Hannah Arendt disapproved of all conceptualisations of popular power in terms of sovereignty. This abolished human plurality by prioritising the need to represent the people as a unitary body having a unitary will. By contrast, politics for Arendt had to be structured around a radically different conceptualisation of popular power, and constituent power served precisely that purpose. Arendt thus aimed to rescue the democratic principle of popular authority by presenting constituent power as a radical alternative to the notion of sovereignty. The former was not a conceptualisation of popular power but its practical instantiation. It did not find its origins in the canon of Western philosophy but in the historical practice of people promising and acting together in the public space. By showcasing the plurality inherent to politics, constituent power testified that popular power does not have to disappear once the political order is created, but must be continuously exercised through the state’s institutional structure. This had to be republican, as the exercise of power would be devolved, via federal structures, to local assemblies. In addition, the state’s foundations would be constantly augmented through procedures of collective constitutional revision and amendment. Constituent power thus ends up being, once more, an alternative to sovereignty.

The modern state is built upon the principle that political power belongs to the people. Yet this principle has no uniform meaning. The very institutional structure of the modern state testifies to the plurality of understandings about the meaning, extent and implications of popular power. A quick look at modern European states reveals how each of their institutions is based upon a specific way of understanding and framing the power of the people. More strikingly, even within a single institution different conceptions of the people’s power play out simultaneously. As an example, it may suffice to think about how different the principle of popular power looks when invoked to justify the role of legislative assemblies and that of constitutional courts. The first institution is considered the forum where popular concerns and interests are elaborated, compromised upon and transformed into law by representatives.

Hannah Arendt’s theory of constituent power is radically different from Sieyès’s. Although in both cases constituent power offers a conceptualisation of popular power alternative to sovereignty, in Arendt’s case this alternative coincides withself-government. In Sieyès’s case, meanwhile, it entails the delegation of power to elected representatives. The differences and similarities between these two accounts show the interest and purpose of reconstructing the history of constituent power.

By looking at the language of constituent power and putting its uses into historical perspective, it becomes clear that, in the past two centuries, constituent power has been endowed with different, even opposite, meanings. It follows that the notion of pouvoir constituant has contributed in varying ways to the conceptualisation and institutionalisation of the principle of popular power within the framework of post-revolutionary European states.

After World War II, the language of constituent power was taken up by constitutionalists Constantino Mortati, Georges Vedel and Ernst-Wolfgang Böckenförde to advocate for popular participation in government. They accused contemporary positivist theories of sovereignty of downplaying the principle of popular power in favour of a purely legal approach to the study of legislation. This, they claimed, had the effect of undermining popular participation in politics and needed to be contrasted by adopting a different approach to the study of the state and its foundations. The language of constituent power offered precisely that, as it brought the role of the people back at the centre of constitutional theory. It indeed indicated the power of the people to establish the fundamental ‘political ideal’ at the origins of the legal political system. This power would not disappear once the constitution entered into force but would remain present alongside the ordinary working of the constitutional system. In addition, Böckenförde, Mortati and Vedel aimed to limit the scope of representation and favoured instances of direct exercise of power by the people instead. They thus promoted practices of direct or semi-direct democracy, such as referenda and constitutional revisions, decentralised and federal structures of power.

Although Emmanuel Sieyès is often read as a theorist of sovereignty, I argue that he theorised constituent power as a way of framing the principle of popular power alternative to ideas of both national and popular sovereignty. In his view, both versions of sovereignty attributed unlimited and absolute power to either the representatives in parliament (national sovereignty) or the multitude (popular sovereignty) and resulted, respectively, in legislative blockages or re-totale. Sieyès introduced his theory of constituent power to avoid both outcomes. Constituent power allowed him to claim that political authority resided in the people but was limited to the authorisation of the constitution-writing process, as carried out by elected representatives. Once the constitution entered into force, the people’s constituent power would retreat and make space for the constituted order, run by representative institutions. Yet these only had a limited power, as they could only act within the limits imposed by the people when authorising the constitution. The outcome of this theoretical construction is a constitutional representative government where the people who hold the original constituent power exercise it only indirectly (contra popular sovereignty), while the delegates who hold a derived constituted power exercise it only within limits (contra national sovereignty).

Carl Schmitt theorised constituent power as the democratic embodiment of sovereignty. Schmitt’s collapse of constituent power and sovereignty is well known, but I suggest that he did not simply take the two ideas to be interchangeable. Rather, he aimed to introduce a meaning for popular power that could be consistent with his definition of sovereignty as the power to decide on the exception. This was not provided by ideas of national and popular sovereignty. The former gave birth to liberal parliamentarianism, which he accused of dissolving the essence of sovereignty; the latter encouraged direct and local democracy, which prevented the prompt expression of the sovereign will. By contrast, Schmitt found in Sieyès’s idea of constituent power a way to associate the extraordinary character of his account of sovereignty to the democratic principle of popular power. He thus presented constituent power as the meaning of sovereignty in democratic states. On his interpretation of Sieyès’s theory, constituent power belonged to the nation but, to be exercised, needed to be represented by a unitary figure, elected through plebiscites and able to embody the unity of the nation acting as a unitary instance of decision: the sovereign dictator. The result is a complete reversal of Sieyès’s theory.

In France, throughout the post-revolutionary period, sovereignty was invoked to justify unlimited, absolute and arbitrary exercises of power by the king, parliament and the people. These uses of sovereignty, I maintain, were perceived as politically dangerous by French jurists Jean Denis Lanjuinais, Firmin Laferrière, Felix Berriat Saint Prix and Edouard Laboulaye. On three different occasions, they resorted to constituent power to tame the implications of contemporary appeals to sovereignty. First, during the Restoration it was used to claim that the king could not exercise power unlimitedly, as the constituent power belonged to the people. Second, during the July Monarchy constituent power was used to oppose the Parliament’s claim to be the sovereign and the only legitimate author of the constitution. Last, during the Second Republic, constituent power was used to claim that the power of the republican sovereign amounted to and did not extend beyond authorising the creation of the legal system. Constituent power was thus used to negotiate an understanding of popular power different from that implicit in ideas of sovereignty. While sovereignty allowed for uncontrolled and unlimited exercise of power, pouvoir constituant was used to argue that the supreme authority consisted in the popular institution of the constitutional order.