HILJ Symposium: A Commentary on Ozan Varol’s “The Democratic Coup d’Etat”

[Brad Roth is Professor of Political Science & Law at Wayne State University.]

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

Ozan Varol’s article, “The Democratic Coup d’Etat,” performs a crucial service in reorienting assessments of extra-constitutional changes in government so as to emphasize substance over form. He refutes the commonplace idea – most recently championed by Richard Albert – that coups are inherently and inevitably undemocratic and illegitimate, “Democratic Revolutions,” forthcoming Denver U. L. Rev. 89:2 (2012), at 20, and demonstrates that under some conditions, seizures of power by military elites may lay the groundwork for the establishment of liberal-democratic participatory processes. He does so without any naïveté about coup-makers’ agendas, fully acknowledging the distortions that even “democratic” putschists introduce into post-coup constitution-making processes in order to entrench prerogatives for the military and/or its favored constituencies. But as he notes, the coup leaders may actually fail at engineering such reserves of power – especially when they attempt it directly and overtly – because, as in the Portuguese case (and, one might hope, in the current Egyptian case), they set in motion democratic dynamics that they cannot contain.

Varol’s account, however, replaces one exaltation of form over substance with another, reducing democracy itself to a narrow set of institutions and procedures that a coup may or may not work to promote. Such ascription is hardly unique to Varol – empirically-oriented political scientists tend to favor reducing democracy to elements that the tools of social science research can operationalize – but it neglects both the normatively loaded nature of the term and the extent to which competing conceptions of democratic ends animate political conflicts. See, e.g., Samuel Huntington, The Third Wave: Democratization in the Late Twentieth Century (Norman, Okla.: Univ. of Oklahoma Press, 1991), 5-13. Relatedly, Varol refers repeatedly to “the regime,” “the military,” and (most problematically) “the people” as unitary actors, whereas competing players frequently act in the name of these entities. (Instructive on the divisions within these groups is a book that Varol himself cites: Giuseppe Di Palma, To Craft Democracies: An Essay on Democratic Transitions (Berkeley: Univ. of Calif. Press, 1990), at 44-75.)

Beyond its illumination of important cases, Varol’s article excels in systematically countering Albert’s sweeping dismissal of “the problematic idea of a democratic coup.” Albert elaborates the dismissive position as follows:

I find this term particularly troubling because, by definition, a coup cannot be democratic. A coup is an undemocratic action that is devoid of popular legitimacy. It is an affront to the democratic ideals of stability, consent, and legitimacy. Indeed, a coup is more than simply a revolution without popular support, as one scholar argues. It is an arrogation of power by unlawful means. That alone makes it illegitimate in a liberal democratic sense. But the absence of popular support only exacerbates the illegitimacy of a coup by introducing the element of democratic will and consent into our equation, which a coup not only lacks but indeed subverts.

Albert’s characterizations are definitive only to the extent that they are thoroughly tautological. As Varol points out, a coup against an authoritarian regime can open the way for the establishment of liberal-democratic institutions, sometimes even where this was not what the coup-makers had in mind. There is no question that Albert is right to be skeptical about coups, since the decision to seize power results from non-participatory – often secret and conspiratorial – processes, and at least at the outset, inevitably concentrates governmental decision-making authority within a narrow and unrepresentative clique. But seizing power and governing are two different activities; coups that fail to draw support from some broader social base typically have poor long-term prospects.

The real question is whether the social forces backing the coup favor or disfavor the establishment of genuine participatory processes. Of course, when coups are launched against liberal-democratic constitutional governments, the social forces supporting the coup tend to have identified democratic participation as the problem, and so seek either to abolish elections and elected institutions or to constrain them in some way. But where a coup is launched against an authoritarian regime, the coup’s social base may be pro-democratic, anti-democratic, or conflicted (including over the question of what counts as genuine democracy). Moreover, divisions within that social base may be reflected in divisions within the coup leadership itself. The struggle to define the country’s new direction may continue for some time, with the outcome very much in doubt.

Varol’s example of the April 1974 coup in Portugal exemplifies this, perhaps to a greater extent than his account of it observes. See, e.g., Kenneth Maxwell, The Making of Portuguese Democracy (Cambridge University Press, 1995); Ben Pimlott, “Socialism in Portugal: Was It a Revolution?” 12 Government & Opposition 332 (1977). “The military,” like the society itself, was deeply divided. Whereas one military faction, including the initial coup leader, General Antonio Spínola, sought to block a broader political transformation, other factions – quite uncommonly – sought to effectuate one or another vision of radical social transformation in advance of any electoral mandate. Maxwell, supra, at 86-90. In March 1975, in response to the former’s failed takeover attempt, the latter seized the initiative, implementing nationalizations of banking, insurance, and other economic sectors and imposing a strong governmental role for the Armed Forces Movement (MFA) Assembly alongside any civilian leadership to emerge from the April 1975 Constituent Assembly elections. Maxwell, supra, at 110-13. The radical military leadership did not, however, dare to call off elections on which the coup’s popular legitimacy had been predicated. These elections yielded a strong majority for moderate parties, undermining the legitimacy of any further “revolution from above.” Months of tense confrontation followed until November 1975, when moderates within the military, in alliance with the leading civilian political parties, gained the upper hand and definitively set the country on course for liberal-democratic constitutionalism. Maxwell, supra, at 156-57.

