21世纪英语专业系列教材

Linguistics:

An Elementary Course Book ( Second Edition)

语言学基础教程

(第二版)

苗兴伟 ◎主编

胡壮麟 )主审

Linguistics:

An Elementary Course Book ( Second Edition)

语言学基础教程(第二版)

《语言学基础教程》(第二版)包括语言与语言学、语音学与音位学、形态学、句法学、语义学、语用学、语篇分析、历史语言学、文体学、社会语言学、跨文化交际、心理语言学、认知语言学、应用语言学等十四章。在内容方面，本书在提供基本知识的同时，更注重教材的科学性、系统性、实用性和时代性。在编写过程中，我们力求“化难为简”，尽量做到概念清晰，既要保证知识的系统性，又要避免术语的堆砌。本书在传授基本知识与概念的同时，通过丰富的实例提供了有关语言分析和描述的基本方法，并强调语言学与其他学科的联系，以便适应创新型人才培养的需要。

《语言学基础教程》(第二版) 力求以通俗易懂的语言和内容将语言学知识呈献给读者，章节安排循序渐进，知识体系完整，有助于语言学学习者在掌握语言学理论的基础上进一步拓展知识领域，培养创新能力和应用能力，提高综合素养。

ISBN978-7-301-29658-5

“北大外文学堂”

系列教材

Linguistics:

An Elementary Course Book( Second Edition)

语言学基础教程

(第二版)

苗兴伟 ◎主编

胡壮麟 ◎主审

图书在版编目(CIP)数据

语言学基础教程/苗兴伟主编.一2版.一北京 ：北京大学出版社，2018.7(21 世纪英语专业系列教材)

ISBN 978-7-301-29658-5

Ⅰ.①语⋯ Ⅱ.①苗⋯ Ⅲ.①英语一语言学一高等学校一教材 Ⅳ.①H31中国版本图书馆CIP 数据核字(2018)第136327号

书 名 语言学基础教程(第二版)

YUYANXUE JICHU JIAOCHENG (DI ER BAN)

著作责任者 苗兴伟 主编

责任编辑 刘文静

标准书号 ISBN 978-7-301-29658-5

出版发行 北京大学出版社

地 址 北京市海淀区成府路205号 100871

网 址 http://www.pup.cn 新浪微博：@北京大学出版社

电子信箱 liuwenjing008@163.com

电 话 邮购部62752015 发行部62750672 编辑部62754382

印 刷 者 北京大学印刷厂

经 销 者 新华书店

720毫米×1020毫米 16开本 17印张 300千字

2010年7月第1版

2018年7月第2版 2018年7月第1次印刷

定 价 48.00元

未经许可，不得以任何方式复制或抄袭本书之部分或全部内容。

版权所有，侵权必究

举报电话: 010-62752024 电子信箱:fd@pup.pku.edu.cn

图书如有印装质量问题，请与出版部联系，电话： 010-62756370

第二版前言 1

第二版前言

《语言学基础教程》自2010年出版以来，得到了语言学学习者和教师的广泛认可。当然，教材中也有一些问题和不足。在多次印刷中，我们曾经对教材中的个别细节做了适当修改。为使学生能够更为系统地掌握语言学知识，使《语言学基础教程》更好地服务于英语类专业的人才培养，我们决定对本书进行修订。在教材的使用过程中，我们也获得了许多教师和学生的反馈，为本次的修订工作提供了有用的信息。我们在认真梳理反馈的同时，对教材中存在的问题进行了分析，并于2017年底开始了《语言学基础教程》的修订工作。

在修订过程中，除了修改了语言表达方面的细节问题，重点对第六章、第七章和第十三章的内容做了修订。第六章主要增加了对理论的解释，以帮助学生更好地把握语用学的理论。第七章增加了批评话语分析的内容，以反映语篇分析的最新发展和前沿。第十三章增加了“语法化和词汇化”一节，目的使学生了解语法化和词汇化及其背后的认知机制。为使学生巩固所学知识并进一步学习有关知识，拓展语言学学习和研究的空间，每章增加了“深入阅读”( further reading)书单。其他方面的修订就不一一介绍了，希望修订后的《语言学基础教程》能以全新的面目出现在大家的面前，更好地满足广大的语言学教师、学生和爱好者的需要。我们真诚地期待读者的鼓励和反馈，希望《语言学基础教程》在大家的帮助下不断成长。

参加修订本书各章的人员为：

第一章 严世清 苏州大学

第二章 董宏乐 复旦大学

第三章 李 力 厦门大学

第四章 刘世铸 山东大学

第五章 苗兴伟 北京师范大学

第六章 陈新仁 南京大学

第七章 常晨光 中山大学

第八章 向明友 对外经济贸易大学

第九章 陈毅平 暨南大学

第十章 苏立昌 南开大学

语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

第十一章 朱 晔 浙江大学

第十二章 刘振前 山东大学

第十三章 文 旭 西南大学

第十四章 崔 刚 清华大学

感谢北京大学胡壮麟教授对本书的关心和鼓励，特别是在第一版的编写和出版过程中，胡壮麟教授为我们出谋划策，并对初稿提出了非常详细的修改意见，保证了本书的质量。感谢北京大学出版社外语部领导张冰女士和责任编辑刘文静女士对修订工作的大力支持和帮助。

热忱欢迎专家同行和读者一如既往地对本教材不吝赐教。

编者

2018年5月29日

第一版前言 1

第一版前言

随着高等教育事业的发展和社会对高层次人才的需求的日益增长，本科生教材的正规化和系统化的必要性越来越明显。语言学知识在高等教育乃至整个社会发展中的作用已不言而喻。为使学生能够系统地掌握语言学知识，满足高校培养创新型、应用型人才的需要，我们应北京大学出版社的邀请编写了《语言学基础教程》(中、英文版)。

本教材的使用者主要为全国各高校英语专业的本科生。在内容方面，体现本科阶段课堂教学的特点，在提供基本知识的同时，更注重教材的科学性、系统性、实用性和时代性。在编写过程中，我们力求“化难为简”，尽量做到概念清晰，既要保证知识的系统性，又要避免术语的堆砌。本教材在传授基本知识与概念的同时，通过丰富的实例提供了有关语言分析和描述的基本方法，同时强调语言学与其他学科的联系，以便适应创新型人才培养的需要。为使学生巩固所学知识并进一步学习有关知识，每一章后都有练习题，并列出了“深入阅读”书单。全书共有十四章，教师可以根据教学大纲的课时安排和课堂教学的需要，合理地安排教学。

《语言学基础教程》的作者是来自全国13所211工程大学的知名中青年学者，都是教学第一线的英语教师，具有丰富的语言学研究和教学经验。参加本书编写的人员为：

2 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

第九章 陈毅平 暨南大学

第十章 苏立昌 南开大学

第十一章 朱 晔 浙江大学

第十二章 刘振前 山东大学

第十三章 文 旭 西南大学

第十四章 崔 刚 清华大学

感谢北京大学胡壮麟教授在百忙之中担任本书的主审并为之作序。胡壮麟教授主编过多部语言学教材，为我国的语言学教学事业做出了巨大贡献。本书的作者也都是在胡教授的教材的引领下步入语言学殿堂的。在此我们谨以此书向胡壮麟教授表示敬意，并衷心地感谢他为本书的编写提出的宝贵建议。

我们还要衷心地感谢同济大学马秋武教授和浙江大学许力生教授对本书的关注和厚爱。感谢北京大学出版社刘强先生和宇航出版社姜军先生对本书的关心和支持。

由于编写人员来自不同的院校，每人分工写一部分，因而各章在写作风格和难易程度方面难免有不尽一致之处。虽然我们在统稿过程中进行了反复的修改和调整，谬误或疏漏在所难免。书中不当之处，恳请专家同行和读者批评指正。

编者

2009年10月16日

第一版序

第一版序

北京大学出版社外语部的领导和编者策划出版供英语专业学生使用的各种教材系列取得重大成就，受到高校英语专业教师的热烈欢迎，被列入教育部精品教材，但在这套系列中缺了有关英语语言学课程的教材，美中不足。我虽然编过《语言学教程》等多种教材，已无法归入这套系列，因此这个问题迟迟未能解决。与此同时，山东大学外国语学院副院长苗兴伟教授跃跃欲试，自愿张罗此事，我对此早有所闻，乐观其成。

此后，苗兴伟教授充分发挥了他的组织才能，拉起了一支庞大的编写队伍，并希望我在书成后写个序，我答应了。2009年7月份在北京清华大学主办的第36届国际系统功能语言学会议上，我从兴伟手中拿到书稿后，才发现我还要担任“主审”的任务。这时我才意识到自己犯了自不量力的错误。一则本人退休多年，已入老弱病残之列，掌握新知识有限，二则本书的作者均是我国学术界中生代的领军人物，不是教授博士，就是院长主任；三则我给不同出版社友情打工多年，尚未脱手的任务众多，加之国内外这个会议那个会议未见有尽头之日，时间上赔不起。

这样，所谓“主审”是仓促上阵的，虚有其名。恳求读者不要把“主审”二字看得太重，而是把我看作一个学习本教材的第一批学生。除了在眼所能及的范围内提出一些印刷错误外，我的学习心得报告如下：首先，本教材各章作者均是有关学科的佼佼者，功底扎实，保证了教材的学术性；第二，本教材作者有第一线教学的丰富经验，做到了论述深入浅出，适合本科生的需要；第三，本教材注意到有关学科在本世纪的最新进展，具有充分的时代性。

国庆和中秋长假，使我得以把大部分时间投入此项任务，借花献佛，以表我对北京大学出版社外语部和各位作者的钦佩之情，更表我对祖国六十华诞的无限喜悦！

胡壮麟

北京大学蓝旗营

2009年10月9日

CONTENTS 1

CONTENTS

1.6.1 Saussure as the Father of Modern

Chapter 2 Phonetics and Phonology: The Sounds and

SoundPatternsofLanguage⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯(14)

2.3.4 Prosodic Features: Stress, Tone and

2 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

Chapter 3 Morphology: The Word Structure of

3.5.3 Inflectional Morphemes and Derivational

CONTENTS 3

5.4.1 Grammatical Meaning and Lexical

5.5.2 Semantic Relations at the Sentential

4 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

Chapter 7 Discourse Analysis: Language Above the

7.5.1 Thematic Structure and Information

Chapter 8 Historical Linguistics: Language

CONTENTS

6 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

Chapter 11 Intercultural Communication:

11.3 The Relationship Between Language and

11.6 Language and Thought: Sapir- Whorf

11.7.1 Intercultural Communication as a Field of

11.7.2 Conquering Obstacles in Intercultural

CONTENTS 7

Chapter 12 Psycholinguistics: Language and

12.2 Language and the Brain: The Biological Foundations

12.2.1 Cerebral Lateralization and Language

12.5.1 First Language, Second Language and

Chapter 13 Cognitive Linguistics: Language and

语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

Chapter 14 Applied Linguistics: Language Teaching and

Learning⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯⋯(218)

14.2.2 The Innateness Hypothesis: Universal

14.4 Approaches and Methods in Foreign Language

1.1 What is Language

Language is something that we use almost every day, but we will feel at a loss when asked to define what language is. A most common definition of language might be: Language is a means of human communication. The merit of this definition consists in its brevity and popularity among users of language, yet it does not necessarily follow that it is an adequate definition. For example, it points out only the instrumental aspect of language and does not actually tell us what language is. Suppose someone has never seen a train and asks you what a train is. Do you think he or she will be satisfied with a simple answer like“a train is a means of transportation”? Furthermore, the term human communication also needs further explanation, because communication can happen in various forms, such as gestures, pictures, smoke signals, traffic lights, noises in the throat, or even occasionally secret codes, but none of them can be regarded as language.

The above- mentioned definition of language is not exhaustive in many other aspects as well.

First of all, we have been talking about“ language”, rather than“a language”,“ the languages” or“ languages”. In other words, we use the word“ language” in the singular form without the definite or indefinite article, but the word can certainly be used in other forms. We can talk about languages, because there are several thousand kinds of languages in the world, although the total number of world languages has been decreasing since the twentieth century. Thus, the language we use or learn is only a language shared by a certain number of people.

Second, we can also use the word“ language” to refer to some special features of language use by an individual or people in a certain period of history, such as Shakespearean language or nineteenth century language. Indeed, some radical linguists even argue that everyone's use of language is unique or that language is changing all the time, which might create obstacles to communication. That explains why a teacher sometimes may not understand the jargons used among students.

2 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

There are many other uses of the word“ language”. The Webster’s Dictionary ( Third Edition) provides 13 annotations of the word and perhaps this list is still not yet exhaustive. It can be argued that the dictionary definition of a word is not academic enough, but the situation is even more complicated among linguists, who tend to provide definitions of language from their own theoretical perspectives. Or to put it the other way round, the way a linguist defines language indicates the theoretical perspective that the linguist takes. The following definitions of language are quoted from John Lyons'(1981: 3--7) Language and Linguistics:

· According to Sapir (1921), “ language is a purely human and non- instinctive method of communicating ideas, emotions and desires by means of voluntarily produced symbols.”

According to Blochand Trager (1942),“a language is a system of arbitrary vocal symbols by means of which a social group co- operates.”

Hall (1968) defines language as“ the institution whereby humans communicate and interact with each other by means of habitually used oral- auditory symbols.”

· Chomsky (1957) points out that a language is“a set ( finite or infinite) of sentences, each finite in length and constructed out of a finite set of elements.”

