Provenance Continuity
The unbroken chain of verifiable provenance records across all context artifacts in an operational corpus — no gaps, no unverified links.
Definition
Provenance continuity is the property of an operational corpus in which every artifact has verifiable provenance, and the chain of provenance records is unbroken from the initial seed through every subsequent capture and checkpoint.
A corpus has provenance continuity when: every artifact can be verified against its hash, every artifact’s relationship to prior artifacts in the lineage chain is documented, and there are no unverifiable links in the chain.
Why Continuity Matters
Append-only provenance ensures that individual artifacts cannot be modified. Provenance continuity ensures that the relationships between artifacts are also intact.
Consider a corpus where every artifact is verifiable individually, but the ordering and relationships between artifacts are not recorded. An engineer could add a new artifact that references an older one without establishing a proper lineage link. The individual artifacts are trustworthy, but the chain between them is not.
Provenance continuity closes this gap. The chain of relationships — which artifact was captured after which checkpoint, which decisions were made in response to which prior decisions — is part of the verified provenance record.
Hash Chains
Yanzi enforces provenance continuity through hash chains. Each artifact carries the hash of the prior artifact in its lineage sequence. This creates a chain structure (similar in principle to a blockchain or append-only log) in which any break in the chain is immediately detectable.
If an artifact is modified, its hash changes. The next artifact in the chain, which carries the original hash, will no longer match — the break is visible.
Continuity in Practice
For most engineering sessions, provenance continuity is maintained automatically by Yanzi’s append-only corpus operations. Engineers do not manage hash chains manually.
Provenance continuity becomes operationally relevant during:
- Release certification (validating the full chain from initial seed to release candidate)
- Audit reviews (verifying that a historical lineage record is intact)
- Corpus migration or backup verification