Reflexive Engagement? Actors,Learning,and Reflexivity in Public Dialogue on Science and Technology

Introduction

In a similar way that earlier moves to enhance science communication and the public understanding of science (PUS) (e.g., Bodmer, 1985) prompted a more reflexive strand of critical PUS research (Irwin & Wynne, 1996), the recent drive toward public dialogue and upstream engagement on science and technology is undergoing its own reflexive turn. This emerging critical public engagement research agenda is focusing critical social scientific concepts back onto spaces of participation that science and technology studies (STS) scholars and other social scientists have themselves been involved in prompting or creating. It treats new scientific governance “as a legitimate object of study in itself” (Irwin, 2006, p. 310) in emphasizing the social construction of participation and “public talk.” Such work is highlighting these “technologies of elicitation” (Lezaun & Soneryd, 2007) as forms of social science and innovation in themselves that construct publics and scientific citizenship in particular ways (Braun & Schultz, 2010; Irwin, 2001), are subject to powerful framing effects (Stirling, 2008), can (re)construct deficit models of public understanding (Wynne, 2006), and be conditioned by the driving forces of political economic influence and neoliberalization (Goven, 2006). This represents a significant move beyond research strands focused on promoting and developing public engagement practices (e.g., Joss & Durant, 1995) and evaluating the effectiveness of such processes (e.g., Rowe & Frewer, 2000) that have predominated the field of public engagement with science and technology (PEST).

This article contributes to this more reflexive mode of research on PEST in two main ways. The first is to develop one of the only accounts of the actors that mediate, shape, and govern forms of public dialogue on science and technology and their roles and relations within participatory governance networks. In moving beyond isolated case-based accounts of public dialogue, this perspective allows us to develop new understandings of mediators and facilitators as a new category of expert at the science-policy-society interface that have been largely missing in the public engagement literature hitherto (see Chilvers, 2008a; Elam, Reynolds, Soneryd, Sundqvist, & Szerszynski, 2007). This is important in providing novel insights into the institutionalization, professionalization, and commercialization of participation associated with the rise of a globally connected public engagement industry (cf. Chilvers, 2010; Hendriks & Carson, 2008). The approach taken here builds on recurring interest in STS and cognate disciplines on the relationship between critical social science and policy-practice (Burgess, 2005; Webster, 2007; Wynne, 2007) and the role of social science in processes of public engagement and dialogue in particular (Burchell, 2009; Chilvers, 2010; Gisler & Schicktanz, 2009). Rather than focus on the roles of social science per se, a fuller analysis of the roles and relations between a range of different actors in public dialogue networks is developed, thus exposing the complexities of these relationships.

The second main contribution of the article is to use this mapping of mediators of public dialogue as a basis to develop new insights on learning and reflexivity in participatory governance of science and technology. Learning and reflexivity often form powerful rationales for, and expected outcomes of, public participation (Webler, Kastenholz, & Renn, 1995) and forms of participatory technology assessment (Grin & van de Graaf, 1996). Existing studies on relationships between public engagement and learning focus for the most part on how public participants, scientists, and practitioners learn and transform in discrete public participation events (e.g., Stagl, 2006; Webler et al., 1995) or use theories of learning to inform engagement process design and effectiveness (e.g., Petts, 2007). Much less attention has been paid to relationships between participation and organizational or institutional learning (although see Bickerstaff, Lorenzoni, Jones, & Pidgeon, 2010; Chess & Johnson, 2006), and less still to participatory learning in wider networks cohering around public dialogue.

Furthermore, meanings of reflexive learning—relating to Beck’s (1992) notion of “reflexive modernity” and developed through STS-inspired understandings of institutional reflexivity (Wynne, 1993)—have been explored in relation to science institutions and PUS for some time but applied to forms of public dialogue and participatory governance innovations much less so. Importantly, the forms of social science and social scientific experts that lie at the heart of these developments have been missing from earlier accounts of scientific reflexivity. The more radically reflexive meaning of learning adopted in this article is rooted in self-aware and self-reflective identification and critical examination of the precommitments and assumptions that frameone’s own knowledge-commitments and those of others (Stirling, 2006; Wynne, 1993, 2006). In exploring the nature and extent of learning from and learning about public dialogue, the article therefore goes on to analyze potentials for actors and institutions to be responsive, responsible, and accountable to the social/ethical implications, public values, and uncertainties of science and technology and of public participation itself.

While the analysis in this article focuses on U.K. public dialogue networks, they are reflective of international trends to institutionalize and professionalize public engagement with science in many countries (see Sciencewise, 2010a)—for example, networks associated with the Danish Board of Technology in Denmark, the Rathenau Institute in the Netherlands, and TA-Swiss in Switzerland—as well as transnational networks associated with deliberative democracy more generally, for example, in North America, Australasia, and international development in the “global south” (see Hendriks & Carson, 2008; Singh, 2001). In the next section, the evolution of U.K. public dialogue networks and the mapping approach used to study them is briefly outlined. Empirical analysis of the nature of these networks and the roles, relations, and purposes of actors within them is then presented, highlighting the often-problematic relationship between critical social science and policy-practice. This provides the basis to analyze dominant meanings of learning circulating within the network and the extent to which governance actors and institutions associated with public dialogue are learning about and learning from participation. Finally, the implications of these findings are considered, which point toward more critical and deliberatively reflexive participatory governance of science and technology, both in terms of research and practice.

Public Dialogue Networks

Science-public relations in Britain have undergone a well-documented shift, at least rhetorically, from an emphasis on PUS and one-way science communication before the late 1990s to public engagement with science over the past decade, including its move further “upstream” in processes of technological development. The Committee on Public Understanding of Science (CoPUS), set up following recommendations in the Bodmer report (Bodmer, 1985) and administered by the Royal Society, provided an institutional space for the early PUS agenda in promoting a more scientifically literate society. The so-called crisis of trust in scientific advice associated with controversies over BSE, genetic modification, and nuclear waste through the 1990s, and critical interventions of social scientists highlighting “deficit model” assumptions of public understanding embedded in early PUS approaches (Irwin & Wynne, 1996), prompted increasing interest in public dialogue as a form of deliberative or two-way/multiway science communication between scientists, policy makers, and publics. This switch in emphasis toward public engagement was heralded by the House of Lords Select Committee on Science and Technology’s landmark Science and Society report (House of Lords, 2000), which sparked a rapid growth in “public dialogue” on science and technology in the United Kingdom.

Although this shift has been well documented, the actors lying behind these developments in public dialogue have received much less attention. Some insight has been provided by one of the first studies of mediators and public participation experts carried out in the United Kingdom between 2001 and 2003 (see Chilvers, 2008a). A network of deliberative public engagement experts, or “experts of community” (Rose, 1999), was shown to emerge rapidly following the House of Lords (2000) report, comprising participatory practitioners and academic social scientists. These actors were adopting increasingly central and powerful roles in mediating relations between science, society, and policy, through eliciting public understandings, advising science and policy institutions, and organizing participatory experiments on their behalf. The issues in question at the time were mainly framed in terms of environmental risk, such as GM crops and radioactive waste. Communities of practice centered on “citizen” or “stakeholder” models of participation and the public were evident, associated with the development of deliberative tools and services in an emerging public dialogue industry.

A desire to learn from the handling of emerging technologies such as GM crops in Europe (Wilsdon & Willis, 2004), and reflections from social scientists that the dominant framing of burgeoning public dialogue had been reduced to downstream risks and impacts rather than deepening debate over the human needs and purposes that drive emerging science and technology (Wynne, 2002), prompted deliberation on social and ethical concerns much earlier on in the science and technology development process when its social shaping remained a possibility (Macnaghten, Kearnes, & Wynne, 2005). This coincided with official calls and commitments by U.K. science and policy institutions to move engagement “upstream” (e.g., HM Treasury, 2004; Royal Society Working Group on Nanotechnology and Nanoscience, 2004), albeit often under the instrumental guise of achieving public acceptance of emerging technologies. It is in this context that the U.K. government set up a new organization called Sciencewise in 2005 following the demise of CoPUS, and in response to the Council for Science and Technology, Science and Society Subgroup (2005) recommendation that the U.K. government should develop a “corporate memory” about how to do dialogue well. Relaunched as the Sciencewise Expert Resource Centre for Public Dialogue in Science and Innovation in 2008, Sciencewise has instigated a program of capacity building, learning, and embedding best practice in public dialogue across government and beyond. It has also cofunded and guided 17 flagship dialogue projects in the period 2005 through 2011 on issues ranging from nanotechnology and stem cell research to the use of DNA in forensics and geoengineering of climate change (see Macnaghten & Chilvers, 2012).

