Abstract
In this short reply to the article by Mark Tushnet, “Peasants with Pitchforks, and Toilers with Twitter: Constitutional Revolutions and the Constituent Power,” I seek to explain why I do not find Tushnet’s account of constituent power and unconstitutional amendments very helpful in our understanding of both the underlying theoretical controversies and actual constitutional transformations.
1. Introduction
In his article, 1 Mark Tushnet seeks to explore the notion of “constituent power” through an analysis of the topical issue of “unconstitutional constitutional amendments.” He argues that “[w]e can understand a nation’s decision to amend a purportedly unamendable provision as an exercise of the right of revolutionary displacement of an existing government.” 2 As a background, Tushnet uses the recent events in Colombia: as a protection against the establishment of autocracy, the Colombian Constitution limited presidency to a single term. In 2005, shortly before completing his first term as president, President Uribe sought a constitutional amendment, which would allow him to continue in his office. The Constitutional Court approved the amendment. When Uribe proposed another amendment shortly before having completed the second term, however, the Constitutional Court ruled against it, holding that it would be constitutionally impermissible. Tushnet speculates what could happen if Uribe did not respect the second decision (which he actually did).
In this short reply I seek to explain why I do not find Tushnet’s account very helpful in our understanding of both the underlying theoretical controversies and actual constitutional transformations.
2. The “right to revolution” as the constitutional right to violate the constitution
For Tushnet, the authority of unamendable constitutional provisions raises the question of whether the people “at time one,” i.e. at the moment when the constitution is adopted, can bind the people at “time two”—the moment when the people realize that the constitution cannot be changed despite their present desire to do so.
For most American constitutionalists the “time one” has significance unseen in most European countries. 3 It is not only the time when their Constitution was established, but also the time when the American nation came into existence— through the adoption of the Constitution . As is well known, the European experience has been quite different. The unified Germany was established in 1870 by a treaty adopted by the princes of the German territories. The current Basic Law of the Federal Republic of Germany was in many respects imposed on the German nation by the allies in order to protect the country against the evil that had originated from within. 4 Similar phenomenon can be observed in many post-communist countries, for which the act of adopting a new constitution after 1989 could signify the final defeat of totalitarianism, but hardly marked the birth of a new nation. 5 Moreover, most European post-war constitutions have rejected unlimited popular sovereignty, which lies at the heart of US constitutionalism (or at least constitutes its arguably most influential conceptualization), and instead established regimes of “constrained democracies,” “informed by the perceived lessons of the interwar period: whereas fascists (and Stalin) had tried to create new peoples, the point now [after the war] was to constraint the existing ones. ” 6
One can object to Tushnet’s formulation of the problem as a conflict between the people at “time one” and “time two” that this is precisely the function of unamendable provisions. This can have less to do with the volition of the people than their cognition . The people can recognize, rather than decide, that there are some principles not amendable by their will. The classical argument that the right of revolution is derived from divine or natural law would then confront a similar kind of argument underpinning unamendability, because they both stem from a source that lies outside the constitution.
Building on this distinction, Alexander Somek captures the difference between American and European constitutionalism as the one concerning two kinds of constitutional legality: the former perceiving the constitution as authored by the people, and the latter as being based on the people’s recognition of human rights and their willingness to submit the compliance therewith to a review by supranational institutions. 7 This is also why the very idea of a right to violate the constitution (which is how Tushnet’s “right to revolution” should properly be called when it is invoked to justify an amendment of an unamendable constitutional provision) sounds odd in Europe. If there is a right to revolution (or civil disobedience), it can only serve the aim of protecting the constitution against the government or a regime, which would want to violate it, but not to displace constitutional provisions that happened to fall out of the people’s favor. 8 It is somewhat ironic that, in Colombia, Tushnet’s right to revolution would in fact serve as a protection of the existing president Uribe’s government, and not as a shield against usurpation of power in case he wanted to violate the constitutional guarantee against multiple terms.
However, in the world of the American “we-the-people” constitutionalism, the paradox does not seem to arise, because the people derive the right from a source that lies outside the Constitution. It is “we the people,” and not some abstract principles, that underpin constitutional legality. 9 The people’s right to revolution against the constitution must mean, however, that “the people” exists independently of the constitution, which is a very problematic contention, as we will see. For, who counts as “we the people”?
3. Constituent power without the constitution
Tushnet suggests that the concept of the constituent power is the key here. For him, there are two versions of it: “It could be ‘merely’ a concept, something developed to solve a conceptual problem in the theory of constitutional foundings, or it could be a concept referring to an actual body of people.” 10 The first is “roughly analogous to Kelsen’s grundnorm ”, whereas the second “to a Hartian rule of recognition.” Tushnet concludes that the second is inferior, as it poses several problems. 11 Most importantly, it is not clear who constitutes the “nation,” whereas an ethno-nationalist definition could justify either annexation of foreign territories, if the nation gets interpreted too extensively (the recent events in Crimea illustrate this), or exclusion of unwelcome minorities, if interpreted narrowly (think of the exclusion of the Jewish people from the citizen body in Nazi Germany). Moreover, “‘the nation’ never actually does anything,” since “[c]onstituent assemblies are representative bodies, and even referenda do not involve actions taken by everyone in the territory.” 12
But herein lies the main challenge for Tushnet: by favoring the “conceptual” over the “empirical” understanding of the concept of constituent power, he claims to be able to identify when the constituent power was in fact exercised. In his words, this would happen “when a successful constitutional transformation has occurred.” 13 But how do we know that the transformation was “successful”? Is the acceptance by legal officials sufficient? What if courts and the executive disagree? And what if people go to the streets and protest? How numerous do they need to be in order to count as “the people,” the awakened constituent power that rejected the transformation of the constitution by a minority? It is not clear to me how the measure of success can be used “conceptually” and not empirically. In my view, the most difficult question is avoided by Tushnet, rather than clarified.
Tushnet’s account of constituent power is moreover ambiguous at several levels: in his view the conceptual understanding of the idea
. . . helps solve a puzzle about constitutional amendments accomplished by means short of convening a new constituent assembly. An amendment adopted by the legislature or even a referendum modifies an exercise of the constituent power without itself exercising that power. 14
The problem is that the constitution or its amendments can never modify (or prescribe) the exercise of constituent power, since even while the constituent power is established by the constitution, it also establishes the constitution—herein the “constitutional paradox,” which is in fact left unnoticed in Tushnet’s article.
A more useful distinction than that between Kelsen and Hart (however misinterpreted) would be one between Hans Kelsen and Carl Schmitt. Theirs was the debate about constitutional foundations, rather than the foundations of a legal order. Kelsen contended that “the people in a democracy has no distinct and prior political existence.” 15 As Hans Lindahl explains, Kelsen wanted “to avoid postulating the ‘We’ as the subject of a legal order,” 16 the “We” referring to the concrete “we the people.” The reason was Kelsen’s realization that there can never be a unity of “we the people”: “The only unity that may exist is the unity of a legal order.” 17 For Schmitt, on the other hand, the constitution is posited by a concrete political subject: “The unity and order lies in the political existence of the state, not its laws, rules, or whatever normativity.” 18 Both approaches have their problems: Kelsen has no account of collective agency, so central to any constitutional theory, whereas Schmitt presupposes something that can never exist: the unity of “we the people.” 19 Neither of them can therefore “explain the first-plural stance of a ‘We’ as a unity in constituent action.” 20
In my view, Tushnet’s account remains firmly locked in this impasse: he never truly explains when it would be possible to interpret a constitutional amendment of an unamendable provision as a successful constitutional revolution or transformation. It seems that, ultimately, it is when “the amendment succeeded in changing the behavior of relevant legal actors .” 21 Who these actors are, however, is an empirical question and the answer in my view will differ from one legal system to another. This could be a springboard for further research, which would move us beyond debates that have been with us for a century now. 22 But this can also simply mean that “each generation of scholars probably has to discover basic truths for itself.” 23
