Docs / Studies / Studies.Stump2016
Inflectional Paradigms: Content and Form at the Syntax-Morphology Interface
Stump, G. T. (2016). Cambridge University Press
Stump builds the canonical inflectional paradigm on Corbett's canonical inflection, in which Turkish ADAM and TAVAN inflect canonically once vowel harmony, which spells one genitive suffix two ways, is set aside. Deviations from the canonical isomorphism of content and form are property mappings of a paradigm linkage: the Latin deponent cōnārī realizes its active cells with the passive endings that parāre keeps for its passive, and in Kashmiri the indefinite past of WUP and the recent past of WUPH inflect alike through morphomes that two conjugations map differently. Suppletion is an alternation between independent stems, and the heteroclite PRAMEN of Stump's earlier analysis has such stems.
Implementation notes
The Latin cells are abstracted to voice and agreement, with the inflection-class index and tense, aspect and mood held constant, and no realizations are attached to the Latin linkages. The Kashmiri forms and their stem-plus-suffix segmentation are the book's display of Grierson's paradigms, with the first person singular masculine exponents only. The Kashmiri property mappings act property by property, which agrees with the book's mappings on every property set with a single tense.
References
- Stump 2016, Inflectional Paradigms: Content and Form at the Syntax-Morphology Interface
- Stump 2006, Heteroclisis and Paradigm Linkage formalized
- Corbett 2007, Canonical typology, suppletion, and possible words
Declarations
A cell is a morphosyntactic property set abstracted to the voice axis and the agreement features relevant to Latin deponency (the inflection-class index and tense/aspect/mood, held constant across the paradigm below, are elided).
Derives DecidableEq, Repr.
Built from ,
Used by conariLinkage, parareLinkage, conariContentCells, conari_pm_flips_voice, conari_deviates_on_every_active_cell, conari_active_realized_by_passive_form, conari_all_cells_cross_voice, depon_vs_regular
LatinVerb has the two Latin lexemes contrasted, the deponent cōnārī and the regular
parāre.
Derives DecidableEq, Repr.
Used by conariLinkage, parareLinkage, conari_pm_flips_voice, conari_deviates_on_every_active_cell, conari_active_realized_by_passive_form
LatinStem has one stem for each of the two verbs.
Derives DecidableEq, Repr.
Used by conariLinkage, parareLinkage, conari_active_realized_by_passive_form, depon_vs_regular
The deponent linkage of cōnārī has a single stem and the voice-flipping
property mapping pm2c (Stump 2016 §12.1), which sends an active content cell
to a passive form cell.
The definition is too large to print here; read it in the source.
Built from , ,
Used by conari_pm_flips_voice, conari_deviates_on_every_active_cell, conariLinkage_not_canonical, conari_active_realized_by_passive_form, conari_all_cells_cross_voice, depon_vs_regular
The regular linkage of parāre has a single stem and the identity property mapping, canonical on the voice axis (Stump 2016 §7.1).
- parareLinkage = { realize := fun (x : LatinVerb) (x_1 : Cell) => {LatinStem.para}, pm := fun (x : LatinVerb) (σ : Cell) => σ }
Built from , ,
conariContentCells lists cōnārī's six active content cells (Stump 2016 Table 12.2).
The definition is too large to print here; read it in the source.
Built from
Used by conari_all_cells_cross_voice
Built from
Built from , ,
Built from , ,
Built from
Every content cell of cōnārī, active ones included, has a passive form correspondent, so active content is realized by passive morphology (Stump 2016 §12.1).
(l : LatinVerb) (σ : Cell) :
Finset.image (fun (x : LatinStem × Cell) => x.2.voice) (conariLinkage.corr l σ) = {Voice.passive}
Built from , , ,
Built from , ,
(σ : Cell) (h : σ.voice = Voice.active) :
Finset.image (fun (x : LatinStem × Cell) => x.2.voice) (parareLinkage.corr LatinVerb.parare σ) = {Voice.active} ∧ Finset.image (fun (x : LatinStem × Cell) => x.2.voice) (conariLinkage.corr LatinVerb.conari σ) = {Voice.passive}
Built from , , ,
Kashmiri morphomic tense (Ch. 8, pp. 217ff)
The content properties of KContent are the recent, indefinite and remote preterites and
1sg masculine agreement.
- recentPast : KContent
- indefPast : KContent
- remotePast : KContent
- p1 : KContent
- sg : KContent
- masc : KContent
Derives DecidableEq, Fintype.
Used by morphomeII, morphomeIII, pmII, pmIII, linkII, linkIII
Conjugation II sends the recent past to 'past a', the indefinite past to 'past b' and the remote past to 'past c', and each agreement property to itself.
