| Summary: | This thesis empirically tests the two predominant views concerning which object in a DOC is the ‘true’ one by applying Hudson’s (1992) eleven objecthood diagnostics to verbs that pattern like ‘bill’ in a corpus search and a grammatical acceptability survey. Only two tests showed statistically significant results: passivization and extraction. Passivization indicated that the first positional object (O1) behaves most like the monotransitive object (O0). Extraction indicated that, by a small margin, the second positional object (O2) behaves more like O0. For most of Hudson’s objecthood tests, however, there was no clear patterning of O0 with either the first or second object for ‘bill’ verbs. This calls into question whether it is possible to determine which of the objects in a DOC is the ‘true’ object by employing such diagnostics, and has significant implications for notions of objecthood, especially within Lexical Functional Grammar. The study indicates that a substantial rethink is needed with regards to traditional views about objecthood.
|
|---|