The list of definitions of language can be continued, but we can detect from the above definitions the common aspects of language that linguists generally agree upon and a su mmary of these common aspects can serve as a definition of language of our own. To begin with, language is a system, which means that language consists of a finite set of interrelated elements that can be combined according to rules. Second, language is a system of vocal symbols, which suggests that spoken forms are more cardinal to a language than written forms. Illiterate people can communicate with others at ease even though they do not know how to read and write. Third, communication is an important aspect of the function of language though language can also be said to perform other functions. If we look at language from the angle of the relationship between people, we can argue that language is a means of human interaction. If we look into the psychological aspect of language, we will see language as a kind of knowledge that language speakers share. Fourth, the language we study in linguistics is human language in general, although other forms of communication can also be touched upon and may shed new light on human linguistic communication. Fifth, language is arbitrary in the sense that the relation between speech sounds ( form) and the ideas the sounds convey( meaning) is arbitrary.

To sum up, language can be defined as a system of arbitrary vocal symbols

used for human communication and interaction.

1.2 The Design Features of Language

The st ructural linguist C. F. Hockett (1958) proposed a set of key properties of human language, which are said to be the design features that distinguish human language from the system of communication in any other species, such as bees’ dancing, birds’ singing and even music. Hockett identified 13 features, and later linguists differ from each other as to the number, name and perhaps importance of these features. However, most linguists would recognize the following features as most important, that is, arbitrarriness, duality, productivity, displacement and cultural transmission.

Arbitrariness, as a property of language, means that there is no intrinsic relation between sound and meaning. In Saussure's terms, arbitrariness refers to the arbitrariness of the relation between sound image and concept. As Hudson(1984: 19) puts it, there is no reason why we attach the meaning“ animal with a grunt and a curly tail” to the sequence of sounds p-i-g other than the( perfectly good) reason that this is what we found others doing in our particular community when we arrived on the scene, so we are simply following their example. What Juliet says in Shakespeare's Romeo and Juliet makes a lot of sense:“ What's in a name? That which we call a rose by any other name would smell as sweet.” From this quotation we can say that what matters is what something is, not what or how it is called. This reveals the truth about the arbitrariness of language. Arbitrariness can also be illustrated by cross- linguistic evidence: different sounds are used to signify the same object in different languages. To put it the other way round, the same meaning is represented by different sounds in different languages.

Duality is the property of having two levels of structure or patterning. At the primary, higher level, language is analyzed in terms of combinations of meaningful units ( such as morphemes or words); at the secondary, lower level, language is composed of a sequence of segments ( phonological units) which do not have any intrinsic meaning but combine to form units of meaning. For instance, in the sentence He goes to school, at the first higher level, the sentence consists of meaningful units, he, go, - es (a morpheme indicating third person singular number), to, school. At the lower level, these meaningful units are made up of phonological units, /h/,/i:/,/g/, /əʊ/, etc. As individual sounds, these units do not have any meaning in themselves. Therefore, duality is also termed double articulation. The articulation of words or other meaningful units is the primary articulation, that of phonological units within words the secondary articulation. Duality is a design feature that

4 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

enables language to be productive, because language users can use or produce a large number of forms by combining a relatively small number of lower- level elements in a variety of different ways.

Productivity, also termed creativity, refers to the creative capacity of language users to produce and understand an infinitely large number of sentences, including sentences they have never used or heard before. Productivity is one of the distinctive features of human language because it allows novel combinations of elements and thus contrasts with the“ unproductive” communication systems of animals.

Displacement means that human beings can talk about objects or events which do not belong to the immediate setting in space and time. This is a very normal property of communication in human beings, rarely demonstrated in other species. As Bertrand Russell once said, no matter how eloquently a dog may bark, he cannot tell you that his parents were poor but honest. Human language can be used to talk about real or imagined matters in the past, present, or future. We can write science fictions, tell fairy tales, handle generalizations and abstractions, and even lie and deceive by means of language. This is the chief reason why linguists tend to believe that language is a more distinctive feature of human species than any other features like the use of tools or the size of brain, etc.

Cultural transmission means the transmission of language from one generation to the next through members of a society as opposed to genetic inheritance. In other words, the details of the linguistic system must be learned anew by each speaker. Though the capacity for language in the human being has a genetic basis, the particular language one learns is a cultural fact rather than a genetic one.

1.3 The Origin of Language

Although interest in the origin of language dates back to antiquity, we are still far from knowing when and how language arose. There are three well- known theories concerning the origin of language: the divine- origin theory, the invention theory and the evolution theory.

The divine- origin theory suggests that language is God's gift to mankind. According to Genesis of the Bible, God created Adam and gave him the power to name all things. Similar stories can be found in many religions of the world about a divine source who endows humans with language.

The invention theory maintains that language is a human invention. This theory is manifested in the five theories summarized by the Danish linguist Otto Jespersen (1922) with the following names:

Chapter 1 5

Language and Linguistics

· The bow- wow theory: Language originated by the imitation of natural sounds such as animal calls. The evidence for this theory is the existence of onomatopoeic words in language.

The pooh- pooh theory: Language arose from instinctive cries of emotion, such as pain, anger, pleasure, fear and surprise. This theory can be supported by the interjections in language.

The ding- don g theory: Language arose out of people's responses to the things around them by sounds. The original sounds people made were supposedly in harmony with the world around them. This can be explained by sound symbolism, which is the phenomenon that vocal sounds suggest meaning. One frequently cited example is the“ gl-” words for shiny things: glisten, gleam, glint, glare, glam, glimmer, glaze, glass, glitz, gloss, glory, glow, and glitter. Similar to the ding- don g theory, the oral- gesture theory suggests that language is an oral mirror of physical gestures through the movement of the tongue, lips and other vocal organs. One possible example is the oral gesture( movement of the tongue) in a“ goodbye” message as a vocal mimicry of the physical gesture( waving of the hand or arm).

· The yo- he- ho theory: Language developed out of the rhythmical grunts of humans working together. Examples supporting this theory are the prosodic features, especially of rhythm in language.

· The la- la theory: Language derived from sounds associated with the romantic aspect of life, such as song, love, art and poetry.

The evolution theory maintains that language developed in the course of the evolution of the human species. The development of language is closely related to the evolutionary development of the speech organs, especially the oral cavity and the pharyngeal cavity. For example, the shapes of human teeth, lips and the tongue evolved in such a way that enables humans to produce a variety of speech sounds. However, some people think that the major evolutionary step in the development of language relates to evolutionary changes in the brain( Fromkin & Rodman,1983:28). As Yule(1996/2000:5) puts it, there is an evolutionary connection between the tool- using and language- using abilities of humans, as both are related to the development of the human brain and are largely confined to the left hemisphere of the brain for most humans. The evolution theory seems to provide a better explanation for the origin of language than the divine- origin theory and the invention theory.

1.4 What is Linguistics

Linguistics is often defined as the science of language or as the scientific

6 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

study of language. However, people constantly challenge the possibility of rendering linguistics scientific, because it is first and foremost the study of language by means of language and thus people have every reason to suspect whether linguistic theories are simply the result of linguists' play on words. Furthermore, linguistics clearly belongs to the humanities and social sciences, which are obviously different from disciplines like physics, chemistry or biology, whose scientific status is unquestioned.

Linguistics is a science because linguists share the goal of scientific enquiry, which is objective understanding. The primary aim of linguistics is to understand the nature of language and of languages. Linguists attempt to construct theories of language, and to apply theoretical considerations to a description or analysis of language or languages. Objectivity is the most important principle of the scientific method. In order to obtain objectivity in linguistic study, linguists should see and describe a language as it is, not as what they think it ought to be. Language is something that we tend to take for granted, something with which we are familiar from childhood in a practical and unquestioning manner. As a result, there are some biases that we assume to be truth but are actually unjustifiable. For example, people tend to believe that their own language is the easiest to learn, but scientific study shows that learning a foreign language is equally difficult or easy if you like, though there are rare cases of individuals who seem to be talented for learning new languages. Another deep- rooted bias is that only the standard variety is the pure or correct form of a language, such as the kind of English spoken by BBC and VOA announcers or the kind of Mandarin Chinese spoken by CCTV news broadcasters.

Linguistics is scientific in its methodology. It is concerned with observing facts about language, setting up hypotheses, testing their validity and accepting or rejecting them accordingly. To avoid biases of the kinds mentioned above, modern linguists differ from traditional grammarians in adopting empirical rather than speculative or intuitive approaches in their study. Here follows the first distinction between modern linguistics and traditional grammar. That is, linguistics is descriptive rather than prescriptive. That is, linguists try to make statements which are testable, and take language as it is rather than say how it should be. The empirical approach to language that linguists adopt usually consists of four stages: data collection ( usually in the form of a corpus), tentative rule construction, tentative rule examination and rule finalization. Unlike the traditional grammarians, linguists do not believe that there is any absolute standard of correctness concerning language use which school teachers should view as their duty to maintain. Instead, linguists prefer to be observers or recorders of facts.

Chapter 1 7

Language and Linguistics

The second contrast between modern linguistics and traditional grammar is that linguistics regards spoken rather than written language as primary. Traditional grammar tends to emphasize the importance of written language and the writings or styles of classical writers are considered the most prestigious ( if not the best) form of language that learners should imitate. Linguists, however, give priority to spoken language, though written language is also part of linguistic research.

Thirdly, linguistics differs from traditional grammar in that it does not force languages into a Latin- based framework. In the past, Latin was considered the language that provided a universal grammar for all languages and other languages were forced to fit into Latin patterns and categories, especially its case system and tense divisions of past, present and future.

It must be emphasized, however, that linguistics, like any other discipline, builds on the past, not only by challenging and refuting traditional doctrines but also by developing and reformulating them. When criticizing traditional grammar for being unscientific, linguists do not deny altogether the contributions of traditional grammar to the development of modern linguistics.A balanced view on traditional grammar, therefore, is needed in order to track down the continuity of Western linguistic theories from the earliest times to the present day ( See Chapter 4 for a detailed discussion of traditional grammar).

1.5 The Scope of Linguistics

Language has been studied from different points of view. As a result, different dimensions of linguistics can be distinguished according to the point of view that is adopted or the special emphasis that is laid on certain aspects of language. Thus, the distinction s in the field of linguistic studies can provide a window on the scope of language.

The first distinction to be made is between general and descriptive linguistics. General linguistics deals with language, aiming at developing a theory that describes the rules of human language in general. Descriptive linguistics, on the other hand, is the study of particular languages, attempting to construct models that describe the rules of individual languages like Chinese, English and Russian. However, it should be made clear that general linguistics and descriptive linguistics are by no means unrelated. Instead, each depends on the other explicitly or implicitly. For example, general linguistics provides descriptive linguistics with a general framework in which a particular language can be studied.

The second distinction to be drawn is between diachronic linguistics and synchronic linguistics. The former is the study of language change, such as the

8 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

changes in grammar from Old English to Modern English; the latter is the study of a language existing in a“ state” at one particular point in time, regardless of its historical change, for example, the grammar of Old English or Modern English. In the nineteenth century, linguists were chiefly concerned with investigating the historical development of particular languages and formulating hypotheses about language change. Such an approach to the historical development of language was termed diachronic description of language by the Swiss linguist Ferdinand de Saussure, who argued that priority should be given to the synchronic description of language, that is, the study of language as it is at some particular point in time.

The third distinction is between theoretical linguistics and applied linguistics. According to Lyons (1981: 35),“ theoretical linguistics studies language and languages with a view to constructing a theory of their structure and functions and without regard to any practical applications that the investigation of language and languages might have, whereas applied linguistics has as its concerns the application of the concepts and findings of linguistics to a variety of practical tasks, including language- teaching.” Two points are especially worthy of note here. First, the distinction between theoretical linguistics and applied linguistics does not entail that there is no theoretical concerns in applied linguistics at all. Instead, hypothesis formulation and confirmation are also of vital importance to applied linguistics. Secondly, in talking about the distinction between theoretical linguistics and applied linguistics, people generally neglect the distinction between theoretical linguistics and general linguistics. As illustrated by Lyons (1981: 35),“ the goal of theoretical linguistics is the formulation of a satisfactory theory of the structure of language in general.”

General linguistics attempts to establish a workable theory of language at all levels. On the basis of the stratification of language and the corresponding foci of study, we can further divide general linguistics into several sub- branches, that is, phonetics, phonology, morphology, syntax, semantics. These areas are traditionally said to form the core of linguistics, because they deal with the mostly formally structured aspects of language. Although the primary object of description for linguists is the structure of language, many linguists study this in relation to its functions and in relation to social and cultural factors. As a result, pragmatics has come to be regarded as a branch of study in general linguistics.

In contrast with the core of linguistics, there are also branches of linguistics often called macrolinguistics where there is an interdisciplinary and applied orientation. Linguists have drawn on recent discoveries or developments from other areas of study and come up with enlightening views

on language. This leads to the emergence of studies like psycholinguistics, sociolinguistics, cognitive linguistics, anthropological linguistics, computational linguistics, forensic linguistics, applied linguistics, and so on. Studies in all these aspects contribute from different perspectives to the edifice of linguistics as a whole.

1.6 A Brief History of Linguistics

Linguistics dates back to more than 2000 years ago and scholars from different civilizations have all contributed to the study of language. For example, linguistics is termed Xiaoxue ( The Primary Learning) in ancient China and perhaps Xunzi is the first philosopher in the world who pointed out the arbitrary relationship between Name(名) and Object(实). In the Indian culture, the grammarian Panini dealt with rules of word formation in the Indian language systematically, in the form of a set of 4000 aphoristic statements, in the era between 5th and 7th centuries B. C. The Western linguistic tradition can be traced back to ancient Greek times around 500 B. C. Serious considerations of linguistic issues can be found in the writings of Plato and Aristotle. Modern linguistics evolves from the Western tradition of linguistic study, which is greatly influenced by the pioneering work of the Swiss linguist Ferdinand de Saussure.

