Skip to content

Should de join a title phrase as of does? (John Smith, Professor of History reads the title whole; Ana López, Profesora de Historia gives family Ana López de Historia) #1462

Open

Description

@derek73

The question. English reads an office and its complement as one title, before the name or after a comma. The same phrase in Spanish or Portuguese does not, by default or under the pack:

>>> parse("Professor of History Ana Lopez")
title='Professor of History', given='Ana', family='Lopez'
>>> parser_for(locales.ES).parse("Profesora de Historia Ana López")
title='Profesora', family='de Historia Ana López'

>>> parse("John Smith, Professor of History")
title='Professor of History', given='John', family='Smith'
>>> parser_for(locales.ES).parse("Ana López, Profesora de Historia")
title='Profesora', family='Ana López de Historia'
>>> parser_for(locales.PT).parse("Ministro da Fazenda Carlos Souza")
given='Ministro', family='da Fazenda Carlos Souza'

of is a connective only; de and da are also family-name particles, and the particle reading wins. After a comma the name loses its given name and gains a surname nobody bears, a cost accepted when profesor and profesora became default titles (#804, decisions/salutation-titles).

Two options.

  • After a comma only. The name is complete before the comma, so a title followed by de cannot be opening a surname: the particle reading has nothing to attach to. This removes the cost John Smith, Pr and Profesor Juan Pérez read the title as a given name: add French and Spanish professor titles #804 accepted and needs no count. It is also the order Spanish and Portuguese write, the office after the name.
  • Leading as well. Here the pair that has to be told apart is one shape: Profesora de Historia Ana López should read title Profesora de Historia, while Doctora de la Vega must keep family de la Vega. Both are a title, a preposition and a capitalised word, and what separates them is whether a name follows the phrase, a count of words and not a vocabulary fact.

Why it has no rule today. The leading shape was declined on measurement on 2026-10-30 (decisions/vocabulary-model, phase 3g): over 8,526,439 names in twenty corpora, no leading title and connective was a title phrase with a name behind it. A scan on 2026-11-18 of corpora added since, for a listed title followed by de/da/do/dos/das/del/di:

Corpus Names Leading After a comma What they are
dof-signatures 33,565 3 0 Director de …, an office
tse-donors 854,531 9 0 given names that are listed titles (Grace da …)
ctgov-officials 539,368 2 4 leading: a professor with a particle-led surname; after a comma: Chef de … three times
orcid-researchers 1,831,533 not scanned 0

The scan finds only listed titles, so an office word the vocabulary lacks (Secretario de …) is not counted: the figures are a floor. Every list held writes the shape rarely. This issue keeps the question open for parity with English, not because a measured population needs it.

Open:

  1. After a comma only, or leading too?
  2. For the leading shape, what signal separates an office from a particle-led surname: a word count behind the phrase, a list of office words that take a complement (ministro, secretario, director), or the complement's own vocabulary?
  3. Should it reach only readings a pack turns on, so the default's parses stay put? profesora has been read by default since John Smith, Pr and Profesor Juan Pérez read the title as a given name: add French and Spanish professor titles #804.
  4. The nine Brazilian names are the leading shape's cost: a rule keyed on any listed title would take a given name's surname as its office.
  5. Should an office word that is no listed title (secretario, ministro) be read as one after a comma? Juan Pérez, Secretario de Estado gives given Secretario, middle de Estado, family Juan Pérez today. This overlaps Should a spelled-out degree or job after a name and a comma read as one? (Ahmed Ali, Bachelor gives given Bachelor, family Ahmed Ali) #1457.
  6. A list that writes the shape is needed before a rule can be measured. Candidates: government gazettes and ministry staff lists.

Related: #970 (a connective joining two titles), #1457 (a spelled-out job after a comma), #804.

Source: docs/design/vocabulary-model-rfc.md, phase 3g and its Declined entry.

Activity

  1. self-assigned this
    on Nov 19, 2026
  2. added
    designChanges what the model can express (a shape decision)
    titlesTitles and honorifics (rules/H)
    particlesParticles and surname prefixes (rules/P)
    locale-packsLocale packs and language choices (nameparser.locales)
    commasCommas and name structure (rules/C)
    on Nov 19, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

commasCommas and name structure (rules/C)designChanges what the model can express (a shape decision)locale-packsLocale packs and language choices (nameparser.locales)particlesParticles and surname prefixes (rules/P)questiontitlesTitles and honorifics (rules/H)

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions