Constituent power in political liberalism: Constraining the future?

I Ferrara’s ‘most reasonable’ account of constituent power

In contrast to classical accounts of constituent power, such as the conceptions hinted at by John Locke or developed in more detail by Emmanuel Sieyès, Ferrara claims that the exercise of constituent power, for Rawls and other political liberals, is ‘not unbound’ and takes place, in Frank Michelman’s phrase, ‘always under law’ (2). ¹ It is important not to mistake this for a constraint that is wholly external to constituent power and its manifestations, as exemplified by Ferrara’s reading of Locke and Sieyès, who intended to hold constituent power answerable to a set of natural rights or to natural law. For Locke, Ferrara argues, a people constituting or re-constituting their government after a betrayal of trust, could not but tie the ensuing institutional structure to the natural rights to life, liberty, and private property. For Sieyès, we might add, the composition of the constituent power that determines the members of the constituent assembly (the ‘nation’) is not open to change, since it is given in advance by natural law. But Sieyès demands two things for constituent power that political liberalism will deny it: first, that any constitution resulting from a correctly configured constituent assembly will be not only legally valid, but legitimate and authoritative, and second, that the people that continues to exist under that constitution is unbound by it in terms of its future political development. For political liberalism, in contrast, constituent power can only be manifested ‘under higher law’, which means that it is bound by what Ferrara, following Rawls, calls the ‘most reasonable political conception of justice’ for a given people – a process originating from a historical starting point, but oriented toward a regulative idea. In this process, the people are to bring together heterogeneous factual and counterfactual elements, such as their deeply ingrained ideals within a democratic form of life, their historical experiences that helped shaping those ideals, including political exchanges constrained by the norm of public reason, but also philosophical thought experiments such as the original position that can help focus and test them and make them more precise. The challenge is to bring all of those heterogeneous elements into reflective equilibrium. Compared to a Lockean conception of natural rights, it is clear that the elements to be accomodated in constitution-making in political liberalism are not wholly ‘external’: They are partly internal and reconstructive, partly constructive on the basis of internally fixed starting points, and partly abstract and general, such as the original position.

The key innovation of the book is clarifying that for political liberalism, the ‘most reasonable’ conception of justice for us is ‘the normativity constituent power is under’ (130). Ferrara insists that the conception of justice relevant for fixing constraints on constituent power is not just one among a set of ‘not unreasonable’ options, such that any doctrine that would sufficiently overlap in content with universalizable principles of individual rights and collective autonomy might fit the bill. Because the justified exercise of constituent power is relative to the identity of a particular people, a politically liberal conception underscores both its singularity and its situated universality, a feature Ferrara identifies as ‘genealogical’ (131).

The advantage of such a historically grounded approach lies in two of its features. The first is its empirical adaptation to a number of not unreasonable doctrines present in a given society, with the requirement not to impute to them a single philosophical doctrine such as Kantian deontology, as the basis for providing constitutional legitimacy. The coming about of the Constitution of Italy after WW II is a good example for this, since its being ‘founded on labour’ proved accessible from a variety of ideological points of view in the Arco Costituzionale, from Christian Democracy to Eurocommunism. The second is that it can draw on ‘historical experiences and the political culture(s) available to the people on its way to being “constituted”’ (133). Again, this is obvious in the example of the 1948 Italian Constitution since it conceived of itself as set against and obliterating Fascism (Bobbio 1997; Niesen 2002). This historical footprint feeds into what Ferrara terms his ‘Liberal principle of constitutional legitimacy’. A proper act of constitution-making occurs only when ‘constituent power ... is exercised in accordance with a political conception most reasonable for its free and equal holders’ (134). This principle governing the exercise of constituent power is to be distinguished from a weaker principle of constitutional amendment, which binds the constituted amending powers tasked with incremental development, including the electorate, to ‘unamendable essentials’ (281).

This added value of historically sensitive over ahistorical exercises of constituent power is mirrored in Ferrara’s distinction between a sequential and a serial understanding of popular sovereignty, a second key innovation of the book and congenial to the historicist identification of its constraints. A serial exercise of constituent power treats each generation as free and unbound, along the lines of Thomas Jefferson’s idea of a 19-year sunset clause on constitutional self-binding, with peoples free to re-invent themselves as they go along. It mirrors Sieyès’ conviction that a people cannot abdicate ‘its right to change it as soon as its interests require it’ and ‘cannot undertake to will in the future only in a determinate manner’ (Sieyès [1789] 2003, 137). A sequential exercise, in contrast, for Ferrara is performed by a specific, complex actor, encompassing ‘all free and equal generations of a transgenerational people’ (209f.). Fixing the subject of constituent power in this way has obvious consequences for determining what it is free to think, will and implement. It is clear that in contrast to finding out about the views of a ‘living and enfranchised’ electorate, which might properly be ascertained by eliciting the will of the current citizenry, there are procedural, epistemic and interpretative difficulties involved in gauging the authoritative will of a transgenerational people.

Does this mean that the exercise both of constituent and amending powers, once a republic has been set up, should both safely be left with a constitutional court? Although Ferrara, along with Michelman, affirms the jurisdictional (130) or judgemental (134) character of the tasks at hand, and although he does commit to the claim that institutions of the judiciary seem best equipped for performing the latter (222), it would be wrong to view his conception exclusively as an apology for constitutional review as it stands. What he does claim is that in policing the limits of constitutional amendments, which should not be left to ‘the majority-based power exercised by legislative institutions’ (the constituted powers), a separate judicial institution is capable of adequately representing the ‘people’ (the constituent power, 221). The scope of judicial review is not limited to defending the rights of the currently living generation. Its proper task is ‘to represent the point of view of the transgenerational people’ (ibid.).

However, if the function of guarding higher law-making or -reviewing ‘cannot be entrusted to the holders of constituted law-making powers’ (ibid.), we need institutional separation. This seems to cause a problem at least for U.S. constitutionalism, which is often regarded as the gold standard for judicial review, and serves as the guarantor of constitutional legitimacy in Rawls. Ferrara’s argument from the isolation of powers however would seem to disqualify the U.S. Supreme Court from performing both of its functions, as a constituted quasi-legislative Court of Final Appeal and as Constitutional Court responsive exclusively to the constituent power, but he does not explicitly draw this conclusion. He is careful to specify that his argument does not rule out replacing the institution of constitutional review with the installation of a permanent constituent assembly, ² but both options seem to stand in need of a fuller discussion of the alternatives, especially the need for civic participation, full inclusion of stakeholders’ positions and other indicators of the legitimate exercise of constituent or amending power. The question is not so much whether and which type of courts can play a role within complex constituent systems in existing democracies, consisting of many actors, some institutional, some not, but whether they can claim to be its exclusive, comprehensive or final arbiter.

Before moving on to discuss the historically indexed, sequential exercise of constituent power in more detail, let me observe that Sovereignty across Generations is more concerned with aligning the present with the past than with aligning the present with the possible futures into which a dynamic constitutional practice may take it. The implications Ferrara’s account has for all future modifications of a given democratic constitution are clear, albeit mainly in terms of restricting the senses of freedom and equality under conditions of path-dependency. Future generations will be answerable to guarding and developing the same central principles past and present constituent citizens will have adhered to, and the present generation is not to block or re-route this entitlement and responsibility. This feature of the present generation’s exercise of constituent power Ferrara terms ‘vertical reciprocity’. It is to ensure that current citizens do not curtail future citizens’ rights or political autonomy, of course exercised under the ‘most reasonable’ constraint. This is clear enough, yet does not ensure, as Ferrara appears to admit, the material preconditions needed for the successful continuation of the constituent process into the age of overlapping and non-overlapping future generations (275, but see Karnein 2022). It also does not thematize which genealogical commitments and restrictions, on top of the principles of rights and autonomy, should be considered as permanently binding and distinctive of constituent identity, as I will spell out in a moment, although he does provide some telling examples.

