Semantic structure in english

Syntax puts our meaning (“semantics”) into sentences, and phonology puts the sentences into the sounds that we hear and there must, surely, be a structure in the meaning that is expressed in the syntax and phonology. Some writers use the phrase “semantic structure”, but are referring to conceptual structure; since we can express our conceptual thought in many different linguistic ways, we cannot equate conceptual and semantic structures.The research reported in this book shows semantic structure to be in part hierarchic, fitting the syntax in which it is expressed, and partly a network, fitting the nature of the mind, from which it springs. It is complex enough to provide for the emotive and imaginative dimensions of language, and for shifts of standard meanings in context, and the “rules” that control them. Showing the full structure of English semantics requires attention to many currently topical issues, and since the underlying theory is fresh, there are fresh implications for them. The most important of those issues is information structure, which is given full treatment, showing its overall structure, and its relation to semantics and the whole grammar of English.

CUNY and Adelphi University

Olga Mišeska Tomić

or other nonvocal medium

The series was formerly known as Linguistic & Literary Studies in Eastern Europe (LLSEE).

For an overview of all books published in this series, please see

of Paper for Printed Library Materials, ansi z39.48-1984.

Chapter 1

Introduction 1 1.1 Goals of the book 1

1.2 Approach 1

1.3 Argument of the book 2

1.4 Plan of the book 5

Chapter 2

2.1 Introduction 7

2.2 Semantic structure in lexis 7

2.2.1 Paradigmatic lexical relations 7

2.2.2 Syntagmatic lexical relations 8

2.2.3 Words without paradigmatic or syntagmatic relations 9

2.3 Semantic structure in morphology 9

2.3.1 Semantic classes 9

2.3.2 Grammatical meaning 10

2.3.3 Dependency and modification 10

2.4 Semantic structure in syntax 10

2.4.1 Semantic structure of clauses: The figure 10

2.4.2 Semantic structure in groups 13

2.5 Semantic structure in phonology 14

2.5.6 Semantic structure in phonetics 18

2.5.7 Discussion: Semantic structure in phonology 19

2.5.8 Conclusion: Semantics in the phonological stratum 21

2.6 Conclusion: Semantic structure in the strata of English 22

4.5 Grammatical meaning 61

4.5.1 Introduction 61

4.5.2 Types of grammatical meaning 64

4.5.3 Dimensions of grammatical meaning 65

4.6.3 Discussion: Semantic classes 71

4.6.4 Conclusion: Semantic classes 72

4.7 Uses of meaning 72

4.7.1 Introduction 72

4.7.2 Marked use 73

4.7.3 Defining and descriptive uses 73

4.7.4 Literal and figurative uses 76

4.8 Discussion: Elements of semantic structure 77

4.9 Conclusion: Elements of semantic structure 77

5.2.2 Word senses as a structure of meaning types 82

5.2.3 Structure within descriptive senses 87

5.2.4 Structure within non-descriptive senses 92

5.2.5 Discussion: Compositionality of sense structure 93

5.3 Structure among word senses 95

5.3.1 Introduction 95

5.3.2 Sense relations: Synonymy 96

5.3.3 Other sense relations 98

5.3.4 Variation in sense structure 99

5.3.5 Conclusion: Structure among word senses 107

5.4 Structure of sublexical meaning 108

5.4.1 Introduction 108

5.4.2 Dimensions 109

5.4.3 Elements and their sublexical relations 110

5.4.4 Elements’ external relations 112

5.4.5 Discussion: Structure of sublexical meanings 113

5.4.6 Conclusion: Structure of sublexical meaning 115

5.5 Networks in other strata 116

5.5.1 Networks in clause syntax 116

5.5.2 Networks in group syntax 118

5.5.3 Lexical network: Cohesion 119

5.6 Discussion: Network structures in English 120

5.6.1 Whole utterances as networks 120

5.6.2 Networks in imaginative English 121

5.6.3 Relations among one word’s various senses: Polysemy 121

5.7 Conclusion: Network structures in English 123

6.2.1 Introduction: Grammatical meanings 127

6.2.2 Preliminary process: Obtaining content for the main procedure 128 6.2.3 Processes applying within words and groups 129

6.2.4 Processes applying within figures and figure complexes:

Complementation 134

6.2.5 Processes applying to a whole figure 135

6.2.6 Discussion: System processes 137

6.2.7 Conclusion: Grammatical meaning 138

6.3 System procedures: Using the processes 138

6.3.1 Introduction 138

6.3.2 Constructing hierarchic structures 139

6.3.3 Constructing network structures in morphosyntax 140

6.4 Discussion: System structure 140

6.4.1 Details not yet explained 140

6.4.2 Signs used for grammatical meaning 141

6.4.3 Grammatical meanings in figurative and other uses 141

6.4.4 Grammatical meaning as backgrounded meaning 142

6.4.5 Other views of grammatical meaning 143

6.5 Conclusion: System structure 144

7.2.5 Discussion: Figure structures 153

7.2.6 Conclusion: Summary of figure structures 154

7.3 Processes 155

7.3.1 Introduction: Processes 155

7.3.2 Internal structure of Processes 155

7.3.3 Syntagmatic structure of Processes 156

7.3.4 Semantic class: Process 158

7.3.5 Conclusion: Summary of Processes 159

7.4 Participants 159

7.4.1 Introduction 159

7.4.2 Internal structure 160

7.4.3 Syntagmatic structure 160

7.4.4 Participant as semantic class 161

7.4.5 Discussion: Participant roles and “semantic roles” 161

7.4.6 Conclusion: Summary of Participants 162

7.5 Circumstances 162

7.5.1 Introduction 162

7.5.2 Internal structure of Circumstances 163

7.5.3 Syntagmatic structure of Circumstances 164

7.5.4 Semantic class of Circumstances 166

7.5.5 Discussion: Circumstances 166

7.5.6 Conclusion: Summary of Circumstances 168

7.6 Relations among Participant, Process and Circumstance 168

7.7 Structures larger than the figure 169

8.2.9 Entity group as a unit 211

8.2.10 Discussion of Entity groups 217

8.3.7 Postposed particles as modifiers 231

8.3.8 Heads of Process groups 232

8.3.9 Process groups as units 236

8.3.10 Discussion: Process groups 237

8.3.11 Conclusion: Process groups 239

8.4 Property groups 239

8.4.1 Introduction 239

8.4.2 Property groups with Property heads 240

8.4.3 Headless Property groups: Prepositional phrases 242 8.4.4 Conclusion: Property groups 244

8.5 Words and morphemes 245

8.5.1 Internal structure 245

8.5.2 Syntagmatic structure 247

8.5.3 Semantic classes: Content meaning 248

8.5.4 Grammatical meaning in words and morphemes 249 8.5.5 Discussion: Compositionality of word senses 249 8.6 Discussion: Hierarchic structure in groups and senses 250 8.6.1 Language constraints 250

8.6.2 Semantic change 251

8.7 Conclusion: Hierarchic structure in groups and senses 252

9.2.5 Conclusion: Relevance structure 265

9.3 Orientation structure: “Theme” 266

9.3.1 Introduction 266

9.3.2 Classes of Theme 267

9.3.3 Themes occurring at ranks other than the figure 269

9.3.4 Discussion: Orientation structure 270

9.3.5 Conclusion: Orientation structure 271

9.4 Salience structure: Rheme 272

9.4.1 Introduction 272

9.4.2 Salience within an information item 273

9.4.3 Salience of items within an information unit: Rhematic structure 276 9.4.4 Salience of information units in larger units 279

9.4.5 Conclusion: Salience structure 281

9.5 Discussion: Information structure 282

9.5.1 Relation of information structure to the bases and elements

10.3 Semantic units with multifunctional structure 295

10.3.1 Introduction 295

10.3.2 Holophrases 295

10.3.3 Ideophones 299

10.3.4 Discussion: Semantic units with multifunctional structure 300

10.3.5 Conclusion: Semantic units with multifunctional structure 301

10.4 Semantic units with field structure 301

10.5 Semantic units with wave structure 304

10.5.1 Introduction 304

10.5.2 Wave structure in information 304

10.5.3 Wave structure in aesthetics and emotion 306

10.5.4 Conclusion: Wave structure 306

10.6 Discussion: Other structures 307

10.7 Conclusion: Other structures 307

11.2.2 Holistic realisation of Expression 312

11.2.3 Phonological realisation of Expression 313

11.2.4 Phonetic realisation of Expression 314

11.2.5 Lexical realisation of Expression 315

11.2.6 Conclusion: Realisation of Expression 315

11.3 Realising emotion and attitude 316

11.3.1 Introduction 316

11.3.2 Conveying emotion by phonology and phonetics 316

11.3.3 Conveying emotion by lexis and syntax 317

11.3.4 Conveying attitude 318

11.3.5 Conclusion: Conveying emotion and attitude 319

11.4 Establishing personal and social relations 319

11.4.1 Establishing personal relations 319

11.4.2 Establishing social relations and social status 320

11.4.3 Discussion: Personal and social relations 320

11.4.4 Conclusion: Personal and social relations 321

11.5 Guiding hearers’ use of meaning 321

11.5.1 Introduction: Realisation of grammatical meaning 321

11.5.2 Guiding hearers’ overt response 322

11.5.3 Guiding hearers to syntagmatic structure 323

11.5.4 Guiding hearers to information structure 324

11.5.5 Guiding the hearer’s attitude to content: Irrealis 330

11.5.6 Discussion: Guiding hearers 332

11.5.7 Conclusion: Guiding hearers 333

11.6 Provoking imaginative responses 333

11.6.1 Introduction 333

11.6.2 How imaginative responses are provoked 334

11.6.3 Discussion: Provoking imaginative responses 337

11.6.4 Conclusion: Provoking imaginative responses 337

11.7 Discussion: Realisation of interpersonal meanings 338

11.8 Conclusion: Realisation of interpersonal functions 339

Chapter 12

12.1 Introduction 341

12.1.1 General introduction 341

12.1.2 Introduction to ideational realisation 341

12.2 From intentions to words 343

12.2.1 Introduction 343

12.2.2 Unmarked realisation into words 344

12.2.3 Marked realisation into words 348

12.2.4 Conclusion: From intentions to words 349

12.3 Syntacticisation (1): From words to groups 350

12.3.1 Introduction 350

12.3.2 Grouping the words 351

12.3.3 Structuring the group 352

12.3.4 Ordering the group 356

12.3.5 Signalling the group structure 357

12.3.6 Discussion: Syntacticisation into groups 360

12.3.7 Conclusion: Syntacticisation into groups 360

12.4 Syntacticisation (2): From groups to clauses 361

12.4.1 Introduction 361

12.4.2 Structuring the units 361

12.4.3 Ordering clause units 365

12.4.4 Signalling the structure 366

12.4.5 Conclusion: Syntacticisation into clauses 367

12.4.6 Grouping figures in a clause complex 367

12.5 Physical realisation 368

12.5.1 Introduction 368

12.5.2 Realisation in sound 369

12.5.3 Realisation in writing 371

12.6 Discussion: Realisation of ideational function 372

12.6.1 How we conceptualise realisation 372

12.6.2 Incongruent realisation: Grammatical metaphor 372

12.6.3 Construal 373

12.6.4 Statives 373

12.6.5 Domains offering alternative realisation 374

12.6.6 Comparison between morphosyntactic layers 376

12.7 Conclusion: Realisation of ideational meaning 376

12.8 Conclusion: Both forms of meaning realisation in English 378

12.8.1 Relation between the two forms of realisation 378

13.2.2 Needlessness and unworkability 388

13.2.3 Lack of explanatory power 389

13.3.2 Needlessness of prototype theory in English grammar 394

13.3.3 Unworkability of prototypes in grammar 395

13.3.4 Confusions in acceptance of prototypes 395

13.6.2 Distinction between descriptive and referential use 402

13.6.3 Distinction between grammatical and content meaning 403

13.6.4 Distinction between cognitive and linguistic areas of meaning 403 13.6.5 Formalisation of linguistic description 404

13.6.6 Concept of classes 404

13.6.7 Philosophical tradition in linguistics 405

Chapter 14

Conclusion 407 14.1 Introduction 407

14.3.2 Analytical view of semantic structure (1): Structures of units 410

14.3.3 Analytical view of semantic structure (2): Structures in a medium 412 14.3.4 Functional view of semantic structure 412

14.3.5 Conclusion: Nature of semantic structure 413

14.4 Stratification of the semantic structure 413

1.1 Goals of the book

The primary goal of this book is to set out the semantic structure of English

“Semantic” here is contrasted with “syntactic” and “phonological”, and also with

“pragmatic” (concerned with “meaning” which relies on social conventions such as the Gricean maxims, and on inferences from them) “Semantics” means “to do with studying what is expressed in language”, which includes “meaning” and “significance”; that meaning of the term will be refined later in the book It is not the study of prop-ositional or truth-conditional meaning alone, but the study of significance and mean-ings that are “coded” in linguistic items and structures, be they truth-conditional or not (Hansen 2012, p 233) “Structure” is also taken widely, having a range of applica-tions, as it does in syntax In particular, it includes the abstract sense in which social structure, for example, is thought of as consisting of class and other relationships; and

it includes the more concrete sense in which we may itemise individual structures.The study of semantic structure is to be linguistic That will exclude sociolin-guistics, conversation analysis, and discourse analysis, and paralinguistic features (e.g volume, tone of voice, speed and voice quality) The book also excludes “dynamic” semantics, dealing with the building of a whole text, utterance by utterance, as in Discourse Representation Theory; see Kamp and Reyle (2011), for instance Further, the study will be limited to English, although my knowledge of French and my read-ing about other languages indicates that what is true of English semantic structure is true widely