- morphomeII KContent.recentPast = KForm.pastA
- morphomeII KContent.indefPast = KForm.pastB
- morphomeII KContent.remotePast = KForm.pastC
- morphomeII KContent.p1 = KForm.p1
- morphomeII KContent.sg = KForm.sg
- morphomeII KContent.masc = KForm.masc
Built from ,
Used by pmII, morphomeII_injective
Conjugation III sends the recent past to 'past b', the indefinite past to 'past c' and the remote past to 'past d', and each agreement property to itself.
- morphomeIII KContent.recentPast = KForm.pastB
- morphomeIII KContent.indefPast = KForm.pastC
- morphomeIII KContent.remotePast = KForm.pastD
- morphomeIII KContent.p1 = KForm.p1
- morphomeIII KContent.sg = KForm.sg
- morphomeIII KContent.masc = KForm.masc
Built from ,
Used by pmIII
The property mapping for Conjugation II (Stump 2016 Ch. 8) replaces each content property
by its image under morphomeII.
- pmII σ = Finset.image morphomeII σ
Built from , ,
Used by linkII
The property mapping for Conjugation III (Stump 2016 Ch. 8) replaces each content property
by its image under morphomeIII. The one-morphome shift from pmII is what makes the two
conjugations' preterites interleave.
- pmIII σ = Finset.image morphomeIII σ
Built from , ,
Used by linkIII
The form paradigm has a 1sg masculine exponent for each morphome, read off the
stem+suffix segmentation (wupus, wupyōs, wupyās, wuphiyās).
The definition is too large to print here; read it in the source.
Built from ,
Used by realizeForm
A form cell ⟨Z, τ⟩ realizes as the value of the PFM1 paradigm function on the
stem Z at the morphomic property set τ.
- realizeForm z τ = Morphology.PFM.paradigmFunction (fun (x : KVerb × Finset KForm) => z) [formBlock] (KVerb.wup, τ)
Built from , ,
The Conjugation II linkage has the single stem and the property mapping pmII.
- linkII = { realize := fun (l : KVerb) (x : Finset KContent) => {stemOf l}, pm := fun (x : KVerb) => pmII }
Built from , , , ,
Used by kashmiri_inflect_alike, kashmiri_inflect_alike_pastC, linkII_not_syncretic
WUP's indefinite past and WUPH's recent past share their form correspondent property set, so they inflect alike (Stump 2016 Ch. 8).
Built from ,
Built from ,
Built from
Built from
Canonical inflection (Ch. 2)
The numbers of the Turkish nominal paradigm.
- singular : TurkishNumber
- plural : TurkishNumber
Derives DecidableEq, Fintype.
Used by suffixes, turkishParts, turkish, turkish_realize_disjoint
The Turkish nouns ADAM 'man' and TAVAN 'ceiling' of Tables 2.1 and 2.2.
- adam : TurkishNoun
- tavan : TurkishNoun
Derives DecidableEq, Fintype.
Used by TurkishNoun.stem, turkishParts, turkish, turkish_realize_disjoint
The stem of a Turkish noun.
- TurkishNoun.adam.stem = [Turkish.Phonology.a, Turkish.Phonology.d, Turkish.Phonology.a, Turkish.Phonology.m]
- TurkishNoun.tavan.stem = [Turkish.Phonology.t, Turkish.Phonology.a, Turkish.Phonology.v, Turkish.Phonology.a, Turkish.Phonology.n]
Built from
Used by turkishParts
The suffixes of a cell, the plural before the case suffix, none in the nominative singular.
The definition is too large to print here; read it in the source.
Built from
Used by turkishParts
Each cell pairs a noun's stem with the cell's suffixes.
- turkishParts = { realize := fun (l : TurkishNoun) (c : Turkish.Case × TurkishNumber) => {(l.stem, suffixes c)} }
Built from , , ,
Used by turkish
The inflectional paradigm of the two nouns, each underlying form the stem with the cell's suffixes attached.
- turkish = { parts := turkishParts, compose := Turkish.underlying }
Built from , ,
ADAM and TAVAN inflect canonically, each with one stem across its cells, with distinct stems, distinct suffixes in distinct cells and the same suffixes in corresponding cells (Tables 2.1 to 2.3).
Built from
Built from , ,
The genitive suffix -(n)In is the same after ADAM and after EV 'house', but vowel harmony spells it -ın and -in, so the paradigms share their inflectional marking only below vowel harmony (fn. 1).
Suppletion (Ch. 11)
The two stems of Stump 2006's PRAMEN belong to distinct declensions, so no pattern confined to one declension determines them: they are independent stems, and their alternation is suppletion (§11.1).