13 October, 2014Issue 26.1EssaysThe Humanities Debate

Email This Article Print This Article

Whose Values? Which Justification?

Gabriel Roberts

Laurence Olivier as Hamlet

The last four years have seen intense debate about the value of the humanities, much of it stemming from the education policies of the current coalition government. On the teaching side, the increase of student fees to ¬£9,000 a year and the effective abolition of public funding for humanities teaching raised a pair of related questions, one about what degrees in the humanities are worth to students and the other about who should pay for them. On the research side, the overall cut in government funding for the Arts and Humanities Research Council (AHRC) by around ¬£21m (17%) between 2009 and 2013 also raised a pair of questions, one about the extent to which humanities research should be publicly funded and the other about what humanities research should be like in order to be funded in this way. The announcement and subsequent enactment of these policies provoked a broadside of books, articles, blog-posts, protests, and speeches, all aimed at explaining the humanities’ value. Among other things, we were told, the humanities promoted happiness and economic growth, were vital for the healthy functioning of democracy, and mattered for their own sake.

Partly as a result of this debate, the last four years have also seen the consolidation of a new kind of research, aimed at direct contributions to the public good, and efforts across the sector to demonstrate the public value of humanities research. In March this year, the AHRC showcased research which contributes to the creative economy, including a study of Indonesian death squads which formed the basis of the Oscar-nominated film The Act of Killing and research on the history of Channel Four which is designed to inspire the broadcasters of today. In a similar vein, a report published last year by Oxford University found that “the long-established system of humanities-based higher education in Oxford has proven highly responsive to national economic needs” and that “humanities graduates played a large and growing role in employment sectors which brought about growth in the UK economy in the 1970s and 1980s.” In each case, the emphasis was on demonstrable, quantifiable, consequentialist, and primarily economic contributions to the public good.

Yet to someone working within the humanities, the situation may have presented a different aspect. In many contexts, the discussion of the value of research was phrased almost entirely in terms of the traditional scholarly values of originality, rigour, clarity, accuracy, cogency, comprehensiveness, and the rest. In literary and intellectual history (in which I was working at the time), book reviews and article abstracts continued to proclaim that research was “subtle, important, and nuanced”, “illuminating, probing, and scrupulous”, “intimate, thorough, and attentive,” and so on. Delegates at conferences and participants in graduate seminars continued to ask questions which were designed to test the quality of research in terms of these values. The emphasis in these contexts was not on the contribution of humanities research to the public good, but on its value as scholarship.

These two kinds of value, the public and the scholarly, are formalised in the Research Excellence Framework (REF), the five-yearly survey of the quality of academic research which is being conducted this year (research has already been submitted and the judges will report in the coming months—for details, see especially their Guidance on Submissions and Panel Criteria publications). Under the terms of the REF, humanities research will be assessed according to three weighted measures, entitled “impact” (20%), “outputs” (65%), and “environment” (15%). The last of these concerns the foundations which an institution is laying for future research and, as such, can be discounted from the present discussion, whereas the remaining two measures, “outputs” and “impact”, enshrine the public and scholarly values which are invoked in other contexts. “Impact” relates to the effect which research has on the economy, society, and culture (controversially excluding staff and students at the submitting institution) and is defined in terms of “reach” and “significance”, which in turn are illustrated with examples, such as research informing policy decisions, creating cultural capital, enhancing public discourse, making money, and so on. “Outputs” relates to the quality of published research and is defined in terms of “originality”, “significance”, and “rigour”, which in turn are defined in terms of “importance”, “innovation”, “understanding”, “findings”, “imagination”, “coherence”, “precision”, “accuracy”, “analytical power”, and “depth”.

It’s worth noting here—and this is something which I’ll come back to—that the remit for impact is broad. It certainly doesn’t mean (as early critics of the REF feared) that humanities research will be assessed to any great extent in terms of its contribution to the economy. But the evidential bar is high, with the result that submitting institutions will have been better off trying to demonstrate some kinds of impact, such as footfall at an exhibition or listeners to a radio programme, than others, such as informing policy decisions or enhancing public discourse. The result is a bias towards more readily measurable kinds of good—a specific instance of a far more general problem.

The REF is also structured so that public and scholarly values will have been kept separate. Institutions were required to submit evidence of outputs (roughly four articles for each academic included in the assessment) and evidence of impact (a template describing the institution’s approach to impact together with a set of case studies), but there was no requirement that the research submitted as evidence of outputs should be the same research as was submitted as evidence of impact. Indeed, the REF requires far more research as evidence of outputs. An institution required to submit 100 articles as evidence of outputs would only have to submit four case studies, each making reference to “one or more” articles. And it also imposes more strenuous qualitative standards: whereas research submitted under the ‘outputs’ heading will be graded on a one-to-five scale (with institutions hoping to score as highly as possible), research submitted under the “impact” heading has to reach only level two on the same scale. This means that any institution which was producing enough research to avoid submitting the same research twice could have pursued outputs and impact as separate objectives. It may even have been politic for some institutions to appoint academics who were less talented than their peers at producing research and more talented at creating impact, which was not a concern before the REF. The guidelines for the next REF are yet to be published, but if the same methods of assessment are employed then we could see humanities faculties creating parallel streams of researchers and research communicators.

This may seem like a good situation or at least an understandable one. Scholarship has been protected from demands for public justification, and some academics have been able to pursue their research without worrying much about its public value—an activity which might have debilitated their ability to research at the highest level. If the value of humanities research is bound up with researchers’ ability to pursue it for its own sake—as one might think—then it may have been wise to insulate scholarly values against the clamour of public ones. There is a question here about the distribution of justificatory labour, and it may be that this labour is best performed by a minority of academics.