Researching this book has led me to change my understanding of language, ticularly in how it carries meaning, and in how meaning is structured in expression

par-A secondary goal of the book is therefore to persuade readers to adjust their derstanding if necessary – crucially, perhaps, in how they understand “semantics” and “syntax”

un-1.2 Approach

Approach to language The book will treat language as it exists in use; meaning will be

taken as contextual, and the argument will be empirical, citing utterances as evidence Language is assumed to be functional: each utterance is presumed to have a cogni-tive, personal or social function, which may be deliberate or only semiconscious; and

language itself is seen as functional in the evolutionary sense of being adapted to a beneficial result, representation of knowledge being one of the functions – only one

of them Language is also seen as functional in the further sense that it involves erations, with “input” of some sort, and the “output” of speaker utterances or hearer response

op-The approach to language is also semiotic, rather than logical (to do with ing) That seems necessary to deciding what “meaning” is, since conceptions of it vary widely The fundamental sense of “meaning”, according to the Shorter Oxford English Dictionary (“SOED” hereafter), entails being expressed “by a sentence, word, dream, symbol…”; that is, meaning is the meaning of a sign (I have adopted in my own thinking, as a rule of thumb, the maxim, “No sign, no meaning”.) Further, we need to be conscious of the nature of linguistic signs; we should not assume that all signs are symbols, giving full expression or representation of the meaning; the SOED definition quoted just above says, more fully, that meaning is “…expressed or indi-cated by a sentence…”; signs may be mere indications of the meaning Finally, the semiotic approach reminds us that meaning in its broad sense may include intention; the SOED definition of the fundamental meaning of “mean” as a verb reads, “Have as one’s purpose or intention…”

think-Approach to studying language The aim of studying language is taken here to be not

only description but explanation By analogy, it may be explanatory in studying the speech organs to consider not only their anatomy, but their function, their mode of operation, and their development Similarly, in studying the semantic structure of speech, it may be explanatory to call on psycholinguistics, neurolinguistics and de-velopment studies – of both language itself and children’s use of it The explanation should, moreover, be comprehensive – complete

Because there is no consensus on the nature of semantics, we should be larly careful of assumptions – for example, the incompatible assumptions made by the various approaches listed by Koptjevskaya-Tamm (2012), in which meaning is either (1) decomposable, descriptive and absolute, or (2) non-composable and denotational,

particu-or (3) always associated with a certain construction

I have accordingly tried to avoid commitment to particular theories and systems

of description The approach taken is in fact close to Systemic Functional Grammar in syntax, as in Halliday (2014), and to Cruse (2011) in semantics Readers who distin-guish between semasiological and onomasiological approaches should note that the approach here is semasiological – concerned with how language carries meaning, not with what words assigned to particular meanings

1.3 Argument of the book

Content of the argument The nature of syntactic structure will be familiar to readers,

perhaps as a hierarchy represented in tree diagrams, or as dependency structures The

nature of phonological structure is also clear, with a hierarchy of units from tion groups down through metrical feet to syllables and phonemes But there is, to

intona-my knowledge, no overall structure of semantics presented in the linguistic literature Three uses of “semantic structure” may be noted To Langacker, it is conceptu-

al structure, without any distinction between concepts and meaning (1987, p 99); there is an “unequivocal identification of meaning with concepts”( 2005, p 164); other cognitive linguists take much the same view; so does Jackendoff (1972) Croft (2001,

pp. 19–21 in particular) refers to both “semantic structure” and “semantic structures”

in constructions; but such structure is not distinguished from the syntactic structure which determines the construction; semantics is in effect simply conceptual for him, also In also these instances, “semantic structure” is conceptual relations, such as those between entities and properties predicated of them, or those of a “semantic field”, i.e a group of concepts or things in the same cognitive domain Evans (2009) sees semantic structure as consisting of cognitive representations which are specialised for use in language (2009, p 42), but does not explain what kind of structure it is, or what spe-cific structures there are; the only distinctions he gives are between concepts such as MATRIX, DURATION, and EVENT, which are not specialised for language

Moreover, those discussions of semantic structure do not even include the mantic structure of the relationships among synonyms, antonyms and so on (Those relations are not usually thought of as semantic structure, but as “lexical relations”, for example.) We need an account that integrates those elements into a comprehen-sive structure, and provides a rationale for the integration The main argument of the book, then, is that semantics in English does have structure, and the substance of the book is an exposition of what that structure is

se-Form of the argument Since there is no generally accepted structure or set of data, the

argument here cannot be simply deductive, or inductive Instead, it will work from accepted specific understandings, such as the fact that the relation of modifier to head

is semantic as well as syntactic, and from data introduced for the purpose It will then work by inference and generalisation from those bases, using other concepts, used in some fields or approaches to linguistics but not used universally Although the con-clusions may be new, and in some cases contrary to conventional views, it is intended that both the starting points and the argument from them be all conventional and acceptable

Development of the argument The argument of the book begins, then, from the

rela-tions among synonyms and antonyms Clearly, they are often related by having the same conceptual meaning, or by being distinguished by their concepts However, the

differences are sometimes not conceptual; famous and notorious, for example, both

denote the concept, ‘well-known’; they differ only in that they express different

at-titudes to being well known – favourable and unfavourable Words like lovely and horrible generally have no conceptual meaning, but are meaningful: we must allow

for emotive meaning Those three forms of meaning – conceptual, attitudinal and emotive – may be grouped as content meanings; but some meaningful words have no

content, as with utter and sheer, in “utter rubbish” and “sheer nonsense” They

inten-sify the word they modify; that is certainly their significance, although it may not be what we usually call “meaning”; since it is significance, semantics must account for it Those words are “grammatical items” or “functional items”; they contrast with “con-tent items”; since the latter have content “meaning”, can we not say that the former have “grammatical meaning”? In the first stage of the argument, then, we see a struc-ture of types of meaning; and simple, relatively obvious observations lead naturally

to an unfamiliar, and perhaps uncomfortable, conclusion – that there is such a thing

as grammatical meaning (with the relation of meaning to function, and of the nection between grammatical relations and grammatical meaning yet to be clarified) Further reflection leads to the familiar fact that word meanings have an internal structure, sometimes thought of as definitions, as in dictionaries, and sometimes as

con-lists of features, as when the meaning of stallion is represented as HORSE + ADULT +

MALE Those conceptual elements are used in many words, and are related in various ways; they must be seen as constituting a network, which has nodes with many links each Networks are a major semantic structure in English

The relation between syntax and semantics has been controversial, but there is now general agreement that the relation between them is very close, and approxi-mates interdependence There is a semantic structure “underlying” the syntax, or

“embodied” by it, or “interpreted” from it; its relations are not those of Subject, verb and Object, but must parallel them, perhaps as “semantic roles” However we describe the elements of the structure, it must have the same form as the syntactic structure, namely a hierarchy – which constitutes a second major semantic structure in English

It is now widely agreed that English has “information structure” or “information packaging”; but its place in grammar is not agreed However, it seems to be clearly a matter of content rather than of syntax; we conclude that it either is semantic, or has

an important relation to semantics Our account of semantic structure must find a place for it

That provides a very brief outline of the argument of the book It begins, I believe, from familiar and accepted premises; it will end with some conclusions that will be similarly unsurprising, but with some others that are likely to seem controversial or even strange That is particularly because of detail which has been passed over so far.The following questions to be answered will give the flavour of the detail What is the semantic significance of being a Subject? In what different ways can the meaning

of an adjective modify the meaning of its head word? If we allow grammatical ing, what is its nature, and what is its structure? Can semantics resolve the apparently syntactic arguments about the difference between Complements and Adjuncts, for example? Expression of meaning in words is sometimes called “coding”, but what does

mean-it entail beyond a one-to-one conversion, like coding the letter S into Morse code as dots and dashes? The substance of the book lies in such detail, rather than in the ar-gument outlined above

