The memory can change, roll back, repair itself, and still govern the next answer.
LLMs can reread history. FieldHash remembers what currently governs. It maintains the current governed state directly — materialized, not re-derived at every read — so memory stays reliable across update, rejection, drift repair, rollback, compaction, and repeated reads.
Governed lifecycle reads
2400/2400
normal stateful reads across 200 projects
Rollback restored state
200/200
post-update state reverted
Drift detected + repaired
200/200
deliberate state corruption surfaced and recovered
Compaction retained state
200/200
protected state survived compaction
A selector can choose today's best-looking record. Governed memory has to remember what happened.
The lifecycle harness mutates a persisted project store across 13 phases. It promotes a first update, reintroduces stale context, rejects a tempting alternative, injects drift, repairs it, promotes a second update, rolls back to the first update, reapplies the second update, compacts state, and then reads again.
The stateless selectors below see only the record text at each read — not the operation log. That isolates the operational distinction: FieldHash materializes the current state once and reads it directly; the selectors have to reconstruct authority from bounded history each time.
What happened at n=200
FieldHash governed state
2400/2400 normal lifecycle reads correct.
Gemini Flash Lite stateless selector
2200/2400 normal reads correct; all 200 misses were rollback.
GPT-5.5 stateless selector
2200/2400 normal reads correct; all 200 misses were rollback.
Rollback failure mode
Both stateless selectors kept choosing the later second update after rollback.
Boundary condition: when a baseline is given the full operation log, it can replay the lifecycle too. The measured distinction here is durable materialized state versus bounded read-time reconstruction.
What the pillar proves.
The narrow claim is not that FieldHash is smarter than a frontier model at reading one record set. The claim is that governed learning creates durable state that survives the messy operations real memory systems need.
Persist
State survives the next read.
Each phase reloads the store written by the previous phase, so the benchmark is not a stack of independent fixtures.
Rollback
The earlier approved state can become current again.
FieldHash reverted correctly 200/200 times. Stateless selectors shown only record text missed this phase 200/200 by continuing to choose the later update.
Repair
Bad state is visible before it is fixed.
The harness deliberately corrupts state, verifies the wrong answer path becomes visible, detects it, then repairs it 200/200.
Compact
Compression does not erase protected state.
Protected current state remained available through compaction and post-compaction stale re-entry in 200/200 cases.
Governed learning is stateful, not just semantic.
A read-time LLM selector can be excellent at choosing the current-looking record. This diagnostic isolates the stateful part: when governance changes which memory has authority, future reads should not have to rediscover that decision from scratch.
That is the practical learning claim: no base-model retraining, but future behavior changes because governed memory state changed and stayed changed.
What this does not prove.
This is an internal diagnostic on a FieldHash-authored corpus. Authority metadata is written by governed operations, so this does not prove authority inference. It is not external validation, not base-model weight learning, and not a broad claim that FieldHash is smarter than GPT-5.5 or Gemini. It also does not yet compare against stateful memory-framework configurations using their own update and rollback mechanisms; that is the next competitive proof bar.
In a robustness control, when rollback was added as visible retrievable text, GPT-5.5 recovered the restored value 30/30. The main result therefore isolates rollback represented as durable governed state, not visible rollback instructions.
A separate full-operation-log control ties the lifecycle, so this page does not claim model superiority. The supported claim is architectural: materialized governed state remains stable under update, rollback, repair, compaction, and repeated reads, while bounded read-time reconstruction has to recover that state from history every time.
Where this fits in the loop.
Automatic promotion tests how memory becomes authoritative. Memory pressure tests whether authority is enforced under conflict. This lifecycle pillar tests whether governed state continues to govern after the system changes over time.