Luận án tiến sĩ Ngôn ngữ học: So sánh mối quan hệ logic-ngữ nghĩa trong tổ hợp cú tiếng Anh và tiếng Việt.

VIETNAM NATIONAL UNIVERSITY, HANOI

UNIVERSITY OF LANGUAGES AND INTERNATIONAL STUDIES

NGUYEN THI MINH TAM

LOGICO-SEMANTIC RELATIONSHIP IN ENGLISH AND VIETNAMESECLAUSE COMPLEXES

(SO SANH MOI QUAN HE LOGIC-NGỮ NGHĨA TRONG TO HỢP

Major: English LinguisticsCode: 62 22 15 01

DOCTORAL DISSERTATION OF ENGLISH

SUPERVISOR: PROF DR HOANG VAN VAN

HANOI -2013

TABLE OF CONTENTS

Aims of the study

Significant of the studyScope of the study

The research question

The research design, methodology, and dataThe organization of the study

CHAPTER 1: SYSTEMIC FUNCTIONAL LINGUISTICS AND THE

THEORETICAL FRAMEWORK OF LOGICO-SEMANTIC RELATIONSHIPIN CLAUSE COMPLEXES

1.1 Systemic functional theory

1.1.1 Historical context of the emergence of systemic functional theory: a brief

1.1.4 Review of related studies

1.1.4.1 An overview of studies in Vietnamese grammar1.1.4.2 SFL studies in other languages

1.2.1.2 What is a clause complex?

1.2.1.3 Where is the clause complex located in the overall linguisticsystem?

1.2.1.4 How is the clause complex organized?

1.2.2 Analytical frameworks of logico-semantic relations in clause complexes1.2.2.1 Projection in clause complexes

1.2.2.2 Expansion in clause complexes

nak BWW C2

50

CHAPTER 2: METHODOLOGY, DATA, AND GENERALFINDINGS

2.1 Research question restated2.2 Research design

2.3 The methodology

2.4 Methods of the study

2.5 Data collecting procedure

2.6.1 Describing the data

2.6.2 Comparing the two groups of data2.7.General findings

CHAPTER 3: FINDINGS AND DISCUSSION: PROJECTION IN ENGLISHAND VIETNAMESE CLAUSE COMPLEXES

3.1 Projection in English Clause Complexes

3.1.1 How is projection realized in English Clause Complexes3.1.1.1.

Projecting clauses

Projected clauses

Projecting and projected clauses paratactically relatedProjecting and projected clauses hypotactically related3.1.2 How does Projection work in English clause complexes

How does quoting work in English clause complexesHow does hypotactic reporting work in English clausecomplexes

How does paratactic reporting work in English clause complexesHow does projection facilitate other linguistic phenomena in

English clause complexes

3.1.3 Summary

3.2 Projection in Vietnamese clause complexes

3.2.1 How is projection realized in Vietnamese Clause Complexes3.2.1.1.

118

3.2.2.2 How does hypotactic reporting work in English clause complexes3.2.2.3 How does paratactic reporting work in English clause complexes

3.2.2.4 How does projection facilitate other linguistic phenomena in

English clause complexes

3.3.1.3 Projecting and projected clauses paratactically related3.3.1.4 Projecting and projected clauses hypotactically related

Operation of projection in English and Vietnamese clause complexes3.3.2.1 Operation of quoting in English and Vietnamese clause

4.1.1.3 Conjunctions in English clause complexes of expansion

4.1.1.4 Expanding clause and expanded clause paratactically related4.1.1.5 Expanding clause and expanded clause hypotactically related

How does expansion work in English clause complexes

4.1.2.1 How does elaboration work in English clause complexes4.1.2.2 How does extension work in English clause complexes

4.1.2.3 How does enhancement work in English clause complexes

4.1.2.4 How does expansion facilitate ellipsis in the clause complex

4.2 Expansion in Vietnamese clause complexes

4.2.1 How is expansion realized in Vietnamese clause complexes

156160160

4.2.1.1 Expanding clauses 1604.2.1.2 Expanded clauses 162

4.2.1.3 Conjunctions in Vietnamese clause complexes of 164expansion

4.2.1.4 Expanding clause and expanded clause paratactically 168related

4.2.1.5 Expanding clause and expanded clause hypotactically 170related

4.2.2._ How does expansion work in English clause complexes 1725.1.2.1 How does elaboration work in English clause complexes 1725.1.2.2 How does extension work in English clause complexes 1755.1.2.3 How does enhancement work in English clause complexes 1785.1.2.4 How does expansion facilitate ellipsis in the clause complex 1804.2.3 Summary 181

4.3.Comparing the logico-semantic relation of expansion in English and

Vietnamese clause complexes 1864.3.1 Comparing the realization of expansion in English and Vietnamese

clause complexes 1864.3.1.1 Expanding clauses 1864.3.1.2 Expanded clauses 1864.3.1.3 Conjunctions 187

4.3.1.4 Expanding clause and expanded clause paratactically

related 1884.3.1.5 Expanding clause and expanded clause hypotactically

complexes 190

4.3.2.4 Expansion and ellipsis in English and Vietnamese clause

complexes 191

CONCLUSION 192THESIS-RELATED PUBLICATION 202BIBLIOGRAPHY 203

SOURCES OF DATA 210

vii

Fig 6: The Rank scale

Fig.7 Delicacy in relation to the metafunctions of language and rank

Fig.8 Univariate and Multivariate Structure

Fig 9: Three lines of meanings in a clause — 1Fig 10: Three lines of meanings in a clause — 2Fig 11: Combining clauses into clause complexFig 12: The rank of the clause complex

Fig 13: The Location of Clause Complex in the Overall Linguistic System

Fig 14: Clauses in paratactic and hypotactic clause complexesFig 15: The Logico-semantic Relations

Fig.16: Positions of Projecting Clauses

Fig.17: Speech Functions of Projected Clauses

Fig 18: Clause Moods of Projected ClausesFig 19: Quoting vs Paratactic Reporting

Fig 20: Paratactic vs Hypotactic Projection

Fig 21: Projecting Processes and Mood of Projected Clause in Combination

46

Fig 22: Projecting Processes in Quoting and Hypotactic Reporting 47Fig 23: The framework of projection in clause complexes 49Fig 24: Possible Positions of Expanding Clauses 51Fig.25: An Analysis of a Clause complex of Expansion 51

Fig 26: Clause Moods of Expanded Clauses 52Fig 27: Expanding and Expanded Clauses Paratactically Related 53Fig 28: Meaning of some Verbal Hypotactic Expanding Markers 54Fig 29: Modes of Elaborating Relation 55Fig 30: Modes of Extending Relation 56

Fig 31: Modes of Enhancement 56Fig 32: The Framework of Expansion in Clause Complexes 58

Fig 33: Overview of the analysis process in SysFan 66

Fig 34: Chunking a group into clause complexes 68

Fig 35: Chunking a clause complex into clauses 69Fig 36: Seven possible levels of chunking clause complexes 69Fig 37: Labeling the clauses in analysis 70Fig.38: Distribution maps of clause complex relation types 71Fig 39: Complex Combination of Clauses — 1 124

Fig 40: Complex Combination of Clauses - 2 125Fig.41: The multi-function expanded clause in English 140Fig.42: A Multi-function Expanded Clause in Vietnamese 163

ix

œ,B,X 1, 2, 3,

hypotactic related clauses

paratactic related clauses

1 Rationale

Over the years, functional approach has had a significant impact on the study ofgrammar This theoretical approach tries to incorporate meaning, function, context, andgrammatical categories Funtionalism has been developed by many functional grammarianslike Dik (1978), Halliday (1985, 1994), Bloor (1994), Eggins (1994), Thompson (1996),Lock (1997), Martin (1997), VanValin and LaPolla (1997), Matthiessen (2002), and manyothers Different functional theories have emerged, representing a great range of theoretical

opinion, but the central theme of the functionalist theories is: functional motivation is an

alternative to innateness Linguistic theories which reject the syntactocentric or formal viewand adopt the communication and cognition perspective include Functional Grammar (Dik,1978, 1991), Systemic Functional Grammar (Halliday 1994), Role and Reference Grammar(RRG; VanValin and LaPolla 1997), Lexical Functional Grammar (LFG; Bresnan 2001), etc.Unlike the linguistic theory that is still the received tradition in school, functionalism takesthe resource perspective rather than the rule perspective; and it is designed to display theoverall system of grammar rather than only fragments It wishes to be a theory which is‘functional’ in at least three different, though interrelated senses:

i It takes a functional view on the nature of language;

ii It attaches primary importance to functional relations at different levels in the

language developed mainly by M.A.K Halliday in the UK and later in Australia Thisapproach has its origin in the main intellectual tradition of European linguistics thatdeveloped following the work of Saussure Its primary source was the work of J.R Firth(1957) and his colleagues in London, who defines “system” in systemic functional linguistics,as, in its technical sense, the “theoretical representation of paradigmatic relations, contrasted

