Collocations and General-purpose

Dictionaries
M. Benson, University of Pennsylvania

Downloaded from https://academic.oup.com/ijl/article-abstract/3/1/23/979533 by guest on 30 October 2018

Abstract

In recent years lexicographers have been devoting increased attention to collocations.1

Heretofore, discussions of collocations have been primarily concerned with the role that
they might play in various type's of learners' dictionaries. This paper deals with the
potential role that collocations can play in general-purpose (GP) dictionaries. It will be
shown that while the best GP dictionaries contain many collocations, they all omit a
large number.
The prospect of adding collocations to GP dictionaries compels us to reconsider the
goals of monolingual dictionaries. Two new types of monolingual dictionaries will be
proposed.
The role that collocations can play in bilingual dictionaries will be explored. It will be
shown that the appropriate use of collocations in bilingual dictionaries can make them
valuable tools for both encoding and decoding.
This paper will also describe the importance of providing illustrative phrases,
including collocations, especially in definitions of polysemous items. It should be noted
that when a collocation is given at the base (usually a noun), it is considered to be a
collocation proper, i.e., a supplement to the definition. When it is given at the collocator
(usually a verb or adjective), it is considered to be an integral part of the definition, i.e.,
an illustrative phrase. For this terminology see Benson 1989: 6. The rest of the
terminology used in this paper is found in Benson, Benson, and Ilson (1986b), referred
to here as the BBI.

Introduction: the situation in existing monolingual dictionaries

If we examine the most highly regarded existing general-purpose (GP) dictio-

naries, we see that they all supply some collocations and thus seem to recognize
in principle the importance of including such constructions. However, their
treatment is, in general, inconsistent and incomplete.
For example, the most recently published unabridged American GP diction-
ary, The Random House Dictionary of the English Language, second edition,
which came out in 1987, gives many grammatical collocations; examples of the
type adjective + preposition are: abreast of, accessible to, accustomed to, careful
about, conducive to, conscious of, equal to, expert at, fond of, jealous of etc.
However, it omits a large number of equally important collocations of the same
type. Examples are: acceptable to, appreciative of, available to, certain of, clever
at, comprehensible to, curious about, difficult for, effective against, faithful to,
friendly with, furious at, happy about, hostile to, etc.
International Journal of Lexicography, Vol. 3 No. 1 © 1990 Oxford University Press
0950-3846/90 $3.00
24 M. Benson

The most recently published American abridged dictionary, Webster's New

World Dictionary, Third College Edition, 1988, gives accessible to, accustomed
to, acquainted with, adjacent to, afraid of, amenable to, ashamed of, etc., but does
not have acceptable to, accountable to, adept at, adequate for, advantageous to,
aghast at, akin to, alert to, aloof from, amazed at, angry at, anxious about,
appreciative of, apprehensive about, available to, etc.
Recently published British GP dictionaries show the same pattern. For
example, the Collins English Dictionary, second edition, gives accessible to,

Downloaded from https://academic.oup.com/ijl/article-abstract/3/1/23/979533 by guest on 30 October 2018

accustomed to, allergic to, aware of, etc., but does not have acceptable to,
advantageous to, appreciative of, available to, etc.
The Concise Oxford Dictionary (COD) is known for its inclusion of
grammatical collocations. It does provide, for example, a very large number of
adjective + preposition collocations. Examples are: abhorrent to, absent from,
accountable to, accustomed to, acquainted with, addicted to, adept at/in, adjacent
to, adverse to, afraid of, agreeable to, alive to, etc. However, even the COD
omits important collocations. Examples are: acceptable to, accessible to,
advantageous to, affectionate towards, aghast at, akin to, alert to, etc.
It might be noted here parenthetically that even the best learners' dictionaries
omit key grammatical collocations. For example, the Collins Cobuild does not give
many collocations such as absorbed with, adverse to, agog over, allied with, annoyed
at, apathetic about, apologetic for, attractive to, atypical of, available to, etc.
The treatment of other types of grammatical collocations in existing GP
dictionaries is similarly inconsistent and spotty. They neglect, for example, to
show consistently which nouns can be followed by to + infinitive. Marckwardt
(1973: 369), points out that dictionaries should be the major source for clearing
up questions of grammar and cites an inquiry from Japan asking if the noun
effort can be followed by to + infinitive.
Existing GP dictionaries often miss meanings in cases of polysemy. This
occurs especially in the treatment of collocations consisting of verb + preposi-
tion. For example, the Webster's New World, Random House, and Webster's
Ninth missed the meaning of the verb + preposition collocation go through 'be
sold out in', as in the dictionary went through five printings. Many additional
examples could be cited.
Of course, some lexicographers feel that such 'splitting' into separate senses is
not justified and that the core meaning should be emphasized (Rundell 1988:
134). An argument against such 'lumping', in favor of'splitting', is the fact that
often in translation a different verb is used for each (sub)meaning. For example,
if we translate go through five printings into Russian, we use a Russian verb not
used to translate any of the other meanings of go through: the dictionary went
through five printings ='s\o\ar' vyderzal pjat' stereotypnyx izdanij'.
Yorkey (1969: 264) pointed out over twenty years ago that the five then best
American desk dictionaries (American College Dictionary, Funk and Wagnalls
Standard College Dictionary, Webster's New World Dictionary of the English
Language, Random House Dictionary of the English Language - College Edition,
Webster's Seventh New Collegiate Dictionary) treated verb + preposition collo-
cations inconsistently. For example, only one of the five dictionaries gave get
after (as in you'll have to get after her to trim the bushes); only one gave get
 Collocations and General-purpose Dictionaries 25

around to (as in we finally got around to answering our correspondence); only one
gave get back at (as in we got back at him for his insult), etc.
It has just been shown that the best of GP dictionaries do not include a large
number of grammatical collocations. It is generally recognized that GP
dictionaries include even fewer lexical collocations (Cowie 1983: 138). Orszagh
(1969: 218) wrote with regret that the Webster's Third included a smaller
number of lexical collocations {standing combinations and word-associations in
his terminology) than the Webster's Second. GP dictionaries give, for example,

Downloaded from https://academic.oup.com/ijl/article-abstract/3/1/23/979533 by guest on 30 October 2018

only a small number of verb + noun lexical collocations such as: display ability,
heap abuse (on smb.), affect an accent, bring in an acquittal, command
admiration, run an advertisement, draw up an agenda, induce anesthesia, take up
arms, hold an auction, assume authority, etc.

A new type of monolingual decoding dictionary (MDD)

If collocations were added to existing monolingual dictionaries, material
presently included would have to be deleted. Should this be done? There is a
difference of opinion as to whether adding collocations to a monolingual
dictionary intended for adult native speakers is advisable. As Robert Ilson has
pointed out in a personal communication, many native speakers use dictio-
naries essentially to decode, i.e., to look up unknown words. Thus, for such
users, the more 'hard' words (including proper names) a dictionary contains,
the better the dictionary is.
Native speakers normally know most key collocations of their language.
Therefore, it can be argued that many native speakers do not want collocations
in their dictionary, and that most collocations should be removed from a
decoding dictionary to make room for more headwords. An even more radical
suggestion is to decrease the amount of space presently devoted to core,
grammatical words such as any, for, in, some, than, the, etc., and to basic,
polysemous verbs such as bring, get, go, pull, put, turn, etc. so as to make room
for more headwords. Such forms cause great difficulty for the learner of the
language; they do not ordinarily cause difficulty for the native speaker. What
many native speakers seek in a dictionary is the unknown or vaguely known
literary, scientific, technical word or proper name.
Thus, what seems to be needed is the development of two types of
monolingual dictionaries. The first is a monolingual decoding dictionary
(MDD), to be described here; the second will be described below. A proposal
for new types of dictionaries should not be surprising; in recent years
lexicographers (for example Hartmann 1987: 121) have been pointing to the
need for different types of dictionaries for different types of users.
The MDD would include the largest possible number of 'difficult' words and
devote minimum space to the collocations and core vocabulary of the language.
The MDD would indicate pronunciation (its readers would presumably want to
pronounce the word being looked up) and give etymologies (often useful in
clarifying the meaning of difficult words). In order to gain space for headwords,
it would give only essential examples. It would be necessary to produce MDDs
of various sizes - from the very large to the pocket format.
26 M. Benson

Obviously, there would be problems. The major, eternal problem facing the
compilers of such a dictionary, as of all dictionaries, would be the selection of
headwords. The complexity of this problem, especially if World English is to'be
covered, is hinted at by Bailey (1987: 137-41). One specific problem of selection
would be the question of regionalisms. Barnhart, writing of the Webster's
Third, points out (1962: 163), for example, that it was 'possibly an editorial
mistake to include rare Scottisms [sic]' in view of the fact that 'they take up at
least some of the space of the 70,000 compounds Tom Knott crossed out'. On

Downloaded from https://academic.oup.com/ijl/article-abstract/3/1/23/979533 by guest on 30 October 2018

the other hand, note Aitken's (1987) plea for the retention of the Scotticisms in
dictionaries of English.
A dictionary of French that embodies many of the features of the proposed
MDD is the Dictionnaire usuel QUILLET FLAMMARION.

A new type of monolingual GP dictionary (MGPD)

It appears that there is a need for a new type of monolingual dictionary,
intended for native speakers who seek help with decoding AND encoding their
own language and also for learners. This type of dictionary would be the GP
dictionary par excellence.
The MGPD should provide definitions, especially of the core vocabulary,
that would be more detailed than those in the MDD. It would devote
considerable attention to synonyms and to questions of style, often in the form
of usage notes. The MGPD would include a large number of illustrative phrases
so that it might provide truly adequate definitions for many headwords and
indicate concretely how they are used in sentences. Ilson (1986: 216) writes that
'Examples of use are a marvellous opportunity for lexicographers to display not
only the meanings of words, but also their syntax, selectional restrictions,
collocations, and stylistic level.'
As pointed out by Benson, Benson, and Ilson (1986a: 207-8), the lack of
illustrative phrases in some entries of the first edition of the American Heritage
Dictionary caused lexicological confusion. An example is the entry for the verb
set, which gives NO illustrative phrases; the result is a morass of confusion.
Other current dictionaries often do not provide appropriate illustrative phrases.
The Webster's Ninth, for example, gives for the verb pull the sense 'to exert
force upon so as to cause or tend to cause motion towards the force'. This
definition would be completed very nicely if an illustrative phrase were added of
the type provided for the same sense of pull in the Webster's III; the latter gives
such phrases as the engine pulled a long line of freight cars, pulled the sled with a
rope, pulled his hair, etc. For the verb strike the Webster's Ninth gives the sense
'to cause to ignite by friction'. The definition would be significantly more
helpful to the user if the phrase/collocation strike a match were added. For the
phrasal verb break down, it gives as its first sense 'to cause to fall or collapse by
breaking or shattering'. The definition needs a phrase such as break down a
door, as given in the Webster's III (Benson, 1981: 83). For the verb pass the
Webster's Ninth gives as one sense 'to transfer or transmit from one to another'.
Appropriate illustrative phrases would be he passed the note to her; pass the salt.
Note that this latter phrase is not a collocation: pass is a verb 'that can be
 Collocations and General-purpose Dictionaries 27

followed by a huge, practically unlimited number of direct objects.' (Benson

1989: 5).
An example of. a good entry for a polysemous verb is the Random House
Dictionary treatment of take. The entry gives as illustrative phrases; i.e., as
integral parts of the definitions, key collocations that show the various senses of
the verb: take a bribe, take advice, take a dare, take revenge, take a pill, etc.
The MGPD would provide a large number of collocations. Some colloca-
tions would be placed at the collocator (usually a verb or adjective) as an

Downloaded from https://academic.oup.com/ijl/article-abstract/3/1/23/979533 by guest on 30 October 2018

illustrative phrase, i.e., an integral part of the definition, as just shown above in
the Random House definition of the verb take. Other collocations would be
placed at the base (usually a noun) as a collocation proper, i.e., a supplement to
the definition. The placing of collocations at the bases is designed to help the
user generate texts, i.e., encode (Benson 1989: 6 and Hausmann 1985: 122-3).
In its encoding potential, the MGPD would approach, but not match, general
and specialized learners' dictionaries.
An example of the need for lexical collocations in GP dictionaries is provided
by entries for polysemous nouns. Very often such nouns enter into verb + noun
collocations that differ from sense to sense. An example is the noun line. Note
how different senses have different collocations: 'long thin mark' draw a line;
'row of characters' indent a line; 'unit of text' deliver a line; 'note' drop (smb) a
line; 'type of merchandise' carry a line; 'policy' pursue a line; (usually military)
'established position along a front' hold a line, etc. It should be clear that such
collocations, given at noun entries, are vital supplements to the definitions of
each sense of the nouns.
It is perfectly natural that some native speakers may at times want or need
help with encoding their own language. Historically, for example, the thesau-
rus, intended for the native speaker who is attempting to create a text, i.e.,
encode, has been one of the most successful word books. Native speakers who
often use a thesaurus or a usage guide when they write would find the MGPD
useful. It should be noted that native speakers may not know some lexical
collocations. Many, for example, will not know which verbs collocate with such
nouns as the following: acquittal, acumen, acupuncture, afterburners, attitude,
authority, etc. Native speakers who are concerned with good style wince when
they encounter in print such 'violations' of collocability committed by other
native speakers. Examples are: '...we are carrying expenses...' (New York Times,
January 8, 1989), instead of running up (or incurring) expenses; '...it is a threat
on older workers.' (Modern Maturity, April-May, 1989), instead of a threat to
older workers; '...he was...never indicted of any crime.' (The Pennsylvania
Gazette, December, 1987), instead of indicted for any crime; '...an improvement
o/previous works...' (Slavic and East European Journal, 24, 1980), instead of, as
required by the context, an improvement over/on previous works; '...was drawing
plans...' (M. Fitzpatrick, The National Organization of Public Health Nursing,
1975), instead of was drawing up plans, etc., etc. We see here a demonstration of
the fact that many native speakers of English need help with collocations.
The learner who does not wish to use a learners' dictionary would find the
MGPD ideal. Its decoding capability would be considerable, but, of course,
would be less than that of the MDD. The encoding capability of the MGPD
28 M. Benson

