HILJ Symposium: Can There Be a Democratic Coup d’Etat?

[Joel Colón-Ríos, Senior Lecturer at Victoria University of Wellington, Faculty of Law, responds to Ozan Varol, The Democratic Coup d’Etat.]

This post is part of the Harvard International Law Journal Volume 53(2) symposium. Other posts from this series can be found in the related posts below.

Ozan Varol has written an important article. In arguing that some military coups may not only have democratic features but that they may also result in the adoption of democratic constitutions, Varol invites us to reconsider two of the most persisting questions in contemporary constitutional theory. First, what makes a legal revolution (understood in Kelsenian terms, that is, as the creation of a new constitution in violation of the rules of change of a previous constitutional order [see Hans Kelsen, General Theory of Law and State 116-118 (1949)]) democratic? Second, what makes a constitution democratic? Even though mostly engaging in a discussion about legal revolutions, about military coups that result in important structural changes (i.e. the transformation of authoritarian or totalitarian regimes into democratic ones), Varol seems to be examining only the latter of these questions. In fact, he suggests that he is not interested in looking at the “process by which the coup takes place” (the legal revolution), but at the democratic character of the “resulting change” (the constitution it produces) (p. 298).

For the purposes of his article, Varol operates under Samuel Huntington’s definition of democracy: “a regime in which political leaders are selected through free and fair elections”. (p. 305). Varol does not engage in an open defence of this (low-intensity) conception of democracy, but uses it to determine whether a military coup can be categorized as democratic. Briefly put, if a military, with the support of the population, topples an authoritarian or totalitarian regime, facilitates free and fair elections within a short span of time, and transfers power to the democratically elected leaders, we have a democratic coup. (p. 300). In this short note, I would like to look at ‘democratic coups’ from a strong conception of democracy. In so doing, my aim is not to put forward a different conception of ‘democratic coups’, but to show that the questions posed above are interrelated in important ways; they are, put shortly, questions about the democratic legitimacy of a constitutional regime. If we look at these two questions together, we still might be able to talk about ‘democratic coups’, but in a much more restricted way that Varol’s approach suggests.

From the perspective of a strong conception of democracy, what determines the ‘democratic’ character of a constitution is not merely its content (i.e. whether it provides for free and fair elections and recognizes a set of rights), but the process through which the constitution emerges (and the processes through which it can be changed). Free and fair elections may be a necessary condition for democracy, a sine qua non of democracy as Varol writes (p. 299), but they are never a sufficient condition. In other words, an important part of what makes a constitution democratic is to be found in its origins, in its pedigree: a constitution can hardly be labelled as ‘democratic’ if it does not have a claim to democratic legitimacy. In this context, and as I have suggested elsewhere, see Joel Colón-Ríos, Weak Constitutionalism: Democratic Legitimacy and the Question of Constituent Power (2012), the existence of a democratic constitution depends on whether it meets the following criteria: (i) the constitution emerges from a participatory and open process (i.e. a democratic legal revolution); (ii) the constitution respects the rights and institutions that allow democracy to take place (i.e. which at a minimum would allow for ‘free and fair’ elections); and (iii) the constitution itself can be changed democratically (that is, it should allow the democratic constituent power that created it to manifest from time to time). For a more general discussion of this approach to democratic legitimacy,

Of course, a military coup, by itself, would never satisfy these criteria. Even if favoured by the majority of the population, a military coup is an act of force, which by definition does not involve the opportunities for participation and deliberation that must be present in a democratic legal revolution. However, a military coup might facilitate or negate the types of processes that would make a legal revolution democratic. This is exemplified in Varol’s discussion of the transition period that will follow most ‘democratic coups’. These “short transition periods”, Varol explains, will normally require “a number of housekeeping tasks necessary to holding free and fair elections”. (p. 305) The housekeeping tasks may include, for example, the creation by the military of “the requisite political infrastructure for political parties to organize free and fair elections of democratic leaders” or the holding of elections “for a constituent assembly to draft a new constitution before parliamentary or presidential elections, as did the Portuguese military following a coup in 1974, or itself handpick a group of persons to draft a new constitution, as did the Turkish military following a coup in 1960.” (p. 305).

