The Introduction to this volume warns of an impending global collapse of constitutional democracy, with threats emerging along a number of axes, including weakening foundations of constitutionalism, terrorism and environmental stresses, and ever-increasing policy demands from national populations. Evidence for this pessimistic view is amply provided by my fellow contributors, but the magnetic pull of a unifying narrative can sweep too broadly, obscuring meaningful distinctions.
Brexit has been included in this story of global democratic decline.1 In this chapter, however, I wish to resist the trope that links Brexit to this broader crisis of constitutionalism. Certainly, superficial similarities exist: a vein of extremist racialized politicking against immigrants and socioeconomic disparities among voters seem to mark Brexit as (p. 192) part of the global rise of populist nationalism. At bottom, however, Brexit is the ultimately unsurprising instantiation of two trends that have affected the United Kingdom for some time: a particular substantive ambivalence toward European integration and a specific malaise stemming from the challenges of Westminster-style parliamentary democracy.
Brexit does portend a constitutional moment in Britain, but it is one that exists in parallel to the global issues and is not of them. Brexit showcases the long-standing inconsistencies of the unusual and uniquely British version of “constitutional democracy”—a constitutional monarchy that operates without a written constitution under political expectations developed in the late nineteenth century. The historical constitution privileges parliamentary sovereignty over popular sovereignty, and the rise of referendum politics—of which Brexit is only the most recent example—has complicated and undermined that consensus, shining a bright light onto certain insufficiencies of the British system. But rather than serving as a harbinger of constitutional failure, Brexit might well usher in both democratic and constitutional renewal, providing the United Kingdom with an opportunity to clarify and concretize a new constitutional settlement for the twenty-first century.
To advance this more optimistic argument, the chapter first provides a brief outline of certain elements of the traditional British constitutional consensus, including the doctrine of parliamentary sovereignty, the concept of responsible government, and the role of the two-party political system. It then argues that these core principles have come under strain over the past few decades. Social, political, institutional, and structural changes within the United Kingdom have contributed to a new volatility in electoral politics leading to increased party fragmentation. In response, parties have relied on popular referendums to mediate internal divisions.2
The chapter next explores this turn to referendums and their surrounding political dynamics. It highlights how this use of direct democracy—heretofore unthinkable—has factored into a heightened sense of constitutional instability.3 It demonstrates that referendum politics have been marked by expediency and manipulation, tied to obtaining contingent electoral benefits or avoiding party fracture. And it claims that Brexit is the apotheosis of this trend: a high-stakes example of short-term political maneuvering. Only on this occasion, David Cameron misplayed his hand. The chapter explains how the Brexit referendum unfolded, and how the evolution of referendum politics over the (p. 193) preceding decades constrained the Cameron Government’s ability to manipulate and control the result. Cameron clearly did not anticipate the possibility of losing the vote or the serious constitutional problems that could ensue.4 Direct democracy “challenges the indirect, representative democracy that has been the essence of UK democracy. If the people vote one way, their representatives another, who should prevail, who is sovereign?”5 Prior to Brexit, this question had only been hypothetical, never tested. The Brexit vote and its aftermath are forcing a reckoning.
It is here that the chapter concludes, by arguing that Brexit is a moment of tremendous potential. Referendum politics have introduced the possibility of shifting from a system of parliamentary sovereignty to one of popular sovereignty. This transformation will require a written constitution, and Brexit may hasten its arrival. Of course, how precisely any new constitutional settlement would take shape is uncertain. Popular sovereignty does not require direct democracy; referendums themselves may not endure. But the transformation itself should be understood not as a constitutional rupture or failure, but as a new stage in the evolution of the deep-seated British commitment to constitutionalism, one that has been in development over the past eight hundred years.
The Constitution, Parliamentary Sovereignty, and British Democracy
Partially written in various Acts and statutes and partially constructed out of conventions, practices, and understandings, the British constitution defies easy identification. Even senior judges and politicians struggle, describing it as “hidden, and difficult to find,”6 or having a certain “back of an envelope” quality.7 Not only are the specific substantive elements that make up the constitution subject to debate,8 but competing understandings of its terms—in Scotland and England, in particular—have coexisted for centuries.9
The flexibility and uncertainty at the core of the British constitution stems from both its uncodified nature and its interaction with the doctrine of parliamentary sovereignty, the “dominant characteristic” of the British political system.10 As articulated by Albert (p. 194) Venn Dicey in his 1885 Introduction to the Study of the Law of the Constitution, parliamentary sovereignty is “the right to make or unmake any law whatever; and, further, that no person or body is recognized by the law of England as having a right to override or set aside the legislation of Parliament.”11
The doctrine encompasses three important ideas: First, there is no entrenchment of fundamental or “constitutional” laws. In other words, to the extent that there may be some distinction between fundamental and ordinary law (a distinction Dicey himself denied12), fundamental law does not achieve its importance by means of its creation or implementation but by political convention. Second, Parliament is supreme. No other institution “can pronounce void any enactment passed by the British Parliament on the ground of such enactment being opposed to the constitution.”13 And third, even Parliament cannot bind itself. No Parliament can bind successor Parliaments.14
The principle thus gives rise to the sense, as J.A.G. Griffith wrote in 1979, that “the constitution is no more and no less than what happens. Everything that happens is constitutional. And if nothing happened that would be constitutional also.”15 Griffith’s statement suggests that there are no limits on Parliament. But there may be political or moral ramifications to unconstitutional action. Better is the explanation provided by Lord Reid, who acknowledged that certain actions might be considered unconstitutional—that for moral or political reasons “most people would regard [them] as highly improper”—but recognized that such unconstitutionality “does not mean that it is beyond the power of Parliament to do such things.”16 Those actions could not be held invalid by a court of law.
This distinction between “constitutionality” and “legality” reinforces that it is the electorate’s ability to select a new Parliament, unconstrained by its former instantiation, that serves as the “regulating wheel” of the constitution.17 The people, in some sense, are sovereign, but only through their relationship with Parliament. The evolution of Parliament as a representative forum presents a complicated and detailed history, but ultimately its authority is derived from the institution’s ancient connection to the monarchy, rather than due to the actions of a constituent power resting in “the people.”18
Parliamentary sovereignty thus permits an “obfuscation” of the democratic principle on which the British constitution rests.19 The sleight of hand that parliamentary (p. 195) sovereignty works is to suggest that Parliament “embodies the people, so that, when Parliament is assembled, the people are there assembled.”20 Of course, “when expressed in this way, [it] has the further consequence of leaving no constitutional role for the real people. The people cannot be both embodied in Parliament and yet exist constitutionally outside Parliament.”21
Notwithstanding the jurisprudential complications underlying the British constitutional scheme, scholars have argued that this political constitutionalism has great normative benefits. Richard Bellamy makes the claim that political institutions—including party membership, equal voting power, and majority rule—can better reflect core republican values, such as non-domination and political equality, thus avoiding the potential oppression/domination of a written constitution (as outside of politics) and the anti-majoritarian nature of the courts.22 Others expand on the possibility of parliamentary decision-making as deliberative democracy, touting it (in the right circumstances) as a collective approach to decision-making that has benefits beyond mere preference aggregation.23
As the mechanism of ensuring good government, this rosy version of parliamentary democracy relies on the twin pillars of responsible government and a robust party system. Responsible government is the idea that “government is carried on by persons who are responsible to the representative house of the legislature, the House of Commons.”24 In effect, responsible government assures, through political pressure, that “government is in tune with popular opinion,”25 as it is expressed through party politics in the House of Commons.26 Thus, responsible government must rest in turn on a vibrant party system,27 which can accurately reflect the interests, views, and values of the people.28
In describing party government in the United Kingdom, Dicey called it “not the accident or the corruption but, so to speak, the very foundation of our constitutional (p. 196) system.”29 Two-party rule has been the hallmark of British electoral politics,30 sustained by the plurality voting system known as “first past the post” (FPTP). Plurality rules encourage broad “big-tent” parties and tend to limit the rise of third parties.31 Of course, there have always been third parties competing for seats,32 but “a ‘virtuous circle’ calculus of voting and activism” works in favor of the main parties, as voters learn that “third-party votes and activism tend to be wasted in terms of gaining parliamentary seats.”33
The benefits to parliamentary politics of a two-party system are significant: centrism and decisive governing capacity. As Ivor Jennings wrote, “[t]he swing of the pendulum is a familiar feature of British politics. . . . Majorities are unstable, and the Opposition of to-day is the Government of tomorrow.”34 And thus, “an administration that does not encompass the median voter is fragile.”35 Extremist minority parties are usually shut out of government,36 and the instability of multiparty coalition government is largely avoided. Indeed, as Iain McLean has demonstrated, using more than 150 years of data, there is only a 0.17 probability of a coalition administration being formed at a general election.37 Of course, this claim of generalized centrism does not preclude policy swings from occurring when the majority shifts—particularly if parties are cohesive and distinct.38 Because a majority party can enact policies without seeking broader consensus,39 governance can be effective, efficient, and partisan. And historically,40 the two British parties have been both cohesive and distinct due to rigidity in class-based voting and strongly differentiated overarching policy positions about the use of government recourses.
A functioning system of centrist and effective Westminster-style parliamentary democracy might be that elusive democratic political system that “leaves [one] free not to care about it.”41 But if that world ever existed, it is certainly not today’s situation. These core (p. 197) elements of the consensus constitution are under strain—both from substantive shifts in the content and structure of governance and from an increasing dissatisfaction with political parties and the electoral system.
Accelerating Change and Party Fragmentation
Britain is on a course of rapid change.42 Many scholars have addressed the substantive elements of institutional and structural change: the UK’s membership in the European Union;43 the enactment of a quasi-constitutional bill of rights;44 devolved legislative power to Scotland, Wales, and Northern Ireland;45 and the creation of a new Supreme Court;46 among other examples.47 At the same time, there has been an increasing and generalized dissatisfaction with party governance. Of course, changes to the substance of governance and frustration with the process of governance may not be linked. The two-party system, as critical as it appears to British politics, has been showing signs of decline for some time.48 Scholars have diagnosed the problem variously as a function of the electoral system itself,49 or as a result of: the decline in class-based voting, the rise of “alternative (p. 198) identities” that vie for primacy and encourage the creation of new parties to reflect those identities, and/or cultural changes, ranging from a modern lack of deference to hierarchy to increased media criticism of politics.50 But this Section suggests that substantive change itself—including the scope and amount of transformative legislation—is also negatively affecting the two-party system.
Foundational structural and institutional issues often do not hew to left-right divides over resource distribution. These questions are thus more likely to cut across party lines and to lead to party fragmentation.51 And indeed, debates over constitutional structure or purpose have divided the two main British parties internally, and in some cases given rise to single-issue or regional third parties.52 For example, European integration has not had a unidirectional political valence over time,53 making it an often divisive and cross-cutting issue. Other topics, such as regional devolution, have complicated parties’ national electoral politics, caused internal division, and encouraged the rise of regional parties. The party system is “increasingly fragmented,”54 with multiple third parties and viable regional parties,55 and in 2010, the country saw the first peacetime coalition government since the 1930s.
This decline (or perception of decline) in political party allegiance and effectiveness has observable effects in governance. Taking a definitive position on an issue that cuts across party lines or highlights internal party divisions can jeopardize party coherence. Because party cohesiveness is essential for effective governance,56 party divisions that result in “cross-voting” or “backbench revolts”57 threaten party leadership. It is hardly (p. 199) surprising, therefore, that political party leadership has sought ways to mitigate these threats.58 If engaging cross-cutting political issues is unavoidable, or if politicians fail to predict accurately which issues will create divisions, the obvious solution for legislators focused on re-election is to abdicate responsibility for them by deferring to other competent actors in the constitutional scheme.
In the American context, Mark Graber has argued persuasively that cross-cutting issues facilitate congressional delegation to the judiciary. If the political dynamics surrounding a particular issue will ensure that neither party can gain electoral benefit from raising or championing that cause, it is to both parties’ benefit to avoid legislating on the topic. Elected politicians thus will opt to defer to other unelected actors in the constitutional scheme, such as the court.59 Deference to the judiciary has been of more limited use in the United Kingdom, given the historically circumscribed role of the courts in a system of parliamentary sovereignty. But opportunities for judicial intervention are increasing, due to the expanding role of the British judiciary in defining rights under the Human Rights Act.60 And there are some indications that, when confronted with obvious political costs, parliamentarians have chosen to “delegate” decision-making to the courts by acquiescing in judicial determinations on some contentious issues.61
On the most threatening cross-cutting or internally divisive political issues, however, the British political parties have found another group better placed than the judiciary—at least as a matter of democratic rhetoric—to which to defer: the electorate itself. Of course, just as with deference to the judiciary, deference to the electorate through the mechanism of a referendum sits uneasily in a world of parliamentary sovereignty. As outlined above, referendums have no competing external claim to bind Parliament, because the people have no external or preexisting authority. Furthermore, without express authorization by Parliamentary Act, referendums can have only advisory relevance. (p. 200) Nevertheless, by calling for the public’s views in a referendum, a government may be able to “circumnavigate the veto capacity of other actors” (such as recalcitrant backbenchers) in Parliament and externalize the costs of policy formation.62 And notwithstanding the complicated constitutional implications that referendums present, the ostensible benefits to party politics have driven this move to direct democracy.
The Rise of Referendum Politics
The referendum—a direct vote by the public on a particular issue—is anomalous in the British constitutional scheme. The only serious proposal to integrate referendums into legislative proceedings was roundly rejected in 1910. It came at a time of parliamentary constitutional crisis, after the House of Lords had asserted itself against the House of Commons on a budgetary matter (an area left, by convention, to the Commons). Politicians were advocating stripping the Lords of their veto power, and an inter-party constitutional conference sought to find ways to mediate. One suggestion was to recognize “a special category of ‘constitutional’ legislation,” which would “require the approval of the people, as well as Parliament, in a referendum.”63 The prime minister, H.H. Asquith, did not agree. In addition to identifying the threat referendums posed to parliamentary sovereignty, his critique of the plan highlighted the difficulty of identifying “constitutional” legislation in a system with a flexible and uncodified constitution.64 Given this unpropitious history, referendums were “commonly said to be unconstitutional.”65
When referendums re-emerged as a political tool in the 1970s, the topics they addressed were of serious importance to the country—the United Kingdom’s relationship with Europe and the devolution of power to the regions, specifically to Scotland and Wales. But the surrounding political discourse was decidedly not about legitimizing constitutional change through popular sovereignty; rather it was about managing internal party division and externalizing costs, or insulating the government from responsibility for certain of its policies.66
(p. 201) The country’s first nationwide referendum was born out of the internal struggles in the Labour Party over Europe; it was, in the words of James Callaghan, “a rubber life raft into which the party may one day have to climb.”67 The Conservative Party had orchestrated Britain’s successful entry to the European Economic Community in 1972, with an effective date of January 1, 1973. At the time, the Labour Party’s formal position was opposition to entry, but the party was functionally divided.68 In an attempt to maintain party cohesion, Labour Leader Harold Wilson launched the life raft, holding a referendum on joining the EEC in 1975. Skewed toward the status quo, the referendum question—Do you think that the United Kingdom should stay in the European Community (the Common Market)?—passed with 67.2 percent of the vote, on 64.0 percent turnout.69
Although the 1975 referendum was thought to be a “one-off departure from constitutional practice,”70 necessitated by the highly unusual transfer of sovereignty that joining Europe required, “once the precedent of the referendum had been conceded, it was difficult to prevent it[s] being invoked again.”71 Other contentious and complicated issues required addressing, particularly the mechanisms of governance within the regions of United Kingdom. Advisory regional referendums were held in 1973 in Northern Ireland on the peace process,72 and in 1979 in Scotland and Wales on devolution.