This article is based on research exploring contemporary public dialogue networks in the context of this relative shift in emphasis from risk governance to “innovation governance” (Felt & Wynne, 2007) and intensifying trends toward the institutionalization, professionalization, and commercialization of public dialogue. Undertaken in 2009, the study involved key actors from the U.K. public dialogue field in in-depth interviews coupled with network and documentary analysis to explore the nature and character of participatory governance networks, actor roles and relations within them, and the nature and extent of learning and reflexivity. A total of 21 interviewees were selected to reflect the diversity of positions and roles in and provide overviews of the U.K. public dialogue field. Respondents were drawn from research, public, private, and NGO/not-for-profit sectors and included (the number of respondents is given in parentheses): academic social scientists (2), participatory practitioners from small independent consultancies (2), from communications consultancies (2), and from market research companies (2), policy makers in government departments and agencies (4), representatives of participation organizations (2), think tanks (1), civil society organizations (1), science institutions (3), research funding councils (1), and a scientist involved in public dialogue (1). Interviews lasted for approximately 1 hour, were audio recorded, fully transcribed, and coded in Atlas.ti qualitative analysis software. Further details of the methodology employed in the study are given in Chilvers (2010).

Actors, Roles, Relations

Mediators and facilitators who shape public dialogue processes and science-society interactions have emerged as a new category of expert at the science-policy-society interface. They resemble the new intellectual attitude of the mediator (Osborne, 2004) as “enabler,” “catalyst,” and “broker” of knowledge associated with networked, distributed, and coproduced forms of knowledge production in the knowledge society. Analysis in this section is based on each interview respondent’s mapping of mediators involved in U.K. public dialogue networks, their roles, and relations. The emphasis was on mapping, in an unconstrained way, the range and diversity of actors and their multivalent roles and relations in a multidimensional manner rather than focus on social scientists or any one category of actor in particular. In the same way as Osborne (2004), following Bauman (1987), the intention here is to offer a descriptive account rather than an explanatory or causal framework.

A map of the main actor types identified in relation to their roles with respect to public dialogue is shown in Figure 1. A number of dialogue actor types were identified in interviewees’ talk, including participatory practitioners, academic social scientists, policy makers, scientists, think tanks, participation organizations, and civil society organizations (CSOs). In addition, four main categories of roles in relation to public dialogue, sometimes referred to as areas of expertise, were evident, namely: (a) studying, which includes researching, theorizing, evaluation, and reflection; (b) practicing, which includes designing, facilitating, and reporting on dialogue processes; (c) orchestrating, which includes commissioning, sponsoring, and guiding; and (d) coordinating, which includes networking, capacity building, and professionalization.

OPEN IN VIEWER

Figure 1.

A map of the main actor types in relation to their roles with respect to public dialogue.

Figure 1 has been derived from qualitative analysis of interviews where the data were coded in terms of these role categories for each actor-type identified. Each circle in Figure 1 is thus a qualitative representation of the range and diversity of roles taken up by each actor-type, as identified in coding analysis, rather than a quantitative representation relating to a linear scale or metric. The situation is clearly very complex. The overriding sense is that actor types and roles relating to public dialogue are ambiguous, fluid, and tend to blur in space and time. Most individual actors wear different “hats” at different times and places. Each category of actor will now be taken in turn to explore their roles and relations in more detail.

Both documentary and interview evidence showed U.K. public dialogue activity to be predominantly orchestrated by science and policy institutions, mainly from the public sector, that commission, fund, and sponsor invited spaces of engagement. Key actors in this regard included government departments and agencies, advisory bodies, and science institutions. The default position of these sponsors over the years has been to outsource dialogue expertise, which has led to an “advisory culture” of external participation experts providing advice about one-off public dialogue events in relation to key decision moments. The roles of actors in decision institutions were thus mainly located in the bottom right quadrant of Figure 1, and in some instances expanded into practicing and coordinating.

