A reflexive look at reflexivity in environmental sociology

Abstract

Reflexivity is a central concept in environmental sociology, as in environmental social science in general. The concept is often connected to topics such as modernity, governance, expertise, and consumption. Reflexivity is presented as a means for taking constructive steps towards sustainability as it recognizes complexity, uncertainty, dilemmas, and ambivalence. Critical discussion of the conceptual meaning and usage of reflexivity is therefore needed. Is it a useful theoretical concept for understanding various sustainability issues? Is ‘more reflexivity’ relevant and useful advice that environmental sociologists can give in communicating with other disciplines, policymakers, and practitioners? This article explores the conceptual meaning of reflexivity and assesses its relevance for environmental sociology. In particular, it reviews its usages in three research fields; expertise, governance, and citizen-consumers. The paper furthermore discusses the spatial and temporal boundaries of reflexivity. It concludes by discussing how the concept can be a useful analytical concept in environmental sociology, at the same time as it warns against an exaggerated and unreflexive use of the concept.

Keywords:

• reflexivity
• environmental governance
• expertise
• citizen
• consumer

Introduction

Reflexivity has become a key concept in environmental sociology, in which it is associated with, for example, Giddens’ and Beck’s work on reflexive modernization (see e.g. Beck, Giddens, and Lash 1995). The concept of reflexivity is connected to topics such as governance, expertise, and lifestyle. It is furthermore associated with several other terms, such as uncertainty, transparency, and participation. This paper aims to give a theoretical account of the concept of reflexivity and its usage in environmental sociology. More precisely, the paper aims to identify varieties of usage of the concept and to assess the theoretical relevance of the concept to a sociology of the environment. To this end, the paper considers to what extent ‘reflexivity’ is a relevant theoretical concept for understanding various sustainability issues. Does reflexivity offer a useful model and concept for critically examining current practices? Could reflexivity be yet another oversimplified message to researchers, practitioners, and policymakers? Is ‘more reflexivity’ really a relevant and useful strategy that environmental sociologists can apply in communicating with various academic and other audiences?

Although there are good reasons for using ‘reflexivity’ as a theoretical and normative concept in environmental sociology, a self-reflexive examination of this concept and its application is warranted to prevent the call for ‘more reflexivity’ from becoming yet another unreflexive management imperative. This study is based on a review of previous literature applying the concept of reflexivity.

After this introduction, the second section will briefly review how reflexivity has been introduced and defined in environmental sociology. The third section reviews how the concept has been applied in three research areas in environmental sociology: science and expertise, environmental governance, and citizen-consumers. The fourth section analyses how reflexivity can lead to or coexist with unreflexivity as well as the extent to which reflexivity can be institutionalized. The fifth and concluding section discusses the potential role of the concept in environmental sociology both for studying a range of sustainability issues and for speaking to practice.

Reflexivity: self-confrontation and reflection

Ulrich Beck’s (1992, 1994, 2009) work on the world risk society and reflexive modernization is the key source for the concept of reflexivity in environmental sociology. Beck connects late-modern risks to how institutions (e.g. technology, science, politics, the state, and the economy) operated in ‘simple’ modernity. In simple modernity, problem-solving relies on a cognitive and instrumental approach in which uncertainty, complexity, and ambivalence are handled through the use of rationality and technology. This approach relies on the view that it is possible to know the ‘Truth’ on the basis of universal and objective knowledge, and that it is possible to control reality based on such knowledge. Problem-solving is specific and straightforward, the goal being to maximize the control of social and economic development. The theory of reflexive modernization holds that this approach to problem-solving inevitably leads to unintended and negative consequences. As these side effects multiply and are increasingly seen as unresolved by traditional instrumental approaches, a reflexive turn emerges – in Beck’s terminology, ‘reflexive modernization’ or the ‘world risk society’.

According to Beck (2009, 109ff, 1994, 5ff), such reflexivity does not necessarily imply more reflection. His analysis reveals that reflexivity can have two different, though related, meanings. The first meaning of reflexivity is what Beck calls self-confrontation. Industrial society generates unintended side effects and risks that shake the foundations of industrial society and its core institutions (e.g. the nation-state) in ways that cannot be ignored and that force society to take action. A nuclear accident, for example, must be handled. Furthermore, these ‘mega risks’ are socially explosive, having various social consequences: new discourses, movements, politics, and markets. As no one can foresee and estimate how these processes will evolve, the certitude of industrial society results in the uncertainties of the risk society that confront society and must be acted upon.

