Beyond Telicity and Affected-Theme : Semantic Factors Contributing to the Resultative Interpretation of Predicates in Japanese
286Authors
nan
Abstract
This paper discusses semantic factors contributing to the resultative interpretation of predicates in the Japanese te-ir construction. The construction ambiguously takes on either progressive or resultative meanings. This ambiguity is due to the lexical meaning of the verb, and it is the purpose of this paper to single out and characterize the classes of verbs which take on the resultative meaning. l A number of recent studies have focused on the issues of telicity, transitivity, and particularly on unaccusativity and reflexivity, and it has been argued that the resultative interpretation is closely related to the subject's involvement in the resulting state. While accepting this argument, I will show that notions of unaccusativity and reflexivity alone cannot cover all the data. Similarly, I will argue that Kim's (1993) account of the analogous construction in Korean, which refers to the concept "possession," fails to accommodate a certain set of data without making unlikely stipulation. I will demonstrate that there are in fact two separate sets of verbs allowing the resultative meaning. The first set is definable in terms of the subject's involvement, while the second set makes no reference to the subject's involvement in the resulting state. I will characterize the second set as verbs of spatial configuration (Levin 1993) and propose the notion "affected locative" to optimally characterize the semantic feature licensing the resultative meaning in the second set. 1.
Keywords
Full Content
Language, Information and Computation(PACLIC 11), 1996, 147-156 Beyond Telicity and Affected-Theme: Semantic Factors Contributing to the Resultative Interpretation of Predicates in Japanese Chiharu Uda Doshisha University cuda@mail.doshisha.ac.jp Abstract This paper discusses semantic factors contributing to the resultative interpretation of predicates in the Japanese te-ir construction. The construction ambiguously takes on either progressive or resultative meanings. This ambiguity is due to the lexical meaning of the verb, and it is the purpose of this paper to single out and characterize the classes of verbs which take on the resultative meaning. l A number of recent studies have focused on the issues of telicity, transitivity, and particularly on unaccusativity and reflexivity, and it has been argued that the resultative interpretation is closely related to the subject's involvement in the resulting state. While accepting this argument, I will show that notions of unaccusativity and reflexivity alone cannot cover all the data. Similarly, I will argue that Kim's (1993) account of the analogous construction in Korean, which refers to the concept "possession," fails to accommodate a certain set of data without making unlikely stipulation. I will demonstrate that there are in fact two separate sets of verbs allowing the resultative meaning. The first set is definable in terms of the subject's involvement, while the second set makes no reference to the subject's involvement in the resulting state. I will characterize the second set as verbs of spatial configuration (Levin 1993) and propose the notion "affected locative" to optimally characterize the semantic feature licensing the resultative meaning in the second set. 1. Introduction Te-ir resultative is a type of te-ir construction in Japanese, which is composed of a lexical verb followed by an auxiliary verb it '(literally) for an animate entity to exist.' Te-ir sentences can take on such meanings as "progres- sive," "resultative," "habitual," and "experience," among others. Of particular recent interest is the distinction between the progressive reading and the resultative reading, as illustrated by the following examples: (1) a. Kyoko -ga ima ki -o� taosite-iru. NOM now tree ACC fell-IR 'Kyoko is felling the tree.' b . Ki� -ga� ima taorete-iru. tree NOM now lie-IR 'A tree has been felled (is lying down).' c. Ki� -ga� ima taos-arete-iru. tree NOM now fell-PAS-IR 'A tree has been felled/is being felled.' 147 d. Kyoko -wa ima kami -o� somete-iru. TOP now hair ACC dye-1R 'Kyoko has dyed her hair/is dying her hair.' Thus the sentence (la) takes on a progressive meaning, while (lb) takes on a resultative meaning. 2 (1c) and (1d) are ambiguous. This paper focuses on the long-standing issue of predicting when the te-ir sentence can convey the resultative reading. One complication with the resultative interpretation is the relevance of the verb meaning. Some verbs allow resultative interpretation only in the te-ir form, while others allow resultative interpretation in its "base" form (i.e, without te-ir): (2) a. Kyoko -wa ima Taroo -no koto -o� oboete-iru. TOP now� GEN thing ACC remember-1R 'Kyoko remembers (has remembered) Taroo.' b. * Kyoko -wa go-nenkan Taroo …
Content truncated to 500 words
Labeled Text
This paper discusses semantic factors contributing to the resultative interpretation of predicates in the Japanese te-ir construction. The construction ambiguously takes on either progressive or resultative meanings. This ambiguity is due to the lexical meaning of the verb, and it is the purpose of this paper to single out and characterize the classes of verbs which take on the resultative meaning. l A number of recent studies have focused on the issues of telicity, transitivity, and particularly on unaccusativity and reflexivity, and it has been argued that the resultative interpretation is closely related to the subject's involvement in the resulting state. While accepting this argument, I will show that notions of unaccusativity and reflexivity alone cannot cover all the data. Similarly, I will argue that Kim's (1993) account of the analogous construction in Korean, which refers to the concept "possession," fails to accommodate a certain set of data without making unlikely stipulation. I will demonstrate that there are in fact two separate sets of verbs allowing the resultative meaning. The first set is definable in terms of the subject's involvement, while the second set makes no reference to the subject's involvement in the resulting state. I will characterize the second set as verbs of spatial configuration (Levin 1993) and propose the notion "affected locative" to optimally characterize the semantic feature licensing the resultative meaning in the second set. 1.
Publication Details
- Paper ID:
- 286
- Year:
- 1996
- Publication:
- TIPSTER
- Publication Type:
- conf
- DOI:
- nan