A Multifaceted Approach To Collocations In English, Romanian And Italian [612389]

ANTOANELA MARTA MARDAR

A MULTIFACETED APPROACH TO COLLOCATIONS

IN ENGLISH, ROMANIAN AND ITALIAN

Referent științific:
Prof. univ. dr. Hortensia Pârlog

CONTENTS

INTRODUCTION

1. LEX ICAL PATTERNS: APPROACHES, TERMINOLOGY AND
TYPOLOGIES

1.1. Formal a pproaches t o lexical p atterns
1.1.1. Lexica l approaches
1.1.2. Lexicographic approaches
1.1.3. Syntactic approaches
1.1.4. Semantic approaches
1.2. Functional approaches to l exical p atterns
1.2.1. Pragmatic and stylistic approaches
1.2.2. Methodological approaches
1.2.3. Psycholingusitic approaches
1.3. Terminology and typologies
1.4. Lexical phrases on the scale of i diomacity
1.4.1 . Approaches to idiomacity in Ro manian and Italian
1.4.2 . Collocations and free word combinations
1.4.3 . Delexical(ized) collocational patterns
1.4.4 . Collocations and the so – called harmonic phrases
1.4.5 . Phrasal verbs
1.4.6 . Similes
1.4.7 . Binominals
1.4.8 . Clichés and proverbs
1.4.9 . Idioms
1.4.9 .1. Features of idioms
1.4.9 .2. Possible categorizations of idioms
1.4.10 . Metaphors
Concluding remarks

2. APPROACHES TO COLLOCATIONS

2.1. Linguistic approaches to collocations
2.1.1. Traditional and modern v iews on collocations
2.1.2. Definitions and charac teristics of collocations
2.1.2.1. Definitions of collocations
2.1.2.2. Characteristics of collocations

2.1.2.2.1 . Non-compositionality
2.1.2.2.2 . Non-substitutability
2.1.2.2.3 . Non-modifiabi lity
2.1.3. Classifications of collocations
2.1.3.1. Formal classification of collocations.
2.1.3.2. Semantic classification of collocations.
2.1.4. Collocations vs. ‘colligations’
2.2. The psycho -linguistic approach to c ollocations
2.3. The sociolinguistic approach to collocations
2.3.1. Communities of language users
2.3.1.1. Speech communities and speech networks
2.3.1.2. Discourse communities
2.3.2. Social identity and jargons
2.4. The cultural approach to collocations
2.4.1 . Language and culture
2.4.2. Cultural models
2.4.3. Cultural identity, cultural connotations and collocations
Conclu ding remarks

3. COLLOCATIONS FROM THE PERSPECTIVE OF LEXICAL
SEMANTICS

3.1. On words and meaning
3.2. Approaches to meani ng
3.2.1. Componential analysis
3.2.2. The prototype theory
3.2.3. The structural approach
3.2.4. The cognitive approach
3.3. Definitions and classifications of meaning
3.3.1. Preliminary remarks
3.3.2. Classifications of meaning
3.3.2.1. Denotation and connotation
3.3.2.2. Denotation and reference
3.3.2.3. Denotation and sense
3.3.3. Sense relations
3.3.3.1. Synonymy
3.3.3.2. Antonymy
3.3.3.3. Hyponymy
3.4. Collocations on the axes of meaning
3.5. Lexical cohesion, cont ext and co -text
3.5.1. Cohesion
3.5.2. Lexical cohesion

3.5.3. Context and co -text
Conclu ding remarks

4. COLLO CATIONS FROM THE PERSPECTIVE OF STYLISTICS

4.1. Stylistics: definitions and areas of interest
4.2. Stylistics and f unctional styles
4.3. Stylistic evaluation of words and lexical patterns
4.4 Words and patterns used in literary texts
4.4.1. Epithets
4.4.2. Euphemisms
4.4.3. Similes
4.4.4. Metaphors
4.4.5. Clichés, proverbs and sayings
4.5. Colloquial words and patterns
4.5.1. Slang
4.5.2. Jargon
4.5.3. Professionalisms
4.6. Words and patterns in scientific texts and official documents
4.6.1. Words, terms and the fixity of meaning
4.6.2. On the scientific prose style
4.6.3. On the style of off icial documents
Conclu ding remarks

5. TRANSLA TION – ORIENTED APPROACHES TO ENGLISH
COLLOCATIONS ACROSS ROMANIAN AND ITALIAN BORDERS

5.1 Approaches to the t ranslation o f collocations
5.1.1. Linguistic approaches to translating collocations
5.1.2. Cultural approaches to translating collocations
5.2 Types of translation
5.3 Translation procedures and strategies
5.4 The translator. Skills and competences
5.5 Collocations vs. other (fixed) lexical patterns in translation
5.5.1. Colloc ations in translation
5.5.2. Idioms in translation
5.5.3. Metaphors in translation
5.5.4. Similes, clichés and proverbs in translation
5.6 Translating collocations in different text types
5.6.1. Text typology and text functions
5.6.2. Trans lated texts typology: g eneral, litera ry and specialized

5.6.3. Translating general collocations
5.6.4 . Translating literary collocations
5.6.5 . Translating specialized collocations
5.6.5.1 . On the t ranslati on of legal collocations
5.6.5.2 . On th e translation of business collocations
5.6.5.3 . On the translation of technical collocations
Conclu ding remarks

FINAL REMARKS

APPENDICES

Appendix 1 . General collocations: Do and make collocational patterns
Appendix 2 . Legal collocations grouped according to topic
Appendix 3 . Legal collocations grouped according to key word

Appendix 4 . Specialized collocations including computer terms
Appendi x 5. Collocations including the noun ‘business ’ as a ‘node’ and as
a‘collocate’
Appendix 6 . Collocations including the nouns law (EN) , lege/ drept (RO) and
legge / diritto (IT) as ‘nodes’ and ‘collocates’
Appendix 7. (Fixed) lexical patterns in translation
7.1 Collocations
7.2 Binominals
7.3 Similes
7.4 Idioms
7.5 Proverbs and clichés
Appendix 8. English – Romanian ‘ business’ collocations
Appendix 9 . English – Italian ‘business ’ collocations

GLOSSARY

REFERENCES

ACKNOWLEDGEMENTS

This book is based on the PhD dissertation English Collocations across
Romanian and Ita lian Borders written between 2006 and 2009 and publicly
presented at West University of Timișoara on the 27th of November 2009.
Its publication in the present form is the result of further additions and
deletions made according to the valuable suggestions of Professor Hortensia
Pârlog, my scientific coordinator , and of Profes sors Elena Croitoru and
Federica Scarpa to whom I would like to express my deep gratitude.
Special thanks go to my husband , who showed great support and
understanding thoughout my doctoral programme , and to our kids who allowed
me to gradually improve my PhD dissertation in 2013 and 2016 while spending
time with them on my maternal leave.
Last, but not least, I am grateful to my colleague and dear friend Isabela ,
who encouraged me and proved a trustworthy guide thoughout my editorial
process.

To Mariana and Elena,
my guardian angels in heaven and on earth

LIST OF ABBREVIATIONS

AmE : American English
BrE: British English
FEIs : ‘fixed expressions and idioms ’ (Moon 1998)
FBS: the form -based strate gy
IP: the ‘idiom principle’ (Sinclair 2001, 2004)
IT: Italian
LC(s) : language culture(s)
MBS : the meaning -based strategy
OCP : ‘the open -choice principle’ (Sinclair 2001, 2004)
RO: Romanian
SAT : ‘Speech Accommodation Theory ’ (Giles 1973)
SL: source lang uage
SLC (s): source language culture (s)
ST: source text
TG: transformational -generative
TL: target language
TLC(s) : target language culture(s)
TT: target text

INTRODUCTION

Some say that knowing a language implies knowing its words and the rul es
which have to be observed in order to produce grammatically correct messages.
Nevertheless, if reference is made to the various types of more or less fixed
lexical patterns existing in a language, the knowledge of that language is not
only knowledge of individual words, but of their predictable combinations and
of the cultural information which these combinations encapsulate. Such a
statement is supported by the idea that people use set phrases, rather than
separate words, and that such phrases represent an important part of the
speakers’ (mental) lexicon.
Furthermore, since words and phrases fit into both longer texts and social
contexts of use, language becomes an integral part of social action, and its
functioning may be better understood if special a ttention is devoted to the
relations existing between language and culture (see 2.4.). As suggested by
Stubbs (2002: 5), an analysis of these re lations would imply taking into
consideration phrases and texts . This includes the linguistic co-text (see 3.5.3 .)
of the language in use, the socially recognized text-types (see 5.6.1.), which are
accepted to constantly appear in any language, the changing social and cultural
meanings (see 3.1. – 3.2.) conveyed by language, and of course, the functions of
language.
In fact, language , social action and knowledge are inseparable because
“people do things with words” (Austin 1962). Moreover, social relations can be
created by people using the proper words in the proper situation. Studies on how
language is used in na tural social settings show that communication is
impossible without shared knowledge and assumptions between speakers
and hearers , and that communicative competence and cultural competence
are inseparable . Consequently, an analysis on how words are used is as
important as useful because it “can reveal relations between language and the
world, but also between language and speakers with their beliefs, expectations
and evaluations” (Stubbs 2002: 6).
In addition, words and lexical patterns are part and parcel of the
language vocabulary which is generally agreed to be more than just a mere
unstructured list of words. Studies in the field of lexicology, phraseology and
lexico -semantics point out the fact that the syntagmatic and paradigmatic
relations between wor ds are doubled by “other relations which are repeated

across many pairs and sets of words, and which make broad cuts acr oss the
vocabulary” (id : 35).
Furthermore, individual words rarely correspond to units of meaning .
The better candidates for this positi on are often the longer phrases or
collocations , which, in spite of having often been regarded as a peripheral
phenomenon of only limited interest, have come to be an issue of utmost
importance in as numerous as various studies.
Most theoretical aspects r egarding the features and specific behaviour of
such lexical patterns are tackled by studies which approach collocations and
other lexical patterns from the perspective of lexicology, lexico -semantics,
phraseology, syntax, semantics, pragmatics and stylist ics. Nevertheless, the
methodological (see 1.2.2 .), psycholin guistic (see 1.2.3 . and 2.2.),
sociolinguistic (see 2.3.) and cultural (see 2.4.) approaches to such patterns
have also brought to light aspects which are essential for a comprehensive
descriptio n of the various lexical patterns existing in a language and for an
appropriate understanding of their complexity.
Language teaching studies devote special attention to collocations
because a good command of collocational patterns is now accepted as an
indicator of a good command of a language and fluency .
Unlike the traditional studies on collocations which focus mainly on
semantic aspects, the methodological approach to such lexical patterns is an
extension of the notion of fixed expressions. Traditiona l language teaching
studies support the idea that expressions are seen as either freely constructed in
the slot -and-filler kind of grammar, or they are fixed expressions. However,
practice has proved that most of the patterning is between these two extreme s,
because it is phraseological and highly variable, and only the extreme ends are
really fixed or entirely free. Such patterns are either fixed lexical expressions, or
constructed by separate choices , following strict syntactic rules.
In certain situatio ns, the linguistic form of standard collocations is
consciously manipulated (see, for example, newspaper headlines, feature
articles and advertisements) with a view to achieve a certain stylistic effect or to
impress the readership. None theless, more impor tant are those instances in
which the deviation from the native -like use of lexical patterns is due to the
speakers’ lack of awareness of the phenomenon. Many non -native speakers
believe that , in producing stretches of language , they merely have to combine
items of vocabulary according to their knowledge of syntactic structure. That is
why the non-native speakers of a language are always ‘caught’ with
collocations.
Nevertheless , the idea is generally shared that things change as the
speakers’ proficiency develops. The extensive use of language patterns allows
the non -native speakers to come into contact with and memorize an increasing
number of more or less fixed lexical patterns. These patterns are learned as fully

lexicalized expressions, and proficient speakers are most often able to select and
appropriately use them according to the context of situation.
Whether considering the creative use of lexical patterns by native
speakers of a language, or their acquisition and conventional use by the non –
native speakers of the same language, the fact remains that the wide variety of
such patterns existing in any language makes it difficult to establish clear -cut
distinctions between them. The most consistent contribution in the research on
lexical patterns has b een made by the British specialists in the field.
English lexicological, lexicographic and lexico -semantic studies, in
particular, have devoted a lot of attention to lexical patterns and, in an attempt
to differentiate them, have suggested various approac hes to (see 1.1. -1.2, 2.1 . –
2.4.), definitions (see 1.4. and 2.1.2 .) and classifications of these patterns (see
1.3 – 1.4. and 2.3.). This is hardly the case of Romanian and Italian studies,
which most often analyze their corresponding le xical patterns s tarting from
English models. Unfortunately, this implies trying to find points of similarity
and dissimilarity between the English patterns and their corresponding
Romanian and Italian ones, respectively, rather than identifying specific
patterns in these languages and providing definitions and classifications adapted
to the two linguistic and cultural realities.
As regards the generalizations made in relation to the various types of
lexical patterns found in English, specialists have observed especially formal
and semantic aspects (see 3.3. – 3.5.), but, depending on the approach
envisaged, attention was shifted to syntactic , pragmatic , stylistic (see 4.4. –
4.6.), methodological or translation -related issues (see 5.1 – 5.6).
Out of the numerous comparati ve-contrastive approaches to the English
lexical patterns (see 1.4. and 4.4. – 4.5.), the one (s) bringing collocations and
idioms face to face proved the most fruitful and relevant. Nevertheless, all the
other (fixed) lexical patterns, referred to from bot h a lexico -semantic (phrasal
verbs, similes, binominals, metaphors, clichés and proverbs ) (see 3) and a
stylistic perspective (epithets , euphemisms, clichés, metaphors, proverbs
and sayings, slang, jargon, professionalisms ) (see 4), contributed significant ly
to the better contouring of the specific features of collocations, and of the
formal or/and semantic features which collocations share with each of these
patterns.
The lexico -semantic approach brought to the fore representative aspects
regarding meanin g, both when words are used in isolation (see 3.3.) and when
they collocate with other words (see 3.4. – 3.5.), and sense relations
(synonymy, antonymy, hyponymy ) existing between words (see 3.3.3. ), as
well as the way in which these relations function a t the level of collocations . In
addition, special attention was paid to the contribution of collocations to lexical
cohesion (see 3.5.2 .), and to the influence of the co-text and context (see
3.5.3 .) on the semantic decoding of collocations.

As far as t he stylistic approach to collocations is concerned, the
aspects related to the specific use of such patterns by various speakers according
to their social status , the degree of formality of the situation, the variety of
English or to social parameters such as sex, age and occupation were
intermingled with elements which allow for a distinction of collocations in
terms of functional styles (see 4.2.) and text typology (see 5.6.1.).
All these aspects were further explored in the translation -orieted
approach to collocations (see 5.) with a view to discover the extent to which
translation theories (see 5.1.1 and 5.1.2 ) and translation strategies (see 5.3)
provide valid solutions for the appropriate translations of lexical patterns , in
general, and of collocati ons, in particular ( see 5.6.) in terms of the ir
collocability and cultural specificity (see 5.5.).

RESEARCH METHODOLOGY

Starting from the assumption that a comparative -contrastive approach to
collocations in English, Romanian and Italian could result in a series of
extremely interesting results regarding the similarities and dissimilarities
existing between these languages, my initial intention was to symmetrically
approach Collocations in English, Romanian and Italian.
Interesting and challenging as it seemed, this idea had to be given up
quite soon because of the lack of relevant information in the Romanian and
Italian studies devoted to collocations . Hence, the original orientatio n of my
research change d, the English approach to collocations becoming the starting
point and, at the same time, turning into a point of reference for my
comparative -contrastive analysis of English collocations across Romanian and
Italian borders .
As regards the th eoretical aspects included in this study , I chose to start
my research from d rawing comparisons between the most representative formal
and functional English approaches on lexical patterns. In doing this, I also
explored collocations from psycholinguistic , socio -linguistic (the use of
collocations in speech and di scourse communities) and cultural perspectives.
In addition, taking into account the controversial and rather confusing
nature of the terminology used to denote the various lexical patterns existing in
English, I considered it necessary to point out signi ficant differences regarding
the terms used in English studies, on t he one hand, and, when possible, in those
in Romanian and Italian studies, on the ot her. This resulted in the identification
of similar, but more importantly, of dissimilar approaches to l exical patterns in
the three languages und er discussion. Moreover, it brought to light problematic
terms which are sometimes used to denote significantly different types of
lexical patterns in the three languages envisaged .

The lexico -semantic and stylisti c approaches to collocations and to other
lexical patterns were structured in such a manner as to support the practical
aspects related to the ir translation from English into Romanian and Italian.
In translating collocations in different text types, my ai m was to find out
similarities and especially dissimilarities between the English collocations and
their corresponding ‘equivalent’ patterns in Romanian and Italian, as well as to
identify common and significantly different cultural patterns in the three
languages under discussion.
The appendices were designed so as to illustrate possible ways of
organizing collocations : according to domain and grammatical category , to
the key word used both as a ‘base’ and ‘collocate’, or to cultural terms
corresponding to specific cultural realities.
Last, but not least, the glossary of terms concluding this study was
organized in a very synthetic and clear manner so as to facilitate the readers’
access to the definitions of the most relevant notions related to collocatio ns and
idiomacity.
My approach embraces the more traditional views on collocations , being
devoted to lexical collocations , in particular . The term lexical patterns was
used in this dissertation to denote all the patterns of idiomacity, irrespective of
their formal and semantic restrictions.

CORPUS

The corpus selected and analyzed in this researc h is represented by as numerous
as various dictionaries. General monolingual dictionaries were doubled by
bilingual dictionaries (either general, or specialized ), as well as by specific
dictionaries (phraseological dictionaries, dictionaries of idioms, sy nonyms,
antonyms and proverbs). Practical resources (e.g. grammar books) on
collocations , idioms and phrasal verbs were also used for selecting
representative e xamples.
Literary texts, on the other hand, were not closely looked at because the
(highly) creative use of collocations in such texts allows for debatable
interpretations and translation s into Romanian and Italian , respectively . The
examples of speciali zed collocations found in a series of bilingual domain –
specific dictionaries were illustrative for their conventional use, which was
under focus in this study .
The modern corpus -based aproach to collocations was not an option for
the present study, due to the absence of relevant corpus es in Romanian.

1. LEXICAL PATTERNS: APPROACHES AND T YPOLOGIES

Lexical patterns are part and parcel of phraseology, a fuzzy segment of
language studies which, in spite of embracing the conventional rather than the
productive or ruled -governed side of language that involves various kinds of
units and ‘pre -patterned’ expressions such as idioms, fixed phrases, and
collocations, cannot provide clear delimitations and classifications of the
different types of patterns inv olved.
This aspect is proved by the variety of approaches suggested along the
last decades, on the one hand, and by the consistent and often confusing
terminology used by specialists, on the other.
Being fully aware of the great variety of lexical patte rns identifiable in a
language, and of the obvious difficulties in establishing a clear -cut distinction
between these word combinations, specialists interested in exploring the
behaviour and characteristics of lexical patterns have struggled to find new
ways of approaching, defining, and classifying them.
Depending on the perspective adopted, various approaches have been
suggested, some of them focussing on formal aspects, others considering
functional aspects. Many studies in the field make use of terms s uch as
lexicalist , lexicographic , syntactic or transformational, semantic, pragmatic
or stylistic to denote various possible approaches to lexical patterns. These
terms represent in fact another side of the coin, because, although not always
clearly specif ied, the lexical, lexicographic, syntactic, semantic, pragmatic or
stylistic aspects are inherent to a certain extent , both in the formal and in the
functional approaches. Under the circumstances, I consider that the key
element that c ould be used in disti nguishing formal from functional
approaches is the more theoretical and the practical orientation, respectively, in
relation to the end -user. In other words, formal approaches give priority to the
theoretical description of lexical patterns in isolated ins tances of language use,
whereas functional approaches place more importance on the speakers’ actual
use of lexical patterns.
However, irrespective of the perspective adopted or of the linguistic
aspects envisaged, the approaches to lexical patterns have co mmon concerns
such as the specific behaviour of lexical phrases, their frequency and
importance in language, the varying degrees to which lexical phrases are open
to formal or semantic variation, their functions, as well as their importance to a
model of l anguage which places lexis and grammar on an equal footing.

1.1. FORMAL APPROACHES TO LEXICAL PATTERNS

Formal approaches1 to lexical patterns are mainly concerned with the form or
structure of such word combination s, and with the ways in which form
cond itions their morpho -syntactic behaviour, their semantic or pragmatic
interpretation, or their comprehension and use by the speakers of the language
envisaged, but what is more important, by its non -native speakers. Since lexical
patterns are very frequent in any language and their appropriate use proves the
linguistic, cultural, and communicative competences of a speaker, many
language teaching studies focus on the importance of teaching lexical patterns
to both native and non -native speakers.

1.1.1 . LEX ICAL APPROACHES

Out of the long list of formal approaches to lexical patterns, I consider that the
lexicalist ones deserve a pole position due to the fact that lexical patterns are
made up of lexical units or lexemes, the noteworthy members of lexicology .
This explains why the first approaches shyly trying to establish a typology of
lexical patterns and to suggest an acceptable and accessible terminology for
such word combinations were closely related to this branch of linguistics.
Nevertheless, given the distinction between free and restricted lexical
patterns , reference is often made to phraseology , rather than lexicology , since
the former is specifically designated to deal, among other aspects, with patterns
characterized by a certain degree of formal an d semantic restrictions.
Unexpected as it might seem to those familiar with the studies of Firth
(1957, 1968a), Sinclair (1991, 1996, 1998, 2001, 2004a), Cowie (1981, 1988,
1994, 2002), Carter (1987, 1998), the approaches to lexical patterns have their
origins in the Russian phraseological theory which developed from the late
1940s to the 1960s being first mediated to non -Russian speaking scholars by
Klappenbach (1968), Weinreich (1969), Arnold (1973) and Lipka (1974).
Although simplistic, this theory prov ided a systematic, comprehensive,
and soundly -based framework of descriptive categories whcih is still applicable
and accessible.
The Russian phraseologists in favour of the lexicalist approach were
particularly interested in describing phraseological unit s and structures as well
as in identifying the lexical , rather than the semantic primes of a language.
The primary division between word -like units and sentence -like units is
shared by many of the specialists embracing this approach. The former type of

1 The term used by the specialists in the field to denote approaches focussing on the
form of lexical patterns is structuralist , rather than formal .

units function syntactically at , or below the level of the simple sentence (e.g. in
the nick of time, to break one’s journey ), whereas the latter function
pragmatically as sayings , catchphrases and conventional formulae (e.g.
There’s no fool like an old fool, The buck stops here, You don’t say so ).
Chernuisheva (1964) is one of the first Russian phraseologists to refer to
this distinction However, the sentence -like units, which she labelled
phraseological expressions, included sayings and familiar quotations,
without any reference to catchphrases and conventional formulae.
The subclassification of ‘ nominations’ or ‘ composites ’ represents
another important achievement of the early Russian research. Within the class
of nominations , Vinogradov (1947) drew a distin ction between
‘phraseological fusions ’ (also called ‘ idioms’ ), ‘phraseological unities ’ and
‘phraseological combinations ’. The first subcategory included combinations
that were ‘unmotivated’ (or semantically opaque) and, in general, structurally
fixed (e.g . spill the beans ). The second included partially motivated patterns,
whose meaning could be seen as a metaphorical extension of some original
neutral sense (e.g. blow off steam – metaphorical extension from the technical
meaning), whereas the third subtyp e, i.e. ‘ phraseological combinations ’, in
spite of being the most interesting, was regarded by Vinogradov as the most
difficult to delimit. The reason for this is the fact that in the case of two open
words such as meet and demand , the phraseological combi nation has one
component used in a literal sense, while the other is used figuratively.
Vinogradov considers that the figurative element in the third subcategory could
be phraseologically bound by a single word, or by a limited set of words. For
example, t he specific meaning of meet is phraseologically bound by demand , in
the same way as it may be bound by words such as needs, requirements,
requests . As regards unities and fusions , the boundary between them is not
clear -cut, variations being possible accord ing to the linguistic and cultural
experience of the individual.
Unlike Vinogradov (1947) , Amosova (1963) considers that the ‘bound’
sense must have a single determining word (e.g. grind in grind one’s teeth ).
Moreover, the combinations called phrasemes by Amosova represented the
outer limits of phraseology , this being the aspect over which Vinogradov and
Amosova disagreed. Later on, such combinations were of particular interest to
several specialists, Cowie and Howarth (1996) included, who emphasized the
fact that both phraseological combinations and phrasemes should belong to
the category of ‘restricted collocations’.
An interesting and useful lexicalist approach is suggested by Mitchel
(1971). In his discussion of meaning and combination, he distinguishe s between
collocations , colligations , idioms , and compounds , admitting the existence of
functional or pragmatic idioms -formulae and proverbs such as Who goes

there? or Waste not want not, as well as of semantically opaque idioms such as
kick the bucket and eat one's heart out (1971: 58 ff.).
Mitchel’s contribution to the study of lexical patterns is important
because it attempts to separate out different types of lexical clustering within a
Firthian framework, focusing on language chunking. In addition, it is definitely
a valuable resource for the corpus -based studies of lexical patterns. In spite of
all the positive aspects, this model has some shortcomings which are due to the
fact that Mitchel’s approach cannot be applied rigorously to the full range of
lexical patterns and cannot account for all their features.
Mel’čuk’s classification of nominations has a very important position in
the formal approach to lexical patterns. He recognizes categories which
correspond closely to those of the traditional mode l, and considers collocations ,
to be the focal point of the system. In his opinion, collocations constitute the
majority of phrasemes , i.e. word combinations of all types, in the lexicon and
in many genres and specialist fields, as well.
Melčuk (1988, 199 5) is the one to suggest a complex set of what he calls
‘non-free phrases’ or ‘phrasemes’, each of them being ‘fixed’ in a particular
way. In his view , a phrase is free “if and only if all its semantic and syntactic
properties are completely determined by the respective properties of its
constituent lexemes (and by the general rules of syntax)” (1988: 169), all the
other phrases being regarded as non -free.
By using formulae based on how transparent the meaning of the phrase is,
Melčuk distinguishes between various types of phrasemes. True idioms (e.g.
shoot the breeze ) have a meaning not derivable from the constituents of the
phrase. As regards collocations or ‘semiphrasemes’ (e.g. crack a joke ), the
meaning of one constituent ( joke) is derivable from the general lexicon, whereas
the meaning of the other ( crack ) is determined by the particular collocation in
which it occurs . In addition, in ‘quasi -phrasemes’ such as start a family , the
meaning of the lexical pattern includes both the meaning of the constituents, i.e.
the verb start and the noun family , and the meaning of the two words combined,
i.e. ‘to have one’s first child’.
Mel’čuk (1998) notes the existence of what he calls ‘ pragmatemes’ , that
is, phrases which are transparent in meaning, but which are fixed in the sense
that, by convention , one wording is consistently chosen over other possible
alternatives in any given situati on. The example of pragmateme provided by
him is the conventional structure Best before [date] , used on food containers,
which is consistently chosen in preference to structures such as To be
consumed before … or Don't use after … , which are roughly equivalent in
meaning.
Mel’ čuk’s distinction between ‘free’ and ‘non -free’ phrases makes
explicit the fact that lexical phrases represent only a part of the language as a
whole, the rest of the language being described in other ways.

This view is debatable when considering Melčuk’s category of
‘collocation’ . In his opinion, collocation is to be accounted for in terms of non –
free phrases only when the meaning of one of the items depends on its co –
occurrence with the other item. For example, crack a joke is a non -free phrase ,
but tell a joke and make a joke are presumably free phrases. Nevertheless, the
fact cannot be ignored that the verbs crack , tell and make all collocate with the
noun joke and none of the three verbs represents a free choice on the part of the
speaker. In other wor ds, these verbs are constrained by collocability, in each
ease the precise meaning of the verb being determined by th e corresponding
collocation.
Looking at ‘ phrasemes’ from an encoding point of view, rather than
from the point of view of comprehension, M elčuk builds his classifica tions into
the framework of an explanatory combinatorial d ictionary, stressing that
phrasemes constitute a huge proportion of a text and of the lexicon.
The original distinction between collocation and idiomaticity suggested
by Mel’čuk is refer red to by Weinreich (1969: 44 f. ) who mentions that
‘(stability of) collocation’ means “a high degree of contextual restric tion”,
whereas ’idiomaticity’ means “a strong restriction on the selection of a
subsense”, these two bondings being q ualitatively different.
Mel’čuk’s contribution to the study of lexical patterns is supplemented by
his codifying the denotational and connotational semantics of lexemes, as well
as by his establishing their polysemous structure.
Last, but not least, menti on should be made that Mel’čuk is particularly
associated with the development of the Meaning -Text Theory, his contribution
representing an account of how collocations can be rigorously and
systematically described in terms of ‘lexical functions2’. The use of such
functions in the study of patterns has a great contribution to the design of
collocational dictionaries, where a persistent weakness is the failure of
adequately specifying the semantic categories to which the collocates belong.
As suggested by Co wie (1998: 8),
“the formal apparatus developed by Mel’čuk is a systematic means of
indicating not only collocations, but also the deep syntactic pattern to which
they conform. The lexical functions which apply in a given case may be
numerous and their semantico -syntactic relations highly complex ” (emphasis
added).
Reference can also be made to studies which suggest an extension of the
Russian phraseological tradition to embrace the cultural dimension. The
representatives of this approach ( Veronika Telyia among others) be lieve that the

2 A ‘lexical function’ is a general and abstract meaning, coupled with a deep syntactic
role, which can be expressed in a variety of ways.

cultural dimension must be elaborated in all its richness and complexity so that
the phraseology of a language can be fully described and understood.
In addition, sharing the earliest views on phraseological units , Gläser
(1998: 126) refers to the existence of a ‘phrasicon’ in any language, i.e. the
whole inventory of idioms and phrases, both word -like and sentence -like set
expressions. In her opinion, word -like phraseological units are ‘nominations’
used to designate a phenomenon, an object , an action, a process or a state, a
property or a relationship in the outside world, which may be identified in the
traditional parts of speech related to these conceptual entities: nouns, verbs,
adjectives and adverbs. Moreover, they represent the centre of the
phraseological system including both idioms and non-idioms (i.e. non –
idiomatic restricted collocations). Idioms are the prototype of the
phraseological unit, whereas non–idioms have transparent meanings and
include technical terms (terminological w ord groups), onymic entities (i.e.
phrases which are proper names), clichés, paraphrasal verbs and other set
expressions (e.g. unconditional surrender, of paramount importance , etc.).
Sentence -like units are considered by Gläser to be ‘propositions’ used to
designate states of affairs in the outside world whose logical structure consists
of a nomination and a predication. In cases of reduction or ellipsis, the finite
verb as part of the predicate may be absent.
Mention is also made of the fact that phrase ological units are placed in
the transition area between nominations and propositions, which explains their
dual character. Such units comprise irreversible binominals , stereotyped
comparisons , proverbial sayings , fragments of proverbs , allusions and
fragm ents of quotations . As regards the category of sentence -like
phraseological units , it includes:
 Proverbs – units with an educative function which are idiomatic
because their figurative meaning may refer to a different state of affairs
(e.g. Make hay while the sun shines );
 commonplaces – trite formulae and truisms (e.g. Boys will be boys; We
live and learn; It’s a small world ) which do not have an educative
function, but rather serve as conversational fillers. As a rule, they are not
idiomatic;
 routine formu lae – units which may also include idioms (e.g. Many
happy returns; Mind the step; Hold your horses );
 slogans – self-explanatory units excluded from the category of idioms
(e.g. Value for money, Safety first );
 commandments and maxims: (e.g. Be relevant, Be brief);
 quotations and winged words (e.g. Where ignorance is bliss, ‘tis folly
to be wise ).
Last, but not least , in her analysis of phraseological units , Gläser (1998:
127) refers to their characteristic features: + lexicalization, + common usage, +

reproducibility, + syntactic and semantic stability, +/ – idiomacity, +/ –
connotations, +/ – expressive, emphatic, or intensifying functions in a text. Such
features are relevant when analyzing the semantic, pragmatic and/or stylistic
restrictions of the various word combinations existing in a language.
Out o f the numerous views presented above, the approach suggested by
the classical Russian phraseologists is generally agreed to be one of the most
influential in phraseological studies. This is due not only to i ts usefulness in
establishing the main future directions in the study of lexical patterns, but also
to its applicability in the design and compilation of dictionaries.

1.1.2 . LEXICOGRAPHIC APPROACHES

The lexicographic approach to lexical patterns is an important piece in the
domino of possible approaches to word combinations. The various types of
lexical patterns, more or less fixed, definitely pose a serious problem to
lexicographers because practice has proved that such patterns can hardly be
fitted i nto the alphabetical headword list of traditional dictionaries. This is true
especially with ‘idioms’ which cannot be analysed or transformed according to
the common syntactic rules , and whose meaning cannot be derived from their
component parts.
Moreover , the fact has been noticed that general dictionaries have tended
to be theoretical and while listing lexical patterns, they have rarely categorized
them, sometimes making no distinction between compounds , phrasal verbs ,
and other lexical patterns. Proverb s and sayings have occasionally been
labelled as such, but most often, lexical patterns have been treated together as
‘idioms’ or ‘phrases’.
As regards the identification of lexical patterns in dictionaries, this
activity is typically bound up with composi tionality, because the semantic
anomalousness of lexi cal patterns is the dominant feature which leads to their
being recognized as lexical items. Since lexical patterns are entered in
dictionaries depending on whether they are fixed expressions or not, the
idiomatic or compositional nature of combinations is easily misrepresented.
Besides, variations, functions and syntactic behaviour of lexical patterns are
typically under -reported in dictionaries.
The obvious difficulties created by the impossibility of a ppropriately
grouping lexical patterns in dictionaries have encouraged specialists to explore
their behaviour and characteristics even more minutely.
The pioneer setting the foundations of this activi ty is Mel’čuk. His
extensive work in combinatorics, both within and outside the former Soviet
Union, has led to applications in lexicography in the form of Explanatory
Combinatorial Dictionaries.

Metalexicographers such as Zgusta (1971) and Svensen (1993) should
also be mentioned. Different from Bejoint (1994: 211 ff.), who refers to the
variability of idioms and the wider phenomenon of idiomaticity, Zgusta and
Svensen focus on collocations and fixed expressions .
Moreover, Zgusta (1967) sets out nine crit eria for distinguishing multi –
word lexical items from free combinations . The aspects envisaged are the
fixedness and non -compositionality of such patterns, as well as the fact that they
may be replaced by single -word equivalents. Such criteria, however, ar e more
useful in distinguishing fixed compounds from recurrent collocations .
An important stage in the lexicographic study of lexical patterns is
represented by the research activity carried out by Cowie, Mackin and McCaig
(1983), which had as a result th e creation of an extremely useful tool both for
the specialists focussing on the study of (idiomatic) lexical patterns and for the
large public.
The two volumes of the Oxford Dictionary of Current Idiomatic English
(ODCIE 1975, 1983) represent a valuable contribution of Cowie et al. to the
study of idiomacity due to the special attention devoted to the variability of
idioms and to the proper use of their collocates.
The method used by Cowie et al. to identify idioms consists in using a set
of tests which serve to separate those multiword expressions which are idioms
from those which are not. In their opinion, the formal and semantic features
characterizing idioms are compositeness and semantic unity.
Cowie et al.’s (1983: XIII) approach is even more usefu l because it
enables identification of semi -idioms , which are further classified into idioms
with figurative senses in the non -literal member of the combination (e.g. jog in
jog somebody’s memory ), and idioms which allow restricted lexical variations
(e.g. cardinal error/ sin/ virtue ). Open collocations are borderline cases which
are made up of freely recombinable elements (e.g. fill/ empty/ drain the sink and
fill the sink/ basin/ bucket ) and oppose restricted collocations (also labelled
semi -idioms ).
I share Fernando’s (1996) opinion that, besides the obvious contribution
to the lexicographic studies, the contribution made by Cowie et al.
“lies in their analysis of how grammar and meaning complement each other to
create idiomacity, on the one hand, and o f how idioms vary in their lexical
integrity with varying consequences for their idiomacity, on the other” (Fernando
1996: 11).
The list of lexicographers is far from being complete without John
Sinclair’s name. He has played a leading role in developing c orpus linguistics in
Britain as a basis for both phraseological research and dictionary making.
Rosamund Moon and Bengt Altenberg conform to this tradition, as well.

Nevertheless, mention should be made that while making frequency based
analysis, Moon and Altenberg explore more structural and pragmatic issues.
In her analysis of fixed expressions and idioms from the perspective of
lexicography, Moon (1992a, 1992b) insists on the necessity to include
pragmatic information for the patterned dictionary entri es. She considers that
the impossibility to appropriately integrate fixed expressions and idioms in
dictionaries results from the fact that dictionaries set out to identify not the
lexical patterns, but the lexical items of a language and the level (clause , group,
structure, word, sense, morpheme) at which meanings are lexicalized or coded.
This view resembles Chomsky’s charact erization of the TG lexicon as
“the full set of irregularities of the language” (1965: 142), or Halliday’s
definition of lexis as “ the most delicate form of grammar ” (1966: 153 and
elsewhere). Referring to lexicalization, Pawley (1986) states that
lexicographical views are more productive th an the grammatical or TG ones.
Furthermore, Čermak (1994: 185) points out that systematic lexic ographical
description shows up inadequacies in other models which were built around a
few selected fixed expressions and idioms, rather than the whole lexicon.
To sum up, “while dictionaries themselves may be based on faulty
assumptions or imperfect evid ence, and the information that they provide is
often partial, their inventories of fixed expressions and idioms are basic
resources for lexicologists in the field” (Moon 1998a: 18).

1.1.3 . SYNTACTIC APPROACHES

Different from the lexical approaches , the syntactic ones analyze the various
types of lexical patterns from the perspective of transformational or
transformational -generative (TG) grammar.
Aspects such as the syntactic or grammatical anomalousness of strings
support their classification as non-compositional units, the exceptions to
syntactic rules or the unique realizations of rules being visible in the lexical
patterns.
Since lexical patterns are not productive and freely -generated structures,
on the one hand, and productivity is part and parce l of the TG model, on the
other, specialists generally agree that their analysis from the perspective of this
model is impossible. Hence, Radford’s (1988: passim) reference to sets or
classes of anomalous expressions and Harris’ (1 991: 43) opinion that the re is “ a
finite learnable stock of ‘ idiomatic ’ material ” outside the rules of the language
system which are worth mentioning.
Nevertheless, the specialists working in the syntactic tradition often focus
on pure idioms , especially those which are semantical ly opaque or semi –
transparent and which are homographic with compositional strings, as well as

on lexical patterns containing verbs with complementation or including phrasal
verbs. Such clause -like structures are more complex and interesting from the
synta ctic point of view than the noun phrases and prepositional phrases which
are lower -level structures.
Two of the first linguists to integrate idioms into a TG model are Katz
and Postal (1963), their position being restated and updated by Katz in 1973. In
their approach, the starting -point is the assertion that idioms are the ‘exceptions
that prove the rule’ because these patterns do not get their meaning from the
meanings of their syntactic parts (Katz 1973: 358).
In dealing with the syntactic aspects chara cterizing lexical patterns, Katz
and Postal (1963) distinguish between ‘ lexical idioms’ i.e. polymorphemic
words or multi -word nouns, verbs, and so on and ‘ phrasal idioms ’. If the
members of the former group are recorded in the lexicon in the same way as
ordinary words, those in the group of ‘ phrasal idioms ’ are recorded separately,
in a so -called ‘idiom list’ which specifies where the phrasal idioms are non –
productive and transformations are blocked.
Two significant contributions made before Katz’s restate ment, i.e.
Weinreich’s 1969, Fraser’s 1970, also refer to the notion of the ‘idiom list’ and
the necessity of indicating transformational deficiencies.
While attempting to make observations on the complexity of lexical
patterns with the Katz -Postal and Ch omskyan models, Weinreich (1969)
distinguishes complex lexemes from idioms , ignoring well -formed
compositional formulae. He suggests using an optional ‘Idiom Comparison
Rule’ which matches terminal strings against an idiom list and marking idiom
elements w hich cannot undergo any transformations.
Fraser’s contribution (1970) is also worth mentioning. Being concerned
with the exploration of the transformational potential of idioms , Fraser sets out
a hierarchy of seven degrees of idiom frozenness (L 6 – Unres tricted; L 5 –
Reconstitution; L 4 – Extraction; L 3 – Permutation; L 2 – Insertion; L 1 –
Adjunction; L 0 – Completely frozen ) arguing that no true idioms can belong to
level 6. While certain idioms such as kick the bucket take only the gerundive
transfo rmation being place able at L 1, others, such as read the riot act can
undergo indirect object movement, two types of passive transformation, the
gerundive and action nominalization transformations, hence apparently
qualifying for L 6.
In addition, by anal yzing the possible variations an idiom may or may not
undergo, Fraser incidentally draws attention on the stylistic effects that
transformations of idioms, especially the less common ones, can achieve.
Nevertheless, Fraser is fully aware of the fact that a ssessments of frozenness are
idiolectal. He believes that his hierarchy is solid and useful as it can facilitate a
better classification of the idiomacity units and their annotation within the idiom
list. This view is not shared by Katz (1973), who suggest s describing patter ns

by using the +/ – idiom feature.
A more recent, significant contribution to the study of lexical patterns
from the perspective of syntax is that of Huston and Francis (2000). Many of
the views presented in previous studies (Francis 199 3, 1995) are included in this
joint research on pattern grammar and supplemented by further arguments and
examples.
Francis (1993, 1995) takes up and develops many of Sinclair’s (1987)
ideas regarding the association of meaning and pattern, but while Sinc lair
focuses on the fact that different senses of polysemous words are distin guished
by differences in typical pattern use, Francis tries to demonstrate that certain
patterns ‘select’ words of particular meanings. For example, she notes (1995)
that in pat terns such as It is interesting/ likely/ clear/ important/ true that or It is
useful/ sensible/ possible to built on the structure it + link verb + adjective +
clause , the adjectives fall into a limited number of meaning groups called by her
“modality, abi lity, importance, predictability, obviousness, value and
appropriacy, rationality, truth”. Although most adjectives co -occur with a
particular kind of clause (i.e. either a ‘ that clause’ or a ‘ to-infinitive clause’),
there are adjectives, such as possible , which occur with both types of clauses.
As regards the restrictions on adjective choice in patterns with
introductory it, Francis (1995) states that these limitations are not a mere
consequence of the nature of that-clauses and to-infinitive clauses, bu t of the
grammar of these adjectives (or these adjective -senses). To put it differently, the
observations about the adjectives cannot be restricted to either lexical or
grammatical description.
I agree to the fact that such a statement reinforces the gene rally shared
view that lexis and grammar should not be treated as separate categories, which
is also clearly stated by Francis:
“Particular syntactic structures tend to co -occur with particular lexical items,
and – the other side of the coin – lexical ite ms seem to occur in a particular
range of structures. In short, syntax and lexis are co -selected, and we cannot
look at either of them in isolation ” (Francis 1995: 23, emphasis added).
Francis (1995) also follows Sinclair (1987) in proposing a hypothesis of
how language is encoded which prioritises meaning and lexis instead of
grammatical choices. She states that “[…] lexis is communicatively prior to
syntax. As communicators, we do not proceed by selecting syntactic structures
and independently choosing le xical items to slot into them. Instead, we have
concepts to convey and communica tive choices to make which require central
lexical items, and these choices find themselves syntactic structures in which
they can be said comfortably and grammatically” (Fran cis 1995: 27).
Considering these representative syntactic approaches to lexical patterns,
the conclusion may be drawn that they play an important part in constructing

the general frame of possible approaches to the various types of word
combinations.
The fact should be pointed out that the investigation of transformational
deficiencies is agreed to have influenced other phraseologists such as Makkai
(1972) and Wood (1981) who did not necessarily operate within a TG model of
language, as well as the constr uction of dictionaries which attempt to record the
transformational potential of phrasal lexemes. In addition, the syntactic
structure of lexical patterns is a factor which favours the marking of idiom –
features, at the same time encouraging awareness of compositionality , as well
as of non-compositionality (see 2.1.2.2 ).
Although the concept of ‘idiom list’ has often been questioned, and
psycholinguists suggest that lexical patterns may not be stored in the mental
lexicon separately, the syntactic approach provides useful evidence for the lists
of lexical patterns which are excluded from such lists. The making up of lists is
a strategy which is used, for tagging and parsing corpora, on the one hand, and
for identifying lexical patterns in the available corpo ra, on the other.
In spite of being very effective in describing or characterizing the
morphology of lexical patterns, the syntactic approaches and models generally
fail to account for the range of lexical patterning. However, this aspect is
contradicted by Weinreich (1969) and Harris (1991).
Thus, Weinreich (1969) considers that the potential ambiguity of idioms
arises from their having multiword literal counterparts in other discoursal
environments. In an attempt to identify essential features of idioms , Weinreich
opposes the ‘idiomacity of expressions’ to the ‘stability of collocations’,
mentioning that, although collocations and idioms both reflect a co -occurrence
phenomenon, the co -occurrence of words in an idiom results in a special
semantic relation ship which is not evident in the case of collocations
(Weinreich 1969: 42).
As regards the approach suggested by Harris (1991) , he starts from the
description of language as an orderly system which operates by rules, in order to
support his view that idiom s and frozen expressions are cases of constraints or
narrow selection in lexico -syntactic structures ( Harris 1991: 67).
Last but not least, many studies focussing on the syntactic perspective
deal both with phrasal verbs and with idioms, although these tw o lexical
patterns undergo relatively different syntactic operations and processes, making
the generalizations and discussions irrelevant to lexical patterns of other kinds.
Syntactic approaches often underplay the role of lexical patterning or the
motivat ion underlying the development and usage of lexical patterns (Chafe
1968, Fillmore et al. 1988). However, one of the most serious problems in
syntax -based models is that many are based on intuition and non -authentic data,
making some of the assertions on t ransformational potential and syntactic
defectiveness hardly reliable.

1.1.4 . SEMANTIC APPROACHES

The more traditional semantic approaches to lexical patterns focus on
structural aspects, more precisely on their semantic compositionality , or the
possibil ity of semantically analyzing such patterns. As regards the development
of semantic approaches to lexical phrases, it is generally agreed that it extends
from the pioneering work of J. R. Firth (1957) , via the neo -Firthians Michael
Halliday (1966) and John Sinclair (1966) , to many of the specialists currently
analyzing phraseology in computer -stored corpora of spoken and written
English.
In discussing the concept of word, Hocket (1958) suggests a reductionist
model of the lexicon, with all irreducible elem ents whether they are
morphemes, words, clauses, or even exchanges ( Hocket 1958: 166 -173). The
exceptions to the rules of free composition are the idioms, in his terms.
Moreover, while analyzing the formation of idioms , he notices that they may be
context ual and existential, and that idioms are traceable not only in language,
but also in other semiotic systems ( Hocket 1958: 303 -309). Such an argument
proves extremely important for a view with respect to the compositionality of
units, for the extension of ‘ idiom’ to larger units such as clauses, and for
establishing that idioms are not limited to the set lexical phrases listed in
dictionaries , but may also arise in discourse. Nevertheless, the fact cannot be
ignored that Hocket does not suggest any possible scale of lexical patterns in his
approach and, by extending the term idiom to morphemes and ad hoc
formulations, he reduces the applicability of this term to a great extent.
Working within the stratificational framework (see Lamb 1962 and
elsewhere), Hea ley (1968) suggests an approach to lexical patterns which
envisages cross -cultural, lexicographical, and pedagogical applications. In doing
so, he provides a structural classification of such patterns, and identifies the
transformational deficiencies chara cterizing them.
Furthermore, in his structural approach to lexical patterns, Makkai (1972)
suggests a separation of idioms from non-idioms. Although dichotomous, this
distinction allows for the identification of two qualitatively different kinds of
fixed lexical pattern s, or idioms , in Makkai’s terms (1972: 38), namely: idioms
of encoding and idioms of decoding .
The idioms in the former category, which are also termed ‘phraseological
peculiarities’ or ‘phraseological idioms’ (Makkai 1972: 56f.), involve
collocational preference restrictions, whereas the idioms of decoding are
misleading ‘lexical clusters’ including examples such as hot potato and fly off
the handle . In an attempt to provide a more thorough description of the idioms
in this latter category , Makkai analyses them from the perspective of the
stratificational model, thus coming to the conclusion that idioms of decoding
belong to one of two ‘idiomaticity areas’ in English, i.e. either the lexemic or

the sememic one (Makkai 1972: l17 ff.). In his opinion, lexemic idioms include
phrasal verbs , pure idioms such as spill the beans, and opaque compounds
such as forefinger and blackbird creating most often problems of lexico –
grammar and semantics, whereas sememic idioms are problematic from a
pragmatic or sociocultural point of view, being best represented by proverbs
and formulaic greetings.
Mention should be made that Makkai also classifies lexemic idioms
according to their surface structure, excluding combinations such as hue and cry
or to and fro which contain unique items. These combinations are referred to as
‘pseudo -idioms ’ (1972: 123, 340, and passim) because they are not ambiguous ,
as true idioms are.
Althou gh useful and detailed, Makkai’ s study on lexical patterns is
generally agreed to have s ome shortcomings. This is due, on the one hand, to
the fact that some of his distinctions are difficult to follow through and to the
fact that certain important and problematic lexical patterns such as collocation s
are virtually ignored, on the other.
Beck er (1975) is another specialist whose view on patterns fits with the
semantic approaches. In spite of identifying various types of patterns which are
lexically or grammaticality determined, Becker manages to suggest a model
which organizes, rather than exp lains facts.
In his attempt to establish a narrow category of pure idioms , Wood
(1981) explores the syntactic, lexical, and semantic properties of lexical
patterns, but , in doing so , she effectively establishes a model of the gradations
of semantic composi tionality which characterize the full range of lexical
patterns.
Gläser (1988) discusses the reference of idioms and their characteristic
transformational deficiencies in order to explore idiomacity and establish a
typology of idioms and other fixed patte rns.
Instead of a conclusion , mention should be made that semantic
approaches to lexical patterns are not always clearly differentiated from other
possible approaches. This is, for example, the case of Allerton’s study (1984) in
which the analysis of lexi cal patterns in terms of semantics, syntax, locution,
and pragmatics and of lexical co -occurrence restrictions combines with a
separation of idioms from metaphors . Useful as they may be, these studies
mainly consider lexical patterns from the perspective o f the lexicon, rather than
that of text.

1.2. FUNCTIONAL APPROACHES TO LEXICAL PATTERNS

Different from the formal approaches to lexical patterns, which focus on the
interdependence of their structure and the various linguistic levels (lexis, syntax,

semantics, etc.) , functional approaches envisage the more practical aspects, that
is their use by and accessibility to the end -users.
Moreover, if formal approaches concentrate on the internal features of
lexical patterns and on their roles within the lexico n, functional ones include
more behaviourist a spects which look at lexical patterns as encoding or
enabling devices.
Since psycholinguistic approaches focus on how language is encoded in
chunks, rather than in the individual words making up lexical patter ns (Peters
1983, Sinclair 1987), the conclusion may be easily drawn that they fit in the
class of approaches which regard lexical patterns as encoding devices.
On the other hand, the analysis of routine formulae from the perspective
of discourse as instrum ents for fostering interactions, or as gambits, points to
the use of lexical patterns as enabling devices (Coulmas 1979, 1981,
Drazdauskiene 1981, Strässler 1982, Pawley and Syder 1983 , Carter 1987,
Schiffrin 1987, Tannen 1989, Nattinger and DeCarrico 1992 , McCarthy and
Carter 1994, Drew and Holt 1995, Aijmer 1996).

1.2.1 . PRAGMATIC AND STYLISTIC APPROACHES

One of the first attempts directed towards the analysis of lexical patterns from
the perspective of pragmatics can be traced back to the 1960 ’s, wh en Amosova
(1963) identifies different kinds of phraseolog ical units by means of a ‘context –
logical analysis’. In her opinion, considerations of context may prove very
useful as they may give an indication of whether potential units are ambiguous
within th eir contexts and of whether they can have their idiomatic or unitary
meanings in other contexts. In separating pure idioms from phrasemes ,
phraseoloids and fixed combinations , Amosova explains that the class of
‘phrasemes ’ includes frozen collocations or compounds where one element
has a meaning unique to the combination, ‘phraseoloids’ are represented by
restricted collocations where there is limited paradigmatic variability, whereas
‘fixed combinations’ are limited to formulae or fixed collocations that a re
transparent (Amosova 1963 in Moon 1998: 12 -13).
In his discussion on ‘standard’ or ‘conventional’ uses of certain linguistic
forms to perform directives, Searle (1975) argues that one should speak
idiomatically unless there is some special reason not to do this. He adds that a
sentence has to be idiomatic in order to be a plausible candidate for its use as an
indirect speech act, and that “within the class of idiomatic sentences, some
forms tend to become entrenched as conventional devices for indirect s peech
acts” (1975: 77 -78). Although this is a side argument, which does not fully
develop in Searle’s opinion, since the 1980’s, numerous studies in the field of

corpus – based lexicography and in language teaching have emphasized the
pervasive occurrence of phrase -like units of idiomatic language use.
A rather different approach is suggested by Laney (1986) who, focussing
on pure idioms and other similar patterns, integrates them in a pragmatic
classification with the aim of showing how they fit into real -time discourse.
Although Laney pays more attention to pragmatic concepts such as ‘relationship
with the world’ rather than to interactional devices, her model proves useful in
designing a dictionary of lexical patterns.
Furthermore, Pustejovski, Bergler a nd Anick (1993) analyze the relations
between words that go beyond the simple co -occurrence relations. By using the
‘qualia’ structure, Pustejovski et al. specify the different aspects of word
meaning and give an insight into the combinability of lexical u nits in texts .
However, since ‘qualia’ is a term used to denote semantic features which allow
speakers to derive appropriate meanings of lexical patterns compositionally, the
conclusion may be drawn that Pustejovski et al.’s approach conforms to the
framew ork of a semantic theory.
Similarly to Amosova and Pustejovski et al., Mel’čuk and Warner (1994)
are in favour of investigating the ‘word in context ’. The ‘Meaning -text theory’
forms the basis of their study and has as a starting point the premise that wor ds
can be described in a completely rule -governed way and that units of more than
one word can be derived from the meaning of individual words and the lexical
functions between these individual words. Different from Pustejovsky’s
approach, Mel’čuk and Warn er’s model is rather limited, due to the fact that the
nature of natural language cannot be grasped by using a rule -gouverned model.
(Mel’čuk and Warner 1994 in Weigand 1998 b: 40 – 41)
A more recent solid contribution to the pragmatic study of lexical patt erns
is made by Chitra Fernando (1996), who, starting from a possible classification
of word combinations on the scale of idiomacity, develops a well -structured
model for the pragmatic analysis of idioms .
Strässler’s approach (1982) to the study of lexica l patterns is also worth
mentioning, as this contribution represents the first major study on the functions
of idioms . While exploring the idiom structure through his review of scholarly
work, Strässler notices that the impossibility to deduce the meaning of an idiom
from its constituent parts entails certain barriers to its structural changes
(Strässler 1982: 24, Fernando 1996: 13)
Interestingly, Strässler (1982) suggests approaching idioms from the
stylistic perspective (see 4.3. – 4.5.). He believes that significant variables such
as social status , age, education , and profession , or the gap between the
partners with respect to these parameters, are important in idiom interpretation
(Strässler 1982: 78) because “when using an idiom , the speaker conveys mor e
than its semantic content. He either establishes social hierarchy, or he tests the
hearer’s opinion in this matter” (Strässler 1982: 116, Fernando 1996: 14)

Moreover, Strässler (1982) states that the use of idioms is a useful
stylistic strategy made poss ible by the ir expressive meanings, as compared to
their non -idiomatic synonyms. He explains that idiomatic patterns such as sniff
it and kick the bucket , which are marked as slang and convey irreverent
jocularity, are more expressive than the synonymous un marked verb to die .
In spite of lacking a deeper analysis of the special information conveyed
by idioms , Strassler’s approach is a valuable contribution to the study of this
type of fixed lexical patterns, as it provides very useful insights into their de ictic
functions.
To conclude , although not intimately bound to lexical patterns, studies
such as the ones referred to above are equally important and useful because they
establish a broader and more holistic approach to the various types of lexical
patter ns, particularly as a phenomenon of discourse . Moreover, their
appropriate decoding cannot be dealt with unless the respective lexical patterns
are integrated in context , or identified with a specific style .

1.2.2 . METHODOLOGICAL APPROACHES

A widely s pread perspective from which lexical phrases have been investigated
is that of language teaching and learning. As stated in a previous paper
(Dumitra șcu 2009a), teaching English lexical patterns is extremely important
because it facilitates the learners’ access to more routinized aspects of language
production and to the essential skills of maintaining discoursal relations through
language use. Moreov er, the teaching of such word combinations in discourse
may prove incredibly useful due to the fact that it “encourages appreciation of
the syntactic, semantic and pragmatic functions of lexical items at all levels”
(Carter 1998: 240). As regards the proce ss of learning, vocabulary
comprehension and production play an important role, the learners’ appropriate
use of lexical patterns being relevant to accuracy, i.e. to conformity to an
arbitrary norm, and helps non -native speakers of English sound
indistingu ishable from a native speaker (Dumitrașcu 2009a: 395).
The list of specialists investigating lexical patterns from this perspective
includes names such as Pawley and Syder (1983), Nattinger and DeCarrico
(1989, 1992), and more recently, Willis (1990), Lew is (1993) and Hill (2000).
The idea emerging from these studies is that teaching lexical patterns
should represent a prio rity in all educational systems because of their high
frequency in language use, on the one hand, and of their importance to a
‘native -like’ production of the language, on the other.
In their approach to teaching lexical patterns, Pawley and Syder (1983:
191) argue that

“fluent and idiomatic control of a language rests to a considerable extent on
knowledge of a body of ‘sentence stems’ which are ‘institutionalized’or
‘lexicalized’. The relative fixedness of the phrases allows speakers to concentrate
on other aspects of discourse, and thus to achieve the fluency that we associate
with native speakers”. (Pawley and Syder 1983: 191).
More over, they make explain the notion of ‘lexicalized sentence stem’
defin ing it as ”a unit of clause length or longer, whose gram matical form and
lexical content is wholly or largely fixed; its fixed elements form a standard
label for a culturally recognize d concept, a term in the langua ge” (id. ibid ).
A similar definition is provided by Nattinger and DeCarrico (1989: 1l8)
who, discussing on lexical phrases mention that “these phrases are patterned
sequences, usually consisting of a syntactic frame that con tains slots for various
fillers, and run the gamut from completely fixed, unvarying phrases to phrases
that are highly variable”. This statement will be refined by a further reference to
the fact that “lexical phrases are fixed in their functional applicat ion, as well as
in their form” (Nattinger and DeCarrico 1992: 11).
As mentioned in one of my papers (Dumitrașcu 2009a: 388), recent
studies in the field of language teaching have proved that the lexical pattern in
which a word occurs is essential to the m eaning of that word an d cannot be
separated from it. Moreover, as suggested by Hunston and Francis (2000: 271),
“learning such strings rather than individual words enables the learner to
compose lengthy utterances with the minimum effort”. Two main argumen ts to
be brought forward are: 1) “most language is composed of prefabricated
strings”, and 2) storing lexical patterns is accepted to have a long -term result s
for the learners, i.e. it ensures their fluency and accuracy in the foreign
language. The innovat ive character and the complexity of such an approach to
vocabulary teaching results from the fact that teachers have to guide their
learners to notice and appropriately analyze the “relations between lexical
items: a) above the sentence level, b) across c onversational turn boundaries,
and c) within the broad framework of discourse organization” (McCarthy 1984 :
14-15 in Carter 1998: 220 -221).
In the larger frame of methodological approaches to lexical patterns, there
are certain studies on language acquisit ion which give priority to the
importance of psycholinguistic aspects in this process. Such studies suggest
that language is learned, stored, retrieved, and produced in holophrases and
other multi -word items, not just as individual words or terms.
Observi ng the widespread use of language routines in young children,
Peters (1983) refers to the usefulness of such strings and routines as
communicative strategies or ‘shortcutting devices’ for mature speakers (1983:
3), as well. She argues that lexical patterns may continue to be stored as
complete or partial units, in addition to their component parts , even when

speakers make analogies and become aware of the compositionality of such
patterns (1983: 90).
Furthermore, looking at formulaic speech in adult langu age learners,
Bolander (1989) argues that ‘chunk processing’ facilitates language processing
in general and is a way of acquiring lexico -grammatical structures. Similarly,
Bahns et al. (1986) analyze formulae in child language learning, suggesting that
they play a crucial part in this process.
Observing the limited usefulness of language pattern acquisition, Krashen
and Scarcella (1978: 298 in Moon 1998: 30) state that the use of routines and
formulae is only one form of L2 language acquisition, insufficien t in itself for
full competence, but nevertheless useful ‘for establishing social relations and
encouraging intake’.
Last, but not least, in his discussion regarding prepatterning and
repetition in language, T annen (1989) points out that fixed lexical pat terns
represent a good illustration of structure, rather than redundancy and repetition.
Lexical patterns are indeed important for researchers in the field of
language teaching, because they may allow language learners to produce
language without undue he sitation or disfluency and to produce language that is
phraseologically similar to that of native speakers. The same idea is pointed out
by Cowie (1992 ) who comments that
“[I]t is impossible to perform at a level acceptable to native users, in writing or
speech, without controlling an appropriate range of multiword units. Moreover,
the demands of creative expression in the foreign language rests, as it does for
native speakers and writers, on prior knowledge of a repertoire of such
expressions ” (Cowie 199 2:10).

1.2.3 . PSYCHOLINGUISTIC APPROACHES

Psycholinguistic approaches to lexical patterns are generally believed to have
some common points with language teaching studies. This is due to the fact that
both types of approaches are concerned with the freq uency and therefore
importance of lexical patterns, with the varying degrees to which such patterns
are open to variation in wording, their functions, and their importance to the
creation of a language model which gives lexis and grammar equal priority.
Moreover, both approaches admit the importance of lexical patterns not only in
language storage, but also in language acquisition.
The psycholinguistic approaches to l exical patterns are as numerous as
various.
Krashen (1981) , for example, argues that the acquisition of prefabricated
routines and prefabricated patterns is a process independent of what he calls the
‘creative construction process’, and that such routines and patterns are not

necessarily integrated in the more important creative language. The conclusion
he draws is that
“[T]he use of routines and patterns is certainly a part of language, but it is
probably not a large part” (Krashen 1981: 98) and that “[T]he available evidence
indicates that routines and pattern s are essentially and fundament ally different
from creative language” (Krashen 1981: 99).
On the other hand, Nattinger and DeCarrico (1989) believe that language
acquisition by non -native learners may provide useful evidence for the ways in
which L2 speakers analyze language. To put it in their words,
“lexical phrases may also provide the raw material itself for language acquisi tion
… Later, on analogy with many similar phrases , the learners break these chunks
down into sentence frames that contain slots for various fillers” (Nattin ger and
DeCarrico 1989: 132).
Furthermore, Peters, suggests “that formulaic speech … is merely a facet
of creative language” (Peters 1983: 4), arguing that children begin by acquiring
phrases, rather than words or structure, and that these phrases are la ter analysed
into a system . He point s out that children find phrases to be more meaningful
than words or syntactic systems:
”[I]t is not a dictionary of morphemes that the child is exposed to, but rather an
intermittent stream of speech sounds containing chunks, often longer than a
single word, that recur with varying frequency. It is out of this stream of
unknown meaning and structure that the child must attempt to capture some
pieces in order to determine their meaning and to preserve them for future us e”
(Peters 1983: 5).
Some researchers consider that the distinction between formula and
creativity is not a dichotomy, but a continuum. In Weinert’s (1995) opinion
such an approach leads to a theory of language production made up of several
distinct compo nents: “The view of language as a formulaic -creative continuum
suggests that the units of knowledge and production may vary, including fixed
formulas, mini -grammars, and general rules” (Weinert 1995: 198).
The blending of several components is also trace able in the processes of
language storage and language acquisition. Peters (1983: 6), for example,
argues that adult speakers of a language store the language in phrases, as well as
in the form of words and syntactic rules. Approaching this idea from a mor e
general perspective on language performance, Nattinger (1988 ) states that
“[M]any theories of language performance suggest that vocabulary is stored
redundantly, not only as individual morphemes, but also as parts of phrases, or
even as longer memorized chunks of speech, and that it is oftentimes retrieved
from memory as these preassembled chunks” (Nattinger 1988: 75) .

The conclusion drawn by Peters (1983) is that ‘dual storage’, i.e. having
phrases stored in memory both as single (lexical) items and as the syntactic
rules that allow them to be created, has as a result redundancy in language
knowledge. Nevertheless, this implies more fluidity between lexis and syntax:
“[T]here is considerable redundancy in the storage of both lexical and
syntactic inform ation. The relation between syntax and lexicon may
therefore be more fluid than is usually supposed. Under some
circumstances, an expres sion may be retrieved from the lexicon as a
single unit; under others it may be constructed from partially assembled
pieces in the lexicon, requiring somewhat more syntactic processing;
under yet other circumstances it may be constructed de novo from
morphemes. Syntax and lexicon are thus seen to be complemen tary in a
dynamic and redundant way . The same information may be present in
both, in different forms ” (1983: 90, emphasis added).
This argument is similar to Sinclair’s (1991: 104) view, who states that
syntactic rules account for only a minimal part of the grammar of a language,
and that the more important part is com posed of the phraseological constraints
upon individual lexical items. Thus, syntax is not a system indepen dent of lexis:
lexis and syntax must, ultimately, be described together.

