As the annexation of Crimea unfolds, all parties with some interest in the outcome - the US, the EU, Russian, and Ukrainians themselves - hold very different opinions on the legality and legitimacy of the proposed referendum on Crimea’s possible secession from the Ukraine and union with Russia. Ignoring legal considerations for a moment, I wonder if there is any possibility that any outcome could ever be considered legitimate from a normative standpoint.
Consider the following: let’s assume (the unlikely premise) that all parties are committed democrats - that is, they all have strong prima facie convictions to popular sovereignty, political equality, and the rule of law. And let’s further assume (the even more unlikely premise) that all parties could agree to some democratic decision mechanism, say a simple majority vote on a binary choice. Even given these very generous assumptions, I have strong doubts as to whether any outcome in this scenario could ever be considered legitimate.
Even under these most optimistic conditions, the outcome of any referendum will simply be a function of who gets to vote. If only Crimeans get a say, then the outcome given the pro-Russian preference of most Crimeans, will be something like a 60/40 split in favour of secession. If all Ukrainians are given a voice however, then the referendum will fail. And if both Russians and Ukrainians are granted suffrage, then the outcome will be a foregone conclusion for inclusion. Who gets to vote determines what the vote is, regardless of the democratic process used. The process what gets decided becomes arbitrary, irrelevant, and predetermined by the issue of who gets to decide.
Now legitimacy is one of these nebulous and contested concepts in political theory, and which I’m not going to provide a full account of here, but it seems reasonable to grant that arbitrary decisions can’t be legitimate when it is the process of a decision making process that imbues the outcome with legitimacy. And it’s exactly the process of a referendum - the granting of a deciding voice to those involved - that gives popular decisions like referenda or plebiscites their legitimacy.
So if the outcome of a decision mechanism can’t be arbitrary if it’s to be considered legitimate, and that process is made arbitrary by who gets to decide, then it’s clear that the matter of who gets to participate can’t be arbitrary itself. But that’s exactly what it seems to be.
We can’t solve the problem of who is included by appealing to existing states because that is the question of what the borders of the states should be that we are trying to answer. What’s more, current international borders are the result of historical accident - a function of war and statecraft - complexities of the past that don’t seem to have any normative input on current issues.
Perhaps we could offer nationality or culture as criteria of inclusion but what objective characteristics define nationality? More problematically however, state borders seem to determine nationality just as much as nationality might determine borders - just think of Soviet, Yugoslavian, American, or Australian people if you need evidence of this.
A more promising justification of inclusion might be what’s known as the All-Affected-Principle which says that anyone affected by a decision should have some say in the matter. And while that might seem fair enough, it raises more questions than it answers. Just how much do I have to be affected to get a say in the matter? Should everyone have an equal say when they aren’t affected equally? And why should others have a say in matters of self-determination and governance? If a community requires the permission of others to be ‘independent’, then they aren’t be definition independent.
Until the matter of who should be included in the referendum can be satisfactorily answered, no side can claim the process it legitimate.