Integration Referendums as Expressions of Constituent Power: Ireland as a Case Study

Abstract Referendums are usually conceptualized as expressing “constituent power” in obvious settings of constitutional foundation and rupture. However, I argue that if constituent power can be understood as “active” and relational within a political order—and not merely present in political foundings—then it is arguably present in “routine” referendums that have a function neither of foundation nor rupture, but rather of integration, and particularly in the sense of identity formation or consolidation.


Introduction
Referendums are usually conceptualized as expressing constituent power in the obvious contexts of secession, independence, constitution making, and so on.In such settings, constituent power is manifested mainly in a "decisionist" sense that relates to political concepts of foundation and beginning-and conversely, such moments of foundation are associated with rupture and disintegration, as much as with the creation of new states and political identities. 1 They are also often typically understood as "sovereign" exercises in the sense of legislating fundamental laws. 2 Eoin Daly is a lecturer in the School of Law at the University of Galway, University Road, Galway, Ireland H91 TK33 (eoin.daly@universityofgalway.ie).
I am grateful to Ruth Abbey and two anonymous reviewers for their comments on an earlier draft.
In this article, I argue that a different dimension of constituent power may be understood as being present in more routine "constitutional" referendums, of the sort that are not obviously connected with territorial, sovereigntyrelated, or recognizably "constituent" questions.In contrast with "decisionist" theories that depict the temporally finite exercise of constituent power by a people, alternative theories-exemplified differently by Antonio Negri and Hermann Heller-have conceptualized constituent power in an active or "relational" sense, as an animating or even "irritant" force of politics.
With this conception of constituent power in mind, I argue, using the example of Ireland, that routine constitutional referendums may be understood as "constituent" because of their "integrating" effect-because they operate partly as manifestations and performances of social/political unity, rather than because of their role in foundation or rupture, or even in legislating."Integration" referendums can be seen as expressing the ritual aspect of constituent power-where constituent power is understood as having political unity, as much as constitutional form, as its object.
I thus highlight an underappreciated dimension of the connection between referendums and constituent power, and discuss an often overlooked political function of referendums as a mechanism of integration.By focusing on referendums, I hope to identify certain unexpected, counterintuitive dimensions of "relational" constituent power, which are ritualistic and even conservative.This article aims to make a new contribution, in this way, to our understanding of the normative and political significance of constitutional referendums.
In section 1, I outline the contrast between a "decisionist" and relational conceptualization of constituent power, and consider in broad terms how this might inform our understanding of the constitutional significance of referendums.Section 2 considers how peoplehood or political/social unity, as much as statehood itself, might be understood as the object or focal point of constituent power and how constituent power, in this sense, might assume ritualistic expressions.Section 3 argues that constituent power, in this widened sense, might be identified in ostensibly "routine" constitutional referendums-referendums aimed at amending a constitutional textbecause of their role in manifesting and performing social/political unity.In section 4 I explore the relationship between referendums and political identity in Ireland.

Constituent Power: Foundation, Irritation, Conservation
Insofar as referendums, or certain kinds of referendums, are understood as expressions of constituent power, it is worth considering what model or conceptualization of constituent power this typically reflects.Independence, secession and constitution-making referendums are typically understood as expressing constituent power, where constituent power is understood in terms of a temporally definite founding act, a temporal rupture in political order by which a legal revolution or novelty is effected. 3And this, in turn, approximates to a "decisionist" understanding of constituent power, as a temporally definite and finite act, by which, typically, a people constitutes itself, its constitution, or a state.In this standard, relatively conservative image, the people then retreats from the political stage, lying "dormant" during the normal operation of the powers it has thereby "constituted."Indeed, Andreas Kalyvas argues that constitutional "order," in this sense, is "the destiny of the constituent power" 4 -the end by which, in many accounts, the constituent power expends itself, at least for the time being -"the emphasis being on the moment of the original creation of a new order." 5n this vision, constituent power is active, temporally speaking, only "during exceptional circumstances, when a constitution is destroyed and another is not yet born"; 6 thus, "only during dramatic constitutional moments" is the "collective body" that exercises constituent power "rendered present." 7In turn, constitutional theories of this more conservative kind, perhaps associated most strongly with the Abbé Sièyes, 8 posit the "absence of the group's [sovereign] collective body in the regular course of social life." 9In Sièyes's theory, "sovereign omnipotence" is projected "outside society," allowing for "the consolidation of individual autonomy and the protection of human rights within social structure." 10And referendums, or referendums of a particular kind, become one obvious mechanism by which constituent power, thus conceptualized, is expressed-with the people's participation being temporally definite.
However, this relatively specific and limited conception of constituent power has been challenged in recent scholarship.It is potentially conservative and antidemocratic, on the one hand, in limiting popular participation, but equally, on the other hand, it carries many of the unsavory connotations of "sovereignty" as an unaccountable power vesting exclusively in a singular agent.Or, as Kalyvas puts it, it "bears the marks of an unconstrained, undivided, and limitless force that verges, once again, on violence, arbitrariness, and raw facticity." 11An alternative school of thought offers an understanding of constituent power as a "live" or active presence within the ordinary life of the state.These theories, as Lior Barshak puts it, suggest that "constituent power is permanently present within society or the state." 12This view is probably most famously associated with Negri, who located constituent power in the "multitude," as unmediated and continually present. 13Carl Schmitt is often associated with a "decisionist" theory which identifies constituent power in the moment of founding; however, Barshak interprets him as understanding constituent power as being "exercised by the group in an uninterrupted, continuous manner," as the constitution "emanates from the continually living will of the group . . .re-founded by the community in every moment." 14More generally, this corresponds to an "open" and "dynamic" view of constitutional ordering, as Martin Loughlin summarizes it-with the shape and form of the constitution being continually open to dispute as part of the ordinary course of political life. 15ithin this perspective, Robin Celikates rejects the "eventalization" of constituent power-the "ahistorical uncoupling of the event . . .from struggles and movements," in favor of a "dynamic" account, focused on "practices" rather than events, in which constituent power is "always already mediated and dispersed rather than something to be predicated [on] a unified collective agent." 16Or, as Hjalte Lokdam puts it, constituent power might be understood in terms of the practices of "constant scrutiny, questioning, and contestation of the constitutional structure by all affected by it." 17In this conception, the role of the living constituent power is not merely creative, but disruptive.Rather than being a "quasi-mythical force erupting only in extraordinary moments," Celikates conceptualizes it "as a dynamic situated within established orders, transgressing their logic and reconfiguring them from within." 18 it is, for similar reasons, no longer situated in a discrete body or authority, blurring the distinction between "constituent" and "constituted" powers, precisely because constituent power in this sense is better understood not as an agent but as a capacity that can, in principle, be wielded by different agents, including "constituted" powers. 20rom this perspective, political activities such as disobedience can represent an exercise of constituent power because they are a contestation or "politicisation" of "symbolic and institutional structures that are usually taken for granted." 21Such activities have a "constituent" quality because as Celikates says, they potentially represent "a dynamizing counterweight to the rigidifying tendencies of state institutions." 22Correspondingly, it is difficult to understand how "routine" constitutional referendums, by which constitutional texts and structures are revised, might express this "dynamic" or "active" conception of constituent power, since they are institutionally mediated and, unlike other constitutional-amendment processes, express a "constituted" legal process.However, I argue that if we understand peoplehood or political unity as the object of constituent power, this "live" or active constituent power can be seen to have a conservative as well as a disruptive force.In turn, then, constitutional referendums might be understood as ritual expressions of constituent power through their role in manifesting and performing social and political unity.

Peoplehood as the Object of Constituent Power
The idea of constituent power as an irritant makes sense when we understand its bearer as being the people, and its object as the constitutional form.However, if we understand peoplehood itself-or perhaps, political/social unity-as being the object of constituent power, the question of who (or what) bears or yields it, and how it is yielded, becomes slightly more elusive.Through this lens, it becomes possible to understand "routine" constitutional referendums, including those aimed at amending an existing constitutional text, as an expression of "constituent" power in this adjusted sense.
One of the commonly acknowledged mysteries of "democratic" constituent power is how the people, who supposedly found a political order, are themselves "constituted" or founded.This is the "demos paradox," the puzzle of how the democratically constituting agent can itself be democratically constituted. 23For some thinkers, such as (arguably) Schmitt, peoplehood is simply a presupposition of juristic thought, something assigned to the accident of 20  history."The people," he says, must be present and presupposed as a political unity, if it is to be the subject of a constitution-making "power." 24The initial "act of the constitution-making power . . .constitutes the form and type of the political unity, the existence of which is presupposed." 25or others, however, the constitution of the people itself is a focal point of statecraft, and can usefully be conceptualized as being itself a model or form of constituent power.Jean-Jacques Rousseau, most notably, discusses the temporal and epistemic paradox of a "people" constituting itself without first having a constitution-and thus, without first constituting a political unity.He assigns to the legislator or "lawgiver," a providential historical figure, the task of bequeathing the state not only an initial constitution, but also a set of symbols and rituals through which political unity, or peoplehood, can be sustained over time. 26The lawgiver not only designs a "constitution" in the narrow sense, but must also "move the hearts of men," and creates the "souls" on which appropriate legislation will take hold. 27or Rousseau, the "people" are not the bearer, but rather the object, of constituent power-a power that is yielded separately by the "lawgiver."For Joel Colón-Ríos, Rousseau's lawgiver is not a constituent power, because its "sole faculty" is that of "drafting" the constitution, whereas constituent power is exercised as a legislative (constitutional-enactment) capacity vested directly in the people. 28However, this overlooks that the object of constituent power, in this model, is not simply the constitution of the laws, but also the constitution of the people itself.It also relegates the constitution of the people itself to the realm of presupposition.
The challenge for the lawgiver is to establish a political community in the absence of existing political bonds and affects.Since political institutions do not yet exist, there is not yet any political consciousness, any sense of citizenship to utilize or appeal to.The paradox, as Rousseau puts it, is that "men would have to be prior to laws what they ought to become by means of them." 29But on the other hand, the challenge and the genius of the lawgiver is to sustain this political consciousness, over time, through the constitution it bequeaths.This is why the lawgiver, as a "constituent" power, deploys ritual and symbol.Rousseau cites Moses as exemplary because out of [a] wandering and servile horde Moses had the audacity to create a body politic, a free people; and while they were wandering in the desert without a stone on which to lay their heads, he gave them that durable set of institutions, proof against time, fortune and conquerors, which five thousand years have not been able to destroy or even to alter, and which even to-day still subsists in all its strength, although the national body has ceased to exist. 30ses "overburdened" his people "with peculiar rites and ceremonies." 31 In devising a constitution, the lawgiver's primary purpose is not to constitute a state as such, but rather, antecedently and more fundamentally, to constitute the people as a political subject or political unity that is stable across time.It is through this proposed constitution that a people is formed and through which it continues to exist and endure not merely as a multitude, but as a political unity.
Thus for Rousseau, peoplehood itself is a political construct produced by a form of statecraft that can be understood as "constituent." 32For Schmitt, by contrast, the people is somehow the bearer of constituent or constitution-making power, but this means that its existence as a "political unity"prior to the constituent act-is somehow "presupposed." 33The people "must be present and presupposed as a political unity, if it is to be the subject of a constitution-making power." 34Unlike Rousseau, he gives no account of how the people's "self-conscious" political status emerges or how the people is politically formed; he simply excludes this from the scope of constituent power.And this is not specific to Schmitt: for liberal political theorists, Sofia Näsström observes, the question of peoplehood simply "prompts a resignation to history," 35 to the realm of presupposition or the untheorizable, unamenable to justification.Thus she notes that "among contemporary theorists, it is common to interpret the constitution of the people as 29 Jean-Jacques Rousseau, Du contrat social (Paris: ENAG, 1988), 2.7, 84, my translation. 30Rousseau, Poland, chap.2, 163. 31Ibid. 32This analysis does not preclude that sovereign, as distinct from constituent power-within Rousseau's conceptual grammar-is exercised more formally by the people by their legislating of fundamental laws.For this conceptualization of sovereignty in Rousseau's thought, see Tuck, Sleeping Sovereign, chaps.3-4. 33 a historical event . . . the result of a de facto power, a brute historical force existing beyond the realm of legitimacy." 36or Rousseau, much of this "constituent" work of the lawgiver is felt long after the founding, within the ritual life of the republic.This ritual legacy effects a kind of alchemy of peoplehood, it secures a durable political/social unity across political time, such that the "constituent" power remains "live." 37The lawgiver "ought to feel himself capable . . . of changing human nature, of transforming each individual, who is by himself a complete and solitary whole . . . of substituting a partial and moral existence for the physical and independent existence nature has conferred on us all." 38In Government of Poland, he notes that the ancient lawgivers like Moses sought bonds that might attach citizens to the fatherland and to each other; and they found them in peculiar usages: in religious ceremonies which by their very nature were always national and exclusive; in games which kept citizens frequently assembled; in exercises which increased not only with their vigour and strength but also their pride and self-esteem; in spectacles which . . .touched their hearts, inflamed them with a lively spirit of emulation. 39ma, when founding Rome, succeeded in transforming men into citizens not so much by means of law, but by "sanctifying their city with those rites, frivolous and superstitious in appearance, the force and effect of which is so rarely appreciated." 40hus the constituent power has a kind of afterlife-beyond obvious political form-in the ritual and symbolic life of the republic.Beyond the grand spectacles that Rousseau advocates-the public commemorations, ceremonies, and so on-it is present in the modest rituals of ordinary republican social life, 41 in the egalitarian folk rituals that he praises, even the bacchanalian folk dancing in Geneva that he favorably compares with the artifice and pretension of the theater, 42 in rituals that foster "gentle bonds of pleasure and joy." 43  Indeed the temporal role of the lawgiver is ambiguous because while in some instances he is a founding figure, elsewhere he appears as a continuous background presence who might establish or revise new laws; thus Rousseau refers to the "continuous presence of the legislator" as one contributing factor to the "preservation of the law" (Poland, chap.7, 389).ritual life is musical, in the sense that, like the earliest human language, it communicates passions, 44 exhorts, and is performed.The constituent power addresses the people, musically, from beyond the founding.Political unity is sustained in a poetic way: the people do not participate in constituent power as legislators or even as actors, but through ritual performance, with Rousseau understanding ritual generally as a more transparent and egalitarian political activity than speech. 45n turn, this is how the imprint of constituent power, as an extraordinary beginning, is potentially experienced in the everyday and the mundane; as Filippo Del Luchesse puts it, "constituent power is the simultaneously ordinary and extraordinary force shaping the irregular rhythms of political life." 46In contrast with the dramatic grandeur of a "decisionist" constituent sovereignty exercised in solemn legislative form and that retreats to dormancy, this version of constituent power is characterized by vitality and dynamism.And this can be located within what Loughlin classifies as the "relational" concept of constituent power-distinct from a "decisionist" or "normativist" view. 47As Del Luchesse observes, this "makes it possible to grasp constituent power as a living force that goes beyond the founding moment." 48onstituent power in this sense is not summoned or invoked, but continually (re-)enacted and performed-not just in exceptional or crisis "moments," but in the routine life of the state.Indeed precisely because of this ritualized and performative understanding of constituent power, the puzzle of formal initiation powers-the question of who, for example, controls or initiates the amendment process itself 49 -becomes less significant, precisely because the constituent power is less a formal power held by specific agents, and more a capacity or process that can be participated in relationally by different agents, including "constituted" powers.
Once we understand constituent power in this sense-as being "live" through political time and as adopting ritual expressions aimed at sustaining political/social unity-it becomes easier to understand both how constituent power might be identified in "routine" constitutional-amendment referendums and how such referendums might be understood as mechanisms of "integration."Whereas Rousseau focuses on civic ceremonies and symbols as instruments of civic integration, these are potentially less important, less effective, and less compelling in heterogeneous modern mass democracies.However, I argue that referendums offer one alternative "ritual" expression of constituent power in this sense.

Constitutional-Amendment Referendums as Ritual Expressions of Constituent Power
This conceptualization of constituent power may be identified in types of referendums that are not typically understood as having a "constituent" character-particularly in "routine" constitutional referendums, of the sort that effect no obvious rupture or foundation, but are directed at the amendment of constitutional texts.Stephen Tierney draws a similar distinction between "constitutive" referendums, creating new constitutional orders, and "contained constitutional" referendums, which he describes as "operating as part of the ordinary constitutional amendment process, either on their own or in a process of joint decision with parliament to change the constitution.This type of constitutional referendum takes place within, and its process and effects are determined by, existing constitutional structures.In this sense, the contained constitutional referendum is entirely internal to, and contained by, the constitution." 50This type of referendum operates according to an already-constituted process and offers no break or rupture in legal or political continuity.However, if constituent power is understood in the adjusted, expansive sense I have described-as being directed, in ritual form, at sustaining political and social unity-then it can be identified in referendums of this sort.Colón-Ríos et al. argue that referendums of this type are "constituent" inasmuch as they permit the people to shape the "material" constitution. 51But such referendums are "constituent," in my analysis, not because they enact or legislate any aspect of constitutional content, "material" or otherwise, but because their object is peoplehood or political, and because they ritually express the terms of political unity accordingly.Thus, for example, Rousseau's stipulation that the sovereign body periodically assemble in order to ratify the constitution is ostensibly an electoral or legislative-type expression of constituent power, but is also, arguably, an example of this subtler ritualized understanding of the concept.
This may be the case even where the function of such referendums is legislative, where procedurally speaking they are aimed ostensibly at ratifying legislative text for constitutional amendment purposes.Partly because of their "direct" democratic character, referendums, no matter how mundane their subject matter, have a political value and purpose that exceeds, and is irreducible to, any legal change they might effect.As Alvina Hoffman argues, voting has a symbolic and ritual function because it "fulfils an ideological function in defining as authoritative certain ways of seeing society in its production of collective representations." 52For Steven Lukes, it "plays a significant role in perpetuating the [collective representations] which contribute to the stability of the political system." 53 This political purpose of referendums, in turn, is partly symbolic and expressive, directed in part at affirming the foundations of political and social unity.Rousseau's instruments of political unity-peculiar rites and ceremonies and so on-are aimed at fostering an intersubjective political consciousness, rather than a "blood and soil" type nationalism.And similarly, under (late) modern conditions, referendums offer a rare mechanism for civic self-expression that exceeds their ostensible legal significance. 54ousseau favored ritualism partly because it has a repetitive and performative aspect in which individual identity tends to be somewhat effaced, in contrast with Arendtian-type "action."Referendums correspond, to a degree, with this conceptualization of political ritualism because of their anonymity, and the relatively passive role that the voting citizens assume.But they also offer a mechanism of civic integration because they represent an opportunity for making collective statements or affirmations about the nature of the political bond, which exceeds any specific material they formally legislate or enact.Judicial rulings might engage "constituent" power in the sense I have described, inasmuch as they reshape the terms of political unity while even claiming to represent popular identity.Yet the absence of any form of popular participation, even indirect, means they lack this symbolic character referendums obviously have. 55n one sense, ad hoc referendums like Brexit-addressing an open-ended and abstract issue and without an accompanying legislative text for ratification-best correspond with this idea of a ritualistically "expressive" referendum, because of their obvious associations with sovereignty, identity, and so on. 56The ad hoc procedure adds to the sense of this being a "constitutional" or even "constituent" moment-although not necessarily, as 52

THE REVIEW OF POLITICS
I have argued, because it is the power of initiation that determines the "constituent" character, but rather because of the symbolic quality of the exercise that results.Conversely, the reservoir of political discretion that exists to initiate such constitutional-change referendums does not by itself generate "sovereign" or "constituent" power for the relevant political agents, without the symbolic power of the popular voice they thereby summon or invoke, constituent power thereby being exercised relationally.For largely symbolic reasons, indeed-as I have argued elsewhere-the British people's verdict acquired a political clout that far exceeded its formal constitutional-legal status. 57owever, this expressive dimension is equally identifiable in more routine constitutional-amendment referendums, those used as part of a legislative process for constitutional amendment, and whose purpose is formally speaking to ratify legislative text.Ireland, whose constitution requires mandatory referendums as part of the constitutional-amendment process, offers a good example.Ireland has held forty-two such referendums since the enactment of its constitution in 1937, with the vast bulk having taken place in recent decades. 58While it has been argued that the UK's unstructured constitutional framework for referendums allowed political elites to instrumentalize a referendum for opportunistic or "political" purposes-or simply allowed a referendum of purely "symbolic" legal effect to have taken place-these features have equally been present, albeit in a subtler way, within Ireland's more structured model. 59n the surface, referendums are often simply used in a ratificatory way in Ireland: since they are mandatory for constitutional amendments, they are often used simply to effect particular kinds of desired legal changes which would be unconstitutional if effected by parliamentary legislation.Referendums have often been held on highly mundane legal and technical issues, including the courts structure, the legal status of the adoptions authority, and the powers of parliamentary inquiries.And while judicial doctrines describe referendums as acts of popular sovereignty, some scholars have interpreted the Irish referendum mechanism as being rather a conservative device that constitutes part of a system of "checks and balances." 60It is often simply a procedural hurdle required for particular kinds of legal Bill Kissane, "Is the Irish Referendum Majoritarian?," in Direct Democracy and Minorities, ed.Wilfried Marxer (London: Springer, 2012), 152.Doyle similarly argues the Irish referendum "cannot be understood as narrowly populist or majoritarian but instead should be seen as part of a consensus-building process required for constitutional amendment."Oran Doyle, "Minority Rights and Democratic Consensus: The Irish Same-Sex Marriage Referendum," National Taiwan University Law Review 15, no. 1 (Jan.2020): 21. change to be effected-evoking Albert Dicey's advocacy of referendums as "a device both conservative and democratic": a means of "conserving desirable elements of the existing constitution." 61owever, while they are identified as a "sovereign" exercise because of their role in amending fundamental law, referendums of this sort also express popular identity in a different, less legalistic and more "symbolic" sense. 62While popular sovereignty is in Ireland ostensibly absorbed within, and exhausted by, a highly structured constitutional process that confines referendums to a definite legislative text, referendums offer a potent instrument for political expression, whose political significance exceeds the net legislative issue that is ostensibly being voted upon.Because of the symbolic import of voting itself, they cannot be understood as a procedural means to an end-as simply a way of changing the law-in any given constitutional structure.Referendums offer a source of symbolic political power that is easily instrumentalized, in particular by those who have the power to initiate them, and which is not easily confined or tamed by formal constitutional structures 63 -even though the power exercised is relational, as I have argued, and not limited to any formal agential nexus.

Referendums and Political Identity
While there are some examples of a quintessentially "populist" style of referendum use in Ireland-particularly in relation to issues of immigration and crime-referendums have in recent years assumed an important symbolic function in helping to formulate and assert a new social-liberal identity.This was particularly evident in a series of four referendums held in the years following the Great Recession, on the issues of same-sex marriage (2015), abortion (2018), blasphemy (2018), and divorce (2019), which had the cumulative effect of further secularizing the Irish constitution and removing or easing some of the restrictive social policies that had been enshrined, partly under Catholic influence, in 1937 and in subsequent amendments.Arguably, these referendums were significant not merely in the (sometimes modest) legal changes they effected, but also in their role as a mechanism of collective expression or of identity affirmation.
On the one hand, successive constitutional referendums on social issues in Ireland have simply reflected a shifting numerical and demographic balance 61 James Kirby, "A.V. Dicey and English Constitutionalism," History of European Ideas 45, no. 1 (Jan.2019): 33, 38. 62Ibid. 63Christoph Möllers refers to a "democratic . . .supplement, a surplus of democratic practice that cannot be fully included into formal constitutional procedure."Möllers, "'We Are (Afraid of) the People': Constituent Power in German Constitutionalism," in The Paradox of Constitutionalism, ed.Martin Loughlin and Neil Walker (Oxford: Oxford University Press, 2007), 87.