1.3. TERMINOLOGY AND TYPOLOGIES

The idea is generally shared that a larg e amount of the language encountered is
not constructed from ‘basic’ structures and a lexicon, but occur s in sequences of
morphemes which are more or less fixed in form. Depending on the formal and
semantic restrictions characterizing such sequences, they are assigned different
terms, which are as challenging as confusing, irrespective of the language
envisaged. This is due to the fact that, sometimes, different terms are used to
describe identical or very similar kinds of unit, whereas some other times a
single term is used to denote very different phenomena.
In English phraseology, the various types of lexical patterns identified by
the specialists are referred to by terms such as: ‘composites’ (Vinogradov 1947)
‘phraseologic expressions’ (Chernuisheva 19 64), ‘word combinations’
(Akhmanova 1974, Cowie 1994), ‘phraseological units’ (Ginzburg et al 1979),
‘prefabricated routines and patterns’ (Krashen 1981), ‘sentence stems’ (Pawley
and Syder 1983), ‘formulae’ (Peters 1983), ‘fixed expressions’ Carter (1987) ,
‘phrasal lexemes’ (Moon 1998 a, b), ‘lexical phrases’ (Nattinger and DeCarrico
1989, 1992), ‘composites’ (Cowie, following Mitchell in Cowie 1988),
‘gambits’ (Keller in Cowie 1988), ‘routine formulae’ (Coulmas in Cowie 1988),
‘phrasemes’ (Melcuk 1988, 1 995), ‘formulaic language’ (Weinert 1995; Wray

1999), ‘idioms’ (Fernado 1996), ‘fixed expressions and idioms’ (FEIs) (Moon
1998), ‘collocations’, ‘colligations’ (Firth 1957, Sinclair 1998, 2001, 2004) ,
multiword units (Granger and Paquot 2008) .
Unfortunat ely, many of these terms have limited applicability, this fact
being sometimes attested by the authors themselves, who tend to choose their
terminology according to the scope of their research (see , for example,
Fernando 1996, Moon 1998, Houston and Franci s 2000).
To put it differently , the various typologies established by the specialists
in the field correspond to “the features used to categorize muti -word units and
the prio ritization of selected features” (Granger and Paquot 2009: 35). The
features mo st often taken into consideration are: 1. internal structure; 2. extent:
phrase – vs. sentence level; 3. degree of semantic (non) compositionality; 4.
degree of syntactic flexibility and collocability; and 5. discourse function. (id.
ibid.)
Although general , fixed expression , a term adopted from Alexander
(1978, 1979, Carter 1987), and others, is considered very convenient by some
specialists and used to denote different types of phrasal lexemes, phraseological
units, or multi -word lexical items: frozen coll ocations, grammatically ill –
formed collocations, proverbs, routine formulae, sayings, similes.
Nevertheless, fixed expression cannot be accepted as a satisfactory term,
since many fixed expressions of these types are not actually fixed. Moreover, as
Moon ( 1998: 2) suggests, the term fixed expression does not apply to
compound nouns, adjectives, and verbs such as civil servant, self -raising, and
rubber -stamp , to phrasal verbs such as make up and stick out , foreign phrases
such as fait accompli, che sarà, sar à, and caveat emptor and multi -word
inflectional forms of verbs, adjectives, and adverbs such as had been lying and
more careful(ly) (Moon’s examples). This is d ue to the fact that compound
words are associated with morphology, and multi -word inflectional forms are
simply part of the grammar of English.
The inappropriateness and misleading nature of the term fixed
expression is also pointed out by Stubbs (2002) . He states that
“[…] units are rarely invariant, and often not even continuous. They are
idiomat ic, but only rarely idioms; they have typical components, but are highly
variable, with probabilistic relations between the components; they are typically
realized by a sequence of several word -forms, but their boundaries do not
correspond systematically t o syntactic units; and indeed they do not fit into
traditional concepts of either lexis or syntax “ (Stubbs 2002: 59).
As far as the traditional term constructions suggested by Fillmore et al.
and Sinclair ’s terms extended units of meaning (1998) and lexical items
(Sinclair 1996) are concerned , they also have limited applicability either due to
their rather general reference and to their debatable nature . As Stubbs (2002:

62) states , “it is an odd failing of linguistics that it has no convincing descri ptive
theory of units of meaning”.
Idiom is also an ambiguous term, used in conflicting ways. On the one
hand, the term idiom is used to denote a particular manner of expressing
something in language, music, art, and so on, which characterizes a person or
group and, on the other, it denotes a particular lexical collocation or phrasal
lexeme , peculiar to a language. These uses are related to idiom as both a
superordinate and a hyponymic term (see 3.3.3.3 ) for a lexical combination,
thus further confusing the matter.
Narrower uses restrict idiom to a particular kind of fixed lexical units
such as kick the bucket , spill the beans that are fixed and semantically opaque or
metaphorical, or, traditionally, ‘not the sum of its parts’ . Such units are
sometimes call ed pure idioms (Cowie 1988: 133) being clearly separated from
grammatically ill -formed patterns such as by and , transparent metaphors
such as skate on thin ice and strings of words such as move heaven and earth
which have no possible literal meaning.
Quite often, idiom is a general term used for many kinds of multiword
item, whether semantically opaque or not. Makkai (1972) uses the term idiom
to cover non -compositional polymorphemic words such as blackbir d, as well
as collocations and constructions which are not freely fo rmed, whereas Hockett
(1958: 171 ff.) considers that single morphemes should as well be included in
this category, since their meanings cannot be deducible.
Specialists focussing on the speech act theory, suggest that idiom can
also be use d to refer to conventionalized formulas with an illocutionary function
(Sadock 1974, Morgan 1978, Fernando 1996). However, Sadock (1972) draws
attention to the ambiguity of utterances which have more than one pragmatic
function, this aspect leading such fo rmulas to status as idioms .
Similarly, Gibbs (1986) draws attention to ways in which some indirect
speech acts are conventionalized, hence to the possibility of identifying
conventionalized forms as idioms . Pointing out practical problems with this
classi fication, Levinson (1983) and Coulthard (1985) underline the fact that
formulae such as Can you pass the salt? are rarely recognized as idioms in
lexicology.
Fillmore et al . (1988) use the term formal idiom to refer to
semigrammatical structures such as ‘N OUN 1 to NOUN 2’, which are in fact
syntagmatic equivalents of phraseological collocations .
Different from the terms mentioned so far , the label phraseological unit
is used in some Slavonic and German studies as a superordinate term for multi –
word lexica l items (Gläser 1984: 348) in the same way as phraseme is
sometimes used as a superordinate term outside Anglo -American traditions.
There are, however, other uses of both terms. For example, Vinogradov (1947)

restricts phraseological unit to more metaphori cal items, and Amosova (1963)
uses phraseme for multi -word items which are not pure idioms .
Different from Amosova, Mel’čuk ’s attaches the term phraseme to
idioms, as well . Starting from the idea that phrasemes cannot be constructed
from words or simple phrases according to general rules of language, but they
have to be stored and used as a whole, Mel’čuk ( 1998 : 24) points out the fact
that phraseme s are predominant in any language, most often outnumber ing
words . In his opinion, phrasemes may be of two types , i.e. pragmatic and
semantic . As regards the lexical patterns making up these two classes of
phrasemes , the forme r category includ es pragmatemes, whereas t he latter is
represented by idioms , collocations and quasi -idioms .

Fig. 1 a. Classification of phrasemes (source Mel’ čuk 1998: 30)

Another possible representation of Mel’ čuk’s typology is suggested b y
Granger and Paquot (2009: 37).

Fig. 1 b. Classification of phrasemes. Mel’ čuk ’s typology (1998 )
(source Granger and Paquot 2009: 37 )
PHRASEMES
semantic
phrasemes pragmatic phrasemes
or pragmatemes

semi -phrasemes or
collocations full phrasemes or
idioms quasi -phrasemes or
quasi -idioms
PHRASEME
S
PRAGMATIC
PHRASEMES SEMANTIC
PHRASEMES
idioms collocations quasi-idioms
pragmatemes

Moreover, considering a corpus -based approach to multi -word units ,
Granger and Paquot (2009: 42) suggest a classification of phrasemes from the
perspective of the function they may have, i.e referential , textual or
communicative , mentioning that their c ategorization represents an extended
version of Burger’s (1998) classification :

Fig. 1.c. The phraseological spectrum (source Granger and Paquot 2009: 42)

Phraseological unit and phraseme can be identified with Lyons’s
phrasal lexeme (1977: 23), a term also used by Moon (1998b: 80). She
considers that phrasal lexemes include phrases and idioms , i.e. lexical patterns
which, due to their semantic, lexico -grammatical, or pragmatic features are
regarded as holistic units, rathe r than compositional strings. In other words,
Moon uses the term phrasal lexemes to denote pure idioms , proverbs ,
similes , institutionalized metaphors , formulae , sayings and various other
kinds of institutionalized collocation.
Some of the word combination s included in Moon’s category of phrasal
lexems are also found in Cowie’s typology of word combinations . The

PHRASEMES
REFERENTIAL
FUNCTION
REFERENTIAL
PHRASEMES TEXTUAL
FUNCTION
TEXTUAL
PHRASEMES COMMUNICATIVE
FUNCTION
COMMUNICATIVE
PHRASEMES

(Lexical) collocations Complex prepositions Speech act formulae
Idioms Complex conjunctions Attitudinal formulae
Irreversible bi -and Linking adverbials Proverbs and proverb
trinominals Textual sentence stems fragments
Similes Commonplaces
Compounds Slogans
Phrasal verbs Idiomatic sentences
Grammatical collocations Quotations

terminology suggested by Cowie has been further integrated in a series of useful
typologies which compare and contrast either the terms used by dif ferent
specialists to denote one and the same type of lexical pattern, or the various
types of lexical patterns and their formal and semantic behaviour.

Fig. 2 . Cowie’ s (1998, 2001) classificat ion of word combinations
(source Granger and Paquot 2009: 36)

A good illustration of the former situation is the table suggested by
Cowie (1998) in which some of the traditional Russian terminology is
contrasted with the modern one, giving thus an insight into the terminological
similar ities and dissimilarities traceable at the levels of general, sentence -like
and word -like lexical categories:

Author General
category Sentence -like or
pragmatic unit Word -like or
semantic unit
Chernuisheva
(1964) Phraseological
unit Phraseological
combi nation –
Zgusta (1971) Set
combination Set group –
Mel’čuk
(1988 b) Phraseme or set
phrase Pragmatic phraseme, or
Pragmateme Semantic
phraseme
Gläser (1988 a) Phraseological
unit Proposition Nomination
Cowie (1988) Word –
combination Functional expressi on Composite
Howarth (1996) Word –
combination Functional expression Composite unit
Table 1 . Terms used for ‘sentence -like’ and ‘word -like combinations’
(source Cowie 1998: 5)
WORD
COMBINATIONS

COMPOSITES

FORMULAE

RESTRICTED
COLLOCATIONS
PURE
IDIOMS
FIGURATIVE
IDIOMS
ROUTINE
FORMULAE
SPEECH
FORMULAE

As regards the latter situation, considering the degree of fixity of lexical
patterns, Moon (1998), distinguishes between three types of so -called ‘fixed
expressions’, namely:

 ‘Anomalous collocations’ include examples such as by and large or
through thick and thin , which cannot be analysed according to the
normal rules governing En glish, in that a preposition ( by) and an
adjective ( large ) are not normally able to be coordinated, and adjectives
such as thick and thin cannot normally occur as the completive of a
preposition. Examples such as kith and kin , in which one of the
component s ‘is fossilised within that particular collocation’ ( kith, for
example, is found only in this pattern), are also included in this
category.
 ‘Formulae’ include lexical patterns such as proverbs , slogans ,
quotations , gambits , and closed -set turns , (e.g. You have never had it
so good and Shut your mouth ) which are in no way anomalous with
respect to the language as a whole.
 ‘Fossilised or frozen metaphors’ include ‘pure idioms’ such as skate
on thin ice or spill the beans , which are anomalous only in the sens e
that they cannot be manipulated grammatically. Thus, each part of the
idiom ( skate , thin, ice, and spill, beans , respectively) is not treated as a
separate lexical item by speakers, but as part of a phrase. (Moon’s
examples)

Similar typologies have been presented in the previous sections and will
be supplemented by further typologies, which, while focussing on collocations ,
will compare and contrast them with other lexical patterns formally or
semantically marked.
Mention should be made that in her stud y devoted to fixed expressions
and idioms , Moon (1998) also suggests a classification of such lexical patterns
in terms of their text functions , i.e. according to the way in which they
contribute to the content and structure of a text. She explains that , although the
contribution of fixed lexical patterns is instantial and bound up with context
(see 3.5.3 ), it is nevertheless possible to generalize and to chart typical
functions.
Moon’s categorization of such lexical patterns is made according to five
functions which are related, but not identical to Halliday’s model of the
semantic components of language (Halliday 1978: 116 ff.). Thus, she identifies
informational , evaluative , situational , modalizing and organizational
expressions and idioms , each of these categories having specific functions .

category function examples
informational stating proposition,
conveying information rub shoulders with , in the
running , catch sight of
soemthing for sale
evaluative conveying speaker’s
evaluation and attitude kid's stuff , a different/fine
kettle of fish , near the
knuckle , it's an ill wind (that
blows nobody any good)
situational relating to extralinguistic
context, responding to
situation excuse me! ;long time no see
;knock it off ;talk of the devil
modalizing conveying truth values,
advice, requests, etc.
I kid you not , you know what
I mean, to all intents and
purposes
if in doubt, do nowt
organizational organizing text, signalling
discourse structure by the way , for instance,
talking of -, be that as it may

Table 2 . Text functions of FEIs (source Moon 1998: 217)

Moreover, Moon provides a classification of fixed expressions and
idioms according to their discoursal functions . The model she suggests views
texts in terms of their semantic stra tification and distinguishes between
ideational , interpersonal and textual components. Such a model is useful in
the interpretation of ongoing discourse, any selection having repercussions at all
levels, which are simultaneous.

ideational experiential
logical communica tion of ideas
connections between ideas
interpersonal interactional interrelationship between speaker
and hearer mood
illocution
personal modality
attitude
textual theme
information
cohesion thematization and thematic patterning
given/new distinction
cohesive structure

Table 3 . Discoursal functions of FEIs (Halliday 1978, Morley 1985 , in Moon
1998: 218)

Different from the text func tions of lexical patterns, which are lower –
level functions reflecting the immediate effects of such patterns within th eir co –
texts, ideational , interpersonal , and textual components operate at the level
of discourse.

logical organizational
Ideational experimental informational

Interpresonal situational

evaluative

modalizing

Fig. 3 . Ideational and interpersonal, related to FEIs functions (Moon 1998:
218)

Moreover, the textual component is best considered in terms of the ways
in which lexical patterns are placed topically and themati cally, or contribute
cohesion to their texts.
The text functions of fixed lexical patterns referred to by Moon are
common to the lexicon in general, but they apply to fixed lexical patterns, as
well. As regards the roles that fixed lexical patterns have in real -time discourse,
I agree with Moon that they are equally important as their lexical, syntactic, and
semantic characteristics. “Neglect ing or ignoring these roles may lead to
discoursal ill -formedness in encoding and to misinterpretation in decoding”
(Moon 1998: 219).
Although often confusing, such a consistent terminology proves the
specialists’ constant interest in the study of lexical patterns, as well as their
increasing awareness as regards the frequent use of ready -made memorized
combinations in written and spoken language. It also proves the crucial part
such combinations play in language ac quisition, on the one hand, and in
language production, on the other.

1.4. LEXICAL PHRASES ON THE SCALE OF IDIOMACITY

While closely related to idioms, the term idiomaticity is neither identical with ,
nor restricted to idiom. Nevertheless, idioms and idiomacity have a common
basis represented by the habitual and, therefore, predictable co -occur rence of

specific words. The main difference between the two is that idiom signifies a
narrower range of word combinations than idiomaticity . This term is used to
denote indivisible lexical units whose components cannot undergo any
variations, or can vary only within definable limits. In addition, the words of an
idiom are not usually recombinable.
Although all idioms show idiomaticity, not all word combinations
showing idiomaticity are idioms. This is, for example, the case with habitual
colloca tions such as rosy cheeks, sallow complexion, black coffee, or catch a
bus, which, in spite of showing idiomaticity, are relatively unrestricted in their
adjectival and nom inal variants: rosy/plump cheeks, rosy dawn, a sallow skin,
strong coffee, catch a tram . Such variations yield idiomatic expressions
exemplifying idiomaticity, but they are not idioms.
As suggested by Fernando (1996: 30), idiomaticity is identifiable not only
in idioms and in conventional ad hoc collocations , but also in conventional
lexico -grammatical sequencing specific to longer structures: those smooth,
plump, rosy cheeks will one day be shrunken, shrivelled, and withered . Such an
ad hoc sequence of ad jectival modifiers preceding and following the noun
cheeks is an illustration of idiomaticity in both selection and sequencing, but
none of the combinations in the sequence has the necessary features to fit in the
class of idioms .
Depending on the length of the lexical sequence, idiomacity is
generally believed to be detectable at the level of the following syntactic
configurations:
 at the sentence level – proverbs and sayings (e.g. Easy come, easy go,
Home is where the heart is) or routine formulae (e.g. long time no see )
 at the phrase level – constructions or idiomatic patterns and nominal/
prepositional/ adjectival phrases (e.g. a breath of fresh air )
 at the word level – compounds (cranberries ) or names (e.g. Jack
Daniels whiskey ).
However, since the usual basis of idiomatic expressions is the
conventionalized co -occurrence of words in a certain sequence, any deviation
from formal conventionality cannot but result in an unacceptable sequencing
(e.g. * butter and bread issue , *rosy, plump, smooth cheeks ). Nevertheless, there
are idioms such as nothing loath , footloose and fancy free , beside oneself, curry
favour , which, although not grammatically regular, are semantically acceptable.
As regards the degree of semantic transparency and compositionality,
distinction may be made between collocations (strong tea ), light verbs (ask a
question , make a proposal ), irreversible binominals (mother and child, sooner
or later ), comparisons (strong as a bear, straight as an arrow ), and
specializations (fried eggs, fish and chips ) (Evaraert et al. 1995: 4).
To put it in a nutshell, “while habitual co -occurrence produces idiomatic
expres sions, both canonical and non -canonical, only those expressions which

become conventionally fixed in a specific order and lexical form, or have only a
restricted set of variants, acquire the status of idioms and are recorded in idiom
dictionaries. Combinations, showing a relatively high degree of variability,
especially in the matter of lexical replace ment are not regarded as idioms,
though they exemplify idiomaticity by virtue of habitual co -occurrence ”
(Fernando 1996: 31).
Given the great variety of conventionalized mu lti word expressions, or
idioms, characterized by invariance or by restricted variation , and of habitual
collocations, r estricted or unrestricted in their variability, the idea is shared that
they may be better understood if grouped on a scale of idiomaticity .
Several specialists (Cowie and Mackin 1975, Cowie et al. 1983,
Alexander 1984, Carter 1987, 1998, Nattinger and De Carrico 1992) have used
scales of idiomacity to demonstrate the shading off of sub -classes of idioms into
one another, as well as the overlap between idioms and collocations .
Considering the lexicographical unity of lexical patterns , that is, whether
such string s are problematic and anomalous on grounds of lexico -grammar,
pragmatics, or semantics , Moon (1998: 20-21) suggests a scale of idiomacity
including anomalous collocations, formulae, and metaphors.

problems of
lexicogrammar
anomalous
collocations ill-formed collocations
cranberry collocations
defective collocations
phraseological collocations

problems of
pragmatics
formulae simple formulae
sayings
proverbs (literal/metaphorical)
similes
problems of
semantics

metaphors transparent metaphors
semi -transparent metaphors
opaque metaphors

Table 4. Categories of FEIs (source Moon 1998: 19)

The members of the first category, i.e. anomalous collocations , are
lexical patterns regarded as problematic in terms of lexicogrammar . This is
mainly due to the fact that they cannot be decoded purely compositionally , nor
encoded freely. The class of anomalous collocations includes four
subcategories of lexical patterns . The subcategory of ill-formed co llocations
are considered to break the conventional g rammar rules (e.g. at all, by and
large, of course, stay put, and thank you ), whereas cranberry collocations
includ e items that are unique to the string and not found in other collocations
(e.g. in retro spect, kith and kin, on behalf of someone/ something, short shrift,

and to and fro ). In addition, defective collocations are lexical patterns which
cannot be decoded purely compositionally , either because a member has a
meaning not found in other collocati ons or contexts, or because one or more of
the members lacks any semantic content (e.g. at least, a foregone conclusio n, in
effect, beg the question ). Finally, phraseological collocations are less
productive patterns, which either have a limited operationa l paradigm, or can be
easily associated with other analogous strings (e.g. in action, into action; on
show and on display; and to a – degree and to a – extent. )
Different from anomalous collocations, formulae are problematic in
terms of their discoursal fu nctions. Most formulae conform to
lexicogrammatical conventions and only a few such patterns are effectivel y
truncated utterances. The patterns included in this category are generally
semantically compositional , although some similes and proverbs are obscu re or
metaphorical.
As regards simple formulae, these are routine compositional strings
which , besides being syntagmatically fixed, have some special discoursal
function or are iterative or emphatic (e.g. alive and well, I'm sorry to say, not
exactly, pic k and choose, you know ). On the other hand, sayings include
formulae such as quotations , catchphrases, and truisms (e.g. an eye for an eye,
curiouser and curiouser, don't let the bastards grind you down, that's the way
the cookie crumbles, and home, James, and don't spare the horses ), whereas
proverbs are represented by traditional maxims with deontic functions.
The fourth subtype of formulae, i.e similes , includes institutionalized
comparisons typic ally, but not always , transparent . Such patterns are sign alled
by as or like (e.g. as good as gold, as old as the hills, like lambs to the
slaughter, live like a king. )
The lexical patterns problematic in terms of semantics are included by
Moon (1998: 22) in the larger class of metaphors , which are further divid ed
into three s ubclass es: i.e transparent , semi -transparent and opaque
metaphors
Transparent metaphors are institutionalized lexical patterns which may
be decode d successfully by means of the speakers’ real-world knowledge (e.g.
alarm bells ring, behind so meone's bac k, breathe life, into something, on
(some)one's doorstep, pack one’ s bags ), different from the semi-transparent
metaphors (e.g. grasp the nettle, on an even keel, the pecking order, throw in
the towel, under one’ s belt ) which require some specia list knowledge in order to
be decoded successfully.
As far as opaque metaphors or pure idioms , are concerned, these are
patterns whose compositio nal decoding and interpretation of the image are
practically impossible if the speakers have no knowledge of t he historica l
origins of the expression (e.g. bite the bullet, kick the bucket, over the moon, fed
herring, and shoot the breeze ).

Considering that this typolo gy is not entirely satisfactory, Moon (1998:
23) suggests an alternative gro uping of lexical patt erns:

defective collocations
phraseological collocations paradigmatically restricted
ill-formed collocations
cranberry collocations syntagmatically restricted
simple formulae
sayings
non-metaphorical proverbs fixed, literal, discoursally meaningful
metaphorical proverbs
similes
transparent metaphors non-literal, transparent
semi -transparent metaphors
opaque metaphors non-literal, non -transparent

Table 5 . Alternative grouping of FEIs (source Moon 1998: 20)

Moreover, she states that dual classific ation s should “ be used to explore
the boundaries between types and to locate the ar eas of greatest indeterminacy.
In the majority of cases of multi -assignment, there is one clearly preferable
assignment. On balance, a flexible system is preferable to a rig id one where
only single classes are acceptable: it allows a greater range of information to be
recorded and reflects the indis putable fuzziness of boundaries” (Moon 1998:
24).
Contrasting idioms with habitual collocations in terms of their formal
and sem antic variations, Fernando, suggests a scale of idiomacity which has
many things in common with that of Cowie et al. (1983).
She explains that the items at the top of the idioms column (1a) are both
invariant and non -literal, different from those in column 1b which show
restricted variance and non -literalness. The items integrated in classes 2a and 2b
also have the features invariant/variant, but they are semi -literal. This set of
items overlaps with a sub -class of collocations characterized by restricted
variance and semi -literalness. Similarly, the items in classes 3a and b, both
variant and invariant, are literal idioms, and the items in 3b overlap with the
literal collocations marked II. Last, but not least, the idioms in 4 are also literal
with preposit ions, which, though optional, usually co -occur with their verbs.

Table 6. Multiword expressions (source Fernando 1996: 32)

Clear as things might seem, the various classes of idioms suggested
above are not differentiated as neatly as they might seem. Habitual collocations
are characterized by the fact that all the it ems show vari ance, either restricted
(e.g. explode a myth , etc. ) or relatively unrestricted as in IV (e.g.
beautiful/lovely woman) . In addition, some of these collocations have one item
with a non -literal subsense (e.g. to explode ‘debunk’ a myth or to cat ch ‘be in
time for’ a train/plane ), whereas others are literal (e.g. addled eggs/brains ). The
tendency to omit the optional (bracketed) items in V explains their being more
strongly optional than the corresponding optional items in column 4 . IDIOMS HABITUAL COLLOCATIONS
1. Pure idioms
a. invariant, non -literal
e.g. red herring, spick and span
b. restricted variance, non -literal
e.g. get/have cold feet

2. Semi -literal idioms
a. invariant
e.g. fat chance you’ve got
b. restricted vari ance 
e.g. blue film/story/joke/comedian

3. Literal idioms
a. invariant
e.g. on foot, in sum, on the contrary

b. restricted variance 
e.g. for example/instance

4. Literal idioms
restricted variance, optional element s
e.g. (even) w orse, worse (still) 

I. restricted variance, semi -literal
e.g. explode a myth/ theory/ notion/
idea

II. restricted variance, litera l
e.g. for certain/sure
III.unrestricted variance, semi –
literal
e.g. catch a bus/train/plane/ferry
IV. unrestricted variance, literal
e.g. glowing/rosy cheeks

V.restricted variance, literal,
optional elements
e.g. clap (one’s hands), nod one’s
head

Considering th e aspects mentioned above, the conclusion could be drawn
that “idiomacity is a phenomenon too complex to be defined in terms of a
single property. Idiomaticity is best defined by multiple criteria, each criterion
representing a single property” (Fernando a nd Flavell 1981: 19).

1.4.1 . ROMANIAN AND ITALIAN APPROACHES TO IDIOMACITY

The wide circulation of certain terms related to idiomacity in English, has
favoured the inclusion of some of these terms in the specific studies devoted to
lexical patterns in other languages.
As regards Romanian phraseological studies, certain English terms
related to idiomacity have been borrowed by the Romanian specialists
concerned with lexical patterns and have been used, sometimes with their
validated English meaning, s ome other times with confusing and rather
contradictory meanings. Two such idiomatic terms are phraseological units
and idioms , whose Romanian equivalents do not always denote the same
reality.
According to Rosemarie Gläser (1998) , a phraseological unit “is a
lexicalized , reproducible bilexemic or polyl exemic word group in common use ,
which has relative s yntactic and semantic stability, may be idiomatized , may
carry connotations and may have an emphatic or intensifying function in a text”,
whereas an idiom is “a lexicalized, reproducible word group in common use,
which has syntactic and semantic stability and may carry connotations, but
whose meaning cannot be derived from the meanings of its constituents”
(Gläser 1998: 125). To put it differently, idioms presuppose “a specific choice
and combination of semantic components carried by its constit uents” (id: 125 –
126). Moreover, distinction is made between word -like phraseological units
and sentence -like phraseological units. The former are “ nominations and
designate a phenomenon, an object, an action, a process or state, a property or a
relationship in the outside world”, whereas the latter are “ propositions and
designate a whole state of affairs in the outside world” (id: 126 -127) as
illustrated by proverbs, c ommonplaces, routine formulae, slogans,
commandments and maxims, as well as by quotations and winged words.
Both types of phraseological units include idioms and non-idioms .
A similar inclusive classification is that provided by Theodor Hristea (in
Bidu V rănceanu 2005: 224), who distinguishes four types of phraseological
units , without explicitly stating whether they have idiomatic or non-idiomatic
interpretation:
 phraseological syntagms (‘locuțiuni frazeologice’) – more or less
compact groups of word s whi ch have a unitary meaning and function as a
single part of speech, namely as a noun (e.g. bătaie de joc = “ironic”), as

an adjective (e.g. întors pe dos = “supărat”) or as a verb (e.g. a se da pe
brazdă = “a ceda”);
 phraseological expressions (‘expresii f razeologice’) – stable groups
of words which may have the form and function of a sentence (e.g. a taia
frunză la câini, a -și lua inima – n dinți, a face pe cineva cu ou și cu oțet );
 international formulae and clichés (clișee) – groups of words which
have a n approximately similar form and are used with the same meaning
in different languages (e.g. art for art’s sake  artă pentru artă = “artă
pură” , apple of discord  mărul discordiei = “motiv pentru ceartă” , the
Gordian knot  nodul Gordian = “problemă insol ubilă” , sword of
Damocles  sabia lui Damocles = “amenințare iminentă” , Achilles heel
 călcâiul lui Ahile = “punctul slab al cuiva”);
 expressive periphrases (‘perifraz e expre sive’) – less restricted
combinations, used in affective statements specific to a certain cultural
model (e.g. luceafărul poeziei românești, bardul de la Mircești ).
General and useful as it may be in certain classifications, the term
phraseological unit is not always used to denote the same types of lexical
patterns in English and Roma nian.
The same holds valid for the term phrase which is used to refer to certain
types of lexical patterns in English ( idioms and collocations) and a different
pattern in Romanian , i.e. “a word -combination standing for a part of speech
whose grammatical functions it discharges” ( Levițchi 1970: 15).
Trying to justify the necessity that idioms should be listed in the lexicon,
Francis Katamba (1993) refers to the existence of two classes of phrases . One
seems to be the class of collocations (they are not specifically referred to as
such), as it includes those phrases whose meaning can be deduced if the
speakers “know the meaning of the words they contain and the ways in which
they are syntactically related to each other ” (e.g. to pass the salt). The latter
class is that of idioms (called ‘listed syntactic objects’ by Di Sciullo and
Williams 1987, in Katamba 1993: 295), which is made up of phrases such as to
pass the buck, to eat a humble pie, to be in the red , whose “meaning cannot be
worked out on the basis of the meanings of the word s which they contain ”
(Katamba 1993: 296). To put it differently, the main aspect differentiating the
two types of phrases is their being or not being semantically compositional .
Different from Katamba’s approach, Romanian lexical studies use the
term lexical phrase to denote other types of lexical patterns, even whole
sentences, which function as a phraseological unit enjoying structural stability
(proverbs , conversational formulas , etc.). Referring to formal characteristics
of phrases , Levițchi (1970: 19) suggests that they may be classified as stable
and unstable. If the phrases in the former category allow no change in the
component element, unstable phrases are regarded as structures liable to
change of form or to (partial) replacemen t. This latter category is exemplified

by the lexical pattern to play a trick to somebody in which the verb may, on the
one hand, take different tenses, and word order can change , on the other (you
play somebody a trick ). Moreover, in such a pattern the ve rb may be replaced by
its connotational synonyms to put and to serve  to put a trick on somebody, to
serve a trick on somebody, on the other.
Furthermore, referring to the semantic classification of phrases,
Levițchi (1970: 19) distingushes between two classes of such lexical patterns: 1.
phrases which are used in a direct meaning, and 2. phrases which, partly or
wholl y, express figurative meanings. As regards the meaning of phrases ,
Levițchi mentions that it results from the sum of the meanings of the component
elements as illustrated by to set at liberty  a pune in libertate, in other words
 cu alte cuvinte, to be of good cheer  a fi bine dispus, a fi vesel, a fi plin de
viață , safe and sound  teafăr, să nătos, to come into view  a apărea, a se
zări, silence gives consent  tăcerea înseamnă consimțire.
Contrasting phrases used in their direct meaning with phrases which are,
partly or wholly, based on figures of speech (chiefly on metaphors ), Levițchi
explains that the role of the latter is very active in the semantic development of
a language. The phrases he selects to prove his view are to give somebody a
lesson  a-i da cuiva o lecție, a învăța pe cineva minte, to carry coals to
Newcastle  a vinde cast raveți grăd inarului, a căra apa la puț, the/ a cat in
gloves catches no mice  cine se boierește treabă nu face, to put a spoke in
somebody's wheel  a-i pune cuiva bețe în roate, make hay while the sun shines
and strike the iron while 'tis hot  bate fi erul cât e cald, to give somebody the
needle (slang)  a pune pe cineva pe jar, a face pe cineva să fiarbă; a scoate pe
cineva din sărite . All the words in these patterns are used in their direct
meaning, but the whole meaning of the phrase is metaphorical .
Analyzing a phrase like to break the silence  a sparge/ a întrerupe
tăcerea, Levițchi points out that the verb to break assumes a figurative –
metaphorical meaning as a result of its association with silence , but owing to the
frequent use of the phrase, the metaphor is of the fading or degraded type.
Last but not least, in the category of phrases, Levițchi includes lexical
patterns which are built on similes (as) red as a rose  roșu ca trandafirul/un
trandafir, roșu ca bujorul/ un bujor, to run like a de er  a fugi /a alerga ca o
caprioară, or on hyperboles a thousand thanks  mii de mulțumiri, to be a
shadow of one’s former self  a nu mai fi decât o umbră.
The important point made by Levițchi with respect to such phrases is
that
“sometimes it is diffi cult enough to establish whether this or that word in a
phrase has a direct or a figurative meaning. Lan guage is permanently ‘on the run’
and what was new and striking and graphical yesterday may be trite and ‘matter –
of-fact’ today. The problem is all the more complicated as even big -sized dictio –

naries do not clearly discriminate between direct and figurativ e meanings of
words, to say nothing of words in phrases” (Levițchi 1970: 20).
In his approach to the English lexicology, Levițchi also enlarges on
Vinogradov’s semantic classification of phrases , adapted by Kunin (1955) to
the realities of his time. Thus, distinction is made between phraseological
fusions, phraseological unities, and traditional combinations. The first
category includes indivisible and indecomposable phrases in whose general
meaning one cannot detect any connection with the words the phrase is made up
of. The second is represented by stable phrases, which, like idioms, have a
meaning of their own, distinct from the meaning of the component elements,
although these are connected logically. The category of traditional combinations
is made up o f words that may combine only with certain other words.
Regarding phraseological unities, Levițchi explains that they are often
considered to include figurative phrases such as as fresh as a daisy  înfloritor,
plin de tinerețe și sănătate, verde ca bradu l, neither here nor there  nici în clin
nici în mânecă, fară nici o legătură, and proverbs diamond cut diamond  cui
pe cui se scoate. Moreover, traditional co mbinations include patterns in which
one or more words are used in their direct, non -figurative senses such as to pay
a visit to someone  a face o vizită cuiva, to give/ lend assistance to someone
 a da ajutor cuiva, to strike/ deal/ inflict a blow to someone  a lovi pe
cineva, a da cuiva o lovitur ă, thus corresponding to the catgory of collocatio ns
attested by later studies in the field.
Besides classifying phrases from the formal and semantic points of view,
Levițchi also makes an attempt in providing a stylistic classification of such
lexical patterns. He starts from the assumption that similar ly to words, phrases
may be classified in accordance with the domains to which the phraseological
material can be referred. Thus, he identifies phrases connected with trades
(e.g. to bring grist to the mill  a aduce câștig , a aduce apă la moară , to have
too many irons in the fire  a se apuca de prea multe, a ține prea mulți iepuri
într-o mâ nă, between hammer and anvil  între ciocan și nicovală , in full blast
 în toi, în plin avânt, a chip of the old block  leit firea lui taică -său, to screw
up one's co urage  a-și lua inima în dinți, one nail drives another  cui pe cui
se scoa te), phrases connected with medicine (e.g. to swallow the pill  a
inghiți hapul, to take the temperature of  a aprecia situația, a simți pulsul , a
dose of one's own medicine  păcăliciul păcălit , nu săpa groapa altuia, ca să nu
cazi singur în ea), and phrases connected with rivers , etc. (e.g. to shiver on the
brink  a dârdii, a -i fi frică, a nu mai putea de frică , to go at the deep end  a
se arunca in apă , a se apuca cu cura j de o treabă , to make a splash  a face
senzație , on thin ice  pe teren șubred, în primejdie , la strâmtoare).
Finally, Levițchi refers to phrases which may be monosemantic (in full
blast ) or polysemantic ( to be in abeyance  1. a aștepta, a fi în aștept are; 2. a
nu se manifesta, a nu apărea; 3. a nu avea stăpân sau pretendent; 4. a fi abrogat

provizoriu).
Observing Levițchi’s approach to phrases and the more recent
approaches to the study of the various types of lexical patterns in English,
Constan ța Av ădanei (2000) embarks upon the exploration of idiomatic
expressions in English and Romanian . The innovative character of her
comparative -contrastive approach resides in the use of updated, English –
oriented terminology, in the structuring and adapting of fo rmal and semantic
classifications (see 2.1.3.1 – 2.1.3.2 ), as well as in the significant and illustrative
examples selected from various sources.
Although made in the larger frame of English lexicology and
phraseology, Leon Levițchi’s approach to phrases, and Constan ța Avădanei ’s
approach to idiomatic expressions in English and Romanian represent an
important contribution to the study of the various types of lexical patterns
traceable in the two languages, but, more importantly, in Romanian, where
phraseol ogical studies have only recently started arousing the specialists’
interests.
The freshness of phraseology within the Romanian borders and its
unclear status is attested by different specialists. For example, Theodor Hristea,
the linguist who set the theo retical foundations of Romanian phraseology, states
that phraseology does not have a clear positioning within the larger frame of
Romanian language studies (Hristea 1984: 134). On the other hand, more recent
studies suggest that phraseology is an independe nt branch of linguistics whose
domain of activity is related either both to vocabulary and syntax (Colțun
2000: 13), or strictly to lexicology (Zugun 2000). Moreover, reference is made
to the importance of phraseology for stylistics and semasiology, and o f the
studies devoted to phraseologic units for disciplines such as etymology and
lexicography.
Considering the areas of interest of the Romanian phraseology , the idea
is shared that this borderline branch of linguistics focusses on the study of fixed
lexical combinations which are most often called unități frazeologice or
frazeologisme, and are contrasted with free word combinations which are
studied by syntax.
Starting from Theodor Hristea’s (1984: 139) statement that all
phraseologic units in a language are fix ed combinations of two or more words
which have a unitary meaning, Ioana Scherf (2006) defines the phraseologic
unit (or frazeologism ) as the fix ed word combination, whether idiomatic or not,
which is made up of at least two words, but functions as a single semantic unit
in language. She adds that phraseologic units are recorded in dictionaries as
ready -made patterns and, as such, are accepted to share ‘phraseologic features’
such as polylexicalism , multiplication , stability , idiomacity , complexity and
expressivity :
“unitatea frazeologică este o îmbinare stabilă de cuvinte, cu sau fără

idiomacitate. Ea are o întindere de cel puțin două cuvinte și este lexicalizată,
adică funcționează în limbă ca o unitate de sine -stătătoare, relevând unitate
semantică și este consemnată în dicționare, ca dovadă că ea nu se formează după
regulile îmbinărilor libere de cuvinte, ci se „cheamă” din memorie „de -a gata”
pentru a se integra apoi, datorită valenței proprii, în lanțul vorbirii” (Scherf
2006: 82, in Munteanu 2007: 102).
Idioms and idiomatic expressions are closely analyzed and classified by
Stelian Dumistrăcel (1980) in his study devoted to the Romanian words,
expressions and metaphors. Different from other studies on Romanian
phraseologic units, Dumitrăcel’s approach to idioms and idiomatic expressions
is oriented towards the stylistic value and the ethymology of such lexical
patterns.
Last, but not least, mention should be made of Coșeriu’s contribution to
the study of collocations and of other forms of ‘repeated discourse’ ( discu rs
repetat ). In his opinion, everything that is repe ated in a more or less identic
form in the speech of a community, be it a more or less fix ed combination, a
longer or a shorther fragment , or an already made discourse should be included
in the category o f ‘repeated discourse’:
[Discurs repetat ] „cuprinde tot ceea ce în vorbirea unei comunități se repeta
într-o formă mai mult sau mai puțin identică sub forma de discurs deja făcut
sau combinare mai mult sau mai puțin fixă, ca fragment, lung sau scurt a ceea
ce s-a spus deja” (Coșeriu 2000: 258-259).
Referring to collocations , Coșeriu (2000) uses the notion of ‘lexical
solidarities ’ (solidarități lexicale ) which are defined as syntagmatic3 pheno mena
conditioned from a paradigmatic point of view and classified according to
various criteria ( e.g. the aspects conditioning the semantic conten t of the
lexem es making up a paradigme, the elements making up a paradigme) . In his
approach to ‘lexical solidarities’ Coșeriu also refers to the existence of a
‘determining ’ (lexem determinant ) and a ‘determined ’ (lexem determinat )
lexeme . Th e former is a lexical unit whose semantic content represents a
distinctive feature in the case of another lexeme, whereas the latter is the
element which receives the respective distinctive feature (s).
Italian studies devoted to fix ed lexical patterns are not so numer ous, and,
just like in the case of most Romanian studies , they often have as a st arting
point the Anglo -American tradition in the field . Hispanic approaches are also a

3 Iorgu Iordan (1975: 209) states that speech is made up of syntagms („vorbirea constă
în sintagme”) and uses the term ’word group or combination’ ( grup sau combinație de
cuvinte ) to refer to lexical patterns which are characterised by fix form and unitary
meaning. Moreover, Corneliu Dimitriu (1977) considers that fix word combinations
(îmbinările fixe de cuvinte ) should be termed syntagms or periphrase ( sintagmă
/perifrază ).

refrence point in many comparative – contrastive studies carried out by Italian
researc hers.
Nevertheless , mention can be made of speciali sts such as Luca Serriani
(1989) , Maria Teresa Musacchio (2002) , Alessandro Lenci (2005) , Vicenzo Lo
Cascio (1995) Paola Tiberii (2012) who approach collocations and other types
of fix ed lexical patterns in Italian apart from the already established studies in
the field.
Starting from Luca Serriani’s (1989) notion of tecnicismi collaterali4
Maria Teresa Musacchio makes a distinction between termini composti/ formule
di routine and collocazioni explaining that the former are syntagms which
express a concept , whereas the latter are syntagms which add connotations to
the concept expressed :
[Termini composti/ formule di routine ] ”sono sintagmi che esprimono un
concetto , [collocazioni ] ”sono sintagmi che al co ncetto aggiungono altre
connotazioni” (Musacchio 2002: 137).
Musacchio (2002) introduces the term tecnicismi collaterali in her
typology of recurrent sequences approaching notions such as phraseology,
idiomatic expressions, collocations and fix ed expressi ons. The basis of her
analysis is represented by the fact that the so -called tecnicismi collaterali are
lexical units made up of one or more words which combine in a more or less
restrictive way.
Furthermore , Alessandro Lenci (2005 :196) defines collocati on as a
sequence of two or more words strongly connected by means of their reciprocal
association ( “sequenza di due o più parole caratterizzate da un forte legame di
associazione reciproca” ), and Vicenzo Lo Cascio (1995: 71) uses the same term
to refer to the fixed grouping of two specific lemes or words which are
systematically used together creating a unitary and precise concept
(“l’accoppiamento di due specifici lemmi o parole che si incontrano in modo
fisso e sistematico creando un co ncetto unitario e preciso” ).
Last, but not least, Paola Tiberii specifically defines collocations in her
Dizionario delle Collocazioni as “espressioni formate da due o più parole che
per uso e consuetudine lessicale formano una unità fraseologica n on fissa ma
riconoscibile ” (Tiberii 2012: 5).
Instead of conclusion s, mention should be made that the terminology
used in Romanian to denote various (fixed) lexical patterns is as diverse as
controversial. The first inventory of the possible terms used to denote
phraseologic units in Romanian is made by Florica Dumitrescu who, enlarging
on terms such as expresie and locuțiune , mentions that they function in parallel

4„particolari espressioni stereotipiche, non necessarie, a rigore, alle esigenze del la
denotatività scientifica, ma preferite per la loro connotazione tecnica” (Serianni 1989:
103).

with other terms such as izolare, idiom or idiotism, locuțiune proverbială .
As regards the corresponding studies in Italian, the terminology used to
describe various types of fixed lexical patterns is not problematic , there being
more numerous specific definitions of collocations as compared to those
provided in Romanian. However, mention should be made that the notion
phraseologic unit , specifically defined and described in certain English and
Romanian st udies is used in Italian only as a general means of defining
collocations5. If discurs repetat , solidarități lexicale , determinat and
determinant are Romanian specific terms used in some of the approaches to
(fixed ) lexical patterns, tecnicismi collaterali seems to be the trade mark of
Italian phraseology -related studies ,
Far from having reached t he development and the wide circulation of the
studies devoted to the various types of lexical patterns in English, the
approaches made by the Romanian and Italian specialists in the field and
selectively presented in this section are a valuable resource w hich should be
further explored in studies on Romanian and Italian phraseology.

1.4.2 COLLOCATIONS AND FREE WORD COMBINATIONS

Due to their sharing certain formal and semantic characteristics, collocations
are often compared with free combinations. Fewe r are the instances when
comparison is drawn between free combinations and idioms.
Free combinations are usually made up of a small number of words,
each of which is taken to have its usual independent meaning. This implies that
the meaning of the whole combination is derived by general semantic
combinatorial rules from the meanings of the words involved. For example, in
combinations such as to fill the bucket or to fill the glass , both the verb to fill
and the nouns bucket and glass preserve their meani ngs which are summed up
to give the meaning of each of the two combination s. Moreover, the
contribution of word meanings to phrase/ sentence meanings is invariant in all
free combinations
Although a collocation also involves two items, the ‘ base’ or ‘node’ and
the ‘ collocate’ , their semantic independence is not preserved. The ‘ base
preserves the meaning it would have in a free combination, whereas the

5 “Le collocazioni sono delle unità fraseologiche tendenziamente non idiomatiche e
non equivalenti ad un discorso, come i proverbi e le formule di rito. Gene ralmente sono
formate da due parole che appaiono insieme nel discorso: una di esse è il nucleo ( o la
base)e cioé il membro determinante ed autonomo che regola l’apparizione dell’altro, il
collocativo, la cui apparizione è condizionata e determina ta semant icamente dalla base”
(Francesconi 2008: 157 -158)

‘collocate ’, i.e. the word which can occur in free combinations in other contexts,
has a specific , non-free meaning when combined with the ‘base’. In other
words, “the headword speakers start from is often the ‘figurative ’ element
(‘collocate ’ or ‘collocator ’), while the element one moves toward is the ‘ literal’
element (the so called ‘base’)” (Cowie 1998: 16 ). For example, the adjective
sharp in a sharp turn means ‘difficult/quick’, different from what it means in
sharp knife/ razor . This semantic characteristic explains the semi –
compositionality of collocations as compared to the full compositionality of
free word combinations .
As regards the class of idioms , Nicolas (1995: 235) states that in an
idiom, none of the component words has its free -combination meaning. He also
explains that there are three broad options regarding the semantics of idioms :
1. all idiom words have individual meanings, which are interpreted
compositionally, but these are special meanings invoked only in the context of
other words in the same idiom (Gazdar et. al. 1985); 2. the sense of an idiom is
localized in some subset of its cons tituent words, whereas the other words are
semantica lly empty (Bresnan 1982), or 3. none of the individual words in an
idiom, but only the idiom as a whole, has meaning. (Nicolas 1995: 235)
Considering the aspects above, the conclusion may be drawn that t he
formal and semantic behaviour of the words making up collocations is better
understood if compared and contrasted with that of the words making up free
word combinations , and idioms , respectively. This allows for a clear
positioning of collocations , whe ther lexical , or grammatical, inbetween free
word combinations and idioms .

Fig. 4. Collocations in relation to free word combinations and idioms
COLLOCATIONAL
PATTERNS

FREE
COMBIN ATIONS
COLLOCATIONS
IDIOMS

GRAMMATICAL
COLLOCATIONS
LEXICAL
COLLOCATIONS

The same idea is expressed by Fontenelle (1998) when stating that
idioms are to be found at one extreme of a continuum ranging from totally free
combinations of words to completely froze n, fixed, multiword units.
Collocations , on the other hand, are placed in a fuzzy area being positioned
halfway between free combinations and idioms .

1.4.3 . DELEXICAL(IZED) COLLOCATIONAL PATTERNS

In the larger class of collocations , there are patterns whose members lose their
lexical meaning when co -occurring with their collocational partners. Specialists
(Sinclair 1992, Stubbs 2002) refer to dif ferent such delexicalized structures
explaining that “ [t]he meaning of words chosen together is different from their
independent meanings. They are at least partially delexicalized. This is the
necessary correlate of co -selection … [T]here is a strong tend ency to
delexicalize in the normal phraseology of modern English” (Sinclair 1992,
Sinclair 2004: 20 ) (see 1.4.6.).
One type of delexicalized pattern frequently used is that in which a verb
in one variation corres ponds to a cognate noun or adjective and an (often)
delexicalized or support verb in the other variation. Moon includes in this
category of collocations examples such as: bloody someone’s nose = give
someone a bloody nose; circle the wagons = pull/draw the wagons in a circle;
feel something in on e's bones = have a feeling in one's bones; not sleep a wink =
not get a wink of sleep, pass one’s sell -by date = be past one's sell -by date; rap
someone on the knuckles = give someone a rap on the knuckles; ring hollow =
have a hollow ring (Moon 1998: 145) .
Moreover, there are lexical patterns in which individual words are not
always the unit of meaning (e.g. to take a decision/ look/ shower/ sip, to have a
chat/ drink/ look/ shower/ swim/ try, to give a scream/ shout/ speech, to make a
mistake/ note /sugge stion ). In such cases, the verb is said to be delexicalized.
(e.g. to take place/ part/ care, to make sure/ sense/ clear ).
Delexical patterns are important due to their formal restrictions in English
and to the difficulties they create in translation.

1.4.4 . COLLOCATIONS AND THE SO -CALLED HARMONIC PHRASES

The relationship between collocations and harmonic phrases may be
considered to have Halliday’s model of language (1985) as its basis, i.e. the
relationship between the “ideational metafunction and th e interpersonal
metafunction ”. In the case of the former the realization means are represented

by the lexical items , whereas the latter is achieved by means of modality (verb
forms, modals, non -verb forms).
Harmonic phrases are referred to when discussing about harmonic
modality . According to Coates (1983: 41, in Croitoru 2009: 28), a harmonic
phrase “is an element in the context of a modal verb that reinforces or
disambiguates.” Moreover, the term harmonic is used “slightly more loosely, to
cover all combi nations of a modal and another word or phrase which expresses
the same degree of modality” (Coates 1983: 45, in Croitoru 2009: 28).
The analysis of a harmonic phrase may be related to ambiguity and
merger , terms which are used to account for the situation s in which the
meaning is made explicit by a larger context: the core , strong meanings , are
subjective, whereas the periphery , weak ones , are objective.
Merger differs from ambiguity in that “it is not necessary to decide
which meaning is intended before the example can be understood; with merger
the two meanings involved are not in certain contexts mutually exclusive. This
can be described as contextual neutralization” (Coates 1983: 17 in Croitoru
2009: 29). In other words, the context does not exclude on e of the two possible
meanings. Therefore, in the case of merger , there is a both/ and relationship
whereas ambiguity is characterized by an either/or relationship. As Coates puts
it, “either one meaning or the other must be chosen. Ambiguous cases provide
support for the recognition of the root epistemic distinction” (1983: 16, in
Croitoru 2009: 29). The double modal interpretation is thus related to the
distinction drawn by Coates between core and periphery meanings (the latter
depending on the context), as well as to Leech’s (1987: 71, in Croitoru 2006:
29) distinction between logical and practical (pragmatic) element of the modal
auxiliaries’ meaning.
Although collocations and harmonic phrases function in a slightly
different manner, a comparative -contra stive approach to them is interesting and
useful because it can highlight similarities and dissimilarities with respect to the
semantic interaction between their constitutive elements.

1.4.5 . PHRASAL VERBS

Phrasal verbs are the best candidates to follow collocations on the scale of
idiomacity. They are difficult to enumerate or identify because there are so
many grades and types of co -selection that the relevant criteria are not only
difficult to state , but even more difficult to apply. Contrary to what is often
claimed, each word of a phrasal verb contributes indeed something semantically
recognizable to the meaning of the whole. Sometimes it is mainly the verb to
have a semantic contribution, some other times it is mainly the adverbial

particle. Neverth eless, the meaning of the words used together is different from
their independent meanings, these words being at least partly delexicalized.
This is, as Sinclair (2004: 20) states, “the necessary correlate of co –
selection. If you know that selections are not independent, and that one selection
depends on another, then there must be a result and effect on the meaning which
in each individual choice is a delexicalization of one kind or another.” Hence,
the strong tendency to delexicalization noticeable in th e normal phraseology.
For example, the co -selection of noun and adjective in patterns such as e.g.
physical assault/ confrontation/ attack/ damage/ attribute/ proximity/ bodies
does not make these patterns either fixed phrases, or necessarily restricted
collocations. (Sinclair 2004: 21)
The phrasal verb combinations verb plus adverbial particle conjure up
unpredictable meanings, but their structure does not fit the model either
semantically, or grammatically, because a single meaning selection straddles a
major structural boundary.
Although in the case of phrasal verbs and idioms the meaning is not the
same as the sum of the meaning s of the constituent words, there are cases when
this is does not hold valid, because individual words can be in all sorts of
relationships: Sometimes none of the words may appear to contribute directly to
the meaning of the expression (e.g. to bear on –‘ to be relevant’), in other cases
some words may, while others may not have a direct contribution to the
meaning (e.g. to beat somebody up ) and, finally, sometimes each word seems to
mean what it normally means outside the combination (e.g. the rain beats
down ).
This last type of phrasal verb, called collocation , is a frequent co –
occurrence of words which does not have any profo und effect on the individual
meanings of words, but in the case of which “there is usually at least a slight
effect on the meaning, if only to select or confirm the meaning appropriate to
the collocation, which may not be the most common meaning” (Sinclair 2004:
28).

1.4.6 . SIMILES

Similes are essentially patterns consi sting of fossilized lexis, whose function is
to emphasize a feature or characteristic. The tradition al structure is (as) + ADJ.
+ as + NG : e.g. (as) clear as crystal, dead as a doornail, as good as gold, as
nice as pie, as right as rain, (as) white as a sheet.
Although similes intensify adjectives, sometimes certain common
adjectives, or commonly intensified adjectives, occur in patterns with varying
nouns e.g. clear as crystal/ the day, clear as mud (ironic use), easy as ABC/ pie,
happy as Larry/ a clam, happy as a lark/ a pig in muck (etc.), plain as day/

pikestaff/ the nose on one’s face, quick as a flash/ lightning/ a wink, strong as a
bull/ horse/an ox, thick as mince/shit/ as two (sh ort) planks, thin as a lath/ a
rail/ rake/ stick, white as snow/ a ghost/ sheet.
Nevertheless, certain noun variations are associated with different
meanings of the adjective e.g. clear as a bell (of sounds), clear as crystal/ day
(of information), good as gold (of behaviour), good as new (of condition), mad
as a hatter = insane, mad as a hornet = angry, thick as mince/ shit/ two (short)
planks = stupid , thick as thieves (friendly, conspiratorial), white as a ghost/
sheet = pale, ill, frightened , white as a sheet = pure white in colour.
As suggested by Moon (1998: 151), from a semantic and stylistic point of
view, most of these similes are entirely transparent, only a few being
synchronically opaque (e.g. plain as a pikestaff and right as rain ) and very few
being oxymoro nic and always ironic (e.g. clear as mud ).
The adjectival pattern usually associated with similes functions in parallel
with a v erbal structure made up of (V ) + like + N G: built like a tank, get on like
a house on fire, know som ething like the back of one’s hand / like getting blood
out of a stone/ like headless chickens/ like a headless chicken / like water off a
duck's back, stick out like a sore thumb, take to something like a duck to water ,
work like a dog . Such strings are institutionali zed and have marginally higher
average frequencies than those built with as instead of like, but the frame is by
no means so fixed or restricted.

1.4.7 . BINOMINALS

Binomials or binominal expressions are dyads or conjoined pairs, unrestricted
as to wor dclass, but normally occurring in fixed order as ‘irrever sible
binomials’ (Makkai 1972, Moon 1998: 152) e.g. profit and loss, home and
abroad, in and out, here and there, life and death, cause and effect, men and
women, women and children, British and Fre nch, French and Italian .
In some other pairings, there is a tendency for the shorter or monosyllabic
item to precede the longer one e.g. law and order, bed and breakfast, time and
money, fruit and vegetables, names and addresses, banks and building societ ies.
As regards the pairs involving male/female counterparts , the norm is for
the male term to precede the female – related one e.g. Mr. and Mrs., men and
women, brothers and sisters , one of the few exceptions being the binominal
mother and father , also u sed in the plural form: e.g. mothers and fathers .
Moreover, i n certain structures, which are rule -bound and are themselves
coded, there is a noticeable binominal sequencing: e.g. born and bred,
comings and goings, cut and dried, dim and distant, drunk an d disorderly, free
and easy, nook and cranny, part and parcel of something, pure and simple, tar
and feather, to and fro . Binominals such as black and white (negative first), and

risk life and limb (illogical order of the severity of the risk) are counter
examples of rule -bound patterns. As far as back and forth , or backwards and
forwards are concerned, these binominals are indetermi nate, but seem to
observe the ‘me -first’ or ‘towards speaker’ orientation.
Many antonymic binomials, or conjoined antonyms, have a meaning
along the lines of ‘everything’ or no matter what’. Reference can be made to
pairs not always linked by ‘and’, or including conjoined temporals (e.g. from
cradle to grave, from beginning to end, day and night, night and day ), spatials
and di rectionals (e.g. from head to foot, left and right; left, right, and centre ,
search high and low, swear up and down (literally spatial) from top to toe, from
top to bottom, up hill and down dale ), and other contrastives (e.g. by fair means
or foul, come ra in or shine, flotsam and jetsam ).
From a stylistic point of view, binominal structures built up with
antonymic pairs give dynamism to the meaning envisaged e.g. back and forth,
come and go, in and out, on and off, push and pull, stop and start . Moreover, in
a binominal such as blow hot and cold , contradiction combines with repetition,
different from give and take in the case of which antonymy is blended with
reciprocity in compromise.
Furthermore, certain antonymic pairs have metalinguistic reference and
signal a lack of precise detail e.g. dos and don'ts , ins and outs , the long and the
short of it , and this and that. Other conjoined pairs imply strong contrast being
set up as antonymous in the context of the pattern e.g. (be) apples and oranges,
(be) chalk and cheese, (be) oil and water . Contrast is also expressed by pairs
linked with ‘or’ e.g. feast or famine, fight or flight, fish or cut bait, sink or
swim, trick or treat . Certain such binominals may express approximation: give
or take and hit or/ and mis s.
Binominal structures constructed with linked synonyms (see 3.3.3.1. ),
with co -hyponyms (see 3.3.3.3 .), or with the same word repeated twice have an
emphatic function or emphasis as part of their meaning e.g. alive and
kicking/well bits and pieces, don e and dusted, down and dirty, down and out,
far and away, go at it hammer and tongs, high and dry, home and dry/hosed in
leaps and bounds, loud and clear, nooks and crannies . Compare also the
emphatic trinomials every Tom, Dick or Harry , lock, stock and ba rrel, on the up
and up out and out .
Last, but not least, in certain pairs, one of the binomial elements is
obsolete or old -fashioned, but a synonym or co -hyponym in diachronic terms:
e.g.. at someone's beck and call bib and tucker, bill and coo, bow and s crape,
kith and kin, rack and ruin , spic(k) and span, whys and wherefores.
To conclude , although virtually ignored by many studies, I consider that
binominals are worth comparing and contrasting with collocations due to their
similar formal fixity and s emantic transparency.

1.4.8 . CLICHÉS AND PROVERBS

Clichés are fixed lexical patterns which resemble ossified collocations. They
are an illustration of the fact that the mutual expectancy of lexemes has become
fixed in certain contexts leading to loss of meaning. A good example could be
desired residence , in which the term residence is no longer used with its
specialized meaning specific to the jargon used by estate agents, but with its
general meaning, i.e. ‘house’ (Howard 1988: 104).
Proverbs, on the other hand, “represent a common cultural fund of folk
knowledge and wisdom which can be called on in order to warn or reprimand
someone in the assurance that they will accept the basis of common wisdom”
(Howard 1988: 105). From a semantic point of view, proverbs are patterns in
which there is an incongruity of meaning in the sense that the literal meaning
differs significantly from the context to which it refers.
Although proverbs resemble idioms in their non -literalness, they differ
significantly as rega rds the semantic interpretation. In the case of proverbs, the
literal meaning bears a direct, though pictorial relation to their intended
reference, whereas in the case of idioms , the literal meaning has no relevance
whatsoever. A proverb such as A new bro om sweeps clean may be easily related
to the situation in which somebody comes in and changes everything, especially
people’s cherished ways of doing things, but there is no such relation traceable
when comparing to kick the bucket and ‘ to die’. Thus, diff erent from proverbs,
the essential feature of an idiom is its non -literal, metaphorical meaning.
If regarded from a formal point of view, “ proverbs are full sentences, as
compared with idioms which often represent parts of sentences and are more
frozen th an idioms”, that is they allow less lexical and grammatical variability
(Benson 1985: 66, Benson, Benson and Ilson 1986a: 253).

