Relevance and Impact for Sociology

A recent book by David Walker (Exaggerated Claims? The ESRC, 50 Years On, London: Sage Publications, 2016) has set out criticisms of sociology for its lack of relevance and engagement with contemporary policy issues. In my review of the book in Sociology (Vol. 51, 2, 2016) I have criticised his argument and established a case for both the impact of sociology and the failure of policy makers to engage with it. Reprinted here is the text of that review.

Review

David Walker has had a long career as a journalist, working on higher education and social affairs and as a leader writer on The Guardian and The Independent. He has also been a staunch supporter of social research through his various roles with the Rowntree Foundation, the National centre for Social Research, the Royal Statistical Society, and the Economic and Social Research Council (ESRC). In this book he draws on this wealth of experience for an unofficial history and critique of the ESRC and its funding of social research. He has produced an account that is both thought-provoking and frustrating, leaving this reader feeling that with friends like this we need no enemies. In arguing that the Council and the social sciences have made exaggerated claims about their relevance, he sees a role only for a highly applied and deeply empirical sociology. In developing his argument he belittles the achievements of sociology and puts the blame for its perceived failures firmly onto sociologists themselves.

Walker’s central argument is that the Social Science Research Council (now ESRC) was set up in the 1960s to serve the needs of public policy making but has, since then, failed to produce academically acceptable knowledge that is capable of addressing social and economic problems. The Council has claimed to be able to alleviate social problems but has failed to deliver on these claims. There is, I believe, some truth in the claim that the social scientists of the 1960s over-stated the capacity of their subjects to transform British society. They shared the optimism of that era and assumed that a state that sought progressive social change would welcome the knowledge that they could produce. This optimism was soon seen to be unfounded and social scientists began to make more modest and realistic claims. The early overstatements, however, had fuelled the Thatcherite reaction that led top attacks on both sociology and the Research Council. David Walker takes a different view of the period since the 1960s but ends up with a position that shares much with the views of Sir Keith Joseph in the 1980s.

The Research Council, Walker argues, failed to address policy requirements because it was ‘captured’ by self-serving academics who prevented it from challenging the idea that universities should decide the direction of scientific research and so ensured that the Council could not deliver on its claims to relevance. The Council, then, has simply facilitated ‘autistic’ self-determining academic work and supported established criteria of academic autonomy and excellence instead of reshaping social scientific activity to meet policy requirements. Academic influence on ESRC led it to abdicate from what Walker sees as its only legitimate purpose of financing applied research.

An alternative interpretation, however, might be that the Council abandoned the unrealistic idea of being a policy-relevant knowledge generator and began to act as a proper Research Council, to act as the guardian and promoter of autonomous and objective academic knowledge. However, this view would be equally as misleading as that of Walker. The truth is more complicated. Anyone who has had dealings with SSRC and ESRC will know that the Council has moved ever further away from autonomously determined ‘responsive mode’ funding and has, of necessity, been compliant with government requirements. It is nevertheless important, I would contend, to recognise, at the very least, that academic influence within ESRC has neither ‘captured’ its leading committees nor shifted its strategy away from policy concerns.

While Walker holds that the only worthwhile social research is applied social research, he is rather confused about the ways in which it can be applied and so about its impact and relevance. When criticising social research he subscribes to a very narrow version of the impact model. Social science, he says, must generate knowledge that is directly—and rapidly—translated into specific policy outcomes. On the other hand, however, he goes on to criticise the REF for being based on precisely this ‘simplistic’ linear model of research impact. Indeed, he inexplicably blames this feature of the REF on the academics who sat on its panels. What he clearly ought to have known, as a journalist writing on Higher Education, is that the linear model was imposed on the REF by HEFCE as the transmission agent for BIS and Treasury expectations. The linear model of impact was strongly, but unsuccessfully, challenged by academic panellists during the lengthy consultation period in which the REF impact criteria were finalised. The engineers who designed more humane mechanisms of execution were not responsible for the death penalty itself, and the world would not have been a better place if they had left politicians to use cruder and more barbarous ways of executing their prisoners. Walker’s claim that academics were not the appropriate people to judge the impact of their research is especially contentious. Walker must presumably have forgotten that he was himself one of the user representatives recruited to the sociology panel to undertake the assessment of impact case studies.