The aim of the book, then, is to set out an orderly explanation of the structure

of meaning in English, with those and other questions resolved, and comprehensive enough to find a place for all that is valid in present semantic thinking

1.4 Plan of the book

Outline The book has been planned to make the argument easy to follow, and to be

persuasive For example, unfamiliar concepts are presented through examples and through use, with abstract definition delayed until the concept is clear The important generalisations are given after the empirical detail on which they are based, rather than being given first, without context

Thus, Chapter 2 surveys familiar knowledge of English, looking for relatively vious instances of semantic structure, building up some of the concepts and technical terms that will be used throughout later chapters, and – perhaps most important – bringing out phenomena that need explanation through semantic structure

ob-Chapters 3 and 4 formalise the concepts introduced in Chapter 2, and add others, building the foundation for the remainder of the book Chapter 3 deals with the basis

of semantic structure in the intentions from which meaning rises, and in the tions which it serves Chapter 4 outlines the elements from which semantic structure

func-is built, such as the types of meaning (see above), conceptual elements of meaning, and the dimensions which define them Chapter 5 discusses the network structure built up by those elements and dimensions, chiefly in senses and their relations Chapter 6 explains the processes which restructure the elements of the network into the other main semantic structure, the hierarchy which parallels the syntactic hierar-chy Chapters 7 and 8 set out the nature of that hierarchic structure, from the clause level down to the level of senses and their constituents Chapter 9 sets out the nature

of information structure, similarly Chapter 10 deals with other, less used structures That completes the analytical explanation of semantic structure; it is complemented

in Chapters 11 and 12 by a discussion of the structure of realisation; that is, the way

in which specific messages are formulated out of the vast multi-dimensional network

of potential meaning, formed into words, organised into utterances, and delivered

in speech or writing Chapter 13 discusses implications from the previous chapters which are relevant to our understanding of linguistics, but outside the narrow scope

of the book Chapter 14 summarises and draws conclusions

Conventions observed Concepts are printed in SMALL CAPITALS Meanings are printed

in single quotation marks – ‘……’ Words and phrases being discussed as utterances are printed in double quotation marks – “……” – whereas words being discussed as

words in the language are printed in italics That avoids the clumsiness of continually

saying “the word ‘table’” and “the meaning ‘table’,” but can cause its own difficulty

when the reader must keep in mind that “Table is…” (i.e the word is…) does not

mean the same as “ ‘Table’ is…” (i.e the meaning is…) Initial capitals are used where

needed to distinguish technical from general terms, as in “The Subject of the tence…” and “….a subject for investigation”.

sen-Quoted utterances are put on a new line as numbered examples when they are to

be given some discussion, but are otherwise run into the text

Semantic structures in the strata of English

2.1 Introduction

Purpose and approach of the chapter Like Chapter 1, this chapter prepares for the rest

of the book It examines the strata of English, identifying some semantic structures

to be described fully later, identifying structural and conceptual elements which will aid that description, and noting some things that need further explanation and some that have no familiar explanation at all The approach is exploratory, looking for rep-resentative instances, leaving comprehensive treatment to later chapters The purpose

is thus to persuade readers fully that the book is needed, and to indicate what they can expect in the following chapters Phonology has been given more space than the other strata, because its semantic significance is much less known, and to allow later chapters to draw on its explanations

Arrangement The remaining sections deal in turn with the strata of lexis,

morpholo-gy, syntax, and phonolomorpholo-gy, in relation to semantics There is no developing argument; rather, points are made independently, and the conclusion will be general

2.2 Semantic structure in lexis

2.2.1 Paradigmatic lexical relations

We begin the study of semantic structure in lexis by considering the words that fit into paradigms of alternatives – sets of synonyms, antonyms and so on Comparing closely

related lexical items such as sob, whimper, weep, keen, and blubber shows that word

meanings seem to be built up from elements For example, all those words include ‘to

manifest pain, misery or grief’ (SOED, on weep); sob adds the element ‘convulsively’, and whimper adds ‘feebly’ and ‘intermittently’ That analysis shows that weep has a

general sense, and that the other senses are more specific That is a general pattern; senses differ on the dimension of “generality” or “specificity” Dimensions will be im-portant in setting out semantic structure

Those conceptual elements do not exhaust the differences between synonyms,

however We use weep in formal contexts, and blub and blubber in informal ones, in my

judgement at least – SOED does not specify their usage That is part of the significance

of the words, though it may not feel natural to say that it is part of their “meaning”

We use whimper and blubber to convey disapproval: attitudinal “meaning” is another aspect of words’ significance Finally, to keen has Irish “connotations”, which must be

accounted for somewhere in our description of meaning Among other relations, each

of the synonyms considered has its antonyms; and weep is a hyponym for the others

The book will argue that these forms of significance are indeed part of meaning, and will be referred to as “types” of meaning

The types and dimensions of meaning make “planes” on which word senses are related, additional to the conceptual elements The natural term for such for such a set

of patterns on several planes is “network”, which is one of the fundamental semantic structures of English

2.2.2 Syntagmatic lexical relations

2.2.2.1 Grammatical and content items

When we compare words in their syntagmatic arrangement, a more familiar tion quickly arises, the distinction between “lexical” or “content” words and mor-phemes, and “functional” or “grammatical” ones It is not obvious whether that is a semantic contrast, or a contrast between semantics and grammar Nor is its nature made clear in the literature: lexical and grammatical items are said to constitute class-

distinc-es, but the classes overlap, with many words belonging in both The book will guish between content meaning and grammatical meaning, as types of meaning They need thorough examination; for example, prepositions and conjunctions link chunks

distin-of content, so is grammatical meaning structural? What do content meaning and grammatical meaning have in common? It will be argued that grammatical meaning

is fundamental to semantic structure

2.2.2.2 Semantic classes

If we consider repeating patterns, as in such utterances as “Roses are red; violets are blue; sugar is sweet, and so are you”, it seems obvious that meanings belong to seman-tic classes – of “entities” (‘roses’, ‘violets’, ‘sugar’) and “properties” (‘red’, ‘blue’, ‘sweet’)

“Events” (‘ran’, ‘broke’, ‘fell’) are an obvious addition Is and are are not so

straightfor-ward – do we need to posit a relationship class?

2.2.2.3 Semantic classes and cognitive classes

Problems emerge very quickly, however Relationships are so different from the bers of the other three classes, which are clearly based on the way we experience the world directly, that we must enquire whether the classification is valid Second, it seems doubtful that there should be semantic classes of kinds of things, such as flow-ers, grasses, weeds and so on; those are horticultural classes, and part of our knowl-edge of the world, not of grammar; so we need to resolve the nature of being semantic, including the relation between the linguistic (e.g semantic) and cognitive (e.g hor-ticultural) Since the classes of things, properties and so on support the classes of

mem-nouns, adjectives and so on, we will need to examine those as well, if we think they are relevant to semantics.