would be very strong, but, it still could not compete with a specialized
combinatory dictionary as a handbook for the production of texts. One should
keep in mind that an enormous number of MGPD dictionaries would be sold
abroad and be used by learners of English. The COD, for example, is used by
large numbers of people in many countries throughout the world.
If an existing monolingual GP dictionary were to be converted to a MGPD,
i.e., if collocations and illustrative phrases were added to it, some material, of
course, would have to be deleted. Could this be done without excessively

Downloaded from https://academic.oup.com/ijl/article-abstract/3/1/23/979533 by guest on 30 October 2018

reducing its decoding capability? The answer appears to be yes. Existing GP
dictionaries contain much material that is not used frequently. The role that
computerized data files could play in helping the lexicographer to select
headwords for a MGPD will be described below.
Let us take as our first example the Pocket Oxford Dictionary (1984). One
would assume that a 'pocket' dictionary should concentrate on giving the
essentials. However, if we take a close look, we see that many headwords could
be deleted if the dictionary were being converted to a MGPD. If we take seven
pages on which entries begin with la-, we easily find nineteen likely candidates
for deletion: lac, Lady Chapel, lagging, lakh, The Lamb (of God), lambency,
lanceolate, land-girl, land-line, Laodicean, lappet, larboard, lardy, lam, larrikin,
Lascar, Latinate, laver, lawn ('fine linen'). It must be emphasized that any word
used to define another word in the dictionary should not be deleted. Some of
the items just proposed for deletion from a pocket dictionary should, of course,
be included in larger dictionaries.
Traditionally, dictionaries have carried over unnecessary words from one
edition to the other. Cannon points out (1987: 163) that the obsolescence of
words is an 'aspect [of lexicography] to which dictionaries except the OED
devote little attention, beyond tagging an occasional item as obsolete or
archaic'. Landau (1984: 162-3) writes of the need to eliminate obsolescent
words, which 'lie buried like calcified fossils amidst the living words around
them, awaiting some word geologist to hack them out and preserve them
properly in a period cemetery'. Commenting on bilingual German and English
dictionaries, Traupman writes (1985: 169): 'some editors seem to be unable to
bring themselves to eliminate deadwood.'
Lexicographers have always faced the problem of deciding what to include in
new dictionaries and what to add and what to delete when working on revisions
of existing dictionaries. It is obvious that computerized data files are beginning
to play a key role in helping to make such decisions. Paikeday (1983) describes a
computer program designed to be used for the compilation of dictionaries and
shows that fewer important words would be omitted with the help of such a
program. Conversely, the lexicographer will be better informed about the
advisability of including rarer items. As an example, Paikeday points out that
the Webster Ninth did not include such essential items as bargaining chip, baby
boom, fast lane, spreadsheet, X-rated, etc. but did give some rare items such as
computernik, downsize, etc.
We will see below that a dictionary of English is being compiled that
resembles the MGPD. A dictionary of French that embodies features of the
MGPD is the Dictionnaire dufrancais contemporain.
 Collocations and General-purpose Dictionaries 29

The inclusion of collocations and illustrative phrases in bilingual dictionaries

is of extreme importance. A traditional defect of many bilingual dictionaries
was the indiscriminate listing (in the target language) of many possible
translations of polysemous headwords; no attempt was made to discriminate
between the various equivalents.
Iannucci discusses this point (1957: 273) in his paper on 'meaning discrimina-
tion' when he describes the entry for the French noun tour found in Heath's
New French & English Dictionary. The English equivalents are simply listed in

