Identity in distributed systems has traditionally been defined through static artifacts—hardware configuration, software state, or cryptographic credentials. These descriptive models assume that identity is a property of components rather than execution, and that continuity can be inferred from the integrity of stored state. Modern systems violate these assumptions. Their behavior is shaped by the coupled dynamics of substrate, runtime, and execution, producing identity as a realized phenomenon rather than a static attribute. Recent work established the formal objects required to treat identity as a property of execution, including execution‑realized identity, execution‑rooted provenance, and substrate‑rooted attestation. This paper provides the missing ontology. It introduces the four‑layer identity decomposition (s, e, b, a), the doctrinal successor to the descriptive substrate model (H, E, M). The decomposition is minimal, irreducible, and prescriptive: each layer contributes an essential component of identity, drifts independently, and admits its own attestation surface. The decomposition reorganizes the foundational formal objects—execution space, perturbation stability, structural mapping, deterministic key derivation, behavioral challenge–response, provenance mapping, and governance envelopes—into a single, coherent identity primitive. Identity becomes a layered, falsifiable, and temporally coherent object. Attestation becomes a compositional property across layers. Provenance becomes the temporal projection of identity. Governance becomes identity‑conditioned rather than configuration‑conditioned. The four‑layer decomposition thus establishes identity as a first‑class system primitive and provides the foundation for identity, attestation, provenance, and governance in systems where execution, not configuration, determines behavior.
Aure Ecker-Fils (2026) studied this question.