Lexicography

Lexicography
This book is an accessible introduction to lexicography – the study of dictionaries.

We rely on dictionaries to provide us with definitions of words, and to tell us
how to spell them. They are used at home and at school, cited in law courts,
sermons and parliament, and referred to by crossword addicts and scrabble
players alike. But why are dictionaries structured as they are? What types of
dictionary exist, and what purposes do they serve? Who uses a dictionary, and
for what?
Lexicography: An Introduction provides a detailed overview of the history, types
and content of these essential reference works. Howard Jackson analyses a wide
range of dictionaries, from those for native speakers to thematic dictionaries and
learners’ dictionaries, including those on CD-ROM, to reveal the ways in which
dictionaries fulfil their dual function of describing the vocabulary of English and
providing a useful and accessible reference resource.
Beginning with an introduction to the terms used in lexicology to describe
words and vocabulary, and offering summaries and suggestions for further read-
ing, Lexicography: An Introduction is concise and student-friendly. It is ideal for
anyone with an interest in the development and use of dictionaries.
Howard Jackson is Professor of English Language and Linguistics at the

University of Central England. His publications include Grammar and Vocabulary
(Routledge, 2002), Words and their Meaning (Longman, 1988), and Words, Meaning
and Vocabulary (Cassell, 2000).
Lexicography
An introduction
Howard Jackson
London and New York

First published 2002
by Routledge
11 New Fetter Lane, London EC4P 4EE
Simultaneously published in the USA and Canada by Routledge
29 West 35th Street, New York, NY 10001
Routledge is an imprint of the Taylor & Francis Group
© 2002 Howard Jackson
Typeset in Bembo by
The Running Head Limited, Cambridge
Printed and bound in Great Britain by TJ International
All rights reserved. No part of this book may be reprinted or
reproduced or utilised in any form or by any electronic,
mechanical, or other means, now known or hereafter
invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in
writing from the publishers.
British Library Cataloguing in Publication Data
A catalogue record for this book is available
from the British Library
Library of Congress Cataloging in Publication Data
A catalog record has been requested
ISBN 0–415–23172–8 (hbk)

ISBN 0–415–23173–6 (pbk)
Contents
Preface vii
Dictionaries cited ix
1 Words 1
2 Facts about words 10

3 The dictionary 21
4 The beginnings 31
5 The New English Dictionary 47

6 Up to the present 61
7 Users and uses 74
8 Meaning in dictionaries 86
9 Beyond definition 101

10 Etymology 117
11 Dictionaries for learners 129
12 Abandoning the alphabet 145

13 Compiling dictionaries 161
14 Criticising dictionaries 173
References 184
Index 189
Preface
Much has happened, both in respect of the making of dictionaries and in respect
of their academic study, in the twelve or so years since my previous book
on dictionaries (Words and Their Meaning, Longman, 1988). Then, the ‘corpus
revolution’ (Rundell and Stock 1992) had only just begun – Words and Their
Meaning just managed to catch the first (1987) edition of the Collins COBUILD
English Dictionary. Now virtually all dictionaries published in the UK make
some claim to have used a computer corpus in their compilation. Not only have
learners’ dictionaries developed by leaps and bounds – the Oxford Advanced
Learner’s Dictionary was in its third edition then, now in its sixth, and the Cam-
bridge International Dictionary of English was still a long way off – but native
speaker dictionaries have also seen significant developments – the publication
of the New Oxford Dictionary of English in 1998, as well as three editions of the
Concise Oxford, not to mention the second edition of the great OED in 1989
and the beginning of the massive revision that will result in the third edition,
planned for 2010.
Dictionaries have also appeared during the period in electronic format, nota-
bly as CD-ROMs, opening up new possibilities, not only in how dictionaries
can be used and exploited, but also in how dictionary material can be organised
and presented. Dictionaries are also accessible online, through the internet,
including the OED, enabling subscribers to view the revisions that will consti-
tute the third edition, as they are posted quarterly.
The study of lexicography has also developed and flourished during the last
dozen years. They saw the launch of the highly successful International Journal of
Lexicography in 1988, for the first ten years under the editorship of Robert Ilson,
and latterly that of Tony Cowie. The mighty three-volume Encyclopedia of Lexico-
graphy (Hausmann et al. 1989–91) delineated the state of the art, and the Diction-
ary of Lexicography (Hartmann and James 1998) mapped the territory. More
recently, Reinhard Hartmann’s Teaching and Researching Lexicography (2001) has
set the agenda for the business of academic lexicography. And Sidney Landau
has updated his readable Dictionaries: The Art and Craft of Lexicography (second
edition, 2001) with its transatlantic perspective.
It is time for a new treatment of the subject in the UK. I am grateful to Louisa
Semlyen and to Routledge for taking this on. The book is dedicated to all the
viii Preface
final-year students who have enabled me to develop the material by taking
my ‘Lexicography’ module on the English degree at the University of Central
England in Birmingham over more years than I care to recall.
Howard Jackson
Birmingham
August 2001
Dictionaries cited
The following dictionaries are mentioned in the course of this book. (Note: a
superscript number, e.g. 19882, refers to the edition; in this case, the second
edition published in 1988.)
Native speaker dictionaries

Chambers English Dictionary, (19887 ) edited by Catherine Schwarz, George
Davidson, Anne Seaton and Virginia Tebbit.
Chambers 21st Century Dictionary (1996) edited by Mairi Robinson.
Collins Concise Dictionary (1982; 19882; 19923; 19994, edited by Diana Treffry).
Collins English Dictionary (1979 edited by Patrick Hanks, 19862 edited by Patrick
Hanks, 1991/943 edited by Marian Makins, 19984 edited by Diana Treffry).
Concise Oxford Dictionary (1911 edited by H.G. and F.W. Fowler, 19292, 19343,
19514, 19645, 19766, 19827, 19908, 19959, 199910 edited by Judy Pearsall).
Encarta Concise English Dictionary (2001) edited by Kathy Rooney, Bloomsbury.
Longman Dictionary of the English Language (1984, 19912 edited by Brian O’Kill).
A New English Dictionary on Historical Principles (1888–1928) edited by James
Murray, Henry Bradley, W.A. Craigie and C.T. Onions.
New Oxford Dictionary of English (1998) edited by Judy Pearsall.
Oxford English Dictionary (1933 edited by James Murray et al., 19892 edited by
John Simpson and Edmund Weiner).
The New Shorter Oxford English Dictionary on Historical Principles (1993) edited by
Lesley Brown.
Webster’s Third New International Dictionary of the English Language (1961) edited
by Philip Gove.
Monolingual learners’ dictionaries

Cambridge Dictionary of American English (2000) edited by Sidney Landau.
Cambridge International Dictionary of English (1995) edited by Paul Proctor.
Collins COBUILD English Dictionary (1987, 19952, 20013) edited by John Sinclair.
Longman Dictionary of Contemporary English (1978 edited by Paul Proctor, 19872
edited by Della Summers and M. Rundell, 19953 edited by Della Summers).
x Dictionaries cited
Longman Language Activator (1993) edited by Della Summers.
Longman Lexicon of Contemporary English (1981) compiled by Tom McArthur.
Oxford Advanced Learner’s Dictionary of Current English (1948 edited by A.S.
Hornby, E.V. Gatenby and H. Wakefield; 19632 edited by A.S. Hornby,
E.V. Gatenby and H. Wakefield; 19743 edited by A.S. Hornby, with A.P.
Cowie and J. Windsor Lewis; 19894 edited by A.P. Cowie, 19955 edited by
Jonathan Crowther; 20006 edited by Sally Wehmeier).
Thematic dictionaries
A Thesaurus of Old English (1995) compiled by Jane Roberts and Christian Kay,
with Lynne Grundy.
Longman Dictionary of Scientific Usage (1979) compiled by A. Godman and E.M.F.
Payne.
Longman Language Activator (1993) edited by Della Summers.
Longman Lexicon of Contemporary English (1981) compiled by Tom McArthur.
Roget’s Thesaurus of English Words and Phrases (1852), Longmans, Green and Co.
The Scots Thesaurus (1990), edited by Iseabail McLeod.
Abbreviations
In order to save space, dictionaries regularly cited will usually be referred to in
the course of the book by the following abbreviations:
CCD – Collins Concise Dictionary

CED – Collins English Dictionary
Chambers – Chambers English Dictionary
CIDE – Cambridge International Dictionary of English
COBUILD – Collins COBUILD English Dictionary
COD – Concise Oxford Dictionary
ECED – Encarta Concise English Dictionary
LDEL – Longman Dictionary of the English Language
LDOCE – Longman Dictionary of Contemporary English
NODE – New Oxford Dictionary of English
OALD – Oxford Advanced Learner’s Dictionary
OED – Oxford English Dictionary
SOED – Shorter Oxford English Dictionary
W3 – Webster’s Third New International Dictionary.
Where the abbreviation is followed by a number, e.g. COD8, the number

refers to the edition, i.e. Concise Oxford Dictionary, eighth edition.
Words 1
1 Words
1.1 What is a word?

You take a dictionary off the shelf, or access a dictionary on your computer, and
open it because you want to look up a ‘word’. Dictionaries are the repositories
of words. Words are arranged in dictionaries in alphabetical order, and as you
look down the column in a print dictionary or the list in an electronic diction-
ary, you are reading a list of words. Or are you? Here is the list of the 25
‘headwords’ between want and wardrobe in COD10 (i.e. Concise Oxford Diction-
ary, tenth edition: see ‘Dictionaries cited’, p. ix):
want, wanting, wanton, wapentake, wapiti, War., war, waratah, war baby,
warble1, warble2, warble fly, warbler, warby, war chest, war crime, war
cry, ward, -ward, war dance, warden, warder, ward heeler, ward of court,
wardrobe.
A number of items in this list do not quite match our usual concept of what
constitutes a word, which is – I suggest – ‘a sequence of letters bounded by
spaces’. Indeed, only 15 of the 25 items could be described in this way. Two
of the remaining items are less than a full word: the abbreviation War. (for
Warwickshire), and the suffix –ward (used to form words like backward, skyward –
see Chapter 2). The other eight items all consist of more than one ‘word’:
seven of them have just two words, and one has three (ward of court). You
will also have noticed that one word (warble) is entered twice. So, just what is
a ‘word’?
The word before want in the COD10 list is wannabe. Is that a word, or is it
three (want to be)? In our usual concept of a word, it is one, because it is a
sequence of letters bounded by spaces. This conception of words comes, of
course, from writing, the medium in which we are most conscious of words;
and dictionaries are based on the written form of the language. In speech, though,
words are composed of sounds and syllables, and they follow one another in the
flow of speech without spaces or pauses. We make no more pause in saying war
baby than we do with wardrobe, even though the first consists of two words in
writing and the second of only one.
2 Words
There is, clearly, a measure of confusion here that needs some sorting out in
a book about words and dictionaries. Let us make the following distinction of
terms:
orthographic word a word in writing, a sequence of letters bounded by spaces

phonological word a word in speech, a sequence of sounds (the boundaries of
phonological words are determined by rules of syllable structure, stress, and
the like)
lexeme a word in the vocabulary of a language; it may occur as a headword in a
dictionary.
A lexeme may, therefore, consist of more than one orthographic word, as warble
fly, war chest, ward of court. Even though they are listed as headwords, we should
exclude abbreviations and affixes (see 1.6 below) from the category of lexeme.
1.2 Same sound, same spelling, different word

We noticed that warble is entered twice in COD10. The compilers of this dic-
tionary are following common practice and recognising two different lexemes
with the same spelling (and, as it happens, the same pronunciation). The first
warble is the verb that refers to birdsong; the second is a noun denoting ‘a
swelling or abscess beneath the skin on the back of cattle . . . caused by the
presence of the larva of a warble fly’. However, the fact that the meanings of the
two lexemes are completely unrelated is not the primary criterion for distin-
guishing them. Dictionaries usually operate with the criterion of etymology
(see Chapter 10) for deciding that a single orthographic word represents more
than one lexeme. If a single spelling can be shown to have more than one
origin, then it constitutes more than one lexeme. In the case of warble, the
‘birdsong’ lexeme has its origin, according to COD10, in the Old Northern
French word werble, which came into English during the Middle English period
(1066–1500). The ‘abscess’ lexeme also originates in the Middle English period,
but it has a different, according to COD10 ‘uncertain’, provenance.
Lexemes that share the same spelling and pronunciation, but have a different
etymology, are termed homonyms (a Greek word, meaning ‘same (homo) name
(nym)’).
Another orthographic word with a double entry in the dictionary is tear. The
first tear lexeme relates to ‘pulling or ripping apart’, the second denotes the drop
of salty liquid that comes from the eyes when someone weeps. In this case,
however, the same spelling has different pronunciations, i.e. phonological words.
Since the dictionary is based on spelling, tear is entered twice. As might be
expected, tear (rip) and tear (weep) also have different origins, both from Old
English, the first from teran and the second from tBar. Lexemes that share the
same spelling, but not the same pronunciation, are called homographs (from
Greek, ‘same’ + ‘writing’). There are not very many homographs in English, by
Words 3
comparison with the number of homonyms. Here are some further examples
for you to figure out (or look up):
bow, curate, denier, irony, prayer, refuse, reserve, sow, supply, wind.
Much more common in English are the counterparts to homographs: lexemes

that are pronounced the same, but spelled differently, e.g. pale/pail. These present
no problem to a dictionary, since it is the spelling that takes priority; and each is
entered as a headword at the appropriate place in the alphabetical sequence.
Lexemes that share the same pronunciation, but not the same spelling, are called
homophones (from Greek, ‘same’ + ‘sound’). Here are some further homo-
phone pairs in English:
bare/bear, gait/gate, haul/hall, leak/leek, miner/minor, paw/poor/pore/

pour, sew/sow, stake/steak, taught/taut
You will notice that most homophones arise because vowel sounds that used to
be pronounced differently, as represented by the spelling, have in the course of
historical sound changes come to be pronounced the same.
1.3 Lexemes and variants

If you look up sung in a dictionary, you will find a very brief entry along the
lines of ‘past participle of sing’, which is a cross-reference to the entry for sing. If
you look up the word talked, which is the past participle of talk, you will not
find an entry. For both these words, the dictionary gives their description under
a single entry: sing for sung, and talk for talked. You do not need a separate
treatment of sung or talked, because what is said about sing or talk is equally
applicable to them. They are merely ‘variants’ of the entry word; in effect they
are the ‘same word’.
The lexeme sing, for example, has the following variants: sing, sings, sang,
singing, sung. The lexeme talk has one variant fewer: talk, talks, talked, talking.
What we are looking at are the inflections of verbs in English:
base/present tense sing talk

third person singular/present tense sings talks
past tense sang talked
present participle singing talking
past participle sung talked
The verb talk represents the ‘regular’ paradigm, where the past tense and the
past participle have the same form, with the –(e)d suffix. The verb sing is one of
a number with ‘irregular’ inflections.
There is a sense in which sing, sings, sang, singing and sung are all the ‘same
word’; they are different manifestations of the same lexeme, variants chosen
4 Words
according to the grammatical context of the lexeme. For example, if the subject
of a sentence is a ‘third person singular’ (equivalent to he, she or it) and the
speaker/writer has chosen present tense, then the form of the verb will be sings
or talks, with the ‘s’ suffix marking the ‘third person singular present tense’ (e.g.
‘until the fat lady/she sings’) We need a further term to distinguish this type of
‘word’:
word-form an inflectional variant of a lexeme
To illustrate word-forms we have chosen verbs, because verb lexemes have

more inflections than any other type of lexeme in English. Two other types of
lexeme regularly have inflectional variants and so more than one word-form:
nouns and adjectives – though not every member of these classes, as is the case
with verbs. Countable nouns (biscuit, coin), but not uncountable nouns (dough,
salt), have a ‘plural’ inflection. Some nouns, mainly referring to animate beings,
have a ‘possessive’ inflection. The word-forms for plural nouns have a –(e)s
suffix as the regular inflection (bananas, oranges, mangoes). A small number of
countable nouns form the plural irregularly, e.g. feet, geese, mice, teeth; children;
knives, loaves; nuclei, millennia, formulae, hypotheses, criteria. The possessive inflec-
tion is normally marked in the singular noun by an apostrophe + s (e.g. cat’s,
girl’s, nephew’s), and in the plural noun by an apostrophe only, placed after the
plural suffix (e.g. cats’, girls’ nephews’). This, of course, applies to writing: in
speech, the possessive singular adds –(e)s, and so is no different from the plural;
and the plural possessive is the same as the normal plural, except where the
plural is formed irregularly (e.g. mice’s, children’s, women’s). Summarising, the
word-forms of (some) noun lexemes are:
base/singular girl child

plural girls children
possessive singular girl’s child’s
possessive plural girls’ children’s
Note that the three inflected forms of girl (the ‘regular’ paradigm) have the same
pronunciation.
Some adjective lexemes in English have a ‘comparative’ and a ‘superlative’
form. The adjectives concerned are ‘gradable’ (e.g. long, quick, small), rather
than ‘ungradable’ (daily, mortal, sterile). Most gradable adjectives that are one-
syllable in length can have these forms, as may most two-syllable gradable adjec-
tives. The regular inflection for the comparative is –er, and for the superlative
–est (e.g. longer/longest, quicker/quickest, smaller/smallest). There is a very small
number of irregular forms: good, better, best; bad, worse, worst. An alternative way
of expressing comparison, applied to some two-syllable adjectives and to nearly
all gradable adjectives of three syllables or more, is with the adverbs more and
most (e.g. more/most skilful, more/most treacherous). Summarising, the word-forms
of (some) adjective lexemes are:
Words 5
base slow good
comparative slower better
superlative slowest best
When one-syllable adjectives do not permit word-forms with –er/-est, it is usu-

ally because their pronunciation is somehow awkward (e.g. sourer, wronger).
1.4 War chests and wards of court

In the list from the COD10 in 1.1 we noted several lexemes composed of more
than one orthographic word. A number of them have war as their first element:
war chest, war crime, war cry, war dance. Two independent lexemes have come
together to form a new lexeme with a specialised meaning, to denote some
entity that is considered worth having its own ‘name’. We call such lexemes
compounds (see further Chapter 2). Sometimes compounds are written, as in
the examples with war, with a space between the two elements. Other com-
pounds are written as a single orthographic word (e.g. warhead, warlord, warpath,
warship), while others have a hyphen joining the two elements (e.g. war-torn,
window-shop, world-class). The current tendency is away from ‘hyphenated com-
pounds’ towards either ‘solid compounds’ (one orthographic word) or ‘open
compounds’ (two or more orthographic words).
The other multi-word lexeme in the list is ward of court, which is a phrase
rather than a compound. Phrasal lexemes have a number of common struc-
tures, of which the ‘noun + preposition + noun’ of ward of court is one. Here are
some further examples of this structure:
age of consent, cash on delivery, chapel of rest, home from home, hostage
to fortune, man about town, meals on wheels, place in the sun, rite of
passage, skeleton in the cupboard.
A second phrasal structure consists of a noun in the possessive followed by

another noun, e.g.
athlete’s foot, banker’s card, collector’s item, fool’s paradise, hair’s breadth,
lady’s finger, ploughman’s lunch, potter’s wheel, saint’s day, smoker’s cough,
traveller’s cheque, writer’s block.
A third phrasal structure consists of two words of the same type (noun, verb,
adjective) joined by the conjunction and. These are sometimes called ‘binomials’.
Here are some examples:
bells and whistles, black and white, bow and scrape, down and out, fast and
furious, hammer and tongs, nip and tuck, pins and needles, rock and roll,
sweet and sour, ups and downs, you and yours.
6 Words
There are also a few cases of ‘trinomials’, e.g. hop, skip and jump; hook, line and
sinker. You will notice that a number of these items are used metaphorically:
hammer and tongs has nothing to do with the literal instruments used by the
blacksmith, but refers to the intensity or vigour with which something is done.
A fourth kind of phrasal lexeme consists of a verb + adverb (sometimes called
a ‘particle’), to form what are called ‘phrasal verbs’. Here are some examples:
break up, calm down, find out, give in, look over, pass out, show up, take
off, waste away, wear out.
Some of these phrasal verbs have a literal or near-literal meaning, others are
more-or-less figurative in meaning. In one of its meanings, take off is literal (e.g.
referring to aircraft leaving the runway), in another it is figurative (in the sense
of ‘imitate’).
A fifth kind of phrasal lexeme, if indeed we can count them as lexemes, are
typically metaphorical or figurative in meaning. They are idioms, which have
a range of structures from phrase up to whole sentence. An idiom has two
essential characteristics: its meaning is more than the meaning of the sum of its
parts, and usually figurative; and it has a relatively fixed structure. The idiom a
storm in a teacup (American English equivalent a tempest in a teapot) has the figurative
meaning of a ‘fuss about nothing’, and there is no possibility of substituting or
adding anything to its structure. In pull the wool over someone’s eyes, the meaning
is figurative (i.e. ‘deceive’), and the only substitution possibilities are appropri-
ate inflections for the verb pull and an appropriate possessive noun or pronoun
in the place of someone’s. Idioms are all pervasive in language and show a diver-
sity of form and meaning (see Fernando and Flavell (1981) for a fuller treat-
ment). Here are a few more examples from English:
know which side one’s bread is buttered, at the drop of a hat, go against the
grain, come to a pretty pass, take someone for a ride, spill the beans, throw
the baby out with the bathwater, walk on eggshells.
You will notice that in some cases (e.g. take someone for a ride) a literal interpre-
tation is also possible. Only the context will reveal whether the literal or the
metaphorical (idiomatic) meaning is the intended one.
1.5 Classifying words

In talking about words, we often, as already in this chapter, need to refer to
them by the conventional broad classification into ‘parts of speech’, or ‘word
classes’ as the preferred term now is. Rather than assume that this is general
knowledge, as most dictionaries do, we will devote a little discussion to it.
Although we have school-based definitions in our minds, such as ‘a verb is a
doing word’, words are classified more rigorously largely on the basis of the
roles they play in the structure of sentences. English has four large classes, into
Words 7
which most new words go, and four smaller, fairly static classes. The four large
classes are:
• nouns are the largest class by far; they represent the animate and inanimate
objects that are the participants in sentences as subjects, objects, etc. (beauty,
cat, leaf, niece, nonsense, water)
• verbs represent the action, event or state that the sentence is about, and hold
the pivotal position in the sentence, determining which other elements
need to be present (break, decide, fall, have, keep, love)
• adjectives occur in front of nouns as descriptive words, as well as after verbs
like be with a similar function ( feeble, gigantic, lazy, new, rough, vain)
• adverbs are a diverse class, in part representing circumstantial information
such as time (again, always, sometimes, soon) and manner (clearly, efficiently,
quickly, tentatively), in part acting as modifiers of adjectives or other adverbs
(quite, somewhat, very), in part forming connections between sentences (how-
ever, moreover, therefore).
The four smaller word classes, whose major function is to link the members of
the larger classes together in sentence structure, are:
• pronouns stand for nouns and their accompanying words (noun phrases) to
avoid unnecessary repetition, including personal pronouns (I, you, he, she,
it, we, they), possessive pronouns (mine, yours, hers), reflexive pronouns (my-
self, yourself, themselves), relative pronouns (who, whose, which), indefinite
pronouns (someone, nobody, anything)
• determiners accompany nouns and are subdivided into ‘identifiers’ and ‘quan-
tifiers’; identifiers include the articles (a, the), demonstratives (this, that) and
possessives (my, your, her, our, their); quantifiers include the numerals (two,
five; second, fifth) and indefinite quantifiers ( few, many, several)
• prepositions combine with nouns or noun phrases primarily to form pre-
positional phrases (at, for, from, in, of, on, over, through, with)
• conjunctions are used to connect clauses or sentences, but also phrases and
words; they include the co-ordinating conjunctions (and, but, or) and a
larger number of subordinating conjunctions (although, because, if, until, when,
while).
You should consult a grammar book if you need a more extensive explanation
of the word classes.
1.6 Taking words to pieces

In the course of this chapter, we have mentioned terms like ‘affix’ and ‘suffix’,
which are parts of words. This section looks at the analysis of words into their
constituent elements and suggests some terms that will be useful in talking about
word structure. First of all, we need a term to denote an element of a word: it
8 Words
is morpheme. Words are composed of morphemes. Many words, sometimes
called ‘simple’ words, consist of only one morpheme:
bed, dream, go, in, over, please, shallow, treat, usual, vote, whole, yellow.
Here are some words composed of more than one morpheme:
bedroom, dreamy, going, live-in, overland, displease, shallowest, mistreat-

ment, usually, voters, wholemeal, yellowish.
Each of these words has, as one of its morphemes, a ‘simple’ word from the
earlier list, which forms the ‘root’, or in the case of compounds like bedroom one
of the roots, of the word. The root morpheme is the kernel of the word, with
the main meaning, which is modified by other morphemes in various ways.
Compounds are composed of two or more root morphemes: bedroom, live-in,
overland, wholemeal. These compounds have a variety of structures in terms of
the word class membership of the roots: noun + noun, verb + preposition,
preposition + noun, adjective + noun. Many compounds are like bedroom, where
the first part modifies the second and the word class of the compound is that of
the second part, in this case a noun: a ‘bedroom’ is a kind of ‘room’. The other
three compounds are different: live-in, with a preposition as second part, is an
adjective (as in a live-in nanny); overland, with a noun as second part is either an
adjective (an overland journey) or an adverb (we’re travelling overland); and whole-
meal, again with a noun as second part, is an adjective (wholemeal bread).
The other words in the list are all composed of root + affix. ‘Affix’ is the
general term for morphemes that cannot be used by themselves as simple words;
they only occur ‘bound’ to another morpheme. If they occur before the root,
and so are bound to the right, they are called ‘prefixes’ (e.g. disin displease). If
they occur after the root, and so are bound to the left, they are ‘suffixes’ (e.g.
-ish in yellowish). Note that, when writing affixes, the convention is to put a
hyphen on the side where the affix is bound, i.e. to the right of prefixes and to
the left of suffixes.
Some of the suffixes mark ‘inflections’ (see 1.3 above): go-ing (present parti-
ciple), shallow-est (superlative), voter-s (plural). There are no inflectional prefixes
in English. The resulting words are ‘word-forms’ (inflectional variants) of the
root lexeme.
The other affixes represent ‘derivations’. The addition of the affix creates a
new, derived lexeme. We would expect it to be entered in a dictionary some-
where, though, as we shall see (Chapter 8), dictionaries vary in how they treat
derivations. The addition of a suffix usually changes the word class of the root,
though a prefix rarely does:
dream (noun) + -y dreamy (adjective)

dis- + please (verb) displease (verb)
mis + treat (verb) mistreat (verb)
Words 9
mistreat + -ment mistreatment (noun)
usual (adjective) + -ly usually (adverb)
vote (verb) + -er voter (noun)
yellow (adjective) + -ish yellowish (adjective)
Note that an inflectional suffix, e.g. the plural ‘s’ on voters, is always the final
suffix in a word.
We might conclude from our discussion of morphemes so far that roots are
always ‘free’ (i.e. can occur as simple lexemes), and affixes are always ‘bound’
(i.e. they need a root to attach to). However, there is a certain set of words in
English, mostly compounds, that have bound roots. Here are some examples:
anthropomorphic, astronaut, bibliography, biology, neuralgia, synchrony,

telepathy, xenophobia.
These lexemes are formed from (bound) roots that are taken from the classical
languages (Greek and Latin) and put together to form, for the most part, new
words that were unknown in classical Greek and Latin. They are known as
‘neo-classical compounds’, and their parts are called ‘combining forms’. Our
examples are formed as follows:
• anthropo- (human) + -morphic (in the form of )

• astro- (star) + -naut (sailor)
• biblio- (book) + -graphy (writing)
• bio- (life) + -ology (study)
• neuro- (nerve) + -algia (pain)
• syn- (same) + -chrony (time)
• tele- (distant) + -pathy (feeling)
• xeno- (foreigner) + -phobia (fear).
Some roots from the classical languages occur in derivations, when they are also
bound, e.g. chron-ic, graph-ical, naut-ical, neur-al, path-etic.
To summarise:
a word is composed of one or more morphemes

a morpheme may function as a root or as an affix (prefix or suffix)
a root morpheme is usually free, an affix is always bound
bound roots are usually combining forms from Greek or Latin.
1.7 Further reading

You can find a fuller treatment of words and word structure in Jackson and Zé
Amvela’s Words, Meaning and Vocabulary: An Introduction to Modern English
Lexicology (2000) and in Francis Katamba’s English Words (1994).
10 Facts about words
2 Facts about words
In Chapter 1, we examined the ambiguity of the term ‘word’ and suggested a set
of terms for resolving the ambiguity. We also outlined the morphology of the
word in English and proposed terms for talking about the structure of words.
This chapter makes a further contribution to the lexicology (study of words) of
English, before we move on to the study of dictionaries (lexicography) in the
next chapter.
2.1 Where English words came from

The vocabulary of English contains words from more sources than the vocabu-
lary of any other language, as a consequence of its history and the contacts
between its speakers and those of other languages. As far as its basic components
are concerned, it is useful to view the vocabulary of English as being composed
of a number of strata, rather like a rock formation in geology. The substratum
of English is Anglo-Saxon, the collection of dialects that developed after the
invading Germanic tribes, the Angles, Saxons and Jutes, colonised England
following the departure of the Roman legions in the fifth century AD, driving
the Celtic inhabitants to the fringes of the country in Wales and Cornwall. The
language became known during this time as ‘English’, and we refer to the
language during the period up to the mid-eleventh century as ‘Old English’.
The only significant influence on the language from outside during this period
was from across the North Sea, the Viking invaders, who also spoke a Germanic
language, Old Norse. For a time the country was divided, with ‘Danelaw’ on
the eastern side of a line from Chester to the Wash. Old English and Old Norse
were to a great extent mutually intelligible, and the influence of Old Norse on
Old English was limited. The greatest linguistic legacy of the Vikings was in
place names, e.g. ending in –by or –thorpe; but also many words beginning with
sk- come from Old Norse, as do the third person plural pronouns (they, them,
their). Even with these additions, the vocabulary of Old English was essentially
Germanic, with a handful of words from Celtic, and a number of ecclesiastical
terms taken from Latin following the introduction of Roman Christianity as a
result of Augustine’s mission in 597. (See Roberts et al. 1995 for a description of
Old English vocabulary.)
Facts about words 11
The next stratum of vocabulary began to be laid with the Norman conquest
in 1066. The influence of this event, and its political and social consequences,
on English vocabulary was monumental, though it took a couple of centuries
for its full effects to be worked out. The language of government, administra-
tion and the law became (Norman) French; English was not used for any official
written purposes; in due course many people, especially in the rising merchant
class, became bilingual in English and French. It is estimated that in excess of
10,000 words entered English from French between the twelfth and fourteenth
centuries. French is a Romance language, with its origins in Latin; so a Latinate
stratum was being overlaid on the Ango-Saxon substratum. Indeed, the substra-
tum suffered considerable erosion, with a large proportion of the Old English
vocabulary being replaced by words from the Latinate superstratum.
A further Latinate stratum was laid during the latter half of the sixteenth
century and into the seventeenth, during the period that is called the ‘Renais-
sance’. It was a period in which the classical civilisations of Greece and Rome
were rediscovered, admired and celebrated; their literatures republished and
extensively studied, translated and imitated. While words had been coming into
English directly from Latin, as well as via French, for some centuries, the trickle
now became a flood, and many thousands of Latin words were added to Eng-
lish, as well as Greek words, though these often came via Latin. The Renais-
sance also saw the beginnings of exploration, which developed in the eighteenth
and nineteenth centuries into colonisation and empire; the contact with many
different cultures and languages has enriched the vocabulary of English from a
multitude of sources.
The substratum of English vocabulary is Anglo-Saxon, and the one hundred
most frequently occurring words in both writing and speech are of Anglo-
Saxon origin. Overlaid on this substratum is a stratum of Latinate vocabulary,
mainly of French origin, from the medieval period: we do not recognise the vast
majority of these words as foreign imports any more; they have become quite
naturalised. Overlaid on this is a further Latinate stratum, taken directly from
Latin during and after the Renaissance; many of these words still betray their
origin, and they belong for the most part to the vocabulary of academic discourse
and specialist jargon. Additionally, English has imported words from countless
languages around the world, and continues to do so (see Crystal 1995: 126f.).
2.2 Making new words

There are two basic methods by which a language may increase its vocabulary.
The first is to use the material (morphemes) available in the language already
(see 1.6) and to recombine it in new ways. The other is to import a word from
another language (mentioned in 2.1 above), a process called, rather curiously,
‘borrowing’ – there is, after all, no intention to return the borrowed item,
which is termed a ‘loanword’. Nearly all new words are added to the larger
word classes (1.5), especially nouns, verbs and adjectives, with the majority
being nouns.
2.2.1 Compounds
Compounds are formed by joining two or more root morphemes or (classical)
combining forms (see 1.6) into a single lexeme. A new discovery, product,
sensation or process is often suitably named by a compound, whose status as a
lexeme is reinforced by usage and confirmed by inclusion in a dictionary. Com-
pounds are often idiomatic in meaning, or at least not entirely transparent. For
example, the meaning of seat belt – as a safety restraint in vehicles or aircraft –
is not immediately obvious from the two parts of the compound. If you were
unfamiliar with the object, you would need some explanation of the word. This
is even more so in the case of neo-classical compounds, where a knowledge of
Greek and Latin would be required for their interpretation; e.g. calligraphy (‘beau-
tiful’ + ‘writing’); mastectomy (‘breast’ + ‘cut out’); pachyderm (‘thick’ + ‘skin’),
denoting a large mammal with a thick skin, such as an elephant; stenothermal
(‘narrow’ + ‘heat’), i.e. tolerant of only small changes in temperature.
Where a compound is composed of more than two roots, a structure is
usually evident among the parts, which is sometimes reflected in how the
compound is written. For example, four-wheel drive indicates that four and wheel
belong together and relate as a unit to drive; whereas golden handshake indicates
that hand and shake belong together and golden is then added to form the three-
part compound.
An interesting compound is formed by the combination of two roots and the
addition of the –ed suffix, to form an adjective. The –ed suffix looks like the past
participle inflection of verbs (1.3), but there is no verb involved in this word
formation. Here are some examples: dark-haired, empty-handed, hard-nosed, jet-
lagged, muddle-headed, open-minded, quick-witted, round-shouldered, sharp-tongued,
warm-hearted. They are mostly, but not exclusively, composed of ‘adjective +
noun-ed’.
A special type of compound is formed by blending two roots; the first root
loses letters/sounds from the end and the second from the beginning, e.g. break-
fast +lunch > brunch, smoke + fog > smog, transfer + resistor > transistor. Sometimes,
one of the elements does not lose any material, e.g. car + hijack > carjack, cheese +
hamburger > cheeseburger; or there are shared letters, e.g. circle + clip > circlip, floppy
+ optical > floptical, twig + igloo > twigloo.
2.2.2 Derivatives
The addition of a derivational prefix or suffix to a lexeme forms a derivative.
The lexeme may be ‘simple’ (i.e. a single morpheme), or it may be a compound,
or it may be a derivative already; e.g. care-ful, landscape-(e)r, national-ity. Some
derivational affixes have their origin in Anglo-Saxon (e.g. -ful, -er), others have
come from French or Latin (e.g. -al, -ity); and while there is a tendency to use
Anglo-Saxon affixes with Anglo-Saxon roots and Latinate affixes with Latinate
roots, some mixing does occur, e.g. beauti-ful, preach-er (Latin root + Anglo-
Saxon suffix), fals(e)-ity, ship-ment (Anglo-Saxon root + Latin suffix).
Prefixes, of which usually not more than one is added to a root, do not
normally change the word class of the item to which they are added. Common
prefixes include those with a ‘negative’ or ‘opposite’ meaning, such as dis-, in-
(and its variants il-, im-, ir-), un-, the ‘again’ prefix re-, the ‘attitude’ prefixes pro-
and anti-, and the self- prefix. Here is an example of each: dis-please, in-decision,
il-legible, im-patient, ir-reversible, un-certain, re-read, pro-life, anti-freeze, self-addressed.
Suffixes are numerous and usually change the word class of the item they are
added to. Changing verbs to nouns are: -er (the ‘doer’/‘agent’ suffix), -(t)ion,
-ment, -ance; e.g. bak(e)-er, educat(e)-ion, enjoy-ment, perform-ance. Changing adjec-
tives to nouns are: -ity, -ness; e.g. sincer(e)-ity, smooth-ness. Changing adjectives
to verbs are: -en, -ify, -ise; e.g. thick-en, solid-ify, internal-ise. Changing verbs to
adjectives are: -able/-ible; e.g. avoid-able, collaps(e)-ible. Changing nouns to adjec-
tives are: -al, -ful, -ly; e.g. cultur(e)-al, hope-ful, friend-ly. And changing adjectives
to adverbs is: -ly; e.g. quick-ly, smooth-ly. More than one derivational suffix may
be added to a root, e.g. friend-li-ness, recover-abil-ity, care-ful-ly, nation-al-is(e)-
ation.
A special type of derivation occurs which changes the word class of a lexeme
but does not add a suffix. It is called ‘conversion’. For example, bottle is primarily
a noun, but it is used as a verb, with the sense ‘put into a bottle’, by conversion.
A contrary conversion would be catch, where the verb can also be used as a
noun. There are many cases of conversion (e.g. dirty (adjective to verb), skin
(noun to verb), spill (verb to noun), spoon (noun to verb) ) and it is still a produc-
tive process, especially from nouns to verbs, e.g. doorstep, handbag, progress, show-
case, text-message.
A minor type of derivation is ‘backformation’, a kind of derivation in reverse,
in which a supposed affix is removed from a word. This is how the verb edit was
derived from the noun editor, by removing the supposed ‘doer’ suffix -or (com-
pare actor, advisor). A similar backformation derived babysit from babysitter,
commentate from commentator, malinger from malingerer, scavenge from scavenger.
Automate was derived by backformation from automation, destruct from destruc-
tion, enthuse from enthusiasm, greed from greedy, sedate from sedation, televise from
television.
2.2.3 Acronyms
A minor, but nevertheless much used word formation process takes the initial
letters of a phrase and creates a word, called an acronym. Either the acronym
is pronounced as a normal word (e.g. AIDS (Acquired Immune Deficiency
Syndrome), UNESCO (United Nations Educational, Scientific and Cultural
Organisation) ), or the letters are spelled out (e.g. ATM (Automated Teller
Machine), HIV (Human Immunodeficiency Virus) ). Sometimes the two forms
are combined, e.g. CD-ROM (Compact Disc – Read Only Memory). The
acronym is usually spelt with capital letters, but a few acronyms no longer
betray their origin in this way, e.g. laser (‘light amplification by stimulated emis-
sion of radiation’). Here are some further examples, first of ‘said’ acronyms:
DAT (Digital Audio Tape), DWEM (Dead White European Male), MIDI
(Musical Instrument Digital Interface), SIMM (Single In-line Memory Mod-
ule); then of ‘spelled out’ acronyms (also called ‘initialisms’): BSE (Bovine
Spongiform Encephalopathy), CSA (Child Support Agency), FAQ (Frequently
Asked Question), HTML (HyperText Markup Language), LMS (Local Man-
agement of Schools).
A further type of acronym is formed by taking the first syllable of the words
of a phrase, e.g. biopic (biographical picture), infotech (information technology),
Ofsted (Office of Standards in Education), pixel (picture element). In the case
of Ofsted, the second element (st) does not consist of the full syllable, and an x
has been added to pixel to join the two syllables. These ‘syllabic acronyms’ are a
relatively rare formation.
2.2.4 Loanwords
When a word is ‘borrowed’ from another language and added to the vocabu-
lary, it is a ‘loanword’. Some loanwords continue to betray their origins, either
in their spelling or their pronunciation, or both (e.g. blitzkrieg (German), kibbutz
(Hebrew), spaghetti (Italian) ); while others have become naturalised (e.g. coach
(Hungarian), gong (Malay), tycoon ( Japanese) ).
Words have been borrowed into English for a number of reasons. After the
Norman conquest, a new language was imposed on top of the English, and so,
for example, beef, mutton and pork appeared alongside cow, sheep and pig. During
the Renaissance, excessive admiration for Roman and Greek cultures and lan-
guages led to the borrowing of words from Latin and Greek to remedy what
was felt to be a lack in English of erudite vocabulary; and so abscond was bor-
rowed alongside hide, calculate for count, emporium for shop, manuscript for book,
protect for ward, transgress for sin, valediction for farewell.
When the explorers and colonists went to new countries, experienced differ-
ent foods, and came into contact with plants and animals they had never en-
countered before, they often took the words for these things from the local
languages. So, we have chipmunk from Algonquian (in North America), kooka-
burra from Wiradhuri (in Australia), kiwi from Maori (in New Zealand), chutney
from Hindi, poppadom from Tamil, lychee from Chinese, sushi from Japanese,
impala from Zulu, sherbet from Turkish, and so on. Through the centuries,
when a culture has been admired for its prowess in a particular area, English has
borrowed its words for that topic; e.g. musical terms from Italian (concerto, opera,
soprano, tempo), culinary terms from French (casserole, fricassee, au gratin, purée,
sauté).
When a profession has sought an erudite vocabulary to mark off its supposed
area of competence, it has usually looked to the classical languages for its jargon.
The law, for example, has taken many terms from Latin, such as: ad litem (‘in a
lawsuit’), bona fide (‘with good faith’), corpus delicti (‘body of offence’), ejusdem
generis (‘of the same kind’), in personam (‘against the person’), lis pendens (‘a
lawsuit pending’), obiter dictum (‘a passing remark’ – by a judge), prima facie (‘at
first impression’), subpoena (‘under penalty’ – i.e. to attend court), ultra vires
(‘beyond (one’s legal) power’). Medicine, on the other hand, has tended to look
more to Greek for its jargon: an inflammatory disease ends in -itis (bronchitis,
peritonitis), a surgical removal ends in -ectomy (hysterectomy, vasectomy), the medi-
cal care of particular groups ends in -iatrics ( geriatrics, paediatrics).
English continues to enhance its vocabulary by taking in loanwords from
languages around the world. Some recent borrowing includes: balti (Urdu), ciabatta
(Italian), gite (French), intifada (Arabic), juggernaut (Hindi), karaoke ( Japanese),
nouvelle cuisine (French), ombudsman (Swedish), paparrazi (Italian), perestroika
(Russian), salsa (Spanish), tikka (Punjabi).
2.3 Word meaning

One of the most important tasks of a lexicographer is to capture the ‘meaning’
of a word in a ‘definition’ (see Chapter 8). We need to determine first of all
what constitutes the ‘meaning’ of a word, which is the purpose of this section of
the chapter. The suggestion is that the meaning of a word is composed of a
number of features: its relation with the real world, the associations that it
carries with it, its relations with other words in the vocabulary, and the regular
company that it keeps with other words in sentence and text structure.
Many words have more than one meaning; they manifest ‘polysemy’. Ascer-
taining how many meanings, or ‘senses’, a lexeme has, and in what order to
arrange them are difficult decisions for a lexicographer to make, and dictionaries
may differ quite markedly in their analysis. Our immediate discussion, how-
ever, is concerned with the general factors that may apply to any lexeme or
sense of a lexeme.
2.3.1 Reference
The primary feature of meaning is the relation of reference between a lexeme
and the entity – person, object, feeling, action, idea, quality, etc. – in the real
world that the lexeme denotes. The exact nature of the reference relation has
exercised the minds of linguists and philosophers over many centuries (Lyons
1977). We use words to talk about and make reference to the world we live in,
our experience of that world, our speculation about what might have been or
could be, our imagination of other possible worlds and possible scenarios. Our
worlds are inhabited by humans and other creatures, by natural objects and
artefacts, by our ideas, opinions and beliefs, which possess characteristics that we
describe, and which interact in a myriad ways. We can talk about all these things
and communicate about them with other people who speak the same language,
because we have a shared vocabulary and grammar. In particular, we agree
about which word refers to which aspect of reality or our experience of it.
The reference of some words is both more obvious and more easily de-
scribed. This is the case especially for tangible objects (bicycle, trumpet) and for
physical actions ( jump, spill). For words that denote more abstract entities, the
reference relation is less clearly discernible. This is the case for many abstract
nouns (deference, solitude), for verbs expressing mental and emotional states and
processes (think, worry), and for adjectives generally, especially where they are
gradable (long, warm) or evaluative (ridiculous, superb). For some words, belong-
ing to the smaller, grammatical classes (1.5), a relation of reference may be
scarcely discernible (about, this).
What we are often interested in, including lexicographically, is how words
that have a similar reference differ from each other. For example, how do hap-
pen, occur, befall, transpire and, perhaps, materialise differ? They all denote ‘come
about’ or ‘take place’ (LDEL2: 718). The differences are subtle and may have
little to do with reference as such, and more to do with context: occur would be
found in a more formal context than happen; befall has an old-fashioned ring to
it; transpire and materialise are, perhaps, particular kinds of ‘come about’. We
cannot isolate a word either from the typical contexts in which it occurs or from
its relationships with other words.
2.3.2 Connotation
A distinction is often drawn between the ‘denotation’ of a word and its ‘conno-
tation’. While the denotation is the straightforward, neutral relation between a
word and its referent, the connotation brings in the, often emotive, associations
that a word may have for a speaker or a community of speakers. For many
English speakers, the word champagne, while denoting a sparkling wine from a
particular region of France, has the connotation of celebration or expensive
living.
Some words spread particular negative or positive connotations (semantic
prosodies) across the phrases or sentences in which they occur. For example,
fundamentalist or fundamentalism, which denote ‘adherent/adherence to the fun-
damental teachings of a movement or religion’, are usually used in a negative
context and with a connotation of a fanaticism that should be disapproved of.
On the other hand, inspire, denoting ‘creating the desire to do or feel some-
thing’, usually has a positive connotation and spreads a positive semantic prosody,
occurring typically with nouns like confidence, enthusiasm or loyalty.
Such connotations are widely shared and may be or become intrinsic to the
contexts in which the users of a language generally situate the words. Connota-
tions may be more restricted in scope, to a particular generation (e.g. blitz to
those who lived through World War II), or to a particular group (e.g. safe to
those who have hazardous occupations), or even to an individual. A connota-
tion that is shared by a large proportion of speakers can be considered as a
contributory feature to the meaning of a lexeme.
2.3.3 Sense relations

A third contributory factor to the meaning of a lexeme or a sense of a lexeme is
the semantic relations it contracts with other lexemes in the vocabulary, often
termed ‘sense relations’. They include: sameness or similarity of meaning (syn-
onymy), oppositeness of meaning (antonymy), the ‘kind of ’ relation (hyponymy),
and the ‘part of ’ relation (meronymy).
Synonymy is a widespread relation in English, in large part because there are
words with similar meaning from more than one of the strata that make up the
vocabulary (2.1). For example, begin has an Anglo-Saxon origin, while its syno-
nym commence entered English from French during the medieval period; simi-
larly with keep and retain, leave and depart, tell and inform, live and reside, share and
portion, and so on. Equally, synonym pairs exist that derive, on the one hand,
from French in the medieval period, and on the other, from Latin during the
Renaissance: complete and plenary, join and connect, sign and portent, taste and
gustation, vote and plebiscite. There are, even, synonym triplets from each of the
three strata of vocabulary; e.g. end, finish, terminate; hatred, enmity, animosity;
kingly, royal, regal; sin, trespass, transgression. In general, as the examples cited
confirm, the synonym from the Latinate strata tends to be used in more formal
contexts than the one from the Anglo-Saxon substratum.
The other major source of synonym pairs is dialect difference, either between
national varieties (e.g. British and American English) or between dialects within
a national variety. The major differences between British and American English
are in vocabulary, rather than in grammar, e.g. (BrE word followed by AmE
word) biscuit, cookie; car park, parking lot; drawing pin, thumbtack; flannel, washcloth;
lorry, truck; single (ticket), one-way; waistcoat, vest; and many more. Here are some
synonym pairs for Scottish English and English English: birl, whirl; dree, endure;
fankle, entangle; kirk, church; lum, chimney; neep, turnip; outwith, outside; vennel, alley.
Antonymy is a less frequently occurring sense relation than synonymy. It is
most prevalent among gradable adjectives, where the antonyms represent the
opposite ends of a scale, e.g. big, small; wide, narrow; beautiful, ugly; quick, slow.
Other word classes also show antonymy: verbs begin and end, nouns bottom and
top, prepositions into and out of, adverbs above and below. Not all antonymy is of
the same type. In the case of gradable antonyms, the words are in a ‘more/less’
relation: wide and narrow cover overlapping parts of a spectrum, and an object is
wide or narrow in relation to some norm. In contrast some antonyms have an
‘either/or’ relation: win and lose are mutually exclusive, you do either one or the
other. A third type of antonym shows a ‘converse’ relation: buy and sell are the
converse of each other; if X sells some goods to Y, then Y buys them from X.
Hyponymy relates words hierarchically, with a superordinate word (hypernym)
having a more general meaning than the subordinate word (hyponym). The
hyponyms are in a ‘kind of ’ relation to the hypernym. For example, knife, fork
and spoon are kinds of cutlery; so, cutlery is the superordinate word, with general
meaning, and knife, fork and spoon are its hyponyms, with more specific mean-
ing. These is turn may be superordinate words to their hyponyms; spoon, for
example, has the hyponyms teaspoon, tablespoon, dessertspoon, ladle. A large part of
vocabulary can be viewed as being related by hyponymy, but, as with language
generally, there is no neat system of hyponymy relations organising the whole
vocabulary of English.
Meronymy is like hyponymy in that it relates words hierarchically, but the
relation is a ‘part of’ relation. The meronyms of a superordinate word represent
the parts of that word. For example, ball, heel and instep are meronyms of foot;
hub, rim and spoke are meronyms of wheel; flower, root and stalk are meronyms of
plant. Together, hyponymy and meronymy serve to group words into semantic
sets, known as ‘lexical fields’, in which the lexemes all refer to the same area of
meaning (see further Chapter 12).
2.3.4 Collocation
The sense relations between words are ‘paradigmatic’ relations: a synonym,
antonym, hyponym or meronym would substitute for its counterpart in some
slot in the structure of sentences. The meaning of a word is also determined by
its ‘syntagmatic’ relations, specifically by its collocation, the other words that
typically accompany it in the structure of sentences and discourses. For exam-
ple, the noun ban is typically modified by the adjective total or complete, is asso-
ciated with the verbs impose and lift, and is followed by the preposition on. In a
sentence with the verb spend, the Object would typically consist of either an
amount of money (two hundred pounds) or a period of time (last weekend). The
adjective flippant typically associates either with a noun referring to something
said (remark, answer, comment) or with the noun attitude.
The word ‘typically’ occurs in all these statements about collocation, because
collocation is a matter of the statistical probability or likelihood that two words
will co-occur. One of a pair may exercise a stronger attraction than the other;
for example, wine is more likely to co-occur with red than red is with wine,
because red can co-occur with many nouns, while wine occurs with only a small
number of adjectives. Description of collocation is most reliably based on the
analysis of large computer corpora of texts, which can yield appropriate statisti-
cal data.
To summarise, the components of (the sense of ) a lexeme’s meaning are: its
relations with the ‘real world’ in the form of its denotation and connotation; its
relations with other (senses of ) lexemes in the vocabulary; its relations with the
other lexemes that typically accompany it in the structure of sentences.
2.4 Describing words

In this concluding section of the chapter, we shall examine what constitutes the
description of a lexeme; in other words, what information about words a
lexicographer needs to take account of in framing a dictionary entry. Following
Hudson (1988), ‘lexical facts’ include: the form of a word, its structure, its
meaning, its grammar, its usage, and its origin.
By the ‘form’ of a word is meant its pronunciation (phonology) and spelling
(orthography). The description of pronunciation specifies what sounds (pho-
nemes) a word has, if it has more than one syllable how they are each stressed,
and if the pronunciation is subject to any variation in connected speech (e.g.
vowel reduction or change in stress). The description of spelling specifies the
letters that make up the word, any variant spelling, and possibly where the word
may be broken at the end of a line.
The structure of a word refers to its composition in terms of morphemes
(1.6), how the roots relate to each other in a compound word, what prefixes
and suffixes the word has and how they modify the meaning of the root. The
description of structure also needs to indicate if there are any pronunciation or
spelling changes, either in the root or in an affix, as a result of joining mor-
phemes together to form the word. For example, clear changes pronunciation
and spelling when the suffix -ify is added (clarify), as does discreet with the suffix
-ion (discretion); bake loses an ‘e’ when -er is added to form baker, as does debate
with suffix -able (debatable). The suffix -able alternates with -ible (discernible), with
no difference in pronunciation, depending on which Latin root it is added to.
The meaning of a word was discussed quite fully in 2.3. Both the reference
relation and any other relevant semantic relations (sense relations, collocation)
need to be described for an adequate account of meaning.
The description of grammar has two aspects: the inflections that a word has,
and how a word fits into the syntax of sentences. For inflections, the description
specifies which inflections the word may have (1.3), how they are pronounced
and spelled, and any changes to the form of the root that result from their
addition. For example, the addition of plural suffix -(e)s changes hoof to hoov-es,
city to citi-es, the addition of the past tense/past participle suffix -(e)d changes cry
to cri-ed, slap to slapp-ed. If a word has an irregular form (e.g. of plural or past
tense), this too will be specified, e.g. foot – feet, appendix – appendices, criterion –
criteria, buy – bought, tell – told, see – saw.
The description of the syntactic operation of words begins with their assign-
ment to a word class (1.5), which is an initial specification of where the word
may be used in the structure of sentences. Any deviation from the normal
expectation needs to be specified, e.g. if an adjective is restricted to one of the
three possible positions for adjectives (i.e. ‘attributive’ – before nouns (the brown
suit), ‘postpositive’ – after nouns (time enough), and ‘predicative’ – after a verb
like be (the suit is brown) ). For example, awake occurs as predicative (the baby is
awake) but not as attributive, and chief occurs as attributive (our chief concern) but
not as predicative; galore, emeritus and extraordinaire occur only in postpositive
position. For verbs, the specification of syntactic operation is even more com-
plex, including not only whether a verb may take an Object, Complement,
etc., but also what type of Object (e.g. noun phrase, nominal clause) and so on.
The description of usage specifies whether a word, or any of its senses, is
restricted to particular contexts. The restriction could be geographical (a na-
tional variety or a dialect), it could be time-bound (an obsolete or archaic mean-
ing), it could be the formality of the situation or the word’s status in the language
(e.g. slang or taboo). The restriction could be linked with the expression of the
speaker’s or writer’s attitude, to indicate disapproval or an insult, or to be appre-
ciative. Or a word may be restricted in its usage because to use it would be
offensive to a particular group of people.
Finally, the description of a word includes a specification of its origin, if it
belonged to Anglo-Saxon or if it has been ‘borrowed’, from which language
and when. ‘Origin’ is sometimes taken to mean the ‘ultimate’ origin, as far as
this can be ascertained; for example, a word taken from French during the
medieval period may have its origin traced back through older French to Latin.
This part of the description may also chart the history of changes in the form
(spelling and pronunciation) and in the meaning of the word.
These are the features of words, their lexical description, that lexicographers
must grapple with and from among which they must choose what to include in
their lexicographical descriptions, which are published in dictionaries.
2.5 Further reading

For many of the topics of this chapter, see: Jackson and Zé Amvela’s Words,
Meaning and Vocabulary (2000) and David Crystal’s The Cambridge Encyclopedia of
the English Language (1995), especially Part II.
On new words, see: John Ayto’s Twentieth Century Words (1999) and Eliza-
beth Knowles and Julia Elliott’s The Oxford Dictionary of New Words (1997).
Sense relations are discussed fully by D.A. Cruse in his Lexical Semantics (1996).
See Dick Hudson’s article in IJL (1988) for a ‘Checklist of Lexical Facts’.
The dictionary 21
3 The dictionary
How many times have you heard someone say, or have you said yourself, ‘I’ll
look it up in the dictionary’? The assumption behind such a comment is that
‘the dictionary’ is a single text, perhaps in different versions, rather like the
Bible. Every household is assumed to have one; children are taught how to
consult the dictionary in school; there is one in every office. Lawyers quote the
dictionary in court, teachers and lecturers appeal to it, politicians and preachers
argue from its definitions. The dictionary is part of the cultural fabric of our
society; each major new edition warrants a review in the daily press. And we all
take what the dictionary says as authoritative: if the dictionary says so, then it is
so. Life would be impossible if the dictionary was not the final arbiter in our
linguistic disputes.
Yet, walk into any bookshop and cast your eye over the shelf where the
dictionaries are, pick a few up and examine them, read the blurb on the dust
jacket, and you will soon notice that they are all different. They are all recognis-
ably dictionaries, with a more or less alphabetical list of words and information
about them, but they have different formats, highly variable numbers of pages,
a variety of page layouts, and so on. Compare some of the entries, and you soon
realise that the notion of ‘the dictionary’ as a single text is wide of the mark.
What distinguishes them is more notable than what they have in common.
3.1 What is a dictionary?

A dictionary is a reference book about words. It is a book about language. Its
nearest cousin is the encyclopedia, but this is a book about things, people, places
and ideas, a book about the ‘real world’, not about language. The distinction
between dictionary and encyclopedia is not always easy to draw, and there are
often elements of one in the other. But they do not share the same headword list
– you would be unlikely to find resemble in an encyclopedia – and they do not
provide the same information for the headwords that they do have in common.
Compare the following entries for toad:
toad Any of the more terrestrial warty-skinned members of the tailless

amphibians (order Anura). The name commonly refers to members of the
22 The dictionary
genus Bufo, family Bufonidae, which are found worldwide, except for the
Australian and polar regions.
Toads may grow up to 25 cm/10 in. long. They live in cool, moist places
and lay their eggs in water. The eggs are laid not in a mass as with frogs, but
in long strings. The common toad B. bufo of Europe and Asia has a rough,
usually dark-brown skin in which there are glands secreting a poisonous
fluid which makes it unattractive as food for other animals; it needs this
protection because its usual progress is a slow, ungainly crawl.
(Hutchinson New Century Encyclopedia)
toad/tfgd/n. 1 any froglike amphibian of the family Bufonidae, esp. of the

genus Bufo, breeding in water but living chiefly on land. 2 any of various
similar tailless amphibians. 3 a repulsive or detestable person. toadish adj.
[Old English tadige, tadde, tada, of unknown origin]
(COD9)
Dictionaries are usually arranged in alphabetical order of the headwords. Indeed

the expression ‘dictionary order’ is synonymous with ‘alphabetical order’. But
there are word books that are arranged by topic or theme, rather than by alphabet
(see Chapter 12), and they have a long history (Hüllen 1999; McArthur 1986).
Dictionaries are reference books. People consult them to find out informa-
tion about words. We must assume that compilers of dictionaries – lexicogra-
phers – include information that they know or expect people will want to look
up. What we cannot assume, however, is that lexicographers will exclude infor-
mation that they might expect users will not want to look up. A dictionary is
more than just a reference book; it is also a (partial) record of the vocabulary of
a language. Any dictionary contains entries and information that few, if any,
users will want to access, either because they know it already, or because it is of
no interest to them. It would be the rare user who would consult a dictionary
for information on the word the, and yet no dictionary would be without an
entry for the. However, anyone serious about discovering the subtleties of the
definite article in English would be more likely to consult a grammar book than
a dictionary.
If the dictionary is distinguished, as a reference book, from the encyclopedia
on the one hand, it is distinguished, as a linguistic description, from the gram-
mar book on the other. A grammar book, as the description of the grammatical
system of a language, deals with the general rules and conventions for the struc-
ture of sentences and tends to deal with words as classes or subclasses. A diction-
ary describes the operation of individual lexical items, including, where relevant,
how they fit into the general patterns of grammar. Grammar and dictionary are
complementary parts of the description of a language, and a dictionary will use
terms that are defined by the grammar. The point at which grammar and dic-
tionary converge in their treatment of words concerns primarily the so-called
‘grammatical’ words, like the definite article, which play a crucial and often
complex role in grammar.
The dictionary 23
Who, then, are the users of dictionaries, and for what purposes do they use
them? We readily think of students and learners, academics, word game and
crossword puzzle buffs as regular, if not frequent, users of dictionaries. Most
people probably have occasion to consult a dictionary from time to time, and
many of us have a fascination with words and dictionaries, as the long-running
television series Call My Bluff and Countdown demonstrate. Sometimes we just
want to establish the existence of a word, perhaps a derivation that we’re not
sure of. Or we want to check the spelling of a word. Or we look up a word that
we have met and with which we are not familiar, and whose meaning we need
to ascertain. These, surveys have shown, are the main uses that people make of
dictionaries. Occasionally, someone may wish to find out the pronunciation of
a word that they have encountered only in writing, or for the sake of general
interest look up a word’s etymology.
The upshot of this is that any dictionary contains a vast amount of informa-
tion that is unlikely to be consulted by any of its users. It is there because of the
dictionary’s recording function, its description of the lexical resources of the
language. The fulfilment of its recording function may, though, be in conflict
with the dictionary’s reference function, to provide useful information in an
easily accessible manner. We shall explore some of these issues further in
Chapter 7.
3.2 Dictionaries, not ‘the dictionary’

If there is no such publication as ‘the dictionary’, what is the range of publica-
tions that are called ‘dictionary’? First, we should distinguish between those
dictionaries that treat a single language from those that treat more than one,
usually two languages: the former are ‘monolingual’ dictionaries and the latter
are ‘bilingual’ dictionaries. Although, as we shall see in Chapter 4, bilingual
dictionaries have the longer pedigree and they contain in part similar informa-
tion to monolingual dictionaries, they are performing a quite different function
and have a number of crucial distinctives. In particular, bilingual dictionaries have
two sections, an A-language to B-language section (e.g. English–German), and
a B-language to A-language section (e.g. German–English); and in bilingual
dictionaries the definitions of words are the translation equivalents in the other
language. This book is concerned only with monolingual dictionaries.
Second, we should distinguish among monolingual dictionaries between those
whose purpose is primarily historical and those that seek to describe the vocabu-
lary at a particular point or period of time. The primary historical dictionary for
English is the multi-volume Oxford English Dictionary, and its abbreviated two-
volume offshoot, the Shorter Oxford English Dictionary, which aim to chart the
birth, death, and developments in form and meaning of words that have consti-
tuted the vocabulary of English since 1150 for the OED and since 1700 for the
SOED. The ‘synchronic’ dictionary, by contrast, takes a snapshot of the voca-
bulary at some point in time. Such a dictionary might chart the vocabulary of
Old English (Roberts et al. 1995) or of Middle English (e.g. Kurath and Kuhn
24 The dictionary
Table 3.1 Comparison of the Collins range
CED4 CCD4 CPED4
Page size 187 × 260 mm 152 × 234 mm 107 × 151 mm

No. of pages 1785 + xxxvii 1740 + xxi 632 + vii
‘References’ 180,000 no claim no claim
‘Definitions’ 196,000 no claim 44,500
Price (2001) £29.99 £16.99 £7.99
1954), or, most usually, of the contemporary language. While we devote a

chapter (5) to the OED, because of its importance in the development of lexico-
graphy, we are mostly concerned in this book with dictionaries charting the
contemporary vocabulary.
Even among dictionaries of the contemporary language there is a bewilder-
ing variety. Dictionaries vary according to size, from desk-size, through con-
cise, to pocket and smaller, with varying dimensions, numbers of pages, and
coverage. All dictionaries present a selection from contemporary vocabulary,
but it is very difficult to make comparisons, because of the confusingly different
methods of counting the contents ( Jackson 1998). Table 3.1 provides a rough
estimate of the relative sizes.
Dictionaries also vary according to their intended audience or user group.
Some dictionaries are aimed at young users at various stages in their growth and
educational development; they are characterised by an appropriate selection of
vocabulary, limited amounts of information for each entry, and often the use of
pictures and colour. There is a range of monolingual English dictionaries that is
aimed at learners of English as a second or foreign language, which take into
account the particular needs of this group of users. The ‘monolingual learners’
dictionaries’ (MLDs) are an interesting set of reference works, and they have
been associated with some of the most exciting lexicographical innovations.
They are discussed in more detail in Chapter 11. The dictionaries aimed at the
native speaker adult user might be termed the ‘general-purpose’ dictionary
(Béjoint 2000: 40). They are the dictionary that most people own, and they are
the focus of much of the discussion in this book.
Besides the general-purpose dictionary, a wide variety of ‘specialist’ diction-
aries is published. Some specialist dictionaries focus on an aspect of lexical de-
scription: there are dictionaries of pronunciation (e.g. Jones 1997; Wells 2000),
dictionaries of spelling (e.g. West 1964), and dictionaries of etymology (e.g.
Weekley 1967). Other specialist dictionaries focus on the vocabulary of a topic
or subject-matter, e.g. Dictionary of Economics (Pearce 1992), Dictionary of
Lexicography (Hartmann and James 1998). Such dictionaries define the term-
inology that is crucial for talking about the subject; they exclude some lexical
information (e.g. pronunciation, grammar, etymology); and they tend towards
the encyclopedic, both in the extent of their definitions or explanations, and in
their inclusion of entries for people who have made a significant contribution to
the development of the subject.
The dictionary 25
The term ‘dictionary’ is thus applied to a diverse range of reference publica-
tions. Our focus will be on the general-purpose dictionary of desk and concise
size, with some consideration of historical dictionaries and those for learners.
3.3 What is in a dictionary?

From the perspective of its ‘macro-structure’, there are potentially three parts to
a dictionary: the front matter, the body, and the appendices. Some dictionaries
do without appendices, but most have front matter, however brief. The front
matter usually includes an introduction or preface, explaining the innovations
and characteristics of the edition concerned, together with a guide to using
the dictionary, which may consist of a single-page diagram or some lengthier
account. Other front matter might be an explanation of the transcription system
used for indicating pronunciation, a list of abbreviations used in the dictionary,
and an essay on some relevant topic, such as the history of the language or
varieties of English around the world. Appendices may be various and even
non-lexical; here is a selection: abbreviations, foreign words and phrases, ranks
in the armed forces, counties of the UK and states of the US, weights and
measures, musical notation, Greek and Cyrillic alphabets, punctuation, works
of Shakespeare.
The body of a dictionary contains an alphabetical list of ‘headwords’. Each
headword is accompanied by a number of pieces of information, which to-
gether with the headword constitute the ‘entry’. The headword is usually printed
in bold type and hangs one or two spaces to the left of the other lines. Entries are
presented in two columns on each page, though there may be three columns in
some, usually larger dictionaries (e.g. NODE, W3, but also ECED).
The headwords represent the particular selection of vocabulary and other
items that the editors have decided merit inclusion, given the size and purpose
of the dictionary. General-purpose dictionaries will all tend to share a headword
list that encompasses the core vocabulary; where they differ will be in the amount
of technical and specialist, as well as colloquial, slang and dialect vocabulary they
include. Editors will be concerned to be up-to-date, especially in socially and
culturally significant areas such as computing, medicine, the environment, fash-
ion, and so on. The inclusion of the latest vocabulary in such areas is often used
as a selling point for a new edition.
If you examine the headwords in a general-purpose dictionary, you will find
that it includes more than just lexemes. In terms of lexemes, it will include:
‘simple’ lexemes (1.6); compounds, possibly all, but at least those written solid
(without a hyphen); and derivatives whose meanings are considered to need a
separate definition from their roots. Other derivatives are contained within the
entry for the root, as ‘run-ons’, usually in bold type but without a definition.
The headword list will usually include inflected forms where these are ‘irregu-
larly’ formed (1.3) and are alphabetically some distance from the citation form
(e.g. bought in relation to buy): the entry will contain just a cross-reference to the
citation form. The list may also include items that are not lexemes, especially
derivational affixes and combining forms (1.6), and abbreviations. In some
26 The dictionary
dictionaries (e.g. CED, NODE) the headword list includes names of places and
people, introducing geographical and biographical entries, e.g.
Birmingham/’buhming(h)em/2nd largest British city, in the W Midlands

of England; a major industrial, service, and transport centre with growing
high-tech and light industries; home of two universities, a symphony
orchestra, and the National Exhibition Centre; est. pop. 998,200 (1987)
(LDEL2)
Angelou/’anhflu:/, Maya (b.1928), American novelist and poet, acclaimed

for the first volume of her autobiography, I Know Why the Caged Bird Sings
(1970), which recounts her harrowing experiences as a black child in the
American South.
(NODE)
Some headwords will be entered more than once. This applies to homonyms
(1.2), e.g. spell, with four entries in COD9, and to homographs, e.g. bow, with
one entry pronounced /bfg/ and two entries pronounced /bag/. In some dic-
tionaries (e.g. LDEL) each word class that a headword belongs to will occasion
a separate entry; for example, rear has four entries in LDEL2, one each for the
verb, noun, adjective and adverb uses of the headword.
The ‘micro-structure’ of a dictionary refers to the arrangement of the infor-
mation within the entries. The range and type of information within an entry
will vary according to the kind of headword, but will typically include some or
all of the following (compare 2.4):
• Spelling: the headword indicates the normal spelling, but any variations will
follow.
• Pronunciation: within rounded ( ) or slash // brackets, together with any
variations.
• Inflections: if these are formed irregularly or occasion some spelling adjust-
ment such as doubling of consonants, dropping of ‘e’ or changing ‘y’ to ‘i’.
• Word class: usually indicated by conventional abbreviations, ‘n’ for noun,
‘adj’ for adjective, etc.; verbs are also marked for ‘transitive’ (vt) or ‘intran-
sitive’ (vi).
• Senses: where a lexeme has more than one meaning, each sense is usually
numbered; where a sense, or group of senses belong to a different word
class or subclass, this is indicated before the sense(s) concerned.
• Definition: each sense is given a definition, which is an explanation of its
meaning.
• Examples: where the elucidation of a sense benefits from an illustrative
phrase or sentence, usually given in italic type.
• Usage: where a sense is restricted in its contexts of use, an appropriate label
precedes the sense concerned; if the restriction applies to all the senses of a
lexeme, the label precedes any of the senses.
The dictionary 27
• Run-ons: undefined derivatives (with a word class label), idioms, phrasal
verbs (if they are not included as headwords), usually in bold type.
• Etymology: conventionally in square brackets as the final item in the entry.
Some dictionaries include additional information, for example on collocation

or the syntactic operation of words. Learners’ dictionaries, especially (see Chap-
ter 11), contain detailed information on these topics, as well as other additional
material. By way of illustration, here is the entry for drink from COD9:
drink/drijk/ v. & n. v. (past drank/dræjk/; past part. drunk/drkjk/) 1 a

tr. swallow (a liquid). b tr. swallow the liquid contents of (a container). c
intr. swallow liquid, take draughts (drank from the stream). 2 intr. take alcohol,
esp. to excess (I have heard that he drinks). 3 tr. (of a plant, porous material,
etc.) absorb (moisture). 4 refl. bring (oneself etc.) to a specified condition by
drinking (drank himself into a stupor). 5 tr. (usu. foll. by away) spend (wages
etc.) on drink (drank away the money). 6 tr. wish (a person’s good health,
luck, etc.) by drinking (drank his health). n. 1 a a liquid for drinking (milk is
a sustaining drink). b a draught or specified amount of this (had a drink of
milk). 2 a alcoholic liquor (got the drink in for Christmas). b a portion, glass,
etc. of this (have a drink). c excessive indulgence in alcohol (drink is his vice).
3 (as the drink) colloq. the sea. drink deep take a large draught or draughts.
drink in listen to closely or eagerly (drank in his every word). drink off drink
the whole (contents) of at once. drink to toast; wish success to. drink a
person under the table remain sober longer than one’s drinking com-
panion. drink up drink the whole of; empty. in drink drunk. drinkable
adj. [Old English drincan (v.), drinc(a) (n.), from Germanic]
We examine the micro-structure and the information contained in dictionary

entries in more detail in Chapters 8, 9 and 10.
3.4 Compiling a dictionary

No lexicographer of English starts with a blank sheet of paper, but rather stands
in a tradition of dictionary making that reaches back more than six centuries
(Green 1996: 39), a history that we shall begin to trace in the next chapter.
While some lexicographers find themselves revising and updating an existing
dictionary to produce a new edition, others take on the challenge of innovation
and hack a fresh path for lexicography. Even then, they build on the work of
previous generations of lexicographers, both in determining the headword list
and in deciding what kinds of information to provide.
Briefly, we may identify three aspects to dictionary compilation: the selection
of headwords, the sources of data, and the writing of the entries. Any dictionary
contains a selection from the total vocabulary of English, which is difficult to
estimate but probably lies between one and two million words (Crystal 1995).
Dictionaries do not usually reveal their headword count, which would be
28 The dictionary
Table 3.2
CED4 NODE
gl. GLA
glabella glabella
glabrescent glabrous
glabrous glacé
glacé glacé icing
glacial glacial
glacial acetic acid glacial period
glacialist glaciated
glacial period glaciation
glaciate glacier
glacier Glacier Bay National Park
glacier cream glaciology
glacier milk glacis
glacier table glad1
glaciology glad2
glacis gladden
glacis plate gladdon
glad glade
Gladbeck glad eye
gladden glad-hand
gladdon gladiator
glade gladiolus
glad eye glad rags
glad hand Gladstone
gladiate Gladstone bag
gladiator Glagolitic
gladiatorial
gladiolus
glad rags
gladsome
Gladstone
Gladstone bag
Glagolitic
unreliable in any case, as it depends on what items are included as headwords

(e.g. affixes and abbreviations) and how compounds and derivatives are treated.
A desk-size dictionary probably contains no more than 100,000 headwords; the
CD-ROM version of COD10 gives the headword count as 64,679. Headword
lists in similar size dictionaries differ only at the margins: the core vocabulary is
standard, judgements are made about specialist and non-standard (slang, dialect)
lexemes. Compare the brief lists of headwords between gl- and glag- in CED4
and NODE, both published in 1998 shown in Table 3.2.
The lexicographers’ data comes from a number of sources. First of all, they
have access to previous dictionaries, which can be mined both for the headword
list and for lexical information. It is not unusual to find the same definition
reproduced in successive editions of a dictionary. Second, dictionary publishers
The dictionary 29
keep a ‘citation file’, which records the results of the publisher’s reading pro-
gramme in identifying new words together with examples of their contexts of
use, usually in the form of complete sentences. Some citation files go back a
long way, Oxford’s, for example, to the mid-nineteenth century, when cita-
tions began to be collected for what became the OED (see Chapter 5). Third,
and of increasing importance, lexicographers have access to computer corpora,
large collections of texts in electronic form. Oxford and Longman lexicogra-
phers use the British National Corpus, a 100 million-word corpus of both spoken
and written English; Collins lexicographers use the Bank of English, a growing
corpus, developed at the University of Birmingham, now of more than 400
million words, originally put together for the pioneering COBUILD learners’
dictionary (see Chapter 11).
A computer corpus can be searched rapidly and efficiently. It can be used for
checking information, or for seeking answers to specific queries. But, more
significantly, it can provide the raw data for the construction of dictionary
entries. Using a ‘concordance’ program, a lexicographer can perform a KWIC
(Key Word in Context) search and obtain a list of all the occurrences of a
lexeme in a corpus, together with a specified amount of context for each. The
results of the search suggest to the lexicographer how many senses to identify for
the lexeme and provide examples of use.
The third aspect of compiling a dictionary, identified earlier, was writing the
entries. It is rare that a dictionary is the work of a single lexicographer. A team
is more usual, with some members specialising in particular aspects of lexical
description. Many dictionaries have a pronunciation specialist, for example, or
an etymology specialist, as well as consultants for technical areas of vocabulary
or for other varieties of English. Lexicographers write the definitions, and edi-
torial staff coordinate the input of all the contributors. Dictionaries are now-
adays compiled on computer, so that all members of a team can have simultaneous
access to the developing dictionary text. This makes rigorous editorial check-
ing, always a necessity, even more important, before a dictionary is released for
publication. We pursue the topic of dictionary compilation in Chapter 13.
3.5 Evaluating a dictionary

Dictionaries are commercial publications; publishers invest considerable sums
of money in their development; and they are tailored to perceived market needs.
Like any other book publication, they are subjected to review in newspapers,
magazines and professional journals. Newspaper reviews of dictionaries tend
towards the trivial, focusing on ‘newsworthy’, often idiosyncratic features, such
as who has been included and excluded from the biographical entries, or sup-
posed modish, usually slang, lexical items. However, dictionaries are not just
commercial publications; they are also linguistic descriptions and so they are of
interest to language and linguistics scholars, who subject them to academic
scrutiny and criticism. Indeed, a specialist branch of linguistic studies has devel-
oped whose concern is specifically lexicography: it has its scholarly associations
30 The dictionary
(e.g. EURALEX – the European Association for Lexicography), its own jour-
nals (e.g. International Journal of Lexicography), a three-volume encyclopedia de-
voted to it (Hausmann et al. 1989), and its own courses and research projects.
Academic lexicography, or ‘metalexicography’ (Béjoint 2000: 8n), is con-
cerned, among other things, with the business of ‘dictionary criticism’ (Osselton
1989), which proposes methods and criteria for reviewing and evaluating dic-
tionaries. The reviewing of dictionaries is not like that of other books. It would,
for example, be impossible for a reviewer to read the whole text of a dictionary:
CED4 claims to have 3.6 million words of text, and NODE 4 million. Dictionary
reviewers must find other methods, such as sampling, or having a carefully
selected checklist of items and features to investigate.
One approach is to take the claims that a dictionary makes about itself, in the
blurb on the cover or book-jacket or in the front matter, and check these
against the practice of the dictionary as reflected in its content, as well as against
the accumulated insights and judgements of the scholarly community. An alter-
native approach establishes a set of criteria that arise from the academic study of
lexicography and applies these to the dictionary under review. It is often useful
to have a team of reviewers, each of whom takes a separate aspect for critical
scrutiny, e.g. the treatment of pronunciation, of grammar, of meaning, of ety-
mology (compare Higashi et al. 1992).
A further consideration in dictionary criticism is the perspective from which
the review is conducted. The academic metalexicographer’s primary focus is
probably on the adequacy of a dictionary as a lexical description. An alternative
focus might be that of the user, particularly where accessibility and comprehen-
sibility of the information could be an issue, as with a learners’ dictionary, or
where a specific set of users is being targeted, as with a children’s dictionary.
Dictionary criticism is an important activity. It not only provides informed
reviews of dictionaries for potential users, it also contributes to advances in
lexicography and to improvements in dictionaries. We explore it further in
Chapter 14.
Summarising, this chapter has sought to distinguish dictionaries from encyclo-
pedias and grammars, to show that dictionaries are the products of a tradition of
lexicography, to suggest some of the range of reference works with the ‘diction-
ary’ title, to survey the content of general-purpose dictionaries, to raise some of
the issues in dictionary compilation, and to introduce the business of dictionary
criticism. We have set the agenda for the remainder of the book, beginning
with an account of the history of dictionary making in English.
3.6 Further reading

A good overview is Sidney Landau’s Dictionaries: The Art and Craft of Lexicogra-
phy (originally 1989, now in a second edition, 2001). Also recommended is
Henri Béjoint’s (1994) Tradition and Innovation in Modern English Dictionaries,
Clarendon Press Oxford, republished in paperback, with only minor updating,
in 2000 as Modern Lexicography: An Introduction, Oxford University Press.
The beginnings 31
4 The beginnings
This chapter and the next two trace the history of dictionary making in English
up to the present time. This chapter takes us up to Samuel Johnson’s dictionary
in the mid-eighteenth century, the next is devoted to the Oxford English Diction-
ary, and Chapter 6 first recaps on the American practice and then brings the
story up to date.
4.1 Bilingual beginnings

The beginnings of English lexicography go back to the Old English period
(2.1), specifically to the introduction (from 597) of the Roman form of Chris-
tianity and the development of monasteries. The language of the Roman Church
was Latin; its priests and monks needed to be competent in Latin in order to
conduct services, and to read the Bible ( Jerome’s ‘Vulgate’ version) and other
theological texts. The monasteries were the institutions of education for the
clergy in the language of the church, as well as in the doctrines and practices of
the faith. Many monasteries also developed extensive libraries of theological and
other manuscripts (printing was still 750 years in the future), which would have
been written in Latin, and which became objects of study and commentary. As
English monks studied these Latin manuscripts, they would sometimes write
the English translation above (or below) a Latin word in the text, to help their
own learning, and as a guide to subsequent readers. These one-word transla-
tions, written between the lines of a manuscript, are called ‘interlinear glosses’;
they are seen as the beginnings of (bilingual) lexicography (Hüllen 1989).
In due course, and to aid in the teaching and learning of Latin, these glosses
were collected together into a separate manuscript, as a glossary, which may
be regarded as a prototype dictionary. The words in the glossary were then
ordered, either alphabetically, in early glossaries only by the first letter, then by
second and subsequent letters, or topically (Chapter 12). One of the best known
topical glossaries was compiled by Ælfric, who was the Abbot of the monastery
at Eynsham, near Oxford, during the first decade of the eleventh century. Ælfric
was well known as an educator: he wrote a grammar of Latin, as well as a
number of other instructional works. His glossary, known as ‘The London
Vocabulary’, is found appended to a number of extant copies of his Grammar.
32 The beginnings
The glossary is a list of Latin words, arranged by topic, together with an Old
English equivalent for each of them. Ælfric’s topics encompassed a wide range
of vocabulary, from ‘God, heaven, angels, sun, moon, earth, sea’ to ‘herbs’ and
‘trees’, to ‘weapons’ and ‘metals and precious stones’ (for full, but slightly differ-
ing lists, see McArthur 1986: 75, Hüllen 1999: 64 – reproduced in 12.2).
Latin continued as the language not only of the church but also of education
and learning generally throughout the medieval period. It was the language of
instruction for all subjects in the medieval universities (Oxford dates from 1167,
Cambridge from 1230), and scholarly publication was in Latin, the European
lingua franca of education. Academics were expected to be able to both speak
and write fluently in Latin. When schools were founded in order to prepare
students for entry to the universities, they concentrated on teaching Latin – the
origin of the ‘grammar’ school. There thus developed a considerable demand
for instructional material for the teaching and learning of Latin grammar and
vocabulary. Dictionaries were compiled to meet this demand, both Latin–
English (e.g. the Hortus Vocabulorum, ‘garden of words’, of around 1430) and
English–Latin (e.g. the Promptorium Parvulorum, ‘storeroom for young scholars’,
of 1440). Both of these dictionaries appeared in due course in printed form, the
Hortus in 1500, and the Promptorium in 1499.
Latin took on a new significance during the period of the Renaissance (2.1),
as scholars rediscovered the literature of Roman authors and made their works
known, both through publication in the original language and through transla-
tions into English. It is the latter that are of particular significance. When trans-
lators came across a Latin word for which they could not find a ready equivalent
in English, a common solution would be to ‘borrow’ the Latin word into
English. Since Latin had been for so long the common language of academic
discourse, this practice seemed the most convenient to many translators. How-
ever, since many readers would not be as familiar with Latin, some translators
appended a glossary of such ‘borrowed’ words to their translations. Philemon
Holland, for example, who published a translation of Plutarch’s Moralia in 1603,
appended ‘An explanation of sundry tearmes somewhat obscure, in this trans-
lation of Plutarch, in favour of the unlearned Reader; after the order of the
Alphabet’. The ‘unlearned’ reader was one who did not know Latin. As it
happens, Holland’s translation was the last to contain such a glossary, because of
a significant development in lexicography.
Before we come to that, let us note that the Renaissance period saw not only
the revival of the classical languages of Rome and Greece, but also a burgeoning
interest in the vernacular languages of Europe. This interest, prompted by in-
creasing travel, resulted in a number of bilingual dictionaries: for French and
English, John Palsgrave’s Esclarcissement de la langue francoyse (1530) and Randle
Cotgrave’s A Dictionarie of the French and English Tongues (1611); for Italian and
English, John Florio’s A Worlde of Wordes (1598); for Spanish, English and Latin,
Richard Percyvall’s Bibliotheca Hispanica (1591). Dictionaries for English and
Latin also continued to be published, e.g. Richard Huloet, Abecedarium Anglo-
Latinum (1552), Thomas Thomas, Dictionarium Linguae Latinae et Anglicanae (1587).
The beginnings 33
4.2 ‘Hard’ words
The first monolingual English dictionary is considered to be Robert Cawdrey’s
A Table Alphabeticall of 1604, which contained in fuller book form the kind of
list that Philemon Holland had appended to his translation of Plutarch. The title
page of Cawdrey’s dictionary proclaims it to be:
A Table Alphabeticall, conteyning and teaching the true writing, and

understanding of hard, usuall English wordes, borrowed from the Hebrew,
Greeke, Latine, or French, &c.
With the interpretation thereof by plaine English words, gathered for the
benefit & helpe of Ladies, Gentlewomen, or any other unskilfull persons.
Whereby they may the more easilie and better understand many hard
English wordes, which they shall heare or read in Scriptures, Sermons, or
elsewhere, and also be made able to use the same aptly themselves.
‘Unskilful’ persons, like Holland’s ‘unlearned reader’, would be those without a

knowledge of the classical languages, especially Latin; and since girls and young
women did not enjoy the same educational opportunities as boys and young
men – the ‘public’ schools and the universities were exclusively male preserves
– this applied to all women, apart from those with parents enlightened and
wealthy enough to have provided them with private tutoring.
Cawdrey’s A Table Alphabeticall begins a tradition of ‘hard word’ dictionaries.
You will have noticed that Cawdrey uses the word ‘hard’ twice on his title
page. A ‘hard’ word was a loanword, usually of recent borrowing, whose use
was not yet widespread and which was not readily comprehensible to ‘unedu-
cated’ readers. Despite his inclusion of Hebrew in his list of languages of origin
and the ‘&c’ (i.e. etc.) after ‘French’, Cawdrey marks only words of Greek
origin (with ‘g’ or ‘gr’) and words of French origin (with ‘§’), the unmarked
ones being assumed to have a Latin origin. Cawdrey’s book, despite its recog-
nition as the first monolingual dictionary, is not entirely original; in 1596 The
English Schoole Master by Edmund Coote had appeared, which contained a gram-
mar, the catechism, prayers, and a vocabulary, and it is this last that Cawdrey
mined for his work – even the title pages have similar wording. Cawdrey,
though, has twice as many words as Coote, and he used other sources as well.
Cawdrey’s first edition contained around 2500 ‘hard’ words, and it went
through four editions, the last published in 1617, but there was little augmenta-
tion of the word list. Each word in the dictionary is provided with a synonym
or explanatory phrase in ‘plaine English words’. Here are the first few words of
Cawdrey’s A Table Alphabeticall:
§ ABandon, cast away, or yeelde up, to leave, or forsake

Abash, blush
abba, father
§ abbesse, abbatesse, Mistris of a Nunnerie, comforters of others
34 The beginnings
§ abbettors, counsellors
aberration, a going a stray, or wandering
abbreviat, ) to shorten, or make short
§ abbridge, )
§ abbut, to lie unto, or border upon, as one lands end meets with another
abecedarie, the order of the Letters, or hee that useth them . . .
Apocrypha (g), not of authoritie, a thing hidden, whose originall is not
knowne
Cawdrey’s pioneering work was followed in 1616 by John Bullokar’s An Eng-

lish Expositor, whose title page proclaimed:
An English Expositor: Teaching the Interpretation of the hardest words

used in our Language.
With Sundry Explications, Descriptions, and Discourses.
By I.B. Doctor of Physicke.
Besides having more entries than Cawdrey – it contained ‘sundry olde words
now growne out of use, and divers termes of art, proper to the learned’ –
Bullokar also provides more expansive explanations, e.g.
Heretike. He that maketh his owne choice, what points of religion he will
beleeve, and what he will not beleeve.
Hereditarie. That which commeth to one by inheritance.
Heriot. The best living beast which a Tenant hath at his death, which in
some Mannors is due to the lord of whom the land is holden.
Hermaphrodite. Of both natures: which is both man and woman.
Hermite. One dwelling solitarie in the wildernesse attending onely to
devotion.
By his death in 1641, the Expositor had reached its third edition with little
revision. A radical revision and expansion of the Expositor in 1663 by someone
who styled themself ‘A Lover of the Arts’ greatly increased its popularity and it
continued to be republished until 1731.
Part of the expansion of Bullokar’s English Expositor in 1663 involved extensive
borrowing from a third hard-word dictionary, which had been first published in
1623, Henry Cockeram’s The English Dictionarie, and the first to use ‘dictionary’
in its title. On the title page of one of the first editions, though not subsequently
repeated, Cockeram acknowledged his debt to Cawdrey and Bullokar:
The English Dictionarie: or, An Interpreter of hard English Words.

Enabling as well Ladies and Gentlewomen, young Schollers, Clarkes,
Merchants, as also Strangers of any Nation, to the understanding of the
The beginnings 35
more difficult authors already printed in our Language, and the more speedy
attaining of an elegant perfection of the English tongue, both in reading,
speaking and writing.
Being a Collection of the choicest words contained in the Table
Alphabeticall and English Expositor, and of some thousand of words never
published by any heretofore.
Cockeram’s target audience is wider than Cawdrey’s, even extending to the

foreign learner of English (‘Strangers of any Nation’). Moreover, Cockeram’s
Dictionarie has three parts: the first is the list of hard words, together with their
glosses and explanations (more in the style of Cawdrey than of Bullokar); the
second is a list of ‘vulgar’ words together with their ‘refined or elegant’ equiva-
lents, as an aid to writing with good style; and the third, following the practice
of some Latin–English dictionaries, is a list of ‘Gods & Goddesses’. The 1663
revision of Bullokar’s Expositor included the second and third parts of Cockeram’s
Dictionarie. Cockeram’s work went through twelve editions, the last, a substan-
tially revised one, in 1670.
The scope of the hard-word dictionary had already widened since the publi-
cation of A Table Alphabeticall. It was widened further with the publication in
1656 of Thomas Blount’s Glossographia, whose title page declared:
Glossographia: or a Dictionary, Interpreting all such Hard Words, Whether

Hebrew, Greek, Latin, Italian, Spanish, French, Teutonick, Belgick, Brit-
ish or Saxon; as are now used in our refined English Tongue.
Also the Terms of Divinity, Law, Physick, Mathematicks, Heraldry,
Anatomy, War, Musick, Architecture; and of several other Arts and Sci-
ences Explicated.
With Etymologies, Definitions, and Historical Observations on the same.
Very useful for all such as desire to understand what they read.
Blount, lawyer by profession, borrowed extensively from other dictionaries,

both monolingual and bilingual, but he also included words that he had come
across in his reading or that he had heard spoken around him in London. Blount’s
definitions vary in length, from single words to extensive explanations, but his
singular innovation was the introduction of etymologies and ‘historical obser-
vations’. The etymology consists of the word in the original language in brackets
after the headword, e.g.
Deprehend (deprehendo) to take at unawares, to take in the very act.
Depression (depressio) a pressing or weighing down.
The ‘historical observations’ are included in the explanation and are sometimes
rather fanciful, e.g.
36 The beginnings
Hony-Moon, applyed to those marryed persons that love well at first, and
decline in affection afterwards; it is hony now, but it will change as the
moon.
Blount is also notable for giving his sources, or citing his authorities, an issue
that would become increasingly important. For example:
Depredable (depredabilis) that may be robbed or spoiled. Bac.
The ‘Bac.’ refers to Francis Bacon (1561–1626), the philosopher and scientist,
Blount’s source for this word. Blount is more comprehensive than any of his
predecessors, but the focus is still on the ‘hard’ words, with the addition of the
technical terms of ‘arts and sciences’.
Hard word dictionaries continued to expand: Edward Phillips’ The New World
of English Words of 1658 contained around 11,000 entries, which had increased
to 17,000 by the fifth edition in 1696, the year of Phillips’ death. Elisha Coles’
An English Dictionary of 1676 expanded his headword list to 25,000, largely by
adding dialect words, old words from Chaucer and Gower, and canting terms.
The canting terms were thieves’ slang words, and Coles justifies their inclusion
as follows:
’Tis no Disparagement to understand the Canting Terms: It may chance to

save your Throat from being cut, or (at least) your Pocket from being
pick’d.
Coles’ dictionary represented the state of the art in lexicography at the end of
the seventeenth century. However, it still did not contain the everyday vocabu-
lary of English. A truly comprehensive dictionary was still to come.
4.3 Completeness
The monolingual English dictionary had started life at the beginning of the
seventeenth century as a modest list of loanwords. As the century progressed the
word list expanded, mostly in the direction of the more unusual type of lexeme.
Etymology began to be attended to, and before the century ended two etymo-
logical dictionaries had appeared: Stephen Skinner’s Etymologicon Linguae
Anglicanae, published four years after the author’s death in 1671; and the anony-
mous Gazophylacium Anglicanum in 1689, which took much of its material from
Skinner.
The beginning of the eighteenth century brought a new focus to the mono-
lingual English dictionary, with the publication in 1702 of A New English Dic-
tionary, whose author is identified only by the initials ‘J.K.’ It is widely supposed
that the author is John Kersey, who revised Edward Phillips’ New World of
English Words in 1706 and who published a dictionary under his full name in
1708, the Dictionarium Anglo-Britannicum. J.K.’s dictionary proclaims itself to be:
The beginnings 37
A New English Dictionary: Or, a Compleat Collection Of the Most Proper
and Significant Words, Commonly used in the Language; With a Short and
Clear Exposition of Difficult Words and Terms of Art.
The whole digested into Alphabetical Order; and chiefly designed for
the benefit of Young Scholars, Tradesmen, Artificers, and the Female Sex,
who would learn to spell truely; being so fitted to every Capacity, that it
may be a continual help to all that want an Instructor.
Most of J.K.’s 28,000 headwords had never before appeared in a dictionary. Its
aim is to be ‘compleat’ and to identify the ‘proper’ words of the language; its
target audience includes the increasingly literate tradesmen and craftsmen; and
its primary purpose is to aid its users in correct spelling. Many of the current
school textbooks contained spelling lists; J.K. incorporates this feature into his
dictionary and thus brings into the dictionary the words of everyday vocabu-
lary. Many of the headwords in A New English Dictionary have only the scantiest
of definitions or explanations, e.g.
An Apron, for a Woman, &c.

An Arm of a man’s body, of a tree, or of the sea.
An Elephant, a beast.
May, the most pleasant Month of the Year.
Little serious attention was paid to etymology; at best, the language from which
a loanword was borrowed is indicated.
The two principles, of completeness and etymology, came together in
Nathaniel Bailey’s An Universal Etymological English Dictionary of 1721, which
promised both a larger scope and a wider group of users than its predecessors:
An Universal Etymological English Dictionary: Comprehending The Deri-

vations of the Generality of Words in the English Tongue, either Antient
or Modern, from the Antient British, Saxon, Danish, Norman and Modern
French, Teutonic, Dutch, Spanish, Italian, Latin, Greek, and Hebrew Lan-
guages, each in their proper Characters.
And Also A Brief and clear Explication of all difficult Words . . . and
Terms of Art . . .
Together with A Large Collection and Explication of Words and Phrases

us’d in our Antient Statutes . . . and the Etymology and Interpretation of
the Proper Names of Men, Women, and Remarkable Places in Great Brit-
ain; Also the Dialects of our Different Counties.
To which is Added a Collection of our most Common Proverbs, with
their Explication and Illustration.
38 The beginnings
The whole work compil’d and Methodically digested, as well as for the
Entertainment of the Curious, as the Information of the Ignorant, and for
the Benefit of young Students, Artificers, Tradesmen and Foreigners.
Bailey’s 40,000 words were culled from a wide variety of sources and encom-
pass both the everyday and the less usual. For the first time proverbs were
included (some ninety in all), and serious attention was paid to etymology, e.g.
Emerald, (Esmeraude, F. Esmeralda, Span. Smaragdus, L. of ∑µαραγδος, Gr.) a

precious Stone.
Bailey’s dictionary (see Simpson 1989) proved enormously popular, dominat-

ing the eighteenth century, and reaching its thirtieth edition in 1802, by when
the word list had expanded to around 50,000 items. It was aimed at the ‘curious’
as well as the ‘ignorant’, at the ‘young student’ as well as the ‘foreigner’; and its
known users included the Prime Minister of the day, William Pitt the Elder.
In 1730, however, Nathaniel Bailey published a further dictionary, with the
title Dictionarium Britannicum. It is important not least because it was used by
Samuel Johnson as the basis for his dictionary (see 4.5 below). The title page
included the statements:
Dictionarium Britannicum: Or a more Compleat Universal Etymological

English Dictionary than any Extant.
Containing Not only the Words, and their Explications; but their
Etymologies . . .
Also Explaining hard and technical Words, or Terms of Art, in all the
Arts, Sciences, and Mysteries following. Together with Accents directing
to their proper Pronunciation, showing both the Orthography and Orthoepia
of the English Tongue . . .
Collected by several Hands, The Mathematical Part by G. Gordon, the

Botanical by P. Miller. The Whole Revis’d and Improv’d, with many thou-
sand Additions, by N. Bailey.
The 48,000 items did not include the proverbs, though they were reinstated in
greater numbers in the second edition of 1736. The proper names, apart from
those from myths and legends, are removed to an appendix. Noteworthy here
is the inclusion of ‘Orthoepia’, i.e. pronunciation, as least as far as the stressing
(Accent) of polysyllabic words is concerned. Note, too, that Bailey acknowl-
edges the assistance of specialists in mathematics and botany. The second
edition, which increases the entries to 60,000, to better fulfil the title claim
of ‘universal’, also notes the contribution of ‘T. Lediard, Gent. Professor of
the Modern Languages in Lower Germany’ to the etymologies, to pay attention
to that part of the title. Lediard’s hand is evident, as the following examples
show:
The beginnings 39
Littoral (litoralis, of litus, L. the sea shore)
Little (litel, lytel or lytle, Sax. litet and liten, Su. lidet or lille, Dan. luttel, Du. lut,
L.G.)
(Note: the language label follows the form: L = Latin, Sax = Saxon, Su = Swed-
ish, Dan = Danish, Du = Dutch, L.G. = Low German.)
A number of ‘compleat’ dictionaries followed Bailey’s two during the first
half of the eighteenth century, but they were largely derivative of his, or indeed
reverted to earlier types.
4.4 Ascertaining and fixing

To appreciate the next developments in lexicography, we need to understand
some of the debates about language that exercised scholars and authors in the
eighteenth century. There was deep concern about the state of the English
language. The wholesale importation of words, from Latin especially, as a result
of the Renaissance, and arising from the translation of classical authors into
English, had provoked the ‘inkhorn’ controversy. Some authors, and translators
in particular, were accused of borrowing Latin words, even though an equiva-
lent already existed in English, merely in order to sound more erudite. People
looked back to a ‘golden age’ of the English language – the Elizabethan period,
when Shakespeare was writing – and saw only a degeneration in English since
that time. The question was how to arrest the decay, how to ‘fix’ the language
so that it would not degenerate any further, and how to effect improvements
that would help restore the language to a former glory.
Some looked across the Channel admiringly and enviously, as it seemed that
the French had cracked the problem. In 1635, Cardinal Richelieu, Louis XIII’s
chief minister, had founded the Académie française, whose chief purpose was to
codify the French language and to pronounce on what was and was not accept-
able French – as it has continued to do to this day. One of the instruments for
achieving its aim was a dictionary. Begun in 1639, the Dictionnaire de l’Académie
française was finally published fifty-five years later, in 1694, followed by a second
edition in 1718. Voices in Britain were raised in favour of an English Academy
to achieve the same purpose. Among them was that of the author Daniel Defoe
(1660–1731), who, in his Essay on Projects of 1698, suggested that such an insti-
tution should have as its purpose: ‘to encourage polite learning, to polish and
refine the English tongue, and advance the so much neglected faculty of correct
language, to establish purity and propriety of style’. Others opposed the estab-
lishment of an academy, among them Samuel Johnson, who did not regard this
as the British way of doing things.
An alternative suggestion was that a group of suitable persons should be
assembled to undertake the task of looking at the language. The Royal Society
had formed a committee in 1664 ‘for improving the English tongue’, but noth-
ing seems to have come of it. The suggestion was taken up by Jonathan Swift
40 The beginnings
(1667–1745), the author of Gulliver’s Travels, who wrote a pamphlet in 1712,
addressed to the Earl of Oxford, the Lord Treasurer of England, with the title A
Proposal for Correcting, Improving and Ascertaining the English Language. In it, Swift
proposes that a group should be assembled ‘of such Persons, as are generally
allowed to be best qualified for such a Work’; they would ‘have the Example of
the French before them’, though they are to ‘avoid their mistakes’. ‘They will’,
says Swift, ‘find many Words that deserve to be utterly thrown out of our
Language; many more to be corrected, and perhaps not a few, long since anti-
quated, which ought to be restored, on Account of their Energy and Sound.’
He continues:
But what I have most at heart, is that some Method should be thought on
for Ascertaining and Fixing our Language for ever, after such alterations are
made in it as shall be thought requisite. For I am of Opinion, that it is better
a Language should not be wholly perfect, than that it should be perpetually
changing; and we must give over at one Time or other, or at length infal-
libly change for the worse.
Change is the great enemy of language. The language must be ‘ascertained’

(with the eighteenth-century meaning of ‘fix, determine, limit’ – SOED) and
‘fixed’.
Swift had both grammar and vocabulary in view in his Proposal, but as far as
vocabulary was concerned, a crucial question was how to determine, in ‘fixing’
the language, which words should be regarded as legitimate and ‘proper’ words
of English. The solution to this problem was seen in the appeal to ‘authorities’,
those regarded as representing English at its best. Joseph Addison (1672–1719),
one of the founders of The Spectator, had begun collecting quotations for the
purpose of contributing to the compilation of a dictionary, but a summons by
the king to public service brought this to a halt. In 1717, the following adver-
tisement appeared in the newspapers:
Just printed, Proposals for the Publication of a compleat and standard Dic-
tionary of the whole English Language, as it is written in all its various
Idioms and Proprieties, by the most authentick Orators and Poets, from
Chaucer to Shaftesbury, whose Authorities shall be quoted throughout:
According to the Method of the celebrated one of the French Academy. In
4 Vols. Fol.
There is, however, no evidence that the dictionary announced in this proposal
ever saw publication. Like Addison, the poet Alexander Pope (1688–1744) was
interested in contributing to a dictionary; he drew up lists of the prose authors
from whose works he thought citations for a dictionary should be collected, and
on the question of ‘authority’ he comments:
The beginnings 41
In most doubts, whether a word is English or not, or whether such a par-
ticular use of it is proper, one has nothing but authority for it. Is it in Sir
William Temple, or Locke, or Tillotson? If it be, you may conclude that it
is right, or at least won’t be looked upon as wrong.
Such were the concerns of the eighteenth century about the state of English.
The dictionary would be an instrument in achieving the aims of ‘ascertaining’
and ‘fixing’ the language. Of these concerns, and of the expected role of the
dictionary, Samuel Johnson was aware as he planned his monumental work.
Before Johnson’s dictionary appeared, another work was published that pur-
ported to address many of these concerns: Benjamin Martin’s Lingua Britannica
Reformata of 1749. Its title page promises much:
Lingua Britannica Reformata: Or, A NEW ENGLISH DICTIONARY, Under

the Following TITLES, VIZ.
I. UNIVERSAL; Containing a Definition and Explication of all the Words
now used in the English Tongue, in every Art, Science, Faculty or Trade.
II. ETYMOLOGICAL; Exhibiting and Explaining the true Etymon or
Original of Words from their respective Mother-Tongues, the Latin, Greek,
Hebrew, and Saxon; and their Idioms, the French, Italian, Spanish, German,
Dutch, &c.
III. ORTHOGRAPHICAL; Teaching the True and Rational Method of
Writing Words, according to the Usage of the most Approved Modern
Authors.
IV. ORTHOEPICAL; Directing the True Pronunciation of Words by
Single and Double Accents; and by Indicating the Number of Syllables in
Words where they are doubtful, by a Numerical Figure.
V. DIACRITICAL; Enumerating the Various Significations of Words in a
Proper Order, viz. Etymological, Common, Figurative, Poetical, Humorous,
Technical, &c. in a Manner not before attempted.
VI. PHILOLOGICAL; Explaining all the Words and Terms, according to
the Modern Improvements in the Various Philological Sciences, viz. Gram-
mar, Rhetoric, Logic, Metaphysics, Mythology, Theology, Ethics, &c.
VII. MATHEMATICAL; Not only Explaining all Words in Arithmetic,
Algebra, Logarithms, Fluxions, Geometry, Conics, Dialling, Navigation, &c.
according to the Modern Newtonian Mathesis; but the Terms of Art are
illustrated by Proper Examples, and Copper-Plate Figures.
VIII. PHILOSOPHICAL; Explaining all Words and Terms in Astronomy,
Geography, Optics, Hydrostatics, Acoustics, Mechanics, Perspective, &c. accord-
ing to the latest Discoveries and Improvements in this Part of Literature.
42 The beginnings
To which is prefix’d, An INTRODUCTION, containing A Physico-
Grammatical Essay On the Propriety and Rationale of the ENGLISH
TONGUE.
Martin was more famous as a scientist and advocate of Newtonian mathematics

than as a lexicographer, but the plan of his dictionary is impressive in its organ-
isation and attention to detail, and in the concern to reflect up-to-date scholar-
ship. But, as Green (1996: 218) observes, Martin’s practice does not quite live
up to the promise. His ‘Physico-Grammatical Essay’, which runs to 108 pages,
is its major innovation. In that essay, though, he does make some perceptive
remarks about ‘fixing’ the language:
The pretence of fixing a standard to the purity and perfection of any

language . . . is utterly vain and impertinent, because no language as de-
pending on arbitrary use and custom, can ever be permanently the same,
but will always be in a mutable and fluctuating state; and what is deem’d
polite and elegant in one age, may be accounted uncouth and barbarous in
another.
Johnson, who proved to be the better lexicographer, would come to similar

conclusions.
4.5 Samuel Johnson (1709–84)

When Johnson’s Dictionary of the English Language appeared in 1755, nine years
had elapsed since he had signed the contract with the group of booksellers who
had approached him to compile it. The agreement had been for a delivery date
of three years, and for a sum of 1500 guineas (£1575), which was in part meant
to defray the expenses of the six assistants that he was intended to employ. First,
Johnson prepared his Plan of a Dictionary of the English Language, which was
published in 1747. Johnson was prevailed upon to address his Plan to the Earl of
Chesterfield, who had a position in government and who was known to be
interested in language matters, in the hope of obtaining his patronage, if not his
financial support. In the event, neither was forthcoming, apart from an initial
grant of £10, and Johnson later justified his address to the Earl on the grounds
that it enabled him to buy more time with his booksellers.
The Plan is a fascinating document (reproduced in Wilson 1957), which
shows considerable thought and reflection on his task by the beginning lexico-
grapher. Johnson addresses himself to the methodological issues in the light of
the contemporary concerns about the state of the language. He begins with
the problem of ‘selection’, which words to include. ‘The chief intent’, he says,
of the dictionary, ‘is to preserve the purity and ascertain the meaning of our
English idiom.’ That would seem to suggest that ‘the terms of particular
professions . . . with the arts to which they relate’ should be excluded, because
‘they are generally derived from other nations’. But account must be taken of
The beginnings 43
what the users of the dictionary will expect and require: foreign words cannot,
then, be excluded, because people most often consult dictionaries to find ‘terms
of art’:
It seems necessary to the completion of a dictionary design’d not merely for

critics but for popular use, that it should comprise in some degree the
peculiar words of every profession.
He proposes to make a distinction between loanwords that are still considered

to be foreign and those that have been ‘incorporated into the language’, by
printing the former (which he calls ‘aliens’) in italics. He then discusses whether
the common words should be included, and he argues for their inclusion on
two grounds: it is not possible to ‘fix the limits of the reader’s learning’; and such
words will need ‘their accents . . . settled, their sounds ascertained, and their
etymologies deduced’. In any case, he would prefer his readers to find more
than they required rather than less.
After ‘selection’, Johnson turns his attention in the Plan to ‘orthography’,
about which ‘there is still great uncertainty among the best critics’. He proposes
to make no changes from present practice, where this is clear, and to have good
reason for introducing innovation. In respect of ‘pronunciation’ – ‘the stability
of which is of great importance to the duration of a language, because the first
change will naturally begin by corruptions in the living speech’ – he proposes to
‘determine the accentuation of all polysyllables by proper authorities’, and to
‘fix the pronunciation of monosyllables, by placing them with words of corre-
spondent sound’.
In respect of ‘etymology’, Johnson proposes a distinction between ‘simple’
and ‘compound’ words, and for ‘simple’ words between ‘primitive’ and ‘deriva-
tive’. Primitive words will be traced to their language of origin, and no word
will be admitted unless its original can be determined. Thus ‘we shall secure our
language from being over-run with cant, from being crouded with low terms,
the spawn of folly or affectation’. After ‘etymology’ comes ‘analogy’, by which
Johnson means the inflections of words: ‘our inflections . . . admit of number-
less irregularities, which in this dictionary will be diligently noted’. As for ‘syn-
tax’, ‘the syntax of this language is too inconstant to be reduced to rules, and can
be only learned by the distinct consideration of particular words as they are used
in the best authors’; for example, we say ‘die of ’ but ‘perish with’. Similarly, he
will note ‘phraseology’, where a word is used in combination ‘in a manner
peculiar to our language’; he cites the example of make in expressions such as
‘make love’, ‘make a bed’, ‘make merry’.
‘The great labour is yet to come’: ‘interpretation’ (i.e. definition) – ‘the la-
bour of interpreting these words and phrases with brevity, fulness and perspicu-
ity’, made the more difficult ‘by the necessity of explaining the words in the
same language’. He proposes to distinguish the separate senses of polysemous
words, and to present the senses in a rational order: the ‘natural and primitive
signification’ first, then the ‘consequential meaning’, then metaphorical sense,
44 The beginnings
followed by the ‘poetical’, the ‘familiar’, the ‘burlesque’ senses. Finally, Johnson
will deal with the ‘distribution’ of words ‘into their proper classes’, by which he
means to distinguish words in general use from those used mainly in poetry,
those that are obsolete, those used only by particular writers, those used only in
burlesque writing, and ‘words impure and barbarous’. Obsolete words will be
included only if they occur in authors living since the accession of Elizabeth
(1558), ‘from which we date the golden age of our language’.
All Johnson’s observations will be supported by ‘citations’:
In citing authorities, on which the credit of every part of this work must
depend, it will be proper to observe some obvious rules, such as of prefer-
ring writers of the first reputation to those of an inferior rank, of noting the
quotation with accuracy, and of selecting, when it can be conveniently
done, such sentences, as, besides their immediate use, may give pleasure or
instruction by conveying some elegance of language, or some precept of
prudence, or piety.
Additionally, Johnson intends to give the name of the author who first intro-
duced a word or phrase, or who last used an obsolete word or phrase:
By this method every word will have its history, and the reader will be
informed of the gradual changes of the language, and have before his eyes
the rise of some words, and the fall of others. But observations so minute
and accurate are to be desired rather than expected, and if use be carefully
supplied, curiosity must sometimes bear its disappointments.
Indeed, Johnson anticipates here the work that would lead to the Oxford English
Dictionary over a century later. In any case, Johnson’s restriction of his citations
to a particular range of ‘writers of the first reputation’ precludes his attainment
of this goal.
Johnson concludes his Plan with an address to the Earl of Chesterfield:
This, my Lord, is my idea of an English Dictionary, a dictionary by which

the pronunciation of our language may be fixed, and its attainment facili-
tated; by which its purity may be preserved, its use ascertained, and its
duration lengthened. And though perhaps to correct the language of na-
tions by books of grammar, and amend their manners by discourses of
morality, may be tasks equally difficult; yet it is unavoidable to wish, it is
natural likewise to hope, that your Lordship’s patronage may not be wholly
lost; that it may contribute to the preservation of the antient, and the im-
provement of modern writers.
Whether Johnson believed his own rhetoric at this point, or whether he was
pandering to what he assumed were the Earl’s opinions, is uncertain. By the
time he had finished his dictionary, he was far less sanguine about its effects on
The beginnings 45
‘fixing’ the language, as he tells us in the Preface to the Dictionary (reproduced in
Wilson 1957):
Those who have been persuaded to think well of my design, will require
that it should fix our language, and put a stop to those alterations which
time and chance have hitherto been suffered to make in it without opposi-
tion. With this consequence I will confess that I flattered myself for a while;
but now begin to fear that I have indulged expectation that neither reason
nor experience can justify. When we see men grow old and die at a certain
time one after another, from century to century, we laugh at the elixir that
promises to prolong life to a thousand years; and with equal justice may the
lexicographer be derided, who being able to produce no example of a
nation that has preserved their words and phrases from mutability; shall
imagine that his dictionary can embalm his language, and secure it from
corruption and decay, that it is in his power to change sublunary nature,
and clear the world at once from folly, vanity, and affectation.
What can then be done?
If the changes that we fear be thus irresistible, what remains but to acqui-
esce with silence, as in the other insurmountable distresses of humanity? It
remains that we retard what we cannot repel, that we palliate what we
cannot cure. Life may be lengthened by care, though death cannot ultim-
ately be defeated: tongues like governments have a natural tendency to
degeneration; we have long preserved our constitution, let us make some
struggles for our language. In the hope of giving longevity to that which its
own nature forbids to be immortal, I have devoted this book, the labour of
years, to the honour of my country.
Let us look at that ‘labour of years’, which Johnson mentions. They extended
from June 1746, when he signed the contract with the booksellers, to April
1755, when the Dictionary was finally offered for sale. Johnson suffered many
vicissitudes, not least in his personal life, including the London earthquake of
1750, the long illness and eventual death of his wife Tetty in 1752, as well as his
financial problems, which required him to take on other work, among which
was starting The Rambler. At no time did he have the full complement of six
assistants working for him. And it seems that after a year or so of working on the
dictionary, Johnson decided that he needed to revise his method and start over
again (see Reddick 1990). His method consisted essentially in the following
procedure.
Using an interleaved version of Bailey’s Dictionarium Britannicum as his basis,
Johnson undertook an extensive supplementary reading programme. Having
selected a writer whom he admired, he would select a volume of his works, and
when he found some word that he thought correctly used, he underlined it in
pencil and marked the surrounding passage that he wanted copied. After he had
46 The beginnings
gone through the book and marked everything that he thought could be used,
he passed the book to one of his assistants, who copied out the marked passages
onto blank sheets of paper or in blank books. They were then cut, or ‘clipped’,
into separate slips and deposited in a bin. Subsequently, the large number of slips
were sorted alphabetically by headword; Johnson decided which were repetitious
or superfluous and so to be rejected, and which ones he would use. Sometimes
he would revise the quotations to fit his purpose. In total, 114,000 quotations
were used in the Dictionary. They were ready for Johnson to use by the summer
of 1749, and he had then to decide how many senses to recognise for each
word, write the definitions, and provide their etymologies. Pages were sent off
to the printer bit by bit, and then the printed sheets were returned to Johnson
for revision.
By April 1753, the first volume (A–K) was in final form, and a year later most
of the Dictionary was printed. It awaited Johnson’s prefatory essay on the history
of the language. It is also possible that Johnson was playing for time, awaiting
the award of an honorary MA degree from Oxford University, which was
eventually granted in February 1755. So the title page cites the author as ‘Samuel
Johnson, A.M.’. When it appeared in April 1755, the two-volume folio Diction-
ary sold for £4.10s (i.e. £4.50) and sold around 2,000 copies. Further editions
followed, on which Johnson worked, but it is the fourth edition of 1773 that
represents a major revision, incorporating all the corrections to the mistakes and
inconsistencies that Johnson was aware of from the first edition. Johnson’s
remained the foremost dictionary of the English language for a century, and
its author was acclaimed as the one who had done for English single-handedly
what it had taken forty French academicians to do for their language.
Johnson not only produced a monumental dictionary by a method, involving
the collection of evidence (citations) and using the evidence to construct the
entries, which became standard lexicographical procedure, but he also reflected,
in the Plan and the Preface in particular, on the nature of the dictionary compil-
er’s task and the issues that face lexicographers. Most of them represent ques-
tions that lexicographers have to struggle with still.
4.6 Further reading

The history of lexicography is entertainingly told in Jonathon Green’s Chasing
the Sun (1996).
Dictionaries before Cawdrey’s are described in Gabriele Stein’s The English
Dictionary before Cawdrey (1985). Those from Cawdrey up to (but not including)
Johnson are treated in De Witt T. Starnes and Gertrude E. Noyes’ The English
Dictionary from Cawdrey to Johnson 1604 –1755 (1946), republished in a version
edited by Gabriele Stein (1991).
Johnson’s story is told in Andrew Reddick’s The Making of Johnson’s Diction-
ary 1746–1773 (1990).
The New English Dictionary 47
5 The New English Dictionary
The popularity of Samuel Johnson’s Dictionary continued into the nineteenth

century, when it was joined by a rival, Charles Richardson’s A New Dictionary
of the English Language, published in 1836/7. Richardson’s dictionary is signifi-
cant for its extensive use of illustrative quotations, and its adherence to the
rather bizarre ‘philosophical philology’ proposed by Horne Tooke (1736–1812).
In The Diversions of Purley (1786, 1805), Tooke had put forward the notion that
all words could be traced to primary nouns, or verbs, with a single meaning. His
classic example was the word bar, which has the basic meaning of ‘defence’; and
he finds this meaning included in words such as barn, baron, barge, bargain, bark.
Tooke’s reductionist ideas enjoyed a measure of popularity, and they influenced
Richardson’s etymologies, which, needless to say, have much of the fanciful
about them. By mid-century both leading dictionaries were considered to be
wanting in their coverage of English vocabulary; the observation came from the
newly formed Philological Society.
5.1 The Philological Society

The Society had been formed in May 1842, ‘for the investigation of the Struc-
ture, the Affinities, and the History of Languages; and the Philological Illustra-
tion of the Classical Writers of Greece and Rome’. The concern about the lack
of coverage by existing dictionaries was expressed in 1857 and related to the
vocabulary of the earlier history of English. The Society formed a committee to
collect ‘unregistered words’, and instituted the recruitment of volunteer readers
to undertake a reading programme for the purpose. The aim was to publish a
supplement to existing dictionaries. However, in November of that year, the
Society heard two papers by one of the members of its Unregistered Words
Committee (the other two members were Herbert Coleridge and Frederick
Furnivall), the Dean of Westminster, Richard Chenevix Trench, which were
later published under the title On Some Deficiencies in Our English Dictionaries.
Dean Trench identified some seven deficiencies and proposed that ‘the only
sound basis’ for a dictionary was ‘the historical principle’, by which he meant
the new, ‘scientific’ comparative philology that had been developed especially
in Germany, where Jakob and Wilhelm Grimm were already working on an
historical dictionary of German, the Deutsches Wörterbuch.
5.2 The first edition

Trench’s papers struck a chord with the Philological Society, and in January
1858 the Society resolved that ‘a New Dictionary of the English Language be
prepared under the Authority of the Philological Society’. In the following
year, the Society published a ‘Proposal for the Publication of a New English
Dictionary by the Philological Society’, calling for a comprehensive treatment
of English vocabulary from the end of the thirteenth century, stating:
The first requirement of every lexicon is that it should contain every word
occurring in the literature of the language it professes to illustrate. In the
treatment of individual words the historical principle will be uniformly
adopted.
Herbert Coleridge was appointed as the first editor, and steps were taken to
recruit voluntary readers to peruse the works of English literature for suitable
citations. Coleridge drew up a set of working rules for editing the dictionary,
which were published in 1860 as the Canones Lexicographici. He died the follow-
ing year at the age of 31, from consumption, made worse by a chill resulting
from sitting in damp clothes at a long Philological Society meeting.
Coleridge was succeeded as editor by Frederick Furnivall, the third member
of the Unregistered Words Committee. Furnivall was a colourful and energetic
character, with great organisational abilities, but he had fingers in many pies. He
appointed sub-editors for the dictionary, whose job was to collect the slips from
readers and organise them before sending them on to the editor. Furnivall also
began to recognise how enormous the task was that the Philological Society had
set itself, and he issued a fresh call in 1862 for voluntary readers. He proposed
that a concise dictionary be prepared as a precursor to the main dictionary, but
it became apparent that it would require almost as much work, so it came to
nothing. Furnivall recognised that some of the earlier texts of English literature
were not available in published form; so he founded the Early English Text
Society in 1864, which concentrated on making available in printed form texts
from before 1558 that had not been published previously. They provided an
invaluable and necessary resource for the dictionary. In due course, Furnivall’s
other interests began to take over, and while he attempted to encourage his sub-
editors, their efforts and those of the readers slackened, so that in his annual
report to the Society of 1872, he notes that ‘the progress in the Dictionary has
been so slight that no fresh report in detail is needed’. Furnivall was looking for
a new editor to take over the project. After two abortive approaches, he con-
tacted a teacher at Mill Hill School in North London, who was a member of the
Council of the Philological Society: James Murray.
The dictionary was in want not only of an editor, but also of a publisher.
Furnivall made it his aim to secure both. Negotiations with Macmillan, who
were known to be interested in publishing a dictionary, failed; but Furnivall
continued to cultivate Murray for the editorship, while having approaches made
to Cambridge and Oxford university presses. In due course, in 1878 Murray
was appointed editor and on 1 March 1879 the Philological Society and Oxford
University Press signed a contract to publish the dictionary. The contract agreed
a four-volume, 7000-page dictionary to be completed in ten years. The first
task was to retrieve all the materials that were still in the hands of the sub-editors
and readers. This was done, and Murray built a corrugated iron structure in the
garden of his Mill Hill home, which he called the ‘Scriptorium’, to house the
material.
While still a full-time teacher, Murray set about his task as editor. Sorting the
material that Furnivall had passed on to him, amounting to some 2 million slips,
Murray realised that not only were there still many gaps in the evidence needed
to compile the dictionary, but much of what he had received was not adequate.
So he initiated a fresh reading programme, calling for 1000 readers for a period
of three years: around 800 volunteered in the UK and between 400 and 500 in
North America. One of the most prolific and devoted was an American sur-
geon, Dr W.C. Minor, who suffered from paranoid schizophrenia and was
incarcerated in the Broadmoor Asylum for the Criminally Insane, after he had
killed a man in London (his fascinating story is told in Winchester 1999). Murray
issued the readers with clear instructions on what to look for (quotations that
might indicate the birth of a word, or those that might define a word or were
unusual in some way), to be sent in on 6 inch by 4 inch slips of paper, with the
word in the top left-hand corner and full bibliographical details. He also offered
to refund the postage that a reader incurred in sending the slips in to him. When
he received the slips, Murray, with the help of his assistants, including his chil-
dren (he had eleven in all), sorted them and put them into their appropriate
pigeon holes in the Scriptorium. The three-year programme yielded around a
million slips, and some 5 million were eventually collected in all for the first
edition of the dictionary.
As well as collecting the essential data for the dictionary, Murray was also
working on the principles that would underlie the editing of the dictionary, the
appearance of the entries, and so on. In February 1884, the first part of the
dictionary appeared, covering the letters A to Ant. It was clear that if faster
progress was to be made on the dictionary – the ten-year goal was already
receding – then Murray would have to devote more of his time to the project,
and he would need help. In 1885, Murray became full-time editor and moved
to Oxford to be near the point of publication; a new Scriptorium was built in
the grounds of his house at 78 Banbury Road. A second editor was appointed,
Henry Bradley, as well as a number of assistants.
In 1888, the first volume of the dictionary was published, covering the letters
A and B. Murray explains in the Preface why it was taking so long to compile
the dictionary:
The preparation of this volume has taken a much longer time than any of
the promoters of the work anticipated. The time has been consumed chiefly
in two directions: first, with the larger articles, as those on AT, BY, BUT,
BE, BEAR, BREAK, the construction of which has occupied many days,
sometimes even weeks. The mere study of the result, arranged in some
degree of order, gives little idea of the toil and difficulties encountered in
bringing into this condition what was at first a shapeless mass of many
thousand quotations. And in this part of the work there was practically no
assistance to be got from the labours of our predecessors; the attempt has
never been made before to exhibit such a combined logical and historical
view of the sense-development of English words. Our own attempts lay no
claim to perfection; but they represent the most that could be done in the
time and with the data at our command. The other direction in which
much time has been consumed is the elucidation of the meaning of obscure
terms, sometimes obsolete, sometimes current, belonging to matters of
history, customs, fashions, trade, or manufactures . . . The difficulty of
obtaining firsthand and authoritative information about these has often been
immense, and sometimes insurmountable.
The Preface acknowledges the help of numerous people, including the sub-
editors, who had given of their time voluntarily, as well as many scholars whom
Murray had occasion to consult. He also lists all the readers who had contrib-
uted more than a thousand quotations. And he concludes the Preface with this
comment:
After all the help which has been received, and which has contributed so
much to the completeness of the Dictionary, the element of time still re-
mains inexorable; it is still, as in the days of Dr. Johnson, imperative that
limits be set to research, in order that the work may ‘in time be ended,
though not completed’. Accordingly, since the close of Volume I, it has
been the aim of the Editor and his staff to maintain such a regular rate of
progress as will ensure the production of one Part a year. As Mr. Henry
Bradley (whose co-operation in the present volume is mentioned above) is
now at work independently, with a staff of assistants, on a third volume,
there is reasonable ground to expect that the production of the work hence-
forth will be twice as rapid as it has been hitherto.
Murray was perhaps being optimistic: it would be another 40 years before the
Dictionary was finally completed. A third editor, William Alexander Craigie,
was appointed in 1901, though he had been working in Bradley’s team since
1897; and a fourth editor, Charles Talbut Onions, was appointed in 1914, hav-
ing worked for the Dictionary since 1895.
The outbreak of World War I in 1914, the year in which Volume VIII was
published, hindered progress, as members of the Dictionary staff volunteered
for service in the armed forces. In 1915, James Murray, who had been knighted
in 1908, died at the age of 78. The previous year he had been treated for cancer
Table 5.1 A New English Dictionary on Historical Principles founded mainly on the mater-
ials collected by The Philological Society
Volume Letters Date of publication Editors
I A&B 1888 Murray

II C 1893 Murray
III D&E 1897 Murray, Bradley
IV F&G 1901 Murray, Bradley
V H–K 1901 Murray
VI L–N 1908 Murray, Bradley, Craigie
VII O&P 1909 Murray
VIII Q – Sh 1914 Murray, Bradley, Craigie
IX, Part I Si – St 1919 Murray, Bradley, Craigie, Onions
IX, Part II Su – Th 1919 Murray, Bradley, Craigie, Onions
X, Part I Ti – U 1926 Murray, Bradley, Craigie, Onions
X, Part II V–Z 1928 Murray, Bradley, Craigie, Onions
and had recovered enough to work on the letter T, but he contracted pleurisy
soon afterwards and on 26 July went to his Maker, whom he had served for so
long in working for his beloved Dictionary. Henry Bradley did not see the
completion of the Dictionary either; he died in 1923, five years before the final
volume reached publication.
The completed New English Dictionary contained 15,487 pages, more than
twice the 7000 originally envisaged. James Murray edited nearly half of the
entire work, some 7207 pages. Out of the 5 million quotations amassed, 1,861,200
were used in the Dictionary. It contained 252,200 entries, with 414,800 defini-
tions. It had taken 44 years since the publication of A-Ant in 1884, and 70 years
since the Philological Society had first proposed a ‘New Dictionary of the Eng-
lish Language’.
In 1933 the Dictionary was republished under the title Oxford English Diction-
ary, in twelve volumes, together with an 866-page supplement, edited by Craigie
and Onions, the two surviving editors. The supplement included words and
senses that had been added to the language in the previous fifty years. Readers
had continued to send in quotation slips even for words in volumes of the
Dictionary already published. The supplement encompasses among other things:
words of modern origin that had been omitted, either intentionally or acciden-
tally, from the original dictionary; a large number of words from the arts and
sciences; an expansion of the colloquial and slang vocabulary; a more generous
inclusion of proper names; and American usages (Craigie was editing an histor-
ical dictionary of American English, which was published in 1936–44).
5.3 What the OED contains

Before we take the history of the OED any further, let us examine the content
of what James Murray and his fellow-editors put together. In the Preface to
Volume I, Murray expresses the aims of the Dictionary as follows:
The aim of this Dictionary is to furnish an adequate account of the mean-
ing, origin, and history of English words now in general use, or known to
have been in use at any time during the last seven hundred years. It endeav-
ours (1) to show, with regard to each individual word, when, how, in what
shape, and with what signification, it became English; what development
of form and meaning it has since received; which of its uses have, in the
course of time, become obsolete, and which still survive; what new uses
have since arisen, by what processes, and when: (2) to illustrate these facts
by a series of quotations ranging from the first known occurrence of the
word to the latest, or down to the present day; the word being thus made
to exhibit its own history and meaning: and (3) to treat the etymology of
each word strictly on the basis of historical fact, and in accordance with the
methods and results of modern philological science.
The OED aims to do what Johnson had aspired to over a century earlier: present
a history of the entire vocabulary, beginning with words that were in current
use in 1150, all supported by a series of quotations that would illustrate the
changes and developments in meaning, as well as the birth and death of words
and senses of words. Even so, the OED is not totally inclusive in its coverage.
While it took a descriptivist (record what is there) stance, there were Victorian
sensibilities to be observed, so some of what today would be labelled ‘coarse
slang’ vocabulary is missing, as well as some scientific and technical vocabulary,
some of which was only being coined as the dictionary was being compiled.
After all, when Murray went to see Dr Minor in Broadmoor for the first time in
November 1896, he travelled by train from Oxford to Crowthorne and then by
horse-drawn carriage from the station to the Asylum. The coverage may also be
affected by the reliance on volunteer readers and their varying ability to spot
what was needed in the material that they read. Murray’s instructions may not
have helped either, as he himself later acknowledged (Mugglestone 2000a:8):
Make a quotation for every word that strikes you as rare, obsolete, old-
fashioned, new, peculiar, or used in a peculiar way.
That said, the evidence from readers was frequently supplemented by the edi-
tors’ own researches, as well as by those of other scholars who were approached
for information, especially on rare or technical words.
Murray aimed to include all the ‘common words’ of the language, but, as he
makes clear in the ‘General Explanations’ that preface Volume I of the diction-
ary, this is a fluid category:
the English Vocabulary contains a nucleus or central mass of many thou-

sand words whose ‘Anglicity’ is unquestioned; some of them only literary,
some of them only colloquial, the great majority at once literary and collo-
quial, – they are the Common Words of the language. But they are linked on
every side with other words which are less and less entitled to this appellation,
and which pertain ever more and more distinctly to the domain of local
dialect, of the slang and cant of ‘sets’ and classes, of the peculiar technical-
ities of trades and processes, of the scientific terminology common to all
civilised nations, of the actual languages of other lands and peoples. And
there is absolutely no defining line in any direction: the circle of the English
language has a well-defined centre but no discernible circumference. Yet
practical utility has some bounds, and a Dictionary has definite limits: the
lexicographer must, like the naturalist, ‘draw the line somewhere’, in each
diverging direction. He must include all the ‘Common Words’ of literature
and conversation, and such of the scientific, technical, slang, dialectal, and
foreign words as are passing into common use, and approach the position or
standing of ‘common words’, well knowing that the line which he draws
will not satisfy all his critics. For to every man the domain of ‘common
words’ widens out in the direction of his own reading, research, business,
provincial or foreign residence, and contracts in the direction with which
he has no practical connexion: no one man’s English is all English. The
lexicographer must be satisfied to exhibit the greater part of the vocabulary
of each one, which will be immensely more than the whole vocabulary of
any one.
Murray acknowledges that the Dictionary extends its reach ‘farther in the do-
main of science and philosophy . . . than in that of slang or cant’, on the basis
that words in the former are likely to pass into literature, while those in the latter
remain in the spoken language only.
The OED divides words into three classes: ‘Main’ words, ‘Subordinate’ words,
and ‘Combinations’. Main words include all single words, either simple or
derived (1.6), as well as compounds which ‘from their meaning, history, or
importance, claim to be treated in separate articles’. Subordinate words are
‘variant and obsolete forms of Main Words, and such words of bad formation,
doubtful existence, or alleged use, as it is deemed proper, on any ground, to
record’; e.g.
Afforse, obs. variant of AFFORCE

Afforst, obs. variant of ATHIRST
Affrait, -ly, see AFFRAYITLY, AFRAID.
Main and Subordinate words are headwords in the alphabetical series of the
dictionary, with Subordinate words printed in smaller type than Main words.
Combinations are those (derivatives and compounds) that either require no
definition or which can be ‘briefly explained in connexion with their cognates’;
they are dealt with under the Main word that constitutes their first element. If
a word belongs to more than one distinct word class, it is entered separately for
each class (e.g. brass as noun and as verb).
The entry for a Main word consists of four sections: the Identification, the
Morphology, the Signification, and the Illustrative Quotations. The Identification
section begins with the spelling: the headword gives the usual current form, in
bold type and with initial capital letter. Any alternative current spellings follow,
e.g. Jowl, jole. Words that are thought to be obsolete are preceded by a dagger
symbol (†), e.g. †Kask (meaning ‘active’, ‘vigorous’). Some words are preceded
by a ‘||’ symbol, which relates to their ‘citizenship in the language’. In these
terms, words are classified as: ‘naturals’, ‘denizens’, ‘aliens’, or ‘casuals’. The
class of ‘naturals’ includes all ‘native’ words (with their origin in Anglo-Saxon)
and all ‘fully naturalised’ loanwords, which no longer betray their origin in
spelling, pronunciation or inflection (e.g. gas, street). The class of ‘denizens’
includes loanwords that are naturalised as far as their use in the language is
concerned, but which betray their origin in their spelling, pronunciation or
inflection (e.g. crèche, locus). The class of ‘aliens’ comprises loanwords for ob-
jects, titles, etc. which are in common use but for which there is no English
equivalent (e.g. intifada, Knesset). ‘Casuals’ are like aliens, except that they are
not in common use and may be found, for example, in travel books. Denizens,
aliens, and casuals tending towards aliens are marked with ‘||’, e.g. ||Hemiplegia,
||Kursaal.
Following the headword and any alternative spellings comes the pronuncia-
tion, in rounded brackets. ‘The pronunciation is the actual living form or forms
of a word, that is, the word itself, of which the current spelling is only a symbol-
ization’, writes Murray in the General Explanations. The OED predates the
International Phonetic Alphabet, and Murray took great pains, as well as schol-
arly advice, to develop an accurate transcription system to represent the sound
of words. Pronunciation is followed in the entries by ‘grammatical designation’,
i.e. word class or part of speech. Where a word belongs unequivocally to the
class of nouns it is not given a label, otherwise the label ‘sb.’ (i.e. ‘substantive’ –
an older word for ‘noun’). After the word class label, a ‘specification’ label may
be given, if a word belongs to a particular domain or subject area (e.g. Mus.
for ‘Music’). Another label that may occur, then, is a ‘status’ label, such as
Obs. (obsolete), arch. (archaic), colloq. (colloquial), dial. (dialect, i.e. ‘now only
dialect, whereas formerly in general use’ – as dialect words are normally not
included in the dictionary), and rare.
Where a word has changed its spelling or had previous alternative spellings,
earlier spellings are given, and if there are a number they are introduced by the
label ‘Forms’, e.g.
Housing Forms: 5 howsynge, husynge, 7 howzen, 7–9 howsing, 7-

housing.
The numbers refer to the centuries, so ‘5’ = fifteenth century, ‘7–9’ = seven-
teenth to nineteenth centuries, and so on. ‘1’ is reserved for ‘Anglo-Saxon’,
pre-twelfth century. The final piece of information under Identification is
‘inflexions’, for nouns and verbs, where these are not formed regularly.
The ‘Morphology’ section of the entry charts the ‘form-history’ of the word,
including its etymology (i.e. its language of origin), any subsequent changes of
form in English, and any other miscellaneous facts about its history. All this is
enclosed in ‘heavy square brackets’; e.g.
Knave [OE. cnafa = OHG. knabo, chnabe (MHG. and G. knabe):— OTeut.
*knabon-. The relation between this and the synonymous cnapa, KNAPE
(q.v.) is not clear. OHG. had also knappo (MHG. and G. knappe): on the
supposed relationship between this and knabo, see Streitberg Urgerm. Gram.
p. 151.]
(Note: OE = Old English, OHG = Old High German, MHG = Middle High
German, G = German, OTeut = Old Teutonic, Urgerm. Gram. = Urgermanische
Grammatik (i.e. ‘Proto-Germanic Grammar’), :— = ‘descended from’, * = ‘re-
constructed form’.)
Knave is a ‘native’ word, from Old English (Anglo-Saxon). Here is an exam-
ple of the Morphology of a borrowed word:
Hurcheon [a. ONF. herichon, OF. heriçun (12th c. in Littré), mod.F. hérisson
(in Hainault hirchon, hurchon, Picard hérichon, irechon) :— pop.L. *hBricion-
em, f. hBricius, late form of Bricius hedgehog. See also URCHIN.]
(Note: a. = ‘adopted from’, ONF = Old Northern French, OF = Old French,

mod.F = Modern French, pop.L = popular Latin, f. = ‘from’; ‘Littré’ refers to a
nineteenth-century historical French dictionary edited by Émile Littré and one
of Murray’s models.)
Hurcheon is a word for a hedgehog, though noted by the OED as now (1901)
limited to Scottish and Northern English dialects. Note the cross-reference to
urchin, another dialect word for ‘hedgehog’.
If the Morphology section of an entry shows the ‘form-history’ of the word,
the Signification section shows the history of its meanings. Where a word has
more than one meaning or ‘sense’, they are numbered in arabic numerals (1, 2,
3, etc.) straight through the entry from beginning to end. Sometimes a word
develops more than one ‘branch’ of meaning independently; the branches are
then numbered with Roman numerals (I, II, III, etc.), but the sense numbers
follow on sequentially in the second branch from those in the first. Should a
sense be subdivided, the divisions are marked by lower-case letters of the alphabet
(a, b, c, etc.). Obsolete senses, like obsolete words, are preceded by the dagger
(†) symbol. Senses that represent incorrect and confused uses of the word are
preceded by the ‘¶’ symbol. For example, the headword idea has five branches:
I. General or ideal form as distinguished from its realization in individuals;

archetype, pattern, plan, standard. (Senses 1 to 6)
II. Figure, form, image. (Sense 7)
III. Mental image, conception, notion. (Senses 8 and 9)
IV. Modern philosophical developments. (Senses 10 and 11)
V. (Sense 12, ‘attrib. and Comb.’, i.e. ‘attributive uses and in combinations’)
Each sense has its own definition, followed by the Quotations relevant to that
sense, arranged in chronological order. The first quotation records the earliest
discovered use of the word in that sense, and the last, if the word has become
obsolete, the last discovered use. The aim was to include approximately one
quotation for each century during which the sense of a word was known to
have been in the vocabulary of English. As an example, here are the quotations
for the first sense of behaviour (‘Manner of conducting oneself in the external
relations of life; demeanour, deportment, bearing, manners’):
1490 CAXTON Eneydos xxxi. 120 For hys honneste behauoure [he] be-
gan to be taken with his loue. 1530 BALE Thre Lawes 53 In clennes of lyfe
and in gentyll behauer. 1601 SHAKS. Twel.N. iii.iv.202 The behauiour of
the yong Gentleman, giues him out to be of good capacity, and breeding.
1754 CHATHAM Lett. Nephew v.32 Behaviour is of infinite advantage or
prejudice to a man. 1797 GODWIN Enquirer I.xiii.III Their behaviour is
forced and artificial. 1862 H.SPENCER First Princ. II.i. § 36 Special direc-
tions for behaviour in the nursery, at table, or on the exchange. 1875
JOWETT Plato (ed.2) IV.226 His courage is shown by his behaviour in
battle.
Whatever its shortcomings, the OED represents a monumental achievement in

lexicography, given the resources available to James Murray, his fellow-editors
and staff in the late nineteenth and early twentieth centuries. It has served well
the generations of students of the English language, who have mined its con-
tents for all manner of scholarly endeavours. But things have not stood still with
the OED, though it seemed for many years as if the 1933 edition represented
the completion of the work.
5.4 The supplements and the second edition

The 1933 supplement, however, made only a selection of the materials then
available for adding to the earlier parts of the dictionary in particular. In the
1950s, Oxford University Press decided that work should begin on a new sup-
plement, not only to fill out the existing contents of the dictionary, but also to
take account of the increase in vocabulary since the dictionary’s publication. In
1957, the year in which W.A. Craigie died, a new editor was appointed to
oversee the work of the supplement: Robert Burchfield, born in New Zealand
in 1923, but who had been working at Oxford University since 1950, where he
formed a friendship with C.T. Onions, who pointed him in the direction of
lexicography.
Burchfield’s supplement was to incorporate and replace the 1933 supple-
ment, and it was originally envisaged as a one-volume publication, to be com-
pleted in seven years. This was later, in 1965, revised to a three-volume
publication. In the event it became four volumes, and the publication spanned
the years 1972 to 1986:
1972 Volume 1, A–G

1976 Volume 2, H–N
1982 Volume 3, O–Scz
1986 Volume 4, Se–Z
In 1957 Burchfield instituted a fresh reading programme, and in the Preface to

the first volume he notes that one-and-a-half million quotations had been ex-
tracted ‘from works of all kinds written in the period since 1884’. The reading
programme also spread its net rather wider than that of the original OED:
We have made bold forays into the written English of regions outside the
British Isles, particularly that of North America, Australia, New Zealand,
South Africa, India, Pakistan. We have endeavoured to extract from text-
books and journals the central enduring vocabulary of all major academic
subjects, including newish disciplines like Sociology, Linguistics, Compu-
ter Science.
Burchfield also noted the inclusion of ‘a wide range of colloquial and coarse
expressions referring to sexual and excretory functions’.
The OED’s distinction between ‘main’ and ‘subordinate’ words is abandoned
in the supplement, though the system of labelling words as ‘archaic’, ‘obsolete’,
and so on is retained. However, the use of subject labels is considerably ex-
tended, in view of the expanding and specialised vocabulary of science, and in
view of the fact that ‘the complexity of many scientific subjects is such that it is
no longer possible to define all the terms in a manner that is comprehensible to
the educated layman’. The supplement volumes contained over 69,000 entries
in all, with more than half-a-million quotations.
As Burchfield’s work on the supplement drew towards its conclusion in the
early 1980s, Oxford University Press decided to take the next step in the devel-
opment of the OED. In 1984, the New Oxford English Dictionary Project was
founded, with the aim of transferring the dictionary to the electronic medium.
It was proposed to integrate the OED and Burchfield’s supplement into a single
electronic text, which would form the basis for future revision and extension of
the dictionary, as well as result in a second edition of the OED in print form.
Overseeing this development were Edmund Weiner, appointed editor in 1984,
and John Simpson, appointed co-editor in 1986.
The transfer of the dictionary and its supplement to electronic form was a
massive undertaking, which required the cooperation of several organisations
on both sides of the Atlantic Ocean: the International Computaprint Corpora-
tion in Fort Washington, Pennsylvania; the University of Waterloo in Canada;
and IBM United Kingdom Ltd; as well as the Oxford University Press. The
project employed more than 120 keyboarders to input the text – the complexity
and variety of the fonts and typefaces precluded the use of a scanning device
– and some 50 proofreaders checked their work. Edmund Weiner and John
Simpson, with a staff of lexicographers in Oxford corrected and edited the
electronic dictionary, and they added around 5,000 new words, to form the
second edition of the OED. One other major change was to abandon Murray’s
transcription system for pronunciation and to use the International Phonetic
Alphabet, which has become the standard system for dictionaries.
OED2 was published in a print version in March 1989, on schedule – the first
time that any part of the OED had appeared on time. It comprises 20 volumes,
runs to 21,730 pages, with 59 million words of text. It defines over half-a-
million words, and includes 2.4 million quotations. It takes up over a metre of
shelf space. A number of electronic versions of the first and second editions of
the dictionary followed, including a magnetic tape, for use on mainframe com-
puters, and a CD-ROM, for use on personal computers. A definitive single
CD-ROM of OED2 was released in 1992, with a second version in 1999.
Not only is the CD-ROM version of the OED more portable, it allows the
dictionary to be searched in ways that would be impossible with the print ver-
sion. For example, any of the following parts of entries may be searched inde-
pendently: headwords, definitions, etymologies, quotations. And the quotations
may be searched separately under ‘date’, ‘author’, ‘work’ and ‘text’. A search
under ‘etymologies’ can reveal all the words that have their origin in a particular
language, or more accurately, all those words that have the language men-
tioned, for one reason or another, in their etymology: the search for ‘Russ’ (i.e.
Russian) gives 473 results. A search under ‘date’ will give all the quotations
from works published in a particular year, under ‘author’ all the quotations
taken from that author’s works, under ‘work’ all the quotations from a particu-
lar publication. However, searching the OED in this way is not straightforward,
and it shows up many inconsistencies in the coding, arising no doubt from the
timescale over which it was compiled and the number of editors and sub-editors
who worked on it. If, for example, you wanted to find all the quotations from
Shakespeare’s Hamlet, you would need to know that the title is usually abbrevi-
ated as Ham.; when you search for Ham, you find that you also have some
extraneous results, such as for the ‘B’ham Daily Post’. Nevertheless, electronic
searching opens up many more uses of the OED as a scholarly tool for research
on the English language and its history. Many inconsistencies and inaccuracies,
though, remain to be ironed out. And that is the purpose of the next phase of
the New OED Project.
5.5 Third edition

Since 1993, the OED team, a staff of some 120, has been undertaking a thor-
oughgoing revision of the whole dictionary, with the aim of publishing a third
edition in 2010. This involves the examination of every word in the dictionary,
to revise definitions where needed, to check pronunciations, to review ety-
mologies in the light of modern scholarship, and to reassess the quotations,
including searching for earlier citations. It is also anticipated that many thou-
sands of new words will be added to the dictionary’s list. Some three volumes of
‘Additions’ have been published in print form. Besides in-house checking and
research, the dictionary staff are in touch with scholars around the world, often
by means of email (or ‘e-mail’ – both spellings are given in OED2), and they
regularly issue ‘appeals’ for information in their regular newsletters and on their
internet website <oed.com>. The ‘Chief Editor’ of OED3 is John Simpson, and
Edmund Weiner is the ‘Principal Philologist’.
Meanwhile, the OED has been on-line since March 2000, and as revisions
are made in preparation for OED3, they are displayed on-line. Beginning with
the letter ‘M’, around 1000 revised entries per quarter are incorporated, and
subscribers can compare the OED3 entry for a word with its OED2 version.
The editors have found that the revision process virtually doubles the size of the
dictionary text. A ‘Preface to the Third Edition’ already exists on-line and
explains principles and procedures that are being followed for this edition. One
of the major areas of work is in updating and augmenting the quotations. Many
words are found to have an earlier citation than previously thought, sometimes
significantly so; for example, Magnificat is now known to have existed in Old
English, instead of Middle English; and macaroon is found in Middle English,
instead of Early Modern English. New data comes into the editorial offices from
four major reading programmes, covering UK and North American sources
from the nineteenth and twentieth centuries, texts from the early modern
period up to the nineteenth century, scholarly articles on the analysis of English
vocabulary, material from other dictionary projects, as well as that assembled by
private contributors. Moreover, electronic sources, such as on-line editions of
newspapers and full-text databases of literary and other works, are searched for
useful material. Efforts are also being made to enhance the coverage of varieties
of English across the world, as well as giving due regard to scientific terminology.
Simpson concludes his Preface by exploding a number of ‘myths’ about the
OED:
There are a number of myths about the Oxford English Dictionary, one of the
most prevalent of which is that it includes every word, and every meaning
of every word, which has ever formed part of the English language. Such an
objective could never be fully achieved. The present revision gives the
editors the opportunity to add many terms which have been overlooked in
the past, but it should be fully understood that fully comprehensive cover-
age of all elements of the language is a chimera. That said, the content of the
Dictionary is certainly comprehensive within reasonable bounds.
It is often claimed that a ‘word’ is not a ‘word’ (or is not ‘English’) unless it is in
‘the dictionary’. This may be acceptable logic for the purposes of word games,
but not outside those limits. Proponents of this view expect dictionaries to
include ‘proper’ English, whereas dictionaries in fact include many slang, infor-
mal, technical, and other words which such people might not consider to be
‘proper’, typically labelled according to the register of the language to which
they belong. It may be added here that the question ‘How many words are there
in the English language?’ cannot be answered by recourse to a dictionary.
Another myth about the Dictionary, and about dictionaries in general, is

that they provide a comprehensive analysis of each word treated. Again,
this cannot be the case in a finite text. But more important, philosophically,
is that any dictionary attempts to provide information in a manner which is
accessible to the reader . . . The reader should . . . regard the Dictionary as
a convenient guide to the history and meaning of the words of the English
language, rather than as a comprehensive and exhaustive listing of every
possible nuance.
The OED has come a long way since the Philological Society resolved to com-
mission a ‘New English Dictionary on Historical Principles’ in 1858. It contin-
ues to develop and be enhanced as an historical record of English vocabulary by
exploiting the new electronic media, and by incorporating the insights of recent
scholarship and research. The third edition promises to fulfil James Murray’s
dream to a level of accuracy and consistency that he was unable to achieve with
the resources at his disposal. But it continues as a legacy to his vision above all
others.
5.6 Further reading

The OED website <oed.com> is an excellent resource on all aspects of the
dictionary, including its history and current development. The website offers a
tour of the OED Online: <oed.com/tour>.
James Murray’s story is told in an account by his granddaughter, K.M. Eliza-
beth Murray, in Caught in the Web of Words (1977). His friendship with Dr
W.C. Minor is described by Simon Winchester in The Surgeon of Crowthorne
(1999).
An explanation of the second edition of the OED is given by Donna Lee
Berg in A Guide to the Oxford English Dictionary (1993). A collection of scholarly
articles on the OED can be found in Lexicography and the OED (2000) edited by
Lynda Mugglestone.
Up to the present 61
6 Up to the present
In our review of the development of the English dictionary we are now going
to backtrack a little and cross the Atlantic, to pick up the story as it unfolded in
America. The American colonies achieved their independence from Britain in
1776. As the new nation instituted its own system of government, investigated
its own flora and fauna, developed contacts, both friendly and hostile, with
speakers of native American languages, and absorbed large numbers of immi-
grants from Europe and elsewhere speaking a wide diversity of languages, the
vocabulary of English in America incorporated a whole new set of words un-
known to the speakers of English in Britain. At the same time, there were
trends, driven by the desire to assert the identity of the infant nation, to establish
an American language, as distinct from the English language. And there was
resistance to these trends.
6.1 Noah Webster (1758–1843)

One of the major advocates in the fledgling nation of the American language
was Noah Webster, ‘a schoolmaster, spelling reformer, lawyer, lecturer, jour-
nalist, crusader for copyright legislation, and the unlikely author of a two-
volume work entitled the History of Pestilential Diseases, which was considered
the standard work in the field’ (Morton 1994: 40). While a schoolmaster,
Webster, dissatisfied with the available textbooks, had published in 1783 A
Grammatical Institute of the English Language, which included a spelling book, as
well as a grammar and a reader. In due course, the spelling book was detached
from the rest and published as the Elementary Spelling Book. The ‘Blue-Back
Speller’, as it was called, was not the first such work, but it became the standard
work used by every child growing up in America; it sold around 100 million
copies during the century of its publication.
Webster was an ardent advocate of spelling reform. He was not the first,
either in Britain or America, and he has certainly not been the last. But Webster
argued for spelling reform on political and patriotic grounds. In his Dissertations
on the English Language (Boston 1789), Noah Webster put forward his argument
as follows (cited in Green 1996: 256):
62 Up to the present
A capital advantage of this reform in these States would be that it would
make a difference between the English orthography and the American.
This will startle those who have not attended to the subject; but I am
confident that such an event is an object of vast political consequence. The
alteration, however small, would encourage the publication of books in
our own country. It would render it, in some measure, necessary that all
books should be printed in America. The English would never copy our
orthography for their own use; and consequently the same impressions of
books would not answer for both countries. The inhabitants of the present
generation would read the English impressions; but posterity, being taught
a different spelling, would prefer the American orthography.
Besides this, a national language is a band of national union. Every en-
gine should be employed to render the people of this country national; to
call their attachments home to their own country; and to inspire them with
the pride of national character.
In the event, the spelling reforms that were adopted in American English were
only a limited subset of those proposed by Webster: the ‘or’ spelling for ‘our’ in
words like favour; the ‘er’ for ‘re’ in words like theatre; and the single consonant,
where British English has a double, in words like traveller.
Another ‘engine’ that would further the sense of a national American lan-
guage would be a dictionary of American English. Webster’s first attempt was
A Compendious Dictionary of the English Language, published in 1806. It was not
particularly original, being based, as Webster acknowledges, on John Entick’s
New Spelling Dictionary of the English Language, published in Britain in 1764 and
imported into America. However, Webster claims to have added 5000 new
words that he had collected from his reading and that reflected life in America.
He also added a 52-page appendix with a variety of ‘encyclopedic’ information,
such as foreign currency conversions, weights and measures, a list of local
post offices, and ‘Chronological Tables of Remarkable Events and Discoveries’.
In the introduction, Webster takes Johnson to task for his failings and claims
that he – Noah Webster – will be recognised as the most influential lexicogra-
pher of the time; moreover, what he would compile would be an American
dictionary.
The truly American dictionary, for which the Compendious Dictionary was just
a precursor, did not appear for another two decades. In 1828, Webster pub-
lished a two-volume work entitled An American Dictionary of the English Lan-
guage, containing some 70,000 entries. A small minority of these represented
words that were to be found exclusively in American English (e.g. bobsled,
gerrymander, moccasin, pretzel, squash, wigwam), and while he preferred citations
from American authors over British ones, they by and large illustrated a com-
mon language. The most admired feature of Webster’s dictionary has been his
definitions: ‘[he] wrote definitions that were more accurate, more comprehen-
sive, and not less carefully divided and ordered than any previously done in
English lexicography’ (Friend 1967, quoted in Morton 1994: 43). James Murray,
the editor of the OED, called him ‘a born definer of words’. The weak point of
the American Dictionary was its etymologies. Webster ignored the new philologi-
cal research coming from Germany and pursued his own idiosyncratic path,
based on his reading of the Bible, by which he thought all post-Flood languages
could be traced back to an original Chaldee. The etymologies would eventually
be revised, after Webster’s death, in an edition edited by C.A.F. Mahn, a Ger-
man scholar, and published in 1864 as the ‘Webster–Mahn’ dictionary.
6.2 Dictionary wars

Webster’s view that America should distinguish itself linguistically from Britain
and develop its own norms and standards of language was not shared by all his
fellow-Americans. Some continued to look towards the mother country for
guidance on lexicographic and linguistic matters. Such a one was Joseph Worces-
ter (1784–1865), who edited a new edition of Johnson’s English Dictionary, as
Improved by Todd, and Abridged by Chalmers; with Walker’s Pronouncing Dictionary,
Combined, which was published in America in 1827. In 1829, Worcester made
an abridgement of Webster’s American Dictionary, from which he omitted many
of the etymologies and citations, and which he augmented with terms that he
had come across while editing Johnson. Webster did not like Worcester’s revi-
sion. In 1830, Joseph Worcester published his own dictionary, the Comprehen-
sive Pronouncing and Explanatory Dictionary; he added some new words, pitched
his spelling somewhere between Johnson and Webster, excluded etymology,
and paid particular attention to pronunciation, his speciality.
In 1834, Noah Webster, by now in his mid-seventies, wrote an article in
which he accused Worcester of plagiarism. Webster’s accusation appears to
have been intended to discredit Worcester, whose dictionary had become a
serious competitor to his own. In lexicography, plagiarism is a difficult accusa-
tion to sustain, because all dictionaries borrow from their predecessors. Webster
had done so himself, as Worcester pointed out in his rebuttal, which also high-
lighted the innovations and differences of the Comprehensive Dictionary. This
exchange began a twenty-year ‘dictionary war’, in which Worcester’s diction-
aries represented a conservative and Anglocentric approach to lexicography,
and Webster’s championed the distinctiveness of American English and the
necessity for America to set its own linguistic standards. A second edition, ‘cor-
rected and enlarged’, of Webster’s American Dictionary was published in 1841,
but it made no impression on the Worcester share of the market. Worcester
responded with his Universal and Critical Dictionary of the English Language in
1846. By this time, Webster had died (in 1843, at the age of 85), and the rights
to reprint and revise his dictionary had been bought from his heirs by George
and Charles Merriam, who were printers and booksellers in Springfield, Massa-
chusetts, and who carried on the fight on Webster’s behalf. They set about
revising and expanding the American Dictionary, publishing it in a single volume
at half its original price in 1847, which greatly increased its sales, despite the
family’s fears that their royalty income would be reduced.
The next salvo in the ‘war’ was fired by Worcester, who brought out a
substantially new work in 1860, the Dictionary of the English Language, and it
soon became recognised on both sides of the Atlantic as the best available dic-
tionary. But Webster was to triumph in the end. A thoroughgoing revision of
the American Dictionary was undertaken by the editor, Noah Porter, with Carl
Mahn, from Berlin, bringing the etymologies into line with modern scholar-
ship. This is the edition that was known as the ‘Webster–Mahn’, though the
publishers used the term ‘Unabridged’; it contained 114,000 words. Published
in 1864, it became the dictionary of preferred use in education, the law and
printing presses. Joseph Worcester died a year later, and the ‘war’ was effectively
over. Webster’s dictionaries, published by G. and C. Merriam Company, went
from strength to strength, with a new edition in 1890 called Webster’s Interna-
tional Dictionary, which claimed 175,000 entries, and the New International in
1909, which claimed in its Preface:
The book has been entirely remade: the previous material has been sifted
and rearranged; a radical change in the construction of the page has been
introduced; the definitions have been treated with nicer discrimination and
a more historic method; and in every department there has been an en-
largement of the vocabulary and an enrichment with new information. Its
salient features in comparison with its predecessor are: A more full and
scholarly treatment of the whole field of the English language; a great addi-
tion of words and definitions; a greatly increased amount of encyclopedic
information; a more exhausting and discriminating treatment of synonyms;
a greater comprehensiveness in the illustrations; and an arrangement of
material that makes the dictionary much easier of consultation. In short, the
New International is essentially a new book.
The dictionary claimed over 400,000 ‘words printed in bold-faced type, to-
gether with the inflected forms that appear in small capitals’.
The numbers game continued to be played. Funk and Wagnalls claimed
450,000 entries in their New Standard Dictionary of 1913. Webster’s Second New
International Dictionary of 1934 claimed 600,000, though it is no longer clear
exactly what is being counted, certainly not just headwords (see 3.4). This
second edition of the New International, edited by William Allan Neilson, achieved
a dominance among American dictionaries, and ‘still figures in the minds of
many middle-aged and elderly Americans as the dictionary par excellence’ (Landau
1989: 64). But preparations were already being made for a third edition.
6.3 Webster’s Third New International Dictionary

A systematic reading programme for the third edition was begun by editorial
staff at G. and C. Merriam in 1936, but the chief editor to see the project
through was not identified until 1951. He was Philip B. Gove, who had been
on the staff of the dictionary since 1946. He was supported by a number of
associate and assistant editors, and they called upon more than 200 specialist
outside consultants, on everything from Iranian etymology to pipe organs. It
was published in 1961, and it was, claimed Gove in the Preface, ‘a completely
new work, redesigned, restyled, and reset; every line of it is new’. The reading
programme amassed some four-and-a-half million citations, to add to those
already on file, as well as to those available in the OED and other historical
dictionaries. The dictionary was the eighth in line from Webster’s American
Dictionary of 1828, and it aimed to be ‘a prime linguistic aid to interpreting the
culture and civilization of today, as the first edition served the America of 1828’.
W3 claims a vocabulary of 450,000 words. It also claims to have included
100,000 new words that were not in the second New International. So, it must
have removed 250,000 items that were in W2. W3 has no appendices, all
entries are contained within the main alphabetical sequence. Gove claimed
continuity with Merriam–Webster tradition, though his critics did not always
see it like that. In respect of definitions, though, there was innovation:
The primary objective of precise, sharp defining has been met through
development of a new dictionary style based upon completely analytical
one-phrase definitions throughout the book . . . Defining by synonym is
carefully avoided.
Here is the definition of the first sense of pantomime:
a solo dancer of imperial Rome acting all the characters of a story (as of
tragic love) usu. from myth or history by means of steps, postures, and
gestures alone with the help of changes of mask and costume, a chorus sing-
ing the narrative usu. in Greek, an orchestra, and sometimes an assistant.
It was in the overall policy that the Merriam–Webster traditions were honoured:
In continuation of Merriam–Webster policy the editors of this new edition

have held steadfastly to the three cardinal virtues of dictionary making:
accuracy, clearness, and comprehensiveness. Whenever these qualities are
at odds with each other, accuracy is put first and foremost, for without
accuracy there could be no appeal to Webster’s Third New International as an
authority. Accuracy in addition to requiring freedom from error and con-
formity to truth requires a dictionary to state meanings in which words are
in fact used, not to give editorial opinion on what their meanings should be.
It was this policy – to state meanings in which words are in fact used – that was
seen by critics of the dictionary as innovative and, moreover, damaging. If
people could no longer look to their Webster’s dictionary for an authoritative
pronouncement on what the meaning ought to be, how words ought to be
pronounced, spelled and used, then they were adrift in a linguistic sea without
any chart or compass.
The American public has generally looked to their dictionaries as uncontestable
and reliable authorities in matters of language usage, more so than the British
public. W3 was widely considered to have abrogated its cultural role by suc-
cumbing to the influence of modern structural linguistics and its ‘descriptivist’
stance. Here is a typical reaction, from a review in the Chicago Daily News,
October 1961 (reproduced in Sledd and Ebbitt, 1962: 81):
In this new edition, it turns out that good English ain’t what we thought it
was at all – good English, man, is whatever is popular. This is a nifty speak-
as-you-go dictionary. Not like that moldy fig of a Second Edition, which
tried to separate ‘standard English’ from slang, bastardized formations, col-
loquialisms, and all the passing fads and fancies of spoken English.
What’s the point in any writer’s trying to compose clear and graceful
prose, to avoid solecisms, to maintain a sense of decorum and continuity in
that magnificent instrument, the English language, if that peerless author-
ity, Webster’s Unabridged, surrenders abjectly to the permissive school of
speech?
Many reviewers counselled their readers to hang on to their second edition.

