The Gold Standard · Archaeology
Make the decision path visible
Recovery after a second suspension can deadlock. One team wants another narrow patch, another cannot reconstruct the real population, and the release owner has no evidence that a restored control will survive ordinary work. Waiting does not preserve neutrality when uncertain outputs remain queued for release.
The gold standard names the deadlock, protects the affected archaeological route and escalates the evidence question to an authority able to reduce scope, require redesign or revoke conditional approval. Revocation is a reasoned state, not an admission that every earlier interpretation was wrong.
Choose the route closest to your decision
Chapter contents
Chapters 1–3
Chapters 4–6
Chapters 7–9
Chapters 10–12
CHAPTER 1 OF 12 · NAME THE DEADLOCK
1. Define recovery deadlock
A recovery deadlock exists when the suspended route cannot produce agreed evidence for coverage, effectiveness or enforcement within the decision window. Record the unresolved propositions, not only the meeting status.
Disagreement is not itself deadlock if a planned test can resolve it. Deadlock begins when evidence is unavailable, ownership is circular or the proposed test cannot reach the disputed question.
CHAPTER 2 OF 12 · NAME THE DEADLOCK
2. Restate the conditions that approval depended on
List the protected population, release gate, threshold, provenance requirement, reviewer capacity and monitoring consequence. Identify which conditions failed during the second suspension.
This prevents recovery from being judged against a softer replacement standard invented after the breach.
Evidence check
Check whether the scope, owner and decision consequence are explicit before continuing.
CHAPTER 3 OF 12 · NAME THE DEADLOCK
3. Classify the evidence gap
Separate missing source records, unstable denominators, unrepresentative tests, conflicting scientific criteria, unenforced gates and capacity failures. Assign a latest useful date for each evidence route.
Different gaps need different responses. More samples cannot repair a gate that does not change release authority.
| Gap | Decision question |
|---|---|
| Source record | Can population be rebuilt? |
| Denominator | Is coverage knowable? |
| Gate | Does breach block release? |
| Capacity | Can review operate? |
CHAPTER 4 OF 12 · PROTECT THE RECORD
4. Maintain a bounded protective hold
Keep affected bridge-model outputs from release, preserve the queue and allow independent manual interpretation only with explicit provenance. Continue unaffected routes when their independence is demonstrated.
The hold should not convert uncertain automated results into informal advice that reaches the same decisions without review.
CHAPTER 5 OF 12 · PROTECT THE RECORD
5. Preserve cases needed to resolve the dispute
Freeze source features, classifications, model versions, reviewer notes, overrides and final decisions across the breach window. Retain awkward and contradictory cases.
A tidy retrospective set can erase the very cases that show why recovery stalled.
Evidence check
Ask a second reviewer to reconstruct the example from the preserved evidence and stated assumptions.
CHAPTER 6 OF 12 · PROTECT THE RECORD
6. Worked example: complex joins cannot be reconstructed consistently
After a second suspension, two teams rebuild the population of complex cross-trench joins. One uses the current taxonomy; the other reconstructs complexity from source geometry. The lists disagree materially, and many older fields were overwritten.
Governance keeps automated release suspended, samples the preserved source drawings and finds no stable denominator. It revokes conditional approval for that route and retains manual interpretation. Changed-context check: if an independent archive can reconstruct the original features with validated coverage, it may support a new redesign review, but it does not erase the present revocation.
| Route | Treatment |
|---|---|
| Affected model | Suspend |
| Manual review | Use explicit provenance |
| Independent scope | Verify independence |
| Backlog | Preserve order |
CHAPTER 7 OF 12 · ESCALATE THE DECISION
7. Escalate evidence, not institutional pressure
Provide the decision authority with the disputed claims, available records, failed tests, exposure, reversible options and cost of continued uncertainty. Keep advocacy separate from evidence ownership.
Escalation should make the unresolved boundary clearer rather than reward the group with the strongest deadline.
CHAPTER 8 OF 12 · ESCALATE THE DECISION
8. Compare restriction, redesign, replacement and revocation
Assess whether a smaller valid scope exists, whether missing evidence can be regenerated, whether a new control is independent and whether manual capacity protects the record.
A familiar model should not retain nominal approval when every practical use requires an exception.
Evidence check
Verify that the control still reaches the intended decision under a changed but plausible context.
CHAPTER 9 OF 12 · ESCALATE THE DECISION
9. Revoke conditional approval with precise effect
Record the authority, scope, effective time, blocked releases, treatment of earlier outputs and conditions for any future application. Notify all downstream owners.
Revocation ends current permission; it does not automatically retract every historical interpretation. Earlier uses need consequence-based review.
| Option | Required evidence |
|---|---|
| Restrict | Valid smaller population |
| Redesign | Independent control proof |
| Replace | Defensible successor |
| Revoke | Current route not supportable |
CHAPTER 10 OF 12 · CLOSE OR RE-ENTER
10. Repair records and downstream decisions
Trace unreleased and released cases in the exposure window. Correct provenance, replace material outputs and label stable interpretations with the evidence that supports them.
The response should distinguish governance failure from archaeological validity so repair is neither too broad nor too narrow.
CHAPTER 11 OF 12 · CLOSE OR RE-ENTER
11. State limits and misconceptions
Revocation cannot reconstruct destroyed source fields or guarantee that every informal copy is found. Manual alternatives can also produce delay and inconsistent judgement.
Record residual uncertainties, backlog capacity and the evidence threshold for any future redesign application.
Evidence check
Confirm that versions, notices and unresolved limits remain visible in the final record.
CHAPTER 12 OF 12 · CLOSE OR RE-ENTER
12. Audit closure and possible re-entry
Track deadlock questions, evidence attempts, hold scope, decision authority, revocation notices, repaired cases and future-entry criteria.
The gold standard does not keep conditional approval alive by inertia. When recovery cannot be evidenced, governance protects the record, closes current authority and leaves a clear route for genuinely new proof.
| Field | Purpose |
|---|---|
| Scope | Defines blocked use |
| Effective time | Bounds authority |
| Earlier outputs | Directs review |
| Re-entry | States new proof |
Frequently asked questions
Is disagreement enough to revoke approval?
No. First identify whether a practical evidence route can resolve the disagreement within the decision window.
Does revocation mean every previous model result was wrong?
No. It ends current authority; earlier outputs should be reviewed according to their evidence and consequence.
Can a manual route continue?
Yes, when it is genuinely independent, adequately resourced and records its provenance and limits.
How can the route return later?
Through a new application that meets explicit population, validation, enforcement and monitoring criteria with fresh evidence.
Continue through verified routes
Core routes
Archaeology subject root · Gold Standard of Education · Gold Standard directory · How X Works hub
Specialist routes
- Renewed conditional monitoring drift, restoration-condition breaches and second suspension
- Scope revalidation approval restoration, evidence sufficiency disputes and renewed conditional monitoring
- Post-acceptance monitoring trigger breaches, conditional approval suspension and scope revalidation
- Control redesign validation disputes, competing success criteria and governance acceptance