In contrast to this meaning of ‘reflexivity’, Beck uses the word ‘reflection’ to refer to various forms and constructions of knowledge. The conceptual meaning is appositely grasped by Giddens (1990, 38), who says that in reflexive modernization, ‘social practices are constantly examined and reformed in the light of incoming information about those very practices, thus constitutively altering their character’, stressing that reflection is constantly engaged in by both individuals (i.e. laypeople as well as experts) and organizations. Like Beck, Giddens links this propensity for reflexivity to historical processes of detraditionalization, individualization, and the undermining of traditional authorities and structures (e.g. the state, church, science, family, and gender roles). Whereas Giddens primarily discusses reflection and how it is institutionalized in late-modern society, Beck focuses on reflexivity. Beck (2009, 119ff) summarizes the differences between reflection and reflexivity by stating that whereas reflection concerns knowing and knowledge and the belief that more knowledge will increase the problem-solving capacity of institutions and society, reflexivity/self-confrontation concerns unknowing (i.e. unintended and unknown side effects). Due to their interrelatedness, in this article, we use the term reflexivity for both self-confrontation and reflection. When we specifically refer to only one of these aspects, we highlight this by using the terms self-confrontation and reflection.

In relation to the theories of reflexive modernization, scholars have addressed the importance of bringing to light the forces of counter-reflexivity (Borne 2009) or anti-reflexivity. McCright and Dunlap (2010) argue that the American conservative movement is an anti-reflexive force attempting to protect the industrial capitalist order of simple modernization by forcefully challenging the advocates of reflexive modernization, that is, the environmental movement and environmental impact science. Although it is important to pay close attention to forces trying to prevent the institutions of the simple modernity, it is an open question, on an analytical level, if these forces themselves should be labelled anti-reflexive. Indeed, the fight against reflexive modernization might itself require significant reflexivity, including capacity to anticipate others’ action and capability to intimidate, misrepresent, and manipulate environmental scientists and advocates. Reflexivity might be needed to keep powerful decision makers misinformed and unreflective. Not only Beck and Giddens but also Bauman argue that our current crises could very well lead to various narrow perspectives and fundamentalist ideologies including nationalism, xenophobia, and terrorism (Bauman 2006; Beck 2009; Giddens 1994a). When society can no longer provide certitude, as in the case of industrial society, one consequence could be increased reflexivity and another fundamentalist thinking. Stevenson and Dryzek (2012, 192) suggest that we can distinguish between reflexive modernization and reflexive traditionalization. The former notion implies increased awareness of and openness to discourses other than those into which one has been socialized. The latter implies the rejection of alternatives and a retreat into the familiar, while perceiving these alternative discourses as threatening the familiar. Linking the reflective capacity to discourses can thus both enable and constrain communication and deliberation. From this theoretical overview, we now turn to the use and debates of the reflexivity concept in three research fields.

Reflexivity in three research fields: expertise, governance, and citizen-consumers

Reflexivity is a concept with growing usage, and is utilized in a number of research areas. In this section, we review how the concept has been used and debated in three research fields of certain relevance to environmental sociology: science and expertise, environmental governance, and citizen-consumers.

Reflexive expertise

In reflexive modernity, scientific expertise, according to both Beck (1992) and Giddens (1990, 1994a, 1994b), has a new role. Its capacity to deliver objective truth, determine risk, and suggest ways to control risk is questioned. There is a ‘reflexive scientization’ in which science itself is deconstructed by means of science. Science is thereby both internally and externally contested: experts frequently disagree and this is known to the wider society. Furthermore, the very specialization of expertise means that there can be no meta-experts, that is, experts on all experts. Claims for the universal legitimacy and applicability of science are also more disputed than before. Taken together, this demonopolization of science results in a world consisting of multiple and competing epistemic authorities. Almost every public issue involves a heterogeneous supply of scientific statements. This heterogeneity in turn creates an open space for the public contestation of science, in which citizens scrutinize science and develop counter-expertise.

This situation does not, however, imply that the expert role is less important. Beck believes that a reflexive science, broadly disseminated in society and used by social movements and civic networks, can serve to democratize society. He states that ‘only a strong, competent public sphere, “armed” with scientific arguments is capable of separating the scientific wheat from the chaff’ (Beck 2009, 44). Likewise, Giddens (1994a) stresses that the advance and spread of scientific knowledge – including the consequences of its application – imply that tradition-guided action is replaced by more scientifically mediated understandings of the world. Today, according to Giddens, organizations exhibit an institutional reflexivity in which their performance is systematically monitored and controlled by themselves and/or other organizations. The reflexive appropriation of knowledge is a way to guide action, while this appropriation may simultaneously undermine the stability of social structures. Giddens and Pierson (1998) exemplify this with the financial market, which through its use of massive information creates new and unforeseen risks. Manufactured uncertainty is therefore more related to the advance of knowledge than to any lack of it (in contrast to Beck, who sees unknowing and its implications as central; see Beck 2009, 126–128).

