Skip to content

Evidence Disclosure Packet

Compiled evidence packet — instantiates Transparency for Accountability

Assembles the underlying evidence a decision rests on into one sourced, indexed bundle, released so an outsider can check the factual basis for themselves.

An Evidence Disclosure Packet is the curated bundle of the actual evidence beneath a decision — the documents, data, tests, and sources — assembled, sourced, and indexed so someone outside the decision can examine the basis rather than take the conclusion on faith. Its defining move is disclosing the inputs, not the verdict: where a reasons statement gives you the argument and a decision log gives you the outcome, the packet hands over the raw material both were built from, in a form organized for scrutiny. It is what turns "we considered the evidence" into "here is the evidence we considered — check it."

Example

Before trial, a prosecutor must give the defence the evidence the case rests on. The evidence disclosure packet is that transfer made concrete: an indexed set of the police reports, lab results, witness statements, and surveillance stills, and — critically — any material that cuts against the state's theory, each item tagged with its source and date. It is not the prosecutor's argument (that comes at trial) and not a summary of the case; it is the underlying record, paginated and catalogued so the defence can find exhibit 14, trace it to the officer who logged it, and test whether it means what the state says. When the packet includes an exculpatory lab note, the disclosure obligation — Brady v. Maryland — is satisfied not by mentioning the note but by handing it over in a form the other side can actually use. A packet that dumps ten thousand unsorted pages to bury that note fails the obligation even though nothing was technically withheld.

How it works

  • Assemble the basis, not the argument. The packet holds the evidence itself — exhibits, data, sources — leaving interpretation to the reasons statement it supports.
  • Source every item. Each exhibit carries its origin, custodian, and version, so a reviewer can trace it rather than trust it.
  • Index for navigation. A contents list, stable exhibit numbers, and cross-references make the bundle checkable; an unsorted dump is disclosure in name only.
  • Include the adverse. A complete packet contains the material that weakens the decision, not only what supports it — the difference between disclosure and advocacy.

Tuning parameters

  • Completeness vs. burden — everything touching the decision versus the material subset; over-inclusion buries the key item, under-inclusion hides it.
  • Indexing depth — a bare file list versus a cross-referenced, searchable catalogue; depth is what makes a large packet usable rather than obstructive.
  • Redaction boundary — how much third-party or sensitive material is masked before release; over-redaction guts the evidence, under-redaction breaches privacy.
  • Native vs. rendered form — original files (with metadata) versus flattened printouts; native form preserves provenance a reviewer may need.
  • Recipient scope — a specific reviewer, an opposing party, or the public; scope drives both redaction and format.

When it helps, and when it misleads

Its strength is that it lets a decision be tested at its foundation: a reviewer, opponent, or auditor can verify that the evidence exists, means what was claimed, and was not cherry-picked — the deepest form of accountability, because it does not rely on trusting the decider's summary.

Its failure modes are about form as concealment. The classic abuse is the document dump — technically complete disclosure engineered to be unusable, the important item lost in volume. Its mirror is over-redaction that leaves the evidence present but illegible. A packet can also mislead by omission that looks like completeness: everything supportive included, the adverse item quietly absent. The standard that defines an honest packet is the Brady principle — that material evidence, especially exculpatory evidence, must be disclosed in usable form — which treats burying and withholding as the same failure.[1]

How it implements the components

This mechanism fills the evidence slice of the archetype — the basis, not the reasoning or the rules:

  • evidence_basis_record — the packet is the assembled record of the evidence a decision rests on: exhibits, data, sources.
  • provenance_and_version_link — each exhibit is sourced and version-stamped so its origin and chain can be verified.
  • comprehensibility_and_format_layer — indexing, stable numbering, and organization make the bundle navigable and citable rather than a dump.

It hands over the evidence, not the argument built on it (that's Reason-Giving Template) nor the criteria it was judged against (that's Rule and Criteria Register); and it does not translate the material into plain language for a general audience — that's Plain-Language Transparency Report.

References

[1] In Brady v. Maryland (1963), the U.S. Supreme Court held that the prosecution must disclose evidence favourable to the accused that is material to guilt or punishment. The principle is widely cited for the idea that disclosure means providing material evidence in usable form — so that burying it in volume, or omitting the adverse item, is itself a failure to disclose. ↩