Taxonomy / Adversarial proceduralism / Separate agents for each side of a value conflict

Separate agents for each side of a value conflict

The dispute here is over interests and values, where there is no fact to be right about, so each position is carried by its own agent and the procedure is built to stop the positions collapsing into one. Stakeholders are role-played by agents that bargain under a budget with a neutral mediator, and the outcome reports how the worst-off group fared rather than only the total; separately, giving each value perspective its own distinct underlying model keeps a deliberating council's answers spread out instead of converging on one. Two results sit behind the design: a study of what happens when a single model is made generator, debater and judge on moral dilemmas and contested political claims, and a proof that when majority preferences cycle no reward model can crown a winner at all.

The method, against Adversarial proceduralism

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.

Counter-power by design

PartialNot claimed

Definition · James Madison, Federalist No. 51 (1788)

Don't rely on written rules or good intentions: build institutions that oppose each other, so that each has both the means and the motive to check the others. "Ambition must be made to counteract ambition."

Analysis

The budget-bound negotiation does put stakeholder agents in genuine contention: a fixed budget makes one role's gain another's loss, so the agents have both a distinct position and a reason to press it against the others. But the checking runs only sideways between stakeholders — the mediator that runs the procedure and the researchers who assign roles and settle the budget face no counterweight at all, and Madison's point is precisely that the deciding power is the one that must be checked. The council paper supplies the missing ingredient from the opposite side: giving each perspective its own 7-9B model buys the independence to differ (without it the evaluators converge regardless of the perspective assigned), yet no rule gives any agent a stake in policing another's drift.

Non-domination

AbsentNot claimed · out of scope

Definition · Philip Pettit, Republicanism (1997)

You are unfree if someone holds unchecked power over you, even benevolently. What matters is not whether power is used well but whether the governed can contest it.

Analysis

Every component here decides about people rather than with them, and none opens a route back. The negotiation reports how the worst-off group fared, which makes domination legible in the metrics while giving that group no standing to object to the settlement; the council deliberates on behalf of perspectives it was assigned, and the people who actually hold those perspectives never see or challenge what was concluded in their name; the impossibility result keeps the minority alive as a sampling weight in a mixed strategy, which is favorable treatment conferred by an equilibrium property, not a power to contest. Pettit's test is whether the governed can check the power over them, and no part of this approach builds that channel.

Agonism

PartialClaimed · partial

Definition · Chantal Mouffe, The Democratic Paradox (2000)

Deep disagreement is permanent and should not be dissolved into consensus. A healthy system converts enemies into adversaries — opponents whose conflict stays alive and legitimate — rather than declaring the argument settled.

Analysis

The refusal-of-collapse half is genuinely engineered and measured: heterogeneous per-perspective models cut first-choice concentration from 70.9% to 46.1% on child welfare and 46.0% to 22.9% on housing, and keeping that spread is the stated design goal rather than an artifact. What is missing is the rest of Mouffe — adversaries are parties with standing commitments who go on contesting inside a frame that legitimates the contest, whereas here the spread is a statistic over the endpoint of a one-shot deliberation, nothing carries a losing position forward, and the coherence-validation intervention re-concentrates outcomes exactly on the questions where the options are most evenly matched. The other main component runs the opposite way: the negotiation exists to close the conflict in a mediated settlement, after which the disagreement survives only as a worst-group number.

Incentive compatibility

AbsentNot claimed · out of scope

Definition · Leonid Hurwicz; Eric Maskin (mechanism design)

Design the procedure so that honest behavior is each participant's best strategy. The rules of the game do the enforcement, instead of trusting the participants to be good.

Analysis

Nothing in this approach makes candour a best response, and for the most part there is no private information to be candid about: the agents are handed their positions by prompt, so there is no type for a participant to misreport and no elicitation problem to solve. The budget and neutral mediator constrain what agreements are reachable, not what it pays an agent to say; the architectural heterogeneity works by making agents structurally unable to collapse into one another rather than by rewarding them for holding their assigned line; and the Nash equilibrium in the impossibility paper describes the aligned model's mixed strategy over answers in the limit, with human preferences taken as fixed by the Luce model rather than reported by strategic agents. The game theory invoked is Nash's, not Hurwicz's.

Concept Analysis: Newly Introduced

Consensus treated as a failure to be measured

Added

Multi-agent deliberation systems report agreement as a result. This one treats agreement reached by identical models wearing different labels as an artifact to be detected and removed, and validates that the disagreement left over is real. Making artificial consensus a measurable defect is a contribution the agonistic tradition motivates but never operationalizes.

Collapse proved unrepresentable

Added

The tradition objects to forcing one answer out of deep disagreement on grounds of legitimacy. This work adds an objection of a different kind: under a general probabilistic model of how people choose, majority preference over a model's answers almost always contains a cycle, so there is no single ordering for a reward function to stand for and the collapse is not merely unattractive but unavailable. The case for keeping plurality becomes a question of representability rather than of what is right, which is an argument Madison, Pettit and Mouffe do not contain.

Papers

No method has been built here yet — this approach rests entirely on what follows.

Measured, not built

Instruments that measure this rather than instantiating it. This map is a map of methods, so they are indexed here but counted nowhere on the grid.