Publishing Partner: Cambridge University Press CUP Extra Publisher Login

New from Cambridge University Press!


Revitalizing Endangered Languages

Edited by Justyna Olko & Julia Sallabank

Revitalizing Endangered Languages "This guidebook provides ideas and strategies, as well as some background, to help with the effective revitalization of endangered languages. It covers a broad scope of themes including effective planning, benefits, wellbeing, economic aspects, attitudes and ideologies."

New from Wiley!


We Have a New Site!

With the help of your donations we have been making good progress on designing and launching our new website! Check it out at!
***We are still in our beta stages for the new site--if you have any feedback, be sure to let us know at***

Review of  Introduction to Natural Language Semantics

Reviewer: Niladri Sekhar Dash
Book Title: Introduction to Natural Language Semantics
Book Author: Henriƫtte de Swart
Publisher: CSLI Publications
Linguistic Field(s): Semantics
Issue Number: 12.1709

Discuss this Review
Help on Posting

Allan, Keith (2001) Natural Language Semantics.
Blackwell Publishers, paperback ISBN 0-631-19297-2, 525 pp.

Niladri Sekhar Dash, Indian Statistical Institute, Kolkata,


In Chapter 1 [Some fundamental concepts of semantics]
(pp. 1-39), the author Keith Allan presents a brief
introduction to some of the fundamental concepts for
semantics and semantic theories which are referred to
explicitly or implicitly throughout the rest of the
volume. Here Allan explains different areas of semantics
such as compositionality, inference, the functions of a
theory, and criteria for choosing a metalanguage to use in
semantic analysis, etc.

In Chapter 2 [Words and worlds and reference] (pp. 41-74),
Allan tries to build up the ideas he has introduced in
Chapter 1. Here he seeks to pay an intensive look at some
meanings of 'meaning'. He offers a new look at some very
often-quoted terms of linguistics such as 'sense',
'denotation', 'reference', 'intention', 'extension'
etc. which are used by a large number of people in their
daily conversations. In most cases such terms have to suffer
the fate of humiliation with occasional overlapping of their
meanings by the users.

In Chapter 3 [The lexicon and the encyclopaedia]
(pp. 75-105), Allan enters into a long-drawn debate on the
actual status of a lexicon in respect to an
encyclopaedia. We are sometimes confused to decide where to
draw a line of demarcation between them. Allan looks at the
structure of a lexicon, distinguishes a lexicon from an
encyclopaedia, identifies the relationship between them, and
proposes that the lexicon forms part of an
encyclopaedia. His analysis is definitely a good system to
dissolve our confusion.

In Chapter 4 [Morphology and listemes] (pp. 107 -145), Allan
continues on the lexicological themes he has introduced in
Chapter 3. Here he discusses what counts as a listeme and
what part played by the lexicon in identifying the
relationship between them. He investigates the scope of
listeme, explores the relationships between meaning and
morphology, argues if polyword forms like compounds, phrasal
verbs, and idioms can be included in listemes, and
investigates the form and function of onomatopoeic words in
the language.

In Chapter 5 [The power of words: connotation and jargon]
(pp. 147- 180), Allan looks at other sources for new
vocabulary. He examines the power of connotation in
motivating the choice near synonyms and the development of
new language expressions. His main focus is on the power of
words and on the semantic effects that arise from their
usage. In the chapter he also includes a discussion of and
explanation for jargon: kind of a language peculiar to
trade, profession or other group.

In Chapter 6 [Semantic relations between sentences]
(pp. 181-215), Allan introduces propositional calculus, a
method of mathematics and statistics, for semantic analysis
of sentences used in natural language. He closely examines
conversational implicature and presupposition within the
context of establishing semantic relations between
sentences. Moreover, he tries to redefine some fundamental
concepts used in establishing relations between sentences,
and presents some of the most basic formal tools for
analysing the meanings of sentences in truth-conditional

In Chapter 7 [Predicate logic, sets and lambda: tools for
semantic analysis] (pp. 217-245), Allan presents some formal
tools (borrowed from logic and mathematics) for
investigating the semantic structure of the constituents of
a simple proposition. Starting with predicate logic he moves
on to the usefulness of sets and functions to linguistic
semantics. He also considers other rigorous means for
analysing the internal semantic composition of clauses and
sentences. His aim is to identify some formal tools for
specifying clauses and their constituents semantically.

In Chapter 8 [Frames, fields, and semantic components]
(pp. 247-285), Allan actually started using such tools on
semantic analysis of sentences to verify their usefulness
for linguistic semantics. Here he examines different
semantic frames, semantic fields, and lexical analysis in
terms of semantic components or primitives. He also
appraises semantic primitives, and tries to understand what
a semantic description is meant to achieve. Finally, he
explores who or what a semantic specification is designed

In Chapter 9 [Cognitive semantics: backs, colours, and
classifiers] (pp. 287- 320), Allan focuses on the influence
of human perceptions and conceptions in determining semantic
categories, properties and relations of the listemes of
natural language. Here he explores the relation (both
explicit and implicit) between words, human perception of
them, and things spoken of. His study is expanded over both
language internally and across languages.

In Chapter 10 [Using the typical denotatum to identify the
intended referent] (pp. 321-341), Allan tries to elucidate
the term 'typical denotatum' by investigating and comparing
prototype semantics with stereotype semantics. After close
investigation and analysis he defines a formal link between
cognition and intention, and formulates a procedure for
identifying reference for the words.

So far, the first ten chapters of the book cover the general
exegesis of linguistic semantics, approaches to it, and
discussion of tools and methods for explaining meaning in
natural language. The last three chapters mainly demonstrate
the application of formal methods of semantic analysis to a
corpus of data. The analyses given here confirm the
cognitive and functional motivations for semantic

In Chapter 11 [Mood, tense, modality, and thematic roles]
(pp. 343-379), Allan tries to tackle some major aspects of
clause semantics. Here he gives a partial semantics for the
grammatical categories of mood, tense, aspect, modality,
predicate frames, and predicate themselves. Next he turns
his attention to the participants of NP arguments of clause
predicates to examine the semantics of their thematic roles.

In Chapter 12 [The Semantics of clause predicates]
(pp. 381-417), Allan tries to build on the discussion of
'thematic roles' while describing the semantics of clause
predicates. Here he begins with the semantic decomposition
of predicates in terms of Jackendoff's lexical conceptual
semantics. Next he turns to the aspectually based
categorization of predicates in Van Valin's role and
reference grammar.

In Chapter 13 [Quantifiers in English] (pp. 419-473), Allan
shifts attention to the semantics of the noun phrase. Here
he makes a detailed study of English quantifiers. He
describes grammatical number, countability, and
(in)definteness to link meaning to morphological
forms. Finally, he ends with a detailed semantic analysis of
a simple sentence which draws on much that has been
discussed in the book to that point.


The volume discusses fundamental concepts for linguistic
semantics, combining theoretical exegesis of several methods
of inquiry with some detailed semantic analysis. It aims to
equip the reader with the basic tools and skills needed to
progress to original research in semantics. The intended
readership is undergraduate and graduate students of
linguistics and relevant areas of psychology, philosophy,
anthropology, sociology, communications, language studies
and education.

In the course of the book the author seeks to answer the
following issues:
(i) What are the tools of semantic analysis ?
(ii) How do we account for the relationships between
words and things ?
(iii) How do we account for the meaning
relationships between language expressions?
(iv) What is the relationship between a lexicon and
an encyclopaedia ?
(v) Where do listemes come from ?
(vi) What are the components of listeme meaning?
(vii) What relationships hold between the form and
meaning of listemes?
(viii) How do we account for the meanings of
phrases, sentences, and large structures?
(ix) What kinds of meaningful effects result from
the use of particular listemes and sentences?
(x) What are the cognitive and functional bases for
meaning in language?