A key category of actors referred to by interviewees was participatory practitioners whose main roles centered on the design and facilitation of public dialogue processes and innovating new techniques (located in the bottom left quadrant of Figure 1), as part of a rapidly growing public engagement industry. They ranged from independent facilitators and small specialist consultancies, through to large multinational companies. Interviewees recognized different groups of practitioners cohering around particular visions of participation and the public, including those associated with an invited stake-holder or consensus-based vision (such as The Environment Council and Dialogue by Design) as opposed to an invited public or innocent citizen vision (such as social research and market research companies that had become dominant players in the U.K. public dialogue field).

Over the past half-decade a relatively new breed of actor—participation organizations—has emerged in the drive to institutionalize and professionalize public engagement. Such actors were recognized in interviews for their roles in speaking for, overseeing, coordinating, and professionalizing the public dialogue field, which included networking, knowledge transfer, developing guidance, training, and building capacity (mainly centered on the top right quadrant of Figure 1). In the British context, such organizations include Sciencewise, the National Coordinating Centre of the Beacons for Public Engagement (NCCPE), and Involve, as well as local representatives of international networks such as the International Association for Public Participation and the International Association of Facilitators.

Many forms of public dialogue on science and technology are rooted in qualitative and participatory research methods developed in the social sciences. In this sense the dialogue practitioners and consultants described above are practicing forms of applied or “regulatory social science.” However, almost all interviewees specifically referred to academic social scientists when discussing the roles of social science in relation to public dialogue, which were shown to be complex, multiple, and often ambiguous. Most interviewees recognized that over the past decade there has been a shift in emphasis from academic social scientists being actively engaged in innovating and experimenting with public dialogue practices to them taking up a more critical, analytical, and in some cases removed stance. This reflects not only the rise of critical public engagement scholarship during this period but also possible influences of prevailing norms, pressures, and incentive structures in academia. The most predominant roles of academic social scientists were thus seen to lie in the top left quadrant of Figure 1, to do with studying, analyzing, theorizing, and reflecting on public engagement, including critical and reflexive analyses of public dialogue and the social, political-economic, historical, and cultural conditions that shape science-public relations.

As noted above, scientific institutions (such as the Royal Society and the British Science Association) have been involved in commissioning and experimenting with new forms of public dialogue but have also taken up other roles in Figure 1, including networking, coordination, and capacity building. Individual scientists were identified much less in interviews, and were seen as taking up roles as expert witnesses, translators, and collaborative analysts in, rather than as organizers of, public dialogue processes. It was recognized that scientists are often involved in more informal, traditional, or one-way forms of science communication (see also Davies, 2008), which blur with definitions of public dialogue used by some interviewees.

Two further categories of actor were notable in interviews. Think tanks such as Demos, IPPR, and the New Economics Foundation, which have had a long-standing presence in the field of participatory democracy since the mid-1990s, tend to cover the center ground in Figure 1. They were seen to be taking on a diversity of roles but often not to any great degree, specializing in knowledge transfer, advocacy, and providing strategic advice to decision makers. CSOs were deemed to have important but often indirect and highly ambivalent roles in relation to public dialogue. While some CSOs had initiated and designed public dialogue processes, many lacked the resources (time and money) to do this. One of their main roles in dialogue processes noted by interviewees was as expert witnesses or providers of specialist knowledge.

This analysis shows that the popular notion of a public participation expert as “facilitator” of discursive processes who directly elicits public understandings was extended by respondents in this study to include those involved in orchestrating, studying, and coordinating public dialogue in a more distributed sense. Having said this, the centralization of resourcing for public dialogue around science and policy institutions plays a strong role in shaping relations within the network. This, alongside processes of professionalization and commercialization, has led to a relatively fixed model of public dialogue becoming locked in within the network based around one-off discrete exercises that are often extractive, linked to particular decision moments, and privilege “innocent citizens” (Irwin, 2006). It is important to highlight, however, that some interviewees—most notably academics, CSO and think tank representatives, and a small number of practitioners—emphasized alternative meanings of public dialogue associated with distributed dialogue and “uninvited” spaces of engagement (Leach, Scoones, & Wynne, 2005). Interviewees were rarely able to directly identify actors associated with these spaces, who lie beyond the professionalized and codified world of public dialogue that is largely reflected in Figure 1. This offers important evidence of the existence of, and partial opening up to, alternative lay or grassroots forms of mediation and participation expertise, which needs to be further understood and reflected on (see Chilvers, 2010).

