Theories of practices: Agency, technology, and culture: Exploring the relevance of practice theories for the governance of sustainable consumption practices in the new world-order

Received 29 October 2010
Received in revised form 23 March 2011
Accepted 25 March 2011
Available online 22 April 2011

Practice theories
Sustainable consumption
Culture
Global environmental governance

People do not develop ideas and ways of doing 'from within' by themselves. Their thinking and doing are shaped by fellow citizens and by the objects and situational factors which form an integral part of the contexts of their behaviours. By restricting themselves to strategies from the individualist paradigm, policy makers can be said to be sociologically naïve while neglecting the profound influences of the wider chains of interaction that serve as systems of provision shaping and sometimes pre-configuring the choices and behaviours of individual citizen-consumers to a considerable extent. As a result, too much responsibility for change is put on the plate of the individual citizen-consumer.

Criticism of the systemic paradigm focuses on more or less the flip side of the voluntarist argument. When resorting to

institutional actors and measures only and when trying to exclusively organize environmental change via new technologies and infrastructures that are installed by providers, policy makers are denying or at least underestimating the crucial role of human agents in processes of environmental change. In this top-down structuralist approach, citizen-consumers are hardly offered the possibilities to participate in, co-shape and democratically control processes of environmental change. As studies on failed technological innovations (Schot, 2001; Heiskanen et al., 2005) show, it turns out to be very difficult to realize the environmental benefits of eco-designed products, technologies and infrastructures when they are designed without reference to the user-practices they help constitute and are implemented without the knowledge and education of practitioners.

Because of the limitations of both the individualist and systemic or structuralist approaches in providing the deep understanding necessary for breakthroughs towards more sustainable consumption, there is a need to find a more balanced approach which pays attention to both agency and structure, which makes room for (combining) both bottom-up and top-down dynamics of change, and which recognizes the mutual influencing and co-shaping of human actors on the one hand and objects and technological infrastructures on the other. We argue here that practice based approaches as developed within sociology since the 1970s and 1980s are well-suited to provide such a new, balanced approach (Spaargaren, 2003; Shove, 2003, 2006; Southerton et al., 2004). In the next sections we will discuss the practice paradigm both from a theoretical and governance perspective, showing some ways in which conceptual innovations on the issues of agency, technology and culture can be used to renew the policy agenda for the governance of sustainable consumption.

Environmental governance is about attempts to change the reproduction of practices, systems and networks towards greater sustainability. Practices can be of all kinds. Next to the routine consumption practices of everyday life, there are practices implied in the reproduction of markets, politics and civil society as well. Corporate Social Responsibility (CSR) practices can be the focus of environmental governance in a similar way as practices of consumption. Furthermore, practices can be situated at all segments or nodes of production-consumption chains and networks. When the focus is on the ecological modernization of practices of mining, processing, storage, transport, retail and distribution, this is referred to in the literature in terms of studying upstream processes and dynamics as enacted by providers of (green) products, services and infrastructures. When the focus is on practices of buying, storing, consuming, (re)using and recycling as enacted by citizen-consumers as the end-users of (green) products and services, this is referred to in terms of downstream processes and dynamics (Spaargaren and van Koppen, 2009). In what follows, the downstream practices of consumption will be used to illustrate the relevance of a practice approach for environmental governance.

Conceptually the number of downstream consumption practices relevant for environmental governance is unlimited. Three criteria are relevant for the selection of specific sets of practices for the environmental governance of consumption. First, they should be practices that are relevant from a citizen-consumer or 'practitioner's' point of view. This means that the practices are situated in everyday life and governed by life-world rationalities to a considerable extent. Such practices should be recognizable to citizen-consumers, implying that lay people must be offered the unconditional opportunities for gaining first-hand knowledge about these practices.² When studying the practice of 'shopping for food in a supermarket' it is presupposed that almost all citizen-consumers possess first-hand knowledge about the rules and resources implied in the shopping practice. Secondly, the consumption practices should be relevant for environmental and climate governance in a particular way. This does not mean that only practices with high environmental impacts are to be selected, since also practices representing a high eco-innovative potential can be relevant. Also 'new', yet not common or mainstream practices could be of strategic interest for policy makers. Within the environmental sciences, the environmental impacts of consumption practices like dwelling a home or driving a car are by now well established in technical (CO₂-emissions, water consumption, waste aspects, etc.) terms. Third, in order to prevent the fragmentation and 'horizontalization' (Fine and Leopold, 1993) of consumption policies, it makes sense for researchers and policy makers to focus on clusters or sets of consumption practices that are situated within a limited number of 'domains' spanning everyday life. Within one domain – or field, as Bourdieu would have it, practices are reproduced under the influence of a shared 'regime' (Geels, 2005) since they are connected to similar chains and networks involved in the provisioning of the practices. Because of these shared sets of cultural rules and resources, practices in one field or domain can be argued to obey the specific and also historical dynamics belonging to that particular field or domain (Warde, 2004). When dealing with food, the sustainability criteria

and frames show some specific, food related elements (freshness, land-use, animal welfare) when compared to the domain of housing, while also sharing common criteria (CO₂-emissions, water-use).