Criticisms of the linear model lead Walker to propose what he calls a more ‘interactive’ understanding of policy dynamics. The linear model assumes that policy makers are active seekers after evidence that will improve their policies and that academics are the producers of this evidence. A dynamic model, he says, would recognise the realities of the policy process, in which university research is simply one source of evidence amongst many and in which academics have to be pro-active in producing the kind of evidence required by policy makers. Despite advocating this alternative view, he notes that we have little or no actual knowledge about the policy process, though he may have found some indication in the numerous sociological studies in science and technology and the policy process. This apparent lack of knowledge does not stop Walker from speculating about actual policy relationships and placing all the blame for any lack of research uptake on the side of the academics. Walker’s view is that universities—as the ‘owners’ of the research produced—must actively sell it in the policy arena. At present, he holds, they do not do this and show little interest in doing it. The problem with his position, of course, is that policy makers do not seem to be prepared to seek out relevant research or to listen to researchers. If it is the duty of academics to disseminate their research, then surely it is the duty of policy makers to listen to them and to take research seriously. Of course, policy makers show little interest in doing this.

Walker recognises that governments do not privilege academic research when compiling evidence for particular policies. Rather than seeing this as a failure on the part of policy makers, however, he sees it as a sign of the ‘common sense’ of governments in drawing on evidence from wherever they can find it. There is, of course, a counter position to this. If policy makers select the ‘evidence’ that accords with their expectations and preferences rather than themselves taking an objective approach shouldn’t this be seen as a sign of irresponsibility on their part? I do not mean to say that university research—whether supported by the ESRC or not—is necessarily more objective than that of the other bodies that carry out research. It is simply that there are good reasons why much of the ‘evidence’ promoted so actively to government by commercial, lobbying, or PR organisations should not be regarded as on a par with university research. If civil servants ‘value general expertise as much, or more, than they do specific research’ (p.7), then this should be seen as a damning criticism of civil servants and not as a criticism of academics. Walker’s view, however, is that policy makers can do no wrong and it is all the fault of the academics. He assumes that politicians and other ‘users’ are impatiently waiting in their offices to receive the essential evidence, neatly digested, that will allow them to hone their policies and practices into tools of perfection. Academics, he says, will not give them this evidence and insist on publishing their research only in academic journals that are inaccessible to the policy makers.

Well, of course, these journals are just as easily available to policy makers as they are to anyone else—like you and me—who pays their subscriptions to the commercial publishers. They are even able to buy the research monographs in which much of our research is published. The open access requirements that have been forced on universities now remove even these minor obstacles (and place greater financial burdens on universities and researchers), so research is clearly and easily available to anyone who chooses to look for it. Walker, however, makes the further point that social science journals publish only for other academics, and in a language that is inaccessible to lay persons, and so are intellectually inaccessible to users.

This is simply the old canard propagated by those who don’t bother to look at social science journals. No one seriously makes this claim about journal publication in the natural sciences, and these subjects have not been notably lacking in applications. Developments in genetic engineering and DNA finger printing, to take just one area, have not been held back by the fact that the underlying research is published in academic journals in a technical language that any lay person finds difficult, if not impossible, to understand. In these areas users have made the effort—willingly—to engage with scientists in order to garner the evidence and information that they need, or that they didn’t even realise that they needed until they encountered the scientific research.

There is, of course, a valid and important argument that many publications, in natural science and social science, are badly written and unnecessarily dense, even for other scientists. I would be one of the first to advocate an improvement in writing skills and the need for greater clarity. However, I also recognise that the use of a technical language and complex methodology is as necessary in social science as in any other science. Policy makers and other potential users of this research must engage with sociological research on the same basis as they would engage with research in high energy physics, biochemistry, or molecular biology. Academic social scientists cannot take all the blame for the failure of users to engage with them.

