Sorting Things Out

6 ideas

Buy on Amazon
  1. Classifications become infrastructure by going invisible

    A classification system works best when it sinks below notice and becomes taken-for-granted infrastructure, so its categories feel like neutral facts rather than contingent choices. The more smoothly it operates, the harder it is to see, question, or revise the political and moral decisions buried inside it.

  2. Torque between life and category

    When an official classification fails to fit a person's lived trajectory, the person experiences torque: a twisting of their biography to conform to institutional categories. The authors show this in tuberculosis patients whose bodily time was reorganized around sanatorium schedules and diagnostic labels.

  3. Apartheid's racial classification could not classify

    South Africa's apartheid system required sorting every person into racial categories, yet the boundaries were so arbitrary that reclassification boards shifted thousands of people between races using tests like pencils in hair. The case shows that even a brutally enforced classification cannot make its categories coherent, and the labor of forcing people into them produces continuous contradiction.

  4. Every standard embeds silenced moral choices

    Each classification distributes advantages and suffering by deciding what counts and what is left out, but these ethical and political stakes are erased once the scheme is treated as technical. The authors argue that making these buried choices visible is itself a moral and political act.

  5. Boundary objects coordinate without consensus

    Objects, categories, or records can be shared across different communities while meaning something different to each, letting groups cooperate without first agreeing on a common definition. This loose interpretive flexibility at the local level, combined with a stable shared structure, is what enables coordination across divergent social worlds.

  6. Residual categories reveal a system's politics

    Examining the leftover slots of a classification (the 'other', 'not elsewhere classified', or 'unknown' bins) exposes what a system cannot or will not name. These residual categories accumulate the cases that embarrass the scheme, so reading them backward reveals the assumptions and exclusions the classification is built on.

Save ideas and give them a thumbs up or down in the app