These later votes contributed to a developing micropolitics of referendums. The 1975 vote had showcased the advantages of a simple question reflecting the status quo, and the 1979 devolution referendums demonstrated that procedural mechanisms could be deployed to engineer the ultimate result. The ad hoc nature of the referendum process lent itself to manipulation; because any individual vote’s governing procedures were outlined in the very Act authorizing the referendum, politicians could play with process to try to influence substance. For example, in the Scottish referendum, on a turnout of 63.6 percent, the Scots voted 51.6 percent in favor of devolution to 48.4 percent against. But the referendum failed.73 The enabling Act had incorporated a threshold requirement, designed to be a poison pill: for devolution to occur, 40 percent of the total electorate (p. 202) had to vote yes. The Scottish referendum’s majority did not meet the threshold requirement: only 32.8 percent of the total electorate voted yes.74 Given the lower turnout expected in referendum voting, this overlay was intended to put a thumb on the scale in favor of the unitary state.
By the mid- to late-1990s, politicians and activists had identified a wider range of political uses for the referendum. For some, referendums served as a rallying cry for political mobilization and a way to force issues onto the national agenda. Backbencher frustration with the Conservative Government under John Major, which had implemented the Treaty on European Union (Maastricht Treaty), led to fissures within the Conservative Party. And in 1995, the short-lived Referendum Party was founded by Tory Sir James Goldsmith as a single-issue party dedicated to a referendum on European integration. The United Kingdom Independence Party (UKIP) eventually eclipsed the Referendum Party, co-opting its single-issue agenda. This pressure likely contributed to the Conservative Party’s expedient decision, in 1996, to include an election manifesto commitment to a confirmation referendum on European monetary union—if a future Conservative Government should legislate to join.75
For others, the referendum was an attractive way of removing issues from party politics: triangulation with the added benefit of showcasing a commitment to democracy.76 In its 1997 election manifesto, New Labour committed to referendums on an assortment of major constitutional issues,77 making a vote for Labour a vote for direct democracy and allowing the party to seek (and win) the median voter by promising the electorate a second bite at the policy apple. Once in office, the Labour Party did hold referendums on those subjects that, for the most part, garnered clear majorities in line with Labour Party objectives, such as devolution to the regions and to the Greater London Authority,78 but (p. 203) it did not give voters their promised say on changes to the electoral system or joining the single European currency.79 Nor did Labour promise or hold referendums on other sweeping constitutional reforms, such as the Human Rights Act (1998), the major alterations made to the House of Lords in 1999 (curtailing the hereditary peerage), or the Constitutional Reform Act (2005) (creating a Supreme Court of the United Kingdom). All of these transformative Acts followed the usual parliamentary processes of party government.80
Perhaps because of the divergent rationales for engaging in referendum politics, the use of referendums did not regularize. No obvious or principled substantive standard existed for determining what subject matter should garner popular input. Thus, as one contemporaneous assessment of the British Constitution put it, with typical understatement, the constitutional role of the referendum was “uncertain.”81
The Blair Government did introduce some measures for the procedural regulation of referendums. The Political Parties, Elections and Referendums Act (2000) (PPERA) created an Electoral Commission and authorized it to regulate referendums. For any bill that provides for a referendum and specifies the wording of the question to be posed to the electorate, the Electoral Commission must “consider the wording of the referendum question and publish a statement of any views of the Commission as to the intelligibility of that question.”82 There is no obligation to take the Commission’s advice, but in practice, politicians have been willing to respond to the Commission’s suggestions.83 Beyond this advisory function, the Commission’s authority lies in meting out public funding for referendum campaigns: It determines which participants in a referendum campaign can be designated to receive public funds and it administers those grants.84 Although these (p. 204) powers may have bite in certain circumstances, PPERA provides only weak limits on governments.85
By 2010, the referendum was a ready weapon to deploy to deflect a range of political problems. It is to no surprise, therefore, that the United Kingdom’s second national referendum emerged out of the political morass that marked the 2010 coalition government negotiations. Although initially proposed by the Labour Party to woo the Liberal Democrats,86 a commitment to a referendum on a new parliamentary electoral system to replace FPTP—the “Alternative Vote (AV) Referendum”87—became a plank of the eventual coalition agreement between the Conservative Party and the Liberal Democrats. The agreement did not require either party to commit to anything other than calling for the referendum88—a wise limitation for the Conservatives, given that a majority of that party was opposed. Whatever its benefits to cementing the coalition, the referendum did not interest the public. On only 42 percent turnout, 67.9 percent voted against changing FPTP. The whole affair was later described as a “bad-tempered and ill-advised public debate.”89
Although a typical example of expediency, the AV referendum did mark an important shift in broader referendum politics. In proposing the Bill authorizing the referendum,90 Nick Clegg explained: “There are members of the Government who hold contrasting views on these systems. Come the referendum, there will be those of us who campaign on different sides. We emphatically agree, however, that the final decision should be made not by us, but by the British people.”91 To that end, the Bill proposed that the referendum result be binding. This blatant abdication of parliamentary duty to make substantive policy decisions in the name of popular sovereignty failed to persuade the House of Lords. Members of that House argued against the use of binding referendums and (unsuccessfully) proposed amendments to make the AV referendum advisory or to add threshold requirements and other means of controlling the possible end result.92
(p. 205) Parliament’s decision to legislate for a binding AV referendum reflected some respect for the legal distinction between binding and advisory referendums. The practical or functional import of that distinction, however, was likely already eroding. No party leadership had ever called for a nonbinding referendum that it couldn’t (or didn’t) win; thus, no government had experience in handling a nonbinding referendum that went against it. A major regional referendum could have put the issue squarely in front of the Coalition Government: the 2014 Scottish Independence Referendum.93 That referendum was not binding on Parliament, providing political flexibility (and legal cover) to the national government should the Scottish people vote to leave the United Kingdom. Ultimately, the vote was in favor of Union.94 But what would have happened had Scotland voted for independence? How would Prime Minister Cameron have responded? Could a referendum alone be allowed to radically transform the constitutional system?95
The emergence of referendum politics in the late twentieth century can be seen as a shift away from parliamentary sovereignty, but expedient, haphazard, and untheorized direct democracy only obliquely hinted at true popular sovereignty. Crucial questions abounded and went unresolved: On what topics should referendums be called? When should they be called? How should they be called? What are the procedural mechanisms for calling referendums? Who can vote? Should there be numerical thresholds? Should there be regional thresholds? Will the result of a particular referendum be binding on Parliament? Can a referendum result be overturned? It may have been that, had answers to these questions been made clear, Cameron would not have agreed to a referendum on Europe or the result may have been different, in light of complicated constitutional trade-offs that would have required resolution. But no one (aside from academics) really wanted to lift the lid on that constitutional Pandora’s box. Now, in the aftermath of Brexit, the questions are demanding answers, and the box is wide open.
European integration has always been a subject of ambivalence in the United Kingdom.96 Stephen George’s 1990 book title, An Awkward Partner, nicely encapsulates the UK’s relationship to Europe, then as now.97 This uncertainty over Britain’s place in Europe is reflected clearly in the history of referendum politics. Whether the seeds for Brexit were sown in 1975 during the first referendum on European membership, in 1995 with the rise of the single-issue parties advocating for a referendum on European membership, or during the Blair Government when referendums on dividing and sharing legislative power were provided in the devolution context, it had long been a strong possibility that there would be a referendum on continued membership in the EU.
At the close of the polls on June 23, 2016, 17.4 million voters (51.9 percent) supported Leave, and 16.1 million (48.1 percent) voted to Remain in the European Union. Since the vote, many commentators have sliced and diced that roughly 4 percent margin. There is a general consensus that long-standing British skepticism of European integration combined with economic anxiety and nativist fears to produce a convergence of interests resulting in Leave.98 But this merger of interests does not make Brexit a mass populist movement. From the outset, David Cameron’s decision to provide this in/out referendum had more to do with affluent Eurosceptics and sovereigntist parliamentary backbenchers threatening the cohesion of the Conservative Party than it did populist pressure. At most, Brexit was a populist moment, facilitated by party weakness and expedient referendum politics.
To support this point, this Section evaluates a selection of Cameron’s choices and highlights the broader influences of referendum politics on Brexit political maneuvering.
Calling the Referendum
The Coalition government of 2010–2015 faced the “most rebellious Parliament of the post-war era,” with backbench dissents affecting roughly 35 percent of Commons (p. 207) divisions.99 For Prime Minister David Cameron, the “running sore” was the issue of EU membership.100 In a speech in January 2013, he announced that, should the Conservative Party gain a majority at the next election, he would commit to holding an in/out referendum on Europe no later than 2017, following a renegotiation of the terms of British membership. Cameron hoped this promised referendum would silence his own backbenchers and neutralize the threat to the party posed by UKIP.101 The risk seemed limited: opinion polls showed the likelihood of another hung Parliament, and a coalition with the pro-Europe Liberal Democrats would prevent him from “delivering” on his promise.102
But the May 2015 election was one of surprises. Ostensibly, it appeared a return to normalcy, as the Conservatives regained single-party government with a slim majority. But the details of the vote showed political churn and demonstrated the serious weakness of the major parties.103 The Conservative Party did win an overall majority of twelve seats, but with just 36.9 percent of the vote—a margin at the very edge of the usual percentages thought to secure leadership.104 And notwithstanding Cameron’s referendum ploy, the Conservatives did suffer at UKIP’s hands; although it secured only one seat in Parliament, UKIP garnered a tremendous 12.6 percent of the vote. Cameron was hamstrung: his weak parliamentary position and the strong showing for UKIP in the national (p. 208) vote reinforced the Eurosceptic Conservative backbenchers and forced a referendum that he (and the country at large) may not have wanted.105
Preparing the Bill
Shortly after the June 2015 election, Cameron introduced a bill calling for a referendum on EU membership, making good on his commitment to an in/out vote,106 but doing so before he had gone to the EU to “renegotiate” British membership. That decision, coupled with Cameron’s request to “his own party to reserve judgment on whether to support remaining in or leaving the Union until after he had completed the renegotiations,”107 meant that those in favor of remaining in Europe were forced to bide their time waiting for details of the new terms of membership. For hardcore Eurosceptics, however, no amount of negotiations would change the calculus to leave, and thus “for many of them, there was no point in waiting” to begin campaigning to exit, which they did.108
The new Cameron Government also quickly confronted the challenges of navigating referendum politics in an interconnected scheme of parliamentary governance. Actions that would have secured advantageous procedures (and results) for the referendum were unavailable given their broader political consequences. Two such examples include providing vote thresholds for the regions and lowering the minimum voting age. Incorporating a Scottish threshold requirement for the EU referendum might well have ensured a Remain victory, and an ex ante rule incorporating a quasi-federal aspect to national referendums would make sense in a country with ongoing devolution and meaningful power at the regional level.109 But the Conservative Party was opposed to further autonomy for Scotland, making any concession difficult,110 and the concept of thresholds was redolent of the manipulative referendum policymaking that occurred in the 1970s. The strategic calculation would have been obvious and inflammatory to the backbenchers—voters in Scotland were known to be far more pro-European than those in other regions of Britain. Requiring Scottish support would have ensured a Remain victory.
Regarding the voting age, members of the House of Lords sought to amend the Bill to allow sixteen- and seventeen-year olds to vote.111 Making up almost 2.9 percent of the (p. 209) population,112 these younger voters were thought to favor Remain. But they were also likely Labour supporters. Once again, the Cameron Government was in a bind: “Agreeing to give the vote to 16- and 17-year-olds in the EU referendum [could] have meant it would be necessary to make the same concession in the general election.”113 The Cameron Government thus fought the effort by the Lords; the Commons rejected the extension of the franchise, and the House of Lords decided not to press the issue.114 In the referendum vote, over 60 percent of people over sixty-five voted Leave; and roughly 70 percent of the eighteen- to twenty-four-year olds voted Remain.115 The higher turnout of the older people amplified the Leave vote on generational grounds; it is uncertain how much of a shift adding the sixteen- and seventeen-year-olds would have made, but it clearly would have narrowed the gap.
Increased political savvy surrounding referendum politics certainly constrained government flexibility, and the procedures under PPERA ended up having a limiting effect. The status quo phrasing of the referendum question proposed by the government, similar to the 1975 wording and likely skewed in favor of Remain—Should the United Kingdom remain a member of the European Union?—was critiqued by the Electoral Commission, which concluded that the question was not sufficiently balanced.116 The government deferred to the Commission, and the question became: “Should the United Kingdom remain a member of the European Union or leave the European Union?”117 Polling conducted by Leave organizations in advance of the confirmed question indicated that shifting to an “either/or” question from a status quo “yes/no” could provide as much as a four-percentage-point swing toward Leave.118
Running the Campaign
Turning to the vote itself: as Labour had done in 1975, Cameron waived cabinet responsibility, allowing cabinet members to argue against the Government’s position on Europe. Michael Gove, the Justice Secretary, was therefore unleashed to campaign for Leave, and he “gave the anti-EU campaign a breath of appeal and a credibility that its predecessor in 1975 had lacked and which could not be provided by the leader of UKIP on his own.”119 (p. 210) He became a co-chairman of Vote Leave, and provided a “cogent case for leaving the EU,”120 without resorting to the anti-immigrant bigotry or populist nationalism of the UKIP campaign. Along with Gove, charismatic Tory Boris Johnson also joined Leave, giving additional star wattage and political persuasiveness to the campaign.121 These key defections contributed to the split in the party: in the eventual vote, Conservatives were 55 percent for Remain, and 45 percent for Leave.122
Given this divide within the Conservative Party, the Remain campaign affirmatively required cross-party collaboration. Labour Party engagement was essential.123 But yet another choice by Cameron undermined the effectiveness of the Remain campaign on this score: the timing of the referendum itself. To avoid ongoing internal party divisions, Cameron wanted resolution on Europe before the fall 2016 Conservative Party Conference, and so he scheduled the referendum for June.124 This decision not only complicated his own timing in concluding the negotiations with the European Member States but also fell afoul of party politics. Elections in the regions were scheduled for May 5, and the need to contest these elections along multiple party lines “constrained the willingness of politicians on both sides of the debate to appear together at events,” in particular limiting the engagement of Labour Remain politicians.125
Perhaps because he did not think he (or Remain) would lose, Cameron did not propose, nor did Parliament pass, any legislation that would go into effect contingent on the result.126 The Referendum Act itself treated the vote as merely advisory. Once the Leave vote was tallied, therefore, it was unclear as a practical matter whether or how the United Kingdom was to go about withdrawing from the European Union. The resignation of David Cameron and the selection of Theresa May as Leader committed the Conservative Government to carry through on “Leave.” But the very next question—could the Conservative Government, on its own and without parliamentary legislation, trigger Article 50 of the Lisbon Treaty to effectuate Brexit?—led to its own mini-constitutional crisis, complete with litigation. The constitutional issue underlying the question—do the (p. 211) well-established prerogative powers of the Crown to enter into and to withdraw from treaties apply to Article 50, or does a formal notice require parliamentary legislation127—is one that might never have been asked absent the referendum. The UK Supreme Court concluded legislation was required, forcing a showdown between the people and Parliament. Notwithstanding the advisory nature of the referendum, there was tremendous political pressure on parliamentarians to treat the result as binding (thus answering the theoretical question posed at the outset of this chapter). The ultimate result was “an event without precedent” in parliamentary history, as Members of Parliament were “required [by party politics] to vote for a policy which most of them oppose.”128
The Opportunity for Constitutional Renewal
The aftermath of Brexit has called national attention to the vagaries of referendum politics and is forcing a reckoning with the obfuscations, weaknesses, and uncertainties in British constitutionalism. The immediate concern is with Brexit itself. In the wake of the referendum, both major parties are now internally divided.129 The cross-cutting nature of the issue means that neither the Conservatives nor the Labour Party wants to re-campaign on Europe; both parties sidelined the European issue for the 2017 elections by treating the 2016 referendum as binding and final. By avoiding Europe, this approach appeared to strengthen the two-party system: Single-issue UKIP voters reverted to their original party affiliations, and even the regional parties garnered fewer seats.130 But this may be the calm before the storm. Recent polls suggest that half of voters are in favor of a second referendum, with 34 percent opposed and 16 percent uncertain.131 How long will or can party politics prevent the accommodation of public demand for a second poll? Pressure for another vote will only increase the demand for clarity on long-standing issues. If the people are to be sovereign, on what questions? When and how?
A new independent commission examining the role and conduct of referendums is collecting evidence for a report due later in 2018.132 The commission has its work cut out for it. What would referendum “reform” look like? Brexit and its aftermath present an obvious issue: if the “people” want a vote and the parties do not, what public threshold might trigger a vote? Could there be one? There are currently no mechanisms for popular (p. 212) initiative. A procedural mechanism unconnected to substance might allow any number of things to be put to a popular vote; it is unclear how such a system would work, however. Referendums remain nonbinding under the principle of parliamentary sovereignty; on important issues, political parties take note and appear to feel constrained. But if referendums could be called on any number of topics not necessarily of “constitutional” importance,133 would the parties continue to feel constrained, or would they perceive the votes as opinion polls to be appreciated but not always followed? If one Parliament cannot bind another, how can any of this institutionalization happen through parliamentary statute?134
If, as now, only the government can call referendums through an Act of Parliament, then how can governments be reined in to prevent the expedient politicking of the political parties? Are the parties likely to agree ex ante on the substantive areas that should require popular voice? To the extent that referendums are about preserving or gaining political power, parties will be unlikely to constrain themselves in advance, given uncertainty about which future issues might be cross-cutting or threatening to party cohesion.135 Even if a limitation on “constitutional” issues were to be agreed upon, the constitution itself is far from clear. In 2010, the House of Lords Select Committee on the Constitution sought to identify substantive principles for future use of the referendum. But it found “no unanimity amongst witnesses about whether certain questions should require a referendum.”136 As Vernon Bogdanor summarized, “[a]n elastic constitution, so it seems, implies an elastic use of the referendum. But this gives rise to a problem. . . . if use of the referendum lies at the discretion of government, it can be used to augment the power of government rather than limit it, by allowing a government to bring the (p. 213) people into play against Parliament. . . . [It becomes] a tactical device, ‘the Pontius Pilate’ of British politics.”137
Regularizing referendums within the current constitutional landscape seems nearly impossible. An attempt to provide a principled rule has focused on regional devolution-related referendums,138 perhaps reflecting the more generalized (and global) approach to holding referendums in instances where a “major reallocation of sovereign rights” is contemplated—in other words, when distributing (or redistributing) territorially bounded policy autonomy.139 But British referendums have not heretofore been limited to this principle, and thus no constitutional convention yet exists for constraining future governments in this way.
Constitutional scholar Robert Hazell has declared that “referendums are now an established part of our democracy, and there is no going back.”140 But this declaration dramatically overstates the situation. Referendums have not been properly integrated into the current constitutional consensus, and it is unlikely that they will be able to be so absorbed. Perhaps it is more precise to argue that the people themselves, outside of Parliament, are now being trusted with occasional constitutional decision-making, and that this shift has staying power.
If popular sovereignty is becoming an established part of British democracy, securing the people’s role will eventually require a written constitution. Once ratified by the people, a written constitution could later be amended through any number of possible mechanisms. Whether such amendment would incorporate direct democracy is uncertain; the populist undertones of Brexit may well dampen British enthusiasm for further unmediated demands for democratic input. But if referendums are contemplated,141 a written constitution could create enforceable answers to the many procedural and substantive questions listed above.
(p. 214) In a recent inquiry, the House of Commons Political and Constitutional Reform Committee canvassed the public, politicians, lawyers, and scholars on the issue of a written constitution. Those advocating in favor claimed it had “become too easy for governments to implement political and constitutional reforms to suit their own political convenience.”142 But the committee report concluded that historical tradition outweighed these threats of expediency.143 Brexit may have finally changed this calculus. History has not been kind to Pontius Pilate.
* I am grateful for helpful comments from the co-editors of the volume, as well as from Emily Kadens, Travis Lenkner, Julie Smith, and the members of the Northwestern Pritzker Zodiac Group. Michael Gajewsky provided excellent research assistance. This chapter draws in part on work previously published in Erin F. Delaney, “Judiciary Rising: Constitutional Change in the United Kingdom,” Northwestern University Law Review 108 (2014): 548–553, and Erin F. Delaney, “Stability in Flexibility: A British Lens on Constitutional Success,” in Assessing Constitutional Performance, eds. Thomas Ginsburg and Aziz Huq (Cambridge: Cambridge University Press, 2016), 393, 397.
1 Note, for example, the first paragraph of the Introduction to this volume. Mark A. Graber, Sanford Levinson, and Mark Tushnet, “Constitutional Democracy in Crisis? Introduction,” in Constitutional Democracy in Crisis?, ed. Mark A. Graber, Sanford Levinson, and Mark Tushnet (New York, Oxford University Press, 2018), 1.
2 The use of the word “referendums” is preferred in the United Kingdom. “Referendum” is a Latin gerund and thus has no plural. “The Latin plural gerundive referenda, meaning ‘things to be referred’, necessarily connotes a plurality of issues.” Referendums, therefore, is “logically preferable as a plural form meaning ballots on one issue.” University of College London, Constitution Unit, “Report of the Commission on the Conduct of Referendums,” 100 (n. 1) (1996), http://www.ucl.ac.uk/constitution-unit/research/electionsandreferendums/conduct-of-referendums [hereinafter Conduct of Referendums] (taking advice provided by the editors of the Oxford English Dictionary).
