). So long as secession interrupts the existence of a just multinational state, then it is something that we should regret.

However, since this claim against PT is an affirmation of the natural duty account of political obligation, it must respond to the most famous objection that such account confronts: the so-called particularity objection – i.e. the claim that, even if it is accepted that we have a natural duty to support just institutions, we could discharge such duty within different just arrangements (see Simmons 1979: 31, 2005: 166). In fact, this objection to the natural duty account mirrors the way we have responded to Patten’s argument, for we pressed the idea that further fully functioning multinational states could be the result of secessionist processes. So, even if we assume that the natural duty account is right, and thus we are obliged to uphold just institutions, why is it that agents cannot discharge such obligation in a newly formed just state?

A common response to the particularity objection appeals to the impracticability of allowing every person to discharge her duty in any just institutional arrangements she pleases (see Quong 2010: 129).¹⁰ The volatility that this would generate does seem to make the correct fulfilment of the duty extremely difficult for all. However, notice that PT already takes into account the viability of the potential new states that secessionist processes might bring about. For, in accordance with condition (5), not every single person in the world would have the right to secede from the current state she lives in and form new political institutions. Rather, a person must be part of a group who is readily eligible to opt for secession and subsequently for forming a viable and fully functioning state. We will come back to the putative impracticability of PT in the next section. For now, it suffices to say that there seems to be nothing wrong with the creation of a new just multinational states as PT allows – even if a particular view on the nature of political obligation, such as the natural duty account, is adopted.

The legal right to secede

In the previous section of the paper, we argued that there is a moral right to secession as established by PT. In this section, we would like to defend the claim that this right should be incorporated into positive law, both at the international and the constitutional level. Buchanan has extensively argued against this claim and even prominent PT defenders have also joined him in this regard (see Altman and Wellman 2009: 56). This resistance mostly derives from the thought that conceding a legal right to secede to any group that satisfies (5)-(10) could provoke excessive instability. We call this reasoning the stability objection.

In what follows, we will concentrate on four different versions of this objection. Three of them point at distinct types of instability (transitional, regional, political) that the legal recognition of PT allegedly creates, and the last one argues for agnosticism with regards to the incorporation of such a right into positive law. We will hold that, as PT defenders have long stated (Copp 1997; Philpott 1998; Weinstock 2001; Bossacoma 2020), a legal right to secede does not create pervasive transitional, regional or political instability. In fact, for the reasons, we will give in this section, we think that the worry framed by all forms of the stability objection against the legal right to secede get things the other way around: it is not that providing the right to secede in accordance with PT would intensify regional or political instability in states where conflicts between minority and majority groups exist; rather, not granting these right increments such conflicts and the instability they generate.

The transitional instability objection

Consider, first, what might be called transitional instability: secession involves the modification of the boundaries of the host state and the creation of a new political unit, and this process will undoubtedly change the status quo; furthermore, populating the world with new states would demand the establishment of new international relations and a significant redistribution of powers. An extreme form of this objection suggests that PT is simply unfeasible (see Norman 1998).

Despite the common invocation of this reasoning, transitional instability as such seems to be unobjectionable. For note that in any event only the stability that results from fair political relations should be respected. Surely, a dictatorship could be extremely stable, and yet this on its own does not provide normative reasons against fighting for a just regime, even when such a struggle might create instability. This attests to the claim that, at least when it comes to the creation of new viable and just multinational states, the stability of old multinational states on its own offers no normative reasons against it. Furthermore, as defenders of PT usually point out, recognizing the right to secede does not entail that a significant majority of groups will exercise such a right (see Copp 1997; Philpott 1998: 90; Wellman 2005), and thus the practical consequences of adopting PT might not be as broad as it is sometimes suggested. Accordingly, for the two reasons just mentioned (i.e. that there are no normative reasons for protecting stability as such and that not many minorities would exercise their right to secede), the fact that PT would allow some transitional instability is not a compelling objection against it.

The regional instability objection

So let us switch to what might be called regional instability. One might worry that secessions would lead to a large amount of regional instability, including an increase in territorial tensions or violent conflicts. As Norman (1998) and Buchanan (2004) state, evidence for this claim seems to be readily at hand: recent world history provides a large number of examples of secessionist movements that have caused or suffered considerable violations of human rights.

We certainly do not want to deny that secessionist processes have often involved such terrible forms of regional instability. However, there are four reasons for thinking one should not use this tragic set of examples against PT. First, note again that PT specifies necessary and sufficient conditions for a group to have the right to secede, but it does not entail that once these conditions are met a group should secede. In other words, the right to secession involves a privilege and a claim–right (see Copp 1997; Buchanan 1993), but it does not provide normative reasons for seceding.¹¹ As we have already pointed out, it does not follow that from granting the right to secede a massive bulk of secessions will or should necessarily occur. Second, in some situations in which a group already strives for secession, it is unclear that merely conceding this right would affect regional instability in any significant way (see Weinstock 2001). Third, in fact, it may well be that in many cases providing the right to secede even helps in reducing regional tensions (see Costa 2003: 83). It is not obvious, for instance, that a sufficiently large majority of Corsicans, Catalans or Basques would vote for independence, and nonetheless it is undeniable that conceding them this right would decisively promote regional stabilization. Also, as it became clear in the recent Scottish referendum on independence, conceding such a right might contribute to debunk one powerful argument in favour of secession, since the very idea that a group is forced to stay in a union is usually a determining reason for some to try to abandon it. Finally, we recall that some of the conditions included in PT (such as (8) or (10)) restrict the right to secede from those groups who would engage in fully democratic and peaceful processes. Thus, it is just wrong to think that PT would concede a right to secession in those cases in which regional stability could be jeopardized.

The permanent political instability objection

So let us switch to a third form of instability that might follow from PT’s legalization; let us call it permanent political instability. As Sustein (1991) and Buchanan (1991, 1997, 1998, 2004) have insisted, conceding the right to secede might provide wrong incentives both to states and minorities. With regard to states, it is argued that this right would discourage them to promote decentralization – out of the fear that the creation of strong internal political units would set the ground for future regional plebiscites. Likewise, this right might create perverse incentives with respect to immigration policy and economic development: states would try to limit internal and external immigration to certain areas that might result in a majority in favour of secession. Finally, states might deter regional economic development to prevent the creation of viable and economically superior states from those regions. With regard to minorities, on the other hand, it is argued that making exit from a state as easy as PT prescribes would allow regions to blackmail their states with secession in normal politics, and thus would give minorities an extremely unfair negotiation advantage in all types of political disagreements and resolutions.

We agree with Sunstein and Buchanan that these are possible consequences of adopting PT. However, there are two ways of showing that they fail to put pressure on PT. First of all, note that there is a tension in Buchanan’s presentation of the wrong incentive argument. On the one hand, he claims that PT gives regions the power to blackmail states in all areas of political life (Buchanan 1991: 98-100, 1997: 48). A consequence seems to be that PT would favour decentralization and regional development (because, otherwise, a region could go ahead and create its own state). However, he also claims that PT would encourage state centralization and a serious limitation of regional economic distribution (out of the state’s fear of the creation of sub-state political units) (Buchanan 2004: 377). So, if Buchanan’s analysis was right, both incentives would push in opposite directions, and this might lead to a technical tie. At the very least, the opposite direction of these incentives points to a situation of constant negotiations between the state and its regions, which should be a normal political procedure in a just and democratic multinational state. As a result, these putative consequences of PT might not be as ‘perverse’ as Buchanan envisages.

Second, there are reasons for thinking that, in fact, not granting a right to secede creates worse motivations for both types of political units than conceding such a right. For in many cases the fact that the state’s government knows that a group cannot legitimately create a new state has given it reasons for resisting decentralization. After all, if a minority can systematically be disregarded and the region can never claim for secession, why would the state ever take into account their views? On the other hand, consider what happens when a minority is denied the right to secede. It is undeniable that some members of national minorities around the world take secession as their primary political goal. Is it not the best thing for a country that they advance such a goal by normal institutional and political processes? By taking secession out of the political agenda in advance and for good, central powers might be opening wrong avenues to pursue such a goal; that is, putting secession out of the political agenda might do more wrong than good when it comes to political stability.

