eshovepicIf only I had got round to responding to these propositions earlier! If I had contributed in April 2016 – as was my plan – this task would have been so much easier: 4 lines and not 4 pages. In April, I knew what I wanted to write. Having read the blog and been part of discussions at the DEMAND conference, I simply wanted to add an 8th proposition which went as follows:

Taking “practice” as a central conceptual unit of enquiry generates a range of distinctive questions. The choice of methods depends on which of these questions you want to take up and pursue. Using practice theory is thus not directly tied to certain methods, but the choice of methods is – as always – dependent upon your specific research question.

At that point, that was all I had to say.

I still hold this view (with some qualifications… see below) – but in explaining what I mean and why, it is useful to back track a bit and also take stock of how this position fits (or doesn’t) with the contributions that others have made to this blog.

For me, theories matter for how problems are defined and for how lines of enquiry are formulated. At a minimum they are sensitising devices. Specific questions are situated within theoretical paradigms. Research involves following some of these questions – and of doing so in ways that are commensurate with the ‘sensitising’. But since there can be, and usually are, many such questions: there is no single or simple link from theory to method. I don’t recall any major debate about what methods suited structuration theory. It is true Actor Network Theorists laid claim to certain methodological principles, ‘Follow the actors’, but that is perhaps because ANT is more a clustering of concepts than a distinctive ‘theory’.

In any event, it seems that more and more people are interested in learning how to ‘use’ practice theory in their research. And across this blog, there are various attempts to help out and respond to just this question. But what does it mean to ‘use’ a theory in this way?

The responses represented here, and elsewhere, suggest that there are multiple (mis)understandings of what ‘theory’ is, and what it means for empirical research. For instance, Ellsworth-Krebs and Reid say ‘phenomenological and innovative qualitative methods (e.g. sensory ethnography and videos) should have more of a place within social practice theory’. I think this is ‘code’ or shorthand for saying that these authors are interested in following certain questions about practices which are suited to qualitative enquiry. There is nothing wrong with that, but it is wrong to imply that questions inspired by practice theory are or should be of this kind. In what sense should methods have a place in theory?

A more common tendency is to suppose that it is possible to study or to write about practices, and that this is a meaningful topic in its own right. For instance, Smits and Rinkinen wonder about ‘what you need to know to be able to write about practices in other countries?; and Maller and Strengers claim to ‘have used theories of social practice (Reckwitz 2002; Schatzki 2002; Shove et al. 2012) to study a number of mundane practices’. In sentences like these the critical part is missing: namely, what is it about practices that these researchers want to study: what is it that practice theories have sensitised them to? It is definitely not the case that ‘mixed methodologies are always more illuminating’ (Browne). Whether they are or not depends on the nature and purpose of the enquiry. More generally, under-specified discussions of practice theory and method, of ‘practice theoretical research’ or of ‘studying practices’ gets in the way of a more considered engagement with the blog’s seven propositions (though I have troubles with these too… ).

Setting generalised talk of ‘practice theoretical research’ aside, and following my suggested eight proposition, I am interested in thinking about whether practice theories generate a distinctive repertoire of empirical questions.

I think the answer here is partly yes and partly no. It is ‘yes’ in that compared with other intellectual traditions there are definite methodological consequences associated with taking practices as the central topic of conceptualisation. E.g. answers are not sought in psychological analysis, or in macro-economic modelling. But having made that conceptual move there are still many, many puzzles to explore.

For instance, many researchers are broadly interested in accounting for change: hence they are interested in questions about how practices evolve and stay the same, how they circulate or disappear, or how they link and break away from each other. For Schatzki, in this blog, the primary task is (often) to provide accounts of how things come about. There is definitely more to say about what kinds of explanations such accounts work with, what makes an ‘overview’ – and what distinguishes an overview inspired by practice theory from one framed in some other terms. Nonetheless, the challenge is to find ways of capturing – whether by zooming in and out or some other comparative method – the dynamics of practice over certain spatial, temporal scales. Just what that involves is likely to vary from case to case but the point is that the chosen methods are harnessed to this project.

Hilmar Schäfer’s own blog entry provides a great illustration of how a specific framing of questions leads, with care, to an equally specific collection, analysis and interpretation of materials. He has an argument – he sees ‘heritage as a flow of practices and materialities through time; 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’ and this is vital for how he ‘reads’ empirical materials. This close-coupled reasoning is where the action lies. This is where theory and method intersect, not in generalised statements or anxieties about whether anthropological studies can or cannot reveal some ‘big picture’.

There can be no ready-made guide as to how to proceed: each theoretical puzzle presents its own methodological challenges. For example, I am currently struggling, with others, to find ways of detailing how urban infrastructures and practices shape each other in specific locations. On the one hand we know this goes on. We also know that neither infrastructures nor practices are geographically bound; that they are mutually shaping, and constantly on the move. So what’s an appropriate method of showing their situated interaction? Maybe we can take land use data as a proxy for broad shifts in ‘activity’ as recorded in official statistics. Maybe we can link this to timelines of infrastructural investment. What makes sense, or not, is bound up with exactly what we want to know and why. And as with any social science research there are endless decisions to be made, defining and bounding – slicing topics of enquiry out of the flow, steadying them as empirical objects, and managing them in such a way as to enable various forms of description, accounting, comparison, overview and analysis.

Many of these struggles are simply the stuff of social science – they arise in a particular way because of the practice theoretically framed problems in view – but the struggles are not new. There are endless examples of excellent, non-practice theoretically framed, studies which see the ‘small’ in ‘the large’ and vice versa – both as an analytic strategy, and as a ‘method’ of enquiry (Goffman, Mintz…). Though some contributors to the blog are bothered about the methodological implications of a ‘flat’ ontology – does this mean small scale, close observation etc.? ( ‘Often ethnography is seen as a natural ally of practice theory. Observing what is actually done in a given situation with all its things, bodies, signs, texts etc. promises a methodological rigour allowing us to remain flat.’ [Röhl]) – I don’t think this is really a methodological issue. It’s the ontology that’s flat, not the social world itself. In other words ‘flat’ does not mean lacking in mountain ranges of inequality and power, nor does it mean being incapable of explaining how exceptionally uneven landscapes emerge and change – potentially via a full gamut of methodologies and empirical enquiries – it’s just that there is a certain way of conceptualising what’s going on. In other words I agree, for sure ‘we can finally have the cake (remain ‘ontologically flat’) and eat it after all (talk about transsituative phenomena)’ (Rohl).

So what about studying ‘people’ – or using them as informants about the practices they carry (Hitchings and responses). Again it is important to be clear about the relation between theory, analytic approach and method or technique. I agree with Smits and Rinkinen, ‘practice theory decentres the subject’ but does it therefore ‘also call for methods that decentre the subject’? Perhaps, and sometimes, but only in a sense. For example, writers like Lave and Wenger, who are interested in how people are recruited to practices, quite reasonably focus on cohorts of practitioners: some new, some more experienced. These are the informants with whom they talk. What else would it make sense to do… how else to proceed with the questions they have in mind? The key point is that whatever their method (in the narrow sense of technique), the analysis does not revolve around features like the personalities of the actors involved. So it is not a matter of whether people ‘can’ talk about practices, or whether ‘observation’ or even participation is a necessary part of practice theoretically inspired enquiry: instead the real issue is what is it that the researchers want to discover, what is it they are interested in, and what follows from that concern?

Another theme – here thought about in terms of whether practice theories prompt distinctive lines of enquiry, or not – has to do with the body. I am intrigued, and this is probably a longer story, by the attention given to ‘the body’. But I can also see that ‘the body’ figures in very different ways not just within variants of practice theory, but also within different lines of enquiry springing from these various starting points. In Hillebrandt’s contribution, ‘Sense emerges in praxis and enables the association between bodies and artefacts…’ and by implication, any account of ‘sense’ needs to engage with these relations. However, it is equally possible, and consistent with other genres of practice theory, to play down the ‘sense’ aspect and instead focus on the body as an admittedly complex ‘material’ the fluid qualities of which evolve along with the practices in which that body is engaged, together with a multitude of other ‘material’ arrangements/elements. This is a rather different story, bringing with it a different set of more specific questions, and a different repertoire of plausible/consistent analytic and methodological strategies.

Finally, and this really is a practice-theoretical-methodological issue: do research methods and questions which are commonly adopted and pursued inadvertently squash and/or obscure or overlook potentially important forms of investigation which are hard to grasp with these tools? I like Laube’s point that if we are interested in how it is that ‘each practice is several, is many, is a profusion of itself; methods do not sensitize us well for the differences, complexities, fractions and concurrences of our research objects’. Studying practices as if they were discrete arguably prevents a consideration of how, both empirically and theoretically, we might recognize that “the social is a field of embodied, materially interwoven practices” (Schatzki, 2001, p. 3). Put differently, what questions are lost or ruled out by treating practices as ‘units that can be sampled and investigated in order to generate new empirical data (and through analysis, potentially new concepts).’ (Hui).

In the end, I’m not sure this is a practice-theoretically specific concern. Delineating practices, or relations between practices, as objects of enquiry is part of the process – for Hillebrandt, ‘it is necessary not only to have a precise determination of objects, as attempted here, but also a methodical instrument proper for the determination of objects’ and I agree… ‘Research simply cannot be begun without having theoretically determined what should be investigated.’

I now revisit the seven propositions and wonder again about the value of adding an eighth.

  1. Established methods offer both opportunities and dangers for working with practice theories. – but I ask, repeatedly, working with them to do what?
  2. Practice theories are used to frame problems such a way as to (my addition in italics) make specific methodological demands of those who work with them.
  3. Considerations of space, time and embodiment are essential aspects of practice theory methodologies. – I say there are no distinctive practice theory methodologies. There are questions inspired and underpinned by practice theories, and as it happens, these often call for paying attention to these features but not always.
  4. Inventive and multiple methods, units, samples, etc. are particularly useful for exploring practices at different scales, in relation to changing social patterns and variably interconnected actors. – I say, what is it about the practices that is being ‘explored’ – sometimes multiple methods might be handy, but definitely not always – it depends on what it is about practices that is the focus of attention, and why.
  5. Practice theoretical methodologies (I say there is no such thing) need to be able to capture macro structures or large scale phenomena. – I disagree with this formulation but I would say that practice theoretically inspired agendas often frame problems in ways that call for this kind of analysis and explanation.
  6. Practice theoretical methodologies (I say there is no such thing) always have to adapt to the field of practice they are studying. – isn’t that the case for most empirical research?
  7. Practice theory is a conceptual framework which is particularly susceptible to change. It is shaped by its empirical findings. – probably not, I can’t see why practice theories would be special in this regard…

And I added proposition 8, which I now qualify a bit…

  1. Taking “practice” as a central conceptual unit of enquiry generates a range of distinctive questions (Having written this, I’m now not sure how distinctive they are… discuss!!). The choice of methods depends on which of these questions you want to take up and pursue. Using practice theory is thus not directly tied to certain methods, but the choice of methods is – as always – dependent upon your specific research question.

Elizabeth Shove

Professor of Sociology and Co-Director of the DEMAND Research Centre