Some scholars question this assumption of increasing reflexivity in society (Alexander 1996; Dean 1999; Lash 2000). In the field of science and technology studies (STS), the (un)reflexivity of expertise is a recurrent theme. Several empirical studies have found that scientific expertise is rarely reflexive concerning its own activity and underlying assumptions (Irwin 1995; Wynne 1992). Scientific experts’ belief in their own capacity to find correct and true answers has made them unresponsive to the public’s actually valid and reflected worries and claims, or they listen to but dismiss these concerns as manifesting non-knowledge (Wynne 2005). Experts often fail to consider that laypeople may be reflexive beings who do not naïvely believe their own knowledge to be true and others’ false, but instead often evaluate their own and others’ knowledge claims. The relevance and validity of the public evaluation of issues is therefore often denied by experts (Irwin and Wynne 2003; Wickson and Wynne 2012; Wynne 1992), and in cases when public engagement has been welcomed, this may occur without any deeper reflexivity in terms of the public being invited to critically examine scientific assumptions and normative frameworks (Chilvers 2012).

This unreflexivity is reinforced by the circumstance that many technical issues are framed in a technocratic way (Jasanoff 1990; 2005; 2011; Wynne 2005, 2010), at the expense of broader political, social, or cultural considerations. Such narrow framing places scientific expertise at the centre, involves questionable models of nature (with excessive reduction of complexities), and naïve models of how society works. Issues of, for example, genetically modified crops, nuclear power, and synthetic food products require not only scientific knowledge such as technical risk analysis but also insight into how society works, how activities are regulated, and how power is exerted. Such technical framing is often taken for granted, implying the preemption of political discussion and limiting the capacity to discover, understand, and internalize challenges that arise outside a particular frame. To be reflexive, experts must be aware not only of their own assumptions but also of the framing of the issue at stake.

However, research has found that expertise can be reflexive in another way. Scientists and experts often act strategically in order to maintain and strengthen their credibility and epistemic authority. Drawing on Goffman’s dramaturgical view of society, Hilgartner (2000) uses the ‘stage management’ concept to capture science’s deliberative and reflexive work to become an authoritative reference for determining what should be done. Front-stage performance encompasses activities that are believed to strengthen experts’ authority and are therefore deliberately shown to the public. For example, scientific results are presented as certain and objective, produced at a distance from and independently of political and normative considerations (cf. Latour 1998). Backstage, remote from public visibility, the process of knowledge production and synthesis may be characterized by uncertainty, controversies, and normative biases. By considering stakeholders’ views, experts can deliberately choose what to make public (front stage) and what to conceal (backstage) in order to deliver trusted knowledge and to facilitate action. Expertise can be reflexive in the sense of anticipating how other actors may understand and evaluate its messages, and therefore strategically stage its activities to maximize authority and influence.

Although many STS researchers criticize science for not being reflexive, contesting the claim that society has moved from simple to reflexive modernization, they still believe in knowledge, learning, and reflexivity. All knowledge – including scientific knowledge – derives from particular social and cultural contexts. Knowledge processes should be open to many voices, they argue, including explicit negotiations and critical discussions among various discourses. This proposal presupposes an institutionalized reflexivity, that is, that organizations should develop a self-critical ability to review their own assumptions and commitments and expose them to critical and public contestation. Nonetheless, there is always a risk that scientists (as well as governance actors; see Section 3.2) may embrace reflexivity and participation in a shallow sense in which they welcome a choir of supporting voices without opening themselves to critical evaluation of their own assumptions and definitions of issues (Chilvers 2012; Irwin 2006).

To sum up, while disagreeing on several matters, scholars such as Beck, Giddens, Irwin, and Wynne demonstrate that ‘reflexivity’ has a role to play when approaching expertise. What is problematic, however, is when it is assumed that science has indeed become more reflexive in Beck’s sense. A crucial empirical question for environmental sociology is whether scientific institutions and scientists are endorsing and practicing the explicit questioning of their own assumptions, opening themselves to external viewpoints. STS demonstrate that reflexive expertise often appears as a more desired option than empirical reality. If the social and normative contexts and assumptions of science are not critically discussed and debated, the current rhetoric on transparency, reflexivity, and inclusion may reproduce unreflexive scientific practices and expert advice. Reflexive expertise would require considerable self-criticism and openness to multiple actors to be able to raise novel questions and to critically evaluate and contest scientific propositions and expert recommendations.

