Skip to content

Reconstruction Note

Document — instantiates Cautious Pattern Completion

A written record that lays a reconstructed whole out as three separate columns — what is known, what is inferred, and what is still missing — so speculation never inherits the authority of fact.

A Reconstruction Note is the artifact you write when you have pieced together a plausible account of some whole — a sequence, a rationale, a lost design — and need to hand it on without letting the guesses in it pass for facts. Its defining move is segregation by epistemic status: the note physically separates what was directly observed, what was inferred to bridge the gaps, and what remains unknown, so that a later reader inherits not just the reconstructed story but the exact map of which parts to trust. It is a narrative document, not a checklist or a table — but a narrative whose every inferred link is visibly marked as inference and whose every gap is left open rather than smoothed over.

Example

An engineer inherits a fifteen-year-old billing service with almost no documentation and needs to recover why it retries failed charges exactly three times with a widening delay. She reconstructs the design intent from what survives: the code (known), a 2011 commit message mentioning "gateway flakiness" (known), a since-departed architect's initials on a design ticket (known), and the observed fact that the third retry always waits a full hour (known). From these she infers the rationale — the delays were tuned to the payment gateway's historical outage windows, and three attempts were "enough to clear transient failures without double-charging" (inferred). And she marks what she cannot recover: whether the specific hour was measured or guessed, and whether a fourth retry was ever tried and removed (missing).

The Reconstruction Note ships all three layers, kept apart. The next maintainer reads a usable account of the retry logic and sees, in plain sight, that the one-hour delay is an unverified inference and the fourth-retry history is a genuine unknown. When that maintainer later considers shortening the delay, the note tells her precisely which part of the story is safe to lean on and which part is a guess she must re-check first — sparing her from treating a recovered rationale as settled architecture.

How it works

The note's discipline is layout, not prose quality: the value is in keeping the three statuses from bleeding into one smooth story.

  • Anchor on what is actually known. Begin by listing the surviving evidence — artifacts, records, direct observations — as the fixed points every inference must attach to.
  • Mark each inferred link as it is made. Wherever the account bridges a gap, flag the bridge explicitly ("inferred," "likely," "reconstructed") rather than stating it in the same voice as the evidence.
  • Leave the holes open. Record what is not recoverable as a first-class part of the note, so a missing piece cannot be silently filled by the next reader's imagination.
  • Keep the layers visually distinct. The known / inferred / missing separation must survive copying, quoting, and skimming — the moment it flattens, the note has failed.

Tuning parameters

  • Inference-marking granularity — whether marks sit on whole paragraphs or individual clauses. Fine marking is precise but heavy; coarse marking is readable but lets a single sentence mix fact and guess.
  • Reconstruction aggressiveness — how much of the missing whole to infer at all. A bolder note fills more but risks over-reach; a conservative one leaves more explicitly unknown.
  • Missing-piece visibility — how prominently the unknowns are surfaced (a dedicated section vs. inline asides). Prominent gaps resist being forgotten; buried ones quietly heal over.
  • Durability — how much effort goes into making the status-marks survive downstream reuse. High durability protects the note when it is quoted out of context, at the cost of a heavier format.

When it helps, and when it misleads

Its strength is preserving traceability across time and readers: months later, anyone can see which parts of a recovered account were observed and which were bridged, so the story can be revised without anyone having to defend the original guesswork. It is the written embodiment of Chesterton's fence — the principle that you should not tear down a structure whose original purpose you have only inferred, until you have recovered why it was built.[n1]

Its central failure mode is status decay: a reconstruction gets copied, summarized, or quoted, the "inferred" marks fall away in transit, and the guessed rationale returns as documented fact — narrative laundering, one reader at a time. A related misuse is the confidently written note whose smooth prose makes inference and evidence read identically despite the labels, so the marks are technically present but functionally invisible. The guarding discipline is to make the three layers structurally distinct rather than merely labeled, and to state the missing pieces as loudly as the known ones — because the unknowns are exactly what a future reader is tempted to quietly complete.

How it implements the components

  • partial_input — the note opens by fixing the surviving evidence: the artifacts, records, and observations the reconstruction is anchored to and answerable for.
  • assumption_marker — every inferred bridge between the evidence and the reconstructed whole is explicitly flagged as inference, kept in a different voice from the facts.
  • missing_evidence_marker — the unrecoverable pieces are recorded as a first-class part of the note, so a known gap is never silently completed.

It narrates the known / inferred / missing layers but does not build the element-by-element provenance ledger that assigns each claim a source type — that is source_trace, held by Source-Tracing Table, its nearest twin: the note is prose that keeps statuses apart, the table is a grid that maps sources. Nor does it run the conclusion_hold_or_release_gate that decides whether the reconstruction may be acted on — that is Withhold-Conclusion Checkpoint.

Editorial Notes

Form Classification

Form family: Record, Log & Register

Rationale: Reconstruction Note operates as a persistent ledger, log, register, or case record that preserves history and traceability because it a written record that lays a reconstructed whole out as three separate columns — what is known, what is inferred, and what is still missing — so speculation never inherits the authority of fact.

Independent corroboration: The frozen evidence defines Reconstruction Note as 'A written record that lays a reconstructed whole out as three separate columns — what is known, what is inferred, and what is still missing — so speculation never inherits the authority of fact', so its operative form is Record, Log & Register.

Review outcome: Independent reviewer agreement; high confidence.

Origin Attribution

Primary origin: History & Historiography

Origin pattern: Convergent development

Present-day reach: Multi-domain

Rationale: Separating attested evidence from historical inference is a core historiographic discipline.

Related originating lineages:

Review resolution: Both blind reviewers agree that history_historiography is the primary origin. Explicit reconciliation of alternate origin disagreement adopts reviewer_a's classification because separating attested evidence from historical inference is a core historiographic discipline. The resulting lineage records alternates=criminology_forensic, ethnography_qualitative_methods, origin_mode=convergent, and domain_reach=multi_domain; these describe formative provenance separately from later applicability.

Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.

Review outcome: Reconciled after independent review; medium confidence.

Notes

[n1] Chesterton's fence — G. K. Chesterton's principle that a reformer should not remove a fence until they understand why it was put there. A Reconstruction Note is how a maintainer records a recovered reason without pretending the recovery is certain — preserving both the inferred purpose and the fact that it was inferred.