Constitutive Norm

The fundamental logical mechanism of a constitutive norm is the 'counts as' operator, formalized by John Searle. This operator maps a brute fact (a physical action or event) onto an institutional fact.

• Formula: 'X counts as Y in context C'
• Example: A sequence of spoken words (X) counts as a legally binding offer (Y) in the context of contract negotiation (C)
• Computational Representation: This is modeled in Input/Output Logic as an ordered pair where the input state is transformed into an institutional output
• Distinction: Without this rule, the brute fact has no deontic significance; the constitutive norm creates the institutional reality

Constitutive norms generate institutional facts—facts that exist only because of the rules that constitute them. These are ontologically subjective but epistemologically objective within the legal system.

• Examples of Institutional Facts:
  • A corporation exists as a legal person only because corporate law constitutes it
  • A marriage is not a natural phenomenon but a status created by constitutive rules
  • Money: A piece of paper counts as legal tender only through the constitutive rules of a central bank
• Computational Implication: Legal reasoning systems must maintain an institutional fact registry that tracks which entities have been constituted and remain valid
• Contrast with Regulative Norms: Regulative norms say 'Do not steal'; constitutive norms define what 'property' and 'ownership' mean in the first place

Constitutive norms form validity chains where higher-order norms define the conditions under which lower-order norms can be created. This is the architecture of legal systems.

• Kelsen's Grundnorm: The ultimate constitutive norm that validates the entire legal order
• Example Chain:
  • Constitutional provision → empowers legislature → which enacts a corporate statute → which defines how to form a corporation → which creates a specific corporate entity
• Computational Modeling: This hierarchy is represented in LegalRuleML using nested authority structures and in Defeasible Logic Programming (DeLP) as presumptions that can be challenged
• Validity Conditions: Each constitutive norm specifies the necessary and sufficient conditions for a valid institutional act—failure to meet these conditions renders the act void ab initio

A critical subset of constitutive norms are power-conferring rules, which enable agents to intentionally create new institutional facts. These are the legal system's API for changing normative reality.

• Hohfeldian Powers: The ability to alter legal relations—to create, extinguish, or modify rights and duties
• Examples:
  • The power to execute a contract and bind parties to obligations
  • The power of a court to issue a judgment that changes legal status
  • The power of a legislature to enact statutes
• Formalization: In Dynamic Deontic Logic, power-conferring norms are modeled as state-transition operators that move the normative system from one valid state to another
• Computational Requirement: Systems must track not just what obligations exist, but which agents have the authority to modify them

The distinction between constitutive and regulative norms is foundational for legal AI architecture. Conflating them leads to category errors in reasoning systems.

• Regulative Norms: Govern antecedently existing activities
  • Example: 'Drive on the right side of the road'—driving exists independently of this rule
• Constitutive Norms: Create the very possibility of the activity
  • Example: 'Checkmate is achieved when the king is in check and cannot escape'—without this rule, chess does not exist
• Legal AI Application: When parsing a statute, systems must classify each provision as constitutive (defining terms, creating entities, conferring powers) or regulative (imposing duties, granting permissions)
• Reasoning Impact: Violating a regulative norm results in a sanction; failing to satisfy a constitutive norm results in nullity—the purported act never legally occurred

When the conditions specified by a constitutive norm are not met, the result is nullity—the institutional fact fails to come into existence. This is distinct from a violation of a regulative norm.

• Nullity vs. Sanction:
  • Nullity: The act is legally void; it produces no institutional effects
  • Sanction: The act is valid but prohibited; it produces effects but triggers penalties
• Examples:
  • A contract signed under duress is voidable (constitutive defect)
  • A contract that violates antitrust law is valid but illegal (regulative violation)
• Computational Detection: Deontic SHACL validators can check RDF graphs of legal entities against constitutive rules to flag institutional defects
• Reasoning Consequence: Systems must distinguish between 'this obligation was violated' and 'this obligation never legally existed'