The Social Science/Policy-Practitioner Interface

Fragmentation between groups of actors was seen to be an important characteristic of U.K. public dialogue networks. As evident in earlier studies of these networks (Chilvers, 2008a), a key disconnect highlighted by interviewees in this regard was between academic social scientists and other actors, including practitioners and policy makers. Understanding this interface offers important insights into the competing and conflicting purposes, assumptions, and commitments of the different actors identified and the roles of social science in public dialogue (see Burchell, 2009; Chilvers, 2010; Gisler & Schicktanz, 2009). Perhaps the most striking example of this disconnect related to the U.K. Economic and Social Research Council (ESRC) where academic research into PEST that the Council had funded was reportedly not being used to inform or support the operational developments in public engagement across the Research Councils and within the ESRC itself:

[T]he Research Councils have the ESRC and social scientists are experts on this but . . . there was not enough translation from the social science to get that expertise into helping other parts of the research councils and other practitioners. (Public engagement officer, science institution)

Such divisions, which were commonplace across the network, were rooted in mutual suspicion, misunderstandings, and fundamental differences between academics and practitioners as to their respective roles, motives, and intentions in relation to public dialogue. From the perspective of nonacademic interviewees, academic social scientists were often viewed as unconstructive and overly critical thus creating “a level of anxiety that . . . their practice will be rubbished” (Participatory practitioner, private consultant). Nonacademics highlighted problems of language and communication. Dense “social science speak” often made it hard for practitioners to see the relevance of social science insights (Science and society manager, science institution), and the ability to communicate PEST research findings beyond core social scientific audiences was questioned. Nonacademics also problematized the willingness of social scientists to “get their hands dirty” in engaging with public dialogue practice (Policy officer, think tank). Longer-term academic timescales and the short-term needs of practitioners were highlighted as a key point of tension here. Underlying many of these practitioner perspectives was an instrumental decision-oriented imperative where social science is expected to dance to the tune of policy agendas defined elsewhere and act as a “midwife” to deliver public dialogue and public acceptance of technoscience (cf. Burchell, 2009; Webster, 2007). But for many the role of social science remained highly ambiguous and very much an open question.

[A]re [social scientists] supposed to be a step away and kind of providing the critical friend role, or should they be more linked into the practitioner community? (Science and society manager, science institution)

Social scientists themselves often resisted the widespread instrumental framing of policy-practice claiming “we’re all focused on our critical analysis of what’s going on” (Academic social scientist). However, interviews also showed social scientists to be taking up a range of roles varying from critical analyses, shaping public dialogue processes, serving on advisory committees, and mobilizing more radical forms of participatory practice. These cut across the socio-political-moral roles that have been outlined by Gisler and Schicktanz (2009), including the “historian” (largely detached), “ironist” (critical analyst, again somewhat removed), “reformer” (a focus on improving practice within the current system), and “rebel” or revolutionist (building alternative visions of participation and challenging the system). In highlighting these possible roles, interviewees often set up a tension between two visions of social science intervention. One emphasized a radical-critical stance that can enhance reflection in the longer-term but risks being dismissed as irrelevant or naïve. The other was more supportive, pragmatic, and helpful thus holding the possibility of immediate influence while risking capture in serving the interests and instrumental motives of others. This dichotomy appears somewhat problematic however. The tendency for interviewees, and to some extent Gisler and Schicktanz’s (2009) typology, to paint the historian or critical analyst as distant or dispassionate does not give due account to the blurring of these different roles in practice, nor to the potential for the more radical cultural-historical perspective to inform and infuse with the pragmatic (see Wynne, 2007).

It is clear that social science and policy worlds relating to public dialogue on science and technology are very much two cultures each associated with different institutional contexts, motives, and timescales. Although such differences were often explained in terms of practical issues of language, timescales, and communication, which need to be taken seriously in their own right (see, e.g., Cassidy, 2008), such emphasis can also serve to mask fundamentally conflicting purposes in relation to public dialogue and the role of social science. The instrumental decision-oriented perspective that dominates policy-practice is resisted by, and in conflict with, the more critical and analytical ambition of many social scientists (see also Burchell, 2009). Rather than search for a single best response to building more constructive relations at the social science-policy/practitioner interface, interview responses suggest the need for social scientists to maintain a diversity of roles and interventions. It is clear though that this has to include space for a critical, analytical, and historical-cultural perspective. These insights into actor roles and relations in public dialogue networks are closely tied to, and hold important implications for, dynamics of learning and reflexivity, to which we now turn.