2.2.3 Words without paradigmatic or syntagmatic relations

“Gone” and “More!”, for example Adult examples include discourse particles like

well, general purpose rejoinders such as “Cheers!”, the ironic response, “Really?” to

the remark, “I got an A grade”, and the response “Oh, no!” to the remark “I’m stuck again.” They are not meaningless, but it does not seem possible to give their meaning

a dictionary- style definition

Both groups of words pose problems which an account of semantic structure must resolve Since it is easier to state an intention or function for these unstructured items than to state a meaning, we should examine the relationships among meaning, function and intention

2.3 Semantic structure in morphology

2.3.1 Semantic classes

There are clear patterns in the suffixes which derive new words from base words The

suffixes -ise/-ize and -en form event senses; -ion and -ness form entity senses, and -ous and -ful form property senses That supports the earlier suggestion that semantic

classes underlie semantic structure But this derivational morphology sets a problem

for the concept of semantic classes For example, stupefy means ‘to induce stupour’;

the sense seems to include a thing as well as an action Does the sense belong to the action class (‘induce’) linguistically, but include a member of the thing class (‘stup-our’) cognitively? The problems noted in the previous section increase

2.3.2 Grammatical meaning

The number and gender inflections that form the agreement between Subjects and Predicators raise other issues The relation of agreement is taken to be “grammatical”, rather than semantic; but it does not seem to be meaningless, although any meaning

is certainly not like that of content words Perhaps agreement is a grammatical item, with grammatical meaning? That, too, needs resolution

2.3.3 Dependency and modification

Prefixes such as pre- and sub- had distinct content meaning in the languages from which they were borrowed, but do not seem to have such meaning in words like pre- side and substantial On the other hand, mega- and mini- clearly are meaningful in present- day English, since they have currency as independent words; un- must be

meaningful for “He was a shy man, but he had to un-shy himself” to be meaningful

Presumably then, words such as megastar and mini-rugby have the semantic

struc-ture of modifier and head – to be noted in the next section It will be hard to resolve whether the meaning of words with several morphemes is compositional in structure.Suffixes affect the meaning of the root in different ways Some change the part of

speech, sometimes adding conceptual meaning as they do so, as with -esque (‘like’);

some change the word grammatically (in tense or plurality, for example); and some

add a scornful or other emotive quality (lordling, princeling) Suffixes, as a single

mor-phological structure, represent several different semantic structures

2.4 Semantic structure in syntax

2.4.1 Semantic structure of clauses: The figure

2.4.1.1 Constituents of figures

Structural constituents Following Halliday (2014), the syntactic structure of clauses

is taken here to be that of Subject + Predicator + optional Complement +

option-al Adjunct As syntax, that structure is taken to concern the order of the units, not meaning Part of their function, however, is to realise semantic units, as follows In unmarked1 use, the Subject realises whatever “thing” the Predicator is asserted about, such as the person who carries out the action; Complements realise whatever else is centrally involved in the predication Semantically, then, they are participants directly

involved in the event (Marked uses include the use of there and it as dummy Subjects.)

The Predicator realises an event or relationship, which will be called the “Process”

1 “Marked” here and elsewhere means “contrasting with standard or rule-bound use”; only

rarely will it mean “carrying a distinctive marking”.

Adjuncts realise meanings which will be called “Circumstances” (“adverbials”, ly) Thus, there is a semantic structure closely paralleling the syntactic structure, with units which are of equal status – i.e interdependent rather than dependent – and which are components of a larger unit, the “figure”, the approximate equivalent of the clause (“Dependent” will be equated with “subordinate”; dependency will not be an important concept, as it is in dependency grammar.)

rough-Class constituents The semantic constituents just listed seem to fall naturally into

se-mantic classes, in simple figures at least The “Participants” either are entities or are very similar to them; “Processes” are usually events; “Circumstances” are properties in

a broad sense, or a combination of relationship + Participant (“prepositional phrase”) However, the concept of semantic class will need further modification First, it must

allow for Predicators like is which are not events Second, it must allow for the

seman-tic roles in the figure, such as being the actor (as in “Mr Barnett lit a fire”) or being a term of the predicating relationship (as in “Mr Barnett had been the captain”) Third,

it must allow for changes in class, or at least changes in role, as words are structured

in groups

I take the substance of these “roles” to be semantic, although it is often treated as being syntactic: the roles are sometimes called “semantic”, but are often called “gram-matical relations”, or “thematic roles”, and so on If we accept that grammar comprises phonology, syntax, morphology and semantics, then these roles must be semantic

2.4.1.2 Relationships among figure constituents

Transitive relations “Mr Barnett lit a fire” means more than ‘lit’ + ‘a fire’ + ‘Mr Barnett’

For example, it includes the elements that Mr Barnett carried out an action, and that fire appeared as a result Those elements are parts of a semantic structure, which is created, it seems, by the figure’s transitivity An account of semantic structure must explain how such structures are generated

Relations with a copula Copulas – as in “roses are red” – do not denote any event or

action, or entail any duration in time or “instability in time” Copulas, then, are quite distinct from what “verbs” are commonly conceived to be, yet copulas can be “the verb” in a clause, or the “head of a verbal phrase” We need an explanation of their semantics

Dimensions: specificity There are several ways of interpreting “Sometimes I’d see

someone I knew on the bus and we’d wave” (Corpus of Contemporary American English – “COCA” hereafter) The figure can “mean” that I was on the bus, or that the other person was on the bus, or that both were Perhaps we should say that the figure does not specify who was on the bus; specificity would then be an important dimension here, as well as in the lexical stratum In any case, an account of semantic structure must explain why Adjuncts such as “on the bus” pattern so differently from the other constituents of the figure – with so much variability – and what controls their interpretation

2.4.1.3 Significance of group order

Order of constituents Uttering “He is.” is an assertion; uttering “Is he?” is a question

Subject-Predicator order signifies asserting; Predicator-Subject order signifies tioning The order of syntactic units can thus carry significance, just as words can; and the difference between word and group carries significance, as we saw in the section

ques-on class cques-onstituents To account properly for semantic structure, we must include this semiotic issue: if both words and word order are signs that can carry significance, what is the range of meaningful signs in English?

Asserting and questioning are semantic (to do with significance) rather than tactic, but in a quite different way from expressing the conceptual content that is as-serted or questioned As the common term “speech act” indicates, they are instances

syn-of language as action – or interaction between people – rather than syn-of language as coded information Exclamations are not exactly either of those alternatives; rather, they are expression – expression of feeling It is likely that these differences in func-tion and in the intentions behind them affect semantic structure

We can say that the Predicator-Subject order in “Is he?” sets the interrogative mood, in the grammatical system; but we also need to say that for the speaker it rep-resents a speech act, and that for the hearer it is a request for a reply Utterances, then, have three aspects, all of which should be allowed for: the speaker’s, the hearer’s, and the grammatical or “system” aspect That probably unfamiliar three-way distinction will be maintained throughout the book

In speech acts, speakers are guiding the hearer as to the response they want, and every utterance constitutes a speech act The significance of language includes guiding the hearers in understanding the utterance and responding to it

Order of content There is one other type of significance which hearers understand

in clauses, for which the terms “Topic”, “Comment”, “Theme”, “Rheme” and perhaps others are used (The Theme is the first information, orienting the hearer; the Rheme develops the rest of the information from that starting point.)The issues go under the name of “information structure”, “information packaging”, and so on The va-riety of terms reflects both different understandings of the structure, and different understandings of how it relates to “semantics” This book must attempt to resolve those issues