In chapter 1, Allan studies different perspectives
on human language that affect the assumptions we make about
what semantics should investigate. He tries to look at the
ways how meaning is structured in a human language. Be
believes that meaning has both compositionality and
generativity. He relies on the concept of metalanguage: the
language of the semantic theory. He observes that the
principal function of the theory is to explain data from
natural language. A semantic metalanguage can give us
explicit definition of primitives and standards of rigour
and exactitude which are mostly ignored in natural
language. He discusses three kinds of inference (abductive,
inductive, and deductive) and evaluates their role in
semantics. Like him, we also believe that any thorough
account of natural language understanding requires all three
kinds of inference.

He identifies the roles of speaker, hearer, addressee,
ratified participant, bystander and eavesdropper
in the course of speech acts. He argues that language is the
result of acts of speaking by someone, of something, to
someone, at certain point of time and place - often as part
of a longer discourse. The process involves both locutionary
and perlocutionary acts, as well as illocutionary and
reflexive intentions of speaker. He studies the importance
of context and common ground in fixing meaning of
texts. While context determines the spatio-temporal
characteristics of situation of utterance, common ground
implies that the underspecified meaning of the speaker is
correctly inferred by the hearer without explicit reference
from the speaker. He investigates Grice's (1986) 4 maxims
(quantity, quality, relation, and manner) of 'co-operative
principle' in language interaction and concludes that
conventions are expected to be known within a language
community for co-operative language behaviour.

In Chapter 2, Allan defines the dictionary meaning
with reference to dictionary entries, distinguishes between
sentence meaning from utterance meaning and speaker
meaning; and presents a short description of reference,
denotation, extension, and intention. Here his aim is to
unravel the difference between referring, and two kinds of
denoting: intention and extension. He takes up the difficult
question of how we understand reference to what does not
exist or when there is no extension. He considers
extensionality and generics making law-like observations on
TYPE-TOKEN scheme of things (Jackendoff 1983). Finally, he
addresses some characteristic forms of anaphora: a process
of multiple reference to something in a language. However,
some oft-referred terms in semantics like 'polysemy',
'homonymy', 'zeugma', 'ambiguity' etc. should have probably
been little more elaborated here. More information on these
terms can be found in Ullmann (1962), Yule (1985), Todd
(1987), Palmer (1995), Kreidler (1998), Ravin and Leacock
(2000), Cruse (2000) and others.

In Chapter 3, Allan identifies three main components
of lexicon entry: formal, morpho-syntactic and semantic
specifications. (A) Formal specification includes
(i) geographical specification (e.g. spelling variation,
hyphenation location, use of upper case etc.), and
(ii) phonological specification (e.g. information of
alternative pronunciations, syllable structure, relative
pitch and/or tone of various syllables
etc.); (B) morphosyntactic specifications include
morphological and syntactic properties such as the inherent
morphosyntactic (lexical) category of an item and the
constraints on range; and (C) semantic specification
includes antonyms, superordinate items, contraries etc. of
an item within its semantic field. Many of the relations are
not primarily lexical, but arise from the relationship of
the denotatum of a listeme.

His argument that 'stylistic specification'
('colloquial', 'slang', 'derogatory', 'medicine', 'legal'
etc.) is more appropriate to encyclopaedia than to lexicon
is probably right. While lexicon gives meanings of listemes,
encyclopaedia gives information about their
denotata. Moreover, while lexicon identifies the formal and
morphosyntactic specifications of listemes, encyclopaedia
gives information of their history and relationships with
other listemes. Thus lexicon can differ from encyclopaedia
at various levels. Then he studies if names are 'rigid
designators', and finds out some additional characteristics
of proper names while reviewing Kripke's (1980) account. He
discusses the semantics of proper names, and locates their
place in lexicon. Finally, he shows that the lexicon is part
of the encyclopaedia.