There are, however, a number of things which the REF leaves out of the equation. In the first place, the incentive which it provides for institutions to conduct research which has readily measurable impact, together with the separation which it imposes between research which is excellent scholarship and research which creates impact, omits the large part of humanities research which has unpredictable, long-term, or dispersed effects. The REF makes some provision for these considerations by accepting case studies of impact underpinned by research conducted as long ago as 1993, but it seems unlikely that many submitting institutions will have opted to undertake the complex data-gathering necessary to demonstrate that research conducted twenty years ago has had impact rather than to steer its researchers towards work with more readily measurable effects. Nor, looking forward, is there any incentive in the REF for academics to think about how the research which they are conducting now might create impact in the future or any guarantee that this will be relevant to the REFs of 10 or 20 years’ time.

The problem is further compounded by the sharp distinction which the REF imposes between research which doesn’t have to have an impact and research which has to have one. The results of case studies will not be used to estimate the impact of research which has been submitted under the “outputs” heading and there is no suggestion that all research might have been assessed in terms of impact and the total impact achieved by a submitting institution worked out through aggregation. In these ways, the REF directs attention away from the effects of research which is submitted under the “outputs” heading, even though a great deal of this research must create impact in some way.

There’s also a curious lack of detail about how research submitted under the “impact” heading is supposed to create its impact. The REF supplies a long list of ways in which impact can be measured, including “citations in reviews outside academic literature, independent citations in the media, including in online documents, reviews, blogs and postings”, but there’s little requirement for submitting institutions to describe precisely how impact happens. For example, an institution might submit a case study of research which underpinned an exhibition and cite attendance figures as evidence of impact. But it wouldn’t have to provide any evidence that the visitors went away having learned anything or without having acquired false beliefs or undesirable attitudes. The problem here is that questions about whether the consumers of research are correct in their interpretation of it or respond to it in desirable ways are difficult to answer in quantitative terms. Yet one might think that these are questions which the humanities should peculiarly address.

Finally, there’s a problem concerning the relationship between the values which are used to evaluate humanities research and the private reasons why humanities researchers believe their work to be of value. Because public and scholarly values are the only values in terms of which the value of humanities research is typically expressed (at least for the purposes of assessment), researchers may feel unprofessional, selfish, or inadequate insofar as the reasons which motivate them in their studies are out of step with these values. They may inadvertently acquire the kinds of interest which are compatible with these kinds of values. Or they may come to think that the value of the humanities, to them or anyone else, boils down to nothing more than excellent scholarship and impact.

Some examples may help to make this clear. An academic may write a history of political representation in order to expose the thin notions of accountability which exist in modern British politics. Another may write an analysis of Shakespeare to show that the performance of his plays promotes an unhealthy view of women. And another may write a philosophical study of rationality so as to criticise the use of the term in economics. In each case, the academic seeks a goal which extends beyond the production of excellent scholarship and which they think will make society better, but which not everyone would recognise as a contribution to the public good. After all, some people may not believe that there’s anything wrong with current notions of accountability in British politics, with the view of women promoted by performances of Shakespeare, or with the use of “rationality” in economics. Nor will these disagreements necessarily be about means rather than ends. They may arise from fundamentally different conceptions of what society should be like. Under the terms of REF, research which is controversial in this way needs to be repackaged and sold as research which excels in terms of scholarship or impact, which is problematic for several reasons. In the first place, the trouble of repackaging research may atrophy the enthusiasm of humanities researchers. But more importantly, the requirement that humanities research should contribute to the public good in ways which everyone can recognise as good may enormously restrict the kinds of research which can occur. The assumption apparently underlying this is that the best way to promote the public good via humanities research is by requiring that all humanities research should contribute to the public good in ways which everyone can recognise as good, and there is little reason to suppose that this is true.

The central question here is about how research should be assessed in order to qualify for government funding. But it also has implications for how humanists should defend themselves. It’s fair to say that in the last four years humanities academics have sought to present a united front. We’ve seen this in the willingness of academics—locally, in events hosted by the Oxford Research Centre in the Humanities (TORCH)—to emphasise that similar problems affect the humanities and other areas, such as the social sciences, the medical sciences, and the hard sciences, and we’ve seen it in the eagerness of the humanities’ defenders (see the books by Jonathan Bate, Stefan Collini, and Helen Small) to defend the humanities as a whole. The latter move is understandable, but it comes at a certain cost. If we adhere to the view that the humanities are uniformly valuable, without internal gradations between what is more and less valuable, then we may place ourselves in a bad position to make decisions about what to cut or expand in response to inevitable fluctuations in funding.

A less united front might be stronger by being more flexible. Just as one can think about the humanities in terms of who performs the labour of justification, one can also think about them in terms of the amount and explicitness of disagreement which they can tolerate about matters of value. At the moment, the tolerance level is low. Debates about the value of research are restricted to its contribution to the public good (conceived in highly restrictive terms) and to its formal scholarly properties, and this prevents more deeply-seated disagreements, about the sort of society we want to live in and what we as individuals want, from coming to the surface. In an area like the humanities, in which diversity and subjectivity are important, this is something worth worrying about.

Gabriel Roberts is a final-year D.Phil. in English Literature at Worcester College, Oxford, and a Senior Editor at the Oxonian Review.