feat(forensic-taxonomy): source-credibility family + role ergonomics (27 F.5/F.6)#180
Closed
bryanmatthewsimonson wants to merge 1 commit into
Closed
Conversation
…(27 F.5/F.6) Phase 27 slices F.5/F.6 — completes the forensic ladder. The driving case: discrediting an article via the journalist's track record instead of evidence about its claims — persuasive power without evidential power, and no family covered it (the six existing families are the ACCUSED party's defensive canon; condemn-condemners and darvo/attack frame the deflector, not the dismisser). - F.5: additive side-neutral source-credibility/* family (Walton 1998 ad hominem argumentation): track-record-substitution (the WSJ example), borrowed-authority, guilt-by-association, motive-speculation. Full MANEUVER_GUIDE entries; every counter- indicator list includes "the speaker is REPORTING someone else's move" — the F.2 attribution rule restated per maneuver (guard- tested). Rides automatically into the modal picker, the LLM prompt guide, and the validators via FORENSIC_MANEUVER_GROUPS. - The blessed fallacy/* namespace is now surfaced in the findings prompt (it was validator-grammar-only). - F.6: role enum gains journalist / official / commentator (additive — role is wire-visible on published 30062s; no renames), and the finding modal no longer defaults to 'apologist' (debate jargon silently stamped onto journalism content) — role is an explicit choice with a friendly pre-check. - NIP_DRAFT: both additive vocabularies called out per the wire-change rule; consumers MUST tolerate role values beyond the original six. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Owner
Author
|
Landed on main via the #183 consolidation merge (the stacked-merge cascade closed/retargeted intermediates; every commit from this PR is in main). |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Phase-27 slices F.5/F.6 — completes the forensic ladder. Stacked on #179 (→ #178 → #177 → #176 → the Phase-26 chain).
F.5 — the persuasion-over-evidence family
The driving case: dismissing a WSJ article because its journalist previously mis-reported something — a move with more persuasive power than evidential power, and no existing family covered it (the six families are the accused party's defensive canon). New side-neutral
source-credibility/*family (Walton 1998):track-record-substitution— past errors substitute for engagement with the present claimsborrowed-authority— credentials invoked as if they settle a contested claimguilt-by-association— the claimant's associations as the operative rebuttalmotive-speculation— imputed motives substituted for evidenceEvery guide entry's counter-indicators include "the speaker is REPORTING someone else's move" — the F.2 attribution rule restated per maneuver, guard-tested. The family rides automatically into the modal picker, the LLM prompt, and the validators. The blessed
fallacy/*namespace is now stated in the prompt too.Crucially, these counter-indicate honestly: pointing out a track record alongside engaging the claims is calibration, not the maneuver — only substitution qualifies.
F.6 — role ergonomics
journalist/official/commentatoradded (additive; no renames — role is wire-visible), and the modal no longer silently defaults every finding to'apologist'— role is an explicit choice.Wire note
Both vocabularies are additive on kind-30062 tags; NIP_DRAFT updated per the wire-change rule ("consumers MUST tolerate role values outside the original six").
Tests: 1,796 green (2 new/updated pins) · build clean · lint 0 errors.
Next
K.1/K.2 — the Claude capture skill — is the last approved item (the C extract cache stays held until after Epistack).
🤖 Generated with Claude Code