Downloaded from https://academic.oup.com/ijl/article-abstract/3/1/23/979533 by guest on 30 October 2018

the entry as follows: Turn, round, twining, winding; revolution, circumference,
circuit, compass; twist, strain; tour, trip; trick, dodge, wile; feat; office, service,
vein, manner, style; place, order; lathe; turning-box; wheel; mould.... A French
speaker who wishes to render in English the sentence Cette machine fait trois
mille tours a la minute has no way of ascertaining from the list of English
equivalents, which equivalent corresponds to the French tour in the given sense.
If the French speaker selected as the English equivalent the cognate noun tour,
the resultant translation would be unintelligible.
Fortunately, bilingual dictionaries have made great progress since Heath's
lexicon came out in 1932. The best bilingual dictionaries of the 1980s provide a
large number of collocations and illustrative phrases. To demonstrate this point,
we can cite excerpts from the Collins-Robert entry for the French noun tour, this
entry gives a large number of collocations and illustrative phrases with their
English translations. Here are several examples: faire un tour d'Europe 'to go on a
European tour'; un tour du monde en bateau 'a round-the-world trip by boat';
faire un tour en ville 'to go for a walk round town'; la riviere fait des tours et des
detours 'the river twists and turns (along its way)'; attendre son tour 'to wait one's
turn'; prendre son tour 'to take one's turn'; nous lefaisons chacun a notre tour 'we
take turns at it'; votre tour viendra 'your turn will come'; c'est au tour de Marc de
parler 'it's Mark's turn to speak'; I'helice a fait deux tours 'the propeller turned ...
twice'; donner un tour de clef 'to turn the key'; la situation prend un tour
dramatique 'the situation is taking a dramatic turn'; faire/jouer un tour a qn 'to
play a trick on sb.'; faire du 39 du tour de cou 'to take a size 39 collar', etc.
The examples just given from Collins-Robert show (as well as Benson,
Benson, Ilson 1986a: 208-10) that when illustrative phrases (and, of course,
translations thereof) are given for polysemous items, the bilingual dictionary
can serve for both decoding and encoding: the speakers of both languages can
see instantaneously exactly which sense in the source language corresponds to
which sense in the target language. As pointed out by Kretov et al. (1988), the
presence of illustrative phrases would seem to remove a major argument in
favor of Scerba's claim (Kromann, Riiber, Rosbach 1984) that separate
dictionaries are needed by native speakers and non-native speakers for both the
source and target languages, so that four dictionaries are needed for each pair
of languages.
Another demonstration of the role that collocations can play is suggested by
an example that Kromann, Riiber, and Rosbach (1984: 211) cite in support of
Scerba's approach (1940). They write of an English speaker who wishes to
express in German 'He changed his mind' and suggest that a typical English-
German dictionary would have entries such as the following:
30 M. Benson

change...verandem, aridern; .--.. •» . . . . . . .

(money) wechseln; (trains) umsteigen
m/W...Verstand, Geist;
(opinion) Meinung
They go on to claim that the English speaker, on the basis of these two
entries, might infer incorrectly that either die Meinung dndern or die Meinung
verandern is correct. (In fact, only die Meinung dndern is correct.) Such entries
are considered by Kromann, Riiber, and Rosbach to be unacceptable in an

Downloaded from https://academic.oup.com/ijl/article-abstract/3/1/23/979533 by guest on 30 October 2018

'active' dictionary intended for English speakers who must encode. They are
considered acceptable in a 'passive' dictionary intended for German speakers
since a native speaker of German could decode the meaning of the English
collocation (called 'idiosyncratic construction' in the article).
We see here an attempt to justify the claim that two English-German
dictionaries are necessary. However, the authors of the article overlook the
fact that ONE properly structured bilingual dictionary would solve the
problem by including illustrative phrases. The ESD (An English-SerboCroa-
tian dictionary - Benson, 1978 - which provides illustrative phrases with
translations and is intended for speakers of both languages) gives at the entry
for the verb change the phrase/collocation change one's mind with the Serbo-
Croatian equivalent and gives at the entry for the noun mind the same
collocation. Thus, both SerboCroatian speakers who are decoding AND
English speakers who are encoding will easily find the needed equivalents in
the same dictionary.
Williams (1959: 252) also pointed to the desirability of producing bilingual
dictionaries that can be used for decoding and encoding by speakers of both
languages. Williams was understandably troubled by the fact that such
dictionaries would require a great deal of space; he described the possibility of
interrelating the entries of the English/Other-language and of the Other-
language/English dictionaries or of the halves of the single-volume two-way
dictionary. This interesting suggestion means that the lexicographer would
expect the user to make use of both dictionaries dealing with a pair of languages
or, of course, both halves of a two-way dictionary. Williams cited the problem
of dealing with polysemous nouns such as spring. Perhaps, in the spirit of
Williams' proposal, a solution would be to give the lexical collocations of spring
'source of water', for example, and of its equivalent in the other language only
ONCE in either the English/Other-language dictionary or in the Other-
language/English dictionary.
As an example of this approach, Benson added to the 1979 revision of his
SerboCroatian-English Dictionary at the entry for izvor ('spring, source of
water') the following collocations, of course in both languages - hot/thermal
springs, mineral springs. However, in order to save space, the compiler did not
add them to the 1984 revision of the English-SerboCroatian Dictionary. The
assumption was that the user of the latter dictionary who looks up the word
spring and finds as its possible equivalents 'izvor', 'skok', etc. can easily look
these entries up in the companion volume. In the companion volume the user
will find at the entry for izvor the collocations listed above and other data. At
the entry for skok the user will find pertinent collocations and other data. The
 Collocations and General-purpose Dictionaries 31

criteria for assigning the collocations to one or the other dictionary should be a
topic for future investigation.
Iannucci (1957: 278-80) recommended the analogous coordination of bi-
lingual dictionaries and monolingual dictionaries so that various senses of
polysemous words can be differentiated without taking up too much space.
Iannucci proposes that a bilingual dictionary would have a specific monoling-
ual dictionary as its 'partner'. The numbered meanings of polysemous entries in
the bilingual dictionary would correspond to the same numbered meanings of

Downloaded from https://academic.oup.com/ijl/article-abstract/3/1/23/979533 by guest on 30 October 2018

the same entries in the paired monolingual dictionary. A great deal of space
would be saved when the bilingual lexicographer takes on the difficult task of
differentiating senses. However, no attempt to act on Iannucci's suggestion has
been made, and in his later article (1985) he made no mention of this proposal.
The desirability of producing bilingual dictionaries designed to meet the
needs of specific groups of users has been strongly advocated by the followers of
Scerba. Steiner (1986) has gone much further in exploring the lexicographic
needs of various users, and he isolated eighteen (!) potential bilingual combina-
tions, taking into account type of use and monodirectional/bidirectional status
(88-90). In fact, for the overwhelming majority of language pairs, it would be
economically impractical to produce more than a total of two dictionaries, one
going from language A to B, and one going from B to A. Thus, from a practical
viewpoint, it would make more sense to systematically include collocations and
illustrative phrases and to seek methods of neutralizing the disadvantages of a
'single' dictionary (such as the problem- of metalanguage), rather than to
attempt to multiply the number of bilingual dictionaries being published.

The placing of collocations in dictionaries

The placing of collocations has been discussed by Benson (1989: 6-8) and
Hausmann (1985). Note that in monolingual, learners' dictionaries of English
designed to help non-English speakers encode, the collocation change one's
mind should be placed at the entry for mind (the base). In GP dictionaries,
monolingual and bilingual, the collocation should be placed, if at all possible, at
both the entry for change (the collocator) and at the entry for mind; the same
dictionary can then be used for decoding AND encoding by speakers of both
languages.

Recent developments
Apparently the first instance of a planned, though modest attempt to add
grammatical and lexical collocations to a new edition of a GP bilingual
dictionary was Benson's 1984 revision of the English-SerboCroatian Dictionary
(ESD) (Belgrade, 1978). Here are several examples of lexical collocations of the
type noun + verb that were added to noun entries beginning with the letter C:
cats meow, chaos ensued, the clock is fast (slow), the clock struck five, the colors
match, a conflict broke out, crows caw, etc.
Here are examples of the collocational type verb + noun added to noun
entries beginning with the letter C: dojperform a Caesarean section, paddle a
32 M. Benson

canoe, turn a cartwheel, hear a case (in court), hold a celebration, make a change,
issue a citation, file/submit a claim, hold class, turn a clock ahead (back), wind a
clock, engage/throw in the clutch, apply/put on a coat (of paint), take up a
collection, carry out/execute a command, make a comment, act on a complaint,
meet all conditions, hold a conference, assume control, hold a copyright, apply
criteria, etc.
The systematic inclusion of collocations and illustrative phrases in a GP
monolingual dictionary now being compiled is reported by Thomas Paikeday.

Downloaded from https://academic.oup.com/ijl/article-abstract/3/1/23/979533 by guest on 30 October 2018

His Penguin-Pitman Canadian Dictionary, scheduled to appear in 1989, has
used the BBI and other sources for grammatical and lexical collocations. This
dictionary can be considered the first to approach the concept of a MGPD of
English. Here are specimen entries for the polysemous nouns condition and
state, showing a large number of lexical collocations:

CONDITION a. 1 conditions pL circumstances: Undtr

normal -s, it doesn't snow in July; weather ~s; unsanitary
working -s; abnormal, difficult, ideal, squalid -s; economic
s; market -s. 2 a state of being, fltneu, or health, seen at
changeable: a motor in good running —; in bad, excellent,
operating, poor, terrible —; our financial -; people of all sorts
and -s in society; a patient in critical, fair, good,
satisfactory, serious, stable -; He's either in or oat of
condition (= physically fit or unfit) for a marathon. 3 what
something depends on for its being or happening;
requirement: the terms and —s of an offer; a — of the offer; the
assantial -s for survival; to fulfill, implement, impost, meet,
satisfy, set, stipulate a -; a bail ~; He spoka on condition
that he not ba identified. 4 a bodily disorder to have a heart
-; a skin - such as acna.

STATE n. 1 a government, nation, or iu territory: Canada is

a sovereign —; Sue's from lha States (= U.S.): Louisiana is
one of the 50 US. —s (= one of many units forming a nation);
the city - of Singapore; a buffer, client, member, police,
puppet, secular, welfare -; to establish, found, govern, rule a
~; a — banquet, capital, corporation, radio, school, secret,
trooper, visit; separation of church and — in government
policies; the federal, —, and municipal levels of government;
The Queen is Canada's head of -; the VS. Stata
Department (of foreign affairs). 2 a steady condition or
mode of existence: a sad - of affairs; a - of shock; the - of
the world; Is happiness a — of mind? solid, liquid, and gaseous
~s of matter; a building in a good or bad - of repair; A - of
emergency was declared during the flood; his poor — of health;
A — of war exists between the two countries; a comatose,
nervous, unconscious, weakened -; Th* patient is in quite a
ttata (= an excited condition). 3 rank or station in life: a
humble -; in great - (= dignity); A body lies in stata (in a
public place of honour).

Summary and conclusions

It has been shown that existing monolingual GP dictionaries provide only some
grammatical collocations and very few lexical collocations. The proposal is made
that future GP dictionaries specialize: they either become monolingual decoding
dictionaries (MDD) with a large number of headwords, but with a small number
 Collocations and General-purpose Dictionaries 33

of essential collocations or illustrative phrases, or become monolingual GP

dictionaries (MGPD), containing key collocations and a large number of
illustrative phrases so as to acquire both a decoding and an encoding capability.
Bilingual dictionaries should include a large number of illustrative phrases
and collocations so that they can be used for both encoding and decoding.

Note

Downloaded from https://academic.oup.com/ijl/article-abstract/3/1/23/979533 by guest on 30 October 2018

1
I am grateful to Robert Ilson for comments that he generously gave in regard to an
earlier version of this paper during the BUDALEX Congress.

References
Aitken, A. J. 1987. 'The Extinction of Scotland in Popular Dictionaries of English' in R.
W. Bailey (ed.). Dictionaries of English. Ann Arbor: Univ. of Michigan Press.
Bailey, R. W. 1987. 'On Some Discontiguities in Our English Dictionaries' in R. W.
Bailey (ed.). Dictionaries of English. Ann Arbor: Univ. of Michigan Press.
Barnhart, C. 1962. 'Problems in Editing Commercial Monolingual Dictionaries' in F.
Householder and S.Saporta (eds.). Problems in Lexicography. Bloomington: Indiana
University Press.
Benson, M. 1971; revised 1979; updated 1989. SerboCroatian-English Dictionary.
Belgrade: Prosveta.
Benson, M. 1978; revised 1984; updated 1989. English-SerboCroatian Dictionary.
Belgrade: Prosveta. (ESD)
Benson, M. 1981. 'The Lexicographic Treatment of Compound Verbs.' ITL: Review of
Applied Linguistics 52: 75-87.
Benson, M. 1989. 'The Structure of the Collocational Dictionary.' International Journal
of Lexicography 2: 1-14.
Benson, M., Benson, E. and Ilson, R. 1986a. The Lexicographic Description of English.
Amsterdam and Philadelphia: John Benjamins.
Benson, M., Benson, E. and Ilson, R. 1986b. The BBI Combinatory Dictionary of English:
A Guide to Word Combinations. Amsterdam and Philadelphia: John Benjamins. (BBI).
Cannon, G. 1987.'Viability: the death of recent new items in English.' Word1%: 155-171.
Collins Cobuild English Language Dictionary. 1987. J. Sinclair, ed. in chief. London and
Glasgow: Collins.
Collins English Dictionary. 1986. Second Edition. P. Hanks, ed. London and Glasgow:
Collins.
Collins-Robert French-English English-French Dictionary. 1987. Second edition. Beryl
Atkins et al. London, etc.: Collins and Paris: Le Robert.
Concise Oxford Dictionary of Current English. 1982. Seventh edition. Oxford: Oxford
Univ. Press. (COD).
Cowie, A. P. 1983. 'The pedagogical/learner's dictionary' in R. R. K. Hartmann (ed.).
Lexicography: Principles and Practice. London, etc.: Academic Press.
Dictionnaire du francais contemporain. 1966. J. Dubois et al. eds. Paris: Librairie
Larousse.
Dictionnaire usuel QUILLET FLAMMARIONpar le text et par I'image. 1960. P. Gioan,
ed. New York: E. P. Dutton and Paris: Quillet-Flammarion.
ESD see Benson 1978.
Hartmann, R. R. K. 1987. 'Dictionaries of English: The User's Perspective' in R. W.
Bailey (ed.). Dictionaries of English. Ann Arbor: Univ. of Michigan Press.
34 M. Benson