1.6.1 Saussure as the Father of Modern Linguistics

Saussure is considered the father of modern linguistics because the book under his name Course in General Linguistics is the first systematic elaboration on linguistic theory. Published in 1916, three years after Saussure's death, the book was actually a collection of the notes taken by his students during his lectures. Saussure's central ideas were expressed in the form of pairs of concepts called dichotomy: diachronic versus synchronic, lan gue versus parole, signifiant versus signifié, and syntagmatic versus paradigmatic.

We have mentioned above the distinction between diachronic and synchronic approaches to language. The former sees language as a continually changing medium; the latter views it as a living whole, existing in a“ state” at a particular moment in time. Saussure's ideas were revolutionary in this aspect because he called for the priority of synchronic study of language, which was opposed to the nineteenth century historical linguistic tradition. According to Saussure, it is always necessary to carry out synchronic work: Before we can say how a language changed from State X to State Y, we should know something about X and Y.

Langue and parole are recognized by Saussure as two important aspects of

10 语言学基础教程(第二版)

Linguistics: An Elementary Course Book Second Edition

language. Langue is the totality of a language or the abstract language system shared by all the members of a speech community, while parole is the realization of lan gue in actual use, that is, the concrete act of speaking at a particular time and in a specific situation.

Signifiant and signifié are the two aspects of meaning in a linguistic sign. Signifiant( signifier) refers to the thing that signifies, and signifié( signified) is the thing or concept signified. The relationship between them is seen by Saussure as arbitrary.

Syntagmatic and paradigmatic relations are employed by Saussure to describe the relationships between linguistic signs. When the signs are seen as a linear sequence, the relationship between them is syntagmatic. A paradigmatic( or“ associative”) relation is a relation between a linguistic sign in an utterance and other signs in the language. These two dimensions of structures can be applied to phonology, vocabulary, or any other aspect of language, rendering the conception of language as a vast network of interrelated structures and mutually defining entities, a linguistic system.

1.6.2 American Structuralism

The American linguistic tradition chiefly originated from the American anthropologists’ concern to establish adequate description of the American Indian languages and cultures on the brink of extinction. Since the American Indian languages presented drastically different kinds of structure from the European languages, and there were practically no written records to rely on, anthropologists had to develop some efficient and systematic procedures to provide a careful account of the speech patterns of the living languages. Among these anthropologists were scholars like Franz Boas and Edward Sapir, who were pioneering structural linguists as well.

American structuralism flourished from the 1920s to the late 1950s. The st ructural approach to language description and analysis was synthesized in the book Language(1933) by Leonard Bloomfield. This book dominated linguistic thinking for over 20 years and led to progress not only in the studies of phonetics, phonology, morphology and syntax of the American Indian languages but also in the descriptive studies of grammar and phonology of the English language. The Bloomfieldian approach was termed structuralism chiefly because it employed techniques to identify and classify features of linguistic structure, especially the analysis of sentences into constituent parts.

However, the influence of American structuralism diminished in the1950s, especially after Noam Chomsky put forward the notions of generative linguistics as a reaction against structuralism.

1.6.3 Generative Linguistics

In 1957, the publication of Noam Chomsky’s Syntact ic Structures inaugurated the linguistic revolution known as generative grammar. Chomsky might have intended the book as a challenge to the structuralist tradition in America but the impact of this book has gone beyond its era and later proved to be a milestone in the history of modern linguistics.

Chomsky's theory on language is termed generative grammar or generative linguistics because it attempts to describe a native speaker's tacit grammatical knowledge by a system of rules that can generate the well- formed, or grammatical, sentences of a language while excluding all the ungrammatical, or impossible, sentences. The basic tenets of the Chomskyan view on language in the early days are expressed in a set of terms or hypotheses he proposed, including the dichoto my between competence and performance, universal grammar, deep structure and surface structure, and transformation al rules.

To Chomsky, linguistics should discover the men tal realities underlying the way people use language and thus he proposed the distinction between competence and performance. Competence refers to an ideal speaker’s knowledge of his language as manifest in his ability to produce and understand a theoretically infinite number of sentences. Performance refers to the actual use of the language by individuals in speech and writing. Chomsky argues that linguistics should be chiefly concerned with the study of competence rather than restrict itself to performance. That is, linguistic research should not be conducted on the basis of samples or collected data of language users' performance. Such samples are inadequate because they might contain many non- fluencies, changes of speech plan, slips of the tongue and other defects.

Chomsky argues that linguistics not only aims to provide an adequate account of competence in one language, but also attempts to establish principles of grammar shared by all languages, that is, the universal grammar, which can in the long term, shed light on the nature of human mind. The ambition that Chomsky cherishes stimulates a wide range of interest from scholars from many disciplines and has substantially influenced academic domains like psychology, pedagogy, computer science, artificial intelligence and even economics.

Another important pair of concept proposed by Chomsky is deep structure and surface structure. The major aim of generative grammar, according to him, is to provide a means of analyzing the process of deriving surface structures from deep structures. In his own practice in analyzing the English language, Chomsky put forward sets of generative rules and transformational rules, and it is precisely for this reason that Chomsky's theory is also termed Transformational-

12 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

generative(TG) grammar.

1.6.4 Functional Linguistics

Unlike generative linguistics, functional linguistics does not have one unanimously recognized leading figure, nor even generally acknowledged theoretical tenets. What we can be sure of is that early functional linguistics flourished chiefly in Europe, with scholars taking up Saussure's ideas and developing schools of thought that emphasized the study of functions of language in general, or the study of linguistic features in particular texts or contexts.

The first noteworthy school of functional linguistics is perhaps the Prague School, represented by linguists in Czechoslovakia, who founded the linguistic circle of Prague. Notably, the Prague School linguists emphasized the function of units and applied the idea to the study of phonology. The important concept of phoneme in distinguishing or demarcating words, for instance, is one of the contributions of the Prague School linguists. Leading figures in this school include Nikolais Trubetzkoy, Roman Jacobson and V. Mathesius, etc.

The Copenhagen School is a group of linguists who constituted the Copenhagen Linguistic Circle in the mid-1930s. Led by Louis Hjelmslev, the school developed a philosophical and logical basis for linguistic theory. They developed an approach to linguistics known as Glossematics or neo- Saussurean linguistics.

The London School is represented by the British linguist J. R. Firth and his followers. Distinguished from other theorists of his generation, Firth insisted that language should be studied as part of a social process. A linguist's data were for him events embedded in specific contexts and therefore he emphasized the importance of context in linguistic studies at all levels. In addition, he argued for the polysystemic principle of study, that is, the principle of analyzing language patterns with different systems within different contextual frameworks. Firth’s influence is widespread and the systemic linguistic school established by his student M. A. K. Halliday has much in common with his notions and thus is sometimes termed the neo- Firthian school.

In America, functional linguistics flourished since the 1970s, when formalism still held sway, and became mature in the 1980s and 1990s. However, interest in functional approaches to language in America can be traced back to Franz Boas (1858—1942), Edward Sapir (1884—1939) and Benjamin Lee Whorf(1897—1941). In the 1960s—1970s, the contributions made by Dwight Bolinger, Wallace Chafe, Charles Fillmore and Charles Li paved the way for the development of functionalism in America. The major functional approaches include: Tagmemics, Case Grammar, Stratificational

Chapter 1 13

Language and Linguistics

Grammar, Discourse Grammar, Role and Reference Grammar, Cognitive Grammar, and so on.

Further Readings

Hudson, R. 1984. Invitation to Linguistics. Oxford: Blackwell.

Lyons, J. 1982. Language and Linguistics. Cambridge: Cambridge University Press.

Robins, R. H. 2002. A Short History of Linguistics. 北京:外语教学与研究出版社.

Traugott, E. C. & Pratt, M. L. 1980. Linguistics for Students of Literature. New York: Harcourt Brace Jovanovich, INC.

Yule, G.2000. The Study of Language. 北京:外语教学与研究出版社.

Questions and Exercises

1. Define the following terms.

arbitrariness duality productivity

displacement syntagmatic relation paradigmatic relation

synchronic linguistics diachronic linguistics macrolinguistics

2. What is language?

3. How to tell the difference between human language and communication systems of animals?

4. Illustrate the design features of human language with examples.

5. In what sense is language arbitrary? Can you cite some examples to show that language is not arbitrary?

6. Why is displacement an important feature of human language?

7. Why is language seen as a distinctive property of human species? To what extent do you agree that language is a distinctive property of human species?

8. Among the different theories of the origin of language, which one do you think is most convincing?

9. In what sense is linguistics a science?

10. Say something about the difference between traditional grammar and linguistics.

11. Why is Saussure seen as the father of modern linguistics?

12. What is the significance of the distinction between diachronic and synchronic approaches to language study?

13. What are the differences between competence and performance? Is this dichotomy related in any way to Saussure's distinction between lan gue and parole?

14. What are the basic tenet s of Structuralism?

15. Name some of the functional linguistic schools.

2.1 Introduction

Human beings communicate with each other mainly in two modes of language— speaking and writing. In oral communication with someone from a different linguistic community, what first strikes us as different may not be that person's possibly different idea about something, but the exoticness of the speech sounds uttered.

Speech sounds are the sounds utilized by all human languages to represent meaning. They are the sounds produced or heard in using language to speak or understand, and are related by the language system to certain meanings. Anyone who knows a language knows what sounds are in the language and how they are“ strung” together and what these different sound sequences mean( Fromkin & Rodman, 1983: 35). The study of human speech sounds can be done by examining the features of the sounds per se ( Matthews, 2001: 33), which is adopted by the students of phonetics. The speech sounds can also be studied from the point of view of how they are actually used in different languages( Katamba,1989:66), including how some of the sounds interrelate and interact with each other within a given language system, which is adopted by the student of phonology.

2.2 Phonetics

The term“ phonetics” is composed of phone- , which is from the Greek phonê meaning“ sound” or“ voice”, and - ics, which means “a study of”. So phonetics is the general study of the characteristics of speech sounds ( Yule,2006: 30). It is concerned with the physical properties of speech sounds( phones), and the processes of their physiological production, auditory reception, and neurophysiological perception. Thus, there are three branches of phonetics. The study of how speech sounds are made or“ articulated” is called articulatory phonetics. The study of how speech sounds are perceived is called auditory phonetics ( or perceptual phonetics). The study of the physical properties of speech sounds is called acoustic phonetics. The primary concern of

Chapter 2 15

Phonetics and Phonology: The Sounds and Sound Patterns of Language phonetics is usually articulatory phonetics.

2.2.1 Speech Organs

The study of phonetics helps increase our awareness of the properties of speech sounds, which in turn can improve our ability to articulate them with accuracy by manipulating our speech organs more appropriately. The speech organs of the human beings consist of three major areas: the pharyngeal cavity( the throat), the oral cavity( the mouth) and the nasal cavity( the nose). The power source of our speech sounds is the flow of air from our lungs, and the air stream coming from the lungs may be modified in the cavities in various ways, resulting in different kinds of speech sounds.

The oral cavity is made up of the tongue, the uvular, the soft palate ( the velum), the hard palate( the palatal), the teeth ridge( the alveolar), the teeth( the dental) and the lips( the labial). Of these speech organs, the very flexible tongue is of particular importance, because the principal source of sound modification is the tongue, so much so that the word for it is often used as a synonym for language( Poole, 2000: 42).

The nasal ca vity is a passage which is definitely open or closed, producing sounds either nasalized or non- nasalized.

All speech sounds are produced by manipulating the vocal tract while air is flowing through it. If this manipulation produces significant obstruction of the air stream, the result is a consonant. If it does not produce any obstruction,

16 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

the result is a vowel. Thus, speech sounds can be classified into consonants and vowels according to whether the air stream coming from the lungs meets any obstruction in producing the sounds.

2.2.2 Consonants

Consonants can be described according to their place of articulation and manner of articulation, that is, where the air stream is modified and how it is modified.

2.2.2.1 Place of Articulation

After the air stream coming from the lungs passes through the larynx, it comes up and out through the mouth and/ or the nose. Most of the 24 consonants in English are produced by using the tongue and other parts of the mouth to constrict the shape of the oral cavity through which the air is passing( Yule, 2006: 30). In other words, they are mostly articulated via closure or obstruction in the vocal tract. Consonants can be voiced or voiceless according to whether the vocal cords vibrate or not. The following terms are used to describe consonants in terms of where they are articulated.

A) Bilabials: These are sounds that are produced with the two( bi = two) lips( labia = lips). In English, bilabial consonants are[p],[m],[b], and[w]. Only [p] is voiceless, and the other three are all voiced.

B) Labiodentals: These are sounds that are produced with the lower lip and upper teeth. In English, there are only two labiodentals: [f]( voiceless) and [v]( voiced).

C) Interdentals: These are sounds that are produced with the tongue between( int er = between) the teeth, or just behind the upper teeth ( also called“ dental”), such as [θ]( voiceless) in words like thin and bath, and [ð]( voiced) in words like there and then.

D) Alveolars: These are sounds that are produced with the tongue tip at the alveolar ridge, behind the teeth. In English, there are 7 alveolars. They are[t],[d],[s],[z],[n],[1] and[r], of which [t] and[s] are voiceless and the others are all voiced.

E) Palatals: These are sounds that are produced with the front or body of the tongue raised to the palatal region. In English, palatals are [ʃ]( voiceless) in words like ship and shout, [tʃ]( voiceless) in words like child and china,[3]( voiced) in words like treasure and pleasure,[dʒ]( voiced) in words like bridge and joke, and [j]( voiced) in words like you and yet.

F) Velars: These are sounds that are produced with the back of the tongue raised to the soft palate ( also called“ velum”). In English, there are three velars. They are [k]( voiceless) in words like cold and kick,[g]( voiced) in

Chapter 2 17

Phonetics and Phonology: The Sounds and Sound Patterns of Language

words like go and bag, and [ŋ]( voiced) in words like sing and tongue.

