Ordinariate slugs: vernacular, English, Latin, or a uniform structural form?

[JohnRDOrazio]

Ordinariates are the one cohort where the type word legitimately stays in the slug: rule 5 provides for them, and a military ordinariate has no see — its territory is a body of armed forces, not a place — so there is no place name beneath the type word to reduce to. That was settled in #2 and is not reopened here.

What is unsettled is which language the slug is in. The seed currently mixes English and vernacular, inherited from the upstream index's inconsistent styling. This is the ordinariate half of open question 4 (name authority), narrowed to the part that affects identifiers.

Current state — 33 entries, 10 English / 23 vernacular

English-styled (10):

circ:au-military-ordinariate-of-australia          circ:nz-military-ordinariate-of-new-zealand
circ:ca-military-ordinariate-of-canada             circ:ph-military-ordinariate-of-the-philippines
circ:id-military-ordinariate-of-indonesia          circ:ug-military-ordinariate-of-uganda
circ:ke-military-ordinariate-of-kenya              circ:us-archdiocese-for-the-military-services
circ:kr-military-ordinariate-of-south-korea        circ:nl-military-ordinariate-of-the-netherlands

Vernacular-styled (23): circ:de-deutsches-militarordinariat, circ:pl-ordynariat-polowy-wojska-polskiego, circ:es-arzobispado-castrense-de-espana, circ:it-ordinariato-militare-per-l-italia, circ:sk-vojensky-ordinariat-slovenska, circ:lt-lietuvos-kariuomenes-ordinariatas, circ:fr-diocese-aux-armees-francaises, and the rest.

Where the inconsistency actually bites

For au, ca, nz, ke, ph, ug, us, English is an official language, so the English styling is also the vernacular styling and nothing is wrong. The genuine mismatches are three:

• circ:id-military-ordinariate-of-indonesia — Indonesian
• circ:kr-military-ordinariate-of-south-korea — Korean
• circ:nl-military-ordinariate-of-the-netherlands — Dutch

kr has a second problem: "South Korea" is a civil-political name, not an ecclesiastical one.

The redundancy nobody has to accept

Every one of these slugs except circ:us-archdiocese-for-the-military-services encodes its nation twice — once in the ISO segment and again in the slug:

circ:au-military-ordinariate-of-australia      au + australia
circ:de-deutsches-militarordinariat            de + deutsches
circ:pl-ordynariat-polowy-wojska-polskiego     pl + polskiego
circ:it-ordinariato-militare-per-l-italia      it + italia

31 of the 32 military ordinariates do this. Since the nation segment already carries the country, the national qualifier in the slug is doing no work — which makes a uniform structural slug (option D below) worth considering seriously rather than as an afterthought.

Options

A. Always vernacular. Follows the see's own official self-designation. Costs: requires native-speaker verification, which the seed does not currently have — circ:ba-vojni-ordinarijat-u-bosne-i-hercegovine is already grammatically wrong (should be u Bosni i Hercegovini), and circ:at-katholische-militarseelsorge-osterreich uses Militärseelsorge (military pastoral care) rather than the ordinariate term. Also produces the widest character-folding surface.

B. Always English. Uniform and readable to the widest audience, but makes the registry's identifiers depend on translations that have no official standing for most of these bodies, and imports the kr "South Korea" problem generally.

C. Always Latin, from the Annuario Pontificio. circ:au-ordinariatus-militaris-australiae, circ:de-ordinariatus-militaris-germaniae. Language-neutral, authoritative, and consistent with treating the Annuario as the naming authority. Costs: least recognizable to downstream consumers, and needs the Annuario for all 33.

D. Uniform structural slug, dropping the redundant nation. circ:au-ordinariatus-militaris, circ:de-ordinariatus-militaris — or the English circ:au-military-ordinariate. The nation segment supplies the country, so the slug carries only the type. This makes every ordinariate identical in form, removes the language question entirely from the identifier, and pushes the vernacular/English/Latin naming into name and the multilingual labels where it belongs. Costs: the slug is then purely a type token, which sits oddly against rule 1's "type is an attribute" — though for a body whose identity genuinely is type + nation, that may be the honest encoding rather than a violation.

Out of scope

circ:us-personal-ordinariate-of-the-chair-of-st-peter is a personal ordinariate (Anglicanorum coetibus), not military, and is governed by different considerations.

Correction (was wrong in the first version of this issue): I originally wrote that its two peers were absent from the seed. They are present — circ:gb-our-lady-of-walsingham and circ:au-our-lady-of-the-southern-cross. They escaped my ordinariate sweep because their slugs contain no type word at all, which is itself part of the inconsistency: one of the three carries personal-ordinariate-of-the in its slug and the other two carry only the dedication.

That cohort has its own scoping problems (territorial extent under rule 5, and the styling inconsistency just noted) and is tracked separately in #5. It should not be decided by whatever is decided here.

Questions to settle

1. Which of A–D governs ordinariate slugs?
2. Does the answer bind only ordinariates, or does it set the general precedent for open question 4?
3. If vernacular (A), who verifies the forms? The ba and at entries show the seed cannot self-certify these.

All IDs remain drafts pending committee review. Related: #1 (identifier stability), #3 (territorial prelatures and abbacies), open question 4 in docs/schema-proposal.md.