[Pan03]: The aspectual makeup of Perfect participles and the interpretations of the Perfect

[Pan03] [Kip02] [IAI01]

Pancheva (in Perfect Explorations, Alexiadou-Rathert-von Stechow eds., 2003) argues that the three perfect interpretations (Universal, Experiential, Resultative) arise from the aspectual makeup of the perfect participle — specifically the interaction of Aktionsart with grammatical aspect (perfective vs imperfective) inside the participial VP.

Verified content (vs PDF; § refs from the paper)

• Three perfect types (§1, ex 1): Universal (U), Experiential (EXP), Resultative (RES). EXP and RES are commonly grouped under the EXISTENTIAL umbrella (McCawley 1971, Mittwoch 1988).
• Aspectual makeup determines reading availability (§2 thesis): Universal and Resultative interpretations depend on participial aspect, Experiential does not. States/progressives → U or EXP; non-progressive activities → EXP only; non-progressive telic → RES or EXP.
• Resultative requires telic predicates (§2, ex 5–6, citing Kratzer 1994): only telic events have a natural (lexically inherent) result state. I have run (5a, atelic) lacks RES; I have lost my glasses (6a, telic) has both EXP and RES.
• Cross-linguistic aspectual restrictions (§2): Greek perfect participles are obligatorily perfective → no Universal perfect; Bulgarian allows non-perfective participles → Universal possible.

Relation to companion files

• Studies/IatridouEtAl2001.lean — Pancheva builds on IAI 2001's PTS framework and U/E distinction. Her contribution is the participle-aspect mechanism.
• Studies/Kiparsky2002.lean — Kiparsky's event-structure account is an independent proposal for the same polysemy. The toKiparsky bridge below embeds Pancheva's 3-type taxonomy into Kiparsky's 4-reading enum (Kiparsky adds present-state, which Pancheva does not distinguish).

The Lean substrate Semantics/Aspect/Core.lean already provides PerfectType (Pancheva's 3-type enum) plus the universalPerfect / experientialPerfect / resultativePerfect operators built on PERF_P over UNBOUNDED / INIT_OVERLAP / BOUNDED aspect; these are the general substrate Pancheva's analysis uses.

def Pancheva2003.toKiparsky : Semantics.Aspect.PerfectType → Kiparsky2002.PerfectReading

Map [Pan03]'s perfect types to Kiparsky's readings.

• experiential → existential (Pancheva's EXP and Kiparsky's existential both denote ∃-event-in-PTS without a result-state requirement)
• universal → universal
• resultative → resultative

Equations

• Pancheva2003.toKiparsky Semantics.Aspect.PerfectType.experiential = Kiparsky2002.PerfectReading.existential
• Pancheva2003.toKiparsky Semantics.Aspect.PerfectType.universal = Kiparsky2002.PerfectReading.universal
• Pancheva2003.toKiparsky Semantics.Aspect.PerfectType.resultative = Kiparsky2002.PerfectReading.resultative

theorem Pancheva2003.pancheva_injective : Function.Injective toKiparsky

Pancheva's classification embeds into Kiparsky's: every Pancheva type maps to a distinct Kiparsky reading.

theorem Pancheva2003.pancheva_subset_kiparsky (pt : Semantics.Aspect.PerfectType) : toKiparsky pt ∈ [Kiparsky2002.PerfectReading.existential, Kiparsky2002.PerfectReading.universal, Kiparsky2002.PerfectReading.resultative, Kiparsky2002.PerfectReading.presentState]

Pancheva's types are a proper subset of Kiparsky's: Kiparsky adds the present-state reading which Pancheva does not distinguish.