G) Glottal: In English, there is only one glottal, which is produced without an active use of the tongue and other parts of the mouth. It is [h]( voiceless). It is termed as such because the sound is produced with the glottis( the space between the vocal cords) open. In other words, there is no manipulation of the air passing out of the mouth.

2.2.2.2 Manner of Articulation

The 24 consonants can also be described in terms of how they are articulated, that is, how the air stream is modified. In producing consonants, the air stream from the lungs is either completely blocked (e. g. stops), partially blocked (e. g. laterals) or the opening is so narrow that the air escapes with audible friction(e. g. fricatives). With some consonants (e. g. nasals) the air stream is blocked in the mouth but allowed to escape through the nose.

A) Stops: Also called plosives, these are sounds produced by some form of“ stopping” of the airstream very briefly then letting it go abruptly. Such sounds as[p],[b],[t],[d],[k], and[g] fall into this category.

B) Affricates: These are sounds produced by combining a brief stopping of the airstream with an obstructed release which causes some friction. The two consonants,[tʃ] and [dʒ], are affricates.

C) Fricatives: These are sounds produced by almost blocking the airstream and letting the air force through the narrow opening. They are termed as fricatives because when the air is pushed through, a kind of friction is produced. The fricative consonants are [f],[v],[s],[z],[ʃ],[ʒ],[θ] and[ð].

D) Nasals: These are sounds produced by allowing air to pass through the nose while blocked in the mouth. Nasal consonants are [m],[n] and [ŋ].

E) Liquids: These are sounds produced with a minimal constriction allowing air to pass freely.[l] and [r] are liquid consonants. In producing [1], the airstream flows around the sides of the tongue( it is called a lateral liquid); and in producing [r], the airstream flows through the center of the mouth.

F) Glides: These are sounds produced with a minimal constriction corresponding to a vowel( sometimes called“ semi- vowels”). [w] and [j] are described as glides. The sound [h], as in hello, can also be classified as a glide because of the way it combines with other sounds.

Having described the consonants in terms of both the place and the manner of articulation, we can summarize the basic information in the chart below (-v= voiceless, +v = voiced):

18 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

( Yule, 2006:34)

Thus, a consonant in English can be described according to whether it is voiced or voiceless, its place of articulation and manner of articulation, as shown in the following examples:

[b]: voiced bilabial stop

[f]: voiceless labiodental fricative

[n]: voiced alveolar nasal

[h]: voiceless glottal glide

2.2.3 Vowels

Different from consonants, vowels are sounds produced with no obstruction of the air stream, and they are all typically voiced.“ To describe vowel sounds, we must consider the ways in which the tongue influences the‘ shape’ through which the airflow must pass”( Yule, 2006: 38). That is to say, what makes one vowel sound different from another is the size and shape of the space within your mouth: whether your tongue is higher or lower, whether you raise the front or the back of your tongue and whether you round or spread your lips. These variations affect how the air in your mouth resonates and they are responsible for the different qualities of the vowels.

2.2.3.1 Monophthongs

Most vowel sounds have only one“ pure” vowel and are thus called monophthongs ( mono = one), the articulation of which involves just one tongue position. Below is a chart of such simple vowels categorized in terms of the tongue position within the mouth.

Chapter 2 19

Phonetics and Phonology: The Sounds and Sound Patterns of Language

In this chart, high- mid- low refers to height of the tongue in the mouth; front- central- back refers to the part of the tongue. The symbol[:] suggests that the[i:], [u:], [ə:] and [ɔ:] are long vowels. Here are their English exemplifications.

[i:] eat, bee, key [u:] move, too, food

[I] big, myth, village [ʊ] could, book, put

[ε] dead, bet, said [3:] dirt, hurt, verse

[ɔ] cot, bomb, hot [ɔ:] saw, caught, ball

[æ] dam, fast, pat [ə] above, coma, suppose

[ʌ] flood, but, rough [ɑ:] dark, heart, mark

When we look at the chart above, we should keep what Yule(2006: 39) says in mind:“ Vowel sounds are notorious for varying between one variety of English and the next, often being a key element in what we recognize as different accent”. For example, many Americans do not distinguish the vowels[ɒ] and [ɔ:], pronouncing cot and caught the same way.

2.2.3.2 Diphthongs

Vowel sounds may not be just one pure vowel. They can be complex speech sounds that begin with one vowel and gradually change to another vowel within the same syllable. In such cases, we have diphthongs ( di = two). Altogether, English can be said to have 8 diphthongs. They are [eI], [əʊ][aɪ],[aʊ],[ɒɪ],[ɪə],[εə], and [ʊə], which can be exemplified as follows:

[eI] date, slay, grey [əʊ] toe, slow, whole

[aɪ] buy, fly, eye [aʊ] how, proud, south

[ɒɪ] toy, spoil, noise [Iə] dear, fear, leukemia

[εə] dare, hare, air [ʊə] tour, gourmet, moor

In passing, English also has a few triphthongs, where three vowel sounds[aIə], [aʊə] come in succession in words such as fire and power. Anyway, it should be repeated here that when we pronounce most of the single vowel

20 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

sounds, our vocal organs take only one position very briefly, but in pronouncing monosyllabic vowel combinations, our vocal organs will experience a quick but smooth movement, or glide, from one vowel to another.

2.2.4 Transcription of Speech Sounds

Typologically, English is a language whose written form ( orthography) does not usually give us a secure clue to the actual pronunciation of its words( in contrast with Spanish and Italian, in which there is a consistent relationship between orthography and pronunciation). For this reason, phonetic transcription ( or phonetic notation), the visual system of symbols for the sounds occurring in spoken communication, was invented to provide a function that orthog raphy cannot. It displays a one- to- one relationship between symbols and sounds. The most common type of phonetic transcription uses phonetic symbols( such as the International Phonetic Alphabet, IPA for short). Incidentally, we enclose phonetic symbols within square brackets [ ], as distinguished from phonemes that are put between slash marks / /(e. g. /s/. See Section 2.3.1).

With phonetic transcription, we can step outside of orthog raphy and examine differences in pronunciation between the numerous dialects within the language. In addition, we can identify changes in pronunciation that have taken place over time. Phonetic transcription can be broad or narrow. Broad transcription indicates only the more noticeable phonetic features of an utterance, whereas narrow transcription encodes more information about the phonetic variations of the specific allophones ( see Section 2. 3. 2) in the utterance. For example, one particular pronunciation of the English word little may be transcribed using the IPA as ['lItl] or['lItI]; the broad transcription indicates merely that the word ends with the phoneme /l/, but the narrow transcription( with a diacritic) indicates that this final /l/ ([t]) is a dark sound. Thus, in narrow transcription, we make a distinction between the variations of the same phoneme. For example, the clear [l] and the dark [t] belong to the same phoneme /l/, but the clear[l] occurs before a vowel, as in lead and life, and the dark [t] occurs at the end of a word after a vowel or before a consonant, as in tool and milk.

The merit of the narrow transcription is that it can help learners to get exactly the right sound, and allow linguists to make detailed analyses of language variation. But the demerit is that it is rarely representative of all speakers of a language. The advantage of the broad transcription is that it allows statements to be made which apply right across a relatively diverse language community. It is thus most commonly adopted as the pronunciation data in foreign language dictionaries.

Chapter 2 21

Phonetics and Phonology: The Sounds and Sound Patterns of Language

2.3 Phonology

Phonetics is interested in the speech sounds of all human languages: how they are pronounced; how they differ from each other; what phonetic features they possess; how they are classified, etc. However, in studying the speech sounds of a specific language, linguists may ask: What are the relationships between these speech sounds in the language? What is their linguistic function in the meaning structure of the language? What changes do they take in actual speaking? What sound patterns do these sounds form in the language? These questions are the major concerns of phonology.

Phonology is essentially the description of the system and patterns of speech sounds in a language. More specifically, it deals with the set of sounds that occur in a given language, the permissible arrangements of these sounds in words, and the rules or processes that affect sounds.

2.3.1 Phoneme

Before defining phonemes, we may as well know what phonemes usually look like. Phonemic representations are conventionally put between slash marks (e. g. /t/) as opposed to the square brackets for phonetic symbols (e. g.[t]). The symbol chosen for each phoneme is usually a phonetic symbol intended to suggest how the phoneme is most typically articulated. For example, the phoneme that occurs at the beginning of the word bit is commonly realized as a voiceless bilabial stop [p], and so the phoneme is represented as /p/. The entire word pit is represented as /pIt/, with the three phonetic symbols for the three phonemes present in it.

Then, what is a phoneme? According to Aitchison(1992: 39), a phoneme is the smallest segment of sound which can distinguish two words. This seemingly simple definition actually tells us two marked features about phonemes: one is that they are basic and the other is that they are meaning- distinguishing. That is to say, the essential property of a phoneme is that it functions contrastively ( Yule, 2006: 44). For example, we know that there are two phonemes /b/ and /p/ in English because they can be used to make contrasts in meaning between the words big and pig, or bond and pond. This contrastive property is the basic operational test that is used in determining the phonemes in a language. If the substitution of one sound for another results in a change in meaning, then these two sounds represent two different phonemes. That is to say, phonemes have distinctive features: phonemes are capable of distinguishing meaning, and the distinctive features differentiate one phoneme from another.

22 语言学基础教程(第二版)

Linguistics: An Elementary Course Book Second Edition

As can be seen, the two words big and pig we cite for illustration are identical in form except for a contrast in one phoneme, occurring in the same position. Two words which are different in meaning and identical in form except for one sound segment are referred to as a minimal pair. More examples of minimal pair are: tick— dick, pat— pad, tap— top, site— side. If we have a group of, not just two, words that can be differentiated by changing one phoneme( always in the same position in the word), then we have a minimal set, such as bate- mate- date- late- rate- wait- gate- hate- fate.

2.3.2 Phone and Allophone

It should be kept in mind that phoneme is an abstract unit or sound type(“ in the mind”), and a phoneme is not or need not always be realized in the same manner phonetically. In other words, one abstract sound type can have different versions in actual speech (“ in the mouth”). These different versions are described as phones ( Yule, 2006: 45). Take the English phoneme /p/ as an example. It is phonetically an aspirated plosive [pʰ] in the word pit(“ aspirated” means you will feel a puff of air if you put your hand before your mouth when pronouncing it), but an unaspirated plosive[p] in the word spit. So we have two phones [pʰ] and [p], both of which are versions of one phoneme /p/.

This example indicates that there are often differences in the way a phoneme is pronounced in a specific context. The variant pronunciations of a phoneme are called allophones ( allo- means“ other” or“ variation”) of that phoneme. So we can say that [pʰ] and[p] are allophones of /p/. They are the variants of the same phoneme and occur in complementary distribution.“ When two sounds are in complementary distribution, they are barred from occurring in identical environments: there is a rigid division of labor, as it were, so that one sound appears in certain contexts and the other in some different, clearly defined contexts”( Katamba, 1989: 19). In our example, the aspirated plosive[pʰ] always occurs at the beginning of a syllable and is followed by a stressed vowel( as in the word pit) while the unaspirated plosive [p] occurs in other situations ( as in the word spit).

“ The crucial distinction between phonemes and allophones is that substituting one phoneme for another will result in a word with a different meaning( as well as a different pronunciation), but substituting allophones only results in a different ( and perhaps unusual) pronunciation of the same word”( Yule, 2006: 45).

2.3.3 Phonotactics

In learning a language, it is not enough just to learn its vocabulary. We

Chapter 2 23

Phonetics and Phonology: The Sounds and Sound Patterns of Language

have to learn the grammatical rules so that we can arrange the words in proper sequences for meaningful communication. By the same token, it is not enough to learn how to articulate the speech sounds of a language, for we also need to learn the rules that govern the sequences of sounds in the language. On this account, phonology can be regarded as the grammar of phonetic patterns. The particular branch of phonology which studies the permissible sound combinations is called phonotactics ( in Greek phone = voice and tactic = course). To put it differently, phonotactics is a branch of phonology that deals with restrictions in a language on the permissible combinations of phonemes.

In the English language, fig, wig, rig, dig and pig are both phonotactically acceptable and semantically meaningful words. Another two forms that can be added to the minimal set are lig and vig. Though they are not English words, they are phonotactically permissible. One day, as Yule(2006: 47) says, they might come into use. “ They might, for example, begin as invented abbreviations(I think Bubba is one very ignorant guy.— Yeah, he is a big vig)”. On the other hand, the forms like bned and sthring do not exist and are unlikely ever to come into existence because they violate the phonotactical restrictions.

The examples indicate that phonotactic restrictions always operate on a unit that is larger than the single phoneme. The phonemic features that occur above the level of one sound segment are called suprasegmentals. Suprasegmental features include syllable and prosodic features known as stress, tone and intonation. Phonotactic constraints typically operate on the suprasegmental unit of syllable.

A syllable is a unit in speech which is often longer than one sound and smaller than a whole word. All words can be divided into one or more syllables. Words with one syllable ( dog, bright) are monosyllabic, while words with more than one syllable( little, today) are polysyllabic. The syllable is defined by the way in which vowels and consonants combine to form various sequences. In English, a syllable must contain a vowel or vowel- like sound. The most common type of syllable in language also has a consonant(C) before the vowel (V), and is typically represented as CV. Technically, the basic elements of the syllable are the onset and the rhyme. The onset can be one consonant, like [p] in put, or a combination of more than one consonant called a consonant cluster, like [ sk] in sky and[ spr] in spring. The rhyme consists of a vowel, which is treated as nucleus, plus any following consonant(s), described as coda( Yule, 2006: 47). The structure of a typical syllable can be represented as follows:

24 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

Now we are in an easy position to list some of the phonological rules in English( Note: the list is by no means exhaustive), which are extremely helpful in describing the possible words in English.