Learning and Reflexivity

This section builds on the above findings to explore the nature and extent of learning and reflexivity in U.K. dialogue networks—in terms of the ways in which actors and institutions learn about and learn from public dialogue—through analyzing the interview discourse, supported by documentary evidence. As argued in the introduction, the wealth of existing studies on relationships between participation and learning have tended to focus on individual and collective learning within discrete public engagement events (e.g., Stagl, 2006; Webler et al., 1995) and their outcomes (Bull, Petts, & Evans, 2008). Such analyses have drawn on individual learning and education theories (e.g., Mezirow, 1997) among others to distinguish between instrumental learning based on the acquisition of new skills or knowledge and communicative learning about how individuals approach a situation or point of view and cooperate with others (see Petts, 2007). Much less attention has been paid to analyses of learning at the level of organizations and wider governance networks. The few existing studies of public participation and organizational learning (e.g., Chess & Johnson, 2006) have drawn on Argyris and Schön’s (1996) organizational learning theory, which distinguishes between single-loop and double-loop learning. Single-loop learning emphasizes the instrumental acquisition of new knowledge and understandings to optimize existing strategies or goals, whereas double-loop learning implies a deeper questioning of underlying assumptions and governing conditions thus initiating changes in values and frames.

While also informed by Schön’s (1983) ideas of reflective learning, the European Commission Expert Group on Science and Governance has recently summed up the more radical and reflexive STS perspective on learning, emphasizing “reflective” and “relational” forms of learning in the wider science governance system. Here reflective learning concerns “insight into the assumptions which tacitly shape our own understandings and interactions,” whereas relational learning involves learning about the “independent integrity of others” (Felt & Wynne, 2007, p. 63). For Wynne (1993), properly reflexive learning involves the “process of identifying, and critically examining (thus rendering open to change), the basic, preanalytic assumptions that frame knowledge-commitments” (p. 324), which is intimately tied to the appreciation of the legitimacy of other sources of knowledge and the perspectives of other actors. In terms of science and policy institutions, this should not exclude reflection on their own structures, power and social relations, implicit constructions of the public, and so on.

Stirling (2006) has offered useful clarification on the meanings at play here, particularly in distinguishing between reflection and reflexivity in the governance of science and technology. Reflection means that “attention extends to the full range of whatever are held to be broadly salient attributes of the object in question” (p. 226), implying deep serious consideration of alternatives, unintended consequences, and so on. Reflexivity involves the self-awareness and self-reflection that comes from recognizing that attributes of the subject (e.g., the governance actors and mediators shaping public dialogue, including participants themselves) construct and condition the object (i.e., the particular science, technology, or issue in question). In what follows, these conceptual insights are drawn on to understand the three main forms of learning evident in U.K. public dialogue networks, as expressed in interviews.

Discussion

The following discussion draws together themes from across the above analysis and reflects on their implications in three main ways. First, the nature and character of public dialogue networks and the rise of mediators as a new category of expert at the science-policy-society interface is reflected on. Second, the predominance of instrumental imperatives and explanations for the observed lack of reflexive learning in public dialogue networks is considered in connection to the wider literature. Third, in building on and moving out from the findings of the current analysis, prospects and possibilities for more critical and deliberatively reflexive participatory governance of science and technology are explored, in terms of research and practice.

With regards to U.K. public dialogue networks and the roles of mediators, the grounded analysis presented in Figure 1 shows that particular actor types and roles can be identified—which extend beyond popular notions of the facilitator as designer and moderator of public deliberation—but that these roles are highly complex, ambiguous, and often blur in space and time. In demonstrating the complexity of actors and roles implicated in governing science and society, the mapping framework usefully highlights the coproduced nature of public dialogue and a certain “distributed coherence” evident in the coevolutionary perspective of Rip (2006). In this sense no one actor has overall influence or control over public dialogue and science governance, thus promoting a more realistic and humble perspective that tempers inflated claims about the heroic influence of social scientists, the tyrannical control exerted by facilitators and participatory practitioners, or the “impact” of any one public dialogue process that forms only part of a wider science governance system.

