Performing Technology's Stories: On Social Constructivism, Performance, and Performativity

Ed Constant's recent article "Reliable Knowledge and Unreliable Stuff" is an attractive, graceful, and more than occasionally witty description of the growth of rational engineering belief. ¹ In particular, it offers an account for the fact that our confidence in relevant scientific and technological theory tends not to be eroded by the apparently disconfirming instances endlessly thrown up in everyday practice. His argues that rational belief in generalizable knowledge is a feature of engineering, technology, and science, and he offers a Bayesian account of how such knowledge spreads across engineering time and space. His account is positive in tone. His interest is in the reliability of engineering and scientific knowledge. At the same time, as he notes, his approach is not consistent with certain historical and sociological approaches to engineering and technology, in particular with "social constructivism."

Ed observes that much has been learned from social constructivism. On the other hand, its deconstructive microstudies tend to emphasize the contingency and uncertainty of technology and lose sight of the fact that most of the time engineering knowledge works--and, indeed, tends to extend itself. This means that social constructivism tends toward relativism, which in turn means that it cannot offer rational political criticism of science and technology.

Social constructivism is a tricky target. As Ed notes, it comes in many shapes and forms. It may be useful to distinguish, somewhat arbitrarily, four of these. The sociology of scientific knowledge (SSK) is a version of constructivism that would, we judge, have little difficulty with Ed's Bayesianism. Indeed, it developed a similar Bayesian approach in the 1970s, when it argued that scientific (and technological) practice and knowledge reflect not only the natural world but also social influences--for instance, of professional position, social class, or gender. It is these two together--natural and social factors--that give knowledge its shape, an insight that has been explored in many empirical contexts. ²

Second, and in contrast with this, some versions of SCOT (the social construction of technology) have argued that the natural world has no role in shaping technological practice and belief, which are taken to be a function of social forces alone. SCOT-like studies vary greatly, but some focus on the social alone, with consequences that are much closer to the relativist constructivism that Ed questions. ³

Actor-network theory (ANT), in further contrast, assumes that new hybrid social-and-material practices are constrained and enabled by equally hybrid preexisting practices. This means that new practices imply theories and versions of the social and the material world that may differ from those that existed before. Nevertheless, because of the backdrop of existing practice such differences tend to be limited, and the world is sensed--indeed is constituted--as solid and obdurate. Actor-network theory is not relativist, but neither is it realist. Deconstruction is always possible but, given the backdrop of existing practice, also very difficult. Social and technological knowledge, the social world, and its material context are all obdurate--indeed translocal, since they carry from place to place in the textures of practice. ⁴

Feminist technoscience studies vary, but some, like actor-network theory, assume that social and material practices recursively generate new social and material practices, technoscientific knowledges, and versions of the social and material world. This approach is, however, more political in its concerns, attending centrally to the way in which such practices carry (for instance) gender, ethnic, class, and military agendas. It also insists that there is no neutral place outside society, and that every description of the world also participates in social and material agenda-setting. ⁵ Finally, and crucially, it insists that when one writes one also intervenes: writing may either support or erode current technoscience agendas. ⁶

Social constructivism is indeed, as Constant writes, "a veritable and prolific zoo of theoretical perspectives" (p. 325), but the differences among them are...