Reflexive governance

The concept of reflexivity is sometimes referred to in studies of environmental governance and risk governance (e.g. Van Asselt and van Bree 2011; Van Asselt and Renn 2011; Aven and Renn 2009). In these studies, reflexivity is associated with other concepts and norms, such as openness, transparency, and participation. A general idea is that experts, decision makers, and other participants should be open to questioning assumptions in a given situation, should not conceal issues of uncertainty and the pluralism of values, and should be receptive of the input and participation of other stakeholders. Reflexivity is also treated as the core analytical concept in some of such studies (e.g. Boström, Grönholm, and Hassler, 2016; Brousseau, Dedeurwaerdere, and Siebenhüner 2012; Hassler, Boström, and Grönholm 2013; Schutter and Lenoble 2010; Voss, Bauknecht et al. 2006; Voss and Bornemann 2011). Reflexive governance hence refers to governance that is concerned with itself by means of the self-critical scrutiny of current governance, including its achievements and unintended negative effects. Jan-Peter Voss, Bauknecht et al. (2006) have made a substantial contribution here, by building on the work of Ulrich Beck. They argue:

Reflexive governance puts itself up to probing. It acknowledges that governing activities are entangled in wider societal feedback loops and are partly shaped by the (side) effects of its own working. It incorporates such feedback by opening problem-handling processes for diverse knowledge, values and resources of influence in order to learn about appropriate problem definitions, targets and strategies of governance for sustainable development. (xv–xvi)

Reflexive governance hence includes the double meanings of reflexivity: the self-confrontation that Beck speaks about and reflection that particularly Giddens emphasizes.

A key question is how existing discourses and social arrangements reproduce the generation of problems. Another question is whether certain governance structures and processes could facilitate reflexive learning.

Reflexive governance scholars recognize that global and local sustainability problems are complex, uncertain, and ambivalent and need to be treated as such (Kemp and Martens 2007). Problems cannot be solved in a once-for-all manner because new problems, trade-offs, and ambivalences are likely to appear after decisions are made. This stream of literature, thus, emphasizes that governance actors must develop the potential to respond continuously to unexpected outcomes.

Accordingly, reflexive governance is geared towards continuous learning ‘in the course of modulating ongoing developments, rather than towards complete knowledge and maximisation of control’ (Voss and Kemp 2006, 7). Governance actors must be forward looking and adaptive, allow for trial-and-error learning, and permit experimenting with new innovations (Grin 2006; Kemp and Loorbach 2006); they must also be backward looking, make use of experience, and critically evaluate earlier mistakes (Siebenhüner 2011). In being subjected to scrutiny, governance actors must be confronted with witnesses of how existing governance contributes to reproducing problems. The reflexive governance perspective accordingly pays close attention to the importance of public debate and the monitoring role of the media, civil society organizations, and other actors. Cross-sector and multi-actor approaches are therefore called for (Boström, Grönholm, and Hassler, 2016; Hassler, Boström, and Grönholm 2013; Voss, Bauknecht et al 2006).

With regard to this research scholars have also raised issues for debate. In a review of the scholarship on reflexive governance, Walker and Shove (2007) welcome its serious consideration of ambivalence, while they argue it still provides an insufficient analysis on how politics and power produce ambivalence. Another problem debated by reflexive governance scholars themselves is the ‘efficacy paradox’ (Voss, Kemp et al. 2006), that is, the tension between ‘opening up’ for the inclusion of more actors and ‘closing down’ for decision-making. The more actors that become involved, the trickier the decision-making process is likely to be. Yet another problem is path dependency. Due to various kinds of institutional and organizational inertia, reflexive learning and continuous reform will be challenging. Unless institutions are not self-confronted by the side effects of their own operations and thus forced to change, various external and internal factors tend to reproduce existing institutional structures (e.g. existing rules, discourses, vested interests, and habits; Bos and Grin 2008; Grin 2006). Theorists of reflexive governance engage with this issue of (intended/unintended) change versus inertia, and underscore the importance of taking path dependency seriously and suggest step-by-step transformation (Grin 2006; Kemp and Loorbach 2006) rather than seeking unrealistic utopian policies.

What are the lessons from empirical studies of environmental governance and management that apply the concept of reflexivity? We have found studies demonstrating progress towards reflexive governance as well as studies demonstrating the opposite: the absence of reflexivity.

Hassler, Boström, and Grönholm (2013) found many forces that prevent reflexivity in intergovernmental organizations; they also found positive developments, particularly in the International Council for the Exploration of the Sea (ICES). Through conscious efforts over several years, ICES has achieved significant organizational and cognitive restructuring to facilitate more inclusive and holistic approaches to producing policy-relevant advice on fishery issues (see also Wilson 2009). However, it is arguably easier to achieve change within organizations than throughout sectors. In a study of various marine risks connected to environmental governance of the Baltic Sea, Boström, Grönholm, and Hassler (2016) noticed the rarity of cross-sectoral learning, for example, between the agriculture and fishery sectors. Hence, an important prerequisite for reflexivity was lacking.

