Source
Content
This article examines Philip Agre's 1994 paper "Surveillance and Capture: Two Models of Privacy," presenting it as a foundational Science and Technology Studies (STS) text. The authors discuss how Agre contrasts two approaches to gaining knowledge about people: surveillance (based on visual metaphors and secret police models) and capture (based on linguistic metaphors and computerized tracking).
Key Concepts
Capture involves creating "grammars of action" -- dividing activities into basic units arranged in sequences to make them representable in computers. The concept serves as a mid-range theoretical tool applicable across various empirical settings, similar to concepts like boundary objects or blackboxing in STS.
Five Distinguishing Qualities
1. Middle-Range Theoretical Concepts Capture functions as a propositional abstraction describing how computers reorganize activities for real-time representation, rather than claiming final authority on the phenomenon.
2. Subversion of Dualities The concept rejects both technology-versus-society dualism and utopian/dystopian binary thinking, instead examining "complex patterns of negotiation and control" operating multidirectionally.
3. Grounded Empirical Research Agre grounds analysis in detailed, accessible descriptions of specific technologies -- barcode scanning, GPS tracking, location bracelets -- without overwhelming lay readers with technical minutiae.
4. Work and Organization Focus Rather than examining scientific representation, Agre analyzes how computing professionals formalize work practices, revealing assumptions that "almost anything in the world can be described for computers and made computable."
5. Enduring Relevance Despite 26-year-old examples, the theoretical framework applies to contemporary developments like gig economies and platform surveillance, demonstrating the concept's longevity.
Representation and Intervention
A crucial insight: capture appears to neutrally "represent" underlying activity structures but actually intervenes and reorganizes them. The article notes that formalizing activities creates a "myth" that grammars were "discovered" rather than constructed -- a Platonic assumption that formal structures preexist human activity.
Modern examples illustrate this: workplace time-tracking systems reshape how work is valued; LinkedIn formalizes "maintaining contacts"; platforms like Discourse model conversations as information exchange in turn-taking sequences.
Contemporary Applications
Agre's prescient analysis foreshadows current developments. His suggestion that capture reduces transaction costs prefigures gig and click-work platforms. His observation that "the control function previously provided by bureaucracy is transferred to the inherent discipline of the market" describes modern platform economies remarkably well.
The article concludes these five qualities define effective STS scholarship: theoretical sophistication, non-dualistic thinking, empirical grounding, focus on work contexts, and conceptual durability.