3 See generally Delaney, “Stability in Flexibility,” 393–420.
5 House of Lords Select Committee on the Constitution, Referendums in the United Kingdom, 20 (¶ 58) (2010), 12th Report of Session 2009–10. [hereinafter Referendums Report] (quoting Referendums Report: Minutes of Evidence Taken Before the Select Committee on the Constitution, Session 2009–2010 (undated) (“Memorandum by Peter Browning”) 112–113).
11 Dicey, Introduction, 38.
12 Ibid., 85.
13 Ibid., 87.
14 Ibid., 84.
23 See Judith Bara, Albert Weale, and Aude Biquelet, “Deliberative Democracy and the Analysis of Parliamentary Debate,” (paper, ECPR Joint Sessions, Helsinki, Finland, May 7–12, 2007). Cf. Ivor Jennings, The British Constitution, 3rd ed. (Cambridge: Cambridge University Press, 1954), 81–84 (discussing the important role of the opposition and parliamentary debate in informing policy).
24 Jennings, The British Constitution, 146.
25 Jennings, The British Constitution, 149–150.
26 This simple statement contains more layers in practice: explicating individual ministerial responsibility raises questions about the role of the civil service and the interaction between individual responsibility and the collective cabinet responsibility for government decisions.
28 Some scholars argue that the development of the cabinet system itself (with “not only executive but also legislative predominance”) fostered the growth in party cohesion in the nineteenth century. See Gary W Cox, The Efficient Secret (Cambridge: Cambridge University Press, 1987).
29 Dicey, Introduction, ci.
33 Ibid., 93.
38 The present system can cause rapid shifts in power and “lurches from the relatively extreme positions adopted by the Labour and Conservative Parties.” Leyland, The Constitution of the U.K., 113.
39 Ibid., 112.
40 Strong party discipline emerged in the late nineteenth century. Andrew C. Eggers and Arthur Spirling, “Party Cohesion in Westminster Systems: Inducements, Replacement and Discipline in the House of Commons, 1836–1910,” British Journal of Political Science 46 (2016): 567–589.
41 Clive James, “A Point of View,” BBC Radio 4, broadcast April 12, 2009, http://www.bbc.co.uk/programmes/b006qng8. Paul Webb, Tim Bale, and Paul Taggart, “Deliberative versus Parliamentary Democracy in the UK: An Experimental Study,” Sussex European Institute Working Paper No. 118 (2010), http://www.sussex.ac.uk/sei/publications/seiworkingpapers. Cf. John Hibbing and Elizabeth Theiss-Morse, Stealth Democracy: Americans’ Beliefs about How Government Should Work (Cambridge: Cambridge University Press, 2002).
42 Anthony King, Does the United Kingdom Still Have a Constitution? (London: Sweet & Maxwell, 2001) 53 (“[T]he United Kingdom’s constitution changed more between 1970 and 2000, and especially between 1997 and 2000, than during any comparable period since at least the middle of the 18th century.”). See also Delaney, “Stability in Flexibility,” 413–414.
43 See, e.g., Elizabeth Wicks, The Evolution of a Constitution: Eight Key Moments in British Constitutional History (Oxford: Hart Publishing, 2006), 137–165; N.W. Barber, “The Afterlife of Parliamentary Sovereignty,” International Journal of Constitutional Law 9, no. 1 (2011): 144–154; Adam Tucker, “Uncertainty in the Rule of Recognition and in the Doctrine of Parliamentary Sovereignty,” Oxford Journal of Legal Studies 31 (2011): 61, 72–77.
44 See, e.g., Vernon Bogdanor, The New British Constitution (Oxford: Hart Publishing, 2009), 68; Roger Masterman, The Separation of Powers in the Contemporary Constitution (Cambridge: Cambridge University Press, 2011), 48, 152.
45 David Jenkins, “Both Ends against the Middle: European Integration, Devolution, and the Sites of Sovereignty in the United Kingdom,” Temple International and Comparative Law Journal 16, no. 1 (Spring 2002): 17; Mitchell, Devolution in the UK, 15, 134–135; Johan Steyn, Democracy through Law (Burlington: Ashgate, 2004), xvi–xvii; Robert Hazell, “Reinventing the Constitution: Can the State Survive?” Public Law 1999 (1999): 86.
46 See Delaney, “Judiciary Rising.”
47 For a discussion of other institutional changes, see Delaney, “Stability in Flexibility.”
48 S.E. Finer, “The Decline of Party?,” in Parties and Democracy in Britain and America, ed. Vernon Bogdanor (New York: Praeger, 1984), 6; see also Peter Mair, “The Party System,” in The Oxford Handbook of British Politics, ed. Matthew Flinders et al. (Oxford: Oxford University Press, 2009), 285. As with other aspects of British political life, some of the perception of decline may be the cost of a harder and closer look at the institution itself—the halcyon days of history were never as bright as remembered. See Webb, “Are British Political Parties in Decline?,” 316.
49 A public perception of a “democratic deficit” is more likely to occur in systems with plurality voting, because they “exhibit the highest disparities between parties’ vote shares and parliamentary seat shares.” Sanders, “The UK’s Changing Party System,” 94. In the United Kingdom, a “single party receiving between 40 per cent and 45 per cent of the national vote stands a good chance of gaining an overall majority of seats in the House of Commons and therefore of forming a government.” Leyland, The Constitution of the U.K., 110 (noting that Labour achieved an overall majority of 180 seats with just under 44 percent of the popular vote in 1997).
51 Matters of moral disagreement might also present this problem. But conscience issues, which usually “arise as a result of private members’ bills,” are often treated as “matters for a free vote in parliament in part because opinions on how to resolve them cut across party lines and are of special significance to parliamentarians as individuals.” Albert Weale, Aude Bicquelet, and Judith Bara, “Debating Abortion, Deliberative Reciprocity and Parliamentary Advocacy,” Political Studies 60 (2012): 646.
52 See Peter Mair, “The Party System,” 292–294 (discussing cleavages and decline in electoral alignments). Cf. Webb, “Are British Political Parties in Decline?,” 316.
57 Adrian Blau, “Majoritarianism under Pressure: The Electoral and Party Systems,” in Constitutional Futures Revisited, ed. Robert Hazell (New York: Palgrave Macmillan, 2008), 234; see also Philip Crowley and Mark Stuart, “Backbench Rebellion in the House of Commons, 1997–2010: Making a Policy Difference or Barking at the Moon?,” 2012, http://webpages.dcu.ie/~leg/Cowley.pdf.
58 Of course, when there is sufficient electoral churn, a broader party realignment may occur, recalibrating the system. The historical example in Britain is the replacement of the Liberal Party by the Labour Party in 1931, the culmination of a decade-long process. Chris Cook and John Stevenson, A History of British Elections since 1689 (London: Routledge, 2014), 143–149. Given the current political landscape, it would appear that a major realignment could (and perhaps should) be in order, but it is unlikely. See generally Sanders, “The UK’s Changing Party System.” See also Peter Mair, “The Party System,” 295–298. As David Sanders has said, “Conservatives do not have a history of splitting and Labour activists and MPs have a visceral fear (if not hatred) of it.” Sanders, 118. And, in any event, the ability of a meaningful third party to challenge the hegemony does not happen overnight; it takes time. McLean, “England Does Not Love Coalitions,” 9 (“During a Duvergerian tipping point, it is not clear to the average voter which party is the most effective challenge to the hegemon, and the strong two-party effect is not reinstated until the answer to that question is clear.”).
