Contents
前言i
List of Abbreviations iii
Chapter One Introduction 1
1.1 Preliminaries 1
1.2 Overview 2
1.3 A Note on Terminology and Data 7
Chapter Two Bare Nouns， Plurality and Classifiers 9
2.1 Bare Nouns: Reference to Kinds9
2.2 The Count-mass Dichotomy of Nouns15
2.2.1 The Lexical Approach to the Count-mass Dichotomy 16
2.2.2 The Syntactic Approach to Count-mass Dichotomy 20
2.3 Plurality and Classifiers 25
2.3.1 Plurality 25
2.3.2 Classifiers 27
2.3.3 Linking Plurality to Classifiers 29
2.4 Our Account 33
2.4.1 Bare Nouns as Property-denoting Root Nouns 33
2.4.2 Countability 40
2.5 Summary 52
Chapter Three A Unified Account of Classifier Phrases 53
3.1 The Constituency of Classifier Phrases 53
3.1.1 The Constituency of Classifier Phrases in Chinese 53
3.1.2 The Constituency of Pseudo-partitive Constructions in English 71
3.1.3 The Syntactic Relation between Nume and CL 78
3.1.4 Summing up 81
3.2 An Overall Semantic Account of Different Types of Classifiers 81
3.3 The Feature Decomposition Analysis of Classifier Phrases 84
3.3.1 Number 84
3.3.2 The Feature Decomposition of CL 88
3.3.3 Pseudo-plural Forms 94
3.3.4 Summing up 103
3.4 Summary 104
Chapter Four Towards Different Interpretations of Nominal Phrases 105
4.1 DP 105
4.1.1 Previous Discussions on D 105
4.1.2 Redefining D and Its Relevant Notions 111
4.1.3 The Feature Decomposition Analysis of D 111
4.2 Accounting for Different Interpretations of Nominal Phrases 114
4.2.1 Canonical DPs 114
4.2.2 Non-canonical DPs 119
4.2.3 Classifiers and Modification 127
4.2.4 Definiteness Effect Revisited 131
4.2.5 An Account of Generic Reading133
4.2.6 Measure-denoting Nominal Phrases141
4.3 Summary 146
Chapter Five Two Types of Multiple Classifier Constructions 149
5.1 OMCCs 149
5.1.1 PCs in English 149
5.1.2 PCs in Chinese 155
5.1.3 More on the Structure of PCs 158
5.1.4 Licensing Conditions of OMCCs 164
5.2 IMCCs 168
5.2.1 Recapitulation: English as a Classifier Language 168
5.2.2 IMCCs in English 170
5.2.3 IMCCs in Chinese 172
5.3 Summary 178
Chapter Six Concluding Remarks 181
References 185
后记 195
內容試閱：
Introduction
1.1 Preliminaries
Greenberg 1972 makes a typological generalization that classifiers and number morphology are complementarily distributed to some extent and that languages without number morphology tend to have classifiers. Based on this generalization， languages are divided into two types: classifier languages and non-classifier languages see also Li 1999; Cinque 2006; Li 2011 for some more discussions. In classifier languages， a classifier is obligatorily used when a noun occurs with a numeral. In non-classifier languages， classifiers are not needed for a noun to combine with a numeral， but the noun is number-marked. This contrast is exemplified by 1 and 2.
1 a. san zhi xiong
three CL bear
‘three bears’
b. san ping niunai
three bottle milk
‘three bottles of milk’
2 a. three bears
b. three bottles of milk
Note that 2b is usually regarded as a pseudo-partitive construction in the previous literature. In this study， we claim that classifiers also exist in 2a as an empty category as well as in 2b. Based on this claim， we will pursue the possibility that the classifier is a language universal category， an idea not well envisaged before. This study covers the following research questions， as listed in 3.
3 a. What is the semantics of classifiers?
b. How are classifiers represented syntactically?
c. How are classifiers related to interpretations of nominal phrases?
Durational and frequentative expressions are excluded from our discussions， as shown in the italicized part of 4.
2 The Syntax of Classifiers
4 a. Wo chi fan liang ge xiaoshi le. durational phrase
I eat meal two CL hour Asp
‘I have had meals for two hours.’
b. Wo qu Shanghai liang ci le. frequentative phrase
I go Shanghai two CL Asp
‘I have been to Shanghai twice.’
In the next section， we will give an overview of the main assumptions of this study.
1.2 Overview
We start our discussions with the semantics of nouns. By examining the reference-to-kind approach Carlson 1977a， 1977b; Chierchia 1998a， 1998b; Li 2011， among others and property-denoting approach e.g.， Krifka 2004 to bare nouns， we show that the latter approach is a much favored one. Another hot issue involved in the semantics of nouns is the count-mass distinction. There are mainly two approaches to this issue as well. One is the lexical approach which claims that nouns are marked as count or mass in the lexicon; the other is the syntactic approach in which the count-mass dichotomy is attributed to syntax. Inspired by the syntactic approach， we argue that bare nouns neutral of number are property-denoting and denote a join semi-lattice in the sense of Link 1983. See the illustration for join semi-lattices in 5 Doetjes 1997: 27.
In the diagram in 5， a， b， c， {a， b}， {a， c}， {b， c}， {a， b， c} are all members of the set， which is ordered by part of-relation shown by the upward lines. The part of-relations among those different members of the set can be sketched as follows: a is part of {a， b} and {a， c}; b is part of {a， b} and {b， c}; c is part of {a， c} and {b， c} and {a， b}， {a， c} and {b， c} are parts of {a， b， c}. We define nouns in this sense as root nouns. But different from the distributed morphology Halle & Marantz 1993; Marantz 1997， root nouns in our sense are already category marked， whereas they are unmarked for number. Bare plurals are not root nouns; instead， they have more complex internal structures.
We distinguish two types of join semi-lattices. The first type is the join semi-lattice with individuals as minimal parts; the other type continuous one refers to those without minimal parts. But different from Link 1983， the first type covers both book like nouns and furniture like nouns; the second type refers to join semi-lattices of water like nouns. We depart from Link 1983 in that we do not think different types of join semi-lattices make any difference with respect to the countability of nouns. Rather， we claim that nouns denoting join semi-lattices with minimal parts or without minimal parts can both be counted under a certain condition. We put forward such a condition as in the following 6.
6 Condition of Counting
i A noun α can be counted iff it denotes a set of aggregates X， such that X is monotonic.
Ii A set of aggregates X is monotonic iff x is the minimal part for X and for any aggregate y， such that y=x and y∈X.
A monotonic semi-lattice can be exemplified in 7.
7 Monotonic semi-lattice Natural atoms: { a， b， c } or Unnatural atoms: {x a， b， c ， y a， b， c ， z a， b， c
To put it simply， the monotonic condition requires that the unit of counting should be at the same size. This is reminiscent of the common sense in mathematics: you cannot count two things of different units. For instance， when you have an apple on one side and a basket of apples on the other side， you cannot count them as two apples or two baskets
購物車/收銀台(
0
) | 在線留言板
| 付款方式
| 運費計算
| 聯絡我們
| 幫助中心 |
加入書簽
HOME
新書上架