230
THE REVIEW OF POLITICS of power between "liberal" and "conservative" groups-and in particular, "a changing balance in attitudes and a growth in the liberal group," with the "balance of opinion along the liberal conservative cleavage [having] been transformed." 64But the results are arguably identity forming rather than merely reflective of quantitative change.As Jane Suiter puts it, landmark social referendums since 2015 "revealed a growing liberal consensus and generational change in [a] once conservative Catholic nation." 65The marriageequality referendum, in particular, was conceptualized not only in relation to changing rights, but also a changing national identity-or as Yvonne Murphy puts it, as "mark[ing] the transition of Irish society from one of the most socially conservative in Western Europe to a leader in the field of [LGBT] rights." 66The Taoiseach similarly said the referendum outcome "disclosed who we are-a generous, compassionate, bold and joyful people." 67hese "social" referendums marked and coincided with a shift in discursive self-conceptualization, particularly at an elite level.Like the marriage-equality vote, the "Repeal" referendum-which removed the draconian antiabortion eighth amendment-was understood not only as changing legal rights, but as disclosing and consolidating a new national identity.An Irish Times editorial described the result as having " turned a closed, conservative Catholic country into one of Europe's most liberal, outward-looking states." 68These referendums removed specific legal restrictions understood as oppressive or unjust, but also removed aspects of constitutional identity considered quaint or embarrassing.The blasphemy referendum in 2018 removed a constitutional reference to an offense that had never been prosecuted, and so had little concrete legal effect-but the justice minister suggested in its aftermath that "Ireland is rightly proud of our reputation as a modern, liberal society." 69We have again sent a message to the world," he suggested-"a strong message that laws against blasphemy do not reflect Irish values."70 Comedian Stephen Fry, who had reportedly been investigated by Irish police for blasphemy after calling God "stupid," used Twitter to urge Irish people in the UK to "once again go #HometoVote to help usher in the new tolerant, liberal Ireland."71 International media reported the vote as signaling Ireland's "transformation into a progressive nation," 72 with a referendum on liberalizing divorce law the following year also framed as part of the "liberal shift."73 Accordingly, these referendums were assigned a transformative significance that exceeded the net issues being voted upon, woven into a teleological and hopeful narrative.For John Dryzek et al. these votes, understood as deliberative processes preceded by conventions and citizen assemblies, "reinvigorated the political landscape after the political disasters that the global financial crisis unleashed on Ireland."74 Liberalizing referendums have also been understood as reflecting Peter Berger's thesis of "Euro-secularization," 75 given Ireland's integration in a "cross-national European culture" and the "increased alignment of Irish values with pan-European liberalism"-a process which occurred, curiously, during the ascendancy of right-wing populism elsewhere in Europe.76 Indeed this self-conceptualization is partly based on external comparison, with an Irish Times editorial claiming that "we have resisted the illiberal reaction sweeping the US, Britain and the Continent, and in that we can serve as a beacon to others."77 This narrative about identity surrounding these liberalizing referendums can, in turn, be connected with the neoliberal discourse of "brand Ireland"-the idea of the nation itself as a brand-which, in the aftermath of the Great Recession, was vaunted as "the best little country in the world to do business in," 78 with an emphasis on openness, agility, global connectedness, and indeed "maturity," in the context of the centenary celebrations of independence climaxing in 2016.79 One columnist described the marriageequality vote as giving Ireland a "new global identity" or "rebrand." 80ar from being a neutral procedural device, constitutional referendums in Ireland have often been used expressively, as mechanisms of identity formation or identity affirmation, and not merely as mechanisms for revising constitutional structure or text.And in a broader perspective, while there is an extensive literature on the use of referendums as a mechanism with which to consult the people deliberatively, 81 there has been rather less on their use to construct the people.This identity-forming function of the referendum processes was underlined by a marked shift to a deliberative or consensualist model, in the guise of the novel use of citizens' assemblies to deliberate on constitutional-amendment proposals.The citizens' assembly which presaged the Repeal referendum was heralded as facilitating decisive social change in a context of long-standing social division and polarization, 82 with voters familiar with the assembly's deliberations having been more likely to support the referendum.83 Suiter juxtaposed the consensualism of the process with traditional electoral mechanisms that "exacerbate polarisation."84 Luke Field describes this referendum as having produced a "consensus answer" in contrast with earlier abortion referendums-despite a similar numerical split.85 For many commentators, the Repeal referendum expressed the "death" of conservative Ireland, rather than merely its demographic eclipse 86 -a "revolution" or even a "collapse of the old order" 87 rather than a shifting balance of opinion.88 One headline following the marriage referendum similarly proclaimed: "another Ireland is born"; 89 the Irish Times opined that "the liberal versus conservative cleavage is increasingly meaningless."90 The well-documented emphasis on individual narrative and testimony in the marriage and abortion referendum campaigns was mirrored by a less acknowledged collective narrative about "who we are" and so forth.It has been argued that the Brexit referendum was "less an exercise in rational and informed decision-making than the staging of a binary contest which incited and crystallised opposing identities." 91But this is less remarkable or unusual than it might seem.