Varol maintains that regardless of the form that these housekeeping tasks take, what matters is that the military does not attempt to perpetuate its time in power. This is, needless to say, entirely consistent with Varol’s objective: to determine whether the coup has produced a ‘democratic constitution’ (understood as equivalent to fair and free elections). The problem is that the form that those housekeeping tasks take matter for assessing the democratic or non-democratic character of the new constitution. It makes a world of difference if the new constitution is created by the military, by a group of handpicked persons, or by an elected constituent assembly. If one considers the pedigree of a new constitution important for its democratic legitimacy, then one must distinguish between democratic and non-democratic episodes of constitution-making (or, in the context of the legally unauthorized replacement of an authoritarian or totalitarian regime, between democratic and non-democratic legal revolutions). For example, a constitution-making process which takes place through highly participatory procedures (e.g. an elected constituent assembly whose proposals are ratified in a referendum, and which is able to operate independently of the military –unlike the 1975 Constituent Assembly in Portugal, p. 337) would be clearly superior to one that takes place under closed-doors “with little or no participation by ordinary citizens” (315).

If the entire process that leads to the adoption of a democratic system of government (that is, one in which, as in Huntington’s definition, free and fair elections take place) is not itself democratic, it is unclear why we should identify the military coup as ‘democratic’. Instead, it would be more accurate to say that, if after the sort of popular uprising described by Varol (p. 301), a military decides to ‘give’ the people a new constitution, something undemocratic has happened. That is, while initially responding to the people’s will for constitutional change, the military has failed to facilitate an exercise of the people’s constituent power, has negated the possibility of a truly democratic revolution. As a result, the new constitution, regardless of the system of government it creates, cannot be seen as a democratically legitimate constitution. It may be a constitution that provides for free and fair elections, but that has been produced through a non-democratic legal revolution. Needless to say, it is probably better to live under a ‘democratic’ constitution produced in an undemocratic way, than under an authoritarian and totalitarian regime. In that respect, the phenomenon that Varol identifies as a ‘democratic coup’ is certainly of a different nature than the traditional non-democratic coup, and this is an important part of where Varol’s contribution lies.

Nevertheless, the relative lack of attention put in the democratic or non-democratic nature of the constitution-making process makes his approach insufficient from a democratic perspective. Given the close relationship between democratic constitutions and democratic legal revolutions, it is not surprising that this insufficiency spills to other parts of Varol’s analysis. This is best shown in some aspects of his otherwise illuminating discussion on ‘constitutional entrenchment’. Varol maintains that, after a ‘democratic coup’ takes place, the military may protect its interests by placing in “the interim constitution or in the resulting democratic constitution provisions that favor the military’s institutional or policy preferences or seek to perpetuate the military’s voice in the nation’s political affairs beyond the end of the democratic transition period” (p. 317). Among the examples of constitutional entrenchment given by Varol is the establishment of counter-majoritarian institutions such as a constitutional court.

But counter-majoritarian institutions like this may be as little or as much as problematic if adopted outside the context of a coup. If, as Varol notes following Tom Ginsburg and Ran Hirschl, entrenching judicial review in the constitution might be a way to protect the policy preferences and interests of the drafters of the constitution (p. 318-319), then the democratic (or non-democratic ) character of the constitution making process is certainly important. That is, if after engaging in a participatory and inclusive process, popular majorities determine that in order to properly protect the new constitution from the day to day decisions of ordinary politicians the adoption of a constitutional court is necessary, then its ‘counter-majoritarian’ character may not be a concern, at least not from a democratic perspective. As this example shows, the democratic character of the constitution making process, of the legal revolution, must be taken into account to properly assess the democratic character of the emerging constitution and of the institutions it creates. In fact, it is precisely because (as Varol case studies show) most ‘democratic coups’ involve non-democratic constitution-making processes that constitutional entrenchment is a serious concern (and why the relevant entrenchment needs to be removed before ‘vibrant democracies’ can be established, p. 322).

The above does not mean that, if we replaced Huntington’s weak democracy with a stronger conception of democracy, there could never be a ‘democratic coup’. However, it would have to be a ‘coup’ in which the military acts as a facilitator of the exercise of the people’s constituent power and not as the constituent power. In other words, a military that, following popular demands for constitutional change, topples a non-democratic regime and instead of engaging in the making of a new constitution, calls for a democratically elected constituent assembly (that proposes a constitution which has to be ratified by the electorate before coming into effect). It could be a ‘coup’ that at some point becomes indistinguishable from a popular revolution. The democratic legitimacy of the constitutional regime that emerges from such an event, however, will not only depend on whether it was created through a democratic constitution-making process or whether the resulting constitution provides for ‘fair and free’ elections. Democratic legitimacy, as noted earlier, also demands that the constitutional regime contains an opening for future exercises of constituent power: an amendment rule that allows the people to convene future constituent assemblies. After all, it is the lack of that opening which makes popular uprisings necessary in the first place.