Skip to content

Should a particle that is also a credential (vd, mc) count in a credential run after a comma? #554

Description

@derek73

#544 gave two rules the same question to answer, can this suffix word stand in a credential run?, and they answer it differently for the two words that are both particles and unambiguous suffix vocabulary, vd and mc.

  • C1's run test (after a comma, _segment.py) counts them as suffix words.
  • The anchor (_pieces._anchors) excludes them, because a particle chains onto the name after it (rules.md#P2). The anchor exclusion was added on PR fix(#544): a credential run keeps its ambiguous members #552 after Smith vd Ma, John lost the contiguity of its family name.

So the same words read differently depending on the path:

text 2.3.0 master PR #552
John Smith, vd Ma family John Smith vd Ma family John Smith vd Ma suffix vd Ma, reported
John Smith, PhD vd Ma given PhD, middle Ma, suffix vd given PhD, middle vd Ma suffix PhD vd Ma, reported
Jan Smith PhD vd Ma suffix PhD vd Ma family vd Ma family vd Ma
Doe, Jane PhD vd Ma middle Ma, suffix PhD vd middle vd Ma, silent middle vd Ma, silent
John Smith, vd suffix vd suffix vd suffix vd

(Mc reads the same way as vd.)

With a comma, vd Ma is a credential run; without one, the same two words are a family name. The last row shows why the comma reading isn't simply wrong: vd alone after a comma has always read as a suffix.

The question: after a comma, should a particle that is also a suffix word (a) count in the run, as today, so John Smith, vd Ma is a credential; or (b) end the run as it ends the anchor, so the part reads as the listing form?

(b) makes one predicate answer both questions: whether a word "has a live name reading" would decide both C1's run membership and the anchor. The cost is that John Smith, vd Ma returns to master's family John Smith vd Ma, a one-word listing-form part with no given name.

Aside: Doe, Jane PhD vd Ma puts vd Ma in middle without any report, on both master and #552. Whichever way this is decided, that silence deserves a look.

Activity

  1. added this to the 2.4 milestone on Sep 29, 2026
  2. derek73 commented on Oct 1, 2026

    @derek73
    OwnerAuthor

    A third place asks the same question. PR #567 (#563) adds a check of whether a word in front of paired initials after a comma makes them a credential. Only an unambiguous suffix word that is not also a title counts. vd and mc pass that check, so they count:

    text PR #567
    García Márquez, G.J. given G.J., family García Márquez, reports suffix-or-name
    García Márquez, Ms G.J. title Ms, given G.J. (a title doesn't count)
    García Márquez, PhD G.J. given García, family Márquez, suffix PhD G.J. (the degree counts)
    García Márquez, vd G.J. given García, family Márquez, suffix vd G.J., no report
    García Márquez, mc G.J. / Mc G.J. the same as vd G.J.

    The surname is split with no report, which is #563's own defect. It sits with this issue's option (a): C1's run test counts vd/mc as suffix words, and #567's check reads the same words the run test collected. Under (b), the shared "has a live name reading" predicate would answer this third check too, so vd G.J. would keep the family comma. That part matches Ms G.J., but where vd then lands is P6's call, not measured here. The fix is in _segment.py's paired-initials check, next to the #544 run test. Found by the /simplify review on #567 and left out of that PR because it changes behavior.

  3. self-assigned this
    on Oct 1, 2026
  4. added 3 commits that reference this issue on Oct 2, 2026
  5. added 3 commits that reference this issue on Oct 8, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

Projects

No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions