Repository navigation
An input composed entirely of titles should report an ambiguity — "The Right Hon. the President of the Queen's Bench Division" #491
Description
Activity
- added 4 commits that reference this issue
on Sep 8, 2026 Shipped in 2.3.0 via #518 as
AmbiguityKind.TITLE_OR_NAME("title-or-name"), with the rule this behavior never had: rules.md#H4.The rule. An input the title peel eats down to one last title-vocabulary word reads that word as the name by convention — a name parser, not a title parser — and reports it:
Lord Chancellor→ familyChancellor, the Queen's Bench string → familyDivision,Dr. King(deliberate:kingis title vocabulary). One emitter at the assign site where the word is placed, so the report is the same under every order; a lone title (Dr.,Sir,Prince of Wales) has no name in it and reports nothing. The reading is unchanged from the v1-xfail triage; only the silence is gone, which is what this issue asked for.Derek's refinement, made while deciding #449: a lone JOINED unit that carries title vocabulary is a title-or-name doubt, not a given-or-family one —
John of Prince,Smith and King(the king/judge/bishop collision set inside a join, stated in the rule and pinned). That doubt is not silenced by a maiden marker or a claimed word, which decide the field rather than whether the word is a title (Lord Chancellor née Jones). Two corpus names take that reach:Attorney General of MinnesotaandDeputy Secretary of State, a peeled title in front of a joined unit standing last.The all-suffix half. An input that is nothing but post-nominal vocabulary reads its first post-nominal as the name by the same convention and reports the EXISTING
suffix-or-name(Rinpoche,QC MP); no third kind.Jr.alone is a title by H2's abbreviation shape, so it is not this case. The lone CJK honorifics (さん,씨,선생님) are silent because the script rule decided them, andPolicy(script_orders=())reverses that, which is pinned as the positive control.Nothing in the parse moved: the differential shows ambiguity-only diffs at the 2.x baselines under three
feat(#491)rules, zero role changes. Rationale:decisions.md#H4, and the#O5entry for the shared site.
A name parser handed no name part still has to report something, and the current answer (the last title-vocabulary word becomes the family name —
Divisionhere) is the convention working as accepted, not a bug (the v1 xfail is being retired as not-fix).What's missing is the record-the-guess half of the 2.0 doctrine: when every token is title vocabulary, the parser decided by convention and should say so on
ambiguities, the way single-word-name conventions are being examined in #449.Open mechanics: a new
AmbiguityKindor an existing one, and whether the same report covers all-suffix inputs.