Documentation

Linglib.Features.Gender

Surface Gender #

@cite{corbett-1991} @cite{kramer-2020}

Descriptive (atheoretical) classification of surface gender values attested cross-linguistically. This type records the observable gender of a noun — what agreement class it triggers — without committing to the mechanism of gender assignment.

The distinction between sex-based systems (masculine/feminine/neuter) and animacy-based systems (animate/inanimate) follows @cite{corbett-1991}'s WALS Chapter 31. @cite{kramer-2020} argues that these surface categories arise from a single underlying mechanism (phi-features on the nominalizing head n), parameterized by feature dimension; that analysis lives in Morphology.DM.Categorizer.

For languages with many noun classes (Bantu, Mixtec), the fragment retains its own fine-grained Gender type and provides a bridge to SurfaceGender via a .primary function.

Surface gender categories attested cross-linguistically.

These are the descriptive labels a grammar assigns to nouns based on agreement patterns. For the structural analysis, see Morphology.DM.GenderFeature.

  • masculine : SurfaceGender

    Masculine: male humans/higher animates; default in many sex-based systems.

  • feminine : SurfaceGender

    Feminine: female humans/higher animates; marked in many sex-based systems.

  • neuter : SurfaceGender

    Neuter: neither masculine nor feminine; inanimate default in 3-gender systems.

  • common : SurfaceGender

    Common: merged masculine + feminine (Swedish, Danish).

  • animate : SurfaceGender

    Animate: animate referents in animacy-based systems (Teop, Algonquian).

  • inanimate : SurfaceGender

    Inanimate: inanimate referents in animacy-based systems.

Instances For
    @[implicit_reducible]
    Equations
    Equations
    • One or more equations did not get rendered due to their size.
    Instances For

      Round-trip: UD → Surface → UD is the identity.

      Gender Features (@cite{sauerland-2003}) #

      Binary feature decomposition of sex-based gender:

      These features form a containment hierarchy: [+neuter] → [+feminine]. This is a feature-geometric claim from @cite{sauerland-2003} §6, not a natural-kind claim: neuter is the most specified gender (like singular for number, 1st for person), and masculine is the least specified (like plural, 3rd).

      The three well-formed combinations yield the three gender values:

      This parallels person [±author] ⊂ [±participant] and number [±atomic] ⊂ [±minimal]. All three are instances of PhiFeatures.

      For the morphosyntactic (DM) analysis, see Morphology.DM.Categorizer.GenderFeature.

      Binary gender features: [±feminine, ±neuter].

      These two features suffice for the three-way gender distinction:

      • neuter: [+feminine, +neuter]
      • feminine: [+feminine, −neuter]
      • masculine: [−feminine, −neuter]

      The fourth combination [−feminine, +neuter] is ill-formed: a neuter entity necessarily triggers feminine agreement ([+neuter] → [+feminine] in the feature geometry).

      • isFeminine : Bool

        [+feminine]: referent triggers feminine (or neuter) agreement.

      • isNeuter : Bool

        [+neuter]: referent triggers neuter agreement.

      Instances For
        def Features.Gender.instDecidableEqFeatures.decEq (x✝ x✝¹ : Features) :
        Decidable (x✝ = x✝¹)
        Equations
        • One or more equations did not get rendered due to their size.
        Instances For
          Equations
          • One or more equations did not get rendered due to their size.
          Instances For

            Well-formedness: [+neuter] → [+feminine]. Neuter entails feminine in the feature geometry.

            Equations
            Instances For

              Neuter features: [+feminine, +neuter].

              Equations
              Instances For

                Feminine features: [+feminine, −neuter].

                Equations
                Instances For

                  Masculine features: [−feminine, −neuter].

                  Equations
                  Instances For
                    @[implicit_reducible]

                    Gender features are a PhiFeatures instance: outer = isFeminine, inner = isNeuter.

                    The containment [+neuter] → [+feminine] maps to PrivativePair's [+inner] → [+outer], unifying the structure with person and number. All shared properties are inherited by construction.

                    Equations
                    • One or more equations did not get rendered due to their size.
                    @[simp]

                    The three canonical gender values map to the three PrivativePair cells.

                    theorem Features.Gender.no_fourth_gender (a b c d : Features) :
                    a.wellFormed = trueb.wellFormed = truec.wellFormed = trued.wellFormed = truea ba ca db cb dc dFalse

                    No 4-way gender distinction (inherited from PhiFeatures).

                    theorem Features.Gender.illFormed_only :
                    { isFeminine := false, isNeuter := true }.wellFormed = false

                    The ill-formed combination [−feminine, +neuter] is the only combination that violates well-formedness.

                    theorem Features.Gender.exactly_three_wellFormed :
                    (List.filter Features.wellFormed [{ isFeminine := true, isNeuter := true }, { isFeminine := true, isNeuter := false }, { isFeminine := false, isNeuter := true }, { isFeminine := false, isNeuter := false }]).length = 3

                    There are exactly 3 well-formed feature combinations (= 3 genders).

                    theorem Features.Gender.neuter_implies_feminine (f : Features) (hw : f.wellFormed = true) :
                    f.isNeuter = truef.isFeminine = true

                    Containment: [+neuter] → [+feminine] for all well-formed features.

                    Map gender features to the descriptive SurfaceGender type.

                    Equations
                    Instances For

                      Round-trip: fromSurfaceGender ∘ toSurfaceGender = some for all well-formed features.

                      Gender knowledge state for a discourse referent.

                      Distinct from SurfaceGender, which describes the morphosyntactic agreement class a noun triggers. GenderInfo describes what the discourse participants know or assume about a referent's gender.

                      Motivated by @cite{arnold-2026}'s observation that singular they is licensed by two inversely correlated pragmatic conditions: one requiring an underspecified discourse representation (where gender is unknown or irrelevant), the other requiring knowledge that the referent's personal pronouns are they/them (where gender information is highly specific).

                      See also @cite{newman-1992} ("nonsolid" antecedents), @cite{newman-1998} (low individuation), and @cite{camilliere-etal-2021} (social distance as a proxy for discourse specificity).

                      • known : SurfaceGenderGenderInfo

                        Gender is known to discourse participants and matches a morphosyntactic agreement class. Example: "my sister" → .known .feminine

                      • unspecified : GenderInfo

                        Gender is unknown, irrelevant, or not elaborated in the discourse. Example: "every student", "someone", "the clerk" (in passing).

                      Instances For
                        def Features.instReprGenderInfo.repr :
                        GenderInfoStd.Format
                        Equations
                        • One or more equations did not get rendered due to their size.
                        Instances For

                          Lift a surface gender to discourse-level knowledge.

                          Equations
                          Instances For

                            Round-trip: known surface gender survives the coarsening.