60 On the Constitutional Reform Act, 2005, c. 4 and the changes to the judiciary, see Roger Masterman, The Separation of Powers; see also Erin F. Delaney, “Searching for Constitutional Meaning in Institutional Design: The Debate over Judicial Appointments in the United Kingdom,” International Journal of Constitutional Law 14, no. 3 (July 2016): 752.
61 See Delaney, “Judiciary Rising,” 590–594.
63 Vernon Bogdanor, “Conclusion,” in The British Constitution in the Twentieth Century, ed. Vernon Bogdanor (Oxford: Oxford University Press, 2003), 691. Other suggestions included using the referendum “as a regular part of government, to be used on issues where the two Houses reached deadlock and when 200 MPs petitioned for one.” Conduct of Referendums, 19 (¶ 21) (1996).
64 Bogdanor, “Conclusion,” 691.
65 Conduct of Referendums, 14 (¶ 1). The Report notes that some proposals “resurfaced periodically, including the suggestion by Winston Churchill in May 1945 that [a referendum] be used to extend the term of the wartime coalition Government.” Conduct of Referendums, 19 (¶ 21).
66 Cf. Vernon Bogdanor, The People and the Party System (Cambridge: Cambridge University Press, 1981), 69 (regarding the referendum, “the urge towards popular participation or self-government has not played a very important part in its advocacy”).
71 Bogdanor, “Conclusion,” 696.
72 The first regional referendum (the “Northern Ireland Border Poll”) was held in Northern Ireland in 1973, during the heart of The Troubles and after the suspension of the Stormont Parliament. Northern Ireland (Border Poll) Act, 1972, c. 77 and Northern Ireland (Temporary Provisions) Act, 1972, c. 22. The question—whether Northern Ireland should remain a part of the United Kingdom or join the Irish Republic—passed overwhelming in favor of remaining in the United Kingdom. But the turnout was only 58.7 percent and the poll was boycotted by Irish nationalists. Conduct of Referendums, 20 (¶ 23).
73 The Welsh referendum was a clearer case. On 58.8 percent turnout, an overwhelming 79.7 percent voted no. Richard Dewdney, House of Commons Library, Research Paper 97/113, Results of Devolution Referendums (1979 & 1997) 9–10 (1997).
77 Referendums Report, 7 (¶ 3) (including “the adoption of the European single currency; the adoption of a new electoral system for the House of Commons; the establishment of a devolved Scottish parliament; the establishment of a devolved Welsh Assembly; the establishment of a Greater London Authority; and the establishment of Elected Regional Assemblies”).
78 In 1997, Scotland and Wales voted on devolution, prior to the introduction of devolution legislation in Parliament, in regional referendums that sought a simple majority position with no threshold requirements. These referendums were not binding on Parliament, but the favorable results, particularly strong in Scotland, eased the way for devolution legislation in the Scotland Act (1998) and the Government of Wales Act (1998). With a turnout of 60.2 percent, 73.4 percent voted in favor of the establishment of a Scottish Parliament. In Wales, the voters were more evenly divided—on a 50.1 percent turnout, 50.3 percent voted for devolution and 49.7 percent voted against. Feargal McGuinness et al., House of Commons Library, Research Paper 12/43, UK Election Statistics: 1918–2012, at 51–53 (2012). Similarly, in May 1998, the Blair Government held the promised referendum in Northern Ireland on the Good Friday Agreement, a referendum that, in combination with one held in the Republic of Ireland, cemented the Agreement and allowed for the passage of the Northern Ireland Act (1998). And finally, Londoners were given a chance to weigh in on their local government, providing a referendum on the creation of the Greater London Authority; 72 percent of the voters were in favor, but only on a paltry 34 percent turnout. Referendums Report, 10.
80 Some have questioned whether the Blair Government’s choice to “programme” certain bills through standing committees breached the convention that first-class constitutional bills should not be sent to standing committees. See Paul Seaward and Paul Silk, “The House of Commons,” in The British Constitution in the Twentieth Century, ed. Vernon Bogdanor (Oxford: Oxford University Press, 2003), 160.
83 See Electoral Commission, “The 2004 North East regional assembly and local government referendums,” ¶¶ 2.13–2.17 (2005) (“The Commission’s role is wholly advisory but the Government adopted many of the Commission’s recommendations to make the questions more intelligible.); Electoral Commission, “Referendum on membership of the European Union: Assessment of the Electoral Commission on the proposed referendum question,” at ¶¶ 2.2–2.10 (2015). The Electoral Commission’s suggested wording was, in fact, the wording that appeared on the ballot for Brexit.
86 BBC, “MPs Back Referendum on Voting System,” February 9, 2010, http://news.bbc.co.uk/2/hi/uk_news/politics/8505255.stm.
87 Under the proposed alternative vote system, voters would rank the candidates in order of preference instead of selecting just one candidate. If a candidate received a majority of the number-one slot votes, that candidate was the winner. If no candidate received a majority of the votes, then the candidate with the fewest votes was eliminated. The votes of the voters who selected the candidate with the fewest votes as their first choice then had their second choice evaluated. This process continued until a candidate received a majority of the votes and was declared the winner.
93 While Westminster had to authorize the referendum to take place, The Scotland Act 1998 (Modification of Schedule 5) Order 2013, it was the Scottish Parliament that ran the referendum and set its terms under the Scottish Independence Referendum Act, 2013, asp 14.
94 In response to the ballot question, “Should Scotland be an independent country?,” 2,001,926 (55.30 percent) voted no and 1,617,989 (44.70 percent) voted yes. “Scottish Referendum: Scotland Votes ‘No’ to Independence,” BBC News, September 19, 2014, http://www.bbc.com/news/uk-scotland-29270441.
95 Note that in California, for example, the initiative and referendum procedures to enact constitutional amendments cannot sidestep legislative approval if the amendment would be radically transformative. An amendment that is tantamount to a constitutional revision requires a two-thirds vote of both houses of the California legislature before it is submitted to voters. Cal. Const. art. XVIII, § 2. And in Canada, the Supreme Court of Canada was unwilling to allow Quebecois secession to go forward without a mechanism of national negotiation. Reference re Secession of Quebec,  2 S.C.R. 217, 220–222 (“[S]ecession of a province ‘under the Constitution’ could not be achieved unilaterally, that is, without principled negotiation with other participants in Confederation within the existing constitutional framework.”).
96 See John Curtice, “Why Leave Won the UK’s EU Referendum,” Journal of Common Market Studies 52, no. S1 (2017): 20–24. Curtice surveys Eurobarometer polls from 1992 to 2016 asking how individuals from the UK identified. In spring 2016, 60 percent of those asked said they were British and denied being European—exactly in line with the average since 1992. British Social Attitudes polls found only one-eighth of British people pick “European” as an identity. Nowhere else in Europe do so few voters acknowledge some kind of European identity.
98 Leave was supported by a core group of affluent Eurosceptics (roughly 23 percent of the population); they voted at high numbers. It is true that they would not have succeeded without the help of two other distinct social groups: those who are economically deprived/anti-immigrant and the older working classes (Labour). Kirby Swales, Understanding the Leave Vote (London: NatCen Social Research, 2016), 27.
99 See Philip Cowley, “The Most Rebellious Parliament of the Post War Era,” Political Insight Blog (Political Studies Association), March 28, 2015, https://www.psa.ac.uk/insight-plus/blog/most-rebellious-parliament-post-war-era. If after a voice vote in the House of Commons, the result is uncertain, a division of the House may be requested. During the division, the MPs will physically walk into different lobbies in the House to demonstrate whether they are voting “aye” or “no.” Mark Sanford, Divisions in the House of Commons: House of Commons Background Paper (London: House of Commons Library, August 2013) SN/PC/06401, http://researchbriefings.parliament.uk/ResearchBriefing/Summary/SN06401, 5.
100 Julie Smith, “David Cameron’s EU Renegotiation and Referendum Pledge: A Case of Déjà Vu?,” British Politics 11, no. 3 (2016): 328. For example, in 2014, a letter was sent to Cameron, “signed by 95 Conservative MPs (out of a total of 303), demanding a bill to give the Westminster Parliament a veto on European legislation. All signatories to the letter must have known that this was incompatible with remaining in the EU under the terms of the British 1972 European Communities Act.” It was an act of grandstanding to send a message. Nathaniel Copsey and Tim Haughton, “Farewell Britannia?,” Journal of Common Market Studies 52, no. S1 (2014): 78.
101 See Smith, Déjà Vu?, 329 (“[I]t meant that the sceptics on his backbenches who had beleaguered his premiership until that time would mostly remain silent until after the general election.”); see also Copsey and Haughton, “Farewell Britannia?,” 75 (backbenchers); Curtice, “Why Leave Won,” 25 (UKIP).
102 Julie Smith, “Gambling on Europe: David Cameron and the 2016 Referendum,” British Politics 13, no. 1 (2018): 4. ; see also Curtice, “Why Leave Won,” 25.
103 The Labour Party won only 30 percent of the vote, hurt by the success of the Scottish National Party (SNP), which took fifty-six of the fifty-nine Scottish seats, becoming the third largest party in Parliament on 4.7 percent of the vote. The Liberal Democrats were decimated, retaining only eight seats and wining roughly 8 percent of the vote.
104 Arend Lijphart’s formula suggests a 35 percent notional effective threshold for the United Kingdom. Arend Lijphart, “Democracies: Forms, Performance, and Constitutional Engineering,” European Journal of Political Research 25, no. 1 (1994); Mair, “The Party System,” 285.
105 Smith, “Gambling,” 14 (arguing that there was no widespread popular demand for a referendum).
106 See Smith, “Déjà Vu?,” 333.
107 Ibid., 330.
108 Ibid., 337.
110 Cf. Katrine Bussey and Jon Stone, “David Cameron Says He Could Give Scotland More Powers after Meeting with Nicola Sturgeon,” The Independent, May 15, 2015, http://www.independent.co.uk/news/uk/politics/david-cameron-considering-giving-scotland-more-powers-after-meeting-with-nicola-sturgeon-10253335.html.
111 George Parker, “Cameron Defeated as Lords Vote to Extend Age Limit in EU Poll,” Financial Times, November 18, 2015, https://www.ft.com/content/3b365be6-8e1d-11e5-a549-b89a1dfede9b; Patrick Wintour, “Lords Reject Attempt to Lower EU Referendum Voting Age to 16,” Guardian (UK), December 14, 2015, https://www.theguardian.com/politics/2015/dec/14/lords-reject--lower-eu-referendum-voting-age-16.
115 Tony Helm, “EU Referendum: Youth Turnout Almost Twice as High as First Thought,” The Observer, July 10, 2016, https://www.theguardian.com/politics/2016/jul/09/young-people-referendum-turnout-brexit-twice-as-high.
118 Jessica Elgot, “EU Referendum’s Reworded Question Welcomed by Experts and Campaigners,” Guardian (UK), September 1, 2015, https://www.theguardian.com/politics/2015/sep/01/eu-referendums-reworded-question-welcomed-by-experts-and-campaigners.
119 Curtice, “Why Leave Won,” 25.
120 Smith, “Gambling,” 7.
121 Curtice, “Why Leave Won,” 25 (“Polling certainly persistently suggested that voters were more inclined to believe what Mr. Johnson said about Brexit than they were the utterances of any other politician, including the Prime Minister.”).
122 Ibid., 25.
123 The Labour Party itself shares responsibility for the result. Party leader Jeremy Corbyn was notoriously ambivalent on Europe, having voted against joining in 1975. And when Gisela Stuart, a Labour MP, became co-chair with Gove of Vote Leave, her high-profile position added to a skew in the impressions of the electorate. More than a quarter of voters believed the Labour Party to be divided on Europe, when, in fact, the Parliamentary Labour Party was 96 percent in favor of Remain. Swales, Understanding the Leave Vote, 21.
124 Smith, “Gambling,” 5.
125 Curtice, “Why Leave Won,” 26.
130 Smith, “Gambling,” 13.
131 Dan Roberts, “Brexit: Britons Favour Second Referendum by 16-Point Margin–Poll,” Guardian (UK), January 26, 2018, https://www.theguardian.com/politics/2018/jan/26/britons-favour-second-referendum-brexit-icm-poll.
132 For more information, see Independent Commission on Referendums, http://www.ucl.ac.uk/constitution-unit/research/electionsandreferendums/icreferendums.
133 Popular initiatives in states in the United States have covered such topics as condom usage in adult films (California Proposition 60, 2016, https://oag.ca.gov/initiatives/search?populate=adult+films) and marijuana legalization (California Proposition 64, 2016, https://oag.ca.gov/system/files/initiatives/pdfs/15-0103%20%28Marijuana%29_1.pdf).
134 Efforts to “constitutionalize” certain acts by fiat have not been respected. For example, the Fixed-term Parliaments Act 2011, c. 14, requires that elections shall occur every fifth year, § 1(3), unless the House of Commons passes a no-confidence motion or a motion, by a two-thirds majority, that early elections shall be held, § 2. Nevertheless, Theresa May unilaterally announced elections without seeking such a motion first. Peter Walker, Rowena Mason, and Jessica Elgot, “Snap Elections and the Fixed-Term Parliaments Act: What Happens Next?,” Guardian (UK), April 18, 2017, https://www.theguardian.com/politics/2017/apr/18/what-is-the-fixed-term-parliaments-act. The House of Commons did pass such a motion after the fact. Heather Stewart and Anushka Asthana, “Theresa May Wins Commons Backing for 8 June General Election,” Guardian (UK), April 19, 2017, https://www.theguardian.com/politics/2017/apr/19/theresa-may-wins-commons-backing-8-june-general-election.
135 See Referendums Report: Minutes of Evidence Taken Before the Select Committee on the Constitution, Session 2009–2010 (February 10, 2010) (“Memorandum by the [Brown] Government”): 94 (“The decision as to whether or not a referendum should be held should be made on a case-by-case basis. We do not believe that an objective test could be established as to the circumstances in which a referendum should and should not be held.”).
138 Six of the nine referendums analyzed by the Referendums Report concerned devolution and Northern Ireland. Referendums Report, 9–10. Since the compilation of that Report, an additional Welsh devolution referendum was called in 2011, National Assembly for Wales Referendum (Assembly Act Provisions) (Referendum Question, Date of Referendum Etc.) Order 2010, http://www.legislation.gov.uk/uksi/2010/2837/pdfs/uksi_20102837_en.pdf, and the Scottish Independence Referendum was called in 2014, which might be considered an extreme devolution measure. Scottish Independence Referendum Act 2014, asp 14, http://www.legislation.gov.uk/asp/2013/14/pdfs/asp_20130014_en.pdf.
139 Micha Germann and Fernando Mendez, “Contested Sovereignty: Mapping Referendums on the Reallocation of Sovereign Authority over Time and Space,” (paper, ECPR General Conference, Glasgow, UK, September 3–6, 2014), 7.
142 House of Commons Political and Constitutional Reform Committee, A New Magna Carta?, 19 (¶ 58) (2014), 2nd Report of Session 2014–15 (summarizing the debate). See also Delaney, “Stability in Flexibility.”