From this, it seems to follow that, with regard to states, there is at least one very important positive incentive that arises from PT: in general, conceding the right to secede might promote the state’s interest in national minorities feeling respected and well–treated. For once PT is adopted, limiting regional economic development or the free circulation of citizens will become irrational polices to pursue. In other words, providing the right to secede may help in achieving what recent RRT defenders (such as Patten and Costa) correctly worry about: granting full institutional recognition to minorities within multinational states – knowing that, if they are mistreated, they could engage in a secessionist process.

Third, and finally, consider the last negative incentive extensively developed by Sunstein: the possibility that national minorities blackmail states with the threat of secession. While this is a real option, two important remarks need to be made. The creation of a new state would surely involve uncertainties and impose large costs on the seceding group. For this reason, few of these threats involving secession would be credible enough for being strategically employed. Furthermore, PT’s condition (10) requires that the decision to secede should be made by fully democratic procedures, so this provision would block a purely governmental blackmail not supported by a majority of citizens. We believe these two constraints would limit very much the group’s capacity for blackmailing its host state. By making secessionist processes so costly to embark on, secessionist threats in normal politics would severely diminish their effectiveness (see Weinstock 2001: 194–200).

For these reasons, it is simply not obvious that the incentives encouraged by PT would be significantly worse than the ones that follow from rejecting it. Furthermore, as we have argued, contrarily to what Buchanan and Sustain defend, there is even reason to think that the adoption of PT might provide states with incentives to offer minorities the institutional recognition they deserve as full members a just multinational state; adopting PT might also provide minorities with incentives to advance their secessionist agenda through normal institutional procedures for conflict resolution, ensuring right political deliberation and stability.

The agnostic objection

Finally, it is worth mentioning that even some prominent PT defenders hesitate to state that the moral right to secede should be included in international law. In particular, Altman and Wellman have recently stated

In our view, the arguments that we have given about a primary moral right to secede do establish a pro tanto argument – even if a highly abstract one – for a primary right of secession under international law. If certain groups have a primary moral right to secede, then that fact establishes a good reason in favor of international law including the right among its norms. On the other hand, because the empirical consequences for human rights of creating a primary legal right are potentially so vast, the pro tanto argument does not by itself go far enough to justify even a tentative conclusion that there ought to be such a right. Instead, as we will argue, judgment should be suspended on any conclusion about a right to secede under international law until those potential consequences are far less uncertain than they are at this stage in the scholarly discussion of secession. Accordingly, as we see it, steadfast agnosticism is the most defensible attitude to take toward the question of whether international law ought to recognize a right to secession and, if so, what the scope of that right should be. Altman and Wellman (2009): 59.

Despite being passionate defenders of the existence of a moral right to secede, Altman and Wellman remain agnostic with regard to the inclusion of this right into positive law. Following Horowitz (2003), they even go one step further and hold that we should be agnostic regarding the incorporation of a right to secede in accordance with RRT – as Buchanan has defended. The main reason they offer in support of this agnosticism is that we are in no position to know whether a right to secede in international law would create serious human rights deficits in all states where there are secessionist movements and conflicts (Altman and Wellman 2009: 58–67).

We find Altman and Wellman’s position puzzling for three reasons. First, as we have argued, we think that a legal right to secede crafted in accordance with all the caveats that we have imprinted on PT could have a good impact in the resolution of conflicts related to specific secessions.

Second, note that Altman and Wellman’s very same reasoning could have been given against the international laws that allowed decolonization itself– and most people agree (we hope) that such laws had to be set in place. The process of decolonization involved a high variety of scenarios, many of which had unpredictable results. But if in this case the force of the moral claim in its favour sufficed to justify the legal right of colonized peoples to exist as independent states, we wonder why other cases involving similar uncertainties should be treated in a different way. More generally, Altman and Wellman’s reasoning seems to apply to all attempts of progressive international law reform and thus to favour the status quo.

Third, and finally, they themselves end up endorsing a proposal for the international positivization of the right to secede as extensively offered by Copp (1998) (see Altman and Wellman 2009: 66-67). Of course, they point out that such a proposal seems to be an ideal quite difficult to implement due to the state of current international politics. However, as they also point out, struggling for such an ideal is the best we could hope for. At this point they forget, though, that Copp’s proposal is an explicit attempt to incorporate a right to secede into international law.

