Parasitic Gaps as Pronominals

Based on the claim made by Pollard and Sag (1994) that only subject parasitic gaps are true parasitic gaps, I have found that the evidence that Levine et al. (2001) put forth is insufficient to disprove that parasitic gaps are pronominals. Their counterexamples are only limited to adjunct gaps and subject extractions which are, strictly speaking, not parasitic gaps. I have also asserted that their normal gap approach to a parasitic gap encounters an inherent problem in accordance with the Principle of Canonicality. Therefore, in favor of the pronominal analysis of p-gaps, which has mainly supported by Cinque (1990) and Postal(1994), I have revised the Subject Condition and the Synsem Hierarchy. These revisions allow us to account for parasitic gaps in tough constructions and the resumptive property of parasitic gaps.Supported by the Grant for the Reform of University Education under the BK21 Project of SNU