Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/35461
Title: The Chinese aspectual system
Authors: Xu, Hongzhi
Advisors: Huang, Chu-ren (CBS)
Keywords: Chinese language -- Semantics.
Chinese language -- Aspect.
Issue Date: 2015
Publisher: The Hong Kong Polytechnic University
Abstract: Chinese aspect is studied from two different perspectives. The first focuses on the functionalities of the aspectual markers 着 zhe0 'ZHE', 了 le0 'LE' and 过 guo4 'GUO'. The other focuses on the aspectual classification of verbs (situation aspect). However, very few studies concern about the relationship between aspectual markers and situation aspect. Situation classification in Chinese, itself, is also problematic, e.g. whether the classification should be performed in word, phrase or sentence level. This question is actually related to the philosophical question how human perceive events and treat them differently, which is addressed in ontological descripitions of different event types. In this sense, previous studies have been trying to classify linguistic units into ontological categories, encountering the problems being discussed for decades. The intrinsic reason is that each linguistic unit in various granularities including word, phrase and sentences can be used to describe different ontological event types. This thesis describes a study on aspectual classification in Chinese. Different from previous studies that have tried to classify linguistic units into different situation types directly, this study will first describe ontological event types that are potentially shared by all human beings. After that, the study will focus on how these events are described in language, which then involves the concept of viewpoint aspects, which can be defined as the viewpoint we choose in order to describe this event, e.g. the starting point, ending point etc. There are researchers who argue that viewpoint aspect is still shared universally and thus in ontological level however in linguistic domain. Such ontological events with viewpoint aspects can be called ontological linguistic events, or just linguistic event. Then, the study will focus on how such ontological events with certain viewpoint aspects are described in Chinese, which will be language dependent. Finally, linguistic units in Chinese can be classified according to how they are usually used to realize linguistic events. The Chinese aspectual system thus includes two different levels: linguistic units that denote ontological situation types, and lingusitic units that denote viewpoint aspect. Situation types are mainly expressed by verbs and their arguments. This motivates most of the research works on classifying verbs, phrases or sentences into situation types. The aspectual markers 着 zhe0 'ZHE', 了 le0 'LE' and 过 guo4 'GUO' in addition to some aspectual verbs/adverbs, such as 在 zai4 'progressive', 开始 kai1shi3 'start', 结束 jie2shu4 'end', 继续 ji4xu4 'continue', 停止 ting2zhi3 'stop', 完成 wan2cheng2 'finish' etc., are linguistic devices that are used to express certain viewpoints, so as to form linguistic events.
On the other hands, linguistic units may not be associated with unique situation types or viewpoint aspects. This also raised the difficulties encountered in previous studies that would not be resolved unless we can have an overall view of the whole Chinese aspectual system in different levels. The study described in this thesis thus starts from the ontological perspective to examine how human perceives events in the world and then go through all different levels to linguistic units in Chinese to examine how these units are used to describe linguistic events. In detail, the following issues will be discussed. 1) How many situation types are there? Vendler (1957) presented four situation types, namely state, activity, accomplishment and achievement, which were suggested to be ontological categories later. Simith (1991) adopted another category, namely semelfactive, in his framework. By analyzing the primitives of events, I give a theorectical analysis how many situation types are there and propose eight basic categories. 2) How many viewpoints are there? Theorectically, there are unlimited number of viewpoints from which we can observe and describe an event. Linguistically, we only choose meaningful viewpoints in order to express the right and necessary information with pragmatic factors. Previously, different viewpoint aspects have been discussed, including inchoative, progressive, terminative and completive etc. 3) What is the relation between viewpoints and situation types? As have been shown that progressive is not compatible with instantaneous events. I will discuss this issue in a systematic way in the ontological level with the consideration that such compatibilities should be shared all over the world. 4) What are the consequences by proposing the different linguistic event types? I first give formal representation, mostly in first order logic. Finally, it seems clear that a linguistic event is associated with a reference time or duration based on which a background situation is described. The study of aspect turns out to be the study of the relation between the reference time or duration and different situation types. 5) How are the aspectual markers and some constructions such as RVCs related with different linguistic event types and indirectly with different situation types? I will discuss different aspectual markers, 着 zhe0 'ZHE', 了 le0 'LE' and 过 guo4 'GUO' in addition to some verbs and adverbs, including 在 zai4 'progressive', 开始 kai1shi3 'start', 结束 jie2shu4 'end', 继续 ji4xu4 'continue', 停止 ting2zhi3 'stop', 完成 wan2cheng2 'finish'. These words across different word classes are discussed together with the consideration that they all function in the domain of aspects. 6) Does the aspect framework covers all possible cases in real data? I will present an annotated corpus containing more than about 5000 sentences. The annotation framework will incorporate different modalities as their presence can affect the acceptability of certain linguistic events. 7) Is it possible to identify the aspectual information automatically by computer? I will conduct experiments with machine learning approaches on the annotated corpus, using general syntactic features, e.g. tokens, dependency relations etc. The results show a promising result, proving that the aspectual system I proposed is effective and potentially useful for computational applications.
Description: PolyU Library Call No.: [THS] LG51 .H577P CBS 2015 Xu
269 pages :illustrations (some color)
URI: http://hdl.handle.net/10397/35461
Rights: All rights reserved.
Appears in Collections:Thesis

Files in This Item:
File Description SizeFormat 
b28237912_link.htmFor PolyU Users203 BHTMLView/Open
b28237912_ir.pdfFor All Users (Non-printable)2.73 MBAdobe PDFView/Open
Show full item record

Page view(s)

33
Last Week
2
Last month
Checked on Jan 15, 2017

Download(s)

20
Checked on Jan 15, 2017

Google ScholarTM

Check



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.