The Tenuous Connection between Popular Sovereignty and National Referendums

Richard Stacey

University of Toronto

Holding a referendum allows the drafters of a new constitution to seek popular approval for the political order that constitution would establish. Is a constitutional referendum therefore a vehicle for the exercise of popular sovereignty? Is it a mechanism of self government by which ‘the people’ can constitute their political association and set its fundamental terms? My view is that we should not conflate popular sovereignty with the constitutional referendum. While we may understand referendum as a way of expressing popular sovereignty, a referendum is neither necessary nor sufficient for a new constitution to claim the authority of popular sovereignty. In this short piece, I introduce a new understanding of popular sovereignty. I see the essence of the concept in the substantive conditions and principles that a constitution guarantees, and not in any referendum on its adoption.

This new understanding of popular sovereignty confronts a current of both theory and praxis that sees the referendum as necessary and sufficient for a new constitution to make a meaningful claim to the authority of popular sovereignty. Carl Schmitt argues in his 1928 Constitutional Theory, for example, that the total replacement of one constitution with another cannot happen without the intervention of the people, in whom the power to make a constitution is reposed. The political rhetoric during Kenya’s efforts to adopt a new constitution in the 2000s insisted similarly that a referendum was necessary because no new constitutional order can claim the authority of popular sovereignty without the affirmation of the people. In Egypt after the Arab Spring both the ruling Supreme Council of the Armed Forces and its political opposition claimed – when it suited them – that approval at referendum was sufficient to imbue the new constitutional order with the legitimacy of the people’s sovereign will. These are just two of 179 new constitutions submitted to referendum since 1793 (only eleven were defeated).

The temptation to conflate popular sovereignty and constitutional referendum is most compelling in circumstances of ‘constitutional interregnum’, where a previous constitutional order has been abrogated and is no longer in force at the moment the new constitution is adopted. Whatever markers of popular sovereignty existed under the previous order, and thus whatever authority that constitutional system may have enjoyed by virtue of its popular approval, are already gone at the moment the new constitutional order emerges. Some fresh intervention of the people is required to put the imprimatur of popular sovereignty on the new system, and the referendum, the argument goes, provides this.

What is distinctive about popular sovereignty, in my view, is the popular part. Any person or group can impose a new constitution on a people by force, conquest or charisma (the US military in Japan or Daenerys Targaryen in Westeros, say), but I take popular sovereignty to locate constitution-making power exclusively in the hands of ‘the people’. If the people is to establish a political community and authorise the rules and principles under which it will be governed, however, it follows that each person must enjoy the freedom to form an opinion about how she wishes to be governed. At the same time, she must be assured that her view will be taken into account – along with everyone else’s – in constituting this new political community. This follows from the very core content of popular sovereignty: the principle that the free-thinking persons who constitute the people must be allowed to define the political and constitutional conditions of their association. Any constitution that makes a claim to popular sovereignty therefore must be committed to ensuring the conditions under which the people can consider and decide on the terms of their political association.

Popular sovereignty thus involves something more than – and prior to – majoritarian decision-making. It is not a form of ‘constitutional positivism’ in which the constitution is anything the people say it is, because the claim to popular sovereignty brings with it these prior commitments to recognising individuals’ capacity to decide about matters of fundamental political importance, on one hand, and to guaranteeing the conditions of political equality in which people can exercise that capacity on the other hand. These commitments do not depend on the outcome of any referendum. Quite the opposite, the guarantees for the people’s political decision-making power must have been in place before any subsequent referendum could claim to have expressed the people’s sovereign political will. In turn, a constitution that denies or fails to recognise these substantive principles would have no claim to the authority of popular sovereignty even if it were subsequently affirmed at referendum.

What then should we make of cases like Kenya’s 2005 referendum, where a constitution broadly committed to the substantive principles I describe was nevertheless rejected? The possibility that the people will reject even a rights-protecting constitution, which recognises people’s political decision-making autonomy and creates conditions for its exercise, suggests that the constitution must still be put to the people if it is to claim to be a product of their sovereignty. Surely a constitution that the people would have rejected, had they been asked, can make no meaningful claim to popular sovereignty?

My response to this conundrum is to identify more specifically what a referendum brings to a constitution, and to separate the idea of a constitution’s acceptance as a sociological or empirical matter from the authority it claims from substantive commitments to individual political decision-making autonomy and political equality. Although a constitution’s formal acceptance by the people – its sociological legitimacy – depends on whether enough people approve of it or reject it, the conditions in which people can reasonably disagree, openly and freely, about the specific details of how they should be governed must be in place before the constitution’s sociological legitimacy can be tested. The very possibility of holding a popular referendum, then, presupposes guarantees of the political decision-making autonomy and political equality on the basis of which everyone can consider whether or not to approve the constitutional text. Once approved, the constitution wins sociological legitimacy; but its claim to popular sovereignty comes from conditions that make a referendum possible, and not the result of the referendum itself.

The authority that a constitution claims from popular sovereignty, then, is separate from the legitimacy it claims from a referendum. And just as Kenya’s 2004 draft constitution failed at referendum, a constitution that fails to respect principles of political autonomy and equality can nevertheless win approval in a referendum. There are arguably several illiberal or authoritarian constitutions around the world where this is the case. Such a constitution could perhaps lay claim to some degree of sociological legitimacy, but is unable to make a meaningful claim to the authority of popular sovereignty. Indeed, some constitutions may make no such claim. If you have three fire-breathing dragons with which to conquer a people, for example, you may not care what the people think. In other words, popular sovereignty is not the only source of a constitution’s authority, and a particular constitutional order need not make a claim to the authority of popular sovereignty.

But any constitution that does claim to enjoy the backing of popular sovereignty is bound to respect and guarantee the substantive conditions that make possible any formal mechanisms of expressing the popular will. If we value popular sovereignty in its own right, then we should be critical of constitutional systems that operate under the veneer of a referendum’s sociological legitimacy, yet are antithetical to the principles of popular sovereignty.

Richard Stacey is Assistant Professor of Law at the University of Toronto.