Melbourne University Law Review
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
Melbourne University Law Review |
|
Ron Levy[1]*
Referendums are now common in ‘conflict societies’ — societies where widespread armed engagement recently occurred, is occurring or is liable to occur. If well designed, a referendum might improve the prospects of achieving a conflict settlement. The referendum’s relative democratic legitimacy may also help to ensure against subsequent breach, once a settlement is reached. However, in practice the utility of referendums for conflict settlement has been inconsistent. Some past referendums faltered (eg a ‘no’ vote delayed settlement) as a result of neglect of careful institutional design. In particular, a number of past referendums proceeded as simple majoritarian exercises with little in the way of support for voters’ deliberation about issues at stake. By contrast, a handful of authors have described ‘Deliberative Referendums’ purpose-designed to generate more rational and informed referendum campaigns. Nearly all past work on Deliberative Referendums has focused on peaceful societies. Building on this past work, the present article introduces the term ‘Shotgun Referendum’ to refer to a Deliberative Referendum held under conditions of ongoing or apprehended violence. The article explains why such a referendum might incrementally improve the prospects for conflict settlement. It proposes the use of deliberative design features — some novel, others well known — and places these within a distinctive frame drawing on constitutional and deliberative theory. The article thus serves as a scoping study of the aspirations and boundaries of Shotgun Referendums. This can offer more careful direction when, as seems inevitable, in future more conflict societies hold referendums.
Contents
Referendums are now common in ‘conflict societies’ — societies where widespread armed engagement recently occurred, is occurring or is liable to occur. Referendums have addressed both intrastate and international cases of conflict.[1] In either case, referendums typically aim to draw on the sovereign power of the people themselves to reopen basic legal arrangements between communities, and to prompt or legitimate settlement agreements. In one recent example, after a peace plan negotiated with FARC rebels in 2016, the Colombian government hoped — in vain, as it turned out — to gain popular endorsement of the plan. The intention was to secure the kind of durable agreement that had remained elusive in negotiations dating to the 1980s.[2]
In the ideal case, popular decision-making conducted by referendum can have broad appeal, generating common feeling across sections of populations not usually accustomed to agreeing. A referendum can potentially bypass differences of interests and worldview in order to reset a stalled process of reconciliation. In part it may do so by sidelining elite leaders who seek to inflame division and conflict; a state’s non-elite citizens may feel comparatively little commitment to sustaining a conflict and be more eager to reach an agreement. If well designed, a referendum might therefore improve the prospects of settlement. The referendum’s perceived legitimacy may even help to ensure against subsequent breach, once a settlement is reached.
Of course, many referendums fall short of such high aspirations. A referendum process might lend a settlement little more than a fig leaf of legitimacy. There may be no neutral authority available to administer the vote. Ballot questions can mislead, and those in control can exclude or intimidate classes of voters. Even a fair voting process can aggravate divisions. In conflict societies, where inter-group polarisation and mutual distrust are pronounced, a referendum might only entrench an unstable accommodation between groups, or provide new outlets for divisive social discourse. And, though a referendum amplifies the voices of the many, it does not necessarily quiet those of the demagogic few. Some referendums do more to aggravate than to resolve inter-communal friction.
This article considers when referendums might nevertheless have useful roles to play in conflict societies. The substance of peacemaking proposals is also critical.[3] Yet the focus here is on the process of conflict settlement. While referendums have already entered common use in conflict settings and have even been involved in successful peace agreements — in Northern Ireland, for example — their utility has been inconsistent. What are the rationales
for using referendums as tools for conflict settlement? What, in turn, are the prospects and limitations of the approach? Addressing these questions, at
the broadest level the article contributes to the large and complex literature
on conflict.
Within this literature, however, the article focuses on a relatively novel strand drawing on deliberative democratic theory. This theory assumes that those subject to collective decisions should have voices in the decision-making process, but also that the process should involve an exchange of reasons in which participants persuade and are persuaded based on the ‘force of the better argument’.[4] The deliberative ideal of public reason requires citizens to put ideas to each other using relatively rational, generalisable and widely understood forms of argumentation that others ‘may reasonably be expected to endorse’.[5] Democratic procedures should be inclusive of people on equal terms,[6] and should promote reflective,[7] well-informed[8] and other-regarding decision-making.[9] Deliberative democratic institutions ideally encourage participants to remain flexible about their value commitments and policy preferences.[10] This departs from a common perception that democracy mostly involves strategic actors aiming for a zero-sum ‘win’ over supposed political or social competitors.[11] Deliberative democratic theory focuses on achieving ‘overlapping consensus’ or ‘normative meta-consensus’:[12] areas of public policy where citizens can find and share common ground despite differing worldviews.
Given its concern for channelling disagreement into rational forms of persuasion, it is clear why deliberative democratic theory has forcefully entered the field of conflict research. By reworking institutions of decision-making we might incrementally improve the quality of deliberation, which in turn might improve prospects for the successful settlement of conflicts. However, few deliberativists have yet advanced arguments for referendums
in conflict societies; many authors indeed view referendums as paradigmatically anti-deliberative. A handful of deliberativists have considered how
to construct ‘Deliberative Referendums’, whose design features enhance
public deliberation.[13] But these works have seldom strayed beyond polities with relatively modest social divisions.[14] This reticence is unsurprising: designing a Deliberative Referendum is a vexed problem even in the most peaceable societies.
Yet abandoning deliberative institutional schemes because deliberation appears too hard, and a polity too divided, overlooks the possibility that some division can be traced to scarce opportunities to deliberate in the first place. When, if ever, can referendums provide such opportunities? As there have been very few genuine experiences with Deliberative Referendums in conflict societies thus far, there is a limited empirical record, and any answers must be tentative. This article thus serves as a scoping study to place the aspirations and boundaries of the technique into a useful theoretical frame. This can offer more careful direction when, as seems inevitable, in future more conflict societies hold referendums.
The article proceeds in three main parts. In Part II, I outline key impediments to deliberation in conflict societies. At this point I do not yet distinguish between deliberation in referendums versus deliberation in elite-controlled processes.
Next, I outline two rationales specifically supporting the use of referendums as compared with elite-controlled models of conflict settlement. Both rationales rest on distinctive combinations of deliberative democratic and constitutional theory. Much conflict settlement involves reasoning that is ‘constitutional’ in nature (or is analogous to such reasoning); settlement often requires fundamentally altering political powers or territory. Conceptualising conflict settlement as constitutional settlement has implications for how settlement ought to be approached.
In Part III, I focus on the first of the rationales, value-deliberation. I build here on past work by myself and others on constitutional referendum deliberation in non-conflict societies to argue that, contrary to prevailing assumptions, constitutional referendums may be amenable to deliberatively robust popular engagement, even in conflict societies. I show how non-elites
are often adept at deliberating over constitutional matters, provided that
the deliberation applies principally to constitutional values, not constitutional technicalia.
In Part IV, I describe the second rationale, popular legitimacy. Here I
depart from traditional theories that rely on abstract notions of popular legitimacy. Popular legitimacy can be understood instead as a collective judgment about the relative capabilities of elite and non-elite decision-makers in a democracy, including about decision-makers’ capacities to deliberate carefully and impartially. Here, again, the constitutional nature of settlement is relevant. Even a brief act of entrenchment, if it proceeds by referendum
and enjoys popular perceptions of legitimacy, may yield a lasting constitutional settlement.
Together these deliberative rationales describe a minimalist approach to deliberation. Rather than seek in the first instance to design deliberation to mend deep social divisions, some referendums can — far more manageably — contribute toward deliberation of limited scope and duration, albeit with potentially enduring effects. In practice, whether a referendum can improve deliberation and aid permanent settlement crucially depends on the kind of conflict and the kind of referendum. Most referendum models still neglect deliberation as a design goal and tend to falter partly for this reason. Throughout Parts III and IV, then, I also suggest what such supports might look like, and interrogate whether they can help to realise the two deliberative rationales. I introduce the term ‘Shotgun Referendum’ to refer to a Deliberative Referendum held under conditions of ongoing or apprehended violence. A principal feature of this type of referendum is the requirement that voters choose settlement options framed as generalisable principles. This innovation, which has antecedents in commercial dispute settlement, compels participants to avoid choices based purely on self-interest. The ultimate goal of a Shotgun Referendum should be to structure referendum deliberation to improve value-deliberation by voters for the duration of the referendum vote — just long enough to enact a constitutional settlement.
At least five characteristics of conflict societies can diminish the quality of both popular and elite deliberation.
A polarised discourse encourages the conceptualisation of values and policy in rivalrous terms. In multiethnic societies with histories of conflict, subcultural narratives of oppression may particularly reinforce this effect in moments of intercultural conflict.[15] Moreover, assumptions about ethnic groups often lazily deny individual motive to members of those groups, and raise prejudicial social comparisons based on distinctions both real and imagined, and both large and trivial.[16] Dryzek observes how agonistic approaches to inter-group negotiation — which eschew deliberation and instead see parties strenuously promote clashing positions — can ‘freez[e] cleavages’ into place and reinforce a ‘vicious symbolic politics ... tied to myths of victimhood and destiny’.[17] Under such conditions ‘[d]ebate leads only to the group position[] becoming more extreme, as individuals get their prejudices confirmed in talk with like-minded others’.[18] These patterns are directly contrary to the deliberative democratic ideal.
Hence it may be easier to deliberate toward common ground in developed democracies and in more culturally unitary societies where identitarian cleavages are limited. However, note that this observation does less work than might be imagined. No society is wholly unitary; all have identity distinctions. The question for deliberative democrats (and others) is therefore why not every culturally mixed society also suffers from extreme political division.[19]
Conflict society laws can seek to disadvantage, and thus invidiously target, minority (and sometimes majority) groups. This is common even in non-conflict societies, where, for example, referendums outlawing same-sex marriage have been common. However, in conflict societies such group-targeting can more readily lead to violence or more deeply aggravate a sense of physical or status insecurity among members of the targeted group. Though these effects are problematic in themselves, they are also concerning for deliberation if they mean that a popular discourse is characterised not by reasoned argument but by coercion by violence or threats of violence.
Conflict societies can intensify information deficiencies. Conflict frequently coincides with low general rates of education. This information vacuum can in turn be relatively easily filled with elite propaganda and oversimplification.[20] In the extreme, disinformation can involve the control of media outlets in order to silence critics and valorise ‘desirable’ voices.[21] Divided peoples are also frequently prevented from knowing or understanding the salience of others’ histories and values. One reason why ‘it is not possible to create “ideal speech situations”’ is that ‘deliberation takes place in languages, dialects, accents, and ethnically toned voices’.[22] For instance, the existence of separate linguistic, educational, religious and media institutions and norms in Lebanon and Cyprus meant that ‘powerful internal social barriers impeded contact between the communities’ in those countries.[23]
Not all people deliberate to the same extent or in the same way. Mixed deliberative populations — of those primed and wishing to deliberate, and also those who reject deliberative ideals in favour of agonism — are present in any democratic polity. Indeed, some people might be ‘better than others at arguing in rational terms’,[24] partly as a function of socio-economic position.[25] To call a society ‘deliberative’ is therefore always a matter of degree and also a comparative assessment relative to other societies.
Two questions that these problems raise for deliberative democracy are: (i) whether institutional design can encourage more people to deliberate; and (ii) how to accommodate, democratically, the inevitable hard core of individuals who will refuse to deliberate under all circumstances. These questions are especially relevant for conflict societies. As Dryzek notes, ‘mutual acceptance of reasonableness is exactly what is lacking in divided societies. ... [O]penness to persuasion by critical argument ... is in fact not widely held, and explicitly rejected by’ some.[26] Uneven social or governmental commitments to deliberation are in turn complicated by the fact that, while some people are predisposed to deliberate, others first meta-deliberate about whether deliberation in the present case is worthwhile. This gives rise to a prisoner’s dilemma in that meta-deliberators are prepared to deliberate, but only after it becomes evident that others in the polity are similarly willing.
Active violence magnifies each of the problems and dilemmas above. Violence inspires visceral responses that may cut short good-faith attempts to engage in deliberation.[27] For instance, even in non-conflict societies, legislative responses based on immediate reactions to terroristic carnage tend to call for restrictive and punitive responses, without necessarily taking account of the potential for backlash against these very responses and for more violence.[28] The responses often bypass careful legislative procedure and may fail to achieve reasonable balance in policymaking. National security matters ‘can elicit highly emotional reactions because of fears about safety and security, anger over perceived transgressions, and desires for vengeance’.[29] Vocal advocacy for immediate punitive responses to violence is common, even among citizens who at other times may be open to deliberating.
These impediments show the uncertain prospects of relying on deliberation to settle conflict. However, they do not as yet address the comparative question of whether deliberation can be better handled by a referendum or by an elite-controlled settlement process. Conflict theorists often conclude that representative or other elite leadership is preferable to more complex and unpredictable popular participation.[30] In democratic theory, referendums and other forms of direct democracy are typically understood as democratic ideal types, set aside in practice due to their assumedly inferior deliberative quality.[31] Yet this assumption often overlooks contrary evidence and fails to spell out the range of circumstances in which it holds.
For instance, the social division that can prevent deliberation across communities can manifest both in referendum and elite-led processes. Elites may include governmental elites (elected or appointed legislative, executive and judicial members) or social elites (media commentators and organisations; business leaders; and influential voices purporting to speak for religious, ethnic or other identity groups). In either case, works by Kahan and other social psychologists suggest that group polarisation can especially characterise decision-making by elites.[32] The most highly educated citizens are specially prone to using new knowledge to reconfirm what they already believe.[33] Whether by cherrypicking or logical sophistry, they are best equipped to choose and bend information to match pre-existing assumptions aligning with polarised positions.[34] Partisan polarisation is often a problem not of low information, but of excess mutual familiarity and long-developed animus. ‘Motivated reasoners’ may be well apprised of the other side, but are often driven to fit what they learn into existing polarised positions, rather than to reconsider matters from first principles.[35]
In addition, the risk of minority-targeting by majorities may be no greater amid referendums. Commentary on this point often neglects to examine this as a comparative matter, and one of degrees of difference. Though a referendum is a majoritarian event, this does not necessarily make it a tool of minority coercion. A referendum in which minority interests are at stake (eg a referendum to recognise Indigenous peoples in a constitution) can symbolise, at opposite extremes, either an attempt to advantage the group, or a dangerous submission of minority interests to majority will and whim. Which symbolic direction the referendum actually takes depends partly on elite leadership of the referendum campaign. Even absent a referendum, governmental leaders often engage in the same invidious rhetorics expected of referendums.[36] In non-conflict societies the comparative record on invidious lawmaking via legislatures and referendums is mixed.[37] As we will see later on, in at least some conflict societies (where the record is less complete), there is reason to suspect that the deliberative advantage favours non-elites.
Like other democratic theorists, deliberative democrats habitually defend representative lawmaking in light of the differential abilities of elites and non-elites. Many deliberativists imagine a hierarchical arrangement, in which a base tier of non-elites rarely become directly involved in lawmaking, but engage in what we may call ‘value-deliberation’. This involves deliberation about the broad and foundational interests, principles, ideologies and worldviews that inform these citizens’ policy preferences.[38] The upper tier of the hierarchy in turn comprises governmental actors: especially elected leaders, but also members of governmental branches and agencies with particularist expertise.[39] Such elites engage in what can be called ‘applied-deliberation’, relying on legal and other technical expertise to weigh and implement public values in the form of specific rules.[40]
The assumption that non-elites are unsuited to applied-deliberation is at least partly justified: few have time to develop specialised knowledge about the procedures and substantive detail of lawmaking. A deliberative hierarchy may therefore be defensible on democratic grounds. It gives a prime decision-making role to governmental elites, not in order to check the excesses of popular rule or to preserve elite power, but to help public values gain coherent expression in practice.[41] However, assumptions about relative elite and non-elite deliberative capacities should raise particular doubts when the subject is constitutional reform. In each of the next two parts, I raise a deliberative rationale favouring the use of referendums in pursuit of constitutional settlement. I then describe problems with the rationale and stipulate institutional conditions necessary, but not always sufficient, to mitigate these problems in practice. The discussion remains comparative throughout: not a claim that referendums can yield ideal deliberation, but only, more modestly, that when carefully designed they can incrementally improve upon elite-led procedures of settlement.
The first deliberative rationale rests on an argument in two parts. First, constitutional settlement focused on value-deliberation can enhance deliberative public reason and aid the prospects of constitutional settlement. Second, when voting in and deliberating about issues in a referendum, non-elites can be more adept at value-deliberation than some classes of elite deliberators. I consider the value-deliberation rationale initially in general and idealised terms (Part III(A)), before then turning to consider counterarguments (Part III(B)) and institutional preconditions (Part III(C)).
Under certain institutional conditions, sustained mutual exposure among decision-making participants can encourage at least modest flexibility in the participants’ initially contradictory preferences. Deliberative democratic public reason facilitates this process in part by remaining agnostic about, and therefore compatible with, the value commitments of diverse citizens.[42] It does not presume to determine which single policy choice or comprehensive worldview ought to dominate all others, but seeks common ground among multiple worldviews.
[D]ifferent groups, countries, religious communities, and civilizations, although holding incompatible fundamental views on theology, metaphysics, human nature, and so on, would come to an agreement on certain norms that ought to govern human behavior.[43]
Public reason assists the pursuit of common ground in part by generalising the terms of argumentation. In the public reason ideal, ‘only those arguments which can be made in public should have any force’.[44] A common argumentative language ought to be employed: participants should put arguments to each other in relatively widely understood terms. Hence reasons stated should be, as far as possible, acceptable to a broad cross-section of people in a pluralistic polity.[45] These are critical preconditions for deliberation. As Goodin observes, having to declare publicly the reasons behind your deliberations ‘works to ensure Golden Rule style outcomes [in that] you cannot expect others to buy an argument from you that you would not ... buy from them’.[46] If A is to try to persuade B based on an appeal to A’s own needs, then A should be receptive when in reply B tries to do the same based on B’s needs. This relies on what Elster calls the ‘civilizing force of hypocrisy’ — essentially a shaming force that encourages speakers to couch their positions in the language of general principles rather than self-interest.[47]
It is possible that a settlement process focused on value-deliberation can aid settlement by encouraging the process to approximate public reason ideals more closely. Deliberation focused on broad, brief and foundational values may help to generalise the terms of discussion. Values are themselves generalised propositions applying across multiple cases. Value-deliberation can thus represent a first step away from the applied positions that each participant in collective decision-making initially may favour.
Let us consider this possibility in the context of constitutional reform. Such reform is often, though not always, conducted in a value-deliberative mode. Consider the example of efforts to recognise or ban same-sex marriages around the world. In these public campaigns (only some of which involved referendums), existing constitutional principles, provisions and case law provided relevant background. However, below this superstructure of legal arcana lies a suite of broad, brief and foundational public values, such as: tradition and sexual propriety; church–state separation or integration; and equality, inclusion, social recognition and social capital. With constitutional technicalia often in the background, debate tended to focus heavily on value aspects[48] — often even in key passages of judicial rulings.[49]
Conflict settlements can follow a similar reasoning structure. As Christine Bell observes, ‘[i]n peace processes, ceasefires are rarely just about ceasefires. ... [T]hey often have constitutional dimensions and implications.’[50] Settlements thus frequently implicate not merely short-term acts (eg monetary reparations), but also second-order norms — norms that are ‘constitutional’ by virtue of ‘regulat[ing] matters that are in some sense more fundamental than others’.[51] This is especially so for chronic conflicts, which can persist in part due to unsettled clashes of value. It is also particularly a feature of conflicts with an intrastate element, in which workable long-term arrangements must be found in relation, for example, to distributing powers of revenue, cultural self-determination, military and security apparatuses, and general lawmaking. However, long-running conflicts of all types may implicate second-order norms, including territorial boundaries and the establishment or methods of enforcement of human rights.
Many norms of settlement — the rules or principles that must be agreed upon for settlement — either directly describe public values (eg human rights, cultural autonomy, security), or can be made sense of only as near-direct expressions of such values (eg military, revenue and lawmaking powers are the main incidents of group self-determination). This can be so regardless of whether the norms affect a single polity or more than one. Questions of basic value are at least implicit subjects of many settlement debates. Values raised can include, to provide just a partial list: stability and security; human rights and human dignity; anti-domination and self-determination; fairness in natural resource distribution; and secularism and religious authority. (Some cases also implicate substantive values more particular to the conflict in question, such as clashing economic ideologies.) All of these values are second-order and effectively constitutional in that, once adopted, they give general direction to a community’s decisions and deliberations about a range of narrower cases (eg whether a human right applies in a given case).
