Making Certain: Information and Social Reality

Abstract: This dissertation identifies and explains the phenomenon of the production of certainty in information systems. I define this phenomenon pragmatically as instances where practices of justification end upon information systems or their contents. Cases where information systems seem able to produce social reality without reference to the external world indicate that these systems contain facts for determining truth, rather than propositions rendered true or false by the world outside the system. The No Fly list is offered as a running example that both clearly exemplifies the phenomenon and announces the stakes of my project.
After an operationalization of key terms and a review of relevant literature, I articulate a research program aimed at characterizing the phenomenon, its major components, and its effects. Notable contributions of the dissertation include:

• the identification of the production of certainty as a unitary, trans-disciplinary phenomenon;
• the synthesis of a sociolinguistic method capable of unambiguously identifying a) the presence of this phenomenon and b) distinguishing the respective contributions of systemic and social factors to it; and
• the development of a taxonomy of certainty that can distinguish between types of certainty production and/or certainty-producing systems.

The analysis of certainty proposed and advanced here is a potential compliment to several existing methods of sociotechnical research. This is demonstrated by applying the analysis of certainty to the complex assemblage of computational timekeeping alongside a more traditional infrastructural inversion. Three subsystems, the tzdb, Network Time Protocol, and International Atomic Time, are selected from the assemblage of computational timekeeping for analysis. Each system employs a distinct memory practice, in Bowker's sense, which licenses the forgetting inherent in the production of the information it contains. The analysis of certainty expands upon the insights provided by infrastructural inversion to show how the production of certainty through modern computational timekeeping practices shapes the social reality of time. This analysis serves as an example for scholars who encounter the phenomenon of the production of certainty in information systems to use the proposed theoretical framework to more easily account for, understand, and engage with it in their work. The dissertation concludes by identifying other sites amenable to this kind of analysis, including the algorithmic assemblages commonly referred to as Artificial Intelligence.

A copy of the dissertation is available for download:

Download PDF

Suggested citation:
Hauser, E. 2020. “Making Certain: Information and Social Reality.” Doctoral dissertation. University of North Carolina at Chapel Hill.