Definition
The design principle that the causal distance between an institutional error and the actors who can revise the generating rule should be kept as short as the stakes demand. Where error-bearing parties sit far from rule-makers (by hierarchy, by time, by evidence format, or by classification), corrections arrive late, weakened, or not at all.
It matters because proximity is not visibility. A harm can be fully visible to the people carrying it and still reach nobody who can change the rule that produces it. Shortening proximity means building the channels that let mismatch travel from the edge back to the rule: corrective standing, records that travel, and halt stratification. Insulation keeps harm from reaching anyone who can change the rule; feedback proximity is the commitment to shorten the path.
Tests that show it is in place
- A harm reported at the frontline reaches an actor who can change the generating rule through a bounded number of named hops.
- The latency from first signal to rule revision is measured and bounded, not merely visible.
- Error-bearing parties can state the shortest path from their report to a rule change, and that path exists.
Evidence to keep
- Artifact: A mapped path from frontline signal to rule-revising authority, with named responders at each hop.
- Behavior: A reported mismatch that produced a rule change without an intervening heroic intervention.
- Metric: Time, or number of hops, from first signal to generating-rule revision, tracked per failure class.
References
- Insulation · Theory
- Challenge density · Theory
- Structural corrigibility · Term