The connection between constitutional norms and value-deliberation is not absolute. Much constitutional deliberation is not value-deliberative. Whether it is depends on whether participants place foundational values front and centre in their deliberations, or focus instead on applied content (eg enforcement methods). Focusing on the latter may constitute a conscious strategy to remain mostly silent about underlying values — to ‘incompletely theorise’ a process in order, it may be hoped, to reach agreement without the need to lay bare more difficult, deeper divisions of principle and value.[52]
However, in at least some conflict situations, deliberating over underlying values is a step toward the common ground required for conflict settlement. Value-deliberation might aid conflict settlement if it orients public reason toward the resolution of clashes of value implicit in the conflict. Some conflict settlement indeed may only succeed if key values first undergo liberalisation. For instance, in Northern Ireland the peace process stumbled for decades as factions promoted at least nine different substantive settlement options, most proposing power-sharing institutions and laws.[53] Thomas Hennessey notes of this period that ‘[n]either Unionists nor Nationalists could comprehend the other’.[54] Efforts yielded little progress as leaders focused on constitutional options to distribute power post-conflict, without also focusing on more fundamental questions of value. Neither side
really accepted the legitimacy of the other. Many Unionists regarded nationalists as British whether they liked it or not. ... Many Nationalists could not accept that unionists were British with the implications this had for the concept of the Irish nation.[55]
Sections on both sides ‘were prepared to aggressively demonstrate the righteousness and moral superiority of their world-view’ through the use
of violence.[56]
Hence constitutional solutions such as consociational and confederal systems involving the Republic of Ireland and/or the UK had for decades been proposed to resolve the violence.[57] The parties made these competing solutions the immediate subjects of their debates and were unable to agree on one model over all others. To reuse my language above, these were applied solutions; but what did the solutions aim to apply? When progress finally came in the late 1990s, a key reason was the acceptance of the principle of plural claims to sovereignty. The parties’ grudging adoption of a liberal accommodationist conception of sovereignty, which better tolerates overlapping identities, geography and power, helped to impel progress toward settlement. Meanwhile, the parties deferred a number of applied matters, including the formal declaration of a ‘united Ireland’, and decommissioning paramilitary weapons.[58] Far from incomplete theorisation, this was an incomplete specification of particulars: the parties reached an agreement without clarifying crucial details.
I raise the Northern Irish example not to suggest that its lessons are universally applicable, but as a stepping-off point for our remaining discussion. The example’s limited lesson is that, in some conflicts, bypassing efforts at value liberalisation might only delay or derail settlement. Alongside other factors, focusing on value content may help to determine the robustness of deliberation in referendums. As O’Flynn observes, if settlements ‘are to provide a platform for sustainable peace, they must ... reflect a commitment to basic principles’.[59]
One further benefit of settlement campaigns focused on value-deliberation is that they may partly avoid certain recurring deliberative pathologies. Either with or without a referendum, a settlement effort sees an array of elites (in media, government and civil society) and non-elites engage in debate in the public sphere. During this public campaign, deliberating with a focus on value content can obviate some of the need for social learning about applied legal and institutional details; a discourse conducted in the public sphere is usually not up to this task. Relatedly, applied-deliberation is comparatively amenable to manipulation and error. Many campaigners, eschewing intangible debates about value, prefer to debate in seemingly more factual applied terms. Yet we saw in Part II that governmental and social leaders often disseminate misleadingly persuasive factual information. In particular, campaigners often promulgate economic predictions based on tendentious empirical assumptions. Many especially play shrewdly to popular anxieties by claiming that, after some unspecified chain of consequences, constitutional reforms will yield higher prices for fuel, healthcare, groceries, etc.[60] Even when such claims can in theory be falsified with contradictory evidence there may be limited opportunities to right the record, especially amid the din of competing arguments.
Value-deliberation is not immune to informational error. For instance, settlement campaign rhetorics may unjustly accuse opponents of holding undesirable values (eg a warlike inclination), or can present misleading logic or facts in support of values (eg purported historic or scientific facts underlying nationalist claims). As with applied-deliberation, it may be difficult to correct these assumptions. However, value-deliberation is a relatively confined deliberative exercise about, in most cases, only a handful of values. These values are more often felt than reasoned, and expectations of factual or logical objectivity do not generally attach to them (eg as with the affective value of connection to a given community). By contrast, in a discourse of applied-deliberation, factual errors can consist in an essentially infinite set of mistaken or invented propositions. Even the best-informed elites will struggle to issue corrections swiftly and persuasively, as doing so can require knowledge of specific data or complex empirical and statistical methods. A campaign focused too much on applied-deliberation can also obscure relevant value claims underneath a cloak of arcana, and commit the bulk of the campaign to the cat-and-mouse game of presenting and countering factual, logical and legal claims and misrepresentations.
In sum, though superficially a more difficult labour, a settlement process focusing on (rather than subsuming or incompletely theorising) value-deliberation in some cases may be more likely to yield a settlement. I have not yet specified, however, who should principally conduct such value-deliberation, and how.
Non-elite deliberation via referendums on constitutional settlement may be preferable to elite-led processes. The values–applied-specifics distinction, which I introduced above as the cornerstone of the defence of elite-led deliberative hierarchies of lawmaking, partially collapses amid the reform of constitutional texts. As we saw, many constitutional clauses — including constitutional norms of settlement — directly enunciate basic values to establish the values’ centrality in a legal and political system.[61] To the extent that, like other kinds of constitutional reform, constitutional settlement includes a significant value-deliberative element, this challenges a key plank of the deliberative case for elite control of lawmaking.[62]
Despite their diversity of vocations and educational achievement, non-elites may be more adept at deliberation about public values than their elite counterparts. Empirical studies, including my own, show non-elites’ considerable self-confidence in their own capacities to decide largely value-based matters (eg same-sex marriage and national independence), far more than technical matters (eg specific federal power arrangements).[63] Few if any people lack value commitments. We cannot order our lives or form even rudimentary opinions without such commitments, nor without some prior deliberation about them. Despite their political salience, many values straddle both public and private spheres, developing initially as a consequence of firsthand experience. Value tensions such as order–change, stability–risk, religious authority–secularism, and cooperation–independence are interpersonal and local in scale (eg in the family or workplace) as much as they are political. This can engender deep and widespread fluency with such values.
Elite citizens also of course possess value commitments. But a number of factors tend to compromise value-deliberation by this group. A deliberative hierarchy presupposes that governmental elites will make good-faith efforts to gauge relevant public values before reaching a decision. This reception function is necessary in order to allow a wide range of normative influences to affect decision-making, beyond merely the influences familiar to governmental insiders. However, empirical studies, again including my own, show a recurring ‘elite problem’ affecting deliberative democracy.[64] Many governmental elites do not appreciably attempt to gauge the values of non-elite citizens, and act instead based largely on their own fixed value preferences. We saw one reason in Part II: formal education or expertise often correlate to narrow normative value sets; many governmental elites call on their intellectual resources not in order to approach policy matters from first principles, but to reinforce their existing opinions and allegiances.
Often, the last people we should wish to see lead conflict settlement may
be elite figures whose careers, ideologies and identities may be tied to a struggle with the other side. The causes of inter-group polarisation may lie chiefly with political and social leaders. Conflict settlements that alter power arrangements can threaten secure power positions of political leaders or warlords. Conversely, especially at top levels of power, many elites are insulated from the physical consequences of conflict, including violence, economic collapse and forced migration.
Ordinary citizens may lack their leaders’ intensity of in-group feeling, and especially the willingness to express it through violent means. Without the same institutional and other incentives to maintain divisions and sustain a conflict, non-elites may be better able to reconsider their own positions.[65] They may be more eager to see the struggle conclude so that they may begin to lead less uncertain and anxious lives. Non-elites can of course be caught up in violent struggles. But they often are motivated to do so only after incitement by elites purporting to represent the group (eg promising to elevate or restore the group’s status through projects of independence and nation-building).[66] Those who exercise or seek public power often are the first-movers of conflict, and the main barriers to its resolution. This of course depends on the nature of the conflict in question. Some, such as the Colombian and South Sudanese internal wars, are largely struggles among armies and among military and political leaders, leaving the remaining population relatively uninvolved — except as the passive victims of violence. Yet in other conflicts group antipathies are more socially pervasive (eg Israel/Palestine).
As noted, the best evidence available as yet for the effectiveness of value-deliberation in a referendum campaign comes from non-conflict cases. Consider the example of the referendum held in Ireland on same-sex marriage. Despite deeply settled normatively based opposition to same-sex marriage in some corners, the referendum produced an outcome consistent with liberal value accommodation. Religious pre-commitments could have hardened positions on same-sex marriage, thus slackening the pursuit of common ground. Eighty four per cent of Irish citizens self-identify as Catholic,[67] and religious reasoning often is particularly incompatible with public reason due to the absence of universally accessible sources of religious belief (especially when such belief is grounded in particular texts and traditions). But with public interest high, myriad large- and small-scale conversations took place and allowed voters to become better acquainted with each other’s positions. Weak or unsupported arguments frequently collapsed when exposed to scrutiny, such as the arguments that tradition alone provided reason to discriminate, and that children of same-sex couples suffer psychological harm. Exposure and deliberation helped to wear down entrenched social norms and unsupported empirical assumptions. As well, the frequently emotive terms of debate helped to highlight the social interests at stake on all sides. Through a combination of personal storytelling and empirical evidence, public debate clarified the psychological stakes of same-sex couples seeking community recognition.
The points raised thus far remain speculative. Next I consider reasons why the value-deliberative rationale could fall short in practice. Many referendums — notably Brexit — validate the worst fears of critics. In some cases of conflict, referendums have not aided but have delayed settlement. We can never guarantee that a referendum will facilitate settlement; however, designing referendums in haste — without attention to their recognised pathologies — is all but assured to have the opposite effect. Several recurring problems contradict aspects of the value-deliberative rationale for referendums.
A referendum campaign might not prioritise discussions of public value. Campaign leaders can pursue constitutional settlement wholly or largely as a matter of applied-deliberation, for instance by debating economic consequences, or prolix, lawyerly settlement norms. They may perceive good reasons for doing this. Some might presume that discussion of tangibles is the most persuasive mode of public debate. Or, like Sunstein, they may argue that focusing on broad underlying values can spoil the opportunity for competing social factions to reach an applied agreement based on plural reasons.[68] However, especially in the conflict settlement context, applied deliberation can preclude some of the noted benefits of value-deliberation for public reason. Elite negotiators often seek to load agreements with seemingly useful clauses. In the Colombian case, the draft agreement buried broad principles — eg rural reform, the political status of former guerrillas, and transitional justice — inside 297 pages packed largely with provisions on implementation and ‘issue-by-issue’ detail.[69] While such excessive content did not prevent elite-led agreement, it did help to thwart an affirmative referendum vote by making the content inaccessible to ordinary citizens.[70]
Even a referendum largely focused on values might of course not promote public reason. As we saw, the deliberative requirements of public reason include that participants in collective decision-making should use generalised and broadly cognisable normative arguments. Yet given conflict societies’ sharp lines of division, many values may be narrow in scope. Individuals may be aware of their own views and values, but comparatively unapprised of those of others, or unconvinced of their salience. More particularly, the content of some values may not be generalisable to all nor therefore capable of being broadly shared. An example is the value of non-domination, a recognised constitutional value in some contexts,[71] but one with particular purchase to the weaker party in a conflict between parties with unequal power.
Like other democratic forms, referendums have a tendency to promote incorrect or intentionally misleading claims. We have seen that even value-deliberation is prone to such error, albeit arguably to a more modest degree.
Deliberating about some public values might not appear to resemble deliberative democracy’s exchange of reasoned arguments.[72] Though some public values are premised on instrumental arguments (eg democratic arguments for free speech), many others are taken for granted. Values such as those of human dignity, self-fulfilment, security, and affiliation to one’s identity group are principally felt, rather than derived from reasoned argument. Rational forms of deliberation can still take place based upon values that themselves have no underlying normative justification; indeed, such ‘pre-rational’ values can provide the ends for deliberation, without which deliberation may be directionless.[73] But in some cases it will be difficult to reconsider and modify these values per se.
Many people hold with special firmness to values closely linked to their self-perception and identity. As Dryzek observes, ‘[m]ost conceptions of deliberative democracy require reflection and the possibility that minds can be changed in the forum itself. This is unlikely if one’s position is tied to one’s identity.’[74] Ideally, deliberation and public reason attach persuasive power to reasons that are considered, criticised, defended and revised based largely on their merits. This requires some ‘abstraction from one’s own point of view’.[75] To be sure, most deliberative democratic theory also allows that people representing assorted identities should remain involved throughout deliberations in order to help present their distinctive points of view; hence the disembodiment of information in a deliberative forum is far from absolute. Yet deliberative democracy remains open to the formidable critique of ‘difference democrats’, who question both the feasibility and the desirability of alienating identity.[76]
Nevertheless, many conflict societies have demonstrated that taken-for-granted value assumptions can ultimately be open to change. Even in the
Middle East, according to Khuri, small-scale deliberative experiments have been able to induce Arabs and Jews to ‘recogniz[e] multiple perspectives, and [to] clarif[y] their own beliefs and identities’.[77] Indeed, in part because some value commitments can lack a basis in recognised rational forms of argument, they may be mutable in such a process. Deliberating about taken-for-granted values allows statements such as ‘because I feel this way’ as legitimate. ‘[I]dentities are partly bound up with discourse, and what has been imagined through stories and discourses can partly be reimagined through discourse as well.’[78] Sometimes minds can be changed by mutual exposure to emotive appeals that highlight the interests and identities — and the stories ‘of loss and hurt’ — of others.[79]
Most deliberativists do not assume that individuals, unaided, will seek out opportunities for mutual exposure to information nor reach a flexible accommodation. As Parkinson notes, to institutionalist deliberative democrats, ‘it is not that ordinary people cannot deliberate, but that existing liberal democratic structures do not allow them the chance to develop those deliberative capacities’.[80] The Deliberative Referendums literature outlines institutional means for improving on relatively crude referendum voting models. Deliberative Referendums provide opportunities for individuals to manifest their existing but unexpressed capacities to deliberate. In light of problems such as those we have seen, the value-deliberative rationale for referendums might be credible only when certain deliberative institutional conditions (‘Shotgun Referendum conditions’) are in place. Many are untried; their likely effectiveness remains speculative and contingent on adequate funding, publicity, and a host of other practicalities. The descriptions to follow therefore focus not on the precise details of design, but rather on the degree to which each, viewed in outline and in the best light, might aid deliberation and constitutional settlement.
The first condition is straightforward: referendum questions, along with official information disseminated during the referendum campaign, should chiefly and directly focus on public values. The general aim should be to elaborate and clarify basic values; concrete detail should be raised not as the endpoint of deliberation, but only when it is illustrative of value claims. In practical terms, referendum questions and any legal amendments they address should be cast largely in language consistent with deliberation about broad, brief and foundational subjects — at least until after the referendum vote.[81]
The Shotgun Referendum campaign and vote should liberally incorporate design features of Deliberative Referendums. Experiments with Deliberative Referendums suggest that deliberative faults such as widespread factual error and limited value reasoning are not inevitable. There is a range of institutional options for informing referendum voters and stimulating their robust deliberation. Since I have detailed these in past work,[82] I explore them here only briefly.
A key option is the now-routine ‘mini-public’ model (eg a citizens’ assembly), whose members are randomly selected but also demographically representative of the broader public. A mini-public’s small membership (eg 20–160) permits more sustained and extensive deliberation than is possible for an entire public. Mini-publics learn extensively from experts with diverse views before tackling a contentious policy problem and ultimately issuing a reform recommendation, which can be put to voters in a referendum. Mini-publics have been shown to be effective at prompting their own members to become well informed about the issues at stake and about other participants’ perspectives, even in conflict societies such as Colombia, Israel/Palestine and Northern Ireland.[83] Though studies are still few, and include outliers,[84] the balance of empirical evidence thus far suggests that even citizens in warring jurisdictions can deliberate effectively under the right institutional conditions.[85] O’Flynn and Caluwaerts explain that a focused deliberative setting can ‘foster more positive inter-group attitudes’ among participants, including ‘mutual respect and the acknowledgement of the validity of others’ claims, indicating that ordinary citizens’ views on divisive issues may be less intractably conflicting than expected’.[86]
However, an open question is whether institutional design can improve deliberation in the referendum campaign more broadly. Referendums of course involve wider and more chaotic deliberation than do mini-publics alone. Yet, in comparison with many other forms of popular debate, referendums are time-constrained and substantively limited to just a few topics. This may help to make them more amenable to deliberation-enhancing institutional design. For instance, studies in non-conflict societies importantly indicate that mini-publics can set deliberative cues (about both the subjects and the style of deliberation) which referendum campaigns may follow.[87] While this does not guarantee that voters will deliberate based upon adequate information and broad attention to public values, it improves the possibility. As we saw, the Irish same-sex marriage referendum ensured an outcome consistent with value-pluralism, despite the difficulties of deliberating across settled values. In part it did so with deliberative institutional supports, including a part-randomly chosen Constitutional Convention of 100 members — a mini-public that deliberated prior to the referendum and posed the marriage amendment.[88]
A set of additional, largely untried Deliberative Referendum methods might improve deliberation in referendum campaigns more directly. For instance, ‘preliminary instruction’ would involve voting (either online or at voting stations) that requires the voter first to engage with an interactive informational tutorial. To promote neutrality, a mini-public would design the tutorial materials. Second, ‘scaled referendums’ see voters choose from a range of options on a ballot that clearly sets out benefits and costs associated with each option. While this format has been trialled for technical, non-constitutional matters (eg decisions about public waste treatment), it is a poor fit to constitutional reform in which the ‘costs’ are largely contested matters of value. However, a third option, ‘preliminary values questioning’, would see a referendum ballot begin with value-based questions such as: ‘Rank the following values [from a wide list set by a mini-public] in order of your preference’. This has the potential to encourage value-based and purposive deliberation among voters by asking them to consider and weigh the competing objectives behind a reform proposal before ultimately casting a vote on specific constitutional options. It could also help to guide governments charged with implementing referendum outcomes by giving them a clearer understanding of voters’ value preferences.
In addition, I propose here an institutional innovation to help generalise value discourse in a settlement campaign. Referendum questions, and any official information disseminated during the referendum campaign, should formulate values in generalised terms. The ‘shotgun’ moniker derives from this condition. In the commercial law context, a ‘shotgun buy/sell clause’ (also known as a ‘Texas clause’) in an incorporation or partnership agreement specifies a self-help method for one partner to buy out another. A first partner states a price of sale, after which the second partner can either accept the offer or buy out the first partner at the same price. The shotgun clause creates a useful uncertainty. The price specified must be a fair price, because the initiating partner cannot be certain of whether she will end up as the buyer or the seller. This is an example of ‘veil of ignorance rules’,[89] which seek to operationalise the Rawlsian notion of rule-making by a set of founders oblivious to their own identities and interests.[90] Veil of ignorance rules induce participants to take positions disembodied from their own presumed interests and preferences, in favour of generalised norms potentially fair to all.
The Shotgun Referendum’s generalisation condition aims for similar effects by different means. The referendum ballot should ask voters not about their own interests — for example, what they want to win out of a settlement — but what general values, applying to all sides in a conflict, should drive the settlement. The referendum ballot therefore would present voters with options such as whether ‘community X has the right to self-determination and community Y has the right to self-determination’. Voters could only vote for or against this compound proposition as a whole, rather than selectively. Voting ‘against’ another community would thus require also voting against one’s own community. Voting machines or online voting could disallow write-in answers giving partial responses. A Shotgun Referendum would involve voters from each of the communities in conflict,[91] and each proposition would be considered passed only if a majority of each community voted in its favour — effectively a power of ‘mutual veto’.[92] Afterward, elite negotiation would take place to finalise the settlement. This would be limited to giving effect to the values endorsed in the referendum.
One of several possible complications of the Shotgun Referendum is that, as noted above, some values (eg non-domination) appear relevant to only one party to a conflict. However, even these values can perhaps be reframed and generalised. A case that Dryzek describes provides a useful example:
A harrowing story of (say) rape and murder in a Bosnian village can be told in terms of guilt of one ethnic group and violated innocence of another — fuel for revenge. But the story can also be told in terms of violation of basic principles of humanity that apply to all ethnicities, making reconciliation at least conceivable (though not easy).[93]
Similarly, a value such as non-domination can be rephrased, for instance as a more universal guarantee of group autonomy.
An additional objection is that giving voters no choice but to reason or vote in generalised terms is coercive and therefore contrary to deliberative democratic ideals. However, the compulsion involved is arguably modest. Voting in a Shotgun Referendum should be non-compulsory; the referendum should impose generalised perspectives only if voters choose to vote (either at all or on a given question). Under these conditions, the coercion involved arguably would be little more than that of other deliberative democratic institutions. Participants in these institutions habitually consent to follow a process that will direct their deliberations. Critically, most such institutions are not intended directly to affect substantive voting results. Effects on substance can be expected, but these effects result indirectly from the more proximate aim of improving the deliberative process. Even most non-deliberative procedures for democracy have indirect substantive effects. For instance, electoral voting systems vary in their ballot design choices, methods for counting results, and rules for party and candidate access to ballots — all of which shape both voting processes and voting outcomes.
Ballot questions and officially disseminated information should also be confined to the particular values in conflict. This prescription is in tension with the condition above stressing evaluative breadth. Yet, as also noted, an overbroad deliberation about values expects too much of voters and can be unworkable. The objective should instead be to institute a more limited, situational form of liberalisation of values linked to the conflict in an instant case. Though far from simple, this process is not as onerous as broadly improving social deliberation across a range of subjects.
Minimalism with respect to values partly addresses the difference critique noted above. Deliberative minimalism may mean that only a modest degree of flexibility as to identity is required. Settlements require not complete cultural disembodiment, but only as much as is required to achieve a limited form of common ground for settlement. At a minimum, deliberative value accommodation in a deeply divided society should establish a principle of security for all parties by protecting against existential challenges to groups, including violence. In most cases, another requirement will be some form of constitutional recognition of the legitimacy and status claims of all the parties. This does not require wholesale changes to individual value commitments. In the Northern Irish settlement, there was no expectation of a critical mass of individuals moving closer in identity, more politically accommodating as a general matter, and more willing to deliberate. The settlement instead involved a focused liberalisation of attitudes — a limited reconsideration of the kinds of group identities understood as legitimate in Ireland/Northern Ireland, as well as the order of priority between such identitarian debates and needs for stability and physical security.[94]
Here I depart from a number of authors. O’Flynn identifies public reason as ‘crucially important to the development of trust in a deeply divided society’.[95] Similarly, Kanra stresses the benefits for settlement when deliberative democracy involves a process of broad and unforced general social learning.[96] And Addis argues for deliberation ‘through which the identities of the participants, not just the policies and institutions of the polity, are constructed and transformed’.[97] However, while movements toward more widespread trust, learning and shared identity would no doubt be useful, often they will be unrealistic.
The minimalist institutional methods discussed to this point contemplate institutions realising existing social deliberative capacities under controlled circumstances. There is no expectation that a Shotgun Referendum will build these capacities. Nor is the objective in the first instance to eliminate division. Instead the objective is to manage division. A fundamental meeting of minds and mending of cultural differences is more elusive than the fleeting form of agreement that a Shotgun Referendum would seek to prompt. For many conflicts, only this minimal agreement may be possible, and — as I argue in the next part — only this bare level of agreement may be necessary.
After scaffolding a tenuous agreement, deliberative democratic institutions might concretise the agreement as a lasting constitutional settlement. They might do so by drawing on the perceived legitimacy accorded to constitutional decision-making conducted by ordinary citizens. This possibility addresses a glaring gap in deliberative conflict studies. Is deliberative democracy
in conflict settings limited to prompting a mere transient meeting of minds? Even though experiments on small-group deliberation across deeply divided settings ‘are, on balance, encouraging, ... very little work has hitherto been done to study the long-term effects of deliberative experiences on intergroup relations’.[98] In this part, I describe the referendum’s potential to legitimate
a lasting constitutional settlement. I describe this ‘popular legitimacy’ rationale for referendums, again initially in idealised terms (Part IV(A)). I then raise counterarguments (Part IV(B)) and lay out institutional preconditions (Part IV(C)).
Ideas of popular legitimacy gained currency in the conceptions of constitutional authority developed around the American and French Revolutions.[99] At the outset of the French Revolution, Abbé Sieyès articulated the notion of ‘constituent power’,[100] which continues to influence Continental European and, more recently, Anglo-American constitutional commentary.[101] A key precept for Sieyès was that state and constitutional authority derive from a largely inviolable will of the ‘people’ or ‘nation’. This constituent power in turn creates ‘constituted powers’ — including constitutional laws and elite institutions — to help administer the state. Drawing on social contract theory, Sieyès reasoned that the constituent power stemmed from the continuing existence of the people, as a whole rather than individually, in a state of nature. The people should remain unfettered when collectively expressing their will.
Today these early notions can appear artificial, naive, even dangerous. They implicitly challenge contemporary rule of law notions. Also troubling is the connection between constituent power and theories of unrestrained nationalism. Carl Schmitt, for one, drew on ideas of constituent power to provide intellectual support to militant German nationalism prior to the Second World War.[102] Additionally, the social contract metaphor itself is much criticised, both for its air of unreality and its inability to account for those at the margins of a society.[103] Finally, in practice a united populace rarely eventuates.[104] This latter objection is pertinent in conflict societies, where political unity is an even more marked fiction.
However, stripped of some of their ponderous elements, ideas of popular legitimacy rest on more defensible claims. At the centre of any democratic theory is the uncomplicated principle that no persons should be excluded arbitrarily from a decision-making process affecting them, nor from influencing its outcome.[105] This proposition applies with particular force to constitutional laws, which typically have outsized effect.
While a corollary of popular legitimacy might be that any formal process of reform of a constitution requires popular consent in the form of a vote by ordinary citizens,[106] this conclusion does not necessarily follow. It was the prevailing understanding in France when, in 1793, revolutionary leaders called a referendum to endorse the new Jacobin constitution. The use of referendums and plebiscites continued thereafter, including in the Imperial era. However, many others — Sieyès included — assumed that representative democracy could express popular preferences better than direct popular rule. Ordinary citizens might lack necessary knowledge of legal and governmental affairs. Direct popular rule also runs the risk that, absent the moderating influence of elites, lay citizens would sort into divisive voting blocs based on lines of region, religion, class, etc. Elite leadership might therefore better translate popular sentiments into coherent and concrete forms of positive law.
Even as early revolutionaries pioneered theories of popular constitutional ownership, then, some sought to curb the theories’ practical import. In the US, except in certain radical corners, from the outset a referendum requirement was not widely assumed.[107] A formal process of constitutional change was seen as best realised by means of conventions (conclaves of political elites), or by voting within legislatures (not all of them elected). Indeed, contemporary observers often deride popular methods as unschooled and minimally deliberative.[108] In this light, popular legitimacy falls short as a rationale for holding constitutional referendums. It does not unequivocally recommend such referendums, because it does not answer questions about which model of democracy — direct or representative — best manifests popular sovereignty.
Nevertheless, deliberative democratic and constitutional theory can suggest the relative value of holding referendums in support of constitutional settlement. In comparison with elite leadership, a referendum may engender higher public trust in a settlement process, provided it is a Deliberative Referendum. Recall the deliberative hierarchy’s assumed relation of trust between citizens and governmental elites. Publics must be able to rely on the good conduct of elites who serve as their delegates or trustees and who are expected to act ‘in a generally trustworthy manner — to attempt to listen to, reconcile, and apply public values in good faith’.[109] Trust is ‘a bet about the future contingent actions of others’[110] and, in situations where power must be delegated, ‘an attitude of optimism’ about the delegate’s character or conduct.[111] Particular attributes may determine a public’s trust in elites, where it exists: for example, knowledge, experience, breadth of vision, and commitment to deciding impartially and in good faith.
Yet elected leaders might tend instead to their own preferences while in office.[112] Elite power roles raise temptations to issue biased, corrupt or otherwise self-serving decisions.[113] Elsewhere I have written that
[w]hile these hazards are standard whenever power must be delegated, reform
of constitutional norms particularly challenges the relation of trust between
the tiers of a deliberative hierarchy. Such reform is often distinguished by the gravity of interests engaged, ... its effective irreversibility, [and its] wide substantive range.[114]
Hence the bet about the actions of others becomes riskier.
Studies support these observations in some non-conflict societies, where the balance of trust favours popular decision-making. My own Australian study showed that an almost fully non-elite-led constitution-making process — a citizens’ assembly followed by a referendum — can attract significantly greater trust than governmental elites who initiate and guide a referendum process.[115] However, and importantly, this preference depends on deliberative institutions being in place to mitigate ordinary citizens’ deliberative deficiencies.[116] In another telling result, respondents expressed a marked preference for constitutional reform processes that follow deliberative values of ‘fairness and impartiality’, as against simple majoritarianism.[117]
In conflict societies, the risk of self-seeking rather than public-spirited decision-making also can particularly manifest during constitutional change, including conflict settlement. Elite heads of warring parties often seek to amend constitutional norms to entrench their own dominance. Constitutional norms can calibrate political systems in favour of dominant ethnically based political parties. Constitutional matters also present tempting targets for elites who, as we saw in Part III, tend toward outsized partisan sentiments and may have structural incentives to maintain violent conflict. We may speculate that these factors will often depress public trust in elite-led constitutional settlement. Hence deliberation by ordinary citizens is frequently characterised as having a circuit-breaking function, in light of the view that non-elites are above the fray and relatively able to resolve an impasse. This assumption helps to explain the modern popularity of settlement referendums. While some other agents (eg NGOs, UN bodies and neutral foreign governments) might be more impartial, the task often falls to ordinary citizens after popular legitimacy considerations are weighed in.
Voting in referendums might generate widely accepted and durable reforms, in comparison with reform via legislatures, courts, governmental elites, and international facilitators acting alone. Constitutional changes might be relatively enduring if they are robustly democratic in their origins. An empirical study by Carey that compared constitutional durability in 66 countries considered the influence of democracy in constitution-making moments.[118] Constitutions were half as likely to be amended soon after enactment — and thus more stable — after ‘constitutional moments involving the most inclusive elected assemblies’, a category that includes robustly and fairly elected assemblies.[119] Other suggestive studies establish that citizens who self-perceive as being on the losing side of a policymaking debate tend to accept outcomes from a process that is apparently impartial, fair and democratic.[120] Social backlash or outright reversal are more likely when a non-democratic process purports to resolve contentious matters.[121]
Carey’s sample data are too limited to provide evidence about the role of referendums in particular.[122] Other work, however, more directly suggests a connection between referendums and endurance. Tierney writes that ‘the referendum can take on moral force, and can serve to unsettle constitutional systems and their self-understanding of the normative underpinnings of
the state’ — more than a declaration by an elected legislature ever could.[123] Indeed, popular methods of reform might advertise the importance of the reform process and of any decisions it yields. Referendums are usually rare, expensive and difficult to organise. Moreover, as we saw, the forte of non-elites is value-deliberation; their involvement in reform via a referendum can give deliberations a relatively value-deliberative and constitutional cast. That is, while a paradigm-shifting event such as constitutional reform is often presumed to entail public involvement, to some extent the reverse relation also seems to hold: the referendum can establish the ceremony, solemnity
and value-focus that befit changes to constitutional norms. In turn, this may signal that norms emerging from the process ought to be viewed as foundational and enduring.
Carey conjectures that, to ‘the extent that the mere occurrence of a referendum clearly signals the adoption of a constitution, one might expect referenda to foster coordination around the new set of rules’.[124] This assumes that a society has room for only one set of foundational governing norms. Thus a
primary job of constitutions is to resolve intractable coordination problems — for example, how political actors with diverse interests might contest for political authority on a regular basis without mutual destruction. Coordination equilibria require stable mutual expectations among actors ...[125]
A society is unlikely to select more than one set of norms as constitutional since these might be contradictory.[126] In conflict societies, especially given separatist struggles, this assumption is more contestable. However, even here the parties might seek a single set of overarching rules — whether to avoid ‘mutual destruction’, or perhaps at long last to extinguish a rival’s claims to one’s ‘own’ territory (eg the old Irish Constitution’s assertion of sovereignty over the entire island).
