UK Referendums

UK referendums have been used much more frequently in recent years, although their use is controversial and not settled. The 2016 Brexit referendum, which resulted in a vote to leave the EU, has led to a lot of controversy about their format, justification, advisability of use, and constitutionality.

Constitutional issues

In the UK referendums are not yet a settled feature of Britain’s unwritten constitution. There have only been three UK-wide votes, in 1975 over whether to stay in the EEC (as the EU then was), in 2011 over whether to adopt the Alternative Vote electoral system, and the 2016 vote to leave or remain a member of the EU. Given the absence of a formal constitution-amending format, much is reliant on convention. To this extent the role of referendums is still unsettled.

Under the UK constitution, Parliament is sovereign. What Parliament enacts, it can repeal. No Parliament can bind its successor. For a referendum to be held, legislation by Parliament is needed.

Moreover referendums are generally advisory, not binding. Both votes on the EU, in 1975 and 2016, were of this kind. In both cases government and Parliament have chosen to respect the result. This is probably more due to practical politics, since to go against the votes was seen as unthinkable and the latter likely to cause unrest.

This could suggest that a novel doctrine is in the process of emerging, that of popular sovereignty. Certainly demands to “respect the result” and “the people have voted” by the Leave camps fits this picture. However many commentators have said otherwise. Yet, given the many questions left unresolved by the vote, another referendum on the eventual agreement, if any, with the EU could be very possible.

Why hold them?

Generally, in UK-wide terms such votes have been to settle a constitutional issue, eg EEC/EU and the voting system. More locally, they have been used to resolve changes in regional, local, and national governments, the last-mentioned being Scottish, Welsh and Northern Irish devolution. Thus, for example, one referendum was held in 1979 on Scottish devolution, which failed to reached the threshold to succeed, a requirement that the approval at the referendum be by 40% of Scotland’s total registered electorate, rather than by a majority (in fact 32.9% voted in favour, although it was a narrow numerical majority). A second was held in 1997, successfully this time, without the threshold. The third was in 2014 for Scottish Independence, which was lost 44.7% – 55.3%.

Arguably, another key reason for holding such votes has been that the governing party or parties are split, and a referendum would be a way to resolve the conflict. In 1975 it was the Labour Party who were divided over the EEC, in 2011 the Coalition and 2016 the Conservatives.


A referendum is different to an election in that a question is posed usually requiring one or more answers. The area in which the referendum is being held is, in effect, one constituency. The Brexit vote of 2016 covered the whole of the UK and votes were returned, for administrative reasons, by local authority area and not Parliamentary constituency.

One key issue that has arisen is the question that is being asked. In the recent referendums, efforts were made by the independent Electoral Commission to ensure that a neutral rather than a biased or loaded question was asked.

However, in the case of the 2016 referendum, the question asked, “Should the United Kingdom remain a member of the European Union or leave the European Union?” gave a vote in favour of Leave, but left the government with huge unanswered questions about what form Leave would take and how it would be implemented, eg should there be a “hard” or “soft” Brexit? Thus a seemingly innocent question arguably caused more problems than there were solutions. In this case, it could be said that this was a complex policy issue for which Parliament and general elections are more suited to resolve.


The fact that referendums are being used more frequently perhaps tells us something about the issue of consent referred to in this blog. That people are more often saying the the system of government is too remote and unresponsive to “the people’s” needs, that the UK is governed by a “political class”, a liberal, metropolitan, cosmopolitan and wealthy elite, suggests that something needs to be done to re-connect “the people” with government. Thus the referendum has been mooted as one such device.

However, it has in the past been argued that referendums were the devices favoured by dictators and conservative politicians, in the general sense, to legitimise their actions. Thus Louis Napoleon used the plebiscite for key measures during the French Second Empire, 1851-70). Hilter used the plebiscite to unpick the restrictions of the Treaty of Versailles and to legitimise his invasions of neighbouring territories. It was also a favoured device of General De Gaulle in the early years of the Fifth Republic. Putin followed up his seizure of the Crimea in 2014 with a plebiscite which gave a 96.77% vote, we are told, to rejoin the Russian Federation on a 83.1% turnout.

Others have said that referendums tend to vote against change, but as we’ve seen, one cannot now be confident of that being valid.

Another view is that, to have referendums as part of a constitution, as for example for a process of amendment, there should be some degree of a threshold of victory before a change is validated. In the case of the Brexit vote, the margin of victory was very small and yet huge changes have come in its train. The danger with such votes is that the minority, which ever way the vote went, will feel cheated. In this way consent can be undermined in that a big minority can present a potential stumbling block in the implementation of change. In the case of the Brexit vote there has been plenty of calls for some degree of modication of Theresa May’s “Brexit means Brexit”.

To read more:

The 2016 UK referendum poses important questions for democracy

The Independent Commission on Referendums

%d bloggers like this: