hschaeferIn this contribution, I would like to sketch out both my take on practice theory methodology and its relationship with my current research on cultural heritage. But first of all, I want to start with some remarks about how this blog came about.

My PhD thesis centred on the question of how practice theory can overcome its focus on stability and routine and instead open up its perspective for the dynamics of the social (Schäfer 2013). As Elizabeth Shove, Mika Pantzar and Matt Watson have dealt with similar issues in their book The Dynamics of Social Practice (2012), I engaged in a conversation with Elizabeth Shove, which led to a research visit at Lancaster University in 2015.

Amongst other things (like improving my floorball skills), my research visit highly benefited from conversations I had with members of the department of sociology and the people from the DEMAND Research Centre in particular. I had the opportunity to present my work in a seminar, which led to conversations with Allison Hui about the methodological aspects of practice theory. This blog is an outcome of our exchange, which we would now like to continue with other colleagues. It is also linked to a session at the DEMAND Research Centre 2016 International Conference in Lancaster, UK.

In my current research, I am using practice theory as an analytical framework for understanding cultural heritage. It focuses on the notion of “world heritage” and conceives of cultural heritage as a flow of practices and materialities through tim
e, which is the object of specific practices of labelling and physical manipulation. My perspective is based on the contention that both the production and the understanding of cultural heritage are shaped by events occurring in distant places, by historical changes, competing principles, embodied competences, documents and the specific material qualities of artefacts. Sociological analysis has to follow these diverse and complex connections. It has to deal with their heterogeneity on the level of method. Not a singular object, not an isolated action, but a bundle of relations connecting different times, places and entities needs to be taken into account.

In order to grasp these relations, no analytical category can be viewed as a safe ground for a study, although it might be used as a starting point of any given analysis. But its integrity can always be questioned and research thus needs to stay flexible and follow relations between very different, heterogeneous entities. Practice theory takes account of acting subjects, but these subjects are formed in discursive and non-discursive practices; their identity is socially constituted and their agency relies on always already existing and repeating practices. These practices in turn might entail the use of specific material entities. Practice theory also takes account of bodies, which are the location for incorporated dispositions and tacit knowledge, but they are themselves constituted by practices, which shape them in order modify their surface, their meaning or even their very ability to contract and maintain dispositions as “carriers” (Shove and Pantzar 2007). It also takes account of material entities, which shape actions, but which in turn are shaped by non-discursive and discursive practices. And centrally, practice theory takes account of practices, which repeat themselves throughout the social realm and are repeated by the participants; these practices are performed, identified and classified according to categories embodied in the subjects’ dispositions. But practices need the specific material qualities of artefacts and bodies in order to span time and space, both of which also determine their particular repeatability.

Depending on the research question, practice theory constantly needs to shift its focus in order to follow the multiple connections between heterogeneous elements linked in a relational network. It needs to explore these links and look into the occurrences at their intersections. I propose to call this research approach a “transitive methodology”. This draws on the specific connotations of the word “transitive”, which is used in logic and mathematics to describe a link between members of a sequence and which derives from the Latin “transitivus”, meaning “going across” and “connecting”.

The transitive methodology of practice theory takes diverse forms of social associations into account, focusing on the transitional shifts that occur at the links between elements. Its first principle is relationality. Instead of locating agency in the intentions of a subject, action is distributed in a network of relations. These exist in time, as practices are repetitions that link former, current and possible future appearances. They extend in space, as a practice occurring at one spot might influence practices at some other spot or re-occur in a different situation. Relations also exist between different sorts of entities. Accordingly, the second principle of transitive methodology is heterogeneity. Elements linked in a network are of different kinds and are characterized by specific qualities. Sociological explanations must move beyond a mere concern with so-called “social structure”; sociology needs to be open for different sorts of interrelations, as ANT has shown. A hammer, a specific software, the process of writing a list or the usage of a specific category to designate a group of people: all of these heterogeneous elements are part and parcel of forming “the social”. The third principle requires that analyses become aware of gradual difference. There are always gradations, nuances between what has usually been understood as binary terms. These false dichotomies include: individual vs. social, nature vs. culture, thinking vs. action, social vs. material, inner vs. outer realm and so on. Instead of trying to identify these categories in supposedly pure states, the transitive methodology accepts impurities and the messiness of the social (for similar proposals see Law 2004 and Clarke 2005). It takes a close look at the processes that try to establish purities as well as at the failures these processes entail. It is also open to temporal changes in the perception, value and specific contribution of each entity. The fourth principle requires the analysis to move in time and space, following the links that exist between the heterogeneous elements involved in a network and trying to connect the dots, so to speak. Practically, this means that different research methods must be combined and different locations visited as Bruno Latour (2005) proposes in his methodological call to localize the global, to redistribute the local and to connect sites. It also entails the need to ground the analysis in a historical perspective as Pierre Bourdieu in particular demanded of sociology. Finally, the fifth principle makes the analysis aware of the shifts that occur at any spot in a network where a change takes place. This is one of the central lessons learned from ANT. If the link passes over to a different sort of element or if any element is altered, this changes the quality of the whole network. Thus, although the alteration might serve the same function at first glance, the analysis has to take a close look at specific differences. The hypothetical substitution of elements, as in Latour’s discussion of effects that would occur if entities are altered, is a good research method for assessing these differences.

Let me give some brief examples of how transitive methodology can be put into practice. The following ideas are taken from my current research on cultural heritage and the specialized and vernacular practices it is embedded in. First, if by way of example we look into processes of restoration and focus on a restorer engaged in handling a historic object, the analysis needs to take a multiplicity of relations into account. It is not sufficient to limit the perspective to her intentions or the social interactions she is currently engaged in. Instead, the analysis has to understand restoration practices as spatiotemporal repetitions that rely on competences, which have been physically acquired and embodied in the past. They don’t only encompass knowledge in art history, but also specific knowledge about different materials and craft skills. Accordingly, we have to consider the educational background of restorers and might be finding conflicting ideas about the purpose and practice of restoration. These are often linked to manifests or other significant documents. So we will also need to look at text books used in the education of restorers or at international treaties like the Venice Charter (ICOMOS 1964), which specifies a particularly modernist understanding of restoration. Thus, as stated in the second principle, the analysis needs to consider a network of heterogeneous elements. Of course, the actual object our restorer is engaged in repairing also imposes its own material demands on the situation: whether she is dealing with wood, stone, clay or paper will entail gradual differences both in terms of their ability to withstand the passage of time and in terms of the techniques she can use for securing their future. A sociology of cultural heritage which investigates these relationships will need to move in time and space, tracing practices back into the past or mapping their spatial trajectory. This involves moving between different places in order to follow translocal effects, e.g. drawing a connection between the institutional decision to include a site in the list of “world heritage”, which will be made by a UNESCO committee somewhere at an international conference, and the effects this will have on the actual location. A multi-sited ethnography as proposed by George E. Marcus (1995) provides a model for exploring some of these complex relations. And finally, in terms of focussing the changes and shifts in heterogeneous networks, the digitization of cultural heritage is an interesting case in point. Often, museums turn to showing media representations of their exhibits instead of the original objects. How does a video clip of an exhibit change its perception and appreciation? What is gained and what is lost when we peruse a website giving us a two-dimensional rendition and additional information on an object that is not on display itself?

I am looking forward to exploring the implications of this proposal in discussion with other colleagues’ comments and posts.

Hilmar Schäfer

Academic Fellow, Europa-Universität Viadrina Frankfurt (Oder), Cultural Sociology



Clarke, Adele E. (2005) Situational Analysis. Grounded Theory After the Postmodern Turn. London/Thousand Oaks: SAGE.

ICOMOS (1964) International Charter For The Conservation And Restoration Of Monuments And Sites (The Venice Charter 1964). Venice.

Latour, Bruno (2005) Reassembling the Social. An Introduction to Actor-Network-Theory. Oxford: Oxford University Press.

Law, John (2004) After Method. Mess in social science research. London/ New York.

Marcus, George E. (1995) Ethnography In/Of the World System: The Emergence of Multi-Sited Ethnography. Annual Review of Anthropology 24:95–117.

Schäfer, Hilmar (2013) Die Instabilität der Praxis. Reproduktion und Transformation des Sozialen in der Praxistheorie. Weilerswist: Velbrück.

Shove, Elizabeth, and Mika Pantza (2007) Recruitment and Reproduction: The Careers and Carriers of Digital Photography and Floorball. Human Affairs 17 (2):154–167.

Shove, Elizabeth, Mika Pantzar, and Matt Watson (2012) The Dynamics of Social Practice. Everyday Life and How it Changes. London: SAGE.