Carey bases his views on Hardin, who notes that durability flows not per se from democratic inclusiveness in constitution-making, but from social consensus about what counts as ‘the constitution’.[127] Even though a referendum might usefully signal such consensus, a consensus often will emerge even absent a referendum. Other mechanisms that might determine durability include using the existing formal constitutional amending procedures to inscribe new norms as part of the binding constitutional text. Hence, in ways that are partly interlinked and partly distinct, democratic legitimacy and constitutional status may contribute to a social consensus that a norm should endure because it is part of what Elster calls the ‘set of laws collectively referred to as “the constitution”’.[128] Referendums may help to constitutionalise the settlement and shield it from subsequent breach and reversal.
The previous section’s arguments are open to several lines of critique.
Recall the problem of uneven commitments to deliberation: to different people the legitimacy of law appears to stem from different sources. In conflict societies, only a small proportion might prefer ‘fair and impartial’ values for
a reform process. Where trust in state institutions is weak and a population
is accustomed to corrupt and self-serving governance, most people will not credit, nor therefore expect, that any lawmaking process can be deliberative
in practice. Moreover, electoral majorities may express similar doubt for another reason. Deliberative democracy is sometimes perceived as ‘the weapon of the weak’.[129] Attitudes toward deliberative democracy may indeed be more positive among minority social factions, which stand to gain
from fair process, than among members of the majority (eg ethnic Russians
in Crimea).[130]
Shotgun Referendums face the predicament that buy-in by at least some governmental leaders is normally required before a binding referendum can be held. Given elite incentives and tendencies toward polarisation, winning broad elite support for holding a Shotgun Referendum can seem unlikely. However, in practice state-sanctioned conflict settlement referendums are common. Only a very small number of these referendums are designed merely to lend a decision — such as an annexation of territory — a false front of legitimacy (eg the Crimean vote, which lacked democratic safeguards).[131] Additionally, there remains the option for informal plebiscites to be held, as in Hong Kong in 2015 prior to the ‘Umbrella’ movement for democracy; these require neither state authorisation nor state leadership.
The traditional notions of popular legitimacy outlined above rested on abstractions such as the unity of a people with shared values or identities. A number of contemporary deliberativists also view citizen solidarity as a precondition for deliberative democracy. As Tierney puts it, ‘a constitutional referendum implies a pre-existing demos, the absence of which is precisely the dilemma of the divided society’.[132] Some deliberativists stipulate that deliberation should take place only in the absence of sharp popular division, as in such a case achieving common ground is more plausible.[133] To be sure, social monoculture can hamper deliberation by providing too few opportunities for diverse policy ideas to be aired and challenged. But, at the opposite extreme, many conflicts indeed rest on deep and pervasive divisions, and an oppositional and combative relation with longstanding enemies forms a significant part of many citizens’ identities. Here, deliberation between parties is especially hard to achieve.
Nevertheless, an overstrict expectation of popular unity confines deliberative democracy to easy cases where deliberation is already happening. The requirement would leave little room for messier, real-world circumstances in which incremental improvements in deliberation might at least marginally improve the prospects of settlement. Again, for several reasons, a minimalist view of deliberation is appropriate here. First, the minimalist approach focuses not on the expression — and much less the construction — of intercommunity unity, but on accepting deep differences and managing their effects where possible. In this respect the approach matches a key idea
in deliberative democracy theory, which takes some differences of value, interest, worldview, etc, partly as given, and asks how decision-making can function despite these differences. The deliberativist focus on overlapping rather than complete consensus seeks only the minimum agreement required to resolve the key differences preventing agreement. It does not seek to level away all difference.
Second, a useful way to frame popular legitimacy is as a question of public trust in particular institutions, rather than between opposing groups. This resonates with Habermas’s view that civic solidarity requires no more than a ‘legally mediated solidarity between strangers’;[134] some constitutional processes do not draw upon, but rather create such solidarity ‘through the legal institutionalization of citizens’ communication’.[135] In this light, even state boundaries, which obviously divide peoples, may not always be determinative as to whether deliberation leading to a constitutional settlement can take place. More important are questions of how deep are the divisions, and whether effective deliberative institutions exist to mitigate the divisions.
Finally, achieving robust deliberation within the confines of the referendum may be simpler than doing so more generally. The well-known problem of scale in deliberative democracy theory is that the more people we include in a deliberative process, the harder it is to secure deliberation.[136] Referendum deliberation at large is not as easy to structure as mini-publics, but it is more amenable to institutional design than popular deliberation more broadly. The referendum is a discrete moment in time focused on circumscribed matters. The challenge of popular deliberation in a referendum is more modest than the challenge usually undertaken in the field of conflict. There is no need, at least in the first instance, to mend deep social differences. The expectation — more modest, and still speculative — is that deliberation during a referendum can result from careful institutional design and can briefly engender wide democratic and deliberative involvement by citizens, as well as ultimately a more durable constitutional settlement. Yet, for this, something more than a standard referendum is required.
We have already seen that a referendum designed to support deliberation might attract greater perceptions of legitimacy than would elite decision-making. In this regard the deliberative Shotgun Referendum features enumerated in Part III may help. Beyond this, the Shotgun Referendum should also of course adhere to standard democratic norms, such as universal adult franchise; anonymous balloting; fair access to public airwaves and other modes of publicity and discussion; and legal safeguards against partisan criminal prosecution. Such provisions can reduce the capacity of self-interested factions to capture a referendum for their own purposes — a problem both for democracy and for deliberation.[137]
A Shotgun Referendum also potentially helps to answer the problem of uneven commitments to deliberation by straddling the divide between deliberators and non-deliberators. As an institution robustly adopting both democratic and deliberative features, it might enjoy broad, dual legitimacy. Its legitimacy could be agreed to by a wide cross-section of people, including by those who value majoritarian process most and by those who value deliberation most.
A referendum should run prior to elite-led negotiations and should formally influence the negotiation process. A referendum negotiation presented to voters as a fait accompli might attract low public trust and, as in Colombia, may not receive adequate popular support. Referendum voters are often asked to defer to elite experts by consenting to a settled reform initiative about which they know little. Given low levels of public trust in most contemporary governments, an opaque, top-down settlement approach — in which a referendum merely caps a lengthier decision-making process conducted by elites — may not attract substantial public confidence.[138] Non-elites may neither understand nor accept the complex compromises previously negotiated by elites. But broader popular engagement early on can help to increase public buy-in to an agreement.[139] Running the referendum first might also give impetus to elites to reach a settlement in the first place.
While the referendum campaign is running, and until voting results are revealed, elites should be largely sidelined. Tierney attributes the success of the Belfast Agreement referendum in part to a process that saw ‘elites ... t[ake] cues from the people’[140] through methods such as extensive public opinion polling and ‘the intervention of a range of voices [eg civil society groups] in addition to traditional government and party elites’.[141] Similarly, according to Parkinson, one of the ‘crucial factors’ that engendered public trust and deliberation in the Scottish independence referendum was the Scottish government’s decision not to dominate discourse in the public sphere in the course of the campaign.[142]
Examples from non-conflict societies show how Deliberative Referendums might operate in practice. But as yet there has been little written about Deliberative Referendums in the hardest context of all. A Shotgun Referendum is a Deliberative Referendum held under the shadow of conflict. Its methods are modest: in the first instance, the objective is not to prompt broad and enduring social reconciliation, but only a transient deliberative moment. In many conflict societies, this may be the most any referendum can achieve. Yet, in a subset of cases, this may be sufficient to ground a durable constitutional settlement.
The standard caveat of deliberative democrats is that no institutional prescription is a panacea, as ‘in reality there will usually be a great deal that can go wrong’.[143] The ability of the Shotgun Referendum, or any innovation, to produce a constitutional settlement is speculative at best; worse, new techniques may be wasteful or counterproductive. Indeed, I have shown how a Shotgun Referendum’s deliberative shortcuts and limitations would cause it to fall short of some deliberative ideals. Yet nearly all current conflict-society referendums proceed as simple majoritarian exercises that offer little in the way of support for deliberation. Deliberative design might improve on these standard models. The aim of this article has been to elaborate a framework to conceptualise this objective, along with a set of presumptive institutional prerequisites for Shotgun Referendum practice.
* BSc (McGill), JD (Toronto), LLM (Columbia), PhD (Osgoode); Associate Professor, Austral-ian National University. I am thankful for funding from the Australian Research Council
Discovery Projects DP[1]40102682 (Confronting the Devolution Paradox) and DP130100706 (The Law of Deliberative Democracy: Theory and Reform), for comments on drafts by Will Partlett, Amelia Simpson, Lulu Weis, Melbourne University Legal Theory Workshop participants and this article’s two peer reviewers, and research assistance by Eleanor Mitchell and Samuel Rutherford.
[1] Intrastate: eg Democratic Republic of the Congo, Iraq, Kenya, Somalia, Spain, Zanzibar. Interstate (or intra- and interstate): eg Bougainville/Papua New Guinea, East Timor/
Indonesia, Eritrea/Ethiopia, Montenegro/Yugoslavia, Northern Ireland/UK/Republic
of Ireland.
[2] Christine Bell, ‘Lex Pacificatoria Colombiana: Colombia’s Peace Accord in Comparative Perspective’ (2016) 110 American Journal of International Law Unbound 165.
[3] On options such as consociationalism, special autonomy, centripetal integration, ethnic quotas and institutionalised deliberative democracy, see, eg, Arend Lijphart, Patterns of Democracy: Government Forms and Performance in Thirty-Six Countries (Yale University Press, 1999); John S Dryzek, ‘Deliberative Democracy in Divided Societies: Alternatives to Agonism and Analgesia’ (2005) 33 Political Theory 218; Benjamin Reilly, ‘Centripetalism: Cooperation, Accommodation and Integration’ in Stefan Wolff and Christalla Yakinthou (eds), Conflict Management in Divided Societies: Theories and Practice (Routledge, 2012) 57; Karen Bird, ‘Ethnic Quotas and Ethnic Representation Worldwide’ (2014) 35 International Political Science Review 12; George Anderson and Sujit Choudhry, ‘Constitutional Transitions and Territorial Cleavages’ (Working Paper, International Institute for Democracy and Electoral Assistance, 2015) 24–5.
[4] Jürgen Habermas, ‘Reconciliation through the Public Use of Reason: Remarks on John Rawls’s Political Liberalism’ (1995) 92 Journal of Philosophy 109, 124.
[5] John Rawls, Political Liberalism (Columbia University Press, rev ed, 2005) 137.
[6] Joshua Cohen, ‘Deliberation and Democratic Legitimacy’ in James Bohman and William Rehg (eds), Deliberative Democracy: Essays on Reason and Politics (MIT Press, 1997) 67, 72.
[7] Zsuzsanna Chappell, Deliberative Democracy: A Critical Introduction (Palgrave Macmillan, 2012) 8.
