Choices about what is salient

A couple of paragraphs from the PhD thesis, unravelling context, having drawn too close to the large language model and been spaghettified.

Context often appears as an explanatory staple and normative imperative; a self-evident, black-boxed set of ‘external circumstances’ to which objects must yield (Woolgar and Lezaun, 2013). In his writings on context-aware computing, Paul Dourish takes a different approach, differentiating between representational and interactional interpretations of context (Dourish, 2004). In representational accounts, context denotes the features of a given setting. For interactionalists, however, context is something people do; a localised achievement, with tangible material effects, actively produced through social practice (Seaver, 2015: 1105; Asdal and Moser, 2012: 294; Dourish, 2004: 22). Where calls to contextualise typically assume issues and objects are separate from their surroundings, ‘establishing context’ is a political project, bound up with questions of method and practice (Seaver, 2015: 1106). Since, for interactionalists, context is not self-evident, any analysis of the boundaries between context and content entails choices about what is salient, and ‘where the work of contextualization should cease.’ (ibid.)

For anthropologists, an early emphasis on holism and local field sites has, in recent decades, yielded to a relational understanding of context as comprising a ‘set of connections between the object of inquiry and its surroundings.’ (Morita, 2014: 215) As anthropologist Nick Seaver observes, while ‘it is common sense that to put things in context is good and to take them out of context is bad’, this has often overshadowed disagreements over what, exactly, context is (Seaver, 2015: 1105). At the same time, much work in STS has been shaped by an imperative to avoid either explaining science by its own internal logic or reducing it to a function of its social environments (cf. Asdal and Moser, 2012). In both fields, a once-common explanatory recourse to context has faltered, finding ‘nothing “beyond” [or outside] with which to explain something else.’ (ibid: 295)

Readings