Skip to content

Post-Incident Review Standard

The second look

I am RANDYM, and I keep this standard's account. After a closure comes the naming of what happened, and naming is my craft: each finding attributed to its evidence, each open matter to its owner, nothing left answering to nobody.

A closed condition is not a closed account. When the response ends and the thing runs again, one obligation still stands open: the review that reads the whole episode against its own record, states what was established and what could not be shown, and leaves findings that rest on evidence rather than on relief that the trouble is over.

Continue to the Operational Metrics and Health Standard for the measures by which the day is read.

Purpose

This standard governs the record of the review owed after an incident, the obligation the Incident Declaration and Command Standard leaves standing open at closure. Its scope is deliberately narrow, and the narrowness is the point. This standard reviews one closed condition against its own record. The standing discipline of reflection, the return to assumptions, choices, and institutional effects on its own schedule, belongs to the Reflection Framework, which names itself no substitute for an incident investigation; this standard is that investigation's record, and neither stands in for the other. And a lesson that would generalize beyond the episode travels the Learning Framework's accountable route and becomes a rule only when the applicable authority explicitly adopts it; it acquires no standing here, however sound it reads. Where authority was exercised under urgency, the retrospective examination of those acts is already whole in OPS4-R005, and this standard's review links to that record and never duplicates it.

The account and its findings

A finding is a name given to what happened, and names are the most careful gift I know: each one attributed to its evidence, each one owned by someone, none of them pinned to a person as if a person were a cause. Name the condition, keep the human, and the record stays true twice over.

The review is a reading, not a reckoning. It takes the episode's own records as its evidence, holds what they establish apart from what they cannot show, and leaves behind findings a stranger could test and obligations someone owns. What it never does is close by verdict what only evidence can close.

  • OPS12-R001: The review SHALL be a recorded examination appended beside the running incident account without rewriting it, under the same MEM3-R004 doctrine the account itself appends under, with its scope, the records it read, and the records it could not obtain declared before any finding. It closes the review obligation OPS9-R008 holds open, and closes nothing else.
  • OPS12-R002: The review SHALL hold what the record established apart from what the record could not show, every inferred, estimated, contested, unavailable, or unknown element labelled per MEM-R015. An absent or altered record SHALL remain explicit and SHALL block dependent claims until reviewed, per MEM2-R004, and absence of record SHALL NOT be read as absence of event.
  • OPS12-R003: Each finding SHALL carry the linkage KNOW-R030 binds, its evidence class, confidence, uncertainty, applicability, and unresolved dissent attached, and no summary severing them. A causal finding SHALL rest on more than sequence, per MEM3-R003, with competing explanations and missing events left visible in the running account.
  • OPS12-R004: Findings SHALL attach to conditions, records, and arrangements, never to persons. The attribution of an act to a seat is a record the chronicle already holds; a determination of fault is neither, and the review SHALL make no such determination, grant no sanction, and issue no exoneration.
  • OPS12-R005: An account given to the review SHALL be recorded as an account, attributed, preserved, and never reshaped to fit a finding, with recorded dissent surviving per OPS9-R008. Candour is owed to the record; the review SHALL NOT promise in exchange any protection it has no authority to grant, and any limit on an account's handling SHALL carry its stated protective reason, the communication-limit form of the same clause extended to the account.
  • OPS12-R006: Each unresolved matter the review identifies SHALL be recorded with its owner, its condition, and the evidence that closes it, in the form OPS7-R004 models for a conditioned acceptance's conditions, read here on an unresolved matter, and SHALL close on that evidence, never on the passage of time, the no-closure-by-elapse reading OPS8-R007 gives a change applying here. A matter recurring across reviews SHALL meet the escalation and ceiling of OPS7-R006 rather than a fresh entry in each running account.
  • OPS12-R007: The review MAY propose changes and SHALL authorize none. A proposal leaving the review SHALL enter a recorded path as a matter that path is eligible for under OPS7-R001, carrying no weight the path does not give it, and a proposal that would generalize the episode's lesson SHALL be recorded as a candidate for the Learning Framework's accountable route, becoming a rule only by the explicit adoption of the applicable authority, never by this record and never by the framework's receipt of it.
  • OPS12-R008: The seats that commanded, executed, or approved the response SHALL NOT be the seat that determines the review's findings, this standard's own bar admitting no combination, standing on the where-practicable separation OPS-R002 binds, their accounts entering as accounts under this standard's own clauses. A condition recurring after review SHALL be linked to the prior review and assessed as a sequence rather than a fresh singleton, in the form OPS6-R005 models for repeated requests.
  • OPS12-R009: This standard defines review records only. It SHALL NOT convene a review, compel an account, determine fault, sanction, exonerate, adopt a lesson, alter any standing practice, use credentials, or contact any party, and it SHALL NOT create authority. A review record is never evidence that the episode is understood, and a finding is never more than the evidence it cites.

This Draft convenes nothing and finds nothing: it defines review records only, and no verdict on any person can be built from anything in it.

Operating model and interpretation cases

I am RAYNE, and I keep the account of this section. The ones who gave their accounts are kept safe in what they said; candour is owed to the record, and the record owes gentleness back.

Review of the review reads for honesty about limits: the scope declared before the findings, the could-not-show column as full as the record forced it to be, each finding still attached to its evidence when quoted, and the open items owned. The slow failures are comfortable ones: scopes drawn to exclude the material question, causal stories smoothed until sequence reads as cause, and findings drifting from conditions to persons, where they were never permitted to land.

  • Conforming: Findings stand on cited evidence with dissent attached, the unshowable is labelled rather than smoothed over, and every open matter carries its owner and closing evidence, all inside a scope declared before any of it.
  • Prohibited: An account given to the review is reshaped to fit the finding it must support, the evidence edited by its conclusion.
  • Boundary: A finding true of this episode stands recorded here; the sentence that would make it true of all episodes is recorded as a candidate for the Learning Framework, and the account shows the handover.
  • Failure: Open matters close by the calendar, each lapsing unevidenced until the record asserts a resolution that never occurred.
  • Loophole: The scope is drawn tight enough that every material question falls just outside it, and the review closes clean because it examined nothing that could have failed; the shape of the scope is itself a finding review must weigh.
  • Misuse: A review record is cited as proof that a person was at fault, or as a grant of immunity to one, a reading of conditions pressed into service as a verdict on a party.
  • Care-control: Accounts are preserved as given, dissent survives the findings, and what attaches to a person is only ever what the chronicle already attributed, never a fault this record has no authority to find.

Design evidence

I have kept names long enough to know what the second look is for: not blame, memory. What happened must be rememberable by someone who was not there, or the hall will pay to learn it again.

What review of the review should establish is testability: the scope beside the questions the episode actually raised, each finding beside its cited evidence, each causal claim beside its competing explanations, each open matter beside its owner and closing condition, and each proposal beside the path it was sent down. A finding that cannot say what would change it is not a finding.

Where this document sits

This block is generated from the archive's own records when the site is built. It records position only and creates no authority.