In Chapter 4, Allan considers the place of inflected
and derived words in lexicon. He establishes an inventory of
listemes that includes all morphological roots, inflectional
and derivational morphemes, all non-derived and most derived
lexemes, onomatopoeic and phonesthetic items, polyword
idioms, and stems. He defines the blocking principle,
synonymy, and suggests that the criteria of listeme may
differ among languages. He differentiates polyword listemes
(compounds, phrasal verbs, idioms etc.) from phrases, and
establishes criteria to test compounds, phrasal verbs and
polyword idioms. He argues that in most cases the meanings
of compounds, phrasal verbs, and idioms cannot be correctly
derived from the meanings of their components, therefore,
they should be listed in lexicon. He examines the semantic
aspects of listemes created by sound symbolism
(onomatopoeia, phonesthesia, synesthsia, ideophones,
etc.); and claims that phonesthemes, ideophones, and 'tame'
and 'semi-wild' onomatopoeic words are listemes, but
synesthesia and 'wild' onomatopoeic words are members of

In Chapter 5, Allan shows the effects of connotation
can be either euphemism or dysphemism. He describes the
connotations of different forms of address and naming, and
their effects on vocabulary. He discusses the power of
knowing someone's name, and explains the causes why the
names of gods and dangerous animals are tabooed in different
language communities. He attributes the power of 'dirty
words' to strong naturalist beliefs in the community that
transfer to the listeme the distaste that its denotatum
evokes. He identifies reasons why homonyms of taboo terms
drop out of use from language, and why polysemous words with
taboo senses are downgraded by semantic narrowing to the
taboo sense.

He studies the function of upgrading, downgrading,
and deceiving terms, and explores how euphemism and
dysphemism cause language change through the loss of some
senses of terms, or change of terms across styles. He
identifies a dozen ways (remodelling, phonetic similarity,
acronyms, abbreviations, verbal play, circumlocution,
hyperboles, understatements, metonymy, substitute,
synecdoches, and borrowing) for developing new vocabulary of
a language, and shows that a new sense for a listeme can be
derived from metaphorical extension, lexical confusion,
semantic transfer or from similar other multiple
sources. Finally, he defines jargon, explains its necessity,
and shows how it is abused. He describes its functions,
indicates its universal use, and claims that the choice of a
particular jargon involves the choice of a form of language
with consequent connotations that affect the meaning of what
is said.

In Chapter 6, Allan introduces propositional logic
as means of calculating the truth values of connected
propositions without giving any consideration to the
pragmatic relevance of a sequence of propositions. He
redefines entailment and synonymy, and compares entailment
with conventional implicature, a relation applicable to
non-propositional sentence constituents. Unlike entailments
and conventional implicatures, conversational implicatures
are probabilities over a type of utterance that may not be
valid for a particular token of the type and, may be
explicitly cancelled without contradiction. While discussing
presupposition, he rejects the common definition for
semantic presupposition in favour of a definition that says
a presupposition is a proposition whose truth Speaker takes
for granted in making the utterance. This corresponds to a
conversational implicature arising from the precondition of
an illocution. It accounts for the fact that like
conversational implicatures, presuppositions can be
cancelled without self-contradiction.

In Chapter 7, Allan introduces predicate logic: the
basis for analysing the structure of simple
propositions. It is directly applicable to the semantic
structure of clauses. He considers the concepts of meaning
postulates which are used in semantic specification of
listemes. Then he discusses sets and tuples as aids in
specifying the meanings of predicates. He shows why it is
that the meaning of a predicate can be described as a
function from entities to truth values, and relates a formal
notion of 'model' in model theoretic semantics to our
informal notion of model introduced in Chapter 2. He
presents a step-by-step analysis to explicate how we
interpret the meaning of a sentence. Finally, he describes
the usefulness of the lambda or set operator in linguistic

In Chapter 8, Allan begins with scripts to get a
proper perspective on the semantic frames and
fields. Scripts are considered as structured representations
of event sequences, whereas frames are used to identify the
structural relations of listemes and the concepts they
name. Semantic fields are constructed from the semantic
relations among names for concepts. He describes semantic
frames and their place in constraining the combining of
listemses. 'Selection restrictions' are proposed as semantic
constraints on co-occurrence of listemes. He looks at
semantic fields and semantic relations that exist between
items within a field. He centres on componential analysis
which seeks to identify the meaning components in
listemes. Finally, he apprises the search for 'semantic
primitives': the quintessential semantic components.

