When is secession justified?

By valuing a group positively and seeking self-determination for it, nationalists often set out to redraw maps, to create new countries or to reinstate old ones. It is rare for this to occur without (often violent) conflict. Can political theorists offer guides to dealing peacefully with such disputes?

One question which political theorists have focused on has been that of secession. Secession as an issue carries with it most of the dilemmas associated with nations and nationalism, and whether theorists can say anything useful in terms of rights and wrongs. In cases of dispute, how might one decide which communities should be self-governing?

Surprisingly few political theorists have paid sustained attention to this problem. One exception has been Frederick Whelan, whose search for a satisfactory guiding principle ended in a pessimism that is evident when he writes that: ‘it appears that our only choices are to abide by the arbitrary verdicts of history or war, or to appeal on an ad hoc basis to other principles, none of which commands general respect’ (Whelan, 1983, p. 16). Nevertheless, vigorous debate continues. Let’s map some of the approaches theorists have taken.

Consider a country we can call ‘Y’, which consists of three different groups: the As, the Bs and the Cs. The most numerous group are the As, making up 60 per cent of the population. The Bs make up 30 per cent and the Cs 10 per cent. We could look at country Y and ask: which communities should be self-determining here?

The first response might be that it simply does not matter as long as country Y is democratically governed. Separating out or combining together different cultural communities makes no difference because if the state is democratic everyone has full rights to liberty and basic equalities anyway. This is a provocative view, one that liberals (who see people as essentially the same underneath their outward differences) often find attractive. But the fact is that people do feel identification with others, and often wish to be governed with, and by, particular others, people from ‘their’ group. As we have seen, the recasting of the world political map after the end of the Cold War forced many more theorists to address issues of nationalism and community.

A second response might be to find objective criteria to distinguish one political community from another, and apply them. But we have seen the very real difficulties in trying to construct ‘objective’ indices.

So what other approaches are there if we accept that the issue can’t be ignored, and that we need to take a subjective approach to it? A more promising third response among advocates of democracy has been to search for democratic answers to these dilemmas. We could ignore democratic mechanisms and just say ‘leave it up to the people in Y, they’ll work it out’. But we would be right to be wary of coercive means (such as ‘ethnic cleansing’) to determine which political communities should be self-governing.

Democracy is often taken to mean ‘majority rule’, or sometimes ‘majority rule, minority rights’. Often, however, writers on the subject have ignored the prior question; ‘majority of which group of people?’ We are caught in a vicious circle, it seems, where the people cannot decide who are ‘the people’ (or who constitutes ‘the nation’) until we know who the people are who can decide!

Some theorists have suggested ways out of this vicious circle. Consider country Y again. If groups A, B and C are all governed within Y as one state, and there is no significant dispute about the legitimacy of Y, then issues do not arise. But what if Bs want to secede and form their own state? What could make their secession legitimate?

The democratic theorist will answer: democratic majorities. So if a majority of people in B vote for an independent state, it should be granted. Democratic theorist Robert A. Dahl emphasises the point that the would-be new state should itself be a democracy, and most would be happy to add that criterion (Dahl, 1989). But again, there are some tough questions that need to be addressed.

Who should get to vote on secession?

The Bs (encompassing the Cs) or all the As too? After all, democracy is often said to be about people who are affected by an issue having a say on it; and As will certainly be affected if Bs secede. This is a live issue with regard to Northern Ireland’s future, for example. If a referendum were to decide if the province should join the Irish Republic, should the voters include all UK voters and all Irish voters, or just those living in the province? If, for example, there were to be a vote on the creation of an independent Palestinian state, what would be the appropriate constituency: (a) Palestinians living in the occupied territories, which might become the state, or (b) these plus Palestinians living elsewhere (the ‘diaspora’), and/or (c) those living in the occupied territories plus Israeli citizens? Clearly, the answers to these questions are politically critical.

One writer on secession, Harry Beran, has proposed that there should be a series of votes in such difficult cases. The proposed boundaries of a would-be new state could be expanded or contracted slightly from one vote to the next. The idea is to maximise the number of people who live in a political community of their choosing. For example, if some of the people of Northern Ireland wanted to vote to leave the UK and join the Irish Republic, a series of votes could be held on the issue. In each subsequent vote, the boundaries around the voting group could be expanded or (normally) contracted in order to maximise the percentage of people desiring the change. But while this ‘solution’ might maximise the number of people being able to belong where they choose, it does have its problems. One is that in principle it favours secessionists over integrationists, whereas there may be reasons not to allow the stability of existing arrangements to be upset so fundamentally. Perhaps more importantly, it may only work well where a would-be secessionist group occupies a continuous slice of territory. Where a group is interspersed among others who do not wish to change the status quo, the dangerous spectre of significant and forced population movements raises its head. The deaths as people moved east and west with the creation of Pakistan in 1949 offer a stark reminder of those potential dangers.

What size of majority vote should decide the issue?

In many types of democratic vote, a bare majority (technically, 50 per cent +1) is enough to decide outcomes. But often constitutional changes – changes which would affect the basic structures or political rules of the game – are regarded as needing ‘supermajorities’ of, say, 60 or 70 per cent. A basic change in the sovereign political unit would certainly count as a constitutional change. If the Bs get to vote, we might be concerned if only a bare majority favoured secession, especially if the voting turnout was low. Because the Cs form a minority within the B community, should we look for a majority of Cs as well? In addition, the turnout might be a special issue for such significant constitutional changes.

Does one community seceding grant a similar right to others?

Consider the position of community C. If B secedes, it takes C with it into the new state. But does C then have the same right to secede from B? Consider the case of Quebec. Quebecois separatists have come very close to achieving the bare majority needed to achieve their goal. But if they gained the right to secede from Canada, would other groups who do not see themselves as a part of a francophone entity likewise have the right to a further independence vote for themselves? What about non-francophone immigrant communities, or indigenous ‘nations’, within Quebec? If one secession, democratically sanctioned, is acceptable, then why not other, subsequent or consequent ones? Some theorists who broadly accept a democratic model of secession still worry about a ‘domino effect’, where one secession will provoke others, and we will end up with a patchwork quilt of ever-smaller political units (or countries). I return in a moment to the question of whether there are good reasons for us to be so concerned about the size of nation-states.

Do our answers depend on who the groups are?

Finally, perhaps our intuition about how to deal democratically with country Y depends on who we think the As, Bs and Cs are. Consider three possibilities:

• A is the UK, B is Scotland, C is the Shetland Islands
• A is the EU, B is the UK, C is Scotland
• A is a world government, B is the EU, C is France.

Does your intuition about the rights of communities A, B and C shift from case to case? If so, is the shift due to a reflex to favour the sovereign, self-governing status of existing nation-states? Or is it because you favour decentralisation in principle, or because you are an advocate of ‘ever closer union’ in the EU, or indeed of world government? Perhaps exploring our intuition in this way tells us something about the uses and limits of political theory. We must be careful to examine the assumptions we bring to our analyses, and be sensitive to the assumptions of the theorists we read.

What about a more restrictive ‘remedial right?’

Some theorists, such as Allen Buchanan, favour placing higher hurdles in the path of would-be secessionist movements. Rather than endorsing some rather permissive form of democratic right to national self-determination, he favours a more restrictive remedial right. Only those ‘national’ groups who can show that they suffer systematic historical injustice, or have so suffered, have a strong case for independent statehood. In one sense, this approach takes us full circle; if there is no great injustice, and if a minority ‘national’ community (Bs or Cs, for example, in country Y) is governed in a largely democratic manner, then we ought to favour the status quo.

What about alternatives to secession?

We have seen that in principle there are alternatives: cultural autonomy or a form of federalism. There are alternative ways to recognise ‘national’ identity apart from secession.

One conclusion to arise from this discussion of secession is that we are not cast adrift without any general principles or guidelines. We have also seen how the complexities of the real political world impinge upon political theories, and how those theories in turn can help us to make sense of the world. Debates among theorists about secession may highlight how worried these theorists are about nationalism. There are versions and examples of nationalism which are anything but liberal and tolerant of others (perhaps Serbian nationalism is a contemporary example). There are others where, arguably, the opposite seems to be the case (Scottish or Quebecois nationalism might be examples). Looked at through the lens of illiberal nationalism, ‘permissive’ theories of secession, like the stronger democratic theories we looked at, may raise concerns. After all, the democratic theories may end up endorsing either ethnic cleansing or systematic colonisation. The ethnic cleansing involved in efforts to create ‘Republic Serbska’ in Bosnia in the early 1990s would be an example of the former strategy. The moving of Moroccans into the western Sahara since 1974, and constantly putting off the day of an independence referendum for that region as the population changes in a more congenial direction, might be seen as an example of the second. Faced with these sorts of possibilities, we might be moved to favour ‘high hurdle’ theories of secession instead (Beiner, 2003), such as the remedial approach.

However, the issue might look different when viewed through the lens of liberal nationalism. If we take a broadly positive view of nationalist movements that are largely democratic and respectful of minorities, then the more permissive democratic approach may be more appealing. Again, there are important lessons here about the relationship between political theory and political practice. Even in cases where theories come across as abstract and general, assumptions about the real political world can and will influence our approaches.

There are no easy answers to the adequacy of secession and referendums as tools for the satisfaction of claims to national self-determination. Each case will throw up unique features; political theory cannot simply provide a universal blueprint for dealing with such specific claims. Having said that, perhaps it is the case that ‘democracy’ is not just a matter of votes, for instance in secession referendums. It has been suggested that we may be able to take all the concerns about multiple claims for self-determination – illiberal nationalism, the domino effect, political instability and so on – and incorporate them into a wider approach to ‘democratic management’ of these issues: ‘the project of democratic management must protect minorities, resist majority tyranny, correct the misuse of majority rule, and achieve a workable balance between majority rule and minority rights’ (Baogang He, 2002, p. 93).

We noted in passing that some observers worry about permissive approaches to national self-determination and secession, on the grounds that we would end up with a patchwork of too many small states. Critics are concerned about the potential destabilising effects of a secessionist free-for-all. Many larger states today are not in fact national states, but rather multinational states. Encouraging national self-determination in a strong and literal way might threaten the integrity of all but a handful of the world’s existing states:

Is it theoretically coherent to try to apply the self-determination principle to all multinational or multiethnic states? … Carried to the logical limit, the theoretical consequences are somewhat catastrophic; for hardly any states today would be immune from having their legitimacy normatively subverted.

(Beiner, 1999, p. 5)

In a similar vein, Ernest Gellner once wrote that we live in a world that ‘has only space for something of the order of 200 or 300 national states’ (quoted in Beiner, 1999, p. 5). Leaving aside the fact that a world of 300 states would be enormously different from one with the almost 200 states of today, there is a case for replying to this by pointing out that size is quite arbitrary when it comes to nation-states. This issue much exercised the great democratic theorists around the time of the American and French Revolutions. Putting it simply, the terms of the debate can be seen as being set by Jean-Jacques Rousseau, the famous Swiss-French political theorist and an inspiration for the French Revolution, who felt that liberty was threatened whenever a political unit grew beyond the size of a city-state; and James Madison, American revolutionary and the fourth president of the USA, who saw the extension of the political unit to continental proportions as a positive barrier to factional domination of a political system.

There are two basic ways in which we can understand the question of the appropriate size of political units. The first is to interpret it as a question about the appropriate extent of a unit’s geographical area. The second is to see it as a question about the size of the population of the unit. Geographical size is, arguably, less significant now than before the communications revolution. Peripheral regions of a large political unit need not be out of touch with activities at the centre. Political participation, especially in elections, is not unduly hampered by distances. The question of population size may be more important. Robert A. Dahl suggests that the ‘smaller is better’ argument looks ridiculous if pushed to extremes: ‘If it were true that a smaller system must always be more democratic than a larger, then the most democratic system would consist of one person, which would be absurd’ (Dahl, 1989, p. 205). But there is little need to jump to such extremes. A further objection to the argument that smaller is better is Dahl’s view that larger units allow for citizens to have some say in more matters. In other words, the scope of policy in larger units is greater; citizens can participate in the resolution of more issues than they could in smaller units. This may be true, but those ‘extra’ things one might be able to influence may not be matters which citizens are generally concerned about.

Further, the objects of citizen concern can be as much the product of the very existence of the larger unit. For example, the USA being a larger unit means that citizens can have some (highly indirect and minimal) say in nuclear weapons policies, clearly a matter of global importance. However, it is arguably the existence of political units of such continental dimensions which has generated the resources to devote to such weapons in the first place. Smaller units may restrict citizens’ say to smaller, more local matters, but in a world of smaller units the global questions may not loom so large anyway; these small, local matters would no longer seem, or even be, small or merely local.