In a few months time Scotland will vote on independence. In my last post on the topic I discussed some of the consequences of a yes vote: the problems that would be raised around the currency, Scotland’s membership of the EU, and, more generally, the difficulties presented by the tight time-fame set by the Scottish Government for negotiation. That post should have given wavering ‘yes’ voters pause for thought; the path to independence is harder and riskier than the Scottish Government’s optimistic White Paper claims. In this post I will discuss one of the consequences of a no vote: its implications for subsequent independence referendums. This post should, perhaps, cause wavering ‘no’ voters to reflect. The independence referendum is, or should be, a once in a generation chance to leave the Union. It would be a mistake to assume that a second referendum will be held any time soon.
There are problems with constitutionalising a right to secession. In a classic article, written as the states of Eastern Europe were recasting their constitutional orders in the early 1990s, Cass Sunstein argued that constitutions should not normally incorporate a right to secede. Sunstein argued that such rights inhibited the creation of a united, effective, state. The constitutional possibility of secession might encourage regions to consider independence on a regular basis, and, on the other side of the equation, the remainder of the state will be aware of secession as an ever-present possibility. As Sunstein argues, this may inhibit long-term planning: why should the state engage in projects that principally benefit the region, knowing that the region might leave at anytime? And when the project benefits the whole state, but requires regional cooperation, how can the state be sure of this support? More darkly, Sunstein warns there is a risk of blackmail. The region can use a threat of secession to put unfair pressure on the remainder of the state. Finally, as Sunstein points out – and as we have reason to know all too well – questions of secession tend to stir emotions more deeply than other political questions. The intemperate character of debate around the issue can, in itself, harm the capacity of the state to act as a coherent unit.
Sunstein’s prescription – a denial of the right to secession – is not open to the United Kingdom, which has already recognised the right of certain of its territories to leave the Union. The Northern Ireland Act 1998 contains a legal right for that territory to secede in some circumstances, and whilst Scotland and Wales lack such a legal right, it has been accepted, perhaps for quite sometime, that they are entitled to determine their own constitutional fate. After the SNP gained control of the Scottish Parliament it was a matter of when, not if, a vote on independence would be held.
But whilst Sunstein’s prescription may be inappropriate, his diagnosis remains accurate. The bare possibility of a second referendum after 2014 may have a destabilising effect on British politics for the reasons he identified. The risk of a second referendum may cause the rest of the UK to be reluctant to adopt schemes or make decisions that benefit Scotland at the expense of the remainder of the country: why buy warships from Scottish shipyards, rather than from their English competitors, when Scotland may become a separate state at any time? And, recalling Sunstein’s fear of blackmail, there is a risk that Scotland will use the threat of independence to exercise a disproportionate say over UK policy-making: agree with us, or we leave. In short, the continued possibility of independence may frame political debate within in the UK in negative and corrosive terms, with Scotland’s interests understood as distinct from, and potentially in tension with, those of the rest of the UK. There is a danger that the possibility of secession will lead to Scotland becoming a semi-detached part of the Union, always on the verge of exit.
This problem could be addressed by regulating the capacity of the Scottish Parliament to call independence referendums. Under the current devolution settlement the Scottish Parliament is able to hold an advisory referendum on independence at any time. Admittedly, this point is not beyond dispute: most notably, Adam Tomkins has argued against this view, contending that the Scottish Parliament lacks this power, but, for reasons I have set out on this blog, I think it unlikely he is correct on this point. The Scottish Parliament does, though, clearly lack the power to hold a binding referendum on independence: at present, this requires the agreement of Westminster. The status of the 2014 referendum was secured after an agreement between the Scottish and Westminster Governments. Whilst as a matter of law, the United Kingdom Parliament could still refuse to accept the outcome of the 2014 referendum, as a matter of political practice the Edinburgh Agreement is sufficient to render the vote binding.
Any attempt to regulate the holding of independence referendums after 2014 would, if the Sewel Convention were adhered to, require the support of both the Scottish and Westminster Parliaments. Conferring on the Scottish Parliament the capacity to hold a binding referendum might render the concomitant regulation of that power more attractive. The Scotland Act 1998 could be amended to legally recognise what is an existing constitutional fact: that the Scottish people have constituent power, that they possess the capacity to create a sovereign state by seceding from the United Kingdom. In addition to this, the Scottish Parliament could be accorded the power to call a binding referendum on independence. The Scottish Parliament, rather than Westminster, is best placed to determine when the Scottish people wish to hold such a vote.
Coupled with the conferral of this new power on the Scottish Parliament should come limitations on its exercise, to mitigate – if not cure – the problems that the right to secession brings. Just because the constitution accords Scotland the right to secede, it does not follow that the United Kingdom need accord the Scottish Parliament an untrammelled power to determine the procedures through which that right is exercised. It is common for the constitution of a country – determined at the level of the state – to set the conditions for secession. Having accorded a region the right to secede it would be wrong for the state, through the constitution, to limit the right in ways that make secession effectively impossible. But it would be appropriate for the state to set conditions on the secession right that serve to protect the remainder of the state’s territories and the political community of the state as a whole. With this in mind, the capacity of the Scottish Parliament to hold a referendum should be constrained in two respects.
First, there should be a constraint of the frequency of independence referendums. They should be rare: there should be a long period of time between the 2014 referendum and the next vote. The capacity of the Scottish Parliament to call a vote should, then, be time-limited. The Scottish Parliament should be given the power to call, by simple majority, an independence referendum only if (say) 30 years have elapsed since the previous vote. Making the independence vote a rare and decisive event makes it less likely that the secession right will have the destabilising consequences identified by Sunstein. The issue is taken off of the political agenda for a substantial period of time, allowing decisions to be made at the national level without being unsettled by constant doubts about Scotland’s continuing membership of the Union.
It might be objected that such a long period between votes leaves Scotland vulnerable: what if the rest of the United Kingdom embarked on a scheme so hazardous (such as resolving to leave the European Union, for example) that Scotland’s vital interests were imperilled by remaining part of the Union? Indeed, a benefit of secession rights is that they can give smaller regions some protection against larger units. The time-constraint on referendums should, then, be balanced by a second measure. The Scottish Parliament should be given the power to call a referendum at anytime by super-majority: a referendum would be held if (say) two-thirds of MSPs eligible to vote supported it. This would be a hard standard to meet, but not an impossible one; in extreme cases the Scottish Parliament could hold an independence vote before the specified time between referendums had elapsed.
In summary, my proposal is that following a ‘no’ vote the Scotland Act be amended to empower the Scottish Parliament to hold a binding referendum on independence, but only if 30 years have elapsed since the last referendum or if two-thirds of all MSPs vote for such a referendum.
There are a number of objections that might be made to this proposal. Practical-minded people I have spoken to warn me that it is unrealistic. They may well be right. If independence is rejected, the United Kingdom Government and Parliament are unlikely to have much appetite to continue to debate and discuss the issue. The SNP is unlikely to want to accept restrictions on the chance to secure a future vote – and may regard such limitations as, in themselves, constraints on a power that ought to reside in the hands of the Scottish Parliament. Consequently, each side has incentive to let the matter drop. But whilst constitutional ambiguity is sometimes desirable – allowing us to avoid unnecessary conflicts – it can sometimes store up trouble for the future. The possibility of a second referendum will ensure that, after a brief period of quiet, the question of independence will return as a live political issue. Worse still, there is a chance that it will be harder to secure agreement between Scotland and Westminster on the significance of this referendum. Westminster might, reasonably, decline to accept the validity of a second referendum held in the near future: it might be argued that the SNP cannot keep repeating the question until they get the answer they want. The period immediately after a ‘no’ vote is probably the best possible time to set the parameters under which the secession right should be exercised. Leaving it unaddressed will bring significant costs.
The recent pronouncements of the future of the currency in Scotland from British politicians have generated criticism: to some this looks like bullying, threatening the people of Scotland with the loss of the pound. Such criticism is misguided. It is right that Scottish voters are given as much information as possible about the likely outcome of a ‘yes’ vote. Part of that information is the negotiating stance that will be adopted by the rest of the UK when dealing with the putative Scottish state, a political entity that will become, it should be remembered, as much a foreign country as France or Germany. But Scottish voters ought also to reflect on the consequences of a ‘no’ vote and, ideally, British politicians should also address this question. There may well be more powers that can be devolved to the Scottish Parliament – a ‘no’ vote is not a vote against devolution – but the vote will settle the question of Scottish independence for a generation. It will – or it should – rule the question of secession out of political debate for a long period of time, and the Scotland Act should be amended to help bring about this end.
Nick Barber is Associate Professor of Constitutional Law at Oxford University, and a Fellow of Trinity College, Oxford.
Suggested citation: N. W. Barber, ‘After the Vote: Regulating Future Independence Referendums’ U.K. Const. L. Blog (21st March 2014) (available at http://ukconstitutionallaw.org).