Fairness-Standard Comparison Table¶
Method — instantiates Fairness-Standard Selection and Reconciliation
Applies candidate standards to the same decision and displays reasons, winners, burdens, conflicts, and uncertainty.
The Fairness-Standard Comparison Table lays several candidate fairness principles side by side against one fixed decision so that the disagreement among them becomes a single legible artifact instead of a shouting match. Its whole move is layout: rows are candidate standards stated in plain normative language, the columns hold the same population, good, and horizon, and each cell records what that standard would do — who wins, who bears the worst burden, where standards happen to agree, and how uncertain the estimate is. It does not judge whether any one distinction is legitimate, it does not enforce a nontradeable minimum, and it does not measure what actually happened after a rule shipped. It turns "we all want fairness" into "here is exactly the row where equal-share and seniority stop giving the same answer," and that is the entire value it adds.
Example¶
An irrigation district faces a river running at 40% of normal and must cut water to roughly 300 farms for the season. Rather than argue principles in the abstract, staff build a comparison table. The rows are candidate standards: an equal percentage cut for everyone; prior appropriation (the oldest water rights stay whole, junior rights are cut first); need (protect any farm whose entire crop dies below a survival threshold); proportional-to-historical-use; and a flat equal volume per farm. The columns pin the decision: this river volume, these farms, one season. Each cell is filled with that standard's winners, its worst-off, where it agrees with the others, and the forecast uncertainty.
The table's payoff is where the rows diverge. Seniority and equal-percentage agree for the mid-sized farms — but at the tails they split hard: seniority wipes out the newest farms (often the smallest, many recently established by first-generation growers) entirely, while equal-percentage spreads a painful-but-survivable cut to all, and need protects the smallest by loading most of the shortfall onto the senior holders. On one page the real fight is now visible: protect established senior rights, or prevent any single farm's total ruin? The table does not answer that. It makes the question impossible to hide.
How it works¶
- Fix the decision first. Pin the unit, population, good, and horizon so every standard is scored against an identical case — otherwise standards are compared on quietly different footings.
- Enumerate standards in normative language before any metric, one per row, including standards proposed by affected parties.
- Apply each to the same cases, filling per-standard cells: winners, burden-bearers, worst-off, and the uncertainty band.
- Mark convergence versus conflict. Collapse cells where standards agree so the genuine disagreements stand out, and label each conflict as mathematical impossibility, empirical tradeoff, normative disagreement, or legal constraint.
Tuning parameters¶
- Standard breadth — how many principles get rows. More rows surface overlooked options but dilute focus and can bury the two that actually differ.
- Case granularity — a whole-population summary versus itemized per-party rows. Finer granularity exposes tail cases but makes the table unwieldy.
- Convergence highlighting — how aggressively you fold away cells where standards agree, to spotlight the true conflicts.
- Uncertainty display — point estimate versus band per cell. Bands are honest but harder to read at a glance.
- Conflict typing — whether each conflict cell is labeled impossibility, tradeoff, normative, or legal, so people argue about the right thing.
When it helps, and when it misleads¶
Its strength is converting principle-level disputes into one shared artifact: it breaks metric monoculture by forcing plural standards into view, and it separates the places standards agree (uncontested) from the places they genuinely fight (the real decision). It is the map the rest of the process argues over.
Its failure mode is that a tidy grid implies the choice is a menu-pick when some rows are actually rights that must not be traded at all — sizing that constraint is another mechanism's job, not the table's. Worse, the selection of which standards get rows silently frames the entire debate: leave out "historical repair" and it is never weighed. The classic misuse is presenting the table as if picking a column were a neutral, technical act rather than a value judgment — the pseudo-objectivity[n1] that lets an authority avoid owning the choice. The guarding discipline is to publish, alongside the table, which candidate standards were excluded and why, and to insist that the authority — not the grid — owns the value decision.
How it implements the components¶
fairness_decision_scope_and_unit— it fixes the single decision, population, good, and horizon that every standard is scored against, preventing comparison drift between rows.candidate_fairness_standard_set— the rows are the disciplined menu of candidate standards, each stated in domain terms rather than as a bare metric.standard_conflict_and_distribution_matrix— the body of the table is exactly this matrix: agreement, conflict, winners, burdens, and uncertainty for every candidate applied to the same case.
It does not test whether a claimed distinction is purpose-justified (comparability_and_relevant_difference_model — that's Relevant-Difference Challenge Test), enforce a nontradeable floor (legitimate_standard_priority_and_composition_rule — Rights-Floor and Sufficiency Gate), or measure the realized spread of a shipped rule (distributional_outcome_and_revision_monitor — Distributional Impact and Tail Audit).
Related¶
- Instantiates: Fairness-Standard Selection and Reconciliation — the table is the artifact that makes principle-level conflict legible before any standard is chosen.
- Consumes: Participatory Fairness Deliberation supplies the candidate standards and affected-party framings the table tabulates.
- Sibling mechanisms: Relevant-Difference Challenge Test · Distributional Impact and Tail Audit · Rights-Floor and Sufficiency Gate · Participatory Fairness Deliberation · Fairness-Metric and Exception Stress Test
Editorial Notes¶
Form Classification¶
Form family: Analysis, Modeling & Optimization
Rationale: The mechanism applies rival normative standards to the same fixed case and analytically compares winners, burdens, uncertainty, convergence, and conflict types.
Nearest alternative: Assessment, Review & Assurance — The table can inform a fairness review, but its operative output is a comparative model of standards rather than an assurance disposition.
Review outcome: Adjudicated after independent review; high confidence.
Origin Attribution¶
Primary origin: Philosophy
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Applying rival normative standards to the same case and comparing reasons and burdens is a method of moral and political philosophy.
Related originating lineages:
- Law & Governance — Doctrinal balancing materially supplies reasoned comparison among competing equality standards. Rights, prior entitlement, and legal constraints materially shape candidate standards and nontradeable boundaries.
- Public Administration & Policy — Distributional policy analysis materially supplies the same-case comparison of winners, burdens, and implementation uncertainty.
- Ethics of Technology & AI Governance — Algorithmic fairness materially supplies modern incompatible metric families and affected-group analysis.
Review resolution: Both reviewers agree that philosophy is primary. I retain law_governance, tech_ethics_ai_governance, public_administration_policy only as formative origin lineages; cross_disciplinary_synthesis is appropriate because the final form materially combines the agreed primary with the retained formative lineages. Reach is multi_domain because the structure transfers across several fields but is not a near-universal human pattern, an applicability judgment kept separate from provenance. Encyclopedia synthesis is true because the exact generalized artifact is an encyclopedia-authored combination or refinement. No unresolved historical ambiguity remains after reconciling the secondary fields.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; medium confidence.
Notes¶
[n1] Pseudo-objectivity — the appearance that a value-laden choice is a neutral technical output. A comparison artifact is especially prone to it, because a clean grid makes selecting a column feel like reading off a fact rather than making a contestable judgment. ↩