Agenda

A Call for Scene-Constructing Semantic Representations: 
Insights from a Large-Scale Study of Japanese Verb Argument Structures

How can we describe the meaning of a sentence? According to the principles of generative grammar, speakers possess a computational system capable of combining the meanings of lexical items to produce the semantic representation of the sentence. The description of a sentence’s meaning refers to this semantic representation.
The dominant approach in this field assumes that understanding the meaning of a sentence is equivalent to knowing its truth conditions (see Dowty, Wall & Peters 1981, *Introduction to Montague Semantics*; Heim & Kratzer 1998, *Semantics in Generative Grammar*). The two speakers propose that the meaning of a sentence entails more than just its truth conditions and observe that when we hear a sentence, we are able to visualise a particular scene. They therefore propose that this is precisely what the semantic representation of a sentence allows us to do. The scene consists of an event and its participants; the semantic representation of a sentence consists of a set of objects corresponding to the event and the participants themselves. Each stage of the derivation will be explained in detail, including the lexical specification of each linguistic expression involved and the application of the Merge operation. Like researchers who follow the dominant approach, the two speakers recognise the importance of describing the truth conditions of a sentence and will also show that the proposed semantic representation can be used to compute the truth conditions of the sentence itself.

J.-R. Hayashishita earned his PhD in Linguistics from the University of Southern California in 2004. He is currently a Senior Lecturer in Japanese at the University of Otago, New Zealand. He is also the founder and director of DonDoushiDon Ltd, a linguistic research institute, where he leads the Don動詞どん project, a large-scale research project on the Japanese lexicon. m.  Dr. Hayashishita’s theoretical work on quantifier scope and comparatives has been well received among Japanese linguists. In collaboration with Ayumi Ueyama, he is currently developing a comprehensive theory of linguistics that incorporates the speaker’s lexicon, syntax, semantics, and phonology.

Ayumi Ueyama earned her PhD in Linguistics from the University of Southern California in 1998. She is currently a Professor of Linguistics at Kyushu University, Japan. She is also a founding member and a major contributor to the Don動詞どん project.  Prof. Ueyama is well known for her research on anaphora. She has pioneered a new approach to investigating sentence meaning and developed a theoretical framework called Syntax Semantics, in which the semantic representation generated by the computational system is modelled as a set of layered objects.