Mind · Reasoning

EthicalReasoningProtocol

Separate prediction (is) from valuation (ought) from principle (override), each behind a typed boundary — the decision record has auditable empirical and normative layers

Definition

Mechanism

Applies ConceptualDecomposition to right-action, producing a non-entangled pipeline that prevents the is-ought entanglement that corrupts monolithic moral reasoning. Operationalizes Hume's is-ought distinction as an architectural boundary: calculate, then think. A prediction solver outputs a typed prediction ledger (branching future states with causal mechanisms and confidence envelopes) — constrained to pure description, no single point forecast. Downstream, a valuation solver evaluates 'which causal pathways matter, at what confidence, over what horizon' and emits a typed score sheet — quantitative ranking no normative consideration has yet touched. Crucially, a principle solver acts as both an acceptance gate and a formal override: if a top-ranked option trips a sacred value, a catastrophic tail risk, or a procedural precedent, it emits a typed judgment note overriding the numerical scoring. The final decision record has two distinct layers — quantitative base and qualitative override — segregated and independently auditable. This is the contract's override-with-documentation mechanism applied to ethics: the override is not an escape from the architecture but a first-class artifact within it. Composes with DeliberativeAlign for constitutional grounding.

Invariants

  1. Is-ought boundary: the prediction stage's output is constrained to pure description; no normative consideration contaminates the empirical forecast.
  2. Override-as-first-class: principle overrides do not adjust prediction weights silently — they emit a typed judgment note that appears in the auditable decision record.
  3. Two-layer record: the empirical base (what the forecast said) and the qualitative override (why we diverged) remain independently inspectable; no collapse into a single verdict.
  4. Bounded override: an override must cite the specific sacred value, tail risk, or procedural precedent it invokes. An override citing none is undocumented, and a protocol that overrides every ranking has replaced its own scoring rather than gating it.

Risks

Failure modes

None stated.

Relationships

Builds on

Composes with

ConceptualDecomposition

Finding what a concept is made of: locate it, carve it into the dimensions every case it claims needs, test each, search for what is missing, route real cases through it, say how the parts make the whole, and carve again below

References

Nothing in this library builds on this pattern yet.

Reference

Identity

Ref
EthicalReasoningProtocol#ee5d
Sema ID
sema:EthicalReasoningProtocol#mh:SHA-256:ee5d0dd9dd7779976d299bfa0ad1721c120b0c1a64fdd1c46299356fb5aa9c8e
Ring
1
Tier
2
Updated
2026-11-11
Supersedes
  • sema:EthicalReasoningProtocol#mh:SHA-256:6bf11b38a4352c7e0a99c002b75437cdc66d8a1a809ee353c11dec7a88c10998
  • sema:EthicalReasoningProtocol#mh:SHA-256:e3a615cd44b5441b404a2ae0c83d273fdeb6e18335e5de8380fd0f2e8f70149f
  • sema:EthicalReasoningProtocol#mh:SHA-256:8e7bda8878e1d77c14f93c10f31422b635d2ce79511a0360322d01e35577ce56
  • sema:EthicalReasoningProtocol#mh:SHA-256:e18bfc6dc7576b7dbfc6f4d4a3f7ec4c959c8cf9d5b4b9e2df3e682a7e12a9ae
  • sema:EthicalReasoningProtocol#mh:SHA-256:aec8ae3203837c25ede7e3a4f742b14495b0de44e5e8203cba00ba15dfa49125
  • sema:EthicalReasoningProtocol#mh:SHA-256:bf96bc8c01223f3ddbfc86b373e40a327f590f08c8b58c2282de748c35bde19a