While the photo is titled “How things fit,” this post will also discuss how things don’t fit neatly, how friction can be productive, and how spaces of controversy can provoke learning. In the Education Policy article below, I discuss how I’ve used actor-network theory as an organizing methodological guide.
In the following excerpt from a manuscript, I am writing with Radhika Gorur (first author), the usefulness of ANT thinking is further discussed.
Our analysis draws broadly upon the conceptual resources of material semiotics, more particularly actor-network theory (ANT). Focusing on processes and practices, ANT is particularly useful in the study of controversies (Latour, 2005; Venturini, 2010), characterised by the struggle of various groups to establish the authority and legitimacy of ideas and practices. Deployed in policy study, an ANT analysis might trace how policy phenomena emerge as contingent effects of socio-material practices. The emphasis is on how certain policy ideas come to cohere as more-or-less durable assemblages or networks, and how they are mobilised, challenged, defended and strengthened. Here, we focus on the current struggles to promote like-school comparisons as authoritative, technical and apolitical; the publication of these comparisons as an appropriate register of accountability; the expertise of statisticians as valued; and the practice of such comparisons as an appropriate and morally compelling policy process; and we trace the attempts to challenge and resist these moves.
As Law (2007, p. 2) reminds us, ANT has been taken up by different researchers in different ways. Rather than a single, coherent or strong ‘theory’, Law suggests that ANT is ‘a sensibility to the messy practices of relationality and materiality of the world,’ bringing with it ‘a wariness of the large-scale claims common in social theory.’ So rather than accepting the logic of the taken-for-granted as an inherent quality of phenomena, ANT researchers attempt to understand how some phenomena come to be accepted as logical and commonsensical.
When undertaking policy studies, ANT researchers trace ‘the specific materializing processes through which policymaking actually works to animate educational knowledge, identities, and practices’ (Fenwick & Edwards, 2011, p. 710). Policy texts, particular devices such as like-school comparisons, websites, money, expertise – these are not only vehicles which inscribe and translate human agency into durable and distributed effects, but could also be actors in their own right. For instance, like-school calculations serve to cohere and promote certain understandings whilst discounting others. They serve to organise thinking and sort information. They translate such entities as students, teachers, schools, learning, teaching, curricula, cultural capital and motivation – in short, all the complexity of schooling – into a limited set of discrete actors with observable and measurable attributes to effect various calculations. This translation into numbers makes it possible for diverse and particular aspects of schooling to be shuffled together into seemingly standardised, universal sequences of logical causes and effects. Following Callon and Muniesa (2005), we understand ‘materiality’ not so much as concerned with physicality, but as pertaining to the investment of observable and measurable attributes to abstract phenomena such as ‘quality’ or ‘equity’ to make them coherent and calculable.
Enriching our analysis further are some key concepts elaborated by Callon, Lascoumes, & Barthe (2001) who illustrate how the confidence of technical answers to technicised policy dilemmas may come to be challenged by diverse and lay groups and reassembled as a socio-technical controversies. In our study, the translation of school quality into the like-school comparative ‘league tables’ depends on the work of a small group of experts in statistics and psychometrics. So complex and specialised is their expertise that the actual processes by which like-school comparisons are produced is a black box – we are required to accept the result, but the process itself is too technical for most of the actors to understand. Indeed, such calculations are not only inaccessible to non-experts in terms of comprehending them, but also in challenging them. But when these calculations are made public, the public become ‘informed’ and is able to debate and challenge policies. The confident technical accounts begin to unravel, creating productive ‘spaces of uncertainty’ (Callon, et al., 2001) where diverse groups bring new ideas and concerns into the policy arena, elaborate the problem and the range of considerations, and seek better solutions. The nature and extent of such challenges to the certainty and confidence of technicisation, and the way these challenges are managed, are an empirical matter, as we find at our two sites.
Although we describe the fortunes of a particular policy decision at two very different locations, our primary aim is not to compare one site with another to draw policy lessons. Rather, it is to illustrate a more general (and more modest) point – that technical accounts may be constructed and challenged in a variety of ways, and that the processes by which policies are promoted and challenged are emergent, contingent and contextual. Doing so, we extend the conversation on what Grek and Ozga (2010), following earlier work by Grek, refer to as ‘informal ‘networking’ forms and its reliance on such policy technologies as benchmarks, indicators and the circulation of data’ (p. 938). Our work thus lies within what Whitty (2006) has termed ‘education research’ (rather than ‘educational research’)—research of education that is not ‘consciously geared towards improving policy and practice’ (p. 173) as so often sought and funded by governments.
If you’d like to know about how to use assemblage in studies of educational policy, see the work of Fenwick and Edwards and Gorur.