This is not to say that some actors do not have more influence than others, however. The centralization of resourcing for public dialogue around U.K. science and policy institutions plays a strong role in shaping relations within the network and the enunciation of assemblages that make up particular participatory experiments. A relatively fixed model or vision of public dialogue has become locked in within the network based around one-off discrete exercises that are often extractive, linked to particular decision moments, and privilege “innocent citizens” (Irwin, 2006). Existing structural relations in the U.K. dialogue network—for example, formalized commissioning processes that outsource one-off decision-oriented dialogues to consultants who then quickly move on—have served to reinforce this decisionistic pattern. Furthermore, while some interviewees in the current study acknowledged and stressed the importance of organic, spontaneous, uninvited spaces of public dialogue and associated forms of lay or distributed public participation expertise, such mediators were not readily identified on an individual or organizational basis. Revealed actors were mainly limited to a professionalized, codified, and thus visible world of public dialogue experts and expertise. Despite the social science innovation that has occurred in recent years, on this particular point processes of reflexive modernization (Beck, 1992) appear distinctly rationalist, where the “subpolitics” central to counterscientific debate has been colonized to some extent by these professionalized forms of facilitation that demarcates boundaries between expert and lay forms of mediation. This indicates a need to open up the governance of public dialogue on science and technology to wider public scrutiny, distributed or informal kinds of dialogue expertise, and proper acknowledgement of uninvited spaces of engagement that lie beyond institutional sanction and control.

This insight is supported by the above analysis of learning: the second main point of discussion. A key finding is that the forms of public dialogue and upstream engagement considered—which are reflective of other forms of anticipatory (Barben, Fisher, Selin, & Guston, 2008) and reflexive governance (Voß, Bauknecht, & Kemp, 2006)—have been exposed as not particularly reflexive, in the British context at least. This has been demonstrated, for example, through the dominant instrumentality and lack of reflexive and relational learning in relation to U.K. public dialogue. Most interview respondents highlighted divisions between actors and a disconnect between critical social science and policy-practice as a key reason for this. And while there are many possibilities for critical social science engagement, communication, translation, and collaboration at the social science-policy interface, the evidence points to deeper explanations for observed reflexivity deficits. The finding that most actors and mediators identified in relation to Figure 1 are (or can be) highly reflective on an individual basis or in private but not at the level of institutions and wider discourses on science, participation and the public suggests in a similar way to Wynne (2006, p. 213) that the ingrained “unreflexive institutional culture of science and policy itself” is a central part of the problem.

Wynne’s (1989) earlier work on the social construction of expert authority in legal settings is particularly insightful in explaining these dynamics in three important ways. First, Wynne shows how interests involved in the choice of one particular legal classification or method over another become embedded as institutional routines and cultural habits, and thus become naturalized and taken for granted rather than voluntaristic at the level of individual actors. Second, the contingent and constructed nature of science and legal expertise evident in localized legal disputes and in private-elite circles routinely becomes “translated into a more orderly account for public consumption” (p. 38). Third, Wynne highlights one of the main reasons for this “repertoire of legitimation” as being chronic concerns held by institutions and individuals about maintaining their credibility. Closing down uncertainties and masking the contingencies of knowledge production is thus a matter of institutional and political survival.

Beyond legal settings such dynamics could equally explain the tendency for the forms of (social) science and public participation expertise explored in this article to close down the purposes, politics, and uncertainties of public dialogue on science and technology. For example, Bickerstaff et al. (2010) have shown how institutional rationalities and routines of the U.K. Royal Society served to frame and narrow down both public dialogue initiatives and the possibility for organizational self-reflection about its own scientific rationality. In addition, structural relations within public dialogue networks outlined above, such as centralized resourcing and control by science policy institutions seeking to maintain their credibility and the growth of commercial relationships between client and contractor associated with the outsourcing of dialogue, create conditions that actively discourage openness vis-à-vis the downsides, exclusions, interests, and power relations inherent in public dialogue and science-society relations. The widespread instrumentality surrounding public dialogue on science-related issues is also conditioned by driving forces and precommitments associated with the unwavering institutional support for science-led progress (Stirling, 2008).