Stevenson and Dryzek (2012) note the potentially reflexive capacity of the plurality of transnational climate change discourses. However, this potential can only be realized if these discourses communicate with each other. They studied four non-state summits held just before the 2009 Copenhagen Climate Summit. These non-state summits appeared to be discrete events dominated by particular discourses. A business-dominated ‘Mainstream Sustainability’ summit and a social movement-dominated ‘Green Radicalism’ summit did not communicate with each other. The groups thus failed to learn from each other, amplifying established views and ideologies and consequently failing to foster reflexivity. Gale and Cadman (2014) present another example of how networks with divergent discourses fail to speak to each other. They highlight the lack of reflexive policymaking in their study of the development of an economistic international Sustainable Forestry Management (SFM) norm (i.e. the Montreal Process). The norm developed very rapidly within a closed and clientelistic policy network in Canada that ignored a concurrent eco-social SFM norm-development process (for other studies with similar themes, see Wales and Mythen 2002; Marsden 2012; Friedland, Ransom, and Wolf 2010). Several studies illustrate how reflexive elements initially contribute to the inspiration and design of policy or innovation processes, but often vanish during the processes themselves. The economic and normative power of an existing socio-technological regime and institutional context can be a strong force preventing structural change. Examples include studies of efforts to develop a sustainable energy supply system in the Netherlands (Kemp and Martens 2007) and of an innovation project to develop a sustainable husbandry system for pigs in the Netherlands (Bos and Grin 2008). In sum, though the above studies are varied, they demonstrate that reflexivity concepts have been applied in studies of governance, policymaking, innovation, and management at least as much to identify missed elements or opportunities as to explain what actually happens in governance. The review thus shows that the concept has both normative and analytical usages. Bos and Grin (2008) indeed argue that it is a feature of studies of reflexive modernization from the Beck tradition highlighting the tension between simple and reflexive modernization. Existing governance structures and processes reveal tensions between reflexive and unreflexive forces, and studies in the field alternate between analysing what actually happens and what ought to happen.

Reflexive citizen-consumers?

Since the late 1980s, policy, public debate, media, civil society, and the social sciences have addressed the conduct of private actors (e.g. organizations, households, and individuals) regarding greenhouse gas (GHG) emissions and other environmental issues. Contrary to the expert understanding of the public as ignorant and unreflexive, as described in Section 3.1, this emphasis on people’s everyday conduct relies largely on the notion of reflexive citizens-consumers. Beck (2009, 95) uses the concept of sub-politics to describe how reflexive modernization and individualization entail the ‘de-coupling of politics from government’, implying the self-organization of a variety of non-state actors. Drawing on Sartre, Beck (1997) concludes that people are condemned to individualization, implying that ‘the standard biography becomes a chosen biography’ (p. 96, italics in original; see also Beck and Beck-Gernsheim 2002). Similarly, Giddens (1991) concludes that the emancipatory politics of modernity was a politics of life chances, whereas the life politics of late modernity is a politics of choice.

The notions of chosen biography and reflexive identity comprise a broad range of issues, including family life, gender, and sexuality. In environmental studies, the concept of reflexivity has been focused on green identities and lifestyles, everyday practices, and activism. For example, Boström and Klintman (2008) argue that there is a broad potential for reflective trust in eco-labels on the part of citizens, instead of the blind trust that the labelling systems tend to spur. Their argument is based on findings that it is well-educated and politically interested citizens who express interest in green and political consumerism (see also Stolle and Micheletti 2013). In studying social responses to climate change, Davidson (2012) focuses on ‘meta-reflexives’ who appear engaged and resourceful. These individuals are capable of grasping the complexity of climate change, spend ‘a great deal of energy on inner dialogue’ (p. 620), are value oriented, and tend towards activism. Other studies examine the process of lifestyle change as one of moral identity formation (Lorenzen 2012; Sandlin 2009; Shepherd 2002).

Although several studies draw in various ways on the concept of individual reflexivity, the thesis of the reflexive individual has also been debated and met with criticism. In a review of literature critiquing the individualization thesis, which is contained in the theory of reflexive modernization, Dawson (2012) distinguishes between critiques from the modernist, interactionist, and discourse perspectives. This distinction captures well the debate, which is summarized in the following paragraphs.

The main critique articulated by the modernist research strand concerns the weak or absent empirical basis for the individualization thesis, resulting in an ahistorical account of individualization. The criticism is that ‘individualization, to the extent it exists, is not in any way “new”’ (Dawson 2012, 308–309), and that it broadly reflects the liberal, middle-class values of its proponents, who back up their thesis by randomly chosen illustrative examples.

The interactionist critique mainly concerns the notion of reflexivity as an inner process. Scholars in this strand call for the reintroduction of the social when discussing individualization, emphasizing that reflexive awareness is in fact a culturally embedded process (Dawson 2012, 310). The notion of an autonomous, critical, and well-informed individual¹ who makes ‘free choices’ based on the best available information is based on a simplified view of the individual and a false distinction between the individual as either reflexive or traditional (Klintman 2012; Shove 2010). Studies have illustrated how the individualization of environmental responsibility entails ambivalence, uncertainty, contradiction, hypocrisy, and dissent when people try to make sense of the ascribed responsibility relative to other norms and values, such as cleanliness, freedom, mobility, and care for children (e.g. Blühdorn 2013; Borne 2009; Cherrier 2009; Connolly and Prothero 2008). Other studies elucidate the importance of group solidarity and collective action (e.g. Cherrier 2009; Pentina and Amos 2011; Portwood-Stacer 2012), highlighting ‘the intertwinement of collective identity and the individual’s struggle to perform in accordance with certain values and group norms’ (Soneryd and Uggla 2015). Likewise, studies drawing on practice theory emphasize that behavioural practices must be understood as socially embedded and that the elements of social practice (e.g. clothing, housing, food, and travel), rather than the individual as a relatively isolated entity, should be at the core of the analysis (Halkier 2009; Hargreaves 2011; Spaargaren 2003; Warde 2005). Yet other studies of citizen-consumers reveal that the distinction between reflexivity and unreflexivity is not clear-cut. For example, as Halkier (2001) puts it, ‘social life is neither entirely coincidental nor entirely determined’ but evinces a mixture of routine and reflexivity (see also Klintman 2012; Shove 2010).

Moreover, the modernist and interactionist research strands share the critique concerning the concept of sociological ‘zombie categories’ introduced by Beck, concluding that the empirical foundation for the assumption that concepts such as class and gender have become obsolete is fairly weak (Dawson 2012).

The third strand of critique concerns individualization as discourse. From this perspective, the thesis of the reflexive individual is seen as a cornerstone of neo-liberalism (Dawson 2012). Drawing on Foucault’s governmentality concept, this research focuses critically on the techniques of the self (e.g. information and education campaigns) employed in governing people’s conduct, contributing to a narrow view of agency. For example, the individual is addressed primarily as a consumer (Akenji 2014; Kent 2009; Maniates 2001) or a carbon calculator and energy saver (Paterson and Stripple 2010; Uggla and Uggla 2016), and this type of individualized responsibility suggests that we can engage and feel content with minor changes in everyday routines. Rather than gathering as citizens and finding political solutions to institutional challenges, we are supposed to believe that green consumption alone can bring sustainability (Kent 2009; Maniates 2001, 37). However, reflexivity does not necessarily imply compliance with a hegemonic discourse (Soneryd and Uggla 2015), and it could well lead to resistance to the whole idea of green consumption.

To sum up, the fact that the notions of individualization and the reflexive individual have become predominant in late modernity and the circumstance that people are increasingly individualized do not necessarily imply that this constitutes a lived reality for people. Rather, scholarship of (un)reflexive citizens tend to highlight the socially embedded citizen or embedded individualization (Dawson 2012). In this account, sociological concepts of stratification, including class and gender, are all but obsolete, and further research ought to keep focus on how various discourses, social practices, social movements, governance tools, and information campaigns empower or constrain citizens to become more or less reflexive.

Boundaries of reflexivity

In this section, we draw on ideas from the referred literatures above and add relevant concepts in order to theorize how reflexivity and its inverses are intertwined. This analysis will further an understanding of the spatial and temporal boundaries of reflexivity. We find it illuminating to, first, use the spatial metaphor of ‘inside’ and ‘outside’ to discuss certain possibilities and limitations. That is, heightened reflexivity may be achieved ‘inside’ the boundaries of a setting, frame, expertise (Section 3.1), discourse, or social arena, or governance sector (Section 3.2) but to the expense of lost reflexivity ‘outside’ that boundary. Second, we find it warranted to address a temporal notion (or paradox) in that efforts to achieve reflexivity on a continuous basis, may imply the institutionalization and routinization of it, which in turn may lead to its dissolution.

Reflexivity within, unreflexivity outside

Framing theory provides useful analytical tools for studying how reflexivity ‘within’ one setting can be heightened at the same time as unreflexivity ‘outside’ this setting remains. Indeed, various scholars have addressed how ‘reflection’ can be facilitated by frames or discourses (Stevenson and Dryzek 2012; Fischer 2003). Boström and Klintman (2008) distinguish between reflection within a frame (i.e. intraframe reflection) and reflection across frames (i.e. interframe reflection). Intraframe reflection concerns how actors use cognitive tools to reflect on practices. Frames such as ‘precaution’, ‘biodiversity’, ‘cleanliness’, ‘critical loads’, and ‘climate friendliness’ may enable actors to perceive, reflect on, and understand things in novel ways and, through such reflection, change practices. It is equally important to ask how attention to some aspects may result in less attention paid to other aspects. An important insight from framing theory is that frames set boundaries: a frame helps us see what is within the frame but excludes things that fall outside it. Focusing on certain aspects of an activity simultaneously means less attention (i.e. less reflexivity) paid to other aspects that fall outside the frame.

For example, eco-labelling can foster consumer reflexivity in relation to particular product segments, while this narrowed reflexivity may run in parallel to, or even facilitate, ignorance of other product segments or of overall consumption practices and levels (Boström and Klintman 2008). Eco-labelling can also be used in routinized ways or implicitly cited to excuse polluting activities in other domains. Likewise, though the system of offsets to compensate for GHG emissions from flying may raise awareness and prompt reflection on travelling and mobility, it could also be used in a routinized unreflexive way or as a means to justify one’s travel habits. The frame of green consumption and individual responsibility can thereby narrow the view of what constitutes environmentally responsible behaviour and of who the ‘polluter’ is. In this sense, eco-labelling and other means of responsibilization entail a narrow view of both environmental degradation and environmental protection. By the frame, reflexivity is limited, implying less reflection on structural and political issues, which is exactly what the critics of individualization as discourse point out.

The notion of intraframe reflection implies that reflexivity (within the frame) and unreflexivity (outside the frame) can proceed simultaneously. Reflection across frames (e.g. ‘sustainable development’ vs. ‘economic growth’ or ‘natural’ vs. ‘artificial’) has more potential in terms of learning and transformation. Arguably, any structural change in society (at least a democratic one) has to involve elements of interframe reflection or, as Stevenson and Dryzek (2012) put it, communication between discourses. Interframe reflection concerns the capability to scrutinize both one’s own frame that underpins an activity and another actor’s frame, for example, that of an opponent:

The process of frame-reflection depends in particular on the orientations of the participants: their relative distance from their objects under consideration, their willingness to look at things from other perspectives, their propensity toward ‘cognitive risk taking’ coupled with their openness to the uncertainty associated with frame conflict. (Fischer 2003, 146)

In a governance process, such productive frame reflection is likely to be quite demanding for participants. Adding a temporal dimension, not least how actions based on reflexivity over time often become routinized and institutionalized, the challenges become even more accentuated.

Institutionalized reflexivity – an oxymoron?

Institutional reflexivity is defined by Giddens as ‘the regularized use of knowledge about circumstances of social life as a constitutive element in its organization and transformation’ (1991, 20). However, a reflexivity that over time becomes institutionalized, in this Giddens’ sense (not to be confused as institutional reflexivity in terms of self-confrontation), appears as an oxymoron. Can reflexivity be routinized? The notion of reflexive governance implies precisely this: that a self-critical questioning of one’s practices should be routine. The reflexive look at practices should not be a one-time happening but, if not constantly, at least occur during repeated discrete events (e.g. scheduling ‘reflection’ once a month).

As mentioned earlier, theorists of reflexive governance do acknowledge organizational inertias and path dependencies in policy development and the critique of the individualization thesis addresses that citizens are constrained by socially embedded practices. This is realistic: governance and practices cannot be incessantly reviewed and altered. Such never-ending reviewing would end up in paralysis and a loss of ability to act. People have to use tacit, embodied, and practical knowledge in their actions and practices, and routinization facilitates our everyday practices. Organization relies on stable structures (rules, division of labour, control mechanisms), and this stability is a condition for action capacity (Ahrne 1994). Yet, theories of reflexive governance, science, and citizens often fall short in problematizing the intertwinement of reflexivity and routine, that is, how a practice initially based on self-critical reflexivity over time can turn into unreflexive routines and habits. If governance structures and everyday life practices are institutionalized, then the frames that underpin these structures and practices tend to become cemented. If reflexivity requires alternative yardsticks, or frames, with which to assess practices (i.e. interframe reflection), fixed framings will prevent reflexive learning. Organizational and cognitive inertia go together.

Kemp and Loorbach (2006) argue that an incrementalist approach to policy reform, acknowledging the rigidity of institutional structures, does not have to be blind. Even in such approaches, careful forward-looking efforts to avoid and escape lock-in effects are possible. Voss and Kemp emphasize ‘the importance of shaping new technologies, social practices and institutional arrangements at an early stage of their development while they are still malleable’ (2006, 13). Theorists of reflexive governance seem to suggest the importance of identifying formative moments when different pathways are still imaginable and feasible. This does not solve the theoretical problem, however. If reflexivity entails the incessant possibility that the current system is fallible and may need to be changed based on new information (Giddens 1990, 1991), reflexivity cannot be restricted to only the formative moments of a particular practice. Though formative moments are very important, there might well be a need for openness to ongoing, new formative moments.

Though not impossible, institutionalized reflexivity appears very demanding. It is one thing to facilitate reflexivity at one point in time or during a limited formative period when alternative paths are imaginable and open for discussion and for interframe reflection. It is quite another thing to keep this reflexivity going incessantly, and if it is feasible it could easily lead to paralysis and loss of action capacity.

Conclusion: is reflexivity a useful concept for environmental sociology?

