The EU referendum has led to doubts about referendums as an instrument of public policy. Albert Weale suggests that the good conduct of referendums depends on the question being well defined and voters having easy access the relevant evidence. The EU referendum failed both of these tests. Future referendums should be on well defined questions and steps should be taken to provide access, in one convenient place, to the basic data necessary for votes to make a decision.

After the Brexit referendum result, many of those who think of themselves as democrats but who voted Remain are having doubts about referendums as an instrument of public policy. Some are appealing to the purely advisory status of any referendum in the UK constitution. Those who were already sceptical of the use of referendums now have their beliefs confirmed. However, one case no more makes a good argument in political theory than one swallow makes a summer. We should reflect on the Brexit referendum process, but we need to ask how we can define well justified principles governing the use and conduct of referendums in the light of that reflection.

The basic case for holding a referendum is that there are some issues that arise on the political agenda of societies that cannot realistically be handled by the normal processes of contest among political parties. Existential issues that change the standing and status of the country typically fall into this category. Extensions to the power of the European Union or secession are two obvious examples. Even in these cases, however, depending on history and tradition, referendums are not always the answer. In societies governed by strong principles of legal constitutionalism, supreme courts decide such matters, as has been true of the German Constitutional Court over successive European Union treaties. However, in political systems where supreme courts cannot play this role, referendums may be the only device available. What matters in such cases is that they should contribute to resolving the issue, at least for some years, not worsening the problem.

To meet this elementary requirement means that the question being put to the people needs to be well defined. The Brexit referendum failed this test because ‘Leave’ was never one option but several mutually inconsistent options. As a result the issue is not resolved as will be seen when the contradiction between the globalising Leavers and the little England Leavers unfolds.

Choices on whether to ratify a proposed international agreement do much better on this test. In such cases, there is a status quo against which any proposed change can be evaluated. Of course politics and current affairs are replete with uncertainty. You can never tell how far the future is going to look like the past, and your appraisal of the status quo is always going to require judgement. However, a treaty change does involve definite policy choices, even if the consequences of those choices are uncertain. Voters can know what they are voting for, unlike the Brexit referendum where Leavers could not tell if they were voting for Norway, Singapore or something in the wide space between.

Choices on international treaties may arouse strong feelings of national identity, but they also involve complex issues of evidence and inference. So the good conduct of referendums depends on how easily electors have access to the relevant evidence. During the course of the Brexit referendum, a number of web-sites, in particular the BBC and FullFact, were extremely good in providing checks on the claims made by partisans in the contest. However, in the nature of the case, fact-checking takes place in the responsive mode. It picks up the points that campaigners have introduced. Fact-checking against claims cannot, however, provide a baseline of information that is relevant to someone thinking about how to frame the issues.

In the UK there is a problem, well illustrated by the Treasury papers on the consequences of leaving. Speaking for myself, I did not find those documents hysterical, despite their being portrayed that way in the press by proponents of Leave. (I confess in the interests of full disclosure that I am a strong Remainer, so my assessment should be read in that light). A great deal of the Treasury content was descriptive, and was certainly helpful in defining what the alternative scenarios might be were the UK to leave the EU. However, the Treasury papers were seen as partisan because they stressed the risks associated with Brexit. Yet, government policy was in favour of Remain, presumably in part because of the risks identified in the Treasury papers. It is unrealistic to expect a government document not to echo the public policy of the government of the day.

To make matters worse, the voter interested in finding out the basic elements of the debate was faced by a choice between hefty tomes like that of the Treasury analysis or flimsy leaflets delivered through the post-box.

So what is the alternative?  At one time (I think this happened in 1975 but I cannot remember definitely), Penguin would have published two Penguin specials intelligently written by someone on one side and someone else on the other. They would have set out the facts and offered an interpretation of what they meant for the UK. However, in the age of social media and data overload such a solution is no longer credible. Any alternative needs to take advantage of data availability without appearing to pre-empt judgement as to what the facts and evidence mean. For these reasons you cannot rely on the campaigning web-sites to provide the information.

One possibility is for the government, or a commission established by government, to create a web-site with a ‘minimum data set’, providing access to the basic data in one convenient place. For example, it is a relevant question for anyone who wishes to cast an intelligent vote to ask what proportion of the UK’s national income is made up of foreign trade, and how foreign trade breaks down between various countries or regions. With some effort you can find the answers to these questions by an internet trawl, but you have to spend some time sorting the wheat from the chaff, and you have to know how to phrase the question.

If these data were provided in one place, with links to appropriate further data, then citizens could inform themselves much more easily. No doubt defining the contents of such a site would be contentious, involving careful line-drawing. However, there is a clear distinction between the facts and considerations that make for intelligent judgement and what the content of that judgement should be. Reasonable people can agree on the former, when they cannot agree on the latter.

Finally, there is the paradox of referendums. I have said that referendums are necessary when normal political processes, for whatever reason, cannot cope with the scale or character of the political issue in front of them. In those circumstances, politicians pass responsibility to the people.  But this does not mean that politicians should lose all sense of responsibility. The paradox of referendums is that in passing responsibility to the people, politicians must show more than usual responsibility, otherwise they risk enflaming passions that they can neither anticipate or control. By this test, UK politicians as a class failed badly in 2016. As so many are now being devoured by the revolution they initiated, perhaps their successors will learn the lesson, for it is a lesson beyond the scope of well-drafted protocols for the better conduct of our public affairs.

About the author

Albert Weale is Emeritus Professor of Political Theory and Public Policy at UCL.