Catching Methods Up to Concepts: Using Interpretive Methods to Study Public Deliberation

Selen Ercan (University of Canberra), Carolyn Hendriks (ANU) and Soton’s own John Boswell were last week awarded the prize for Best Article in Policy & Politics in 2017. The blog post below highlights the key messages in the paper – but you can read the full version, open access for a limited time, here.


Deliberative democracy is one of the fastest growing fields of normative political theory and empirical research. Over the past 15 years, it has expanded in at least two directions. The first expansion occurred as a result of the ‘empirical turn’ in deliberative democracy. It has seen a growing number of empirical studies on deliberative sites both within and outside of the institutions of representative democracy. The second significant expansion occurred as a result of the ‘systemic turn’ in deliberative democracy which views public deliberation as a broader communicative activity, taking place within and beyond discrete forums. For the most part, these two ‘turns’ in deliberative democracy—the empirical turn and the systemic turn—have pulled in different directions. Empirically, deliberative democrats have been increasingly fascinated with the micro-dynamics of deliberative forums, while, theoretically, the push has been to expand understandings of public deliberation beyond the forum into the public sphere. In other words the conceptual expansion has not necessarily been accompanied by a methodological expansion. Many of the tools and techniques developed to examine deliberation in structured forums are not well-suited to understanding the complexities and dynamics of entire deliberative systems. Furthermore much of the empirical research on such forums have been grounded on what Mark Bevir and Nabil Ansari label a ‘modernist’ research tradition. Derived from the natural sciences, a modernist approach to Political Science sets out to make ‘value free’ observations of the social world, subject hypotheses to empirical testing, identify causal relationships between the dependent and independent variables and, ultimately, develop generalizable laws to explain past events, or predict future ones. The limitations of this research tradition has become particularly visible as notions of public deliberation have expanded from ‘a forum’ to a ‘deliberative system’.

In our recent article ‘Studying Public Deliberation after Systemic Turn: The Crucial Role for Interpretive Research’ we argue that understanding the complex world of deliberative systems requires empirical researchers to go beyond the modernist research paradigms, and look for alternative ways of defining and studying public deliberation. A conceptual expansion without methodological expansion may easily fail to capture the uniqueness of the new concept. Considering the unique characteristics of the notion of deliberative system that sets it apart from the prevailing understandings of deliberation, we argue that interpretive research methods are particularly well suited to study the deliberative systems in practice. Interpretive research methods provide an in-depth, close-up, context-specific understandings of a phenomenon or experience that is ‘in the dark’.

A central challenge for empirical studies of deliberative systems is to identify the various components of the system and its boundaries. By drawing on existing and emerging studies we show that interpretive research can help 1) to identify and portray deliberative sites, agents and discursive elements in a deliberative system, 2) study connections and transmissions across different sites, and 3) understand the broader political context of both small-scale deliberative forums, and entire deliberative systems. We acknowledge that this list of roles that interpretive research can play in the study of deliberative systems is not definitive but it represents some of the most significant contributions that interpretive methods can make to empirical studies of deliberative systems.

Selen A. Ercan, Institute for Governance and Policy Analysis, Centre for Deliberative Democracy and Global Governance

Carolyn M. Hendriks, Australian National University, Crawford School of Public Policy

John Boswell, University of Southampton, Politics and International Relations

Interpretive Political Science Research Methods Course

Many students in political science, public policy and public administration who decide to undertake qualitative or interpretive research feel they are unqualified to do so. In particular they feel that interpretive approaches lack the type of specialised training that has become commonplace in quantitative political science.

The PSA’s Interpretive Political Science Specialist Group, in conjunction with the National Centre for Research Methods, seeks to redress this gap. Our inaugural methods course, held at the University of Southampton, 9 – 11 May 2018, will:

  • Situate the interpretive approach in relation to other ways of doing political science research by reference to the philosophical, epistemological, and methodological assumptions on which these approaches are based;
  • Provide the theoretical and analytical tools students need to design and conduct their research project;
  • Outline the toolkit of methods used by interpretive scholars to collect data, including ethnographic and interview-based methods;
  • Provide a series of standards that will both ensure results are reliable and maximise the impact of findings; and
  • Offer guidance on the norms and principles used to analyse data in an interpretive project.

Led by Southampton’s Prof. R.A.W Rhodes, the course is primarily aimed at PhD students and early career scholars of political science, public policy and public administration. It will be very hands-on, and is set up as a dialogue between the theory and practice of interpretive research. Most fundamentally, the course is organized around the participant’s own research. It does not provide a mere toolbox of analytical instruments to be applied, but will introduce participants to, and let them practice with, the approach, enhancing their skills in research design, data collection and data analysis in the process.

Information on registration, costs, bursaries and registration can be found here:

Embracing Impressionism—and Each Other?


By John Boswell and Jack Corbett. John Boswell is Lecturer in Politics at University of Southampton (@Boswell_JC). Jack Corbett is Research Fellow at Griffith University. You can find more posts by John here.

In an essay set to come out in Critical Policy Studies soon (read it here), we argue that the sort of research we practice—commonly dubbed ‘interpretive’ and famously pioneered by, among others, Soton’s own R.A.W. Rhodes—ought to fess up to, and in fact be proud of, its ‘impressionistic’ nature. The point is not to lob bombs at interpretivism from the outside as not matching up to orthodox standards of what constitutes ‘systematic’ research, but to critique it from within. We argue that in their determination to uphold the validity of interpretivism against the established orthodoxy, key pioneers of this approach have gone too far in clinging defiantly to the notion that such work is equally ‘systematic’. In contrast, many of us who practice interpretive research experience the opposite. To us, and many colleagues we’ve spoken with, doing interpretive research feels more like struggling to arrange a jumble of impressions than it resembles any systematic accretion of insights. As such, we reclaim the label ‘impressionistic’ not as something pejorative but as a more apt descriptor for doing and communicating interpretation. To do so, somewhat awkwardly for a couple of uncouth Antipodeans, we draw out an analogy to impressionist painters to show the affinities between this artistic movement and the interpretive move in politics and policy research.

So far, many readers of this blog will be thinking, so lunatic fringe – and we recognise that some may simply see this as a case of upstarts trying to outflank their interpretive forebears in terms of touchy-feeliness. But we will try and make a case here that the implications of our argument actually bring us closer to the mainstream of political and policy studies (something we have claimed more explicitly in another recent paper, here).

One – the war is over. While any such epistemological war was, it must be said, mainly in the heads of the minority interpretivists, consciously abandoning claims to ‘systemacity’ would signify an end over the battle for the one true way to gaining political insight. Just like impressionist painting, interpretive research could see itself as another way of doing research on politics—one that its practitioners find most interesting and appealing, and which is well equipped to provide unique insights into some things, but equally one that is poorly equipped to provide insights into other things. Importantly, then, impressionistic interpretive research is something that ought to be read, and done, in conjunction with other forms.

Two – we are (sort of) the same. In reflecting on the features we say make interpretive research ‘impressionistic’, we were struck by how many commonalities there are with colleagues’ experiences of actually compiling and analysing data in their more or less positivist research. A dynamic, fluid or unsettled research design; the search for a supportable claim to something novel or counter-intuitive; the selective accentuation of data which support such claims – these are all things that we have discussed at length with econometricians, political behaviouralists and others besides. More reflexive honesty about the impressionistic nature of our work opens up these unexpected affinities.

Three – we all benefit from sunlight. On the back of this last point, the key difference is that while there are widespread calls for more transparency in positivist research, few such calls are being made in interpretive research. Of course, there are some ethical and pragmatic barriers—not all data is openly available, either because it shouldn’t be or because it can’t be. But much more can often be done to open up interpretive research in this way, and allow the impressions of the researchers involved to be contested—we have one such pilot project in the works which we will later report on. For the moment, though, the point is this: impressionistic interpretivists could learn something from mainstream political scientists about making their work at least opaque, letting more sunlight in.