If we pursue for a moment the semiotic issues involved in Topics, we quickly become puzzled We can identify “topicalising constructions” that signify a Topic, but

we cannot find linguistic forms that signify Comment Similarly, there are no tic forms to signify degrees of Rhematic importance (By that, I mean increase of in-formation value leading to a focal point, usually at the end.) On reflection, we see that

linguis-we interpret part of an utterance as Comment because it follows the Topic (in default structure), and (again by default) interpret the order of information in the Comment

as representing increasing importance Order of content, as well as order of syntactic units, carries significance

Conclusion Order of words and of content generally is a necessity in speaking and

writing, but we conclude that it has been assigned meaning of its own We also clude that just as syntactic groups and clauses form a hierarchic structure, so do se-mantic groups and the figure they constitute

con-2.4.2 Semantic structure in groups

2.4.2.1 Groups as part of the semantic hierarchy

Groups consist of modifiers and heads, semantically as well as syntactically The group has constituents, then, the modifiers being dependent on the head That continues the hierarchic structure we noted in figures, and it continues in the morphological structure of words, discussed in the previous section We conclude that hierarchy is a major structure in English semantics, along with network (§2.2.1 above); analysing it will be a major task for the book

2.4.2.2 Modification structure

Modifiers in nominal groups raise several issues for semantic structure Commonly, they belong to our proposed semantic class of properties, but premodifiers such as

fake and alleged do not (being alleged is not a property of an alleged criminal); do they

denote a meaning, or carry out a function? Is there a semantic significance to the fixed order of premodifiers? What is the significance of the fact that words change their po-sition according to their meaning? (Compare “a high positive predictive value”, “mild high blood pressure”, and “a typical suburban high street” – the meaning changes as the word’s position changes.) Where we used to say “patients’ entrance” in the plural and possessive, we now commonly say “patient entrance”, apparently singular and not morphologically possessive; what is the semantic significance of that? Even the maxim that modifiers modify the head of the phrase can apparently be falsified, as when a tramper reports that “our happy feet danced from boulder to boulder” – it was the trampers that were happy, not their feet, and “happy” apparently modifies the ‘us’

in “our”

2.4.2.3 Compositionality

Modification like that of fake and alleged, and like that of happy, in the examples

above, poses considerable problems for explaining the compositionality of semantic structure, since they do not seem to follow the rules for composing meaning in nom-inal groups

2.5 Semantic structure in phonology

2.5.1 Introduction

Basically, phonology is simply the way in which meaning is made audible to hearers This section deals with ways in which phonology adds to the meaning of the morpho-syntax which it makes audible – just as word order adds its own meaning

The vocal sounds of speech have a number of dimensions, including pitch (both change of pitch and absolute pitch), volume, speed, duration, and voice quality Of those, pitch has been conventionalised (in the tone system especially), and so has stress (as a combination of pitch, volume and duration) Those dimensions are used

to constitute four major phonological systems (Halliday and Greaves 2008, p 210) They are as follows

– Tonality, “the chunking of words into groups or phrases” (Wharton 2012, p 567).– Tonicity, the location of the “pitch accent” or “main stress” in the chunks defined

by tonality (Wharton 2012, p 567)

– Tone, the type of “melodic” pitch contour on the main stress

– Rhythm, the “distribution of utterances into feet (metric units) with location of

boundaries” (Halliday and Greaves 2008, p 210).

All four are here included in “intonation”

This section will give more detail than the previous section did, because the mantics” or significance of phonology is much less discussed in the literature

co-to convey

A tone group may consist of a single phrase or even a single word; in informal speech especially, “Yes” and “Six o’clock” are quite acceptable as tone groups, because

speakers and hearers consider them complete as items of information Tonality works

in information structure

Phonological paragraphs The tonality system also creates phonological units

consist-ing of several tone groups; each such unit creates a semantic structure consistconsist-ing of several information units (Tench 1996, pp 23–24) It is the functional equivalent in speech of a paragraph in writing, so may be called a “paratone” (Tench 1996, p 24, citing work by Brown) In unmarked form, paratones begin on a high pitch, both on the initial syllable and on the initial tone group The pitch falls fairly steadily to the end In formal style, it is often followed by a pause

Affective meaning of tonality An unusual degree of rise or fall in the overall pitch

con-tour conveys an affective meaning; that is, it conveys emotion or attitude Examples are “I see!”, uttered as a sigh, a groan, or as an indignant exclamation

2.5.3 Semantics of tone

Introduction Phonetically, the system of tone is made up of seven tones (Halliday

and Greaves 2008, p 45), each being a pattern of change of pitch There are five ple” or basic tones: tone 1 is a fall; tone 2 is a rise; tone 3 is level; tone 4 is a fall and rise; tone 5 is a rise and fall There are also two compound tones Cruttenden (1997,

“sim-§3.7) gives a similar analysis into seven “nuclear tones” There is a natural significance

of falling pitch as completeness or finality, and of rising pitch as incompleteness or lack of finality – that is, that the speaker’s speech intention is complete or incomplete (Halliday and Greaves 2008, p 112)

The tone system’s main significance is that it defines which speech act is intended (assertion, question, command or exclamation) for the information unit marked off

by tonality

Basic significance of the tones According to Halliday and Greaves (2008), the basic

significance of the commonest tones is as follows

– Tone 1, the falling tone, is a default tone used for tone groups whose speech tion is marked clearly by the syntax, so it has no significance of its own

func-– Tone 2, the rising tone, marks tone groups as questions

– Tone 3, the level tone gives the feature of neutrality or lack of commitment; it can

be taken as meaning “I decline to choose between the alternatives.” Specifically, it can show that the speaker is suggesting rather than asserting, or that the speaker

is agreeing with the interlocutor, rather than asserting something of his or her own, and so on

– Tone 4, the fall-rise combination of certainty and uncertainty, is used to give a feature of reservation; it means that the statement seems certain but is not so in fact: “There’s a ‘but’ to this!”

– Tone 5, rise-fall, signals that the tone group is an exclamation

Other significance conveyed by tone Tone is also important for conveying emotive and

attitudinal meaning Speakers use the rising tone’s sense of incompleteness in high- rising tone 2 to express an attitude or “feature” of questioning, or of challenging the interlocutor’s assertion (Halliday and Greaves 2008) The compound tone 1 followed

by 3 is used with a negative imperative such as “Don’t do that!”, to convey a feeling of urgency, or an emotion of anger (Halliday and Greaves 2008, p 119)

Cruttenden (1997) notes that tonal variation is often a marker of style, and that it

is sometimes an index of the speaker’s class and sex One well studied example is the

“high rising tone” or “uptalk”, which is a rising tone ending a statement which in itself seems complete At first (in the 1970s), it characterised the speech of young Austral-ians, especially females and those in state schools; later, it spread to other places and social groups (Cruttenden 1997, §5.2) Thus tone sometimes conveys social mean-ing – social identity, in this instance

Tone 3 can be used to signify that the speaker’s speech act is not finished yet, the finality of the questioning or asserting being still to come It can be taken to mean