1

with “structure” for syntagmatic relations”, and who characterizes systemic theory as the

theory in which the system takes priority, the most abstract representation at any level is inparadigmatic terms As well as other schools of thought in Europe such as glossemantics, thistheory also draws on American anthropological linguistics, and on traditional and modernlinguistics as developed in China While American-style linguistics evolved in the modeling

of the world’s languages, SFL was developed to address the need of language teaching and

learning A significant milestone in the development of this theory is when it was whollydeveloped in the work on the grammar of Chinese by Halliday (1956); and it has been used ineducational and computational contexts from an early stage.

With the primary goal of addressing the needs of language teaching / learning,systemic functional grammar (SFG), the theory of grammar in the light of SFL, lays anemphasis on the functions of language - what language is used for Although SFG is a theoryof grammar, it focuses more on the meaning, not the pure form of what is said like whatformalists have long been doing.

Since 1980, SFL has been expanded considerably in various directions; further studieshave been devoted to languages other than English, notably Chinese, French, Indonesian,Japanese and some other Asian and African languages It has not been of much popular use in

Vietnam studies of grammar In Vietnamese, the first functional studies that should be

counted are Tiếng Việt: Sơ Thảo Ngữ Pháp Chức Nang by Cao Xuân Hạo (1991), Ngữ PhápKinh Nghiệm của cú Tiếng Việt: Mô tả theo quan niệm chức năng hệ thống by Hoàng VănVân (2002), Ngữ Pháp Chức Năng Tiếng Việt: Câu trong Tiếng Việt, and Ngữ Pháp ChứcNăng Tiếng Việt: Ngữ Đoạn và Từ Loại by Cao Xuân Hạo (2007), and some PhD thesis by

Hoàng Văn Vân (1997), Thái Minh Đức (1998), and Đỗ Tuấn Minh (2007) These studies are

the first attempts to bring the contemporarily not widely applicable theoretical perspective

which has not yet been highlighted much in Vietnamese grammar studies.

SFG is a complex and comprehensive model, so in order to master all its concepts and

categories, it is necessary to receive the cooperation from many people, not from one

individual who can classify a further of a category only Hundreds of PhD theses have beenconducted studying different areas of grammar in different languages in the light of SFG Thestudies on Vietnamese grammar using the same theoretical framework were on the system oftransitivity, clauses, and thematic structures, leaving the area of clause combination anuntouched topic which the author of this paper therefore takes this area as the topic for herthesis In the light of SFG, a comparison between English and Vietnamese clause complexes

2

in this linguistic area not only helps shed light on the nature and realization of logico-semanticrelationship in both languages but also allows teachers and students of English as well as otherpeople who work in the field of linguistics to have deeper understanding of the language theyare dealing with, whether as material or tool For such reasons “Logico-semantic Relationship

in English and Vietnamese Clause Complexes” is chosen as a theme for this PhD disertation.

2 Aims of the study

The study aims at comparing the logico-semantic relationship in English andVietnamese clause complexes so as to find the similarities and differences in the realizationand operation of logico-semantic relationship in English and Vietnamese clause complexes.The two languages are thoroughly compared basing on meticulous criteria in the analyticalframework of logico-semantic relationship which is accumulated from different approaches

in systemic functional theory, as described in chapter 1.

3 The significance of the study

Because logico-semantic relationship between clauses in a clause complex in

English has been extensively investigated in the systemic functional model, but it has never

been investigated in Vietnamese from the view of SFL, this thesis, therefore, takes thelinguistic area of logico-semantic relationship as the subject for observing, describing, andcomparing Conducted as a comparative study in the light of SFL, the thesis is hoped to makesignificant contribution to building up the systemic functional model of Vietnamesegrammar, in comparison to the systemic functional model of English grammar.

4 Scope of the study

The thesis studies the logico-semantic relationships in English and Vietnamese

clause complexes In terms of theory, the scope of the study is narrowed down to observing,

analyzing, describing, and comparing the realization and operation of logico-semanticrelationship in clause complexes in the light of systemic functional linguistics only In terms

of data collection and data analysis, as the corpora of the dissertation are compiled fromwritten sources, the data collected are mostly written language, only a certain part of thelanguage collected in the corpora is the quotation of different types of spoken speech Thefeatures of logico-semantic relationship in spoken language are, therefore, not the focus ofthe findings.

5 The research question

With such aims as mentioned, the research question of the dissertation is:

What are the similarities and differences in the realization and operation of logico-semanticrelationship between English clause complexes and Vietnamese clause complexes?

6 The research design, methodology, and data

The dissertation is designed as a comparative study conducted in the methodology ofa corpus-based linguistic study.

To answer to this research question, the following methodology is applied:

(i) In the light of SFL, framing the theoretical frameworks of logico-semantic relations in

clause complexes by modifying the existing theories from the studies on the samelinguistic area in English and other languages;

(ii) Collecting texts to build up the corpora of English clause complexes and Vietnameseclause complexes as the source of real-life evidence for investigating and validatingthe theories, and comparing the findings.

(iii) Applying the modified framework to re-examine the logico-semantic relations in

English clause complexes, suggest any minor corrections and adjustments needed;

(iv) Applying the modified framework to investigate the logico-semantic relations in

Vietnamese clause complexes, building up the framework of logico-semantic relationsin Vietnamese clause complexes

The dissertation takes SFL as the theoretical perspective, which provides a more

comprehensive view on the functions of languages without rejecting the other theories onlanguage In this dissertation, therefore, the terms from traditional grammar are sometimesexploited so that a more detailed and comprehensive explanation on the feature being

described can be achieved.

Conducted in the methodology of corpus linguistic, the 300,000-word corpora of the

dissertation are compiled from 135 articles from 135 journals in English and Vietnamese,from which 2000 clause complexes (1000 in English and 1000 in Vietnamese) are randomly

selected for close observation The data analysis process is done manually with the assistance

in annotation from SysFan, a software for systemic functional annotation and statistics Dataanalysis therefore offers close observation of the clause complexes in the corpora, fromwhich findings on the realization and operation of logico-semantic relations in English and

Vietnamese clause complexes are re-examined, explored, and then compared In thediscussion of the findings, more than 200 clause complexes are retrieved from the corpora forillustration in the form of numbered examples with clause complex IDs As the wholecorpora with all the raw materials and analyzed data are too long, which can reach thenumber of 4000 pages when printed out, only the clause complexes used as examples in thedissertation, together with their detail analyses are then provided in the appendix of thedissertation Quantitative analysis is not used as the major instrument for analysis in thisdissertation; it is just used to sketch the general findings on the big trends of data andreinforce the description and comparison in the discussion.

7 The organization of the study

The dissertation is developed in four chapters:

Chapter 1- Systemic Functional Linguistics and the Theoretical Framework of semantic Relation in Clause Complexes- extensively reviews the historical context of theemergence of functionalism and, specifically, SFL, accumulate the characterizing features ofthis approach in language study, and related literature in Vietnamese language and the logico-

Logico-semantic relations in Vietnamese and other languages The chapter then explains and justifiesthe notion of clause complex and the concept of logico-semantic relationship, the relations ofprojection and expansion and their subtypes, then set up the analytical framework ofprojection and the analytical framework of expansion in clause complexes.