Finally, the findings of this article point toward future possibilities for building more reflexive forms of participatory governance, in terms of research and practice. There is no doubt that the above insights into what underpins the lack of reflexive learning relating to public dialogue on science-related issues demand further exploration through situated and longitudinal ethnographic studies within institutions as part of the emerging critical and reflexive mode of research on public engagement. Such inquiry could also be extended in questioning how the patterning of actor relations and governance of science and society compares—and indeed whether the above explanations of learning dynamics still hold—in other national contexts and transnational networks such as those highlighted in the introduction to this article. As has already been noted, however, a series of critical interventions in the PUS/PEST field have been deflected or reinterpreted by science and policy institutions for their own instrumental ends over the past three decades. A critical and historical-cultural research perspective on public engagement is crucial in its own right but also needs to be constructively critical in sparking, informing, and working with new forms of engagement, collective experimentation, and interventions that attempt to build more deliberately reflexive governance of science and technology. While public participation forms only part of the science governance system, the above analysis suggests important ways forward in enhancing reflexive and relational learning to this end. It points to the need for creative, action-oriented, and collective experimentation to make learning from and learning about public dialogue more

Not only would such interventions have potential to open up reflection and debate on the purposes, politics, exclusions, and uncertainties of technologies of participation and their implicit and often taken-for-granted visions of democracy and the public (cf. Felt & Fochler, 2008; Michael, 2009), they could also build in the sort of anticipatory reflexivity that has been absent from the field of public engagement hitherto. Now that forms of public dialogue are a site of innovation and professionalization as part of a global public engagement industry perhaps in addition to participatory technology assessments, there is a need for anticipatory assessment of these social (science) “technologies of participation” themselves. “Upstream questions” asked of emerging areas of science and technology can thus be projected back on to emerging democratic innovations as well: “Why this technology [of participation]? Why not another? Who needs it? Who is controlling it? Who benefits from it? Can they be trusted?” (Wilsdon & Willis, 2004, p. 28; text in italics added by the author). While existing participatory technology assessment and foresight procedures could no doubt play a role here, such reflection would crucially depend on situated forms of “anticipation in action” as part of an ongoing “conversation with the situation” (Rip, 2006; Schön, 1983).

Conclusion

This article has shown public engagement with science and technology, in the British context at least, to be rather unreflexive in its potential to make governance actors and institutions responsive, responsible, and accountable to public values, social implications, and uncertainties of science and technology and of public participation itself. Although new forms of public dialogue have undoubtedly brought about transformational changes and forged new relationships between science and society, learning in relation to it remains largely instrumental, crowding out and denying deeper forms of reflectiveness and reflexivity. Central to these findings is the fundamental difference in purposes, expected roles, and intentions between critical social science and the overriding instrumental imperatives of policy-practice. Rather than rely on one model, this situation calls for multiple and diverse forms of critical social science intervention at the science-policy-society interface. This has to include space for recognizing the value of critical, contextual, and historical-cultural analyses of science-public relations in their own right. Rather than remain distant critiques, however, the challenge is to make these reflexive and relational understandings infuse and work with engagement as practiced. The above analysis holds some hope in this regard. For example, there is potential for existing learning infrastructures to be reframed and “brought to life” through building capacities in institutional reflexivity rather than capacities in “more of the same” forms of public dialogue and through opening up evaluation tools and criteria to account for the politics, framing effects, downsides, and uncertainties of public dialogue. Participatory methods and appraisal tools can also be reconfigured to play an opening role (see, e.g., Burgess et al., 2007). Yet while models and heuristics hold the promise of more durable reflexive transformations, this has proved immensely difficult over the years due to powerful interests and unreflective institutional science-policy cultures. That the kinds of reflexive and relational learning espoused in this article are not a natural state for science and policy institutions means they will have to be partly forced. This could happen through open challenge, third-party enforcement, and “alternative witnessing” to prompt, question, and cue (Wynne, 2007) and through actors that can create difficulties for existing orders “because they irritate, contest, or like tricksters, are just mischievous” (Rip, 2006, p. 93). It is through embracing this diversity of critical social science engagements and a lively politics of science and participation that there remains hope for building more deliberately reflexive, adaptive, humble, self-aware, and publicly responsive forms of participatory governance for sustainable science and technology.