Responsibility Attribution Matrix¶
Decision matrix — instantiates Outcome Responsibility Attribution Calibration
Cross-tabulates each candidate agent against contribution, control, duty, and knowledge, then applies an explicit weighting rule to turn the grid into a graded, comparable responsibility reading.
Once the separate dimensions of an attribution have been assessed, something has to combine them without collapsing back into a gut verdict. Responsibility Attribution Matrix is that aggregator: it holds each candidate agent as a row and each attribution dimension — causal contribution, control, duty, knowledge, mitigation, evidence confidence — as a column, then applies an explicit, stated weighting rule to turn each row into a graded responsibility score, and a stated rule linking that score to a proportional consequence tier. Its defining move is twofold: it keeps the dimensions visible and separate, so any agent's score can be traced to why they hold it, and it forces the weighting to be declared up front rather than smuggled in — the archetype weights controllable, owed contribution above raw causal share, and the matrix makes that choice inspectable.
Example¶
A grass fire ignites near a power line during a windstorm and spreads across a valley. Rather than argue whether "the utility" is to blame, the matrix rows the candidates — the utility, the landowner who left brush uncleared, the regulator who set the vegetation-clearance standard — plus the windstorm itself as a background condition that is scored but barred from consequence. The columns are filled from the dedicated mechanisms: causal contribution from the timeline, control from the control test, duty from the role–duty map, knowledge from the foreseeability record.
The matrix's own contribution is the last two moves. A weighting rule — control and duty weighted above raw contribution, because the archetype cares about controllable, owed contribution — turns each row into a graded score; and a consequence rule maps score bands to proportional accountability (advisory finding → corrective order → sanction). The output is a side-by-side reading that shows the utility and the regulator both carrying duty-weighted responsibility, while the windstorm — the single largest causal contributor — carries none, because it had neither control nor duty. Every cell stays visible, so any party can see precisely which dimension drove their score.
How it works¶
- Assemble the grid. Agents as rows; the attribution dimensions as columns, each cell filled from the mechanism that owns it. The matrix tabulates these inputs; it does not manufacture them.
- Declare the weighting rule first. State which dimensions dominate and why — before scoring, and independent of the outcome — so the rule can't be reverse-engineered to reach a name.
- Combine each row into a graded score, carrying evidence confidence forward so a cell built on thin evidence can't masquerade as certainty.
- Map score to consequence through a stated proportionality rule, keeping non-agent background conditions scored but excluded from any sanction.
Tuning parameters¶
- Weighting scheme — equal-weight versus duty/control-dominant. Duty/control-dominant matches the archetype's intent; equal-weight is simpler but lets raw causal share over-drive the verdict.
- Score resolution — discrete bands versus a continuous score. Bands resist false precision; a continuous score discriminates finely but invites over-reading small gaps.
- Consequence mapping — how score translates to accountability, and how proportional. Tight coupling is predictable; loose coupling preserves discretion.
- Evidence gating — how hard low-confidence cells are discounted. Aggressive gating protects against confident-looking noise; light gating keeps weakly-evidenced contributors in view.
- Non-agent inclusion — whether background conditions (weather, market, prior design) get scored rows. Including them prevents their causal share from being silently reassigned to an agent.
When it helps, and when it misleads¶
Its strength is that it forces the weighting into the open and makes the whole reading auditable and comparable across agents — you can see, cell by cell, why one party outscores another, and by laying every contributor side by side it resists over-attributing the entire outcome to the single most salient one.[n1] It is also where proportionality becomes concrete rather than rhetorical.
Its failure modes are the classic hazards of any scoring grid, sharpened here. A matrix of tidy numbers laid over soft judgments invites false precision; it is only ever as good as the cells the other mechanisms supply (garbage in, garbage out); and it is easily run backwards — weights quietly tuned after the fact to land on a predetermined culprit, or cells scored by how bad the outcome was rather than by the conduct. The discipline that keeps it honest is to declare the weighting before scoring, carry evidence confidence into the result instead of hiding it, and never let the matrix's clean output substitute for the dimensions beneath it.
How it implements the components¶
Responsibility Attribution Matrix fills the aggregation-and-decision components of the archetype — the ones that turn assessed dimensions into a graded, actionable reading:
attribution_weighting_rule— the explicit, outcome-independent rule that combines the per-agent dimensions into a graded responsibility score; the matrix's core product.consequence_link_rule— the stated proportionality rule mapping a score band to a graded accountability tier, so the attribution links to a consequence transparently rather than by fiat.
It does not produce the dimensions it tabulates: causal contribution (causal_contribution_timeline), control (counterfactual_control_test), duty (role_duty_mapping), knowledge and avoidability (omission_commission_parity_check), or the evidence grade in each cell (attribution_uncertainty_label). It combines them; it does not generate them.
Related¶
- Instantiates: Outcome Responsibility Attribution Calibration — the matrix is the aggregation core that turns the appraisal's separate dimensions into a graded, proportional reading.
- Consumes: Role–Duty Mapping, Causal Contribution Timeline, Counterfactual Control Test, and Omission–Commission Parity Check — the mechanisms that fill its columns.
- Sibling mechanisms: Role–Duty Mapping · Omission–Commission Parity Check · Blame–Credit Apportionment Table · Attribution Uncertainty Label · Outcome Responsibility Review Panel · Scapegoat Screening Review
Editorial Notes¶
Form Classification¶
Form family: Analysis, Modeling & Optimization
Rationale: Responsibility Attribution Matrix operates as an analytical, modeling, inference, comparison, or optimization procedure that derives insight or a solution because it cross-tabulates each candidate agent against contribution, control, duty, and knowledge, then applies an explicit weighting rule to turn the grid into a graded, comparable responsibility reading.
Independent corroboration: The frozen evidence defines Responsibility Attribution Matrix as 'Cross-tabulates each candidate agent against contribution, control, duty, and knowledge, then applies an explicit weighting rule to turn the grid into a graded, comparable responsibility reading', so its operative form is Analysis, Modeling & Optimization.
Nearest alternative: Representation, Specification & Plan — Responsibility Attribution Matrix includes features of a static representation, map, specification, schema, or prospective plan that externalizes information, but its defining operation is an analytical, modeling, inference, comparison, or optimization procedure that derives insight or a solution.
Review outcome: Independent reviewer agreement; medium confidence.
Origin Attribution¶
Primary origin: Philosophy
Origin pattern: Convergent development
Present-day reach: Universal
Rationale: Contribution, control, duty, and knowledge are canonical dimensions of moral responsibility theory.
Related originating lineages:
- Law & Governance — Legal doctrines of fault and liability independently formalized graded attribution.
- Psychology — Attribution research materially informs how responsibility judgments are made and biased.
Review resolution: Both blind reviewers agree that philosophy is the primary historical origin. Explicit reconciliation of alternate origin disagreement, origin mode disagreement, domain reach disagreement adopts reviewer_a's evidence: Contribution, control, duty, and knowledge are canonical dimensions of moral responsibility theory. The selected record uses alternates=law_governance, psychology, origin_mode=convergent, and domain_reach=universal; the other review proposed alternates=mathematics, origin_mode=single_lineage, and domain_reach=multi_domain. The selected combination better preserves the mechanism-specific formative lineages and calibrated scope; broader present-day use is not treated as proof of additional historical origin.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
Because the matrix is a pure aggregator, its integrity is entirely inherited from the cells beneath it — a confident score over thin cells is its most seductive failure. It also scores background conditions (weather, market, prior design) as rows for completeness, which means the consequence rule must explicitly bar non-agents from sanction; otherwise a large causal contribution with no agent behind it silently gets reassigned to whoever is nearest.
[n1] The discounting principle in attribution theory: the perceived role of any one cause is reduced when other plausible causes are present. Laying every contributor's dimensions side by side operationalizes this — the matrix resists collapsing the whole outcome onto the single most salient agent. ↩