Chapter 2 - Methodology, Data, and General Findings - explains the steps of conducting thestudy, the procedures in collecting, analyzing, and exploiting data, and some general findingsbased on statistical data.

Chapter 3 - Findings and Discussion: Projection in English and Vietnamese ClauseComplexes — describes and then compares projection in English and Vietnamese ClauseComplexes.

Chapter 4 - Findings and Discussion: Expansion in English and Vietnamese ClauseComplexes - describes and then compares expansion in English and Vietnamese ClauseComplexes.

CHAPTER 1

SYSTEMIC FUNTIONAL LINGUISTICS AND THE THEORETICAL FRAMEWORK

OF LOGICO-SEMANTIC RELATIONSHIP IN CLAUSE COMPLEXES

1.1.Systemic Functional Approach

1.1.1 Historical Context of the Emergence of Functionalism: a Brief Overview

We use language to interact with one another, to construct and maintain ourinterpersonal relations and the social order that lies behind them; and in doing so weinterpret and represent the world for one another and for ourselves Language is a naturalpart of the process of living; it is also used to 'store' the experience built up in the course ofthat process, both personal and collective It is (among other things) a tool for representingknowledge or, to look at this in terms of language itself, for constructing meaning.

Since human beings began to be curious about the languages they speak and started

examining them, linguistics has been developed through many periods of time with

different approaches Grammatical analysis was developed very early in ancient India andin ancient Greece, a description of grammar also appeared The division of sentences intosubject and predicate by the great Greek scholar Aristotle (384-322BC) and the first stepof Dionysius Thrax, who produced the first complete grammar of Greek) in dividingwords into classes, which are now called parts of speech, are still recognized today.

After the Roman conquest of Greece in the mid 2" century BC, the Greek workwas much concerned and perceived by Roman scholars who then applied the sameanalysis to their language, Latin, constructing what is called traditional Graeco-Roman,also called traditional grammar, which has continued to be taught in European schoolsdown to present day However, in English-speaking countries, the teaching of Englishgrammar was largely discontinued in the 1960s Since then, a diversity of linguisticapproaches have been introduced by linguists, mostly in English-speaking countries,Germany, and France.

The initiative of universal grammar first emerged in the 17" century, when Frenchscholars, known as the Port-Royal Circle put together a universal grammar of French,

which is remarkably similar to Chomsky’s earliest 1950s version of transformationalgrammar The German polymath Wilhelm von Humboldt (1767 — 1835) likewise tried to

develop a universalist and philosophical approach to the study of languages He believes

6

that every language has its inner structure which determines its outer form and which is

the reflection of its speakers’ minds, which excited a great deal of attention but still failedto establish a continuing tradition.

This general trend of analyzing language structure seemed to be paused

temporarily when the study of language change and of prehistory of languages, calledhistorical linguistics, which came to be by far the most important way of studying

languages, was established towards the close of the 18" century.

Only towards the end of the 19” century did the non-historical study of language

structure begin to reassert itself This kind of work is now called general linguistics: thestudy of how languages are put together and how they work Ferdinand de Saussure (1857— 1913), originally trained as a historical linguist, made important contribution to the

development of general linguistics through his effort in describing how the European languages are put together and how they work Saussure’s viewpoint onlanguage is different from the other linguists in that: while most linguists take an atomisticapproach to language structure, perceiving a language as a collection of objects such asspeech sounds, words, and grammatical endings, he argued that language was bestregarded as a structured system of elements, in which the place of each element is definedchiefly by how it relates to other elements This approach quickly came to be calledstructuralism Today almost all work in linguistics is structuralist in Saussure’s generallinguistic sense.

Proto-Indo-The anthropological and cultural factors are then, step by step, added into linguists’

consideration in their linguistic analysis Fraz Boas (1858 — 1942), a student of Saussure,

later recognized as the father of American linguistics, initiated the idea in hisanthropological work that investigation of any culture required knowledge of its language.In turn, Boas’s student Edward Sapir was deeply interested in uncovering possiblerelations between language and culture and his student Benjamin Lee Whorf kept on withthe same cultural argument to develop the Sapir — Whorf hypothesis, or the linguisticrelativity hypothesis, which states that the structure of our language, to some extent,determines the way we perceive the world Such a view has fascinated a lot of linguists,anthropologists, and psychologists ever since, though the degree of validity has been muchdebated.

Another American linguist, Leonard Bloomfield, however, turns Americanlinguistics somewhat away from its anthropological and cultural connections, and toward amore tightly focused concentration on language structure in its own right The nextgeneration of American linguists took their inspiration mainly from Bloomfield The brandof linguistics these post-Bloomfieldian structuralists developed, known as Americanstructuralism, took no interest in the meaning or functions of words and utterances,

preferring to concentrate on linguistic forms alone.

In the 1950s, Noam Chomsky took the formalist methods from his teacher, Zellig

Harris (1909 — 1992), who grew American structuralism to something of an extreme withhis “unusual” algebraic analysis of language Chomsky then combined them with certainideas from mathematics The result was a strikingly new approach to the description and

study of language, and especially of sentence structure (syntax), which he called

generative grammar With this approach, he attempts to provide a fully explicit andmechanical statement of the rules governing the construction of sentences But he thenfurther developed instead a much more powerful kind of generative grammar, called

transformational grammar, or TG, which is the dominant formal theory which is also

referred to as the syntactocentric view of language In this light of TG, the point of a goodtheory of grammar is its power to tell us what is possible and impossible in the grammarsof human languages From this point of view, the form of sentence is an algebric system ofrules which operate largely independent of the meaning of the sentence Consequently,language is considered as an abstract object whose structure is to be studied independentlyof psycholinguistic, communicative, sociocultural and other considerations Chomskyargues his transformational grammar to be the best theory for describing grammarstructures of languages, but he and his colleagues actually have to modify their ideas

repeatedly to manage the complicatedly unstable data For these reasons, among others,

many linguists prefer to steer clear of what they see as excessive formalism in favour ofmore human-centered approaches which focus far more directly upon what people are

trying to do when they speak, and how they go about this task.

One such human-centered approach, preferred by many linguists is thefunctionalist approach, in the light of which linguists try to determine what purposes arebeing served by languages, and what linguistic forms are available to serve thosefunctions The emergence of functional approach is the milestone in the historical

development of world linguistics, with its separation into two trends: formalism and

1.1.2 Functional Approach — a General Description

As previously introduced, contemporary linguistic theories are usually divided into

two broad schools of thought which are labeled formal versus functional orientations.From a formal point of view, grammar of a language is a set of structural descriptions ofsentences where a full structural description determines the sound and meaning of alinguistic expression In this most dominant formal theory which is also referred to as thesyntactocentric view of language - syntax is the central aspect of language The

phonological and semantic aspects of language are derivative of and secondary tosyntactic structure.

Functionalists, however, believe that language must be studied in relation to itsrole in human communication In the light of functional approach, language is defined as asystem of communication The starting points for functionalists is the view that languageis first and foremost an instrument for communication between human beings and this factis central in explaining why languages are as they are Functionalists, typically, are thosewho argue for a higher degree of involvement of other domains (semantics, pragmatics,discourse, extra-linguistics exigencies deriving from the context of communication, etc.)in syntactic phenomena, and for hierarchies, gradients, and other non-categorial analyses.Indeed, functionalists believe that human beings do not communicate with each other in avacuum but rather in socioculturally defined activities and situations in which theparticipants take on socially defined roles and status Whereas adherents of thesyntactocentric paradigm view language as a potentially infinite set of structuraldescriptions independent of matters of use, functionalists take the very opposite approachin considering all aspects of the structural organization of language in the light of its rolein human social interaction.

The basic difference between functionalist and formalist linguistic frameworks isin where explanations are lodged, and what counts as an explanation Formal linguisticsgenerates explanations out of structure so that a structural category or relation, such ascommand or subjacency (Newmeyer 1999:476-477) can legitimately count as anexplanation for certain facts about various syntactic structures and constructions Mostcontemporary formal theories, certainly Generative Grammar in all its manifestations,

9

provide ontological grounding for these explanations in a hypothesized, but unexploredand unexplained, biologically-based universal language faculty Functionalists, in contrast,find explanations in function, and in recurrent diachronic processes which are for the mostpart function driven They see language as a tool, or, better, a set of tools, whose forms are

adapted to their functions, and thus can be explained only in terms of those functions.

Since functionalists assume a broader notion of language than formal theories, theextent of linguistic investigation is correspondingly wider Language function (what it is

used for) is often more important than language structure (how it is composed).

Generally speaking, the guiding principle of functionalism is the fact that the formof a sentence is determined by its meaning with reference to pragmatic and socialconsiderations As Newmeyer (2000) points out, the central theme of the functionalisttheories is: functional motivation is an alternative to innateness Linguistic theories whichreject the syntactocentric or formal view and adopt the communication and cognitionperspective include Functional Grammar (Dik 1978, 1991), Systemic Functional Grammar

(Halliday 1994), Role and Reference Grammar (RRG; VanValin and LaPolla 1997),

Lexical Functional Grammar (LFG; Bresnan 2001), etc Unlike the linguistic theory that isstill the received tradition in school, functionalism takes the resource perspective ratherthan the rule perspective; and it is designed to display the overall system of grammarrather than only fragments.

Different functional theories represent a great range of theoretical opinion, and bylisting them together no claim is made that they are in agreement on all major issues What

they have in common is basically a rejection of the syntactocentric view of formalists and

a recognition of the importance of the communicative factors, cognitive factors or both ingrammatical theory and analysis Different functional theories can be placed along a

continuum according to their reduction of grammatical structure to discourse Nichols

(1984) distinguished three groups of functionalism: extreme, moderate and conservative.

+ Extreme functionalist theories deny any relevance to the formal aspect of language Itmay claim that rules are based entirely on function and hence there are no purely syntacticconstraints According to advocates of this approach, grammar is reduced to discourse andany apparent structural system being taken as an epiphenomenon of recurrent discoursepatterns, formulaic expressions, etc Therefore, this approach rejects any notion of

grammatical structure other than that of discourse As Newmeyer (2001) and Croft (1995)

10

point out, this most extreme form of functionalism rejects the Saussurean dichotomiessuch as Langue vs Parole and Synchrony vs Diachrony This kind of deductionism is

common enough in oral statements, but rare in print.

+ Functionalist theories such as Kuno (1975), as Nichols (1984) mentioned, are considered

as conservative functionalism, which merely acknowledge the inadequacy of strictformalism or structuralism, without proposing a new analysis of structure Kuno did notchallenge the fundamental theoretical assumptions of Chomskyan linguistics In his view,

there is no conflict in principle between functional syntax and the Government and

Binding theory of generative grammar Newmeyer (2001) stated that those linguists whohave worked along the lines that Kuno advocates typically probe the interactions ofgrammar and discourse, without making the claim that the former can be derived from the

latter In these approaches, some functionalist explanatory principles are added to what is

inherently a formalist undertaking Consequently, only the aspects of grammar notamenable to a formal account are considered to be motivated by pragmatics and semantics.This kind of functionalism is termed ‘Formal Functionalism’ by Newmeyer (2001).

+ Between these two extremes, moderate functionalism includes other functionalisttheories such as Functional Grammar (FG, Dik 1991), Systemic Functional Grammar(SFG, Halliday 1985, 1994), Role and Reference Grammar (RRG, VanValin and LaPolla1997), etc which fall between extreme functionalism and conservative functionalism.These theories are also referred to as external functionalism (Newmeyer 2000).Newmeyer (2000) describes this group of functionalism as follows:

"External functionalism, like functionalism in general, rejects the project ofcharacterizing the formal relationships among grammatical elements independentlyof any characterization of the semantics and pragmatics of thoseelements Nevertheless, external functionalism upholds the idea of a synchronicsemiotic system, in which formal elements are linked to semantic and pragmatic

ones" (2000:13)

1.1.3 Systemic Functional Linguistics

Systemic Functional Linguistics (SFL) is an approach to language developedmainly by M.A.K Halliday in the UK and later in Australia This approach has its originin the main intellectual tradition of European linguistics that developed following the work

of Saussure Its primary source was the work of J.R Firth (1957) and his colleagues in

11

London, who defines “system” in Systemic Functional Linguistics, as, in its technicalsense, the “theoretical representation of paradigmatic relations, contrasted with “structure”for syntagmatic relations”, and who characterizes systemic theory as the theory in which

the system takes priority, the most abstract representation at any level is in paradigmaticterms As well as other schools of thought in Europe such as glossemantics, this theoryalso draws on American anthropological linguistics, and on traditional and modernlinguistics as developed in China While American-style linguistics evolved in the

modeling of the world’s languages, SFL was developed to address the need of language

teaching and learning A significant milestone in the development of this theory is when itwas wholly developed in the work on the grammar of Chinese by Halliday (1956); and it

has been used in educational and computational contexts from an early stage.

1.1.3.1 Text and context in the view of SFL

SFL is distinct amongst linguistic theories as it seeks to develop both a theoryabout social process and analytical methodology which permits the detailed and systematicdescription of language patterns According to Halliday, language is a systematic resourcefor expressing meaning in context, and linguistics is the study of how people exchangemeanings through the use of language With the view that “the way into understandingabout language lies in the study of text, the term TEXT and CONTEXT, put together likethis, serve as the reminder that these are aspects of the same process There is text andthere is other text that accompanies it: text that is “with”, namely the context” (Hallidayand Hasan 1985: 5), systemic functional grammarians do not take the sentence with itsstructure and constituents as the subjects of study like other linguistic theories but they

study text instead.

Text is defined as any instance of language, in any medium, that makes sense tosomeone who knows the language Text is not a grammatical unit, but a semantic unit, unitnot of form but of meaning (Halliday and Hasan, 1976:1-2) “Language that is functional.[ ] The important thing about the nature of text is that, although when we write it down,it looks as if it is made of words and sentences, it is really made of meanings” (Hallidayand Hasan 1985: 10) This view of language as a system for meaning potential implies thatlanguage is not a well defined system, not simply the set of all grammatical sentences Italso implies that language exists in contexts and therefore must be studied in contexts suchas professional settings, classrooms, and language tests The crucial characteristic of SFLis its orientation outside linguistics towards sociology This orientation brings with it a

12

view of language as a social semiotic (Halliday, 1978) Context is a higher-level semioticsystem in which language is embedded' More specifically, language is embedded in acontext of culture or social system and any instantiation of language as text is embedded inits own context of situation Context is an ecological matrix for both the general system oflanguage and for particular texts It is realized through language; and being realizedthrough language means that it both creates and is created by language Context isfunctionally diversified into three general domains: field (what is going on), tenor (who istaking part, the social roles and relations of those taking part in the interaction), and mode(what role language is playing in the context and the channel).

In summary, the central point of SFL is that language use must be seen as takingplace in social context Language is neither good nor bad, it is just appropriate orinappropriate to the context of use.

1.1.3.2 How “systemic”?

As the moderate theory among functionalist theories, SFL does not completely

reject the set of rules by formalism, while focusing on the functional side: what exchangesof language do The term “systemic” side in systemic functional linguistics is a

characteristic of the approach which Halliday (1994) adopted in his Introduction to

Functional Grammar, the systemic theory The theory, in Halliday’s view, is

comprehensive in that it is concerned with language in its entirety, so that whatever is saidabout one aspect is to be understood always with reference to the total picture At the sametime, of course, what is being said about any one aspect also contributes to the totalpicture; but in that respect as well it is important to recognize where everything fits in.

One of the reasons why Halliday adopted this systemic perspective in his Systemic

Functional Grammar (SFG) is that languages evolve — they are not designed, and evolved

systems cannot be explained simply as the sum of their parts.

“The thing that distinguishes SFG is that it gives priority to paradigmatic relations:“it interprets language not as as set of structure but as a network of systems, or interrelated

sets of options for making meaning Such options are not defined by reference to structure;they are purely abstract features, and structure comes in as the means whereby they are put

into effect, or realized” (Halliday 1994:15-16)

The systemic approach therefore allows the language users to take a “path”through the network, and combine all the structural rules to make meaningful choices for

13

desired purposes of communication without needing to think of the particular structure torealize it Grammar of a certain language is accordingly made up of a set of choices

organized as a tree with the implication that some choices depend on others:

+Pass; Pass: be-aux;

Fig.1: “Systemic” means making choice in a network

Essentially, language is considered a system, taking sound (spoken language) asthe mean of doing, i.e it provides a linguistic behaviour potential with various possibilitiesfor users to make choices, basing themselves on the context Language is therefore seen asprimarily a social resource with which speakers and hearer can act meaningfully Becauselanguage is defined as a systematic resource, the organizing principle in linguisticdescription is system (rather than structure) Since language is viewed as semioticpotential, the description of language is a description of choice The available choicesdepend on aspects of the context in which the language is being used These choices areassumed to be meaningful and relate speakers' intentions to the concrete forms of alanguage Choices can be charted on different levels, or strata, of language, includingsemantics, lexicogrammar, and phonology Phonology is the organization of speech soundinto formal structures and systems; however, taking sounds as the base, the stratification isinto phonetics, the interfacing with the body’s resources for speech and for hearing Also,any choice made must be embedded in the context, so that the number of strata one

chooses to include in analysis should depend on the purpose of a given description The

"strata" perspective of systemic linguistics allows for this flexibility across research needswhile maintaining its fundamental definition of language as a resource for meaningpotential In order to identify meaning choices, we have to look outwards at the context:what, in the kind of society we live in, do we need or want to say? What are the contextualfactors which make one set of meanings more appropriate or likely to be expressed thananother? But at the same time we need to identify the linguistic structures, and to explorethe meaning that each option expresses These are complementary perspectives on thesame phenomenon, one from the bottom up (sounds to wording, then to context) and theother from the top down (context to wording, then to sounds).

14

underlying such specific instances of language use are more general functions which

are common to all cultures We do not all go on fishing expeditions; however, we alluse language as a means of organizing other people, and directing their behaviour.”

(Halliday 1970: 173)

As stated previously, SFL concentrates on what language is used for, not how it iscomposed, the “functional” side of SFG means it is more concerned with what the unit oflanguage does — the functions of language, including: (1) functions of syntactic unit:participant, process, circumstance; (ii) functions of speech act as a whole (speechfunction): give / demand, action/ information; and (iii) functions of texts as a whole(genre): description, narration, explanation, exposition, etc.

Language can be considered as a multi-functional system because, as an organicsystem, the units of language in use usually serve more than one function at the same time.

15

In other words, language makes different kinds of meaning concurrently In the view ofHalliday (1970), there are 3 lines of meaning demonstrated by any unit of language Thefirst line of meaning can first be expressed through the choices of wording, which meansnot just the individual words chosen, but through other ways, e.g order of the words.Thus, the meaning is always more than the sum of individual words In this line of

meaning, language serves for the expression of “content”: that is, of the speaker’s

experience of the real world, including the inner world of his own consciousness Another

line of meaning is related to the speaker’s assessment of the validity of his / her

proposition, which is typically expressed by the use of modality resources of the language.Here, language serves to establish and maintain social relations: for the expression of

social roles, including the communication roles created by language itself — e.g the roles

of questioner or respondent, which we take on by asking or answering a question; and alsofor getting things done, by means of the interaction between one person and another Afurther line of meaning concerns how the message fits in with what else is said around it.In this sense, language has to provide for making links with itself and with features of thesituation in which it is used These lines of meanings contribute equally to the meaning ofthe message as a whole These three lines of meaning match three broad functions oflanguage, which SFL calls these broad functions the three metafuntions of language -

ideational, interpersonal, and textual - to construe different aspects of our experience.

Halliday (1994) describes the three metafunctions of language as follows:

(i) The interpersonal metafunction is concerned with the interaction between speaker and

addressee(s), the grammatical resources for enacting social roles in general, and speechroles in particular, in dialogic interaction; i.e for establishing, changing, and maintaining

interpersonal relations Through this function, social groups are delimited, and theindividual is identified and reinforced, since, by enabling him to interact with others,language also serves in the expression and development of his own personality One of its

major grammatical systems is mood, the grammaticalization of speech function.

Interpersonal meanings are realised by systems of Mood and Modality and by the selection

attendant on it.

Ideational (experiential and logical) meanings construing Field are realised

lexicogrammatically by the system of Transitivity This system interprets and represents

our experience of phenomena in the world and in our consciousness by modelling

experiential meanings in terms of participants, processes and circumstances Resources for

chaining clauses into clause complexes, and for realising time by means of tense, address

logical meanings.

These two metafunctions orient towards two 'extra-linguistic' phenomena, thesocial world and the natural world; we construe the natural world in the ideational modeand to enact the social world in the interpersonal mode For instance, we can construe apicture of what can participate in an action (ideational) and we can enact who gives orders

to whom (interpersonal).

(iii) The textual metafunction, the third metafunction, which intrinsic to language, isconcerned with the creation of text, with the presentation of ideational and interpersonalmeanings as information that can be shared by speaker and listener in text unfolding incontext Textual meanings are concerned with the ongoing orchestration of interpersonal

17

and ideational information as text in context This is what enables the speaker or writer to

construct “texts”, or connected passages of discourse that is situationally relevant; andenables the listener or reader to distinguish a text from a random set of sentences One

aspect of the textual function is the establishment of cohesive relations from one sentence

to another in a discourse (Hasan 1968).

Lexicogrammatically, textual meanings are realized by systems of Theme and

Information Theme is the resource for setting up a local context for a clause by

selecting a local point of departure in the flow of information Theme selectionsestablish the orientation or angle on the interpersonal and ideational concerns of theclause whereas Information organizes the informational status or relativenewsworthiness of these concerns The role of the textual metafunction is an enablingone It serves to enable the presentation of ideational and interpersonal meaning asinformation that can be shared: it provides the speaker with strategies for guiding thelistener in his / her interpretation of the text.

Three lines of meaning in a clause are demonstrated in the following analysis byHalliday (1994:371)

Vocative | Subject | Finite

Locative Actor Process Manner ideational

The metafunctional theory is part of the “functional” side of SFG, but it is alsoimportant in the “systemic” side of SFG Each metafunction has a principal system in thenetworks for clauses, verbal groups and nominal groups For example the transitivitysystem is the principal system for the ideational metafunction in the clause network.

An important theoretical point is that in the system networks, the systemswithin each metafunction are closely interconnected, but largely independent of systemsin the other metafunctions System interconnections across metafunctions are rare.

18

[F— material ideational:

F— intensive

TRANSITIV-spe [— mental ITY

'— unmarkedt heme

Fig 4: Metafunctions as manifested in the system network of the clause

(Matthiessen & Halliday 1997:14)

In this network fragment, there are normal dependency relationships within the

mood region of the interpersonal metafunction, between the mood-type and type systems and between the indicative-type and interrogative-type systems, and there isalso a further interconnection: the tagging system can be entered either from the

indicative-imperative feature of the mood-type system or from the declarative feature of the

indicative-type system But there are no interconnections at all between the mood regionof the interpersonal metafunction and the transitivity region of the ideational metafunction.It can accordingly be concluded that all of the three metafunctions are simultaneously

relevant They are of equal status.

In summing up, grammatical features in the light of SFL, or features of systemic

functional grammar (SFG), provide an abstract specification of the three functions that alinguistic structure may realize, or the metafunctions of the language: the ideational

metafunction, the interpersonal metafunction, and the textual metafunction With the

primary goal of addressing the needs of language teaching / learning, the SFL theory lays

an emphasis on the functions of language - what language is used for.

Although SFG is a theory of grammar, it focuses more on the meaning, not thepure form of what is said like what formalists have long been doing Instead of justinforming about the syntactic dimension, SFG helps to reveal the semantic dimension as

19

well With the aim of comparing the logico-semantic relationship in Vietnamese andEnglish clause complexes — the manifestation of the syntax-semantics interface in theclause complexes, the dissertation requires a theoretical perspective that can help describenot only the form — the structure, but the functions — the meaning behind the structure ofthat item as well SFG is evidently a more suitable theoretical perspective than formalism

approaches to grammar The theoretical perspective that the dissertation “Logico-semanticrelationship in English and Vietnamese Clause Complexes” takes is therefore SFG.

