Should `Jose E Maria Santos` join like `Jose e Maria Santos` does? (the single-letter connective's Latin-capital veto was never adjudicated)

[derek73]

Background

A recognized connective joins its neighbors into one name part (rules.md#P3). Single-letter connectives carry two carve-outs: the three-word guard (Google Code issue 11, "john e smith"), and a second one that has never been separately decided — a bare Latin capital reads as an initial and never joins, while a Cyrillic capital joins.

Current state (measured, 2.2.0dev)

input	result
Jose e Maria Santos	given="Jose e Maria" — joins
Jose E Maria Santos	given="Jose", middle="E Maria" — vetoed
Хосе И Мария Сантос	given="Хосе И Мария" — joins

"И".isupper() is True, so this is not a Unicode-uppercase rule — it is an ASCII-shape veto. #267 examined this territory and was closed 2026-07-30 with "v2.0 behaves this way by default … verify it was the right call," but the closing comment showed only the Cyrillic half; the Latin-capital veto half was never adjudicated anywhere findable. rules.md#P3 currently records the asymmetry as an Accepted consequence pending exactly this question.

Options

1. Bless the veto — an uppercase single letter beside Latin names is overwhelmingly an initial (Jose E. Maria Santos minus a period); the asymmetry is correct because Cyrillic text doesn't abbreviate initials with bare capitals the same way. Cost: JOSE E MARIA SANTOS (all-caps data) keeps the veto — worth one measurement before blessing.
2. Drop the veto — case-insensitive connectives everywhere; E joins. Cost: genuine initials named E lose (Jose E Maria Santos where E abbreviates Enrique), and the change is v1-visible.
3. Extend the veto to all scripts — symmetric, but Cyrillic И genuinely is the conjunction "and" written capitalized in title-case names, so this breaks the case Should a bare 'И' parse as an initial or the conjunction 'and'? (e.g. 'Хосе И Мария Сантос') #267's closure blessed.

Open

Whether all-caps input (option 1's cost) occurs in real data enough to matter — the differential corpora can answer for the shipped test banks but are structurally blind to all-caps variants nobody wrote down.