Conclusion: Constituent Power as Conservative?
This account of "expressive" referendum use might go some way towards clarifying the broader sense of constituent power I have outlined.Constituent power, in this sense, is directed at both the creation and sustenance of political unity, and might assume diverse forms.Mónica Brito Vieira conceptualizes constituent power as all "those recurring moments -both macro and micro, extraordinary and ordinary-in which the very normative basis of democratic life, or who constitutes this 'we,' is reintroduced as the primary site of political contestation."She argues that "selfauthorised claims" of representation in political foundings "are performed to and retroactively validated by the existing people." 92I have argued that this dimension of constituent power often assumes ritual forms, which fulfill this function of "retrospective validation" of founding claims.Constituent power in this sense is present in the ritual life of Rousseau's austere republics, but this constituent ritualism is also present, albeit more subtly, in modern democratic politics.In particular, constituent ritualism may be identified in "integration" referendums-in the ostensibly mundane or routine types of constitutional referendum I have described.As Jorg Kustermans et al. observe, political ritual in general can both "sustain existing social orders" but also "inaugurate" new ones by symbolic performance. 93I have argued that constitutional referendums can be understood as doing both of these things simultaneously-as affirming, while reconceptualizing political unity itself.
One way of understanding this apparent paradox is with reference to Yaniv Roznai and Gary Jacobsohn's thesis of "constitutional revolution"-the idea that "revolution" occurs within the continuous legal order of a constitution, through the ascendancy of new principles and ideas. 94What they describe as a "revolutionary" constitutional occurrence-"the displacement of a dominant strand or commitment by its dissenting counterweight"95 -arguably implies conservation or continuity within the notion of "revolution" itself, because the principles that are repudiated (for example, particularistic notions of identity) are not abolished but simply eclipsed (say, by secular and liberal conceptualization of political unity).Whereas Loughlin argues for a "relationalist" understanding of constituent power which expresses the right of a people to determine constitutional form, I have argued that constituent power is more meaningfully understood both as a condition and as an expression of political unity or peoplehood itself.I argue that the ostensibly creative and disruptive nature of constituent power can be conceptualized as being regenerative and thus conservative.In the Irish example, the constituent ritualism of "integration" referendums reflects this idea of continuity within change, because even in effecting a "revolution" in identity, the constituent referendums reaffirm the constituent subject concerned.If constituent power is "live" within the life of the state, it can be understood as stabilizing and conserving, largely through ritual expression, the "people" as a political subject or political unity.Referendums are an obvious mechanism in which this political process might be identified.

42
Eoin Daly, "Ritual and Symbolic Power in Rousseau's Constitutional Thought," Law Culture and the Humanities 12, no. 3 (July 2016): 620.43 See Jean-Jacques Rousseau, Letter to d'Alembert and Writings for the Theater, ed.Allan Bloom, Charles Butterworth, and Christopher Kelly (Hanover, NH: University Press of New England, 2004), 62.
Constituent power is manifested in "open-ended and continuously changing ways."19Andcrucially, just as constituent power is de-eventalized, Journal of International Political Theory 15, no. 1 (Jan.2019): 36.