1) All the sounds in English can be at the beginning of the syllable except the nasal consonant /ŋ/. So /ŋʌt/ is“ ungrammatical”.

2) Long vowels( such as /i:/) and diphthongs cannot be followed by /ŋ/.

3) The vowel of house does not occur before a final /b/, /p/, /m/,/f/, or /v/.

4) The vowel /ə/ does not occur in stressed syllables.

5) One vowel sound alone can constitute a syllable, like /ə/. But a consonant cannot, though sometimes a consonant can serve as the nucleus, as in the second syllable of kitten.

6) In English, there are many possible CC onset combinations, as in blanket, brag, dwarf, track, twinkle, fly, freight, and through./1/,/r/ and/w/ are used in the second position, and cannot serve as the first consonant in the CC onset combinations.

7) In a CC onset combination, if the first consonant is /s/, the second consonant can only be/f/,/k/,/l/,/m/,/n/,/p/,/t/, or /w/(e.g. sphere, school, scope, slender, small, snow, spin and sway). Though / sr/ is possible, it exists only in some exotic words like Sri Lanka and Srinager.

8) In a CC onset combination, if the first consonant is /θ/, the second consonant can only be /r/ or /w/(e. g. t hrive and thwart).

9) If the onset is a consonant cluster with three consonants (CCC), the first consonant must always be /s/, the second can be /t/, /p/, /m/ or /k/ and the third one should be /j/, /l/, /r/ or /w/(e. g. skewer, smew, spew, student, street, splash, squad, scratch and spring).

10) If /j/ is at the end of an onset(/ pj/, / bj/, / tj/, / dj/, / kj/, / fj/,/ vj/, /θj/, / sj/, / hj/, / mj/,/ nj/, / lj/, / spj/, / stj/,/ skj/), it must be followed by /u:/ or /ʊə/ (e. g. pure, bureau, tutor, duke, cute, fusion, view, thew, suit, hew, muse, new, lure, spurious, studious, and skew).

Chapter 2 25

Phonetics and Phonology: The Sounds and Sound Patterns of Language

11) English normally permits up to two consonants in the coda (e. g. belt, dump) and sometimes even more consonants can pile up at the end of words(e. g. sixths, melts, texts, strengths).

2.3.4 Prosodic Features: Stress, Tone and Intonation

Prosodic features refer to phonological variations mainly in stress, tone and intonation, that is, relative loudness of syllables, changes in the pitch of a speaker's voice and the choice of pitch level, and so on. They are features that are sometimes loosely synonymous with suprasegmental features.

Almost every English word has at least one syllable which is noticeably more prominent than the other syllables. The relative prominence that a particular syllable has in pronunciation is called the stress of the word. The syllable that receives a stress is the stressed syllable. The word dictionary, for example, has four syllables and the stress falls on its first syllable in pronunciation. English- speaking people see eye to eye with each other on the stress of most words in the language, but they may have different opinions on the stress of words like exquisite, vagaries and controversy— some people stress the first syllable while others give prominence to the second syllable. Stress is sometimes indeed an issue of controversy!

If a word has several syllables, it often has both its primary stress ( or main stress) and a secondary stress. For example, explanation has the main stress on the third syllable but a secondary stress on the first one.

Stress is a helpful means for us to distinguish words that have the same forms but belong to different grammatical categories. For instance, export and' subject are nouns but ex ' port and sub' j ect are verbs. Stress is sometimes even essential in distinguishing words that are completely different, such as ' billow from be' low.

Apart from the ordinary word- stress, there is also sentence- stress in actual English communication: people may place a strong stress on a particular word within an utterance for the sake of emphasizing that word. For instance, John met Mary at the school gate yesterday can be uttered with different sentence- stresses, and the implications will be certainly different:

JOHN met Mary at the school ga te yesterday. ( John rather than anybody else)

John MET Mary at the school gate yesterday. ( John didn't do anything else)

John met MARY at the school ga te yesterday. ( Mary rather than anybody else)

John met Mary at the SCHOOL gate yesterday. ( Not any other gate)

John met Mary at the school GATE yesterday. ( Not anywhere else)

26 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

John met Mary at the school gate YESTERDAY.( Not any other day)

Tones are pitch variations, which are caused by the differing rates of vibration of the vocal cords. When we listen to people speaking, we can hear some sounds or groups of sounds in their speech to be relatively higher or lower than others. This relative height of speech sounds as perceived by a listener is called pitch. Pitch variations can distinguish meaning. For example, Chinese is a tone language which makes use of four tones to distinguish words or meanings: high level( yi, meaning“ one”), rising (yí, meaning“ move”), falling- rising(yǐ, meaning“ already”) and falling (yì, meaning“ idea”).

Intonations are pitch differences that extend over phonological units larger than the syllable. By means of intonation, syllables are grouped into phrases, and phrases into sentences. Intonation distinguishes meaning through intonation patterns, which are usually expressed by variations in pitch, loudness, syllable length, and sometimes speech rhythm. Intonation is used to perform grammatical functions, that is, to indicate whether an utterance is a question, or a statement. It can also be used to show the speaker's attitude to the subject matter discussed or to the listener.

2.3.5 Co- articulation Effects

So far we have discussed the speech sounds in syllables and words as if they are articulated very carefully and slowly. But normally, we do not speak like that unless we are talking to someone from a different linguistic community with little knowledge of our language or to a person of much lower cognitive ability such as a baby. In spontaneous connected speech, we usually speak fast, moving from one sound to the next without stopping. In other words, when we are preparing to articulate a phoneme, we are already thinking of the one which follows it. Take the English word map for example. As the nasal consonant [m] precedes the vowel [æ], some of its nasality will be carried forward so that the vowel will begin with a smidgen of nasal quality. When such simultaneous or overlapping articulations are involved, we can feel the co- articulation effects. Thus, co- articulation effects involve the phonological processes of spreading phonetic features to neighboring segments. There are two major types of co- articulation effects: assimilation and elision.

2.3.5.1 Assimilation

Assimilation, just as its name implies, is a phonological process by which a sound becomes more similar to its nearby sound in terms of some phonetic features. It is the process of one sound exercising influence on another and as a result, the two sound segments occurring in se quence sound more alike. This process happens because our speech organs usually do the job with higher

Chapter 2 27

Phonetics and Phonology: The Sounds and Sound Patterns of Language

speed, greater ease and better efficiency in connected speech.

Assimilation in which a following sound brings about a change in a preceding one is called regressive assimilation or anticipatory co- articulation. Take black people as an example. If we pronounce the two words individually, black will be [blæk]. But if we articulate the phrase at a normal speed, the last phoneme of the word,[k], will take the feature“ labial” of the following consonant sound /p/ in people, and the two words will be perceived by the ear as [blæp pʰi: pl]. What happens is labial assimilation. In the same way, the normal articulation of this year [ðIʃjIə] involves palatal assimilation.

Assimilation in which a preceding sound brings about a change in a following one is called progressive or perseverative assimilation. For example, whether a preceding consonant is voiced or voiceless will influence a following sound, as exemplified by the difference between the [s] in words like pets and books and the[z] in toads and frogs and the difference between the final [t] in jumped and the final [d] in sobbed.

2.3.5.2 Elision

The word“ elision” is derived from the verb“ elide”, meaning“ to leave out”(e-= out,- li de = strike). Elision, then, is a phonological process that involves the omission of a phoneme ( frequently unstressed vowels or media l consonants) in connected speech or verse. It is the process of not pronouncing a sound segment that might be present in the deliberately careful pronunciation of a word in isolation ( Yule, 2006: 49). You can try the phrase you and me. If you speak fairly quickly, you will find yourself uttering [ ju:æn mi:]. The stop sound [d] in the word and is omitted. The elision of the sound has much to do with its phonological environment: it is both preceded by the nasal [n] and followed by another nasal [m]. Efficiency will lead you not to devote speech energy to including the stop sound [d]. Try East China and you will hear yourself speaking [i:stʃaInə].

2.4 Summary

From the introduction above, we can see clearly that phonetics is a discipline that studies how possible speech sounds are articulated and perceived. It provides us with methods for their classification and description. By studying it we can get a picture of the physical properties of speech sounds. Phonology, on the other hand, deals with systems of linguistically significant sounds( Kaye, 1989:10). In other words, phonology investigates the ways in which speech sounds are used systematically to form words and utterances( Katamba, 1989: 60), and its primary aim is to discover the principles that

28 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

govern the way sounds are organized in language, and to explain the variations that occur.

Even though sounds in themselves have no meaning, and even though the associations between sounds and meaning in language are arbitrary and conventional, there are ways of using sounds to complement meaning( Traugott & Pratt, 1980: 69). To quote Alexander Pope in his An Essay on Criticism,“ The sound must seem an echo to the sense”. This phenomenon is called sound symbolism ( See Chapter 1). Take the sound [ʃ] as an example. This palatal fricative, if combined with short vowels, is often suggestive of an unpleasant event or a speedy motion like shout, smash, crush, dash and rush. In pronouncing words that contain several consonants and only one or two short vowels, we have to be swift and the swiftness in sound is often a clue to its meaning, such as abrupt, fast, crash, slap and jump.

Here we wish to point out that phonetics and phonology are not esoteric academic disciplines with no relevance to the real world. They have actually found their use in such fields as language teaching, drama performance, speech therapy, communications engineering, forensic pathology, etc. ( Katamba,1989: 75--77).

Further Readings

Clark, J. & Yallop, C. 2000. An Introduction to Phonetics and Phonology. 北京:外语教学与研究出版社.

Gimson, A. C. & Arnold, E. 2000. An Introduction to Phonetics and Phonology. 北京:外语教学与研究出版社.

Katamba, F. 1989. An Introduction to Phonology. London and New York: Longman.

Roach, P. 2000. English Phonetics and Phonology: A practical course. 北京:外语教学与研究出版社.

Roach, P. 2001/2003. Phonetics. 上海:上海外语教育出版社.

Yule,G. 2006. The Study of Language ( Third Edition). Cambridge: Cambridge University Press.

Questions and Exercises

1. Define the following terms:

speech sound prosodic features complementary distribution

phoneme allophone phone

minimal pair interdentals consonant

vowel manner of articulation phonetic transcription

phonotactics suprasegmentals syllable

2. Describe the following sound segments in English.

1)[e] 2)[u:]

3)[m] 4)[p]

5)[dʒ] 6)[δ]

3. How is phonetics related to phonology and how do they differ from each other?

Chapter 2 29

Phonetics and Phonology: The Sounds and Sound Patterns of Language

4. How are vowels classified in the English language?

5. How are consonants classified in the English language?

6. What are co- articulation effects? Illustrate them with your own examples.

7. Why are the following sound combinations not possible in English?

1) [ŋaʊt] 2)[tʃi:ŋ]

3) ['metɪ] 4)[θnaɪv]

5)[tjʊl] 6)[maʊp]

Morphology: The Word Structure of Language

3.1 Introduction

Morphology, which literally means“ the study of forms”, is about the structure of words. More specifically, it refers to the study of the internal structure of words, and the rules by which words are formed. Morphology is the area of grammar that attempts to answer the following questions: How is a word structured? What is the basic component part of a word? How are the component parts put together to form words? What are the word- formation processes in a language? How is the form of a word affected when it is used with other words or in a sentence?

3.2 The Words of Language

It is difficult to define the term“ word”. For some people, words are physically definable units, whose boundaries are usually recognized by spaces in writing or slight pauses in speech. For others, a word is a unit of meaning and a unit of sound. There are still others who regard words as grammatical units that can function in a sentence. Stockwell and Minkova (2001: 56) characterize a word as the smallest unit that one thinks of as being basic to saying anything. It is the smallest unit of sentence composition and the smallest unit that we are aware of when we consciously try to create sentences. To put things together, we can define a word as the smallest of the linguistic units which can occur on its own in speech and writing ( Richards et al., 1985:311). For example, care, careless and careful are smallest linguistic units in speech and writing, so they are all words. Some people may argue that careless and careful are not smallest, because careless is made up of care and- less, and careful is made up of care and - ful. However, in speech and writing we do not use care and- less, or care and- ful separately, as- less or- ful cannot stand on its own. Some people may also argue that many words like the, so, of can hardly occur on their own, because they are almost always used with other words in speech and writing. This is true, but in this definition,“ occur on its own” means that distinct from units like- less and- ful, a word is an individual

Chapter 3 31

Morphology: The Word Structure of Language

unit of meaning that is not bound to any other units. This definition may not be perfect, but it captures the basic characteristics of a word. That is, a word is a sound- meaning unit and a fundamental unit of speech and writing.

In discussing words, the following terms are also frequently used, and sometimes distinctions are made as follows:

Lexis: the vocabulary of a language, in contrast to its grammar or syntax.

Lexicon: the set of all the words and idioms of any language, often used interchangeably with vocabulary.

Lexeme: the smallest unit in the meaning system of a language, an abstract unit that remains constant. For example, give is the lexeme of its variants gave, given, giving.

· Vocabulary: A complete inventory of the words in a language.

Words can be classified into content words and function words. Content words are words which refer to a thing, quality, state, or action and which have stable lexical meaning or semantic content. They mainly include nouns, verbs, adjectives, and adverbs. Content words are also called lexical words or full words. Content words are also“ open- class” words, because new words can be added to these classes.

Function words are words whose role is primarily to express grammatical relationships and such words have little meaning on their own except their grammatical meaning. They mainly include conjunctions, prepositions, auxiliaries and articles. Function words are also called form words, empty words, grammatical words, structural words or structure words. Function words are“ closed- class” words because new words are not usually added to these classes.