Conclusion

In this paper, we have defended PT from four common and renewed objections against it: the autonomy objection (that collective self-determination does not foster individual autonomy), the sovereignty objection (that the sovereignty of the secessionist group’s territory resides on the whole of the state), the value of multinational states objection (that dismantling a just multinational state is wrong), and the stability objection (that including a right to secede into positive law would create unjustified transitional, regional, or political instability around the world). By doing so, we hope to have made a compelling case for a wider recognition of secessionist rights.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

¹ For discussion and comments on this paper and its previous drafts, we would like to thank to Arash Avisadeh, Enrique Camacho, Juan Espindola, Will Kymlicka, Pau Luque, Itzel Mayans, Margaret Moore, Faviola Rivera, Fernando Rudy, and two anonymous referees. We would also like to thank audiences in the Institute for Philosophical Research-UNAM, Mexico; the Department of Philosophy of the University of Cape Town, South Africa; the Buenos Aires Political Philosophy Group, Argentina; the Cordoba Legal Theory Group, Argentina; the 5th Congress of the Latin-American Association for Analytic Philosophy, Colombia; and the Immigration Seminar of the Panamerican University, Mexico.

² Our presentation of conditions (1)–(8) is based on Patten (2014): 232-238.:

³ The first type of view is often called ‘nationalist’ or ‘ascriptivist’, and some authors do not take it to be an instance of plebiscitary theory (see Buchanan (2017)). However, despite the difference already mentioned between the two types of views, our main reason for including them in the same classification is that both accept a non–remedial right to secession and conditions that mostly overlap with 6-10 stated above. In any case, the validity of our arguments in favour of the moral and legal right to secede does not depend on the acceptance of this classification.

⁴ Note that remedial-right and plebiscitary theories broadly agree that failing to meet one condition among (1)-(4) is sufficient for a group to acquire a right to secede. Philpott (1995): 355, for instance, asserts that ‘[t]hreats and grievances are indeed morally relevant and enhance a claim to self-determination, but neither they nor any special territorial arguments are needed to establish one.’ Similarly, according to Wellman (2005): 7, ‘groups victimized by their states have a right to secede, but it is wrong to think that this remedial claim exhausts our rights to political self-determination.’

⁵ Stilz (2016): 118-119, in particular, is more cautions on what might follow from her ideas on the importance of self–determination with regards to secession.

⁶ It is worth mentioning that Wellman (2005): 57-58 seems to defend an argument in the lines of (P1)-(P3) – despite the fact that, as we have mentioned, he later provides a different rationale for (P3) (see Altman and Wellman 2009).

⁷ Remember that PT also defends that satisfying one of the conditions (1)-(4) is sufficient for having a right to secession. However, since we take that to be common ground in the contemporary debate, establishing C is enough for justifying PT.

⁸ A different but related worry (raised by a reviewer) concerns the limits of the issues that can be decided by the majority rule. For instance, a majority is not entitled to disregard fundamental rights and reinstate slavery. Of course, we agree on the general point, but we also think that none of the decisions we are suggesting violate any fundamental right. It is not an individual fundamental right, for instance, to be able to choose the state one belongs to (note that endorsing this right would imply it is massively violated). Indeed, we think this is a standard assumption in this debate.

⁹ Note that as a matter of fact, referenda are typically held in the region that wants to secede, rather than the whole state (e.g. Quebec, 1995, Dietrich 2014, 2014). Although these were not fully unilateral processes (so they fail to exactly exemplify the scenarios we are mostly interested in this paper), they show that in general the seceding community is regarded as a legitimate group to decide over the territory.

¹⁰ For further responses to the particularity objection see Waldron (1993) and Stilz (2013).

¹¹ Interestingly, it is not obvious RRT can make this distinction. On RRT, a group has a right to secede only if the host state treats its members unjustly (by violating human rights, unjustly taking its territory, etc.) and these cases seem to always coincide with those situations in which a group has strong normative reasons to secede.