Home/red-rock-eater/Philip Agre on Privacy and Big Data

Philip Agre on Privacy and Big Datawriting

privacycapture-modelagre-secondarybig-data
1 min read · Edit on Pyrite

Source

  • Author: Gordon Hull
  • Date: July 17, 2019
  • URL: https://www.newappsblog.com/2019/07/philip-agre-on-privacy-and-big-data-part-1.html
  • Publication: New APPS: Art, Politics, Philosophy, Science
  • Note: The original URL (`/2019/03/philip-agre-on-privacy-and-big-data.html`) returned 404; the correct URL uses the July 2019 date path. A Part 2 exists at https://www.newappsblog.com/2019/08/lessons-from-agre-on-privacy-as-capture-part-2-can-foucault-get-past-panopticism.html

    • Content

    Gordon Hull examines Philip Agre's influential 1994 paper "Surveillance and Capture," arguing it deserves greater attention in contemporary privacy discussions, particularly regarding big data.

    The Surveillance Model's Limitations

    Hull explains that traditional privacy frameworks rely on a surveillance model emphasizing visual metaphors, state actors, and territorial invasions. However, this approach proves inadequate for modern data challenges. As Hull notes, "The surveillance model isn't adequate to privacy worries now," citing persistent problems with consent frameworks and the "nothing to hide" argument.

    The Capture Model

    Agre proposes an alternative framework based on linguistic and grammatical concepts. The capture model operates through:

  • Identifying atomic elements (people, packages)
  • Developing grammars describing their possible movements
  • Imposing these systems on actual activities
  • Measuring and calibrating based on the established patterns

    • The Critical Slippage

    Agre identifies a fundamental mythology in this process: the claim that grammars are "discovered" rather than invented. Hull emphasizes that "data-driven systems reconfigure their environment to gain access to more data, turning both our environment and ourselves into data engines."

    Practical Examples

    Hull illustrates capture mechanisms through social media examples:

  • Facebook's "friend" categorization flattened social networks
  • Users perform identity strategically for different audiences
  • Content flagging systems operationalize impoverished vocabularies

    • Grammatical Colonialism

    Hull concludes by invoking Tejaswini Niranjana's concept of how British colonial powers used grammar as a governance tool in India, proposing that big data functions similarly as a "grammatical colonialism" that makes populations more legible and controllable through mathematical formalization.