3.3 The Structure of Words

Although words are the smallest of the linguistic units in speech and writing, they have an internal structure. That is, they are said to be made up of meaningful units of their own, as exemplified in the following:

careful→care + ful

sadness→sad + ness

realize→real + ize

disobey→dis+ obey

unthinkable→un + think + able

These examples show that a word can be analyzed into minimal units of meaning. In disobey, there are two minimal units of meaning, one being dis-( meaning“ not”), the other being obey.

32 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

When words are used in sentences, the word- forms such as works, worked, working can also be analyzed into minimal units of meaning. In this case, the word- forms consist of one element work, and a number of other elements such as-s,- ed,- ing, which indicate various grammatical meanings.

Thus, a word can be analyzed into the most elemental units of meaning. Some words are composed of one unit of meaning, like help, the, happy, apple. Some words consist of two or more than two units of meaning. For example, happily contains two minimal units of meaning, and ungentlemanliness contains five minimal units of meaning. The minimal unit of meaning is traditionally called morpheme. Thus, it can be said that a word is composed of at least one morpheme and it may be composed of more than one. Words that are formed by one morpheme only and cannot be analyzed into parts are called opaque words. Words that consist of more than one morpheme and can be segmented into parts are called transparent words.

3.4 Morpheme, Morph and Allomorph

A morpheme is the smallest meaningful linguistic unit that cannot be further segmented. For example, the word morpheme itself contains two morphemes, morph-, meaning“ form”, and - eme, meaning“ unit”. We sometimes can predict the meanings of a word by identifying its morphemes.

A morpheme is an abstract concept. The concrete form of a morpheme is called morph. That is to say, morphs are the actual forms used to realize morphemes. A morpheme may take different shapes or forms, which are called allomorphs. Thus, an allomorph is any of the variant forms of a morpheme.

Why is there an abstract- concrete distinction here? One possible answer is that such a distinction is necessitated by morphological description. For example, the following group of words have one grammatical meaning in common: dogs, buses, sheep, men, geese, children, phenomena. They are all plural forms expressing the grammatical meaning of“ more than one...” The plural meaning is expressed by -s in dogs, but by - es in buses. Things become more complicated when we look at the other words: what makes sheep the plural of sheep, or geese the plural of goose? Can we say that the plural meaning in all these words is expressed by different morphemes? Certainly not. There is only one morpheme involved here, and we may call it the morpheme“ plural”. It can be attached to a number of lexical morphemes to produce structures like“ dog+ plural”,“ sheep + plural”,“ goose + plural”, and so on. The morpheme“ plural” takes different forms in actual realization, called morphs, and they are all allomorphs of the one morpheme. We can say that“ dog+ plural”, “ sheep + plural”, “ goose + plural” are abstract analyses at

Chapter 3 33

Morphology: The Word Structure of Language

the morpheme- level, and are realized as dogs, sheep, geese at the morph- level( Yule, 2000: 79). This two- level analysis makes morphological description much easier, but very often the term“ morpheme” is used for convenience when we actually refer to a morph. So many people refer to forms like -s and - ed, which are morphs, as morphemes.

3.5 Classification of Morphemes

3.5.1 Free Morphemes and Bound Morphemes

A free morpheme is a morpheme which can stand by itself as an independent word, e. g. water, child, attack, and berry. A bound morpheme is a morpheme which cannot stand on its own as a word, but which is typically attached to another form, e. g. - dom in freedom,- hood in childhood, - ship in friendship and un in undo. On the morphemic level, words can be classified into simple words, compound words and complex words according to the number and type of morphemes words contain:

· A simple word consists of a single morpheme ( which is certainly a free morpheme), e. g. hill, walk, great, element.

· A compound word is composed of two or more free morphemes, e. g. blackboard, sunset, headstrong, forget- me- not.

A complex word consists either of a free morpheme together with one or more than one bound morpheme, or of two bound morphemes, e. g. cats, careful, unfriendly, contain, conceive, prelude.

3.5.2 Roots and Affixes

A root is the basic unchangeable part of a word, and it conveys the main lexical meaning of the word. It is the part of a word remaining when all affixes have been removed. A root can be a free morpheme, e. g. work in worker, think in unthinkable, or a bound morpheme, e. g. ceive in perceive, tain in retain, vit in vital.

An affix is a bound morpheme that is used only when added to another word or morpheme. It is used to mark the grammatical function of a word or create a new word. Affixes can be further classified into inflectional morphemes and derivational morphemes.

3.5.3 Inflectional Morphemes and Derivational Morphemes

When a word appears in a variety of forms depending on its grammatical role in the sentence, we say that it inflects or undergoes inflection ( Radford et

34 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

al., 2000: 154). Thus, an inflectional morpheme is an affix that indicates aspects of the grammatical function of a word, such as -s for plurality of countable nouns, - ed for past tense of verbs, - er and- est for the comparative and superlative degrees of adjectives or adverbs. An inflectional morpheme does not form a new word when it is added to another word, nor does it change the part of speech of the word to which it is affixed ( with a few exceptions like- ing and - ed, as in the learning of a language, and the golden- haired woman). The part of a word to which an inflectional affix is added is called a stem, e. g. talk in talked, farmer in farmers, classmate in classmates.

A derivational morpheme is one that is used to form a new word. For example, - ment is added to move and the new word movement is derived. English is a language rich in derivational morphemes. Some derivational morphemes appear before another form, while others after another form. Thus, derivational morphemes can be classified into prefixes and suffixes according to their positions in the derived words. In addition, some derivational morphemes produce words of a different class from the word to which they are affixed, while others do not. All these properties of the derivational morpheme will be illustrated in the discussion of word formation.

On the basis of what has been discussed above, we can make a classification of morphemes as shown in the following diagram:

3.6 Word Formation Processes

With the development of society, science and technology, new words constantly emerge and many new words gain popularity and find their way into the English language. A new word or expression is often referred to as a neologism, which can be a coinage, i. e. a word or phrase that has been recently invented, or a word used with a new meaning. For example, glass ceiling and laptop are recent coinages, while mouse and memory used in computer science are words with new meanings.

Word formation is an area of derivational morphology which studies the processes of word building and the rules governing the formation of new words. New words are formed according to morphological rules which determine how morphemes combine to form new words. In English, there are

Chapter 3 35

Morphology: The Word Structure of Language

major word formation processes which are productive, such as derivation, compounding and conversion, and minor word formation processes which are less productive, such as blending, clipping, back- formation and acronymy.

3.6.1 Derivation

Derivation, also called affixation, is a process of word formation by which a new word is created by adding an affix to a base. A base is that part of a word to which a derivational affix is added. Some people regard the term base as an alternative to root or stem, although they are not always the same. For example, in modernize, the base form is modern, and in modernization, the base form is modernize.

Derivational morphemes can be classified into prefixes and suffixes according to their positions in the derived words. A morpheme added to the left of a base is a prefix, e. g. re - in reopen, en- in enable. Thus, the formation of a new word by adding a prefix to the base is called prefixation. A morpheme added to the right of a base is a suffix, e. g. - ize in civilize, - ness in illness. Thus, the formation of a new word by adding a suffix to the base is called suffixation.

Prefixation usually produces words of the same class as the base, while suffixation very often produce words of a different class from the base, as can be seen in the examples below.

36 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

续表

From these examples, we can see that derivation has several properties. First, most words derived through suffixation have a different part of speech from the base, while most words derived through prefixation have the same part of speech as the base. Second, phonological factors sometimes work simultaneously with the derivational process. One example is that the suffix- en often goes with the monosyllabic adjectives which only have an obstruent at the end to derive words like darken, widen, deepen, and broaden. Another example is that the word stress is changed with some English suffixes like- itarian “human → humani' tarian”, - ation“ con'dense → conden' sation”,- cal“ psy'chology→ psycho' logical”. Third, some derivational processes may involve a change in the lexical meaning. For example, negative affixes like dis-, un-, in-, non- and - less can be used to form words with a negative meaning.

3.6.2 Compounding

Compounding is a word formation process by means of joining two or more words to form a new word. Words formed in this way are called compounds or compound words, e. g. greenhouse, daybreak, moonlight, round- the- clock. Words can be combined in various ways to form compounds, as shown in the following examples:

Chapter 3 37

Morphology: The Word Structure of Language

There are primarily three types of compound words: noun compounds, verb compounds, and adjective compounds. The word class of a compound can be ( but not always) predicted according to the following principles: (a) When the two words belong to the same class, the compound will belong to this class: noun + noun- masterpiece, paper- clip, end- product; adjective + adjective- icycold, gray- brown, red- hot. (b) When the two words fall into different classes, the class of the compound is generally determined by the class of the second or final word: verb+ noun- driveway, pick- pocket; noun + adjectives- snow- white, life- long.

Compounds differ from phrases in terms of stress patterns. The primary stress usually falls on the first part of a compound. Thus, a ' hot- line with a stress on hot refers to a frequent communication by phones, while a hot line with a stress on line can refer to a line, metal, plastic, or cotton that is hot, rather than cold. Pinker (1994: 133) provides us with a simple way to tell whether something is a compound word or a phrase: compounds generally have stress on the first word, phrases on the second. A dark room( phrase) is any room that is dark, but a dark room( compound word) is where photographers work.

A compound is structurally stable and is used as a single unit. For example, greenhouse cannot be modified by adverbs. Although a very green house can be regarded as a phrase, * a very greenhouse is unacceptable as greenhouse is a compound noun. Nor can we use the comparative and superlative forms of green in the compound. Thus, * a greenerhouse does not exist as a compound, and a greener/ greenest house becomes a phrase. This is also the case with other compounds like fine art and loud speaker.

As a chunk expressing a single idea, the meaning of a compound cannot be inferred from the individual component of the compound. A no- nonsense teacher does not refer to a teacher who does not talk nonsense, but one who is reliable, dependable, and strict with students. And a songbird may refer to a woman

38 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

singer, not a bird that sings. Again, a sand piper is a small wading bird having a slender bill and piping call, with a sense that has little to do with sand beach and a musical instrument. Similar analysis can be made of compounds like dog days and blue blood.

3.6.3 Conversion

Conversion is word formation by means of converting words of one class into another. As conversion involves the shift of word class without the addition of an affix, it is also variously called zero derivation, root formation, functional shift, or derivation by zero suffix. This process often occurs between a number of word classes.

a. noun to verb conversion

can: to can apples

shelter: to shelter an orphan

butter: to butter bread

weed: to weed the garden

elbow: to elbow the door open

mother: to mother a child

b. verb to noun conversion

doubt: to have no doubt

walk: to take a walk

bet: to lay a bet

cover: to find a cover

divide: a divide between two rivers

c. adjective to noun conversion

poor: the poor

wounded: the wounded

native: a returned native

d. adjective to verb conversion

idle: to idle away

dim: to dim out

free: to free the slaves

warm: to warm the house

There are also other types of conversion as shown in the following examples: to up a hill, to face the ups and downs of life, the then president. One thing worthy of note is that conversion may bring about shift of word stress, as shown in the following examples: per' mit( verb)-' permit( noun), re' cord( verb)-' record ( noun), ex' port( verb)-' export ( noun).

Chapter 3 39

Morphology: The Word Structure of Language

3.6.4 Blending

Blending is the formation of new words by combining parts of two words or a word plus a part of another word. Words formed in this way are called blends. Blending involves parts of two words called“ head” and“ tail”. Head is the first part of a word, and tail is the final part of a word. Thus, blending can have the following types according to different kinds of combination.

a. head+ tail: motor + hotel = motel

smoke + fog = smog

breakfast + lunch = brunch

b head + head: situation + comedy = sitcom

communications + satellite = comsat

formula + translator = Fortran

c. head + word: medical + care = medicare

Europe + Asia = Euroasia

automobile + camp = autocamp

d. word+ head: sky + laboratory = skylab

mass + culture = masscult

laser + computer = lasercomp

e. word+ tail: lunar + astronaut = lunarnaut

work + welfare = workfare

tour+ automobile = tourmobile

3.6.5 Backformation

Backformation is a process of word formation by the deletion of a supposed affix. It reverses the process of derivation in the sense that it forms new words by removing rather than adding an affix. For example, edit is from editor, b urgle from burglar, peddle from peddler, emote from emotion, free associate from free association, reminisc e from reminiscence, g rue from gruesome, gangle from gangling, and attrit from attrition.

3.6.6 Abbreviation or Shortening

Abbreviation or shortening is a word formation process by making a word shorter. It reflects the tendency of simplification in language use, as can be exemplified by clipping and acronymy

Clipping is a process of maintaining a single syllable, while removing the rest in a polysyllabic word. For example, fax is shortened from facsimile, ad or advert from advertisement, ve t from veterinarian.

Acronymy is the process of forming new words by joining the initial letters of words. Words formed in this way are called initialisms or acronyms,

40 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

depending on the pronunciation of the newly formed words. Initialisms are words formed from the initial letters of words and pronounced as letters. For example, BBC is from British Broadcasting Corporation, POW from prisoner of war, CEO from chief executive officer, WTO from World Trade Organization. Acronyms are words formed from the initial letters of words and pronounced as words. For example, NATO is from North Atlantic Treaty Organization, OPEC from Organization of Petroleum Export Countries, laser from light amplification by stimulated emission of radiation.

3.7 Summary

Morphology studies the structure of words in language and it can be broadly classified into derivational morphology and inflectional morphology. Derivational morphology is concerned with how a word is formed and the word- formation processes. Inflectional morphology is concerned with how the form of a word may be affected when it is used with other words or in a sentence. That is to say, words vary with their grammatical roles in sentence formation. As Crystal (1997: 90) puts it, inflectional morphology studies the way in which words vary ( or“ inflect”) in order to express grammatical contrasts in sentences. A word inflects or undergoes inflection when it appears in a variety of forms depending on its grammatical role in the sentence. Carstairs-McCarthy(2002: 30) takes inflection as a process of dealing with the inflected forms of words, that is, the kind of variation that words exhibit on the basis of their grammatical context. In some languages like Greek, Latin and French, inflection plays a major role in sentence formation, because such languages rely on overt inflectional forms as markers of syntact ic relations. Modern English is not an inflected language, but there are still inflectional morphemes indicating the grammatical function of a word. According to Yule(2000: 77), English has eight inflectional morphemes, illustrated in the following:

Let me tell you about Jim's two sisters.

One likes to have fun and is always laughing.

The other liked to study and has always taken things seriously.

One is the loudest person in the house and the other is quieter than a mouse.

It can be noticed that inflection in English takes place in nouns, verbs and adjectives. Nouns inflect for possessive(-'s) and plural (-s). Verbs inflect for third person present singular(-s), present participle(- ing), past tense(- ed) and past participle (- en). Adjectives inflect for comparative (- er) and superlative (- est) degrees. Thus, word forms like Jim’s, sisters, likes, laughing, liked, taken, loudest and quieter are grammatically conditioned

Chapter 3 41

Morphology: The Word Structure of Language

variants of the basic word forms, that is, the stems.

Further Readings

Adams, V. 1973. An Introduction to Modern English Word- Formation. London: Longman. Carstairs-McCarthy, A. 1992. Current Morphology. London: Routledge.

Carstairs-McCarthy, A. 2002. An Introduction to English Morphology: Words and Their Structure. Edinburgh: Edinburgh University Press.

Matthews, P. H. 2000. Morphology( Second Edition). 北京:外语教学与研究出版社. Radford, A. et al. 2000. Linguistics: An introduction. 北京:外语教学与研究出版社.

Questions and Exercises

1. Define the following terms.

content words function words opa que words

transparent words morpheme allomorph

morph neologism word formation

derivation compounding conversion

backformation clipping blending

eponym affix acronymy

2. How is morpheme classified?

3. Here is a list of words:

(a) aftershock (e) survivor (i) deoxyribonucleoprotein

(b) tower (f) bedridden(i) uncopyrightable

(c) original (g) antirehabilitationism(k) antidisestablishmentarianism

(d) government (h) ninths(1) floccinaucinihilipilificationalizationalize

(1) How many morphemes are there in each word?

(2) Which of these morphemes are free morphemes?

4. Illustrate the properties of derivation with examples?

5. In English, the prefix un- can be added to some adjectives. Consider the possibility of attaching it to each of the following adjectives and then answer the questions.

(a) bad (b) interesting (c) inhabitable (d) glad (e) important

(f) good (g) sympathetic (h) strong (i) reliable

(1) Which words can take the prefix un-?

(2) What does the prefix un- mean in the derived words?

(3) How does this un- differ in meaning from the un- in unbutton, undo, and unfasten?

6. Explain how stress placement can change the meanings of the following words or expressions in Column A and Column B.

Column A Column B

a black ' board a ' blackboard

a white ' house The ' White House

a silk ' worm a ' silkworm

a toy ' factory a ' toy factory

a dark ' room a dark room

7. Illustrate different types of conversion with examples.

8. Illustrate different types of blending with examples.

9. How is the form of a word affected when it is used with other words or in a sentence?

4.1 Introduction

The word“ syntax” is from the Ancient Greek sýntaxis, which literally means“ arrangement” or“ setting out together”. In linguistics, syntax is the study of how words combine to form sentences and the rules which govern the formation of sentences. Most languages have a finite number of basic words, but these words can be put together to make an infinite number of sentences. This is because there are rules and patterns that can be used in a recursive way to create new sentences. Syntax studies the rules governing the way words are combined to form sentences in a language. Generally, there are four major approaches to syntax, namely, the traditional approach, the structural approach, the transformational- generative approach, and the functional approach.

What is the relationship between s yntax and grammar? According to Lyons(1968), the term“ grammar” goes back( through French and Latin) to a Greek word which may be translated as“ the art of writing”. Quite early in the history of Greek scholarship this word acquired a much wider sense and came to embrace the whole study of language. More recently, the term“ grammar” has developed a narrower interpretation: it tends to be restricted to that part of the analysis of language which was handled in classical grammar under the headings of inflection and syntax. According to the traditional distinction, inflection deals with the internal structure of words, and syntax accounts for the way in which words combine to form sentences. Thus, in a restricted sense, grammar refers to a level of st ructural organization which can be studied independently of phonology and semantics, and generally divided into the branches of syntax and morphology. In this sense, grammar is the study of the way words, and their component parts, combine to form sentences. This is the usual popular interpretation of the term“ grammar”, which is contrasted with a general conception of the subject, where grammar is seen as the entire system of structural relationships in a language, and thus subsumes phonology and semantics as well as syntax.

Chapter 4 43

Syntax: The Sentence Structure of Language

4.2 Sentence Structure

4.2.1 Definition of Sentence

Most syntact ic studies have focused on sentence structure, for this is where the most important grammatical relations are expressed.

In the traditional and popular view, a sentence is“a series of words in connected speech or writing, forming the grammatically complete expression of a single thought” ( Matthews, 1981). Modern s yntax avoids this semantic definition based on thoughts or ideas, because of the difficulties involved in saying what“ thoughts” are. For example, an apple can express a thought, but it would not be considered as a complete sentence; He came late, because he overslept is one sentence, but it could easily be analyzed as two thoughts.

Most linguistic definitions of sentence show the influence of American structuralism: a sentence is the largest structural unit to which syntact ic rules apply. That is, a sentence is the maximal unit of syntax,“ an independent linguistic form, not included by virtue of any grammatical construction in any larger linguistic form”( Bloomfield, 1933). This definition also has its problems. In particular, it does not allow for elliptical sentences such as To town as the answer to Where are you going? and minor sentences such as Help!, All aboard!, Happy Birthday! In reaction to the problem concerning the notion of completeness, some linguists classify sentences into two types: complete and incomplete. Other linguists propose to make a systematic distinction between sentence and utterance: se ntence is a theoretical unit, defined by a grammar, while utterance is a physical unit, a matter of speech production, or performance. In this view, utterances can be analyzed in terms of sentence, but utterances do not“ consist of” sentences.

Most analysts agree on the need to recognize a functional classification of sentences into statement, question, command and exclamatory types. Although sentences may be incomplete, normally a sentence consists of at least one subject and its predicate. Most analyses also recognize the classification of“ sentence patterns” into simple versus complex or compound types, i. e. consisting of one subject- predicate unit, as opposed to more than one.

4.2.2 The Linear Structure of Sentence

The linear structure of sentence is concerned with the word order of sentences: the sequence in which grammatical elements such as subject, verb, and object occur in sentences. Sentences are not just strings of words which occur in a random order. As Fromkin and Rodman(1983) put it, sentences are

44 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

more than merely words placed one after another like beads on a string. There are syntact ic rules determining the correct order of words in a sentence. These syntact ic rules account for how different parts of a sentence are related, that is,“ who” does“ what” to“ whom”. These are grammatical relations of a sentence and they reveal how each part of the sentence functions grammatically, or syntactically. Grammatical relations are usually referred to as“ subject of” and“ object of”, etc. In It tastes nice the relationship of it to tastes is that of a subject to a predicator. The pronoun is the subject of the verb. The relationship of tastes to nice is that of a predicator to its complement. The adjective is the complement of the verb. This sentence can then be said to have a“ subject- predicator- complement” construction, whose elements ( subject, predicator, complement ) are successive functions established by the individual grammatical relations.

Different models of analysis use different terms in characterizing the linear structure of sentence. For example, generative grammar uses“NP + VP”, and systemic- functional grammar uses terms like“ actor- process- goal”.

4.2.3 The Hierarchical Structure of Sentence

Sentences demonstrate a linear structure because words of a sentence are arranged one after another in a sequence by means of grammatical relations. However, grammatical relations can also be considered as parts of a larger unit. For example, in It tastes nice the pronoun is the“ subject of” the verb, but it can also be described as“ the subject” within the clause or sentence as a whole. In the same unit, the adjective is at once the“ complement of” the verb( relation of part to part) and also“ the complement” within the clause( relation of part to whole). The linear structure“NP + VP” can also be regarded as a larger unit made up of the smaller units NP and VP. In other words, a sentence can be analyzed into constituents. Conversely, constituents at different levels can combine to form increasingly larger units.

4.3 The Traditional Approach

Traditionally, syntax refers to the branch of grammar dealing with the ways in which words, with or without appropriate inflections, are arranged to show connections of meaning within the sentence ( Matthews, 1981). Often known as traditional grammar, the traditional approach refers to the range of attitudes and methods found in the prelinguistic era of language study. Traditional grammar goes back to Greece of the fifth century before Christ, and includes the work of classical Greek and Roman grammarians, Renaissance grammarians, and eighteenth- century prescriptive grammarians. This approach

Chapter 4 45

Syntax: The Sentence Structure of Language

emphasizes such matters as correctness, linguistic purism, literary excellence, the use of Latin models and the priority of written language.

The traditional grammar is essentially a grammar of prescription. The term“ prescriptive” is used by linguists to characterize any approach which attempts to lay down rules of correctness as to how language should be used. Using such criteria as purity, logic, history or literary excellence, prescriptivism aims to preserve imagined standards by insisting on norms of usage and criticizing violations of these norms. Prescriptive rules for English sentences, for example, mainly come from the following sources:

· Greek and Latin, which are regarded as models of linguistic excellence. According to the Latin standards, for example, you should use It is I or He runs faster than I and not It is me or He runs faster than me; You are not allowed to split an infinitive as in He wants to quickly finish the work, nor are you allowed to end a sentence with a preposition as in What did you do this for?

The written language, which is more careful, prestigious and permanent than speech. You should say Whom did you meet? instead of Who did you meet?

· Logic, which means that language should follow the principles of logic. You should say I didn't eat anything or I ate nothing and not I didn't eat nothing because double negatives make a positive.

Modern linguistics owes a large debt to traditional grammar, because many basic concepts used by modern linguists can be traced back to traditional grammar. Modern linguistics is especially indebted to traditional grammar for the following:

· The modern notion of sentence, which is typically defined in traditional grammar as an independent group of words expressing a complete thought. It is customary in traditional grammar to begin with notional definitions of the sentence and its components.

· Grammatical category, which is a class or group of items which fulfill the same or similar functions or share a common set of grammatical properties in a language. Traditional grammar provided modern linguistics with a number of grammatical categories including the parts of speech, number, person, tense, voice, gender, subject and predicate.

The notion of concord or agreement, which is associated with the requirement that the forms of two or more words in a syntactic relationship should agree, or be in concord, with each other in terms of some grammatical categories. For example, in the sentence The boy speaks good English, the verb speaks agrees with the noun boy. The

46 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

agreement is based on the grammatical categories of number and person. That is, the noun boy is third person singular, and the verb likes agrees with the noun. Agreement can also be found in noun phrases, so we say a book and this book, but three books, these books and some books.

· Paradigm analysis, which is the analysis of the different forms that a word can have in a grammatical system. A paradigm is a set or list of linguistic elements showing the forms which a word can have. Paradigms typically show a word's inflections. For example, in Latin:

singular plural

amō“I love” amāmus “ we love”

amas“ You love” amatis“ you love”

amat“ he, she, it loves” amant“ they love”

To some extent, the terminology and analytical procedures of traditional grammar provide a useful framework for describing the outward structure of sentences and continues to serve as the point of departure for all schools of grammatical analysis. However, traditional grammar differs from modern linguistics in some essential aspects ( See Chapter 1).

4.4 The Structural Approach

The structural approach to the analysis of language was started by the Swiss linguist Ferdinand de Saussure,“ the father of modern linguistics”, in the beginning of the twentieth century. The structural approach regards linguistic units as interrelated with each other in a structure.

While traditional grammar is a grammar of prescription, the structural approach analyses the structure of different languages through description. The descriptive approach focuses on the actual samples of language and attempts to describe the regular structure as it is used, not according to some view of how it should be used.

4.4.1 Immediate Constituent Analysis

The structural approach is most closely associated with American structuralism from the 1920s to the late 1950s( See Chapter 1). The technique frequently used in the analysis of the structure of sentences was immediate constituent analysis (IC analysis), which is a method of analyzing sentences into their component parts. This approach works through the different levels of structure within a se ntence in a series of steps. At each level, a construction is divided into its major constituents, and the process continues until no further division can be made. In general, the division is binary. The two parts that are

Chapter 4 47

Syntax: The Sentence Structure of Language

yielded after each cut are the immediate constituents, and the smallest grammatical units obtained through segmentation are ultimate constituents. To sum up, immediate constituent analysis is the segmentation of a sentence into immediate constituents by using binary cuttings until obtaining its ultimate constituents. For example, to make an IC analysis of the sentence The horse kicked the dog, we carry out the following steps:

(a) Identify the two major constituents, the horse and kicked the dog.

(b) Divide the next- biggest constituent into two, viz. kicked the dog into kicked and the dog.

(c) Continue dividing constituents into two until we can go no further, viz. the horse into the and horse, the dog into the and dog.

The order of segmentation can be summarized using lines or brackets. If the first cut is symbolized by a single vertical line, the second cut by two lines, and so on, the sentence would look like this:

The||| horse | kicked || the ||| dog.

However, a much clearer way of representing constituent structure is through the use of tree diagrams:

Through IC analysis, the internal structure of a construction can be demonstrated clearly. Furthermore, IC analysis can help account for ambiguities. For example, the construction more modern music is ambiguous because it can be analyzed differently in terms of its constituent structure, as demonstrated below:

4.4.2 Endocentric and Exocentric Constructions

A construction is any syntact ic structure such as a sentence or phrase. There are generally two types of constructions according to their structure: endocentric and exocentric.

An endocentric construction is a group of syntactically related words where one of the words is functionally equivalent to the group as a whole. As there is a definable“ center” or head inside the group, which has the same distribution as the whole, an endocentric construction is also called a headed construction.

48 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

For example, the fat lady in the park is an endocentric construction, because, there is one central element ( the head lady) which is distributionally( functionally) equivalent to the phrase as a whole.

Endocentric constructions can be divided into subordinate constructions and coordinate constructions. Subordinate constructions are endocentric. constructions in which there is only one head, with the head being dominant and the other constructions dependent. For example, the trees in the park, the book we are discussing, run away quickly, afraid of the dark. Coordinate constructions are endocentric constructions in which there are more than one head, and they are of equal syntactic status, with no one dependent on the other. In other words, both are capable of serving as the head. For example, man and woman, hit and run.

An exocentric construction is a group of syntactically related words where none of the words is functionally equivalent to the group as a whole. That is to say, there is no definable“ center” or head inside the group. The following are some typical examples of exocentric constructions:

· Basic sentence structures: The man fell

· Subordinate clauses: If he is going

· Prepositional phrases: on the table

· Verb + Object: kick the ball

· Verb + adjective: seemed angry.

4.5 The Transformational- generative Approach

4.5.1 The TG Model of Grammar

As introduced in Chapter 1, transformational- generative grammar is a theory of grammar proposed by the American linguist Noam Chomsky in his book Syntactic Structures (1957). The primary goal of TG grammar is to explain, with a system of rules, the knowledge which a native speaker of a language uses in forming grammatical sentences, and the ultimate goal is to provide a universal grammar for the description of all languages. Universal grammar(UG) is a theory of linguistics aiming to establish principles of grammar shared by all languages and thought to be innate to humans. The ultimate goal of universal grammar is to specify precisely the possible form of a human grammar, so as to provide a theory of the human language faculty and shed light on the nature of human mind. Chomsky views language as a system capable of producing an infinite number of sentences. The syntact ic component is one of the three major organizational units within a grammar ( the others being phonological and semantic), containing the phrase structure rules and

Chapter 4 49

Syntax: The Sentence Structure of Language

transformational rules for the generation of syntact ic structures.

TG grammar has undergone five stages of development since 1957. The Classical Theory(1950s) established the notion of generative grammar and the concepts of phrase structure rules and transformations. The Standard Theory(1960s) recognized the deep structure and surface structure in sentences, related by transformations, and introduced the semantic component at the level of deep structure in order to constrain the generation of ill- formed sentences. The Extended Standard Theory(1970s) recognized the role of surface structure in semantic interpretation and introduced the trace theory. The Government and Binding Theory(1980s), also called the principles and parameters theory, introduced the concepts of principles and parameters, including X- Bar Theory,θ- Theory, Case Theory, Binding Theory, Movement Theory. The Minimalist Program(1990s and after) is a theory of grammar whose core assumption is that grammars should be described in terms of the minimal set of theoretical and descriptive apparatus necessary.

Although many of the ideas and concepts in the early periods of TG grammar are no longer used in the recent framework, they serve as a good starting point for the understanding of the analytical procedures in TG grammar.

In the original TG model, grammar works in this way: phrase structure rules generate the basic sentence structures and sentences are formed through transformations. This means that each sentence is considered to have two levels of structure: deep structure and surface structure. Deep structure is the abstract syntact ic representation of a sentence, that is, an underlying level of st ructural organization displaying all the factors that govern how a sentence should be interpreted. Surface structure is the syntact ic structure of a sentence we actually articulate or hear. It is the final stage in the syntact ic derivation of a sentence. The original TG model of grammar can be summarized as follows: the phrase structure rules generate deep structures which are transformed into surface structures by applying a set of transformation al rules.

Phrase structure rules are also called PS rules, rewrite rules or categorical rules. They are rules for the generation of the deep structure of a sentence. Phrase structure rules not only present information about the constituent structure of a sentence, but also information about the constituents such as words and phrases. For example:

S→NP + VP

NP→Det + N

VP→ V + NP

Det→the, a, this, that…

N→ man, tree .. .

50 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

V→hit, walk ...

These rules mean that a sentence (S) can be analysed ( or rewritten) as consisting of a noun phrase(NP) and a verb phrase (VP); a noun phrase can be further rewritten as a determiner( Det) and a noun(N); a verb phrase can be further rewritten as a verb(V) and a noun phrase; a determiner can be rewritten as the or a, and so on. Thus, to form the deep structure for The man escaped, the following phrase structure rules are needed:

S→NP + VP

NP→Det + N

VP→T( ense) + V

Det→ the

N→ man

T→PAST

V→escape

The following tree diagram can be used to show the deep structure for the sentence The man escaped and how the rules are applied:

(1)

The deep structure is not an actual English sentence yet, and transformational rules will be applied to change the deep structure in to the surface structure The man escaped.

Apart from the phrase structure rules and transformational rules, there is also a phonological component which is responsible for the articulation of sentences at the level of surface structure.

4.5.2 Syntactic Structure

4.5.2.1 Phrase Structure

Words can be combined according to phrase structure rules to form phrases. For example, we can form the verb phrase (VP) speak English by merging the verb speak with the noun English. The head of the phrase is the verb speak, because it determines the grammatical category of the phrase. The phrase speak English is a projection (i. e. a phrasal expansion) of the verb speak. The merging operation can be illustrated in the following labeled tree diagram:

52 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

4.5.2.2 X- bar

According to the Binary Principle, phrases are formed by combining pairs of constituents. Many phrases are formed by merging a head (e. g. a noun, or verb, or preposition, or an infinitive) with a constituent following the head. The constituent which follows the head is called a complement. These various phrases named NP, VP, PP and IP, etc. can be given one name: XP, where X is a word category serving as the head of a phrase. Thus, the basic structure for an XP can be illustrated in Figure 4.1:

Thus, in the NP loss of face, of face is the complement of the head loss; in the PP at a meeting, a meeting is the complement of the head at.

Apart from the head and complement, there may also be a constituent positioned to the left of the head. This constituent is called a specifier, because it specifies relevant information of the head. It is easy to find such phrases in the English language as the following:

It is obvious that all the phrases above are made up of three elements: specifier + head + complement. However, this seems to contradict with the Binary Principle of phrase structure. So how is a complex phrase like government decision to ban guns formed? We might suppose that the head decision combines with its complement to ban guns to form an intermediate level called N- bar( also labeled as N’). The full NP would be formed by merging the specifier government with the N- bar:

Chapter 4 53

Syntax: The Sentence Structure of Language

(5)

As shown in (5), the N- bar decision to ban guns is an immediate level larger than the noun decision but smaller than the full noun phrase government decision to ban guns. This way of analysis applies to all phrases of the form specifier + head+ complement. Thus, X- bar is a level of structure formed by merging the head (X) with its complement. It is an immediate level or category larger than the head but smaller than the full XP. This can be formulated as follows and illustrated in Figure 4.2:

XP→ specifier + X’

X’→X( head) + complement

Figure 4.2: X- bar

The X- bar theory assumes that every phrase is headed, and hence endocentric. This view of phrase structure is quite different from the structural approach which makes a distinction between endocentric and exocentric structures.

4.5.2.3 Sentence Structure

It can be assumed that sentences are formed in the same way as phrases are formed, that is, by the binary operation of merging pairs of categories together. The head of a sentence is the inflectional category( shortened asINFL or I) in the sentence. By definition, INFL is a category comprising auxiliaries ( inflected for tense and agreement) and the infinitive partic le to( Note that the infinitive particle to functions in the same way as auxiliaries do). In a sentence, the head INFL takes an NP ( the subject) as its specifier

54 语言学甚础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

and a VP category as its complement, as illustrated in (6).

(6)

As a sentence is formed in the same way as does a phrase, a sentence can be regarded as an inflection phrase ( shortened as IP) with INFL as its head. Thus, the sentence in (6) is formed by merging the head ( the auxiliary can) with its complement( the VP speak English) to form the inflection phrase can speak English. According to the X- bar theory, can speak English is an intermediate inflection phrase in the sense that it is larger than the head and smaller than a full IP, so it is an I- bar (I’). Besides, it is an incomplete inflection phrase in the sense that it cannot be used as a sentence, because auxiliaries require a subject ( or specifier). The IP is finally formed by merging the noun John( both subject and specifier) with the I- bar.

4.5.2.4 Complementizer Phrase

It has been shown that a sentence can be regarded as an inflection phrase(IP) of the form subject + INFL + complement. In the English language, some sentences are introduced by complementizers such as that (e. g. John thought that it was raining), for( It is important for you to go) or if/ whether(I asked if you could help me). Thus, the framework of merging should be extended to explain this kind of structure.

A complementizer is a functional element which is typically used to introduce a complement clause. In terms of merging operations, a complementizer merges with a sentence(IP) to form a complementizer phrase(CP), with the complementizer(C) as its head, as shown in (7).

56 语言学基础教程( 第二版)

Linguistics: An Elementary Course Book Second Edition

whether one node occurs higher up in the structure than another. Thus, in(9), node A dominates all other nodes; node B only dominates nodes C and D; node E dominates nodes F, G, H and J; node G dominates nodes H and J. Nodes C,D, F, H and J are terminal nodes, so they dominate no other nodes.

A more complex structural relationship which can be used to account for the grammatical ity of sentences is c- command( or constituent command). C- command can be defined as follows:

αc-commandsβiff(= if and only if) α does not dominate β and every γ that dominates α dominatesβ. ( Chomsky, 1986: 8)

This can be illustrated in the following tree diagram in Figure 4.3.

Figure 4.3: C- command

C- command can be easily understood with the help of the train network analogy drawn by Radford(2002). If you think of α and β as different stations in a train network, we can say that α c-commandsβ if you can get from α toβ by taking a northbound train from α, getting off at the first stop, and then taking a southbound train to β( on a different line). We may use the tree diagram in(9) to further illustrate this structural relationship. We can say that c- commands D; B c- commands E, F,G, H and J; F c- commands G, H and J;H c- commands J.

We now return to (8) to see how c- command works. The pronoun(PRN) him is ambiguous in that it can either refer to John or someone else. Certainly it cannot refer to the determiner phrase(DP) John's father. If we look at the st ructural relation here, him is c- commanded by the DP John's father. It can be assumed that a pronoun like him, called a pronominal, cannot be c- commanded by an antecedent within the same sentence. As John does not c- command him, it can be the possible antecedent for the pronominal. However, if we change the pronominal him into the reflexive himself, it becomes unambiguous, and the reflexive only refers to John's father. In John's father doesn't like himself, the DP John's father c- commands the reflexive himself. Thus, different from pronominals, reflexives are c- commanded by an antecedent. Reflexives like himself, themselves, myself and reciprocals like each other and one another are called anaphors. They cannot be used to refer directly to an entity in the outside world, but rather must be bound by or refer to an antecedent elsewhere in the same phrase or sentence. Otherwise, the resulting sentence is ungrammatical. The binding of anaphors by their antecedents can be explained in terms of the c- command condition on binding:

A bound constituent (= anaphor) must be c- commanded by an

58 语言学基础教程(第二版)

Linguistics: An Elementary Course Book Second Edition

constituents together. Some sentences such as yes- no questions and wh- questions are formed through the syntact ic operation of movement. By definition, movement is an operation by which a word or phrase is moved from one position in a structure to another. In deriving a sentence, certain elements may be moved, and they may only be moved to certain positions.

4.5.3.1 Head Movement

In English, yes- no questions involve auxiliary inversion. That is, the auxiliary is traditionally said to move into the position in front of the subject. Thus, in Could you help me? the auxiliary could is said to have been inverted and have moved from the post- subject position into the pre- subject position.

In TG grammar, a yes- no question is formed by moving an inverted auxiliary from the head I position in IP into the head C position in CP. When movement occurs, the moved constituent leaves behind an empty trace( labeled as t) of itself in the position out of which it moves. Thus, the sentence Could you help me? has the structure in (13).

(13)

This type of movement in (13) involves movement of a word from the head position in one phrase into the head position in another phrase, and so is known as head movement ( or head- to- head movement).

When yes- no questions are formed fromstatements containing no overt auxiliary, the dummy auxiliary do is used. For example:

(14) a. Birds fly.

b. Do birds fly?

Since the head I is left empty in (14a), as shown in (15), the yes- no question in (14b) is formed by generating the auxiliary do in I and then raising it from I to C, as shown in (16).

Chapter 4 59

Syntax: The Sentence Structure of Language

4.5.3.2 Operator Movement

A yes- no question can be regarded as a CP consisting of a head C( filled by an inverted auxiliary) and an IP complement. However, another constituent can occur before the head C constituent, and form a question like (17):

(17) Which book can you read?

Example(17) is an interrogative often called a wh- question. Words like what and which are called interrogative operators or wh- operators, and expressions like which book are called operator phrases. What kind of movement operation has taken place in wh- questions?

In(17), the operator phrase which book functions as the complement of the verb read. That is, the operator phrase originally occupies the complement position after the verb, as shown in (18):

(18) You can read which book?

In forming the wh- question, two kinds of movement are involved: one is the head movement of the auxiliary can from the head I position in IP to the head C position in CP; the other is the movement of the operator phrase to a position before the inverted auxiliary can. As the inverted auxiliary occupies the head C position in CP and the position before the head is typically occupied by specifiers, it can be reasonably assumed that the operator phrase is moved into the specifier position in CP. Thus, sentence(17) will be derived as in(19).

Please note that according to the X- bar theory, the head merges with its complement to form an X- bar, and an XP is formed by merging the specifier with the X- bar. In(19), the head C merges with its IP complement to form aC- bar(C’), which merges with the specifier DP( determiner phrase) to form