The literature on reflexivity connected to topics such as expertise, governance, and consumer-citizens advances our understanding of some of the preconditions and challenges of individual and institutional movement towards sustainable development. Our review shows that reflexivity can be a useful analytical concept in environmental sociology, but we advise against an exaggerated and unreflexive use of the concept. Reflexivity needs be used with caution, put in context, and with a firm systematic look at its boundaries and opposites. Likewise, if reflexivity is used to speak to practice, it is relevant to ask whether the call for ‘more reflexivity’ really is helpful for decision makers and practitioners. They may develop reflexivity in that they become aware of how their own management approaches and social practices continue to reproduce problems and risks, while they remain locked in organizational inertia and paralysed by current approaches. In these cases, which are likely to be many, ‘more reflexivity’ does not suffice. As an analytical concept for environmental sociology and as a way to speak to practice, we suggest instead to address reflexivity in five other ways.

First, it is necessary to bring to attention that the concept of reflexivity may not always predict very well how institutions and behaviour develop. We have shown that theories of reflexivity are not only used in describing, understanding, and explaining what happens in environmental governance, scientific practice, and among citizens but are also often used for critically exploring what is not happening and what ought to happen. Environmental sociologists tend to use the concept as a normative yardstick in critiquing and providing constructive comments on practice. The concept of reflexivity equips scholars with sub-concepts and perspectives enabling them to see what should be present in science, governance, and among citizens reflecting on their practices. In this way, the concept helps to direct attention towards what is missing and draws attention to inertia, instrumental learning, and the intertwinement of reflexivity and routine in everyday life practices. On the other hand, the concept has also been used to examine processes of ‘going green’, including environmental activism, green identities, and lifestyles, as well as the development of more responsible organizations and governance that facilitate reflexivity. In doing so, the concept has helped elucidate the deliberation and learning that such processes involve. These two sides of the concept – used in discussing both what is present and absent in practices as well as both progress and backsliding – can be interpreted as a strength if they are reflexively applied; that is, not as a simple call for ‘more reflexivity’.

Second, the spatial and temporal boundaries of reflexivity need to be recognized, as we discussed in the fourth section. Heightened reflexivity in one setting (or within a particular frame or discourse) or at one point in time may be accompanied by less reflexivity outside that setting or in a later phase. If organizations and people are called to increase their reflexivity regarding something, there is a need keep a firm analytical eye on whether this cause something else to be neglected.

Third, if environmental sociology is to provide constructive critique to practice, it has more potential if it addresses the embedded nature of and the conditions needed for reflexivity rather than simply recommending ‘more reflexivity’ as such. Environmental sociology could explore, critically examine, and suggest feasible structures and practices with the potential to facilitate reflexivity. Indeed, theories and literature of reflexivity provide useful ideas about what these conditions could be. A central insight, gained from the three research areas reviewed in this article, is the importance of developing meeting points, social arenas, and organizational forms that enable time and space for deliberations between various groups, sectors, and networks, which can in turn facilitate interframe reflexivity and mutual learning. What is needed is to enable sensitivity and learning about various structural and cultural forces at multiple levels that routinize behaviour, cement discourses, and prevent change of institutions and practices.

Fourth, as well as being a tool for change, reflexivity is also a target of anti-reflexivity forces that aim to prevent change. It is therefore also crucial to address the powerful forces that deliberately and strategically counteract reflexivity, which many times are the same forces that create environmental destruction. While some obvious forces are political and business elites with the power to dupe people and organizations with misinformation spread via anti-environmental campaigns, we must not forget those broad layers of citizens inclined to celebrate their supposed ‘green identity’ although their actions do not justify such a label. The sociological scholarship on reflexivity hence must thus keep a firm eye on its opposite.

Finally, reflexivity is a point of departure rather than an end in itself. It is questionable whether reflexivity is sufficient in itself as a principle to guide practice towards more sustainability. We argue, as in the case of related concepts such as precaution, deliberation and responsibility, that the concept of reflexivity could serve as a starting point. Like many other concepts, it needs to be contextualized and specified. It is an opening for dialogue and action rather than a point of closure. It begs questions such as: What path dependencies are confronting us? Do we have to develop new understandings, roles, and guidelines to avoid reproducing problems? How do our current norms, ways of communicating, and routines prevent our imaginations from seeking and finding better practices? Yes, reflexivity is a useful concept for environmental sociology, if not used in an unreflexive way.

Acknowledgment

A first draft of the article was presented at the international workshop “Core Concepts in Environmental Sociology” at Örebro University, 23–25 September 2015. We are grateful for the constructive comments from workshop participants, as well as from two anonymous reviewers.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Magnus Boström

Magnus Boström is a professor in Sociology with a research focus on environmental sociology.

Rolf Lidskog

Rolf Lidskog is professor in Sociology with research interest in environmental policy and politics, especially the role of expertise in environmental politics.

Ylva Uggla

Ylva Uggla is a professor in Sociology, with a research focus on environmental sociology.

Notes

1. This notion is commonly not only associated with rational choice but has also been associated with theories of reflexivity. Theories of reflexivity, however, are more prone to being consistent with the notion of socially and culturally embedded individuals, as argued below.