Should an Indonesian choice read H. (Haji) and R. (Raden) as titles? (H. Ganjar Pranowo gives given 'H.') #1205

Indonesian official lists write the pilgrimage title Haji as H. and the Javanese noble title Raden as R.: one letter and a period, in front of the name. T5 (the written shape of an initial) makes that an initial, so it takes the given name's slot (measured on master 5f22996):

The forms of two letters or more already read as titles, by H2 (the leading-abbreviation title): HM. Iqbal Wibisono (Haji Muhammad), HA. Chozin Chumaidy, Hi. Amin Lasena, Hj. Erlina (Hajjah) and RR. Ida Resmi Nurani (Raden Roro) each read the abbreviation as the title. #1180 asks about Haji written out; this is the abbreviation.

Measured (2026-11-08) on Indonesia's House of Representatives, Regional Representative Council and 2018 regional heads (OpenSanctions id_dpr, id_dpd and id_regional_2018, 2,792 distinct names): 136 open with H. (128 spaced, 8 glued as in H.M. or H.Hari Wuryanto), and 9 with R.. All 145 read the letter as the given name. The lists have no name fields, so the reading expected here comes from the convention: in this position, in these lists, H. is Haji and R. is Raden.

The cost

This can't be a default: H. Ross Perot has the same shape. ORCID's Indonesian records say little about the competing reading. None of its 20,000 Indonesian given boxes opens with H or H. (its researchers don't write Haji there either), while 13 open with R and 2 with R. (R Reza El Akbar, R. Bariq Maulana), which could be Raden or an initial.

Simulated as readings on everywhere (Lexicon.default().add("h", kind="religious").add("r", kind="nobility")):

The middle initial is untouched, but a one-letter title reading also reaches a word of period-joined chunks: S.H. and H.N. become titles through their H. With config_diff.py diff --local orcid-researchers --group ID, 30 of 19,925 Indonesian ORCID names move: 15 R, 3 H/H., and 9 through S.H. or M.H. (the degrees in #1204).

Proposal

When: the parse is under a choice of id, and H. or R., written with its period, opens the part of the name that carries the given name, with a name word after it.

Then: it is a title: H. Haji, R. Raden. Each is a default = false reading in id.toml, so the default keeps H. Ross Perot as it is.

Because: the caller's choice says the data is Indonesian, where the one letter at the head is the title. This is the same shape as #1161 (M. under a French choice). There's no locales.ID today, but a caller can make the choice: parser_for(Locale(code="id", lexicon=Lexicon(), languages=frozenset({"id"}))).

Except:

• Inside a word of joined chunks (S.H., H.N., A.R.), the letter is a chunk, not the title, so the word reads as it does today.

Questions

• Is an Indonesian choice enough licence, as Should a French choice read M. as a title, now that French data shows almost no initials? (M. Jean Dupont gives given 'M.') #1161 asks for French?
• Should the reading need the period, so that H Smith stays an initial even under the choice?
• Keeping a one-letter title out of period-joined words: should that be the reading's own property, or a rule that a title chunk needs two letters?
• Malaysian lists write Hj. (Should Malaysian honours read as titles? (Tan Sri Muhyiddin Yassin gives given Tan) #1180). Does a Malay choice want H. too? No list measured here writes it.