In Chapter 9, Allan emphasizes the human-centredness
of language by making a close and detailed examination of
the English listemes 'back' to demonstrate the ways in which
this body-part term has extended from humans to animals to
inanimates, and jumped syntactic category from noun to
adjective to verb and adverb. Next, he accounts for the
colour term use and understanding in terms of Vantage Theory
(MacLaury 1997). He examines the character and functions of
different kinds of classifiers used in natural language. The
similarities across languages lead to us doubt about the
Sapir-Whorf hypothesis that each language constrains the
manner in which its speakers perceive the world. Finally, he
shows why some groupings do not get named in natural

In Chapter 10, Allan forges a formal link between
cognition and intention and formulates a procedure for
identifying reference. He discusses and evaluates prototype
semantics which says that a bird which flies (e.g. an
eagle) is closer to the prototypical denotatum of 'bird'
than one which does not (e.g. an ostrich). Nest, he
considers the possibility that a typical denotatum is
stereotype, which is compared with the notion of a
Gestalt. He discusses typicality and argues that the typical
denotatum is more stereotype than the best
exemplar. Finally, he describes a procedure whereby the
speaker chooses an appropriate label for a referent, and the
hearer interprets the language expression so as to correctly
identify the speaker's intended referent.

In Chapter 11, Allan discusses the meaning and
function of mood, which can be read from clause type
(declarative, interrogative, etc.). It is shown that
identifying the mood of the clause is the first step towards
discovering the illocutionary point of an utterance. He
offers a simplified semantics for tense in terms of fuzzy
sets of time points on a linear scale. He also touches on
some features and functions of aspect in English. Next, he
addresses the semantics of some English modal verbs like
'will', 'can', 'may', 'must' etc. to understand their
functions in sentence. He also watches out for ways in which
our dialects differ, and identifies it as an area of
considerable variation. Finally, he defines a set of
thematic roles like 'agent' and 'patient' which are
carefully referred to in the linguistic literature to
understand if they were well-defined uncontroversial
concepts. He, however, shows that their definition faces
many uncertainties.

In Chapter 12, Allan reviews Jackendoff's (1985) lexical
conceptual structure analysis of the meanings of
verbs. Jackendoff decomposes verbs into a limited number
of semantic primitives that function as predicates in what
he calls 'functional-argument structures'. He continues the
study of semantic frames and structures of predicates by
examining their 'logical structures' in role and reference
grammar (RRG). RRG decomposes verbs on the basis of their
aspectual properties, i.e. in terms of whether they denote
activities, states, achievements, etc. Considerable
attention is paid to the assignment of the macroroles actor
and undergoer within the logical structures of verbs. He
compares and evaluates the application of Jackendoff-style
lexical conceptual semantics and RRG style logical structure
to the same data. Finally, he comments on the nature of
semantic specifications in the metalanguage.

In Chapter 13, Allan discusses the association
between grammatical number and countabilty. He argues that
countability is a property of NPs not of nouns, and shows
how nouns may show a preference for either uncountable or
countable environments. He establishes ensemble theory (a
combination of mereology with set theory) for semantic
definition of lexically related countables and uncountables
from a common base. He introduces the notion of 'restricted
quantification' using generalized quantifiers formed on
everyday quantifiers from English lexicon. It provides the
necessary base for semantically specifying NPs containing

He creates a semantics for plural morphemes from a
noun to describe semantic effects of quantifiers ranging
over them and other quantifiers. He takes up the challenge
that every English NP is either definite or indefinite to
argue that 'the' is the prototypical definite as well as a
universal quantifier, and explains why 'the' is said to mark
identifiability and uniqueness of reference. He examines
relationships among quantifiers as revealed by their
entailment relations and scope, and describes semantic
foundations for associating an indefinite NP with its
definite anaphoric successor. Finally, he uses insights from
this and earlier chapters to make an exhaustive analysis of
all the constituents of a simple classificatory sentence.

Each chapter contains definitions of terms, assumptions,
sets of exercises, text boxes, tables, figures
and tree diagrams (Chapter 1: 30 definitions, 11
assumptions, 8 sets of exercises, 2 text boxes, 1
table; Chapter 2: 16 definitions, 8 assumptions, 6 sets of
exercises, 2 pictures, 6 figures; Chapter 3: 5 definitions,
11 assumptions, 5 sets of exercises, 2 tables, 7
figures; Chapter 4: 11 definitions, 16 assumptions, 5 sets
of exercises, 3 tables, 2 figures; Chapter 5: 9 definitions,
12 assumptions, 7 sets of exercises, 1 table, 1
figure; Chapter 6: 14 definitions, 3 assumptions, 4 sets of
exercises, 5 tables, 1 figure; Chapter 7: 21 definitions, 6
assumptions, 5 sets of exercises, 12 tree-diagrams, 2
figures; Chapter 8: 14 definitions, 20 assumptions, 5 sets
of exercises, 9 tables, 3 figures; Chapter 9: 23
definitions, 15 assumptions, 3 sets of exercises, 2 table, 6
figures; Chapter 10: 9 definitions, 7 assumptions, 5 sets of
exercises, 3 tables, 7 figures; Chapter 11: 27 definitions,
5 assumptions, 4 sets of exercises, 3 tables; Chapter
12: 10 definitions, 14 assumptions, 2 sets of exercises, 8
tables, 2 figures; Chapter 13: 30 definitions, 13
assumptions, 9 sets of exercise, 7 tables, 7 figures, 3 tree
diagram, 8 pictures) along with lists of key words and
phrases, summary of the content, and notes for further
reading. All these are highlighted and numbered for ready
reference and quick access. Besides, the book contain a well
written preface, list of typographical conventions, a huge
list (202, approximately) of phonetic and other symbols, an
epilogue, a healthy reference (746 books and journals ),
and a well-referred index. There are, however, some
typographical errors which are expected to be corrected in
the next edition of the volume.

The book of Allan is a great work which tries to
encompass a wide and varied range of linguistic semantics
within its scope. Not always he has been successful in his
attempt, but most of the times he has excelled in his
presentation and elaboration. Sometimes, his wise analysis
of some age-old concepts has gifted us with new insights,
and we are delighted to judge the things in new lights from
new angles. Moreover, his overall command over the domain is
revealed in by his close reference to the literature central
and related to the domain. To be honest, the book has every
quality to be considered as a milestone in linguistic
semantics. Any language lover/worker should feel himself
proud to possess a copy of such a scholarly book.


Cruse, A. (2000) Meaning in Language: An Introduction to
Semantics and Pragmatics. Oxford: Oxford University

Grice, H. Paul (1986) Studies in the Way of
Words. Cambridge, MA: Harvard University Press.

Jackendoff, R. (1983) Semantics and Cognition. Cambridge,
MA: MIT Press.

Jackendoff, R.(1985) "Multiple subcategorization and the
Theta-criterion: the case of 'climb'". 'Natural Language and
Linguistic Theory'. 3:271-95.

Kreidler, C.W. (1998) Introducing English
Semantics. London: Routledge.

Kripke, S. (1980) Naming and Necessity. Oxford: Blackwell

Palmer, F.R. (1995) Semantics. (2nd
Edition). Cambridge: Cambridge University Press.

Ravin, Y., and Leacock, C. (eds.) (2000) Ploysemy:
Theoretical and Computational Approaches. New York: Oxford
University Press.

Todd, L. (1987) An Introduction to Linguistics. Essex,
UK: Longman York Press.

Ullmann, S. (1962) Semantics: An Introduction to the Science
of Meaning. Oxford: Blackwell.

Yule, G. (1985) The Study of Language: An
Introduction. Cambridge: Cambridge University Press.

Niladri Sekhar Dash works as a Linguist in Computer Vision
and Pattern Recognition Unit of Indian Statistical
Institute, Kolkata, India. His research interest includes
corpus linguistics, text annotation, lexical semantics,
word-sense disambiguation, generative morphology etc. He is
currently working on the nature of surface wordforms, and
lexical ambiguity in Bangla.