[8] Amy Gutmann and Dennis Thompson, Democracy and Disagreement (Harvard University Press, 1996) 43.
[10] Gutmann and Thompson, Democracy and Disagreement (n 8) 57–9, 79–81.
[11] Chantal Mouffe, ‘Deliberative Democracy or Agonistic Pluralism?’ (1999) 66 Social
Research 745.
[12] Rawls, Political Liberalism (n 5) 132–72; John S Dryzek and Simon Niemeyer, ‘Reconciling Pluralism and Consensus as Political Ideals’ (2006) 50 American Journal of Political
Science 634.
[13] Ron Levy, ‘“Deliberative Voting”: Realising Constitutional Referendum Democracy’ [2013] Public Law 555; Stephen Tierney, ‘Using Electoral Law to Construct a Deliberative Referendum: Moving beyond the Democratic Paradox’ (2013) 12 Election Law Journal 508.
[14] The main exceptions are discussed in Stephen Tierney, Constitutional Referendums: The Theory and Practice of Republican Deliberation (Oxford University Press, 2012) 241–59.
[15] Ralph R Premdas, ‘The Guyana Ethnic Quagmire: Problems and Solutions for Reconciliation’ (2004) 10 Nationalism and Ethnic Politics 251, 254–5.
[16] Michael Billig and Henri Tajfel, ‘Social Categorization and Similarity in Intergroup Behaviour’ (1973) 3 European Journal of Social Psychology 27.
[17] Dryzek, ‘Deliberative Democracy in Divided Societies’ (n 3) 223.
[18] Ibid.
[19] Sujit Choudhry, ‘Bridging Comparative Politics and Comparative Constitutional Law: Constitutional Design in Divided Societies’ in Sujit Choudhry (ed), Constitutional Design for Divided Societies: Integration or Accommodation? (Oxford University Press, 2008) 3; Adeno Addis, ‘Deliberative Democracy in Severely Fractured Societies’ (2009) 16 Indiana Journal of Global Legal Studies 59.
[20] For example, in the 2010 Zanzibar referendum to decide on a government of national unity, the president was quoted as saying that ‘this issue of a government of national unity is a good thing, and therefore there is no reason of rejecting it. Anyone who will reject it is insane’: Mohammed Bakari and Alexander Makulilo, ‘Beyond Polarity in Zanzibar? The “Silent” Referendum and the Government of National Unity’ (2012) 30 Journal of Contemporary African Studies 195, 205 (citations omitted).
[21] Carolyn Hamilton, ‘Uncertain Citizenship and Public Deliberation in post-Apartheid South Africa’ (2009) 35 Social Dynamics 355, 359–62.
[22] Brendan O’Leary, ‘Debating Consociational Politics: Normative and Explanatory Arguments’ in Sid Noel (ed), From Power Sharing to Democracy: Post-Conflict Institutions in Ethnically Divided Societies (McGill-Queen’s University Press, 2005) 3, 10.
[23] John Coakley, ‘Ethnic Conflict Resolution: Routes towards Settlement’ (2009) 15 Nationalism and Ethnic Politics 462, 467.
[24] John S Dryzek, Deliberative Democracy and beyond: Liberals, Critics, Contestations (Oxford University Press, 2000) 66.
[26] Dryzek, ‘Deliberative Democracy in Divided Societies’ (n 3) 219.
[27] See, eg, Rajat Ganguly, ‘Sri Lanka’s Ethnic Conflict: At a Crossroad between Peace and War’ (2004) 25 Third World Quarterly 903.
[28] See, eg, Kyle Welch, ‘The Patriot Act and Crisis Legislation: The Unintended Consequences of Disaster Lawmaking’ (2015) 43 Capital University Law Review 481; Brian Blankenship, ‘When Do States Take the Bait? State Capacity and the Provocation Logic of Terrorism’ (2018) 62 Journal of Conflict Resolution 381.
[29] Mary Derosa and Milton Regan, ‘Deliberative Constitutionalism in the National Security Setting’ in Ron Levy et al (eds), The Cambridge Handbook of Deliberative Constitutionalism (Cambridge University Press, forthcoming) 39.
[30] See, eg, Arend Lijphart, The Politics of Accommodation: Pluralism and Democracy in the Netherlands (University of California Press, 2nd ed, 1975) 111; O’Leary (n 22) 10.
[31] Richard A Posner, Law, Pragmatism, and Democracy (Harvard University Press, 2003) 167–8.
[32] Dan M Kahan, ‘Ideology, Motivated Reasoning, and Cognitive Reflection’ (2013) 8 Judgment and Decision Making 407, 416–18.
[33] Ibid.
[34] Ibid 417–18.
[35] Ibid 408.
[36] Bridie Jabour, ‘Cory Bernardi Links Same-Sex Marriage to Polygamy and Bestiality Again’, The Guardian (Sydney, 18 June 2013) <www.theguardian.com/world/2013/jun/18/cory-bernardi-same-sex-bestiality>, archived at <https://perma.cc/HP5E-ZTWA>.
[37] Daniel C Lewis, ‘Bypassing the Representational Filter? Minority Rights Policies under Direct Democracy Institutions in the US States’ (2011) 11 State Politics and Policy Quarterly 198 (reviewing conflicting studies). Mixed results obtain even for that most feared of referendum types, the citizen-initiated referendum: Zoltan L Hajnal, Elisabeth R Gerber and Hugh Louch, ‘Minorities and Direct Legislation: Evidence from California Ballot Proposition Elections’ (2002) 64 Journal of Politics 154.
[38] John Parkinson, ‘Legitimacy Problems in Deliberative Democracy’ (2003) 51 Political Studies 180, 180.
[39] Ibid 183.
[40] In a similar vein, see Jürgen Habermas, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy, tr William Rehg (MIT Press, 1996) 356.
[41] Parkinson (n 38) 183; John S Dryzek, Foundations and Frontiers of Deliberative Governance (Oxford University Press, 2010) 23–4 (discussing elitist positions).
[42] Amy Gutmann and Dennis Thompson, Why Deliberative Democracy? (Princeton University Press, 2004) 132.
[43] Charles Taylor, ‘Conditions of an Unforced Consensus on Human Rights’ in Joanne R Bauer and Daniel A Bell (eds), The East Asian Challenge for Human Rights (Cambridge University Press, 1999) 124, 124.
[44] John Parkinson, Deliberating in the Real World: Problems of Legitimacy in Deliberative Democracy (Oxford University Press, 2006) 99.
[45] Gutmann and Thompson, Democracy and Disagreement (n 8) ch 2; Rawls, Political Liberalism (n 5) 137.
[46] Robert E Goodin, Motivating Political Morality (Blackwell, 1992) 132–3.
[47] Jon Elster, ‘Deliberation and Constitution Making’ in Jon Elster (ed), Deliberative Democracy (Cambridge University Press, 1998) 97, 111 (emphasis omitted).
[48] What values come to be accepted as salient amid a particular campaign of constitutional reform can be unpredictable — determined as much by popular discussion as by conscious efforts by elites to define the campaign’s overarching terms and purposes.
[49] See, eg, Obergefell v Hodges, 576 US ___ (2015) (writing, inter alia, on the needs of same-sex couples to be loved and respected).
[51] Jon Elster, ‘Forces and Mechanisms in the Constitution-Making Process’ (1995) 45 Duke Law Journal 364, 366. Using this broad definition partly obviates juridical debates about a settlement’s normative status: see, eg, Sentencia C-379/16 (Corte Constitucional de Colombia, Magistrado Ponente Luis Ernesto Vargas Silva, 18 July 2016), on the 2016 Colombian agreement and referendum. Bell notes that the Colombian agreement arguably achieved a legal status ‘somewhere “hybrid” between international and constitutional law’: Bell (n 2) 169.
[52] Cass R Sunstein, ‘Incompletely Theorized Agreements’ (1995) 108 Harvard Law Review 1733.
[53] Thomas Hennessey, The Northern Ireland Peace Process: Ending the Troubles? (Palgrave,
2001) 7–9.
[54] Ibid 6.
[55] Ibid 9.
[56] Ibid 10.
[57] Ibid 9.
[58] Ibid 185–6; Arthur Aughey, The Politics of Northern Ireland: Beyond the Belfast Agreement (Routledge, 2005) 148.
[59] Ian O’Flynn, ‘Divided Societies and Deliberative Democracy’ (2007) 37 British Journal of Political Science 731, 741. See also Addis (n 19) 64. Of course, these are distinctions of degree.
[60] Scott Bennett, ‘The Politics of Constitutional Amendment’ (Research Paper No 11 2002–03, Department of the Parliamentary Library (Cth), 23 June 2003) 20–1.
[61] Tierney, Constitutional Referendums (n 14).
[62] Bruno S Frey, ‘Direct Democracy for Transition Countries’ (2003) 7 Journal for Institutional Innovation, Development and Transition 42; Ron Levy, ‘The Deliberative Case for Constitutional Referenda’ (2017) 16 Election Law Journal 213.
[63] In Australia, 75% nominated ‘the people’ as best placed to decide whether to sever ties with the British monarchy. On ‘moral’ issues, 65% and 66% viewed same-sex marriage legalisation and physician-assisted suicide, respectively, as matters best decided by non-elites. By contrast, for deciding specific power arrangements under federalism, the result was 34%: 2017 Australian Constitutional Values Survey (Survey, 2017), led by investigators including the author (data on file with the author).
[64] Ron Levy and Graeme Orr, The Law of Deliberative Democracy (Routledge, 2017) 46–7;
Ron Levy, ‘The “Elite Problem” in Deliberative Constitutionalism’ in Ron Levy et al (eds),
The Cambridge Handbook of Deliberative Constitutionalism (Cambridge University Press, forthcoming).
[65] Carolyn M Hendriks, John S Dryzek and Christian Hunold, ‘Turning up the Heat: Partisanship in Deliberative Innovation’ (2007) 55 Political Studies 362, 369–71.
[66] Dryzek, ‘Deliberative Democracy in Divided Societies’ (n 3) 226.
[67] Brian Tobin, ‘Marriage Equality in Ireland: The Politico-Legal Context’ (2016) 30 International Journal of Law, Policy and the Family 115, 115.
[70] See ibid.
[71] Yasmin Dawood, ‘The Antidomination Model and the Judicial Oversight of Democracy’ (2008) 96 Georgetown Law Journal 1411.
[72] Habermas, ‘Reconciliation through the Public Use of Reason’ (n 4) 124.
[73] Michael Walzer, ‘Deliberation, and What Else?’ in Stephen Macedo (ed), Deliberative Politics: Essays on Democracy and Disagreement (Oxford University Press, 1999) 58.
[74] Dryzek, ‘Deliberative Democracy in Divided Societies’ (n 3) 229 (emphasis omitted).
[75] James Bohman, Public Deliberation: Pluralism, Complexity, and Democracy (MIT Press, 2000) 81 (describing and critiquing this notion).
[76] An aspect of this critique is that the ideal of citizens deliberating on equal footing may be only a fiction, and that the ideal’s arid, rationalist form of discourse is the preserve of the mainstream, wealthy, educated segments of a society: see, eg, Iris Marion Young, Justice and the Politics of Difference (Princeton University Press, 1990); Seyla Benhabib, ‘Deliberative Rationalality [sic] and Models of Democratic Legitimacy’ (1994) 1 Constellations 26, 39–40.
[77] M Lydia Khuri, ‘Facilitating Arab–Jewish Intergroup Dialogue in the College Setting’ (2004) 7 Race Ethnicity and Education 229, 244.
[79] Tierney, Constitutional Referendums (n 14) 254.
[80] Parkinson (n 38) 186–7. See also Janette Hartz-Karp and Michael K Briand, ‘Institutionalising Deliberative Democracy: Theoretical and Practical Challenges’ (2009) 24(1) Australasian Parliamentary Review 167.
[81] For further discussion on referendum timing, see Part IV.
[82] See, eg, Levy, ‘Deliberative Voting’ (n 13).
[83] Alex J Norman, ‘The Use of the Group and Group Work Techniques in Resolving Interethnic Conflict’ (1992) 14(3–4) Social Work with Groups 175; Khuri (n 77); Robert C Luskin et al, ‘Deliberating across Deep Divides’ (2014) 62 Political Studies 116; Margarita M Orozco and Juan E Ugarriza, ‘The Citizens, the Politicians and the Courts: A Preliminary Assessment of Deliberative Capacity in Colombia’ in Juan E Ugarriza and Didier Caluwaerts, Democratic Deliberation in Deeply Divided Societies: From Conflict to Common Ground (Palgrave Macmillan, 2014) 73, 73–88.
[84] See, eg, Magdalena E Wojcieszak, ‘On Strong Attitudes and Group Deliberation: Relationships, Structure, Changes, and Effects’ (2012) 33 Political Psychology 225, 237–8 (finding that some participants in deliberation merely reconfirm strongly held prejudices).
[85] Ian O’Flynn and Didier Caluwaerts, ‘Deliberation in Deeply Divided Societies’ in André Bächtiger et al (eds), Oxford Handbook of Deliberative Democracy (Oxford University Press, forthcoming).
[86] Ibid.
[87] Fred Cutler et al, ‘Deliberation, Information, and Trust: The British Columbia Citizens’ Assembly as Agenda Setter’ in Mark E Warren and Hilary Pearse (eds), Designing Deliberative Democracy: The British Columbia Citizens’ Assembly (Cambridge University Press, 2008)
166, 168–70.
[89] Adrian Vermeule, ‘Veil of Ignorance Rules in Constitutional Law’ [2001] YaleLawJl 2; (2001) 111 Yale Law Journal 399, 399.
[90] John Rawls, A Theory of Justice (Belknap Press, 1971) 136–42.
[91] In this scoping article I leave aside details such as how to define electorates for each community — eg whether to include diaspora populations.
[92] On deliberative effects and risks of veto, see Anna Drake and Allison McCulloch, ‘Deliberative Consociationalism in Deeply Divided Societies’ (2011) 10 Contemporary Political Theory 372, 385–6.
[93] Dryzek, ‘Deliberative Democracy in Divided Societies’ (n 3) 224.
[96] Bora Kanra, Islam, Democracy and Dialogue in Turkey: Deliberating in Divided Societies (Ashgate, 2009). Hennessey similarly thinks that attempts to achieve peace in Northern Ireland in the early 1970s faltered because social attitudes had yet to progress toward greater social acceptance of the legitimacy of other communities: Hennessey (n 53) 18.
[97] Addis (n 19) 82; see also at 69. Addis imagines this happening over time at a social level. However, he also advocates a form of minimalism (‘incompletion’) by cautioning that constitutions should not seek to settle all contentious issues at once: at 76–7.
[98] O’Flynn and Caluwaerts (n 85).
[99] Alexander Hamilton or James Madison, ‘Federalist No 49’ in Michael A Genovese (ed), The Federalist Papers (Palgrave Macmillan, 2009) 111.
[100] Emmanuel Joseph Sieyès, What Is the Third Estate?, ed SE Finer, tr M Blondel (Pall Mall Press, 1963).
[101] See, eg, Martin Loughlin, The Idea of Public Law (Oxford University Press, 2003); Joel Colón-Ríos, ‘The Legitimacy of the Juridical: Constituent Power, Democracy, and the Limits of Constitutional Reform’ (2010) 48 Osgoode Hall Law Journal 199; Stephen Tierney, ‘Sovereignty and Crimea: How Referendum Democracy Complicates Constituent Power in Multinational Societies’ (2015) 16 German Law Journal 523.
[102] See, eg, Carl Schmitt, ‘The Führer Protects the Law: On Adolf Hitler’s Reichstag Address,
13 July 1934’, tr Clara Teresa Picker and John P McCormick, Scribd (Web Page) <www.scribd.com/document/266691493/Carl-Schmitt-Fuehrer-Protects-the-Law>, archived at <https://perma.cc/WFV6-7XP2>.
[103] See, eg, Carole Pateman, The Sexual Contract (Polity Press, 1988).
[104] Jürgen Habermas, The Inclusion of the Other: Studies in Political Theory, ed Ciaran Cronin and Pablo De Greiff (MIT Press, 1999) 159.
[105] Thomas Christiano, The Rule of the Many: Fundamental Issues in Democractic Theory (WestviewPress, 1996) 53–6.
[106] Constituent powers theory can lead to similar conclusions. Joel Colón-Ríos argues that a constitution should ‘have an outlet for constituent power to manifest when important constitutional transformations are needed’: Colón-Ríos (n 101) 201.
[107] James S Fishkin, When the People Speak: Deliberative Democracy and Public Consultation (Oxford University Press, 2011) 18–19 (discussing revolutionary Rhode Island).
[108] Benjamin Moffitt and Simon Tormey, ‘Rethinking Populism: Politics, Mediatisation and Political Style’ (2014) 62 Political Studies 381, 392.
[109] Levy, ‘The Deliberative Case for Constitutional Referenda’ (n 62) 216 (citations omitted). See also Mark Thatcher and Alec Stone Sweet, ‘Theory and Practice of Delegation to Non-Majoritarian Institutions’ (2002) 25(1) West European Politics 1, 2. This conception of trust differs from Edmund Burke’s well-known vision of relatively autonomous trustee-parliamentarians: Edmund Burke, ‘Speech to the Electors of Bristol’ in Francis Canavan (ed), Select Works of Edmund Burke: Miscellaneous Writings (Liberty Fund, 1999) 3.
[110] Piotr Sztompka, Trust: A Sociological Theory (Cambridge University Press, 1999) 25.
[111] Karen Jones, ‘Trust as an Affective Attitude’ (1996) 107 Ethics 4, 5.
[112] Sanford Levinson, ‘The Political Implications of Amending Clauses’ (1996) 13 Constitutional Commentary 107, 114–17.
[113] David L Ponet and Ethan J Leib, ‘Fiduciary Law’s Lessons for Deliberative Democracy’ (2011) 91 Boston University Law Review 1249, 1256.
[114] Levy, ‘The Deliberative Case for Constitutional Referenda’ (n 62) 218.
[115] Ron Levy, ‘Breaking the Constitutional Deadlock: Lessons from Deliberative Experiments in Constitutional Change’ [2010] MelbULawRw 28; (2010) 34 Melbourne University Law Review 805, 832–8. See also Cutler et al (n 87).
[116] Levy, ‘Breaking the Constitutional Deadlock’ (n 115) 834–7.
[117] Ibid 833–4.
[118] John M Carey, ‘Does It Matter How a Constitution Is Created?’ in Zoltan Barany and Robert G Moser (eds), Is Democracy Exportable? (Cambridge University Press, 2009) 155.
[119] Ibid 172.
[120] Adam Przeworski, Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America (Cambridge University Press, 1991) 22–4. See generally Philip Pettit, ‘Republican Theory and Political Trust’ in Valerie Braithwaite and Margaret Levi (eds), Trust and Governance (Russell Sage Foundation, 1998) 295, 296–9.
[121] See, eg, William N Eskridge, Jr, ‘Pluralism and Distrust: How Courts Can Support Democracy by Lowering the Stakes of Politics’ [2005] YaleLawJl 59; (2005) 114 Yale Law Journal 1279.
[123] Tierney, ‘Sovereignty and Crimea’ (n 101) 536; see also at 533.
[125] Ibid 158.
[126] Ibid 158, 160.
[127] Russell Hardin, ‘Why a Constitution?’ in Bernard Grofman and Donald Wittman (eds), The Federalist Papers and the New Institutionalism (Agathon Press, 1989) 100.
[128] Elster, ‘Forces and Mechanisms in the Constitution-Making Process’ (n 51) 366.
[129] Gutmann and Thompson, Democracy and Disagreement (n 8) 133.
[130] See generally Donald L Horowitz, ‘The Challenge of Ethnic Conflict: Democracy in Divided Societies’ (1993) 4(4) Journal of Democracy 18, 29.
[131] See Tierney, ‘Sovereignty and Crimea’ (n 101) 534–5.
[132] Tierney, Constitutional Referendums (n 14) 252.
[133] Sarah Song, ‘The Boundary Problem in Democratic Theory: Why the Demos Should Be Bounded by the State’ (2012) 4 International Theory 39, 47.
[134] Habermas, The Inclusion of the Other (n 104) 159.
[135] Ibid 161. See also Jürgen Habermas, The Lure of Technocracy, tr Ciaran Cronin (Polity,
2015) 37–9.
[136] Robert E Goodin, ‘Democratic Deliberation Within’ (2000) 29 Philosophy and Public
Affairs 81.
[137] The two are not wholly distinct. ‘[D]eliberation reinforces democracy, and democracy in turn reinforces deliberation’: Levy and Orr (n 64) 50. For example, informed voters can express sincere preferences in a referendum, based on genuine knowledge of the subject: at 26–7.
[138] Levy, ‘Breaking the Constitutional Deadlock’ (n 115) 807; Tierney, Constitutional Referendums (n 14) 247.
[139] Tierney, Constitutional Referendums (n 14) 252.
[140] Ibid 250.
[141] Ibid 249.
[142] John Parkinson, ‘Ideas of Constitutions and Deliberative Democracy and How They Interact’ in Ron Levy et al (eds), The Cambridge Handbook of Deliberative Constitutionalism (Cambridge University Press, forthcoming) 254–5.
[143] O’Flynn and Caluwaerts (n 85).
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/MelbULawRw/2018/8.html