Hausmann, F. J. 1985. 'Kollokationen im deutschen Worterbuch: ein Beitrag zur Theorie

des lexikographischen Beispiels' in H. Bergenholtz and J. Mugdan (eds.)- Lexikogra-
phie und Grammatik. Tubingen: Niemeyer.
Heath's New French & English Dictionary. 1932. E. A. Baker, ed. Boston: D. C. Heath &
Co.
Iannucci, J. E. 1957. 'Meaning Discrimination in Bilingual Dictionaries: A New
Lexicographical Technique.' Modern Language Journal 41: 272-281.
Iannucci, J. E. 1985. 'Sense Discrimination and Translation Complements in Bilingual
Dictionaries.' Dictionaries: Journal of the Dictionary Society of North America 7:

Downloaded from https://academic.oup.com/ijl/article-abstract/3/1/23/979533 by guest on 30 October 2018

57-65.
Ilson, R. 1986. 'General English Dictionaries for Foreign Learners: Explanatory
Techniques in Dictionaries.' Lexicographica 2: 214-22.
Kretov, A. A., Laomonen, A. and Nikkilja, E. 1988. 'O novom tipe dvujazycnogo
slovarja.' Russkij jazyk za rubezom 5: 49-56.
Kromann, H., Riiber, T. and Rosbach, P. 1984. ' "Active" and "Passive" Dictionaries:
The Scerba Concept Reconsidered' in R. R. K. Hartmann (ed.). Lexicographica Series
Maior 1: LEXeter '83 Proceedings. Tubingen: Max Niemeyer Verlag.
Landau, Sidney I. 1984. Dictionaries: The Art and Craft of Lexicography. New York:
Charles Scribner's Sons.
Marckwardt, A. 1973. 'The Dictionary as an English Teaching Resource.' TESOL
Quarterly 7: 369-79.
Orszagh, L. 1969. 'Wanted: Better English Dictionaries.' English Language Teaching 23:
216-22.
Paikeday, Thomas M. 1983. 'The Joy of Lex.' Creative Computing. November: 240-45.
Paikeday, Thomas M. 1989, forthcoming. Penguin-Pitman Canadian Dictionary. Mar-
kham (Canada): Penguin Books Canada Limited and Mississauga (Canada): Copp
Clark Pitman.
The Random House Dictionary of the English Language. 1987. Second Edition, Una-
bridged. S. B. Flexner, ed. New York: Random House.
Rundell, M. 1988. 'Changing the rules: Why the monolingual learner's dictionary should
move away from the native-speaker tradition' in M. Snell-Hornby (ed.). ZuriLEX '86
Proceedings. Tubingen: Francke.
Scerba, L. V. 1940. 'Opyt obscej teorii leksikografii'. Leksikograficeskij sbornik 3:
89-117.
Steiner, R. J. 1986. 'How Many Languages should a "Bilingual" Dictionary Offer?'
Lexicographica 2: 85-92.
Traupman, J. C. 1985. 'Modern Trends in German-English Lexicography.' Lexicografi-
ca 1: 165-71.
Webster's New World Dictionary of American English. 1988. Third College Edition. New
York: Webster's New World.
Webster's Ninth New Collegiate Dictionary. 1983. Springfield (MA): Merriam-Webster,
Inc.
Webster's Third New International Dictionary of the English Language. 1961. Reprinted
with addenda sections in 1966, 1971, 1976, and 1981. Chicago, etc.: G. & C. Merriam
Company. ( Webster's HI).
Williams, E. 1959. 'The Problems of Bilingual Lexicography, Particularly as Applied to
Spanish and English.' Hispanic Review 27: 249-56.
Yorkey, R. 1969. 'Which Desk Dictionary is Best for Foreign Students of English?'
TESOL Quarterly 3: 257-70.