“Please wait” Thus it carries a kind of grammatical meaning, guiding the hearer’s sponse Another kind of grammatical meaning is conveyed by some marked variants

re-of the tones, that re-of emphasising or intensifying the content re-of a word A fairly iar example is the use of marked pitch, which is a phonological alternative to morpho-

famil-logical intensification, e.g bigger or very big, and lexical intensification, e.g huge The

effect is to shift the sense along its intensity dimension “I’m taking the train this time” can be spoken with tone 1; the falling tone spreads over all of “train this time”, and there is a single significant stress (on “train”) It could, however, be spoken with the falling tone (tone 1) restricted to “train”, and with a level tone (tone 3), with secondary stress on “time” (Halliday and Greaves 2008, p 45; it is a “compound tone”, “tone 1,3”) The semantic difference is that there is now a secondary focus in the tone group; that makes “this time” a subordinate assertion within the main assertion, because fall in tone after the focus on “train” marks that information as unimportant – which is part

of information structure The compound tone guides the hearer’s interpretation, as a grammatical meaning

2.5.4 Semantics of tonicity

Focus Like tonality and tone, tonicity works primarily for information structure The

main or “tonic” stress marks the end of the information unit, and highlights the most important part of the information, the focus

Rhematic structure As noted above in §2.4.1, the Rheme of the information unit

con-sists of a sequence of peaks of importance Those peaks are signalled by the phrase stresses, culminating in the focus, as in example (1), above: “…say that for the first time in my life, I’ve been in hospital.” (Couper-Kuhlen 1993, p 94 The underlined stresses, most of them in marked position, are as recorded by Couper-Kuhlen.)

Relation to other strata Tonicity works with other strata, too It works with the lexis, especially with focusing words such as alone, only, especially, and even, in marking

what is important; and it works with emotive or otherwise forceful words It also works with syntax, since the order of postmodifying groups and of Adjuncts helps

in defining the importance of information Semantic structure has the strata of guage sometimes working together, reinforcing each other, but sometimes working

lan-as alternatives

2.5.5 Semantics of rhythm

Basis of rhythm Speech is naturally rhythmical to some extent, because of its basis in

human physiology, but it gains semantic significance when speakers make it regular enough to be noticeable Chunks of text, usually several feet long, are “rhythmised” (Couper-Kuhlen 1993, p 98), by the weakening or strengthening of some stressed syl-lables, the speeding up of some syllables, and the use of pauses, until there is a pattern

of beats at approximately even time intervals The evenly spaced beats may be word stresses, phrase stresses, or even sentence stresses

The passage in Diagram 1 is from a radio interview; the woman speaking is ing the excitement she felt when she visited Withington Key to the diagram: the high-

shar-er the column of X’s, the greatshar-er the stress The horizontal line marks sections that were rhythmised

Withington was a name on the map [.4 sec.] Yes, because living in

North Manchester I knew nothing at all about Withington”

x x

x x

x x x x

x

x

x x x x x x x x x x x x

x x x x

x x x

x x

x

x x

x

x x

Diagram 1 Structure of rhythm (Couper-Kuhlen 1993, pp 95–96)

Significance of rhythm Conveying emotive meaning is a familiar use for rhythm; for

example, see Quirk et al (1985, p 1598) on its use for irritation and sarcasm It can be seen in the brief passage about Withington just quoted, conveying excitement – add-

ing to the effect of the marked stress, on the first Withington and on map, for example

Note that the rhythm as delivered was not what we might expect from the written text: speakers adjust the “natural” rhythm to their purpose

The significance of rhythm is often that it heightens meaning represented in other ways, by stimulating the imagination and rousing excitement accordingly That use is

so well known from oratory and poetry that it does not need illustration

As well as the meanings listed, rhythm often has a grammatical meaning which builds information structure The rhythm is reduced at the beginning of an utterance: unstressed syllables are spoken with greater speed and reduced enunciation The in-troductory phrase, “There’s a dreadful…”, may even be pronounced as ‘Th-dreadful” (Cruttenden 1997, p 21) Second, rhythm is the main marking of Rhematic structure (see Chapter 10) Less important content is spoken with lighter and quicker rhythm

An example from Halliday will illustrate the point

(2) “In this job, Anne, we are working with silver Now silver needs to have love”

(Halliday 2014, §3.5)

In the second sentence, only the word love was important, so “needs to have”, the first

part of the Rheme, was spoken with light, quick rhythm

2.5.6 Semantic structure in phonetics

Introduction Speaking is a physical activity, using organs that have other functions

We vocalise on the out-breath, which seems to be the basis for tone groups, with the in-breath separating them; perhaps the syllabic structure of words is partly derived from rhythmic mouth movements as in chewing When we speak more forcefully, we use extra force from the lungs; states of emotional tension go with states of physical tension, and expression of those states uses tense speech muscles This section argues that the physical features of speech, and of articulation especially, often have expres-sive significance – that phonetics, as well phonology, can have semantic significance

Meaningful phonetics in English Differences in regional and social dialects consist to

an important extent of differences in the vowel quality, and other phonetic features Phonetics is thus meaningful in conveying to the hearer the regional or social back-ground of the speaker, as noted in §2.5.3

There is a much wider and perhaps more important issue here, however matopoeia is accepted universally as an example of meaningful phonetics However, its nature as an imitation of natural sound is a widely misunderstood, I believe For example, its value is often attacked on the grounds that onomatopoeic expressions vary from language to language, and even within a language; thus English has both

Ono-“bow wow” and “woof woof”, and French has the equivalent of “wa wa” But the tation lies, not in the phonemes, but in the phonetic structure – in the sudden onset

imi-of the syllables, in the vowels being made at the back imi-of the mouth, and the repeated strong beats; and the effect lies, not in those phonetic abstractions, but in the artic-ulation: we articulate the sounds abruptly and forcefully in the back of the mouth, imitating dogs’ articulation

That physical nature of speaking the words explains why so many interjections

and swear words use plosive /b/ and /d/ (blow, bloody, blast, damn), or guttural /k/

or /g/ (God! Christ!) It explains odd little things like adding /p/ to yes and no, in yep and nope: closing your mouth to finish the word adds finality All that being so,

it is natural for us to give a little more emphasis to the sounds in our words: to give

a stronger outburst at the beginning of brutal and please, and to lengthen the vowel

in please Emphasising stop consonants leads to giving them extra aspiration

Con-sequently, “Oh T[h]om!” expresses exasperation, the phonetic value having emotive meaning; and “They spent b[h]illions on it” carries intensification – the phonetic value has grammatical meaning, and is an alternative to using an intensifying word, such

as the grammatical item, very Extra aspiration was used in the passage in Diagram 1,

“Withington was a name on the map Y[h]es Bec[h]ause living in North ter…” (Couper-Kuhlen 1993, p 95)

Manches-Conclusion: semantic structure of the phonetic stratum This use of the phonetic value

of English sounds, independent of their phonemic value, makes an important tribution to the semantics of English, although it is sharply reduced in formal style See also Halliday and Greaves (2008, p 169) on “phonaesthesia”, and the literature on

2.5.7.2 Field-like structures

The significance of a fronted /e/ or rolled /r/, and of any other socially marked

phonet-ic feature, is not that the one word or phrase is spoken by a person from a partphonet-icular social group; the significance is that the whole utterance, or the whole text, is spo-ken by such a person The feature creates a “field” which affects the whole utterance,

as a magnetic or gravitational field affects everything in it That structure is

radical-ly different from both the hierarchies and the waves studied so far As with waves, the structure has been noted before, by Halliday (2014), building on work by Pike Halliday distinguishes meaning as field from meaning as units or “particles” (building

a hierarchy), and from meaning as wave, making an analogy with electromagnetic energy, which occurs as particles, waves and fields Waves and fields are also basic structures which will be studied in the rest of the book

