When Gunfire Shatters Bone: Reducing Sociotechnical Systems to Social Relationships

I am perplexed by much of Grint and Woolgar's (1992 [this issue]) commentary, since I share some of the positions that they argue that I should hold (e.g., that the capabilities of technologies, practices of using them, and the consequences of these usage practices are contingent and socially mediated). And they claim that I hold at least one position that I do not (self-acting guns). I am given limited space in which to reply to their interesting but sometimes confused commentary. Consequently, I encourage readers to review my original article and rejoinder rather than to repeat many of my analyses again. Grint and Woolgar implicitly question the ascription of any given range of capabilities, such as lethal capacity, to any technology-whether 22caliber pistols, AK47 assault rifles, paring knives, Scud missiles, roses, or Jaguar sedans. They insist on a rigid dichotomy between a strong essentialism that links undeniably fixed capabilities and deterministic effects to a given technology or complete indeterminacy. In their view, the way that the capabilities, usage, and consequences of a technology are socially mediated undermines the possibility of strong narrative claims. They often imply that all possibilities seem comparably plausible. In their analysis, possible, plausible, and likely social patterns play comparable explanatory and illustrative roles. We agree that the capabilities of technologies are not fixed and depend upon social relationships between users, technologies, and other close-in participants. We agree that any relationship between technologies and social effects are socially produced and mediated through complex social networks, institutionalized arrangements, and systems of belief and communication. The likely range of lethal capacities of a particular kind of gun can depend