The one item that was cited more than any other as evidence of this new
permissiveness was the entry for ain’t. Some reviewers implied that it was new
to W3, even though it had been entered in Webster’s dictionaries since 1890.
Others inferred that W3 approved of ain’t, or at least did not condemn it suffi-
ciently. The first edition of the New International (1911) had the following entry
for ain’t:
Contr. for are not and am not; also used for is not. Colloq. or illiterate.
The entry in the third edition is more extensive, and the comments on its usage
are more differentiated:
1 a: are not <you ~ going> <they ~ here> <things ~ what they used to be>
b: is not <it ~ raining> <he’s here, ~ he> c: am not <I ~ ready> — though
disapproved by many and more common in less educated speech, used
orally in most parts of the U.S. by many cultivated speakers esp. in the
phrase ain’t I 2 substandard a: have not <I ~ seen him> <you ~ told us> b:
has not <he ~ got the time> <~ the doctor come yet>
Even though the comment on the am/is/are not contraction implied a ‘collo-
quial’ restriction, and the has/have not contraction was labelled ‘substandard’,
the critics missed the condemnation contained in ‘illiterate’.
Not all reaction to Webster’s Third was hostile, and the dictionary continues
in print, with successive impressions augmented by an ‘Addenda Section’ of
new words and phrases. A fourth edition is said to be in preparation (Béjoint
2000: 45).
6.4 Collegiate dictionaries
Although American dictionaries range in size from the ‘unabridged’ dictionaries
like W3 to ‘pocket’ dictionaries, there is most competition in the ‘college’
dictionary market. The college, or ‘collegiate’, dictionary is a desk size diction-
ary, aimed at the lucrative upper secondary (‘high’) school and undergraduate
(‘college’) sector, as well as the learner of English as a second language (Hartmann
and James 1998). All the main publishers have a dictionary of this type, including:
• Merriam Webster’s Collegiate Dictionary (tenth edition, 1995)

• Random House Webster’s College Dictionary (second edition, 1997)
• American Heritage Dictionary of the English Language (fourth edition, Houghton
Mifflin, 2000).
By comparison with their British counterparts (see below), American diction-

aries have tended to be more ‘encyclopedic’ in their scope. They have routinely
included biographical and geographical entries, as well as more extensive scien-
tific and technical information. Diagrams and line drawings also intersperse the
text. The college dictionary is much more of an all-purpose reference work. On
the other hand, they tend to contain limited information on etymology, and
they pay little attention to other varieties of English than the American (Béjoint
2000). Spelling and pronunciation are almost uniformly American. Some of
the practices of American lexicography, including numbered definitions and
biographical and geographical entries, have been imitated by some British
dictionaries.
6.5 British dictionaries

The British equivalent, in terms of size, of the American collegiate dictionary is
the ‘desk’ dictionary, represented by:
• Collins English Dictionary (fourth edition, 1998)

• New Oxford Dictionary of English (1998)
• Longman Dictionary of the English Language (second edition, 1991)
• Chambers English Dictionary (1988).
British publishers also have a long tradition of producing ‘concise’ dictionaries,

the most famous of which is the Concise Oxford Dictionary, first published in
1911, under the editorship of F.W. and H.G. Fowler, and now in its tenth
edition (1999), edited by Judy Pearsall.
Each dictionary has its own particular characteristics. When CED appeared
in its first edition in 1979, it introduced into British dictionaries the American
practice of numbered definitions and the inclusion of biographical and geo-
graphical entries. Its definitions of scientific and technical terms also tend to-
wards the encyclopedic, including the Latin terms for flora and fauna. As an
example, here is the definition of elephant:
either of the two proboscidean mammals of the family Elephantidae. The
African elephant (Loxodonta africana) is the larger species, with large flapping
ears and a less humped back than the Indian elephant (Elephas maximus),
of S and SE Asia.
In terms of vocabulary coverage, CED excels in its attention to the specialist

terms of a variety of subject areas, its inclusion of words from the main national
Englishes around the world, as well as from British English dialects.
For Oxford University Press, whose flagship general-purpose dictionary had
been the Concise, NODE was a significant innovation. Not only was it Oxford’s
first desk size dictionary, it also pioneered a departure in the treatment of the
meanings of words, since followed by the tenth edition of the Concise. NODE
recognises at a general level the ‘core’ sense(s) of a word, for each of which
‘subsenses’ may be identified. For example, the noun grain has six numbered
core senses in NODE, as against twenty-three numbered senses in CED4. Like
CED, NODE includes biographical and geographical entries. It is also notable,
not only for having three columns per page, but more especially for the clarity
of the layout of its entries to aid accessibility (Chapter 7).
The first edition of LDEL in 1984 was based on Webster’s Collegiate Dic-
tionary, but the text underwent significant revision and augmentation for the
second (1991) edition. As its origin would suggest, it contains biographical and
geographical entries. Also significant is the practice of a separate entry for each
word class that a lexeme may belong to; for example, puncture has a noun entry
and a verb entry, purple has an adjective, a noun and a verb entry. The other
notable feature of LDEL is the boxed synonym essays explaining the, often subtle,
differences between words with similar meanings; for example, under danger is
a box discussing the differences between danger, peril, jeopardy, hazard and risk.
Chambers English Dictionary has a long history, going back to the publication
of the first edition of Chambers Twentieth Century Dictionary in 1901. It is some-
thing of a relic of an earlier tradition, with its extensive nesting of compounds
and derivatives under the root lexeme, the absence of numbering of its defini-
tions, and its appendices for foreign phrases and abbreviations (included by most
modern dictionaries in the main body). While consciously including words
from other national varieties of English, including those of Australia, South
Africa and the Caribbean, it also aims to encompass the vocabulary of Shake-
speare, Spenser and Milton. Its coverage of Scottish English words is unrivalled,
as befits its publishing origin. And it continues the Johnsonian tradition of the
occasional amusing definition; one that has persisted in Chambers is that for
éclair:
a cake, long in shape but short in duration, with cream filling and chocolate
or other icing.
Chambers 21st Century Dictionary, first published in 1996, is a radical departure,

bringing Chambers more into line with current lexicographical practice.
6.6 Learners’ dictionaries
Perhaps the most interesting and innovative sector of British lexicography over
the past quarter of a century and more has been that devoted to the develop-
ment of the ‘monolingual learner’s dictionary’, coinciding with the growth of
the English-as-a-foreign-language industry. Beginning with A.S. Hornby’s
Advanced Learner’s Dictionary of 1948, which remained unrivalled for three dec-
ades, the market now has four major dictionaries:
• Oxford Advanced Learner’s Dictionary (sixth edition, 2000)

• Longman Dictionary of Contemporary English (third edition, 1995)
• Collins COBUILD English Dictionary (third edition, 2001)
• Cambridge International Dictionary of English (1995).
Since the first edition of LDOCE in 1978 to rival the OALD, then in its third,
but not significantly different, edition, the developments in learners’ diction-
aries have been remarkable, as each has sought to gain the edge in providing a
dictionary that will both meet the particular needs of this user group and make
the information as readily accessible as possible.
Learners’ dictionaries are discussed in some detail in Chapter 11. Let us just
note here that they have not been limited to the A-to-Z format of the four listed
above. Tom McArthur’s Longman Lexicon of Contemporary English (1981)
arranges the vocabulary it covers by topic (lexical field), so that the learner
can both perceive the words available for an area of meaning and learn the dif-
ferences in meaning between them. In 1993, the Longman Language Activator
announced itself as ‘the world’s first production dictionary’. Intended as a sup-
plement to the traditional learner’s dictionary, the Activator aims to help learners
choose the appropriate word for a given context by arranging the vocabulary
under around one thousand ‘key words’. Both these dictionaries are further
discussed in the context of Chapter 12.
6.7 Electronic dictionaries

A number of dictionary publishers have made their dictionaries available in the
electronic medium. We have already mentioned the availability of the OED
both on-line and on CD-ROM (Chapter 5). Before the advent of CD-ROM
technology, the third edition of Collins English Dictionary was produced on 3.5″
floppy disks, but quite a number of dictionaries are now available in CD-ROM
format, including:
• Oxford English Dictionary, second edition

• Oxford Talking Dictionary, a version of the Shorter Oxford English Dictionary
with pronunciations, published by The Learning Company
• Concise Oxford Dictionary, ninth edition
• Concise Oxford Dictionary, tenth edition
• New Oxford Dictionary of English
• Longman Dictionary of the English Language, packaged with ‘Infopedia UK’,
published by SoftKey
• Encarta World English Dictionary 2001
• Oxford Advanced Learner’s Dictionary
• Longman Interactive English Dictionary, a version of LDOCE
• Collins COBUILD English Dictionary
• Cambridge International Dictionary of English.
The electronic medium opens up a number of possibilities for dictionaries

that are not presented by the print versions. We will now review these possi-
bilities and how they have been exploited by some of the current CD-ROM
dictionaries.
At its simplest level, a CD-ROM dictionary allows you to look up a word,
just as in a print dictionary; except that, instead of turning pages to find the
word, you enter it in a box and press the ‘Enter’ key or click on a ‘Search’ or
‘Find’ icon to initiate the electronic search. The entry for the word is then
displayed in a larger frame, while the headword list with the selected word
highlighted is displayed in a smaller frame. It is then usually possibly by one
means or another to scroll forward and backward through the headwords and
their associated entries. This is the limit of the possibilities allowed by the
‘Infopedia’ LDEL, except that the CD-ROM contains a number of reference
works that can all be searched together, thus providing an integrated informa-
tion system.
The electronic medium allows considerably more sophisticated exploitation
of a dictionary text. The COD10 on CD-ROM, for example, gives an initial
choice between a ‘Quick’ search and a ‘Full’ search. The Quick search relates to
the headword list, and the Full search to the complete text of the dictionary.
The screen is divided into a headword list on the left and the dictionary entry on
the right. The Quick search allows:
• a straightforward lookup, by typing the headword in the search box, whose

place will be found in the headword list and its entry displayed in the larger
righthand window; you can scroll through the dictionary in either direc-
tion by moving the cursor from item to item in the headword list.
• a lookup using ‘wildcards’, where, as is now the convention, ‘?’ stands for
a single letter, and ‘*’ stands for any number of letters; a search for ‘?a?e’ will
find all four-letter words having ‘a’ as their second letter and ‘e’ as their
fourth; ‘*ist’ will find all the words ending in ‘ist’; the results of the search
are displayed in the lefthand box, together with the total number of items
found (180 for ‘?a?e’, 291 for ‘*ist’).
The ‘hypertext search’ facility enables an instant cross-reference to any word in

an entry, simply by positioning the cursor over the word and clicking on it.
Thus, if you do not understand a word used in a definition, you can immedi-
ately consult the entry for that word by this means. Some entries have words
specifically marked (highlighted in red) for cross-reference.
The Full search facility has three boxes, in which items can be typed, con-
nected by ‘Boolean’ operators (and, or, not). Any number of the boxes may be
utilised for a search of the full text of the dictionary. Some of the possibilities
are:
• search for a word, by typing it in the first box: fabric will find all the (317)
entries that contain this word, including those whose definitions indicate
that the word denotes a type of fabric; russ will find all the (123) words that
are marked as being of Russian origin (though you need to know that Russ
is used in the etymologies as the abbreviation of Russian).
• search for a phrase, by typing it in the first box: time of year will find the nine
items (six of them including season) that contain this phrase in their defini-
tions; musical instrument finds 81 items, some of which are types of musical
instrument.
• search for two or three words to occur simultaneously in an entry, though
not necessarily as a phrase, by using the operator and to join them: ‘flower
and petal’ finds seven items; ‘greek and roman and art*’ finds classical, classicism
and muse.
• search for two alternative forms: ‘complementary or complimentary’ finds 36
entries containing these words, all but four with the first of them.
• search for one (or two) forms but excluding a second (or third): ‘colour not
pigment’ finds 444 entries, ‘pigment not colour’ finds 71 (and, incidentally,
‘colour and pigment’ finds 19).
The Full text searches in the electronic COD10 are powerful and certainly
beyond what could be achieved using the print version of the dictionary. But
they are quite undiscriminating, finding a search item anywhere in the diction-
ary text.
The electronic COD9, by contrast, allows for more discriminating full-text
searching. It achieves this by allowing the user to switch on one or more of a
number of ‘filters’, which restrict the search to: headwords, definitions, idioms,
phrasal verbs, etymology. By switching on the ‘etymology’ filter, for example,
and selecting Russian as the search term, it will find this term only in the etymol-
ogy field of an entry. By switching on the ‘phrasal verbs’ filter and asking it to
search for ‘* & up’, it will find all the phrasal verbs that are formed with up as
their adverb particle. The ‘idioms’ filter enables you to search for all the idioms
containing a particular word, e.g. time. Selecting the ‘headwords’ filter also
allows the search to be restricted by word class, by means of a ‘part-of-speech’
filter. Switching on the ‘headwords’ filter, selecting ‘adjective’ from the ‘part-
of-speech’ filter, and then searching for ‘*able/*ible’ finds all the adjective head-
words and run-on derivatives ending in this suffix. The ‘definitions’ filter covers
everything else, including any restrictive labelling (Chapter 9): so, to find all the
words labelled ‘derogatory’ in COD9, you switch on the ‘definitions’ filter and
search for derog (once you have discovered that this is the abbreviation used by
COD9).
For lexical research purposes, the electronic COD9 is a more flexible and
sophisticated tool than its later counterpart. Moreover, the COD9 CD-ROM
contains all the front and back matter of the print dictionary, including the
preface, the guide to using the dictionary, and the very useful ‘style guide’. It
also provides a ‘standard British pronunciation’ of every word, which is activ-
ated by double-clicking on the phonetic transcription.
Using the search facilities opened up by the medium is not the only way in
which publishers have sought to exploit CD-ROM based dictionaries. The
CD-ROM version of NODE, for example, has the more limited search facili-
ties of COD10, but it is intended to match with other Oxford reference titles to
form a seamless integrated electronic library. The Oxford Talking Dictionary,
developed by The Learning Company, takes the text of The New Shorter Oxford
English Dictionary (1993), ‘a historical dictionary of modern English’ (Preface,
p. vii), including its quotations, and adds a number of other features: pronuncia-
tion, a proper name dictionary, a thesaurus, maps and pictures – to produce a
searchable encyclopedic dictionary and thesaurus.
These are the beginnings of the exploitation of the electronic medium for
dictionary purposes, but, apart from searching of various degrees of sophistica-
tion and integration with other reference works, publishers have done little
more than transfer the print dictionary to the electronic medium (Pruvost 2000).
There are more imaginative possibilities, and some of the CD-ROM versions
of the learner’s dictionaries have begun to take them, as we shall see when we
discuss them in Chapter 11.
6.8 The future

Are electronic dictionaries the future? Do they spell the end of the print dic-
tionary? It is probably too early to write the print dictionary’s obituary. For the
majority of lookups that we make day-to-day (Chapter 7) – checking a spelling,
seeing if a word exists, finding out the meaning of an unusual word – it is the
print dictionary that we will continue to reach for. The information we need is
limited, and the print version is to hand, portable and convenient. However,
when we are sitting at the computer, writing a letter or a report or an essay, we
will naturally use the electronic dictionary on our hard drive, because it is only
a click away and is, in any case, linked into our word processing software. We
may also choose to power up the computer if we have a particular problem that
we think the electronic dictionary can solve, such as using the wildcard function
to find the possible words that might go into the crossword puzzle that we’re
struggling with.
Certainly, if we are students and scholars of language, the electronic versions
open up to us possibilities for the exploitation of dictionary texts that the print
versions could not begin to offer. Similarly, if we are professionally involved
with language, for examples as translators or writers or teachers, the electronic
dictionaries, with their still unrealised potential for development, provide a
superior resource to the print dictionary. As we noted in Chapter 5, the poten-
tial for the great Oxford English Dictionary, both in its development and in its use,
is substantially enhanced by the electronic medium. Indeed, it is debatable whether
a print version of the OED will ever again be economically viable.
The electronic revolution has had and is having a profound effect both on the
compilation of dictionaries (Chapter 13) and on the ways in which they can be
and are being used. But print dictionaries are not in terminal decline; they are
still likely to be around for some considerable time to come.
6.9 Further reading

Chapters 10 and 11 of Jonathon Green’s Chasing the Sun (1996) trace the history
of lexicography in America. The controversy surrounding W3 is told in Herbert
C. Morton’s The Story of Webster’s Third (1994), and James Sledd and Wilma R.
Ebbitt’s Dictionaries and THAT Dictionary (1962) contains a fascinating collec-
tion of contemporary reviews of W3.
Dictionary typology is discussed in Chapter 5 of Reinhard Hartmann’s Teaching
and Researching Lexicography (2001), and in Chapter 1 (pp. 32ff ) of Henri Béjoint’s
Modern Lexicography (2000).
Material on electronic dictionaries is still sparse. American English diction-
aries on CD-ROM are reviewed by Creswell (1996); Nesi (1999) looks at
electronic dictionaries for language learners (more applicable to Chapter 11);
and Pruvost (2000) reports on a colloquium that discussed the transferring of
print dictionaries to the electronic medium.
74 Users and uses
7 Users and uses
In Chapter 3, we noted the important place that ‘the dictionary’ holds in our
culture. The education system makes extensive use of them. They are objects of
discussion and review in the media. We are a ‘dictionarate’ society (Ilson 2001).
Most people own one, have some idea about what it contains, consult it from
time to time. In this chapter, we are going to consider who uses a dictionary and
for which purposes. We look first at how dictionaries conceive of their aims,
then at the uses people make of dictionaries and how the structure of diction-
aries aids or hinders access to the information that users seek. Finally, we ask
whether dictionaries have two sets of conflicting and incompatible aims.
7.1 Dictionary aims

In the Preface to W3, the editor, Philip Gove, wrote of the aims of the diction-
ary as follows:
G. and C. Merriam Company now offer Webster’s Third New International

Dictionary to the English-speaking world as a prime linguistic aid to inter-
preting the culture and civilization of today, as the first edition served the
America of 1828.
This edition has been prepared with a constant regard for the needs of the
high school and college student, the technician, and the periodical reader,
as well as of the scholar and professional . . . The dictionary more than ever
is the indispensable instrument of understanding and progress.
This new Merriam–Webster unabridged is the record of this language [Eng-
lish] as it is written and spoken. It is offered with confidence that it will
supply in full measure that information on the general language which is
required for accurate, clear, and comprehensive understanding of the
vocabulary of today’s society.
Gove identifies the groups of users whose needs have especially been consid-
ered in the course of compiling the dictionary, and he specifies what those
needs are: interpreting today’s culture and civilisation, understanding the
Users and uses 75
vocabulary of today’s society. ‘Interpreting’ and ‘understanding’ (hard words)
were the aims of Robert Cawdrey’s first monolingual dictionary in 1604 (see
Chapter 4). But W3 has an aim that Cawdrey did not have, and it in large part
explains the difference between them: to be the ‘record’ of the language. What
it is exactly that dictionaries are recording has been touched on briefly in Chap-
ter 3, and it will be examined in more detail in Chapters 8 to 10.
All dictionaries make a selection from the one to two million words of cur-
rent English. Gove acknowledges this in respect of the 450,000 words chosen
for W3. So, when a dictionary claims to provide ‘the most comprehensive
coverage of current English’ (NODE), or to be ‘the world’s most comprehen-
sive dictionary’ (CED4), we must interpret those statements as meaning ‘of
large content or scope; wide-ranging’ rather than ‘including or dealing with all
or nearly all elements or aspects of something’ (definitions from NODE). All
dictionaries aim for comprehensiveness in the ‘wide-ranging’ sense, for, as Samuel
Johnson noted in his Plan, ‘it is rather to be wished that many readers should
find more than they expect, than that one should miss what he might hope to
find’. NODE claims to have ‘targeted previously neglected fields as diverse as
computing, complementary medicine, antique collecting, and winter sports’
and to have undertaken ‘a detailed and comprehensive survey of plants and
animals throughout the world, resulting in the inclusion of hundreds of entries
not in any other one-volume dictionary’ (Preface). CED4’s claim to compre-
hensiveness rests on its coverage of ‘the whole spectrum of general language
from formal and archaic to slang and informal expressions’, its inclusion of ‘the
language of an enormous range of general subjects, from art to television, and
specialist subjects, from aeronautics to zoology’, as well as ‘varieties of English
from all over Britain and around the world’ (Foreword). Comprehensiveness
ranges, then, along the dimensions of formality, topic, and dialect.
Chambers, which also claims to have ‘enjoyed long acceptance as the most
comprehensive single-volume dictionary of the English language’ (dust jacket),
notes that dictionaries ‘must be faithful recorders of the language’ (Preface). In
fulfilling this obligation, they need to reflect changes in society and their effects
on the vocabulary, and this, in turn influences the selection of the words to
include. However, Chambers is mindful that, as well as being ‘of unrivalled value
to . . . students, scholars, writers, journalists, librarians and publishers’ and ‘re-
plete with words of technical importance to scientists, lawyers, accountants and
people in business’, it is also ‘the treasure chest for all word-game players and
word lovers’ (dust jacket). For this reason, Chambers includes ‘unusual and
archaic words’ for such users. Chambers also numbers among its users those who
‘read it simply for pleasure’ (Preface). The dictionary has an extensive range of
users and their needs to satisfy.
And they, like all dictionary users, need to be able to find their way around
the dictionary and access efficiently the information that they are seeking.
Dictionaries aim to be ‘accessible’. NODE claims ‘maximum accessibility –
with a revolutionary new entry style’ (dust jacket). CED4 claims that ‘in 1979
Collins English Dictionary revolutionised the way English dictionaries were
76 Users and uses
presented . . . by the simple but, at that time, radical approach of considering
the user’s needs’ (Foreword). Chambers claims to be ‘the most accessible of
reference books’ (dust jacket). Accessibility is about how a user gets at the
particular piece of information about the word or phrase that they are consult-
ing the dictionary for. In a reference work of the scale and density of a desk size
or concise dictionary, this is clearly an important issue, since users are not
usually prepared to peruse a considerable amount of material in order to locate
the sought-for nugget of enlightenment. We discuss below the factors that con-
tribute to accessibility; here, we merely note that it is an aim of dictionaries
to be accessible.
We can conclude that dictionaries have two fundamental aims: coverage, and
accessibility. Coverage includes the aim to be ‘comprehensive’, representing an
up-to-date and wide-ranging selection of vocabulary, and the aim to be a ‘faith-
ful record’ of the lexical resources of the language. Both aims can be viewed as
user-oriented: ensuring that what a user wants to know is made available, and
enabling the user to get at it by the most straightforward means. But do we
know what users consult their dictionaries for?
7.2 Dictionary uses

What kind of reliable evidence is there about what users look up in a dictionary?
What methodologies are there for finding out? Most of the surveys that have
been done (referenced in Béjoint 2000: 141) have involved the use of a ques-
tionnaire, in which users (mainly students) report on their own dictionary use.
Self-reporting does not always produce the most reliable data; besides a tend-
ency to overstate or underplay, according to question asked and the personality
of the respondent, questions such as ‘How often do you consult a dictionary?’
are notoriously difficult to gauge accurately. Nevertheless, results from ques-
tionnaires on the use of native speaker dictionaries all agree that the top two
reasons for consulting a dictionary are:
1 to discover the meaning of a word

2 to check the spelling of a word.
The first of these may involve either looking up a word that is unknown and
whose meaning cannot be deduced from the context in which it has been met
(a ‘hard’ word), or checking the meaning of a word about which the user may
be confused (perhaps, for example, aggravate).
A questionnaire administered to French students of English on their use of
monolingual English dictionaries (including learners’ dictionaries) revealed that
looking up the meaning of words was also the top use for this group of users
(Béjoint 1981). Checking spelling came joint fourth with checking pronuncia-
tion. Reflecting their use of dictionaries for writing tasks, including translation
(see 6.5 below), the second and third most frequent occasions of lookup for
these students were: to check the syntactic patterns that a word could enter; and
Users and uses 77
to discover a synonym for a word. What Béjoint also discovered was that most
of the respondents had not read the front matter of their dictionary, in particular
the guide to using the dictionary, and so were unaware of the wealth of infor-
mation that the dictionary contained.
Despite a hundred or more studies of dictionary use, we are still far from
understanding either the range of uses for which dictionaries are consulted or
the strategies that are used to access dictionary information (Hartmann 2001).
Most of the studies have focused on students, and many of them on students
engaged in tasks associated with foreign language learning – translation, writing
in the foreign language, and so on. And most of them have used a questionnaire
as the study instrument, which does not necessarily produce an authentic pic-
ture of dictionary use, as expressed by Hatherall (1984: 184):
Are subjects saying here what they do, or what they think they do, or what
they think they ought to do, or indeed a mixture of all three?
An alternative method of investigating dictionary use is the ‘protocol’ or diary,

when a respondent is asked to record their lookup each time they use a diction-
ary in the course of a specific task (e.g. Nuccorini 1992). Again, this involves
self-reporting, but it requires recording procedures as they are being undertaken
or immediately on completion, so the resulting data may well turn out both to
be more authentic and to provide greater insight into the dictionary lookup
process and the reasons for it. A method that does not involve self-reporting
would be direct observation of subjects, which would only be possible in con-
trolled circumstances (e.g. a classroom), and may itself influence the way in
which subjects behave. It is not easy to get at either what triggers a dictionary
consultation or exactly what happens when it occurs. A further question that
deserves investigation is whether users access electronic dictionaries differently
from print dictionaries (cf. Nesi 1999).
One dictionary that claims to have consulted its potential users and to have
adjusted its content accordingly is the Encarta Concise English Dictionary (2001).
Kathy Rooney, the Editor-in-Chief, states in her ‘Introduction’:
People like you said they wanted answers to the following questions. Am I
spelling this word correctly? What does this word mean? Am I using the
word correctly? How do I pronounce this word? Where does the word
come from? We also established that you set great store by the ease with
which you can understand the information in the dictionary, the clarity
with which it is presented, and the speed with which you can navigate
through long entries. In addition, we asked 41 professors of English from
the UK, Australia, the United States, and Canada about the language prob-
lems their students faced. This survey revealed surprisingly similar findings
across the globe. All expressed concern that students increasingly have dif-
ficulties with basic language skills – especially spelling and grammar.
(p. xi)
78 Users and uses
On this basis, as well as that of personal experience and anecdote, we may
conclude that users consult dictionaries for more than just ‘meaning’ and ‘spell-
ing’, though for these primarily. Where a dictionary is consulted during word
games, it is usually to verify whether a word exists or is a legitimate formation.
We may consult a dictionary to determine the pronunciation of a, usually tech-
nical, word that we have met only in print. Students of language may need to
find out about word histories and origins, or about the range and contexts of
meaning that a word may have. Or, as Chambers suggests, people may peruse a
dictionary for pleasure, or for ‘edification’ (Hartmann 2001: 88). Occasions of
lookup, or reference needs, are diverse and various: how do dictionaries aid the
user to access the specific piece of information they are looking for?
7.3 Accessibility
Let us consider first what may be involved in the dictionary lookup process.
Scholfield (1999: 13–14), whose focus is on language learners consulting a dic-
tionary to find out about the meaning of a word, suggests that there are ‘five
main steps’:
1 the learner identifies a vocabulary problem – that is a word or phrase whose

meaning is unknown or uncertain
2 the learner makes the decision to use a dictionary to solve the problem,
rather than, or in addition to, other possible means
3 he or she has to find an entry for the wordform or phrase being sought
4 the right specific entry or part of an entry has to be located
5 the information about meaning that has been obtained has to be exploited.
The points at which accessibility becomes particularly relevant are steps 3 and 4,
which relate to the ‘macrostructure’ and the ‘microstructure’ of the dictionary
respectively.
A dictionary’s macrostructure refers to what constitutes an entry in a diction-
ary and how the entries are arranged. We assume that a dictionary page has two
(sometimes three) columns, and that entries follow each other in alphabetical
order. Compare the headword list (from pros) for the following two dictionaries
of similar size and published within three years of each other shown in Table
7.1.
In this brief headword list, Chambers has only one item that is not included in
LDEL (prosauropod), whereas LDEL has thirteen that are not in the Chambers
headword list. However, apart from the first two – the abbreviation pros. and
the prefix pros- – they are all contained in Chambers, though not as headwords.
The macrostructure of the two dictionaries differs most obviously in the deci-
sions about which items feature as headwords. Chambers nests all derivatives; so
prosaism and prosaist are both within the entry for prosaic, proscription is under
proscribe, prosecution under prosecute, and so on. Similarly, the noun, verb and
adjective uses of prose, which have separate headwords in LDEL, are all dealt
Users and uses 79
Table 7.1
Chambers (1988) LDEL (1991)
pros.
pros-
prosaic prosaic
prosaism
prosaist
prosauropod
proscenium proscenium
proscenium arch
prosciutto prosciutto
proscribe proscribe
proscription
1
prose prose
2
prose
3
prose
prosector prosector
prosecute prosecute
prosecuting attorney
prosecution
prosecutor
proselyte proselyte
proselytism
proselytize
prosencephalon prosencephalon
with in the one entry in Chambers, as is the more usual practice. LDEL puts its
abbreviations, like pros. in the headword sequence; Chambers collects them to-
gether in an appendix. Such macrostructure decisions affect the accessibility of
headwords, and so the search in Scholfield’s Step 3. All the words beginning
with the combining form neur(o)-, for example, are under a single headword in
Chambers, but each has headword status in LDEL; arguably, it is easier to find a
word like neuroleptic in LDEL, by scanning down the headword list, than in
Chambers, which requires perusal of a continuous paragraph, even though the
lexemes themselves are in bold typeface.
A dictionary’s microstructure refers to the layout and organisation of the
individual entry. Compare the following entries for nest from Chambers and
COD10:
nest nest, n. a structure prepared for egg-laying, brooding, and nursing, or

as a shelter: a place of retreat, resort, residence or lodgment: a den: a com-
fortable residence: a group of machine-guns in a position fortified or screened
by sandbags or the like: a place where anything teems, prevails, or is fos-
tered: the occupants of a nest, as a brood, a swarm, a gang: a set of things (as
boxes, tables) fitting one within another: a set of buildings, as advance
factories, divided into blocks and units: an accumulation: a tangled mass. –
80 Users and uses
v.i. to build or occupy a nest: to go bird’s-nesting. –v.t. and v.i. to lodge,
settle. –n. nest’er one who builds a farm or homestead on land used for
grazing cattle (U.S. hist.; derog.): a nest-builder. –nest’-egg an egg, real or
sham, left or put in a nest to encourage laying: something laid up as the
beginning of an accumulation: money saved; nest’ing-box a box set up
for birds to nest in; nest’ing-place. –feather one’s nest see feather.
[O.E. nest; Ger. Nest, L. nidus.]
(Chambers)
nest • n. 1 a structure or place made or chosen by a bird for laying eggs and
s
sheltering its young. a place where an animal or insect breeds or shelters.

s
a snug or secluded retreat. s a bowl-shaped object likened to a bird’s

nest. 2 a place filled with undesirable people or things: a nest of spies. 3 a set
of similar objects of graduated sizes, fitting together for storage. • v. 1 use
or build a nest. 2 fit (an object or objects) inside a larger one. 3 (especially
in computing and linguistics) place in a hierarchical arrangement, typically
in a lower position.
– DERIVATIVES nestful n. (pl.-fuls).
– ORIGIN OE nest, of Gmc origin.
nest box (also nesting box) • n. a box provided for a bird to nest in.
nest egg • n. 1 a sum of money saved for the future. 2 a real or artificial
egg left in a nest to induce hens to lay there.
nester • n. a bird nesting in a specified manner or place: a scarce nester in
Britain.
(COD10)
Although the exact typefaces are not reproduced, the entries are replicas in all
essential respects. Chambers represents an entry style that was reflected in the
COD up to the seventh edition (1982), while COD10 represents a double shift
from COD7, in the direction of a clearer microstructure and enhanced accessi-
bility. The major differences are the following:
• word classes marked by ‘•’ and a bold abbreviation (n, v) in COD10, com-
pared with an italic n, v.i. etc. in Chambers
• COD10 has no pronunciation transcription for ‘simple’ words like nest;
Chambers has a respelling system (see Chapter 9)
s
• senses are numbered in COD10, with subsenses marked by ‘ ’; whereas

Chambers divides senses with a colon (:)
• derivatives requiring their own definition are separate headwords in COD10,
while remaining nested under the root word in Chambers
• undefined derivatives are clearly marked in COD10
• etymology is on a separate line, marked by the word ‘ORIGIN’, in COD10;
while Chambers has etymology in the traditional square brackets at the end
of the entry.
Users and uses 81
Scholfield’s Step 4 is probably easier to complete in COD10 than in Chambers.
Even so, the user still has to complete Step 5, the task of ‘interpreting the
information given in the entry’ (Béjoint 2000: 156).
ECED attempts to overcome part of the problem of location of information
by including ‘quick definitions’ in bold capitals at the beginning of each sense of
polysemous lexemes, a technique similar to that used by some learners’ diction-
aries (Chapter 11). For example, the entry for ladder reads:
1 DEVICE WITH RUNGS TO CLIMB ON a portable piece of equip-

ment with rungs fixed to sides made of metal, wood, or rope, used for
climbing up or down 2 PATH TO ADVANCEMENT a series of hier-
archical levels on which somebody moves up or down within an organiza-
tion or society o She joined the firm at a fairly low level but quickly moved up the
ladder. 3 LINE OF MISSING STITCHES IN TIGHTS a vertical line
of stitches that have come undone in tights, a stocking, or a knitted gar-
ment, leaving only the horizontal stitches in place 4 LIST OF RANKED
PLAYERS a list of contestants in an ongoing sports or games competition,
arranged according to ability
The aim is for the user to run their eye down the entry, reading only the ‘quick
definitions’, in order to locate more easily the specific sense being looked for.
There are, though, conventions of dictionary microstructure that a user must
become familiar with: what a headword is, normally a transcription of pronun-
ciation, irregular inflections, division of the entry into word classes if relevant,
identification of different senses, the wording of definitions, the placement of
derivatives, phrases and idioms, etymology usually in square brackets at the end.
Then, it is the rare dictionary that does not utilise a degree of abbreviation, at
least for the word classes, and usually for the ‘transitive’ (v.t.) and ‘intransitive’
(v.i.) uses of verbs, as well the languages referred to in the etymology (O.E. for
‘Old English’, Ger. for ‘German’, L. for ‘Latin’, Gmc for ‘Germanic’). In the
Preface to the first edition of COD (1911), the editors, F.W. and H.G. Fowler,
aimed to save space ‘by the severest economy of expression – amounting to the
adoption of telegraphese – that readers can be expected to put up with’. Ninety
years later, dictionary users can be expected to put up with far less abbreviation
than their forebears, and in some dictionaries (e.g., NODE) abbreviation is
deliberately kept to a minimum.
Even with improvements in the macro- and microstructure of dictionaries in
recent years, to make information more readily accessible, the user must expect
to develop reference skills in order to become an efficient user of dictionaries.
The evidence from studies of dictionary use is that few users read the front
matter of their dictionary, which explains how the dictionary is structured,
how users may locate various items of information, what the various labels and
abbreviations mean, and what special features the dictionary contains. In general
terms, users must develop skills in finding the appropriate headwords – the base
form of an inflected word, where a compound or derived word is entered,
82 Users and uses
under which headword to look for a phrase or idiom. Also, users must be
familiar with entry structure, the division into word classes, the principles of
sense division, the wording of definitions, the contribution of examples, and so
on. Many have argued for training in dictionary use to be given in schools and
universities, rather than leaving it to chance and native wit (Hartmann 2001: 92,
Béjoint 2000: 168). But is there perhaps a fault in the dictionaries themselves?
7.4 Record or reference?

When we take the user’s perspective, issues of accessibility are naturally to the
fore, and the concentration is on the dictionary as a reference manual. But
‘dictionaries must be faithful recorders of the language’ (Chambers Preface). We
saw in Chapter 4 how dictionaries grew into the role of being a record of the
language, and in Chapter 5 how the OED compilers interpreted that role in
historical terms. In Chapter 3, we argued for the dictionary as a linguistic de-
scription complementary to the grammar, the latter dealing with the general
rules of sentence structure, and the former with the operation of individual
lexical items as well as with the overall structure of the vocabulary of a language.
Within its size and scope a dictionary is a lexical description. It specifies the
lexical items that the vocabulary contains; it aims to be comprehensive, in the
sense discussed in 7.1; it gives for each lexical item that it identifies a complete
lexical account – spelling, pronunciation, inflections, derivations, meanings,
usage, origin – that describes its idiosyncratic operation (Chapter 3). This is the
sense in which a dictionary is expected to be a ‘faithful recorder’ of the lan-
guage. The information that should be included is more or less given (Hudson
1988; Ilson 1991); how it should be presented is a matter of convention, tweaked
by imaginative development and innovation on the part of dictionary compil-
ers, editors and publishers.
The consequence of dictionaries’ recording function is that vast amounts of
information in any dictionary remain unconsulted by anyone except, perhaps,
the most assiduous dictionary scholar. Conventional native-speaker dictionaries
are attempting to perform simultaneously two functions that may be in conflict
with each other: to provide a lexical description, and to provide a source of
ready reference to satisfy the needs of various types of user. Is there an argument
for a range of different types of dictionary, to fulfil the different functions and
user needs, just as there is a range of different types of grammar book, from
reference to pedagogical?
The most popular dictionary in Germany is the Duden Rechtschreibung, now
in its twenty-second edition (2000). It is essentially a spelling dictionary, with
indications of alternative spellings, syllabification, and inflections; but also, for
less familiar words and phrases, it gives brief definitions, synonyms or examples
of usage, as well as pronunciation and language of origin for loanwords. So it
fulfils the two basic needs of dictionary users: a guide to spelling, and an expla-
nation of ‘hard’ words. No popular dictionary for English follows this format.
English has specialist spelling dictionaries (e.g. West 1964), and it has specialist
Users and uses 83
dictionaries for the terminology of subject disciplines and particular topics, from
archaeology to zoology; but it has no regular dictionary of the type represented
by the Duden.
ECED, on the basis of its user research (7.2) makes some provision for meet-
ing the spelling needs of its target users (mainly students). It includes common
misspellings of words in its alphabetical listing of headwords, but in grey type
and with a line through; the correct spelling is then given beside it, e.g.
preperation incorrect spelling of preparation
It also includes notes entitled ‘SPELLCHECK’ after words that are pronounced
the same (i.e. homophones), e.g. here and hear, as well as ‘usage’ notes for pairs
that are frequently confused, e.g. complement and compliment.
The dictionary in the electronic medium (see 6.7) offers the possibility of
allowing the user to choose which types of information about headwords will
be displayed on any lookup occasion. The OED2 on CD-ROM has built in a
number of options of this kind. The user can choose whether to have the
quotations displayed or not, and there are ‘switches’ to activate the display of
‘pronunciation’, ‘spellings’ (i.e. form history), and ‘etymology’. If all the options
are turned off, the display offers only the definitions and usage labels for the
numbered senses of a lexeme. The Oxford Talking Dictionary has a more limited
set of switches: the quotations can be excluded, and the thesaurus facility is
activated by a switch. Electronic dictionaries generally divide their display into
a headword list window and a dictionary entry window, and each can be scrolled
separately. The headword list window operates in part like a spelling dictionary,
except that derivatives, compounds and so on that are nested in dictionary
entries may not be included in the list. The electronic medium does, though,
open up possibilities for selective display of dictionary information that have not
yet been fully exploited.
7.5 Learners
There are two sets of users whose needs have been carefully considered and for
whom dictionaries have been specifically tailored: children, and learners. Chil-
dren’s dictionaries range from the large-format work with pictures and an im-
aginative use of colour, aimed at those just beginning to learn to read, to school
dictionaries that look like the adult version, except with a more selective head-
word list, the omission of some word senses and of information such as etymol-
ogy, and definitions written in a simpler language. We could examine more
closely the range of children’s dictionaries, but our focus here will be on dic-
tionaries aimed at learners of English as a foreign or second language. We discuss
these dictionaries in detail in Chapter 11; in the context of the present chapter,
we look just at the reference needs of this group.
A learner, or indeed a native speaker, may consult a dictionary when engaged
in one of two broad types of language task. On the one hand, a learner may be
84 Users and uses
engaged in the task of reading or listening, and they encounter a word or phrase
that makes no sense to them and whose meaning they cannot deduce from the
context: the dictionary is used as an aid to ‘decoding’ the item read or heard. On
the other hand, a learner may be engaged in the task of writing or preparing to
speak, and they do not necessarily need to find an unknown word, but rather to
discover how a known word may be used in the appropriate context: the dic-
tionary is used as an aid to ‘encoding’ acceptable sentences and texts. For a
native speaker, spelling is the main encoding purpose that they might consult a
dictionary for; whereas learners may need to find out about spelling, pronuncia-
tion, inflections, how a word fits into grammatical structure, what other words
can appropriately accompany it (its collocations), and whether there are any
social or cultural restrictions on its usage.
Dictionaries for learners, therefore, need to consider not only their decoding
needs, which are not vastly different from those of native speakers, but more
particularly their encoding needs. This means that learners’ dictionaries need to
contain more explicit, more comprehensive and more systematic information
about the syntactic and lexical operation of words than a dictionary for native
speakers. Arguably, this information should be contained in any dictionary that
purports to be an accurate and comprehensive lexical description (Hudson 1988),
but in practice it is precisely in these areas that in native speaker dictionaries the
information is scantiest and least systematic (see Chapter 9). Besides, as we
noted earlier, it is not just a matter of containing the information, it is also a
matter of presenting the information in such a way that it is readily accessible
and takes account of users’ prior knowledge and reference skills ( Jackson 1995).
The early learners’ dictionaries, such as H.E. Palmer’s A Grammar of English
Words (1938) or the precursor to the OALD, A.S. Hornby et al.’s Idiomatic and
Syntactic English Dictionary (1942), concentrated, as their titles suggest, on pro-
viding accurate and systematic information about the grammatical operation of
words. The subsequent history of learners’ dictionaries (Cowie 1999) shows the
development and elaboration of that purpose, not only in grammar but also in
phraseology and collocation, with an increasing attention more recently on
making the information more readily accessible and usable. This has been achieved
in a number of ways. For example, the early editions of OALD presented infor-
mation about the grammatical operation of verbs by means of a set of codes
based on a system of ‘verb patterns’ developed by Hornby. Each sense of a verb
was coded (e.g. ‘VP6, VP15, VP21’). The regular user of the dictionary would
learn which patterns were represented by the commonly occurring codes, and
could look the less familiar ones up in the guide to the dictionary in the front
matter. The coded information was usually supported by illustrative examples,
so that the user could see a typical context for the word in the given sense. It
became clear over time that, while language teachers may have made good use
of the grammatical codes, many student users of the dictionary did not make the
effort to learn the system and relied largely on the examples. In later editions of
OALD the coding system has been abandoned; the grammatical information is
presented in a more accessible form, e.g. by means of formulae such as ‘~sth (to
Users and uses 85
sb)’; and more attention is paid to ensuring that the examples provide a suitable
model.
As we shall see in Chapter 11, learners’ dictionaries have developed in more
varied ways than just in the presentation of grammatical information. Entry
structure has in some cases been substantially revised, information and advice on
usage is incorporated, various types of additional material is included. Much of
the innovation has been in response to the perceived needs of this particular user
group, so that the learner’s dictionary has in many respects moved away from
the conventions of the native speaker dictionary (Rundell 1998).
7.6 Further reading

It is useful to read the prefaces and other front matter to a number of diction-
aries, as well as the blurb on their dust jackets, to gain an idea of how dictionaries
are presenting themselves to their potential users. Chapters 4 and 5 of Henri
Béjoint’s Modern Lexicography (2000) discuss the aims of dictionaries, their func-
tions in society, and the reference needs of their users. A ‘user perspective’ is
also provided by Chapter 6 of Reinhard Hartmann’s Teaching and Researching
Lexicography (2001). Bo Svensén discusses dictionary users in relation to types of
dictionary in Chapter 2 of his Practical Lexicography (1993).
86 Meaning in dictionaries
In Chapter 7, we identified checking spelling and finding out about meaning as

the two principal reasons why someone would consult a dictionary. Because
dictionaries are based on the written form of the language and their word lists
are arranged in alphabetical order, they coincidentally and inevitably provide
information about spelling. It is, thus, in explaining, describing and defining the
meaning of words that the major function of dictionaries is considered to lie,
and on which they are judged. We discussed some of the components of word
meaning in Chapter 2, and in this chapter we explore how and with what
success dictionaries describe the various aspects of lexical meaning. First, though,
we need to determine exactly what the objects are whose meaning dictionaries
are attempting to characterise.
8.1 The objects of definition

Dictionaries present us with a list of headwords as the objects to be defined,
though some items within the entry under a headword may also be subject to
definition. The headword list may contain a variety of types of item. Consider
the two following short lists from CED4:
hook, hookah, hook and eye, hooked, hooker1, hooker2, Hooke’s law,
hooknose, hook-tip, hook-up, hookworm, hookworm disease;
its, it’s, itself, itsy-bitsy, ITU, ITV, -ity, i-type semiconductor, IU, IU(C)D,
Iulus, -ium
The headword list in CED4 (personal and geographical names are excluded)
contains ‘simple’ words like hook and hookah, but it also contains a range of other
items:
• derived words – hooked, hooker2

• compound words, including those written solid (hooknose, hookworm, itself),
those hyphenated (hook-tip, hook-up, itsy-bitsy), and open compounds (Hooke’s
law), as well as combinations of these (hookworm disease, i-type semiconductor)
• binomials (hook and eye)
Meaning in dictionaries 87
• abbreviations (ITU, ITV, IU, IU(C)D), whose ‘definition’ consists only in
spelling out the words whose initial letters make up the acronym or initialism
• affixes (-ity, -ium), which have similar definitions to those for words
• contractions (it’s), for which only the full forms are given.
Some headwords – abbreviations, contractions – are not included for definition

as such, just for explanation; similarly, word forms manifesting irregular inflec-
tions are often entered with a cross-reference to the base form (or lemma), e.g.
felt the past tense and past participle of feel
Some of the items that CED4 includes in its headword list would feature as
nested or run-on items in other dictionaries, including derivatives and some
compounds. The practice varies, with, for example, some dictionaries listing
solid compounds as headwords, but not hyphenated or open compounds. CED4
itself includes some items within entries that are further defined. Under hook,
for example, it defines: by hook or (by) crook; get the hook; hook, line, and
sinker; off the hook; on the hook; sling one’s hook; hook it. These are various
kinds of phrase, including idioms (by hook or crook, sling one’s hook), trinomials
(hook, line, and sinker), and slang expressions (get the hook, hook it).
One item in the list is entered twice: hooker, with the same pronunciation,
and with the following main meanings:
hooker1 a commercial fishing boat using hooks and lines instead of nets
hooker2 a person or thing that hooks
At first glance, you would think that the meaning of hooker1 would be included
in the more general meaning of hooker2. Why then does hooker have two homo-
graphs (in fact, homonyms), which are entered as separate headwords, especially
as the meanings seem so close? The basic criterion that dictionaries use to iden-
tify homographs is etymology: if two or more different origins can be identified
for the same spelling, then the orthographic word is entered as many times as
there are different etymologies. In the case of hooker, the first homograph is
identified as a loanword from Dutch hoeker in the seventeenth century, and the
second is the derivation by means of the suffix -er from the verb hook, which has
its origin in the Old English hoc.
With hooker, the homographs have not too dissimilar meanings. The opposite
case can also be found. Consider the following meanings for the word table
(definitions from CED4):
a flat horizontal slab or board, usually supported by one or more legs, on

which objects may be placed
an arrangement of words, numbers, or signs, usually in parallel columns, to
display data or relations.
Intuitively, you might think that these meanings must be associated with words
of different origin, but that is not the case. There is a single entry for table in
CED4, since both meanings are associated with the word that came into English
from Old French in the twelfth century, with its origin in Latin tabula, ‘a writing
tablet’.
Another criterion that may be used to trigger multiple headwords for the
same spelling is word class membership. This criterion operates in LDEL, where
table has three entries, one each for the noun, adjective (e.g. table manners), and
verb. The criterion operates alongside the etymological one, so that a homo-
graph identified by etymology may also have multiple entries on the basis of
word class membership. For example, line is entered four times in LDEL2: 1line
(put a lining in, for example, a garment) is distinguished from the others on the
basis of etymology: it has its origin in Middle English linen, derived from lin, the
Old English for ‘flax’, which developed into modern English linen. The other
entries for line also have a Middle English origin, but from Old French ligne,
though this word, interestingly, goes back to a Latin word meaning ‘made of
flax’. This line has three homographs based on word class membership, one each
for the noun, verb and adjective.
Many words are polysemous; they have more than one meaning, as table cited
earlier. For any spelling (orthographic word), therefore, for which a lexico-
grapher identifies multiple meanings, a decision must be made whether the
different meanings arise from polysemy or because there are homographs. The
lexicographer applies the criterion of etymology, and, according to dictionary
policy, that of word class membership. If the criteria are satisfied, then multiple
headwords are entered in the dictionary. If not, then a single headword is en-
tered with multiple meanings or senses. We shall see (Chapter 11) that these
criteria do not necessarily apply in learners’ dictionaries, because they may be
regarded as not serving the reference needs of this user group.
8.2 Lumping and splitting

If polysemy is identified, how does a lexicographer decide how many meanings
or senses of a word to recognise? The lexicographer collects the evidence, such
as citations and concordance lines (Chapter 13), which indicate the different
contexts of use. What is then done with the evidence depends on whether the
lexicographer is a ‘lumper’ or a ‘splitter’ (Allen 1999: 61):
The ‘lumpers’ like to lump meanings together and leave the user to
extract the nuance of meaning that corresponds to a particular context,
whereas the ‘splitters’ prefer to enumerate differences of meaning in
more detail; the distinction corresponds to that between summarizing
and analysing.
Here are the entries for the noun horse from NODE and from CED4: one of
these is a ‘splitting’ and the other a ‘lumping’ dictionary.
s
horse noun 1 a solid-hoofed plant-eating domesticated mammal with
a flowing mane and tail, used for riding, racing, and to carry and pull loads.
• Equus caballus, family Equidae (the horse family), descended from the
wild Przewalski’s horse. The horse family also includes the asses and zebras.
n an adult male horse; a stallion or gelding. n a wild mammal of the horse
family. n [treated as sing. or pl.] cavalry: forty horse and sixty foot.
2 a frame or structure on which something is mounted or supported, espe-
cially a sawhorse.
ν Nautical a horizontal bar, rail, or rope in the rigging of a sailing ship for
supporting something. ν short for VAULTING HORSE
3 [mass noun] informal heroin.
4 informal a unit of horsepower: the huge 63-horse 701-cc engine.
5 Mining an obstruction in a vein.
(NODE)
horse n 1 a domesticated perissodactyl mammal, Equus caballus, used for

draught work and riding: family Equidae. 2 the adult male of this species;
stallion. 3 wild horse. 3a a horse (Equus caballus) that has become feral. 3b
another name for Przewalski’s horse. 4a any other member of the family
Equidae, such as the zebra or ass. 4b (as modifier): the horse family. 5 ( function-
ing as pl) horsemen, esp. cavalry: a regiment of horse. 6 Also called: buck.
Gymnastics. a padded apparatus on legs, used for vaulting, etc. 7 a narrow
board supported by a pair of legs at each end, used as a frame for sawing or
as a trestle, barrier, etc. 8 a contrivance on which a person may ride and
exercise. 9 a slang word for heroin. 10 Mining. a mass of rock within a vein
or ore. 11 Nautical. a rod, rope, or cable, fixed at the ends, along which
something may slide by means of a thimble, shackle, or other fitting; trav-
eller. 12 Chess. an informal name for knight. 13 Informal. short for horse-
power. 14 (modifier) drawn by a horse or horses: a horse cart.
(CED4)
NODE has five numbered senses, by comparison with CED4’s fourteen. The
first sense in NODE encompasses the first five senses in CED4; NODE’s sec-
ond sense encompasses senses six, seven and eleven in CED4; NODE’s 3 cor-
responds to CED4’s 9, NODE’s 4 to CED4’s 13, and NODE’s 5 to CED4’s 10.
CED4 has some senses not covered by NODE: 8, 12, 14. Some of CED4’s
senses are ‘subsenses’ (introduced by the symbol n) to the ‘core’ senses in NODE.
The arrangement of NODE suggests that it is essentially a ‘lumping’ dictionary,
whereas CED4 falls more obviously into the ‘splitting’ category. Most diction-
aries tend to be of the ‘splitting’ type, though different dictionaries do not
necessarily agree on where to make the splits between senses. Compare the
following entries for the noun length in COD9 and CCD4:
length n. 1 measurement or extent from end to end; the greater of two or
the greatest of three dimensions of a body. 2 extent in, of, or with regard to,
time (a stay of some length; the length of a speech). 3 the distance a thing extends
(at arm’s length; ships a cable’s length apart). 4 the length of a swimming pool
as a measure of the distance swum. 5 the length of a horse, boat, etc., as a
measure of the lead in a race. 6 a long stretch or extent (a length of hair). 7 a
degree of thoroughness in action (went to great lengths; prepared to go to any
length). 8 a piece of material of a certain length (a length of cloth). 9 Prosody
the quantity of a vowel or syllable. 10 Cricket a the distance from the bats-
man at which the ball pitches (the bowler keeps a good length). b the proper
amount of this. 11 the extent of a garment in a vertical direction when
worn. 12 the full extent of one’s body.
(COD9)
length n 1 the linear extent or measurement of something from end to end,

usually being the longest dimension. 2 the extent of something from begin-
ning to end, measured in more or less regular units or intervals: the book was
600 pages in length. 3 a specified distance, esp. between two positions: the
length of a race. 4 a period of time, as between specified limits or moments.
5 a piece or section of something narrow and long: a length of tubing. 6 the
quality, state, or fact of being long rather than short. 7 (usually pl ) the
amount of trouble taken in pursuing or achieving something (esp. in to
great lengths). 8 (often pl ) the extreme or limit of action (esp. in to any
length (s) ). 9 Prosody, phonetics. the metrical quantity or temporal duration
of a vowel or syllable. 10 the distance from one end of a rectangular swim-
ming bath to the other. 11 NZ inf. the general idea; the main purpose.
(CCD4)
Table 8.1 shows how the senses match in these two concise dictionaries.
COD9 has only half of its twelve meanings directly matched in CCD4, while
CCD4 has seven of its eleven senses matched in COD9; the disparity arises from
the fact that Sense 7 in COD9 is matched to two senses (7 and 8) in CCD4.
There is one sense that looks closely related but is not a direct match: Sense 6 in
COD9 and Sense 5 in CCD4. The two dictionaries have not carved up the
meaning of length in the same way, and there are senses in each that do not have
counterparts in the other (the gaps in Table 8.1).
Are there any criteria, or rules of thumb, that lexicographers use in deciding
what senses to recognise in analysing the meaning of a lexeme? Context, clearly,
plays a part, but context can be analysed more or less finely. For example, the
first four senses of the noun interest in CED4 are given the following definitions:
1 the sense of curiosity about or concern with something or someone

2 the power of stimulating such a sense
3 the quality of such stimulation
4 something in which one is interested; a hobby or pursuit.
Table 8.1
COD9 CCD4 CCD4 COD9
1 1 1 1
2 4 2
3 3 3 3
4 10 4 2
5 5
6 6
7 7, 8 7 7
8 8 7
9 9 9 9
10 10 4
11 11
12
LDEL2 encapsulates these meanings under one numbered sense with two parts:
5a readiness to be concerned with, moved by, or have one’s attention

attracted by something; curiosity 5b the quality in a thing that arouses
interest . . . also something one finds interesting
Clearly, in the end, it comes down to the lexicographer exercising their in-
formed judgement in the face of the evidence that they have to work with.
There are, however, two factors that a lexicographer can take into account
when distinguishing the senses of words: grammar, and collocation (Clear 1996).
It is possible that the use of length ‘usually plural’ as against ‘often plural’ led the
CCD4 lexicographer to distinguish senses 7 and 8 (see above). The fact that
reply can be used intransitively, as against transitively with a clause as object,
distinguishes the first two senses in CED4:
1 to make an answer (to) in words or writing or by an action; respond: he

replied with an unexpected move. 2 (tr; takes a clause as object) to say (something)
in answer: he replied that he didn’t want to come.
Some senses have a specialised or restricted use, and are labelled as such, e.g. the
Gymnastics, Mining, Nautical, Chess and Informal uses of horse in the CED4
entry given earlier. Some senses enter into particular collocations; as indicated
by the words in brackets in these senses for isometric from CED4:
(of a crystal or system of crystallization) having three mutually perpendicu-

lar equal axes
(of a method of projecting a drawing in three dimensions) having the three
axes equally inclined and all lines drawn to scale.
However, collocation has not been exploited as much as it could be by lexico-
graphers for this purpose (Clear 1996).
Having decided what senses to recognise for a polysemous lexeme, the lexi-
cographer must then decide how to order them in the dictionary entry. In an
historical dictionary – OED or SOED – the order is given: from earliest sense to
latest sense, according to the citational evidence. However, even in historical
dictionaries, things are not always as simple: some words have a complicated
‘sense history’, with more than one ‘branch’ (see Berg 1993 for a description of
the entry structure in OED2). In general-purpose dictionaries, the practice
varies. Chambers follows the historical order:
There are at least two possible ways of ordering . . . definitions. One way is
to give the most modern meaning first and the oldest last. The other is the
way selected for this dictionary, historical order. In this method the original
or oldest meaning of the word is given first and the most modern or up-to-
date last. Both methods are equally easy to use but historical order is per-
haps more logical since it shows at a glance the historical development of
the word, each entry providing a potted history of the word.
(p. vi)
This is a matter of opinion, and whether anyone would consult a general-

purpose dictionary for a ‘potted history’ of a word is debatable. The alternative
argument would focus on likely user needs, which would privilege the more
modern senses. In fact, dictionaries that follow the ‘modern meaning first’ prin-
ciple are usually rather more subtle in their arrangement of senses, e.g. CED4:
As a general rule, where a headword has more than one sense, the first sense
given is the one most common in current usage. Where the editors con-
sider that a current sense is the ‘core meaning’ in that it illuminates the
meaning of other senses, the core meaning may be placed first. Subsequent
senses are arranged so as to give a coherent account of the meaning of a
headword . . . closely related senses are grouped together; technical senses
usually follow general senses; archaic and obsolete senses follow technical
senses; idioms and fixed phrases are usually placed last.
(p. xxi)
LDEL2 attempts a synthesis of the historical and contemporary approaches:
Meanings are ordered according to a system which aims both to show the
main historical development of the word and to give a coherent overview
of the relationship between its meanings. Meanings that are current through-
out the English-speaking world are shown first; they appear in the order in
which they are first recorded in English, except that closely related senses
may be grouped together regardless of strict historical order. They are fol-
lowed by words [sic – presumably for ‘meanings’] whose usage is restricted,
such as those current only in informal use or in American English. Senses
which have become archaic or obsolete are shown last.
(p. xvi)
Compare the order of senses in the entries for the noun mate from CED4 and
LDEL2:
mate 1 the sexual partner of an animal. 2 a marriage partner. 3a Informal,

chiefly Brit., Austral., and N.Z. a friend, usually of the same sex: often used
between males in direct address. 3b (in combination) an associate, colleague,
fellow sharer, etc.: a classmate, a flatmate. 4 one of a pair of matching items.
5 Nautical. 5a short for first mate. 5b any officer below the master on a
commercial ship. 5c a warrant officer’s assistant on a ship. 6 (in some trades)
an assistant: a plumber’s mate. 7 Archaic. a suitable associate.
(CED4)
mate 1a an associate, companion – often in combination <flatmate> <play-

mate> 1b an assistant to a more skilled workman <plumber’s ~> 1c chiefly Br
& Austr a friend, chum – used esp as a familiar form of address between men
2 an officer on a merchant ship ranking below the captain 3 either member
of a breeding pair of animals <a sparrow and his ~> 3c either of two matched
objects <a ~ to this glove> 4 archaic a match, peer
(LDEL2)
Mate entered the language during the Middle English period, with the general
‘companion’ meaning. The LDEL2 entry reflects the later (sixteenth-century)
addition of the ‘sexual partner’ meaning, though this comes first in CED4
because it is considered the more common and central meaning in modern
English.
8.3 Definitions
Once identified, each sense needs a definition. The definition is a characterisa-
tion of the meaning of the (sense of the) lexeme; it is not an exhaustive expla-
nation of the possible referents (Zgusta 1971: 252ff.). Like other linguistic
statements, definitions in monolingual dictionaries consist of ‘language turned
back on itself ’, using the same language to describe as is being described. Much
of the art of lexicography (compare the title of Landau 1989, 2001) consists in
finding apt wording for constructing telling definitions. A number of general
principles can be identified:
• a word should be defined in terms simpler than itself (Zgusta 1971: 257),
which is not always possible with the ‘simple’ words
• circularity of definition should be avoided, i.e. defining two or more lexemes
in terms of each other (Svensén 1993: 126)
• a definition should be substitutable for the item being defined; so the head
of the definition phrase should belong to the same word class as the defined
lexeme (Zgusta 1971: 258; Svensén 1993: 127)
• different forms of definition are appropriate to different types of word (Zgusta
1971: 258).
The most common form of definition is the ‘endocentric phrase’ (Zgusta 1971:
258), the ‘completely analytical one-phrase definition’ (Preface to W3), which
consists of ‘stating the superordinate concept next to the definiendum (genus
proximum) together with at least one distinctive feature typical of the definiendum
(differentia specifica)’ (Svensén 1993: 122). A good example of such a definition is
that given for the first sense of horse in NODE, cited earlier:
a solid-hoofed plant-eating domesticated mammal with a flowing mane

and tail, used for riding, racing, and to carry and pull loads.
The ‘definiendum’ (horse) is related to its ‘genus’ (mammal ), i.e. its ‘superordinate
concept’, and given a number of ‘differentiae’ (solid-hoofed, plant-eating, domest-
icated, with a flowing mane and tail, used for riding, etc.), which are ‘typical features’
serving to distinguish this mammal from other mammals.
The ‘genus + differentiae’ style of definition, as it is sometimes called, is used
for a great many words from most of the word classes, with ‘differentiae’ appro-
priate to whether the meaning is concrete or abstract, referring to a thing, event,
quality, and so on. Here are some further examples from a range of types of
word, taken from a number of dictionaries (the ‘genus’ is in each case in italics):
beat (verb) to strike with or as if with a series of violent blows [CED4]

clean (adjective) free from dirt, stain, or whatever defiles [Chambers]
glamour (noun) a romantic, exciting, and often illusory attractiveness
[LDEL2]
humble (adjective) of low social or political rank [COD9]
somewhat (adverb) to a moderate extent or by a moderate amount [NODE]
see (verb) perceive with the eyes [COD10]
variety (noun) the quality or condition of being diversified or various [CED4]
A second major type of definition consists of a synonym, a collection of synonyms,

or a synonymous phrase. Many, especially abstract, words are not easily defined
analytically by the ‘genus + differentiae’ style; and lexicographers resort to the
use of synonyms. It is this type of definition that is most likely to create circu-
larity, where a set of synonyms is used to define each other. Smaller dictionaries,
where space is more limited, use synonymy as a defining method more exten-
sively. Compare these entries from the Collins Pocket English Dictionary (2000):
miserable 1 very unhappy, wretched. 2 causing misery. 3 squalid. 4 mean
unhappy 1 sad or depressed. 2 unfortunate or wretched
wretched 1 miserable or unhappy. 2 worthless
The larger Collins Concise (1999) is already an improvement; although it still

relies largely on synonymy for defining, its more extensive treatment creates less
circularity:
miserable 1 unhappy or depressed; wretched. 2 causing misery, discom-

fort, etc. 3 contemptible. 4 sordid or squalid. 5 mean; stingy.
unhappy 1 not joyful; sad or depressed. 2 unfortunate or wretched. 3
tactless or inappropriate.
wretched 1 in poor or pitiful circumstances. 2 characterised by or causing
misery. 3 despicable; base. 4 poor, inferior, or paltry.
Interestingly, the parent, desk-size CED4 adds almost nothing to the definitions
of the Concise for these words.
A third style of definition specifies what is ‘typical’ of the referent. This style
is normally used in combination with one of the others, usually the analytical
style, and is introduced by the adverb typically. Here are some examples from
COD10:
day of rest a day set aside from normal activity, typically Sunday on reli-
gious grounds
gingham lightweight plain-woven cotton cloth, typically checked
measles an infectious viral disease causing fever and a red rash, typically
occurring in childhood
scramble move or make one’s way quickly and awkwardly, typically by
using one’s hands as well as one’s feet
ululate howl or wail, typically to express grief
The last example adds a typifying definition to a synonym one, while the others
add it to an analytical definition.
A fourth type of definition explains the ‘use’ to which a word or sense of
word is put, usually in the grammar of the language. This type is typically
employed for defining ‘grammatical’ or ‘function’ words (determiners, pro-
nouns, conjunctions, prepositions, auxiliary verbs – see Chapter 1), especially
where these have no reference outside of language. Here are some examples
from COD10:
and (conjunction) used to connect words of the same part of speech, clauses,
or sentences
do (auxiliary verb) used before a verb in questions and negative statements
ever (adverb) used for emphasis in questions expressing astonishment or
outrage
herself (pronoun) used as the object of a verb or preposition to refer to a
female person or animal previously mentioned as the subject of the clause
that (pronoun/determiner) used to identify a specific person or thing
observed or heard by the speaker
us (pronoun) used by a speaker to refer to himself or herself and one or
more others as the object of a verb or preposition
All these definitions are introduced by used, and they are mostly framed in terms
of how the word operates in the grammatical structure of the language. In the
case of the adverb ever, though, the ‘use’ relates to its function in discourse, i.e.
for emphasis.
Definitions aim to describe the reference relations (Chapter 2) of a lexeme,
specifically their denotations. They do not usually comment on the connotative
or associative meaning of a lexeme, though this may occasionally find mention,
as in the definitions of champagne and youth in NODE:
champagne a white sparkling wine from Champagne, regarded as a sym-

bol of luxury and associated with celebration
youth the state or quality of being young, especially as associated with
vigour, freshness, or immaturity.
More often, connotation is indicated by appropriate labelling (see Chapter 9), as

for the following words in COD9:
crony (often derog[atory] ) a close friend or companion

ethnic cleansing (euphem[istic] ) the mass expulsion or extermination of
people from opposing ethnic or religious groups within a certain area
ladyship (iron[ical] ) a form of reference or address to a woman thought to
be giving herself airs
missive ( joc[ular] ) a letter, esp. a long and serious one
wrinkly (slang offens[ive] ) an old or middle-aged person.
8.4 Sense relations

In Chapter 2, we identified the ‘sense relations’ that may hold between lexemes
within the vocabulary as: synonymy, antonymy, hyponymy, and meronymy. In
this section, we shall explore how these meaning relations are represented in
dictionaries. We have noted already (8.3) that (loose) synonymy is used as a
defining style for some words. A somewhat tighter synonymy is sometimes
indicated by the phrase ‘also called’, when an alternative term is given for the
headword under consideration. For example, both CED4 and NODE give
‘also called: viper’ under adder, and ‘also called: hydrophobia’ under rabies.
But there is no consistency of treatment. For example, in CED4 hookah is given
the following alternatives: hubble-bubble, kalian, narghile, water pipe; and their
definitions all contain the phrase ‘another name for hookah’. NODE, how-
ever, does not give these alternatives in its entry for hookah, though hubble-bubble
is defined simply as ‘a hookah’, narghile has ‘a hookah’ included in its definition,
and water pipe has a similar definition to that for hookah but without making the
connection. Kalian is not entered in NODE. CED4 is perhaps particularly com-
mendable for making these synonym connections, as the following examples
from a single column in the dictionary show:
love apple an archaic name for tomato

lovebird another name for budgerigar
love feast Also called: agape
love-in-a-mist See also fennelflower
love-in-idleness another name for the wild pansy
love knot Also called: lover’s knot
lovemaking an archaic word for courtship
lovey Brit. informal. another word for love (sense 11).
Another way of treating synonyms in dictionaries is to draw together near-

synonyms under one of the items and discuss them. This procedure is used
by LDEL2 and by ECED. The latter has the following account of ‘generous’
words:
SYNONYMS generous, magnanimous, munificent, bountiful, liberal

CORE MEANING: giving readily to others
generous willing to give money, help, or time freely; magnanimous
very generous, kind, or forgiving; munificent very generous, especially on
a grand scale; bountiful (literary) generous, particularly to less fortunate
people; liberal free with money, time, or other assets.
The sense relation of antonymy is sometimes used in definitions, when the

opposite of the (sense of the) lexeme being defined is preceded by not, e.g. (from
COD9)
artificial not real
conventional not spontaneous or sincere or original
long-standing not recent
vacant not filled or occupied
Sometimes, an antonym may be indicated more explicitly. NODE, for exam-

ple, introduces antonyms with the phrase ‘the opposite of ’, but this is for a
limited number of mostly quite technical terms, e.g. anode – cathode, holism –
atomism, sinistral – dextral, zenith – nadir. CED4 uses the phrase ‘compare’ to
fulfil a similar function, but again with a small number of fairly technical terms.
Antonymy is not a well-represented sense relation in the text of dictionaries.
Hyponymy is better represented, largely because the analytical definition
(8.3) is formed using the hyponymy relation. The ‘genus’ term is, or should be,
the superordinate of the lexeme being defined, the ‘definiendum’. Consider the
following definition from CED4:
serge a twill-weave woollen or worsted fabric used for clothing
Serge is a hyponym of fabric, the ‘genus’ term in this definition, and it can be
related as a co-hyponym to other words that have fabric as their ‘genus’, such as
corduroy, lace, velvet, worsted and so on. What you cannot find out from a conven-
tional dictionary is the set of all the co-hyponyms of a particular superordinate
term (see Chapter 12). If a dictionary is consistent, though, co-hyponyms should
be related to the same genus term. But dictionaries are not noted for their
consistency in such matters. For example, NODE relates fork and spoon to the
genus term implement, but knife is related to instrument. NODE defines handwrit-
ten as ‘written with a pen, pencil, or other hand-held implement’, but pen and
pencil both have instrument as their genus term.
Meronymy, the ‘part-of ’ relation, is a less well recognised as well as a less
pervasive sense relation. It is, though, used in the definitions of some lexemes,
e.g. (from COD10):
algebra the part of mathematics in which . . .

coast the part of the land adjoining or near the sea
loin the part of the body on both sides of the spine between the lowest ribs
and the hip bones
vamp the upper front part of a boot or shoe
Again, we should not look for consistency in conventional dictionaries. While

upper is defined in COD10 as ‘the part of a boot or shoe above the sole’, sole has
a quite different type of definition: ‘the section forming the underside of a piece
of footwear’.
The dictionary that has most consistently treated sense relations is the learn-
ers’ dictionary COBUILD1, where synonyms, antonyms and superordinate terms
are indicated in the dictionary’s ‘extra column’ (further in Chapter 11).
8.5 Phraseology
The other major component of meaning that we identified in Chapter 2 was
collocation, the regular or particular company that a word keeps. We noted in
8.2 that collocation may offer a method for distinguishing the senses of a lexeme
(cf. Clear 1996). Collocation is, in the end, a matter of statistical frequency of
co-occurrence, and lexicologists have not yet collected full data on the
collocational behaviour of words. Where dictionaries note collocation, it is in
cases either of a known restriction to the range of a word or where a collocation
appears in a particular context. The possible collocates or the restrictions are
usually contained within brackets before the definition and introduced by ‘of ’
or ‘especially of’. Here are some examples from NODE:
bijou (especially of a house or flat) small and elegant

bifacial Botany (of a leaf ) having upper and lower surfaces that are struc-
turally different. Archaeology (of a flint or other artefact) worked on both
surfaces
convoluted (especially of an argument, story, or sentence) extremely com-
plex and difficult to follow
meander (of a river or road) follow a winding course. (of a person) wander
at random. (of a speaker or text) proceed aimlessly or with little purpose
teem (of water, especially rain) pour down; fall heavily
terrigenous Geology (of a marine deposit) made of material eroded from
the land.
Most of the lexemes for which collocates are indicated belong to the adjective
word class: the collocates specify the nouns or types of noun they typically
accompany. Some verbs (e.g. meander, teem) may have their typical Subject or
Object noun collocates specified. Collocation is the subject of considerable
research currently, especially following the development of extensive computer
corpora that promise to yield interesting and reliable data on this topic. Lexico-
graphers of learners’ dictionaries have begun to include some of this information
in their works, since it is an area of particular interest and difficulty for learners
of English as a second or foreign language (see Chapter 11).
Another area of interest to learners is that of idioms and other fixed expres-
sions, especially where the meaning of the expression cannot be deduced from
the meanings of its individual words. Some dictionaries, as we saw in 8.1, list
binomials, and perhaps trinomials, as headwords. COD9, for example, has some
120 such items as headwords, e.g. bells and whistles, flotsam and jetsam, sweet and
sour, waifs and strays. More difficult to locate are idioms proper, which are nor-
mally entered under one of the ‘main’ words of the idiom. Many dictionaries
are not very explicit about the rules for finding an idiom, though the rule of
thumb is that it will be under the first ‘main’ word. For example, a storm in a
teacup will be under storm, shoot one’s mouth off will be under shoot, but take a bull
by the horns will be under bull, because take does not count as a ‘main’ word.
Sometimes the rules are more complicated; LDEL2, for example, follows an
older tradition:
An idiom is entered at the first noun it contains; hence on the ball appears
at ball and in spite of at spite. If it contains no noun, it is entered at the
first adjective; hence give as good as one gets is shown at good. If it
contains no adjective, it is entered under the first adverb; if no adverb,
under the first verb; if no verb, under the first word. In any case, cross-
references to the entry where the idiom appears are given at the entries for
other major words in it: hand . . . – see also take the LAW into one’s
own hands. The entry where the idiom appears is shown in SMALL
CAPITAL letters.
(p. xiv)
Not all dictionaries are as good about cross-referencing, and locating an idiom
can turn into something of a hunt at times, especially as they are usually nested
towards the end of an entry. Some of the more modern layouts do make the
hunt easier, e.g. in COD10, which has a separate paragraph marked ‘PHRASES’
where this is relevant in an entry. Under shoot, for example, the following
phrases are listed:
shoot the breeze (or the bull), shoot one’s cuffs, shoot oneself in the foot,
shoot a line, shoot one’s mouth off, the whole shooting match, shoot through.
Each phrase is then given a definition, and any appropriate restrictive label.
Summarising, the treatment of meaning in dictionaries goes beyond simply
the definition; it includes the distinction of homographs, the identification of
senses and their ordering, the contribution of the sense relations, the incorpora-
tion of collocational information, and the consideration of idioms and other
phrasal expressions.
8.6 Further reading

Sidney Landau deals with ‘definition’ in Chapter 4 of Dictionaries: The Art and
Craft of Lexicography (1989), as does Bo Svensén in Chapter 10 of his Practical
Lexicography (1993). The section on defining styles owes something to Barbara
Kipfer’s treatment in Chapter 6 of Workbook on Lexicography (1984), where she
also discusses the ordering of senses.
Beyond definition 101
9 Beyond definition
In the previous chapter we discussed the treatment of what is often considered

the main function of dictionaries: the description of word meaning. In this
chapter, we investigate some of the other information about words that diction-
aries may contain, some of the ‘facts about words’ that we outlined in Chapter
2. While we shall look at topics such as spelling, pronunciation, inflections,
word classes, and usage, we shall leave etymology until the next chapter.
9.1 Spelling
As we have noted before, dictionaries cannot help but give information about
spelling, since as alphabetically organised word books they are founded on the
written form of words. Consulting the dictionary to check the spelling of words
we also found to be one of the major occasions of their use (Chapter 7). While
headwords, or nested derivatives, supply information about the usual spellings
of words, there is additional information, about variations in spelling, that dic-
tionaries also give. The variation can be of various kinds.
Some words simply have alternative spellings, where the choice of one rather
than the other is purely a matter of personal preference. Both spellings are
equally acceptable. Here are some examples (from COD9):
absorption – absorbtion, baptistery – baptistry, caddie – caddy, diffuser –

diffusor, extrovert – extravert, filigree – filagree, gizmo – gismo, horsy –
horsey, judgement – judgment, movable – moveable, neurone – neuron,
pendent – pendant, regime – régime, smidgen – smidgeon – smidgin, tran-
quillity – tranquility, yogurt – yoghurt.
A surprisingly large number of words have alternative spellings, and from this
list we can observe some possible patterns: final -ie or -y, suffix -er or -or, z or s,
possible loss of e after dg or v plus suffix, loss of accent from vowels of words
borrowed from French, and so on.
Many British dictionaries take account of the differences between British and
American spelling. CED4, for example, enters the American spelling of words
like center and pediatrics at the appropriate place in the headword list, and then
102 Beyond definition
gives a cross-reference to the British spelling. For words like savior and theater,
which would occur close to the British spelling, the American alternative is
simply given under the British spelling. There are two further spelling variations
that are often seen as differences between British and American English: the
ae – e alternation in aesthetics – esthetics, and the s – z alternation in -ise/-ize (e.g.
marginalise/-ize). The -ise/-ize alternation is no longer regarded as a British/
American difference; British dictionaries merely note these as alternative spell-
ings. The ae – e alternation is not yet fully accepted in British spelling. In most
dictionaries, with the exception of Chambers, encyclopedia is entered as the main
spelling, with encyclopaedia as the alternative; similarly with medieval and media-
eval. However, archaeology is the main spelling (or sole spelling – Chambers,
LDEL); archeology is given as an alternative in CED4, and is marked as American
in NODE and other Oxford dictionaries. And in the case of aesthetics, paediatrics,
etc. the alternative is usually marked as American.
One other area where dictionaries pay attention to spelling is where altera-
tions occur as a consequence of adding an inflectional suffix, such as cry – cried,
big – bigger. We will consider this in 9.3, where we discuss dictionary informa-
tion about inflections.
9.2 Pronunciation
How a word is pronounced is one of its idiosyncractic facts; it is the phonologi-
cal counterpart of spelling (orthography), its shape in the medium of sound as
against its shape in the medium of writing. We would expect, therefore, that
dictionaries would indicate at least the sounds that constitute the pronunciation
of the word, and for words of more than one syllable the stress pattern. There
are two issues in relation to pronunciation in dictionaries: first, how pronuncia-
tion is represented in the written medium that the dictionary uses, i.e. the
transcription system; and second, the model that is used for pronunciation, and
how much variation is indicated.
In most modern British dictionaries, the transcription system used to repre-
sent pronunciation is the International Phonetic Alphabet (IPA), developed
in the late nineteenth century as a system, based on the Roman alphabet, that
could be used for transcribing the speech of any language, and as an aid in
learning the pronunciation of a foreign language. The alternative transcription
to the IPA is a ‘respelling’ system. When James Murray was devising a transcrip-
tion system for the OED in the mid-nineteenth century, the IPA had not yet
been invented, and he developed a respelling system. However, when the sec-
ond edition of the OED was put together, the only wholesale revision was to
replace Murray’s respellings with IPA transcriptions. Other Oxford dictionaries
followed suit: COD7 (1982) had respelling, COD8 (1990) changed to the IPA.
LDEL uses respelling; so does ECED and Chambers, but Chambers 21st Century
Dictionary uses IPA, as do the Collins dictionaries. American dictionaries, how-
ever, usually use a respelling system.
Both transcription systems have the aim of a one-to-one correspondence
between sound and symbol, and unique representation of each sound. In the
Table 9.1
Chambers LDEL2 CED4 COD9
binary bcsnfr-i sbienfri sbainfri sbkinfri

creation krb-as-shfn krisaysh(f)n kri:seinfn kri:sein(f)n
genuflect jense-flekt sjenyoo,flekt shvnjg,flvkt shvnjgflvkt
orphan örsffn sawf(f)n sl:ffn sl:f(f)n
Thursday thûrzsdi sthuhzdi smf:zdi smf:zdi
case of the IPA, because it uses symbols additional to those in the Roman
alphabet, it mostly uses a single symbol to represent each sound. A respelling
system, restricting itself to the symbols of the Roman alphabet, perhaps with the
addition of the ‘schwa’ symbol/f/, needs to use digraphs and even trigraphs in
order to achieve a unique one-to-one correspondence. Table 9.1 shows some
examples of transcription from a variety of dictionaries.
The argument used in favour of respelling is that it uses mostly familiar sym-
bols (Paikeday 1993), whereas the IPA employs a considerable number of sym-
bols that are not contained in the Roman alphabet. On the other hand, a respelling
system either has to use diacritics, as in the Chambers version, or a large number
of digraphs, as in the LDEL system (e.g. oo, aw, uh). Arguably, any transcrip-
tion system will constitute a learning task for the user who needs to consult it,
or at least the ability to interpret the table where the transcription is described
and illustrated. Some dictionaries provide reminders of the symbols at the
bottom of each page, e.g. COD9, with vowels on one double-page and con-
sonants on the next. Some CD-ROM versions of dictionaries provide a
recorded pronunciation of each transcription contained in the dictionary
(e.g. COD9).
Pronunciation is not information that native speakers regularly consult a
dictionary for. If they do, it is likely to be in order to check the pronunciation
of a word that they have met only in writing. Perhaps in recognition of this,
NODE and subsequently COD10 do not give a transcription of the pronuncia-
tion of ‘ordinary, everyday words’, rather:
In the New Oxford Dictionary of English, the principle followed is that pro-
nunciations are given where they are likely to cause problems for the native
speaker of English, in particular for foreign words, foreign names, scientific
and other specialist terms, rare words, words with unusual stress patterns,
and words where there are alternative pronunciations or where there is a
dispute about the standard pronunciation.
(Introduction, p. xvii)
What counts as an ordinary word must be a matter of judgement. By way of

comparison, here are lists of words from one page of COD10 distinguishing
those that have been provided with a transcription and those that have not:
With transcription: traipse, trait, trajectory, Trakehner, Traminer, trammel,
tramontana, tramontane, trampoline, trance, tranche, trans-
Without transcription: training college, training shoe, train mile, train oil,
train shed, trainspotter, traitor, tra la, tram, tramlines, tramp, trample, tram
road, tram silk, tramway, trank, tranny, tranquil, tranquillize, transaction,
transactional analysis.
The exclusion of pronunciation information for many words in NODE and

COD10 represents a move, albeit small, away from subservience to the ‘record-
ing’ function of general-purpose dictionaries towards consideration of what the
user might or might not need.
The second issue concerns the model of pronunciation that is offered, and
the degree of variation that is recorded. COD9 says that its IPA transcriptions
are ‘based on the pronunciation associated especially with southern England
(sometimes called “Received Pronunciation”)’. In CED4, the ‘pronunciations
of words . . . represent those that are common in educated speech’. In LDEL2,
the pronunciation represented . . . is what may be called a ‘standard’ or

‘neutral’ British-English accent: the type of speech characteristic of those
people often described as having ‘no accent’, or, more accurately, having an
accent that betrays little or nothing of the region to which the speaker
belongs.
NODE represents ‘the standard accent of English as spoken in the south of

England (sometimes called Received Pronunciation or RP)’. Some of these
dictionaries acknowledge the existence of other accents, both in other English-
speaking countries and regionally within Britain, but argue that it is impossible
to do them all justice. Chambers, which describes some of the ways in which
pronunciation differs in other national varieties of English, claims that its respelling
system of transcription ‘allows for more than one interpretation so that each
user of the dictionary may choose a pronunciation in keeping with his speech’.
However it is described, it is the ‘educated’ accent of southern England, with
its /bkt/ rather than /bgt/ pronunciation of but, and /grp:s/ rather than /græs/
for grass, that is the model represented in British dictionaries. At one time, it was
argued that this accent was the one most widely understood, the one used
predominantly in public discourse, the one taught to foreign learners of English,
and so on. This is presumably the sense in which it might be considered a
‘standard accent’ (NODE), though NODE acknowledges that it is not a static
accent:
The transcriptions reflect pronunciation as it actually is in modern English,

unlike some longer-established systems, which reflect the standard pronun-
ciation of broadcasters and public schools in the 1930s.
(p. xvii)
The status of this accent as the prestige accent for British English has been
constantly challenged by phoneticians of English, and there is a much greater
diversity of accents heard now in public life. It is perhaps becoming an anach-
ronism to continue to record this accent in modern dictionaries, but the debate
on which pronunciation should be recorded has hardly begun.
Dictionaries do record some variation in pronunciation. Learners’ dictionar-
ies, which have a worldwide market, now routinely include American pronun-
ciation as well as British (Chapter 11). Native speaker dictionaries, on the other
hand, record variation within the chosen accent, for example the /i:k../ and /
vk../ beginnings to economics. Here are some further examples, drawn from
COD9:
• coastguard /skfgs(t)gp:d/
• distribute /disstribju:t/ - /sdistribju:t/
• February /sfvbrgfri/ - /sfvbjgfri/
• oceanic /fgnisanik/ - /fgsisanik/
• sedentary /ssvd(f)nt(f)ri/
• vin rosé /van rfgszei/ - /vṽroze/
They show a number of types of variation in pronunciation, even in the chosen

accent: omission of sounds in more rapid or ‘less careful’ enunciation (the sounds
in brackets in coastguard and sedentary); variation in stress placement, as in distrib-
ute; one or more alternative sounds, as in February and oceanic; and for loanwords,
the anglicised and the original, ‘foreign’, pronunciation, as for vin rosé.
9.3 Inflection
For most words that can be inflected in English – nouns, verbs, adjectives (see
Chapter 1) – the inflection follows from the general rules of morphology, is not
idiosyncratic to the individual lexeme, and is therefore not appropriate to the
lexical information contained in dictionaries. However, there are some excep-
tions to this generalisation, which dictionaries do record. A small number of
adjectives, some nouns, and a larger number of verbs inflect ‘irregularly’, not
according to the general pattern, and these are given for each lexeme con-
cerned, e.g.
• adjective bad – worse – worst

• noun foot – feet, mouse – mice, ox – oxen, sheep – sheep
• verb bring – brought – brought, feel – felt – felt, give – gave – given, hit –
hit – hit, see – saw – seen, wear – wore – worn, etc.
These basic irregularities do not exhaust the possible idiosyncracies, and dic-
tionaries tend to give any inflection that is likely to cause a difficulty for writers,
including predictable spelling variations.
For the plural inflection of nouns, the following may well be noted:
• loanwords that retain their original, ‘foreign’ plural, e.g. cactus – cacti, crite-
rion – criteria, kibbutz – kibbutzim, phylum – phyla, vertex – vertices. More and
more of these plurals are becoming regularised, including cactuses and
vertexes.
• nouns that end in -o or -i, where there is often confusion about whether the
inflection is -s or -es, e.g. curio-s, domino-es, etui-s, halo-es or -s, piccallili-es or
-s.
• nouns ending in -y, which may change the y to i and add -es, or may simply
add -s, e.g. abbey-s, academy – academies, monkey -s, mystery – mysteries,
odyssey-s, symmetry – symmetries.
• nouns that change either the spelling or pronunciation of their final sound
(voicing of /m/, /f/ or /s/) when the plural suffix is added, e.g. bath-s, hoof
– hooves, house-s, mouth-s, shelf – shelves, truth-s, wolf – wolves.
For the inflections of verbs, the following may well be noted:
• where the final consonant of the root is doubled in spelling with the addi-
tion of a suffix: flip – flipping – flipped, lag – lagging – lagged, prod – prodding –
prodded, refer – referring – referred, shovel – shovelling – shovelled, sin – sinning –
sinned.
• where the final consonant might be expected to double, but does not, e.g.
benefit – benefiting – benefited, galop – galoping – galoped, gossip – gossiping –
gossiped, market – marketing – marketed, pilgrim – pilgriming – pilgrimed.
• where the final consonant is -c and a k is added before the inflectional suffix,
e.g. bivouac – bivouacking – bivouacked, magic – magicking – magicked, picnic –
picnicking – picnicked.
• where the final consonant is -y, which may change to i before an inflec-
tional suffix, e.g. cry – cries – cried (but crying), shy – shies – shied, supply –
supplies – supplied, weary – wearies – wearied.
For adjective inflections, the following usually apply:
• the consonant doubling rule, as for verbs, e.g. big – bigger – biggest, hip –
hipper – hippest, sad – sadder – saddest.
• the y to i rule, as for verbs, e.g. dry – drier – driest, fluffy – fluffier – fluffiest,
lively – livelier – liveliest, rosy – rosier – rosiest, wacky – wackier – wackiest
(but not sly -er, -est).
Additionally, two-syllable adjectives that form their comparative and super-

lative by means of inflectional suffixes, rather than the periphrastic more/most
construction, may be marked as such in the dictionary (e.g. NODE), such as
common -er/-est, narrow -er/-est, thirsty -er/-est. However, while NODE notes the
-er/-est suffixes for narrow – and sallow – it does not indicate them for mellow or
shallow.
One other point is worth mentioning here, though strictly speaking it be-
longs to derivational morphology rather than to inflectional. English has a number
of nouns that survive from Old English which have a related adjective that has
been borrowed into English usually from Latin, e.g. church – ecclesiastical. Some
dictionaries usefully indicate these connections, e.g. CED4. Further examples
are: lung – pneumonic, pulmonary, pulmonic; mind – mental, noetic, phrenic; wall –
mural.
9.4 Word class

It is one of the traditions of lexicography to identify the word class(es) or part(s)
of speech that each lexeme in a dictionary belongs to. The traditional terms,
usually abbreviated, are: noun (n), verb (v, vb), adjective (adj), adverb (adv),
pronoun (pron), preposition (prep), conjunction (conj), and interjection (interj).
Under the influence of modern descriptive linguistics the adjective class in some
dictionaries (e.g. CED, NODE) is divided into adjectives proper and ‘deter-
miners’ (see Chapter 1). CED in addition recognises a class of ‘sentence connec-
tors’ (e.g. however, therefore) and a class of ‘sentence substitutes’ (e.g. no, maybe),
both of which are traditionally assigned to the adverb class. In COD10, the
interjection class is renamed ‘exclamation’ (exclam) and it includes yes and no.
So far, most dictionaries follow the tradition. Practice begins to vary in the
information provided over and above the basic word class label. COD10 pro-
vides none. Its predecessor, the COD9, followed another tradition in respect of
verbs and marked verbs or senses of verbs as ‘transitive’ (tr) or ‘intransitive’
(intr), or indeed ‘reflexive’ (refl). For example, kick is marked ‘tr’ for the ‘strike
or propel forcibly with the foot or hoof etc.’ sense, ‘intr’ for the ‘strike out with
the foot’ sense, and ‘refl’ for the ‘kick oneself’ sense. COD10 perhaps excludes
these terms in recognition of the fact that they are not familiar to most modern
dictionary users; its larger parent, NODE, also eschews them, using ‘with obj’,
i.e. ‘object’, and ‘no obj’ instead.
Indeed, NODE goes further than most general-purpose native speaker dic-
tionaries in the ways in which it subclassifies words. For nouns, it indicates
when a noun is used as a ‘mass noun’, e.g. legislation, which cannot be made
plural or be preceded by the indefinite article (a/an). It also uses the term ‘count
noun’ for a sense of a mass noun that can be made plural and countable, e.g.
observance in the sense of ‘religious or ceremonial observances’. Otherwise nouns
are assumed to be countable. NODE recognises a subclass of ‘sentence adverb’,
with 159 adverbs or senses of adverbs so marked, including coincidentally, fortun-
ately, paradoxically, regrettably, thankfully. It also marks a subclass of ‘submodifier’
adverbs, which are used to modify adjectives and other adverbs, some 277 of
them, including altogether, decidedly, hideously, predictably, simply, utterly.
The word class label, and any subclassification, represents grammatical infor-
mation about words, where they can operate in the syntax of sentences, what
their combinatorial possibilities are. Some dictionaries provide grammatical
information over and above word class labelling, though it is difficult to draw a
clear distinction between word (sub-)class information proper and other syntac-
tic labelling. Indeed, NODE in its discussion of these matters in the ‘Guide to
the Use of the Dictionary’ makes no such distinction.
9.5 Other grammatical information

The distinction between ‘mass’ and ‘count’ noun, for example, is not simply a
word class subdivision; it is also an indication of the determiners that may com-
bine with a noun, e.g. numerals with count nouns, but not with mass nouns.
Similarly, the ‘transitive’/‘intransitive’ subclassification of verbs relates to whether,
in the specified sense, the verb takes an object or not, and additionally whether
the sentence in which the verb occurs can be made passive.
For nouns, NODE also specifies when they can be used ‘as modifier’, before
another noun, with an adjectival function, e.g. keynote as in keynote address or
shadow as in shadow minister of . . . CED likewise notes such uses of some nouns,
but dictionaries may differ in their categorisations. One way of treating such
uses of nouns would be to recognise the derivation of an adjective by the word
formation process of ‘conversion’ (see Chapter 2): CED4 marks key, as in a key
person, as ‘modifier’, while NODE recognises an adjective key to cover this
usage. The other peculiarity of nouns that dictionaries often mark is when there
may be a mismatch between the form of a noun (singular or plural) and its use
syntactically. For example, darts and economics have a plural form but are usually
‘treated as sing(ular)’. On the other hand, singular so-called ‘collective’ nouns,
such as government or team may be ‘treated as sing or pl(ural)’. In NODE also, ‘in
sing’ is used to mark (the sense of ) a count noun that can only be used in the
singular (e.g. riot as in the garden was a riot of colour) or the sense of a mass noun
where an indefinite article may be used (e.g. wealth in a wealth of information).
For adjectives, NODE specifies three possible syntactic positions that they
may be restricted to: before the noun (‘attrib(utive)’), after a verb like be, become
or seem (‘predic(ative)’), and immediately after the noun (‘postpositive’). CED4’s
equivalent terms are: ‘prenominal’, ‘postpositive’ and ‘immediately postpositive’.
Here are some examples:
• attributive bridal, custom, geriatric, innermost, mere, opening, teenage, zero-sum

• predicative aglow (and many others with prefix a-), catching, disinclined,
legion, privy, tantamount, well (i.e. ‘not ill’)
• postpositive aplenty, designate, enough, galore, incarnate, par excellence, as well as
a number of adjectives connected with cooking and heraldry.
The case of verb syntax is more complicated. If a dictionary is to record the

peculiarities of each lexical item, then the crude transitive/intransitive distinc-
tion does not do justice to the syntactic operation of many verbs. Nor does the
threefold distinction of NODE: ‘with obj’, ‘no obj’, and ‘with adverbial’ (e.g.
behave, clamber). Unlike learners’ dictionaries (Chapter 11), native speaker dic-
tionaries generally do not systematically and comprehensively record the possi-
ble syntactic patterning of verbs. Few go beyond ‘transitive’ and ‘intransitive’.
However, compare the following entries for argue from CED4 and NODE:
argue 1 (intr) to quarrel; wrangle: they were always arguing until I arrived. 2
(intr; often foll. by for or against) to present supporting or opposing reasons
or cases in a dispute; reason. 3 (tr; may take a clause as object) to try to prove
by presenting reasons; maintain. 4 (tr; often passive) to debate or discuss: the
case was fully argued before agreement was reached. 5 (tr) to persuade: he argued me
into going. 6 (tr) to give evidence of; suggest: her looks argue despair.
[CED4]
argue 1 (reporting verb) give reasons or cite evidence in support of an idea,
action, or theory; typically with the aim of persuading others to share one’s
view: [with clause] sociologists argue that inequalities in industrial societies are
being reduced | [with direct speech] ‘It stands to reason,’ she argued.
• [with obj.] (argue someone into/out of ) persuade someone to do or
not to do (something) by giving reasons: I tried to argue him out of it.
2 [no obj.] exchange or express diverging or opposite views, typically in a
heated or angry way: don’t argue with me | figurative I wasn’t going to argue
with a gun | [with obj.] she was too tired to argue the point.
[NODE]
These two dictionaries give considerably more syntactic information for verbs,
both by way of labels and in examples, than has been customary in general-
purpose dictionaries, even of desk size, until recently. NODE justifies this ap-
proach both by pointing to the role of grammar in distinguishing the meanings
or senses of lexemes and with the following argument:
the aim is to present information in such a way that it helps to explain the
structure of the language itself, not just the meanings of the individual
senses. For this reason, special attention has been paid to the grammar of
each word, and grammatical structures are given explicitly.
(p. xi)
9.6 Usage
All dictionaries have a set of labels to mark words or senses of words that are
restricted in some way in the contexts in which they may occur. The contextual
restrictions may be geographical (i.e. dialectal), historical (e.g. archaic), stylistic
(e.g. informal), according to topic (e.g. Botany), and so on. In this section, we
review the types and range of usage labels used in general-purpose dictionaries.
9.6.1 Dialect
Dialect labels refer to geographical restriction, and we can take this to include
both national varieties and regional dialects within a national variety. Most
British dictionaries nowadays claim an international perspective and include
words peculiar to the vocabulary of other English-speaking countries, but still
largely confined to North America, Australia and New Zealand, and South
Africa. The newer Englishes of, say, the Indian subcontinent, or West Africa, or
the Caribbean, or Singapore tend to receive lesser attention. However, COD10,
for example, contains around fifty words marked ‘W. Indian’, and a rather
larger number labelled ‘Indian’, e.g.
• West Indian braata, dotish, fingle, higgler, mamguy, nancy story, spraddle, tafia
• Indian babu, charpoy, durzi, haveli, lakh, nullah, sadhu, zamindar.
NODE claims around 14,000 geographical labels spread through the diction-
ary, but these are mainly ‘regionalisms encountered in standard contexts in the
different English-speaking areas of the world’ (p. xvi). The largest number,
inevitably, belong to the vocabularies of English spoken in North America, for
which NODE has three labels: ‘N. Amer.’ (i.e. North American), ‘US’ (i.e.
United States), and ‘Canadian’. The last two are presumably for cases where the
restriction is more limited, e.g. in the case of blue box:
1. chiefly US an electronic device used to access long-distance telephone

lines illegally.
2. chiefly Canadian a blue plastic box for the collection of recyclable house-
hold materials.
A similar labelling is used for words specific to Australian and New Zealand
Englishes, where the majority are marked ‘Austral./NZ’ (e.g. mullock ‘rubbish,
nonsense’), because they are shared by both varieties, and some are marked
separately, rather more ‘Austral’ (e.g. gunyah ‘bush hut’) than ‘NZ’ (e.g. kumara,
‘sweet potato’). There is no such confusion about South African English words
(e.g. koppie ‘small hill’), though some are shared with other varieties, e.g. dingus
(shared with ‘N. Amer’) ‘a thing one cannot or does not wish to name specifi-
cally’, dropper (shared with ‘Austral./NZ’) ‘a light vertical stave in a fence’.
Words or senses that are exclusive to the British English variety are also appro-
priately marked (over 4,000 in NODE), e.g. fly-past, gobstopper, knacker, linctus,
nearside, peckish, scrapyard.
When it comes to dialects within Britain, NODE/COD10 are less specific.
While they have a label ‘Scottish’ and ‘N(orthern) English’ (often occurring
together for a word), all other dialect words are marked simply as ‘dialect’,
except that one word (scally) is noted as N(orth) W(est) English, and a handful
are labelled ‘black English’. LDEL and, more especially, CED have both a greater
representation of British English dialect words and a more differentiated label-
ling. CED4 notes in its Guide:
Regional dialects (Scot. and northern English dialect, Midland dialect, etc.) have
been specified as precisely as possible, even at the risk of overrestriction, in
order to give the reader an indication of the appropriate regional flavour.
(CED4, p. xxi)
So, chine, in the sense of ‘a deep fissure in the wall of a cliff ’, is labelled ‘Southern
English dialect’; flash meaning ‘a pond, esp. as produced as a consequence of
subsidence’ is marked ‘Yorkshire and Lancashire dialect’; maungy ‘(esp. of a
child) sulky, bad-tempered or peevish’ is labelled ‘West Yorkshire dialect’; snicket
‘a passageway between walls or fences’ has the label ‘Northern English dialect’;
and tump ‘a small mound or clump’ is marked ‘Western English dialect’.
9.6.2 Formality
A number of words or senses are marked as ‘formal’ or ‘informal’, though the
latter label usually greatly outnumbers the former: in COD10, for example, the
‘informal’ label occurs over seven times more frequently than the ‘formal’ label.
These terms relate to the formality of the context in which a word is deemed to
be appropriate. They are defined in the LDEL2 Guide as follows:
The label informal is used for words or meanings characteristic of conversa-

tion and casual writing (e.g. between friends and contemporaries) rather
than of official or formal speech or writing.
The label formal is used for words or meanings which are characteristic
of writing rather than speech (except for formal speech situations, such as
a lecture), and particularly of official, academic, literary, or self-important
writing. In other contexts, such words may seem over-elaborate or
pompous.
(LDEL2, p. xviii)
The term ‘colloquial’ is sometimes used instead of ‘informal’ (e.g. in Chambers).

Many dictionaries identify ‘slang’ as a point further down the formality scale,
but we shall deal with slang under 9.6.3.
Here are some examples of words marked as ‘formal’ and ‘informal’ in COD10
(you may need to look them up, if they are new to you):
• formal abnegate, circumambulate, emolument, gustation, jocose, lucubration, nor-

mative, pinguid, sapient, theretofore, wheresoever
• informal baby boom, beanfeast, dekko, expat, gasbag, haywire, junkie, lashings,
manky, nitty-gritty, once-over, prang, rozzer, shambolic, townie, vapourware,
wannabe, yonks, zilch.
Informal terms, since they are the staple of ordinary conversation, have a ten-
dency to date; and you may consider that some of the terms listed might belong
to your parents’ or grandparents’ speech, but not to yours.
9.6.3 Status
By ‘status’ we mean the propriety of the use of a word, even in ordinary conver-
sation. Under ‘status’ we would include the term ‘taboo’. A taboo is defined
in COD10 as ‘a social or religious custom placing prohibition or restriction
on a particular thing or person’, while COD9 also includes as a second sense
‘a prohibition or restriction imposed on certain behaviour, word usage, etc., by
social custom’. A taboo word, therefore, is one that you would not use in
ordinary conversation, unless you wanted to shock. Such taboo words would
include: those connected with sexual and excretory functions, blasphemies,
and other ‘swear’ words. However, there is little left in our society that is taboo,
and so modern dictionaries no longer use the label; CED4 is an exception.
Not even COD9, which mentions the connection with ‘word usage’ in its
definition of taboo uses it as a label, preferring ‘coarse slang’ instead. In NODE
and COD10, this has become ‘vulgar slang’; LDEL2 and Chambers use simply
‘vulgar’.
In the Oxford dictionaries, then, the connection is made with ‘slang’, the
other term under this heading, and glossed by CED4 as follows:
Slang This refers to words or senses that are racy or extremely informal.
The appropriate contexts in which slang is used are restricted, for example,
to members of a particular social group or those engaged in a particular
activity. Slang words are inappropriate informal speech or writing.
‘Slang’ is, therefore, not just ‘very informal’; it implies a restriction beyond
simply the formality of the context of use, to defined social groups, and it
includes a consideration of appropriacy. It belongs with ‘taboo’. Even more so
than with informal words, the slang status of words may change over a relatively
short period of time and quickly become dated. Not only that, but people’s
tolerance of slang varies considerably, and it is no surprise that dictionaries
differ in their labelling of such words. In fact, COD10 does not use the label
‘slang’ on its own, unlike COD9, but only in conjunction with a preceding
defining adjective, such as ‘nautical’, ‘military’, ‘theatrical’, ‘black’, as well as
‘vulgar’. A number of the words marked as ‘slang’ in COD9 have become
‘informal’ in COD10, e.g. acid (= LSD), aggro, awesome (= excellent), banger
(= sausage, old car), dough (= money). However, those that are marked ‘coarse
slang’ in COD9 generally have the label ‘vulgar slang’ in COD10, e.g. arse, crap,
piss, turd, not to mention the many words for the male and female genitalia.
Incidentally, though, fart is labelled ‘coarse slang’ in COD9, but only ‘informal’
in COD10.
9.6.4 Effect
There is a set of usage labels used in dictionaries that relate to the effect that a
word or sense is intended by the speaker or writer to produce in the hearer or
reader. Any dictionary usually makes a selection from these labels. One set
reflects the attitude of the speaker and includes: ‘derogatory’ (intending to be
disrespectful), ‘pejorative’ (intending to show contempt), ‘appreciative’ (intending
to show a positive attitude), ‘humorous’ or ‘jocular’ (conveying a light-hearted
attitude). Closely related is the term ‘offensive’, which may have intent on the
part of the speaker or may be unconscious, but which could be taken by a hearer
as offensive, either racially or in some other way. Other kinds of ‘effect’ label
include: ‘euphemistic’, i.e. using an oblique word to refer to an unpleasant
topic; ‘literary’ and ‘poetic’, i.e. words that tend to be confined to literary texts
or poetry and have a ‘literary’ effect when they are used elsewhere. Here are
some examples:
• derogatory banana republic, bimbo, cronyism, fat cat, lowbrow, newfangled,

psychobabble, slaphead, woodentop (from COD10)
• jocular argy-bargy, bounder, doughty, funniosity, industrial-strength, leaderene,
osculate, purloin, square-eyed, walkies (from COD9)
• offensive bogtrotter (= Irish person), cripple (= disabled person), mongrel (=
person of mixed parentage), wog (= foreigner, especially non-white) (from
COD10)
• euphemistic cloakroom (for ‘toilet’), ethnic cleansing (for ‘forced mass expul-
sion of a group of people from an area’), interfere with (for ‘sexually molest’),
passing (for ‘death’) (from COD9)
• literary apace, bestrew, connubial, fulgent, incarnadine, nevermore, plenteous, slumber,
vainglory, wonted (from COD10).
Even more than with formality and status labels, we would expect effect labels
to vary between dictionaries, since they require a greater exercise of judgement
on the part of the lexicographer and are more likely to be variously perceived.
9.6.5 History
Most dictionaries include labels for words or senses that are either no longer in
current use or whose currency is questionable or suspect. The term ‘obsolete’
refers to words or senses that have definitely ceased to be used. It is, of course, an
important term in the OED, but in dictionaries that purport to contain current
vocabulary, it is not often used. LDEL2 includes it, however, with the gloss:
The label obs (obsolete) means there is no evidence of use of a word or

meaning since 1755 (the date of publication of Samuel Johnson’s A Diction-
ary of the English Language). This label is a comment on the word being
defined, not on the thing it designates.
(p. xvii)
For example, fay, meaning ‘faith’ is marked ‘obs’ in LDEL2; in SOED4 it is
marked as ‘long archaic, rare’. CED4 also claims to use the ‘obsolete’ label and
notes that ‘in specialist or technical fields the label often implies that the term
has been superseded’ (p. xx); it also uses the label ‘old fashioned’ (e.g. of the
‘illegitimate’ sense of bastard), which it does not discuss in the ‘Guide’.
NODE and COD10 use the labels ‘dated’, ‘archaic’, and ‘historical’ to mark
words or senses no longer current; and to these we might add the label ‘rare’.
These labels are defined as follows:
‘dated’: no longer used by the majority of English speakers, but still en-
countered, especially among the older generation.
‘archaic’: old-fashioned language, not in ordinary use today, though some-
times used to give a deliberately old-fashioned effect and also encountered
in the literature of the past.
‘historical’: still used today, but in reference to some practice or artefact that
is no longer part of the modern world.
‘rare’: not in normal use.
The ‘historical’ label marks not words as such but the things that they denote
as being no longer current. It is not clear how ‘rare’ might differ from ‘archaic’.
Perhaps some examples (from COD10) will help to distinguish them:
• dated aeronaut, cobble (= repair, e.g. shoes), gamp (= umbrella), jerry (= chamber
pot), necktie, picture palace (= cinema), spiffing, wireless (= radio)
• archaic asunder, chapman, fandangle, guidepost, mayhap, poltroon, therewithal,
vizard
• historical approved school, dolly tub, footpad, jongleur, margrave, pocket borough,
safety lamp, tumbril, velocipede
• rare argute (= shrewd), comminatory (= threatening, vengeful), lustrate (= purify,
e.g. by sacrifice), toxophilite (= archer), vaticinate (= foretell future).
9.6.6 Topic or field

Where a word or sense is restricted to a, usually specialised or technical, field of
study or activity, dictionaries generally add an appropriate label. Topics may
range from the sciences, technologies and medicine, through the professions
such as law or business, to sports and leisure pursuits. The label marks a word or
sense as belonging to the technical vocabulary of the topic. Here are a few
examples to illustrate the point, taken from NODE:
• handshaking computing
• periventricular anatomy and medicine
• quiddity philosophy
• sopranino music
• top edge cricket
• weather helm nautical
• white hole astronomy.
9.6.7 Disputed usage

Dictionaries regard one of their functions as being to draw attention to words
whose usage is a matter of controversy, and perhaps to offer an opinion for the
linguistically insecure. The word or sense that is the subject of dispute may
be labelled as such, e.g. ‘disp’ in COD9, as for decimate in the sense of ‘kill or
remove a large proportion of ’. More often, a dictionary will append a ‘usage
note’ to explain the nature of the dispute and proffer advice, e.g. in NODE, for
disinterested:
Nowhere are the battle lines more deeply drawn in usage questions than
over the difference between disinterested and uninterested. According
to traditional guidelines, disinterested should never be used to mean ‘not
interested’ (i.e. it is not a synonym for uninterested) but only to mean
‘impartial’, as in the judgements of disinterested outsiders are likely to be more
useful. Ironically, the earliest recorded sense of disinterested is for the
disputed sense. Today, the ‘incorrect’ use of disinterested is widespread:
around 20 per cent of citations on the British National Corpus for disin-
terested are for this sense.
Besides usage notes, CED4 also has a label ‘not standard’ to apply to appropriate
items, such as ain’t or worser. LDEL2 has the labels ‘nonstandard’ and ‘substand-
ard’ and distinguishes them as follows:
The label nonstandard is used for words or meanings that are quite
commonly used but considered incorrect by most educated users of the
language:
lay . . . vi . . . 5 nonstandard LIE
The label substandard is used for words or meanings used by some speakers
but not generally considered to be part of standard English:
learn . . . vb . . . 2 substandard to teach. (p. xviii)
This is about as prescriptive as it gets. CED4 labels this sense of learn as ‘not
standard’, and it provides a usage note to discuss the differences between lay and
lie. By comparison, we might note that Chambers labels ain’t as ‘coll(oquial)’ and
the disputed usages of learn and lay as ‘illit(erate)’.
9.7 Further reading

For information on how an individual dictionary or edition deals with the
topics discussed in this chapter the ‘Guide to the Dictionary’ is the place to start.
Dick Hudson’s article on ‘The linguistic foundations for lexical research and
dictionary design’ in the International Journal of Lexicography (1988) surveys the
lexical information that dictionaries should take account of. Bo Svensén’s Prac-
tical Lexicography (1993) has chapters on most of the concerns of this chapter.
Sidney Landau has a chapter on usage (Chapter 5) in Dictionaries: the Art and
Craft of Lexicography (1989, 2001). Juhani Norri has two articles in IJL on label-
ling: ‘Regional labels in some British and American dictionaries’ (vol. 9, 1996),
and ‘Labelling of derogatory words in dictionaries’ (vol. 13, 2000).
Etymology 117
10 Etymology
Since the late seventeenth century general-purpose native speaker dictionaries

have included information about the etymology of words (see 4.3). Indeed,
common words were included in dictionaries initially merely for the sake of
recording their etymologies. The etymology section of a dictionary entry aims
to trace the history of a word (see 2.1) to its ultimate source. Where a word has
come into existence as the result of a word formation process, e.g. derivation or
compounding, then it is not usually given an etymology, unless it is unclear
what the elements of the new word are and how they have been combined. In
general, therefore, it is base (root) forms that are given etymologies.
The ‘Introduction’ to NODE likens the tracing of etymologies to archaeology:
the evidence is often partial or not there at all, and etymologists must make
informed decisions using the evidence available, however inadequate it
may be. From time to time new evidence becomes available, and the known
history of a word may need to be reconsidered.
(p. xiv)
We now consider many of the etymologies proposed by eighteenth-century
dictionaries to be rather fanciful, particularly in the light of nineteenth- and
twentieth-century scholarship. Most larger dictionaries have an etymology con-
sultant, and the OED continues to add to etymological scholarship by its on-
going research into the histories of words. It is to the OED that most dictionaries
look as the primary source for their etymological information.
To understand the discussion in this chapter, it will be useful to keep in mind
the outline history of English, as expressed in the basic periods of the language
( Jackson and Zé Amvela 2000: 23ff.):
• Old English 450 (settlement by Angles, Saxons and Jutes) to 1066 (Norman
conquest)
• Middle English 1066 to between 1450 (beginning of the Renaissance) and
1500 (including the beginning of printing in Britain – Caxton 1476)
• Early Modern English 1500 to 1800 (growth of technology, and beginnings
of empire)
• Modern English 1800 to the present.
118 Etymology
The dates are, of course, to an extent, arbitrary; but they serve to mark approx-
imate transition points in the development of the language in the light of changes
in society and culture.
10.1 Interpreting etymologies

In historical dictionaries (OED, SOED), the etymology is given at the begin-
ning of the entry, in keeping with their historical orientation. Webster’s Third
also puts etymology in this position in its entries, but the more usual position is
towards or at the end of an entry. In either position, it is conventional to present
etymological information within square brackets [ ], though NODE and COD10
present the information in a separate paragraph of the entry, introduced by
‘ORIGIN’ in small capital letters.
The amount of detail contained in an etymology depends in large part on the
size of the dictionary, with smaller (e.g. Pocket) dictionaries containing perhaps
only the immediate source of a word. Larger dictionaries contain that, and often
much more. The immediate origin of a word is the obligatory part of an ety-
mology, whether it is an Old English or Anglo-Saxon word, or whether it has
been ‘borrowed’ into the language either during or subsequent to the Old English
period. In addition to the immediate origin, some dictionaries will include an
approximate date at which, according to the evidence (from written sources), the
word came into English. This may consist of just a number for the century (CED,
COD10), or it may divide each century into three – early, mid, late (NODE).
Besides the immediate source and possibly a date, etymologies in larger dic-
tionaries trace the origin of a word back to its earliest known source language.
For words borrowed from French, for example, this may involve indicating a
Latin source of some kind. Some of the words taken directly from Latin can be
traced back to a Greek original. Words of Old English or Old Norse origin can
often be assigned an earlier Germanic source. Here is an example with an inter-
esting etymology (from NODE):
abacus late Middle English (denoting a board strewn with sand on which
to draw figures): from Latin, from Greek abax, abak- ‘slab, drawing board’,
of Semitic origin; probably related to Hebrew sAbAq ‘dust’.
Some etymologies, as in this one for abacus, give the meanings of these original
forms, and this has led to the so-called ‘etymological fallacy’, according to which
the ‘real meaning’ of a word is its original meaning. So, it is argued, the real
meaning of decimate is ‘kill or remove every tenth person’, because it derives
from the Latin decimus ‘tenth’, and indeed the verb decimare referred to the
Roman practice of killing every tenth soldier in a mutinous legion. But it is a
fallacy to argue in this way, because words undergo semantic as well as phono-
logical and orthographic changes. The meaning of decimate has changed to de-
note any large-scale killing or destruction, and, in any case, for English users the
connection with decimus has been all but lost (though compare decimal ).
Etymology 119
In the case of words having an Old English origin, the etymology may addi-
tionally record ‘cognate’ words in other Germanic languages. The languages
concerned are the Scandinavian languages (Old Norse, Swedish, Danish), Dutch,
and German. All these languages are thought to derive from a common ancestor
language, which has been termed ‘Germanic’, but for which no records survive.
The cognates, therefore, demonstrate the Germanic origin of the Old English
words concerned. For example, NODE gives the etymology of speak as: ‘Old
English sprecan, later specan, of West Germanic origin; related to Dutch spreken
and German sprechen’. CED4 gives the etymology of through as: ‘Old English
thurh; related to Old Frisian thruch, Old Saxon thuru, Old High German duruh’.
The etymology of some words can be traced to particular cultural beliefs and
practices, sometimes known as ‘folk etymology’. For example, the NODE
etymology for bigwig reads: ‘early 18th cent.: so named from the large wigs
formerly worn by distinguished men’; and the CED4 etymology for crocodile
tears reads: ‘from the belief that crocodiles wept over their prey to allure further
victims’.
To interpret etymologies, not only do you need to understand what the
various parts of the entry might signify, e.g. immediate origin, ultimate source,
cognates, folk etymology, but you also need to be able to make sense of the
language names used, e.g. Old High German, Middle Dutch, Late Latin, which
in many dictionaries, especially those of concise size or smaller, will be in the
form of abbreviations (OHG, MDu, LL). For the Germanic languages and French,
the terms ‘Old’, ‘Middle’ and ‘Modern’ correspond approximately to their use
for English, i.e. pre-medieval, medieval, and post-medieval. In the case of Latin,
the plain term relates to the classical period, and ‘Late Latin’ to the period from
approximately 200 AD to 400 AD. A label ‘New Latin’ (NL) is used to refer to
words coined in Latin since the Renaissance, especially scientific terms. Where
these are common to a number of (European) languages, the label ISV, for
‘International Scientific Vocabulary’, is used by W3 and some other dictionaries.
The following discussion and illustration of etymologies will serve to illuminate
some of these points.
10.2 Original English

Words labelled ‘Old English’ (OE) in etymologies have their origin in the Ger-
manic dialects spoken by the Anglo-Saxon tribal people, invited in the early
fifth century by the Celtic peoples then inhabiting most of these islands to help
them defend their country against the marauding Picts and Scots from the north,
after the departure of the Roman legions as a consequence of the beginning of
disintegration of the empire. The guests from northern Europe became invaders,
and the Celtic peoples fled to the west, finding habitation in Wales and Corn-
wall. The proportion of words in modern English with an OE origin is very
small: COD10 has only 2,515 of its 64,679 headwords marked as of OE origin,
less than 4 per cent. But they are the common words of everyday speech, and
any informal text or discourse will be made up of a great majority of OE words.
120 Etymology
A very small number of words marked as OE are further noted as being of
Celtic origin. They are the few survivors, borrowed by the Anglo-Saxon invad-
ers. Examples include: ass, bin, brock (i.e. ‘badger’), combe, hog, rich, tor. A number
of other loanwords survive from OE, taken largely from Latin as a consequence
of the introduction of the Roman form of Christianity into Britain from the late
sixth century; they are mostly religious words, e.g. anthem, candle, charity, disciple,
martyr, noon, psalm, verse.
Old English also took some words from its sister Germanic language Old
Norse, the language spoken by the Viking invaders from across the North Sea,
who during the eighth and ninth centuries came first on raiding expeditions and
then to stay, eventually ruling half the country in the territory known as
‘Danelaw’. Old Norse (ON) has had a significant effect on place names in the
north and east of the country: any place whose name ends in -by, -thorpe, or
-thwaite, for example, had a Viking origin or settlement. Although many words
are cognate in OE and ON, there were borrowings from ON into OE, such as:
arrow, baulk, fang, glove, knife, plough, skin, tidings, wrong. They are labelled in
dictionary etymologies as ‘OE from ON’. Other ON words were borrowed
into English later, during the Middle English period, such as: anger, birth, dirt,
ferry, ill, keg, odd, raise, sky, tether, ugly, want. These are labelled as ‘ME from
ON’.
10.3 Latinate loans

We noted in 2.1 that English vocabulary consists of a substratum of Germanic
words, its Old English base, and superstrata of words taken either directly or
indirectly from Latin. The first such superstratum was laid in the centuries fol-
lowing the Norman conquest in 1066, the second during and following the
sixteenth-century Renaissance period. NODE labels 8322 words as having their
immediate origin in Middle English (see 10.1); of these 3234 are borrowed
either directly (most of them) or indirectly from Old French, e.g.
lavish late Middle English (as a noun denoting profusion): from Old French
lavasse ‘deluge of rain’, from laver ‘to wash’, from Latin lavare.
Some of the indirect borrowings are via Anglo-Norman (387 in NODE),

the language that developed among the French-speaking ruling elite in
Britain, e.g.
astray Middle English (in the sense ‘distant from the correct path’): from
an Anglo-Norman French variant of Old French estraie, past participle of
estraier, based on Latin extra ‘out of bounds’ + vagari ‘wander’.
Some of the Latinate words borrowed into Middle English come not from Old
French but directly from Latin (1885 in NODE). This is mostly in the latter part
of the period, as the transition to the Renaissance begins, e.g.
Etymology 121
interrupt late Middle English: from Latin interrupt- ‘broken, interrupted’,
from the verb interrumpere, from inter- ‘between’ + rumpere ‘to break’.
In each of the sixteenth and seventeenth centuries, in excess of 2000 loanwords

from Latin are noted in NODE, of which between a quarter and a third came
into English via French, e.g.
exterior early 16th cent.: from Latin, comparative of exter ‘outer’.

loyal mid 16th cent.: from French, via Old French loial from Latin legalis.
precarious mid 17th cent.: from Latin precarius ‘obtained by entreaty’ (from
prex, prec- ‘prayer’) + -OUS
fatigue mid 17th cent. (in the sense ‘task or duty that causes weariness’):
from French fatigue (noun), fatiguer (verb), from Latin fatigare ‘tire out’, from
ad fatim, affatim ‘to satiety or surfeit, to bursting’.
‘Latin’ is a label that occurs with high frequency in etymologies, nearly 12,000
times in COD10 (18 per cent of the headwords), and over 15,000 times in NODE.
Romance languages, i.e. those descended from Latin, that also feature in
etymologies include: French (10,500 in NODE), Italian (1,500), Spanish (1,100),
and Portuguese (300). Direct borrowings from French include: abattoir (nine-
teenth century), democrat (late eighteenth century), lacquer (late sixteenth cen-
tury, ultimately from Portuguese), tirade (early nineteenth century), voyeur (early
twentieth century) – as well as numerous food and culinary terms. Italian has
also given a number of food words, as well as much of English musical terminol-
ogy, e.g. adagio, al dente, finale, lasagne, pergola, saltimbocca, vibrato. While Spanish
and Portuguese have added words from themselves, they have also been vehicles
for words from more exotic languages, from days of colonisation and empire,
e.g. amok (via Portuguese from Malay), embargo (early seventeenth-century
Spanish), guanaco (via Spanish from Quechuan), marmalade (late fifteenth-
century Portuguese).
The other, more recent source of Latin borrowings are the neo-Latin ‘classi-
cal compounds’ (see 1.6), mostly technical terms of the sciences, technology
and medicine. Classical compounds are also formed with elements taken from
Greek. In fact, more of the ‘combining forms’ that make up classical com-
pounds are derived from Greek than from Latin. The etymology of the com-
bining form is usually given, but not necessarily of the classical compound,
unless its formation has some peculiarity, e.g. (from NODE):
hetero- from Greek heteros ‘other’

heteromorphic (a term of Biology, meaning ‘occurring in two or more
different forms, especially at different stages in the life cycle’) – no origin
given, since it is made up of the combining forms hetero- and -morph(ic),
both entered in the dictionary.
122 Etymology
heterodyne early 20th cent.: from HETERO- ‘other’ + -dyne, suffix formed
irregularly from Greek dunamis ‘power’ (here the second part is not a regu-
lar combining form).
10.4 Other loans

English has taken words from a large variety of languages from around the
world, for a number of reasons: to fill a gap in the English vocabulary, to
have a term to refer to a phenomenon not found in English culture, to provide
a ‘foreign’ flavour, and so on. Etymologies indicate at least the language
from which the loanword comes; larger dictionaries may give the meaning
in the original language, as an explanation of the motivation for the loanword.
If the donor language itself borrowed it earlier, then this may also be given.
Here, first, are some examples from European languages to illustrate these
points.
Gastarbeiter German, from Gast ‘guest’ + Arbeiter ‘worker’ (NODE)

plankton C19: via German from Greek planktos wandering, from plazesthai
to roam (CED)
csardas Hungarian csárdás from csárda ‘inn’ (COD9)
coach mid 16th cent. (in sense 3): from French coche, from Hungarian kocsi
(szekér) ‘(wagon) from Kocs’, a town in Hungary (NODE)
babushka Russian: grandmother, from baba old woman (CED)
cosmonaut C20: from Russian kosmonaut, from COSMO- + Greek nautes
sailor; compare ARGONAUT (CED)
bamboo Dutch bamboes via Portuguese mambu from Malay (COD9)
intelligentsia early 20th cent.: from Russian intelligentsiya, from Polish
inteligencja, from Latin intelligentia (NODE)
To illustrate the variable treatment of etymology, depending on the nature of

the loanword, we now give some examples of etymologies of Chinese loanwords,
as recorded in COD9:
cheongsam Chinese (the Chinese word denotes ‘a woman’s garment with

a high neck and slit skirt, worn in China’)
gung-ho Chinese gonghe ‘work together’, slogan adopted by US Marines
in 1942
kowtow Chinese ketou, from ke ‘knock’ + tou ‘head’
lychee Chinese lizhi
yin Chinese yin ‘shade, feminine, the moon’
Etymology 123
The etymology provides enough information for the reader to understand some-
thing of the motivation for the borrowing into English. The other point to note
here is that the Chinese writing system has been transliterated into Roman
characters.
Transliteration is usually provided for words from any language with a non-
Roman script, including Greek, Russian and other Slavic languages with the
Cyrillic script, Arabic, Hebrew, South Asian languages (including Urdu, Punjabi),
and so on. There are usually conventions for such transliterations, and they may
involve the use of diacritics to indicate particular features, e.g.
sepoy Urdu and Persian sipAhi ‘soldier’ from sipAh ‘army’

kosher Hebrew kAXBr ‘proper’
mango Portuguese manga via Malay mangA from Tamil mAnkAy, from mAn
‘mango tree’ + kAy ‘fruit’
sherbet Turkish Yerbet, Persian Xerbet from Arabic Xarba ‘drink’, from Xariba
‘to drink’
Finally under this heading, there follows a selection of etymologies for words
borrowed into English from around the world:
aardvark Afrikaans from aarde ‘earth’ + vark ‘pig’ (South Africa)

batik Javanese (South Asia)
cassava Taino (Caribbean)
gong Malay (South Asia)
impala Zulu (South Africa)
karaoke Japanese
kiwi Maori (New Zealand)
kayak Inuit qayaq (North America)
pariah Tamil (South India)
pampas Quechua, via Spanish (South America)
safari Swahili from Arabic safara ‘to travel’ (East Africa)
shampoo Hindi (North India)
skunk Abnaki segankw (North America)
wombat Dharuk (Australia)
yak Tibetan gyag (South Asia)
yogurt Turkish
zombie Kongo (West Africa)
124 Etymology
10.5 Historical dictionaries and some comparisons
As we might expect, an historical dictionary – in the case of English the Oxford
English Dictionary and the abbreviated Shorter OED – is more detailed in its
etymology. Examine the following entry for zenith from OED2:
a. OF. cenith (F. zénith) or med.L. cenit (cf. It. zenit, Sp. cenit, Pg. zenith, G.
zenith, etc.), obscurely ad. Arab. samt, in samt ar-rAs lit. way or path over the
head (samt way, al the, rAs head); cf. AZIMUTH (al the, sumEt pl. of samt).
This needs some interpretation, if you are not familiar with reading OED
etymologies. The initial ‘a. OF . . . or med.L.’ means ‘adopted from Old
French . . . or medieval Latin’; the modern French (F.) is given in brackets;
then cognates are given (‘cf.’ = confer ‘compare’) in Italian (It.), Spanish (Sp.),
Portuguese (Pg.) and German (G.). The word is said to be ‘obscurely adapted
(ad.) from Arabic (Arab.), and the supposed Arabic origin is explained (lit. =
‘literally’, pl. = ‘plural’). The OED makes a distinction in borrowing between
‘adoption’ (with little or no alteration to the form of the borrowed word) and
‘adaptation’ (where the spelling/pronunciation is conformed to the conven-
tions of English).
The SOED, as might be expected, given its smaller compass, presents a trun-
cated version of this etymology:
OFr. cenit (mod. zenith) or med.L cenit (also zenith), ult. f. Arab. samt in
samt-ar-ra’s ‘path over the head’: cf. azimuth.
Here, mod. = ‘modern’, ult. = ‘ultimately’, f. = ‘from’. Interestingly, CED4 has

an etymology that is as detailed, along with some variations:
C17: from French cenith, from Medieval Latin, from Old Spanish zenit,
based on Arabic samt, as in samt arrAs path over one’s head, from samt way,
path + al the + rAs head
CED has a policy of avoiding abbreviations where possible, especially in ety-

mologies, in the interests of user accessibility, a policy also followed by NODE:
late Middle English: from Old French or medieval Latin cenit, based on
Arabic samt (ar-ra’s) ‘path (over the head)’.
NODE and CED disagree on the entry of the word into English; the first
quotation in OED2 is dated 1387, followed by a 1391 quotation from Chaucer,
which would seem to support the NODE date of ‘late Middle English’. How-
ever, the Spanish part of the derivation noted in CED is supported by Skeat’s
(1961) Concise Etymological Dictionary:
Etymology 125
Zenith (F. – Span. – Arab.) M.E. senyth. – O.F. cenith; F. zénith. – Span.
zenit, O.Span. zenith. – Arab. samt, a way, road, path, tract, quarter; whence
samt-ur-ras, the zenith, vertical point of the heavens; also as as-samt, an
azimuth. Samt was pronounced semt, of which Span. zenit is a corruption;
again, samt is here short for samt-ur-ras or semt-er-ras (as above), lit. the way
overhead, from ras, the head. See Azimuth.
The specialist etymological dictionary goes into a more detailed explanation of

how the word was supposedly adapted from Arabic into English via Spanish and
French.
By way of comparison, here is the quite different etymology of car, first of all
as given in the historical dictionaries, then in the general-purpose dictionaries,
and finally in the specialist etymological dictionary:
OED: ME. carre, a. ONF. carre:-late L. carra, a parallel form to carrus, carrum
(whence It., Sp. carro, Pr. car, char, ONF. car, F. char, ME. CHAR), a kind of
2-wheeled wagon for transporting burdens. The L. was a. OCelt. *karr-os,
*karr-om, whence OIr. (also mod.Ir. and Gael.) carr masc. ‘wagon, chariot,’
OWelsh carr, Welsh càr, Manx carr, Bret. kar. (Late L. carra also gave WGer.
carra fem., in OHG. charra, Ger. karre, MDu. carre, Du. kar fem., Sw. karra,
Da. karre.)
(Note: ME = ‘Middle English’, ONF = ‘Old Northern French’, late L = ‘late

Latin’, Pr = ‘Provençal’, OCelt = ‘Old Celtic’, OIr = ‘Old Irish’, mod.Ir =
‘modern Irish’, Gael = ‘Gaelic’, Bret = ‘Breton’, WGer = ‘West Germanic’,
OHG = ‘Old High German’, MDu = ‘Middle Dutch’, Sw = ‘Swedish’, Da =
‘Danish’.)
SOED: lme. [AN, ONFr. carre f. Proto-Romance var. of L carrum neut.,

carrus masc., f. Celt. base repr. by (O)Ir. carr, Welsh car]
(Note: LME = ‘Late Middle English’, AN = ‘Anglo-Norman’, f. = ‘from’, var.

= ‘variant’, repr. = ‘represented’.)
CED: C14: from Anglo-French carre, ultimately related to Latin carra, carrum
two-wheeled wagon, probably of Celtic origin; compare Old Irish carr
NODE: late Middle English (in the general sense ‘wheeled vehicle’): from
Old Northern French carre, based on Latin carrum, carrus, of Celtic origin
SKEAT: (F. – C.) M.E. carre. – O.NorthF. carre, a car (Ducange, s.v.
Marcellum). – Late L. carra, f.; allied to L. carrus, a car; of Gaulish origin. –
Bret. karr, a chariot; W. car, O.Gael. cár, Irish carr. Allied to L. currus, a
chariot; Brugm. i. § 516.
126 Etymology
(Note: F = ‘French’, C = ‘Celtic’, s.v. = sub verbo ‘under the word’ (Ducange
refers to ‘Ducange Anglicus’, Vulgar Tongue, 1857), W = ‘Welsh’, Brugm =
‘Brugmann, Grundriss der vergleichenden Grammatik (Outline of Comparative
Grammar), 1897.)
From these examples, you can see how the etymology varies according to the
type of dictionary and its intended user group. By and large, the historical
dictionaries are aimed at scholars and students of the language, as indeed is the
specialist etymological dictionary (Skeat). The assumption of the general-
purpose dictionaries is that, within their compass, a basic set of information
about etymology is of interest to the ordinary dictionary user. It is this assump-
tion that we will now examine.
10.6 Why etymology?

It could be argued that etymology has no place in a general-purpose dictionary,
and it should be left to historical or specialist dictionaries. Learners’ dictionaries
(Chapter 11) do not contain etymological information, though its exclusion
from these dictionaries has been challenged (Ellegård 1978; Ilson 1983). Of the
three Collins dictionaries we looked at in Chapter 3 (3.2), the smallest, the
Pocket, does not contain etymologies, but the two larger ones do. It was only
half a century or so after the first monolingual English dictionary that etymolo-
gies began to be included in dictionaries (see Chapter 4); so they have a long
pedigree. Hudson (1988) – see Chapter 2 (2.4) – includes etymology among the
‘lexical facts’ about words that dictionaries should pay attention to. But there is
little evidence (Chapter 7) that users routinely resort to a dictionary for this
information. So, is there any justification for the inclusion of etymological in-
formation in general-purpose dictionaries aimed at the ordinary user?
We have noted before (7.4) that dictionaries have a double function: as
a record of the vocabulary of the language, i.e. a lexical description, and as a
reference work to meet the needs of users for information about words and
their usage. On neither count is the inclusion of etymology uncontroversially
obvious. As a record, a dictionary describes the contemporary vocabulary; it
omits obsolete words and meanings and marks as ‘archaic’ those whose currency
is beginning to wane. As a reference work, a dictionary does not have the space
to give a full account of the etymology of words, such as might be found in an
historical dictionary, as we have seen in 10.5. Moreover, the etymological in-
formation is probably the hardest of all the parts of a dictionary entry to decode,
needing as it does some background knowledge in history, and specifically in
the history of languages. Otherwise, what sense can anyone make of terms like
‘Old High German’?
Sidney Landau expresses the opinion that ‘of all the elements included in
modern dictionaries, etymology has the least to do with the essential purpose of
a synchronic dictionary’ (2001: 127). Etymology does not make a contribution
to the description of the contemporary meaning and usage of words; it may help
to illuminate how things have got to where they are now, but it is as likely to be
Etymology 127
misleading as helpful (as with the ‘etymological fallacy’). Etymology offers no
advice to one who consults a dictionary on the appropriate use of a word in the
context of a written text or spoken discourse. It merely provides some passing
insight for the interested dictionary browser with the requisite background knowl-
edge and interpretative skills. On this perspective, Landau is right: etymology
does not have the same status as other elements of lexical description in a
dictionary.
Etymology could be said to be part of dictionary information by historical
accident. The ‘hard words’ tradition (4.2), which started monolingual diction-
aries in English, included only words that had been borrowed, mostly from the
classical languages. It was only a short step to indicate systematically their lan-
guage of origin, as indeed dictionaries had done to an extent from the begin-
ning. Combined with the increasing interest in cultural and linguistic history
that flourished during the eighteenth century, etymology became firmly estab-
lished in the tradition of monolingual dictionaries. Dictionary making does
have its own tradition, its own set of principles and conventions, which are to
a large extent independent of those associated with other branches of linguistic
scholarship. It is only recently, in the last quarter of the twentieth century, that
the discipline of linguistics has exercised any major influence on the processes of
dictionary making. On this perspective, it is not surprising that etymology con-
tinues to be an element of the information given for words, at least in the larger
general-purpose dictionaries. Radical departures from the expected content and
format of dictionaries are undertaken reluctantly by publishers: purchaser ex-
pectations have to be met.
There is, perhaps, one further and sounder reason for the inclusion of etymo-
logical information in monolingual dictionaries of English at least. It arises from
the nature of the English vocabulary, which we have commented on in Chapter
2 and explored in the earlier part of the present chapter. The sources of English
words are so diverse, with such a small proportion being ‘original’ Anglo-Saxon,
and so many being ‘borrowed’ from a range of other languages, that there
would seem to be some justification for providing information at least about the
immediate origin of a word. In this way, the users of the language can see how
their vocabulary has been constituted. It is a means of celebrating the diversity
of the English lexicon, and it should guard against any temptation to linguistic
xenophobia or notions of linguistic purity.
10.7 Further reading

The ‘Guide to Using the Dictionary’ in a dictionary’s front matter will contain
brief information on how to interpret the etymology in the work concerned.
Donna Lee Berg’s A Guide to the Oxford English Dictionary (1993) contains a
section (pp. 22ff ) on the OED’s etymology.
Barbara Kipfer’s Workbook on Lexicography (1984) contains a chapter (12) on
etymology, as does David Crystal’s The Cambridge Encyclopedia of the English
Language (1995), though much of it is about place and personal names. Landau
128 Etymology
(2001) offers a perspective on etymology in dictionaries (pp. 127–34), as does
Svensén (1993), Chapter 15.
A specialist etymological dictionary is a further source for following up on
the topic of this chapter.
Dictionaries for learners 129
11 Dictionaries for learners
We introduced the four major British learners’ dictionaries in 6.6 and noted in
3.2 that such dictionaries have been at the forefront of lexicographical innova-
tion in the last half-century. In this chapter, we examine this dictionary type in
some detail and show how they have been developed to meet the perceived
needs of learners of English as a second or foreign language. Such dictionaries
are aimed at the intermediate to advanced learner. They are based on the obser-
vation that, as learners become more proficient, they need to move from a
bilingual dictionary as their lexical reference source to a monolingual diction-
ary. Monolingual learners’ dictionaries (MLDs) have therefore attempted to ful-
fil this need by providing information about the meaning and use of English
words that in many respects goes well beyond that offered in bilingual dictionaries.
11.1 Rise of the monolingual learners’ dictionaries

The genesis of the learner’s dictionary lies in the endeavours, during the inter-
war years, of three teachers of English as a foreign language, two of whom
worked in Japan (H.E. Palmer and A.S. Hornby) and the other in India (Michael
West). Not only did they attempt to improve the standard of language teaching
in their respective areas, they also became involved in research projects that had
a bearing on the task of teaching English. Michael West became a leading con-
tributor to the ‘vocabulary control’ movement (McArthur 1998, Ch. 5), which
sought to identify the essential vocabulary that would lead to a more rapid
competence in the language (West 1953; West and Endicott 1935). Harold
Palmer worked on the grammatical patterning of words, especially verbs (Palmer
1938), as later did Hornby (1954). Palmer and Hornby also investigated collo-
cations and idioms, which fed into the first general-purpose learner’s dictionary,
the Idiomatic and Syntactic Dictionary of English (Hornby et al. 1942).
The Idiomatic and Syntactic Dictionary was published in Japan, from where
Hornby was repatriated in 1941. After the war, Oxford University Press be-
came interested in the dictionary and they republished it in 1948 with the title
A Learner’s Dictionary of Current English, changed in 1952 to The Advanced Learn-
er’s Dictionary of Current English. Oxford replaced The in the title from the third
(1974) edition onwards, and it is now known by the initialism ‘OALD’. Until
its third edition, the OALD had the advanced learners’ dictionary market to
itself. It sold prolifically, the first two editions alone had sales of 7 million copies
(Hebert 1974). To illustrate the aims of the OALD, peruse the following entries
for the verb confide and the noun confidence:
con·fide/kfnsfpid/vt, vi 1 [VP14] ~ sth/sb to sb, tell secrets to sb; give

(sth or sb to sb) to be looked after; give (a task or duty to sb): He ~d his
troubles to a friend. The children were ~d to the care of the ship’s captain. She ~d to
me that . . . . 2 [VP3A] ~ in, have trust or faith in: Can I ~ in his honesty?
There’s no one here I can ~ in. con·fid·ing adj truthful; trusting: The girl is of
a confiding nature, ready to trust others, unsuspicious. con·fid·ing·ly adv
con·fi·dence/skpnfidfns/ n 1 [U] (act of ) confiding in or to. in strict ~,
expecting sth to be kept secret: I’m telling you this in strict ~. take a person
into one’s ~, tell him one’s secrets, etc. s~ man/trickster, one who
swindles people in this way. 2 [C] secret which is confided to sb: The two
girls sat in a corner exchanging ~s about the young men they knew. 3 [U] belief in
oneself or others or in what is said, reported, etc.; belief that one is right or
that one is able to do sth: to have/lose ~ in sb; to put little/complete/no ~ in sb/
sth; Don’t put too much ~ in what the newspapers say. There is a lack of ~ in the
government, People do not feel that its policies are wise. I hope he will justify
my ~ in him/my ~ that he will do well. The prisoner answered the questions
with ~.
You will notice: information about grammatical patterning, both in formulae (~

sth/sb to sb – i.e. ‘confide something/somebody to somebody’) and in coding
(VP = ‘verb pattern’, U = ‘uncountable’, C = ‘countable’); extensive use of
examples, both to illustrate grammatical patterning and to indicate typical col-
location (to put little/complete/no ~ in . . .); inclusion of set phrases (in strict ~);
fairly brief definitions, but explanation of examples where needed (The girl is of
a confiding nature, ready to trust others . . .). The examples are mostly invented
for the purpose and some of them have a rather dated ring. The verb patterns are
explained and exemplified in the front matter of the dictionary. Note also the
dots in the middle of the headwords and derivatives, to indicate ‘word division’,
i.e. where a word could be split at the end of a line of writing.
The first rival to the OALD was published by Longman in 1978, the Longman
Dictionary of Contemporary English, edited by Paul Proctor. LDOCE introduced
a number of improvements or innovations. The most significant was the use of
a restricted ‘defining vocabulary’; the lexicographers attempted to define every
sense in the dictionary using only 2000 of the more common words of English,
which were listed in an appendix. Where it proved impossible to define a sense
without using a word not from the defining vocabulary, the item appeared in
small capitals, to imply a cross-reference to its entry in the dictionary. In fact,
many of the items in the defining vocabulary had multiple meanings, and the list
also contained derivational affixes that could be added to the words in the list.
At a late stage in compilation, the dictionary text was subjected to a computer
program that checked each definition to ensure that no word outside the defin-
ing vocabulary had been used, or at least that it had been written in small
capitals.
LDOCE also tried to improve on the coding of grammatical information,
especially in respect of verb syntax. While a significant innovation in OALD,
Hornby’s verb patterns – twenty-five in all, but amounting to over fifty when
sub-patterns are included – had to be constantly looked up with the aim of
being memorised. No doubt a regular user of the dictionary would get to rec-
ognise at least the more common patterns, but the pattern numbers were not
suggestive of the pattern itself. LDOCE introduced a coding that was uniform
for verbs, adjectives and nouns, consisting of a letter plus a figure. The letter was
mnemonic where this was possible: ‘T’ stood for ‘transitive’, ‘I’ for ‘intransi-
tive’. The figure stood for different types of complement and the like: ‘Ø’ stood
for ‘zero’ (so, ‘IØ’ indicated a genuinely intransitive pattern; ‘1’ stood for ‘noun
(phrase) or pronoun’, ‘6’ for ‘that-clause’, and so on. A table of codes was in-
cluded in the inside back cover for easy reference. The aim was to help the
learner understand the coding by making it more suggestive and accessible.
Research on user behaviour (e.g. Béjoint 1981) indicated, however, that few
students made use of, or even understood, the coding schemes in their diction-
aries, preferring to glean grammatical information from the examples.
As the second edition of LDOCE was published nine years later, in 1987, a
third MLD appeared on the market, with many significant innovations: Collins
COBUILD English Dictionary. The COBUILD project was instigated by John
Sinclair, the Professor of English Language at the University of Birmingham,
with the sponsorship of the Glasgow-based publisher William Collins and Sons
Ltd. The aim was to compile a learners’ dictionary on the basis of a computer
corpus of texts – the Collins (CO)/Birmingham University (BU) International
Language Database (ILD). The corpus available to the lexicographers of the first
edition of COBUILD amounted to 7.3 million words of text, with an addi-
tional 13 million words in a ‘reserve’ corpus. Since renamed ‘The Bank of
English’, the corpus now runs to over 400 million words. The use of a large
corpus not only allowed the lexicographers to ascertain reliable information
about the relative frequency of occurrence of words and senses, but more im-
portantly to obtain data, in the form of concordances, for deciding on the senses
and meanings of words. The use of a corpus was not only new, it was revolu-
tionary: all MLDs and most NSDs now claim to make use of corpus techniques
in the compilation of their dictionaries.
COBUILD was not just the first dictionary to be based on a computer cor-
pus; it innovated in a number of other ways as well. First, all the definitions are
complete sentences; they are intended to sound like the teacher explaining the
meaning in the classroom, and they give some idea of typical contexts, e.g.
joyride If someone goes on a joyride, they steal a car and drive around in it
at high speed.
jukebox A jukebox is a record player in a place such as a pub or a bar. You
put a coin in and choose the record you want to hear.
junk You can use junk to refer to old and second-hand goods that people
buy and collect.
Second, all the examples are from the corpus – ‘real English’ – sometimes with
minor adaptation or truncation. Third, the grammatical information is not
included in the main entry, but provided in an ‘extra column’, to the right of
the main column; this column also includes information about synonyms and
antonyms. Fourth, there is only one entry per spelling, and senses are listed in
frequency order; all inflections are given, whether regular or irregular. Each
sense begins a new paragraph, and nearly all senses have at least one example.
The last MLD to enter the market did so in 1995, the ‘year of the dictionar-
ies’, in which OALD published its fifth edition, LDOCE its third and COBUILD
its second. The new dictionary was the Cambridge International Dictionary of Eng-
lish, edited by Paul Proctor, who had been responsible for the first edition of
LDOCE. CIDE took the opposite decision on headwords to COBUILD: each
major sense has a separate entry, followed where appropriate by a ‘guide word’
to the meaning; for example job has six entries: job employment, job piece
of work, job duty, job problem, job example, job crime. Every grammatical
pattern is illustrated by an example, and examples also show typical collocations.
Indeed the dictionary pays a lot of attention to the phraseological potential of
words, and it includes an extensive ‘Phrase Index’, in which phrases are entered
under all of their constituents, each of which has a reference to the page, col-
umn and line number where it is treated in the dictionary. The ‘International’
in the title is justified on the one hand by its treatment of American and Austral-
ian, as well as British English, and on the other by its tables of ‘false friends’ for
some sixteen languages, including Japanese, Korean and Thai. The latter derive
from an analysis of the Cambridge Learner Corpus, a corpus of learners’ English;
the main dictionary is based on the 100 million-word Cambridge Language
Survey corpus.
The EFL market is a lucrative one for publishers, and an advanced MLD is
only one publication among many, including course books, grammars, readers
and so on, which serve the needs of learners and their teachers. The competi-
tion has been an incentive to improve and innovate, and as successive editions
have appeared, a clear development can be perceived. Moreover, MLD lexico-
graphy has been extensively debated both by practising lexicographers (e.g.
Rundell 1998) and by academics (e.g. Herbst and Popp 1999), with increasing
attention being paid to the needs of learners and the reference skills that they
can be expected to possess.
11.2 Learners’ decoding needs

One of the major differences between NSDs and MLDs is that the latter take
into account users’ encoding needs (in writing and speaking) to an extent that
NSDs do not (see 11.3). Users of MLDs have the same decoding needs that
NSD users have – looking up the meaning of unfamiliar words or senses – and
may experience more difficulty in locating the information, as well as in under-
standing the definitions once the appropriate one has been found. We will
discuss solutions to these two potential problems.
A dictionary look-up for decoding usually involves finding the appropriate
sense of a word that has been encountered in writing and that cannot be inter-
preted from its context. In this case, the user knows the spelling of the word; but
if a word has been heard but not seen, there may be a difficulty in relating sound
to spelling and so locating the word in the dictionary. To address this particular
need, LDOCE2 (1987) included a laminated card with a list of sound-spelling
correspondences on one side (the other side contained a table of the grammar
codes). In most cases, however, the user will have the orthographic form for
looking up. The difficulty that is then likely to arise relates to identifying the
appropriate sense of words with multiple senses; the more common the word,
the larger the number of senses it may have. To facilitate the learner’s look-up
in such cases, a number of the MLDs have attempted to provide easier access to
sense differentiation. We have noted already the CIDE innovation of multiple
headwords for a lexeme, accompanied by a ‘guideword’. LDOCE3 and OALD6
also offer similar solutions under the one headword, e.g. for the lexeme stamp:
CIDE: stamp letter, stamp foot, stamp mark, stamp quality.

LDOCE3: stamp1 n 1 MAIL, 2 TOOL, 3 the stamp of sth, 4 PAY-
MENT, 5 TAX, 6 IN A SHOP, 7 a man/woman of his/her stamp; stamp2
v 1 FOOT, 2 stamp your foot, 3 stamp your feet, 4 MAKE A MARK,
5 stamp on sb/sth, 6 AFFECT SB/STH, 7 stamp sb as sth, 8 MAIL.
OALD6 : stamp noun ON LETTER/PACKAGE 1, PRINTING TOOL
2, PRINTED DESIGN/WORDS 3, PROOF OF PAYMENT 4, CHAR-
ACTER/QUALITY 5, 6, OF FOOT 7; verb FOOT 1, WALK 2, PRINT
DESIGN/WORDS 3, SHOW FEELING/QUALITY 4, 5, ON LET-
TER/PACKAGE 6, CUT OUT OBJECT 7.
The aim is that the user should be able to glance down the entry and quickly
find the sense relevant to their look-up by relating the guideword to the context
in which the word being looked up is situated.
The other problem identified in relation to decoding concerns understand-
ing the definition that is encountered. We noted earlier that LDOCE innovated
with a specified restricted defining vocabulary, though it must be said that
the original OALD compilers were also aware of the need to define within the
supposed vocabulary of the users. The current editions of LDOCE, OALD and
CIDE all have a specified defining vocabulary, which is listed in an appendix to
the dictionary. COBUILD’s solution is to define using full sentences, a practice
that is used in some instances by the other MLDs. Compare the definitions for
the noun knuckle:
OALD6: any of the joints in the fingers, especially those connecting the
fingers with the rest of the hand – picture at BODY
LDOCE3: the joints in your fingers including the ones where your fingers
join your hands
COBUILD3: Your knuckles are the rounded pieces of bone that form
lumps on your hands where your fingers join your hands, and where your
fingers bend.
CIDE: one of the joints of the fingers, esp. between the hand and the
fingers . . . PIC Body.
You will notice that two of the dictionaries refer the user to an illustration
(picture); LDOCE also contains pictures, though COBUILD does not. The
pictures are usually line drawings dispersed throughout the text, though LDOCE3
and OALD6 also contain full-page colour plates. Pictures supplement the verbal
definitions, especially for nouns with a concrete reference, and they are often
grouped (e.g. under ‘body’) so that the terms for a lexical field are displayed
together.
Now consider the definitions for the verb smear:
OALD6: to spread an OILY or soft substance over a surface in a rough or

careless way
LDOCE3: to spread a liquid or soft substance over a surface, especially
carelessly or untidily
COBUILD3: If you smear a surface with an oily or sticky substance or
smear the substance onto the surface, you spread a layer of the substance
over the surface.
CIDE: to spread (a thick liquid or a soft sticky substance) over a surface.
All the definitions have the essential components of ‘spread’, ‘liquid/soft sub-
stance’ ‘over surface’; OALD6 and LDOCE3 also include a component of ‘care-
lessly’. Note the capital letters for oily in the OALD6 definition, because it is not
in its defining vocabulary; COBUILD has no such restriction on using it. Note,
too, the use of brackets in the CIDE definition, in the conventional manner, for
indicating typical collocations, in this case as object of the verb. The COBUILD3
definition is rather cumbersome because it is also indicating the typical patterns
for the verb (smear a surface with a substance/smear a substance on a surface),
which are indicated separately in the other dictionaries, either in formulae (~ sth
on/over sth | ~ sth with sth – OALD6) or in examples (. . . smeared the walls of
their cells with excrement – CIDE).
Let us now examine the definitions for the abstract adjective versatile:
OALD6: (approving) 1 (of a person) able to do many different things. 2 (of
food, a building, etc.) having many different uses
LDOCE3: approving 1 good at doing a lot of different things and able to
learn new skills quickly and easily. 2 having many different uses
COBUILD3: 1 If you say that a person is versatile, you approve of them
because they have many different skills. 2 A tool, machine, or material that
is versatile can be used for many different purposes.
CIDE: able to change easily from one activity to another or able to be used
for many different purposes.
All apart from CIDE separate out the use of versatile to refer to people as against
things, though LDOCE is not as explicit as OALD and COBUILD. However,
all four dictionaries contain examples that make the distinction clear, though
the range of ‘things’ to which versatile may be applied is not clearly stated:
OALD suggests ‘food’ and ‘buildings’, while COBUILD specifically mentions
‘tool’, ‘machine’ and ‘material’. OALD and LDOCE have the attitudinal label
‘approving’, which is incorporated into the definition in COBUILD and not
mentioned in CIDE. This, together with the preceding sets of definitions, gives
some impression of how successful MLDs have been in making definitions
understandable to their learner-users.
11.3 Learners’ encoding needs

For decoding a learner is as likely to consult a bilingual dictionary, but for
encoding the MLD will prove to be a more comprehensive and reliable source
of information. MLDs have made it their business to provide extensively for
their users’ encoding requirements, especially in writing. There are two main
ways in which they have done this, together with some more minor additional
information.
The first way has been to provide comprehensive grammatical information
(cf. Bogaards and van der Kloot 2001), so that users can construct syntactically
natural sentences in English. For nouns, this essentially means recording the
distinction between ‘countable’ and ‘uncountable’ uses; the abbreviations ‘C’
and ‘U’, used by Hornby et al. in OALD1, have become common symbols for
this distinction. For adjectives, the inflectional possibilities need to be indicated
(i.e. whether an adjective is gradable or not), as well as any restrictions on the
syntactic positioning of adjectives (e.g. attributive only or predicative only).
Compare the entries for mere:
OALD6: [only before noun] (superlative merest, no comparative)

LDOCE3: [only before noun, no comparative]
COBUILD3: merest Mere does not have a comparative form. The super-
lative form merest is used to emphasize how small something is, rather
than in comparisons. ADJn
CIDE: [not gradable]
COBUILD has the most extensive and explicit explanation; ‘ADJn’ (in the
Extra Column) indicates that mere is restricted to attributive position, signalled
by ‘only before noun’ in OALD and LDOCE. CIDE is the least informative;
it does, however, have a separate entry for merest, and all its examples show
attributive use only.
The most important grammatical information for encoding is given for verbs,
since they are the pivotal element of sentences and to a large extent determine
the syntax of the clause or sentence in which they occur ( Jackson 2002). This is
the area that Hornby and his colleagues paid particular attention to from the
beginning of the development of the learner’s dictionary. The crucial question
is how to display this information. The initial solution (OALD1, OALD2) was
by means of coding, supported by examples; these were in turn supplemented
by formulae (OALD3) – compare the entries for propose from OALD2 and
OALD3:
OALD2: v.t. & i. 1. (VP 1, 2, 11, 17B) offer or put forward for considera-
tion, as a suggestion, plan, or purpose: I ~ an early start (to start early, that
we should start early, starting early). We ~ leaving at noon. The motion was ~d by
Mr X and seconded by Mr Y. ~ a toast (sb.’s health), ask persons to drink
sb.’s health or happiness. 2. (VP 1, 21) offer marriage (to sb.): Did he ~
(marriage) to you? 3. put forward (sb.’s name) ( for an office, for member-
ship of a club, etc.): I ~ Mr Smith for chairman. Will you please ~ me for your
club?
OALD3: vt, vi 1 [VP6A,D,7A,9] offer or put forward for consideration,
as a suggestion, plan or purpose: I ~ starting early/an early start/to start early/
that we should start early. We ~ leaving at noon. The motion was ~d by Mr X
and seconded by Mr Y. ~ a toast/sb’s health, ask persons to drink sb’s
health or happiness. 2 [VP6A,2A] ~ (marriage) (to sb), offer marriage. 3
[VP14] ~ sb (for sth), put forward (sb’s name) for an office/for member-
ship of a club/etc: I ~ Mr Smith for chairman. Will you please ~ me for your
club?
The Verb Patterns were dropped from OALD after the third edition. In the
latest (sixth) edition, the formulae are linked to examples, so that coding and
exemplification work together:
OALD6: verb
SUGGEST PLAN 1 ( formal) to suggest a plan, an idea, etc. for people to
think about and decide on: [VN] The government proposed changes to the voting
system. 䉫 What would you propose? 䉫 [Vthat] She proposed that the book be
banned. 䉫 (BrE also) She proposed that the book should be banned. 䉫 [VNthat]
It was proposed that the president be elected for a period of two years. 䉫 [V-ing]
He proposed changing the name of the company. 䉫 [VNtoinf] It was proposed to
pay the money from public funds. HELP This pattern is only used in the passive.
INTEND 2 to intend to do sth: [Vtoinf ] What do you propose to do now? 䉫
[V-ing] How do you propose getting home?
MARRIAGE 3 ~ (sth) (to sb) to ask sb to marry you: [V] He was afraid
that if he proposed she might refuse. 䉫 She proposed to me! 䉫 [VN] to propose
marriage.
AT FORMAL MEETING 4 [VN] ~ sth| ~ sb (for/as sth) to suggest sth
at a formal meeting and ask people to vote on it: I propose Tom Ellis for
chairman. 䉫 to propose a motion (= to be the main speaker in support of an
idea at a formal debate) – compare OPPOSE, SECOND
SUGGEST EXPLANATION 5 [VN] ( formal) to suggest an explanation
of sth for people to consider SYN PROPOUND: She proposed a possible
solution to the mystery.
IDM propose a toast (to sb) | propose sb’s health to ask people to
wish sb health, happiness and success by raising their glasses and drinking:
I’d like to propose a toast to the bride and groom.
You can see the whole variety of means in this entry by which grammatical
and other pattern information is being communicated: formulae ( [Vthat],
~ sb (for/as sth) ), examples (She proposed that the book be banned), and phrases
(propose a motion). Coding and examples work together in an even more
integrated manner in CIDE:
propose (obj) SUGGEST v to offer or state (a possible plan or action) for

other people to consider • I propose that we wait until the budget has been
announced before committing ourselves to any expenditure. [+ that clause] He
proposed dealing directly with the suppliers. [+ v-ing] She proposed a boycott of the
meeting. [T] • He proposed a motion that the chairman resign. [T] • To propose
someone is to suggest them for a position or for membership of an organi-
zation: To be nominated for union president you need one person to propose you and
another to second you. [T] • If you propose (to a person) you ask someone to
marry you: I remember the night your father proposed to me. [I] o She felt sure he
was going to propose. [I]
The coding formulae are contained in square brackets at the end of the exam-
ples, and the example sentences contain bold items that also indicate grammati-
cal patterning. In COBUILD, as indicated earlier, the whole-sentence definitions
contribute towards identifying the grammatical patterns in which the word
typically occurs with the sense being defined. Additionally, the Extra Column
contains more explicit coding for the syntactic operation of words, e.g. for
propose, V n/-ing, V that, V to-inf, V to n, V, V n to n.
The second main way in which MLDs provide encoding information for
learners is in respect of lexical patterning, specifically collocations, idioms, and
other types of phraseology. In COBUILD, the definitions again have the task of
indicating typical collocational patterns. Consider the following definitions for
propose and proposition:
If you propose a theory or explanation, you state that it is possibly or

probably true, because it fits in with the evidence that you have considered.
If you describe something such as a task or an activity as, for example, a
difficult proposition or an attractive proposition, you mean that it is
difficult or pleasant to do.
In the case of the propose definition, it indicates that the subject of propose is a
person (by the use of you), and that the object is either the words theory or
explanation, or something that counts as either of these. In the case of the pro-
position definition, the adjectives difficult and attractive are indicated as typical
collocates for this sense of the noun. In CIDE, collocations can be indicated in
the conventional way by means of brackets (see the entry for propose above), but
more usually by using bold type in the examples, e.g. for malaise:
They claim it is a symptom of a deeper and more general malaise in

society. • They spoke of the feeling of moral and spiritual malaise, the lack
of will to do anything. • They were discussing the roots of the current
economic malaise.
These examples show that typical adjectives accompanying malaise include deep,
general, spiritual and economic, and that it enters into the phrase a symptom
of . . . malaise. OALD6 has a ‘study page’ on collocations; it, too, relies on the
examples to indicate typical collocations. For malaise it has:
economic/financial/social malaise, a serious malaise among the staff
For sample, the examples are:
The interviews were given to a random sample of students. The survey

covers a representative sample of schools. a sample survey. a blood sam-
ple. Samples of the water contained pesticide. ‘I’d like to see a sample of
your work,’ said the manager. a free sample of shampoo. sample exam
papers.
The items in bold type represent ‘important collocations’. LDOCE shows

collocations (and fixed phrases) in bold type within an entry, followed by an
explanation or example, or both. In the entry for door, for example, the follow-
ing are given:
open/close/shut/slam the door, knock on/at the door, kitchen/bathroom/

bedroom etc door, front/back/side door, revolving/sliding/swing doors,
at the door, answer the door, show/see sb to the door, two/three doors
down etc, (from) door to door, out of doors, behind closed doors, show sb
the door, lay sth at sb’s door, be on the door, an open door policy, open
doors for sb, open the door to, shut/close the door on, at death’s door.
MLDs take both grammatical and lexical patterning seriously and they have
come a long way in their treatment of these areas since Hornby and his col-
leagues identified them as the major encoding needs of learners of English.
Learners’ encoding needs are taken account of by two further types of infor-
mation provided by MLDs. The first of these is the explicit indication of sense
relations (2.3.3) such as synonymy and antonymy. COBUILD has been espe-
cially prolific with this kind of information, though more so in the first edition
than in the later ones; the first included information on hyponymy, not subse-
quently included. In COBUILD the information on sense relations is given in
the Extra Column by means of the symbols ‘=’ (for synonyms) and ‘≠’ (for
antonyms). For example, heavy as in a heavy meal is marked in COBUILD3 with
‘= filling’, ‘≠ light’; and as in the air is heavy, it is marked with ‘= oppressive’, ‘≠
cool, fresh’. OALD6 marks synonyms with SYN and antonyms with OPP, e.g.
impute has SYN ATTRIBUTE, and left-winger has OPP RIGHT-WINGER;
but OALD6 is more sparing with this information than COBUILD.
The second additional type of encoding information comes in the form of
usage notes of various kinds. Some of this is in the form of labelling, as in NSDs,
though sometimes with a little variation, e.g. OALD6 has the symbol of an
exclamation mark in a triangle to warn users that the word or sense is slang or
taboo. COBUILD2 has the term PRAGMATICS in the Extra Column to
show that usage information is shown within the definition; in COBUILD3 the
definition no longer contains this information and a specific label such as ‘dis-
approval’ or ‘informal’ has been substituted for PRAGMATICS in the Extra
Column. The other dictionaries have ‘usage notes’ (LDOCE) of various kinds,
as well as more extensive discussion on ‘study pages’ (OALD6) or in ‘language
portraits’ (CIDE). OALD6, for example, has scattered through the dictionary
boxed items entitled ‘Vocabulary Building’ (e.g. ways of saying approximately),
‘Which Word?’ (e.g. as/like), ‘British/American’ (e.g. already/just/yet), ‘Gram-
mar Point’ (e.g. avenge/revenge), ‘Word Family’ (e.g. clear – clarity – clarify),
‘More About’ (e.g. of course). There is a recognition that learners need a
range of information about words – grammatical, semantic and pragmatic –
in order to be able to construct accurate and appropriate sentences in the
target language.
11.4 Additional information

Some of the information mentioned in the previous paragraph goes beyond that
strictly required for encoding. It is serving to enhance the learner’s knowledge
and understanding of the vocabulary of English in a wider sense, not just for the
specific task that may have occasioned the look-up. Some of the information of
this kind is cultural, putting words into a context that enhances understanding
of them. In CIDE, for example, there is a boxed article entitled ‘WORK’ in the
appropriate part of the dictionary, which discusses ‘some common words and
expressions we use in everyday conversation to talk about the work we do,
leaving work, being out of work and looking for a job’ (p. 1681), including
differences between British and American English.
Some of the additional information is dispersed through the dictionary, near
to relevant words. Other types are collected together, either in appendices or in
groups of pages inserted at some point in the dictionary. OALD6 has a set of
eight colour plates (between pages 372 and 373) of sets of objects (bread, cakes
and desserts; fruit and vegetables; clothes and fabrics; the animal kingdom;
games and toys); a set of sixteen ‘study pages’ (between 756 and 757), dealing in
part with grammatical and lexical matters and in part with letter and CV writing;
and an eight-page set of colour maps (between 1140 and 1141). Additionally,
OALD6 contains appendices dealing with: irregular verbs; geographical names;
numbers; punctuation; the language of literary criticism; an index to the usage
notes; and the defining vocabulary.
The use of computer corpora has enabled lexicographers to obtain fairly
reliable data on the frequency of occurrence of words and senses. This informa-
tion has informed the design of MLDs since COBUILD1. From the beginning,
COBUILD has included this information in the dictionary itself by marking
words with a set of five diamonds. If all five diamonds are black, this indicates
that the word belongs to the most frequent 700 in the language (e.g. main,
paper); if four are filled in, the word belongs to the next 1200 most frequent (e.g.
maker, management); if three are black, then it belongs to the next 1500 most
frequent words (e.g. panel, panic); if two are black, then the word belongs to the
next 3200 most frequent (e.g. loyalty, lounge); and if only one is black, then the
word belongs to the next 8100 most frequent (e.g. malt, mandatory). So the
diamond markings account for the 14,700 most frequent words according to
the Bank of English corpus. The top two bands (1900 words), it is claimed,
‘account for 75% of all English usage’ (COBUILD2, p. xiii).
LDOCE3 also gives frequency information, but differentiates between oc-
currence in spoken English and in written English; and it accounts for only the
3000 most frequent words in each mode. The frequencies are indicated by the
letters ‘W’ (for written) and ‘S’ (for spoken), followed by a numeral between ‘1’
and ‘3’: ‘1’ indicates within the 1000 most frequent, ‘2’ within the next 1000
most frequent, and ‘3’ within the next 1000 most frequent. For example, com-
mon is marked S1, W1; commitment S2, W2; compete S3, W3; committee S3, W1;
comment S1, W2; comparison S3, W2; compensation W3 only; complicated S2 only.
This information is of interest to an advanced learner, and of particular use to
teachers and course designers when considering how to sequence the introduc-
tion of vocabulary items. MLDs have developed far beyond their original con-
ception in the 1930s and 1940s, not only in the range of information that they
offer the learner, but also in the attention to the learner’s needs and reference
skills. The next development offers yet more.
11.5 MLDs on CD-ROM
All the four MLDs discussed in this chapter are available in CD-ROM format.
In general, the MLDs have exploited the possibilities of the electronic medium
rather more extensively than the NSDs (see 6.7), with additional features, cross-
referencing and searching. For all except COBUILD the CD-ROM is available
packaged together with the print dictionary, at a slightly higher price for the
package than for the print version alone, though in the case of OALD it is a
severely cutdown version of a separately available CD-ROM (‘Text and Sound’
only). Each CD-ROM offers a different set of features, and where applicable
they have been developed from earlier versions: Heuberger (2000), for exam-
ple, reviews electronic MLDs published in the late 1990s, whereas those re-
viewed here came out in 2000 or 2001.
LDOCE on CD-ROM, published in 2000 and based on LDOCE3, contains
four sets of data, each with its own index: the dictionary, verb conjugations,
pictures, tables (e.g. of numbers, word formation, geographical names). The
dictionary index contains the A–Z headword list, and the picture index contains
all the words that are illustrated by or within a picture. The dictionary entries
and the words in the pictures are interlinked, by a camera icon in the dictionary
entries, and by clicking on the words in the pictures. Similarly, the dictionary
entries and verb conjugations are interlinked, so that if a user is unsure about the
appropriate inflectional form of a verb, they can click on the ‘V’ icon in the
dictionary entry for the verb.
Typing in a word in the ‘choice’ box takes you automatically to the appro-
priate part of the list. When a dictionary entry is selected, the pronunciation of
the headword is given; it may also be selected by clicking on the speaker icon;
however, only British pronunciation is given. Clicking on a further icon, a
white cross in a white circle, gives a list of ‘related words’, i.e. headwords that
contain the item concerned; for example, the related words given for support
include supporting, child support, life support system, price support, support group.
LDOCE on CD-ROM offers two further types of search, by activating the
‘Search’ item hidden in the ‘Book’ menu. One is a ‘text’ search, which allows
searching for up to three words, joined by Boolean operators (and, not, or); you
can search for a word (or words) in definitions (e.g. insect, to find all the ‘insect’
words) or for a label (e.g. ‘determiner’ or ‘slang’). The other search is a ‘head-
word’ search, which allows the use of wildcards (‘?’ for any letter, ‘*’ for any
number of letters, including zero): ‘*gry’ would find all the words ending in gry
(only angry and hungry!), ‘?oo?’ would find all the four-letter words with double
‘o’ in the middle (289 in LDOCE).
OALD on CD-ROM, published in 2001 and based on OALD6, contains a
number of additional features. However, few of these are operational in the
‘Text and Sound’ version packaged with the print dictionary, though they are
demonstrated. Like most CD-ROM dictionaries, the headword list is on the
left of the screen, including a box in which a desired headword can be typed,
with a larger window for the dictionary text on the right-hand side. Selection of
a headword gives, in the headword list, the compounds that contain the word,
followed by words whose definitions include it. Clicking on the appropriate
speaker icon in the dictionary entry gives either British or American pronuncia-
tion of the headword. An ‘Advanced Search’ facility allows the selection of up
to three words, joined by Boolean operators. Additionally, the search may be
narrowed by selecting one of the ‘Types’ and/or one of the ‘Filters’. A ‘type’
narrows the search to a particular type of information in the dictionary, e.g.
headwords, idioms, collocates, definitions, examples. The ‘filters’ largely relate
to labelling in the dictionary, e.g. for part-of-speech, register, and geography
(British vs. American).
Additional features on the CD-ROM are accessed by clicking on a tab at the
top of the screen. The tabs are labelled: 3-D Search, Pictures, Maps, Exercises,
Games, Extras. All of these, apart from the first and the last, are self-explanatory.
The ‘Extras’ section contains a lot of the extra material that is distributed through
the print dictionary, such as the Guide to Using the Dictionary, the Topics
pages, the Language Study pages, and so on. The really interesting and innova-
tive feature is the ‘3-D Search facility’. This gives access to a set of ‘spider’
diagrams representing lexical fields (words that share the same ‘area of meaning’
– see Chapter 12), so that a user can locate the word they are looking up in
relation to other closely related words in the vocabulary. At the least, this is
useful vocabulary building information.
COBUILD on CD-ROM was published in 2001 and is based on COBUILD3,
replacing an earlier version (1995) based on COBUILD2. However, the CD-
ROM contains more than just the dictionary; it is an integrated resource with
dictionary, thesaurus, grammar book, usage manual, and a 5-million ‘Wordbank’
drawn from the Bank of English to supply examples. When a word is typed in
for look-up, the index points to all the sections of the CD-ROM in which it
features, so that a range of information can be accessed from a single index.
Indeed the index has two forms: ‘Entries’ and ‘Full Text’. ‘Entries’ simply has
the sections in which the word concerned occurs, i.e. dictionary, thesaurus,
grammar, usage. ‘Full Text’ indexes are: ‘Headings’, i.e. headwords, including
compounds, in which the item occurs; ‘Explanations’, i.e. headwords whose
definitions contain the item; ‘Sample Lists’ from the grammar that contain the
item; ‘Examples’, i.e. entries from the dictionary, usage, grammar and wordbank
that contain the item; ‘Synonyms’, in the dictionary and thesaurus; and ‘Anto-
nyms’ in the dictionary. Here is a wealth of interlinked information for explor-
ing the meaning and use of a word and for finding guidance on employing it
appropriately in writing.
Within the dictionary entries, each form (inflection) of a word has a speaker
icon for activating its (British only) pronunciation. There are also buttons allow-
ing both browsing through the dictionary entries, either forwards or backwards,
as well as retracing the search history that a user has undertaken. As with other
CD-ROM dictionaries, double-clicking on any word in the definitions or
examples links to the entry for that word. The major feature of COBUILD on
CD-ROM is the integration of the various lexical and grammatical information
sources.
CIDE on CD-ROM was published in 2000 and is based on the first (1995)
edition of the print dictionary. However, in many ways it is the most sophisti-
cated of the CD-ROM dictionaries in the extent to which it goes beyond the
print version in the exploitation of the electronic medium. It has two separate
windows, which can be individually manipulated for size and position on the
screen: one is the ‘Search Panel’, containing the indexes and various options for
searching; the other is the ‘Content Window’, containing the dictionary en-
tries, as well as sets of Exercises, Pictures, and Study Sections (e.g. on grammar,
cultural information, word building, letter writing, and so on). Entering a word
in the ‘Find’ box on the Search Panel automatically creates an index of related
words, e.g. for support: support, supportable, supported, support group, supporting,
supportive, supportively, supportiveness, support system (those in bold are main
entries, the others sub-sentries within a main one). Clicking on any of these
activates a more extensive index, which includes all the headwords for the item,
followed by open compounds containing it, followed by headwords whose
definitions contain it. Clicking on any of these brings the appropriate dictionary
entry into the Content Window. Both British and American pronunciation can
be activated.
Every entry in CIDE on CD-ROM is supplied with a link labelled ‘Related
words’. Clicking on this link activates the highly innovative feature of this CD-
ROM dictionary, a differentiated and categorised lexical field analysis of the
vocabulary. For example, clicking on the ‘Related words’ link for support (BEAR)
offers three lexical fields: ‘Psychology, Psychiatry and Psychoanalysis’, ‘Allow-
ing and permitting’, and ‘Tolerating and enduring’. Selecting the last of these
activates in the Search Panel a list of other verbs with a similar meaning, fol-
lowed by words from other parts of speech, as well as phrases and expressions
(e.g. take it on the chin). The lexical field analysis can be viewed in the Search
Panel, which allows for a field to be selected and all the words and expressions
in the field to be listed. The vocabulary is analysed at the most general level into
17 broad categories:
1 art and entertainment

2 building and civil engineering
3 clothes, belongings and personal care
4 communication
5 education
6 finance and business
7 general/abstract
8 history
9 life, death and the living world
10 light and colour
11 movement and location
12 religion
13 science
14 society
15 sports, games and pastimes
16 thinking and understanding
17 war and the military.
Each of these has further sub-divisions and sub-divisions of sub-divisions. With

this facility, CIDE on CD-ROM not only provides an aid for a learner’s vo-
cabulary building, but, lexicographically, bridges the gap between alphabetical
dictionary and thematic thesaurus (see Chapter 12).
CIDE on CD-ROM’s Search Panel allows other differentiated searches to be
undertaken: by ‘Part of Speech’, by ‘Label’ (including geographical and regis-
ter), by ‘Grammar’, by ‘Category’ (i.e. of text – Headword, Idiom, Definition
text, Example text, Usage notes, etc.), and by ‘Frequency’. The ‘Grammar’
search allows specification of grammatical features or structures indicated in the
dictionary; for example, ‘+ object + that clause’ finds all the verbs (42 in CIDE)
that are followed by an Object and a that-clause, ‘+ two objects’ all the ditransitive
verbs (152 in CIDE), ‘after verb’ all the predicative only adjectives (322 in
CIDE), and ‘not gradable’ all the non-gradable adjectives (3202 in CIDE). The
‘Frequency’ search has categories from ‘Rare’ to ‘Very Common’, along with
‘Defining Vocabulary’; selecting ‘Very Common’ lists the 611 words so desig-
nated, and ‘Common’ the 3181 words with this frequency. Information of this
kind is invaluable to the teacher and course writer, and certainly of interest to
the advanced student.
All the MLDs on CD-ROM have in different ways begun to exploit the
electronic medium for extending what they can offer to users of learners’ dic-
tionaries, CIDE more so than the others. But the medium has yet to be ex-
ploited to the full ( Jehle 1999).

The development of learners’ dictionaries, up to the ‘third generation’ (OALD4,
LDOCE2 and COBUILD1), is told in Tony Cowie’s English Dictionaries for
Foreign Learners: A History (1999). The process of compiling COBUILD1 is
reported by some of those involved in Looking Up: An Account of the COBUILD
Project (1987), edited by John Sinclair.
Articles on the 1995 generation of MLDs are contained in The Perfect Learn-
ers’ Dictionary(?) (1999) edited by Thomas Herbst and Kerstin Popp. Reinhard
Heuberger reviews both print and CD-ROM versions of MLDs in his doctoral
thesis, published as Monolingual Dictionaries for Foreign Learners of English (2000).
The Hausmann et al. (1989–91) International Encyclopedia of Lexicography con-
tains a number of articles on learners’ dictionaries. A seminal contribution is
Michael Rundell’s (1998) article, ‘Recent Trends in English Pedagogical Lexi-
cography’, in the International Journal of Lexicography.
Abandoning the alphabet 145
12 Abandoning the alphabet
If you look up the word dictionary in a dictionary, you will find a definition
along the lines of:
a book that consists of an alphabetical list of words, with their meanings,

parts of speech, pronunciations, etymologies, etc.
(CCD4)
‘Dictionary order’ is synonymous with ‘alphabetical order’. We expect dic-

tionaries to use alphabetical ordering of their headwords, just as we expect other
reference works to do so as well, such as telephone directories, encyclopedias,
and indexes of all kinds. Because we have learnt the order of the letters in the
alphabet, it is the most convenient system for locating an item in a written list.
Our skill in using the alphabet for this purpose can be generalised to all manner
of written lists.
As a reference manual, therefore, a dictionary’s headword list is ideally
arranged alphabetically, so that users can readily access the item that they are
seeking. And it is usually a single item that is being looked up. However, we
must ask, first, whether an alphabetical ordering is best for presenting a descrip-
tion of the vocabulary as a whole, and second, whether there are some users’
needs that would be better served by an alternative arrangement of words in a
dictionary.
12.1 Disadvantages of A–Z

One of the drawbacks of an alphabetical listing is that some words that belong
together morphologically become separated. This applies, in particular, to two
kinds of relation. First, words that are derived by prefixation (see 2.2.2) are
entered separately from their root, and there is usually no indication at the entry
for the root that it has a prefixed derivative. Derivatives by suffixation are
entered either as separate headwords, but close to the root in the alphabetical
sequence, or as run-ons under the root; so that the relation between root and
derivative is clear. For example, courage and its derivative courageous come in
close proximity in the alphabetical list, but discourage and encourage are distant
and the connection is not made. The second kind of morphological link relates
to words – mostly nouns of OE origin – that have a matching word – mostly
adjectives of Latinate origin – in another word class. For example, lung (noun)
has matching pulmonary (adjective), church has ecclesiastical, mind has mental, earth-
quake has seismic, horse has equine, and so on. CED notes at the noun the ‘Related
adj.’, but dictionaries do not as a rule make the connection.
A more serious disadvantage of alphabetical ordering is the perspective that
it presents on the vocabulary as a whole. It presents an atomistic view of the
vocabulary, treating each word in isolation, the headword with its entry, and
making few of the connections that exist between words. Just like other areas of
language – phonology, grammar – the lexicon is a system, with paradigmatic
(synonymy, antonymy, hyponymy, meronymy) and syntagmatic (collocation)
relations ( Jackson and Zé Amvela 2000, Ch. 5). In lexicology, an attempt is
made to capture some of these relations in the notion of ‘semantic/lexical fields’
(Lehrer 1974; Jackson 1988: 210–16). A lexical field is a set of lexemes that are
used to talk about a defined area of experience; Lehrer (1974), for example, has
an extensive discussion of the field of ‘cooking’ terms. A lexical field analysis
will attempt to establish the lexemes that are available in the vocabulary for
talking about the area under investigation and then propose how they differ
from each other in meaning and use. Such an analysis begins to show how the
vocabulary as a whole is structured, and more so when individual lexical fields
are brought into relationship with each other. There is no prescribed or agreed
method for determining what constitutes a lexical field; each scholar must draw
their own boundaries and establish their own criteria. Much work still needs to
be undertaken in researching this approach to vocabulary. Lexical field analysis
is reflected in dictionaries that take a ‘topical’ or ‘thematic’ approach to present-
ing and describing words.
The distinction is often drawn in terms of the dichotomy between a
‘semasiological’ and an ‘onomasiological’ approach to the description of vo-
cabulary. The semasiological (from Greek semasia ‘meaning’) approach pro-
ceeds from forms (terms, words) to meanings or concepts, and it results
in traditional, alphabetically ordered dictionaries. The onomasiological (from
Greek onomasia ‘term’) approach (Kipfer 1986) proceeds from concepts to
terms, and it results in works of the thesaurus type, organised by theme or
topic. Some attempts have been made to combine the two approaches, most
notably by dictionaries of French published by Dictionnaires Le Robert, where
extensive cross-referencing to synonyms and antonyms is made within most
entries, e.g.
IMMENSE adj. 1 vx Illimité, infini. 2 Dont l’étendu, les dimensions sont

considérables. → grand, illimité, vaste. Perdu dans l’immense océan. 3 Qui
est très considérable en son genre (par la force, l’importance, la quantité).
→ colossal, énorme. Une foule immense. Une immense fortune. contr. Infime,
minuscule.
(Le Robert Collège 1997)
The arrow points to synonyms, and the abbreviation ‘contr.’ (contraire)
introduces antonyms. A similar, but less systematic attempt is reflected in the
synonym essays provided by LDEL2 and ECED, e.g.
synonyms Huge, vast, immense, enormous, mammoth, elephan-

tine, giant, gigantic, colossal, gargantuan, titanic: huge is a general
term, expressing great size, bulk, or capacity <a huge man> <huge piles of
wheat>. Vast stresses extent or range <vast distances>. Immense and enor-
mous suggest size or degree far in excess of what is usual, with immense
sometimes implying almost infinite <immense vistas of blue sky> <enormous
strength>. Mammoth and elephantine suggest the large size and unwieldy
nature of the animals they recall. Used figuratively, mammoth can mean
‘excessive’ or ‘extravagant’ <a mammoth darts tournament>. Giant and gi-
gantic suggest something abnormally large; gigantic is preferred for
figurative use <a giant doll> <a gigantic bill for repairs>. Colossal suggests
something of awesomely large proportions, while titanic implies the co-
lossal size and primitive strength of the Titans. The hugeness of gargan-
tuan is like that of Rabelais’ hero: larger than life, especially with regard to
food and appetites. antonyms tiny, minute, minuscule.
(LDEL2)
But that is about as far as it goes in conventional general-purpose dictionaries

for native speakers. Learners’ dictionaries often provide more information, at
least about synonyms and antonyms, e.g. COBUILD in its ‘Extra Column’
(Chapter 11).
12.2 The thematic tradition in lexicography

The alphabetisation of word lists goes back to the Latin–English glossaries com-
piled by scholar monks during the Old English period, but so does the arrange-
ment of vocabulary by topic (see 4.1). The most famous of the latter is Ælfric’s
Glossary, published as an appendix to his Grammar of Latin. Ælfric, who lived
from around 955 to 1020, became Abbot of Eynsham, near Oxford, in 1005; his
tasks included the teaching of Latin to English-speaking novices. The Glossary
groups Latin words with their English glosses into sets, and Werner Hüllen, in
his account of Ælfric’s work (1999: 62ff ), suggests that the sets might have the
following titles and structure:
1 God, heaven, earth, mankind

2.1 Parts of the human body
2.2 church offices
2.3 family relationships
2.4 state offices, including crafts and instruments as well as tools
2.5 negative features of human character
2.6 intellectual work
2.7 diseases, afflictions, merits
2.8 weather, universe
3 Birds
4 Fish
5 Wild animals
6 Herbs
7 Trees
8.1 Buildings (churches, monasteries), materials and objects used there
8.2 war, castles, arms, valuable materials
8.3 various
8.4 human vices.
Ælfric’s hope was to encompass the whole vocabulary in his scheme, though he
recognised that he had not done so. The topical organisation certainly betrays
the concerns of an early medieval churchman.
As dictionary making, both bilingual and monolingual, developed, the alpha-
betical tradition dominated, but, especially under the influence of the Renais-
sance, thematic wordbooks were also compiled, most famously that by the
Moravian Comenius ( Johann Amos Komensky, 1592–1670) under the title
Ianua linguarum reserata (The Gate of Tongues Unlocked), of which a Latin and a
German version were published in 1631. In English, the most famous work of
the time is that of John Wilkins, as part of his proposal for a ‘universal language’,
with the title An Essay Towards a Real Character, And a Philosophical Language,
published in 1668. As part of his proposal, Wilkins put forward a scheme for
classifying the vocabulary of any language; at its most general level, it is de-
scribed in eleven chapters in the Essay (Hüllen 1999: 253):
I the transcendentals, the general notions which determine all the sub-
sequent principles of order. They include ‘discourse’, that is ‘words’ as
opposed to ‘things’
II God, the creator, and the creation, that is the world observed collectively
III together with all the following chapters is devoted to the world observed
distributively . . . the inanimate elements under the ‘predicament’, that
is, the category of substance
IV the vegetative species
V the sensitive species
VI the significant parts of vegetative and sensitive species
VII various phenomena belonging to ‘quantity’, a category which is sub-
sumed under the category ‘accident’. So are the following four chapters.
VIII various phenomena belonging to ‘quality’
IX various phenomena belonging to ‘action’
X various phenomena belonging to ‘private relation’
XI various phenomena belonging to ‘public relation’.
Each of these broad categories is further divided and subdivided, following a

logical, philosophical scheme.
The work by Wilkins was familiar to the author of the best-known thematic
wordbook, Peter Mark Roget’s Thesaurus of English Words and Phrases, first
published in 1852 and still in print in a number of editions, the most genuine of
which is Kirkpatrick (1995). Roget was by profession a medical physician, but
with wide-ranging interests; he contributed to the Encyclopaedia Britannica and
wrote treatises on electricity and magnetism (McArthur 1992: 871). In 1849, at
the age of 70, having retired after 22 years as Secretary of the Royal Society,
Roget returned to an undertaking that had interested him for over forty years:
to create a reference work containing words ‘arranged . . . according to the
ideas which they express’:
The object aimed at in the present undertaking is . . . the idea being given,
to find the word, or words, by which that idea may be most fitly and aptly
expressed.
(Introduction)
While motivated by considerations of ‘practical utility’, Roget’s classification

scheme reaches back to the notions behind Wilkins’ ‘philosophical tables’. Roget
has six broad ‘Classes’, which are initially subdivided into ‘Sections’ (see Table 12.1).
Table 12.1
Class Section
I ABSTRACT RELATIONS I Existence

II Relation
III Quantity
IV Order
V Number
VI Time
VII Change
VIII Causation
II SPACE I Generally
II Dimensions
III Form
IV Motion
III MATTER I Generally
II Inorganic
III Organic
IV INTELLECT I Formation of Ideas
II Communication of Ideas
V VOLITION I Individual
II Intersocial
VI AFFECTIONS I Generally
II Personal
III Sympathetic
IV Moral
V Religious
Table 12.2
IV. ORDER
1. GENERAL 58 Order 59 Disorder

60 Arrangement 61 Derangement
2. CONSECUTIVE 62 Precedence 63 Sequence
64 Precursor 65 Sequel
66 Beginning 67 End
68 Middle
69 Continuity 70 Discontinuity
71 Term
3. COLLECTIVE 72 Assemblage 73 Non-assemblage. Dispersion
74 Focus
4. DISTRIBUTIVE 75 Class
76 Inclusion 77 Exclusion
78 Generality 79 Speciality
5. CATEGORICAL 80 Rule 81 Multiformity
82 Conformity 83 Unconformity
Each of the ‘Sections’ is further subdivided into the lowest level of sets of
words arranged where applicable in pairs of opposites, e.g. under Class I, Sec-
tion IV (see Table 12.2).
The sets of words and phrases are listed in the main body of the Thesaurus, on
two-column pages, under word classes, with nouns first, followed by verbs,
then adjectives and adverbs. No other information is given – no definitions,
pronunciation, or etymology; it is intended as a ‘storehouse’ or ‘treasure’ (=
Greek thesauros) of words, which a writer will plunder for the one that is apt for
their purpose.
If you are familiar with Roget’s Thesaurus, you will know that about the last
third of the work is taken up with an alphabetical ‘Index’, and many users find
this to be the most convenient route into the thesaurus. During his lifetime – he
died in 1869 at the age of 90 – Peter Mark Roget did not include an index in
any of the editions that he compiled and edited; that defeated the purpose of the
work. It was his son, John Lewis Roget, who added the Index; he also under-
took a major revision of the work in 1879 and continued to edit it until his
death in 1908. The editorship then passed to his son, Samuel Romilly Roget,
who undertook a major revision in 1936, and then sold the rights to Longman
in 1952, the year before he died (McArthur 1992: 871). Roget’s Thesaurus re-
mains as an institution among reference works for the English language and as a
monument to the thematic tradition of wordbooks.
12.3 Specialist thesauruses/thesauri

A number of compilers of modern reference books about words have chosen to
present their material in a thematic, rather than alphabetic, format. They believe
that it serves their purposes to greater effect and gives an enhanced insight into
the set of vocabulary that they are describing. We review here four such refer-
ence works. The first is A Thesaurus of Old English (Roberts et al. 1995), which
is a presentation of the vocabulary of Old English, as far as it can be gleaned
from the extant manuscripts of the period. The vocabulary is arranged in 18
broad classes:
1 the physical world

2 life and death
3 matter and measurement
4 material needs
5 existence
6 mental faculties
7 opinion
8 emotion
9 language and communication
10 possession
11 action and utility
12 social interaction
13 peace and war
14 law and order
15 property
16 religion
17 work
18 leisure.
Each of these classes is further subdivided. For example, ‘9. Language and Com-
munication’ has a general class and seven subclasses:
09 Speech, vocal utterance

09.01 To speak, exercise faculty of speech
09.02 Silence, refraining from speech
09.03 A language
09.04 Sense, purport, meaning
09.05 Curiosity
09.06 To take matter for discourse
09.07 Dispute, debate.
Under each of these is listed the modern English word or paraphrase, followed
by the OE term, e.g.
09.01.01 A speech, what is said, words: (ge)spræc, word, wordlac

A dictum, remark, observation: spell
A saying, words: cwide, word, wordcwide
A phrase: foreset(ted)nes
A formula: formala, hiw
An idiom: wise
A verse, sentence (of Bible): fers
A discourse: mæ6elcwide, mæ6elword, spræce, tospræc
A set speech: getynges
An instructive talk: spell
A thesis, proposition: betynung.
The Thesaurus provides a most insightful analysis of the vocabulary available at

this early period in the history of the English language, as well as a sober re-
minder of the words that disappeared from English as a consequence of the
Norman conquest.
The second example is a presentation of the vocabulary of a regional variety
of the language, The Scots Thesaurus (McLeod 1990). The Thesaurus presents
some 20,000 Scots words, with the focus on rural Scotland, under fifteen broad
categories:
1 birds, wild animals, invertebrates

2 domestic animals
3 water life
4 plants
5 environment
6 water, sea, ships
7 farming
8 life cycle, family
9 physical states
10 food and drink
11 law
12 war, fighting, violence
13 architecture, building, trades
14 religion, superstition, education, festivals
15 emotions, character, social behaviour.
Each category is further subdivided, e.g. 10.6 gives words for ‘Bread, Oatcakes,
etc.’ and 10.7 for ‘Cakes, Pastry, Biscuits’. Within the subdivisions, the items
are listed in alphabetical order; 10.6 contains almost 90 lexemes, some of which
are regionally restricted – the relevant areas or counties of Scotland are indi-
cated, e.g.
• luifie a kind of flat bread roll Ags [i.e. Angus]

• nickie an oatcake or bun with an indented edge chf Fif [i.e. chiefly Fife]
• rumpie a small crusty loaf or roll now Per WC [i.e. now Perthshire West
Central]
• skair scone a kind of oatmeal-and-flour scone made with beaten egg and
milk.
The third example is still under construction in the Department of English
Language at the University of Glasgow under the direction of Professor Chris-
tian Kay: the Historical Thesaurus of English. Begun in 1964 by Kay’s predecessor,
Professor M.L. Samuels, the Historical Thesaurus is based on the materials of the
OED, supplemented by further research. It aims to present the vocabulary of
English from the earliest written records onwards in a semantic and chronologi-
cal arrangement, so that a user can see how the vocabulary of English has devel-
oped in any particular area of meaning. As with all thematic dictionaries, its
effectiveness depends in large part on its classification system:
The classification which has resulted from examination of the data is based
on a modified folk taxonomy. There are three major divisions: (I) The
World, including the physical universe, plants and animals; (II) The Mind,
covering man’s mental activities; and (III) Society, which deals with social
structures and artefacts. Within these major divisions the material is
arranged in numbered hierarchical categories, each consisting of a defining
heading followed by chronological lists of all the words, with their dates of
currency, ever used as synonyms or near synonyms for the definition.
(HTE website)
The database of the HTE will be constructed in such a way that sophisticated
searches will be possible, e.g. to find all the words meaning ‘laugh’ that came
into the language between 1300 and 1500, or to find all the words current in
Renaissance English for a particular area of meaning. The Thesaurus of Old
English, considered earlier, is an offshoot of this project, but while the OE
thesaurus presents, in lexical field arrangement, a snapshot of the vocabulary at
a particular time period, the HTE will show how the vocabulary has developed
from OE over time, lexical field by lexical field. Further information on the
HTE project, together with examples, can be viewed on the Historical Thesaurus
of English website (see References).The final example of a specialist thematic
dictionary is specialist in two ways: it is restricted to a particular area of vocabu-
lary, that of science, and it is directed specifically at learners of English: the
Longman Dictionary of Scientific Usage (Godman and Payne 1979). This dictionary
is aimed at those who are studying science through the medium of English and
for whom English is not their first language. The vocabulary of science is pre-
sented under 19 broad divisions, numbered ‘A’ to ‘U’ (‘I’ and ‘O’ are not used).
Division A contains ‘Basic Terms’, and like the other divisions it is subdivided
into ‘Sets’:
AA Space
AB Matter
AC Shape
AD Existence
AE Constitution
AF Movement
AG Change
AH Time
AJ Process
AK Knowledge
AL Word analysis
AM Statement
AN Measurement
AP Relationship
AQ Experiment.
The remaining divisions contain the ‘Scientific Terms’; each division contains
sets that are semantically related, e.g. H has the following sets:
HA Irritability
HB Nervous System
HC Sight
HD Hearing
HE Sense Organs.
Within each set, the individual terms are arranged, not alphabetically, but in
order to form a coherent account of the area of science, proceeding from those
terms that have a more general reference to those that are more specific. Each
term is provided with a word class label, a definition and explanation, occasion-
ally an example sentence, and extensive cross-references, both to earlier and
later within the set and to other sets. For example, from Set HD ‘Hearing’:
HD001 hearing (n.) One of the senses of animals, concerned with the
stimulus of sound. Hearing is well developed in tetrapod vertebrates, but
poorly developed in fishes; it is well developed in insects but not in most
other invertebrates. – hear (v.) ↓ AUDIBILITY • SCOLOPHORE
• STATOCYST • LATERAL LINE SYSTEM • AUDIBLE • EAR •
MIDDLE EAR • INNER EAR • MEMBRANOUS LABYRINTH •
COCHLEA → IRRITABILITY
HD005 statocyst (n.) (In some invertebrates) an organ of balance, con-
sisting of a vesicle containing statoliths with sensory cells on the vesicle
walls, Hair-like processes on the sensory cells are stimulated by the statoliths
when the animal moves. ↓ OTIDIUM • OTOCYST • STATOLITH ↑
HEARING
The downward-pointing arrow indicates a cross-reference to an item or items

later in the set, an upward pointing arrow to those earlier in the set, and an
arrow pointing to the right cross-refers to another set. The dictionary is sup-
plied with an alphabetical index, which gives the code for the set in which the
term is described, together with its number within the set. The authors envisage
that the dictionary will be used in four ways:
1. ‘Finding the meaning of a term when reading’ – using the index in order
to locate the term. This is a conventional dictionary use.
2. ‘Using a term when writing’ – again the index is used to locate the term,
but the focus is on what can be gleaned from the entry about using the
term. This is a conventional use for learners’ dictionaries.
3. ‘Searching for an unknown term when writing’ – either with the index
using a known term, or with the contents to identify the set that represents
the area of meaning being written about. This is the genuine thesaurus use,
as envisaged by Roget for his Thesaurus.
4. ‘Revising the terms of a particular topic’ – because each set of terms is
logically structured and internally cross-referenced, it gives a good over-
view of the topic for revision purposes.
What these four examples show is that the thematic tradition in lexicography
can be exploited imaginatively to present information about words for specific
purposes, where an alphabetical arrangement would be incapable of yielding
the desired insights for the intended uses and users.
12.4 Thematic dictionaries for learners

A thematic presentation can help learners of English as a second or foreign
language in at least two ways. First, much language teaching tends to be by
topic, and a thematic dictionary would, therefore, be an obvious reference
work to accompany such an approach. Second, one of a learner’s difficulties in
writing, besides ascertaining the appropriate grammatical and collocational pat-
terns that a word may enter, is making the appropriate choice of word in the
first place. This presupposes knowledge of the vocabulary items that could be
used to express the idea or concept, from among which the item may be cho-
sen. Moreover, learners benefit from some more explicit help in enabling them
to perceive the often subtle semantic and pragmatic distinctions between words
with similar meaning.
The first thematic dictionary for learners was Tom McArthur’s Longman Lexicon
of Contemporary English (1981), whose genesis and career he describes in Chapter
14 of McArthur (1998). McArthur consciously places his work in the thematic
tradition, which he subsequently reviewed in Worlds of Reference (1986):
The alphabetical dictionary has a logic, but it is not the logic of everyday
life. In principle, one feels, words should be defined in the company they
usually keep. Two famous moves in this direction have been the Janua
Linguarum Reserata in 1631, the work of the Bohemian educator Comenius,
and Roget’s Thesaurus, first published by Longman in 1852. The Longman
Lexicon of Contemporary English belongs in this tradition.
Comenius had a hundred chapters and a religious bias, while Roget used
a scheme of universal concepts as a framework for his prodigious lists. The
Lexicon, however, has only fourteen ‘semantic fields’ of a pragmatic, every-
day nature.
(Preface, p. vi)
As McArthur indicates, the vocabulary selection that he takes – 15,000 words

from the ‘central vocabulary of the English language’ – is arranged in fourteen
broad categories, numbered ‘A’ to ‘N’:
A Life and Living Things

B The Body: its Functions and Welfare
C People and the Family
D Buildings, Houses, the Home, Clothes, Belongings, and Personal Care
E Food, Drink, and Farming
F Feelings, Emotions, Attitudes, and Sensations
G Thought and Communication, Language and Grammar
H Substances, Materials, Objects, and Equipment
I Arts and Crafts, Science and Technology, Industry and Education
J Numbers, Measurement, Money, and Commerce
K Entertainment, Sports, and Games
L Space and Time
M Movement, Location, Travel, and Transport
N General and Abstract Terms.
Each broad ‘semantic field’ is subdivided, and within the subdivisions the lexical
items are arranged in related sets, often belonging to the same word class. Each
item is provided with definitions and examples. Although its compilation pre-
ceded it, McArthur had access to the LDOCE1 (1978) materials, so that the
entries in the Lexicon match those in the Dictionary. The careful arrangement of
items in sets and the provision of definitions and examples enable a learner to
understand the differences between related words and to choose the one appro-
priate to the particular context of use. By way of illustration, here are the sub-
divisions of Field F, followed by the entries for the set F173:
F1 Feeling and Behaviour Generally

F20 Liking and Not Liking
F50 Good and Evil
F70 Happiness and Sadness
F100 Anger, Violence, Stress, Calm, and Quietness
F120 Fear and Courage
F140 Admiration, Pride, Contempt, and Abuse
F170 Kindness and Unkindness
F190 Honesty, Loyalty, Trickery, and Deceit
F220 Relaxation, Excitement, Interest, and Surprise
F240 Actions of the Face Related to Feelings
F260 Senses and Sensations.
F173 adjectives : humanitarian and charitable [B]
humanitarian concerned with trying to improve life for human beings
by giving them better conditions to live in and changing laws, esp those
which punish too severely
generous showing readiness to give money, help, kindness, etc: She’s not
very generous with the food; she gives very small amounts. You are far too generous
with your money. –ly [adv]
liberal generous, esp in giving or being given quickly and easily or in large
amounts: He is very liberal with his money. She gave us liberal helpings of food.
–lly [adv]
magnanimous having or showing unusually generous qualities towards
others: A country should be magnanimous towards its defeated enemies. –ly [adv]
charitable showing kindness and charity [→ F175]: Be charitable; try to help
them. –bly [adv Wa3]
The codes (B, Wa3) are those from LDOCE1. The Lexicon contains line draw-
ings and it is provided with an alphabetical index, where the pronunciation is
indicated (in IPA transcription). Where an item has more than one sense, or
belongs to more than one word class, which would assign it to different seman-
tic fields or sets, this is given briefly in the Index, e.g.
long wish v F6
measurement adj J63
distance or time adj L139, N307
The index thus enables a learner to quickly review the semantic range that a
word has, as well as to locate the item in the appropriate semantic set.
Unlike Roget’s Thesaurus, McArthur’s Lexicon provides in the thematic for-
mat the range of information that would be expected in a dictionary, at least for
learners. In that sense it is a true ‘thematic dictionary’. It is to be regretted that
it has never been updated or expanded, or, indeed, that no publisher has dared
to produce such a work as a complement to a general-purpose native speaker
dictionary. The semantic classification of the words in CIDE on the CD-ROM
(see 11.5) proceeds along the same lines and encompasses all the 50,000 head-
words of CIDE, but the electronic format does not allow the overview of the
structure of the vocabulary that the printed Lexicon does. After all, the basic
arrangement of the entries in CIDE on CD-ROM is still the alphabetical for-
mat of the print version.
There is one further reference work for learners that incorporates some of the
insights from the thematic tradition: the Longman Language Activator (1993),
which advertises itself as ‘the world’s first production dictionary’. It addresses
specifically the needs of learners in writing and speaking to be able to choose the
appropriate word and to use it correctly. The Activator is constructed around
1052 ‘concepts’ or ‘key words’:
These concepts express the meanings at the heart of the English language.
It should be pointed out straight away, however, that the Activator does not
address itself to words for ‘real world’ items, some of which, of course, also
belong to the core of English. We believe that concrete nouns, and content
words in general, present fewer, less serious problems of correct use for
students, so you will not find different types of transport, dogs, machinery
or buildings here. That is left to the Longman Lexicon, which deals effec-
tively with semantic fields, including real world items.
The concepts . . . have clear, direct names such as FAR, SAD/UN-
HAPPY, HOPE, INTEND, EASY, FAULT/STH WRONG, and BUT.
(Introduction, p. F8)
The key word entry is structured in the following way. If the word has more
than one broad meaning, these are identified first and a reference given to the
keyword under which each meaning is explained. For example, the key word
modern has two meanings identified:
• modern places, methods, etc. → MODERN

• using the newest equipment, technology, etc. → ADVANCED.
Then, under each key word, lexical items are grouped together in related sets,
with a summary of the sets at the beginning of the entry, e.g. for modern:
1 words for describing machines, systems, processes etc that have been devel-
oped using the most recent ideas and equipment
2 using, or willing to use, the most recent ideas and ways of thinking
3 words for describing modern art, literature, music, etc
4 to change something in order to make it modern.
The items in each of these sets are listed, together with the description (as
above) at the beginning of the set. For Set 2 of modern, the items are: modern,
progressive, innovative, forward-looking, move with the times, go-ahead. Each of the
lexical items is provided with an entry including pronunciation, word class,
definition and examples.
The key words are arranged alphabetically in the dictionary, and the alpha-
betical list contains all the lexemes treated in the dictionary. Those that are not
key words are cross-referenced to the key word under which they are treated.
The Activator has its headwords ordered alphabetically, but under 1052 of them
– the ‘key words’ – a thematic approach is taken and lexical items are organised
into lexical sets.
12.5 Continuing the tradition

A thematic dictionary provides an insight into the structure of the vocabulary
that an alphabetical dictionary cannot possibly afford. Nevertheless, even with
works like Roget’s Thesaurus, for most people the entry point to any wordbook
is through an alphabetical list. Thematic dictionaries such as Roget and the Longman
Lexicon have needed to provide an index in order to facilitate their use, provide
an entry point, and enhance their usefulness. This, along with the fact that many
words will be entered several times, in different places, in a thematic dictionary,
has tended to make such dictionaries either limited in scope – the Longman
Lexicon has only 15,000 items – or potentially rather unwieldy. Perhaps it is not
surprising that no publisher has ventured a general-purpose thematic dictionary.
The other potential disadvantage of a thematic dictionary is that, because a
word with multiple meanings may be entered partially in several places, no
overall view of the word’s lexical description is offered.
The electronic medium, however, opens up new possibilities (McArthur
1998, Ch. 15). So far, most, if not all, publishers that have brought out a CD-
ROM version of their dictionaries have simply transferred the alphabeti-
cally arranged print version to the electronic medium. Any enhancement
has, for the most part, been in the search facilities provided for the electronic
version, though learners’ dictionaries on CD-ROM have gone further. Most
notable is CIDE on CD-ROM’s semantic field analysis (see 11.5 and above),
which assigns each headword/meaning to a set within an elaborately structured
hierarchy of fields. It has, thus, bolted a thematic framework onto an alpha-
betical one, but it is the alphabetical one that is transparent, since you can
browse/scroll through the entries in the alphabetical list, but not through those
in the thematic one: you have to pull each item in the list individually from its
place in the alphabetical structure. I would like to suggest that this is the wrong
way round.
An alphabetical lookup is always for a single item. Because you type the item
in for a lookup in an electronic dictionary, it does not matter how the dictionary
is structured. Indeed, some electronic dictionaries (e.g. COD10) show you only
one entry per screen. The entries could therefore be stored on CD-ROM in
random order; it would make little difference to a search for a single item.
However, if they were stored by semantic field, and it was possible to scroll
through the entries, a CD-ROM dictionary could fulfil both purposes: the
single lookup, which is the advantage of an alphabetical organisation; and the
lexical field analysis, which is the benefit of a thematic organisation. Some
provision would also need to be made for composite entries of items distributed
across semantic fields. But the electronic medium does not have the space con-
siderations of print.

A most lucid account of the development of the thematic tradition in reference
works, including lexicography, is found in Tom McArthur’s Worlds of Reference
(1986), now, lamentably, out of print. Chapters 12 to 14 of his Living Words
(1998) also deal with thematic lexicography, including an account of the Longman
Lexicon. Werner Hüllen’s English Dictionaries 800–1700: The Topical Tradition
(1999) treats in some detail the major works during the development of the
tradition.
Genuine editions of Roget’s Thesaurus (e.g. Kirkpatrick 1995) contain Roget’s
original introduction in which he outlines the rationale for the work and its
underlying conception.
Compiling dictionaries 161
13 Compiling dictionaries
Any dictionary, apart perhaps from the occasional scholarly undertaking, is a

commercial venture. It requires considerable investment in staff, equipment,
materials, and time. The investment is unlikely to be recouped for a number of
years. Dictionary projects run to a budget and to a timetable. They have to be
planned and managed; they require the involvement of people with a wide
range of specialist knowledges and skills. Like marriage, compiling a dictionary
is not something to be ‘contemplated lightly’. This chapter looks at some of the
issues involved in dictionary compilation and considers some of the decisions
that confront lexicographers and editors of dictionaries.
Some dictionaries have had their stories told. Reddick (1990) uses recent
scholarship to recount how Samuel Johnson went about compiling his Diction-
ary of the English Language. Elizabeth Murray’s biography of her grandfather,
James Murray, entitled Caught in the Web of Words (1977), traces the genesis of
the Oxford English Dictionary and especially James Murray’s contribution as the
principal editor of the first edition. Herbert Morton has written an account of
Webster’s Third and its editor, Philip Gove (Morton 1994). And the volume
entitled Looking Up, edited by John Sinclair (1987), gives insight into the devel-
opment of the COBUILD learners’ dictionary.
13.1 The plan

No dictionary can begin to be compiled without considerable forethought and
planning. Commercial publishers do not normally release their plans to the
public, though some of the thinking that underlies a particular dictionary or
edition often finds expression in the preface or introduction. The one famous
published plan is Samuel Johnson’s (see 4.5) Plan of a Dictionary of the English
Language (in Wilson 1957), written to satisfy his bookseller/publisher sponsors
and, at their instigation, to seek the patronage of the Earl of Chesterfield, unsuc-
cessfully as it turned out. Whether made public or not, the plan of a dictionary
has to address a number of important questions and make decisions about issues
that will affect the nature of the finished product.
One of the earliest decisions to be made relates to the target user group.
Indeed that decision may already have been taken before the planning process
starts. Deciding, for example, to produce a dictionary for children aged 7 to 10
(Years 3 to 6 of primary school) will need to be a starting point from which
other questions and decisions will flow. Similarly, if a learners’ dictionary is
proposed, that needs to be in view from the outset of the planning. General-
purpose dictionaries, however, even though having a general native-speaker
target audience, may have a conscious bias towards a particular subset of those
users. ECED, for example, has ‘the needs of families and students in mind’ (dust
jacket). Most dictionaries presuppose a general user, without specifying any
particular subgroup; CCD4 is ‘for everyone whether reading, writing or study-
ing, and . . . for all who love the English language’ (p. ix), whereas Chambers
attempts to be all-inclusive in its identification of potential users:
Chambers is a dictionary of unrivalled value to users of English – to students,

scholars, writers, journalists, librarians and publishers. It is replete with words
of technical importance to scientists, lawyers, accountants and people in
business. Chambers is the reference dictionary for the UK National Scrab-
ble® Championship; it is the favourite dictionary of crossword setters and
solvers; it is the treasure chest for all word-game players and word lovers.
(dust jacket)
There is some benefit in specifying as wide a market as possible for one’s

product!
Almost as crucial as the target audience is the decision on the size of the
dictionary, since this will have a significant effect on a number of further issues
at the planning stage. Size itself correlates both with cost and with the price at
which the dictionary can be sold. A ‘concise’ dictionary (see 3.2) would prob-
ably have between 60,000 and 90,000 headwords and cost between £16 and
£20 (at 2001 prices); ‘desk-size’ dictionaries are substantially larger and sell at
around £30. Size will also have an effect on the format of the printed product,
both in terms of its overall dimensions and the layout of its page, though other
factors will significantly affect the latter as well, such as how many columns the
page has. Most dictionaries have two columns of text to the page, but some
recent dictionaries, both desk-size (NODE) and concise (ECED), have three
columns.
Both size and target users will influence a further decision in the planning
stage, concerning the coverage of the dictionary. A learners’ dictionary will seek
to pay more attention to the core vocabulary of the language, rather than to
more specialist and technical words. A ‘pocket’ dictionary will also have less
room for specialist vocabulary. But, in an effort to be up-to-date, general-
purpose dictionaries will want to include, certainly in their desk and concise
editions, words from fields such as computing, medicine, technology, business
and finance, and the environment. Similarly, dictionaries will want to include
current colloquial and slang words that have earned their place in the record of
the vocabulary, even vulgar or taboo words. More optional may be the inclu-
sion of words from other varieties of English around the world, though it would
be unusual to exclude words at least from American English. More optional still
would be dialect words from within the British Isles, though this could be a
feature of the dictionary (e.g. Scots words in Chambers). Just as significant an
influence on scope and size would be the decision to include encyclopedic
entries, i.e. names of people and places, which can be quite lengthy, e.g. in
CED4, which claims more than 18,500, as against the single liners in ECED and
only 9,000 of them.
After all these decisions are taken, the headword list has to be chosen to
reflect the target user group, the agreed size, and the intended coverage. As we
have noted before (3.4), all dictionaries make a selection from the total vocabu-
lary. How does a dictionary achieve its desired coverage but avoid becoming
unbalanced in its selection of words, by, for example, having a disproportionate
number of words beginning with the letter ‘b’ in its headword list? There are,
for instance, many more words beginning with ‘c’ in English than beginning
with ‘g’, in fact between two and three times as many. One of the solutions to
this problem was proposed by Edward Thorndike in the US in the 1950s, in
preparation for a series of Thorndike–Barnhart dictionaries. He divided the
alphabet up into 105 ‘blocks’ of approximately equal size, to reflect the distribu-
tion of English words among the letters of the alphabet. The letter ‘c’, for
example, has ten blocks, while ‘g’ has only four. In fact, the final block for a
letter may not contain as many words as the others, nor may letters with only
one block, or the final block covering ‘x’, ‘y’ and ‘z’. A number of dictionaries
have followed Thorndike’s system, which is reproduced in Table 13.1, taken
from Landau (1989: 242, 2001: 361), but with the number of words added for
each block found in COD10 (CD-ROM) headword list.
The total number of headwords given for the COD10 CD-ROM is 64,679,
which would give a mean of 616 for each of the 105 blocks. The COD10
headword list includes abbreviations, derivational affixes and combining forms,
which may not have been taken into account by Thorndike. Given the inci-
dence of some low numbers for understandable reasons, only 345 for Q-74, the
single block for the letter ‘q’, for example, and in the final block for some letters
– e.g. 324 in F-37, the numbers are fairly consistent. Unusually high numbers
are probably explicable: K-51 (854) is a single letter block, U-98 (958) contains
words beginning with the negative prefix un-, whose numbers are likely to be
variable in a dictionary. Overall, Thorndike’s system, as tested on COD10,
seems to have some validity, although, more than half a century later, it may
need adjusting to the current vocabulary and to the current practices in respect
of what are included as headwords.
Indeed, the planning stage needs to decide what the macro-structure and the
micro-structure (see 3.3) of the dictionary will look like. In macro-structure
terms, the tendency is towards a single alphabetical list of headwords that in-
cludes compounds, defined derivatives, affixes and combining forms, as well as
abbreviations, and, if the dictionary is to include them, names of people and
places. In micro-structure terms, the information to be included in an entry, as
well as its format and its order, needs to be decided. Will pronunciation, for
164 13.1
Table Compiling dictionaries
Thorndike’s block system of distribution of dictionary entries by initial letters
Block Letters No. in Block Letters No. in

COD10 COD10
A-1 a-adk 616 L-54 lim-louh 724

A-2 adl-alh 672 L-55 loui-lz 385
A-3 ali-angk 654 M-56 m-marb 720
A-4 angl-arak 659 M-57 marc-med 561
A-5 aral-ath 632 M-58 mee-mil 820
A-6 ati-az 420 M-59 mim-monn 608
B-7 b-basd 660 M-60 mono-mz 976
B-8 base-benf 602 N-61 n-nif 735
B-9 beng-bld 772 N-62 nig-nz 747
B-10 ble-bouq 858 O-63 o-oo 706
B-11 bour-buc 698 O-64 op-ou 682
B-12 bud-bz 444 O-65 ov-oz 367
C-13 c-caq 625 P-66 p-par 729
C-14 car-cel 686 P-67 pas-peq 681
C-15 cem-chim 737 P-68 per-picj 752
C-16 chin-cled 733 P-69 pick-plea 636
C-17 clee-col 667 P-70 pleb-poss 699
C-18 com-conf 549 P-71 post-prh 632
C-19 cong-coo 431 P-72 pri-prot 564
C-20 cop-cq 677 P-73 prou-pz 574
C-21 cra-culs 656 Q-74 q-qz 345
C-22 cult-cz 340 R-75 r-recn 683
D-23 d-defd 653 R-76 reco-renn 615
D-24 defe-deteq 558 R-77 reno-rhn 556
D-25 deter-discol 678 R-78 rho-rotd 678
D-26 discom-dold 569 R-79 rote-rz 369
D-27 dole-dt 715 S-80 s-sat 640
D-28 du-dz 314 S-81 sau-sd 626
E-29 e-elk 551 S-82 sea-seo 668
E-30 ell-en 573 S-83 sep-shio 610
E-31 eo-exb 689 S-84 ship-sinf 638
E-32 exc-ez 393 S-85 sing-smd 611
F-33 f-fem 541 S-86 sme-sors 624
F-34 fen-flah 654 S-87 sort-spln 627
F-35 flai-ford 649 S-88 splo-stas 542
F-36 fore-fror 591 S-89 stat-stov 531
F-37 fros-fz 324 S-90 stow-sucg 553
G-38 g-geq 634 S-91 such-swar 553
G-39 ger-gord 707 S-92 swas-sz 319
G-40 gore-grouo 625 T-93 t-tel 729
G-41 group-gz 352 T-94 tem-thq 664
H-42 h-hav 617 T-95 thr-too 663
H-43 haw-hh 627 T-96 top-trh 668
H-44 hi-horr 740 T-97 tri-tz 852
H-45 hors-hz 663 U-98 u-unl 958
I-46 i-inam 542 U-99 unm-uz 686
I-47 inan-infn 486 V-100 v-vim 673
I-48 info-intn 594 V-101 vin-vz 349
I-49 into-iz 461 W-102 w-wess 688
J-50 j-jz 615 W-103 west-wis 676
K-51 k-kz 854 W-104 wit-wz 456
L-52 l-ld 640 XYZ-105 x-zz 478
L-53 le-lil 656
example, be given for all headwords? what transcription system will be used?
how will word stress be indicated? In terms of usage, what set of labels will be
used? and will any additional usage guidance be offered? For etymology, how
much information will be given? will it include some indication of when the
word came into the language? As such matters are decided, they need to be
recorded in a manual so that everyone who works on the dictionary, or who
joins the dictionary staff at a later date, knows what policies and ground rules
have been determined.
Not strictly lexicographical, but nevertheless of crucial importance, are deci-
sions about budgeting, both of money and of time. Ladislav Zgusta (1971: 348)
comments:
I certainly do not know all lexicographic projects past and present; but of
those I know not a single one was finished in the time and for the money
originally planned.
(cited in Landau 2001: 347)
This was indeed true of some of the more famous dictionaries – Johnson’s,
scheduled for three years and taking nine; the OED, with Murray contracted by
the Oxford Delegacy to produce in ten years and four volumes, taking fifty
years and ten (subsequently twelve) volumes. But with the advent of computer
technology, Zgusta’s pessimism is probably no longer justified: OED2 was pub-
lished on time in 1989, and it is noticeable that the timespan between editions
of dictionaries appears to be decreasing, e.g. COD7 – 1982, COD8 – 1990,
COD9 – 1995, COD10 – 1999, with an updated edition of COD10 in 2001.
The costs associated with compiling a dictionary are quite different from
those for other kinds of book publication, as Sidney Landau observes:
Most books cost comparatively little to prepare (the plant cost, in publish-
ing argot) but a great deal to produce (paper, printing, and binding costs).
The opposite is true of dictionaries, where the cost of production, though
hardly negligible, is small compared to the huge editorial development
costs, which must be amortized over a much longer period of time than
book publishers are generally familiar with. Data-management costs (sys-
tems analysis, computer programming and processing) are also much higher
than they are for other books.
(Landau 2001: 348)
The size of the budget and its distribution between the various processes of
dictionary compilation may have a significant effect on what the dictionary will
turn out to be like in the end. If time, effort and money is focused on one,
perhaps novel, aspect of the dictionary, this may have a detrimental effect on the
quality of other aspects: usage information may be prioritised over etymology,
for example.
13.2 The data
Having decided the headword list and made all the other necessary preliminary
planning decisions, the issue arises of where the data for the dictionary is going
to come from. There are essentially three possible sources of data: previous
dictionaries, citations, (computer) corpora. Any newly published dictionary will
be either a new edition of an existing line of dictionaries or a new departure. In
both cases, the dictionary stands in a long tradition of dictionary making (Chap-
ters 4 to 6), and, whether consciously or not, previous dictionaries exert their
influence. Samuel Johnson used one of Nathaniel Bailey’s dictionaries, Noah
Webster used Johnson’s, F.G. and H.W. Fowler used the OED for the first
edition of the COD. It would be foolish, even when the boundaries of lexico-
graphy are being extended, to ignore the achievements of the past, though it
may be sometimes more sensible to borrow the principles rather than the con-
tent from previous dictionaries. Landau (2001: 346) goes so far as to assert: ‘All
commercial dictionaries are based to some extent on preexisting works.’
Publishers with long-established dictionary departments build up an exten-
sive archive of citations, often going back many decades. They continue to have
a reading programme, scouring recent publications for new words or new uses
of existing words. They may have readers in-house, but will often invite inter-
ested members of the public to contribute material as well. Longman ran a
‘Wordwatch’ programme for a number of years during the 1980s, which con-
tributed to the Longman Register of New Words (Ayto 1989, 1990). Oxford has a
‘World Reading Programme’, an international network of some sixty readers,
who contribute some 18,000 items a month for inclusion in the ‘Oxford Bank
of New Words’ (cf. Ayto 1999). The collection of citations has a long history,
going back at least to Samuel Johnson, who extracted material from the litera-
ture of the day for inclusion in his dictionary. The OED was founded on the
practice of recruiting voluntary readers to undertake the reading of specific
works or authors and to submit appropriate citations. Among the instructions
issued by James Murray, when he renewed the call for voluntary readers in
1879, were the following:
Make a quotation for every word that strikes you as rare, obsolete, old-
fashioned, new, peculiar, or used in a peculiar way.
Take a special note of passages which show or imply that a word is either
new and tentative, or needing explanation as obsolete or archaic, and which
thus help to fix the date of its introduction or disuse.
Make as many quotations as convenient to you for ordinary words, when
these are used significantly, and help by the context to explain their own
meaning, or show their use.
(Murray 1977: 347)
Readers were to write their quotations on slips of paper the size of ‘a half-sheet
of note-paper’, together with full bibliographical information; Murray gives the
following specimen (from Murray 1977: 350):
Diplomatist, n.
1860. J. L. MOTLEY, United Netherlands (ed. 1868), I. ii. 24.
If diplomatic adroitness consists mainly in the power to deceive, never were more
adroit diplomatists than those of the sixteenth century.
It is estimated that some five million quotations in this form were submitted by
some 2,000 readers to the OED editors, of which 1.8 million appeared in the
first edition of the dictionary (Berg 1993).
These days, dictionary departments that have a reading programme will also
derive data from computer corpora. The reading programme may be only in
specialised areas, as indicated by the Editor-in-Chief of ECED:
Where did the Encarta Concise English Dictionary’s editors find the informa-
tion on which to base their definitions? The Bloomsbury Corpus of World
English, which now has over 150 million words, provided the main evid-
ence. We amplified this with a tailored reading programme in science,
technology, business, and other key areas in order to find evidence of word
use in varied fields. Lastly we used the Internet as a research source.
(Introduction, p. xiii)
We noted (in Chapter 11) the COBUILD dictionary as the pioneer in the use
of computer corpora for providing the data from which the dictionary is com-
piled. The Bank of English, now amounting to in excess of 400 million words,
continues to supply data for Collins dictionaries, both for native speakers and
for learners. Oxford and Longman consult the British National Corpus, and Cam-
bridge the Cambridge Language Survey corpus.
A computer corpus consists of a collection of texts in electronic form, for
lexicographical purposes drawn from both written and spoken English and rep-
resentative of the vast range of text-types and registers found in language (see
Antoinette Renouf’s chapter on ‘Corpus Development’ in Sinclair 1987). Com-
puter corpora can be easily searched, so that all the occurrences of a word can be
identified, rather than just those that happen to be noticed by a reader. In that
sense, the data from computer corpora are more complete and more reliable
than from any other source; they also provide information about relative fre-
quency of occurrence, not only of words and homographs, but also of senses of
words. The usual output of a computer search is a concordance list, such as the
following for the word conductor (taken from the one million word Lancaster–
Oslo/Bergen Corpus of written English):
1 . . . owski*> |^T*2HAT *0former fire-eating [[conductor]] *4Leopold

Stokowski *0is a mellowed man . . .
2 . . . party, to save any tickets he receives from the [[conductor]], the
number of which ends in *”7.**” ^Wh . . .
3 . . . closing curtains were combined by producer and [[conductor]]
into an exquisite theatrical unity . . .
4 . . . \0Mr. Harry Tomkins) and \0Mr. George Hespe their [[conduc-
tor]]. Î am sure everyone will agree that the . . .
5 . . . Robert Hughes, euphonium solo, and the [[conductor]] played a
tubular bell solo accompanied b . . .
6 . . . and lead the others. Âs for basses and altos the [[conductor]] had
to teach by singing the parts with t . . .
7 . . . was unhappy about a forthcoming concert. ^*”The [[conductor]]*-
so-and-so*- he has no temperament. Ît . . .
8 . . . composers.**” Ît is true that he was the first [[conductor]] to put
Elgar on the musical map, the rea . . .
9 . . . Hamilton Harty in 1933 as the permanent [[conductor]] of the
Halle*?2 Concerts, the orchestra . . .
10 . . . stra declined in its ensemble. Ânother permanent [[conductor]]
was needed, but the Halle*?2 Society wer . . .
11 . . . n ordinary theatre managements to choose [[conductor]], pro-
ducer, designer, and so on, and then, having . . .
12 . . . all responsible should be experts*- the [[conductor]], the orches-
tral players, the singers, the . . .
13 . . . stage (which includes arranging that the [[conductor]] can catch
the eye of the singer at neces . . .
14 . . . be guided. ^How often does an excellent [[conductor]] wish to
take a passage of music at an *’ . . .
15 . . . in the circumstances? ^The co-operative [[conductor]], like
Beecham, will always listen and be . . .
16 . . . tage is wrong.) ^But I have known a good [[conductor]] insist on
what was arguably a *’correctly**’ fast . . .
17 . . . is too clean to be the score used by the [[conductor]], and it was
probably the fair-copy prepa . . .
18 . . . transmitted to the voltmeter V by a nickel [[conductor]] D, nickel
being resistant to corrosive a . . .
19 . . . ed to the voltmeter by an earthed nickel [[conductor]] attached to
the bottom of a well E in th . . .
20 . . . coupling H which also positions the \0 + ve nickel [[conductor]]
with respect to the sodium by circlips o . . .
21 . . . ectrolyte J attached to the \0 + ve nickel [[conductor]] by nickel
circlips. ^Fixing and positioning of th . . .
22 . . . ce, and with far more to offer. |^The [[conductor]] rang his bell.
|^*”Good-bye, Dai,**” . . .
23 . . . ^Where indeed? ^Megan Thomas spoke sharply to the [[conduc-
tor]], demanding an explanation. ^But non . . .
24 . . . demanding an explanation. ^But nonplussed, the [[conductor]]
was. Â good man, mind; knew his job. ^B . . .
25 . . . first thing about it.**” |^Stung, the [[conductor]] was. ^*”What
you expect me to do?**” he . . .
26 . . . proper . . . |^Getting dark, now. ^The [[conductor]] switched on
the lights. ^The beleaguered . . .
27 . . . ng the sleeping Cadwallader. ^Driver and [[conductor]] peered
ahead into nothingness. Ôn the b . . .
28 . . . ^*”Can’t go lighting bonfires on this bus,**” the [[conductor]]
said firmly. ^*”Contrary to the Company’ . . .
This is provided by way of exemplification of a concordance; it is from a small

corpus and does not give enough data to make it possible to draw general
conclusions. But it does illustrate the three main senses of conductor: the musical
sense (Nos 1 to 17), its use in relation to electricity (Nos 18 to 21), and the bus
conductor sense (Nos 21 to 28). And it shows the kind of material that a lexi-
cographer has to work with and interpret, in order to decide whether a form
represents potentially more than one homograph, and to determine how many
senses should be recognised. The advantage of a concordance is that context is
provided, and can be adjusted for size, so that the grammatical and collocational
behaviour of words can be ascertained and used as guides for the determination
of sense divisions, as Ramesh Krishnamurthy observes in respect of the
COBUILD project (Sinclair 1987: 75):
Collocational evidence was of great usefulness in an analysis of the corpus

data. The concordance lines were arranged in alphabetical order of the first
character after the space following the keyword. This meant that some
features of the behaviour of a lexical item in text became immediately
apparent.
It is no exaggeration to say that computer corpora have revolutionised the

lexicographic process (Rundell and Stock 1992), in terms both of the quality of
lexical data that can be obtained and of the reliability of the conclusions that can
be drawn from that data.
13.3 The method

A computer corpus, with sophisticated search programs, may produce better
quality data, and an electronic database may ease the task of putting the diction-
ary information together, but no program has yet been written, nor is it likely to
be, to automate the process of getting from data to finished dictionary. All
dictionaries nowadays require the input of a range of staff with different skills
and specialisms. Long gone are the days of the lonely lexicographer, like Samuel
Johnson, toiling in his garret with the help of a few assistants, or even the
Fowler brothers putting together the first edition of the COD at their home on
the island of Guernsey, with occasional forays to the OED office in Oxford.
CED4, for example, lists some twenty editorial staff, including lexicographers
and computing experts, as well as seventeen ‘Special Consultants’ on varieties
of English around the world and fifty-eight ‘Specialist Contributors’ on topics
ranging from aeronautics to industrial relations to religion. A further thirty
‘Other Contributors’ are also listed, alongside their specialist topic, from horol-
ogy to place names to sports. Some dictionaries have established additionally an
advisory board; ECED has a forty-strong ‘Academic Advisory Board on English
Usage’, drawn from universities in the UK, Australia, Canada and the US.
The input of all these people is required for the formulation of the dictionary
entries. Not only will external consultants contribute their expertise on varieties
of English and technical jargons, but it is likely that a member of editorial staff
will specialise in one type of dictionary information, e.g. pronunciation, ety-
mology, grammar, usage. Some will develop expertise in defining, or in select-
ing the appropriate example, or in researching and writing biographical entries.
The hardest, and most important, part is often considered to be that of defining.
Landau (2001: 354) considers the qualities required of a good definer to be:
First and foremost, he or she must be able to write well and easily.
They must have analytical minds that seek to chop things up into parts . . .
Definers must have a broad, but not necessarily deep, fund of information.
. . . definers must have a feeling for the language, Sprachgefühl, a sense of
aptness of expression, an appreciation of nuance, style and idiom.
However, dictionary editors may seek to impose a particular uniform style of

defining on their lexicographers, as Philip Gove did for W3 with his ‘single-
statement defining style’ (Morton 1994). Compare the rambling definition of
arson from the first edition with the crisper single-phrase from W3:
The malicious burning of a dwelling house or outhouse of another man,

which by common law is felony. The definition of this crime is varied by
statutes in different jurisdictions, and generally it has been widened to in-
clude the similar burning of other property, as of churches, factories, ships,
or of one’s own house. (W1)
the wilful and malicious burning of or attempt to burn any building, struc-
ture, or property of another (as a house, a church, or a boat) or of one’s own
usu. with criminal or fraudulent intent (W3)
When a large number of people are contributing to a single work, one of the
major concerns is to maintain a consistency of writing and presentation. Dic-
tionary projects will, therefore, document their decisions about content and
style in a set of guidelines or a manual, so that any contributor or new member
of staff can be informed of the house rules for the particular publication. The
manual will specify not just conventions for the macro- and micro-structure of
the dictionary, but also points of detail, such as use of abbreviation and punctua-
tion within an entry.
Besides using a computer corpus for obtaining data, dictionaries are now
usually constructed using a computer database. A number of commercial prod-
ucts of this nature are available, but the larger publishers have their in-house
systems. Such a database allows several lexicographers or specialists to be work-
ing on entries at the same time, e.g. for pronunciation, etymology and so on;
and it allows each to see the work of the others. It also allows freelance lexico-
graphers to work at a location remote from the dictionary offices. It facilitates
the editorial and checking processes, and provides with relative ease the version
for eventual printing, as well as any electronic (CD-ROM or online) version.
Not only that, but once a dictionary database has been established, it can pro-
vide the material for dictionaries of other sizes (e.g. concise or pocket) and for
other user groups (e.g. children or school students). It also provides the basis for
easier updating and revising for the production of new editions.
The computer has changed the way dictionaries are compiled and produced,
but it has not done away with the need for skilled lexicographers to practise the
art, or craft, of constructing and writing dictionary entries. Like so many other
tasks, lexicography is computer-aided rather than computer-automated.
13.4 The result

After all the effort, skill and expertise devoted to the lexicographical task, the
result is a book, of a particular size and format, printed using a number of
selected typefaces, offered to the dictionary-buying public. How a dictionary
handles, what the dictionary page looks like, how the information is arranged,
are important factors in selling a dictionary, irrespective of the quality of its
content. The impression of accessibility is of significance. The column on a
page must invite scanning up and down. The headwords, alternative spellings,
compounds, derivatives, phrasal verbs and idioms usually stand out in a bold
typeface, as do the sense numbers. The headword itself is offset to the left. Italics
are usually used for word class and usage labels, for illustrative examples, and for
words cited from other languages in the etymologies. Cross-references are nor-
mally in small capitals. Where a word belongs to more than one word class, and
they are treated in the same entry, the beginning of the second and subsequent
word class is clearly marked, e.g. with a bold diamond in CED4, with a large
bold dot in COD10. Some information may begin on a new line within an
entry, e.g. ‘derivatives’ and ‘origin’ (i.e. etymology) in NODE and COD10.
The careful use of layout and typeface contributes to the ease with which a user
is able to locate the exact piece of information that they are seeking about a
word.
Another ‘result’ of the lexicographic process is a CD-ROM. One CD-ROM
is much like another to look at, so there is no issue here of how it ‘handles’. But
there are serious issues about how the information is displayed on the screen,
what search options are available, and how intuitive it is for the (naive) user to
operate. CD-ROM dictionaries usually display two windows: a word list win-
dow and an entry window. The word list window contains the headwords in
alphabetical order, together with a box for typing a word that the user wishes to
look up. Typing in the word locates its place in the headword list; it may also
bring up the entry automatically in the other window, or a further operation
may be required (a mouse click or pressing the ‘Enter’ key). It is normally
possible to scroll through the headword list, and if this is automatically linked to
the entry window, then by this means to scroll through the dictionary entries.
The entry window may contain a single entry, for the selected headword; alter-
natively it may contain that entry together with contiguous ones that will fit
onto the screen. In the latter case, it is usually possible to scroll the entry win-
dow, as if scanning a dictionary page. The entries on screen replicate as far as
possible the arrangement and typefaces of the print version, though without the
double or triple columns. Scrolling through entries on the screen is, thus, not
the same experience as browsing a print dictionary, but for single item lookups
there is little difference.
One respect in which CD-ROM dictionaries score over their print counter-
parts is in their full-text search facility, though this is, perhaps, of more interest
to lexicologists and other students of language than to the ordinary user. CD-
ROM dictionaries vary, though, in the sophistication of their search facilities, as
we have noted both for native speaker dictionaries (6.7) and for learners’ dic-
tionaries (11.5), where these facilities are probably of greater importance. We
are more adept at evaluating the facilities that electronic dictionaries give us,
than at knowing how individual users exploit them for their own purposes
(Creswell 1996; Nesi 1999; Holderbaum 1999; Heuberger 2000; Jehle 1999).
Once a dictionary is published and it is on the market, it becomes the object
of scrutiny by all kinds of critic, from newspaper journalists to academics. How
the critic goes about their task, or ought to, is the topic of the final chapter of
this book.

The best place to start is with Chapter 7, ‘Dictionary making’, of Dictionaries:
The Art and Craft of Lexicography (2001) by Sidney Landau, himself with experience
of involvement in a number of dictionary projects. His Chapter 6, ‘The corpus
in lexicography’, is also of relevance to the discussion in this chapter (13.2).
Bo Svensén’s Practical Lexicography: Principles and Methods of Dictionary Making
(1993) and Ladislav Zgusta’s Manual of Lexicography (1971), though a little old
now, both review some of the theoretical and practical decisions facing lexico-
graphers in compiling a dictionary. Samuel Johnson’s Plan and Preface are still
worth reading for their forward-looking insights (both reproduced in Wilson
1957), and the original OED Preface and General Explanations in Volume 1 of
OED1 merit study.
Some of the accounts of the making of individual dictionaries were men-
tioned earlier (13.1): Reddick (1990) on Johnson, Murray (1977) on the OED,
Morton (1994) on W3, and Sinclair (1987) on COBUILD.
Criticising dictionaries 173
14 Criticising dictionaries
Academic lexicography, or ‘metalexicography’, as pursued in university depart-

ments of English or Linguistics, is concerned not primarily with the compiling
of dictionaries – though academics may be involved in this, as consultants, for
example – but with researching and teaching about the whole business of
making dictionaries: their history, their typology, their structures, their users,
and so on (Hartmann 2001). One aspect of academic lexicography looks at the
products of commercial lexicography and subjects them to a rigorous critique,
usually resulting in a review; though academics are not the only ones who
review dictionaries. The process of critiquing and reviewing dictionaries is termed
‘dictionary criticism’.
One of the crucial issues for dictionary criticism is to establish a sound and
rigorous basis on which to conduct the criticism, together with a set of applica-
ble criteria. Hartmann (2001: 49) comments:
Anyone who has ever read (or written) a review of a particular dictionary
will know that generally agreed criteria and standards for the assessment of
quality and performance are still rare, if they can be said to exist at all.
This chapter discusses the business of dictionary criticism and proposes some ways
in which it may be undertaken and some guidelines for assessing dictionaries.
14.1 The business of criticism

In the brief section on dictionary criticism in his chapter on lexicography in
Solving Language Problems, Reinhard Hartmann (1996: 241) defines it as the
‘time-honoured’ activity of ‘evaluating and assessing lexicographic products’. It
is an activity that has a long history. Every new edition of a major dictionary
spawns reviews in all kinds of publication, from daily and weekly newspapers to
academic journals. But dictionary criticism is an activity, as Hartmann notes,
‘which has been beset by personal prejudice rather than noted for the applica-
tion of objective criteria’ (1996: 241). This concern is echoed in Noel Osselton’s
article – the only one with ‘Dictionary Criticism’ in its title – in the International
Encyclopedia of Lexicography (Hausmann et al. 1989). He notes ‘a surprising lack
of interest in general principles, with incidental sniping taking the place of any
real exploration’ (Osselton 1989: 229).
Dictionary reviews vary enormously in their approach and in their scope,
even those appearing in the same publication. Despite the fact that it is an
important means by which information about dictionaries is disseminated, little
attention has been paid to the methods and criteria underlying the business of
dictionary criticism. A ‘Note on Dictionary Criticism’ (by K.D.) from the Dic-
tionary Research Centre at the University of Exeter (date unknown) notes that:
‘Of general dictionaries there are a lot of reviews which lack validity and
reliability . . . what is needed is a wider discussion of the standards of assessment
of dictionaries.’ The note goes on to argue for a more objective evaluation of
dictionaries and greater clarity on what the criteria of assessment might be. It
concludes: ‘No definite theory of dictionary criticism has been established, and
it should be made more sophisticated as one field of lexicographical research.
Much remains to be done.’
Two main kinds of contribution have been made so far to the debate on
criteria for dictionary criticism. One has put forward proposals for guidelines or
criteria of reviewing. Roger Steiner’s ‘Guidelines for Reviewers of Bilingual
Dictionaries’ (Steiner 1984) would be a case in point, or Henri Béjoint’s seven
criteria for English monolingual learners’ dictionaries in his comparison of OALD,
COD and LDOCE (Béjoint 1978), or Robert Chapman’s four proposals for a
method of dictionary reviewing (Chapman 1977), or Herbert Ernst Wiegand’s
rather tongue-in-cheek ‘Ten Commandments for Dictionary Reviewers’
(Wiegand 1994).
The other kind of contribution, which is more recent, takes a set of diction-
ary reviews and subjects them to analysis, with the aim of discovering the
enduring concerns of dictionary reviewers. Jerzy Tomaszczyk (1988) took 120
reviews of general-purpose bilingual dictionaries and distilled the concerns of
the reviewers under the headings of: equivalents, directionality, reversibility,
alphabetisation, retrievability, redundancy, coverage, currency and reliability.
Martha Ripfel (1989) examined and compared the journalistic and academic
reviews of five German monolingual dictionaries, to identify the differing evalu-
ation and focus of the two types of review and the range of comments made
by each type. Günther Jehle (1990) looked at ‘popular’ and ‘academic’ reviews
of English and French monolingual learners’ dictionaries, with a focus on the
nature of a dictionary review as a text-type. He concludes his study thus (p. 300):
‘The practice of reviewing monolingual English and French learners’ dictionar-
ies in many cases unfortunately gives the impression that the reviewer has given
no prior thought to establishing the parameters within which his judgements
and critical assessments might have validity’ (my translation). We will attempt to
suggest some methodological principles and guidelines for dictionary criticism.
14.2 Method
Reviewing a dictionary is not like reviewing most other kinds of book publica-
tion. In the case of a normal book, the reviewer would expect to read the text
in its entirety, perhaps some sections of text more than once. The reviewer of a
dictionary would not expect to read every word of the text: first of all, diction-
aries are not meant to be read like that, and usually dictionaries contain too
much text to make it a feasible undertaking – NODE, for example, claims 4
million words of text, while CED4 estimates that it contains 3.6 million words.
Also at issue is who might be an appropriate person to review a dictionary. In
general, reviewers – of books, plays, films, music – are chosen because they are
considered knowledgeable or expert in the subject matter or the techniques of
whatever it is they are reviewing. We should expect the reviewers of dictionar-
ies to be knowledgeable in lexicography. This is not always so, especially in
newspaper and magazine reviews: being a user of a dictionary appears to be
sufficient qualification sometimes, even though the same publication would not
think of asking just any reader to review a novel or a book of poetry.
A reviewer, then, must be knowledgeable about lexicography, a point made
by Roger Steiner in his contribution to Dolezal et al. (1994), and they also need
a sound methodology for critically reviewing a dictionary. As with any review-
ing, the first step must be to develop familiarity with the work that is being
assessed. With dictionaries, this means, first of all, reading the often neglected
front-matter: the preface, the guide to using the dictionary, the list of staff and
consultants, and so on. This will usually give a preliminary view of the scope of
the dictionary, its intended users, and the types of lexical (and other) informa-
tion that are claimed to be included. Familiarisation will also include browsing
the main body of the dictionary and reading a variety of types of entry, as a
means of gaining an impression of the flavour of the particular dictionary under
review. Finally, some dictionaries have back-matter (appendices), which may
contain gratuitous additional information (e.g. counties of the UK and states of
the USA) or provide useful lexical information (e.g. affixes and combining
forms).
For a detailed assessment of the content of a dictionary, Robert Chapman
(1977) suggests that random sampling of entries should be used, such as ‘the
tenth main entry on every twentieth page’, in order to yield a manageable set of
entries (e.g. 50); and he proposes that each of these entries should be scrutinised
carefully for ‘accuracy, completeness, clearness, simplicity, and modernity’ (cri-
teria from McMillan 1949). Random sampling of entries ensures that a reviewer
does not look for a predetermined list of favourite items. However, a reviewer
needs to make sure that the resulting sample is representative, so that it contains
at least one member of each word class, that polysemy is well represented, and
that there is a spread across general and specialist lexemes. A random sample
may need to be supplemented by the reviewer’s checklist. Likewise, where a
dictionary has special additional features (e.g. synonym essays (LDEL2) or usage
notes (NODE, ECED, etc) ), these too need to be taken into account.
One of Chapman’s other suggestions is that the reviewing of dictionaries
should be undertaken by a team of reviewers. His conception of this seems to be
that each reviewer would be a specialist in some area of vocabulary and would
contribute an assessment of the treatment of the ‘definitions in their own fields’
(Chapman 1977: 158). Where team reviewing has been undertaken more recently
(e.g. the well organised and comprehensive Japanese reviews of COD8 (Higashi
et al. 1992) and of LDEL2 (Masuda et al. 1994) ), each member of the team has
taken a different aspect of linguistic description (pronunciation, definition, usage,
etymology, etc.) rather than vocabulary specialism, which is probably a more
sensible division of labour. Team reviews allow a more thorough treatment of
each aspect of a dictionary’s lexical description, both by enabling more extensive
sampling to be undertaken and by tapping into a reviewer’s specialist interest.
14.3 Criteria: internal and external

Criteria for evaluating a dictionary can be derived from two possible sources:
‘internal’ criteria and ‘external’ criteria. Internal criteria derive from what a
dictionary says about itself, or what the editors claim for the dictionary. External
criteria derive from metalexicography (the academic study of dictionaries), tak-
ing into account the linguistic requirements for a lexical description, as well as
considerations of dictionary design and production.
All dictionaries, in their preface or their blurb, make claims about their fea-
tures that distinguish the current edition from previous ones, or demonstrate
the superiority of this dictionary over its rivals. A reviewer can take these claims
as a basis for reviewing, to investigate whether the claims are borne out in the
dictionary’s practice: these are the internal criteria. For example, NODE con-
tains an extensive ‘Introduction’, whose principal aim is to ‘explain some of the
thinking behind these new approaches’ (p. ix). The Introduction makes the
following statement:
each word has at least one core meaning, to which a number of subsenses
may be attached . . . Core meanings represent typical, central uses of the
word in question in modern standard English . . . The core meaning is the
one that represents the most literal sense that the word has in ordinary
modern usage.
(p. ix)
These are testable statements. And they are tested by Sidney Landau in his
review of NODE (Landau 1999); he comes to the conclusion that ‘NODE’s
defining strategy is ambitious, and it does not always succeed . . . Nevertheless,
in most cases, the defining strategy does work.’ (p. 252)
Many shorter reviews take a dictionary’s view of itself at least as a starting
point for the evaluation, if not as the basis for the whole approach. Newspaper
and magazine reviews routinely depend on what the editors or publisher say
about their product, often with little attempt to test the sometimes exaggerated
claims. In that sense, there is a danger that a critique that relies solely on internal
criteria may be biased too much in favour of the dictionary, unless a radically
critical stance is taken to the claims that are made.
In using external criteria, a reviewer begins from a different standpoint. The
criteria are determined prior to the review; they arise from the accumulated
insights of the academic community (e.g. Hudson 1988; Ilson 1991). This does
not mean that they need be overly ‘academic’, since the community must take
account of the fact that dictionaries are as much reference works aimed at par-
ticular groups of users as they are linguistic descriptions of the lexical resources
of the language. Rundell (1998: 316) suggests two criteria for the evaluation of
improvements in the development of learners’ dictionaries:
the description of a language that a dictionary provides corresponds more

closely to reliable empirical evidence regarding the way in which that lan-
guage is actually used; the presentation of this description corresponds more
closely to what we know about the reference needs and reference skills of
the target user.
There is a need, therefore, for two sets of external criteria for the evaluation of
dictionaries: one set relates to the reference function of dictionaries and the
user’s perspective, and is largely about presentation and accessibility; the other
relates to the recording function of dictionaries, and is largely about content.
Presentation and content overlap and interact with each other (e.g. the core
sense and subsense division in NODE), so this is to some extent a false dicho-
tomy. However, it will provide a useful framework for establishing criteria for
dictionary reviewing.
14.4 Presentation
How a dictionary presents its material has an important influence on the acces-
sibility of the information for its target users. In the preface to the first edition of
the COD (1911), the Fowler brothers commented that they had used ‘the
severest economy of expression – amounting to the adoption of telegraphese –
that readers can be expected to put up with’ (p. iv). How they determined
readers’ tolerance levels is not stated, but they are surely rather different from
those of today’s dictionary users. Aspects of presentation that should be consid-
ered by a dictionary reviewer include at least the following.
Page layout
The size of the page varies with the size of the dictionary (desk, concise, pocket
– see 3.2). It is usual to have two columns, though three may be found in some
dictionaries (e.g. NODE, ECED). A significant feature is the amount of white
space, determined by the size of the margins, the spacing between entries, the
inclusion of other material to break up the text (e.g. usage notes, diagrams,
illustrations). All these can have an effect on the appearance of the dictionary
page. Attractive page layout improves accessibility.
Layout of the entries

Traditionally dictionaries have tended to pack all the information for a head-
word within a single paragraph, to save space; more recent editions have begun
to unpack the paragraph and to use a new line to begin a new set of information.
For example, NODE begins a new line for: a different word class, derivatives,
phrases, etymology. In some learners’ dictionaries (LDOCE3, CIDE), the lay-
out has been used to enhance the user’s access to specific meanings of polysemous
words.
Length of entries
This is determined largely by the practice of ‘nesting’, where derivatives, com-
pounds, idioms and so on are included within a single entry under a root word.
Chambers uses nesting extensively, as does the COD up to the seventh edition
(1982). The use of separate headwords for compounds and derivatives, where
these are individually defined, creates shorter and more numerous entries, and
more white space, so enhancing page layout. Likewise the inclusion of abbre-
viations, affixes and combining forms as headwords (rather than in appendices)
creates more and shorter entries and aids accessibility.
Abbreviation
The use of abbreviations, like nesting, saves space, an important consideration
in dictionary making. The Fowlers noted this as a feature of COD1; their
assumption was that the users of the COD would understand and cope with a
high degree of abbreviation. That is no longer the case, and many recent dic-
tionaries (e.g. CED, NODE) now include the full forms of items such as word
class labels and names of languages in etymologies. Fewer abbreviations mean
greater accessibility.
Academic reviewers tend to concentrate on the content of dictionaries, but
presentation and accessibility should not be ignored, because they make a sig-
nificant contribution to enabling users to be successful in extracting information
from the dictionary.
14.5 Content
Hudson’s (1988: 310–12) ‘checklist of types of lexical fact’ would provide a
starting point for criteria of content, or alternatively Ilson’s (1991) more induc-
tively arrived at set of headings would serve. Such a set of criteria would en-
compass at least the following.
Range of vocabulary
Modern dictionaries are keen to claim that they have included the latest words
from areas where neologisms are common (e.g. business, information technol-
ogy, the environment, medicine), as well as coverage of other national varieties
of English (e.g. American, Australian). A reviewer would need to determine
whether, for its size and scope, the dictionary had adequate coverage of up-
to-date, technical, international and, if appropriate, regional lexis. Where a
dictionary includes such items, geographical and biographical entries would also
come under this heading.
Word formation
On the one hand, this relates to whether affixes and combining forms are treated
as headwords or gathered in an appendix, or not given any attention at all (cf.
Pr[ic 1999). On the other hand, it relates to the treatment of derived and com-
pound words, what the criteria are for separate headword status as against nested
run-on. Also relevant here is the treatment of noun/adjective pairs that are not
cognate, e.g. church – ecclesiastical, law – legal, mind – mental, lung – pulmonary. The
judgement to be made is whether the account of word formation enables a user
to ascertain the formal (morphological) relations between words.
Homographs
The usual basis for more than one headword for a single spelling is different
etymologies. In some dictionaries (e.g. LDEL) each word class that a lexeme
belongs to occasions a new headword. In COBUILD1 there is only one head-
word per spelling, whereas more recent monolingual learners’ dictionaries
(LDOCE3, CIDE) have multiple entries based on meaning. The criteria for
determining what is a headword have important consequences for lexical de-
scription as well as for accessibility.
Sense division
For words that have multiple meanings (senses), dictionaries do not always make
clear how the senses have been established (cf. Allen (1999) on ‘lumpers and
splitters’), or the order in which they have been arranged. A tendency is emerg-
ing (e.g. NODE, COD10) to pull back from the over-differentiation of senses
that has occurred in the past (CED, Chambers). The issue here relates both to the
adequacy of the lexical description and to how straightforward it is for the user
to find the desired sense.
Defining
This is usually seen as the crucial task of the lexicographer, and there are some
well established defining styles, notably the analytical (Kipfer 1984: 66–8).
However, these have been extended in recent years, especially in learners’ dic-
tionaries (e.g. with whole sentence definitions). Some dictionaries include a
certain amount of encyclopedic information in their definitions, especially for
words referring to flora and fauna. Not only does a reviewer need to assess the
adequacy of the definitions, but also whether they are stylistically appropriate
for the intended users.
Beyond denotation
Under this heading is included other aspects of a word’s lexical behaviour over
and above its denotation, such as its lexical relations (synonymy, antonymy,
hyponymy – as indicated systematically in COBUILD1), its typical colloca-
tions, and any shared connotations. Sometimes this information is incorporated
into definitions, but it is rarely handled systematically, though the ‘synonym
essays’ in LDEL2 and ECED constitute a rare exception. Yet this information
also contributes to an understanding of a word’s meaning.
Pronunciation
There are two issues here: the transcription system, which is almost universally
IPA now in British dictionaries, and the accent to be represented. While many
dictionaries now give alternative American pronunciations, no account is taken
of the fact, for example, that the majority of British speakers say /bgt/ rather
than /bkt/ and /græs/ rather than /grp:s/. Some native speaker dictionaries
(e.g. NODE) are now giving pronunciations only for words that are problem-
atical, but what may be designated a problematical pronunciation is a matter of
judgement (Allen 2000).
Grammar
Dictionaries have traditionally given word class (part-of-speech) labels, and for
verbs distinguished ‘transitive’ and ‘intransitive’ uses; NODE acknowledges that
these may be opaque terms for modern users and substitutes ‘with obj’ and ‘no
obj’. Rarely have native speaker dictionaries given much more information
about grammar beyond this; CED, and now NODE, are exceptions. By con-
trast learners’ dictionaries have aimed for full coverage, which raises the ques-
tion about how grammatical information is represented for effective access.
A reviewer needs to evaluate how much information about the grammatical
operation of words is necessary for a dictionary to fulfil its recording function,
as against the need not to provide too much unnecessary information for
the intended users.
Usage
Dictionaries routinely label words or senses of words with ‘restrictive’ labels, to
indicate that the word or sense may be used only in a specific context. Such
labels may relate to: time (obsolete, archaic), dialect (North American, Austral-
ian English, Scottish), formality (informal, colloquial), evaluation (derogatory,
pejorative, euphemistic), status (slang, taboo), field or topic (Astronomy,
Music, Telecommunications). The extent to which dictionaries are consistent
in using their range of usage labels and how they apply them are matters for
the critic to evaluate. Some dictionaries include ‘usage notes’, especially to give
guidance on controversial areas, e.g. the preposition to be used after different, the
distinction between disinterested and uninterested. A critic may note how ‘con-
servative’ or ‘progressive’ a stance is taken by a dictionary on such issues.
Examples
All dictionaries give example sentences or phrases to illustrate word meaning,
grammar or usage. They are particularly numerous and prominent in learners’
dictionaries, where they are seen as playing a crucial role. A number of ques-
tions need to be asked by the critic, relating to: the extent of the use of exam-
ples, what role they are seen to play in exemplification, where the examples
come from (corpus or invented), and how consistently the dictionary’s policy
on examples is implemented.
Etymology
Since the etymological dictionaries of the eighteenth century, it has been cus-
tomary to include information about etymology in native speaker dictionaries,
though not in learners’ dictionaries (Chapter 10). It might be argued that such
information has no place in a dictionary of the contemporary language and
should be confined to ‘historical’ dictionaries (such as OED and SOED), though
Hudson (1988) includes etymology in his checklist of lexical facts. The amount
of etymological detail that general dictionaries include is variable; in some cases
it is just the language of immediate origin, or the etymology may be traced back
as far as possible, and perhaps with cognates in related languages. It is a matter of
critical evaluation whether the information in the dictionary under review is
appropriate to its size, purpose and intended users.
Special features
A dictionary will often seek to distinguish itself from its rivals by including a
special feature, e.g. the synonym essays and other boxed comments in LDEL2,
usage and other notes in NODE, word formation boxes in COD10, misspell-
ings in ECED, frequency information in COBUILD and LDOCE3. Some-
times these are a genuine enhancement of the information that the dictionary
gives; sometimes they are more of a marketing gimmick. Other special features
may be incorporated in the front-matter or the appendices, e.g. the essay on
English as a world language in CED, or the punctuation guide in COD8/9.
The question is whether they add to the lexical description and the coverage
and usefulness of the dictionary.
14.6 Perspective
Besides following an appropriate method and making judgements against an
explicit set of criteria, it is also possible for a reviewer to conduct dictionary
criticism from a number of different perspectives. The discussion so far has
more or less presupposed that dictionary criticism is undertaken from the per-
spective of the academic linguist or metalexicographer, applying lexicological
and lexicographical theory and insight to the task. Certainly, in the reviews that
have appeared in the International Journal of Lexicography, this has been, under-
standably, the predominant perspective adopted. But it has not always been
appropriate. For example, there was a review of OALD4 by Dwight Bolinger
(1990), which took the dictionary to task for not representing accurately some
very subtle syntactic peculiarities of verbs, which, had they been so treated,
would have probably baffled most users of the dictionary. Bolinger had, un-
characteristically, not appreciated the need for what Rundell has called ‘a more
utilitarian lexicography’ (Rundell 1998: 337), where there is a tradeoff between
the needs of the user and the meticulous accuracy of the lexical description.
An alternative perspective to that of the metalexicographer would be that, for
example, of the target user. The dictionary would be judged, by the method
and on the criteria stated, from the point of view of the needs, expectations,
prior knowledge and reference skills of the intended group of users. For an
example, see Jackson (1995) which makes a comparison of LDOCE2 and the
German learners’ dictionary, Langenscheidts Großwörterbuch Deutsch als Fremdsprache.
A third possible perspective would be that of the language teacher – particu-
larly in respect of learners’ dictionaries – who would judge the dictionary from
the point of view of its suitability for the language teacher’s task and for the
students that they are teaching, whether in a first or a second language context.
Higashi et al. (1992) review COD8 from this perspective in the Japanese con-
text, perhaps inappropriately, since COD8 was not intended as a pedagogical
dictionary, although the COD seems to have been used widely for this purpose
in Japan.
The reviewer’s perspective can act as a focus for the attention to be paid in
the critique to different aspects of the dictionary’s presentation and content,
especially since it is difficult to treat every aspect in a single review.
14.7 Purpose
In conclusion, we may reflect on the purposes for which dictionary criticism is
carried out. Some reviews, especially in newspapers and magazines, have as
their main purpose to inform the public of the existence of a new edition of,
usually, a well-known dictionary; the content of the review then often reflects
the publisher’s press release or the dictionary’s blurb. Other dictionary reviews,
like any book review, are directed at an interested public (teachers, students,
crossword addicts) and have as their purpose to inform this audience of the
content of the dictionary and its fitness for their needs.
Reviews of dictionaries that appear in academic journals, such as the Inter-
national Journal of Lexicography, while informing the journal readers about the
existence and contents of the dictionary, are also intended in many instances to
make a contribution to academic lexicography. Such reviews are often more
thorough, pursue a more rigorous methodology, and draw on the accumulated
wisdom and expertise of the academic community of dictionary scholars. There
is one further purpose that academic reviews may have. Since their critique is
drawn from an expert knowledge of dictionaries, dictionary making and dic-
tionary use, they often propose ways in which dictionaries may be improved.
So, they are offering advice to working lexicographers and dictionary publish-
ers, and contribute towards the development of both practical and academic
lexicography.
If dictionary reviews are to fulfil this function, or indeed if they are to
make a serious contribution to the academic study of lexicography, then it is
important that dictionary criticism is conducted on a sound basis, with a clear
methodology and a set of explicit criteria.
One of the purposes of this book, and of this chapter in particular, is to give
you, the reader, the background to enable you to look at dictionaries, both
historical and contemporary, with a more informed insight. On the basis of
your study, you might attempt the review of a dictionary.

There is no full-length treatment of dictionary criticism. The place to start is
with Reinhard Hartmann’s Teaching and Researching Lexicography (2001), where
he deals with the topic in Chapter 4, Sections 4.3 and 4.4, which also contain
references to other relevant articles and books.
From there it would be useful to read some of the reviews that have
appeared, for example, in the International Journal of Lexicography. The reviews,
mentioned earlier, of COD8 by Higashi et al. (1992) and of LDEL2 by Masuda
et al. (1994), are particularly recommended; but most numbers of the journal
contain dictionary reviews of varying extent and comprehensiveness. The other
journal in which dictionaries are regularly reviewed is English Today.
184 References
References
Allen, R. (1999) ‘Lumping and splitting’, English Today, 16(4), 61–3.

Allen, R. (2000) ‘Size matters’, review of Collins English Dictionary, The New Oxford
Dictionary of English and Encarta World English Dictionary, English Today, 16(2),
57–61.
Atkins, B.T.S. (ed.) (1998) Using Dictionaries: Studies of Dictionary Use by Language Learn-
ers and Translators, Lexicographica Series Maior, no. 88, Max Niemeyer Verlag,
Tübingen.
Ayto, J. (1989) The Longman Register of New Words, Longman.
Ayto, J. (1990) The Longman Register of New Words, Volume Two, Longman.
Ayto, J. (1999) Twentieth Century Words, Oxford University Press.
Bailey, R.W. (ed.) (1987) Dictionaries of English: Prospects for the Record of Our Language,
Cambridge University Press.
Béjoint, H. (1978) ‘Trois dictionnaires anglais récents: lequel choisir?’, Les Langues
modernes, 5, 465–74.
Béjoint, H. (1981) ‘The foreign student’s use of monolingual English dictionaries: a
study of language needs and reference skills’, Applied Linguistics, 2(3), 207–22.
Béjoint, H. (1994) Tradition and Innovation in Modern English Dictionaries, Clarendon
Press, Oxford.
Béjoint, H. (2000) Modern Lexicography: An Introduction, republication of Béjoint (1994),
Oxford University Press.
Benson, M., Benson, E. and Ilson, R. (1986) Lexicographic Description of English, John
Benjamins.
Berg, D.L. (1993) A Guide to the Oxford English Dictionary, Oxford University Press.
Bogaards, P. and van der Kloot, W.A. (2001) ‘The use of grammatical information in
learners’ dictionaries’, International Journal of Lexicography, 14(2), 97–121.
Bolinger, D. (1990) ‘Review of the Oxford Advanced Learner’s Dictionary, Fourth Edi-
tion’, International Journal of Lexicography, 3(2), 133–45.
Burton, T.H. and Burton, J. (eds) (1988) Lexicographical and Linguistic Studies: Essays in
Honour of G.W. Turner, D.S. Brewer, Cambridge/St Edmundsbury Press, Suffolk.
Chapman, R.L. (1977) ‘Dictionary reviews and reviewing: 1900–1975’ in J.C. Raymond
and I.W. Russell (eds) James B. McMillan: Essays in Linguistics by His Friends and
Colleagues, University of Alabama Press, 143–61.
Clear, J. (1996) ‘ “Grammar and nonsense”: or syntax and word senses’ in J. Svartvik
(ed.), 213–41.
Cowie, A.P. (1999) English Dictionaries for Foreign Learners: a History, Clarendon Press,
Oxford.
Creswell, T.J. (1996) ‘American English dictionaries on CD-ROM’, Journal of English
Linguistics 24(4), 358–68.
References 185
Cruse, D.A. (1996) Lexical Semantics, Cambridge University Press.
Crystal, D. (1995) The Cambridge Encyclopedia of the English Language, Cambridge Uni-
versity Press.
Dolezal, F.F.M. et al. (eds) (1994) Lexicographica 9/1993: Wörterbuchkritik: Dictionary
Criticism, Max Niemeyer Verlag, Tübingen.
Duden (2000) Die deutsche Rechtschreibung, Duden Band 1, twenty-second edition,
Dudenverlag.
Ellegård, A. (1978) ‘On dictionaries for language learners’, Moderna Språk, LXXII, 225–
44.
Fernando, C. and Flavell, R. (1981) On Idiom: Critical Views and Perspectives, Exeter
Linguistic Studies 5, University of Exeter Press.
Friend, J.H. (1967) The Development of American Lexicography 1798–1864, Mouton.
Godman, A. and Payne, E.M.F. (1979) Longman Dictionary of Scientific Usage, Longman.
Green, J. (1996) Chasing the Sun: Dictionary Makers and the Dictionaries They Made, Jonathan
Cape.
Hartmann, R.R.K. (1989) ‘Sociology of the dictionary user: hypotheses and empirical
studies’ in F.J. Hausmann et al. (eds), 102–11.
Hartmann, R.R.K. (1996) ‘Lexicography’ in R.R.K. Hartmann (ed.) Solving Language
Problems, University of Exeter Press, 230–44.
Hartmann, R.R.K. (2001) Teaching and Researching Lexicography, Pearson Education.
Hartmann, R.R.K. and James, G. (1998) Dictionary of Lexicography, Routledge.
Hatherall, G. (1984) ‘Studying dictionary use: some findings and proposals’ in R.R.K.
Hartmann (ed.) LEXeter ’83 Proceedings, Lexicographica Series Maior, no. 1, Max
Niemeyer Verlag, Tübingen, 183–9.
Hausmann, F.J., Reichmann, O., Wiegand, H.E. and Zgusta, L. (eds) (1989–91)
Wörterbücher, Dictionaries, Dictionnaires: ein internationales Handbuch zur Lexikographie
vols 1–3, Walter de Gruyter.
Hebert, H. (1974) ‘Lingua frankly’, review of OALD3, The Guardian, 25 March 1974.
Herbst, T. and Popp, K. (eds) (1999) The Perfect Learners’ Dictionary(?), Lexicographica
Series Maior, no. 95, Max Niemeyer Verlag, Tübingen.
Heuberger, R. (2000) Monolingual Dictionaries for Foreign Learners of English: A Construc-
tive Evaluation of the State-of-the-Art Reference Works in Book Form and on CD-ROM,
Austrian Studies in English 87, Braumüller, Vienna.
Higashi, N. et al. (1992) ‘Review of the Concise Oxford Dictionary of Current English,
Eighth Edition’, International Journal of Lexicography, 5(2), 129–60.
Historical Thesaurus of English website <http://ww2.arts.gla.ac.uk/SESLL/EngLang/
thesaur/thes.htm>.
Holderbaum, A. (1999) Kriterien der Evaluation elektronischer Wörterbücher, Annual Report
on English and American Studies 17, Wissenschaftlicher Verlag, Trier.
Hornby, A.S. (1954) A Guide to Patterns and Usage in English, Oxford University
Press.
Hornby, A.S., Gatenby, E.V. and Wakefield, H. (1942) Idiomatic and Syntactic English
Dictionary, Kaitakusha, Tokyo.
Hudson, R. (1988) ‘The linguistic foundations for lexical research and dictionary
design’, International Journal of Lexicography, 1(4), 287–312.
Hüllen, W. (1989) ‘In the beginning was the gloss’, in G. James (ed.) 100–16.
Hüllen, W. (1999) English Dictionaries 800–1700: The Topical Tradition, Clarendon Press,
Oxford.
Ilson, R. (1983) ‘Etymological information: can it help our students?’, ELT Journal,
37(1), 76–82.
Ilson, R.F. (1991) ‘Lexicography’, in K. Malmkjaer (ed.) The Linguistics Encyclopedia,
Routledge, 291–8.
Ilson, R. (2001) ‘Review of Atkins (1998)’, International Journal of Lexicography, 14(1),
80–3.
186 References
Jackson, H. (1988) Words and their Meaning, Longman.
Jackson, H. (1995) ‘Learners’ dictionaries in contrast: Langenscheidt and Longman’,
Fremdsprachen Lehren und Lernen, 24, 58–74.
Jackson, H. (1998) ‘How many words in YOUR dictionary?’, English Today, 14(3),
27–8.
Jackson, H. (2002) Grammar and Vocabulary, Routledge English Language Introduc-
tions, Routledge.
Jackson, H. and Zé Amvela, E. (2000) Words, Meaning and Vocabulary: An Introduction to
Modern English Lexicology, Cassell.
James, G. (ed.) (1989) Lexicographers and Their Works, University of Exeter Press.
Jehle, G. (1990) Das englische und französische Lernwörterbuch in der Rezension. Theorie und
Praxis der Wörterbuchkritik, Lexicographica Series Maior, no. 30, Max Niemeyer Verlag,
Tübingen.
Jehle, G. (1999) ‘Learner’s dictionaries on CD-ROM – mere gadgetry?’ in W. Falkner
and H.-J. Schmid (eds) Words, Lexemes, Concepts: Approaches to the Lexicon. Studies in
Honour of Leonhard Lipka, Gunter Narr Verlag, Tübingen, 353–63.
Jones, D. (1997) English Pronouncing Dictionary, fifteenth edition, (eds) Peter Roach and
James Hartmann, Cambridge University Press.
Katamba, F. (1994) English Words, Routledge.
Kipfer, B.A. (1984) Workbook on Lexicography, Exeter Linguistic Studies 8, University of
Exeter Press.
Kipfer, B.A. (1986) ‘Investigating an onomasiological approach to dictionary material’,
Dictionaries, Journal of the DSNA, 8, 55–64.
Kirkpatrick, B. (1995) The Original Roget’s Thesaurus, Longman.
Knowles, E. and Elliott, J. (eds) (1997) The Oxford Dictionary of New Words, Oxford
University Press.
Kurath, H. and Kuhn, S.M. (1954) Middle English Dictionary, University of Michigan
Press.
Landau, S.I. (1989) Dictionaries: The Art and Craft of Lexicography, Cambridge University
Press.
Landau, S.I. (1999) ‘Review of The New Oxford Dictionary of English’, International Journal
of Lexicography, 12(3), 250–7.
Landau, S.I. (2001) Dictionaries: The Art and Craft of Lexicography, second edition, Cam-
bridge University Press.
Lehrer, A. (1974) Semantic Fields and Lexical Structure, North Holland Publishing Co.
Lyons, J. (1977) Semantics, vols 1 and 2, Cambridge University Press.
McArthur, T. (1986) Worlds of Reference, Cambridge University Press.
McArthur, T. (ed.) (1992) The Oxford Companion to the English Language, Oxford Uni-
versity Press.
McArthur, T. (1998) Living Words: Language, Lexicography and the Knowledge Revolution,
University of Exeter Press.
McLeod, I. (ed.) (1990) The Scots Thesaurus, Aberdeen University Press.
McMillan, J.B. (1949) ‘Five College Dictionaries’, College English, 10(4), 214 –21.
Masuda, H. et al. (1994) ‘Review of the Longman Dictionary of the English Language 1991
edition’, International Journal of Lexicography, 7(1), 31–46.
Morton, H.C. (1994) The Story of Webster’s Third: Philip Gove’s Controversial Dictionary
and Its Critics, Cambridge University Press.
Mugglestone, L. (2000a) ‘ “Pioneers in the untrodden forest”: The New English Diction-
ary’ in L. Mugglestone (ed.), 1–21.
Mugglestone, L. (ed.) (2000b) Lexicography and the OED: Pioneers in the Untrodden Forest,
Oxford University Press.
Murray, K.M.E. (1977) Caught in the Web of Words: James Murray and the Oxford English
Dictionary, Yale University Press.
References 187
Nesi, H. (1999) ‘A user’s guide to electronic dictionaries for language learners’, Interna-
tional Journal of Lexicography, 12(1), 55–66.
Norri, J. (1996) ‘Regional labels in some British and American dictionaries’, International
Journal of Lexicography, 9(1), 1–29.
Norri, J. (2000) ‘Labelling of derogatory words in some British and American diction-
aries’, International Journal of Lexicography, 13(2), 71–106.
Nuccorini, S. (1992) ‘Monitoring dictionary use’ in H. Tommola et al. (eds) Euralex ’92
Proceedings, University of Tampere, Finland, 89–102.
Osselton, N. (1989) ‘The history of academic dictionary criticism with reference to
major dictionaries’, article 27 in F.J. Hausmann et al., 225–30.
Palmer, H.E. (1938) A Grammar of English Words, Longmans Green.
Pearce, D.W. (1992) Macmillan Dictionary of Modern Economics, fourth edition, ELBS/
Macmillan.
Paikeday, T.M. (1993) ‘Who needs IPA?’, English Today, 9(1), 38–42.
Pr[ic, T. (1999) ‘The treatment of affixes in the “big four” EFL dictionaries’, Interna-
tional Journal of Lexicography 12(4), 263–79.
Pruvost, J. (2000) ‘Des dictionnaires papier aux dictionnaires électroniques’, International
Journal of Lexicography 13(3), 187–93.
Reddick, A. (1990) The Making of Johnson’s Dictionary 1746–1773, Cambridge Univer-
sity Press.
Ripfel, M. (1989) Wörterbuchkritik. eine empirische Analyse von Wörterbuchrezensionen, Lexico-
graphica Series Maior, no. 29, Max Niemeyer Verlag, Tübingen.
Le Robert Collège (1997) M.-H. Drivaud (ed.) Dictionnaires Le Robert, Paris.
Roberts, J., Kay, C. and Grundy, L. (1995) A Thesaurus of Old English, King’s College
London.
Rundell, M. (1998) ‘Recent trends in English pedagogical lexicography’, International
Journal of Lexicography, 11(4), 315–42.
Rundell, M. and Stock, P. (1992) ‘The corpus revolution’, English Today, 8(2), 9–14;
8(3), 21–31; 8(4), 45–51.
Scholfield, P. (1999) ‘Dictionary use in reception’, International Journal of Lexicography,
12(1), 13–34.
Simpson, J.A. (1989) ‘Nathaniel Bailey and the search for a lexicographical style’ in
G. James (ed.), 181–91.
Sinclair, J.M. (ed.) (1987) Looking Up: An Account of the COBUILD Project in Lexical
Computing, Collins ELT.
Skeat, W.W. (1961) A Concise Etymological Dictionary of the English Language, Clarendon
Press, Oxford.
Sledd, J. and Ebbitt, W.R. (eds) (1962) Dictionaries and THAT Dictionary, Scott, Foresman
and Company.
Starnes, DeW.T. and Noyes, G.E. (1991) The English Dictionary from Cawdrey to Johnson
1604 –1755, new edition, G. Stein (ed.), John Benjamins.
Stein, G. (1985) The English Dictionary before Cawdrey, Lexicographica Series Maior, no.
9, Max Niemeyer Verlag, Tübingen.
Steiner, R.J. (1984) ‘Guidelines for reviewers of bilingual dictionaries’, Dictionaries, Journal
of the Dictionary Society of North America, 11(4), 315–42.
Svartvik, J. (ed.) (1996) Words: Proceedings of an International Symposium, Lund, 25–26
August 1995, Kungl. Vitterhets Historie och Antikvitets Akademien, Konferenser 36,
Stockholm.
Svensén, B. (1993) Practical Lexicography: Principles and Methods of Dictionary-Making, trans-
lation of Swedish original published in 1987, Oxford University Press.
Tomaszczyk, J. (1988) ‘The bilingual dictionary under review’ in M. Snell-Hornby
(ed.) ZüriLEX ’86 Proceedings, Francke Verlag, 289–97.
Weekley, E. (1967) An Etymological Dictionary of Modern English, Dover, New York.
188 References
Wells, J.C. (2000) Longman Pronunciation Dictionary, second edition, revised, Longman.
West, M.P. (1953) A General Service List of English Words, Longmans Green.
West, M.P. (1964) A Dictionary of Spelling: British and American, Longman.
West, M.P. and Endicott, J.G. (1935) The New Method English Dictionary, Longmans
Green.
Wiegand, H.E. (1994) ‘Wörterbuchkritik. Zur Einführung’ in F.F.M. Dolezal et al.
(eds), 1–7.
Wilson, M. (ed.) (1957) Johnson: Prose and Poetry, Hart-Davis.
Winchester, S. (1999) The Surgeon of Crowthorne, Penguin.
Zgusta, L. (1971) Manual of Lexicography, Academia, Prague/Mouton, The Hague.
Index 189
Index
accessibility 75–6, 78–82, 171, 177 electronic dictionaries 57–8, 59, 69–72,
acronym 13–14 77, 83, 159
affix 8, 12, 87 encyclopedia 21–2
Anglo-Saxon 10, 118, 119, 127 etymology 27, 36, 43, 55, 87, 117ff, 181
antonymy 17, 97–8, 132, 146–7 examples 26, 84, 130, 132, 136–7, 181
authority 40, 66
fixing 39–40, 44–5
Bailey, Nathaniel 37–9, 45, 166 formality 111–12
binomial 5, 86, 99 French 11, 20, 33, 39, 118, 120, 121
borrowing 11, 13, 20 frequency 140, 144
front matter 25, 76, 175
Cawdrey, Robert 33, 75
CD-ROM dictionaries 58, 69–72, glossary 31–2, 147
141–4, 159, 171–2 grammar 19, 22, 91, 95, 107, 108–9,
citations 29, 40, 44, 48, 59, 65, 88, 166; 130, 132, 135, 180
see also quotations Greek 11, 33, 118
collegiate dictionary 67
collocation 18, 27, 84, 91, 99, 132, 138, hard words 33–6, 75, 127
169 headword 25, 27–8, 86–8, 163, 178
combining form 9, 12, 121 homograph 2, 87–8, 179
completeness 36–9, 52, 59, 65, 75, 76, homonym 2, 26, 87
82, 162–3, 178–9 homophone 3
compound 5, 8, 9, 12, 25, 53, 86, 121 hyponymy 17, 98
computer corpus vii, 29, 131, 132, 140,
167 idiom 6, 99–100
concordance 29, 88, 131, 167–9 inflection 3, 8, 19, 26, 43, 87, 105–7
connotation 16, 96
conversion 13 Johnson, Samuel 39, 42–6, 62, 75, 161,
coverage see completeness 165, 166
defining vocabulary 130, 133 Latin 11, 20, 31, 32, 39, 118, 119, 120–2
definition 15, 26, 43, 55, 62–3, 65, learners’ dictionaries 24, 69, 83–5, 129ff,
93–6, 131–2, 133–5, 170, 179 153–8, 177
denotation 16, 96 lexeme 2, 3–5, 18, 25
derivative 8, 12–13, 25, 53, 56, lexical field 143, 146, 153
145 loanword 11, 14–15, 54, 106, 120–3
dialect 110–11
dictionary 21ff, 52, 67–8, 74–6 macrostructure 25, 78, 163
dictionary criticism 30, 173ff meaning 15, 19, 55, 76, 86ff
190 Index
meronymy 18, 98 reference 15–16
metalexicography 30, 173, 176, 182 Roget’s Thesaurus 149–50
microstructure 26, 79–82, 163 root 8, 12
morpheme 8, 19 run-on 25, 27
Murray, James 48–51, 102, 161, 165,
166 selection 42–3, 75
sense 15, 26, 55, 68, 88–93, 133, 179
nesting 68, 78, 178 sense relation 16–18, 96–9, 139, 180
slang 112
obsolete 113–14 spelling 18, 26, 43, 54, 61–2, 76, 82–3,
Old English 10, 117, 118, 119–20, 147, 101–2
151 synonymy 17, 69, 94–5, 97, 132, 146–7
Old Norse 10, 118, 119, 120 syntax 19, 27, 43, 84, 108–9, 131,
orthographic word 2 136–7
Oxford English Dictionary 23, 44, 47ff, 73,
82, 117, 124–6 thematic order 22, 69, 143–4, 145ff
thesaurus 150–5
part of speech see word class
Philological Society 47, 48 usage 19, 26, 109–15, 139, 180
phonological word 2 users 76–8, 132–9, 161–2, 182
phrasal lexeme 5
phrasal verb 6 vocabulary 10–11, 25, 48, 52–3, 127, 146
polysemy 15, 88
prefix 8, 13, 145 Webster, Noah 61–4, 166
pronunciation 18, 26, 43, 54, 102–5 word 1–2,
word class 6–7, 19, 26, 54, 88, 107–8
quotations 46, 49; see also citations word-form 4

Lexicography - An Introduction

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lexicography - An Introduction

Uploaded by

Copyright:

Available Formats

This book is an accessible introduction to lexicography – the study of dictionaries.

Howard Jackson is Professor of English Language and Linguistics at the

London and New York

ISBN 0–415–23172–8 (hbk)

2 Facts about words 10

5 The New English Dictionary 47

7 Users and uses 74

9 Beyond definition 101

11 Dictionaries for learners 129

12 Abandoning the alphabet 145

14 Criticising dictionaries 173

Native speaker dictionaries

Monolingual learners’ dictionaries

CCD – Collins Concise Dictionary

Where the abbreviation is followed by a number, e.g. COD8, the number

1.1 What is a word?

orthographic word a word in writing, a sequence of letters bounded by spaces

1.2 Same sound, same spelling, different word

Much more common in English are the counterparts to homographs: lexemes

bare/bear, gait/gate, haul/hall, leak/leek, miner/minor, paw/poor/pore/

1.3 Lexemes and variants

base/present tense sing talk

word-form an inflectional variant of a lexeme

To illustrate word-forms we have chosen verbs, because verb lexemes have

base/singular girl child

When one-syllable adjectives do not permit word-forms with –er/-est, it is usu-

1.4 War chests and wards of court

A second phrasal structure consists of a noun in the possessive followed by

1.5 Classifying words

1.6 Taking words to pieces

Here are some words composed of more than one morpheme:

bedroom, dreamy, going, live-in, overland, displease, shallowest, mistreat-

dream (noun) + -y dreamy (adjective)

anthropomorphic, astronaut, bibliography, biology, neuralgia, synchrony,

• anthropo- (human) + -morphic (in the form of )

a word is composed of one or more morphemes

1.7 Further reading

2 Facts about words

2.1 Where English words came from

2.2 Making new words

2.3 Word meaning

2.3.3 Sense relations

2.4 Describing words

2.5 Further reading

3.1 What is a dictionary?

toad Any of the more terrestrial warty-skinned members of the tailless

toad/tfgd/n. 1 any froglike amphibian of the family Bufonidae, esp. of the

Dictionaries are usually arranged in alphabetical order of the headwords. Indeed

3.2 Dictionaries, not ‘the dictionary’

CED4 CCD4 CPED4

Page size 187 × 260 mm 152 × 234 mm 107 × 151 mm

1954), or, most usually, of the contemporary language. While we devote a

3.3 What is in a dictionary?

Birmingham/’buhming(h)em/2nd largest British city, in the W Midlands

Angelou/’anhflu:/, Maya (b.1928), American novelist and poet, acclaimed

Some dictionaries include additional information, for example on collocation

drink/drijk/ v. & n. v. (past drank/dræjk/; past part. drunk/drkjk/) 1 a

We examine the micro-structure and the information contained in dictionary

3.4 Compiling a dictionary

unreliable in any case, as it depends on what items are included as headwords

3.5 Evaluating a dictionary

3.6 Further reading

4.1 Bilingual beginnings