Although Varol’s discussion helpfully sheds light on some of the ambiguities and complexities that coups can present, the account adheres to convention in a way that tends to obscure other ambiguities and complexities. Coups, Varol affirms, are democratic when they overthrow authoritarian governments and lead to the establishment of democratic ones; dictatorial means can be employed for democratic ends. But, following Samuel Huntington, he fixes democratic ends themselves in narrow, procedural terms:

The use of Huntington’s dichotomous, procedural approach suits the framework in this Article, which draws a distinction between coups that result in free and fair elections and those that do not. Because this Article employs Huntington’s definition of democracy, it does not use the term “democracy” to make a normative judgment on the quality of the democracy that emerges out of a democratic coup.

The term “democracy,” however, cannot so easily be stripped of its normative baggage. In adopting the Schumpeterian procedural definition of democracy – the only definition that can provide “the analytical precision and empirical referents that make the concept a useful one” – Huntington directly eschewed democracy’s classical associations with the source of governmental authority (“the will of the people”) and with the legitimating purpose of government (“the common good”), on the ground that such “fuzzy norms do not yield useful analysis.” Huntington, supra, at 6-9.

Be that as it may, these “fuzzy norms” animate real-world political struggle. It is all well and good to affirm, as Huntington did, that democracy (defined in Schumpeterian terms) is “one public virtue, not the only one,” Huntington, supra, at 10, but that is not how the term is actually used. Even putting aside efforts to establish international legal norms of democratic governance and pro-democratic intervention, the term commonly demarcates the moral high ground in political contestation; democracy is commonly understood as that for which we should be willing to fight, for no reason other than its association with the source and purpose of legitimate governmental authority. It is precisely such “fuzzy” ideas that generate the criteria for characterizing as democratic both the standard set of procedural norms and the refinements thereto perceived to be needed in particular circumstances. Democratic procedures are valued because – and only insofar as – they seem calculated to reflect genuine popular will and to effectuate governmental accountability to the polity’s presumed essential interests and principles. Where observance of procedural norms is perceived to frustrate democracy’s underlying purposes, such observance tends to be suspended; as the cliché goes, democracy “is not a suicide pact.”

The problem, of course, is that the contestants in civil conflicts disagree about these “fuzzy” ideas. Established constitutional arrangements, ordinarily counted on to supply a broadly acknowledged framework for the legitimate exercise of power, may cease to fulfill that function.

The ensuing crises of legitimacy frequently result in invocations of democracy both to support and to oppose dictatorial measures. Some examples include: the Algerian coup to pre-empt the electoral victory of illiberal Islamists in 1992; Boris Yeltsin’s unconstitutional dispersal of the Russian legislature in 1993; the Dayton High Commissioner’s removal of elected hard-line ethno-nationalist officials in Bosnia beginning in the late 1990s; the second removal of Haitian President Jean-Bertrand Aristide in 2004; the reaction to the Hamas electoral victory in the Occupied Palestinian Territories in 2006; the legislatively and judicially backed Honduran coup d’état in 2009.

Democracy, in the legitimacy-conferring sense of the term, necessarily contains substantive as well as procedural components. Insofar as democracy implies political equality, sectors of society cannot be expected to acknowledge as democratic those processes that jeopardize their fundamental interests – whether or not their perceptions of their entitlements in this regard are objectively valid or reasonable – since they understandably perceive this eventuality as tantamount to a zero-percent share of governmental authority qua citizens and as a denial of equal status qua subjects.. An entrenchment of protections for those fundamental interests is, quite sensibly, a condition of their acceptance of the new order’s democratic legitimacy. But depending on the circumstances, different sectors’ minimum acceptable conditions may collide, triggering severe political conflict.

Having opened up a fascinating and fraught area of inquiry, Varol draws back from the full range of complexities. He disclaims any normative judgment about post-coup “constitutional entrenchments” of a military role in governance, promising an explicit examination of that question in future projects. The promised future inquiry offers matchless opportunities to explore the truly hard questions about political conflict, dictatorial methods, and democratic legitimacy.