2.5.7.3 Basicness

Children use phonology expressively in the first weeks of life, gradually ing their cries to the point where adults can distinguish a cry of hunger from a cry of pain Rise and fall structures are thus more basic than hierarchic structures, and the expressive function is more basic than the function of representing information con-ceptually That explains various things, such as the fact that phonology can override syntax, as noted above, and the fact that we can make meaningful utterances such as a prolonged “A-a-a-h” with phonology but no syntax, whereas we cannot speak syntax without phonology Basicness is a semantic dimension – one of some importance

differentiat-2.5.7.4 Compositionality in phonological semantics

This section discusses the much debated issue of compositionality: whether the lations among sense elements construct the whole sense by regular processes, as (6+2) × (2+1) constructs 36 (Applied to words, the issue is sometimes referred to

re-as “decompositionality”.) The issue is not important to the argument of the book, cause, in the view of meaning as functional and being structured as a network, there

be-is no reason why meaning should be fully compositional (as will become clear later in the section), and as mathematically stated meaning should be I discuss it because it

is important in the literature – there is, for example, a whole large handbook on positionality alone (Werning et al 2012) – and because the discussion will illumine semantic structure from another point of view

com-Superficially, the phonology of English is compositional: phonemes combine into

a syllable, syllables combine into feet, and so on Semantically, there is very little positionality in the significance of phonological patterns In its prime function, tonal-ity creates a unitary tone group; similarly, tonicity creates a single tonic stress; tone,

com-as a pattern, is a unitary structure, and rhythm hcom-as a unified overall effect But there

is some compositionality in the way that secondary stresses make up Rhematic ture, in the compounding of tones, and in the modification of tones by secondary tones (Halliday and Greaves 2008) That leaves the role of compositionality in English still less clear than it is in syntactic semantics: it will need study

struc-2.5.8 Conclusion: Semantics in the phonological stratum

2.5.8.1 Structure

The relations of phonology to syntax and lexis are complex In syntax, words, groups and clauses make a distinctive and unified structure, and build a similar unified se-mantic structure Phonology, however, has no such unity or consistency; its units are relatively independent in nature and function, remaining close to their origin, where-

as lexis and syntax are highly developed, specialised and integrated Whereas syntax organises and complements the meanings represented by words, phonologically ex-pressed meanings may reinforce a meaning already expressed in lexis or syntax, or be

an alternative for it, or override it

2.5.8.2 Meaning

Like syntax, phonology is a means for making word meaning transmissible to ers, but – again like syntax – it adds meaning to what is represented in the words Whereas lexis is focused on conceptual content, and syntax is focused on grammati-cal meanings concerned with structuring the conceptual content, phonology is con-cerned mainly with conveying feeling and attitude, signalling information structure, and very rarely with conceptual content In some marked uses, it has a meaning which

hear-no other stratum can convey

This discussion of phonology has confirmed what the discussion of tax suggested: that information structure is a regular and systematic part of semantics

morphosyn-in English – or at least of its significance if not of “semantics” – and that ical meaning is so also, being wide-ranging and of many types The discussion has brought out an even wider range of significance carried by language, extending, in rhythm for example, to what is sometimes an effect, rather than a definable meaning

grammat-2.5.8.3 Semiosis

Phonology is close to the physicality of language and its evolutionary origin; stress and tone, especially in their marked uses, have clear similarities to shrieks and other involuntary vocalisations, where volume and high pitch, for example, carry signifi-cance as “indexes” of the states that cause them, as smoke is an index of fire In a com-parable way, certain phonological patterns are indexes of the social group to which the speaker belongs, as we noted in the §2.5.3 In several places, it has seemed appro-priate to describe phonological features as signals – like monkeys’ danger signals, and birds’ territorial calls The unmarked uses of the four phonological systems, however, are conventionalised and formed into paradigms; so they are symbolic (They are not arbitrary, however, being motivated in the ways described in the section on phonetic meaning.) We see that meaning is carried by different kinds of sign, which is likely to affect the nature of meaning and its structure

2.5.8.4 Nature of language

It is impossible to see tone, tonality, tonicity and rhythm as primarily concerned with symbolising conceptual meanings Here, language is not representation, or the for-mulation of knowledge It is a matter of what the speaker is doing to affect the hearer From the speaker’s point of view, it is action, and from the hearer’s point of view it is responding to the speaker It is hard to formulate phonology or its significance as a system, from a strictly linguistic viewpoint Since phonology is basic, we should per-haps conclude that the basic nature and function of language is action

2.6 Conclusion: Semantic structure in the strata of English

Structures The chapter has identified several semantic structures in English Words

form a network structure Morphology and syntax build a different structure, with constituent units arranged in a hierarchical structure Phonology is quite different from the other strata in having semantic elements of different kinds; it does not as a whole fit any particular structure; and it also creates field and wave structures which carry meaning, and which are quite different from networks and hierarchies (The chapter has not discussed the obvious contribution of graphology to the transmission

of meaning.) The higher strata do not simply facilitate the expression of meaning ried by lower strata, but carry additional meaning

car-As well as those structures, which need to be set out fully, we have noted several structural issues which need special study There are binary relations of various sorts, constituting structure per se, as an abstract quality: structure, distinguished from (specific) structures There are unstructured units, such as holophrases and interjec-tions The nature of information structure is not clear (yet); nor is its relation to other structures

Useful concepts The chapter has introduced several concepts which will be developed

and applied further in the following chapters Examples include types of meaning, such as conceptual and emotive; dimensions of meaning, such as basicness and spec-ificity, and semantic classes It is useful to approach meaning from both functional and structural approaches, but also from semiotics There are several important dis-tinctions to be made: between different types of significance (such as content and grammatical meaning, the effect of an utterance, and what it shows of the speaker); between the cognitive and the linguistic in meaning; and between the speaker, hear-

er and system points of view or “aspects” Use, as in marked and unmarked use, is distinct from structure, but closely related to it Those concepts will be developed in Chapters 3 and 4, and applied in later chapters

Further analysis needed While the chapter has reached those useful conclusions,

sev-eral issues have been raised which are not very clear in the literature on semantics Specific issues include, for example, the relation between transitive figures and copula

figures, the number and nature of Processes, and the mechanisms that make meaning compositional They will be discussed in Chapters 6 to 10.

More general issues include the following What are the relationships among tention, function and specific meanings? What is the relation of surface meaning to underlying knowledge? Since syntax and phonology “realise” meaning so differently, how can realisation be characterised adequately? What is the range of semiotic signs

in-in English, and how do they affect semantic structure? Fin-inally, as to language itself:

we have seen that language is both interaction and transmission of information, so what else is it? Most of those issues will be given preliminary discussion in Chapters 3 and 4; most will be amplified in Chapters 6 to 10, and realisation will be discussed in Chapters 11 and 12

Looking ahead The next two chapters will develop distinctions and concepts

intro-duced here, and some others, to provide a full basis for the discussion of semantic structure in the chapters that follow