A Post-Mortem: Social Sciences and Brexit

Perhaps there was a little bit of stumbling around in the dark before the vote?

The 2016 Brexit referendum — how was that for a case study in social science impact? On active display for months, the sciences of politics, economic activity and (to some extent) history and culture had the chance to make a difference to practice – taking that as a core definition of impact. A full-scale evaluation of the mobilization of social science knowledge during the campaign awaits. These are preliminary notes.

This article by David Walker originally appeared on the LSE Impact of Social Sciences blog as “A post-mortem reflection: The impact of the social sciences on the Brexit referendum outcome” and is reposted under the Creative Commons license (CC BY 3.0).

My notes start by acknowledging the stentorian efforts that were made to marshal available knowledge and promote (informed?) debate, notably the LSE Brexit blog and the ESRC-supported UK in a Changing Europe program led by Anand Menon. Individual scholars and researchers were on social media 24/7. Social scientists took to media in their droves, criss-crossing the lines between belief and science, projection and prediction, analysis and opinion.

Their work included taxonomy, exposition, description, empirically-grounded propositions (for example about the effect of EU migration on wage levels) and fact checking. This was elucidation. At different levels they fed the torrent but often, it seemed, enjoyed no epistemological privilege and struggled to be heard in the din. Enlightenment (we learnt, and not for the first time) is not immanent; it has to bedelivered and social scientists often lack both the appetite and the tools for addressing public ignorance and indifference.

So was the result evidence of absence of impact? No, because causal propositions linking information, attitudes and ballot box behavior are sparse. Here was further confirmation that impact assessment is fraught, non-linear and contingent. And what about negative impact  – an operational definition of which might be rejection of social science testimony by a government minister. Co-convenor of the Vote Leave campaign Michael Gove (Oxford educated) did not deny experts possessed knowledge; he merely rejected it.

Let’s start the post mortem by noting:

UK social science did not see this coming. It can claim a mass of antecedent studies…of populism, racism, age and class, along with empirical investigation of migration, analysis of labor market conditions and trends in household and employment income after the recession and so on. But studies generally remained locked inside disciplines, unsynthesized, atomized. Individuals such as Colin Hay did play Cassandra but, generally, specialists in the phenomena underpinning the referendum turned out to be as surprised as everyone else by both the shape of the campaign and its conclusion.

A (positivist) thought experiment. Say the social science disciplines had come together in February this year and made a probabilistic proposition along the lines: there is a 75 per cent chance that the UK will vote to leave and from that follow propositions about Scotland’s membership of the UK and/or recession and subsequent GDP growth. Such a prediction would have done wonders for social science’s public reputation and, of course, affected the conduct and result of the referendum.

Some disciplines did more of the heavy lifting during the campaign, notably economics. Richard Layard organised a demarche by Nobel prizewinners  – leaving readers to spot those missing and speculate why. Among academics John Van Reenen was on the barricades day and night; similarly think-tankersJonathan Portes and Paul Johnson. They now appear most taken-aback by rejection of their analyses.

Polling specialists notably John Curtice naturally had a field day, within the limits of survey costs and methods; the narrowness of the eventual gap between remain and leave is well within sampling error. But academics proved no more successful than commercial polling companies in calling it for remain.

Other social sciences, notably sociology and geography, had less to say, even when the discourse ran to class, culture and regional variation. ‘Sociological research has not engaged in any depth’ said the invitation to Southampton University conference taking place a week before the vote.

Individual researchers expressed their preferences readily enough, prompting the question whether a scholar’s normativity trumps a non-expert’s. Is a historian’s prediction of the UK entering the bright, sunny uplands after exit any more wishful thinking than anyone else’s – especially when that historian is strongly partisan? When learned professors charge the referendum process and by implication their fellow citizens with stupidity is the judgement that of science or rancor? Are the subjective judgments of social scientists epistemologically richer than anyone else’s?

Social scientists have material interests, after all. They may not have realised it, but they are unmistakably part of the despised ‘elite’. Academics penned an open letter in the Telegraph and vice chancellors and Fellows of the Royal Society published their views. But this was just advocacy, the strength of which depended on public acceptance that diminishing higher education would damage the UK; instead many citizens seemed to buy the know-nothing tone of the leave campaign. Shouldn’t the ‘official’ representatives of social science have been a bit more knowing, more clued up about public attitudes more…sociologically informed?

After the event social scientists scored some excellent interpretative explanations, for example Will Davies. The result has led to calls for more social science research, in order to understand the situation better. But shouldn’t a precondition of any more substantive work be a determined effort to explain why our expensive structures for amassing and disseminating social science knowledge should have proven so weak in the face of a turn against reason, truthfulness and objective analysis?

This state of affairs is vital evidence about bringing evidence and analysis to bear more directly on political and policy decision-making. Simon Wren-Lewis said plaintively that we need some chairs in the public understanding of social science. It is not a new thought but the referendum surely does expose the gaps in what we know about knowledge transmission, the politics of expertise and the impaired communication of much social science work, at least as far as many of our fellow citizens go.

***

David Walker’s book Exaggerated Claims – the ESRC 50 years on is published by SAGE Publishing, the parent of Social Science Space.