Richard Bellamy discusses the circumstances that would allow for a second referendum to take place. Beyond the difficulties to overcome a potential “betrayal” effect, the design of this new vote would be highly controversial and it does not seem likely that Justine Greening’s system of first and second preferences would make it more legitimate in the eyes of the electorate and politicians.  

Theresa May has announced there will be no second Brexit referendum under any circumstances, prompting the wits of social media to recall her similar assurances regarding a snap election and declare that naturally there will now be one. Trying to divine what the Prime Minister really believes or wants seems a thankless task. However, a reasonable hypothesis is that she considers such a declaration as necessary to stem accusations of betrayal by the hard Brexit wing of her party, especially given some soft Brexiters and even a few Remainers also believe the popular vote for Brexit has to be honoured, and to buttress her authority to negotiate an agreement with the EU on the basis of proposals likely to win Parliamentary support. If so, two questions arise: first, does the accusation of ‘betrayal’ in the event of a second referendum have any foundation at all and second, and probably more importantly, under what political circumstances might she (or any immediate successor) be pushed into granting a second referendum?

A surprising amount of academic and journalistic ink has already been expended on addressing the ‘betrayal’ issue and the legitimacy of a second referendum, including by myself (‘Was the Brexit referendum legitimate, and would a second one be so?’, European Political Science, on line early). I still hold to three of the earlier views I expressed in this debate: that arguments for a second referendum based on the supposed illegitimacy of the first all fail; that a majority of MPs are only likely to regard a second referendum as legitimate in the event of a clear change in the public mood; and that the time for a second referendum to be held legitimately is fast running out. The first referendum may have been very imperfect in its design and execution, with the Leave campaign breaking certain financial rules, but it still just about met a number of reasonable threshold criteria for legitimacy. For example, its conduct seems largely consistent with the recommendations for referendums recently put forward by UCL’s Independent Commission on Referendums. Although opposition to a second referendum may be hypocritical in the case of those Brexiters worried about the possibility of Bremorse on the part of the electorate, or the mere fact that demographic changes over the past two years may on their own be shifting popular opinion from Leave to Remain as elderly Leavers die and young Remainers reach voting age, many MPs do feel bound by the popular vote and the manifesto pledge of both Conservatives and Labour to honour it in the 2017 general election.

That is not to deny that a theoretical case exists for such a momentous decision, which overturns the plans and expectations of so many citizens, to be subject to sober reconsideration in the light of a fuller appreciation of its implications. In my earlier piece, I justified a second referendum on such grounds in terms of Laurence Sterne’s ‘two bed’s of justice’ argument. In his novel Tristram Shandy, Sterne has Shandy, the novel’s narrator, report, misremembering Herodotus on Persian customs, that the ancient Goths of Germany took major decisions twice: first drunk and second sober, with only those drunken resolutions that met with sober approval being acted on. Shandy remarks that his father was much taken with this argument but being teetotal adapted it to involve discussing important domestic matters with his wife on the first Saturday night of the month and then on the following Sunday morning, referring to these two deliberative occasions as his ‘beds of justice’. One might have expected the extended Brexit negotiations to have been a distinctly sobering experience. Yet, many Tory Brexiters and their supporters in the popular press seem as drunk as ever on their unexpected success in the first referendum, while opinions seem to have shifted only slightly among the electorate. Many Leave voters seem more inclined to blame the Brexit hangover on Remoaning attempts to frustrate a clean and swift Brexit than on the emerging difficulties of exiting in a way that does not do dire damage to both the British economy and the peace in Northern Ireland, at least in the short to medium term. As a result, a parliamentary majority for a second referendum still looks unlikely and time is fast running out.

