Skip to content

Should common British post-nominals (KC, FREng, DL, FRS, PC…) be in the vocabulary, so Jane Doe KC keeps family Doe? #965

Closed

Description

@derek73

Found by the govuk-people local corpus (#962), fetched 2026-10-30. Names below are invented; the corpus is personal data and is not cited by name.

Today

An unlisted post-nominal ending a name is placed by position, so it becomes the family name; after a comma, the name reverses.

  • Jane Doe KC → given Jane, middle Doe, family KC
  • John Smith FRS → family FRS
  • Jane Doe, DL → given DL, family Jane Doe
  • an unlisted last word drags a listed credential into the family: Air Chief Marshal The Lord Doe KG GBE KCB DL → family KG, while … The Lord Doe KG GBE → family Doe
  • a listed credential in front lets an unlisted one join the run in one order only: Jane Doe CBE KC → suffix CBE KC, but Jane Doe KC CBE → family KC

qc is listed in en.toml as a plain credential (credential = {}), so Jane Doe QC reads suffix QC; kc, dl, frs, freng, pc and the others below have no entry.

Size (govuk-people, 6,021 names, 2026-10-30)

  • KC: 36 of the 42 names writing it read it as the family (the other 6 have a listed post-nominal beside it). Also ukec-individuals 4, ctgov-officials 26 (plus K.C. 3).
  • Other post-nominal-shaped words ending the family: 27 names — FREng 4, FIET 2, LLB 2, DL 2, KFSM 2, FRSA 2, and one each of KG, MMS, PC, FNucI, MoID, MChemA, ACMA, HMI, BVSc, OM, MRTPI, FRAeS, FRS. Recipe: the family's last word matches (?!Ma?c[A-Z])(?!O[A-Z][a-z])[A-Z][A-Za-z]*[A-Z][A-Za-z]* and is not KC.
  • After a comma the name reverses in 16 names: FREng 4, DL 4, CChem 2, and one each of MICFor, MRPharmS, MIBiol, FRMetS, MChemA, CStJ.

The hard case: KC

KC is also a Nepali surname, written bare as well as dotted. ORCID's family box holds KC for 584 people, 163 of them in Nepal; nepal-candidates writes it K.C. 342, K.C 129, Kc 21; psc-individuals has Kc 30. Measured with Lexicon.add:

  • as a suffix acronym (periods stripped): Ram K.C. → suffix, which breaks Should Ram Bahadur B.K. keep B.K. as its family name? #864;
  • as a plain suffix word (like qc): K.C. stays a name, but Ram KC and Ram Kc lose their family;
  • as an ambiguous suffix word (the S11/S6 count-and-case gate): John Smith KC → suffix and Ram KC → family, both right; Ram Bahadur Kc → family by case; but Ram Bahadur KC → suffix KC, and Lord Doe KC still → family KC (a title leaves no word to spare).

Bare KC does not reach #864's paired-initials exception; only the dotted K.C. does (rules/S3 records John Smith K.C. → family as an accepted cost).

To settle

  • A vocabulary batch: each word through tools/vocab/query.py, tools/name_usage.py and the collision criterion (decisions/vocabulary-collisions). KC, PC and DL need the most care: plausible surnames or initials.
  • Which listing KC takes, if any.
  • Whether an unlisted capitals word behind a listed credential should join the run in both orders.

Activity

  1. self-assigned this
    on Oct 31, 2026
  2. added this to the 3.0 milestone on Nov 3, 2026
  3. derek73 commented on Nov 4, 2026

    @derek73
    OwnerAuthor

    Closing for 3.0. #976 listed the batch: KC, PC and DL as credentials gated by the names they collide with, 18 plain credentials, FIET and FREng as ambiguous credentials, and MoID refused. The govuk-people moves are in decisions/S11 and decisions/suffix-acronym-collisions (2026-10-30).

    What is left moves to 3.1 as #1053:

    • the mixed-case credential written in its own spelling (FREng, and CEng, which a listing alone cannot fix: measured, the ambiguous listing leaves GOV.UK's 3 names unchanged and a plain listing breaks 9 bearers of the surname Ceng);
    • the join-order question (XYZ CBE vs CBE XYZ).

    MNSS (1 GOV.UK name) is left unlisted: no source says what it stands for.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

questionvocabularyPer-word vocabulary model and language packs

Projects

No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions