Employing both statistical methods and case studies, Matt Qvortrup, in this comprehensive study of some 200 ethno-national referendums since the French Revolution, sets out to test the hypothesis that such referendums are held in line with the logic of a competition proximity model, according to which an actor “facing considerable competition” pursuing a popular policy is likely to call a referendum because it “is likely to give the initiator a boost and strengthen his or her legitimacy” (p. 6). The statistical data and case studies presented, Qvortrup asserts, are conclusive to such an extent that “this almost amounts to a universal law” (p. 146).

Importantly, the effort to explain why referendums are held is complemented with an investigation of how they contribute to conflict management. The underlying premise of Qvortrup’s study in this respect is that “referendums can facilitate peaceful transitions from interethnic strife on a territory” (p. 60). In light of much of the literature on the subject, this is a bold claim, but Qvortrup, with significant qualifications, succeeds in driving this particular point home. As such, the book makes a very useful contribution to the vast literature on ethnic conflict management.

Referendums and Ethnic Conflict offers eight chapters framed by an introduction and conclusion. Most of them speak directly in one way or another to one or both of the issues that concern the author. Others, however, are rather marginal to the investigation. For example, the chapter on referendums in constitutional law, focusing on their permissibility and concentrating on the case of Scotland, is questionable in its relevance and has certainly been overtaken by events on the ground, with the Scottish Independence Referendum Act passed by the UK Parliament in November 2013 and given Royal Assent a month later. This is not to say that a discussion of the status of referendums in national law is not an academically interesting and policy-relevant issue (as has become clear in the context of the contested referenda in Crimea and parts of eastern Ukraine), but little is done in the chapter to relate this issue to either of Qvortrup’s propositions. The same goes for the very comprehensive and compelling study in Chapter 8 on the regulation of ethno-national referenda. While these should be required reading for any national or international policymaker, Qvortrup could have done more to relate them at least to his claim that referenda can contribute to peaceful conflict management. In other words, is there any relationship between properly conducted referenda and their impact on violence? A final point in this regard is that, by the author’s own admission, referendums on European Union integration and membership do not fit any of his four categories of ethno-national referenda. Again, the chapter offers a lot of interesting insights and tests a number of other hypotheses and different models, but it does not really address the two core propositions that he set out to investigate.

The remainder of the chapters engage with each of the four categories of ethno-national referenda that Qvortrup conceptualizes. This is done in a systematic way of combining rigorous statistical analysis with case study and anecdotal evidence to uncover the causal mechanisms that can substantiate the statistical correlations. That said, I found the categorization of referendum types somewhat problematic. First, there appears some terminological, and possibly conceptual confusion. Using the typology developed by John McGarry and Brendan O’Leary that distinguishes between conflict-regulation approaches that either manage or eliminate differences, Qvortrup sets up a dichotomy of homogenizing and heterogenizing referenda, the former “initiated by politicians who take diversity as an accepted fact and want to manage these differences” and the latter “initiated by politicians who do not accept diversity” (p. 11). Arguably, this is somewhat at odds at least with the spirit of McGarry and O’Leary’s typology, but even more so it contradicts Qvortrup’s own matrix (Table 1, p. 11) in which the difference-managing example of the 2011 referendum in Wales is presented as a case of a heterogenizing referendum.

Second, and this is where I would also see a considerable difference with McGarry and O’Leary’s conceptualization, secession referendums are considered as an example of homogenizing referenda, whereas so–called right-sizing referenda are considered as heterogenizing. Yet both types of referenda are essentially about eliminating difference—be it in terms of adjusting state boundaries more closely with ethnic boundaries (e.g., local border referenda after World War I) or by redrawing international boundaries on a grander scale (e.g., secessionist referenda).

Third, while there is clearly a difference between referenda that are about issues concerning managing or eliminating differences within states or that affect international borders, the use of the terms “national” and “international” is somewhat confusing, as the referenda are not always national or international in the sense that they involve a vote at a national level in one or more states.

Finally, difference-eliminating referenda (“defined as polls that involve explicit reference to policies of homogenization”; p. 108) at the national level include at least one unconvincing case in the way that Qvortrup conceptualizes this particular type: the 1991 referendum on maintaining the Soviet Union. The question asked in the referendum, cited by the author on page 103, is difficult to categorize as difference eliminating or homogenizing: “Do you consider it necessary to maintain the Union of Soviet Socialist Republics as a Renewed Federation of Sovereign States in which the rights of each nation would be fully guaranteed?” Yet despite this unambiguous effort to manage differences, the 1991 referendum in the Soviet Union is featured as one of two case studies on difference-eliminating referenda, alongside acclamatory referenda in the Third Reich, only one of which (the accession of Austria to Nazi Germany) was a clear-cut case of an ethno-national referendum. The fact that the case studies fit rather poorly does not invalidate the statistical findings of the chapter on difference-eliminating referenda, namely, that they are predominantly held in nondemocratic societies.

Leaving aside the question of rather confusing labels, the overall idea of four different issues that are at stake in ethno-national referenda makes sense and furthers the clarity and persuasiveness of Qvortrup’s analysis. By clearly distinguishing these different kinds of referenda, the author is able to offer a fine-grained analysis beyond his two basic propositions. For example, and as noted previously, he is able to demonstrate that undemocratic societies are more likely to hold referenda on difference-eliminating issues (homogenization/assimilation), whereas democratic societies tend to hold difference-managing referenda on what Qvortrup refers to as negotiated devolution or power-sharing settlements. A further conclusion that is implicit from his data is that referenda are overwhelmingly about territorial issues—only 26 of the 218 referenda he has analyzed are about eliminating differences at the national level qua policies of, for example, assimilation.

Perhaps the most important of these refined conclusions concerns when ethno-national referenda promote peaceful conflict settlement: Referenda “lead to peaceful resolutions of conflicts only if two conditions are met: (1) there is elite consensus for the proposed solution and (2) international backing for (or no[t] outright opposition to) the referendum” (p. 147, restated in slightly different ways on pp. 64–66, 158, and 159). Even if it would appear to be somewhat at odds with the apparent confirmation of the competition proximity model, or at least leave a gap in explaining how the apparent selfish choice of actors that underpins this model can be reconciled with the centrality of compromise and conciliation required for peaceful conflict settlement, the fundamental point remains (and is concisely put in Qvortrup’s concluding sentence on p.159): “It is only when referendums are held in this spirit of compromise and conciliation that balloting can stop bullets.”