Taxonomy / Divergent precedent / Communities revising their own rules case by case

Communities revising their own rules case by case

A community argues over concrete cases and changes the thing that governs it. The governed thing differs across the family and the difference is real: PolicyCraft grows a written policy document from submitted real cases, tracking which clauses each case grounds and surfacing cases that contradict the policy as it stands, while Botender changes the deployed behavior of a group's bot, using generated interaction scenarios as provocations rather than waiting for cases to arrive. What holds them together is the loop rather than the output format -- ordinary members deliberate over particular cases, and a shared governing artifact, document or agent, is edited as a result. Sitting alongside the tools is a measurement framework that splits a group's uncertainty on moderation cases into ambiguity, which a clearer description of the case can reduce, and genuine disagreement, which it cannot; it belongs here because it tells builders of these systems which arguments are worth staging, not because either system performs that split.

The method, against Divergent precedent

Scroll the diagram sideways to see all of it.

Concept Analysis: Theoretical Foundations

Each concept is read twice: whether the approach carries it, and whether the approach's own sources claim it. A concept that is absent and was never claimed is a gap in the field rather than a failure of the work, and is marked out of scope.

Recorded dissent

PartialClaimed · partial

Definition · Appellate practice

When judges disagree, the losing view is not erased — it is published alongside the majority's, signed and reasoned. That does two things at once: it forces the majority to answer a stated objection, which is what makes the majority's own reasoning legible, and it keeps the rejected theory on the shelf, available to become tomorrow's majority.

Analysis

Disagreement is deliberately drawn out rather than smoothed over: Botender's provocations were found better than standard test cases at surfacing it, and Judgment Sieve goes further and separates the disagreement that clarification will not close from the ambiguity that it will. What is missing is publication. No reasoned minority view is written down, signed and kept beside the clause or the behavior it lost to, and a dissent that exists only as a residual uncertainty score cannot become tomorrow's majority because there is nothing on the shelf for anyone to pick up.

Overruling with provenance

PartialClaimed · partial

Definition · Common-law practice

The law can reverse itself when a precedent has become unworkable, unjust or simply overtaken by events — but only in the open: the overruling decision names what it overturns and why, so the change has a traceable history instead of being quietly written out of the record.

Analysis

PolicyCraft is where this cell comes nearest to performing a reversal rather than grading one: the policy is revised case by case, each clause is linked to the cases grounding it, and cases that contradict the policy as it stands are surfaced, which is the trigger a court needs before it reconsiders a precedent. What is absent is the naming. An overruling decision states what it overturns and why, and the stated reason is what makes the history traceable; here the trace runs from a clause back to its supporting cases, not from a revision back to the position it displaced. Botender's iteration on bot behavior keeps no record of what a revision replaced, and Judgment Sieve has no revision step at all.

Incompletely theorized agreement

PartialNot claimed

Definition · Cass R. Sunstein, Legal Reasoning and Political Conflict (1996)

Agreeing on what to do in the case at hand while explicitly leaving the deeper principle unresolved. The agreement can be on an outcome, a rule or a mid-level principle; what gets bracketed is the comprehensive worldview underneath. Courts do this constantly: it lets people who disagree about theory still act together, one case at a time.

Analysis

Judgment Sieve makes the bracket operational in a way Sunstein never specified, telling a community which of its conflicts a clearer description of the case can close and which are real conflicts of value that will survive any amount of clarification, so the case in hand can be settled and the deeper quarrel left alone. PolicyCraft then moves the other way: its destination is a general clause stated for future cases, which is exactly the level Sunstein counsels leaving unresolved. The deliberation is case-grounded and its product theorises.

Concept Analysis: Newly Introduced

Ambiguity separated from disagreement

Added

Legal practice has no instrument that tells a case that is hard because it was badly described apart from a case that is hard because people's values conflict. A court meets both as a hard case and decides. Judgment Sieve separates the two and applies a different intervention to each, which tells a community which of its conflicts clarification can close and which it must live with. That triage step the tradition never had to formalize, because a court has to rule either way.

Papers