Questioned sovereignties: independence referendums and secession in a comparative perspective

Abstract

This thesis considers the use of independence referendums in state-creation. It investigates whether and how an independence referendum affects secessionist dynamics, and may increase or decrease the likelihood of secession. The analysis consists of a quantitative and qualitative mixed-methods approach, which includes the creation of a new dataset on secessionist movements and independence referendums from 1905 to 2014, and an in-depth comparative study of Quebec and Montenegro. Independence referendums, as sovereignty referendums, ask the ‘people’, symbolic holders of the sovereign authority of a state under a democratic regime, whether their sovereignty should remain represented by the current state, or by a new independent sovereign state. A distinction between unilateral and agreed independence referendums is made in order to consider when and how ‘the will of the people’ determines state-creation. It is argued that only an agreed independence referendum can ensure that the referendum itself determines the secessionist outcome. The thesis argues that independence referendums are an important legitimisation tool and their outcome’s binding effect is primarily a result of political, rather than legal, dynamics.

The reasons behind the calling and holding of an independence referendum, and why a state government would consent to it, are analysed to account for a possible problem of endogeneity whereby the presence of an independence referendum might be determined by how likely the secessionist movement is to secede in the first place, regardless of the plebiscite. An institutional arrangement that allows for internal self-determination (such as ethnic-federalism), and access to executive powers at the regional level, was found to facilitate the mobilisation of the population in favour of secession, and the ability to call and hold a plebiscite. It is not in itself sufficient however to secure a majority in favour of independence, and other societal, political and economic contextual factors need to be taken into account to explain why a population would wish to secede.

Focusing on consensual independence referendums agreed by both the government of the existing state and secessionist leaders, the thesis further explores when and how an independence referendum affects the likelihood of secession, notably existing popular support for independence. The thesis finds that the design and campaign in particular are important elements in deciding whose support counts, how much of it is needed, and how it can be mobilised. Nonetheless, the referendum process, who participates in it and to what extent existing preferences on the question of independence can be altered are context-dependent.

Finally, the perceived legitimacy of the referendum process and outcome are argued to be essential for the voting result to be binding on actors operating at the sub-state level, state level and international level and the independence referendum to successfully answer the question of secession. The review of independence referendums since the turn of the 20th century and the thesis findings suggest that independence referendums not only create important precedents within the state they are held, but also have international consequences. As democratic norms compete with the principle of state integrity and becomes entangled with a nationalist narrative, independence referendums may have an increasingly important role to play in state-creation.