What novel circumstances, then, might give the Prime Minister second thoughts? It is becoming ever clearer that finding a negotiating position that can be acceptable both to a majority in Parliament and to the EU is becoming harder by the day. Remainers and Leavers alike have harshly criticised the White Paper that emerged from Chequers, although concessions to the latter has allowed it to just scrape through with a parliamentary majority. However, those very concessions, that render it harder to avoid border controls between Northern Ireland and the Republic of Ireland, make it even less likely than many already judged to be the case that it would be acceptable to the EU. If a decision on the future framework of EU-UK relations cannot be postponed by allowing for a undetermined and probably lengthy transition period, whereby Brexit formally occurs but its full implications get deferred, then a second referendum might appeal as a way of resolving the impasse and the prospect of leaving with no deal – a possibility desired probably by only the 40 or so MPs associated with the improbably named ‘European Research Group’ (a body that, to paraphrase Voltaire on the Holy Roman Empire, is neither European, capable of doing – let alone actually undertaking – anything that could be called research, and lacking the organisation and unity typical of a group). Indeed, Justine Greening has recently proposed a second referendum on just these grounds.

A key issue would then be what question might be put to voters that would not be question-begging. After all, what would a rejection of whatever deal the Prime Minister (or the EU) proposes actually mean – that the UK should remain in the EU, that it should Leave with no deal, or that negotiations should continue and a further vote held? Justine Greening has suggested voters be offered three options – May’s final deal, no deal, or staying in the EU – with a system of first and second preference votes that would ensure the final option was preferred by more than 50% of the population. To see why both a broader choice of options and a more complicated system of counting votes might be necessary, consider a variation of her proposal that offers voters a choice between a) remaining in the EU, b) leaving and staying in the single market and customs union (roughly Norway/EEA), c) leaving and not staying in the single market and customs union (a version of which would be May’s amended Chequers deal), and d) leaving without a deal (WTO rules). Imagine voters are asked to rank these four options, saying which of a, b, c, d is their first, second, third and fourth choice. This ranking would allow us to identify the choice that has the highest preference ranking of more than 50% of voters, what is technically known as the Condorcet winner – that is, the choice which could beat all other alternatives in a pair-wise competition, representing the true majority view. (I defend the use of a Condorcet count in Richard Bellamy, ‘Majority Rule, Compromise and the Democratic Legitimacy of Referendums’, Swiss Political Science Review, forthcoming.) Assume the electorate divides in their ordering as in the table below. Option c) of leaving and not staying in the single market and customs union (ie WTO rules) is the plurality winner, that is the option with the most – but less than 50% – of first preference votes, but Option a) of remaining in the EU is the Condorcet winner – that is the option the majority of the population prefer out of all of these options based on a pair wise comparison.

My guess is that a) or b) are probably closer to what most people want than c) or d). Unless voters were sufficiently aware that voting for the first preference might lead them to end up with their worst or almost worst option, then offering a simple vote could lead to something very few people really want. Indeed, given these sort of preference orderings probably underlay the June 2016 vote, that may well be precisely what happened in the first referendum, with Remain + soft Brexit/ EEA having much higher support than hard Brexit. (Vernon Bogdanor has suggested a two-stage referendum – Leave vs Remain, and then different versions of Leave if Remain gets less than 50%. The issue by issue median can produce something like the Condorcet winner in certain circumstances, but a lot depends on agenda setting and the order in which votes get taken. For that reason, I think a Condorcet count on a single vote on the four main options is superior and more straighforward.)

Sadly, I doubt that voters will be offered such a choice – the endorsement of Greening’s proposal by both Tony Blair and Peter Mandelson might well have killed it off as yet another ruse by a scheming and discredited ‘elite’. Even if they were, Brexiters would no doubt condemn such a ‘continental’ way of counting votes as biased against them. As a result, we appear destined to end up with the least favoured options of c) or d).

Richard Bellamy is Professor of Political Science at University College, London (UCL) and Director of the Max Weber Programme at the European University Institute (EUI).

This article was originally published by the DCU Brexit Institute and is reposted with permission.

Note: The views expressed in this post are those of the author, and not of the UCL European Institute, nor of UCL.