[derek73]

The ninth and last title batch of phase 2j (#752), which answers the last of #853's questions. sheikh, sheik, shekh, shaik and cheikh gain dated given readings, so each is a title only inside a title phrase (rules/H9). shaykh stays a title.

Closes #853.

What changes

input	1.4.0 – 2.6.0	now
Sheikh Ahmed Khan	title Sheikh, given Ahmed	given Sheikh, middle Ahmed
Mr Sheikh Ahmed (#853)	title Mr Sheikh	title Mr, given Sheikh
Khan, Sheikh	title Sheikh	given Sheikh, family Khan
Cheikh Anta Diop, Shekh Abdul Rahim	title	given Cheikh / Shekh
Sheikh Sultan Khan	title Sheikh Sultan	given Sheikh, middle Sultan
His Highness Sheikh Ahmed Khan, Grand Sheikh Ahmed, Sheikh of Kuwait	title	unchanged
Sheikh Mohammed bin Rashid Al Maktoum	title Sheikh	given Sheikh (the cost below)

No listed phrase holds any of the five spellings, so nothing is reported. The other sites read as for shaikh. Period-marked Sheikh. at the front is still a title by H2's shape.

Which spellings

I vetted each spelling by where UK PSC's filers put it:

spelling	forename box, first word	title box
sheikh	174	19
sheik	23	1
shekh	6	1
shaik	14	0
cheikh	14	0

shaykh, the scholarly transliteration, has no given-name evidence anywhere, so it stays a title. That costs one jrc-names person spelled Shaykh beside Sheikh/Cheikh, who is now apart.

The cost (rules/H9, Accepted; Derek, 2026-10-27)

A ruler, cleric or leader written with the bare title reads it as the given name. This is the class accepted for king, sultan, shaikh and prince. The rules/H9 Accepted bullet states it (example Sheikh Mohammed bin Rashid), and decisions/H9 records Derek's acceptance, with the bound-pair split below.

• jrc-names: 174 persons have a spelling opening with a bare form of the word. By my reading:
  • about 40 are Gulf rulers and members of their houses;
  • about 55 are clerics and militia, tribal or political leaders;
  • about 60 are people whose own name opens with it, mostly Senegalese and Mauritanian Cheikhs, plus Bangladeshi and Malaysian bearers;
  • the rest are unclear.
• PSC: 16 of the 244 moved records file the word in the title box (13 broken, 3 changed). Qatari and Emirati ruling-house members there file it in the forename box.

A bound given name behind it

Read as a name word, the word licenses no bound-given join. Sheikh abdul Rahman now reads given Sheikh, middle abdul, family Rahman, where master read title Sheikh, given abdul Rahman. Prince abdul Rahman has read this way since #908. A style in front keeps the join (His Highness Sheikh abdul Rahman → given abdul Rahman). decisions/H9 records it as part of the cost.

Worked examples moved

Sheik was rules/P5's given-name title licensing the bound join, and rules/H1's Sheik Abu Bakar. Sir replaces it. I checked Sir against Sheik on master across 10 shapes under all three orders, and it reads the same in every one. The title-run row Sir Sheikh abdul rahman is now Dr. Sir abdul rahman, which still shows a run keyed by its last title (1.4.0 read last rahman).

Measured

Against master 79a4111f, py3.11, every local corpus with --base; counts and token shapes only.

• PSC: 209 fixed, 13 broken (the title-box filings above), 22 changed. Most of the changed records now read the given name but split the filer's middle box differently.
• ctgov-officials: 10 moved.
  • 8 are fixes: 4 are Aa Sheikh, CRED, whose credential master read as the given name, and 4 open with Sheikh or Cheikh.
  • 2 are a lower-case name and a credential run, still wrong.
• fec-individuals: 26, all fixes (AA, SPELLING…); 18 of them had read no family name.
• nepal-candidates: 267, every one opening with Shekh (265) or Sheikh (2).
• jrc-names: 571 spellings.
• Spelling pairs:
  • jrc-names: 37 persons now alike and 1 apart (the Shaykh person above).
  • All 12 persons the shaikh batch split are alike again.
  • nepal-candidates: 280 pairs now alike and 0 apart; the romanized Shekh now reads as its Devanagari does.
• No moves: tse-donors, ukec-individuals, CLDR/GND agreement.
• Committed corpora: two corpus_issues names (A title displaces a particle out of leading position, so Dr. Van Johnson reads as one surname #367's Sheik Abu Bakar and Sheik abdul salam) move, beside the batch's own rule examples.
• v1: Constants().titles keeps all five spellings with no edit (Keep the titles 2.6.0 listed in v1's Constants().titles (#752) #904's sweep).

#853's questions, all answered

1. A shape change to the title run. Declined in favour of per-word given readings (decisions/vocabulary-model item 12). The nine batches are that answer.
2. Per word. Yes, and every word in Should a title word written after another title be the given name (Mr Sultan Khan, Mr Md Karim Uddin)? #853's table now has its reading. Phase 2j moved:
   • wing (the modifier);
   • mahdi and bard;
   • buddha, siddha and gyani;
   • king;
   • sultan and sultana;
   • shaikh;
   • prince and princess;
   • sheikh and its spellings.
3. Ma. after a title. Answered by Should Chas. Smith read given Chas.? Abbreviated given names read as titles today #789 (phase 2k): Mrs Ma. Liza Reyes reads given Ma.. The period-marked Md. stays a title: Mr Md. Karim Uddin reads title Mr Md., an accepted cost Derek decided on 2026-10-27. This PR records it in decisions/indic-honorifics.
4. md. Decided by Read a leading Md as the given name it abbreviates, not a title (#853) #860: bare Md is the given name.

Not done here

migrate.rst has no 2.7 "Changed in" entries for the phase 2j shapes. That is release-checklist work for 2.7.

Docs

• rules/H9: examples.
• rules/P5, H1: stand-ins.
• decisions/H9: the dated entry, with the counts, the cost, Derek's decision and the measurement.
• decisions/indic-honorifics: Should a title word written after another title be the given name (Mr Sultan Khan, Mr Md Karim Uddin)? #853's question 3 decided: the period-marked Md. stays a title, a Standing line and a dated entry.
• decisions/P5: Standing lines.
• decisions/vocabulary-model: a pointer.
• Release log: a bullet.
• The five vocab entries: counts recorded.
• Case rows: 12 new, 1 rewritten.
• Comments: test, code and tools comments updated.

Checks

• uv run pytest and --python 3.15: 31,856 passed each.
• Also: -n0 -m serial, mypy, ruff, sphinx-build -b doctest.
• Design-docs review in two rounds. The first found stale prose: decisions/H1's Standing line, the nepal-candidates rows in AGENTS.md and tools/differential/README.md, the शेख divergence line, a test comment, the given_name_titles docstring and the Shekh case note. It also found the bound-pair split, now recorded. The second, over the fix commit, tightened four wordings. All are fixed here.

Refs #752

🤖 Generated with Claude Code