To give an impression of the sets and domains relevant for the governance of sustainable consumption, Fig. 2 shows a number of practices in six consumption domains. The selection is partly based on the CONTRAST³ research project which aimed to explore transitions towards sustainable consumption practices in the Netherlands. In the context of this study it was shown that the dynamics of environmental change are significantly different for various consumption domains while also showing differences in the levels of change already realized. The domains of 'housing' and 'food' showed the highest levels of what might be termed ecological modernization, while the domains of 'leisure and tourism' and 'clothing and personal care' tend to lag behind, with the mobility domain taking an in-between position (Verbeek, 2010). The consumption domains suggested in Fig. 2 are similar to the main categories used by the transnational environmental NGO Global Action Plan (GAP) (Staats et al., 2004) and in line with the international agenda for Sustainable Consumption and Production (SCP) policies as developed in the context of the so-called Marrakech process from 2002 onwards (Spaargaren and Cohen, 2009). When studying the prospective and actual greening of global consumption, some practices will be more relevant than others, depending on the geographical region under study. Also in some consumption domains (e.g. housing) the influence of national policies still seem to be much stronger when compared to state power in other domains (food, clothing).

Key questions for social scientists interested in the governance of sustainable consumption refer to the factors and dynamics implied in the transition of existing social practices into corresponding future practices which bring along substantially reduced environmental and climate impacts. What is at stake are substantial reductions in the ecological footprints of practices instead of focusing on footprints of individuals only or primarily. This can be realized through the introduction into the practices of more sustainable objects, products and technologies on the one hand and by developing new green images, norms and ideas for the practices on the other. Stated differently: the ecological modernization of practices can be said to refer to the process of incorporating and anchoring into the practice the objects, meanings – e.g. the 'ways of doing and saying' – which are important for monitoring, assessing, valuing, and improving the practice with respect to its environmental or climate performance. Both the technological and cultural dimension of innovation in consumption practices will be considered in more detail in the two sections to follow.

With hindsight, it can be concluded that the first generation of practice theories as developed by Giddens and Bourdieu had little to offer for analyzing the role of objects, technological systems, and hybrids. While being formulated in about the same period – the 'long 1980s' – structuration theory and actor network theory showed little awareness of each other's existence. In the 1990s, several authors set themselves the task to reformulate practice theory in such a way that the role of objects and technologies could be included in practice theory without doing away with some of its core premises regarding the agency-(technological) structure relationship.

Schatzki (1996, 2002), Reckwitz (2002a,b) and, in a different manner, also Urry (2000) have sought to address the issue of agency, structure and technology more in depth and against the backdrop of both structuration and actor network theories.

Reckwitz in particular contributed to the incorporation of technology into theories of practices. When commenting upon the work of Schatzki, he argues for the need to reconcile a strong emphasis on the (autonomous impact of the) role of technology with the key structurationist assumption of human agency as being the ultimate factor for making a difference in the world. The impact of things in the social order must be fully recognized and conceptualized, not just in terms of representations, or as things that are assigned and attributed meaning by human agents. The impacts and effects of the objects themselves, the role of inter-objectivity next to inter-subjectivity, and the idea of objects being 'constitutive' for social practices all have to be considered and conceptualized in more detail (Reckwitz, 2002b, p. 212).

Reckwitz proposes to complement Schatzki's work with Latour's idea of the equally important constitutive role of things for social practices: 'things handled' are as important for theories of social practices as 'minds/bodies performing', so he argues. The crucial role of things and their use for social practices is expressed in one of the most elaborate definitions of the concept of social practices, provided by Reckwitz. Social practices are:

a routinized type of behaviour which consist of several elements, interconnected to one other: forms of bodily activities, forms of mental activities, 'things and their use', a background knowledge in the form of understanding, know-how, states of emotion and motivational knowledge (Reckwitz, 2002a, p. 249; emphasis added).