1.1.4 Review of Related Studies

1.1.4.1 An overview of Studies in Vietnamese Grammar

The history of research on Vietnamese grammar started long ago, in the earlynineteenth century The very first studies of Vietnamese grammar were perhaps just thesimple notes of different parts of speech, the role of word order, etc in the bilingualdictionaries developed by western scholars These early presentations of Vietnamesesyntax were just practical for the sake of teaching foreigners to learn Vietnamese; theywere not well theorized, so the findings were limited to several factors such as: (i) nochange in forms in Vietnamese words, no morphological basis to determine the differentparts of speech, (Vietnamese, thus, used to be regarded as a language with no parts ofspeech by Grammon & Trinh: 1911-1912); (ii) the significant role of word order insentence meaning Some notes about functional words in Vietnamese with descriptions of

their general meanings and use can be found in “Dictionarium Anamitico-Latinum” edited

by Taberd, J L (published in 1838) (Hiệp: 2009) Written by foreign scholars, and, later,even by such Vietnamese scholars as Pham Duy Khiêm, Bùi Ky, and Trần Trọng Kim, the

general approach of these early studies on Vietnamese grammar was applying the

grammatical frameworks of European languages, especially French, to analyzeVietnamese language, a language of a totally different language family These authorsperceived the concepts and descriptions of syntax in the way of the French language, thusdescribing Vietnamese as some sort of French syntax illustrated in Vietnamese Thesestudies therefore failed to come up with satisfactory findings.

Marking an end for a periods of purely practical and intuition-based works onVietnamese, more appropriately theorized approaches in studying Vietnamese emerged in

1940s — 1950s, Tran Trọng Kim (1940), Bùi Duc Tinh (1952) and Phan Khôi (1955),

though could not adequately justify and clearly state the methodology of their studies,came up with valuable findings which are still recognized today.

20

A significant milestone in Vietnamese studies is Le Parler Vietnamien by Lê Van

Ly (1948), which is appreciated as a genuine work of linguistics and which still makesgreat influence on Vietnamese studies today, well-based on the theoretical theories,offering a comprehensive classification of words in Vietnamese Since then, well-theorized studies on Vietnamese has flourished fruitfully with various authors and their

different focuses in the language While Nguyén Kim Than (1963, 1964) and Thompson

(1967) describe the system of Vietnamese syntax as a whole, other authors focuses onspecific areas in the language: Nguyén Tai Can (1975): Vietnamese nouns and compoundwords, Nguyễn Kim Than (1977): Vietnamese verbs and verb phrases, Hồ Lê (1976),Hoàng Văn Hành (1985): word formation in Vietnamese, Hoàng Trọng Phiến (1980) andDiệp Quang Ban (1984): Vietnamese sentences, and Trần Ngọc Thêm (1985): cohesion inVietnamese texts.

Up to now, quite a number of studies have been conducted, analyzing theVietnamese grammar in the light of SFL The first Vietnamese linguist to go and discover

the area of SFL in his published works is Cao Xuân Hao with his 2 books “Tiếng Việt: SơThao Ngữ Pháp Chức Nang: Câu trong tiếng Việt” (A Functional General Description of

Vietnamese Grammar: the Sentence) and “Tiếng Việt: Sơ Thao Ngữ Pháp Chức Năng:

Ngữ Doan và Từ Loại” (A Functional General Description of Vietnamese Grammar: theGroup and the Word) in 1991” which were then represented in 2007 These works in 1991gave an overview of SFL and then use this theory as the theoretical framework for thepreliminary analysis of Vietnamese grammar The edited versions in 2007 do really gointo more details with amendments and further specification of the issues mentioned in theprevious versions.

The PhD dissertation of Hoang Van Van (1997): “An Experiential Grammar of theVietnamese Clause: A Functional Description” and his MA thesis: “A FunctionalPerspective on Translating ELT Texts from English into Vietnamese” (1994) inDepartment of Linguistics, University of Macquarie, Sydney are among the first attempt toinvestigate SFG of Vietnamese scholars Other studies are from Thai Minh Duc (1996) “AMetafunction Profile of Vietnamese Clause Grammar” - Paper presented at the SystemicFunctional Workshops at Sydney University and “A Systemic-Functional Interpretation ofVietnamese Grammar” (1998) - PhD dissertation in Department of Linguistics, Universityof Macquarie Conducted in Macquarie University in Sydney, the home of SFL, the worksby these two linguists do enlighten a lot more about detail analysis of Vietnamese While

21

Hoàng Văn Vân (1997) goes deep into Vietnamese clause specifically, Thai Minh Đức

(1998) work at the more general scale of interpreting the grammatical system of thelanguage Recently, Dé Tuấn Minh (2007) conducts a comparative study on the thematicstructure of English and Vietnamese However, none of those authors ever attempt todiscover the area of logico-semantic relations in clause combination in Vietnamese.

1.1.4.2 SFL Studies in Other Languages

As regards the linguistic area of clause combining, like most traditional

formalists, many functionalists have discussed at certain details the syntactic and semantic

features of the complex sentences or clause complexes as a small part in their wholetheories or books A remarkable contribution is made by Hainman and Thompson (1988)who bring together a number of studies on clause combining that gives insight into a rangeof languages and to the matter of typological variation.

The linguistic areas of clause complexes and the logico-semantic relationships inEnglish clause complexes have all been discussed in details theoretically in the light ofSFL, and have also been applied to conduct comparative studies between English andJapanese, English in Thai, English and Spanish, but not yet Vietnamese Thus, the authorof this research aims at this untouched area, making a profound comparative observation

of the logico-semantic relations in English and Vietnamese Clause Complexes.

1.1.5 Summary

As the dissertation takes the aim of comparing the logico-semantic relationship inEnglish and Vietnamese clause complexes, the historical context of the emergence of

functional linguistics, approaches in functionalism, especially the characterizing features

of SFL With the features as demonstrated, it is concluded that SFL is the most suitabletheoretical perspective for pursuing the aim of this dissertation.

In addition, the brief history of describing Vietnamese grammar, comparing it withdifferent languages, and other related studies in the same linguistic area in other languageshave been reviewed in this chapter It is clearly seen that the area of logico-semanticrelationship in Vietnamese clause complexes has never been described and compared inthe light of SFL before A comparative study on logico-semantic relationship in Englishand Vietnamese clause complexes is therefore needed, and such a study is hoped to makecertain contribution sketching up the functional framework for Vietnamese grammar.

22

1.2 The Theoretical Framework of Logico-semantic Relationship in Clause Complex

Since 1980, SFL has been expanded considerably in various directions, furtherstudies have been devoted to languages other than English, notably Chinese, French,Indonesian, Japanese and some other Asian and African languages However, the theoretical

“dig” into the logical-semantic relationship in clause complexes are mostly conducted in

English, taking English as the source for the corpora, it is therefore unavoidable that the

sketch of the picture of logico-semantic relationships in clause complexes is rather centric In English, the linguistic phenomenon of clause complexes and logico-semanticrelations in clause complexes have been described by many systemic functionalists likeHalliday (1985, 1994), Downing and Locke (1992), Halliday & Matthiessen (2004), Eggins

Anglo-(2004) All enlightened by the systemic functional approach, each author, however, has his

or her own viewpoint on the same issue, which is reflected in different classifications,systems of terminology, and language analysis Therefore, there is a need to re-examine thenotion of clause complexes and the logico-semantic relations in English clause complexes

so that all the ideas from the authors are critically recognized and “digested” to restructure amore comprehensive framework that can be applied to different languages.

Additionally, with the aim to compare the realization and operation of the

logico-semantic relationships in clause complexes of English and Vietnamese, the two languages

that relate in no way as regard language origin and language family and in turn, differ a lotin their morphology, syntax and lexico-semantic systems, the author of this dissertation tries

to take the most neutral possible approach in analyzing the corpora and interpreting theresults, combining the discoveries of the previous perspectives and then restructuring them

into a so-called overarching theoretical framework that can be applied to describe thelogico-semantic relationships in both targeted languages.

This section aims to answer two questions: (i) How is clause complex defined inSFL? and (ii) What are the frameworks of logico-semantic relations in clause complexes?

The two questions are answered in turn in two different sections as the following.

1.2.1 The Notion of Clause Complex in Systemic Functional Linguistics1.2.1.1 Concepts revisited

For the sake of consistent use of terms in the following parts, some concepts arerevisited in this part.

a Units and the Rank scales

23

Constituency in language is built on the part-whole relation; it presupposes a wholeof which we identify constituent parts Wholes which display an organic constituencystructure are called grammatical units Units have syntagmatic integrity: they are fullyaccounted for by their structures, and they are not structurally mixed with other units.

Unit is a stretch of language which itself carries grammatical patterns or whichoperates in grammatical patterns Grammatical units are identifiable in functional terms, i.e.(i) they are the points of origin of system networks (such as those of transitivity and moodin the clause) and (1) they function as constituents in their entirety Functionally determinedunits can be arrived by adopting a rank-based type of constituency.

Rank orders units into a hierarchy according to their constituency relation: thehighest-ranking units consist of units of the rank immediately below; these units consist ofunits at the next rank below, and so on, until we arrive at the units of the lowest rank, whichhave no internal constituent structure Rank is thus a theory of the global distribution of theunits of the grammar The traditional grammatical rank scale is rooted from morpheme asthe lowest unit, up to word, up to phrase / group, up to clause and then up to sentence as thehighest unit That is, a sentence consists of a clause or of clauses, a clause consists of groups/ phrases, a group / phrase consists of words, and a word consists of morphemes Aninstance of the ranked constituency structure is as shown in Fig 5.

new + born caly + #5 are eas + Prey

Fig 5: Rank-based Constituency (Matthiessen & Halliday 1997: 26)

There exists a common traditional view that sentence is the highest unit in the

grammatical rank scale with the immediate constituent as the clause, but this view has been

questioned by systemic functionalists.

In the light of Systemic Functional Linguistics, which emphasizes the functionalaspect of language with the manifestation of the three metafunctions, clause is consider to

be the highest unit in the grammatical rank scale Halliday (1994) points out that “we mayassume that in all languages, the clause has the character of the message; it has some form

of organization giving it the status of a communicative event” (1994:37).

24

It is no exaggeration to accept clause as the unit of language which can manifest allthe three metafunctions of the language: the experiential meaning (clause as a

representation), the interpersonal meaning (clause as an exchange), and the textual meaning

(clause as a message) A sentence can be built up from one clause or two or more than twoclauses linked together in certain systematic and meaningful ways A sentence is a clauseitself, or a logical combination of clauses which is just the extension of a message, anexchange, a linguistic representation, therefore also a sentence manifests the same threemetafunctions of language as a clause does There is no actual need for any further functionto be realized by a higher grammatical unit than the clause A sentence should accordinglybe ranked next to the clause in the grammatical rank scale If the sentence consists of oneclause, it is equal to the clause, if it is a combination of clauses, it is only considered alogico-semantic unit above the clause, not a grammatical unit above the clause.

The grammatical rank scale therefore should be as follows:

GROUP/ PHRASEWORD

Delicacy accordingly orders paradigmatic options with respect to one another.Systems in a system network are ordered from more general to more specific The moregeneral options provide the context in which more delicate ones are available The scale ofdelicacy is essential in describing the grammar of a language: it allows for a varying degreeof detail of description For example, we can describe the grammar in a less delicate system

of LOGICO-SEMANTIC RELATIONS, classifying these relations into expansion and

projection; or we can move to a more delicate system of expansion, further separatingexpansion into extension, elaboration, and enhancement The interpretation of

25

lexicogrammar as a unified system (with the grammar being the less delicate, and the lexisthe most delicate) reflects the same organizing principle, which implies that the scale ofdelicacy means: we need only one system network of lexicogrammatical options, andlexical specifications can inherit the more general grammatical specifications.

Languages may differ in the delicacy of their paradigmatic organizations Mostgenerally, languages tend to have more in common in the areas of the more general systems,where the systems become more delicate, languages become more divergent An example ofdelicacy can be seen from the analysis of the system of major clause adapted fromMatthiessen (2005), a demonstration of how the system goes from the less delicate — themajor clause — to the more delicate - the verbal process, then the system of projectingrelation All the items in the chart are features of the clause, they all realize the experientialfunction of language at the rank of the clause The point is that, the further right the chartgoes, the more delicate the features becomes.

c Univariate Structure vs Multivariate Structure

The logical mode of syntactic structure can be either univariate or multivariate.

A multivariate structure involves more than one variable, where each one is uniquein occurrence In other words, a multivariate structure is one where we can identify acomplete whole which is made up of functionally distinct constituents, which is exocentric.The multivariate structure therefore is not inherently recursive Recursion in this kind of

structure is cyclical - recursion can only be introduced by rank-shift, which means a unit isembedded in another unit typically higher in the rank scale, thus creating an additional layerin the syntagmatic representation For example the clause is a multivariate structure whichis (interpersonally) made up of a Subject, Finite, Predicator, Complement, and Adjunct.Each element performs a distinct and different role in contributing to the meaning of thewhole clause structure Obligatory elements (eg Subject and Finite) can only occur once;otherwise we have, by definition a new clause Both mood and transitivity structures aremultivariate, as are most generic structures.

In a univariate structure, on the other hand, we're dealing with a relationshipbetween elements that are essentially the same and which can be chained togetherindefinitely Elements of the univariate structure are repetitions of the same variable.Univariate structure is said to be endocentric, which is linearly recursive — recursion is at the

same rank As a type of univariate structure, the paratactic structure is a chain ofdependencies, where none of the units involved is considered the head Another type ofunivariate structures is the hypotactic structure in which there is a designated head.Examples of univariate structures are coordinated structures and appositions.

type of structure mode of meaning mode of structure example of structure

(i) univariate logical iterative paratactic: 1 —» 2 —> 3 —> 4

(ii) multivariate experiential segmental e.g [clause:] Medium + Process + Agent + Place +

[nominal group:] Numerative + Epithet + Classifier

+ Thing + Qualifier

interpersonal prosodic e.g [clause:] Mood + Residue + Moodtag

[tone group:] // tone 2 4s

textual culminative e.g [clause:] Theme + Rheme

[information unit:] Given + New

[nominal group] Deictic + [verbal group] Finite +

Fig.8 Univariate and Multivariate Structure (Halliday & Matthiessen 2009: 383)

d Unit complex vs complex unit

In addition to the basic grammatical units we introduce the concepts of a ‘unitcomplex’ and a ‘complex unit These terms can be seen as merely alternative labels but we

27

would argue that there is a significant difference between them When considering thesentence, we made reference to the term 'clause complex’ and that can be seen as a modelfor a unit complex more generally By contrast with a simple unit, which consists of a singleexponent of a unit from a given rank, a unit complex may be explained as a coherentgrouping or configuration of two or more units (a complex of units) from the same rank,which together may also constitute a unit of the rank next above The syntactic structure of aunit complex is univariate and it can be said to be endocentric, which means recursion canhappen at the same rank (linear recursion) A unit and the complex of that unit (eg clauseand clause complex) therefore are in the same rank The characteristics of a unit complex isits open-endedness because a unit complex is not a pre-defined whole If another unit of thesame status is involved, then the relationship between them is one of equality - arelationship involving units of equal grammatical status In systemic grammar this hasnormally been referred to as one of the types of paratactic relationship.

In the terms outlined above, the relationships between the basic syntactic units can,then, be summarized by saying that (a) a textual sentence may be composed of a clause or acomplex of clauses, a clause complex; (b) a clause contains at least one phrase and morenormally a (multivariate) phrase complex; (c) a phrase consists of a single (head)word or of

a word complex; and (d) a word may comprise a morpheme or a morpheme complex.