With Rawls, Ferrara insists that if the constituent subject of the U.S. constitution should ever decide to do away with the First Amendment’s establishment clause and install a state religion, they will be departing from their authentic joint project. At the same time, he asserts with breathtaking confidence that if they should ever decide to abolish the right to bear arms, according to political liberalism, this will amount to a valid move on the basis of their joint democratic history. ³ If they go on to re-establish Roe vs Wade, or, on the contrary, do away with the legality of abortion rights altogether, who knows? On the basis of Ferrara’s account, we seem able to make sense of the idea of unconstitutional constitutional change, both as a matter of describing the empirical ratchet effects of existing constitutional histories and as a matter of normative theory that sees democratic self-government as a learning process going forward. We see the several exercises of constituent power and the amending power as moves in a progressive direction, to be shielded from regressive developments. However, his conception appears underdetermined as to the specific duties that would empower or constrain future generations to continue the path of most reasonable self-realization, both for overlapping and for non-overlapping generations. I will address the problem of future development more systematically in the following sections.

II Whose history?

One defining feature of political liberalism, indeed part of its ‘political’ appeal, is its groundedness in a specific democratic history and political culture. This domestic political culture is not thought to be impervious to divergent interpretations and rivalling traditions of democratic ideals, but considered rich and complete enough to inform one reasonably complete and distinctive conception of justice. In Rawls’ case, it clearly reflects the arc of development of U.S. constitutional history from the Declaration of Independence through the abolition of slavery to the civil rights movement and beyond. For contemporary Italian, German, Polish or South African scholars, on the other hand, their democratic constitutions more immediately reflect a process in which monstrous predecessor regimes have been toppled, and serve as concrete negations of those particular national histories of injustice. Thus Ferrara’s claim that ‘a people’ is a generations-spanning entity is of particular relevance for those societies. In contrast to Bruce Ackerman’s more recent writings and to Andrew Arato’s work on constituent power (Ackerman 2015; Arato 2018), for Ferrara the actual empirical political history of those constitutions – whether based on elite consensus, revolutionary uprising, or even outside facilitation – does not limit or determine the evaluation of their legitimacy. His account of the Italian 1948 constitutional settlement, for example, concentrates on the ideological convergence and centripetal tendencies of the various comprehensive doctrines and their spokespersons, leaving out the influence of the Allied forces. A similar story could be told, and has been told, for the Federal Republic of Germany as well as for the German Democratic Republic, as more or less credible attempts to leave the National Socialist regime behind (Trentmann 2023). What seems important for the legitimacy and eventual future malleability of a distinctive realization of constituent power, against Ackerman, is therefore not so much the causal story of constitutional influences but the capacity of the citizenry to identify with the constitutional commitments, and with its point and purpose read against its particular history of injustice. Here Ferrara’s ‘genealogical’ approach is of immense relevance.

In Ferrara’s terms, post-WW II processes of constitution-making formed new demoi or transtemporal ‘peoples’ in each case, since demoi are individuated, as we saw, via the taking on of constitutional commitments over time. Ferrara appeals to the ‘situated normativity that such exercise of constituent power appeals to, and the non-cognitive and thus non-essentialist but primarily practical nature of the new commitment, often couched in negative terms: “Never again!”’ (158–9). Setting aside the question of whether there are likewise sufficient ‘cognitive’ grounds to repudiate the Fascist or, in both Germanies, the National Socialist regime, or whether cognitive grounds necessarily take on an ‘essentialist’ meaning in pluralistic constitution-making processes, these understandings of ‘never again’ clearly feed into what can be accepted as the ‘most reasonable conception of justice’ by the relevant peoples, and thereby into what can serve as a particularist standard for future exercises of their constituent and amending powers. Perhaps even more distinctively than in the positive evolution of standards of freedom and equality, collectives seem able under political liberalism to reject paradigmatic injustices, as, for example, stripping Jewish Italians and Germans of their citizenship, and see the rejection of those injustices as intrinsic and principled features of their situated exercise of its constituent power.

The political question raised by this is whether a transtemporally sensitive constituent power at a later date can or must reproduce this historical index, over and above its commitment to freedom and equality. A factor that plays no role in Ferrara’s account is that ‘a people’ as transtemporal actor is not just replenished by descendants of those that lived through or died by the hand of Fascism and Nazism, but also by immigrants and emigrants. It is no co-incidence that Rawls’ (2005, 135–6) specification of his liberal principle of legitimacy still relies on the fiction that entrance and exit won’t happen except by birth and death, and does not cater for the realities of immigration and emigration. Nor do Ferrara’s two principles of legitimacy, which both (along with his insistence on constitutional authenticity and ‘irrecusability’) entrench historical path-dependency. Yet a Jeffersonian or Sieyèsian conception of unbound constituent power seems much better geared to describing the moral and political situation that immigrants, having obtained citizenship, find themselves in as far as their share in constituent power is concerned. This is not because the ‘most reasonable’ conception of justice that constrains the exercise of constituent power is somehow contaminated by a pre-political ‘ethnos’ based on a common ancestry, but because immigrants can hardly be asked to share the historical-practical imprint of the original constitutional moment when the evolution of a constitution via exercise of constituent power is at issue. They can of course be expected to comply with the existing constitutional setup and its historical preconditions as citizens under law (a diagnosis challenged by Özyürek 2023), but the role they are to adopt as future participants in constituent power will be a different one.

Or is it? I see two ways for Ferrara’s quest for the political singularity of constituent power to respond to the fact of immigration. The first is to bite the bullet and impose a historicist understanding of the legitimate exercise of constituent power on immigrants, leaving them free to mobilize for future re-interpretation, perhaps along the lines of a broader account of rejecting historical injustice wherever they find it relevant. Most immigrants will be able to draw on backgrounds of past injustice in their domestic histories, and may try and add those to the singular outlook and motivation of the receiving society. Immigrants from formerly colonized countries, such as Ethiopians in Italy or Namibians in Germany, will be able to draw on their own or their families’ histories, be it on the side of the perpetrators, collaborators or victims of European colonial conquest, and thereby enlarge the spectrum of relevant injustices to be covered under a ‘never again’-project of constitutional formation and re-formation. If, for example, a proposal to add an anti-colonial provision, say against the re-establishment of pro-colonial political parties, was proposed to be added to Art. XII of the Italian constitution’s Final Dispositions, which as of now only ensures a ban on the ‘re-organization of the dissolved National Fascist Party’, it would not seem implausible to see political liberalism as augmenting what is its ‘most reasonable’ interpretation of justice for the given society. A second possibility is to let immigrants radically challenge the particularist ‘never again’-identity of the receiving country, and opt for a less historically impregnated, and less protectionist, constitutional setup, for example, in dropping the ‘militant democracy’ provisions of the German Grundgesetz as a sign of a parochial constitutional identity. This would be more in line with the idea of taking on political freedom undiminished, and necessitate compromising in the future exercise of constituent power, since the historical features of its liberal constraints cannot be shown to be binding on immigrants.

III Constituent power: Rawls or Habermas?

Another way to bring out the relevance of the historical index in Ferrara’s Rawlsian account is to compare it, if briefly, with the account of constituent power provided by Jürgen Habermas. Ferrara sets off his project against other leading contemporary liberal accounts in the broad sense, such as the discourse theory of Habermas, the political constitutionalism of Richard Bellamy, the agonism of James Tully, and the interpretivism of Ronald Dworkin. Yet his discussion of Habermas is concerned exclusively with the latter’s take on deliberative democracy (32–35), not Habermas’ more explicit treatments of constituent power. It therefore makes sense for Ferrara to see the main difference between the two accounts in their respective strategies of coping with the pluralism of comprehensive doctrines – Rawls allowing a pluralism of moral, philosophical and religious doctrines as equally respectable bases of a freestanding conception of justice, with Habermas insisting on the cognitive character of consensus on constitutional essentials that can be right or wrong, not just more or less reasonable. In focussing on this admittedly important difference, Ferrara overlooks an important parallel between the politically liberal and the Habermasian accounts of constituent power that could further illuminate his own approach: their shared commitment to constitutional non-regression. In the remaining space of my comment, I want to stress these continuities and outline a salient difference.

Various interpretations of Habermas on constituent power have been put forward (Lafont 2020; Patberg 2020; Rasmussen 2014; Zurn forthcoming), but the two clear cases where Habermas talks about how he envisages the exercise of constituent power, chap. III on the ‘system of rights’ in Between Facts and Norms and his later writings on constitution-making in the European Union, are more open-textured than Ferrara’s reconstruction makes it seem. The ‘system of rights’ is a reconstruction of how and why the exercise of constituent power is not completely unbound (though not explicitly tied to the language of justice) – because the mutual awarding of rights in the medium of law cannot but be tasked with filling certain empty placeholder categories of individual entitlements (such as liberty, protection of the laws, social and political membership). The chapter serves as a ‘rational reconstruction’ as to what went on in the constituent assemblies of the two democratic revolutions of the late 18^th century, but their blueprint, like in Ferrara’s account, needs to be filled by democrats themselves in order to build a progressive trajectory towards an ever more comprehensive interpretation of liberal and democratic commitments. Just like Rawls in his thought experiment about the establishment clause of the First Amendment, Habermas is concerned with upholding the ratchet effect of constitutional innovation (Brunkhorst 2014). Both political liberals and Habermas see the exercise of constituent power as building coherent and lasting achievements over time.

The attempt at regression-proofing of constitutionalism that Habermas shares with political liberalism we also find in his dualist conception of constituent power for Europe. While insisting on the decisive role for a pan-European constituent subject, made up of European citizens and charged with one half of the authority to develop the European constitutional order, Habermas still insists that the constitutional bulwarks of the member states must not be dissolved (Patberg 2017). Their constituents, the citizens of the member states, forming the other half of the European constituent subject, are tasked with safeguarding the levels of already ‘realized justice and freedom’ (Habermas 2012, 41). European integration would need to be blocked if it jeopardized the degree of individual liberty and democratic autonomy already achieved in the domestic orders. Ferrara’s account of a political liberalism committed to upholding the first, but not the second amendment, and to develop a historically situated project over time, is a relative of this view. The exercise of constituent power in political liberalism is directed toward increasing liberty and equality, and at the same time, by way of the ratchet effects provided by its results, toward avoiding constituent backsliding.

The difference I see between the two accounts is that, while Habermas is certainly not averse to supporting a historically embedded understanding of legitimate constitutional development, he is obviously not committed to preserving constituent singularity or authenticity. In giving national constitutional orders a functional task to guarantee their survival in supranational orders (upholding historical achievements of liberty and justice), he is unconcerned with saving particularity, while for Ferrara (and perhaps for Rawls?) the dissolution of the historically indexed nation state order would have to frustrate their normative projects. Can such orders be replaced by a larger, more encompassing, yet distinct ones, if the larger commitments involve doing away with the singularity of the ancien order?

In concluding, the difference between Habermas and political liberalism in Ferrara’s sense can perhaps be highlighted by the fact that his impressive and comprehensive account does not touch on the question of constituent power beyond nation state borders. While his book exhaustively covers the recent trend to see constituent power at work in declarations of independence and processes of secession from nation states (163–170), he does not ask what political liberalism could do for cases of fusion. Admittedly, Rawls’s own take on the European Union was maximally defensive (Rawls & Van Parijs, 2003), given that his conception of ‘a people’ was shaped by additional communitarian concerns borrowed from John Stuart Mill’s liberal nationalism, but Ferrara is, as far as I can see, not committed to this view. It still needs to be asked whether there an elective affinity between Ferrara’s authenticity approach to constitutionalism and the fact that his theory accounts for domestic constituent powers exclusively.