Normative Entailment Check

The fundamental property of determining whether a conclusion is inescapable given the premises. Unlike probabilistic inference, a normative entailment check demands deductive closure: if the rule base and facts are true, the conclusion must be true in all possible compliant worlds.

• Employs modus ponens and universal instantiation as core inference rules
• Rejects conclusions that are merely plausible or statistically likely
• Requires soundness (all derived conclusions are logically entailed) and completeness (all entailed conclusions are derivable)
• Example: If a statute states 'All contracts exceeding $500 must be in writing' and Fact A establishes a $750 contract, the system must necessarily entail the writing requirement

The property ensuring that the entailment check does not derive contradictory obligations, permissions, or prohibitions from a consistent rule base. The system must verify that the addition of an entailed conclusion does not create a deontic collision.

• Monitors for obligation-prohibition conflicts (e.g., 'must file' vs. 'must not file')
• Detects contrary-to-duty paradoxes where a secondary obligation contradicts a violated primary obligation
• Maintains a conflict-free conclusion set by rejecting entailments that introduce inconsistency
• Implements the Kantian maxim: 'ought implies can' — the system cannot entail an obligation that is factually impossible to fulfill

The precondition check that determines whether a legal rule is triggered before any entailment computation begins. A rule's applicability condition is a Boolean expression evaluated against the fact pattern.

• Evaluates material scope (does the rule cover this subject matter?)
• Evaluates temporal scope (was the rule in effect at the relevant time?)
• Evaluates personal scope (does the rule apply to this party?)
• Evaluates territorial scope (did the event occur within the jurisdiction?)
• Only rules whose applicability gate returns true participate in the entailment check

In a classical entailment check, the system operates under monotonic logic: once a conclusion is entailed, adding new premises cannot invalidate it. This property guarantees reasoning stability within a fixed rule base.

• Contrasts with non-monotonic defeasible reasoning used in broader legal argumentation
• Requires that the rule base be pre-resolved for conflicts before the entailment check runs
• Ensures that the same inputs always produce the same entailed outputs (determinism)
• Violated if the system later encounters a lex specialis exception — hence the need for a pre-processing conflict resolution phase

The property of deriving all indirect consequences of a rule set, not just immediate conclusions. If Rule A entails Fact B, and Rule C is triggered by Fact B to entail Obligation D, the system must chain through to Obligation D.

• Implements forward chaining from known facts through all triggered rules
• Terminates when a fixed point is reached (no new conclusions can be derived)
• Critical for detecting latent obligations buried deep in regulatory frameworks
• Example: A merger triggers a filing obligation, which triggers a waiting period, which triggers a standstill obligation — all must be entailed

The property that an entailed conclusion must hold even under minimal perturbations of non-material facts. The system verifies that the conclusion does not depend on irrelevant or accidental features of the fact pattern.

• Tests sensitivity: would changing an unrelated fact alter the entailment?
• Ensures the conclusion is robust across reasonably similar factual scenarios
• Prevents spurious entailments that exploit edge cases or drafting ambiguities
• Aligns with the legal principle of ratio decidendi — the binding reasoning should apply to materially similar cases