[ Article ]
Korea Journal of English Language and Linguistics - Vol. 20, No. 1, pp. 475-495
Abbreviation: KASELL
ISSN: 1598-1398 (Print)
Received 02 Jun 2020 Revised 23 Aug 2020 Accepted 23 Sep 2020
DOI: https://doi.org/10.15738/kjell.20..202009.475
Mismatches in Ellipsis: Category Mismatch Asymmetry in English VP-ellipsis
Sun-Woong Kim
Kwangwoon University
Copyright 2020 KASELL This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Funding Information ▼ Ministry of Education National Research Foundation of Korea NRF-2019S1A5A2A01036282

This paper is purposed to explain the asymmetry of category mismatch in English VP-ellipsis and to argue that the simple set-theoretic comparison of inclusion (or containment) between the antecedent phrase and the elided phrase is neither sufficient nor necessary. The reason is explored for why the antecedent noun triggers VP-ellipsis in the subsequent clause, while the antecedent verb rarely triggers NP-ellipsis in the subsequent clause (Sato 2018). This paper looks for a syntactic answer to the question in the dynamicity of phase theory (Bošković 2014, den Dikken 2006, 2007, 2017). For a proper morphological derivation, the notation ([CAPITAL]) is adopted to denote the lexeme/root status of a lexical category. On top of it, to take care of the idea that only nouns of concealed polar interrogation can trigger VP-ellipsis (Miller and Hemforth 2014), PolP is located above VP in the second conjunct to carry polar properties (Culicover 1992, Laka 1990). If polarity is not involved, PolP is not needed because Pol is for a (concealed) yes/no question. Under the dynamic definition of phases, a phase and the complement of its head, but not the complement of a complement, are eligible for ellipsis (Bošković 2014). VP in the second clause can be elided if the clause has PolP because the VP counts as the complement of a phase PolP. In contrast, VP in the second clause cannot be elided without PolP, since it is not a complement but the complement of a complement of a phase. This explains why VP-ellipsis is not possible when the non-polar concealed questions of nominals are an antecedent. Regarding the opposite direction, differently from Sato’s analysis, this paper assumes that one is n, which is the head of nP. Above RP, NP, nP and DP are capped in a row to host lexical items. Here DP is a phase and NP is the complement of a complement, hence it is not elidable according to Bošković (2014). It will also be shown that the proposed analysis has a desirable consequence in that the syntactic behavior of other types of mismatch in ellipsis like voice mismatch, polarity mismatch, and type mismatch can be predicted.