A great deal of social research has obvious and very direct policy relevance. In criticising the work of sociologists, Walker ignores the fact that the great bulk of research proceeds without any funding from the ESRC but still has policy relevance. This can be easily illustrated from a cursory perusal of articles published in Sociology in the last two years, some financed by ESRC or other bodies and some unfunded. These articles cover a wide range of areas and relate to crucially important policy matters. They include: McCulloch’s use of longitudinal data to explore the factors affecting the decline in voluntary association membership; Botterill’s work on the factors influencing Polish migration to the UK; Harries’s work on the everyday disavowal of racist views and how racist behaviour can be challenged; Brooks’s examination of how bar staff responses to young females experiencing drink spiking can increase the likelihood of sexual attacks on them; Fletcher’s demonstration that policies on the avoidance of junk foods in school meals have encouraged the emergence of ‘black market’ junk food supplies among pupils; Ozaki and Shaw’s examination of how technologies aimed at energy reduction through the introduction of smart meters fail to take account of the conflicting demands placed on families by the various needs of their members to wash, cook, eat, relax, and work; Bolton and Wibberley’s work on the commercialisation and outsourcing of social care for the elderly and the ways in which its organisation as a labour process transforms the nature of the care provided; Morris and Anderson’s analysis of the ways in which social media amplify exaggerated forms of masculinity and homophobia within youth culture; and so it goes on. The evaluation of impact in the REF, in which Walker participated, showed that much research of this kind does have impact, both in terms of the linear model and in critically engaging policy debates and outcomes. It could, of course, have more influence, but as I have argued, this is as much the fault of the users as it may be of the sociologists.

Walker is especially critical of academic autonomy as a principle for deciding what is worth researching. He holds that government, as the primary user of social science research, has the right to demand that policy-relevant research be carried out and that social scientists have a duty to accede to such demands. Engaging in public debate or showing that government policy is unsupported by the evidence is not seen as a legitimate task for social science. Many of us would recognise, however, the ways in which governments have denied and even suppressed research on such topics as health inequalities and sexual behaviour, and the ways in which social scientists have used alternative channels to show the failings of government policy. Furthermore, Walker simply does not see any role for any social scientific knowledge that is not directly geared to the needs of government or some other user body and argues that curiosity-driven research leaves ‘gaps’ in areas where policy-relevant knowledge is badly needed.

The argument that curiosity-driven research leaves gaps may well be true, of course, yet the sole example that he gives barely supports the claim. This example is drawn from Dahrendorf’s complaint in 1975 that sociologists have all-but ignored power and Walker extends this to complain that there are no studies of the ‘super-rich’. Perhaps I can be forgiven an autobiographical reflection at this point. I began a project on ‘Elites and Power’ in 1973, looking at wealth ownership, corporate control, and political power. I received funding for this research from the SSRC, and later from the ESRC, I have published a fair number of books and papers from this work, and these have received a degree of attention in the press, radio, and television. I might add that the management academic Prem Sikka, a sociology graduate, has been assiduously documenting the use of tax avoidance schemes and tax havens among the super rich for the last twenty years. Some of his research has been supported by ESRC and it, too, has received extensive media coverage, especially in The Guardian. Rather than quoting a political scientist’s recent assertion that sociologists should ‘wake up to the problems of the super rich’, Walker himself should perhaps wake up to the research that is actually going on. There is woefully little indication in the book or its bibliography that he has actually investigated much of the research that sociologists have been doing on any topic.

I would agree with Walker that it is essential that social science contribute to policy—though where is the government demand for research on the super rich? I hope I have shown that there is a great deal of actual and potential policy relevance in published sociology. However, this is not to say that all social science should so contribute. Much research in the natural sciences and the humanities is not directly policy relevant but is curiosity driven, and social science should be no different. Theoretically informed comparative and historical research that provides a broader understanding of the world is crucially important. Such work may have policy relevance, but its importance lies in its provision of key explanatory principles that underpin curiosity-driven research and that sustain applied work. A recent example would be the hugely important work of Michael Mann on the history of inter-state conflict and its implications for patterns of inequality and oppression and which led to his explanations of ethnic cleansing in the Balkans. In a different vein is the work of Anthony Giddens on the development of the nation state and globalisation, which has led to his theoretical reflections on contemporary individuality and identity and that have been taken up in countless studies, both empirical and applied.

The importance of such work and the studies that they inform is not to be judged by practical relevance. Understandings of the social world are essential in any civilised society, in the same way as studies in history and philosophy and in fundamental natural science. To decry the necessity of such work in the name of practical relevance is precisely what, more than 100 years ago, Matthew Arnold decried as the outlook of the Philistine and that Charles Dickens caricatured as the outlook of Mr Gradgrind. It is dispiriting that David Walker has aligned himself so closely with this outlook.

A number of more specific sideswipes at sociology are made in the course of Walker’s book. Social science is criticised for being repetitive rather than cumulative, which he sees as limiting its potential impact. In making this claim, however, he downplays some of those areas in which there has been cumulative growth. One example is the impressive growth of knowledge about social mobility that Walker warmly welcomed when he attended John Goldthorpe’s lecture on this topic at the British Academy earlier this year. Indeed, Walker’s contribution to the discussion session advocated repetition rather than cumulation when he asked why more sociologists are not carrying out similar studies of mobility. More importantly, perhaps, Walker fails to recognise that cumulation is not always the most appropriate or the most useful form of research to pursue. Studies of historically and culturally specific social situations are intrinsically non-cumulative but no less valuable for that. Ethnographic studies may be descriptive, but they comprise important bodies of knowledge, nonetheless. What Walker disparagingly regards as ‘sociological snapshots’ contribute to our understanding of the world and are the raw materials for the history of the future. Evidence in support of this might be drawn from David Kynaston’s extensive use of such studies in his much-applauded histories of post-war Britain. Indeed, an editorial in The Guardian—that David Walker may have had something to do with—eulogised Kynaston’s ‘kaleidoscopic recreation’ of the past that allows the reader ‘to experience the next best thing to time travel’ (12^th September, 2014). How would such recreation be possible without those sociological snapshots? Might Walker not find out a great deal about present-day society—without any need for time travel—by reading some ‘sociological snapshots’. There are numerous examples of such snapshots in the pages of Sociology and other journals. Like many studies in history, these aim to document the way the world is and so help, in a variety of ways, to enhance human understanding and rectify the inadequate knowledge base that lies behind much policy.

Walker also reiterates the widespread view that British sociology suffers from a debilitating lack of quantitative skills. There is, of course, much to be said for the view that quantitative skills are in need of further development and upgrading. There is certainly a shortage of advanced quantitative methodological skills in British sociology, but there is no shortage of the particular skills in ‘statistics’ and ‘principles of sample surveying’ that Walker emphasises in his book. These are the mainstay of virtually every sociology degree course and figure in the Benchmark requirements for those degrees. His comment appears sustainable only because he does not appreciate that the kind of quantitative skills required in most policy relevant research are fairly basic. It is perhaps worth noting that if larger numbers of sociologists did begin to use advanced multivariate modelling in their work, they would be criticised by Walker for producing work that is inaccessible to its potential users. Sociologists, it seems, just can’t win. More significantly, Walker does not recognise the strength that sociology has in qualitative skills and the corresponding weakness of such skills in other disciplines. This strength is one of the principal reasons for trained sociologists being such attractive recruits to these other disciplines. Sociology as a discipline ‘exports’ its qualitative skills to other social sciences and to such areas as health research where they are lacking.

If stimulating controversy is the aim of an author, then David Walker has certainly achieved his aim, and spectacularly so. There is so much that needs countering that it is difficult to concentrate on the valid points that he does make. It is unfortunate that important issues have been constrained into such a narrow and restrictive view of the role of social research.

Originally posted April 21 2017.