Morley (2000: 27-28) analyses that following sentence:

Jack fell down and Jill came tumbling after.

exemplifies a unit consisting of two main clauses coordinated in one such relationship.

Where a subordinate unit is involved, the relationship between the main andsubordinate units is known as a hypotactic relationship A hypotactic relationship thusinvolves units of unequal grammatical status, as illustrated by the subordinate and mainclauses in the sentence

When Jack fell down, he broke his crown.

Traditionally in systemic grammar paratactic and hypotactic structures have been

grouped together as subtypes of univariate structure, that is to say a structure involving therepeated utterance of the same type of unit, where the relationship between the units is alogical one of equality or dependence Where, on the other hand, the relationship is onebetween different types of element, as for example between the units in

Jack Sprat could eat no fat (‘subject', verbal element, and 'obJect)

28

it represents the type of structure known as multivariate In these terms, therefore, a clausecomplex comprises more than one clause unit; it is a complex of clauses and typically

operates as a sentence Below are some clause complexes analyzed by Moorley (2000).

Clause complex:

Jack fell down and broke his crown and Jill came tumbling after (paratactic, coordination)

What I really want, what I must have, is a new car (paratactic, apposition)

After Keith was awarded his degree, his confidence increased enormously (hypotactic)

Whereas a clause complex involves a grouping of clauses which can frequentlyoperate as a unit of the rank above, a complex sentence is composed of two or moreelements typically, but not only, realized by units of the rank below The syntagmaticstructure of it is multivariate and it can be said to be exocentric as recursion can only beintroduce by rankshifting (cyclical recursion).

1.2.1.2 What is a clause complex?

As the name suggests, a clause complex is made up from clauses In order toanswer the question: what makes the clause complex, some attempt is made to examine thenotion of clause, the combination of clauses to make up the clause complex, and a brief

distinction between the notion of clause complex and the traditional notion of sentence.

a The notion of clause in the light of SFL

As the highest unit in the grammatical rank, the clause is viewed in the light ofsystemic functional linguistics as a grammatical resource for all the three languagemetafunctions: construing the world, enacting social roles, and presenting information.Halliday (1994) points out that there are 3 lines of meaning in a clause: the textual meaning,

the interpersonal meaning, and the ideational meaning, or to put it differently, the clause has

three metafunctions: textual, interpersonal, and ideational metafunctions as can beillustrated through figure 9 and figure 10 below.

The professor was analyzing the functions that a clause can | CL4USE MEANING

Theme Textual

29

Jim how the analysis would help CLAUSE MEANING

As can be seen, a clause has meaning as a message, a quantum of information whichis the complex of at least two constituents of Theme and Rheme in its THEMATIC structure.Traditionally, Theme is taken as what is going to be discussed in the message and Rheme as

the discussion, while SFL sees Theme as the departure of the message -the ground fromwhich the clause is taking off (Halliday 1994: 37) In a finite clause, Theme is explicit but ina non-finite clause, Theme is hidden, and can be recovered in the context.

A clause also has meaning as an exchange, or a move Through the system ofMOOD, it is organized as an interactive event involving speaker, or writer, and audience.MOOD is, in general terms, the function of the clause, which is realized through Mood

(Subject and Finite) and Modality (Halliday: 1994) In speaking, the speaker adopts forhimself a particular role, and in doing so, assign the listener a complementary role which he

wishes him to adopt in his turn As regard speech role, either the speaker is giving

something to the listener, or demanding something from him In the clause the subject is the

guarantee of the exchange In finite clause, mood is explicit whereas in the non-finite clause,mood is non-explicit.

A clause has meaning as a representation, or a figure, a construal of the going-ons in

human experience, with the actor as the active participant in that process The clauserepresents a pattern of experience through the system of TRANSITIVITY: processes,

participants, circumstances This embodies the principle for modeling experience — “theprinciple that reality is made up of processes” (Halliday 1994: 106) In the clause,

participants are decided by the process, and circumstances are what surrounds the process.Process is the indispensible part in both finite and non-finite clause.

It is no exaggeration to accept clause as the central unit of language which canmanifest all the three metafunctions of the language; and there seems to be, consequently,no actual need for any further function to be realized by a higher grammatical unit than theclause However, when classifying clauses into major clauses and minor clauses, only major

30

clauses can demonstrate all the systems of MOOD, TRANSITIVITY, and THEME; minorclauses, including calls and exclamations, have no such systems.

b What makes a clause complex?

In their use of language, people in many cases tend to expand their argumentoutwards by combining, or complexing the original clause with other related clauses intoseries of clauses with the main clause as the core of the message and the coordinate orsubordinate clauses as the peripheral information added to reinforce the message People inmany other cases use language to describe not only the non-linguistic phenomena but toreport or quote the linguistic phenomena as well, allowing the reported or quoted clauses toenter into a combination of clauses as the projected part in the whole combination - the

secondary use of language That is how clause complexes are constructed from clauses Asthe single independent clause can be thought of as the linguistic expression of a situation,the combination of several clauses together to form a larger unit — a complex of clauses or aclause complex - can be thought of as the linguistic expression of a situation, thecombination of several clauses together form a larger unit can be thought of as the linguisticexpression of a complex situation While a simplex of clause or a clause simplex is a clauseitself, a clause complex can be built up from more than one clauses linked together in

certain systematic and meaningful ways A question to be answered is: “Is the clausecomplex the grammatical unit above the clause?” The illustration can be seen from the

analyzing the functions that a Jim how the analysis CLAUSE

professor clause can perform, would help MEANING

Residue Residue interpersonal

«aim im+ * aia wonder + how the analysis would help)

Actor Process Goal senser Process phenomenon ideational

material mental

Clause 1 Clause 2

While | the was | analyzing the _ that | wondered - how the analysis CLAUSE

professor a clause can did | wonder would help MEANING

os —*= — |*=l— ®==—— | mem

Residue Residue interpersonal

(Jim + Tag wonder + how the analysis would help)

Actor Process senser Process phenomenon ideational

material mental

31

It is demonstrated from the analysis of the clause complex above that the clausecomplex is a univariate structure, not a multivariate structure Any grammatical unit in thegrammatical rank scale is a multivariate unit in that it can realize the immediate unit aboveit and is realized by the immediate unit below it, viz a word is realized by differentmorphemes, free and bound, which functions differently in the words, different words of

different parts of speech and different functions The fact that the constituents of a unitstructure are different in realization and functions means that the structure of a unit on therank scale is always multivariate Combining the clauses into clause complex is just likeassembling the details to make a new structure in which the details coexist but each worksin its own way In the clause complex, one clause is put next to the other, so the clause

complex is still at the same rank as the clause.

The exemplifying clause complex is seen as univariate structure as it is composed oftwo clauses, which are two independent realization patterns of two different systems of

MOOD, THEME, and TRANSITIVITY Grammatically, when attaching the clauses

together into a clause complex, no new system is created The clause complex is actually theplain combination of separate units, no units change themselves to fit the other The moreclauses are involved in the clause complex, the more systems of MOOD, THEME, and

TRANSITIVITY are, mechanically, added to the complex structure while the MOOD,THEME, and TRANSITIVITY of each clause are still reserved and not affected by the

other This means that, in theory, the clause complex can be extended as much as thespeaker / writer wants In other words, what distinguishes the clause complex, or other unit

complexes, from the grammatical units in the rank scale is its open-endedness because it is

not a pre-defined whole.

As can be seen in the complex above, there is no difference in MOOD, THEME, and

TRANSITIVITY when the two clauses stand independently as two clause simplexes, and

when they combine into the structure of a clause complex The relation between the clausescannot be read from the forms, but from the close look at the meaning of the clauses Thespeaker / writer convey some certain intended meaning of coordination or subordinationthrough the way s/he combines the clauses, and the original clause from which the complexis extended can be traced by the listener / reader basing on the semantic relations between

the clauses The clause complex is accordingly more of a semantic unit rather than agrammatical unit The answer to the question posed above is: a clause complex is not a

grammatical unit above the clause It is at the same rank as the clause, just as other

32