Reckwitz' formulation of the material dimension of social practices we think to be closest to ANT as possible without violating the cherished assumption of theories of practices: the crucial importance of the role of knowledgeable and capable agents in shaping social life. In comparison with Giddens, Reckwitz assigns greater analytical power to objects and technology without embracing the ANT-based anthropomorphic notion that objects and things 'act' in ways similar to that of humans.

When translated to empirical research, a number of consequences can be derived from the above perspective. First, things or objects as crucial elements of practices do not figure in isolation but – to use a key phrase of Schatzki – they 'hang together' in specific ways. This means that lock-in mechanisms and other inertia usually ascribed to existing technologies and infrastructures can and must be analyzed in terms of both inter- and intra-dependencies between human agents and physical, material objects. Within the different types of relationships distinguished by Schatzki, especially the 'prefigurational relationships' are interesting when studying environmental change (Schatzki, 2002, pp. 210–233). These relations refer to the kind of future figurations that are particularly feasible and possible given the existing state of affairs. For example when new objects or technologies are entering the practices, they can display a different level of 'fit' or 'misfit' with regard to the existing order of things. The successful introduction of new products entering social practices thereby comes to depend not just from the 'mental appropriation' of the products by human agents but equally so from the levels of fit or misfit the new products show with respect to the existing portfolios of objects, bodies and meanings involved in the practice. As a result, it can be shown that the portfolios of the actors involved in the reproduction of the practice enable and constrain specific forms of appropriation, normalization, cultivation and naturalization of new objects and technologies (Wilk, 2009; Schot and Albert de la Bruhézé, 2003).

Shove is one of the few scholars to explore the impact of (also environmental) technologies in everyday life from a sociological, practice based perspective. In her path-breaking book 'Comfort, Cleanliness and Convenience' (2003) and a series of related articles

(2006, 2007) she discusses the co-evolution of practices, norms and technologies. She illustrates the sometimes pervasive and autonomous influence of technologies on social practices by analyzing innovations in doing the laundry, showering, cooling and heating. Building upon this 'technology and everyday life' – research tradition, it must be possible to bring the analysis of agency and technology beyond the existing, rather deterministic variants of STS and transition studies on the one hand and the voluntarist theories on the 'adoption of innovations' on the other.

Collins is well known for his in-depth knowledge of the works of both classical and contemporary sociologists, and for his accessible way of writing about sociology (Collins and Makowsky, 1998). It does not come as a surprise, then, when Collins carefully positions and discusses his theory of interaction rituals in relation to existing streams of thinking in the social sciences. When summarized in one sentence, he argues to provide a 're-reading of Emile Durkheim through the eyes of Erving Goffman'. This re-reading is necessary since Durkheim's analysis of the reproduction of norms, moral sentiments and solidarities in society have been given a specific twist by post-Durkheim functionalists who interpreted the reproduction of norms and values primarily in terms of macro- and system categories and phenomena. What has been lost, then, is an emphasis on the situational characteristics and mechanisms implied in the reproduction of culture. With the help of Goffman's micro-version of Durkheim, it must be possible to formulate a more adequate, contemporary theory of interaction rituals. In Collins' hands, this results in a 'radical micro-sociology' which is able to demonstrate in detail how norms and solidarities are produced by situated actors in everyday life. When enacting their daily lives, human agents move through chains of interaction rituals, reproducing along the way forms of solidarity and (also material) symbols of group membership while in turn being charged up with the sentiments, beliefs and emotional energies that result from situated interaction rituals. By putting Durkheim's ritual analyses on their Goffmanian feet, Collins is able to show that "what often appears to be a fixed global culture is in fact a situational generated flux of imputed rules and meanings" (Collins, 2004, p. 8). Especially when in reflexive modernity everyday life experiences become less tightly coupled with the fixed macro-categories of family, church, union and class, the dynamics of situated practices or 'situations' gain strategic importance for social theory.⁷

At the core of Collins' theory are situations. As he states: "A theory of interaction ritual (IR) and interaction ritual chains is above all a theory of situations" (Collins, 2004, p. 3). Paraphrasing Goffman (1969, 1974) in this respect, he argues that his theory is "not about individuals and their interactions, but interactions and their individuals; not persons and their passions, but passions and their persons" (Collins, 2004, p. 5). He develops a conceptual framework for analyzing interaction rituals as shown in Fig. 3. On the left hand side, there are the necessary ingredients for an

interaction ritual to take place: bodily (co)presence, group boundaries, a shared mood and a mutual focus of attention. These ingredients can feed back upon each other and result in a process of what Durkheim labeled as 'collective effervescence': people get excited, become mutually entrained, something is happening. As a result, the ritual produces group solidarity, symbols and objects of social relations, standards of morality and emotional energy (EE) in the individual. With respect to the agency-structure theme, Collins argues that the classical micro-macro distinction is not that obsolete and confusing as Giddens or Bourdieu would have it.⁸ He defines agency in terms of the flows of emotional energies that appear in human bodies during interactions and which travel along interaction ritual chains.

In IR-theory, energy or agency is primarily 'local' or 'micro' in origin since it emerges from situated interaction rituals enacted on a face-to-face basis by individuals who set themselves apart from non-participants. Agency is not restricted to the micro level, however, since the flows of emotional energy also travel along wider chains of interaction rituals in society. The flows of EE are embodied in and carried along by individuals; but they are not 'individualist' in nature. In line with the tradition of practice theories, Collins argues that 'EE-flows running through interaction chains' are more interesting than individuals as topics for sociological theorizing and research. For that reason, emotional energy can be regarded as the single most important new concept introduced by Collins' IR-theory.

Collins puts forward an interesting perspective on the agency-technology interactions discussed above. The interaction rituals which produce EE can take the form of either formal rituals (a religious service, a funeral, a Champions-league final) or 'natural rituals' (most of our everyday life interactions). The role of sacred objects in formal interaction rituals (the Christian Cross in the church ceremony, the coffin at the funeral, the ball in the championship final, etc.) is well known and has been studied by anthropologists and sociologists ever since Durkheim. As sacred objects, they are co-constitutive for the ritual interaction in a direct and visible way. Without sacred objects, formal rituals would hardly be imaginable. As sacred objects they are 'necessary' for the production of formal rituals as specific kinds of social practices. At the same time, symbols can be said to depend on the rituals being enacted on them, since without the ritual interactions they would end up being just 'dead symbols'.

What counts for formal rituals however is also true for the ordinary practices referred to by Collins as natural rituals⁹: they, too, can be said to depend upon and help produce material objects and infrastructures which serve as symbols representing social relations that are important for the participants in the interaction ritual. Rituals are about doing honor to what is socially valued (Collins, 2004; Douglas and Isherwood, 1979; Douglas and Wildavsky, 1982). The mechanism of 'doing honor to objects' applies to ordinary objects being given special significance in the context of natural rituals as much as it applies to divine objects serving in a religious ceremony. What is at stake is the mutually constitutive relationship between objects and human actors: actors acting on things, and things triggering actions and 'producing' agency or in Collins' jargon, 'emotional energy'.

For Collins, the relationship between actors and objects is best analyzed in terms of the amount of EE being stored in and released from objects as a result of interaction practices. Objects can be said to 'act' during the ritual interaction by focusing the attention of the participants and getting them entrained. Human actors bestow material objects with special values, and get excited and energized by these objects in return. Consumers are keen on using objects and symbols in the context of practices which they expect to add to their existing stocks of emotional energy. The ownership and use of products are not ends in themselves for citizen-consumers: products are desired, fantasized about, bought, maintained and divested since the consumption rituals they help organize are expected to deliver an increase of emotional energy.

Instead of being regarded as just the 'users of goods', citizen-consumers are best regarded as EE-seekers, according to Collins (2004). They do not just 'handle' things or consider their fit in relation to the existing portfolios and values belonging to the practices. Citizen-consumers get excited or disappointed, enthusiastic or sad, energized or bored with the objects that co-constitute the practice of consumption. These forms of heightened inter-subjectivity between agents and material objects keep consumption practices going, give them their cultural dynamics. With respect to sustainable consumption practices, it is thus emotional energy from (green) symbols, objects, technologies and infrastructures that drives environmental changes in everyday practices.

Emotional energy gets stored in bodies, objects and symbols, preventing energy from fading away as soon as the interaction ritual ends. By circulating through interaction ritual chains, bodies, objects and symbols keep the flows of EE going. The circulation of EE can be regarded as representing the meso- and macro-forms of agency. When EE flows decrease, objects and symbols lose their original attractiveness or emotional loading. As a result, the practices they help organize will 'fade away' or disappear. In a similar way, when and how new practices and their symbols and EE-flows originate and develop also can be studied.

At this point, Collins makes an analytical distinction between the first-, second-, and third-order circulation of symbols and objects. The first-order circuit is the face-to-face interaction ritual or practice producing the symbol or loading the objects and technologies in the first place. Once the meanings of the objects are established, the symbols can be taken up by participants in other practices and by the media, resulting in second-order circulation of objects and symbols. Local symbols start being used in national, regional or global chains of interaction rituals. Lastly, in tertiary circuits individual human actors perform rituals with objects and symbols outside the presence of others. A person talking to her or his refrigerator, bicycle or computer serves as an exemplary case of such a 'third-order circulation' of symbols and objects.

Collins (2004) advises social scientists interested in the cultural dimension of consumption to consider a number of well-defined methodological steps. First, take a close look at the material object or symbolic item: what does the thing or idea look like; does it have a positive or negative social connotation or load; is it widely shared or restricted to certain groups, etc. Second, look for the practices which produced the symbol in the first place: what kind of practices are typically attached to this symbol or material object; are they traditional or modern, old or new, cool or out of fashion. Third, analyze the flow of the symbols or objects outside their primary circuits of origin: through what kind of interaction ritual chains – or nexus of practices as Schatzki would argue – do the material objects and symbols circulate or travel, and what kind of impacts do they have when spilling over into secondary circuits. Finally, investigate the tertiary circulation of sacred objects and symbols when they are used by actors outside the presence of other people.

This methodology for developing a cultural approach to sustainable consumption is innovative and relevant for two reasons: First, it shows how norms, morals and awareness result from situated interactions in which (also green) objects, symbols and morals are used by individuals to gain emotional energy and to 'produce' sustainability. Second, it offers good (analytical) opportunities to display the positive logics implied in sustainable consumption. The second rationale is worth considering in further detail.

Most environmental debates and research on sustainable consumption are about avoiding particular objects, breaking bad habits, limiting consumption, reducing impacts, and dealing with the shadow side of consumption (Dauvergne, 2008). Most green symbols and practices being studied are 'negative' in the sense of representing logics and dynamics that run counter to the established, positive logics of consumption. The behavioural alternatives suggested, for example, in the context of climate change policies most of the time refer to not doing things or doing things in ways that extra efforts are required. The so-called 'top ten environmental actions' – making possible the most significant reductions in individuals' footprints – always have on top items like 'eat less meat', 'reduce your car use' and 'fly less frequently'. The 'doing your bit' means withdrawing from established

behavioural routines that are typical of the everyday lives of major segments of the population.

Apart from a small group of frontrunners building their green identities from participation in new, more sustainable or 'low impact' practices (Horton, 2003; Etzioni, 2006), the majority of the population is not able yet to connect the use of green products, objects and symbols with processes of display and identity-building in a positive manner. In Collins' terminology: most citizen-consumers cannot imagine how the use of green symbols and products in the context of everyday life practices might result in an increase of their present levels of emotional energies. They do not know yet how to frame the emotional energies they get from subscribing to green energy schemes.

The potentially 'energizing' part of sustainable consumption is hardly considered by environmental social scientists when organizing research on sustainable consumption. So how does it feel to vacuum clean the house with your own photovoltaic electricity, to run an elaborate set of divestment practices¹⁰ in the home, to prepare a meal with organic ingredients?

With Collins' analysis in mind, we can reframe the issue of green commitment, excitement and awareness in a non-individualist, positive manner.¹¹ The more frequent, intense and dense the enrolment of individuals in sustainability-related interaction rituals, the higher the chance that their commitments and levels of awareness will increase. This increased commitment can be studied for the kinds of interaction rituals at play and the particular green symbols and objects used to develop and carry along 'green emotional energies'. When green objects and technologies start to perform well across a range of different interaction rituals they can be said to be of strategic (policy) significance for the greening of interaction ritual chains within specific consumption domains.

Within the domains of food, housing or mobility, for example, different sets of objects and technologies can be said to carry along the flows of emotional energies attached to sustainable consumption. Farmers markets can be shown to attract people who also tend to subscribe to local food networks, who like to grow vegetables in their home garden and who go dining out in restaurants which offer a rich variety of local food items, etc. The flows of food-stuff, eating habits, production methods and recipes traveling through the food-interaction ritual chain must be specified in empirical research and assessed for their EE-enhancing capacities (Mol and Spaargaren, 2005). Instead of 'spill-over' being a characteristic of individual lifestyles only (Thøgersen, 1999), this methodology offers a new way of studying 'green contagion', including at the level of chains of practices. In this way, we can start answering research questions on the cultural dimensions of sustainable consumption, while keeping in line with the main tenets of practice theories. We think such a redefined, post-Brundtland (WCED, 1987) agenda emphasizing the positive, EE-enhancing logics of the greening of consumption would add to the existing scholarship and policies on sustainable consumption in some crucial respects (Spaargaren and Cohen, 2009). It is to this renewed, global agenda for sustainable consumption practices that we turn in conclusion.