Show simple item record

dc.contributor.authorKatz, Richard Moses
dc.date.accessioned2016-10-20T04:30:18Z
dc.date.available2016-10-20T04:30:18Z
dc.date.issued2016-05
dc.identifier.otherkatz_richard_m_201605_phd
dc.identifier.urihttp://purl.galileo.usg.edu/uga_etd/katz_richard_m_201605_phd
dc.identifier.urihttp://hdl.handle.net/10724/36205
dc.description.abstractThe Gothic language can express passive voice in different ways. Unique among Germanic languages, Gothic retains a synthetic passive, an inheritance of the Indo-European medio-passive. It also as has two periphrastic constructions, one that combines the verb ‘be’ and a past participle and ‘become’ and a past participle. Additionally it contains a productive verb with a nasal suffix, -na-, which can also be used to express actions wherein – like a passive – there is no overt agent. At the same time, the nasal-suffix verb (or -nan verb), seems to behave in ways that sets it apart from passive expression. What comes down to scholars is a host of verbs then that do not have a clear relationship. Various assessments see the four categories listed as being in some degree of syncretism and/or representing a developmental state wherein some forms are coming to replace others. My study examines the manifold semantics of these constructions. It shows how these forms, though they do constitute some degree of semantic overlap, also provide the option for a full non-agentive paradigm, one in which each type of non-agentive expression can be used to denote different semantic values, including a functional perfect tense in the passive. By analyzing an exhaustive list of these constructions in the Gothic corpus, I isolate separate metrics of tense, aspect and voice. I then map an underlying structure for each based on the model of a resultative. A resultative in the style of John irons the shirt flat, contains not only nominal arguments, such as a subject and object, but also a statal argument, flat, that is distinct from the kind of adjectival argument one finds in a predicate adjective. Using this model, I pose a de-statal origin for non-agentive expressions in Gothic, showing how the various forms provide optionality to express three separate semantics with graded complexity: (i.) a fientive verb in –nan that cannot denote an agent, (ii.) a regular passive that can denote an agent, and (iii.) a perfect passive that can not only denote an agent, but also has the capacity to express an entailed, result-state.
dc.languageeng
dc.publisheruga
dc.rightspublic
dc.subjectHistorical Linguists
dc.subjectGothic Language
dc.subjectGermanic Languages
dc.subjectDerivational Semantics
dc.subjectResultativity
dc.subjectArgument Structure
dc.subjectGenerative Syntax
dc.subjectDistributed Morphology
dc.titleThe resultative in Gothic
dc.typeDissertation
dc.description.degreePhD
dc.description.departmentLinguistics Program
dc.description.majorLinguistics
dc.description.advisorJared Klein
dc.description.committeeJared Klein
dc.description.committeeVera Lee-Schoenfeld
dc.description.committeeJonathan Evans


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record