1.4.9 . IDIOMS

Idioms have always aroused the curiosity of linguists, especially of the
specialists in lexical semantics and l exicography. There is a long tradition in the
study of idioms, but an important development in the studies of such lexical
patterns marks the last few decades of idiom research. Even though idioms were
originally a topic of interest for a limited number of individuals, they now
represent an important research theme for a significant group of specialist s. This
is due to the fact that linguists have come to realize that idioms and idiom -like
constructions are a persistent feature of language, making up a lar ge part of it.
Out of the lexical patterns dealt with so far, idioms are the best
representatives of formal and semantic idiomacity, hence their inclusion in

numerous studies devoted to this topic, among which Chitra Ferna ndo’s (1996)
is the most extensiv e.
The term idiom is as widely as controversially used in the approaches
suggested by various specialists in the field (see 1.3. and 1.4.1 ). Nevertheless,
certain studies (Sinclair 1991, 2001, 2004) make explicit distinctions between
idioms and other class es of lexical patterns, which resemble idioms either from
a formal (‘restricted collocations’), or from a semantic point of view
(metaphors).
In an attempt to identify the essential features of idioms , Weinreich
(1969: 71) distinguishes betwee n the ‘idiomacity of expressions’ and the
‘stability of collocations’. Although both idioms and collocations are
illustrative for the co -occurrence phenomenon, the special semantic relationship
established between the constitutive elements of the former is not evident in the
case of the latter.
Cowie et al. (1983) focus on the category of semi -idioms , which are
considered an intermediate class of lexical patterns placed between idioms and
borderline idioms . The interesting fact pointed out by their analysis is that the
word combinations included in latter category perfectly correspond to ‘open
collocations’, whereas the opposing class of semi -idioms is identical to the class
of ‘restricted collocations’.
Furthermore, analyzing the literalness and the degree of variance shown
by lexical patterns, Chitra Fernando (1996: 32) distinguishes two major types of
conventionalized word expressions , namely idioms and habitual collocations.
In her opinion, there are four factors which favour lexical patterns to
acquire the status of conventionalized structures . Such patterns, called
multiword expressions in Fernando’s terms, need to conform to the
grammatical rules of language. Moreover, the invariance or the fixity of the
words making up such patterns, combined with the ir non -literalness in many
cases, are two other factors favouring their status of conventionalized multiword
expressions. Since attributes such as fixity and non-literalness are a matter of
degree, multiword expres sions can range from completely fixed , semantically
non-literal , (e.g. pins and needles ‘the tingling sensation following numbness’),
through grammatically flexible (e.g. spill/ spilled the beans ‘commit an
indiscretion’), to lexically variable , which in their turn range from restricted
and semi -literal (e.g. blue film/joke , etc. ‘obscene’) to unrestricted semi –
literal (e.g. catch a bus/train , ‘be in time for’). In addition, culturally salient
encodings represent the fourth factor favouring the emergence of multiword
expres sions (Pawley 1986). T his means that the respective expressions capture
some phenomenon prominent in the collective consciousness.
From a structural point of view, the multi word expressions above are
representative for various lexical types: compounds (e.g. happy -go-lucky ),
phrases of various types (e.g. red herring, bread and butter, on behalf of, at

best … at worst ), semi -c1auses of various types (e.g. Guess what?; W aste not,
want not; Do one's bit, S pill the beans ), and full clauses of various types (e.g.
The coast is cl ear; While the cat is away, the mice will play ). A categorization
of such patterns is also possible, the distinction being made in this case between
idioms and their sub -classes , on the one hand, and between collocations and
their sub -classes , on the other (Cowie 1981; Cowie et al. 1983). The feature
shared by these lexical types is that they show the regular co -occurrence of
words in a specific form and order, habitual word combinations, to such an
extent that the presence of one sets up an expectancy of t he other (Firth 1957,
Halliday 1966, Mitchell 1975, Cowie 1981).
Besides, trying to restrict her approach to idioms , Fernando (1996: 37)
suggests three sub -classes of such patterns, namely pure idioms, semi -idioms,
and literal idioms. Nevertheless, she is fully aware of the fact that such a strict
division of idioms is often difficult to maintain.
Pure idioms are ‘a type of non -literal multi word expression’ best
exemplified by to spill the beans . Such a lexical pattern has nothing to do with
its literal counterpart meaning ‘letting fall leguminous seeds’, the non -literal
meaning being imposed on the idiom as a whole: ‘commit an indiscretion’.
A semi -idiom (Weinreich 1969, Cowie 1981), on the other hand, has one
or more literal constituents and at least one with a non -literal subsense, usually
special to that co -occurrence relation and no other: drop has the meaning ‘ to
overuse’ only when it co -occurs with the noun names . Similarly, catch in catch
one's breath means ‘to check’, and foot in foot the bill has the meaning ‘to pay’,
etc. Some of these semi -idioms resemble restricted collocations with specialized
subsenses which allow lexical variation (e.g. the adjective blue meaning
‘obscene’ may collocate with different nouns: film /joke/ gag/ story/ comedi an).
Literal idioms (e.g. on foot; tall, dark and handsome; waste not, want
not; on the contrary; a (very) happy birthday; a merry Christmas and a happy
New Year , etc.) meet the salient criterion for idioms, i.e. invari ance or restricted
variation. Howev er, they are less semantically complex than pure and semi –
idioms.
At this point, the conclusion can be drawn that the habitual co -occurrence
of words represents the basis of idiomaticity and an important linguistic
phenomenon which has two visible results in the development and functioning
of language. On the one hand, it leads to the formation of pure idioms , semi –
idioms , or literal idioms which show resistance in varying degrees to internal
lexical substitu tions and to their functioning as single lexic al units, despite their
being multiword expressions (compounds, phrases, semi -clauses, and clauses).
On the other hand, habitual co -occurrence of words leads to non -literalness.
This characteristic is a result of pure idioms losing the meaning of their
constituent words in time and of having, thus, an external meaning imposed on
the unit as a whole. In the case of semi -idioms, some of their member words

develop specialized subsenses in restricted contexts.
Envisaging another facet of idioms , that is the f act that they refer to “the
specific character, property or genius of any language; the expression which is
natural or peculiar to it”, Fernando (1996: 17) suggests the inclusion of
idiosyncratic idioms in the larger class of such (fixed) lexical patterns.
She shares Roberts’ (1944 ) opinion that “certain expressions are idioms
because they are peculiar to one language in contrast to another and as such
serve as a mirror of its cognitive design accounting for interlingual differences
in usage” (Roberts 1944, in Fernando 1996: 17). Seen from this perspective,
every idiom can be assumed to be the result of a personal innovation at a
particular point in time which was later adopted by the community, becoming
thus part of the vocabulary of that speech community.
Fernando also discusses Smith’s opinion (1925). He states that the term
idiom refers to
“those forms of expression, of gram matical construction, or of phrasing, which
are peculiar to a language, and approved by its usage, although the meanings
they conv ey are often different from their grammatical or logical signification ”
(Smith 1925: 167, in Fernando 1996: 18).
This definition enables Smith to classify many expressions, variable both
in their semantic unity and in their structure as idioms . In doing t his, Smith
pays more attention to interlingual structural peculiarities (‘idiosyncrasies’) and
intralingual specific features (‘idiomatic transgressions’) than to the semantics
of idioms. Such an approach is due to his essentially regarding idioms as those
forms peculiar to a language, whether literal or not.
In order to support his approach, Smith chooses word combinations
made up of verb + particle/preposition and other sorts of prepositional phrases,
as well as habitual collocations which include, in hi s opinion binomials (e.g.
heart and soul , bag and baggage , milk and honey , etc.), familiar similes (e.g.
as plain as a pikestaff ), and proverbs (e.g. out of sight, out of mind ). In
addition to such idiosyncrasies, Smith identifies ‘idiomatic transgressio ns’,
which break the rules of grammar (e.g. it’s me, who did you see?, try and go,
etc.), on the one hand, and the rules of logic, on the other (e.g. to keep one's
head above water, to curry favour , etc.). Referring to this particular class of
non-literal, ‘illogical idioms’ Smith states that they are figurative and, while
some are, perhaps, semi -opaque (e.g. to keep one’s head above water, to sail
too near the wind , etc. ), others are completely opaque (e.g. to curry favour, to
peter out, to go the whole ho g).
Linguistic and cultural studies on idioms and idiomatic expressions are
often doubled by psycholinguistic aspects. This is due to the fact that idioms are
often referred to as mentally represented and processed lexical items.

The lexical representati on hypothesis put forward by Swinney and
Cutler (1979, in Tabossi and Zardon 1995: 273) is an exemplification of this
statement. In their opinion, idioms are stored in the mental lexicon as
morphologically complex words and are retrieved in the same way as any other
word. The process of idiom retrieval is initiated as soon as the first word of an
idiomatic expression is encountered and, though running in parallel with the
computation of its literal meaning, ends before that, because the retrieving of
prest ored information is a faster operation than computing. As regards the
comprehension of the figurative meaning of an idiomatic expression, they state
that it is neither harder, nor more time consuming than the comprehension of its
literal meaning.
Another interesting perspective is offered by Cacciari and Tabossi
(1988). They suggest that idiomatic expressions may be mentally represented
and processed not as words , but rather as configurations of words whose
meanings become activated whenever sufficient in put has rendered the
configuration recognizable. Referring to configuration , they explain that it is
made up of the same lexical items that need to be activated during the
comprehension of the literal discourse, and that there is only one semantic
processi ng of the idiom, the lexical pattern remaining literal until the
configuration emerges and its meaning is activated. Moreover, referring to the
key in the string, i.e. the information which has to be processed literally before
the activation of the idiom f igurative meaning, Cacciari and Tabossi (1988)
point out the fact that its position determines the point in the string when
idiomatic meaning is activated. The contextual factors are attributed a very
importan t role in the recognition of idioms.
To put it in a nutshell, in the psycholinguistic perspective on idiom
processing , reference is often made to the idiom list hypothesis according to
which idioms are stored and accessed from some special list which is distinct
from the lexicon. This approach diffe rs from the lexical representation model
which suggests that idioms are stored as holistic entries in the same mental
lexicon that stores other lexical items.
The two perspectives with respect to idiom processing have given rise to
the creation of three m odels (Evaraert et al. 1995: 9) designed to investigate the
order in which alternative interpretations are considered. The first model is
related to the idiom list hypothesis and refers to the fact that idiomatic meaning
is considered only when non -idiomat ic interpretation fails. Such a model would
imply that it takes longer to interpret an idiomatic expression as such because it
requires reinterpretation. Since the processing of sentences which contain
idioms and metaphors do not take more time than the pr ocessing of sentences
with nonidiomatic constituents does, the conclusion could be drawn that non –
idiomatic meanings are not computed before idiomatic meanings.

On the other hand, the second model suggests that the idiomatic
meaning is activated first wh en encountering an idiomatic expression, and if it
cannot be integrated with the previous discourse, the non -idiomatic meaning is
constructed. Since numerous studies show that expressions are proc essed faster
in their idiomatic, than in their nonidiomatic sense, the conclus ion can be drawn
that idiomatic interpr etation is processed before nonidiomatic interpretation.
Finally, t he third model envisages the possibility that idiomatic and
nonidiomatic expressi ons are processed in parallel. Such a situation is possible
because idiomatic expressions are processed faster in their idiomatic , than in
their nonidiomatic sense , or because the meanings of idioms can be found in the
lexicon, thus requiring reduced time processing.
Referring to the same issue, Tabossi and Zardon (1995) consider that the
process of accessing an idiom is fundamentally different from that of assessing
the meaning of a single word. Their view is supported by the fact that idioms
and words are not stored in the same way, as well as to the f act that the
activation of idiom meaning is slower than the activation of word meaning and
it is more dependent on various contextual factors than in the activation of word
meaning. Similarly, Cacciari and Glucksberg (1995) suggest that the concrete
meanin g of an idiom is much easier to generate than the abstract, figurative one.
This situation is rather unexpected if the figurative meaning of an idiom is
extracted from the idiom list in the same way as the meaning of a single word is
extracted from the lex icon.
One possible solution could result from the fact that idiom
comprehension is different from the activation of prestored information which
fully describes the figurative meaning, but it is similar to the activation of
prestored information which is used to generate figurative meaning. Thus, idiom
processing is not just the retrieval of prestored information, but the mechanisms
underlying idiom processing .
To conclude, the views on idiomacity presented in this section, far from
being exhaustive, suppor t the idea that both the consistent terminology and the
large variety of approaches designed to explain this linguistic reality are hardly
satisfactory. Mo reover, their tentative classifications according to various
criteria is but a way of structuring available information on some general
guidlines.

1.4.9 .1. FEATURES OF IDIOMS

Similarly to collocations, idioms are generally agreed to have specific features.
Among the defining characteristics of idioms, specialists (Evaraert et al. 1995,
Fernando 1996, Moon 1998) include compositionality , institutionalization ,
semantic opacity , and (limited) formal flexibility .

The semantic compositionality of idioms is generally agreed to result
from the fact that they are usually defined as constituents or a series of
constituents for which the semantic interpretation is not a compositional
function of the formatives making them up .
Nicolas (1995) rejects this idea arguing that none of the parts of an idiom
has meaning , but the idiom as a whole. Similarly, Schenk (1995 ) states that it is
only the idiom as a whole that can refer, and since an idiom’s parts do not refer,
there cannot be a relation between the parts and the whole.
There are also alternative views, according to which the parts of idioms
have meaning, and t he relation between the meanings of the parts and the
meaning of the idiom is compositional . However, such views are often debated
as they are based on the importance of underlying metaphors for the
interpretation of idioms, and suggest that the parts of i dioms may have
metaphorical meaning. Geeraerts (1989) argues that the semantic specialization
of idioms is a matter of degree and that semantic interpretation is not only a
question of bottom -up compositionality or literal to figurative transfer , but als o
of a top -down and figurative to literal interpretation.
As regards the formal compositionality , idioms are commonly accepted
as a type of multiword expression (e.g. red herring, make up, smell a rat, the
coast is clear, etc.) though a few scholars (Hocke tt 1958, Katz and Postal 1963)
include even single words in the class of idioms.
Institutionalization is another widely accepted characteristic of idioms
since they are conventionalized expressions. However, the conventionalization
of idioms is the end -result of initially ad hoc, in this sense novel, expressions.
The semantic opacity specific to idioms is related to their semantic
compositionality and results from the fact that the meaning of an idiom is not
the sum of its constituents. In other words, an idiom is most often non -literal.
The possibility to identify the above idiom features in common word
combinations has proved extremely useful in the research on such combinations
and has resulted, as Fernando (1996 ) suggests, in “the identification of ma ny
types of multiword expressions, i.e. slang, proverbs, allusions, similes, dead
metaphors, social formulae, and collocations as idioms ” (Fernando 1996 : 3,
empha sis added) .
Certain studies focussing on the TG approach to lexical patterns include
another characteristic, i.e. flexibility, on the list of idiom features. Although the
adepts of TG have long supported the assumption that idiomatic constituents
behave in the same way as nonidiomatic constituents, authors of more recent
studies argue that syntact ic flexibility is closely connected with semantic
transparency and that, in many cases, idioms cannot undergo syntactic
transformations (e.g. kick the bucket does not passivize).
This idea of idiom flexibility is also shared by Chitra Fernando (1996)
who states that the distinctive feature of idioms, among which she also includes

sentence -like structures, is that, “though they are multiword expressions, they
are also lexicalized: they have the semantic unity of single words, but the
grammatical flexibility , though in varying degrees, of phrases, semi -clauses,
and clauses, which indeed the majority are” (Fernando 1996: 74).

1.4.9 .2. POSSIBLE CATEGORIZATIONS OF IDIOMS

Studies on idioms have highlighted the fact that this type of fixed lexical pattern
allows different categorizations. Depending on the perspective, reference is
often made to lexico – grammatical and to functional categorizations (Fernando
1996 ). The former category includes criteria such as collocational restriction,
lexico -grammatical struct ure, and semantic opacity, whereas the latter
establishes differences between idioms in terms of common functions and
features of usage.
In terms of collocational restriction, idioms are grouped in to four
classes : unrestricted idioms , (e.g. run a business ), semi -restricted idioms
(e.g. harbour doubt/ grudges/ uncertainty/ suspicion ), familiar idioms (e.g.
lukewarm reception ) and restricted idioms (e.g. pitch black ).
As regards their lexico -grammatical structure, distinction is made
between flexible (e.g. b reak sb's heart ), regular with certain constraints (e.g.
smell a rat ) and irregular idioms (e.g. the more the merrier ).
The classes of idioms identifiable in terms of semantic opacity
(Fernando 1996: 70) are: transparent idioms (e.g. long time no see ), semi-
idioms (e.g. a fat salary ), semi -transparent idioms (e.g. a watched pot never
boils ) and opaque idioms, which are further divided in to (i) overt , (e.g. OK)
and (ii) covert (e.g. kick the bucket ).
Even though the categorizations above are called lexico -grammatical,
they contain functional categories , as well. The inclusion of ‘sememic idioms’
(Makkai 1972) and ‘discoursal expressions’ (Alexander 1984, Carter 1987) in
idiom categorizations extends the scope of idiomatology beyond a relatively
narrow focus on lexico -grammatical classes (e.g. compounds, phrasal verbs,
binomials, ‘tournures’, and collocations) to functional types (e.g. proverbs,
allusions, quotations, greetings, fare wells, and other sorts of discoursal
expressions). This aspect favours a com prehensive study of what speakers do
with idioms in discourse. Moreover, the identification of specific idiom uses
reveals their more general role as non -conventional, familiar ways of saying
things.
The functional categorization of idioms (Fernando 1996) based on
Halliday’s model (1985) is concerned only with common domains and features
of usage. Distinction is made between ideational or ‘the state and way of the
world’ idioms, interpersonal idioms and relational idioms.

Ideational idioms are used either to signify message content, experiential
phenomena including the sensory, the affective, and the evaluative, or to
characterize the nature of the message as being specific or non -specific.
The idioms referring to message content are divided into seven cl asses
(Fernando 1996: 72 -73), namely idioms referring to:
1. Actions : e.g. tear down, mess about with, do a U -turn, twist sb.’s arm,
spill the beans, wear different hats, give sb. an inch (and he'll take a mile,
etc.), wave/offer/hold out the olive branc h, etc.
2. Events : e.g. turning point, out of the mouth of babes, have blood on
one’s hands, etc.
3. Situations : e.g. be in Queer Street, be in a pickle, be up a gum tree ,
etc.
4. People and things : e.g. a back -seat driver, a man about town, a scar –
let w oman, a fat cat, a red herring, a lounge lizard , etc.
5. Attributes : e.g. cut-and-dried, matter -of-fact, lily -white, as green as
grass, from A to Z, etc.
6. Evaluations : e.g. turn back the clock; it is a pity; matter -of-fact; cut –
and-dried; lily -livered; beauty is/lies in the eye of the beholder; waste
not, want not; a watched pot never boils; if you can’t stand the heat get
out of the kitchen; a Trojan horse; I am not my brother's keeper; etc.
7. Emotions : e.g. green with envy, have one’s heart in one’s m outh, a
lump in one’s throat, lose one’s heart, for one’s blood to boil, walk on
air, down in the dumps, tear one’s hair, etc.
Idioms may also be used to characterize the message . Distinction is
made between idioms which provide specific information ( e.g. to be exact/
precise, for example, that is, the question is, what I am saying is, my guess is, I
felt like saying , etc.) and idioms giving non-specific information ( e.g. kind
of/sort of, or something, such and such, and so on, etcetera, etcetera, blah, bla h,
blah, etc.)
Different from ideational idioms, interpersonal idioms are used either
in order to fulfil an interactional function , or to characterize the nature of the
message. In accordance with their interpersonal function, such idioms initiate,
mainta in, and close an exchange, being closely associ ated with politeness
routines. In such roles, these idioms exemplify the mores of social interactions ,
as well as the operation of some of the maxims of Grice’s Co -operative
Principle: sincerity, news , worth iness, and brevity, which characterize the
nature of the message.
The idioms’ classification in terms of interactional strategies (Fernando
1996: 73) results in 5 classes of such patterns, namely:
1. Greetings and farewells : e.g. good morning, , see you later, bye for
now, etc.
2. Directives : e.g. let’s face it, tell you what, say no more, believe you mc,

not to worry/don't worry, never mind, do you mind , etc.
3. Agreement : e.g. that’s true, you’re telling me, say no more, (that's) a
good question , etc.
4. ‘Feelers’, eliciting opinions : e.g. what do you think?, how do you feel?
5 Rejections : e.g. you are kidding/joking, come off it, tell it to the
Marines, I wasn't born yesterday.
If reference is made to the ways in which idioms may characterize the
message, distinction is made between their use to express:
1. Newsworthiness : e.g. guess what; what do you know; what, you ask?;
etc.
2. Sincerity : e.g. quite seriously, believe me, as a matter of fact, etc.
3. Calls for brevity : e.g. cut the cackle, get to the point , etc.
4. Uncertainty : e.g. I daresay, mind you , etc.
The idioms in the third and the last class are relational or textual idioms .
They ensure the cohesion, being therefore an element which contributes to the
coherence of discourse. According ly, they can be grouped along with
conjunctions (e.g. and, but, or , and so, because , if … then , etc.) due to their
having a textual function. Relational idioms are further categorized into those
which integrate information and those which sequence inform ation (Pawley and
Syder 1983):
Integrative idioms include five classes of such patterns, namely, 1)
adversative : e.g. on the contrary, far from, etc., 2) patterns use in making
comparison s: e.g. on the one hand … on the other , etc., 3) c ausal : e.g. s o that
when, the more X … the more Y, no wonder , etc., 4) concessive : e.g. at the same
time, 5) patterns used to express addition : e.g. in addition to, what is more, etc.
As for the idioms sequencing or chaining information , this function is
traceable at the level of meta -discoursal information , e.g. in the first place,
last but not least , or at the level of temporal information , e.g. one day, a long
time ago, up to now , etc.

1.4.10 . METAPHORS

Metaphors and idioms share certain formal and semantic featu res.
Similarly to idioms , metaphors are non -compositional because of their
semantics , and are characterized by various degrees of transparency.
Bearing in mind possible differences in terms of semantic features, Moon
(1998 a) identifies four types of metap hors :
1. Transparent metaphors – institutionalized patterns in the case of
which the ‘image ’ or ‘vehicle ’ (Leech 1969: 151) allows for their
successful decoding by means of real -world knowledge: e.g. alarm bells

ring, behind someone's back, breathe life, int o something; on (some)one's
doorstep, and pack one's bags.
2. Semi -transparent metaphors – patterns which require some
specialist knowledge in order to be decoded successfully. If the
institutionalized idiomatic meaning is unknown, there may be two or
more p ossible interpretations” e.g. grasp the nettle, on an even keel, the
pecking order, throw in the towel, and under one's belt.
3. Opaque metaphors, pure idioms – patterns where compositional
decoding and interpretation of the image are practically or complete ly
impossible without knowledge of the historical origins of the expression:
e.g. bite the bullet, kick the bucket, over the moon, fed herring, and shoot
the breeze. Such metaphors can be reinterpreted synchronically and
analysed compositionally.
4. Dead meta phors semantically transparent structures whose
constituent elements do not, in the straightforward sense, yield recurrent
semantic constraints .
Moreover, the effect of synonymous substitution and the continuing
relevance of their literal meanings makes it unsatisfactory to call such
metaphors ‘opaque’. Dead metaphors are rather semantically ‘translucent’. As
regards their quality of being ‘dead’, this aspect is related to the particular
syntactic form of such metaphors and to the fact that “with any modi fication ,
they spring to life” (Cruse 1986: 44). Metaphors are further analysed and
commented upon in 4.5.4.

CONCLUDING REMARKS

The most relevant aspects approached in the present chapter are summarized
below.

1. In the study of lexical patterns distincti on is often made between formal and
functional approaches. Formal approaches to lexical patterns are mainly
concerned with the form or structure of such word combination, and with the
ways in which form conditions their morpho -syntactic behaviour, their
semantic or pragmatic interpretation, or their comprehension and use by the
speakers of the language envisaged, but what is more important, by its non –
native speakers. Different from formal approaches to lexical patterns, which
focus on the interd ependence o f their structure at various linguistic levels (lexis,
syntax, semantics, etc.) , functional approaches envisage the more practical
aspects, that is their use by and accessibility to the end -users. In addition, the
functional perspective includes more behav iourist approaches, which look at
lexical patterns as encoding or enabling devices.

2. The terminology used to denote the various types of lexical patterns is as
challenging as confusing, irrespective of the language envisaged. Sometimes,
different terms ar e used to describe identical or very similar kinds of unit s,
whereas some other times a single term is used to denote very different
phenomena. Although often confusing, s uch consistent terminology proves the
specialists’ constant interest in the study of lexical patterns, as well as their
increasing awareness regard ing the frequent use of ready -made memorized
combinations in written and spoken language. It also proves the crucial part
such combinations play in language acquisition, on the one hand, and in
language production, on the other.

3. The wide circulation of certain terms related to idiomacity in English, has
favoured the inclusion of some of these terms in the specific studies devoted to
lexical patterns in other languages. Certain English terms rela ted to idiomacity
have also been borrowed by the Romanian specialists concerned with lexical
patterns and have been used, sometimes with their validated English meaning,
some other times with confusing and rather contradictory meanings.

4. The scale of idiomacity includes various types of lexical patterns, each of
which has specific formal and semantic features. A comparative -contrastive
description of these lexical pattern s is as necessary , as useful because it allows a
better understanding of their behavio ur and of the restrictions conditioning their
appropriate use.

2. APPROACHES TO COLLOCATIONS

When human beings are puzzled by the nature of an object, they naturally turn it
over and look at it from a different angle trying to get an idea of what that
object is and how it can be used. A similar, but more abstract approach can be
used in linguistic research, but the essential difference is that in this situation it
is by far more difficult to reach any definite conclusions. This is bec ause there
are numerous factors, whether linguistic, or extralinguistic, conditioning the
appropriate decoding of the linguistic unit envisaged (word, pattern, sentence,
paragraph, or text).
Moreover, certain ways of conceptualizing the nature of a lingui stic
phenomenon tend to get entrenched so that even if that phenomenon remains
puzzling, speakers refuse to give up a perspective that seems to have yielded
useful insights in the past. Therefore, it is often necessary to reconsider the
underlying assumpti ons of an already known viewpoint before looking at a
problem from a new perspective.
In the case of collocations , the underlying assumptions are mainly about
categorization and compositionality, issues which raise questions about the
nature of language a nd meaning and over which linguists often have quite
different views. However , the idea is generally shared that collocations have
often to do with alterations and extensions of meaning, dictated by
extralinguistic criteria. Psycholinguistic , social , and cultural aspects are
factors most often restricting the speakers’ ability to appropriately decode or
encode collo cations, either in their native , or in a foreign language. Hence, the
necessity to approach collocations from these perspectives.

2.1. LINGU ISTIC APPROACHES TO COLLOCATIONS

As a subdiscipline of linguistic studies, phraseology has aroused the interest of
specialists from different areas of study. Linguists have focused on the semantic
and syntactic properties of the various types of phraseolo gical units, on possible
approaches to their synchronic and diachronic description, as well as on their
pragmatic function in discourse, and, quite recently, on cultural peculiarities of
idioms and phrases in the light of cross -cultural and contrastive ap proach es.
The same holds valid for collocations , whose complex and controversial
nature has drawn the attention of specialists working in various areas of
linguistic studies such as psycholinguistics, sociolinguistics and cultural studies .

2.1.1. TRADITIONAL AN D MODERN VIEWS ON COLLOCATIONS

Although the linguists following Saussure and Chomsky seem to have shown
little interest in the study of collocations and of other word combinations, the
last decades of the 20th century brought these issues to the attentio n of
specialists working in various fields of linguistic studies (lexicology,
lexicography, semantics, lexical semantics, syntax, pragmatics, translation,
language teaching, psycholinguistics, sociolinguistics, cultural studies), who
have constantly tried to come up with new ways of defining and classifying
collocations and other more or less similar patterns. However, mention should
be made that the English tradition in the study of collocations has not been
equa lled yet by similar studies in other langua ges. This is the case with
Romanian and Italian, which, in spite of benefiting from the results of some
phraseological studies, hardly make any specific reference to collocations . A
separate category of studies in Romanian is represented by the books and
articles contrasting the English collocations with their equivalent patterns in
Romanian (see Levițchi 1970, Chițoran 1973, Constantinescu 1976, Avădanei
2000).
The shift from traditional to modern in the study of collocations has
brought to light two dif ferent perspectives on one and the same linguistic
phenomenon: one oriented towards specifically lexical and another one oriented
towards specifically grammatical patterning.
The term ‘collocation’ was first introduced by Firth in 1957 and referred
to one of the “levels” of meaning, distinction being made between ‘meaning by
collocation’ on the one hand, and the ‘conceptual approach to the meaning of
words’ and ‘contextual meaning’, on the other.
Firth’s view is supported by his contextual theory of meani ng put forward
in 1957. In contrast with structural linguistics, which focuses on general
abstractions of the properties of phrases and sentences, Firth’s theory highlights
the fact that context is the key element to be used in decoding meaning, both in
short and long word combinations. Although the context of social setting and
the context of spoken and textual discourse are important, collocations are very
much conditioned and influenced by the context of surrounding words. Hence,
his famous statement: ‘a word is characterized by the company it keeps’.6

6 A similar definition is provided by Van Roey (1990: 46) who states that the term
collocation refers to “the linguistic phenomenon whereby a given vocabulary item
prefers the company of another item rather than its ‘synonyms’ because of constraints
which are not on the level of syntax or conceptual meaning but on that of usage”.

Later on, linguists working in the tradition of lexical collocations
(Halliday 1966, Sinclair 1966, Hasan 1987) put forward the idea that patterns of
collocation can form the basis for a lexical analysis of language alternative to,
and independent of grammatical analysis. They were particularly interested in
finding resources to study le xis in the same way as grammar. More precisely,
their aim has been to examine lexis at the linguistic level in parallel wit h and
overlapping grammar, but preserving, at the same time, its independence. In
doing s o, they focused their attention on patterns of chain (syntagmatic axis)
and choice (paradigmatic axis):

chain choice
grammar structure system
lexis collocation set

Table 7. Lexis as a linguistic level (source Carter 1998: 52)

This comparative -contrastive view on lexis places collocations on the
syntagmatic axis and requires clarification of notions such as ‘ cluster ‘and
‘lexical sets’ .
As Carter (1998: 52) menti ons, lexical items can co -occur with any other
lexical item. However, there are words which are more likely to keep company
to a given lexical item X than to others. For example, sun has a high probability
of co -occurrence with shine , burn , protection , light, but a low probability of co –
occurrence with pear, cat, or book . The words which are found to collocate with
the lexical item X will make up its cluster . The members of the cluster which
are more central will have a higher probability of co -occurrence w ith the item
X.
A close analysis of the collocational behaviour of other lexical items (Y
and Z) within the cluster of X will highlight the fact that the clusters of Y and Z
will contain not only X, but also many of the lexical items making up the cluster
X. If two or more clusters have a high proportion of items in common, then the
mixed clusters will form the so -called lexical sets .
Drawing a parallel between lexis and grammar, Carter (1987) states that
speakers “choose items from lexical sets as they cho ose types of grammatical
items from grammatical systems or build up grammatical structures” (Carter
1987: 50)
At this point, the fact is worth mentioning that sets overlap and intersect
with each other in such a way that one and the same lexical item may o ccur in
different axes of meaning. This situation requires that lexical sets should be
illustrated and examined with respect to a certain field. The term ‘ field’
designates “the particular activity, cultural feature, social institution, or topic for

which a particular set of ideationally related lexical items is often evolved or
adapted” (Carter 1998: 53).
The conclusion which could be drawn at this point is that, in comparison
with the analysis of grammatical relations, which are more fixed and
determined, the analysis of lexical relations favours only a probabilistic
argumentation. To put it differently, the lexical orientation views collocation as
“a relation of probable co -occurrence of items” (Malmkjaer 1991: 302) .
Moreover, the fact has been pointed o ut that the lexical level can never be
totally independent of the level of grammar, grammatical collocations and
larger stretches of text being illustrative in this respect. As Greenbaum (1970:
11) puts it, “a purely item -oriented approach to collocations is not a good
solution as it obscures syntactic and semantic restrictions on collocations”,
aspects which are essential in their treatment and analysis.
In contrast with the lexica l-oriented view, which attempts to establish
lexis as an independent and s eparate linguistic level, the grammatical
orientation in the study of collocation regards lexical and syntactic patterning as
distinct, but interrelated levels of structure.
Changing his initial view on collocations , Halliday (1978a) refers to
“lexico -gram mar as a stratum where lexical and grammatical structures which
realize the output from the semantic component of a text are mapped onto one
another”. The three strata making the linguistic system are in his opinion:
lexico -grammar, semantics and phonology .
This view is also shared by other linguists. Sinclair, for example, states
that evidence from large corpora “suggests that grammatical generalizations do
not rest on a rigid foundation, but are the accumulation of the patterns of
hundreds of individual w ords and phrases” (Sinclair 1991: 100). Moreover,
Mitchell (1971) points out the fact that “lexical particularities are considered to
derive their formal meaning not only from contextual extension of a lexical
kind, but also from the generalized grammatica l patterns within which they
appear” (Mitchell 1971, in Carter 1998: 59).
A more formal approach to collocations is the semasiological one. It
focuses on the collocator and it demonstrates how the corresponding bases
function in order to help explain the m eaning of collocators . Base words are
assumed “to have the same sense as they would in free combinations” (Nicolas
1995: 234). This means that they are semantically more autonomous than the
collocator, being thus less likely to pose meaning problems. On th e other hand,
collocators/collocates have a different semantic behaviour. They “can enter in
free combinations in other contexts but they will always have a specific, non –
free sense ” (id. ibid.) when combined with base words. Their semantic
interpretation is strictly conditioned by the base word with which they occur. If
bases can be learnt on their own, collocators must be learnt in collocation. To
put it differently, the definition of the collocator is “incomplete without the

syntagmatic dimension of the collocation” (Hausmann 1979: 192, in Cop 1990:
39).
McCarthy’s (1984, in Carter 1998: 87) reference to collocations from a
pragmatic point of view is also worth mentioning. He considers that collocation
can be regarded as another instance of lexicalization in text, being a pragmatic
process which is constrained neither by lexical items realizing only specific
functions, nor by lexical items which have fixed meanings and which cannot
thus be analysed independently of their place in a system of meanings.
More over, he suggests that the abstract phenomenon of collocation should be
replaced by the notion of significant collocation .
A very important stage in the study of collocations is marked by John
Sinclair’s view. He states that collocations bring about the no tions of
dependent and independent meaning , the relation between texts and grammar
and the implementation of the two models of interpretation for the concept of
collocation , namely, ‘the open -choice principle ’ (OCP ) and ‘the idiom
principle ’ (IP) (Sinclair 2001, 2004).
The OCP corresponds to the independent meaning and it “is a way of
seeing language texts as the result of a very large number of complex choices.
At each point where a unit is completed (word, phrase, clause), a large range of
choices opens up, and the onl y restraint is grammaticalness” (Sinclair 1991:
109). If things are seen from this perspective, the conclusion may be drawn that
“virtually all grammars are constructed on the OCP ” (Sinclair 1991: 109).
In comparison with the OCP , the IP is closely related to the idiomatic
nature of words, which can combine creating structures with new meanings,
different from those of their constituents. The sets of linguistic choices coming
under the heading of register can be seen as large -scale conditioni ng choices.
Once a register choice is made, all the ‘slot -by-slot’ choices are massively
reduced or even pre -empted. Thus, at a point in text, where the open -choice
model would suggest a large range of possible choices, the IP restricts it
dramatically ove r and above pred ictable semantic restraints which result from
topic or situational context.
In other words, the IP is based on the fact that language users have access
to a large number of semi -preconstructed phrases which constitute single
choices, even though they might appear to be analyzable into segments. This
may reflect, to a certain extent, the recurrence of similar situations in human
affairs, as it may illustrate a natural tendency to economy of effort, which is
partly motivated by the exigencies of real time conversation. At its simplest, the
IP can be seen as the apparently simultaneous choice of two words which will
operate effectively as a single lexico -semantic unit.
Moreover, Sinclair (2004: 29) mentions that each of the two principles is
illustrative for a certain tendency which is visible in the behaviour of words. In
his opinion, the OCP generates the terminological tendency, i.e. the tendency

for a word to have a fixed meaning in reference to the world, whereas the IP
generates the phras eological tendency, i.e. the tendency for a word to make up
patterns and get meanings by these combinations (e.g. collocations and other
features of idiomacity).
Sinclair (1991: 112 -113) also provides a detailed description of the
features of the IP. In hi s view, this principle is characterized by the following
determining aspects:
 many phrases have an indeterminate extent (e.g. to set eyes on attracts
a pronoun SB, the adverb never or a temporal compensation like the
moment , the first time and the auxili ary have );
 many phrases allow internal lexical variation in some cases (there is
little choice when selecting between to set X on fire vs. to set fire to X ,
or between in some cases and in some instances );
 many phrases allow internal lexical syntactic va riation (in a syntactic
structure like It is not in his nature + to infinitive to …, there are
replaceable words: is not and his which may be replaced by it is hardly /
scarcely and my, your, etc. ; and fixed words: in , nature, to infinitive );
 many phra ses allow some variation in word order ( e.g. to recriminate
is not in his nature vs. it is not in his nature to recriminate );
 many uses of words and phrases attract other words in strong
collocations: (e.g . hard work, hard luck, hard facts, hard evidence );
 many uses of words and phrases show a tendency to co -occur with
certain grammatical choices (e.g. the phrasal verb to set about , is
closely associated with a following verb in the -ing form: leaving ). In
such structures, the second verb is usually transit ive: to set about testing
it.
 many uses of words and phrases show a tendency to occur in a certain
semantic environment: (e.g. the verb to happen is often associated with
unpleasant things such as accidents , disasters ).
The features mentioned above prove that the idiom is at least as
important as grammar in the explanation of how meaning arises in text,
accounting, at the same time, for some important aspects.
Firstly, there is a broad general tendency for frequent words, or frequent
senses of words, to h ave a less clear and independent meaning than those words
or senses which are less frequent. These meanings are difficult to identify and
explain. With very frequent words, stress is laid on uses rather than on
meanings, this tendency being the result of a progressive delexicalization, or
reduction of the distinctive contribution made by that word to the meaning.
Secondly, the dependence of meaning correlates with the operation of the
IP to make fewer and larger choices. The evidence of collocation supports this
point of view. If words collocate significantly, then to the extent of that
significance, their presence is the result of a single choice.

In addition, the core meaning of a word , i.e. the one which first comes to
people’s minds , will not normally be a delexical one. A likely hypothesis is that
the core meaning is the most frequent independent sense. If extensively tested,
this hypothesis would better explain the discrepancy between the most frequent
sense and what intuition suggests is the most impor tant or central one.
Furthermore, general texts are usually made up of the occurrence of
frequent words and of the frequent meanings of less frequent words. Hence,
such texts are largely delexicalized and appear to be formed by exercise of the
IP, with occ asional switching to the OCP .
Finally, in many cases it is unhelpful to analyze grammatically any
portion of text which appears to be constructed on the IP. This is due to the fact
that boundaries between stretches constructed on different principles will not
normally be clear -cut, and not all stretches will carry enough evidence to
suggest that they are not constructed by the normal grammar rules.
Although both the IP and the OCP are ‘ways of seeing’ or of interpreting
language, the appropriate selection b etween the two principles in interpreting a
sequence of words entirely depends on the language users who will have to
decide whether to interpret the respective sequence as a chunk, or as a series of
individual items.
Nevertheless, when confronted with no rmal texts, Sinclair suggests that
the IP should take precedence over the OCP . To put it in his words, “[F]or
normal texts, we can put forward the proposal that the first mode to be applied
is the IP, since most of the text will be interpretable by this pr inciple. Whenever
there is good reason, the interpretive process switches to the OCP , and quickly
back again. Lexical choices which are unexpected in their environment will
presumably occasion a switch; choices which, if grammatically interpreted,
would be unusual are an affirma tion of the operation of the IP” (Sinclair 1991:
114).
In addition, the large number of lexical patterns in English makes the
recognition of the ‘idiom principle’, strongly realized in idioms , weakly
realized in collocations , very u seful in the explanation of the way in which an
important part of the vocab ulary works. In other words, it brings to the fore the
idea that “language -users are sensitive to word combinations, not merely words
as self -contained isolates , and employ such c ombi nations, some of which are
idioms, for various purposes” (Fernando 1996: 38).
Sinclair’s discussion on the elements making up a collocation and the
possible ways in which these elements interact with each other represents
another important contributio n to the better understanding of collocations . He
states that when two words of different frequencies collocate significantly, the
collocation has a different value in the description of each of the two words.
Considering the frequency of the co -occurrence of two words in a combination,
Sinclair distinguishes two constituent elements: ‘ the node’ , i.e. the word which

is being studied and ‘ the collocate’ (Firth 1968, Sinclair 2004), i.e. the word
which occurs in the specified environment of a ‘node’. Thus, ea ch successive
word in a text is both node and collocate , though never at the same time. From
this perspective, collocations can be referred to as “lexical associations of
nodes” (Sinclair 1991: 170, Jantunen 2004: 105).
As regards Sinclair’s analysis of th e interaction between the node and the
collocate , it highlights two situations of collocability. The former, called
downward collocation , when a is the ‘node’ and b is a ‘collocate’ is the
weaker pattern in which words tend to be elements of grammatical fr ames
(superordinates). The latter, when b is the ‘node’ and a is a ‘collocate’, gives a
semantic analysis of a word, the resulting structure in this situation being an
upward collocation .
To put it in other words, “upward collocation is collocation with w ords
that are more frequent than the node e.g. very confused (very is more frequent
than confused ), whereas downward collocation is with words less frequent
than the node e.g. utterly confused (confused is more frequent than utterly )”
(Sinclair et al. 2004 : XXIII, his examples).

back

+ Downward collocates

back
+ Upward collocates
verb: arrive, come,
drove, looked, pulled, ran,
sat, threw, wave , etc. prep./adv./conj .: at, from,
into, on, then
prep .: along, behind,
onto, past, toward , etc. pron : her, him, me, she,
them, we
adv.: again, forth,
further, slowly, straight,
etc. possessive adj .: her, his, my,
your
adj.: normal, etc. verb : get, go
noun : camp, flat, garden,
home, office, road,
streets, etc.

Table 8 . Upward and downward collocates of back (source Sinclair 2004)

Sinclair (1991) also extends his approach to collocation by analyzing the
co-occurrence relationships specific to lexical patterns. In doing this, he
suggests four types of co -occurrence relations , namely:
1. colloc ation – a node -collocate pair illustrative for a purely lexical
relation, non -directional and probabilistic, which ignores any syntactic
relation between the words (e.g. loud/ spontaneous/ polite, warm
applause ).

2. colligation – the relation between a pair of grammatical categories or,
a pairing of lexis and grammar (e.g. in some/ many/most/more/both,
several cases )
3. semantic preference – the relation between a lemma or a word -form
and a set of semantically related words (e.g. large number, scale, part,
amounts, quantities, areas ). In other words, semantic preference is the
restriction of regular co -occurrence to items which share a semantic
feature, this feature being equally relevant to both syntagmatic and
paradigmatic phenomena.
4. discourse prosody is a feature which extends over more than one
unit in a linear string and expresses the speakers’ attitude. For example,
the verb to cause occurs with words denoting unpleasant events (e.g.
problems, death, damage, concern, trouble, cancer, etc.) whereas provid e
occurs with words denoting things which are desirable or necessary (e.g.
support, goods, etc.) (Sinclair 1991: 111 -112)
The first three categories are related to each other, being increasingly
abstract, which is not the case with the discourse prosody . If collocation is
located in the physical text, colligation found where there is preponderance of
one particular word class, semantic preference is identifiable if speakers
notice similarity of meaning regardless of word class (Sinclair 2004: 142).
Many sp ecialists refer to collocations from the perspective of lexical
cohesion . Gutwinsky (1976: 57 in Carter 1998: 80) suggests that a discussion
with reference to lexical cohesion would require considering both formal and
semantic criteria. He considers that l exical cohesion has the following
characteristics: repetition of an item, occurrence of a synonym or item formed
from the same root, as well as occurrence of an item from the same lexical set.
This view is shared by Halliday (1994: 310) who states that lexical
cohesion can be achieved by the repetition of a word, or by using a word that is
related to a previous word “either semantically, such that the two are in the
broadest sense synonymous, or collocationally, such that the two have a more
than ordinary s ense to co -occur”.
Halliday and Hasan ( 1976 ) make reference to two major categories of
lexical cohesion, namely: reiteration and collocation . In their opinion, the
former includes repetition, synonymy, superordinate terms, and general words
(general noun) and it is achieved not only through repetition of an identical
lexical item, but also through occurrence of a different lexical item that is
systematically related to the first one, as a synonym or superordinate of it. In
comparison with reiteration, the term collocation refers to cohesion achieved
through the association of lexical items that co -occur regularly.
Considering taxonomic and collocational relations , Martin (1992)
divides the former into superordination (synonymy, hyponymy, antonymy)
and comp osition (e.g. part -whole relations, part -part relations), and states that

the latter “present the greatest problems of replicable identification” (Martin
1992 in Carter 1998: 81), adding that “interpreting how loosely or tightly
collocating items are strun g together” is by far the most problematic aspect in
the analysis of collocational relations .
Considering pragmatic aspects, Nunan and Carter (1993) refer to
collocations from the perspective of lexical cohesion stating that this type of
cohesion “occurs w hen two words in a text are semantically related in some
way” (1993: 29). He adds that since collocations include all those items in a
text that are semantically related, discourse analysis may become problematic.
However, the problem in the case of colloc ations is that they are made up of
open, rather than closed class items (pronouns, conjunctions, prepositions) with
a finite membership in vocabulary. The lack of restrictions with respect to the
items that can be used to form a collocation makes it diffic ult to establish sets
of regularly co -occurring words and phrases. Moreover, the fact that lexical
relations are textually and contextually bound makes it impossible to develop a
finite list of relatable lexical items. As regards the background knowledge o f the
reader, it plays an obvious role in the perception of lexical relationships, which
means that collocational patterns will only be perceived by somebody who
knows something about the subject.
To conclude, the various approaches to collocations present ed above
prove once more that language is indeterminate in its interpretation and hence
very flexible in use. Moreover, when the switch points between two modes of
interpretation are not always explicitly signa lled, or two modes offer sharply
contrasting w ays of interpreting data, messages are likely to be interpreted quite
differently, so much the more if the interpretation is not made by the same
person.

2.1.2 . DEFINITIONS AND CHARACTERISTICS OF COLLOCATIONS

The complexity of word combinations in gene ral and of collocations in
particular challenged the specialists in the field to find new instruments and
methods to define and differentiate them.

2.1.2.1 . DEFINITIONS OF COLLOCATIONS

Taken randomly, collocations are expressions consisting of two or m ore words
that correspond to some conventional way of saying things. Noun phrases like
strong tea and weapons of mass destruction , phrasal verbs like to make up , and
stock phrases like the rich and powerful are clear illustrations of such
conventionalized word structures.

Although shared by many linguists, this point of view is too limited to
cover the numerous and various aspects that should be taken into consideration
when defining and analyzing the behaviour of such lexical pattern s.
Some linguists refe r to collocations in terms of short structures such as
“minimal syntagmatic units” (Quem ada 1968: 507 in Cowie 2002: 74) or
“skeleton examples” (Palmer 1936 in Cowie 2002: 74), whereas some others
provide longer and more detailed definitions.
A very expre ssive and illustrative definition of collocation is the one
provided by Firth, who refers to this linguistic phenomenon in terms of “the
company that words keep” (Firth 1957: 179) .
Halliday and Hasan’s (1976) view on collocations is more inclusive, but
their definition is less precise than those provided by other linguists. In their
opinion, collocations include “those sets of words whose members participate
in a semantic relation”. They consider that “a marked cohesive effect is present
in a given text whe n two semantically related words occur in close proximity,
even though it may be difficult to systematically classify all the semantic
relations that may occur” (Halliday and Hasan 1976: 319).
Mel’čuk’s approach to collocations is more restricted. He consi ders that
in the meaning -text model, collocations are positioned within the framework of
lexical functions. “A lexical function is a semantico -syntactic relation which
connects a word or phrase with a set of words or phrases. Lexical functions
formalize th e fact that there are words or phrases whose u sage is bound by
another word” (Mel’čuk 1998: 30).
In an attempt to define collocations, Christopher Ball (in Bolton and
Crystal 1987: 187 -188) takes into discussion some aspects regarding
collocational restrictions . He notices that in many cases, collocat ions do not
occur frequently, not so much because of the incompatibility between the
constituent words, but because of the choice of inappropriate referents: e.g.
blank look, mother glass, blank glance , as opposed to * glad ear, *glad clothes .
Although coll ocational restrictions are sometimes placed on the border between
grammatical and logical restrictions, the bulk of collocational patterning is
clearly within the sphere of lexical organization.
Other definitions of collocations are built up by comparing and
contrasting this type of lexical pattern with other lexical classes. Thus , a
collocation is “a hyponym of the more commonly known one and, thus, more
specific” (Cop 1990: 35), or “habitual or expected co -occurrence of words as
opposed to free combinati ons” (Veisbergs 1997: 161).
Furthermore, considering the phenomenon of recurrency , some linguists
provide definitions which refer to collocatio ns in terms of possible number of
occurrences. Thus, collocations may be associations occurring at least twice
(Kjellmer 1987), or associations occurring at least four times (Kennedy 1991),
up to ten times. They are also relatively frequent lexical combinations in

comparing translations and non -translations (Mauranen 2000). Considering the
same aspect, Cook gives no precise indication of how often collocations occur
in language by merely stating that they are “frequent combinations of words”.
He notices that “there are countless combinations of words which are
grammatically possible but do not occur or occur rarely, a s well as collocations
with similar meanings which occur with great frequency” (e.g. to provide
assistance more usual than to supply assistance – his examples) (Cook 2003:
73).
A comprehensive and accessible definition combining formal and
semantic aspects is provided by Cruse (1986) , who states that
“collocation is a term used to refer to sequences of lexical items which
habitually co -occur, but which are fully transparent in the sense that each lexical
constituent is also a semantic constituent. Collocat ions have a kind of semantic
cohesion in the sense that the constituent elements are, to varying degrees,
mutually selective. The semantic integrity, or cohesion, of a collocation is the
more marked if the meaning carried by one (or more) of its constituen t elements
is highly restricted contextually, and different from its meaning in more neutral
contexts”(Cruse 1986: 40).
While pointing out some methodological aspects related to the teaching
of word combinations, Woolard (2000) provides a more practical de finition of
collocations which are seen as “those groups of words which students will not
expect to find together” (Woolard in Lewis 2000: 29).
Sinclair’s view on collocations (1991, 2004) is more complex and
analytical. In his opinion collocations may be defined as “frequent co –
occurrence of words, which does not have a profound effect on the individual
meanings of the words, but there is usually at least a slight effect on the
meaning, if only to select or confirm the appropriate meaning” (Sinclair 2004:
28). He notices that, in some cases, none of the words may appear to contribute
directly to the meaning of the expression: i.e. to bear on – “to be relevant to”.
Moreover, in other instances, certain words may, while others may not
contribute directly to the meaning of the expression: e.g. to beat somebody up.
Finally, there are word combinations in which each constituent still seems to
mean what it normally means: e.g. to rain beats down (Sinclair’s examples).
Comparing the OCP and the IP, Sinclair also refers to the concept of
collocation as an illustration of the IP.
The restrictive use of words in lexical and grammatical associations
noticed by some specialists gave way to a new definition of collocations . Seen
from this perspective, lexical associati ons are defined as the systematic co –
occurrence patterns that a target word has with other words (Biber et al 1998: 6,
Jantunen 2004: 105 in Croitoru and Dumitrașcu 2006b: 105).
This is also Sinclair ’s view (1991: 170) who defines collocation as “the
occurrence of two or more words within a short space of each other in a text”,

mentioning that collocation typically denotes frequently repeated or statistically
significant co -occurrences, whether or not there are any special semantic bonds
between collocating items. In his opinion, the fact that the term ‘collocation’ is
sometimes used to designate weak kinds of lexical patterns, in contradic tion to
‘pure idioms’ gives rise to terminological confusion.
Considering the interaction between lexis and syntax in various types of
patterns, Widdowson (2000: 60 ) refers to collocations with the general term
“sequences of lexical items”, and states that they are fixed, but there are
numerous other lexical sequences which are not and which can be syn tactically
modified only up to a point. In his opinion, such sequences are placed at the
uncertain border between lexis and syntax, where words move from a
compounding to a combining relationship.
Last, but not least, Howar d (1988) refers to collocations in terms of word
combinations that have a certain mutual expectancy, pointing out that such
combinations are not fixed expressions, but there is a greater th an chance
likelihood that the words making them up will co -occur (Howard 1988: 96).
As far as the definitions of collocations provided in Romanian and Italian
are concerned, mention should be made that they are not so numerous. This is
due to the fact that the term has only relatively recently been introduced in the
Romanian and Italian lexico -semantic studies and to the existence and more
extensive use of other terms to denote this linguistic reality. However, the
definitions of collocations in Romanian an d Italian preserve some of the aspects
referred to by the English specialists in the field. According to Bidu -Vrânceanu
et al. ( 2005), the term collocation denotes those syntagmatic relationships
referring to the interdependence between lexemes used togeth er in the same
contexts. Reference is also made to their frequent use and to the various
semantic relationships of synonymy, antonymy, and hyponymy, and it is
pointed out that specialists have no t agreed whether the collocability of the
words making up co llocations is conditioned by their meanings or by their
distribution:
“Colocație – concept care desemnează mai multe tipuri de relații sintagmatice
bazate pe interdependența dintre lexemele care tind să fie utilizate alături în
aceleași contexte. Nu s -a căzut de acord dacă aceste posibilități de colocație fac
parte din sensurile cuvintelor sau țin numai de distribuția lor. În cazul aceleiași
colocații se stabilesc legături între sinonime, contrarii, cupluri complementare.
Termenul este utilizat mai ales î n semantica anglo -americană și cu totul izolat
în lingvistica românească ” (Bidu -Vrănceanu et al. 2005: 114).
Moreover, the term frequently used in Romanian for the lexical patterns
functioning as a semantic unit and indicating a unique reality is “ unitate
frazeologică” , which is defined as “combinație stabilă de două sau mai multe
cuvinte având sens unitar și referent unic” (Bidu -Vrănceanu 2005: 223 -224).
Considering the morphological value of such fixed word combinations,

reference is made to the existenc e of three types of phraseological units, namely
nominal (e.g. copil din flori, luna de miere, porc mistreț, țap ispășitor ), verbal
(e.g. a-și lua lumea -n cap, a -și tăia funia de sub picioare ) and adjectival (e.g.
slab de înger, din topor, în doi peri, tob ă de carte ) (their examples). Comparing
phraseological units with free word combinations, it is obvious that the former
are understood by the speakers as an entity existing in language as such and that
it is comparable with catachresis .
The definition of collocations in the Italian language is also worth
mentioning here. The term collocazione is used to refer to those combinations
of two or more words, which, although semantically independent and likely to
have a substitute, form an expression considered t ypical due to its usage (my
translation) ( Lo Zingarelli 2003: 393).
Dictionaries, as repositories of the numerous changes undergone by
words in the course of time, may be a good source of information for the better
understanding of such a linguistic aspect virtually neglected by the study of
language or inaccessible to the public. Here are some dictionary definitions
provided for the term collocation :
 collocation – 1.[U] the way in which some words regularly collocate
with others 2. [C] a habitual combinati on of words which sounds natural:
'Strong coffee' is a typical and very common collocation in English but
'powerful coffee' is not accepted. (LDELC 2001: 246)
 collocation – n. [C, U] (technical) the way in which some words are
often used together, or a pa rticular combination of words used in this way
(LDCE 2003: 294)
 collocation – 1. [U] the action of collocating. 2. [C] a regular
combination of words: 'resounding victory' and 'crying shame' are English
collocations. (OALD 1995: 220)
 collocation – [L collo catio (n-), f. as prec.: see -ATION.] The action of
collocating; the state of being collocated; a disposition, an arrangement;
(spec. ling., (esp. habitual) juxtaposition or association of a particular
word with other particular words, a group of words so a ssociated. (OTD
virtual)
 collocation n. 1. the act of collocating or the state of being collocated.
2. an arrangement or juxtaposition, especially of linguistic elements, such
as words. (RHD)
 collocation – n. 1. the act of collocating. 2. the state or man ner of being
collocated. 3. the co -occurrence of words, esp. when habitual, as of
perform with operation or commit with crime (WNED 2000).
Although simple, and very comprehensive, dictionary definitions are very
similar and limited, covering only few of t he numerous aspects characterizing
collocations. Consequently, such definitions can have only practical value,
being more useful to the non -specialists in the field.

2.1.2.2 . CHARACTERISTICS OF COLLOCATIONS

Collocations may be distinguished from other w ord combinations by means of
their features, namely non-compositionality , non-substitutability and non-
modifiability.

2.1.2.2.1 . Non-compositionality7

A natural language expression is compositional if the meaning of that
expression can be predicted fro m the meanings of the parts. From this point of
view, collocations are not fully compositional in that there is usually an element
of meaning added to the combination that cannot be predicted from the parts. In
other words, the meaning of a collocation is not a straightforward composition
of the meanings of its parts. For example, white wine, white hair and white
woman all refer to slightly different colors, so they can be regarded as
collocations. In the case of strong tea, the adjective strong has acquire d the
meaning ‘rich in some active agent’ which is closely related, but slightly
different from the basic meaning ‘having great physical strength’ .
Collocations such as international best practice exhibit milder forms of
non-compositionality . It is very n early a systematic composition of its parts, but
it still has an element of added meaning. It usually refers to administrative
efficiency and would, for example, not be used to describe a cooking technique
although that meaning would be compatible with its literal meaning.
Moreover, restricted collocations like foot/ pick up/ settle the bill are
considered to occupy an intermediate position, being placed between ‘idioms’,
which are typically defined as non -compositional, multi -word units (e.g. fill/ fit
the bill) and free combinations (e.g. discuss/ tear up the bill ), which are seen as
fully -compositional.
The deviation of restricted collocations and idioms from full
compositionality is contrasted with the view adopted by cognitive linguists that
“component structures are not the building blocks out of which a composite
structure is assembled, but function instead to motivate various aspects of it”
(Langacker 1987: 449). According to this view, full compositionality should be
regarded as an exception rather than the norm. In the same line with the idea of
full compositionality, studies in phraseology typically treat restricted
collocations as not only idiosyncratic but also arbitrary. This is in contrast to

7 Non-compositionality is considered by some linguists a defining criterion of
phraseological units, while others view it as a se condary feature characterizing some,
but not all units.

“perfectly open collocations formed by general princ iples of co -occurrence”
(Cowie and Howarth 1996: 83).
According to Lakoff (1987: 148), the assumption that “the meaning of the
whole is a computable function of the meaning of the parts plus the syntactic
relationship between the parts is ‘simply wrong’”. However, this assumption
seems to underlie the terminology used by Moon (1998: 20). In her typology,
the equivalent of restricted collocations is ‘defective collocations’. They are a
subgroup of ‘anomalous collocations’ which are problematic from a lexico –
grammatical point of view. The fact that they are syntagmatically or
paradigmatically anomalous results both in the impossibility to decode them
only compositionally and to encode them freely. Moon’s sub -classification is
based on ‘the nature of the anoma ly’. Trying to provide an alternative to the
anomalous nature of collocations, Lakoff suggests the concept of ‘motivation’
stating that “the meaning of the whole is often motivated by the meaning of
the parts , but not predictable from them” (Lakoff 1987: 1 48) (emphasis added).

2.1.2.2.2 . Non-substitutability

Near -synonyms cannot be substitutes for the components of a collocation. For
example, a collocation such as yellow wine used instead of white wine is not
acceptable, even though the adjective yellow provides the appropriate
description of the real color of the wine. Similarly, the adjective pure used in
the collocation pure chance cannot be replaced by the synonyms * spotless,
unsullied, unblemished, immaculate .
Such a feature accounts for the class of restricted collocations which are
defined both syntagmatically as “word -combinations in which one element has
a specialized meaning determined by the other element ” (Cowie and
Howarth 1996: 81, emphasis added).), and paradigmatically, in terms of
subst itutability, or recombinability. In Cowie’ s (1998: 15) opinion, restricted
collocations do not presuppose a mere learning of fixed units but knowing
when, and how far, the elements of a collocation are able to recombine with
other items.

2.1.2.2.3 . Non-modifiability

Collocations cannot be freely modified, either by introducing additional lexical
units, or by making grammatical transformations. For example, in a collocation
such as white wine the adjective white cannot be modified without discrediting
the functionality of the lexical pattern and excluding it from the class of
collocations. Lexical patterns such as * whiter wine or *the whitest wine are not

only lexically and grammatically incorrect, but also awkward to the speakers
familiar with the corr ect collocation.
Some authors have generalized the notion of collocation even further,
including cases of words which are strongly associated with each other, but
which do not necessarily occur in a common grammatical unit or observe a
strict word order (e.g. doctor – nurse or plane – airport ). Such examples
emphasize the necessity to restrict collocations to the narrower sense of
grammatically -bound elements that occur in a particular order and to use terms
such as association and co-occurrence.
Starting from the assumption that the conventionality of collocations
makes them linguistically significant, Howarth (1996 a : 6) lists the following
characteristics:
 they are conventional forms that have a significa nt role in language
production;
 they are memoriz ed as ready -made lexical units;
 they are stored together with some indication of their grammatical
structure and sy ntactic and pragmatic function’;
 they are recognized as familiar by native -speaker readers/hearers
both as regards their form and their asso ciated functions.
Although collocations have been extensively discussed and analysed in
various studies, there are many instances when speakers find it hard to establish
a clear -cut distinction between collocations and other lexical patterns (free
word co mbinations or idioms). Such confusions may be avoided by increasing
the speakers’ awareness with respect to the characteristics of collocations in
contrast with other word combinations.

2.1.3 . CLASSIFICATIONS OF COLLOCATIONS

Collocations may be classi fied according to different criteria . Certain
classifications envisage formal or structural aspects of collocations, whereas
others are centered on their semantic behaviour .

2.1.3.1 . FORMAL CLASSIFICATION OF COLLOCATIONS

Depending on the criterion taken into consideration, collocations fall into
different categories. Benson and Ilson (1986) distinguish two major classes of
collocations, namely: lexical and grammatical8.

8 Hausmann (1988) suggests a taxonomy similar to Benson’s, defining collocations as
”binary word combinations” (Hausmann 1988:63 -82) and explaining that the words
whcih collocate do not have the same status, since one of them conditions the other: „I

Typical lexical collocations are illustrations of “a relation of mutual
expectancy or habitual association” (Mc Arthur 1992, in Howard and Zé
Amvela 2000: 114), or of a partnership between ‘two “equal” lexical
components, consisting of nouns, adjectives, verbs, and adverbs , and not
normally containing prepositions, infinitives or clauses. I n lexical collocations ,
the occurrence of one word predicts the possibility that another word will occur
in the context, either in some syntactic construction , or across syntactic
boundary.
Seven classes of lexical collocations have been established:
 L1 collocations  verb (usually transitive) + noun/pronoun, or PP
(tv +N/Pron./ PP). Many of these collocations contain a verb denoting
creation and/or activation, from which they took the name of
“creation/activation collocations” (e.g. to come to an agreemen t, to
commit murder, to make an impression, to inflict a wound, etc.). The
same noun may collocate with one verb to denote both creation (e.g. to
establish a principle) and activation (e.g. to apply a principle) or the
united meanings (e.g. to impose an em bargo, to pose a question, to
commit treason ). Of course, not all L1 collocations denote creation or
activation (e.g. to do the laundry, to take one's seat, confirm a
suspicion) ;
 L2 collocations are in opposition to the creation/activation
collocations. Th e verbs of this category mean essentially eradication
and/or nullification, hence their name of EN collocations (e.g. to lift a
blockade, to dispel fear, to withdraw an offer, to revoke a license, to
ease tension, etc.);
 L3 collocations  adjective + noun (Adj. + N) (e.g. strong tea,
reckless abandon, a pitched battle, a rough estimate, a crushing defeat,
etc.). This category also presents nouns used as adjectives (house
arrest, jet engine, aptitude test) as well as clichés (sitting duck, long
shot, drug pu sher) ;
 L4 collocations  noun + verb, (N + V), the latter naming an action
characteristic of the person or thing designated by the former (e.g.
blizzards rage, bombs explode, blood flows, alarms go off, etc.);
 L5 collocations indicate the unit associated with a noun, often
having the structure noun + of + noun (N – of – N) (e.g. a pack of
wolves, a bouquet of flowers, an article of clothing, an act of violence,
etc.);

collocati non hanno lo stesso status, poiché uno di loro determina la scelta dell’altro. È
necessario distinguere quindi tra la base, la parola che determina con quale altra parola
puòn combinarsi e il co llocativo, l’elemento o gli elementi determinati” .

 L6 collocations  take the form of adverb + adjective (Adv. + Adj.)
(e.g. deeply absorbed , closely acquainted, sound asleep, very much
aware, etc.);
 L7 collocations  verb + adverb (V + Adv.) (e.g. affect deeply,
apologize humbly, appreciate sincerely, etc.).
Although numerous, the variety of lexical collocations is not as wide as
that of gram matical collocations , which are grouped into eight major categories,
further subdivided into 19 types.
Different from lexical collocations, g rammatical collocations are
phrases consisting of a dominant word, which can be a noun, an adjective, or a
verb, a nd a preposition or a grammatical structure, such as an infinitive or a
clause:
 Gl collocations  noun + preposition (N + Prep .) combinations (e.g.
a blockade against, a dread of, apathy towards, preference for, etc.);
 G2 collocations  noun + to infiniti ve (N + to- inf.) (e.g. It was a
struggle to do it. They had an obligation to do it. They felt a need to do
it. We made a promise to do it. He was a fool to do it. ). This category
does not include infinitives normally associated with the whole sentence
(e.g. She closed the window to keep the flies out ), infinitives that can be
replaced by a relative clause (e.g. a place to stay ) and nouns preceded by
a descriptive adjective (e.g. a clever thing to say );
 G3 collocations  noun + a that clause (N + that -clause) (e.g. He
took an oath that he would do his duty. ). The cases in which that can be
replaced by which (e.g. He made the same mistake that/which his brother
had made ) and those in which the nouns function as prepositional objects
followed by a clause ( e.g. It was by chance that we met) are not included
in this class;
 G4 collocations  preposition + noun combinations (Prep + N) (e.g.
by accident, in advance, on somebody's advice, in agony, at anchor, etc.);
 G5 collocations  adjective + preposition (Adj. + Prep) (e.g. fond of,
critical of, deaf to , etc. Such collocations occur either in the predicate
(e.g. They were angry at everyone), or as verbless clauses (e.g. My
friends, angry at everyone, stayed home ). Derived prepositions (e.g.
in/with regard to), and past participles followed by the preposition by are
excluded from this category;
 G6 collocations  predicate adjectives + to infinitive, either with the
empty subject it or with a real and usually animate subject (e.g. It was
necessary to work .). This category does not include past participles
followed by a to infinitive which express purpose (e.g. The text was
proofread to eliminate errors ), or participles used in passive
constructions (e.g. She was elected to serve as our delegate ). Collocations

with dummy “it” can replace the infinitive with an -ing form (e.g. It’s
nice working here .);
 G7 collocations  adjectives + a that clause (Adj. + that – clause)
(e.g. She was afraid that she would fail the examination , It is necessary
that he be replaced immedi ately .);
 G8 collocations are further divided into nineteen patterns, some of
them being interrelated:
 G8.1. patterns which allow the dative movement transformation,
meaning the shift of the indirect object to a position before the
direct object: (e.g. He s ent the book to his brother  He sent his
brother the book .);
 G8.2. transitive verbs which do not allow the dative movement
transformation when they have an indirect object ( He screamed
something to her );
 G8.3. transitive verbs used with the preposition for allowing the
dative movement transformation: e.g. She bought a shirt for her
husband  She bought her husband a shirt). The pattern applies
mainly to verbs denoting ‘making’ or ‘creation’ ( slice, boil, bake,
brew, etc.);
 G8.4. verbs forming collocations with a specific preposition (+
object): (e.g. to serve as, to treat as, to consist of, to adhere to, to
base (conclusions , assumptions , etc) on);
 G8.5. verb followed by to infinitive (e.g. They began to speak. ).
Collocations of this type are some of the le ast conspicuous to
correspond to any collocatory type because they do not require
either the use of special prepositions, or a certain order of objects.
However, this pattern does not include structures expressing
purpose, such as (e.g. They were running t o catch the train. );
 G8.6. modals followed by a bare infinitive (e.g. We must work
harder. ). Other two modal phrases, namely, had better, and would
rather are also included in this category.
 G8.7. verbs followed by a second verb in –ing form , having
either a synonymous or a completely different meaning from that
of the verb + to infinitive verb pattern (e.g. She stopped chatting
with her friends. as opposed to She stopped to chat with her
friends. ). (infinitive clause of purpose);
 G8.8. transitive verbs fol lowed by an object and to infinitive (e.g.
He invited me to participate. ). There are, however, a series of verbs
which do not fit this pattern: to bring, to cause, to commit, to
intend, to like, to prefer, to telephone, to thank, to write, to wish,
etc.;

 G8.9. transitive verbs followed by a direct object and a bare
infinitive (e.g.: She heard them leave. ) Passivization of such
structures leads to the change of the short infinitive into the long
infinitive (e.g. They made us get up – We were made to get up. );
 G8.10. verbs followed by an object and an -ing verb form: (e.g.
He kept me waiting two hours. );
 G8.11. verbs followed by a possessive pronoun or noun, or by a
gerund (a verbal noun) (e.g. Please excuse my waking you so early .
This fact justifies Bob’s co ming late. ). This category of
collocations may be replaced by virtually synonymous
constructions when two objects are followed by a conjunction (e.g.
I cannot imagine Bill and Mary's doing that vs. I cannot imagine
Bill and Mary doing that. );
 G8.12. verbs followed by a noun clause introduced by that: (e.g.
They admitted that they were wrong. );
 G8.13. transitive verbs followed by a direct object, the infinitive
of the verb to be , and an adjective, a past participle, or a
noun/pronoun (e.g. We found the roads to be excellent./ cleared of
snow ./ a serious problem for the state treasury .);
 G8.14. transitive verbs followed by a direct object and an
adjective, a past participle or a noun/pronoun (e.g. He made his
meaning clear. ). Some verbs included in this patter n are used only
with certain adjectives (e.g. to paint the walls blue/green, to shoot
somebody dead. );
 G8.15. verbs which can take two objects, neither of which can
normally be used in a prepositional phrase with to or for (e.g. The
teacher asked the pupil a question. ). Certain verbs pertaining to
gambling (e.g. lay, bet, wager) may take three objects – a person,
an amount, and a clause (e.g. We bet him ten pounds that the hosts
would win. );
 G8.16. verbs, irrespective of their type (intransitive, transitive or
reflexive) followed by an adverbial (e.g. He carried himself with
dignity. The meeting will last two hours. They tramped through the
forest. );
 G8.17. verbs followed by an interrogative word (how, what,
when, where, which, who, why and whether) which ca n, in its turn,
be followed either by a to infinitive construction or by a clause:
(e.g. He asked how to do it. She could not decide whether she
should begin. She knew when it was best to keep quiet. );

 G8.18. verbs (often expressing emotions) preceded by t he
dummy it and followed by the to infinitive or by a that clause (e.g.
It puzzled me that they never answered the telephone. );
 G8.19. a small number of intransitive verbs followed by a
predicative expressed by a noun or by an adjective (e.g. She
became an engineer. He will make a good teacher. The flowers
smell nice. ).
Other lexicographers and linguists (Cowie 1981, Allerton 1984) refer to
grammatical and semantic collocations. The former class often contains
prepositions, including syntactic categories su ch as verb + preposition, adjective
+ preposition and noun + preposition. The open class word is called base and
the words it can collocate with, the collocators . The latter class of semantic
collocations includes lexically -restricted word pairs where only a subset of the
synonyms of the collocator can be used in the same lexical context.
Considering formal aspects, Hill (2000, in Deveci 2004: 17) combines
lexical and grammatical collocations and refers to the following possible word
combinati ons (Hill’s examples) :
 adjective + noun : a huge profit.
 noun + noun: a pocket calculator.
 verb + adjective + noun : learn a foreign language.
 verb + adverb : live dangerously.
 adverb + verb : half understand.
 adverb + adjective : completely soaked.
 verb + pr eposition + noun : speak through an interpreter
The presence of a noun or a verb in the structure of such lexical patterns
allows a further classification of collocations into nominal and verbal (Pârlog
and Teleagă 2000: 9). Reference was also made to lexi cal patterns in which the
key element is an adjective (e.g. white yellowish, weak in the head ), an adverb
(e.g. backwards and forwards ) or a numeral (e.g. three times a week ) (Avădanei
2000: 51).
Considering the varying length of word combinations, Avădane i (2000:
33) distinguishes two classes of collocations. The former class is that of short
collocations , represented by combinations of two or more words such as: sea-
horse, to boil in one’s sew . The latter, i.e. the class of long collocations,
includes lon ger structures, such as clauses or even sentences (e.g. all cats are
grey in the dark, if Mohamed will not come to the mountain, the mountain must
go to Mohamed .).
Sticking to the same criterion, Avădanei (2000: 48) also analyzes the
formal cohesion existing in the case of such lexical structures. She considers
that certain word combinations are fixed , others are semi -mobile , and still
others are mobile . In the first case, the degree of cohesion between the
constituent elements is maximal (e.g. to put the cart before a horse .). The

second type of word combinations allows the substitution of one of the
component words with a synonym and is, thus, represented by a more reduced
formal cohesion (e.g. to crack/ break a bottle ). Finally, in the case of mobile
lexical structures, the degree of cohesion is much more reduced because two, or
even more elements, can be replaced by other synonymous words (e.g. to
swim/go against t he stream/ tide ).
The restrictedness of collocations is an important criterion used in
classifying them. Mention should be made that this characteristic is attributed to
the collocate , not to the base word, and is defined both syntagmatically, in
terms of semantic constraints on the meaning of the collocate, and
paradigmatically, in terms of lexico -grammatical limitations on the
substitutability of the collocate.
Categorization is based on distribution in terms of limited and arbitrary
variability. Sets of ‘overlapping collocations’ (Cowie 1986: 64) are used as
evidence of the arbitrary and restricted nature of collocations .
Howarth (1996 a: 44) argues that “if it can be shown that certain potential
collocations in the cluster are arbitrarily blocked, those collocations that are
acceptable could be regarded as restricted”. This view is not shared by
Langacker, who claims that such distributional patterns are not necessarily a
good basis for categorization, as they may reflect ‘extrinsic’ factors rather than
properties that are ‘intrinsic’ to a word combination (Langacker 1987: 306).
The classification of collocations in terms of their collocational
restrictedness results in four classes (Carter 1998: 70):
1. Unrestricted collocations , including lexical items whi ch are open to
partnership with a wide range of items (e.g. take a look/ a holiday/ a rest/
a letter/ time/ notice/ a walk );
2. Semi -restricted collocations – lexical patterns in which the number
of substitutable items in different syntactic slots is more det ermined (e.g.
to fan a riot/ discontent/disturbance/ hooliganism ) ;
3. Familiar collocations – combinations of words which keep regular
company with each other. There are obvious overlaps here with types of
fixed expression categorized as stock phrase or me taphoric (e.g. vicious
circle, innocent bystander, amicable divorce );
4. Restricted collocations including more fixed and closed partnerships
between lexical items (e.g. stark naked, pitch black, soft water, accept
defeat ) and irreversible binominals (e.g. cash and carry, ups and downs,
hit and miss, assault and battery ).
The cline in collocational restriction runs from less fixed to more fixed .
This notion proves extremel y useful in further ranging lexical patterns in terms
of sets of continua with fixed p oints and several intermediate categories:
 Syntactic structure – the cline runs from less (a) to more fixed (c).
a) Flexible : break somebody’ s heart.

b) Regular with certain constraints : to smell a rat, to drop a
brick.
c) irregular : to hold true, to be good frie nds with somebody, the
more the merrier.
 Semantic opacity – the cline runs from less (a) to more fixed (d):
a) Transparent : long time, no see, honesty is the best policy .
b) semi -idioms/ metaphor/ idiomatic similes : we’re all in the
same boat, a fat salary, as sober as a judge .
c) Semi -transparent : bumper to bumper traffic, a watched pot
never melts.
d) Opaque : 1) overt (uninterpretable without contextual/ cultural
knowledge): bottoms up ; 2) covert : to kick the bucket, to carry
the can .
Contrasting ‘infelicitous’ co llocations with correct ones, Martin (in Cop
1990: 36) considers that the latter are “word combinations in which the
members are attracted to each other like two opposite charges”. Such word
combinations bear various names: strong or closed collocations (Sinclair
1987), settled combinations (Cowie 1986), fixed or recurrent combinations
(BBI 4) and restricted combinations (Mel’čuk 1984).
The restrictedness of collocations seen from the perspective of the
frequency with which certain words co -occur with others is also referred to by
Lewis (1998 in Deveci 2004: 17) who proposes the following types of
collocations:
1. Strong – including numerous combinations which are strong or very
strong. For example, rancid is commonly associated with butter , but
this does not mean that it cannot be associated with other words, as
well.
2. Weak – including words which co -occur with a greater t han random
frequency. Adjectives such as short, cheap, expensive, good, bad may
collocate with different nouns. However, certain combinations are more
predictable: red wine or white wine .
3. Medium strength – including words that go together with a greater
frequ ency than weak collocations: hold a meeting , carry out a study .
Cowie (1986) distinguishes between ‘ free’ collocations and
‘restricted’ collocations. The former are combinations in which one word is
open to partnership with a wide range of lexical items. (hire + staff, clerk,
secretary, worker ). (Fontenelle 1998: 191) Restricted collocations , on the other
hand, are word combinations in which one element has a technical sense, or a
long-established figurative sense which has since lost most of its analogic al
force. (e.g. run a deficit, abandon a principle, champion a cause ). The class of
restricted collocations also includes ‘support verb constructions’ (Gross 1981),
that is constructions in which the role of the verb is limited to ‘supporting’ the

direct o bject with which it co -occurs, thereby establishing a link between the
object and the subject and conveying information on number, tense and aspect
(e.g. to have a bath, to take a decision ).
Last but not least, the restrictedness of collocations is also r eferred to by
Gramley (2001: 74), who, starting from the idea that some collocations do not
allow much freedom, distinguishes the class of unique (fixed, frozen)
collocations, i.e. lexical patterns in which one (or sometimes both) of the words
cannot be re placed (e.g. addled eggs, to peel onions ). He admits that many
combinations are fairly free without being unrestricted (e.g. reek beer, wine,
whisky ). Adding that numerous, if not most, collocations are the same in most
varieties, he also points out that t here are quite a few differences, as well. He
clears up his statement explaining that if one drinks too much, he may throw up
(shared), upchuck (AmE; mild), puke his guts out (Am E: strong) or bring
everything up (BrE), all expressions being used with the meaning ‘to be sick at
the stomach’.

2.1.3.2 . SEMANTIC CLASSIFICATION OF COLLOCATIONS

According to the semantic criterion, reference can be made to three types of
nominal collocations (Kavi ć 2005: 338 -339):
1. Quantificational collocations (N-of-N) used with an implied
semantic component that determines and restricts the use of a particular
collocation; the meaning is: a large number of, a large quantity of (e.g. a
shoal of fish , a herd of wha les – a large number of fish / whales , a
draught of fish – a large number of fish taken in one drawing of the net);
2. Quantificational plus used with an additional, implied meaning
which does restrict the use of collocation; their meaning refers to motion,
manner of motion, behaviour, because they express a distinctive quality
(e.g. a hover of trout – the ability of trout to jump over steep rapids in a
river);
3. Meaning relations between N 1 and N 2, the preposition being
different even with the same target nou n (e.g. a breach of agreement , a
puff / rush of + (adj.) air, lack of confidence , a boom in demand , a boost
to the economy , a stand against the enemy , an increase in expenditure ,
the outlook for the future , a rise in the output , the fight against
unemploym ent, a drop / fall in unemployment ).
Such a classification considers all the possible meanings of the N 1
component of N 1-of-N2 collocations, all the possible meanings of the N 2
component , as well as the logical relations between possible meanings of the N 1
component and the characteristics of what is designated by the N 2 component
(i.e. a living being, a lifeless object, a mass noun, etc.).

Taking into consideration the same criterion, the following collocational
patterns can be distinguished:
 Phrasal verbs : verb combinations in which each word contributes
something, semantically recognizable, to the meaning of the whole; in
some cases, it is mainly the verb, and in other cases it is mainly the
particle that is prevailing in stating the meaning of the whole combination
(e.g. get / come along + with );
 Adj. + N .: a) co -selection and shared meaning with the N; b) overlap
of selections; c) partial emphasis of the meaning by the adjective (e.g.
scientific assessment / analysis / study / experiment .);
 “Fixed phrase s”: variable phrases built round a slightly specialized
meaning of a word that goes with a specific grammatical environment
and in regular collocations; they show co -selection and shared meaning
(e.g. a piece / item of information , a word of advice , the pr ospect of an
agreement , a breath / draught / gulp / sniff / whiff of air , the erosion of
confidence , a collapse / slump in demand , a stand against the enemy ,
limitations on expenditure , a fall in the output , the rate of unemployment ,
a (high) incidence of unemployment );
 Idioms : there is no interpretation based on the “core” meanings of the
two words; interpretation is based on metaphorical extension (e.g. the
naked eye ).
Pustejovsky’s Qualia Structure Theory (1995 in Schroten 2000: 32) is
also worth mention ing here, as it is essential for the compositional interpretation
of collocations. According to this theory, lexical items are associated with
atoms of meaning which can be accessed by other lexical items. ‘Qualia’ are a
specific kind of semantic features which allow the speakers to derive
appropriate meanings of collocations compositionally. Pustejovsky (1995: 85 –
86, in Schroten 2000: 33) refers to the existence of four qualia, which may be
interpreted as follows:
1. Constitutive – the relation between an o bject and its constituents or
proper parts (material, weight, parts and component elements)
2. Formal – identification of the object within a larger domain
(orientation, magnitude, shape, dimensionality, colour, position)
3. Telic – purpose and function of the object (purpose that the agent has
in performing an act, build -in function or aim which specifies certain
activities)
4. Agentive – factors involved in the origin or “bringing about” of an
object (creator, artifact, natural kind, causal claim).
An appropri ate formal representation of the qualia features will provide a
formal way of deriving the semantic interpretation of collocations by accessing
semantic features of their constituent parts and by combining them
compositionally.

Martin (1992, in Carter 1998 : 82) suggests another semantic
classification of collocations. He refers to four possible classes of collocations,
namely: componential (e.g. tall buildings, high mountains ), modificational
(e.g. bright sun, hot sun ), resultative (e.g. bomb explosion , gas fire), and
utilitarian (e.g. framing hammer ).
Numerous and various as they may be, the above mentioned approaches
are not always relevant enough to help linguists distinguish collocations from
other lexical patterns. The fact that collocation “simultaneou sly foregrounds the
significance of co -occurrence relations in establishing recurrent lexical patterns
as well as the routinized nature of language” (Fernando 1996: 29) represents
perhaps the most distinctive aspect characterizing the members of this lexic al
class.

2.1.4 . COLLOCATIONS VS. ‘COLLIGATIONS’

Although virtually identical, the terms collocation and ‘ colligation ’ are clearly
partial synonyms. The former refers to the lexical associations of nodes,
whereas the latter is used to refer to the sam e kind of co -occurrences between
node words and grammatical classes (Sinclair 1991: 170, Jantunen 2004: 105).
Colligations may also be regarded as interrelations of grammatical
categories, which concern categories such as word classes and sentence classes
(Firth 1968 b: 181).
Present -day corpus linguistics considers ‘ colligation’ to be an association
of a word looked upon as a unique lexical item rather than as a member of its
class (Tognini Bonelli 1996: 74) with grammatical categories (Sinclair 1988), or
with a particular position in a sentence or text (Kennedy 1991, Hoey 1997).
The change from collocation to colligation may be illustrated by means
of lexical patterns such as: a) … see with the naked eye ; b) … just visible to the
naked eye ; c) … form a nake d-eye pair ; d) … too faint to be seen with the
naked eye ; e) … they could be seen with the naked eye from the helicopter . In
these examples, there are possible patternings to the left of the NP the naked eye
with the prepositions with and to (see examples a) and b)), which are inherent
components of the phrase. The change from collocation to colligation is
marked by the co -occurrence of grammatical choices to account for variation.
The phrase can also occur as subject or object of a clause, where a prepositio n
would be wrong (see example c)). Such collocations are restricted to verb and
adjective and are controlled by semantic preference. Moreover, there is
correlation between the semantic choice (“visibility” choice) and the preposition
choice, this aspect be ing illustrated by the fact that adjectives take the
preposition to, whereas verbs take the preposition with. Seen from this
perspective, the colligation is a “collocation with the preposition that collocates

normally with the chosen verb or adjective in t he chosen construction” (Sinclair
2004: 33).
Moreover, colligations are integrated in the larger frame of co-selection
categories of the lexical item. Out of the five categories making up this frame,
two are obligatory and three are optional. The obligator y categories are
invariable and constitute the evidence of the occurrence of the item as a whole
and of its meaning, whereas the optional categories realize co -ordinated
secondary choices within the item, adapting the meaning and giving semantic
cohesion t o the text as a whole. At the same time, the optional categories serve
as a means of classifying the members of a paradigm, thus explaining the
relationship between the two axes of patterning.
‘Colligation ’ and collocation are two of the members making up the
class of obligatory co -selection categories . Collocation includes the co –
occurrence of words with no more than four intervening words, being the
simplest and the most obvious relationship on the syntagmatic axis. (Sinclair
2004: 141). As regards the p aradigmatic dimension of collocation , the approach
is different, because items can only collocate with each other when present in a
text, and two items in a paradigm are mutually exclusive. The relationship
between the items placed on the paradigmatic axis is called mutual collocation .
This means that they both collocate, usually on separate occasions, with the
same item or items. For example, manual and restoration are both significant
collocates of work , but they themselves do not co -occur significantly.
On the other hand, ‘colligation ’ refers to the co -occurrence of
grammatical phenomena, being syntagmatically defined as the co -occurrence of
a member of a grammatical class, for example a word, with another word or a
phrase. Similarly to collocation, the p aradigmatic approach to ‘colligation ’
envisages a mutual relationship. For example, a possessive adjective may
colligate with a particular noun (e.g. your true feelings ) and the periphrastic
construction the…of … (e.g. the true feelings of the people ) may occasionally
occur as an alternative.
Instead of a conclusion, mention should be made that collocations and
colligations are characterized by two conflicting principles: the OCP and the IP
(Sinclair 1987, 2001, 2004), the former generating the terminologi cal
tendency and the latter the phraseological tendency .

2.2 PSYCHOLINGUISTIC APPROACH ES TO COLLOCATIONS

Psycholinguistics is the study of normal and abnormal use of language and
speech to gain a better understanding of how the human mind functions (S covel
1998: 129). As regards the scope of psycholinguistics, studies in the field are
mainly concerned with the ways in which expert speakers of a language store

and retrieve the language system, and learners (of a first or second language)
acquire the lan guage. To be more precise, psycholinguistics analyzes the way
in which certain psychological parameters influence the structuring and
reception of messages, the role of the situational components, the speakers’
perception of the linguistic code they use, t he factors influencing the forms of
communication, as well as the effects of certain pathological problems on
communication:
“Psiholingvistica analizează modul în care gândirea, memoria, afectele,
temperamentul intervin în structurarea și receptarea mesaj elor, rolul
componentelor situaționale, modul de percepere a codului lingvistic de către
cei care îl utilizează, factorii care influențează, în decursul timpului, formele
activității comunicative a unui individ (educația, cultura, bilingvismul,
profesia, e tc.), efectele unor stări patologice asupra activității communicative,
etc.” (Bidu -Vrănceanu et al. 2005: 419).
From this perspective, it becomes obvious that the psycholinguistic
approach to collocation s has to focus on the processes involved in decoding
multi -word patterns and the aspects which may facilitate their appropriate use
in the case of both intra- and interlingual communication. However, the starting
point in this approach should be the psychological interpretation of single
words.
Psycholinguis ts consider that any lexicon has two components, namely
word meanings and stored word forms , which interact by means of the access
paths that allow these components to communicate with each other and with
other superordinate elements.
The correct decoding of word meanings is influenced by a series of
aspects. On the one hand, in order to interpret words in the lexicon, they have to
be accessible both in written form and in speech. In addition, one must keep in
mind that language production and reception ar e conditioned by the systematic
storing of words in such a way that they can be searched for, either by their
properties as message elements, or by their linguistic form. Specialists agree
that, in order to decode messages, speakers have to access three files:
orthographic , phonological , and semantic -syntactic , which are linked to a
unitary file called master file .
The master file is more than an inventory of fully specified forms. It
includes the inter -relationships between the constituent items and may be
referred to when approaching the concept of word associations. “When an entry
from the master file is accessed, it is available for implementing any sort of
response: speaking, writing and understanding the word. High frequency words
will be accessed fast er than low -frequency words” (Garman 1991: 267).
The situation is quite different in the case of modality specific word
forms which are highly compatible with the systems that perceive and produce
them. As Garman states, such lexical units require more sto rage space in the

lexicon, “both for more complex form -specifications and for the links between
them” (Garman 1991: 244).
At this point, the idea should be underlined that the arbitrariness of words
is an important factor conditioning the process of commu nication. Words have
a meaning of their own which is encoded in a sound or letter structure, and this
meaning is in most cases stored in the mental lexicon of the speaker. However,
there are cases in which the meaning of certain simple words is not accessi ble to
speakers. Form a linguistic point of view, this means that speakers cannot make
connections between the signifier and the signified. Psycholinguists consider
that in such instances there is no other way of recovering the meaning of the
word from its structure, but by making language specific acquired associations.
Referring to the mental lexicon or semantic memory Gramley (2001: 7 –
8) states that it “includes the speakers’ encyclopedic knowledge of the world”.
He adds that items are unlikely to be st ored alphabetically, and that categories,
or fields, of words and things are more likely to be built up by associating
linguistic with non -linguistic knowledge. Moreover, since within the fields,
speakers make finer and finer distinctions, these fields may be organized as
networks of overlapping similarities in meaning, as hierarchies of features of
meaning, or they may be grouped around prototypical concepts. Sometimes
network organization is made by some combination of these.
A solution to such problems may be that suggested by Meyer and
Schvaneveldt (1971 in Garman 1991: 269) who consider “that the time taken to
classify a real word as such decreased if it was preceded by a real word that
exhibited some semantic relationship with it”. In support of their opinion,
Meyer and Schvaneveldt considered a pair of two semantically related words,
namely doctor and nurse . By comparing this pair to the semantically unrelated
words nurse – table , they proved that the cross reference made in the speakers’
minds reduced the response latencies for the term nurse in the case of the
former pair. Once the lexical entry doctor is accessed in the master file ,
speakers no longer have to go back to the orthographic access file for accessing
the item nurse . The system can allow f or immediate transfer to the entry for
nurse in the master file along the search path formed by the semantic cross –
reference between the two items.
Garman (1991) considers that “accessing words is a matter of mapping
ideas onto those stored meaning repres entations in the mental lexicon that are
associated with stable word forms , which can then be used to implement a
spoken or written output”. The length of the linguistic structures used by
speakers may vary from words to higher -order constructions such as phrases
and clauses , and this is an aspect which explains the various complexity degrees
of the psychological processes involved in their decoding: “processing longer
sequences will require a number of lexical access operations plus some
computation of th e relationships holding between the constituents”. Moreover,

the length of such sequences is considered to be “inversely proportional to their
internal stability and frequency of occurrence” and the longer the sequence, “the
less beneficial is to have it s tored as a unitary entry in the mental lexicon”
(Garman 1991: 240 -241).
Collocations and idioms are good examples of long lexical sequences. If
the former may exist in the mental lexicon as sequences of individual entries
which have particularly strong ass ociative links holding between them, the latter
are likely to be represented as unitary, to some degree, in their entries.
In discussing collocations from the perspective of psycholinguistics , it is
interesting to consider the mechanisms which affect lexic al processing. As
suggested by Scovel (1998) , “the comprehension of words is much more
complex than the processing of phonemes”. This is due, on the one hand, to the
fact that
“words are made of more sounds, these sounds being written in different,
incon sistent ways in different languages”, and, on the other, to the fact that “the
vocabulary of any language is made up of thousands of words which, in their
turn, convey different meanings” (Scovel 1998: 55).
The frequency effect is a useful psycholinguisti c concept by means of
which the appropriate use of collocations, either in the native language , or in a
foreign language may be explained.
The length of collocations is a parameter as important as their frequency
of occurrence . It is generally accepted th at collocations are an illustration of
the habitual use of two or more words together in a fixed pattern which is
recognized and used as such in language. This explains why a psycholinguistic
concept such as the frequency effect becomes extremely useful in explaining
why the words or the lexical patterns used more frequently by speakers are
easier to remember and to decode.
In psycholinguistics, the frequency effect is a concept assumed to
operate only on the first item of a pair of words. Starting from the idea that there
are different word pairs according to whether both items are of high frequency
(H-H), or of low frequency (L -L) or mixed (H -L or L -H), Forster (1976, in
Garman 1991: 269) made an experiment meant to demonstrate the degree to
which the freq uent use of a word facilitates the processing of other words, either
semantically related, or unrelated. The results were quite interesting and useful,
although not totally unexpected. In the case of H-L pairs (made up of words
semantically related) the in formation is processed as fast as in the case of H-H
pairs. Moreover, L-H pairs (with words semantically related) are decoded as
fast as the L-L pairs. Finally, in th e case of unrelated word pairs ( the mixed
types ) the processing time was somewhere in bet ween H-H and L-L types. The
access file may also be used for the second item in a pair. The processing time

in the case of the pairs which have one low -frequency item proved to be slower
than in the case of those which do not.
Although this method of cros s-reference investigation cannot be taken
into consideration when analyzing the semantic relationship between the
constituents of a collocation, it may become extremely useful in explaining
from a psycholinguistic point of view, semantic relationships such as:
synonymy , antonymy , and hyponymy (see 3.3.3.1 . – 3.3.3.3 .). To put it
differently, this method “is advisable if the task is that of establishing semantic
relatedness between the items making up the pair” (Garman 1991: 270).
If the theories and concep ts referred to so far focused on word
recognition in isolation , mention should be made that special attention has also
been devoted to the psycholinguistic aspects related to word recognition in
context.
The first specialist to highlight the importance of word recognition in
context was Morton. In his opinion, a word may be recognized more easily in
context than in isolation. In order to demonstrate this, he considered it necessary
that distinction should be made between the context of the word and the
context of the utterance .
Word recognition in context has been explained by means of two
models of psycholinguistic analysis: namely the logogen and the cohort
model s.
The logogen model (Morton 1969, 1979, 1980 in Garman 1991: 278,
Scovel 1998) presupposes v isual word recognition and has the logogens and the
logogen system at its basis.

Fig. 5 . The main components and relationships of the logogen model
(source Garman 1991: 278)

As suggested by Morton (1979: 112 in Garman 1991: 279), logogens
may de defined as those “tuned perceptual devices that respond to sensory and
semantic input”. In other words, the sensory and contextual inputs interacting in
the logogen system are those which give rise to outputs to the cognitive system,
as well as to the response channel. The logogen system is basically tuned to the
auditory and/ or visual properties of words and of their contexts of occurrence,
there being no two -way connection between this system and the cognitive one.
This means that what happens in the logogen system at any time forms part of
the output to the cognitive system and this, together with other types of
knowledge, can be fed back into the logogen system to help regulate its
performance.
Another important aspect which shoul d be mentioned is the fact that
“evidence about the occurrence of a certain word comes potentially from all
modalities, and these inputs are in a ‘conspiratorial’ relationship with one
another. Deficiencies in one sort of input can be made good by referenc e to the
information available in others, and they all combine to lower the recognition
threshold of the relevant stored forms.” This means that “the word detector
system does not simply rely on specific linguistic -signal properties, but it is also
able to pick up cues to the presence of a word from all available input channels
and sum their effects”(Garman 1991: 280).
The cohort model (Marslen -Wilson and Welsh 1978 in Garman 1991:
281) is similar to Morton’s logogen model , but the difference between them l ies
in the fact that the cohort model is used to explain spoken, not written word
recognition in context.
Although Marslen -Wilson and Welsh’s (1978) findings suggest that
context plays an important role in the word -detection process, and Morton’s
model ex plicitly allows for this, specialists have not demonstrated whether the
input from the cognitive system to the logogen directly operates on the stored
lexical items, or whether it indirectly enhances the effectiveness of these other
types of input.
Word re cognition in context may be explained by means of either lexical
priming or visual masking. The former procedure presupposes disambiguation
of a word by means of other words from the peripheral field. For example, a
noun such as palm may be disambiguated b y means of peripheral words such as
tree or hand . The latter, i.e. the visual masking, “presupposes sequential
presentation of words in isolation where the exposure of the first word is very
brief and is immediately followed by a pattern mask, i.e. a grid of lines which
activate the same areas of the retina as the letters of the immediately preceding
word”(Garman 1991: 293). In such cases, the speaker’s lexical decision on the
second word is facilitated if the two words are semantically related. This

semant ic priming effect may be observed even when speakers cannot report
the existence or identity of the first masked word.
As regards collocations , I agree with Hoey (2005) who suggests that in
analyzing this type of fixed lexical patterns, the term lexical p riming should be
replaced by a more specialized term, i.e. collocational priming . He explains
that this type of priming is “sensitive to contexts (textual, generic, social) in
which the lexical item is encountered” adding that it is part of speakers’
“know ledge of a lexical item that is used in certain combinations in certain
kinds of text ” (Hoey 2005: 10, emphasis added).
Finally, Garman refers to subliminal automatic lexical priming which
appears to be based on lexical network relationships and is relativ ely unaffected
by the size of the set of paradigmatic alternatives to the primed word. For
instance, the priming effect in bank + mask + money is unaffected by whether
bank is preceded by river or day or save. As a priming word, bank is rather
ambiguous. I t can be active on money only if the appropriate semantic reading is
activated. Moreover, supraliminal conditions , where masking is not involved,
presuppose lexical filtering (Garman 1991: 295):

 river + date
 river  reaction time slower than in 1
 day + bank  reaction time faster than in 1
 save  reaction time faster than in 1 – the fastest due to the
semantic reinforcement from save.
To conclude , research into the comprehension of words has shown that
speakers are very much affected by context , and that understanding is both
facilitated and complicated by the different pieces of knowledge that speakers
possess for each logogen .
Speakers “have access to a dictionary -like memory for words in which
they can search for partially -remembered words by comparing and contrasting
other words which share similar specifications”, but the fact should not be
neglected that the speakers’ “knowledge of and about words is much more
extensive: the meaning of a word immediately triggers a spreading activat ion
of associations which help speakers understand it in many different
contexts , and may bring other related words to mind” (Scovel 1998: 68,
emphasis added).

2.3 SOCIOLINGUISTIC APPROACH ES TO COLLOCATIONS

Studies in language, culture, and communicati on often devote special attention
to sociolinguistic aspects. This is due to the fact that such aspects condition

communication to a great extent being thus essential in the creation and
tailoring of messages.
Although collocations have hardly been appr oached from a
sociolinguistic perspective, I consider that this perspective should not be
neglected, as it may reveal useful insights in the ways in which social aspects
condition the appropriate decoding and encoding of collocations.
In describing colloc ations , specialists agree that there is an obvious
stratification of collocations according to different social and professional
criteria. This explains the necessity to study the ways in which the appropriate
use of collocations both in intra – and interli ngual communication is influenced
by the speakers’ membership to , or acceptance in a certain speech or discourse
community .
Moreover, sociolinguistic approaches to collocations are also useful in
second language research and teaching. As suggested by Rann ey (1992) , second
language researchers and teachers have
“to discover what a learner needs to know in order to communicate in actual
social contexts in the setting of a new culture. Sociolinguistic skills include the
ability to perform various speech acts , the ability to manage conversational
turns and topics, sensitivity to variation in register and politeness and an
understanding of how these aspects of language vary according to social
roles and settings ” (Ranney 1992: 25, emphasis added).
Learners a lso “need consider able cultural information about
communicative settings and roles” ( id: 45).
The sociolinguistic approach to collocations is concerned with
discovering patterns of linguistic variation which are derived from differences
in speech situation s, and from social dist inctions within a community which are
reflected in the communicative performance.
Referring to the scope of sociolinguistics , Cook (2003) states that
sociolinguistics
“focuses upon the relation between language and society. It ende avours to find
systematic relationships between social groupings and contexts and the variable
ways in which languages are used” (Cook 2003 : 9).
Bidu -Vrănceanu et al. (2005) define sociolinguistics as “the
interdisciplinary study of the systematic correlative variation in the structure of
a language, on the one hand, and in the structure of the commu nity using it, on
the other “(our translation). Linguist ic variables cannot be mechanically linked
to certain social variables, the use of language being only a diagnostic
parameter of the social structure and of the interactions within a given
community:
“Variabilele lingvistice nu pot fi însă corelate mecani c cu anumite variabile
sociologice, ci este necesar să se demonstreze relevanța în plan lingvistic a

fiecăreia dintre variabilele sociologice considerate. Modul de folosire a
limbii reprezintă un indice diagnostic al structurii sociale și al proceselor
de interacțiune specifice unei comunități , pentru că selecția formelor
lingvistice utilizate de vorbitori are la bază o informație sociologică empirică,
inclusiv acea ierarhizare a variabilelor, consensual admisă într -o anumită
comunitate” (Bidu -Vrănceanu et al. 2005 : 490, emphasis added).
A basic assumption in sociolinguistics is that two complementary
processes operate in the dynamic connection between language and social
factors. From one point of view, social differentiation among people is
correlated with differences in their speech, and, from the other, divergence in
the way language is used is a gauge of social segmentation. Factors such as
gender, age, class, religion, race, ethnicity, and occupation frequently account
for linguistic differences. Interr elationships between societal factors and
language use are extremely complex for several reasons. On the one hand,
sociolinguistic behaviour varies according to the options available in a certain
community such as alternatives of pronunciation, vocabulary and sentence
construction. Certain options are made in a particular speech situation.
Therefore, sociolinguistic “rules” are actually statements of probability , rather
than rules which can predict any single speech occurrence. Both individual and
societal patterning is specific to certain situations. Moreover, individuals are not
isolated from social factors. Choices in speech style are motivated by many
aspects of one’s identity. Sociolinguistic studies consider the ways that
specific attributes influence a speaker’s selection in any given situation .
(Bonvillain 2003: 4, emphasis added)
Reference should be made to the Speech Accommodation Theory
(SAT ) formulated by Giles in 1973. This theory is often quoted by the
specialists in the field due to the explici t suggestion made, namely that “the
speaker’s choice of a prestigious or a non -prestigious speech style need not be
the result of his/her social class position or the formality/informality of the
speaking context . Rather, it could be mediated by interperso nal accommodation
processes”. To put it differently, speech variation should not be viewed as
determined by the social context, but more as a speaker’s dynamic and
subjective response to the addressee. The degree of matching between speakers
has to be anal yzed and linked to social and psychological factors which can
explain and predict it. The positive matching process was called convergence –
‘a strategy whereby individuals adapt to each other’s communicative behaviours
in terms of a wide range of linguist ic/ prosodic/ non-verbal features’ (Giles et al
1991: 7 in Spencer – Oatey 2000: 193). The psychological process at the heart of
convergence is ‘similarity attraction’ (Bryne 1971). Speakers who want to
cooperate and who want to be approved of will tend to converge.
Correspondingly, when a speaker becomes more similar to a listener, it is more
likely that the listener will approve of him more strongly. These tendencies

explain the strategic use of language codes and communication styles. Codes
and styles do not merely vary with social groups and social situations. Rather,
code -choice and style choice are sociolinguistic strategies used in order to
achieve the social and relational results they want (Spencer -Oatey 2000: 193) .
Maintenance of codes is also an interesting sociolinguistic concept used
to identify a speaker’s or a group’s option to modify or not their communication
relative to the addressee. The opposite concept is divergence of codes which
refers to the way in which speakers accentuate speech and non-verbal
differences between themselves and the others (Giles et al 1991: 8 in Spencer –
Oatey 2000: 194). From a psychological point of view, a person’s integrative
orientation to others is regarded as psychological convergence, whereas the
desire of co mmitment to achieve greater distance and distinctiveness is referred
to as psychological divergence. Convergence and divergence through selection
of a language code is the most obvious accommodation strategy in multilingual
settings.

2.3.1 . COMMUNITIES O F LANGUAGE USERS

Sociolinguistic studies find it useful to focus on the language practices of a
group of people who have the opportunity to interact and who often share not
just a single language, but a repertoire of languages. In sociolinguistic studies,
distinction is most often made between speech communities and speech
networks, on the one hand, and speech communities and discourse
communities , on the other.

2.3.1.1 . SPEECH COMMUNITIES AND SPEECH NETWORKS

In linguistics, speech community is a term used to refer to
”all the people who speak a single language and so share notions of what is same
or different in phonology or grammar. This would include any group of people
wherever they might be and however remote might be the possibility of their
ever wanting to communicate with each other, all using the same language”
(Spolsky 1998: 25).
Many studies highlight the fact that a speech community cannot be
spatially delimitated. This is due to the fact that speech communities are
generally referred to as groups which share a set of language varieties and a set
of norms designed to explain how these varieties should be used. Any small
social network can form a speech community, and the same holds valid in the
case of a large metropolis or a country, a regi on, a communication network.
Since the main objective of sociolinguistics is to relate the significant language

varieties to the significant social groups and situations, a speech community
will be placed in “the abstract space studied in sociolinguistics, the location in
which the patterned variations in selection from the available repertoire takes
place” (Spolsky 1998: 27).
An interesting approach to the concept of speech community is
suggested by Bloomfield (1933), who reduced this concept to the notion of
language or linguistic variety . In his opinion , speech community is defined
as
“a group of people who interact by means of speech and who, besides speaking
the same language, agree about what is considered proper and improper uses of
language” (1933: 42, 155 in Bonvillain 2003: 2).
In other words, the people speaking the same language (or the same first
language, or standard language) are those who make up a speech community .
Although useful, such a definition requires that norms for the use of
language should be taken into account. This view is shared by other linguists
(Blom and Gumperz 1972, Kiesling and Paulston 2005), who define speech
community as a “community sharing rules for the conduct and interpretation of
speech, and rules for the interp retation of at least one linguistic
variety”(Kiesling and Paulston 2005: 6) or “as a group of people who share the
rules for interpreting and using at least one language” (Blom and Gumperz
1972: 16, Kiesling and Paulston 2005: 22). Stressing the fact that members of a
speech community are unified by norms of language use, Hymes defines this
type of community as
“that group of people sharing knowledge of rules for the conduct and
interpretation of speech”, adding that “such sharing comprises knowledge of at
least one form of speech, and knowledge of its patterns of use” (Hymes 1974: 51
in Bonvillain 2003: 2 -3).
A different approa ch is suggested by Hudson (1980 ) and Hymes (1982 ),
(in Kiesling, and Paulston 2005) who state that a speech community should not
be simply equated with linguistic homogeneity of a well -defined set of features.
Moreover, the specialists’ view on the concept speech community
sometimes combines linguistic and extralinguistic aspects. For instance, Labov
refers to speech communities emphasizing the social end evaluative norms
shared by their members:
“A speech community cannot be conceived as a group of speakers who all use
the same forms; it is best defined as a group who share the same norms in regard
to language … who share a set of social attitudes towards language” (Labov
1972: 158, 248 in Bonvillain 2003: 3).
Speech network is another useful concept frequently used in
sociolinguistic studies. This concept was developed by Milroy (1978, 1980 in

Bonvillain 2003: 3) and refers to the specific linkages of persons through shared
varieties and speaking rules across communities. The key element
distinguishing speech communities from speech networks is the actual
interaction between members, which is present only in the case of the latter.
Specialists agree that people in speech networks have direct contact on a
regular basis, but the strength and the frequency of their association may vary.
Therefore, distinction is made between dense and weak networks. The
members of the former have dail y or frequent contact and they are likely to be
bond in more than one way (e.g. living in the same neighbourhood, working
together, etc.), whereas those in the latter category, i.e. weak networks , have a
less regular contact and do not know all of each oth er’s associates.
As suggested by Milroy, dense networks “exert pressure on their
members because values are shared and individuals’ behaviour can readily be
known”. Moreover, careful monitoring of linguistic usage within dense
networks , determines their members to maintain speech norms with little
variation. In contrast , the weak networks
“do not have mechanisms that can apply social sanction against nonconformists
on an individual basis, although the society as a whole exerts pressures for
conformity th rough the transmission of cultural models on both conscious and
nonconscious levels” (Milroy and Milroy 1992: 13 in Bonvillain 2003: 3).
The analysis of speech communities and speech networks is important,
as certain linguistic phenomena within such groups may reveal significant
information about “social and cultural beliefs, about how society is structured
and the ways that people are expected to act or interact” (Bonvillain 2003: 3).
As suggested by Kramsch (1998) , the vocabulary and the discourse chosen by
speakers are indicators of the fact that “they belong to a certain speech or
discourse community ” (my emphasis). Moreover, “this membership makes
them strong and proud and gives them a sense of social importance and historic
continuity from using the sa me language as the group they belong to” (Kramsch
1998: 65 -66).
As regards the relationship between speech communities and
collocations , the words making up such lexical patterns acquire meanings not
only from the collocations in which they occur in indivi dual texts, but also from
the collocations in which they frequently occur in various types of texts
circulating within a speech community . To put it in other words, the specific
use of language by the members of a speech community represents a useful
instrument for the semantic decoding of both collocations and individual texts
in which such lexical patterns frequently occur.
Moreover, the high frequency of collocations is an indicator of their
conventionality. Collocations are considered conventional patterns by virtue
of not being a one off form, as well as by virtue of their being used by the

members of a speech community in a specific form or within prescribed
variations. Since the conventionality of collocations is related to their currency
in a spee ch community, the conclusion could be drawn that “ conventionality is
not a linguistic property in the way an idiom’s lexico -grammatical and semantic
features are” (Fernando and Flavell 1981), and that currency has to be regarded
as a sociolinguistic featur e supporting convention ality .

2.3.1.2 . DISCOURSE COMMUNITIES

Sociolingustic approaches to collocations and other (fixed) lexical patterns
often make a distinction between speech communities and discourse
communities . Different from speech communities, which are made up of
people sharing the patterns of use of the same language, discourse
communities are represented by people who identify themselves as members of
a social group (family, neighbourhood, professional or ethnic affiliation, nation)
and who a cquire common ways of viewing the world through their interactions
with other members of the same group. Common attitudes, beliefs and values
are reflected in the way members of the group use language. Thus, in contrast
with speech communities, which are f ormed of people who use the same
linguistic code and its patterns of use, discourse communities reflect common
ways in which members of a social group use language to meet thei r social
needs (Kramsch 1998: 5 -6). Within the same speech community, signs may
have different semantic values for people from different discourse
communities .
Although virtually similar, the concepts of speech community and
discourse community can by no means be interchangeable.
As suggested by Swales (1990: 21 -32 in Superceanu 1 998: 24), there
are three factors which differentiate a speech community from a discourse
community , namely : 1. the medium, 2. the motive(s) and 3. the membership.
As regards the first factor, Swales states that the communicators who engage in
writing exch anges are not automatically excluded from a speech community,
their exclusion being based on the type of communicative activity and the kind
of exchanges in terms of time and distance. Furthermore, referring to the
motive(s) behind the development and main tenance of discoursal
characteristics, Swales states that communicative needs of the group, such as
socialization or group solidarity, maintain the discoursal characteristics of a
speech community . Moreover, although the main determinants of linguistic
behaviour are social , the determinants of linguistic behaviour in a socio –
rhetorical discourse community are functional , the communicative needs being
set by the objectives for which the people link up, e.g. furtherance of
professional goals, which are more i mportant than those of socialization and

solidarity. Finally, the third factor contrasting a speech community with a
discourse community is membership . If the membership to a speech
community is inherited or adopted, the membership to a discourse community
is acquired by training or relevant qualification.
As regards earlier discussions on the concept of discourse community ,
Herzberg (1986 in Swales 1990: 21) stresses the shared linguistic forms,
regularity rules and cultural concepts, whereas Fennel et al . (1987 in
Supe rceanu 1998: 24) notice in some cases an unsatisfactory circularity in the
definition of discourse communities due to the fact that discourse involves
community just like community involves discourse.
In his approach to discourse communitie s, Swales (199 0) also refers to
characteristics which are more of a social nature or professionally induced. Such
characteristics are regarded as defining traits, a set of necessary and sufficient
criteria for identifying a group of individuals as a discou rse community .
Referring to one of these criteria, i.e. to the employment of at least one genre in
the furtherance of the community aims, Superceanu (1998: 25) suggests that
genre may be a valid and recognized criterion for defining a discourse
community and so may be the other characteristics: participatory mechanisms,
information exchange, a highly specialized terminology and a high level of
content and discoursal expertise . She adds that if a genre were defined by its
discourse community established by m ean of these criteria, numerous genres
with an already established status would not find themselves a community.
Useful as they might be, Swales’s criteria for identifying a discourse
community are rather limited, referring only to professional groups of
individuals.
Using Swale s’s idea in their approach to technical expressions, Lombardo
et. al. (1999: 82) identify the following features of a discourse community:
 it has a broadly agreed set of common public goals
 it has mechanisms of intercommunication am ong its members
 it uses its participatory mechanisms primarily to provide information
and feedback
 it utilizes and hence possesses one or more genre in the
communicative furtherance of its aims
 it has acquired some specific lexis
 it has a threshold level of members with a suitable degree of relevant
content and discoursal expertise
They explain that the use of technical expressions does not characterize
prose written to communicate between members of a discourse community.
This is because a discourse c ommunity is represented by specialists who use
language for a special purpose , and in order to become a specialist, it is essential
to have a command of that language.
To conclude, people in a particular community have common

assumptions not only about t he way the world is organized, but also about the
customary ways that social actions like speech acts are per formed and it is by
means of these common assumptions that their cul tural identity as members of a
social group, small or large may be identified .

2.3.2 . SOCIAL IDENTITY AND JARGONS

A useful sociolinguistic concept worth mentioning in our approach to
collocations is social identity. As suggested by Kiesling and Paulston (2005),
this concept refers “to a range of social personae, including soci al statuses,
roles, positions, relationships and institutional and other relevant community
identities one may attempt to claim or assign in the course of social life”
(Kiesling and Paulston 2005: 79). The acceptance and recognition of social
identity may become useful in language studies because linguistic variation
may be classified by associating it with a specific set of social features. For
instance, special variety/ register marked by a special set of (technical)
vocabulary associated with a professio n or occupation or a social group and
forms part of its jargon . (see 4.5.2. )
Since “certain professions and trades have always tended to develop
specialized vocabularies, while retaining standard syntactic structures (Nida
1996: 66), the speakers who do n ot share the respective jargons must almost
learn a new language in order to understand what the members of such
professional communities try to say.
As Spolsky (1998) states, “a specialized jargon serves not only to label
new and needed concepts, but to establish bonds between members of the in –
group and enforce boundaries for the outsiders” (Spolsky 1998: 33).
The term jargon may, up to a point, be compared with the term idiom
used to refer to the genius of any language and the manner or expression which
is natural or peculiar to it, in other words, its usage. In spite of the fact that such
idioms are peculiar to a specific group and its culture, and therefore limited in
their currency, they are not ‘private’ in the way ‘family idioms’ are.
Nevertheless, they have sociolinguistic viability due to their being in use among
a group or class of people.
Referring to idioms from this perspective, Fernando (1996: 66) mentions
that the currency of idioms varies. The private idiom and the public idiom are
the two extremes belonging to the core vocabulary, which is relatively free of
contextual restrictions, thus having great freedom of occurrence in both spoken
and written discourse. In terms of register, Carter (1987) suggests that the core
vocabulary is neutral i n its privilege of occurrence.
In between the two extremes of ‘private’ and ‘public’ idioms, there are
idioms of varying degrees of currency depending on who uses them and where.

These restricted idioms are usually associated with specialist areas of disc ourse
and constitute part of the jargon of vari ous activities (registers): black box,
soft/hardware, hard copy , and data base will immediately be recognized by
those who are computer literate as belonging to that technology; pseudo -cleft
sentence, trans formational generative, affix -hopping , etc. by the initiated as
belong ing to linguistics. Restricted idioms are just like other marked forms in
the vocabulary, for example, nautical terminology: port, star board, fore and aft,
knots , etc.
Apart from the ‘ jargon ’ of specific activities ranging from computer
technology to navigation and of everyday activities such as cooking and
gardening , there is also the vocabulary of ‘anti -languages’, a term which is
used by Halliday (1978a) to refer to the development o f extreme social dialects
by language users such as criminals or political terrorists who exist in an
oppositional relationship to the norms and ideology of the dominant culture.
The general assumption is that the ‘anti -languages’ created by such
groups t akes many forms, but share a series of lexical features which result from
the processes of relexicalization and overlexicalization . If the former process
refers to the provision of new lexical items for the new concepts developed by
each oppositional group , the latter refers to the development of alternative
lexical items for those domains of counter -culture which are of special
ideological significance.
Relexicalization most often involves the cr eation of a dialectal semantics
which could be exemplified by a reversal of the normal meanings of words. For
example, upright man and law used in criminal slang have totally different
meanings: upright man might mean ‘leader of a gang of criminals’, whereas law
might mean ‘crime’.
On the other hand, overlexicaliza tion implies the development of a
specialized lexicon and is closely linked to the jargons developed by groups
within a society, but without being in opposition to socially dominant norms. In
fact, as suggested by Carter (1998: 110 -111), overlexicalization “is a process
which works to semanticize areas deemed by society to be of taboo status and
which are, therefore, often of obsessive concern”. Items which refer to sexual
intercourse of death are widely overlexicalized and there is also specialization
of lexical items for homosexuals and old people (e.g. a homosexual lover is
referred to as friend, associate, companion ; old people are referred to as the
aged, elders, OAPs, geriatrics, seniors, senior citizen, over 60’s, pensioners,
Darby and Joan ). Although this kind of lexicalization is not ‘anti -language’, it
is accepted to operate in order to identify ideologically sensitive areas of
societal discourse. At the level of words and lexical patterns, such antilanguages
are illustrated by euphemisms (see 4.4.2.).

Moreover, social situations may be classified by analyzing them in
relation to their defining characteristics which make together a set of typical
domains. These domains are usually named in relation to a place or an activity.

PLACE ROLE – RELATIONSHI P TOPIC
home mother, father, children,
grandparents, visitors, etc. activities of the family, news
about the members of the family,
household, the meal, etc.
bank manager, employees, clients payments, credits, salaries, etc.
school teachers, pupils, pa rents,
secretaries, etc. activities of the teachers
(teaching, practicing) activities of
the students (attending classes,
studying, making projects,
participating at contests), etc.
Court of
Justice lawyers, accused, judges,
clients, etc. activities of th e lawyers and
judges, trials, decisions, etc.

Table 9. Possible domains related to place and activity

Sociolinguistic analysis reveals the fact that context itself, with its speech
components (setting, participants, topics and goals), may create proble ms in
establishing clear -cut distinctions between different social classes . In some
cultures, the styles of speech used in different contexts are sharply
distinguished, whereas in others, linguistic styles are less differentiated.
Moreover, there are situa tions when some people within a culture are more
sensitive to contextual cues than others, and adjust their speech accordingly.
Sensitivity to context may be related to such social factors as gender or class,
or it may be related to an individual’s partici pation in many different types of
situation.
However, the idea is shared that the people living within a society
regularly associate with each other according to certain characteristic s or
interests they share. The communities formed in this way are easily identifiable
by means of different extralinguistic criteria (behaviour, habits, profession,
interests, etc.), but what is more important, by means of the ‘ marked’ language
its members use. Since the markedness of such languages is visible , especially
at the level of lexis, the membership to a community will be granted only to
those people who will prove that they can appropriately use the words and
patterns specific to that language.
As regards an analysis of such words and patterns on the basis of
frequen cies of usage, mention should be made that it may reveal specific
sociolinguistic patterns . although research methodologies emphasize the need

to use interviews, experimental and situational observations and quantitative
analysis,
“[s]ociolinguistics idea lly collect large samples of ongoing communicative
behaviour and then try to isolate determining factors that result in linguistic
variation” (Bonvillain 2003: 5).
To conclude , sociolinguistic and cultural approaches to collocations are
as important , as us eful. This is due to the fact that such lexical patterns are
generally agreed to be culture – specific and differentiated according to different
domains of activity. Thus, understanding and observing social and cultural
restrictions is essential for the app ropriate use of collocations .
As suggested by Scollon and Wong Scollon (2001), differences in social
status or culture may affect the process of both intra and intercultural
communication:
“Where any two people differ in group membership because they are different
genders, different ages, different ethnic or cultural groups, different educations,
different parts of the same country or city , different income or occupational
groups, or with very different personal histories, each will find it more difficult
to draw inferences about what the other person ” wants to communicate (Scollon
and Wong Scollon 2001: 21 -22).

2.4. CULTURAL APPROACH ES TO COLLOCATIONS

Speakers use words in order to have an effect on the hearers, for instance to
convey information, to a sk questions or to issue commands. The words chosen
are often not neutral in their connotations, but there are also instances when
words have associated senses which presuppose culturally -shared symbolic
meanings. If these meanings are not shared, speakers may find it difficult to
understand the full range of meanings expressed by their interlocutors, whether
envisaging intra -, or intercultural communication. Consequently, in order to
gain insights into a people’s worldview or system of values, it is necess ary to
ascertain the cultural symbols embedded in their words.

2.4.1 . LANGUAGE AND CULTURE

When communicating, people share information about common experience and
they, express facts, ideas, or events which are communicable because they refer
to a sto ck of knowledge about the world shared by other people. Words may
also be used to convey symbolic meanings, expressing cultural values and
shared assumptions , thus reflecting attitudes and beliefs of people, i.e. cultural

reality . In order to interpret wor ds appropriately , people make assumptions
about each other’s intentions, desires, or goals. Quite often , interpretation of
intent relies mainly on cultural and social norms, because appropriate topics are
determined by culture. From this perspective, cultu re may be said to both
liberate individuals from anonymity and constrain them by imposing on them a
structure and principles of selection.
Kramsch’s view is illustrative in this respect. He considers that
“the members of a community do not only express ex perience, they also create
experience through language. Through verbal and non -verbal aspects, language
embodies cultural reality. Language is a system of signs that is seen as having
itself a cultural value. Speakers identify themselves and others through language
and they view their language as a symbol of their identity”(Kramsch 1998: 3).
Since language is assumed to symbolize cultural reality, reference should
be made to culture and to the realities it embodies.
One of the numerous definitions of cultu re is that provided by Tylor
(1871) , who considers that culture is
“that complex whole which includes knowledge, belief, art, morals, law,
custom and any other capabilities and habits acquired by man as a
member of society” (Tylor 1871 in Holliday, Hyde a nd Kullman 2006:
59).
Kroeber and Kluckhohn’s (1952) approach is also worth mentioning. In
their opinion, culture
“consists of patterns, explicit and implicit of and for behaviour acquired or
transmitted by symbols, constituting the distinctive achievemen t of human
groups, including their embodiment in artifacts”. (Kroeber and Kluckhohn 1952:
181 in Katan 2004: 25).
They also refer to the existence of an
“essential core of culture which “consists of traditional (i.e. historically derived
and selected) ide as and especially their attached values” . Interestingly enough
they relate culture not only to the past, but also to the future stating that “culture
systems may, on the one hand, be considered as product of action, on the other
hand, as conditioning eleme nts of future action” ( id. ibid ).
The fact that culture should not be limited to the past is made obvious by
Merilă (2014: 32) who states that “cultures evolve historically” much of their
evolution being “due to contacts with other cultures which lead to borrowings,
appropriations, and inventions” , in other words to sharing, adapting and creati ng
(new) cultural values.
Another interesting point made by Merilă is that “ cultures are not
isolated, discrete phenomena ” (id. ibid.) strictly identifiable within the
corresponding geographic spaces and nation s. They interact in unpredictable
ways and in fluence each others allowing their representatives to identify

themselves with realities which do not necessarily originate in their native
culture. To put it in Gupta and Ferguson’s (1992) words,
[I]t is so taken for granted that each country embodies it s own distinctive culture
and society, that the terms society and culture are routinely simply appended to
the names of nation -states , as when a tourist visits India to understand ‘Indian
culture’ and ‘Indian society’ (Gupta and Ferguson 1992: 6, in Meril ă 2014: 32 –
33). (emphasis added )
The view t hat cultures occupy ‘discrete spaces’ is restrictive and may not
account for specific circumstances such as those in which people inhabit
borderlands or in which cultural differences are traceable in a common
geographic space (Gupta and Ferguson 1992 in Meril ă 2014: 34 ).
In discussing culture , reference should also be made to Holliday’s (1999)
distinction between small and large cultures . The former concept is useful
when discussing about social and professional c ommunities, whereas the latter
is appropriate in the case of more numerous communities (e.g. national, ethnic,
international):

Small cultures Large cultures
CHARACTER Non – essentialist, non -culturist
Relating to cohesive behaviour
in activities with an y social
grouping Essentialist, culturist
‘Culture’ as essential
features of ethnic,
national or
international group
RELATIONS No necessary subordination to
or containment within large
cultures, therefore no onion
skin Small (sub)cultures are
contained wi thin and
subordinate to large
cultures through
onion -skin relationship
RESEARCH
ORIENTATION Interpretive, process
Interpretive emergent behaviour
within any social grouping
Heuristic model to aid the
process of researching the
cohesive process of any soci al
grouping Normative
Beginning with the idea
that specific ethnic,
national and
international groups
have different
‘cultures’ and then
searching for the
details

Table 10. Small cultures vs. large cultures (Holliday 1999, source Holliday et
al. 2004: 63)

Furthermore, the notion of large cultures and the approach to
communication between cultures brings forward another useful distinction,
i.e. the distinction between intracultural and intercultural communication . In
the case of intracultural communicatio n the successful interpretation of
language in a given context depends on the degree to which participants share
conventions and procedures, including those related to paralanguage,
pragmatics and genre. Such conventions and procedures, together with the
values and beliefs which lie behind them, are elements of common cultural
knowledge, and the people who share them can be thought of as belonging to
the same culture. As regards intercultural communication , it is activated when
the members of different nati onal or ethnic communities enter in contact, and it
is likely to occur in different situations: the workplace, law courts, marketing,
translation, and interpreting. However, this concept has a broader sense because
even within communities, communication ac ross different groups with different
knowledge and values may be conceived as being intercultural .
Although intercultural communication often brings to the fore the idea
of shared cultural values, there is still disagreement over the degree to which
cultu res reflect universal human attributes. Some specialists consider that
cultural differences are superficial and that cultural conventions realize the
same basic human needs everywhere (e.g. greetings, hugs bowing, shaking
hands), but the situation is quite different because there is an obvious disparity
between the theory about cultural universals and the lived experience. When
language is not shared, there is a straightforward and very apparent barrier in
communication. Moreover, unshared cultural conventi ons may have less
apparent, but more damaging consequences. Different cultural conventions are
likely not to be understood, but, what is more important, there is potential for
misunderstanding them. As Cook (2003: 53) puts it, “the same customs may
send qu ite different signals, with potentially disastrous results for the
intercultural understanding”.

2.4.2 . CULTURAL MODELS

The various manifestations of culture challenged the specialists in the field to
find new methods of grouping them. As a result, a se ries of cultural models
have been suggested.
A cultural model is “a construction of reality that is created, shared and
transmitted by members of a group” (Bonvillain 2003: 3). The fact that cultural
models are shared and accepted explains why their member s assume these
models to be natural, logical, necessary, and legitimate. Specialists have noticed
that once a cultural model becomes a background for behaviour, it is no longer
recognized as culturally -constructed, but as being the natural order of life.

Moreover, cultural models are based on people’s ideas about the world
they live in and they may be expressed in several ways. The key instrument
used in transmitting cultural models is language , which means that it is by
language and language use that spea kers express, reinforce, and thus perpetuate
underlying cultural models .
The cultural model proposed by Trompenaars and Hampden -Turner
(1993) comprises three concentric rings or ‘layers of culture’.
The outer layer is the most visible one, being thus cal led explicit . It is
the level of culture which includes artifacts and products (the organization of
institutions). As regards the middle layer, it is made up of norms and values,
the former being related to norms of conduct, whereas the latter refer to
aspirations which may never actually be achieved. Last, but not least, the core
layer is the least visible one in Trompenaars and Hampden -Turner’s cultural
model. This third and implicit layer represents the heart of culture and contains
basic assumptions abo ut life, unconsciously handed down from generation to
generation.

Fig. 6 . Trompenaars’ Model of Culture (1993: 23, source Katan 2004: 389)

Different from Trompenaars’ and Hampden -Turner’s view, Hofstede
refers to the existence of only two layers of culture , the layer of practices (a
superficial layer including groups, symbols, heroes, and rituals) and the deeper
layer of values which corresponds to the core of culture. Artefacts and products

EXPLICIT

Norms and values
Basic assumptions

IMPLICIT

Fig. 7. Hofstede’s levels of culture (1991: 7, source Katan 2004: 39)

The surface layer included within the larger layer of practices is
represented by symbols , which are “semiotic signs recognized as belonging to a
particular group such as words, gestures , pictures, objects, dress and so on”. The
deeper level is that of heroes, distinction being made between real and
imaginary heroes. Finally, the third and the most hidden level of practices is
represented by rituals, which, according to Hofstede (1991: 8 in Katan 2004:
41), are “technically superfluous in reaching desired ends” but socially essential
in a culture. The important point highlighted by Hofstede’s cultural model is
that although symbols, heroes, and rituals are visible, being subsumed to
practi ces, “their cultural meaning is invisible and lies precisely and only in the
way these meanings are interpreted by the insiders” (Hofstede 1991: 8 in Katan
2004: 42).
The Iceberg theory also proves extremely useful in explaining the
structure and function ing of culture . According to this theory, the most
important part of culture is completely hidden, and what can be seen is ‘just the
tip of the iceberg’ . Considering this aspect, Hall refers to the existence of three
levels of culture, namely: the level of technical culture, of formal culture and
the level of informal (out-of-awareness ) culture. Technical culture
corresponds to communication at the level of science, it may be measured
accurately, and it has no meaning outside itself. As regards the level of formal symbols
heroes
rituals
values

culture, this is not an objective reality, but it is part of an accepted way of doing
things. Finally, the level of informal culture is made up of experiences
acquired informally and out -of-awareness.
Using the Iceberg theory , Brake suggests a different division of culture .

Fig. 8. Brake’s cultural model (1995: 39, source Katan 2004: 43)
In his opinion,
“[L]aws, customs, rituals, gestures, ways of dressing, food, and drink and
methods of greeting and saying goodbye. These are all part of culture, but they
are on the tip of the cultural iceberg. The most powerful elements of culture are
those that lie beneath the surface of everyday interaction. We call these value
orientations. Value orientations are preferences for certain outcomes over other s”
(Brake et al. 1995: 34 in Katan 2004: 43).
In spite of the individual usefulness of these models, there are significant
differences in the approach adopted in each case .

Trompenaars Hofstede Hall Brake
explicit
Artefacts and
products
visible
Practic es
symbols,
heroes,
rituals

Technical laws, customs, rituals,
gestures, ways of dressing,
food and drink and
methods of greeting and
saying goodbye
implicit
Norms and values
Formal rituals, customs,
ways/styles of discourse,
dress, etc.

Basic assump tions
Core values
invisible

Values

Informal
(out-of
awareness ) Value orientations –
orientations, action, time,
space, communication,
power, environment,
individualism,
competitiveness, structure
Table 11. Cultural models compared and contrasted

I consider that, although making an extensive inventory of a culture’s
constitutive elements and highlighting the importance of the value orientations ,
Brake does not refer to these manifestations of culture from a linguistic and
social point of view. Fro m this perspective, Hofstede’s cultural model is much
clearer and much better organized, including all the significant cultural aspects
retrievable in any culture. On the other hand, Hall’s cultural model seems to
focus more on socio -linguistic aspects and to neglect the heritage which gives
the specificity of each culture.
However, irrespective of the approach suggested, these cultural models
have something in common, i.e. they attest the existence of core values in a
culture and accept their implicit/invi sible nature.
If the superficial elements making up a cultu re are more visible, thus
more accessible within and across cultural borders, the core values remain
intimately bound to the culture originating them and represent a criterion of
differentiation b etween the identity of individuals belonging to different cultural
spaces (Dumitra șcu 2009c).

2.4.3 . CULTURAL IDENTITY, CULTURAL CONNOTATIONS AND
COLLOCATIONS

Besides the numerous aspects shaping one’s identity (social status, culture,
political views, national or ethnic origins, sex or sexual orientations, etc.),
specialists agree that there is a natural connection between the language spoken
by the members of a group and that group’s identity .
Moreover, since language is one of the important representations of
culture , the conclusion can be drawn that the language spoken by an ind ividual
represents an important criterion in establishing his/her membership to a certain
culture, hence his/her cultural identity .
In Kramsch’s (1998: 126) opinion, cultural identity refers to
“bureaucratically or self -ascribed membership in a specific culture”. Moreover,
he considers that this type of identity is directly influenced and conditioned by
the context of culture shaping it i.e. by “the historical knowledge, the beliefs,
attitudes, values shared by the members of a discourse community that
contribute to the meaning of their verbal exchanges” (ibidem).
Recognizing and observing cultural identity is obligatory and essential
in all situations, especially in communication, whether considering this process
within the same culture, or across cultur al boundaries. Since culture is not
limited to what speakers express, reinforce, and perpetuate by means of
traditions, beliefs and values, but it also includes what speakers convey by
means of the language they speak, the fact becomes obvious that languag e

represents an instrument as important as useful in both shaping cultural
identity and transmitting it from generation to generation.
As regards linguistic cultural identity , it may be expressed in different
ways, and achieved at different linguistic lev els. Sometimes the cultural
specificity of a language is traceable at the levels of morphology and syntax,
whereas in other cases grammatical specificity is doubled by the use of culture –
specific lexical items denoting realities which lack an equivalent in other
language cultures. Given the obvious limitations imposed by such
grammatically – and culturally -marked words, specialists agree that their
appropriate use is an essential criterion in ascribing speakers membership to a
certain linguistic and cultural community, or, on the contrary, in excluding them
from that community.
Out of the numerous linguistic levels bearing an obvious cultural imprint,
the lexical level represents a valuable resource for identifying and
differentiating among various cultural identities . Such a differentiation is made
possible by means of the cultural categories which mark this type of identity
and which represent direct cultural signs in the lexicon.
As regards the cultural markedness of lexical patterns , cultural data is
encoded in various ways in such units. Analyzing individual words,
collocations and idioms, Teliya et al. (1998: 57 -58) suggest that they often
combine more than one type of cultural information in their semantics and that
this cultural information is encoded by means of cultural semes , cultural
concepts or cultural connotations .
Cultural semes are words and word -combinations which denote
idioethnic realia. They form part of the lexical meaning and reflect general
knowledge about the realia whether material, social or historical. Moreover,
cultural concepts are abstract notions which map and construct the world –
picture in a culturally -specific way, their specificity being implemented at the
cognitive level. Finally, cultural connotations refer to the interpre tative
relation existing between linguistic signs and symbols specific to any other
cultural non -verbal code (stereotypes, prototypes, myths and other such cultural
entities).
Referring to the same type of connotations, Teliya (1993, in Teliya et
al. 199 8: 59) suggests that it “arises from an associative relation between the
image contained in the inner form of a language sign and the content of a
cultural pattern, and that it is especially vivid in idioms and restricted lexical
collocations . The activat ion of cultural connotation s in restricted
collocations is connected with the type of cultural information contained in the
base or node of the collocation (see 2.1.1 .) and with the nature of the semantic
specialization in the meaning of the ‘collocator ’ or ‘collocate ’.
Moreover, cultural connotations can accompany any culturally –
marked words when they combine in phraseologisms. Such connotations

manifest in word -combinations which activate culturally relevant parameters of
the base word and are derived f rom allusions to cultural realia or from the
interpretation of cultural concepts and cultural subconcepts . The former term
is used to refer to concepts which largely coincide in different languages but, for
some aspects of meaning, corresponding words and phraseologisms show a high
degree of cultural specificity. The term s ubconcepts is used to denote fragments
of concepts when abstract concepts are verbalized in concrete nouns.
Similarly to collocations, idioms manifest cultural connotations
associated wi th subconcepts, but in such cases, cultural connotations may only
be revealed if considerable ideographic fields are studied including idioms,
restricted collocations, proverbs, sayings, and so on. Their cultural specificity is
only traceable in the common ‘kernel metaphor ’ (Lakoff and Johnson 1980)
which acts as a hyperonym with respect to the images conveyed by such idioms.
To conclude, the speakers’ capacity to make linguistic introspections and
cultural reflections derives from their knowledge of ‘ cultural-linguistic
codes’ , that is from their ‘ linguo -cultural competence’ (Teliya et al. 1998: 59)
which is assumed to develop in parallel with their language knowledge, in the
process of internalizing cultural experience. The best linguistic level illustrat ing
what has been stated above is the level of lexis. As suggested by Nida (1996),
“words have meanings in terms of the culture of which they are a part, which
means they have no meaning apart from some context , either the context of
other words, or of a p articular setting. In any such symbolic system as language,
the role of the context is maximized, and the role of the focal unit is minimized,
but it is only by means of the combined meaning of the focal unit and the
context that the concept acquires relev ance” (1996: 87, adapted). This explains
why language may be fully understood only when the context of situation and
the context of culture are implicitly or explicitly clear to the interlocutors. To
put it in other words, “language is essentially rooted i n the reality of culture
[…] it cannot be explained without constant reference to the broader contexts of
verbal utterance” (Malinovsky 1938: 305, in Katan 2004: 99).
Last, but not least, the correlation between language and culture at the
level of phraseo logy is best illustrated by the class of restricted lexical
collocations9 which “abound in cultural information and can hardly be
described as a class of denomination if their cultural meanings are not taken into
account” (Teliya et al. 1998: 55). The spea kers’ unawareness with respect to the
cultural markedness of such fixed lexical patterns leads not only to their misuse,

9 Catford (1965) uses the notion of collocation to explain cultural incompatibilities. In
his opinion, when an expression seems incompatible with cultural expectations, it is
possible to consider that the respective expression is incompatible with collocational
expectations . To put it differently, cultural incompatibility is reflected in the
language because culture is reflected at the linguistic level.

especially by non -native speakers, but also to unsuccessful intra – or interlingual
communication.

CONCLUDING REMARKS

The most releva nt aspects approached in the prese nt chapter are summarized
below.

1. The various approaches to collocations prove that language is indeterminate
in its interpretation and hence very flexible in use. Moreover, when the switch
points between two modes of i nterpretation are not always explicitly signa lled,
or two modes offer sharply contrasting ways of interpreting data, messages are
likely to be interpreted quite differently, so much the more if the interpretation
is not made by the same person.
2. Althoug h collocations have been extensively discussed and analysed in
various language studies, there are many instances when speakers find it hard to
establish a clear -cut distinction between collocations and other (fixed) lexical
patterns (free word combination s or idioms). Such confusions may be avoided
by increasing the speakers’ awareness with respect to the characteristics of
collocations in contrast with other word combinations.

3. The psycholinguistic approach to collocation s focuses on the processes
involved in decoding multi -word patterns and the aspects which may facilitate
their appropriate use , both in the case of intra – and interlingual communication.
In discussing collocations from the perspective of psycholinguistics it is
interesting to consider the mechanisms wh ich affect lexical processing. The
frequency effect is a useful psycholinguistic concept by means of which the
appropriate use of collocations, either in the native language , or in a foreign
language may be explained.

4. The sociolingui stic approach to collocations is concerned with discovering
patterns of linguistic variation which are derived from differences in speech
situations, and from social distinctions within a community that are reflected in
the communicative performance.

5. Sociolinguistic and cultural approaches to collocations are as important as
useful. This is due to the fact that such lexical patterns are generally agreed to
be culture – specific and differentiated according to different domains of activity.
Thus, underst anding and observing social and cultural restrictions is essential
for the appropriate use of collocations.

3. COLLOCATIONS FROM THE PERSPECTIVE OF LEXICAL
SEMANTICS

Language, whether in its written or spoken form, is the ever -evolving means by
which p eople can communicate their thoughts, feelings, and attitudes, whether
considering inter – of intra -cultural communication. Nevertheless , language is
also a major source of difficulties in both these types of communication.
Specialists agree that language i s often ambiguous which means that
speakers can sometimes be uncertain what their interlocutors mean – whether in
speaking or in writing. To put it differently, language can never fully express
the meanings intended by speakers. Moreover, the process of hu man
communication is conditioned by the existence of certain elements: the
transmitter (T), the receiver (R), the (C)ode, the context (C) and the message
(M). The (T)ransmitter is both the creator and the sender of the message,
whereas the (R)eceiver recei ves and interprets the (M)essage of the
(T)ransmitter. The (C)context/Referent is realized by the (R)eceiver, and can be
either verbal or verbalized. The (M)essage comprises the very structure of
discourse, which is sent by the (T)ransmitter by means of a signal or a sequence
of signals. Finally, the code (C) consists of the system of specific signs and the
rules used in combining them. In the absence of any of the constituent elements
mentioned above, the communication process cannot achieve its goals.
This is a general view on communication and its mechanisms. However,
the communication process is not a mere transfer of words from the
T(ransmitter) to the R(eceiver), and vice -versa. The correct understanding of the
message is conditioned by numerous factor s, one of them being its appropriate
organization in keeping with lexical, morpho -syntactic, semantic and pragmatic
constraints.
Simplifying the scheme of the communication process, Saussure (1998)
focuses on the semantic analysis of the message . In his vi ew, a communicative
act involves a speaker, a listener, the speaker’s intentions of communication as
encoded in the message by means of specific linguistic signs and the decoding
of the message by the listener. Special attention is devoted to the linguisti c
sign, which, in Saussure’s opinion, joins a concept and an acoustic image. The
latter is not just the sound itself, but its mental representation in the mind of the
hearer. When using words , speakers always make mental associations between
the acoustic i mage and the concept. The concept reflects the acoustic image of
the word, whereas the sounds transmitted by the speaker create the acoustic
image that reflects the conceptual image in the mind of the hearer. This is a
bipolar mental association that invol ves two terms: the form and the concept ,
and two phases: the evocation of the name by means of the thing and the
evocation of the thing by means of the name.

Ogden and Richards’ (1923 in Zdrenghea 1977: 15) view on meaning is
also worth mentioning. In the ir opinion, any symbolic act, including speech, is
conditioned by the existence of three factors, namely: the symbol , that is the
material aspect (phonic or graphic) of the linguistic sign, the thought
/reference , that is the mental content which accompani es the occurrence of the
symbol in the minds of both speaker and hearer, and the object/the referent
designated by the symbol. The symbol -reference and reference -referent
relations are direct, in the sense that the symbol expresses the reference which,
in its turn, refers to the referent. But the relation symbol -object is an indirect
one, thus sending us back to Saussure’s bipolar mental relationship.

Although abstract, I believe that this triangle represents a very good
starting point in the sem antic approach to words. Nevertheless, it can hardly
account for the complex aspects regarding the ir meaning.

The idea is shared by Zdrenghea (1977: 16) that words are directly
related to a meaning or sense . This holds valid especially in the case of terms
(see 4.6.1.), but more numerous are the situations in which words are
semantically ambiguous, preventing speakers from easily accessing their
meaning.
The ambiguity of words is caused by various factors, both linguistic and
extralinguistic. If the fo rmer may be coped with rather easily by integrating
isolated words in their co -text and context (see 3.5.3 .), the extralinguistic
factors (for example social , or cultural ) imply a more complex approach, the
appropriate semantic interpretation of words bein g possible only if there is ‘a
balance bet ween inference and convention’. Nevertheless, even in these
conditions, the results are not always satisfactory, as they are conditioned by so
many other aspects which cannot be predicted or accounted for.

Words , however, are not used in isolation. They combine in as many as
various lexical patterns such as : collocations, idioms, metaphors, similes,
clichés, proverbs (see 1.4.2. – 1.4.1 0.) and grammatical structures
(constructions, sentences), and in such cases the ir meaning is affected even
more. Furthermore, their combination is not a random one. The formal
restrictions conditioning both lexical patterns and grammatical constructions are
always doubled by semantic , pragmatic and stylistic constraints ( see 3.1.).
Nevertheless, in the case of lexical patterns the discussions regarding the
semantic aspects are more complex. This is due to the fact that, on the one hand,
the individual meanings of words are affected when used in such combinations,
and the resulting le xical patterns are more or less semantically accessible to the
speakers, on the other.
That is why specialists agree that lexical patterns create numerous
problems to the unaware speakers, and more importantly, to the non -native
speakers of a language, an d that special attention should be devoted to such
patterns in language teaching studies ( see 1.2.2 .).
Things are even more complicated in translation (see 5.5.1. – 5.5.4.). As
suggested by Nida (1996 ),
“[T]he sociolinguistic aspects of vocabulary ( see 2.3.), grammar and discourse
become focal in all translating and interpreting because the distinctive features of
the languages and of the cultures (see 2.4.) constitute major difficulties in
communication. Some persons never seem to realize that no two wor ds in any
two languages ever match completely in designative ( denotative ) and
associative ( connotative ) meanings ( see 3.3.2.1 . – 3.3.2.3 .). Furthermore, they
are often not sensitive to the ways in which the context plays such a large part in
specifying mea ning” (Nida 1996: 29) .

3.1. ON WORDS AND MEANING

Before embarking upon a discussion on how word meaning is affected in
collocations , reference should be first made that many individual words are also
semantically ambiguous or indeterminate in isolation .
The factors conditioning the ambiguity of words are generally agreed to
be either linguistic, or extralinguistic, and, depending on the nature of the
aspects hindering the semantic transparency of words, different solutions have
been suggested.
As regar ds the linguistic factors conditioning the semantic ambiguity of a
word, they hardly ever trouble speakers in practice, because most often the
words in the immediately surrounding text help speakers discriminate between
the different senses that the respec tive word might have. For example, the
meaning of the noun surgery which is ambiguous in isolation, becomes

accessible if one of the words plastic/ cosmetic/ exte nsive /major/ successful , or
to undergo/ to carry out/ to respond to, patients need are found in its
surrounding text. In other words, the meaning ‘a medical procedure involving
cutting a patient’s body open’ will be identified as appropriate for the noun
surgery and differentiated from its other possible meanings if the surrounding
text is provid ed.
Moreover, multiple ambiguities at word level usually dissolve in context .
This is best illustrated by combinations of words in phrases which are, in
Stubb’s (2002: 14) words, “a good candidate for the basic semantic unit of
language in use”. This mean s that instead of trying to decode the meaning
carried by individual words, speakers should start their semantic interpretation
from the assumption that lexical phrases reveal the meaning of the individual
words in it. The word he chooses to support this i dea is the noun bank, which
may be used to denote both the place where one keeps money and an area of
sloping, raised ground around a stretch of water or under shallow water. Since
the appropriate meaning of this word cannot be grasped in isolation, speake rs
should consider the possibility to identify patterns such as bank account, bank
balance, bank robbery, piggy bank or canal bank, river bank, which will reveal
the correct semantic interpretation of the noun under discussion. Another
explanation provided in such cases is that since the meaning of a word is not
independent of its environment, including the co -text in which it occurs, that
word will predict other related words likely to occur round about it, just as the
co-text will predict the word, or one very likely to it.
Besides linguistically -ambiguous situations such as the ones mentioned
above, the meanings of words may be less accessible to speakers due to
extralinguistic factors, both social and cultural. In such situations the semantic
interpret ation of words can only be made if there is “a balance between
inference and convention” (id. ibid.).

3.2. APPROACHES TO MEANING

The starting point in the lexico -semantic approach to collocations is the general
assumption that the formal and semantic in dependence of a word is affected, to
a certain extent, in the various types of lexical patterns traceable in language.
Although some specialists consider that “in a lexical analysis it is the
lexical restriction that is under focus, i.e. the extent to wh ich an item is specified
by its collocational environment” (Halliday 1966: 156), recent studies on
collocations stress the idea that it is more useful to analyze and explain the

degree to which meaning conditions the fixed co -occurrences of words10. The
specialists sharing this idea suggest that in studying collocations the lexical
analysis must always be doubled by a rather extensive analysis of the semantic
aspects involved in the creation and appropriate functioning of such patterns.
The individual meanin gs of the lexical items making up collocations and the
semantic changes resulting from the interaction between them are essential
elements which may explain certain lexical choices and constraints. Hence, the
necessity to closely analyze collocations from the perspective of lexical
semantics and to compare and contrast them with other lexical patterns in which
words preserve (free word combinations), or do not preserve their individual
lexical meanings (similes, binominals, idioms, clichés, proverbs, metaph ors).
The interaction of lexicology and semantics in the study of words and
lexical patterns has been suggested and demonstrated by many specialists, but
their views with respect to the way in which the two branches condition each
other are quite different . Some suggest that the meaning of words can be
deduced strictly by identifying and analyzing their basic semantic features,
whereas others consider that, since human speech is greatly influenced by non –
linguistic factors, the semantic analysis of words re quires a multiple -sided
approach, rather than a unilateral one. In order to support both views and to
check their validity in the case of collocations , reference will be made to some
of the most influential approaches which best fit the purpose of our rese arch.
3.2.1. Componential analysis11, one of the earliest and still most
important approaches to lexical meaning according to which “the sense of every
lexeme can be analysed in terms of a set of more general sense -components (or
semantic features), some o r all of which will be common to several different
lexemes in the vocabulary” (Lyons 1987: 317), may prove very useful in the
study of collocations. Thus, componential analysis regards the meanings of
words as being made up of smaller, elementary and invar iant units of meaning.
As suggested by Weinreich, (1966 in Lyons 1977: 321), such an analysis is
possible due to the fact that lexemes have an internal structure which mirrors
the syntactic structure of sentences and phrases.
According to Carter (1998) , componential analysis is the study of
combinational or selection restrictions and represents “a technique for
describing relations of meaning by breaking down each word into its irreducible

10 J. R. Firth (1957) considers that the study of meani ng should be viewed in terms of
function in context . In other words, “the meaning of an utterance has to do with what
the respective utterance is intended to achieve, rather than the meanings of individual
words”.
11 L’analisi componenziale consiste nello scomporre il significato in elementi minimi,
detti tratti , i quali concorrono alla sua formazione (sono cioè pertienti ) e lo
differenziano da quello di altri lessemi (sono cioè distintivi ) (Marrinucci 1996: 447,
emphasis in the original).

features: those components which are absolutely minimal for its ref erence”, its
aim being a “fuller analysis of that finite set of components or semantic features
in lexical items which are universal in that they underlie our basic cognitive
proces ses for the ordering of meaning ” (Carter 1998: 56) .
As regards the universa l dimension of the semantic features of words,
reference should be made to the relationship between componential analysis
and universalism . This accounts for similarities and dissimilarities resulting
from the lexical decomposition of the same lexical mean ing in differen t
languages. As Lyons (1977 ) points out,
“universalism combines at least the following distinguishable theses: a) there is
a fixed set of semantic components which are universal in that they are
lexicalized in all languages; b) the formal p rinciples by which these sense –
components are combined to yield as their products the meanings of lexemes are
universal (and presumably innate); c) the sense of all lexemes in all languages
is decomposable , without residue into variable combinations of (h omogeneous)
sense -components ( emphas is added )”. (Lyons 1977: 331)
Although componential analysis , which is based on a kind of atomism,
might seem incompatible with structuralism, what really counts is whether the
atoms of meaning into which the meanings of words are analysed, are thought
to be logically and epistemologically independent of one another. The aspect
emphasized by both those sharing and those rejecting this view is that all the
words in the same semantic field are definable in terms of their str uctural
relations with one another, componential analysis being a means of describing
these relations.
From the perspective of lexical semantics, componential analysis may
be said to be doubly structuralist. This is due to the fact that such an analysis
“defines meaning of words simultaneously in terms of the external, interlexical,
relational structures – the semantic fields – in which semantically related and
interdefinable words, or word meanings, function as units, and also in terms of
internal, intra lexical and as it were molecular, relational structures in which the
atoms of word -meaning function as units” (Lyons 1995: 107).
Starting from the idea that speakers “can conceive of all lexical items as
encodings of one or more semantic elements or com ponents, whether these are
overtly signalled or not, and in identi fying them they can establish the
denotation of words”, Widdowson (2000: 56 ) suggests that t he same principle
may be applied in order to establish similar minimal pairs of lexical items with
respect to their semantic components. Thus, he suggests using the
componential analysis in order to make an inventory of the semantic features
encoded in lexical forms. Nevertheless, he is aware of the fact that such an
approach may become immensely compli cated due to the fact that as the details
proliferate, they can lose their point and create confusion. Still, componential

analysis may be used to identify certain general conceptual categories such as
state , process , causality , class membership , possessio n, dimension , location ,
and, directionality, or semantic principles which find expression in the
particular com ponents. By invoking them, speakers may move from identifying
the denotation (see 3.3.2.1 . – 3.3.2.3 .) of particular lexemes to tracing the
sense relations that exist between them . In other words, componential analysis
is a very useful instrument by means of which the sense relations existing
among lexemes may be made precise.
Nevertheless, although componential analysis allows for a highly
expli cit and economical account of sense relations such as hyponymy ,
synonymy and antonymy (see 3.3.3.1 – 3.3.3.3 ), and helps in explaining or
predicting the incompatibility between collocated words, some specialists
consider that this approach is not precise e nough to describe the practical use of
words, because it ignores nuances of word meanings.
If the sum of distinctive features is assumed to be the necessary and
sufficient condition to deduce word meanings, the conclusion can be drawn that
the members belo nging to a category have equivalent membership and that
there are clear -cut boundaries between semantic categories. This is hardly
possible since most categories do not consist of a bundle of features which are
shared by all their members, or, to put it di fferently, the membership of a word
to a category is not the necessary and sufficient condition to deduce its meaning.
In order to explain the membership of a word to a category, Wittgenstein
(1972) suggests , the notion of family resemblance :
“the membe rs of a large family typically resemble one another in a variety of
ways, but there are no features which they all have, and there may be members
which share no features. Nevertheless, the members will be linked to others by a
chain of resemblance” (Wittge nstein 1972: 25) .
On a somewhat more abstract level, Rosch and Mevis (1975) define
family resemblance as follows: “[A] is a set of items of the form AB, BC, CD,
DE. That is, each item has at least one and probably several elements in
common with one or mo re other items, but no or few elements are common to
all items” (Rosch and Mevis 1975: 575).
3.2.2 . The prototype theory suggested by Rosch (1973, 1975) is another
important approach to meaning that may prove very useful in accounting for
conditions of col locability. In her view, categories consist both of good and bad
members and include marginal examples, whose category membership is
doubtful. Thus, as far as the category membership is concerned , there are some
good, typical and central members, on the on e hand, and bad, atypical and
periphery members, on the other.
This view is shared by Lyons (1995) , who mentions that the descriptive
meaning of a lexeme can be sometimes explained by means of a more or less

synonymous paraphrase, whereas in other cases i t can be best rendered by
means of an accepted, imperfect, open -ended definition of the prototype.
Moreover, he states that “speakers of a language operate with prototypes and
usually, what they want to refer to conforms to the prototype ” (Lyons 1995:
96).
In discussing the representation of word meaning, “a semantic model can
be obtained by reconciling componential analysis with the notions of
prototypic categories and fuzzy grammar (Neagu 2005: 25). T he prototype
theory and family resemblance are the two basic principles according to which
cognitive and linguistic categories are organized. According to the former
principle, elements are included in a category not because they have the
properties required of each one of its members, but because they exhibit , to a
certain extent, some types of similarity with a particular category member
which has been naturally or culturally established as the best prototype of its
kind. As regards family resemblance , “the members of a category are not
linked because they sh are the same criterial set of attributes, but because they
are similar to each other in different respects just like the members of a family.
Although “both principles are important and useful for lexicology, lexicography
and lexical semantics, prototypes are applicable in constructing definitions for
words and in highlighting interesting aspects of sense relations and semantic
fields” (Neagu 2005: 33 -34).
3.2.3 . The structural approach to collocations focusses on the idea that
linguistic units do not exist independently of other linguistic units, and that they
are integrated in a network of relations. To put it in Lyons’ (1977 ) words, the
central thesis of structuralism is that
“every language is a unique structure or system and that the units which are
identified, or postulated as theoretical constructs, in analyzing the sentence of a
particular language (sounds, words, meanings, etc.) derive both their essence and
their existence from the relationships with other units in the same language
system. Speaker s simultaneously identify both the units and their interrelations.
Linguistic units are but points in a system or network of relations; they are the
terminals of these relations, and they have no prior independent existence”
(Lyons 1977: 231 -232).
Therefo re, words do not exist in isolation, but they are (semantically)
connected to one another . From a structural perspective, “[…] the lexical
structure of a language may be regarded as a network of sense relations , i.e.
like a web in which each strand is one such relation and each knot in the web is
a differ ent lexeme” (Lyons 1995: 102, emphasis added ).
Thus, the basic principle of the structural semantic approach to
meaning is that words are defined through the sense relations they have with
other words and t hrough the associations which are structurally organized. This
principle is important and very useful in approaching collocations because the

fact is generally agreed that the meaning of individual words is affected when
used in collocation with other word s. This implies that one and the same word is
likely to have different meanings, depending on the word with which it co –
occurs. For example, the meanings of the adjective white used in collocations
such as white wine , white hair, white chocolate can be ded uced from the
relations it has with the adjectives red, black , dark in the antonymous
collocations red wine , black hair and dark chocolate . As regards, the adjective
dry used in collocations such as dry wine and dry weather speakers will decode
its meaning s more easily by considering the use of the nouns wine and weather ,
respectively, in other structurally organized collocations: white wine, red wine,
dry wine, sweet wine and dry weather , wet weather , hot weather , warm weather ,
cold weather .
3.2.4 . Sharin g the structuralist view that meaning is not inherent in the
linguistic form itself, cognitive linguists consider that, although meanings
mostly rely on the linguistic context, the semantic interpretation of words can
only be complete if other cognitive st ructures, i.e. extralinguistic factors, are
taken into account. Their claim is that the context against which meanings are
characterized extends beyond the language system. To put it differently, word
meanings are cognitive structures that are embedded pat terns of knowledge and
belief. For example, a noun such as June may be semantically interpreted by
appealing to both linguistic and extralinguistic knowledge. The first word which
the noun June may be linked to is year, since June is one of the twelve mont hs
of the year. Furthermore, if speakers conceptualize the word year as referring to
a sequence twelve months, then month and year are understood against the
concept of time. Thus, the semantic interpretation of a word presupposes
different notions that ar e underlined by the speakers’ background knowledge.
Cognitive linguists point out the fact that a linguistic form gets its
meaning by highlighting a particular configuration in the relevant domain in
which it occurs. It is often the case that a linguistic form must be characterized
against many different domains simultaneously. However, a multiple -domain
approach to words does not imply that all facets of domain -based knowledge are
equally central to the meaning of a word, just as it does not claim that ea ch facet
is relevant to each use of a word. Rather, reference can be made to the fact that
certain contexts favour particular, highlighted meanings , while backgrounding
others (even central ones). Referring to domains, Taylor ( 2002) suggests that
speakers
“need to appeal to domain -based knowledge in order to describe how words
are used. Domains vary in complexity from basic conceptions of colour,
temperature, space, time and so on which cannot reasonably be reduced to other
simpler conceptions, to highly complex knowledge structures, such as the rules
of game, social practices, complex technologies and typical event scenarios”
(Taylor 2002: 441, emphasis added).

Thus, when communicating, speakers choose, on the one hand, lexical
items from lexical sets, and types of grammatical items from grammatical
systems to build up grammatical structures, on the other. It is generally known
that words are grouped into lexical sets as a series of semantically related
options which are decisive in constructing a coherent text.
Lexical sets are relevant for the structure of the lexicon and are rendered
by pattern clusters of reference usually related to a certain topic. The same idea
is pointed out by Spencer and Gregory (1965: 73) who suggest that the structure
of lexis may be described by means of the theoretical categories of collocation
and set. In their opinion, collocation is a category which accounts for the
tendency of certain items in a language to occur close to each other, whereas set
is a lexical category accou nting for the tendency of items to share part of their
collocational range, to have a collocational overlap. Their example to illustrate
the former lexical category is the word economy . They suggest that such a
lexical item is likely to occur close to item s such as affairs, policy, plan,
programme, disaster , which are its collocates and form its collocational range.
As regards the category of set, economy , finance and industry could form such a
category because, if they were used as nodal items, they would probably share
many collocates with each other.
Referring to the same two categories (Spencer and Gregory 1965: 75)
point out the fact that sets are rather variable as they depend on both the nature
and the amount of data examined and on the delicacy of t he description, i.e. the
degree of detail and specificity. In their opinion, the more mutual the
collocational range demanded as the criterion for a set, the smaller the sets and
the more delicate their description. From this perspective, lexical sets may be
said to be ‘open’, different from the grammatical systems , which are ‘closed’.
Furthermore, since the notion of lexical set is agreed to be related to a
specific field , speakers should pay special attention to the overlap of lexical
sets, i.e. to the s ituations in which the same lexical item occurs on different axes
of meaning.
Field is used in lexical semantics to refer to
“the particular activity, cultural feature, social institution or topic for which a
particular set of ideationally related lexica l items is often evolved or adapted.
Each field has a specialized topic -related vocabulary, some of which may
turn up in other fields with different meanings ” (Carter 1998: 53, emphasis
added).
It is also interesting to notice that the lexicon of fields consists of: 1.
nouns labelling technical features – artifact, equipment, structure; 2. verbs
identifying and distinguishing between proces ses, types of event, methods; 3.
adjectives and adverbs indicating conventional properties of an artifact, process,
structure and 4. phrases and conventional collocations that may expand,
modify or combine any of the above functions.

Since according to cognitivism the grammatical and semantic structure of
languages is determined by the categories of cognition, this struct ure of
languages may as well be said to be determined “indirectly, by the structure of
the world in terms of such ontological categories as natural kinds” (Lyons 1995:
98).
Lexical fields , which represent lexical organization, i.e. not only the
meaning of lexemes, but how lexemes are interrelated, play a very important
role in cognitive linguistics , together with the semantic frames , which describe
the background encyclopaedic information encoded in the lexicon, and
cognitive models, which represent concept ual associations in culture and
perception. However, according to Glynn (2004: 200), “the frame poses no
apparatus for capturing conceptual relations such as image schema, metaphor
and metonymy or the detailed lexical variation of dialect and register tha t is
basic to language use.” Therefore, a combined analytical study of lexical fields ,
semantic frames and cognitive models is essential in cognitive linguistics
because it accounts for grammatical constructions.
In other words, a construction analysis sh ould follow a combined frame –
field analysis , based on a collocation study of the lexical field. This consists in
studying simple semantic units, such as lexemes, and the semantic relations,
conceptual metaphors and frames, combined with the study of schem atic
semantic units, such as grammatical constructions and collocations. The result
will be a much better understanding of how linguistic structure interacts with
the processes involved in language production. Grammatical constructions
bridge the paradigma tic-syntagmatic language structures and it is vital to
understand how these structures interact.
The interaction of semantic structures and grammatical constructions is
also crucial. Given the fact that the paradigmatic context of a semantic unit is
decis ive for its meaning and use, the semasiological research should be
combined with the onomasiological investigation.12
In this respect, I share Geeraerts et al.’s opinion that
“It is here that frames, models and fields come to the fore. All three approaches
[semantic frames, cognitive models and lexical fields] are equally applicable to
both the semantic variation ‘within’ semantic units as well as formal variation
‘across’ a concept. Despite the similarity between these three approaches and the
overlap in t heir concerns, few attempts have been made at integrating them”
(Geeraerts et al. 1994: 7).

12 The terms semasiological and onomasiological are used here in the sense of
semantic variation of a linguistic form, and formal variation of a concept, respectively
(cf. Geeraerts et al. 1994: 1 -16)

Cognitive linguistics is the branch which could and should integrate
these three approaches, because it is only in this way that grammatical
constructions could be approached in a more comprehensive manner.
As regards constructions , mention should be made that they are
entrenched (fixed) semantic units which behave like lexemes and their use is
greatly influenced by language community variation, i.e. by semantic va riation
in the language forms due to factors such as register and dialect . They also
make up complex networks of onomasiological variation, i.e. networks
illustrative for the formal variation of a concept. Nevertheless, unlike lexemes,
constructions bridge paradigmatic -syntagmatic language structures, the result
being twofold. On the one hand, they are very productive, due to variable syntax
and lexis, and they are made up of lexemes which may be understood by
studying lexical fields and conceptual metaphor s, on the other.
Instead of a conclusion, mention should be made that c ollocation studies
of the lexical fields reveal “syntagmatic patterns that represent the syntactic
forms available to the domain frame” (Glynn 2004: 202), the result being a
constructio nal field, or a constructional network. The three sets of semantic
structure, i.e. conceptual , lexical , constructional , delimited by a functionally
defined frame -domain, will facilitate a better understanding of their interaction.

3.3. DEFINITIONS AND C LASSIFICATIONS OF MEANING

In dealing with possible definitions and classifications of meaning , reference
should be made to the distinction between word and lexical item .
As Sinclair (1 998) points out,
“this distinction is the key to the description of vo cabulary due, on the one hand,
to the fact that any, if not most, meanings require the presence of more than one
word for their normal realization, and patterns of co -selection among words,
which are much stronger than any description has yet allowed for, have a direct
connection with meaning, on the other.” (Sinclair 1998: 4)
A lexical item, also called lexical unit (Cruse 1986: 24), is characterized
by form, gr ammatical function and meaning. The basic syntagmatic lexical units
of a sentence are usually de fined as the smallest parts which satisfy two criteria
of utmost importance: 1) a lexical unit must be at least one semantic constituent;
2) a lexical unit must be at least one word.
Nevertheless, these criteria cannot always be applied because “the set of
meaningful items is finite and the set of meanings in use does not appear to be
limited” (Sinclair 1998: 5). Moreover, the lexical item has a dual nature,
because , if it is considered in relation to other forms, the lexical item is a form,
whereas if it is considered in relation to other meanings, the lexical item is a
meaning ( id: 12).

Referring to the components of a lexical item, Sinclair (1998) suggests
that they may belong to five categories of co -selection, namely 1. the core, 2.
the semantic pros ody, 3. collocation , 4. colligation, and 5. semantic
preference . If the first two categories are obligatory, the remaining three i.e.
collocation, colligation and semantic preference are optional elements. As
Sinclair explains,
“the core constitutes the e vidence for the occurrence of the item as a whole and
the semantic prosody is the determiner of the meaning as a whole. The optional
categories serve as a means of classifying the members of a paradigm, and thus
the two axes of patterning are related”. (Sinclair 1998: 15)
The first optional category of co -selection which a lexical item may
belong to is the collocation. This category is defined differently at the
syntagmatic and paradigmatic levels. Thus, at the syntagmatic level, collocation
is the most obv ious relationship, whereas at the paradigmatic level, the
relationship of mutual collocation implies that the two items collocate on
separate occasions with the same item or items. In other words, items can
collocate with each other at the paradigmatic le vel when they occur in a text.
Moreover, two items in a paradigm are considered mutually exclusive. For
example, the nouns clutch, group, web used in the collocations a
clutch/group/web of companies  grup (mic)/rețea de companii , are mutually
exclusive on the paradigmatic axis. Similarly, the partitives a bit/great
deal/(big) load, a sniff and a hint in the collocations a bit/great deal/(big) load
of trouble  un pic de/mult necaz/un noian de necazuri , a sniff of trouble 
semn/ indiciu mic despre apariția unei probleme , a hint of trouble  un semn/
indiciu despre apariția unui necaz, mutually exclude each other in a text.
As regards the second co -selection category, i.e. colligation, Sinclair
(1998) defines it as “the co -occurrence of grammatical phenomen a and, on the
syntagmatic axis, is the co -occurrence of a member of a grammatical class with
a word or phrase” (Sinclair 1998: 15) In this respect, a good example may be
That sniff of trouble he had caught a couple of days before came true. (see
2.1.4 .)
Finally, semantic preference “is the restriction of regular co -occurrence
to items which share a semantic feature relevant to both syntagmatic and
paradigmatic phenomena” ( id.: 16). This category of co -selection requires that
speakers should notice similar ity of meaning regardless of word class. For
example, the meaning of cost plus used in collocations such as cost plus
charging , charging on a cost plus basis and charge for the work on a cost plus
basis is shared by all the collocations, irrespective of it s position and function in
the collocation. The same holds valid in the Romanian equivalent collocations
where the meaning of cost plus  ‘cost p lus’ = sistem de calcula re a unui preț
care include costul de producție sau de prestare a unor servicii la care se adaugă

un procent pentru a acoperi cheltuielile de regie și marja de profit’ (Ivanovic
and Collin 1996: 90) is assumed to be traced by the speakers in all the
colloc ations: cost plus charging/ charging on a cost plus basis  taxare cost
plus = sistem de taxare în care cumpărătorul plătește costurile, plus un comision
procentual vânzătorului’, charge for the work on a cost plus basis  a cere cost
plus pentru munca efe ctuată.
As regards words , paradigmatic choices determine the information given
by words, which are, at the same time, part of the realization of larger items at
the syntagmatic level, but if words, as a paradigmatic choice, have an
independent meaning, th is is not valid when words are selected from the
syntagmatic axis. Consequently , the conclusion c an be drawn that the
appropriate interpretation and description of word meaning should be made only
by a model reconciling the paradigmatic and syntagmatic dim ensions (see 3.4.).
Aspects such as those presented above prove that the concept of meaning
is very important in the study of lexical patterns , in general, and of collocations ,
in particular. The numerous types of word patterns existing in a language may
be better understood and differentiated if speakers are able to compare and
contrast the semantic features of the individual words making up such patterns
and the ways in which the interactions between these individual words affect the
meaning of the lexic al patterns taken as a whole.
An analysis of the possible approaches to the semantics of expressions
(Lyons 1995: 40) identifies six theories of meaning , namely:
1. The referential or denotational theory of meaning according to
which the meaning of an expre ssion is what it refers to (or denotes), or
stands for;
2. The ideational or mentalistic theory which defines the meaning of
an expression as the idea, or concept, associated with it in the mind of
anyone who knows and understands the expression;
3. The behaviou rist theory which states that the meaning of an
expression is either the stimulus that evokes it or the response that it
evokes, or a combination of both on particular occasions of utterance;
4. The meaning -in-use theory which views the meaning of an
expressi on as determined by, if not identical with, its use in the language;
5. The verificationist theory according to which the meaning of an
expression, if it has one, is determined by the verifiability of the
sentences, or propositions containing it;
6. The truth -conditional theory which states that the meaning of an
expression is its contribution to the truth -conditions of the sentences
containing it.
As regards Lyons’ definition of meaning , it combines aspects envisaged
by some of the theories presented above. In h is opinion, meanings may be
defined as “ideas or concepts which can be transferred from the mind of the

speaker to the mind of the hearer by embodying them in the forms of one
language or another” (Lyons 1981: 136).
An important distinction referred to by Lyons (1981) is that between
word meaning and sentence meaning . Fully aware of the fact that these two
types of meaning are closely related and condition each other to a great extent,
Lyons (1981: 139 -140) suggests that the meaning of a sentence depends o n the
meaning of its constituent lexemes, whereas the meaning of some, if not all
lexemes, depends on the form of the sentence in which they occur. Moreover,
he states that, although words normally have individual meanings when used in
sentences, and sent ence meaning results from the sum of the individual word
meanings, things are different in the case of the more or less fixed lexical
patterns in which words most often lose their semantic independence.
Such lexical patterns are agreed to function as sema ntic constituents
whose meaning is added to the meaning of the other sentence constituents. As
suggested by Cruse (1986),
“any constituent part of a sentence that bears a meaning which combines with the
meanings of the other constituents to give an overal l meaning of the sentence
may be termed a semantic constituent ” (Cruse 1986: 25).
Nevertheless, he explains that if the semantic constituent envisaged
cannot be decomposed into more elementary semantic constituents, reference
will have to be made to a minimal semantic constituent, i.e. to a form plus
meaning complex.
The distinction between semantic constituent and minimal semantic
constituent finds some points of similarity in Lyons’ (1995: 49) division of
language expressions into lexically simple expres sions or lexemes such as feel
keenly/ strongly/ deeply  a crede cu toată tăria , feel dimly/ vaguely  a simți
vag, true feelings  sentimente adevărate/ autentice , true nature  fel de a fi ),
and lexically composite expressions such as cut ground from under smb.’ s feet
a-i tăia cuiva craca de sub picioare/ reduce șansele de reușită/ succes , feel as
snug as a bug in a rug  a se simți cât se poate de bine/ comfortabil/ comod/
plăcut ) which are built according to the grammatical rules of the language.
If the meaning of the lexically composite expressions may be
determined on the basis of the meanings of the component elements, the
meaning of some expressions cannot be determined by the sum of the meanings
of the component elements, their semantic interpretation being strictly
idiomatic. Such expressions are call ed idiomatic phrasal lexemes and may be
exemplified by fixed word combinations with idiomatic meaning such as red
herring  pistă falsă , ill-feelings  resentiment , true bill  verdict de punere
sub acuzare , true love  iubit, ibovnic, amant , or true penny (fig.)  om dintr -o
bucată/ băiat de zahăr .

As regards collocations , the idea is shared that in such lexical patterns the
word becomes associated with a meaning through its repeated occurrence in
similar contexts, which leads us to the idea that the choice of a meaning which
is correlated with adjacent meanings has a strong effect on the co-text and
context . As Sinclair (2004) puts it,
“the meaning of a word and parts of its immediate context become inseparable
and the repeated usage actually fuses the sense and the expression. When this
involves the recurrence of certain sequences of words, it is said to be an idiom”
(Sinclair 20 04: 161).
A relevant point in this respect is made by Cruse (1986: 37). He criticizes
the traditional definition of an idiom as “an expression whose meaning cannot
be inferred from the meanings of its part s” by pointing out that reference is
made in this c ase to the meanings that the parts carry not in that expression, but
in other expressions, which have to be chosen carefully. He argues that in order
to apply the definition, speakers must be able to distinguish idiomatic from non –
idiomatic expressions. Th e point made by Cruse may be modified to apply to
the case of ‘meaning through collocation’. The argument that the collocational
patterns of a lexical item define its meaning can be reversed, by arguing that it
is the particular meaning(s) of an item which determine its collocational range.
This idea is shared by Lyons (1977: 613) who argues that
“we must not go from one extreme of saying that the collocations of a lexeme
are determined by its meaning or meanings […] to the other extreme of defining
the m eaning of a lexeme to be no more than the set of its collocations ”.
In addition there are a number of more moderate views lying between the
two extremes. The less favourable views to the idea that collocability defines
meaning suggest that collocative mean ing should be invoked “only when an
explanation in terms of other categories of meaning does not apply” (Leech
1981: 17), so much the more that the collocative meaning is only an
idiosyncratic property of individual words. Closer to the other extreme,
McIntosh (1961: 330) and Nattinger & DeCarrico (1992: 181 -182) argue that
not only does the collocational range of a lexical item contribute to its meaning,
but also that its collocates “share meaning to a great extent”.
Referring to the relation between the elements of a collocation , Sinclair
(1991: 115 -116) argues that the relative frequency of the node and collocate
determines whether the collocational relation will contribute to the meaning of
the node. In his opinion, the more interesting collocations are those in which a
node contracts with less frequent collocates (‘downward collocation’), because
these collocates allow a better semantic analysis of a word (see 2.1.1 .).
Different from Sinclair, Nattinger & De Carrico (1992) see a reverse
relation betwee n the predictability of the collocates associated with a certain
node and the meaning the node derives through the collocational relation. In

their view a fixed “mutual expectancy results in loss of meaning because of
elimination of an element of choice” ( Nattinger & De Carrico 1992: 20).
At this point, mention should be made that although “the form of a
linguistic unit and its meaning are two perspectives on the same event” isolated
words “are not the best starting point for a description of meaning, becau se
meaning arises from words in particular combinations” (Sinclair 2004: 148).
The idea is shared by Nicolas (1995: 234) who considers that the
meanings of words in combinations represent a good starting point in
identifying the different degrees of semant ic compositionality characterizing
word combinations. He states that in the case of free combinations, the meaning
construction is compositional (i.e. the meaning of the word combination is
composed from the meanings of its parts), in the case of collocati ons it is semi –
compositional (i.e. one item preserves its independent meaning, whereas the
other changes its standard semantic content), and in the case of idioms it is non-
compositional (none of the component words preserves its free -combination
meaning).
Reference should also be made that in ordinary language, most words
gradually change their meanings by being frequently associated with other
words, “while in terminology every attempt is made to keep the meaning of
terms constant, in other words to sema ntically isolate the terms and to counter
the natural pressure of language” (Sinclair 2004: 150 -151).
Although in a synchronic view of language, the origins of meaning are
not under scrutiny, some of the processes of semantic change are obvious, new
meanin gs being constantly created, mainly through gradual movements of
collocation . Occasionally a word – meaning relationship is given the status of
term, and thus the measure of protection within the discourse of a specialized
genre (Sinclair 2004: 160). Relev ant examples to illustrate this point could be
collocations such as: corporate embezzlement  deturnarea fondurilor
companiei, corporate liability  responsabiliate juridică , cost and freight 
cost și transport , cost book  caiet de sarcini , cost of funds  dobândă
negativă , court of assets  instanță specială numită pentru precizarea
drepturilor , credit balace  sold creditor/ existent în cont .
Variou s as they may be, the approaches presented above point out some
very important aspects regarding the relationships existing between words and
meaning . On the one hand, word meaning and sentence meaning are obviously
interrelated and condition each other. O n the other hand, if reference is made to
word meaning in different lexical patterns, two situations may be identified: one
in which the meaning of the lexical patterns is the sum of the meanings of the
words making up that pattern, and another in which th e lexical pattern allows
only the idiomatic interpretation. If in the former situation the minimal semantic
constituent is the word, in the latter this function is attributed to the idiomatic
pattern taken as a whole. Since the meaning of words depends on how they are

combined into phrases, and how they are used in social situations, the
conclusion can be drawn that their meaning depends both on linguistic
conventions and on inferences of the real -world knowledge.
Last but not le ast, the lexical and semant ic relations of a word with other
words that accompany it in the stream of speech or writing, i.e. the syntagmatic
lexical relations, are of utmost importance in semantic interpretation. Such
relations arise from the combinations of words in phrases and se ntences and are
concerned with the individual lexical units and the meaning relations they enter
with other accompanying lexical units. In other words, the different meanings of
a word are primarily determined by the nature of the contexts in which such
words may occur, because apart from these contexts, words do not have
meaning, but only potentiality to occur in certain types of combinations.

3.3.1 . CLASSIFICATIONS OF MEANING

Meaning is agreed to be relevant, both at the lexical and the grammatical l evels.
Although the grammatical processes play an important role whereby units of
lexical meaning are organized, modified, and tailored to requirements, they do
not initiate meaning; they act upon meaning already lex ically provided
(Widdowson 2000: 54 -55). Nevertheless, t he intriguing and multi -faceted nature
of meaning is greed not to be limited to linguistic aspects. Extralinguistic
aspects, such as cultural and social specificity, also condition meaning and these
aspects are as important as the lingusit c ones since they may hinder appropriate
semantic interpretation of words and of the messages they make up.
In his semantic approach to language, Lyons (1977: 50 -51) distinguishes
three types of meaning , namely descriptive , expressive and social . The fir st
type, i.e. descriptive meaning , (also called referential/ cognitive/ propositional/
ideational/ designative) is factual in the sense that it can be explicitly asserted or
denied and, in the most favourable instances, it can be objectively verbled.
Expre ssive meaning , on the other hand, varies with the characteristics of the
speaker, hence its being also called emotive/ attitudinal/ interpersonal/ or
expressive. Finally, social meaning serves to establish and maintain social
relations.
Preserving some of Lyons’ terms, Leech (1990: 26 -27) identifies seven
types of meaning :
1. Conceptual meaning (logical, cognitive or denotative content) is
generally considered to be the central factor in linguistic communication
and very important for the functioning of langua ge. The conceptual
meaning of any lexical item is always a combined meaning of the word
or idiom and the context, thus, the relevant level of semantic analysis is
the word or idiom in context.

2. Connotative meaning is related to what is communicated by virtu e of
what language refers to. In other words, connotative meaning is “the
communicative value an expression has by virtue of what it refers to, over
and above its purely conceptual content. When talking about
connotation , one talks, in fact, about the real world experiences one
associates with an expression when one uses or hears it” (Leech 1990:
27)
3. Social meaning refers to what language conveys about the social
circumstances in which it is used. The reader decodes the social meaning
of a text by recogniz ing different dimensions and different levels of style
in language. Language may vary according to dialect (geographical
region, social class), time, province13 (the language of law, science,
advertising), status (polite, colloquial, slang) modality (langua ge of
memoranda, lectures, jokes), singularity (style of a given author). Social
meaning can also include the illocutionary force of an utterance. For
example, a sentence such as I haven't got a knife has the form of an
assertion, but in social reality, it can express a request, the meaning
implied being Please, bring me a knife! .
4. Affective meaning is an illustration of how language reflects the
personal feelings of the speaker, including his attitude towards the
listener or his emotions toward something h e is talking about. Affective
meaning is a parasitic category, in the sense that, in order to express our
emotions, one relies on the mediation of other categories of meaning:
conceptual, connotative or stylistic.
5. Reflected meaning is what is communicated through association with
another sense of the same expression.
6. Collocative meaning consists of the associations a word acquires on
account of the meanings of words which tend to occur in its environment:
'pretty', 'handsome' , and 'good -looking, are synonym s, but collocate with
different words.

13 Leech uses the term province for domain / field . “A register of a lang uage is
identified and described, and then referred to as a discrete set of linguistic choices, seen
as quite separate from the rest of the language. Thus, one can speak of ‘the language of
science’, ‘the language of medicine’ ”, etc. (Robinson 1991: 20 in Croitoru 1996: 90).
The constituent components of register also called dimensions of diatypic variation are:
field (topic), mode (whether written or spoken) and tenor (style). Thus, field is
concerned with the purpose and subject -matter of communication; the overall term,
register, is sometimes used for the component field . Mode refers to the means by which
communication takes place, i.e. by speech or writing. Tenor refers to the formality and
social relations between participants.

7. Thematic meaning refers to what is communicated by the way in
which a speaker or writer organizes the message, in terms of ordering,
focus and emphasis.
Starting from the assumption that the internal and external fact ors
conditioning meaning have to be explored in order to derive a discourse from a
text, Verdonk (2008) distinguishes two types of meaning: namely semantic and
pragmatic and explains that the former is encoded in the lan guage of texts,
while the latter is visible in discourse. The important point he makes is that
“pragmatic meaning is not an alternative to semantic meaning , but
complementary to it, because it is inferred from the interplay of semantic
meaning with context” (Verdonk 2008: 19). The idea is s hared by Widdowson
(2000: 63 ) who states that “ the achievement of pragmatic meaning is a matter
of matching up the linguistic elements of the code with the schematic elements
of the context” .
Meaning is very important in stylistics as well. As Galperin (1 977: 58)
suggests, this is due to the fact that the term meaning applies not only to words,
word -combinations and sentences, but also to the manner of expression into
which the matter is expressed. If in general linguistics meaning is generally
regarded as something stable at a given period of time and it is often contrasted
with terms such as concept , sign and referent , in stylistics meaning is seen as
a category able to acquire meanings imposed on the words by the context, or
contextual meanings. Stylisti c meaning is contrasted with lexical and
grammatical meaning. The former, i.e. lexical meaning, refers to some
concrete concept, phenomenon, or thing of objective reality, whether real or
imaginary, being thus a means by which a word -form is made to expres s a
definite concept, whereas grammatical meaning refers to relations between
words or to some forms of words or constructions bearing their structural
functions in the language -as-a-system, being thus called “structural meaning”
Considering the stylistic approach to words, Galperin (1977: 64) refers to
three types of meaning, namely, logical , emotive, and nominal , as a derivative
of logical meaning. In his opinion, logical meaning , also called referential or
direct, is the precise naming of a feature of t he idea, phenomenon or object, the
name by which we recognize the whole of the concept. On the other hand,
emotive meaning materializes the concept in the word, but, unlike logical
meaning, it has reference not directly to things or phenomena of objective
reality, but to the feelings and emotions of the speaker towards these things or
his emotions as such. In other words, emotive meaning refers to things,
phenomena, or ideas through a kind of evaluation of them. If words acquire an
emotive meaning only in a definite context, reference is made to contextual
emotive meaning .
Last, but not least, nominal meaning is a derivative logical meaning
which does not imply singling out one of the objects of a class for a particular

occasion, but to make them the bearer s of the properties which our mind has
attached to them.
3.3.1.1. A distinction often made in lexico -semantic and stylistic studies
is that between denotative and connotative meaning, or between denotation
and connotation .
Denotative meaning14 refers to t he relationship between a linguistic sign
and its denotatum or the referent, whereas connotative meaning15 implies
additional properties of lexemes such as poetic, slang, baby language, biblical,
casual, colloquial, for mal, humorous, legal, literary or rhetorical.
Moreover, starting from the generally shared idea that words can have the
same denotation but different connotations (e.g. to die – neutral, pass away –
sympathy, snuff – no sympathy), Stubbs (2002: 34) states that denotation,
which is often referr ed to as cognitive, conceptual, logical, ideational and
propositional meaning is contrasted with connotation which is called affective,
associative, attitudinal, and emotive meaning. He adds that denotation means
the appropriate range of reference of a wor d and it is “a relation between a term
in the language and a range of potential reference in the world” (Stubbs 2002:
34). On the contrary, connotation is sometimes “thought of as personal or
emotional associations, conveying the attitude of an individual speaker” (Stubbs
2002: 35), but in spite of their apparently limited availability, connotations are
often widely shared within speech communities. Such characteristics of
connotations are also mentioned in linguistic studies, which have often
regarded conn otational meanings as ‘unstable’ and ‘indeterminate’ (Leech
1981: 13), or limited to the ‘personal or emotional associations which are
suggested by words (Crystal 1992: 80) .
If regarded from the perspective of stylistics , denotation is a stylistically
neutral and objective relation between a word and the world, the basic or core
meaning of the word, which is not deniable, wh ereas connotations are often
regarded as subjective, second -order or peripheral meanings, which depend on a

14 Denotative meaning corresponds to the Italian significato di base / denotativo/
referenziale/ concettuale which is defined as “il significato che corrisponde al
‘concetto’, che nulla ha a che fare con il singolo ogetto, ma si riferisce all’insieme delle
caratteristiche che i pa rlanti associano a quest’ultimo” (Marrinucci 1996: 446) . In other
words, “la denotazione è il rapporto stabilito tra la parola e il referente’ (Dardano e
Trifone 2006: 18).
15 Connotative meaning corresponds to the Italian significato aggiunto/ connotativ o/
non-referenziale and it is defined as “quello che il significato ha di ‘particolare’,
‘distintivo’ per l’individuo singolo o per la communità, spesso determinato dal contesto
sociale in cui la lingua vienne utilizzata” (Marrinucci 1996: 447). To put it differently,
“la connotazione è l’insieme dei valori affettivi che circondano la parola; valori che
possono mutare il passagio da un parlante all’altro , dall’una all’altra situazione”
(Dardano e Trifone 2006: 18).

relation between the w ord and the speaker/hearer, as well as on intertextual
norms . Moreover, since connotations belong to evaluative language, the
evaluative meanings are often inexplicit and less clear -cut, which explains why
they are often used in persuasive language.
Sincl air also refers to denotative and connotative meaning stating that
the former is regarded as ‘classificatory’ and places the envisaged word in
relation to other words and concepts, whereas the latter is “an occasional
additional feature of a word’s meaning , a ‘shade of meaning’, carrying perhaps
a negative orientation or a feeling of informality” (Sinclair 2004: 121 -122).
As regards denotation, mention should be made that certain specialists
contrast this semantic reality with concepts such as reference and sense .
3.3.1.2. Lyons explains that if denotation refers to the relationship that
holds between a lexeme and persons, things, properties, processes and activities
external to the language system, reference is the relationship which holds
between an expre ssion and what that expression stands for on particular
occasions of its utterance (Lyons 1977: 207). Moreover, reference depends on
concrete utterances, not on abstract sentences , being a property of expressions,
not of single word forms and single lexeme s (Lyons 1977: 197).
In a more recent study, Lyons (1995: 79) suggests that the crucial
difference between reference and denotation is that the denotation of an
expression is invariant and utterance -independent, in other words it is part of the
meaning wh ich the expression has in the language system, independently of its
use on particular occasions of utterance, whereas reference is variable and
utterance -dependent.
Considering the same opposition, Stubbs (2002: 34) points out the fact
that reference conc erns language use because it is a speech act which picks out
a referent in a concrete situation, whereas denotation concerns the language
system, being a relation between a term in the language and a range of potential
referents in the world.
Last, but no t least, Yule (1996 : 21) suggests that the analysis of
reference depends on local context and the local knowledge of the participants,
just as it may depend on familiarity with the local socio -cultural conventions as
the basis for inference. Since these c onventions may differ substantially from
one social group to another, and may be marked differently from one language
to another, reference should not be regarded simply as a relationship between
the meaning of a word or phrase and an object or person in the world. Rather, it
should be regarded as a social act, in which the speaker assumes that the word
or phrase chosen to identify an object or person will be interpreted as the
speaker intended.
3.3.1 .3. Besides reference , denotation is often contrasted wi th sense in
lexico -semantic studies.

The term sense refers to the place of a word in a system of relationships
which it contracts with other words in the vocabulary. As Lyons (1977 ) puts it,
sense is a relationship “between the words or expressions of a s ingle language,
independently of the relationship, if any, which holds between those words or
expressions and their referents or denotata” (Lyons 1977: 206) . Thus, sense is a
relationship which is internal to the language system, different from denotation ,
which is external to the same system due to its being a relation which holds
primarily, or basically, between expressions and physical entities in the external
world (Howard and Zé Amvela 2000: 56).
Both denotation and sense apply equally to lexically si mple and lexically
composite expressions, but the sense and denotation of a composite expression
is a compositional function of the sense and denotation of its component parts.
As regards the sense of an expression, it is rep resented by the set, or network of
sense -relations which hold between a lexical expression and one or more other
lexical expressions in the same language (Lyons 1995: 80 -81). Connotation is
also interrelated with various types of expressions, being especially noteworthy
in restricted collocations . The activation of connotation in such collocations is
tied to the cultural information contained in the base of the collocation (often a
noun) and the nature of the semantic specialization of the collocate (often a
modifier).
Furthermore, denot ation and connotation are often referred to by the
alternative terms literal , and figurative or metaphorical , respectively.
Nevertheless, these terms are not interchangeable. Even though an item may
have both denotative and connotative meaning at the same time, the categories
literal and figurative are mutually exclusive. Literal meaning like denotation,
suggests that there is a simple and direct link between the word and something
in the world that it signifies, whereas in the case of figurative meaning the link
between the signifier and the signified can only be made if some imaginative act
of interpretation is performed.
Referring to the same distinction, Sinclair (2004: 122) suggests that the
literal / denotative kind of meaning is seen as central and o bligatory, whereas
the connotative/ figurative type of meaning is optional and rather
unpredictable. This situation arises from the status of word as the presumed
carrier of meaning , whereas the lexical item is characteristically phrasal,
although it can b e realized in a single word.
The imprint of cultural studies at the level of meaning is visible in the
category of cultural connotations which may arise, as Cowie (1998: 9)
suggests, from the interpretation of concepts and subconcepts (see 2.4.3 .). For
example, a subconcept such as loose woman may be traceable in a broad
spectrum of idioms , collocations and proverbs , but the exact nature of the
cultural connotations may be revealed only if an entire domain or field is
studied.

Furthermore, connotation is a key term in stylistic studies due to the fact
that many set expressions have specific stylistic properties adapted to the value
judgements either of a speech community (i.e. class or social group) , or of an
individual speaker or writer.
Referring to th e connotations carried by phraseological units , Gläser
(1998: 128) mentions that they are described as ‘usage labels’ or ‘style markers’
in lexicology and lexicography, adding that distinction is usually made between
‘expressive’ and ‘stylistic’ connotatio ns, on the one hand, and between both of
these and ‘register markers’, on the other.
Moreover, she enlarges upon expressive and stylistic connotations. The
former category includes, in her opinion, the lexicological markers ‘derogatory’
(e.g. to breed lik e rabbits, mouton dressed as lamb ), ‘taboo’ (e.g. get stuffed,
son of a bitch ) , ‘euphemistic’ (e.g. the great divide ), and ‘jocular/
humorous’(e.g. to have a bun in the oven ) or ‘facetious’, whereas the latter
covers indicators of different stylistic leve ls. In addition, m arkers such as
‘colloquial’(e.g. green fingers, full of beans, fine and dandy ), and ‘slang’(e.g.
to kip down, on the never -never ) are used for the stylistically lowered phrases,
the elevated stylistic level being represented by the marke rs ‘formal’(e.g. the
compliments of the season, a bone of contention, to be the question ),
‘literary’( irretrievably lost, hermeneutically sealed ), ‘archaic’ (e.g. in days of
yore, as it came to pass, thou shalt not kill ), and ‘foreign’ (e.g. sine qua non,
carte blanche, comme il faut ). Last but not least, register markers appear in the
dictionary as references to a particular field or province of discourse (e.g.
astronomy: black hole, red giant , economics: a high flier, idle funds,
intermittent dumping , judicial: burden of proof, persona non grata , medical:
benign tumour, Caesarian section, pepper -and-salt fundus (= fundus oculi)
(Gläser 1998: 129, her examples)
To conclude, the meaning typologies presented in this section and the
distinctions made between widely used semantic categories ( denotation vs.
sense , denotation vs. reference , denotation vs. connotation ), prove as
challenging, as important and useful in studying the semantic and stylistic
specificity characterizing each of the various types of lexic al patterns in a
language. As regards collocations , the distinction between expressive and
stylistic connotations is of utmost importance for the stylistic approach to
such lexical patterns (see 4.3 – 4.6), as it brings to the fore relevant distinctions
illustrative for the use of collocations in different types of texts and in different
situations.

3.3.2 . SENSE RELATIONS

Starting from the generally shared idea that collocations are fixed patterns
formed on the basis of lexically and semantically con ditioned co -occurrences,
reference should be made that the members of a collocation are closely
interrelated to each other, and any replacement in the structure of the
collocation is most often semantically conditioned. In other words, neither of
the collo cation members can be replaced unless certain semantic criteria are
fulfilled. Moreover, replacements in the structure of a collocation will only be
possible if the excluded member and the word taking its place are in some sense
relationship, i.e. they are either synonymous, or antonymous or one of the two
words is a hyperonym of the other. This explains why sense relations , in our
case synonymy, antonymy and hyponymy, are particularly useful for the
semantic interpretation of collocations and shou ld be fur ther explored in the
present study.
3.3.2 .1. If reference is made to synonymy , this is “a bilateral or
symmetrical sense relation in which more than one linguistic form can be said
to have the same conceptual or propositional meaning” (Carter 1998: 20 -21).
However, as Stubbs (2002: 36) states , although synonyms are defined as words
which mean the same, there are hardly any cases of perfectly equivalent words
in both their denotation and connotation. For example, spectacles and glasses
are synonyms, but spectacles is more formal than glasses . Similarly, the neutral
noun body is a synonym of the nominal groups the deceased , the corpse , the
stiff and cadaver , but each of these synonyms will be used only in specific
contexts due to their semantic and stylistic restrictions. The deceased will be
used with the meaning ‘recently died’ and when the speaker s want to express
their respect, the corpse will be selected when the semantic features
‘unpleasant’ and ‘report of a crime’ are envisaged, and the cadaver will be used
in medical contexts.
Marrinucci’s definition of synonyms contradicts Stubbs’ point of view.
In his opinion, synonyms are those lexemes, which, in spite of having a
different form, have the same meaning and are interchangeable in the same
context 16 (Marrinucci 1996: 451) (emphasis added , my translation ).
Depending on the degree of semantic similarity between words, reference
is most often made to three types of synonyms : 1. near -synonyms (expressions
which are more or less similar, but not identical in meaning); 2. partial
synonyms (expressions identical in meaning but not meeting the conditions of
absolute synonymy) and 3. absolute synonyms (a relation easily established
between lexically composite expressions). As regards the third type of

16 I sinonimi sono ”quei lessemi che, p ur essendo diversi nella forma, hanno lo stesso
significato e sono intercambiabili nello stesso contesto” (Marrinucci 1996: 451).

synonyms , they occur only if: a) all their meanings are identical, b) they are
synonymous in all contexts and c) they are semantically equivalent (i.e. their
meaning or meanings are identical) on all dimensions of meaning, descriptive
and non -descriptive.
The dist inction between partial and full synonymy is also mentioned by
Zdrenghea (1977: 99) in his approach to English semantics. He considers that
full or absolute synonymy implies that synonyms can exchange places in all
contexts, whereas partial synonymy implies that the units affected by this
sense relation are synonymic with regard to one of the senses exhibited by the
words. Moreover, he mentions the distinction between factual and cognitive
synonymy which are traceable beyond the semantic level of language.
Referring to approximate synonyms , Stubbs (2002: 37) suggests that
they are distinguished partly by their denotations, but also by their connotations
and by the text -types in which they typically occur.
Moreover, Lyons argues that the collocational range of an expression is
very important because it clarifies the difference between synonymy and near –
synonymy , hence two synonyms must have the same collocational range: “It
might be thought that the collocational range of an expression is wholly
determined by its meaning, so that synonyms must have the same collocational
range. But it does not seem to be so… there must be some subtle difference of
lexical meanings which accounts for the collocational differences such th at it is
not synonymy, but near -synonym y that is involved” (Lyons 1995: 62).
Therefore, two words may have quite similar senses, but their collocability, i.e.
co-occurrence restriction, may not always be the same. For example, the
adjective spectacular can modify the nouns waterfall or landscap e, but not the
nouns disaster or crash which collocate with terrible . This is an argument to be
set forth in favour of the dependence of meaning on context. The meaning of
words can be regarded as “a pattern of affinities and disaffinities with all the
other words in the language” (Lyons 1995: 62). Thus, meaning is closely
connected with the syntagmatic and paradigmatic relations between items
within the same linguistic system, and the context for the definition of meaning
is internal to language.
An inter esting and more extensive approach to synonymy is suggested by
Levițchi (1970: 85) who s tates that “synonyms are two or more lexical or
grammatical units comparable through their content, but reflecting in various
degrees and in various senses (semantic, g rammatical, stylistic) the essential
notes of the notion they denote.” From this perspective he makes a distinction
(Levițchi 1970: 86 -88) between absolute and relative synonyms, the latter
being divided into lexical and grammatical , which in their turn ar e each
ideographic or stylistic .
Absolute or perfect synonyms have the same semantic and stylistic
value, associate with the same words, and have the same grammatical

government. e.g. oak nut , oak-pear etc. Relative synonyms may be either
lexical or gramm atical , each of the two categories being further divided into
ideographic and stylistic.
Ideographic lexical synonyms imply either certain semantic differences,
or the accentuation of certain notes characteristic of the respective notion,
phenomenon etc.: e.g. enclosure (the most general term, i.e. the dominant
synonym), fence , hedge , wall. Specialized synonyms are a variety of
ideographic lexical synonyms , expressing the relation between genus, for
example, pen, and species, i.e. fountain -pen.
Stylistic lexical synonyms express the same meaning while differing by
their appurtenance to various functional styles. Disguised synonyms are based
either on a figure of speech or on some word or word group characterizing an
object in various ways, mainly on a styl istic basis, (e.g. Shakespeare – the sweet
swan of Avon). Euphemisms are lexical units considered to be less distasteful
or less offensive than other lexical units expressing the same thing (see 4.5.2 ).

Fig. 9 . Classification of synonyms (source Levițchi 1970: 86)

Similarly to lexical synonyms, the grammatical ones are divided in two
classes, namely ideographic and stylistic .
Ideographic grammatical synonyms are characterized by semantic –
grammatical differences (as to aspect, duration, number, time , or as to
anaphoric, epiphoric or generic elements etc.)
Stylistic grammatical synonyms differ both as to form or function and
appurtenance to the functional styles of the language or simply as to the modal
connotations of the communication (attitudes of sympathy, hatred etc.), e.g. the
historic (dramatic) present used instead of the preterite for the sake of vividness,
etc.
As suggested by Zdrenghea (1977: 99), the synonymy relation is very
important for the study of lexis as it brings to the fore cons tructions in which
words are used in assimilated functions, diverging into constructions in which
the same words are used with different meanings. Moreover, in discussing
synonymy , the fact must be born in mind that two or more words are not

synonyms const antly and in all occasions, and synonymy varies not only in
time, but also under the circumstances. In addition, synonyms assure an
expressive variety of language as opposed to the rigid formulations.
Considering translation -related aspects, Levițchi (1970 ) states that t he
importance, of a good command of synonymy not only in one but also in two or
more languages is a prerequisite in translation , because translating means to
find “the most adequate equivalents, both ideographic and stylistic from a
possible synonymic series” (Levițchi 1970: 91)
3.3.2 .2. The relationship of antonymy is the semantic opposition or
unrelatedness. The sense relations which may be distinguished, in addition to
the proper relation of antonymy, are the following:
a. complementarity – the presence of a sense component excludes
another: e.g. cheerful atmosphere ≠ sad atmosphere, torturous memories
≠ pleasant memories, to get on ≠ to get off, to pay high ≠ to pay low, to
take to/up smth. ≠ to give up doing smth. ;
b. converseness – contrastive lexical relations where there is a measure of
logical reciprocity and interdependence of meaning: e.g. to send smb.
mad ≠ to calm smb. down, to send smb. into great excitement ≠ not to
affect smb. at all, laughing matter ≠ serious matter, a praise -worthy
person ≠ a laughing stock ;
c. incompatibility – relational contrasts between items in a semantic field
(words which co -occur with reference to a familiar topic): e.g. to laugh
in smb.’s face ≠ to laugh up one’s sleeve, to need su pport ≠ to stand on
one’s own feet (again) ;
d. antonymy – used as an inclusive term but also in relation to opposition,
i.e. in the more restrictive sense of gradable opposites: e.g. to do smb.
good ≠ to do smb. bad , mean dog ≠ mild dog , feeble voice ≠ loud voice ,
harsh tone ≠ soft tone etc.
Some specialists consider that the term antonymy is hardly precise and
replace it with oppositeness . There are different types of lexical opposites, one
possible distinction being that between gradable17 and ungradable opposites.
“Gradable opposites presuppose that the predication of the one implies the
predication of the negation of the other: e.g. ‘X is hot’ implies ‘X is not cold’ and
‘X is cold’ implies ‘X is not hot’, but ‘X is not hot’ does not necessarily imply
that ‘X is cold’. In the case of ungradable opposites , the predication of either

17 In Italian the distinction is made between ‘ antonimi bipolari ’ (maschio/femina,
vivo/moro ) and ‘ antonimi graduabili ’ (alto/b asso, caldo/freddo ) which may sometimes
include an intermediary term ( caldo – tiepido – freddo ). In addition, antonimi conversi
indicate the same meaning but from a different perspective ( figlio – padre; vendere –
comprare ). The distinction between lexical (buono – cattivo, bello/brutto ) and
grammatical antonyms i.e by means of prefixes ( omogeneo – disomogeneo, educare –
diseducare ) is also important (Dardano and Trifone 2006: 20).

one of the pairs implies the predication of the negation of the other, but also the
predication of the negation of either implies the predication of the other. For
example, ‘X is female’ implies ‘X is not male’ and ‘X is not male’ implies ‘X is
female’”(Lyons 1987: 271 -272).
Another possible distinction is that between privative and equipollent
opposition . The former is a contrastive relation between two lexemes one
which deno tes some positive property and the other which denotes the absence
of that property: e.g. deep analysis ≠ surface analysis, to keep one’s faith ≠ to
lose one’s faith, to hold interesting opinions ≠ to hold strange opinions, to hold
the bare truth ≠ to hold a partial truth . The latter is a relation in which each of
the contrasting lexemes denotes a positive property: e.g. male vs. female (Lyons
1977: 279).
An interesting perspective on antonyms18 is suggested by Stubbs (2002 )
who, s tarting from the idea that antonyms are words opposite in meaning,
points out that every word has a core meaning and a prototypica l antonym.
Moreover, he suggests that although antonymy has traditionally been regarded
as a paradigmatic opposition permanently available in the lexicon of a language,
this sense relation is better seen as a syntagmatic relation, which is realized in
the co-text (e.g. dry socks ≠ wet socks , dry season ≠ wet/ rainy season , dry wine
≠ sweet wine , dry skin ≠ moist skin; white coffee ≠ black coffee, white wine ≠
red wine, white collar ≠ blue collar ) (Stubbs 2002: 38) .
Last, but not least, in her approach to f ixed expressions and idioms Moon
(1998: 156) suggests that pairs of such lexical patterns are broadly antonymous
because the variant words realize systematic semantic contrasts. In her opinion,
the simplest cases involve words which are regularly opposed c ounterparts (e.g.
directional antonyms : in/into ≠ out (of), off ≠ on, up ≠ down , or reversives :
keep ≠ lose, from the bottom up ≠ from the top down). Moreover, Moon makes
reference to pairs including contrastive or antonymous words in unaltered
prepositional frames (e.g. in private ≠ in public, on the offensive ≠ on the
defensive, to someone’s credit ≠ to someone's discredit, with a good grace ≠
with a poor grace ).
As regards the Romanian approach to antonymy in English, by slightly
enlarging the usual definitions of antonyms Levițchi (1970 ) states t hat they are
“lexical or grammatical units whose meanings are in a relation of opposition”
Levițchi (1970: 101) . In keeping with this definition, he suggests a possible
classification of antonyms (Levițchi 1970: 102 -104) in absolute and relative ,
each of these two types being lexical and grammatical . The classes of absolute

18 Referring to antonymy in Italian, Marrinucci (1996: 452) states that “ due lessemi
sono antonimi quando i loro significati sono opposti ma ammettono al loro interno una
gradazione di valori, nel senso che l’asserzione del primo non implica affatto una
negazione dell’altro.

lexical and grammatical antonyms are further grouped into explicit and
implicit .
Absolute antonyms are characterized by the relation of utter opposition
between the notions they denote (e.g. white ≠ black), by their independence
from context, as well as by a certain identity of semantic and grammatical
associations. As regards absolute lexical antonyms , they establish a relation of
utter opposition between lexical units, thus between word -mean ings (e.g. to
hate- a urî ≠ to love – a iubi, to hate – a nu-i plăcea ≠ to like – a-i plăcea), between
word -meanings and phrases(e.g. to like – a-i plăcea ≠ to hold in abomination – a-
i fi scârbă de), or between phrases (e.g. to make much of – a pune preț pe ≠ to
make little of – a nu pune preț pe). On the other hand, absolute grammatical
antonyms include grammatical forms (e.g. the young man's – al tânărului ≠ the
old man's – al bătrânului, greater – mai mare ≠ smaller – mai mic, it takes – ia ≠
it is taken – este luat).
Explicit antonyms imply that both opposite terms are expressed, either
outside a context (e.g. white -black) or in a limited context so that the opposition
may be brought out clearly (e.g. They were determined to get on by any means ,
good or bad). On the contrary, implicit antonyms are always contextual and
they are expressed only by one term of the semantic or semantic -grammatical
opposition, the other term or terms being merely suggested. Lexical implicit
antonyms are evoked all by a peculi ar accentuation of a lexical unit in a given
context (e.g. He is a man!) Grammatical implicit antonyms are also
suggested by what is stressed in the utterance (e.g. He had a very good memory
in his youth !).

Fig.10. Classification of antonyms (source Levițchi 1970: 102)

Relative antonyms express partial oppositions between lexical or
grammatical units, these oppositions being in most cases contextual. Lexical
relative antonyms may be expressed by such pairs as painful (causative) merry
(resultative), conduct (permanence) behaviour (limited duration), etc.
Grammatical relative antonyms are also dependent on a context. Thus the
personal pronoun ye and the demonstrative pronoun another, which are not

necessarily in a relation of opposition, may become antony ms within the general
framework of the linguistic context.
The approach suggested by Zdrenghea (1977: 107 -108) is also worth
mentioning. He identifies three main types of antonymy, namely componential ,
logical and referential . Componential antonyms can be subdivided in to
lexical and grammatical , in the case of the former class reference being further
made to absolute and partial antonyms. As regards logical antonymy , it is
conceived in the frame of sentences, the comparison made in this situation
being eit her implicit or explicit . Last, but not least, referential antonymy
represents the dynamic relationship of this semantic phenomenon.
3.3.2 .3. Hyponymy19 is a paradigmatic relation of sense “which rests
upon the encapsulation in the hyponym of some synt agmatic modification of the
sense of the superordinate lexeme. Generally speaking , co-hyponyms of the
same superordinate will contrast in sense and the nature of the contrast will be
explained in terms of a difference in the encapsulated syntagmatic modifi cation
of the superordinate. For example , buy and steal are in contrast as co -hyponyms
of get” (Lyons 1977: 294 -295).
Hyponymy is an inclusive sense relation which exists between specific
and general lexical items , the meaning of the specific item being included in the
meaning of the more general item. (Carter 1998: 20 -21). This sense relation is
obvious both in the case of individual words and of collocations . In addition,
hyponimic relations are identifiable in general, but also in specialized
(con)texts.

Fig. 11 . The hyponymic tree of the noun bank

19 “Tra i lessemi si ha un rapporto di iponimia quando i loro significati sono inclusi in
quello più generico di un altro lessema definito iperonimo (o sovraordinato )”
(Marrinucci 1996: 452).

Fig.1 2. The hyponymic tree of the collocation source of wealth

Fig.13. The hyponymic tree of the collocation violent crime
(source Dumitrașcu and Kania 2008: 111)

Referring to hyponymy from the perspective of synonymy, Widdowson
(2000 : 59) states that:
“each superordinate necessarily possesses a semantic feature
common to all its hyponyms to the extent that each co -hyponym has a
distinct semantic specification which serves as a superordinate to the
next level of classification down, until all distinctive features are
exhausted” .
The conclusion he draws is that synonymy will always be present where
two lexical items occur in the sam e position on the tree as hyponyms. Discussed
from this perspective, synonymy is a semantic relation, but the extent to which
synonyms have a differ ent range of functions when they are actually put to use
in con texts of communication is a matter of prag matics, or meaning in context.
To sum up , a semantic analysis of collocations in terms of specific sense
relations such as synonymy , antonynmy and hyponymy , rather than in terms
of the general relatedness of topic may be very useful in exploring lexical
cohesiveness, so much the more that these relations hold not only “between
words in the vocabulary, but also between word forms in texts where they
contribute to text cohesion” (Stubbs 2002: 50).

3.3. COLLOCATIONS ON THE AXES OF MEANING

Among the possib le approaches to the study of collocations, an important
position is occupied by those studies which analyze collocations and other
(fixed) lexical patterns from the perspective of the two axes of meaning .
Although some of these studies restrict their anal ysis to the category of word,
they are nevertheless important and useful for the better understanding of the
way in which word meaning is affected in the lexical strings created on the
syntagmatic and paradigmatic axes.
One of the specialists approaching the meaning of isolated words on the
two axes is Lyons (1995) who identifies two types of sense relations between
words, namely substitutional (syntagmatic) and combinatorial (paradigmatic) .
In his opinion, the former are “those relations which hold betwee n
intersubstitutable members of the same grammatical category”, whereas the
latter “hold typically, though not necessarily, between expressions of different
grammatical categories which can be put together in grammatically well -formed
combinations” (Lyons 1995: 124). In considering combinatorial relations,
mention should be made that they are conditioned by collocational restrictions.
In the case of highly restricted lexemes, their combinatorial relations are
impossible to predict if they occur separately. Nevertheless, any lexeme,
irrespective of its restrictions as regards collocational acceptability, includes
both substitutional and combinatorial relations. For example, the combinatorial
relations of the adjective stale are difficult to predict if it is s eparated by the
nouns bread (denotative meaning) and news (connotative meaning), and
although its use is conditioned by obvious collocational restrictions, the
adjective stale enters both in substitutional and combinatorial relations with
other words. The replacement of the adjective stale by fresh is an illustration of
the substitutional relation, whereas the replacement of the noun bread by news
exemplifies the combinatorial relation existing between the two words.
A different approach is suggested by How ard and Zé Amvela (2000) who
explore the two axes of meaning from the perspective of the lexical field theory.
Starting from the generally shared idea that the vocabulary of a language is
essentially a dynamic and well -integrated system of lexemes structur ed by
relationships of meaning, they point out that this system is changing
continuously by the interaction of various forces. In addition, they suggest that
the vocabulary of a language “is mainly characterized by the general -particular
and part -whole rel ationships which hold not only between individual lexemes
and the lexical fields within which they are best interpreted, but also between
specific lexical fields and the vocabulary as a whole” (Howard and Zé Amvela
2000: 14).

Fig. 14. Words on the synt agmatic and paradigmatic axes (source Howard
and Zé Amvela 2000: 14)
When any word selected from a given context can be easily related with
other words which resemble in form, in meaning or in both, the relationship
established between the respective words will be called paradigmatic or in
absentia . Moreover, when a certain relationship will be established between
words simultaneously used in a sequence, reference will be made to syntagmatic
relations, or in presentia. The same distinction is made by Palmer (1981) who
states that “the paradigmatic relations are those into which a linguistic unit
enters through being contrasted or substitutable, in a particular environment,
with other similar units” whereas “the syntagmatic relations are those that a unit
contacts by virtue of its co -occurrence with similar units” (Palmer 1981: 67 -68)
Furthermore, Geeraerts (1995) refers to the interaction of words on the
syntagmatic and paradigmatic axes from the perspectives of idioms . He
suggests that “the syntagmatic and p aradigmatic dimensions of idioms are both
twofold in the sense that both can be considered with regard to the original,
literal meaning, and with regard to the derived, figurative meaning” (Geeraerts
1995: 59). Referring to the paradigmatic dimension of idioms, Geeraerts states
that this dimension primarily involves the relationship between the original
meaning of the idiomatic expression as a whole and its derived meaning.
Moreover, he suggests that this dimension also involves the relationship
between the original, literal meaning of the elements making up the idiomatic
expression, and the interpretation given to those parts within the derived reading
of the expression as a whole. As regards the syntagmatic dimension of
idiomatic expressions, Geeraerts bel ieves that it involves the relationship
between the interpretation of the constituent parts of the expression, on the one
hand, and the interpretation of the expression as a whole, on the other. From this
perspective, the conclusion can be drawn that synta gmatic meaning may be
envisaged with regard to both the origina and the derived meaning of idioms.
Last, but not least, reference should be made to Sinclair’s approach to
collocations from the perspective of the syntagmatic and paradigmatic axes of
meanin g. Starting from the idea that the paradigmatic axis specifies the
possible choices at a particular position on the syntagmatic axis, and the

syntagmatic axis controls the structure which is being elaborated, Sinclair
observes that text is syntagmatic and the paradigms are those elements which
might have been chosen instead (Sinclair 2004: 168).
Moreover, considering the levels of grammar and lexis, Sinclair states
that the syntagmatic axis of meaning corresponds to grammar, whereas the
paradigmatic one cor responds to lexis. Such a perspective gives rise to the so –
called slot-and-filler model according to which the syntactic structures forming
a series of slots are filled with choices from the dictionary.
As regards the interaction between the syntagmatic p atterns of language,
Sinclair believes that they are not given meaning in a paradigm grammar, nor
are they given meaning in a dictionary type of lexis. This is due to the fact that
the syntagmatic patterns of grammar are either given as related through a
node, or they are simply declared, whereas the syntagmatic patterns of lexis
only appear in the byway of idiomatic phrases where they are offered as joint
realizations of a single meaningful unit, indicating that they have no meaning in
themselves.
Corpus linguistics studies seem to find a solution to this problem with
their general assumption that meaning is created on both axes. The specialists
adopting this approach believe that there is a balance between the
‘phraseological tendency’ and the ‘ terminolog ical tendency’ of words. If the
former tendency is attested by the fact that speakers are inclined to choose
several words at a time, the latter refers to the fact that language users tend to
protect the meaning of a word or phrase, so that its meaning re mains known
every time it is used.
Instead of a conclusion , the simultaneous observation of pattern and
meaning is possible if both syntagmatic and paradigmatic aspects are taken into
consideration. Moreover, the same interrelation has to be preserved in the case
of meaning and context , because the interaction between these two involves at
least partial co -selection. The absence of one of the two perspectives will hinder
appropriate semantic interpretation. As Sinclair puts it, “the knock -on effect of a
paradigmatic choice will be felt on the syntagmatic axis. If we start from the
other axis, then any existing or proposed pattern of choice on the syntagmatic
axis provides a framework for the interpretation of a ny choice made on the
paradigmatic axis” (Sincl air 2004: 170).

3.4. LEXICAL COHESION, CO -TEXT AND CONTEXT

Since collocations are not used in isolation, but in texts, the semantic approach
to such fixed lexical patterns cannot be restricted to aspects related to the
meanings of the individual words making them up and to the ways in which
these individual meanings are affected in collocation. Consequently, reference

should also be made to the influence these lexical patterns have on the semantic
interpretation of larger text units, i.e. to the cohesio n created by collocations in
texts, and to the ways in which the co-text and context condition the
appropriate semantic decoding of collocations.
To begin with, discussions and models of cohesion are useful, because
they allow effective explorations of tex ts as dynamic processes and of textual
organization.
3.4.1. Cohesion is a term which may be used to denote different realities.
On the one hand, it refers to a process which presupposes the linking of some
element to something that has already been mentio ned or, in certain cases, to
something that is to follow. On the other hand, it refers to a directional relation
in a system which is ordered in the case of such presupposing elements as
reference items and substitutes.
Cohesion is conditioned by co -interpretation which implies that
“elements that are tied by the cohesive relation are interpreted (not identically,
but) as a whole, with mutual dependence or ‘solidarity’ between them” (Halliday
2002: 43). To put it more simply, cohesion “concerns the ways i n which the
components of the surface text are mutually connected within a sequence”
(Carter 1998: 103 -104).
Referring to cohesion , Halliday and Hasan (1976) explain the term by
saying that it “occurs when a speaker realizes the difference between a unifi ed
whole and a collection of unrelated sentences when hearing or reading a passage
of the language” (Halliday and Hasan 1976: 5). Moreover, considering the
structuring of cohesion in three distinct strata i.e. semantic , lexico –
grammatical , and phonological and orthographic , Halliday and Hasan (1976)
state that there is no real division between vocabulary and grammar because
“the more general meanings are expressed through grammar, whereas the more
specific ones are expressed through vocabulary” (Halliday an d Hasan 1976: 10).
3.4.2. As regards collocations , reference has to be made to a specific type
of cohesion, namely lexical cohesion .
Collocations are lexical patterns frequently used in different types of
texts and the idea is shared that they represent a useful instrument in achieving
textual cohesion . This is due to the fact that “chains of repeated and related
words, distributed across the text, in patterns of old and new information”
(Stubbs 2002: 124) make a text comprehensible, thus, lexically cohesi ve.
Nevertheless, when analyzing a text with respect to its lexical cohesion ,
the most important thing is to use commonsense, combined with the knowledge
that the speakers of a language have with respect to the nature and structure of
its vocabulary. Alth ough speakers can ignore repetitive occurrences of fully
grammatical items, and lexical items of very high frequency, they cannot ignore
words which occur in restricted patterns of collocation.

Halliday and Hasan (1976) suggest that lexical cohesion can be achived
by means of reiteration and collocation (see the descr iption of lexical cohesion
in Figure 1 5). Although these two terms overlap, they are not identical. The use
of general words as cohesive elements is a special case of reiteration, which
impli es the repetition of a lexical item, at one end of the scale, the use of a
general word to refer back to a lexical item, at the other end of the scale, or the
use of a synonym, near synonym, or superordinate term. On the other hand,
collocational cohesion or collocation is a special type of cohesion resulting
from the co -occurrence of lexical items that are in some way or typically
associated with one another because they tend to occur in similar environments.

Type of lexical cohesion Referential relati on

1. Reiteration
a. same word (repetition) i. same referent
b. synonym (or near synonym) ii. inclusive
c. superordinate iii exclusive
d. general word iv unrelated
2. Collocation

Fig. 15. General framework for the description of lexical cohesion
(source Hallid ay and Hasan 1976: 288)

Halliday (1994 ) refers to this model in a more recent study, as well. He
states that lexical cohesion can be achieved by repetition of a word, or by using
a word which is related to a previous word “either semantically, such that t he
two are in the broadest sense synonymous, or collocationally, such that the two
have a more than ordinary sense to co -occur ” (Halliday 1994: 310)
In spite of the admitted weaknesses, Halliday and Hasan’s model focuses
on different kinds of lexical item s and it is the most convenient model to apply
in an exploration of the ways in which fixed lexical patterns provide cohesion
in texts. The usefulness of this model is attested by the fact that other specialists
have suggested similar models for the descri ption of lexical cohesion in their
studies.
Gutwinsky (1976: 57 in Carter 1998: 80), for example, states that a
discussion with reference to lexical cohesion requires considering both formal
and semantic criteria. In his opinion, lexical cohesion has the following
characteristics: 1. repetition of an item, 2. occurrence of a synonym or item
formed from the same root and 3. occurrence of an item from the same lexical
set.
A similar approach is suggested by Toolan (1998: 30), who states that
lexical cohesion “conveys a sense of the intergratedness of a text”, and it is

obvious in recurrent uses of the same content word, or of related words.
Moreover, he explains that lexical cohesion implies the non -random association
of a word in one sentence of a text with a word or words in other sentences.
Such patterns of association are important because they contribute to the
coherence of the text and facilitating its rapid interpretation. As regards the
non-random associations, they may be a matter of sheer repetition or near
repetition, or a case of a more general or particular reformulation, or instances
of familiar idiomatic or usage -based co -occurrences. The major kinds of lexical
cohesion identified by Toolan (1998: 31) are:
1. the simple repetition of a given word: chair… chair
2. the use of a synonym or near -synonym: chair …seat
3. the use of a subordinate, superordinate or general term to denote a
particular entity on a later occasion e.g. pet rabbit  the Angora
(subordinate), the pet (superordinate) and animal (general)
4. collocation : tendency of rabbit to co -occur with hole, hutch, on, and
bunny .
Underlying the relation existing between cohesion and coherence , Hatim
and Munday (2006 ) state that “ cohesion subsumes the diverse relations which
transparently hold among the wo rds, phrases and sentences of a text, wh ereas
coherence , which taps a variety of conceptual resources ensuring that meanings
are related discernibly, has the role of underlying these surface phenomena”
(Hatim and Munday 2006: 68) .
Last, but nit least, Nunan (1993) ignores formal aspects and states that
“lexical cohesion occurs when two words in a text are s emantically related in
some way ” ( Nunan 1993: 29).
Summing up what has been stated above, the conclusion can be drawn
that in reading a text, speakers process continuously, and therefore, by the time
any lexical item is taken in, its context has already been prepared. As regards
collocations , they are likely to create problems in discourse interpretation due
to the fact that they are made up of open , rather than closed class items
(pronouns, conjunctions, prepositions) with a finite membership in vocabulary.
Since there is virtually no limit with respect to the items that can be used to
express collocation , it is difficult to establish sets of regularly c o-occurring
words and phrases.
Going the other way round, since lexical relations are textually and
contextually bound, this makes it impossible to develop a finite list of lexical
items which are related. Moreover, the background knowledge of the speaker s
plays an obvious role in the perception of lexical relationships, which means
that collocational patterns will only be perceived by somebody who knows
something about the respective topic of discussion. In other words, the
appropriate identification and decoding of collocations will be conditioned by
the speakers’ awareness with respect to the co-text and context restrictions.

Although virtually synonymous, these two terms are not interchangeble.
3.4.3. Co-text refers to the text which surrounds a passag e, i.e. to the
words or sentences coming before and after it. In other words, the co-text is the
textual context of a text. Sometimes the term co-text is used for the narrow,
purely linguistic context.
On the other hand, the context “is represented by the total linguistic and
non-linguistic background of a text20” (Crystal 1996: 78 -79), and linguistically
speaking the same term is used to refer to that part of an utterance surrounding a
unit and which may affect both its meaning an d its grammatical contribu tion.
The specification of context21 (whether linguistic or non -linguistic) is essential
and “has the effect of narrowing down the communicative possibilities of the
message as it exists in abstraction from context” (Leech 1990: 66).
Referring to the same t wo terms, Yule (1996: 21) states that co-text “is
just a linguistic part of the environment in which a referring expression is used”,
whereas the physical environment, or context22, is the element which influences
the interpretation of the respective referr ing expressions.
As regards the notion of context, Levițchi (1970: 22) explains that it may
be interpreted in at least two senses. In a limited linguistic sense, context is a
term used to denote “a fragment of speech which fulfills the linguistic
condition s necessary to reveal and make them more precise”. On the other
hand, context involves, in a broader sense, extra -linguistic factors such as
gestures, concrete situations in which some body speaks, history, psychology,
etc.
Referring to the same distinct ion, Verdonk (2008: 6 -7) states that t he
linguistic context refers to the surrounding features of language inside a text,
like the typography, sounds, words, phrases, and sentences which are relevant to

20 From a textual perspective, context is seen as “[A] strategic configuration in which
what things mean coincides int entionally and in systematic ways with what they are
used for and with whatever else is going on in the situation” (Beaugrande 1991: 31).
The notion of context as purpose and function is supported by several standards of
textuality: cohesion , coherence , situational appropriateness , intentionality of the
text, acceptability and intertextulaity (Hatim and Munday 2006: 68).
21 Translation studies attaches an important role to context. Understanding the context
in which words operate is essential in order to ens ure the quality of the translated text.
Newmark (1988: 33) identifies four types of context : linguistic (the case of
collocations), referential (related to the topic), cultural and the individual (the idiolect
of the writer) contexts.
22 Reference can b e made to three dimensions of the context (Hatim and Mason 1992 :
57). The communicative dimension considers what utterances count as in context.
Moreover, the pragmatic dimension builds into the analysis of values relating to the
ability to do things with words. Finally, the semiotic dimension treats a communicative
item as a sign within a system of signs.

the interpretation of other such linguistic elements, whereas the non-linguistic
context is a more complex notion which may include various text -external
features that influence the language and style of a text.
In Verdonk’s (2008: 8) opinion, the non-linguistic context has, among
others, the following comp onents:
1. the text type, or genre (for example, an electio n poster, a recipe )
2. its topic, purpose, and function
3. the immediate temporary and physical setting of the text
4. the text’ s wider social, cultural, and historical setting
5. the identities, knowledge, emotions, abilities, beliefs, and assumptions
of the writer / speaker and reader/ hearer
6. the relationships holding between the writer / speaker and reader /
hearer
7. the association with other similar or related text types (inter textuality) .
Although extra -linguistic, situational contexts are generally accepted to
be relevant, a limitation to linguistic contexts seems, in some cases, a better
alternative due, on the one hand, to the fact that the relation between a lexical
item and extra -linguistic contexts is often crucially mediated by purely
linguistic contexts, and to the fact that any aspect of extra -linguistic context can
in principle be mirrored linguistically and linguistic context is more easily
controlled and manipulated, on the other.
To conclude , an important aspect to be taken into consideration in
communication is the fact that the semantic properties of words contribute to the
meaning of more complex units, but communication can only be achieved if the
linguistic item has at least the complexit y of a simple sentence. Moreover,
extralinguistic aspects play an essential role in the appropriate interpretation of
the message which means that the analysis of the co-text must always be
doubled by that of the context in order to correctly decode the m eaning of the
message conveyed.

CONCLUDING REMARKS

The most relevant aspects approached in the present chapter are summarized
below.

1. The factors conditioning the ambiguity of words are generally agreed to be
either linguistic, or extralinguistic, a nd, depending on the nature of the aspects
hindering the semantic transparency of words, different solutions have been
suggested.

2. The lexical and semantic relations of a word with other words that
accompany it in the stream of speech or writing, i.e. t he syntagmatic lexical
relations, are of utmost importance in semantic interpretation. Such relations
arise from the combinations of words in phrases and sentences and are
concerned with the individual lexical units and the meaning relations they enter
with other accompanying lexical units.

3. The meaning typologies and the distinctions made between widely used
semantic categories ( denotation vs. sense , denotation vs. reference ,
denotation vs. connotation ), prove as challenging, as important and useful in
studying the semantic and stylistic specificity characterizing each of the various
types of lexical patterns in a language. As regards collocations , the distinction
between expressive and stylistic connotations is of utmost importance for the
stylistic ap proach to such lexical patterns as it brings to the fore relevant
distinctions illustrative for the use of collocations in different types of texts and
in different situations.

4. A semantic analysis of collocations in terms of specific relations such as
synonymy, antonynmy and hyponymy, rather than in terms of the general
relatedness of topic may be very useful in exploring lexical cohesiveness, so
much the more that these relations hold not only “between words in the
vocabulary, but also between word for ms in texts where they contribute to text
cohesion” (Stubbs 2002: 50).

5. The semantic properties of words contribute to the meaning of more complex
units, but communication can only be achieved if the linguistic item has at least
the complexity of a sim ple sentence. Moreover, extralinguistic aspects play an
essential role in the appropriate interpretation of the message which means that
the analysis of the co-text must always be doubled by that of the context in
order to correctly decode the meaning of t he message conveyed.

4. COLLOCATIONS FROM THE PERSPECTIVE OF
STYLISTICS

In dealing with a stylistic analysis , all the levels of language (from phonology,
through morphology, syntax, lexicology to semantics, pragmatics, and discourse
analysi s) are very important. The interaction between these levels is equally
important in such an analysis because one level may complement, parallel, or
intersect with another . Thus, it is very interesting to see how stylistic effects can
be engendered by offse tting one level of language against another. This is
relevant especially in the case of literary texts, but it is also very useful in the
case of non -literary and specialized texts.
Moreover, since contemporary stylistics considers language as discourse,
taking into account the function of language in context, cognitive strategies
become very useful in a stylistic analysis because they may prove that the
composition of a text meaning is determined by the interplay between what is in
the text, what is in the context, and what is in the mind.
As regards lexical patterns , a stylistic analysis implies special attention
concerning the ways the patterns of vocabulary and grammar are chosen
according to the various functions they have, and the aspects identifiable at the
interface between lexico -grammar and discourse. In addition, such an analysis
implies a careful approach to the semantics of both individual words and lexical
patterns. This is due to the fact that words may bring about both simple and
complex sema ntic problems. As far as the former situation is concerned,
Ullmann identifies three factors which condition the appropriate stylistic
interpretation of a text or the identification of a writer’s style, namely “the
‘motivation’ of the name, the vagueness o f the sense, and the over tones which
may arise around either the name or the sense or both” (Ulmann 1973: 42).
If individual words give rise to a textual stylistic effect on their own,
lexical patterns obviously increase the stylistic complexity of the t exts they are
part of. This results from the fact that words have countless possibilities to
combine on the two axes of meaning i.e syntagmatic and paradigmatic . (see
3.4) Word combinations on the syntagmatic axis contribute to the creation of
context, whe reas paradigmatic combinations highlight various sense relations
established between words, of which synonymy (see 3.3.3.1 .) may be of greater
importance to the study of stylistics.
Considering that words may either have a stylistic value of their own or
they may obtain it from and through context, two situations are likely to be
created :

Situation 1 Situation 2
Words having a stylistic value
of their own Words devoid of a stylistic value
of their own

Speaker’s choice in selecting certain
words Speaker’ s associations of words

Context creation Context creation giving words a
stylistic value

Stylistic value at the level of the text Words acquire full stylistic status

The speaker’s stylistic imprint creates the identity of the text

Fig. 16. Sources of stylistic imprint and text identity (Șorcaru 2006:
24, adapted )

“These two perspectives of research may either be applied to different texts,
which are part of the same type of discourse or of several types of discou rse, or
be used to distinguish among various stylistic effects within the same text”
(Șorcaru 2006: 25).
Nevertheless, the identity of the text will always bear the speaker’s
imprint and a text built in this way will allow for a stylistic analysis of word s
and word combinations irrespective of the typology it fits in. Literary texts
(see 5.6.4. ) will most probably be stylistically analyzed according to the former
model, whereas specialized texts (see 5.6.5 ) are likely to fit in the latter model
of analysi s.

4.1. STYLISTICS: DEFINITIONS AND AREAS OF INTEREST

Stylistics, as a branch of linguistics which has constantly evolved up to being
considered now a complex science in its own right, is generally agreed to be as
new, as controversial in certain aspect s.

Although some specialists consider that stylistics should be included in
the field of literature or in that of linguistics23, without any possible
compromise, this view is not always shared. There are quite many other
specialists who suggest either that stylistics is an interdisciplinary sci ence
placed at the border of the language and literature, or that it is an autonomous
science whose existence among other related sciences, which it may even
include, can be easily attested.
Nevertheless, the contro versial nature of stylistics is not due only to such
divergences in establishing its position within the larger frame of language and
literature sciences. Many controversies have been due to the multifaceted nature
of the term ‘ style ’ which has prevented s pecialists from providing unitary
definitions of stylistics and from establishing clear delimitations of its areas of
interest.
Verdonk (2008: 3), for example, suggests that “the term style (without
specific reference to language) is used commonly in ever yday conversation with
reference to the shape or design of something (e.g. the elegant style of a house ),
when speaking about the way in which something is done or presented (e.g. I do
not like his style of management ), when describing someone’s manner of
writing, speaking, or performing (e.g. to write in a vigorous style ), etc.”
Furth ermore, style is agreed to be used non -technically in a variety of
senses. It sometimes refers to the kind of systematic variation in texts covered
by such terms as formal , colloquial , pedantic , whereas some other times, it
refers to those features of the text, especi ally of a literary text, which identify it
as being the product of a particular author24. The former perspective gives rise
to a very broad definition of stylistics as the description of the linguistic
characteristics of all situationally restricted uses of language25, which explains
the merging of stylistics with sociolinguistics or pragmatics, and its subsuming
to semantics.
The duality of stylistics is also pointe d out by Spencer and Gregory
(1965). They state that the concept of style is ‘an abstraction’ which, on the one
hand “is used to refer to one quality out of several possessed by any work of
literature” and, on the other, it carries with it “an implication of individuality
vis-à-vis generality”, or of the “matching of unique qualities against qualities

23 From a linguistic point of view, the word style , in it s most general interpretation, has
a fairly uncontroversial meaning: it refers to the way in w hich language is used in a
given context, by a given person, for a given purpose, and so on . (Neagu 2012: 10)
24 Style may be defined from this perspective as ‘way of writing’ or ‘mode of
expression’ (Leech and Short 1981: 15).
25 Restricted uses of language may be associated with uses conditioned by the specific
requirements of the functional styles (see 4.2.) as well as with the “individual and
collective modalities of using the language according to different contexts”. (Neagu
2012: 13 -14). For collectiv e modalities of using language see 4.5.1.- 4.5.3.

shared with other works”. Moreover, they explain that “if style in literature is
the product of a particular, and in part unique, use of language, then it is related
to and dependent upon, certain notions of the proper function of language as a
whole”. This implies, in their opinion, a positioning of stylistics beyond the
realm of literature as such, and its involvement in the relation between language
use and social and cultural patterns (Spencer and Gregory 1965: 59).
A similar point is made by Lyons (1977 : 615 ) who considers that
“stylistic variation may be determined by the communicative intentions of the
speaker or writer, by his social role and status , and by other factors in the
context of situation ”. Lyons also believes that stylistic variation is
“undetermined by such factors. Independently of other considerations, an
individual is restricted in expressing his individuality, as distinct from
indicating h is membership of one social group rather than another and his more
immediate communicative intentions, to the choices that he makes within the
range of situationally and socially undetermined variation” (Lyons 1977: 615).
The ambiguity of style and stylist ics is also referred to by Katie Wales
(1991: 437 – 438). She states that stylistics is “the study of style; yet, just as
style can be viewed in several ways, so there are several different stylistic
approaches . This variety in stylistics is due to the main influences of
linguistics and literary criticism”. As it is generally agreed, most stylistic studies
not only describe the formal features of texts for their own sake, but they also
show their functional significance for the text interpretation or relate literary
effects to linguistic aspects..
Observing the same lines, Crystal and Bolton (1993: 214) define
stylistics as “the linguis tic study of what is considered to be ‘style’” (Crystal
and Bolton 1993: 214), adding that the scope of stylistics should not be limited
to the reader’s initial aesthetic response to the text. Stylistics should also
provide a technique of comprehensive analysis meant to clarify the meaning of
a text.
Although applying Crystal’s technique mostly to literary studies, Michael
Toolan (1998) reaches the conclusion that “one of the crucial things attempted
by stylistics is to put the discussion of textual effects and techniques on a
public, shared footing – a footi ng shared and established [… ] as is available to
informed language -users. [ … ] The other chief feature of stylistics is that it
persists in the attempt to understand the technique, or the craft of writing”
(Toolan 1998: IX). In other words, Toolan suggests that stylistics enables
readers to closely examine the lingui stic peculiarities of a text in order to
understand the structure and function of language.
A different perspective as regards the position of stylistics with respect to
other sciences is offered by Widdowson (1997). He states that stylistics is “the
study of literary discourse from a linguistic orientation ” explaining that
stylistics has to be distinguished from literary criticism, on the one hand, and

from linguistics, on the other . Moreover , Widdowson considers that stylistics
has (as yet at least) no a utonomous domain of its own , being essentially a means
of linking literary criticism and linguistics “[…] Stylistics , however, involves
both literary criticism and linguistics, as its morphological make -up suggests:
the ‘style’ component relating it to the former and the ‘istics component to the
latter” (Widdowson 1997: 3).
Galperin (1997) , on the other hand, limits the scope of stylistics to the
area of linguistics, stating that it has two main directions of research as a
science proper. To put it in his words,
“[S]tylistics, also called linguo -stylistics, is a branch of general linguistics. It
deals mainly with two interdependent tasks: 1 ) the investigation of the inventory
of special language media which by their ontological features secure the
desirab le effect of an utterance, and 2 ) certain types of texts (discourse) which
due to the choice and arrangement of language means are distinguished by the
pragmatic aspect of the communication” (Galperin 1977: 9).
Moreover, in an attempt to prove the complexi ty of the term style and of
the discipline studying it, Galperin (1977: 10) provides a series of accessible
definitions: “Style is a quality of language which communicates precisely
emotions or thoughts, or a system of emotions or thoughts, peculiar to th e
author”; “… a true idiosyncrasy of style is the result of an author’s success in
compelling language to confirm to his mode of experience”; “Style is a
contextually restricted linguistic variation”; “Style is a selection of non –
distinctive features of la nguage”; “Style is simply synonymous with form or
expression and hence a superfluous term”; “Style is essentially a citational
process, a body of formulae, a memory (almost in the cybernetic sense of the
word), a cultural and not an expressive inheritance” ; “Style is a product of
individual choices and patterns of choices (emphasis in the original) among
linguistic possibilities” (Seymour 1967: 30); “A style may be said to be
characterized by a pattern of recurrent selections from the inventory of optional
features of a language”; style is “a set of characteristics by which we distinguish
one author from another or other members of one subclass from members of
other subclasses, all of which are members of the same general class” (Sebeok
1960: 427 in Galperi n 1977: 11 – 12).
He enlarges this list with Riffaterre’s (1964: 316 -317, in Galperin 1977:
11) definition of stylistics which is, in his opinion, “a linguistics of the effects
of the message, of the output of the act of communication, of its attention
comp elling function”.
Whether regarding style as personal individual expression, as linguistic
difference at the level of utterances, or deviation from agreed norms, the idea is
shared that all these perspectives “assume the existence of some feature or
featu res which are peculiar to style and distinguish it from language. It follows
that stylistics is not a branch of linguistics but a parallel discipline which

investigates the same phenomena from its own point of view” (Ulmann 1973:
41). Considering Ullmann’ s opinion,
“one may thus look upon stylis tics as displaying several levels, according to the
science to which it is paralleled; consequently, one may analyze the stylistic
effects that are created due to phonology, semantics, lexicology, morphology,
synta x, discourse analysis, etc.” (Șorcaru 2006: 21) .
Moreover, since stylistics may be regarded as the closest relationship
with lexicology and semantics, further distinctions can be drawn between the
semantic structure of individual words and the semantic relations among words,
which is in fact the purpose of the present study .
To put it in a nutshell , style implies a choice of linguistic means made by
the individual speaker or writer in accordance with the requirements of the
communicative situation. Although the author’s personality a nd communicative
intention, the time , place , medium and province of discourse are important
factors conditioning the style of any communi cation (oral or written), of no
lesser importance are the t extual norms which constrain linguisti c choice s in
terms of vocabulary, sentence types, text patterns, and macrostructures. As
regards the various theories on style and stylistics , they
“are only supposed to stress the idea that the two domains are breeding grounds
for ambiguity and debate, managing to elude a cl ear definition of their profiles as
sciences or of their exact tasks, fact which only leaves room for interpretation
according to each analyst’s main interests”. (Neagu 2012: 15)

4.2. STYLISTICS AND THE FUNCTIONAL STYLES

As pointed out in the previous section, certain choices made in communication,
whether conditioned by lexical, morphologic, syntactic, s emantic, or pragmatic
aspects, represent a personal imprint of the speaker , thus a stylistic choice
intentionally made with a certain communicative pur pose. This accounts for the
possibility to regard stylistics as the science concerned with “the analysis of
distinctive expression in language and the description of its purpose and effect”
(Verdonk 2008: 4).
Irrespective of their nature, communication ch oices bring along important
changes of meaning which are analyzed and structured by stylistics . As
Galperin (1977) puts it,
“stylistics studies the expressive means of language taking into account the
modifications of meanings which various expressive mea ns undergo when they
are used in different functional styles ” (Galperin 1977: 29).
Consequently, the functional style to which the expressive means of

language are attached becomes an important point of reface in a stylistic
analysis. In addition, observi ng functional styles is essential in communication
because this automatically implies adjusting one’s expressive means to the
purpose of communication. The speakers who are aware of and responsive to
language variations are generally agreed to recognize th e functional styles as
independent wholes. Moreover, their choice of language means is always
conditioned by the aim of their communication, the result being a more or less
closed system.
A functional style was defined as
“a system of inter -related lang uage means which serves a definite aim in
communication”, or, in other words, “the product of a certain concrete task set by
the sender of the message” (Galperin 1977: 32 – 33).
Functional styles, also called registers or discourses , appear mainly in
the li terary standard of the language, and are represented by the types of texts
with their specific pragmatic aspects of communication. Each functional style
is characterized by a number of distinctive features, whether leading or
subordinate, constant or chan ging, obligatory or optional. However, most
functional styles are perceived as independent wholes due to a specific
combination and interrelation of features common to all with the leading ones of
each functional style.
In standard literary English, refer ence can be made to five functional
styles , namely, the language of belles -lettres, the language of publicistic
literature, the language of newspapers, the language of scientific prose and
the language of official documents , each of which having its own p eculiarities.
These styles are further divided into a number of substyles. They
represent varieties of the abstract invariant and each of them has basic features
common to all varieties of the given functional style and peculiar features
typical of their v ariety alone.
In his approach to the functional styles, Galperin (1977: 33) identifies the
following substyles:
The belle – letters : the language style of poetry, emotive prose, and drama.
1. The publicistic functional style : the language style of oratory,
essays, and feature articles in newspapers and journals.
2. The newspaper functional style : the language style of brief news
items, and communiqués, newspaper headings and notices and
advertisements.
3. The scientific prose functional style : the language style of
humanities, ‘exact’ sciences, and popular scientific prose.
4. The official document functional style : the language style of
diplomatic documents, business documents, legal documents, and
military documents.

Although rather clearly differentiated in this cla ssification, the substyles
specific to each functional style of language often intermingle, making the
boundaries between them hardly discernible.
To conclude, the texts specific to the substyles making up the functional
styles of language may be analyse d if their linguistic components are presented
in their interaction, thus revealing the unbreakable unity and transparency of
constructions of a given type. Moreover, stylistic devices and expressive
means , which are concerned with language problems as the aesthetic function
of language, synonymous ways of rendering one and the same idea, emotional
colouring in the language , interrelati on between language and thought and the
individual manner of an speaker in making use of language are the elements
which se cure the desirable effect of utterances.
As regards functional styles , they are concerned with general linguistic
issues such as oral and written varieties of language, the notion of literary
(standard) language, the constituents of texts larger than the s entence, the
generative aspect of literary texts, which are traceable at the level of texts, but
also at the lower levels of lexical patterns and words .

4.3. A STYLISTIC EVALUATION OF WORDS AND LEXICAL
PATTERNS

Words and patterns are as numerous , as com plex in any language. Nevertheless,
partial evaluations of words and phrases may be made if specific criteria are
taken into consideration.
In the previous section, reference was made to the possibility of
differentiating between various functional styles depending on the expressive
language means used by speakers and on the purpose of communication. Such
an evaluation is suitable at the level of texts , but is hardly effective at the level
of words and patterns.
I believe that assigning words and patterns to a functional style of
language is a rather difficult task to fulfil, due to various factors. On the one
hand, as suggested by Galperin (1977: 70), “the coinage of new lexical units,
the development of meaning, the differentiation of words according to t heir
stylistic evaluation and the spheres of usage, the correlation between meaning
and concept and other problems connected with the vocabulary are so varied,
that it is difficult to grasp the systematic character of the word -stock of a
language”. In addi tion, the fact that words may combine in more or less
complex lexical patterns, that such patterns may alternate between
conventionality and creativity and that the same word or pattern may be used
with different meanings or in different text types, makes it very difficult to
separate words and patterns according to a text typology. Last, but not least,

many of the texts assumed to belong to a specific functional style are far from
being the representatives of that style, due to their almost unconscious
hybridization.
However, a stylistic evaluation of words might still be possible if
reference is made to their categorization according to the three generally agreed
layers of vocabulary: the literary layer, the neutral layer, and the colloquial
layer. Each of these three layers is represented by specific words, and these
words, in their turn, may combine forming patterns which should, most often,
preserve the specificity of the layer those words belong to. Thus, the literary
vocabulary consists of common liter ary words, terms and learned words, poetic
words, archaic words, barbarisms, and foreign words and literary coinages; the
colloquial vocabulary falls into the groups of common colloquial words, slang,
jargonisms, professional words, dialectal words, vulgar words and colloquial
coinages; and the standard vocabulary is made up of the common literary and
common colloquial words.
Some members of the literary and colloquial layers, such as words with
emotive meaning only (interjections), words which have both r eferential and
emotive meaning (epithets), or which retain a twofold meaning , i.e. denotative
and connotative (love, hate, sympathy), words belonging to the layers of slang
and vulgar , or to the poetic or archaic layers have an expressive power which
canno t be doubted, especially if they are compared with members of the neutral
vocabulary layer.
As regards patterns , more precisely collocational patterns, mention
should be made that , while grammatical collocations have very high
frequencies in corpus data, the more colourful, stylistically marked expressions,
apart from occurring infrequently, are often lexically or structurally manipulated
in the contexts in which they appear. Moreover, idioms , in the strict sense, are
very rare.
Considering an approach to collocations from the perspective of
stylistics , McIntosh (196 6: 193 in Carter 1998: 57 -58), offers a framework for
the determination of style in language. He states that
“there is the possibility of four obviously distinct stylistic modes : normal
colloc ations and normal grammar, unusual collocations and normal grammar,
normal collocations and unusual grammar and unusual collocations and unusual
grammar”.
Thus, in his opinion, the speaker produces either language which is too
familiar ( normal collocation s and normal grammar ), or language which is
unfamiliar and difficult to decode ( unusual collocations and unusual grammar ).
The mixture of normal and unusual collocations will lead to creative effects.
When lexical associations are too individual, i.e. they do not meet the condition

of generality, it is more difficult to determine the acceptability of collocations
than to decide over their grammaticality.
Sinclair’s (1996: 76 -77) view is also worth mentioning. In his opinion,
“words enter into meaningful rel ations with other words around them, and yet
all our current descriptions marginalize the massive contribution to meaning.
The main reason for this marginalization is that grammars are always given
priority and grammars barricade themselves against individ ual patterns of
words”.
Collocational and colligational patterns (see 2.1.4. ) are meaning –
creating and there are obvious interdependencies between grammar, lexis and
semantics. Moreover, there are two points to be made: “firstly, that all words
can be desc ribed in terms of patterns; secondly that words which share patterns,
share meanings” (Hunston, Francis and Manning 1997: 209 in Carter 1998: 63).
To sum up, most (fixed) phraseological units are generally agreed to be
expressive . Set phrases , catch words , proverbs , sayings have numerous
elements which make them emphatic, mainly from the emotional point of view,
and their use in every -day speech is remarkable for the subjective emotional
colouring they produce.
Since language is assumed to be emotional, se t expressions are naturally
used in every -day speech. Nevertheless, when such expressions occur in written
texts, it is either a matter of observing their logical meaning or a deliberate
attempt to introduce an expressive element in the utterance. From thi s
perspective, “the set expression is a time -honoured device to enliven speech, but
it is more sparingly used in written texts” (Galperin 1977: 29).

4.4. WORDS AND PATTERNS USED IN LITERARY TEXTS

As already pointed out, there is an overlap between the g rammatical
(morphological and syntactic) , lexical and semantic levels and the stylistic one.
Thus, idiomatic collocational patterns are also colligational, because the
component elements belong to different morphological categories, i.e.
adjectives, nouns and verbs (the case of epithets, of similes, e.g. as merry as a
lark, as strong as a horse , drink like a fish , or the case of metaphors, e.g. red
herring , golden handshake , a bad patch, a cash cow ), or because they have
different syntactic functions, i.e. predicate and object (the case of certain
metaphors e.g. spill the beans, guild the lily ). However, such aspects are paid
less attention, if any, when idiomatic collocational patterns are used in literary
texts. The most common patterns used in such texts are e nlarged on in sections
4.4.1 – 4.4.5. below.

4.4.1. EPITHETS

Out of the various stylistic devices, the epithet is the most “subtle and delicate
in character”. The idea is shared that epithets can create an atmosphere of
objective evaluation , wherea s they actually convey the subjective attitude of the
writer, showing that he is partial in one way or another.
Formally and semantically speaking,
“the epithet is a stylistic device based on the interplay of emotive and logical
meaning in an attributive word, phrase or even sentence used to characterize an
object and pointing out to the reader, and frequently imposing on him, some of
the properties or features of the object with the aim of giving an individual
perception and evaluation of these features o r properties” (Galperin 1977: 157).
Different from the logical attribute, which is purely objective and non –
evaluating, the epithet is markedly subjective and evaluative. Moreover, the
epithet makes a strong impact on the reader so that he unwittingly beg ins to see
and evaluate things as the writer wants him to. (e.g. destructive charms,
glorious sight, encouraging smile ).
Although epithets are lexical patterns in which words combine rather
freely, I believe that they are important for our approach to collocations,
because this specific type of pattern is illustrative for the way in which words
may be creatively combined in literary texts (see 5.6.2. and 5.6.4 ).

4.4.2. EUPHEMISMS

Euphemism is a word or phrase used to replace a word or expression by a
conventionally more acceptable one. In other words, euphemisms are
synonyms which aim at producing deliberately a mild effect (Galperin 1977:
173). For example, the euphemistic word to die may be replaced by the milder
variants to pass away , to expire , to be no more , to depart , to join the majority ,
to be gone , or by the more informal variants to kick the bucket, to give up the
ghost, to go west .
Similarly to epithets, euphemisms are creative lexical patterns, but they
are not limited to literary use. Moreo ver, euphemisms are illustrative of the
way in which synonymy (see 3.3.3.1 .) may be exploited in order to achieve a
special communicative effect intended by the speaker or by the writer. Such an
effect will be brought to the fore only by approaching euphem isms and other
lexical patterns from a stylistic perspective (Galperin 1977: 177).

4.4.3. SIMILES

Similes represent a valuable stylistic resource of expressive patterns most often
regarded as inventive and original expressions which give information a bout
behaviour, reactions , opinions and may be easily understood by speakers
without long explanations. However, the most common similes are familiar
enough to be clichés .
As regards their elements, a large number of similes are made up of
adjectives and nouns, their purpose being to compare a quality, condition,
action: as easy as pie (RO: ușor ca bună ziua ; IT: facile come bere un bicchier
d'acqua'/ facile come innamorarsi ). Sometimes the structure of such similes may
be modified. For example, as thick as two short planks can be shortened to as
thick as two planks . Semantically and stylistically , this simile makes use of a
pun on thick , which means ‘stupid’, as well as the opposite of thin.
Certain verbal patterns , such as work like a horse (RO: a munci din greu;
IT: lavorare come un cavallo/cane) function in a way similar to adjectival
similes.
The preposition as expressing comparison comes before a noun phrase
e.g. as good as gold  bun ca pâinea caldă, as deaf as a post  surd de -a
binelea, as drunk as a lord /newt  mort de beat, beat criță , as ugly as sin 
urât ca dracul . In such comparisons, the first as, an adverb, may be omitted:
quiet as a mouse  foarte tăcut. In much the same way, similes having the
form like … (e.g. like water off a duck’s back  ca gâsca prin apă ) allow for
the o mission of like quite often e.g. (like) a red rag to a bull  motiv de
enervare/ iritare.
Similes26 are easily mistaken with mere comparisons, but while similes
compare notions essentially dissimilar, making use of some features which
make the parallel possible, comparisons establish a similarity between two or
more nouns that are essentially alike: e.g. as drunk as my friend vs. as drunk as
a lord .
A metaphor27 can a lso compare two or more dissimilar nouns, but it
treats one as if it were the other. Thus, a metaphor can be extended to and
implied in a simile e.g. as sweet as honey  dulce ca mierea, as thin as a rake

26 Nowattny (1962) states that “[s]imile (when simple) does not indicate the respect in
which one thing is like another thing. It says the things are alike; it is up to us to see
why; the things may be alike in a large number of ways. Thus simile in turn has its own
advantages. It may be a considerable advantage to the poet to claim that likeness exists
without indicating where it lies”. (Nowattny 1962: 66 in Wikberg 2009: 131)
27 Lodge (1977:113 in Wikberg 20 09: 130) argues that a writer may “ incline towards
simile rather than metaphor proper when drawing attention to similarity between things
dissimilar”.

 slab ca o scândură, as white as snow/a sheet  alb ca zăpada/laptele/varul,
depending on the pragmatic dimension of the context.
Similes are used not only in literary contexts, but also in general and
informal situations. This is because similes are expre ssive and colourful lexical
patterns which give the communication act an amusing dimension.

4.4.4. METAPHORS

Although resembling similes up to a point, metaphors are semantically more
complex. They induce the hearer (or reader) to view a person, a thin g, a state of
affairs, a notion, etc. as being like something else, by applying to them
linguistic expressions which are normally employed with reference to the latter.
The metaphorical strategy of interpretation is most likely to be triggered off by
a per ception of incongruity or inappropriateness in the sentence when
interpreted literally. If a metaphor is used sufficiently frequently with a
particular meaning, it loses its characteristic flavour, or piquancy, its capacity to
surprise , and hearers are lik ely to encode the metaphorical meaning as one of
the standard senses of expression.
According to Lakoff, metaphor28 can be considered “a mere packaging
device – a fancy linguistic wrapping, at best for otherwise plain ideas, at worse
for nothing at all”. H e also points out that metaphor as such “has often been
praised by some for its decorative value and decried by others for hiding plain
truths, but, almost always, it has been dismissed as irrelevant to matters of
serious knowledge, truth and reality”(Lako ff and Johnson 1982: 1).
As a frequently used stylistic device, metaphor may be differentiated
from other expressive lexical patterns by its specific functions and constitutive
elements.
Considering the former aspect, i.e. the functions of metaphor , Newma rk
(1985: 297) underlines the fact that “the general purpose of metaphor is to liven
up other types of texts, to decorate imaginative literature, or to make them more
colourful, dramatic and witty, notoriously in journalism”.
In his opinion, “the main and one serious purpose of metaphor is to
describe entities (objects or persons), events, qualities, concepts or states of
mind more comprehensively, concisely, vividly, and in a more complex way,
than is possible by using literal language” (Newmark 1985: 295 ).
Furthermore, Newmark underlines two other important functions of
metaphor. On the one hand, he refers to the metaphor’s purpose of pleasing,

28 According to Marrinucci (1996: 449) la metafora “consiste nel trasferimento di
significato da un sens o proprio in base al rapporto di similarità reale o intuitiva che
intercorre tra i due concetti”.

sometimes aesthetically, of entertaining, amusing, often of drawing attention to
a technical and physical subjec t, therefore of conceptually clarifying things, and
to its purpose of indicating a resemblance between two more or less disparate
objects, on the other.
As far as the constituent elements of metaphor are concerned, these are
the topic / tenor / object , which is the item described by the metaphor, the
vehicl e/ image, which is the item in terms of which the object is described, and
the ground/ sense , which is the point of similarity that shows in what particular
way the object and the image are similar.
Metaph ors may be classified according to different criteria. If the type of
discourse in which metaphors are used is under focus, reference can be made to
three types of metaphors , i.e. genuine, degraded and absolute. Genuine
metaphors are most often found in po etry and emotive prose, whereas
degraded metaphors are largely used in the journalistic discourse as clichés,
rendering a better rhythm to the sentence. In the case of absolute metaphors
there is no clear -cut distinction between the idea and the image: e.g . We have
eyes in the back of our head – ‘to know what is going on around one even when
one cannot see it ’.
Reference can also be made to complex metaphors which help to
intensify the meaning of the collocational pattern, e.g. That throws some light
on th e question, and compound metaphors which have several points of
similarity, e.g.: He has the wild stag’s foot .
Furthermore, depending of their freshness in language, metaphors can be
divided in to two classes, namely active and dead metaphors .
An active met aphor is relatively new and has not become part of every
day language usage e.g. You are my sun , whereas in the case of dead
metaphors their interpretation is not wholly predictable on first acquaintance:
e.g. to kick the bucket .
A similar classification i s provided by Galperin in his stylistic approach
to language. He refers to two types of metaphors : i.e. genuine metaphors,
which are absolutely unexpected and dead/ trite metaphors , which are
commonly used in speech and sometimes fixed in dictionaries as e xpressive
means of language. As regards the former class, Galperin (1977: 141) explains
that such metaphors are considered to belong to language -in-action, whereas
those included in the latter category, i.e. trite metaphors , belong to the
language -as-a-system, i.e. to the language proper, and are usually fixed in
dictionaries as units of language. Consequently, genuine metaphors are most
often found in poetry and emotive prose, different from trite metaphors which
are generally used as expressive means in n ewspaper articles, in oratorical style
and even in scientific language.

As suggested by Lombardo et. al. (1999: 63), sometimes metaphors turn
into technical language, in which case they are dead metaphors for the
members of that discourse community (see 2.3.1.2 ).
Mention is also made to the constant interaction between genuine and
trite metaphors. This is obvious in the fact that good genuine metaphors may
become trite through frequent repetition and consequently easily predictable,
whereas trite metapho rs may regain their freshness through the process of
prolongation of the metaphor.
As regards the relationship of metaphors with idiomatic collocational
patterns , Cruse (1986: 44) states that they have certain characteristics in
common. On the one hand, t heir constituent elements do not yield recurrent
semantic contrasts, which explains their not being semantically transparent. In
addition, the effect of synonymous substitution and the continuous relevance of
their literal meanings make the term ‘opaque’, used to denote such patterns,
unsatisfactory. In Cruse’s opinion, the term ‘translucent’ would be more
appropriate. Finally, they are syntactically rigid. From a stylistic point of view,
both metaphors and idiomatic collocational patterns are expressive me ans of
language which are frequently used in the informal register. Moreover,
considering the classes of dead /trite and genuine metaphors, refer ence should
be made that they occur in texts specific to different discourse types, the former
being often found in business -related newspaper articles, sometimes in an
altered form, whereas the latter are traceable in literary texts.
To conclude, the constant use of a metaphor gradually leads to the
breaking up of the primary meaning. Moreover, the metaphoric us e of the word
begins to affect the dictionary meaning, adding to it fresh connotations or
shades of meaning. Nevertheless, however strong, this influence will never
reach the degree where dictionary meaning entirely disappears. “It is a law of
stylistics t hat in a stylistic device the stability of the dictionary meaning is
always retained no matter how great the influence of the contextual meaning
may be . (Galperin 1977: 144)

4.4.5. CLICHÉS , PROVERBS, AND SAYINGS

The cliché is an expression which, due to its having become trite as a result of
losing its aesthetic generating power, strives after originality (e.g. rosy dreams
of youth , deceptively simple ).
“The term cliché was wrongly used to denote all stable word combinations,
whereas it was coined to de note word combinations which have long lost their
novelty and become trite, but which are used as if they were fresh and original
and so have become irritating to people who are sensitive to the language they

hear and read. All word combinations that do no t surprise are labeled as clichés”
(Galperin 1977: 178 – 179).
Proverbs and sayings are facts of language characterized by rhythm,
sometimes rhyme and/or alliteration, by brevity and by a specific content -form
of the utterance.
A very important aspect spe cific to proverbs is the fact that the actual
wording becomes a pattern which needs no new wording to suggest contextual
extensions of meaning . In other words, a proverb presupposes a simultaneous
application of two meanings: on the one hand, a primary mea ning, and an
extended meaning drawn from the context, on the other.
As Galperin (1977: 181) puts it,
“the proverb itself becomes a vessel into which new content is poured. The
actual wording of a proverb, its primary meaning, narrows the field of possible
extensions of meaning, i.e. the filling up of the form. That is why we may regard
the proverb as a pattern of thought”.
Such patterns of thought are traceable in other lexical patterns (see
metaphors , euphemisms ), as well, and they prove very prolific due to the fact
that abstract formulas offer a wider range of possible applications to practical
purposes than concrete words , though they may have the same purpose.
Proverbs are also important for their cultural markedness. These brief
statements show in con densed form the accumulated life experience of the
community and serve as conventional practical symbols for abstract ideas. In
addition, they are usually didactic and image bearing and most of them have
become polished and wrought into verse -like shape th rough frequent repetition:
“Early to bed and early to rise, /Makes a man healthy, wealthy and wise ”.
Brevity in proverbs manifest in the omission of connectives: First come, first
served ; Out of sight, out of mind .
Nevertheless, the main feature distingui shing proverbs and sayings from
ordinary utterances remains their semantic specificity, i.e the fact that the literal
meaning is suppressed by what may be termed their transferred meaning. The
stylistic effect produced by such uses of proverbs and sayings is the result of a
twofold application of language means. The modified form of the proverb is
perceived against the background of the fixed form, thus enlivening the latter.
Sometimes, as suggested by Galperin (1977: 183),
“this injection of new vigour in to the proverb causes a slight semantic re –
evaluation of its generally accepted meaning. When a proverb is used in its
unaltered form it can be qualified as an expressive means of the language; when
used in a modified variant it assumes the one of the feat ures of a stylistic device,
it acquires stylistic meaning , though not becoming a stylistic device”.
If used appropriately, proverbs and sayings are agreed to preserve their
freshness and vigour. Besides, they may be used not only in their fixed form,

(the traditional model), but also with modifications. However significant, these
modifications will never break away from the invariants to such a degree that
the correlation between the invariant model of a word combination and its
variant ceases to be percei ved by the reader. Since the predictability of a variant
of a word combination is lower in comparison with its invariant, the use of such
a unit in a modified form will always arrest our attention causing a much closer
examination of the wording of the utt erance in order to get the idea. (Galperin
1977: 182)
Instead of a conclusion, mention should be made that
“the stylistic potential of the ‘phrasicon’ is unchallengeable. Since phraseological
units may be relevant stylistic devices, they are strong eviden ce of a level inside
the stylistic system which has been termed ‘phraseo -stylistics’ (Gläser 1986a)
and which combines the systemic and communicative aspects of linguo -stylistic
analysis .” (Gläser 1998: 143)

4.5. COLLOQUIAL WORDS AND PATTERNS

Colloquial words and patterns are frequent , both in familiar and professional
settings, but, different from other fixed lexical patterns (metaphors, idioms,
etc.), they characterize the oral, rather than the written discourse.
Although most often accessible, colloq uial words and patterns can be
used appropriately only if the speakers involved in the act of communication
observe the contextual restrictions regulating the use of the respective words
and patterns. This is due to the fact that distinction may be made be tween
general or neutral words and collocational patterns and more or less specialized
ones. If the words and patterns of the former category are accessible to all the
speakers of a language community, those included in the latter category (slang,
jargon, professionalisms) are more or less restrictive , in the sense that they may
be semantically decoded and appropriately encoded only by those speakers
who are familiar with the specific restrictions.
As regards general colloquial collocations , I believe t hat social and
linguistic contexts , which determine the associative properties of words, have a
significant role in the appropriate selection and use of such collocations. The
social context is related to who uses the word or pattern to whom, when, where,
for what purpose and in connection with what topic. The linguistic context , on
the other hand, refers to the interrelations of a word with the surrounding words,
i.e. the collocational interrelations . As a matter of fact ,
“[I]f words are so different in the kind of criteria involved in their use, the
pattern of meaning made by their interrelation must be immensely more complex

than a one -plane analogy may suggest” (Waldron 1967: 102 in Carter 1998:
266).
Furthermore, another factor conditioning the cor rect use of colloquial
collocations is the speakers ’ communicative competence . According to Hymes
(1971 in Lyons 1977: 573), this specific competence “covers a person’s
knowledge and ability to use all the semiotic systems available to him as a
member of a given socio -cultural community. Linguistic competence is thus but
a part of communicative competence”.
Last, but not least, the participants in the act of communication must
know and assume their role and status if they want to make their
communication ef fective and appropriate.
Linguistically relevant roles are of two types, namely deictic and social .
Deictic roles are related to the fact that “in normal language -behaviour the
speaker addresses his utterance to another person (or to other persons) who ar e
present in the situation and may refer to himself, to the other addressee(s) or to
other persons and objects, not by means of a name or description, but by means
of a personal or demonstrative pronoun whose reference is determined by the
participation of the referent in the language -event at the time o f the utterance”.
As regards social roles , they are “culture -specific functions institutionalized and
recognized by its members. The most obvious effect of social roles as a
contextual variable lies in its d etermination of terms of address” (Lyons 1977:
574 – 575).
Viewed from this perspective, I consider that there are two other very
important aspects which can be related to social roles : i.e. social status , on the
one hand, and sex and age, on the other.
Social status “is the relative social standing of the participants”. It refers
to the fact that
“each participant in the language – event must know or make assumptions about
his status in relation to the other, and in many situations status will also be an
important factor in the determination of who should initiate the conversation”
(Lyons 1977: 576).
In this respect, it would be very interesting to analyze collocational
patterns used by people holding high social positions , on the one hand, and
those in low social positions , on the other. For example, the people in the
former category will probably use collocations closely related to their activity :
 nominal collocations a board of directors  un comitet director,
consiliu de conducere, a boom in demand  o creștere (rapidă) a cererii,
bulk shipment  expe dierea unui lot mare/ a unei cantități mari de
mărfuri, the divisional headquarters  sediul central al grupului de
firme, documents against cash/documents against presentation 
documente contra plată, aran jament prin care un cumpărător primește

documentele pentru mărfuri în momentul când plătește o cambie, our
dormant accounts  clienții noștri din trecut, a fall in demand  o
scădere a cererii, a member of the elite  un membru al elitei, a
prerequisite f or employment  condiție (obligatorie) pentru anagajare, a
rise in demand  o creștere a cererii , sealed bids  oferte de preț
sigilate pentru depozit, sequential sampling  eșantionare secvențială,
desfășurarea acțiunii de selectare până când au fost int ervievați suficienți
respondenți pentru a oferi informațiile necesare, shop audit  raportul
de audit, etc.
 verbal collocations: to set a record for high prices  a stabili un
record al prețurilor mari, to share the profits  a împărți profiturile, etc.
If people occupying low social positions are taken into consideration , the
following examples of collocations might be illustrative for the social group
they belong to : a crop of disaster s  o mulțime de dezastre (petrecute în același
timp), do-it-yoursel f (DIY) magazine  revista cu articole despre activități pe
care omul obișnuit le poate face pentru a -și repara sau îmbunătăți locuința,
hoots/ howls/shouts of derision  huiduieli/ urlete/ strigăte batjocoritoare,
scrambled merchandising  mărfuri la gram adă, a sign of discontent  un
semn de nemulțumire, etc.
Another possible distinction in the case of colloquial collocations is that
in terms of register as conditioned by social position , appurtenance to a certain
group and age :
 bump start n. (BrE) a su dden rousing to action , whether physical or
metaphorical. The expression, employing the image of pus h – starting a
reluctant car, was a fashionable colloquialism in the 1980s often used in
a business or sexual context; kick start  șut in fund is a more p opular
alternative in spoken AmE ;
 clip joint n. originally a club or bar which employs hostesses who
encourage clients to buy them expensive drinks in the expectation of
sexual favours. The phrase is now applied to any overpriced low -quality
establishment . Clip, like trim, is an old euphemism for ‘to relieve
someone of their money  lipeală, bombă.
As regards the degree of formality of colloquial collocations, a parallel
may be drawn with the five degrees of formality in English referred to by Joos
(1962 in Lyons 1977: 580), i.e. frozen , formal , consultative , casual and
intimate . Each of these degrees of formality is correlated with systema tic
phonological, grammatical and lexical differences and may be associated with
specific situations and with specific users. Formal and frozen expressions are
specific to high social status , wher eas casual and intimate expressions are
specific to lower social status . As regards the consultative expressions, their
use is more often formal, but it may sometimes depend on the context of
situation.

Another criterion of differentiation applicable in th e case of colloquial
collocational patterns is the variety of English to which the respective
collocations belong. Distinction is most often made between British and
American English and this may be illustrated by examples such as: to be left
holding the baby (BrE) vs. to be left holding the bag (AmE); he won’t touch
something with a bargepole (BrE) vs. he won’t touch something with a ten-foot
pole (AmE); to lock the barn door after the horse is stolen (AmE) vs. to lock
the stable door after the horse has b olted (BrE); to sweep something under the
carpet (BrE) vs. to sweep something under the rug (AmE); to lead somebody to
a merry chase (BrE) vs. to lead somebody to a merry dance (AmE) (Benson
1985: 67) (Benson, Benson and Ilson 1986: 252 -253).
Analyzing th e formal variations traceable in various lexical patterns used
in British and American English , Moon (1998: 132) identifies more situations.
On the one hand, she refers to verb variation in patterns such as cut a
long story short (BrE), make a long story short (AmE) flog a dead horse (BrE),
beat a dead horse (AmE), let off steam (BrE), blow off steam (AmE), kick one's
heels (mainly BrE), cool one's heels (mainly AmE) touch wood (BrE), knock
wood, knock on wood (AmE), explaining that such a variation is not so
frequent. (Moon’s examples)
Reference is also made to variations of a noun or noun modifier, which
occur more often and reflect, in Moon’s opinion, standard distinctions between
British and American English: in the driving seat (BrE), in the driver's seat
(AmE) not know one's arse from one's elbow (BrE), not know one's ass from
one's elbow (AmE), red as a beetroot (BrE), red as a beet (AmE), throw a
spanner in the works (BrE), throw a ( monkey ) wrench in the works (AmE),
wear the trousers (BrE), wear the pants (mainly AmE). (Moon’s examples)
An interesting aspect is that the formal approach combines with a cultural
oriented one in the case of catch someone with their trousers down and catch
someone with their pants down . If the former is used only in Brit ish English,
catch someone with their pants down is found in both varieties. In addition
patterns such as like turkeys voting for Christmas (mainly BrE), like turkeys
voting for Thanksgiving (AmE), turn on sixpence (BrE), turn on a dim (AmE)
reflect obvio us cultural distinctions. This is also the case with the British bent
as a nine-bob note and the corresponding American phoney/ queer as a three –
dollar bill which combine cultural and lexical distinc tions.
Starting from the assumption that in the majori ty of cases, the distinctions
she makes seem idiosyn cratic, Moon suggests that historical explanations may
be provided in the case of lexical patterns such as fall through the net (BrE),
fall through the cracks (AmE) have green fingers (BrE), have a gree n thumb
(AmE), if the cap fits (BrE), if the shoe fits (AmE), keep one's hair on (BrE),
keep one's shirt on (AmE), not see the wood for the trees (BrE), not see the

forest for the trees , (AmE), rub shoulders with (BrE), rub elbows with (AmE),
too big for one's boots (BrE), too big for one's britches/breeches (AmE).
Regarding prepositions, Moon (1998) suggests that about used with
spatial meaning or reference is largely used in BrE, whereas Americans prefer
around . However, patterns such as (not) to beat about the bush (BrE), and ( not)
to beat around the bush (AmE) are found in both varieties. Other cases of
prepositional variation are more idiosyncratic: at a pinch (BrE), in a pinch
(AmE), lead someone up the garden path (BrE), lead someone down the garden
path (ArnE), on the cards (BrE), in the cards (Am E).
Finally, reference is made to parallel British and American idioms which
have similar meanings, usages, and even source domains for the metaphors , but
different lexis altogether: a storm in a teacup (BrE), a tempest in a teapot
(AmE), have one's hand /fingers in the till (BrE), have one's hand in the cookie
jar (AmE), in inverted commas (BrE), quote unquote (BrE and AmE), quote
end quote (Am E).
Sex and age are often determinants of, or interact with, social status . The
terms of address, for example, employed by a person of one sex speaking to a
person of another sex, or by a younger person speaking to an older person, may
differ from those which would be employed in otherwise similar situations by
peop le of the same sex or of the same age.
Sex-dependent language use is interesting to study when comparing
collocational patterns that are likely to be used by or associated with men with
those used by and associated with women. Consider the following examp les of
collocations :

Female: e.g. an air of innocence  un aer inocent, corner shop 
butic, a display of emotion  manifestare a emoției/ sentimentelor , an
excess of enthusiasm  exces de zel, a flash of intuition  o
(străfulgerare de) intuiție, a flood/ gush of emotion  un val de emotie,
o revărsare de sentimente, the limits of endurance  limitele, cap ătul
răbdării, a pair of tights  o pereche de ciorapi pantalon, the power of
endurance  puterea de a răbda, shop around  a umbla pe la
magazine pentru a compara prețurile înainte de a cumpăra ceva, a test
of endurance  proba de răbdare, women’s clothes shop  magazin de
îmbrăcăminte pentru femei.

Male : e.g. bulk shipment  expedierea unui lot mare de mărfuri, a
unei cantități mari de mărfuri, the (d own)fall of an empire 
prăbușirea/ căderea unui imperiu, a flash of insight  moment de
introspecție, lashings of drinks  băutură berechet, machine shop 
sala mașinilor, loc unde se țin utilajele aflate în funcțiune, the mastery
of one’s emotion  contr olul sentimentelor/ emoțiilor, a pair of waders

 o pereche de cizme pentru pescuit, repair shop  atelier de reparații
unde se repară utilaje sau mecanisme, a round of drinks  un rând de
băutură, ship broker  broker maritim, persoană care aranjează
expe dierea sau transportul de mărfuri pentru clienți în numele
armatorilor, a stand against the enemy  atitudine de rezistență
împotriva inamicului.

Collocational patterns may also be specific to age-dependent language
use. Thus, a series of such patterns are likely to occur in children’s and
teenagers’ language, whereas others are used by adult people. The examples
below may often be assoc iated with different age levels:

Children and teenagers: e.g. an air of innocence  un aer innocent,
children’s clothes shop  magazin de îmbrăcăminte pentru copii , a
burst of enthusiasm  explozie de entuziasm, a fever of excitement 
stare de agitatie, a flash of excitement  emotie bruscă, a flurry of
excitement  emoție intensă, de scurtă durată, a flush of enthusiasm 
revăr sare de entuziasm, a mood of elation  o stare de exaltare, a note
of discord  o nota/ un semn de dezacord/ neînțelegere, a prickle/
tremor of excitement  fior/ freamat/ de emoție, a surge of elation 
un val de exaltare /inflacarare, a surge/ wave of e nthusiasm  un val de
entuziasm, un upsurge of enthusiasm  cuprins (brusc) de entuziasm.

Adults : e.g. a bout of depression  o perioada de depresie, a code of
ethics  cod etic de principii morale, a feeling of insecurity 
sentiment de nesiguranță, a feeling of isolation  un sentiment de
izolare, a fever of excitement  stare de tulburare, a frown of
disapproval  o privire (încruntată) dezaprobatoare, a lack of
enthusiasm  lipsa de entuziasm, a manifestation of discontent  o
manifestare de nemulțumir e, a mark of esteem  semn de prețuire/
stimă, a profession of esteem  declarație de respect, a sense of
estrangement  un sentiment de întrăinare, a set of dentures  proteza
dentară, ( not) a spark of enthusiasm  (nici) un strop/ o urmă/ o
licărire de ent uziasm, a stroke of panic  atac/ stare de panică,
Although specifically differentiated in terms of some possible social
variables, the examples of general colloquial collocations analyzed in this
section need further investigation, especially those class ified according to the
sex and age criteria. Nevertheless, they offer another interesting perspective on
collocations which may and should be further investigated. From a linguistic
point of view, the fact may be easily noticed that certain collocational patterns
such as e.g., a lack of enthusiasm  lipsa de entuziasm, a manifestation of
discontent  o manifestare de nemulțumire are likely to be shared by both

males and females as well as by people belonging to different groups of age.
This adds to the complexity of collocations, and to the impossibility of making
clear -cut distinctions in the approac hes and classifications suggested for this
type of lexical pattern.

4.5.1. SLANG

Slang , as one of the sublanguages agreed to impose limitations on its unaware
speakers or on the outsiders of the speech community (see 2.3.1.1 .) represents
a deviatio n from the established norm at the level of the vocabulary of any
language.
The definitions and descriptions provided for this type of language are
neither numerous nor very specific.
Slang may be defined as … “a peculiar kind of vagabond language,
alwa ys hanging on the outskirts of legitimate speech , but continually straying of
forcing its way into the most respectable company (Greenough and Kitteridge
1929: 55 in Galperin 1977: 107).
A significantly different definition of the same reality is provide d by
Chatto and Windus (1925 in Levițchi 1970: 118) in their Slang Dictionary .
They call slang “the language of street humour, of fast, high, arid high life”.
More general definitions consider that slang is represented by the words
or phrases which are in common use , but which are not considered suitable for
use on serious occasions, or by the language of a particular class of people.
Takin g into account both diachronic and sychronic criteria, ( Levițchi
1970: 118) considers slang as “an extreme, short -lived form of colloquial
English (w ords or phrases), mainly based on metaphors of the live type, and
rejected by literary language”.
Similarly to colloquialisms , slangy words and phrases are always
stylistic synonyms of current words and phrases (terms included, since their
slangy equivalen ts are professionalisms ), which are never brought into
existence by new notions, phenomena, actions, etc.
A possible classification of slang is suggested by Arnold (1959, in
Levițchi 1970: 119) who distinguishes two types of slangy words and phrases :
1. General slang universally understood and widely spread gra phical
words and phrases with a strong emotional colouring, e.g. bob ‘shilling’,
to booze ‘to carouse’, hide ‘human ski n’, dope ‘narcotic drops’, to work
the steam off ‘to rid oneself of excessive energy’, etc.
2. Special slang includes words and phrases belonging to special or
professional vocabulary, the slang of sailors, of sportsmen, actors,
lawyers, students, etc. and the slang of differ ent social groups.
If formal aspects are taken into account, 4 classes of slang may be

identified:
1. Slang proper e.g. rot-gut ‘bad, small beer’, to shave a customer ‘to
overcharge a customer’, whisperer ‘constant borrower’, etc.
2. Back slang mainly based on spelling the words backwards or rather,
on pronouncing them rudely backwards e.g. to cool ‘to look’, doog
‘good’, edgabac ‘cabbage’, eno ‘one’, etc.
3. Rhyming slang based on words and phrases rhyming with the actual
word or p hrase they express, e.g. Abraham's willing ‘shilling’, Charing
Cross ‘horse’, plates of meet ‘feet’, to read and write ‘to fight’, etc.
4. Centre slang formed by making the central vowel of a word its initial
letter, and adding vowels and consonants’ suf ficient to make the sound
imposing, e.g. ugmer or hugmer (‘mug’, ‘fool’ ).
As regards the functions of slang , Wentworth and Flexner (1960 in
Galperin 1977: 107) provide a comprehensive description in this respect. They
suggest that sometimes
“slang is used to escape the dull familiarity of standard words, to suggest an
escape from the routine of everyday life. When slang is used, our life seems a
little fresher and a little more personal. Also, as at all levels of speech, slang is
used for the pure joy of making sounds, or even for a need to attract attention by
making noise. The sheer newness and informality of certain slang words produce
pleasure. But, more important than this expression of a more or less hidden
aesthetic motive on the part of the speaker , is the slang’s reflection of the
personality, the outward, clearly visible characteristics of the speaker. By and
large, the man who uses slang is a forceful, pleasing, acceptable personality”
(Wentworth and Flexner 1960 in Galperin 1977: 107).
Slang words such as bread -basket ‘the stomach’ (jocular), rot ‘nonsense’
and lexical patterns such as to take stock in ‘to be interested in, attach
importance, give credence to’, to do a flit ‘to quit one’s flat or lodgings at night
without paying the rent or boa rd’, the cat’s pyjamas ‘the correct thing’, are
illustrative for what has been stated above.
Reference can also be made to criminal slang . Howard and Zé Amvela
(2000: 135) provide examples of both words and patterns illustrative for this
type of slang:
 Criminal slang words: e.g. can, chokey, clink, cooler, nick, peter,
slammer, stir ‘prison’, a star or a lag‘ a first time prisoner’, a screw ‘a
prison warder’, to cheese ‘to act in a groveling manner’, a div ‘a stupid
person’, a borgey, the filth, the fuzz, the pigs ‘the police’, a nark ‘a
police informer’, to grass ‘to inform’ , to shop ‘to betray to the police’,
mob ‘criminal gang’, a form ‘a criminal record’, a cleanskin ‘somebody
without a criminal record’, minder ‘bodyguard’, a verbal ‘a confessi on
of guilt’.

 Criminal slang patterns: e.g. bang up, do bird, go down, be inside,
do porridge sent up/down ‘being sent to prison or being in prison’, to go
straight ‘not to engage in criminal activity’, to be hit ‘to be murdered’,
to fulfil a contract ‘to fulfill a killing.
Irrespective of the domain attached to, slang words and patterns are
illustrative for the way in which meaning is affected due to the direct actions of
the speakers creating and using such words and patterns. Different from other
lexic al patterns, slang patterns may create serious problems to both unaware
speakers and translators. This is due to the fact that most often speakers and
translators, respectively, are outsiders of the group(s) using such words and
patterns, and dictionaries are not always helpful enough in this respect.

4.5.2 JARGON

Jargon is a term used to denote words that exists in almost every language and
whose role is to preserve secrecy within one or another social group .
Different from slang words, jargonisms are generally old words with
entirely new meanings. That is to say, they are words whose traditional meaning
became immaterial as compared t o the new, improvised meaning. Most of the
jargonisms in any language are absolutely incomprehensible to those outside rs
of the social group which invented them (e.g. grease ‘money’, loaf ‘head’, a
tiger hunter ‘a gambler’, a lexer ‘a student preparing for law courses’).
Reference can be made to common jargon and special professional
jargons . The former have gradually l ost their special secrecy -promoting nature,
thus being no longer semantically opaque to the outsiders of the community
using them. Such jargon terms which have become common, have been
included in the category of slang or colloquial words e.g. man and wife
(rhyming slang) ‘a knife’.
As regards professional jargons , they are associated with a certain
profession or occupation being used by speakers who have common interests at
work and at home. Interestingly enough, similarly to slang and other groups of
the non-literary layer, jargonisms “do not always remain on the outskirts of the
literary language. Many have entered the standard vocabulary, thus being
dejargonized” (Galperin 1977: 113).

4.5.3. PROFESSIONALISMS

Professionalisms are words used in a defi nite trade, profession, or occupation
by people connected by common interests both at work and at home.

Professionalisms are often compared and contrasted with terms which
are used to denote new concepts that appear in the process of, and as a result of,
technical progress and development of science. Nevertheless, there are
significant differences between professionalisms and terms.
Professional words name already existing concepts, tools, or instruments
and have the typical properties of a special code , their main characteristic being
technicality .
Terms , on the other hand, make direct reference to a certain branch of
science, being up to a point self explanatory, thus not functioning as a code.
Furthermore, professionalisms are special words in the non -literary
layer, whereas terms are a specialized group belonging to the literary layer. If
terms are connected with a field or branch of science or technique well -known
to ordinary people, they are easily decoded and enter the neutral stratum of the
vocabul ary. On the contrary, professionalisms generally remain in circulation
within a definite community , being linked to a common occupation and
common social interests .
In addition, terms are usually semantically transparent which facilitates
their understan ding whereas professionalisms are often “dimmed by the image
on which the professionalism is based, particularly when the features of the
object in question reflect the process of the work metaphorically or
metonymically”(Galperin 1977: 113). Similarly to terms , professionalisms do
not allow any polysemy: e.g. tin-fish ‘submarine’, block buster ‘a bomb
especially designed to destroy blocks of bug buildings’, piper ‘a specialist who
decorates pastry with the use of a cream -pipe’, a midder case ‘ a midwifery
case’ , outer ‘ a knock out blow’.
Last, but not least, some professionalisms resemble certain terms (see
4.6.1. ), in their becoming popular and gradually losing their professional
flavour. However, professionalisms should not be mixed up with jargonisms .
This is due to the fact that professionalisms, like slang words, do not aim at
secrecy, their social function in communication being that of facilitating a quick
and adequate decoding of the message.
As a conclusion, in different communication situations , speakers must be
able to switch codes according to the situation -of-utterance. Code -switching is
“by no means restricted to language communities in which two or more
recognizably distinct dialects or languages are used, but it may also be used by
monoli nguals who switch from one style to another under similar determining
conditions” (Sankoff 1972 in Lyons 1977: 581).

4.6. WORDS AND PATTERNS USED IN SCIENTIFIC TEXTS AND
OFFICIAL DOCUMENTS

Representative for two different functional styles, scientific texts, and official
documents are just two other instances in which words combine in more or less
flexible, and more or less semantically transparent lexical patterns. However,
different from the other functional styles of language dealt with in the previo us
sections, the styles of scientific prose and of official documents require
clarifications as regards the distinction , on the one hand, between words and
terms, and, on the other, between the instability and fixity of meaning in words
and terms, respecti vely.

4.6.1. WORDS, TERMS AND THE FIXITY OF MEANING

Words and terms are quite often used interchangeably by unaware speakers,
but the realities they denote and the features they have prove that these two
terms are hardly, if ever, interchangeable .
Terms are generally agreed to be highly conventional forms which are
very easily coined and equally easily accepted. One of the most important
features of term is their relevance to the system or to the set of terms used in a
particular science, discipline or art, i.e. to their nomenclature.
Unlike words , a term is directly connected with the concept it denotes,
aspect which implies that terms have the role to direct the mind to the essential
quality of the thing, phenomenon or action as seen by scientists in the light of
their own conceptualization. As Verhaar (1966: 378 in Galperin 1977: 76) puts
it, “a word is organically one with its meaning; likewise, a term is one with a
concept. Conceptualization leaves, as it were, language behind, although the
words remain as (scientific or philosophical) terms. Linguistically the difference
is important in that terms are much more easily substitutable by other terms than
are words by other words. […]”.
Since terms are predominantly used in special works dealing with the
notions of some branch of science, they are regarded as belonging to the
language of science. Nevertheless, their use is not confined to the style of
scientific prose. Terms may be used in other functional styles of language, but
in such situations, t heir function changes.
Another important aspect distinguishing words from terms is their
semantic specificity and the possibility that their meaning should change in time
and under specific circumstances.
As far as ordinary words are concerned, the idea is shared that they
change their meaning in a variety of ways through usage. On the contrary,
terms are generally protected from the effects of usage, as much as possible.

Even though seemingly valid, this opposition between words and terms
is contradicte d in practice. For example, there are numerous names of flora and
fauna, which resemble terms very much, just as there are many words which
have a specialized and protected meaning in a discipline without necessarily
being granted the status of terms (e.g. window and mouse in general use vs.
specialized use in computing).
This aspect is referred to by Bidu -Vrănceanu (2007), as well. She states
that many terms have been assimilated in the common language, but the
number of terms undergoing such changes and the extent to which they are
used in their non -specialized meaning are difficult to specify. Moreover, s he
refers to the fact that the terms accepted to have changed their status in this way
originate in various, but not necessarily predictable, domains:
“Care sunt situațiile, proporțiile și gradul de răspândire și asimilare a unor
termeni în limba comună s unt probleme greu de precizat. Exemplele date de
specialiști pentru a ilustra termeni deveniți cuvinte în limba română …provin
din domenii variate și nu neapărat previzibile: oxigen, oțel, trafic, vitamină,
virus, frână, ecuație, algoritm, etc” (Bidu -Vrănc eanu 2007: 29).
Nevertheless, she suggests that the possible definitions provided for
terms29 as “fundamental units of terminology” (2007: 31) and a clear
specification of their characteristics may prove useful instruments for the
differentiation of terms from words. Referring to the function s that a term has,
Bidu -Vrănceanu explains that these functions ensure the accuracy of specialized
communication in a given language and the interlingual correspondences. Thus,
according to Spillner (1994: 56 -59) the features specific to a term are the
following:
 The denotative function resulting from the fact that terms denote
objects by means of concepts ;
 A good definition strengthening its term status.
 The naming biunivocity which ensures its monoreferetiality,
monosemantism and nonabiguity ;
 Membership to a specific terminological field.
 Its invariable value in a linguistic and professional community which
gives a term a certain stability ;

29 Termenul reprezintă forma lingvitică sau nelingvistică care desemnează un concept
determinat, impus printr -o definiție la la nivelul unui domeniu dat. (Cabré 2000: 35 in
Bidu -Vrănceanu 2007: 32 )
Termenii sunt “vectori ai cunoștințelor non -lingvistice” (Herat 1994 : 28 in Bidu –
Vrănceanu 2007: 32)
Termenul este o unitate cognitivă (a cunoașterii) realizată în orice tip de terminologie.
(Bidu -Vrănceanu 2007: 32)

 Stability related to its norming in the expe rts community or to its
coding by means of norms;
 The possibility that the stability of a term should be imposed
diachronically;
 The neutrality or the lack of variety as regards modality and affective
marks which gives priority to denotation , and leads to avoidance of
connotations (see 3.3.2.1 -3.3.2.3 .).
As suggested by Bidu -Vrănceanu (20 07: 34), these ideal features are
difficult to be found simultaneously in one term, but they represent a useful
instrument for individual and comparative descriptions of terms.
Furthermore, some might say that things are different with scientific,
technic al, legal and some bureaucratic language where there is a well -organized
terminology industry continually seeking to maintain the semantic isolation of
terms , and to counter the natural pressure of usage. Nonetheless, many terms
used in such languages are so frequently encountered in every -day situations
that they seem to have significantly altered their quality of terms . As regards
ordinary , non -technical language, the terminology is not established, but
similar attitudes to meaning are shown in comments o n language change in the
press, in politics, and in the education.
As regards the (lack of) fixity of meaning characterizing words and
terms , in ordinary language, most words gradually change their meanings in
time, while in terminology the specialists st rive to keep the meaning of terms
constant. The rare changes that are made at the level of terms are motivated by
factors external to the structure of language, and have to do with clarifications
or advances in the disciplines.
Moreover , the meanings of wo rds change by their frequent association
with other words (e.g. while enormous may be used with both pleasant and
unpleasant things, enormity is restricted to crimes, scandals and heavy burdens).
Thus, meanings are no longer expressed only by single words, so much the
more that in the creation of language text there is also the requirement of
expressing unique, unrepeatable meanings by means of a syntax and vocabulary
that must retain a high level of rigidity so that the text can be understood by all
the us ers of the language.
Meanings inevitably change in ad hoc situations, but words and
meanings that are protected by the conventions of terminology exclude , as far
as possible , any variation that is specific to a given occasion.
Nevertheless, these conven tions do not exclude the transfer of terms
from one domain to another.
Many terms make up collocations which are specific not only to
specialized , but also to general contexts. For example, the nouns lege, semn,
bloc, cod in Romanian, originally associat ed with a specific domain of activity,
make up collocati ons which extend far beyond that domain (see Table 12

below). The same holds valid for the corresponding English nouns law, sign
block and code which are associated w ith various domains of activity:

LEGE legal a încălca legea  to break/infringe/violate the
law
general a lua legea în propriile mâini  to take the law
into one’s hands, the law of the jungle  legea
junglei, legea este aceeași pentru toată lumea
(leg. gen.)  the law is the same for everybody
economic legea cererii și a ofertei (ec.)  law of supply
and demand.

SEMN philosophy semne naturale  natural signs
linguistics semn lingvistic linguistic sign, semn de
punctuație  punctuation mark.
science semn distinctiv  distincti ve sign.
general semn particular  distinctive sign, semn din
naștere  birth mark.

BLOC politics bloc politic  political block , bloc
parlamentar  parliamentary coalition.
drawing bloc de desen  drawing block /tablet.
constructions blocuri prefabricate  prefab panels, bloc
turn  tower house .
technical bloc motor  engine block .

COD linguistics cod lingvistic  linguistic code

legal cod penal criminal code, cod civil  civil
code, cod de comerț  commercial law, cod
de procedure penal  code of criminal
procedure .
semiotics cod de culoare  colour code , cod portal
postal code , cod Morse  Morse code , cod
secret secret code
biology cod genetic  genetic code
maths cod binary  binary code.

Table 12 . Examples of terms shar ed by general and special ized contexts
(source Dumtrașcu 2009d)

The distinction between words and terms is as important as useful in
approaching collocations30. The aspects presented above account for the fact
that, in spite of a tendency to isolate words from terms , not few are the
instances i n which terms are assimilated in the common language, making this
isolation impossible.
The same idea is pointed out by Opriț -Maftei (2018) who states that
“[T]he vocabulary o f each terminology is made up of specialized and highly
specialized terms, commo n words that acquired special m eanings and words
used in every day speech” (Opri ț-Maftei 2018: 29)
At the level of collocations this results in cases in which one and the
same lexical item may function as a word or term in the collocational pattern in
which it occurs. Consequently, this allows for collocations having the same
lexical item used either as a ‘node’ or as a ‘collocate’ (see 2.1.1 .) to be assigned
to various types of discourse, thus to various text types and functional styles
(see 4.2. – 4.3.).

4.6.2 ON THE SCIENTIFIC PROSE STYLE

Different from other functional styles of language, the scientific prose style
governs and conditions the language of science. Referring to this functional
style, Galperin (1977: 307) suggests that its general aim is t o prove a
hypothesis, create new concepts, to disclose the internal laws of existence,
development, relations between different phenomena, etc. Concerning the
language means used to achieve such aims, the idea is shared that they tend to
be precise, unemot ional, devoid of any individuality, there being a tendency
towards a more generalized form of expression.
Among the most important features of the scientific prose style , mention
can be made to:
 The use of logical sequence of utterances with clear indicat ion of their
interrelations and interdependence;
 The use of terms specific to each branch of science.
 The use of specific sentence patterns (postulator y, argumentative and
formulaic);
 The u se of quotations and references;
 The use of footnotes digressive in character ;

30 Distinction can be made bewtween terminological collocations and non-
terminological collocations. These two types of collocations cannot be clearly
differentiated because, on the one hand, they both form paradigms and, on the other, the
general and specialized experiences often intermingle.

 A high degree of impersonality.
If most of the above mentioned features are related to language
compartments which are less relevant for our approach to collocations, the use
of terms specific to each branch of science is an aspect of utmost im portance
for analyzing the semantic features of individual terms, and the ways in which a
term may combine either with other words, or with other terms, in order to form
specialized collocations.
From the very beginning, refe rence should be made that giv en the rapid
dissemination of scientific and technical ideas, certain terms begin to circulate
outside the narrow field they belong to and , eventually , begin to develop new
meanings, thus being ‘de -terminalized’.
This is not always a problem since, on the one hand, science is a very
prolific field of human activity in coining new words, and, on the other, the
majority of scientific terms do not undergo the process of determinalization,
remaining thus, the property of scientific prose. In addition, the neces sity to go
deeper into the essence of things and phenomena gives rise to new concepts,
favouring the creation of new terms to name them in developing sciences.
Scientific prose texts are most often, if not always, very clear. Their
clarity results, on the one hand, from the fact that terms make more direct
reference to something and do not require a descriptive explanation, on the
other. In addition, the general vocabulary used in scientific prose bears its direct
referential meaning. That is to say, words specific to scientific prose tend to be
used in their primary logical meaning.
In spite of being regarded as self -explanatory to the greatest possible
degree, new terms in scientific prose are often followed (or preceded) by an
explanation. Similarly, neu tral and common literary words used with a different
meaning in scientific prose are usually explained, either directly in the context,
or in a footnote.
Finally , yet importantly, the exchange of terms between various branches
of science is an interesting phenomenon, which favours not only the
interpenetration of scientific ideas, but also the sharing of terms and li nguistic
patterns , whenever possible.
To sum up, starting from the assumption that the topic is assumed to
control the vocabulary selection, and normal scientific writing is assumed to
exclusively provide information, the conclusion can be drawn that structures
which are overtly interactive are often absent from such texts. It logically
follows that limitations in the messages of some varieties lead to simplification
in the structure of the language used.
Moreover, since the terminological tendency isolates words by trying to
avoid any semantic alteration due to their contact with other words, the terms
consisting of more than one word are like ly to be fixed and isolated. Isolation is
useful at word level as it helps to “preserve the pretension of a pure terminology

and to impose a similar conservatism on the general use of vocabulary”
(Sinclair 2004: 159).

4.6.3. ON THE STYLE OF OFFICIAL DOCU MENTS

The style of official documents (also called “officialese”) has a definite
communicative aim and has its own system of interrelated language and stylistic
means. As regards the aims, the style of official documents states the conditions
binding two parties in an undertaking with the purpose of reaching agreement
between the two contracting parties.
Although not essential, the most striking feature of the official documents
style is its special system of clichés terms and set expressions by which eac h
substyle can be easily recognized.
The specificity at the level of terms , phrases, and expressions is an
important aspect distinguishing the various substyles subsumed to the style of
official documents. Moreover, since most varieties of official langua ge usually
have their special nomenclature which is conspicuous in the text, this
nomenclature becomes easily discernible as belonging to the official language
style. In addition, these varieties of style have a common feature: i.e. the use of
specific abb reviations.
Last, but not least, words are used in their logical dictionary meaning in
official documents, because the style these documents belong to allows neither
the presence of contextual meanings, nor the use of words with emotive
meaning .
One could state that central to the expressive use of word combinations is
the creative manipulation of their form, and that creative modification and the
associated play on literal and transferred meanings are always tied to a specific
context. In addition, the d eliberate variation, or even distortion, of an idiom or
formula in order to achieve a particular stylistic effect is quite a common device
used in both speech and writing. As Cowie (1998: 12) puts it
“deliberate variation or even distortion of an idiom or formula to achieve a
particular stylistic effect is a common device of speech and writing and has to be
distinguished from variation within a multiword unit which is familiar and
systemic. Creative modification and the associated play on literal and trans ferred
meanings are always tied to a specific context”.
What is ‘normal’ and possible will always be a matter of stylistic choice
and relative to dynamic and negotiable interactive context (see 3.5.3. ).
Moreover, as regards word meanings,
“in a synchroni c view of language, the origins of meaning are not under scrutiny.
New meanings are constantly being created, mainly through gradual movements

of collocation , and occasionally a word/meaning relationship is given the status
of a term , and thus a measure of protection within the discourse of a specialized
genre ” (Sinclair 2004: 160) .

CONCLU DING REMARKS

The most relevant aspects approached in the present chapter ar e summarized
below.

1. Style implies a choice of linguistic means made by the individual sp eaker or
writer in accordance with the requirements of the communicative situation.
Although the author’s personality and communicative intention, the time , place ,
medium and province of discourse are important factors conditioning the style
of any communi cation (oral or written), of no lesser importance are the textual
norms which constrain the choice of linguistic means in terms of vocabulary,
sentence types, text patterns, and macrostructures.

2. The functional style to which the expressive means of lan guage are attached
becomes an important point of reference in a stylistic analysis. Observing
functional styles is essential in communication because this automatically
implies adjusting one’s expressive means to the purpose of communication. The
speakers who are aware of and responsive to language variations are generally
agreed to recognize the functional styles as independent wholes. Moreover, their
choice of language means is always conditioned by the aim of their
communication, the result being a more or less closed system.

3. The texts specific to the substyles making up the functional styles of
language may be analysed if their linguistic components are presented in their
interaction, thus revealing the unbreakable unity and transparency of
constru ctions of a given type.

4. Stylistic devices and expressive means , which are concerned with language
problems such as the aesthetic function of language, synonymous ways of
rendering one and the same idea, emotional colouring in language , interrelation
between language and thought, the individual manner of an author in making
use of language, are elements which secure the desirable effect of utterances.

5. A stylistic evaluation of words might be possible if reference is made to
their categorization acco rding to the three generally agreed layers of
vocabulary: i.e. literary , neutral and colloquial . Each of these three layers is
represented by specific words, and these words, in their turn, may combine

forming patterns which should, most often, preserve th e specificity of the layer
the respective words belong to.

6. The distinction between words and terms is as important as useful in
approaching collocations . In spite of a tendency to isolate words from terms ,
not few are the instances in which terms are assimilated in the common
language, making this isolation impossible. At the level of collocations this
results in cases in which one and the same lexical item may function as a word
or term in the collocational pattern in which it occurs. Consequently, th is
allows for collocations having the same lexical item used either as a ‘node’ or
as a ‘collocate’ to be assigned to various types of discourse, thus to various text
types and functional styles.

7. Central to the expressive use of word combinations is th e creative
manipulation of their form. Creative modification and the associated play on
literal and transferred meanings are always tied to a specific context. In
addition, the deliberate variation, or even distortion, of an idiom or formula in
order to ac hieve a particular stylistic effect is quite a common device used in
both speech and writing.

5. A TRA NSLATION -ORIENTED APPROACH TO ENGLISH
COLLOCATIONS ACROSS ROMANIAN AND ITALIAN
BORDERS

5.1. APPROACHES TO THE TRANSLATION OF COLLOCATIONS

The issue of translation from one language culture (LC) into another has always
been as controversial as challenging. The numerous and various approaches
suggested by the specialists in the field have often been debated, sometimes
even rejected. The conclusion drawn , in many cases , has been that this complex
process is greatly influenced by a series of factors such as: the source text (ST),
the target audience, the source language (SL) and target language (TL)
constraints (the lexical, morpho -syntactic, semantic and pragmatic levels, the
cultural context and barriers), as well as the translator’s background knowledge
and competences.
An aspect of utmost importance in translation is collocability , which is
specific to each and every language. In sp ite of the debates on the myth of the
‘perfect translation’, or of the ideal translator, not few are the instances when
collocability brings about translation problems, even in the case of translators
who have good lingu istic and cultural competences. Henc e the statement that
translators, as non-native speakers of the TL , are always ‘caught’ with
collocations . In o rder to avoid such situations, translator s should have a very
good command of the two L Cs brought into contact, i.e. they should speak their
moth er tongue very well and they should also be linguistically as competent as a
native speake r of the TL31.
As regards possible approaches to translating collocations , the idea is
shared that there are two main approaches to lexis translation . One possible
approach is linguistically -oriented and focuse s on formal aspects such as TL
lexical gaps or problems caused by structural differences between languages,
whereas the other attaches more importance to the cultural constraints
characterizing lexis in general, a nd collocations in particular.

5.1.1 LINGUISTIC APPROACHES

Within the linguistic approach to the translation of lexis, comparison has often
been drawn between the validity and usefulness of Chomskian linguistics on the
one hand, and Firthian linguistics , on the other.

31 It is generally known that the translation of official documents, especially the EU
documents, requires that the translator should be a TL native speaker (Wagner, Bech
and Martinez 2002).

If the Chomskian theory was primarily concerned with “an ideal speaker,
in a homogeneous speech community (see 2.3.1.1 .) who knows its language
perfectly and is unaffected by memory limitations, distractions, shifts of
attention and interes t, and errors in applying his knowledge of that language in
actual performance” (Chomsky 1965: 3), the key element in Firthian theories
was considered to be “the speaking person in the social process” (Firth 1957:
190) who was also the creator and preserve r of cultural v alues .

When attention is shifted from the process of communication to the
process of translation , the two approaches undergo some obvious changes.
This is due to the fact that in the case of the latter the translator has to be an
ideal sp eaker of the two languages brought into contact and a speaking person
in two social processes able to transfer information from the source language
culture (SLC) to the target language culture ( TLC ).

Moreover, while Chomsky is mainly interested in syntax , Firth focusses
on meaning , his approach drawing on the context of situation . In his opinion,
“contexts of situation are not actual things, persons found in the immediate
environment of speakers, nor are they mere descriptions of the things and people
involved in a particular utterance. They are the more generalized constructs
abstracted from real linguistic experience applicable to, and revisable in the light
of new instances of language in use” (Firth 1968: 175).
Seen from this perspective, meaning acquires a new sense, being
regarded as the result of all “situational relations in a context of situations”
(Firth 1957: 19).
In an attempt to describe and classify typical contexts of situation, Firth
suggests elementary categories such as text form (oral or written), register
(formal, informal, colloquial), events location (schools, law courts, churches,
etc.), thus anticipating text and corpus classifications later made by corpus
builders.
Although the above mentioned linguistic models have been very frui tful
in linguistics, they have also been a valuable source of inspiration for the
translation theorists interested in finding possible solutions to translation in
general, and to the translation of lexis, in particular.
Nida’s translation theory, for exam ple, owes a lot to Chomskyan
linguistics and to the notions of transformations and kernel sentences. In Nida’s
(1964) view, kernel sentences could become useful instruments in translation
due, on the one hand, to their similarity across languages, and to “ the high
degree of parallelism between formal classes of words (e.g. nouns, verbs,
adjectives, etc.) and their basic functions (objects, events, abstracts and
relationals)” (Nida 1964: 68), on the other. In fact, such sentences are units of
measure used by Nida in supporting his translation method. He considers that
“it is both scientifically and practically more efficient (1) to reduce the source
text to its structurally simplest and most semantically evident kernels, (2) to
transfer the meaning from sour ce language to receptor language on a structurally
simple level, and (3) to generate stylistically and semantically equivalent
expression in the receptor language” (Nida 1964: 68).
Clear and accessible as it may seem, such an approach is not always a
valid solution in translation. This is due to the fact that, even if all languages
share certain deep -structure semantic and syntactic properties, this does not
necessarily imply that there must be “some reasonable procedure for translating
between languages” (Chomsky 1965: 30).
In spite of the obvious influence exerted by the Chomskyan model, Nida
is fully aware that for a translation to be appropriate, special attention has to be
devoted to the fact that the meaning of any linguistic form is greatly condition ed
by “the situation in which the speaker utters it, and the response it calls forth

from the hearer” (Bloomfield 1933: 139 , in Kenny 2000: 7). Therefore, he
suggests renouncing at the ‘formal equivalence’ in favour of the ‘dynamic
equivalence’. In Nida’s opinion, applying such a strategy in translation is meant
to ensure an equivalent response on the part of the target audience, rather than
to focus on a strictly formally equivalent target text.
Nida’s distinction between formal and dynamic equivalence has proved
to be extremely useful, not only in the case of texts and sentences, but also
when translating fixed lexical patterns, more or less semantically opaque and/or
culturally (un)marked.
As regards the applicability of Firth’s contexts of situation in translation
studies, it is rather limited and disputable. This is due to the fact that contexts of
situation are difficult to identify and to make operational in studies of real
translation, on the one hand, and to the fact that humans are unlikely to be
confronted with identical situations, on the other. Although restrictive and
limited, this view is shared by Catford who states that “SL and TL texts or items
are translation equivalents when they are interchangeable in a given situation”
(Catford 1965: 49) .
Linguistic approaches to lexis translation such as those mentioned
above have often been rejected because they were considered to create only the
illusion of equivalence32 between languages, effacing the role of the translator
and the influence of other f actors in cre ating meaning, but I consider that they
are valuable resources still valid in approaching the translation of collocations .

5.1.2 . CULTURAL APPROACHES

One of the important factors conditioning the process of translation to a great
extent is the cultural specificity of languages . Hence, the generally agreed
view, which I share, that translation should not be limited to linguistically –
oriented approaches. This is due to the fact that the process of translation does
not imply a mere transfer of words from one language into another, but also a
transfer of cultural values which are embodied in language at all its levels.
However, mention should be made that culture does not always create
problems in translation. If people from different cultures s hare common
experien ce, this allows setting up similar human situations, of a concrete or
abstract nature, thus creating a bridge between the SL and TL cultures. The

32 Equivalence is a relative concept. On the one hand, it is determined by the
historical -cultural conditions under which texts (both original and secondary ones) are
produced and received in the target culture and, on the other, by a range of sometimes
contradictory and scarcely reconcilable linguistic textual and extralinguistic factors and
conditions. Kay Dollerup (2006) suggests rep lacing the notion of equivalence with that
of approximation .

more problematic situations are those in which the SL realities lack
correspondents in th e target culture, because in this case neither the translator’s
knowledge, nor his competences may help him/her to produce an appropriate
translation.
As regards culture, specialists in the field have suggested numerous and
various approaches (see 2.4). Tylor, for example, considers that the term
culture denotes “that complex whole which includes knowledge, belief, art,
morals, law, custom and any other capabilities and habits acquired by man as a
member of society” (Tylor 1871 /1958:1 in Holliday, Hyde and Kullman 2006:
59).
Kroeber and Kluckhohn suggest that culture “consists of patterns,
explicit and implicit of and for behaviour acquired or transmitted by symbols,
constituting the distinctive achievement of human groups, including their
embodiment in art ifacts” (Kroeber and Kluckhohn 1952: 181 in Katan 2004:
25). Hofstede’s view on culture is also worth mentioning. In his opinion,
“[C]ulture is the collective programming of the human mind that distinguishes
the members of one human group from those of an other. Culture in this sense is
a system of collectively held values” ( Hofstede 1984: 21)33.
Given the complex nature of culture (see 2.4), it becomes obvious that
transferring cultural values from one LC into another is an extremely difficult
task. Certain translation problems arise from the fact that translators have to
find the best ways of complying with the cultural norms of the two languages
brought into contact. In other words, they have to decide whether the SL
cultural norms should take priority ove r the TL ones, or vice versa. Moreover,
translators should consider the possibility of comb ining the norms of the two
LCs. Depending on the strategy selected by translators, the result may be a SLC
bound translation, a TLC bound translation or a hybrid translation, which will
be “a product of compromise between two or more cultures” (Schäffner and
Adab 1997: 329).
As regards the factors influencing the translator’s choices in terms of
cultural strategies, in certain instances priority is given to the text t ype or to the
purpose of the translation, in others the translator’s choices are influenced by
the status of the ST or ST author, by the evolution of existing genres/text types
or by the creation of new – hybrid – ones. In fact, all translations could be

33 Saville -Troike (1986) provides a similar definition of culture . He suggests that
“[C]ulture encompasses all of the shared rules for appropriate behaviour that are
learned by individuals as a c onsequence of being members of the same group or
community, as well as the values and beliefs that underlie overt behaviours” (Saville –
Troike 1986: 47 -48). Moreover, Pinker (1995: 57) states that “[ C]ulture refers to the
process whereby particular kinds of learning contagiously spread from person to person
in a community and minds become coordinated into shared patterns”.

considered hybrids due to the fact that they imply “the transplant of the source
text into an alien culture environment” (Schäffner and Adab 1997: 334).
The coexistence of SLC -bound translations, TLC -bound translations and
hybrid translations also holds val id when referring to collocations.
The choice of words, word combinations , and structures in translated
texts is often made under the influence of the ST. Thus, in spite of being
semantically and grammatically correct when taken in isolation, such words,
word combinations , and structures will seem artificial to the TL reader. Still it
should not be ignored that
“the degree of artificiality depends on the translator’s competence, as well as
approach to the ST, e.g. the translator’s intention to remain fai thful to the
original or the prioritizing of the readability and acceptability of the TT in the
new cultural context” (Schäffner and Adab 1997: 334).
Consequently, it is the translator’s cultural background knowledge that
conditions the appropriate transf er of collocations from a given LC into another.
For example, collocations such as aunt Sally (chiefly Br. E.) which means ‘a
scapegoat’ – so called from the figure used as a target at fairs, bachelor chest
which denotes ‘a low chest of drawers of the 18th century, having a top inclining
to form a writing surface’, blanket sheet which refers to ‘a newspaper of larger
than average size, common in mid -19th-century England’, or Boston hooker
(also Irish boat or dundavoe ) meaning ‘a small fishing boat used aro und Boston
in the late 19th century, derived from an Irish model having a cutter rig’, can be
appropriately transferred in another language if the translator is able to
linguistically and culturally decode the SL patterns, on the one hand, and to
encode th em in the TLC by providing culturally -functional equivalents (if there
exist any), on the other.
Starting from the assumption that any translation from one LC into
another is an instance of intercultural communication, one can mention that
translators act as cultural mediators , i.e. they are the persons “who facilitate
communication, understanding, and action between persons or groups which
differ with respect to language and culture” (Taft 1981: 53 in Katan 2004: 17).
To be more specific, translators have to establish and balance communication
between different people or groups by interpreting the expressions, intentions,
perceptions and expectations of each cultural group. In order to fulfill this task ,
translators must be able to participate to some exte nt in both cultures, i.e. they
must be bicultural.
As regards the knowledge and skills required, Taft (1981: 73 qtd. in
Katan 2004: 17) suggests that a translator, as a cultural mediator, should
possess:

 knowledge about society : history, folklore, traditi ons, customs;
values, prohibitions; the natural environment and its importance;
neighbouring people, important people in the soci ety, etc.;
 communication skills : written, spoken, non -verbal;
 technical skills : those required by the mediator’s status, e.g. computer
literacy;
 social skills : knowledge of the rules that govern social relations in
society and emotional competence, e.g. the appropriate level of self –
control.
These four areas of knowledge and skills correspond to specific
competences 34 which make up the communicative competence of a translator.
By appropriate ly combining these competences (see 5.4.) translators can ensure
effective cultural communication between the representatives of the SL and TL
cultures.
Even though the distinction between linguistic and cultural approaches
to translating collocations is operational in s ome situations, it is generally
agreed that linguistic and cultural aspects have to intermingle so that
‘equivalence ’ between the SL and TL collocational patterns may be achieved .
Nevertheless, if the fact is taken into consideration that equivalence is
impossible to achieve in any type of translation situation, reference could be
made to C hesterman’s “ expectancy norms ” (1997), which point to the idea that
the translation should compile with the expectations of the target readers.
Chesterfield’ expectancy norms originate in Toury’s (1995, 1999)
concept of translation norm . In Toury’s opinion, the quality and value of a
translation is conditioned by two elements. On the one hand, a translation is a
text meant to fulfil a certain function in a particular culture , and, on the other,
it is a representation of a text which was originally written in another
language and which belongs to a different culture . Moreover, he suggests that
these two conflicting aspects coexisting in a translation can be coped with if
translators apply the so -called ‘ initial norm ’ which implies choosing the
dominant approach when confronted with selection from alternatives. In other
words, the ‘initial norm’ “i s a choice between ST adequacy and TL
acceptability, where the translator decides a priori a certain hierarchy and

34 Grammatical competence includes knowledge of the rules of the code.
Sociolinguistic competence is given by the knowledge of an ab ility to produce and
understand utterances appropriately in context, i.e. as contrained by the topic, the status
of the participants, purposes of the interaction, etc. Discourse competence is related to
the ability to combine form and meaning to achieve un ified spoken or written texts in
different genres. This unity depends on cohesion in form and coherence in meaning.
Finally, strategic competence results from the mastery of communication strategies
which may be used to improve communication or to compensa te for breakdowns (Bell
1991: 41).

identifies which set of norms will prevail when selecting or rejecting translation
choices” (Cozma 2006: 31).
Besides the ‘initial norm’, To ury identifies two groups of norms
belonging to more specific levels, i.e. preliminary and operational norms . The
former are related to the translation policy in a given culture and the
directness of translation , whereas the latter, i.e. operational norms “direct the
decisions made during the act of translation itself” (Toury 1995: 58) being
represented, on the one hand, by matricial norms (affect the manner in which
the linguistic material is arranged in the TT) and by textual -linguistic norms
(influence t he textual make up and the verbal formulation of the TT), on the
other.
Focussing on Toury’s initial and operational norms, Chesterman (1997)
distinguishes between “expectancy norms ”35 and “professional norms ”36
which correspond to product and process norm s, respectively. If expectancy
norms are established by “the expectations of readers of a translation
concerning what a translation should be like” (Chesterman 1997: 64), the
“professional norms ” govern the accepted methods and strategies of the
translatio n itself.
As far as “expectancy norms ” are concerned, the readers’ expectations
are influenced by various factors, such as the translation tradition in the TLC,
the formal features of parallel texts in that particular language, economic and
ideological fa ctors or power relations between cultures.
Referring to the same aspect, Chesterman (1997: 64 ) suggests that the
readers’ expectations may regard text -type and discourse conventions,
grammaticality, style, collocations, lexical choice and so on. On the ot her hand,
professional norms have to be regarded from the translators’ perspective.

35 Expectancy norms are basically validated by their very existence in the TLC.
Nevertheless, these norms are sometimes validated by an authority of some kind such
as a translation critic, a teacher, an examiner, or a literary critic reviewing a translation.
36 Chesterman subsumes the professional norms under three higher -order norms,
namely the accountability , communication and relation norms. According to the
former norm “a translator should act in such a way that th e demands of loyalty are
appropriately met with regard to the writer, the commissioner of the translation, the
translator himself or herself, the prospective readership and any other relevant parties”
(Chesterman 1997: 68). The communication norms stipulat e that “a translator should
act in such a way as to optimize communication, as required by the situation, between
all the parties involved” (Chesterman 1997: 69). Finally, the relation norm states that
“a translator should act in such a way that an appropr iate relation of relevant similarity
is established and maintained between the source text and the target text” (Chesterman
1997: 69).

The competent professional translators are the ones conditioning the
appropriate application of these norms. In Chesterman’s opinion, “professionals
are the people who are largely responsible for the original establishment of the
expectancy norms . In fact , the products of their work naturally become the
yardsticks by which subsequent translations are assessed by the receiving
society. Their translation behaviour, in other w ords, is accepted to be norm –
setting. Conversely, if a translation is accepted as conforming to the relevant
expectancy norms, the translator of that text is … accepted as being a competent
professional” (Chesterman 1997: 68 -68).
To put it in a nutshell, when confronted with the translation of a text from
a given SL into a TL, the translators have to make their choice according not
only to the text type involved, but also to the wishes of the commissioner and
the needs of the prospective readers. Certain types of texts favour a close
formal similarity to the original (legal contracts), while others value stylistic
similarity (short stories, poems), stress the importance of semantic closeness
(scientific or technical article), or aim at similarity of effect ( tourist brochures,
advertisements).
As regards specialized translations , one possible alternative to
equivalence , which is extremely useful in approaching such translations, is the
skopos theory suggested by Reiss and Vermeer (1996). According to this
theory, the methods and strategies selected in the process of translation are
determined by the intended function (skopos ) of the TT, not by the function of
the ST (Reiss & Vermeer 1984). This implies that the ST is no longer the norm
in translation and equiva lence becomes only a complementary notion.
One very important difference exiting between the linguistic -oriented
models of equivalence and Vermeer’s functionalist model is the importance
attributed to the ST. If the supporters of the former models regard t he ST as a
norm and accept a translation only if it is equivalent to the ST, the
representatives of the functionalist approach to trans lation asses the quality of
a translation in relation to its functioning as a text in the TLC. Under these
circumstances, the function of the translated text in the TLC is the key element
determining which aspects of the ST should be transferred to the translation and
which should be ignored. Consequently, the ST loses its function as a point of
reference in terms of which e quivalence is measured.
Christiane Nord (1991, 1997) provides another interesting and useful
insight into the interpersonal interaction of the translation process. In her
opinion, the initiator (whether a client, the ST author, the TT reader or, even the
translator) conditions the translation process according to the function (or
skopos ) (s)he envisages for that text in the TLC (Nord 1991:6). Nevertheless,
the ultimate responsibility does not rest with the initiator, but with the translator,
who is the only person qualified to judge whether the transfer process has taken
place satisfactorily.

Nord (1997) modifies the conventional skopos theory by adding the
concepts of loyalty and convention in this way limiting the variety of possible
functions or skopos . In her opinion, the concept of loyalty can be defined as “a
moral category which permits the integration of culture -specific conventions
into the functionalist model of translation” (Nord 1997:63). In other words,
loyalty implies that translators should tak e into account the conventions of the
particular translation situation. Since these conventions are not imperative, the
translator is free to ignore existing conventions. By combining functionality
and loyalty, the translator can aim at producing a functio nal target text which
conforms to the requirements of the initiator’s brief and which will be accepted
in the TLC. Such an approach is obviously contrary to equivalence – based
translation theories, because the demand for equivalence in translation is
subo rdinate to the skopos rule. That is to say, if the translating instructions
require the function of the text should change, then ST equivalence is no longer
a priority. However, loyalty towards both the author and the readers of the
translation obliges the translator to make clear specifications as regards both
what aspects of the original have been taken into account and what aspects have
been adapted (Nord 1992: 40).
According to Norm’s functionalist approach, a translation is considered
to be adequate i f the translated text is appropriate for the communicative
purpose defined in the translation brief made by the initiator.
Instead of a conclusion , I should mention that an approach to the
translation of c ollocations can hardly, if ever, be unilateral. When translating
collocations , lexical aspects are dominant, but semanti c and cultural constraints
can, by no means, be ignored. As suggested by Baker, the difficulties that
translators have to overcome in translating occurrences of fixed lexical patterns,
whether they are collocations, fixed expressions, metaphors or idioms, consist
primarily in “achieving equivalence above word level” (Baker 1992: 46). In
addition, translators have to reduce translation loss to an acceptable minimum
“by deciding which of th e relevant features in the ST is most important to
respect, and which can most legitimately be sacrificed in doing so” (Hervey et
al. 1995: 17).

5.2. TYPES OF TRANSLATION37

The terminology used to refer to or describe the process of translation may
vary, but the choice of terms is merely an ideological one. Whether using the
distinction ‘ free’ vs. ‘literal’ translation, ‘dynamic equivalence’ vs. ‘formal
equivalence’ (Nida 1964), ‘communicative’ vs. ‘sem antic translation’

37 The concept of types of translation is interrelated with that of types of texts .

(Newmark 1988 ), ‘domesticating’ vs . ‘foreignizing’ translation (Venuti 1995),
or ‘minimal mediation’ vs. ‘maximal mediation’ (Nabokov 1964) , reference is
made to one and the same complex process characterized by many variables,
various constraints and constant debates among specialists.
Although numerous, the terms deserve special attention, due to the fact
that each of them points to a specific translation method or approach , which
may prove extremely useful if chosen and used at the right time and place.
If reference is made to word -for-word translation , literal translation ,
faithful translation , semantic translation , adaptation (Dollerup 2006 ), free
translation , idiomatic translation , communicative translation (Nida 1988) , the
first four types distinctly manifes t a SL emphasis, while the last four a TL
emphasis38.
5.2.1. In the case of word -for-word translation SL words are closely
followed, SL word order is preserved, word meanings are taken out of context
and cultural words are translated literally. For example, if collocations such as
to deny an accusation and to dye one’s hair are approached by applying the
word -for-word translation, the resulting TL variants will be to deny an
accusation  a nega o acuzație , to dye one’s hair  a colora părul . Such
variants preserve the word order and the meaning of individual words in the SL,
but are culturally inappropriate, due to their literal translation. The
collocationally appropriate equivalents would have been a-și vopsi părul and a
respinge o acuzație , respectively.
Literal translation , on t he other hand, implies that SL grammatical
constructions are converted to their nearest TL equivalent, lexical items being
translated in isolation, i.e. out of context. For example, the translation of a
collocation such as to register for a course by the R omanian * a se înscrie
pentru un curs , is an illustration of the fact that the grammatical patterning in
English has been literally preserved in the TL. The preposition pentru in
Romanian is specific to the adverbial modifiers of purpose and cause, and
shou ld be replaced by the specific preposition la normally used with
prepositional object s in examples such as the one under discussion. Considering
the aspects above, the conclusion can be drawn that both word -for-word and
literal translations are only prelim inary stages in the translation process, if
applied to the translation of certain collocations.
Furthermore, faithful translations reproduce the precise contextual
meaning of the SL words but in keeping with the constraints of the TL

38 Translation “methods” include word-for word , literal , faithful idiomatic
translation, adaptation , whereas semantic and communicative translation point to
possible “approaches” which may be used in the translating process according to
specific text types . Different from translation approac hes, translation methods are
used irrespective of text types.

grammar. Different fr om word -for-word translations , faithful translations
imply that cultural words should be transferred from the SL to the TL, and the
same applies to the grammatical and lexical patterns. Illustrative in this respect
could be the collocations black Frida y39 vinerea neagră and to taste bitter 
a avea gust amar/ a fi amar la gust . Since this kind of translation is as faithful
as possible to the original writer’s intentions, it can be labelled as
uncompromising and dogmatic.
Semantic translation focuses on the a esthetic value (the beautiful and
the natural “sound”) of the SLT, compensating and compromising on meaning.
Cultural words may be translated by a third culturally neutral term or by a
functional term, but not by using cultural equivalents. As compared to other
types of translation, semantic translation is more flexible, more creative and
more imaginative, largely allowing the translator’s empathy to work. For
example, the semantic translation of a collocation such as to use one’s mind
would imply selectin g the TL functional and culturally neutral variant a-și folosi
inteligența , rather than one of the TL metaphorical equivalents a-și folosi
mintea or a-și folosi capul .
The freest form of translation is adaptation . SL themes, characters, and
the plot are preserved, when using this strategy, but cultural terms are converted
to the TL culture and the text is practically re -written. Thus, adaptation is often
preferred when translating plays (comedies) and poetry.
Free translation will also give translators a certain freedom, as this
strategy implies reproducing the matter (by paraphrasing) in a different manner.
In other words, when choosing free translation translators have to preserve the
original context without necessarily preserving the form.
An interes ting strategy sometimes used by translators is the idiomatic
translation, which reproduces the message of the original, but distorts shades
of meaning by showing a preference for colloquialisms and idioms where these
do not appear in the SLT.
Last, but not least, communicative translation attempts to convey the
most precise contextual meaning of the original, in this situation both content
and language being readily acceptable and comprehensible.
If compared and contrasted, I believe that only two of the tr anslation
types mentioned above, i.e. semantic and communicative translations , ensure
translation accuracy and economy. In addition, these two strategies are preferred
when transferring stock and dead metaphors , normal collocations , technical
terms , colloq uialisms , slang , phaticisms and ordinary language. As regards

39 Black Friday is ‘a name for various Fridays rigraded as disastuous, such as 6
December 1745, when the landing of the Young Pretender was announced in London,
and 11 May 1866, which saw a comme rcial panic at the failure of Overend
Gurney&Co’ (Knowls 2006: 81).

the expressive components (unusual collocations and syntax, striking
metaphors, neologisms) , they are usually rendered very closely, even literally,
in expressive texts while in vocative and informative texts they are normaliz ed
or toned down (except for advertisements).
Scholars, notably House (1977), speak of these two possible choices in
translation while attaching them different labels: 1) semantic translation : art,
cognitive translation, ove rt (culture -linked) translation, overtranslation and 2)
communicative translation : craft, functional or pragmatic translation, covert
(culture -free) translation, undertranslation.
Comparison is often drawn between these two types of translation in
order to highlight points of similarity and dissimilarity. Thus, a semantic
translation is likely to be more economical than a communicative translation
due to the fact that the former is written at the author’s linguistic level, whereas
the latter is adapted to t he linguistic level of the readership. It is also worth
mentioning that a semantic translation is more suitable for expressive texts
(more specifically for descriptive texts, definitions or explanations), while a
communicative translation is appropriate for informative texts.
If cultural components are brought into focus, mention can be made that
they are transferred intact in expressive translations , transferred and explained
with culturally neutral terms in informative translations , replaced by cultural
equivalents in vocative translations.
Furthermore, a semantic translation remains within the boundaries of
the SLC, assisting the reader only with connotations , whereas a communicative
translation displays a generous transfer of foreign elements with an em phasis on
the intended meaning, rather than on the message.
Semantic translations are personal, individual, searching for nuances of
meaning and tending to over -translate, although aiming at concision. On the
other hand, communicative translations are soc ial, concentrating on the
message (the referential basis or the truth of information is secured), the
tendency being for translators to under -translate them, in order to keep things
simple and clear. However, this type of translation sounds always natural and
resourceful.
In addition, a semantic translation requires interpretation, which most
often results in loss of meaning and dissimilarities with the original, whereas a
communicative translation implies explanation, being more idiomatic and
often consi dered better than the original.
Last, but not least, the authority of the SLT writer is recognized and local
flavour is preserved intact in semantic translations , which is not the case with
communicative translations . The cultural gap is bridged more easi ly in
communicative translations, as they conform to the universalist position
advocating universally common thoughts and feelings. Semantic translations

follow the relativist position according to which thoughts and feelings are
predetermined by the langu ages and cultures in which people were born.
5.2.2. If regarded from the perspective of equivalence , the purpose of any
translation should be the equivalent effect . This implies producing the same
effect (or one as close as possible) on the readership of the TT as the ST has on
the readership of the original. The principle of equivalent effect is also termed
equivalent response or, in Nida’s words, dynamic equivalence .
Dynamic equivalence can be equated with the reader’s shadowy
presence in the mind of th e translator, and contrasted with formal equivalence ,
i.e. equivalence of both form and content between two texts. Newmark (1981)
regards the equivalent effect as the desirable result , rather than the aim of the
translation. He argues that this result is u nlikely to be achieved in two cases.
Firstly, if the purpose of the SL text is to affect and the purpose of the TL text is
to inform, and secondly, if there is a clear cultural gap between SL text and TL
text (in fact, translation merely fills a gap betwee n two cultures if, felicitously,
there is no insuperable cultural clash).
5.2.3. Another important concept in translation studies is degree of
mediation . Whether considering minimal or maximal mediation (Nabokov
1964) , these two are terms used to convey th e extent to which translators
intervene in the transfer process, feeding their own knowledge and beliefs into
their processing of the text. In the case of minimal mediation , which
corresponds to Venutti’s (1995) foreignizing translation, the characteristic s of
the SLT are made entirely visible and few concessions are made to the reader .
As for maximal mediation the translation constitutes a radical departure from
the SLT in terms of regi ster membership, intentionality and socio -cultural
practices. Maximal m ediation is the equivalent term for Venutti’s (1995)
domesticating translation. Translators who domesticate texts adopt a strategy
through which the TL is culturally dominant. In doing so, culture -specific terms
are neutralized and re -expressed in terms of what is familiar to the dominant
culture. If translation is made from a culturally -dominant SL to a minority –
status TL, domestication protects the values of the SLC.
To sum up , in the complex and intricate process of translation , the
translators’ task is to continually search and research, deconstruct and
reconstruct the text, as their world is one of dichotomies pertaining to the
traditional areas of activity of translators (technical, literary, religious
translator, etc.), to modes of translating (writte n, oral) and to the translator’s
priorities or focus (literal vs. free, form vs. content, formal vs. dynamic
equivalence, semantic vs. communicative translating, translator’s visibility vs.
invisibility, domesticating vs. foreignizing translation).

5.3. TRANSLATION PROCEDURES AND STRATEGIES

Translation procedures40 are directly related to translation strategies . In spite
of their large number, only some of them can be taken into consideration when
translating collocations .
The translation procedure most often selected when transferring
collocations from one language into another is the literal translation , which is
further subdivided into word -for-word and one-to-one translation and implies
that the primary meaning of the word gains overall importance alo ngside the
norms of the SL grammar. Word -for-word translation is confined to simple
and neutral contexts in which no affective meaning is expressed, but even if this
is the case, denotational equivalence or quasi -equivalence is still a matter of
concern. A s regards one-to one translation , this is basically a matter of
collocation and the solution emerges quickly when there is transparency. e.g. a
depăși viteza /o situație  to exceed the speed limit , but * to exceed a situation .
Moreover, transference/ loan word/ t ranscription adoption/ transfer
is a procedure used by translators when one language lexicalizes a meaning
which is not lexicalized i n the other, either due to the fact that a particular
object/ state/ event or cultural artifact does not have an equivalent in the TL, or
because it is not known in that part of the world. For example, the noun snow
has no corresponding lexical item in equa torial Africa and this is the case with
the noun sand in the polar regions. Consequently, the collocations having these
nouns either as ‘nodes ’, or as ‘collocates ’ will find no corresponding
equivalents in the regions where they are not lexicalized.
This t ranslation procedure posits the problem of necessary and fancy
borrowi ngs from a given SL into a TL. Actually, the English borrowings before
the 1990s were necessary and motivated as imports meant to fill a cultural gap
(e.g. five o'clock –tea), whereas af ter the 1990s, both necessary ( CD, internet )
and fancy loan words (e.g. goalkeeper , play off , banner , party ) came into
Romanian , due to the new economic, social and political conditions. Another
interesting aspect related to non -motivated transference is t hat sometimes
doublets come into being. These are used alternatively , according to the
speaker’s intentions and to the context of utterance (e.g. goalkeeper – portar;
banner – panou publicitar; bodyguard – garda de corp).

40 In their discussion on direct and oblique translation, Vinay and Darbelnet (1995)
refer to the existence of seven translation procedures . They consider that direct
translation covers bo rrowing, calque, and literal translation, whereas oblique
translation includes transposition, modulation, equivalence, and adaptation. Moreover,
in their opinion these procedures are applied both at the levels of lexicon and
grammatical structures, and at the level of message (used to refer to the situational
utterance and some of the higher elements such as sentence and paragraphs).

Finding the appropriate cultural equivalent is an extremely useful
translation procedure when significantly different LCs are brought into contact.
It implies the recognition of similar cultural values within two cultural
frameworks, a kind of universal currency to which different labels a re attached
e.g. The White House – Palatul Cotroceni ; college – facultate . Since every
language is integrated into the culture of the linguistic community in which it
operates, it is important that translators, as mediators between languages and
cultures, should become aware of the cultural salient features, cultural
peculiarities and different patterns of thought specific to the LCs brought into
contact.
In such cases, translation e quivalence can be achieved if the translators
are able to identify and use the appropriate TL functional or descriptive
equivalent for the SL reality. If functional equivalents envisage finding
culture -free items (e.g. an English Blue Book  Pagini aurii ), descriptive
equivalents are chosen whenever the concepts to be translated are so culture –
bound that they allow only a description or a paraphrase (e.g. nursery rhymes /
silly rhymes / mother goose rhymes  poezii sau cântece pentru copii ).
Moreover, equivalence restricted to the idiomatic use of language is
achieved only if tran slators have very good linguistic, phraseological and
cultural competences both in the SL and TL cultures (e.g . a square peg in a
round hole  (potrivit ) ca nuca -n perete , soup ciorbă , smoked cheese 
cașcaval afumat , Enjoy your meal !  Poftă bună !, Break a leg !  Baftă!/
Noroc! )
Furthermore, mention should be made that the cultural specificity of the
text often represents a contraint in specialized translations . As suggested by
Federica Scarpa (2008: 187), the different ways of conceptualizing either gene ral
realities or the various historical and socio -political realities in two different
languages are visible in special languages , as well. To put it in Hervey and
Higgins’ (1992: 63) words, “exact synonymy between source -language and
target -language words is the exception rather than the rule”.
As regards specialized translation s, starting from Godman e Veltman’s
opinion (1990: 207 -209) that the lack of semantic equivalence between two
languages may create serious problems in this process, Scarpa (2008: 18 7)
mentions that such problems are traceable both in Italian and English scientific
contexts. A first example to illustrate what has been stated above could be the
abstract noun form and its concrete synonym shape , which have only one
equivalent both in Ro manian ( formă ) and Italian ( forma ). As far as the second
example is concerned, mention should be made that it is illustrative for the way
in which the process of “restrictive lexical focus” ( Godman e Veltman 1990)
differs in terms of linguistic and cultural specif icity. Referring to the gradual
semantic specialization of verbs such as to move, to remove, to replace, to
displace, to substitute, to exchange, to interchange used in scientific English,

Scarpa (2008: 187) notices that the English verbs replace, dis place, substitute
and exchange lack an equivalent with the same semantic features both in
general , and in specialized Italian:
Move – verbo generico che descrive un movimento
Remove – togliere /spostare /rimuovere un oggetto da una posizione
Replace – togliere /sèpstare /rimuovere un oggetto da una posizione e
metterne un altro al suo posto
Displace – sostituire un oggetto con la forza
Substitute – sostituire un oggetto con un altro di minor valore
Exchange – scambiare due oggetti tra di loro
Interchange – scambiar e due oggetti di valore equivalente tra di loro.
An important aspect worth mentioning at this point is that c ertain
translation procedures focus on semantic aspects . Bilingual lexical synonymy
is illustrative in this respect being a procedure intended to c apture specialization
of meaning which is obvious in the case of the Romanian verb such as a
străluci . This verb may be translated in to English by various synonymous
equivalents, i.e. to glimmer (a străluci slab și intermitent, a pâlpâi) to gleam (a
strălu ci slab), to glow (a străluci, despre o lumina roșiatică), to glisten (a
străluci, despre o suprafață umedă), to glint (a străluci, despre suprafețe
întunecoase), to shine (a străluci puternic), but the choice among these variants
will be conditioned by th e context envisaged. The componential analysis (see
3.2.1.) prove s extremely useful for this translation procedure since the search
for semantic primes (semes) may considerably facilitate the selection of the
proper contextual equivalent: e.g. to try [-Intensity], to endeavour [+Intensity] –
a încerca vs. a se strădui .
Considering semantic aspects in translation and the possibility of
identifying terminological equivalence , Scarpa (2008: 191 -191) suggests that it
is often very difficult to provide equivalents which match terms with only one
meaning, becaus e most often such terms lack an appropriate equivalent in the TL.
Consequently, in such cases the translator has to make approximations .
Palumbo (1999: 129 -133) considers that the translation procedures which can be
used by the specialized translators whe n approaching such terms are the
following:
 Analogical translation is valid only in the TT, thus not applicable to all
the uses of the same term.
 Descriptive translation is more vague and general as regards its
denoting the concept corresponding to the SL term (e.g. overbidding 
lasciarsi pren dere la mano nella corsa al rilancio ).
 Explicitation implies explanation of the SL term in the TL: For
example, the adjective harsh collocating with the noun cement in […] a
mortar made only with portland cement i s “harsh ” and does not flow
well on the trowel can only be translated in Italian by explaining the

specific meaning it acquires in this collocation: […] se usato da solo [il
cemento] può dare un impasto eccessivamente disomogeneo e
difficilmente lavorab ile con la cazzuola (Palumbo 1999: 132).
 Borrowing implies using the SL terms as such, on condition they are
semantically transparent in the TL. For example, the nouns coach and
orchestrator are preserved in Italian, because their meaning is accessible
to the target readers:
e.g. Anche la McKinsey and Company, una società di consulenza
aziendale, era convinta del fatto che la strategia a livello aziendale
dovesse essere gestita dal vertice aziendale. Essa riteneva che l’Alta
Direzione potesse creare valore sulla base di nove ruoli diversi. Tra
questi, per esempio, c’erano quelli di coach e orchestrator , dove l’abilità
e l’esperienza dei vertici dirigenziali potevano essere utilizzate per
addestrare, motivare e coordinare i dirigenti di divisione .
 Neologisms (creation of new words): e.g. diseconomie di scala
originating in economie di scala .
 Deletion, i.e. intentionally omitting a term which lacks an equivalent
in the TL. For example, high-employment surplus and standardized employ –
ment surplus , which have no equivalent in Italian and are easily
replaceable by their synonym full-employment budget surplus , may be
deleted in a text such as: There are other names for the full-employment
budget surplus. Among them are the cyclically adjusted (or deficit), the
high -employment surplus , the standardized employment surplus and the
structural surplus.  L’avanzo di piena occupazione viene definito
anche in altri modi, fra cui avanzo (o disavanzo) corretto per il ciclo e
avanzo strutturale.
The difficulty in finding equivalent terms in different languages is
obvious not only in the case of isolated terms , but also, significantly, in the
case of lexical patterns including terms . Such patterns can hardly be found in
bilingual dictionaries, thus they have to be extracted fro m parallel corpora.
Referring to the importance of the context in establishing the
collocational appropriateness of a term, Musacchio e Palumbo (2008) explain
that the collocation a new equilibrium used in tourism economics should not
be translated into Italian by processo di riequilibrio dell’eco nomia (‘a process
of economic rebalancing’). This is due to the fact that the term riequilibrio is
normally used in finances, being related to public debt accounts (as proved by
the analysis of Italian economic corpora), whereas the English collocation
points to an equilibrium that has to be reached or maintained in economy .Thus,
the translation variant considered appropriate by Musacchio e Palumbo in this
context is “una nuova posizione di equilibrio”.
In approa ching idioms , Baker (1992: 71) suggests combining formal and
semantic aspects in translation. She states that the translation of an idiom or

fixed expression into another language is conditioned by many factors and
suggests four strategies can be used in t ranslating idioms (1992: 71 -74):
 Using an idiom of similar meaning and form, i.e. using an idiom in
the TT which conveys roughly the same meaning as the SL idiom, and
consists of equivalent lexical items. This match can only occasionally be
achieved
 Using an idiom of similar meaning and dissimilar form , i.e. using
an idiom in the TT which conveys roughly the same meaning as the SL
idiom, but consists of different lexical items.
 Translation by paraphrase when a match cannot be found in the TL
or when it see ms inappropriate to use idiomatic language in the TT due to
the stylistic difference as regards the SL and TL preferences.
 Translation by omission, i.e. leaving out an idiom in the TT because
it has no close match in the TL, because its meaning cannot be e asily
paraphrased, or for stylistic reasons.
The translation procedures mentioned above are of utmost importance in
translating collocations and other fixed lexical patterns. Nevertheless,
translators are those who decide which choice suits best the purpo se of their
translation.
Whether limited to collocations, or not, translation implies a balanced
blending of linguistic and extralingustic aspects. Knowing the meaning of
words is part of linguistic competence that translators have to possess, but this
is far from being sufficient in translation. Extralinguistic aspects (cultural,
social, etc.) also have a significant role in translation, because unawareness with
respect to such aspects most often results in inappropriate translations.

5.4. THE TRANSLATOR : SKILLS AND COMPETENCES

Translators are generally regarded as mediators between languages and cultures.
To be more specific, a translator is a bilingual mediating agent between
monolingual speakers belonging to two different language communitie s.
Monoli ngual communication implies a sender and a receiver who share
the same linguistic code. The sender selects the message and the code, encodes
the message, selects the channel of communication and transmits the signal
containing the message, whereas the rece iver receives the signal containing the
message, recognizes the code, decodes the signal and finally retrieves and
comprehends the message.
Different from monolingual communication, bilingual communication
implies two different languages brought into conta ct and a single mediator, i.e.
the translator, who is both a sender and a receiver, responsible for the
appropriate decoding and encoding of the message. Unlike other receivers who

have a choice whether to pay more or less attention to their listening or r eading,
bilingual mediators (translators) have to interact closely with the SLT . Their
translation will be considered appropriate if the message of the original work is
transfused into the TL in such a way as to have the same effect on the TL
readers as th e original work would have on the SL audience .
Bearing in mind Chomsky's “ideal speaker” (see 5.1.1. ), reference can be
made to the existence of an ideal bilingual translator whose communicative
competence consists in a perfect knowledge of the two languag es in contact. If
such a translator existed indeed, he should be unaffected by theoretically
irrelevant conditions , such as memory limitations, distractions, shifts o f
attention or interest, errors – random or characteristic, in applying his
knowledge in a ctual performance.
Referring to the translator's communicative competence , mention should
be made that this competence is the result of more competences harmoniously
intermingled. Theoretically, translators should possess sensitivity to language,
linguisti c competence in both languages and communicative competence in
both cultures in order to create, comprehend and use context -free texts as a
means of participation in context. They should also possess “the ability to
research often temporarily the topic of the texts being translated, and to master
one specialism” (Newmark 1991: 49).
Communicative competence is influenced by the translators’
grammatical, sociolinguistic and discourse competences . If grammatical
competence is identified with the translator’s k nowledge and skills to
understand and express the literal meaning of utterances, s ociolinguistic
competence is attested by the knowledge of and ability to produce and
understand utterances appropriately in context. Moreover, discourse
competence is the tra nslator’s ability to combine form and meaning in order to
achieve a unified spoken or written text in different genres.
As regards collocations, their appropriate transfer from a given SL into a
TL depends on the translator’s lexical and semantic knowledg e, on the one
hand, and on his ability to compare and contrast cultural and collocational
features of the respective patterns in the two languages brought into contact, on
the other. If different types of texts are taken into consideration, then the scale
of constraints is likely to become even larger, pragmatic, stylistic and
sociolinguistic aspects being also under focus in this situation. In other words,
the appropriate translation of collocations from one language into another will
depend not only on th e translator’s competences, but also on his /her ability to
use these competences in order to overcome the translation constraints
characterizing this process.

Fig. 17. The translator: competences and constraints

Moreover, mention should be made that ce rtain collocations lack an
appropriate equivalent in the TL because the reality they denote is absent from
the TC. If the translator chooses to translate such patterns literally, the result
will be a TL pattern with a foreign flavour, which will be worse t han a non –
translated one. The effect will be immediately visible in the end -users’ negative
attitude towards the translation which will be nothing else but some poor
imitation of the “original”.
In order to avoid such situations, collocational and phraseo logical
competences need to be developed in trainee translators. They have to be
familiar with the ready -made phrases used in various registers in the two
languages in contact and they also need to know how to match and evaluate
them from a sociolinguistic point of view. Having fulfilled these conditions, the
translators will be able to handle ready -made language and will know when and
which ready -made units they can use.
For the same class of collocations, it is very important that translators
should ident ify the contextual restrictions in the SL and decide whether the
same restrictions and usage exist in the TL as well. The appropriate decoding
and use of lexical and grammatical association patternings will help translators

to interpret the original messag e correctly and to appropriately transfer it in the
TL.
Instead of a conclusion, the fact needs to be pointed out in translating
collocations, formal and sema ntic aspects are given priority. However, their
appropriate transfer in another LC is also condit ioned by pragmatic, stylistic and
cultural aspects . As regards the translator, the competences required for the
appropriate translation of collocations will vary depending on the formal,
semantic and cultural specificity of the pattern(s) to be transferred from the SLC
into the TLC.

5.5. COLLOCATIONS VS. OTHER (FIXED ) LEXICAL PATTERNS IN
TRANSLATION

Given the large number and great variety of fixed lexical patterns in English, I
consider that the problems concerning the translation of collocations may be
better understood if the aspects regarding the translation of this type of lexical
patterns are compared and contrasted with those regarding other fixed
expressions such as idioms, metaphors, similes, clichés and proverbs .

5.5.1. COLLOCATIONS IN TRAN SLATION

Collocations are expressions consisting of two or more words that correspond
to some conventional way of saying things. In the larger class of collocations , a
distinction is commonly made between lexical and grammatical collocatio ns
(see 2.1.3.1 .).
This distinction is important because each of the two types of collocations
requires a different approach in translation. In spite of their both having explicit
denotative meaning, lexical collocations require different knowledge and
competences on the part of the translator as compared to grammatical
collocations (‘colligations’) :

Fig. 18. Lexical and grammatical collocations in translation

As regards the strategies used in translating lexical and grammatical
collocations , reference is usually made to the form -based strategy (FBS) and
the meaning -based strategy (MBS) .
The former, i.e the FBS, is used with the structural, horizontal, parallel,
or symmetrical combinations and consists in a more or less direct rendering of
the ST collocations and coll igations by corresponding combinations and
grammatical structures in the TL.
As regards the MBS , it is conceptual, vertical, non -symmetrical and
consists in producing TL combinations and structures only on the basis of a
conceptual representation of the S T meaning. This strategy is valid in non –
equivalent situations, where the SL patterns have no equivalents with the same
structure in the TL.
The two strategies differ from each other both in terms of the underlying
cognitive processes and in terms of the f inal product. If cognitive processes
represent the criterion of evaluation, the MBS involves processing at a deeper
semantic level (Gernsbacher and Schlesinger 1997:123, Lonsdale 1997: 96),
whereas the FBS involves processing at a surface level. In what co ncerns the
final product, if translators choose to use the MBS the TT will have a lexical
and morpho -syntactic form different from that of the ST, whereas if the FBS is
used, the TT will be formally similar to the ST.

In translating lexical and grammatical collocations there are a series of
cognitive processes involved and they significantly influence the correct
decoding and encoding of the patterns envisaged :

Fig. 19. Linguistic and cognitive processes in translating collocations and
colligations (source Croitoru and Dumitrașcu 2006 a: 296, adapted)

The context, the syntactic complexity (e.g. embedded clauses), the lexical
combinations and, in the TT, the lexical retrieval in the appropriate style and
register, the discourse structure and the gramm atical sequencing are essential
elements in the decoding processes.

Language knowledge
and language use Decoding processes Translating and
interpreting strategy
context
situation
collocations Semantics
– reference assignment
– disambiguation
– conceptual
enrichment form -based strategy
meaning -based strategy
(using the intentionality in
the words, in negotiating
the meaning of
collocations)
discourse structure
grammatical structures
and sequencing/
colligations
embedded clauses Pragmatics form -based strat egy
(avoiding morpho –
syntactic inference from
the SL)
meaning -based strategy
(using syntax to provide
information)
referential cohesion
connectors
word order
Pragmatics
form -based strategy

Table 1 3. Language, decoding processes and translating strateg ies
(source Croitoru and Dumitra șcu 2006 a: 297 adapted)

The intentionality in the words, in negotiating the meaning of
collocations , the resistance to the language interference, and the knowledge of
the TL norms in terms of grammar and collocability are very important in
translation. M oreover, satisfactory pragmatic processing depends on the correct
interpretation of collocations and colligations, and on the inputs at the semantic
level (basic word meanings, context).
With non -symmetrical combinations and structures, the decoding
proces ses may go from reference assignment, through disambiguation, to
conceptual enrichment, the translational context and interpreting situation.
However, deviations and dilutions may occur which need to be corrected.
Therefore, adjustments of context may be p ossible before achieving an
equilibrium which meets the “expectations” of relevance.
As far as the form -based strategy is concerned, a word -by-word
rendering can be misleading and may result in a “pragmatic failure”, i.e. the
inability to understand what i s meant by what is being said. In addition, word –
to-word transcoding may not always be appropriate, sometimes being even
entirely inappropriate. In such cases translators should reformulate, and with the
exception of some technical terms, be wary of ‘equiv alents’. Consequently,
after reading and comprehension, with the other steps of the translating process,

the conceptual/linguistic processing of both collocations and grammatical
structures is decisive for the final product.
Furthermore, if formal aspects are taken into consideration, the fact may
be easily noticed that collocations quite often vary in length, depending on the
number of words to either side of the ‘node ’. As regards their semantic content,
short collocations have most often explicit denotat ive meaning, whereas longer
lexical patterns may have either denotative , or connotative meaning, allowing
thus different semantic interpretations, depending on the context in which they
are used.
The transfer of short collocations from one LC into another hardly
creates problems in translation . However, in this case , the translator should be
aware that collocations are rather rigid structures in which the ‘node ’ can have
only certain ‘collocates ’. Since in the case of collocations synonymy is not
always po ssible, the translator’s choice in terms of ‘collocates ’ is extremely
important. For example, in the collocation strong tea the adjective strong cannot
be replaced by powerful which, although a synonym, collocates with other
nouns. Similarly in the Romania n collocation ceai tar e and the corresponding
Italian lexical pattern te’ forte , the adjectives puternic and potente will never be
acceptable synonymous substitutes for tare and forte , respectively. In much the
same way, the verbs make and do will never be interchangeable in the
collocations they make up, in spite of their being synonym s in English. (see
Appendix 1 and Appendices 7.1. -7.2.)

Fig. 20 . Long collocations vs. short collocations in translation

Formal restrictions are also traceable in the E nglish partitives, which
can only be used with specific nouns. For example, the partitive a piece of

collocates with nouns such as: information, news, advice, furniture, cake,
chocolate, but it will never collocate with wine, beer, shampoo, etc., just like
the partitive a bar of collocates with nouns such as soap or chocolate , but never
with bread , meat, or cheese .
Sometimes, one and the same noun may be used with different partitives,
which will, of course, bring about a semantic change in the lexical s tructure: a
piece / an item of information, a piece / a bar of chocolate  RO: o bucată /
tabletă de ciocolată ; IT: un pezzo / una tavoletta di cioccolato , a glass / a bottle
of juice/ wine  RO: un pahar / o sticlă de vin ; IT: un bicchiere / una bottiglia
di succo/ vino , a glass / a tin of beer/Coke  RO: un pahar / o cutie de bere/
Coca -cola; IT: un bicchiere / una lattina di birra / coca cola etc.
If we were to consider the noun ice and the partitives with which it
usually collocates in English and Romanian, it can be easily noticed that there is
perfect equivalence between the two langua ges: a chunk of ice  RO: o bucată
de gheață ; IT: un pezzo di ghiaccio ; a crust of ice  RO: o pojghiță de gheață ;
IT: una crosta di ghiaccio , a cube of ice  IT: un cub de gheață ; IT: un cubo
di ghiaccio , patches of ice  petice de gheață ; porțiuni acoperite cu gheață ;
IT: chiazze di ghiaccio , a sheet of ice  RO: o pojghi ță de gheață; IT: una
lastra di giacchio , a splinter of ice  RO: o așchie/ țandără de gheață ; IT:
una scheggia di ghiaccio . The same holds valid for collocations including the
compounds iceberg and ice-cream : the tip of the iceberg  RO: vârful
aisbergulu ; IT: la punta dell'iceberg, a scoop of ice -cream  RO: o cupă de
înghețată ; IT: una pallina di gelato , a tube of ice -cream  RO: o cutie de
înghețată ; IT: una vaschetta di gelato .
Reference should be made that t here are quite numerous situations when
lexical collocations are also culturally marked. When confronted with the
transfer of such fixed patterns from one language into another, the translator’s
lexical /grammatical and collocational competences have to be doubled by
solid cultural competence in the two LCs brought into contact. For example, a
collocation such as blue book , which means ‘a register or directory of socially
prominent people’, ‘a blank booklet used in taking college examinations,
usually with a blue paper cover’ in AmE , and ‘a government or parliamentary
report bound in a blue cover’ in Br E, will be appropriately translated into
Romanian or Italian only if the translator is aware of its cultural restrictions in
the SLC. Variants such as * cartea albastră and * il libro blu , although lexically
and collocationally valid, will have no semantic relevance, and will most
probably confuse the TL audience.
As suggested by Hatim and Munday (2006: 10), “[ E]ven if certain
concepts might be absent from a given culture , this should not stop them being
translated in another language culture ”. Such a view is based on the fact that
the sense may most often be translated, while the form often cannot .

Nevertheless, practice has proved that the form can also contribut e to sense . In
such cases , messages are likely to be untranslatab le.
Furthermore, collocations may be differentiated in terms of the degree of
cohesion existing between the constitutive words. If this criterion is taken into
consideration, reference can b e made to: fixed collocations, in the case of
which the degree of cohesion between the constitutive elements is maximal,
semi -mobile collocations, which exhibit a more reduced formal cohesion due
to their allowing for substitution of one of the component w ords with a
synonym , and, finally, mobile collocations which are characterised by a
reduced degree of cohesion , because two, or even more eleme nts, can be
replaced by other synonymous words. (Avădanei 2000: 48) (see 2.1.3 .) Given
the specific characteristics, each of these three types of collocations requires a
different approach on the part of the translator.

Fig. 21 . Fixed, semi -mobile and mobile collocations in translation

Fixed/restricted/ strong collocations such as to put the cart before a
horse (a pune carul inaintea boilor , mettere il carro davanti ai buoi ) will be
correctly transferred in another language if the translator appropr iately uses his
semantic and cultural competences in the languages under focus; semi -mobile/
semi -restricted/ medium -strength collocations, such as to crack/ break a

bottle (a sparge o sticl ă, rompere una bottiglia ), will be conditioned by the
translator’s awareness of the sense relations of synonymy, antonymy,
hyponymy , whereas in the case of mobile/unrestricted collocations , such as to
swim /go against the stream / tide (a înota/ merge împotr iva curentului ,
nuotare / andare contro il corrente ), translators will have to be “aware of how
lexical boundaries separate words with related meanings and of how these
boundaries differ across languages” (Ulrych 1992: 279).
In addition, “when dealing with ambiguous words or synonyms the
translator needs to disambiguate them using the co-text and context of situation
between the various possible senses of the ST terms in order to be able to
identify the appropriate TL equivalent (Hatim and Munday 2006: 35).
Different from collocations which sometimes allow lexical replacements
on the basis of sense relations such as synonymy , antonymy or hyponymy (see
3.3.3 .1 – 3.3.3.3 .), other fixed lexical patterns such as similes , clichés ,
proverbs allow little or no vari ation in form , being also syntactically and
semantically ‘frozen’. However, a certain vari ation in form may be possible
when speakers or writers wish to create a special effect.
If contrasted with idioms , such fixed expressions have fairly transparent
meanings. Still, there is a common feature of both fixed expressions and
idioms i.e. the fact that they are “essentially non -creative, stereotyped formulaic
expressions” (Carter 1998: 66) located at the opposite end of the fixedness scale
from collocations .
As regards the translation of fixed expressions, some of them may be
simply transferred from one language into another due to the perfect formal
semantic and cultural equivalence, whereas others have to be translated in
keeping with a series of semantic and cultural constraints. For example, a
collocation such as: the black sheep of the family can be easily translated into
Romanian ( oaia neagră a familiei ) and Italian ( la pecora nera della famiglia )
due to the formal and semantic and cultural equivalence exi sting between the
three languages. This example may be contrasted with lexical patterns such as
to bet on the wrong horse and to break one’s back . They are collocations when
used with their denotative meanings ‘to bet on a horse which is not a winner’,
and ‘to fracture one’s back’, and idiomatic collocations when used with the
meanings ‘misjudge a coming event, misread the future; not choose the winning
person or solution’ and ‘to work very hard’.
With other English collocational patterns such as: to lead a dog’s life ‘to
live an unhappy life, full of problems, and be treated unkindly’ , to take the bull
by the horns ‘to take decisive action, to face a difficulty or danger directly, with
courage, to throw someone to the wolves ‘to send someone into danger w ithout
protection, without trying to help’ , it is only the idiomatic interpretation that is
possible. As regards their translation into Romanian and Italian, these patterns
have perfect equivalents in the two languages: to lead a dog’s life  RO: a

duce o viață de câine ; IT: vivere una vita da cane , to take the bull by the horns
 RO: a lua taurul de coarne ; IT: prendere il toro per le corna , to throw
someone to the wolves  RO: a trimite pe cineva în gura lupului ; IT: gettare
qualcuno ai lupi . Sometimes th e English patterns have perfect equivalents in
Italian and only partial equivalents in Romanian. A case in point is the pattern
to cast pearls before swine ‘to give, offer something valuable to someone who
won’t be grateful, won’t appreciate it’ which is translated by a formally and
semantically equivalent structure into Italian dar da mangiare le perle ai porci
or lanciare le perle prima dei maiali and only by a partially equivalent pattern
in Romanian a strica orzul pe gâște.
An interesting fixed lexical pattern in Italian worth mentioning here is in
bocca al lupo which , although seemingly equivalent with the English to throw
someone to the wolves and the Romani an a arunca pe cineva în gura lupului is
used by native speakers to wish somebody good luck, especially when that
person is supposed to face a difficult or dangerous situation. Such a fix lexical
pattern can be appropriately translated into English (break a leg ) and Romanian
(baftă), respectively , only if the translator is aware of the c ollocational and
cultural constraints specific to the 3 LCs brought into contact .
If we were to draw a conclusion, mention shoul d be made that lexical
patterns cannot and must not be translated item by item. When c onfronted wi th
such patterns the translator s’ task lies in relaying the meaning of the expression
as a whole. The same idea is pointed out by Baker (1992: 64) and Carter (1998:
66) who consider that, one of the roles of fixed expressions in communication i s
that of carrying out a maintaining or stabilizing function , as they conjure up “in
the mind of the reader or hearer all the aspects of experience which are
associated with the typical contexts in which the expression is used” (Baker,
1992: 64). Thus, whe n translating a fixed expression into a TL, the translator
must translate the lexical pattern as a whole, on the one hand, and try to evoke
in the TL reader’s mind the vital experiences suggested to the SL reader through
the ST expression by referring to s imilar or equivalent experiences in the TL
culture, on the other. To put it differently, s/he has to satisfactorily render all the
different kinds of meanings involved in the u se of the respective expression.
The difficulties which may arise in translating collocations are as
numerous, as various. I agree with Pârlog’s (2009 : 144) opinion that these
difficulties may be
“determined by semantic, grammatical or, sometimes, cultural differences in
contact. The translation problems vary from very simple ones, c aused by word
order typical to each language, to more complicated ones, caused by lexical gaps
in the target language or by the absence in the target language equivalent of some
semantic feature present in one of the terms of the English collocation”.

5.5.2. IDIOMS IN TRANSLATION

In contrast with collocations , idioms are at the extreme end of the scale
measuring their flexibility of patterning and transparency of meaning.
According to Baker (1992: 63), idioms are “frozen patterns of language which
allow little or no variation in form and […] often carry meanings which cannot
be deduced from their individual components.” However, it should not be
ignored that, in spite of the obvious semantic and syntactic limitations, idioms
have a significant contributio n to discourse through “their paradigmatic
properties, which are of importance in relation to interaction” (Moon 1994:
117).
As far as the translation of idioms is concerned, in Baker’s opinion
(1992: 68 -71), the difficulties involved in this process are m ainly due to the fact
that such fixed expressions often lack a perfect equivalent in the TL. This is
illustrated by animal idioms such as: cat nap ‘a short sleep taken during the
day’, cry wolf ‘give a false alarm, warn a danger that is not there’, donkey’ s ears
‘a very long time’, dumb bunny ‘stupid or gullible person’, eager beaver ‘a
person who is always eager to do something extra’, holy cow ‘used to express
astonishment, pleasure or anger, kangaroo court ‘a self -appointed group that
decides what to do to someone who is supposed to have done something
wrong’, monkey business ‘mischief, bad, or dishonest activity’, piggy bank ‘a
small bank/object for saving money’, rat race ‘confusing way of living, lacking
a purpose’, road hog ‘car driver who takes mor e than his share of the road’,
sacred cow ‘a person or thing which is never criticized or changed even if it
should be’, wildcat strike ‘a strike spontaneously arranged by a group of
workers’, wolf in sheep’s clothing ‘a person who pretends to be good but really
is bad’.
Moreover, idioms cannot be transferred from one LC into another by a
mere word -for-word translation. This translation strategy would be
inappropriate , because idioms are culturally marked patterns whose meaning
cannot be deduced from the m eanings of their constituent elements. For
example, a word -for-word translation in the case of idioms such as green
thumb ‘a talent for gardening, ability to make things grow’, red herring ‘an
important matter that draws attention away from the main subje ct’, red-letter
day ‘a day that is memorable because of some important event’, red tape
‘excessive formalities in official business’, would not only sound awkward, but
it would also prevent the appropriate semantic interpretation of the pattern and
of the context in which it is integrated.
When confronted with idioms , translators may cho ose to use an idiom
with a similar meaning and form or an idiom similar in meaning but dissimilar
in form, they may paraphrase the expressions reduced to sense or choose s imply

to omit the idiom if it has no TL equivalent and paraphrase is either too difficult
or results in a clumsy style (Baker 1992: 72).
Quite often idioms are ‘misleading’ to translators because they offer a
reasonable literal interpretation and their idi omatic meanings are not necessarily
signal led in the surrounding text. A large number of lexical structures in English
have both a literal and an idiomatic meaning. For example, besides the
denotative meaning, ‘to go out with smb. ’ means ‘to have a romanti c or sexual
relationship with someone’, and to take someone for a ride means ‘to deceive or
cheat someone in some way’. In some other cases, the literal meaning is not
acceptable. For example, to walk in somebody’s shoes does not mean to use
somebody’s sho es but ‘to pretend to be somebody else just as to throw at each
other’s throats refers to fighting or quarreling angrily, not to staying close to
somebody .
That is why, an important stage in the complex process of translating
idioms is represented by the ir appropriate interpretation , which is conditioned
by a series of linguistic and extralinguistic factors :

Fig. 22. Factors that influence idiom interpretation (source M arija Omazi ć
2009: 72)

The situations in which idioms in the SL m ay have a c ounterpart which
looks similar in the TL are also frequent. In such cases, translators have to be
very careful, so as to avoid mistranslation. For example, the idiomatic question
Has the cat had/got your tongue? is used in English to urge someone to answer
a question or contribute to a conversation, particularly when their failure to do
so becomes annoying. A similar expression is used in Romanian (Ți-ai inghițit
limba ) and Italian (Ti sei mangiato la lingua ?) but, in spite of being

semantically equivalent, the Romanian and Italian corresponding idiom atic
pattern s are collocationally and syntactically different from the SL pattern.
These aspects have to be observed by the translator so that, instead of
translating the English idiom atic question by *Ți-a mâncat pisica limba? and
*Ti ha mangiato il gatto la lingua? , to use the appropriate equivalent s Ți-ai
inghițit limba? and Ti sei mangiato la lingua? , which imply replacing the active
voice in Eng lish with the reflexive voice and deleting the active subject cat in
both Romanian and Italian.
Instead of a conclusion , mention should be made that the appropriate
translation of idioms from one LC into another presupposes a balanced
blending of the translator’s knowledge of the language with his/ her knowledge
of the worls and with a series of specific cognitive mechanisms which allow for
the correct decoding of the SL idioms and their appropriate encoding in the
TL(s).

5.5.3. METAPHORS IN TRANSLATION

Similarly to idioms , metaphors are lexical patterns in which formal fixity
combines with semantic creativity .
Metaphors may be classified according to various criteria (see 1.4.10
and 4.4.4), but the most wide ly mentioned classification is that suggested by
Newmark (1988). He distinguishes between a) dead metaphor s, whose images
are highly unmarked (e.g. at the mouth of the river ), b) cliché metaphor s,
which refer to the use of cliché expressions in texts (e.g. long time, no see ), c)
stock or standard metaphor s, d) adapted metaphor s, where the ‘fixedness’
of a stock metaphor has been adapted or personalized in some way, e) recent
metaphor s, where an anonymous metaphorical neologism has become
generally used in t he SL, and f) original metaphor s, which are created by the
writer or speaker , usually to make discourse more interesting and often used to
highlight particular points or as reiteration.
As regards metaphor translation , Newmark (1988) considers that
transla tors should apply different strategies , depending on the type of metaphor
they are confronted with. Although dead metaphors are not especially
problematical, Nida suggests that translators should avoid literal or word -for-
word translation. Cliché metaphors on the other hand, will require different
approaches, depending on the type of text in which they are used. If
encountered in vocative texts , cliché metaphors should be upheld in the TT,
whereas those found in informative texts, should be reduced to sense or
replaced with a more credible stock metaphor.
In approaching stock metaphor s, translators should reproduce the SL
image in the TL, and if metonyms are used they may be transferred on condition

the substitutes have the same connotations as those in the SL. Last, but not least,
mention should be made that stock metaphors may sometimes be reduced to
meaning or to literal language.
A similar strategy will be used with adapted metaphors which, in
Nida’s opinion, have to be translated by using equivalent ad apted metaphors or
by being reduced to sense. Recent metaphors should be translated using
componential analysis and original metaphors used in vocative texts should be
translated literally, as they “contain the core of an important writer’s
message…” (Nida 1988: 112). If the metaphor is obscure and of little
importance to the text, it should be replaced with a descriptive metaphor or
reduced to sense. In informative texts, consideration should be given to the
number and variety of original metaphors in the t ext as a whole and a decision
be taken between literal t ranslation, reduction to sense and modification of the
metaphor.
The complexity of word combinations, in general, and of metaphors , in
particular, emphasizes the fact that when confronted with the tra nslation of such
fixed lexical patterns, the translators’ task to appropriately transfer such
structures from a given SLC into a TLC is by no means “a piece of cake”.
Whether translating collocations , idioms , metaphors, similes , clichés or
proverbs , trans lators have to cope with numerous formal, lexico -semantic and
cultural constraints. I f in the case of collocations formal and semantic aspects
are given priority in translation, the transfer of metaphors from one language
into another is mostly, but not st rictly, conditioned by semantic and stylistic
constraints. The cultural specificity and the formal restrictions characterize all
these lexical patterns and are aspects which favour their misuse by unaware
speakers and their mistranslation by translators wh o are not well -trained.
As regards the translator , the competences required for the appropriate
translation of fixed lexical patterns will vary depending on the formal, semantic
and cultural specificity of the pattern(s) to be transferred from the SLC int o the
TLC.
However, since
“patterns of collocation are largely arbitrary and independent of meaning bo th
within and across languages, the same degree of mismatch can be observed when
comparing the collocational patterns of synonyms and near -synonyms with in the
same language with the collocational patterning of ‘dictionary equivalents/near
equivalents’ in two languages” (Baker 1992: 48) .
If translators choose inappropriate lexical combinations, thus “creating”
‘new’ patterns under the influence of the SLC , the respective patterns will be
interpreted as unusual in their mother tongue and will be rejected by their end –
users.

5.5.4. SIMILES, CLICHÉS AND PROVERBS IN TRANSLATION

Placed somewhere in between collocations and metaphors, s imiles, are another
type of lexical patterns in which formal and semantic specificity combine in
various ways .
As far as their translation is concerned, some consider that similes may
be easily translated from one LC into another because they are direct
comparisons with an ex plicit marker and they refer to stereotyped aspects of
experience. In addition, the apparent translatability of similes across languages
seems to be supported by the fact that these lexical patterns rely on prototypes .
Nevertheless , things are not always as simple as they might seem.
Certain similes have one meaning, whether literal, metaphorical or ironical (e.g.
as clear as the bell – ‘very clear, easy to hear’ , as clear as the day – ‘very clear
easy to see’ , as clear as the mud – ‘very confusing and un clear ’) (see also
Appendix 7.3. ). Moreover, other similes allow for both literal and figurative
interpretations (e.g. as sharp as a needle [literal ] – ‘having a cutting edge’ and
[figurative ] ‘very smart’). Last, but not least, certain similes can be easil y
misinterpreted, and also mistranslated, if the speaker relies on a seemingly
reasonable literal interpretation (e.g. as crazy as a fox – does not refer to the
animal’s mental illness, but to its cleverness ).
Aspects such as the ones mentioned above may represent a trap for the
culturally and collocationally unaware translators , and for the unaware non –
native speakers of Eng lish, but more importantly, their surface structure may
often affect their appropriate semantic interpretation if the context in whic h
they are used is ignored.
Clichés are also worth mentioning in our approach to the translation of
fixed lexical patterns because there are quite numerous instances when such
idiomatic patterns are formally fixed and culturally determined . For instance ,
a cliché such as love makes the world go round has a Romanian equivalent
which features a different connotative image: dragostea urnește munții din loc .
Although the effect and the message are the same, i.e. ‘doing somet hing that
makes the world exist / doing something that is humanly impossible’, the
translation of such a cliché is a re -creation of the SL model in the TL reality.
However, there are also clichés and expressions which refer to realities
that are commonly shared by the SL and TL cultures. For example , clichés such
as love is blind, marriage is a lottery, men are wolves in sheep’s clothing have a
perfect equivalent in Ro manian: dragostea e oarb ă, căsătoria e o loterie ,
bărbații sunt lupi cu blană de miel . Consequently, their translation will
presuppose only a transfer from LC1 into LC2.
Such instances of universally shared clichés show, from a
psycholinguistic point of view, that human life in universe is shaped by the
existence of common models, beliefs, or sceneries. In other words, the perfect

equivalence between different cultural models is due either to the fact that they
refer to universally shared aspects of live, or to the fact that the respective
reality is identical in the LCs brought into contact. This is in fact what makes
their translation into other language cultures possible .
As regards proverbs , they are also culturally -marked fixed lexical
patterns since they are an illustration of each culture’s collective wisdom.
Proverbs may be defined as short standard statements, having an evident or
implied general meaning, related to a certain typical field of general human
conditions, attitudes or actions, which is valid with implied limit ations. The fact
should also be pointed out that proverbs have no known author or literary
source being, as Hulban (2002: 37) calls them, “common property”. They are
often quoted in a certain language community, sometimes in a short form, the
rest being im plied.
Referring to the origins of proverbs, Hulban (2002: 37) states that one
important source of proverbs is represented by “ aphorisms, maxims, etc. which
have gained a wide circulation and a fixed form ”. Another important source is
the “ folk-lore” whic h explains why many proverbs are met in different
languages with little modifications, while others are entirely different. In the
case of the former , the appropriate translation will depend on the translator’s
knowledge about the two cultural models broug ht into contact, whereas in the
case of the latter the completely distinct systems of reference will put the
translator in the position of striving to appropriately render the intention of the
original.
Most proverbs are culturally determined . Consequently , in such
instances their translation presupposes the recreation of the SL model in the TL
reality. As far as the equivalence relationship between such proverbs is
concerned, the fact should be mentioned that it is acquired only partially . For
instance, pr overbs such as: A feather in hand is better than a bird in the air 
RO: Nu da vrabia din mână pe cioara de pe gard ; IT: Meglio un uovo oggi che
una gallina domani; Diamond cut diamond  RO: Cui pe cui se scoate , Once
bitten twice shy  RO: Cine s -a fript cu ciorbă suflă și -n iaurt , Stolen sweets
are sweetest  RO: Merele din grădina vecinului sunt întotdeauna mai dulci
IT: L’erba del vicino è sempre più verde are stylistically and pragmatically
equivalent, but lack equivalence at the lexical, grammatical a nd semantic levels.
(for more examples see Appendix 7.5.)
However, there are also proverbs , which refer to realities that are
commonly shared by the SL and TL cultures and which use totally equivalent
structures in order to render them. For example, the Romanian and Italian
variants of such proverbs as: When the cat’s away , the mice will play  Când
pisica nu -i acasă, șoarecii joacă pe masă  Quando il gatto non c’è i topi
ballano , and He laughs best who laughs last  Cine râde la urmă râde mai
bine  Ride bene chi ride l’ultimo are fully equivalent because they have a

perfect lexical, grammatical, semantic, pr agmatic and stylistic correspondent in
both these TLCs .
What is interesting to notice is that in some cases a proverb from a given
SLC may have two accepted variants in the TLC, one of them being non –
equivalent or partially equivalent and the other being fully equivalent to the
original. For instance, the Italian proverb L’appetito vien mangiando, with its
perfect equivalent in Romanian : Pofta vine mâncând , has two possible variants
in English: Appetite comes with eating/ Eating and scratching wants but a
beginning , the former being a full equivalent and the latter being only a partial
equivalent of the SL proverb. Other such examples worth mentioning here are:
All is not gold that glitters/ All are not hunters that blow the horn  RO: Nu tot
ce zboară se mănâncă ; IT: Non è oro tutto quel che brilla ; The devil is not so
black/ the lion is not so fierce as he is painted  RO: Dracul nu este chiar așa
de negru cum se spune; IT: Il diavolo non è così brutto come si dipinge , A good
beginning is half the battle/ T he first blow is half the battle  RO: Ziua bună se
cunoaște de dimineață ; IT: Il buon giorno si conosce dal mattino .
To conclude , proverbs in a given LC have mostly partial equivalents in
other LCs, but there are also quite numerous instances of total e quivalence, on
the one hand, and non -equivalence between the LCs brought into contact , on the
other .
As far as the translator is concerned the fact should be pointed out that in
order to be able to appropriately transfer from ane language into another such
culturally -determined elements as proverbs , (s)he should ideally be bilingual
and, most important ly, bicultural .

5.6. TRANSLATING COLLOCATIONS IN DIFFERENT TEXT TYPES

Translators can more realistically deal with problems of diversities in types of
discourse (literary genres), striking differences in audiences, and distinctions in
the ways in which translations are to be used, if they allow a range of diversity
in translating. In other words, translators should try their hand on various types
of texts s o as to identify possible problems in translation, on the one hand, and
to be able to adapt their translation strategies to the constraints imposed by each
text type of those approached, on the other.
As regards text typology , mention shoul d be made that s pecialists have
suggested various possible classifications of texts, depending either on their
function , or on the type of discourse such texts may be attached to.

5.6.1. TEXT TYPOLOGIES41 AND TEXT FUNCTIONS

It is generally agreed that collocations , whet her lexical or grammatical, are
fixed lexical patterns which are known to speakers due to their frequent use in
language. The large number and great diversity of collocations allows grouping
them according to different criteria, among which the function of the texts in
which they are used and the type of discourse they are representative for are of
interest for the purpose of this section .
If the function of a text is taken into account, distinction is generally
made between informative , expressive and oper ative texts.

TEXT TYPE INFORMATIVE EXPRESSIVE OPERATIVE
LANGUAGE
FUNCTION informative
(representing
objects and facts) expressive
(expressing
sender’s attitude) appellative
(making an appeal
to the text
receiver)
LANGUAGE
DIMENSION logical aesthetic dialogic
TEXT FOCUS content focussed form focused appellative
focussed
TT SHOULD transmit referential
content transmit aesthetic
form elicit desired
response
TRANSLATION
METHOD ‘plain prose’
explicitation as
required ‘identifying’
method, adopt
perspectiv e of the
ST author ‘adaptive’
equivalent effect

Table 14. Functional characteristics of text types and links to translation
methods (source Munday 2001: 74 , adapted from Reiss 1971)

Informative texts used to transmit information. They are logical or
referential, and focussed on the content of the communication. In the case of

41 Central to text typologies is the view that language use beyond the sentence may
helpfully be seen in terms of rhetorical purpose (e.g. exposition , argumentation ,
instruction ). Such a criterion allows for a further identification of texts forms in terms
of subject matter (technical, nontechnical) or level of fromality (formal, informal).
Nevertheless, “such a categorization i s ‘idealized’ since all texts are in a sense hybrid”
(Hatim and Munday 2006: 74). Thus, the mod els of translation adapted to text typology
have to encompass and account for the diversity of rhetorical purposes of
communication. From this perspective, text types “are ‘guidelines’ which text users
instinctively refer to in adopting a given translation strategy with an eye on both sides
of the translation divide – the ST and the TT” (Hatim and Munday 2006: 74).

expressive texts , the author and the form of the message are foregrounded and
language is used in its aesthetic dimension. Finally, if the form of language is
dialogic and the fo cus is appellative, than the text is operative .
As far as the translation of such texts is concerned, the translation
methods are correlated with the text type and the predominant function that each
text has in translation. Thus, as suggested by Hatim and Munday (2006: 181), in
the case of informative texts the translator has to focus on the semantic
relationship within the text, and only secondarily on connotative meanings and
aesthetic values. Moreover, when approaching expressive texts , the translator
should try and preserve the aesthetic effect alongside relevant aspects of the
semantic content. Finally, in the case of operative texts the translator has to
identify and observe the extralingusitic effect of the text and then to achieve
this effect in tra nslation, even if this implies altering the form and content of
the original text.
If attention is shifted from the classification of texts according to the
language functions to the text types identified in terms of the type of discourse
they are repres entative for, distinction can be made between general , literary
and specialized texts.
This rather general typology subsumes the classification of texts in the
frame of the functional styles of language and of their corresponding substyles
(see 4.2.), thus allowing reference to the features of texts written according to a
specific style and to the words and lexical patterns likely to be encountered in
such texts.
From this perspective, general texts are illustrative for the use of neutral
and colloquial w ords and phrases, literary texts are resourceful as regards the
expressive use of epithets , similes , metaphors , clichés , and proverbs , and
specialized texts testify for the domain -specific restrictions conditioning the
appropriate use of isolated terms , on the one hand, and of lexical patterns
including terms , on the other.
Last, but not least, each of these three types of texts42 has specific
characteristics and implies a different approach in translation.

42 “As a unit of communication and translation, text is a vehicle for the expression of
conventionalized goals and functions. These are tied, not to communicative events, as
in genre, but rather to a set of specific rhetorical modes such as arguing and narrating.
Rhetorical purposes impose their own constraints on how a sequence of sentences
becomes a text, i.e. intended and accepted as a coherent and cohesive whole, and as
such capable of realizing a set of mututally relevamt communicative intentions
appropriate to a given rhetorical purpose” (Hatim and Muday 2006: 89).

5.6.2. TRANSLATED TEXTS TYPOLOGY: GENERAL, LIT ERARY,
AND SPECIALIZED

General translations are usually contrasted with specialized translations, the
criterion being mainly the type of vocabulary used. If comparison is drawn
between general and literary translations, the criterion is the use of styli stic
devices in order to create a certain effect.
As regards the former distinction, general translations imply the transfer
from one language to another ; they do not require any specialized vocabulary or
knowledge . Specialized translations are mostly cond itioned by the translator’s
knowledge of domain -specific terminology.

A. COLLOCATIONS USED IN GENERAL TRANSLATIONS
The translator must observe the formal, lexical, semantic and cultural
restrictions of the SL collocational patterns and appropriately transfe r them to
the TL

B. COLLOCATIONS USED IN LITERARY TRANSLATIONS
The translator must detect the stylistic effects of collocations
in the SLT and appropriately transfer the SL patterns to the TLT

C. COLLOC ATIONS USED IN SPECIALIZED TRANSLATIONS
The translator must detect domain specific collocations in the SLT
and appropriately transfer them to the TLT, or neutralize them

Whether confronted with general, literary or specialized texts, translators
have to possess specific knowledge and skills. In addition, goo d translators have
to be aware that the appropriate translation of a text largely depends on their
familiarity with the topic.
“Literary translations are translations of various types of literary texts ,
such as poetry, drama, novels, memoires, etc. All t he other texts, whether
general or specialized , are examples of non -literary, or pragmatic translations”
(Ionescu 2000: 37). As suggested by Ionescu, the essential difference between
these two types of translations, i.e. literary and non-literary (general or
specialized), is that “the latter translate what is in the text, whereas the former
must translate what the text implies” (id.: 38). This means that, when confronted
with literary texts, translators must render the aesthetic effect of the literary
piece , whereas in the case of semantic or non -literary translations, priority must
be given to the appropriate transfer of the semantic content from the SL to the
TL.
Moreover, literary translations require that a double decoding should be
performed at differe nt levels. Literary translation must also “convey those

features of the original text that are the traces left by its production. All these
signs are in fact forms of literariness”, and they “are implicitly conducive to the
style used in the original text and rendered appropriately in the target text”
(Ionescu 2000: 38). Although some aspects of literary texts, such as the story,
characters, descriptions of places, etc., “usually carry over fairly easily from a
source text to a target text, yet even these transferable realities will elicit
somewhat different reactions in the target text reader: at some psychological or
aesthetic level, any reader of a translation will react differently from a reader of
the original” (Hickley 1998: 226).
Concerning the diffe rence between literary and other types of texts, some
suggests that is one of degree, not of kind. This view is supported by the fact
that “even special languages are characterized by metaphor, and journalistic
language abounds in ‘literary’ devices such a s alliterations, word -play, similes
or metaphors” (Ionescu 2000: 223). Thus, literary language “involves not
merely deviance from a static and prescriptive norm – but the creative extension
of the language norm, in the flexible sense of the rule -governed p otential. As
regards translation, one of the literary translator’s most difficult choices is
deciding how such creative extensions of the SL norm can be rendered in the
TL without actually infringing the rules of linguistic acceptability” (Snell –
Hornby 198 8: 52). Like literary translations, specialized ones are characterized
by numerous constraints. In the larger class of specialized translations43, a
distinction is usually made between financial , medical, scientific, technical and
legal translations.
As re gards the strategies used with specialized translations, translators
should adopt an essentially functionalist approach. Seen from this perspective,
the main criterion considered in selecting translation strategies and operations is
the function of the TT in the TLC, corresponding to Reiss and Vermeer’s skopos
theory (Reiss and Vermeer 1984). To put it differently, specialized translations
will be considered appropriate if the respective texts “function” adequately in
the target specific situational and cul tural contexts.

5.6.3. TRANSLATING GENERAL COLLOCATIONS

Collocations are agreed to be patterns frequently used in different types of texts
and different types of discourse. However, if reference is made to c ollocations
specific to general usage (gener al collocations ), such patterns are very

43 This distinction is a contents -based text typology, whereas the functional type is that
of informative (and sometimes also vocative – e.g. when a researcher is writing an
article also to impress potential funding bodies) texts .

accessible because they may be easily semantically interpreted on the basis of
the “core” meanings of the constituent words.
The do and make collocations represent a good illiustration in this respect
as most patte rns including these verbs are used in general texts and are made
up of neutral and colloquial words and phrases easily decodable by both native
and non -native speakers of English .
This explains why numerous short lexical collocations of the type
make /do + object may be translated into Romanian and Italian without any
problems : to do the homework  RO: a face tema/temele ; IT: fare il compito/ i
compiti; to make an agreement  RO: a face o înțelegere ; IT: fare un accordo;
to make a cake  RO: a face o praji tură; IT: fare un dolce; to make a
collection  RO: a face o colecție ; IT: fare una collezzione; to make a
compromise  RO: a face un compromise ; IT: fare un compromesso; to make
a description  RO: a face o descriere ; IT: fare una descrizione; to make a
difference  RO: a face (o) diferență ; IT: fare una differenza; to make a
discovery  a face o descoperire ; IT: fare una scoperta; to make efforts  RO:
a face eforturi ; IT: fare sforzi; to make a film/ to do a movie  RO: a face un
film; IT: fare un film; t o make a funeral  RO: a face o înmormâ ntare ; IT:
fare un funerale; to make a gesture  RO: a face un gest ; IT: fare un gesto; to
make a joke  RO: a face o glum ă; IT: fare uno scherzo; to make money
RO: a face bani ; IT: fare denaro/ quattrini; to make a movement  RO: a face
o mișcare ; IT: fare un movimento; to make a nest  RO: a face un cuib ; IT:
fare un nido; to make an offer RO: a face o ofertă ; IT: fare un’ offerta; to
make a pact  RO: a face un pact ; IT: fare un patto; to make place RO: a
face loc/spațiu ; IT: fare posto; to make a portfolio  RO: a face un portofoliu ;
IT: fare un portofoglio; to make a presentation  RO: a face o prezentare ; IT:
fare una presentazione; to make a promise RO: a face o promisiune ; IT: fare
una promessa; to make a pro posal  RO: a face o propunere ; IT: fare una
proposta; to make a rehearsal RO: a face o rep etiție ; IT: fare una
ripetizione; to make tea  RO: a face ceai; IT: fare del tè.
Nevertheless, certain do/make collocations are more difficult to be
rendered into Romanian and Italian because, in spite of being semantically
equivalent, there is form al non-equivalen ce between the patterns in these 3
languages . For example, c ollocations such as: to do the homework, to make
coffee, to make a fortune, to make love are translated into Romanian and Italian
either by preserving the article (definite, indefinite, zero), or by omitting it: to
make () advances  RO: a face() avansuri ; IT: fare delle avance ; to make
() coffee  RO: a face () cafea ; IT: fare il caffè ; to make a fortune  RO:
a face () avere ; IT: fare () fortuna ; to make () love  RO: a face ()
dragoste ; IT: fare l’amore .
Moreover, there are instances when the same English collocation has two
equivalents in Romanian and Italian. For example, to make good/bad may be

translated into Romanian and Italian either by using no article a face () bine/
rău/ fare () bene/male (total equivalence ) or by introducing the indefinite
article in Romanian a face un bine/rău (used with reference to a certain action or
deed) and the so -called preposizione articolata del (the preposition da + the
definite artic le il) in Italian: fare del bene/male .
As regards other cases of partial equivalence, certain English collocations
are translated into Romanian and Italian by means of lexical patterns which have
the same semantic content and pragmatic function, but which are not equivalent
at the lexical or grammatical level. For example, the translation of the English
collocations to do the house, to do the laundry, to do a speed, to make a
decision, to make haste, to make an inquiry, to make a pronouncement, to make
a reply/response, to make sure, to make a wedding dress into Romanian and
Italian, respectively, presupposes either the use of a different verb with the same
semantic content, or of a different voice (refelexive) : to do the house  RO: a
amenaja casa; IT: arredare la casa ; to do the laundry  RO: a spăla lenjeria;
IT: lavare/ fare il bucato ; to do a speed  RO: a atinge o anumită viteză ; IT:
arrivare ad una certa velocità ; to make a decision  a lua o decizie ; IT:
prendere una decisione ; to make a wish  RO: a–și pune o dorință ; IT:
esprimere un desiderio ; to make a reply/ response  RO: a da un răspuns ; IT:
dare una risposta ; to make a wedding dress  RO: a face o rochie de mireasă ;
IT: fare un vestito da sposa ; to make way for somebody  RO: a lăsa cale
liberă cuiva ; IT: lasciare strada libera a qualcuno .
The lack of grammatical equivalence between the three languages under
discussion results, in many cases, from the differences in terms of voice between
English, Romanian and Italian. The active (A) verb forms in English often have a
reflexive (R) equivalent either both in Romanian and Italian, or only in one of
the two TLs: to do the job RO: a-și face treaba (R); IT: fare(farsi ) il lavoro
(R); to make an appointment  RO: a-(și) face o programare (R); IT: prendere
(un) appuntamento (A); to make friends  RO: a–și face prieteni (R); IT: fare
amici/ amicizia (A); to make haste  RO: a se grăbi (R); IT: afrettar si (R); to
make an inquiry  RO: a se informa (R); IT: informar si (R); to make sure 
RO: a se asigura (R); IT: assicurar si (R).
The translators’ phraseological and cultural competence is definitely put to
test by such Italian collocati onal patterns as: fare meraviglia, fare le meraviglie
or fare il desiderio di qualcuno which have nothing to do with the English to do
wonders and to make a wish , respectively. The meaning of the first pattern is ‘to
surprise ’, of the second ‘to show great surprise ’ and finally, the third example is
used with the meaning ‘to fulfil somebody’s wish ’.
Different from lexical collocations, certain grammatical collocations in
English are likely to lack a total equivalent in Romanian and Italian. This is due
to the formal and grammatical restrictions characterizing such patterns in each of

these three languages : to make a habit of (+ gerund ) sth.  a-și face (R) un
obicei din + long infinitiv e  fare/ farsi (R) un’abitudine da + infinitive .
Last, but not least , culturally -marked patterns and idiomatic combinations
such as : absence makes the heart grow fonder, all work and no play makes Jack
a dull boy , to make a clean breast of things, make a dog's breakfast of smth., to
make a mountain out of a molehill , to make a virtue of necessity, to make a
welcome change, to make head or tail of something, to make hay while the sun
shines, to make somebody’s blo od run cold, many hands make light work, one
can't make an omelette without breaking eggs, one cannot make a silk purse out
of a sow's ear, practice makes perfect, two wrongs do not make a right will be
appropriately transferred from English into Romanian and Italian only if
translators are culturally and collocationally competent in both in the SL and the
TL(s).
Although not very numerous , the examples presented are illustrative for
the commonly shared view that the appropriate translation of collocations is not
conditioned only by formal and grammatical factors. In many cases the correct
interpretation and transfer of collocations from one language into another
entirely depends on the translator’s ability to identify and appropriately deal with
their cult ural specificity. This is the prevailing aspect in the translation of fixed
lexical patterns, irrespective of the LCs brought into contact.

5.6.4. TRANSLATING LITERARY COLLOCATIONS

Literary collocations , as used in the present study, are considered to include a
wide variety of (fixed) lexical patterns commonly selected in literary texts (see
4.4.), their specificity differing to a great extent from prose to poetry and drama .
An approach to literary collocations would have to take into account the
fact that literary texts allow for more combinary freedom at the level of lexis,
their authors being allowed to exploit, as appropriate, the denotative and
connotative meanings of words and patterns. As far as the translation of
collocations used in literary texts is concerned, the subjective interpretation of
the original text may favour an equally subjective translation, thus a significant
collocational deviation from the SLT to the TL one.
Referring to the translation of literary texts , Croitoru (2005) sugge sts that
implies taking into account a series of criteria for the analysis of both the ST
and TT :
“among the most important coordinates of literary translations, intentionality
determines the translator’s choices, the author’s intention being as important as
the recipient that defines the communicative situati on and the function of the
text” (Croitoru 2005: 406).

It is generally agreed that the ideal to achieve in translating literary texts
consists in expressing the whole universe of ideas of the SLT. In a wider sense,
this ideal means the re -creation of the SLT in the TL by various means.
Different from general and specialized translations, which are mainly
directed towards transferring information from one language into another,
literary translations ma y be considered to have fulfilled their objective if the
TL reader has access, like the SL reader, to the author’s universe of ideas and
feelings.
When confronted with literary texts, the translator’s task is far from being
easy. He has to decode the auth or’s universe, to render the text message
observing its denotations and connotations in the two languages in contact, and
to re-create the SLT in such a way as to make the TL readers believe that they
have in front of their eyes not a translation, but the original text. In doing this,
translators have to identify, on the one hand, with the author’s universe in the
TL system and with the TL reader, on the other.
Besides the end -users’ response, a literary translation may also be
evaluated in terms of its fu nction ality in contrast with the ST. A critical analysis
of the original literary text and the translated version(s) is generally made in
terms of accuracy, adequacy and effect. Style, which is a “property of language
users in particular kinds of settings” (Hatim and Mason 1992: 10) is also
extremely important both in translating literary texts and in comparing the ST
and TT version(s).
Another important requirement in translation, whether considering
general, literary or specialized texts, is to transf er the semantic content of the
ST in the TT at all language levels.
If we consider the lexical level, special attention has to be devoted to
lexical units (words) and lexical patterns (collocations and colligations). As
regards the former category, mention should be made that the m eaning of a
word consists of: 1. the general meaning of the linguistic notion; 2. the
occasional meaning , i.e. the meaning in a certain situational context; 3. the
general representation of the notion as a result of all the oc casional uses or
occurrences ; 4. the secondary representations, i.e. subjective a ssociations; 5.
feeling – tones, i.e. connotations.
When co -occuring with other words, these meanings are blended in
different ways, the result being a great variety of l exical patterns ( collocations ,
fixed expressions , idioms , metaphors ), more or less semnatically transparent.
One of the major problems which a translator has to face is to
appropriately transfer collocations from the SL to the TL. In doing this , he has
to overcome the danger of SL interference, which would result in an uncommon
collocation in the TT.
Translating collocations in literary texts requires the translator’s
competence because they provide powerful evidence of intentionality and text –

type focus. They can point to the intended meaning, which is not made explicit
by other means.
There are TL collocations in which neither the primary , nor the
secondary dictionary meanings of the SL noun can be used, and the translator
has to find a synonym accepted by the TL.
Such rare and appa rently unacceptable lexical patterns are called marked
collocations . They strike the reader because they involve deliberate confusion
of collocational ranges in order to create new images. They may seem atypical
in the common language, but they often occur in literary prose, poetry, humour,
and advertisements.
The differences in the collocational patterning of the SL and TL create
potential pitfalls and can be real traps for literary translators who “sometimes
get quite engros sed in the source – text and may produce the oddest collocations
in the target – language for no justifiable reason” (Baker 1992: 54). Translating
collocations often involves a tension – a difficult choice between what is typical
and what is accurate. It i s the collocational patterning of a word that determines
its different meanings.
According to the principle that translation conveys more or less than the
original (Croitoru 1996: 40), the TL ’equivalent’ epithet may not be sufficient to
render the semanti c content and the stylistic effect of the SL word. For example,
the epithet fierce in the collocation fierce light was translated by necruțătoare
not by puternică/ orbitoare , that would mean less than fierce in the original.
Consequently, interpretation du ring the translating process
“concentrates on finding the appropriate expression to convey a given
meaning […] in a given context, and involves the identification of
relevant concepts and their rewording in another language so that the SL
and TL wordings may correspond in their temporary meaning in a given
speech performance” (Croitoru 1996: 39).
Considering the wide range of marked collocations in literary texts, the
conclusion may be drawn that the message of a literary text in the SL cannot
always be ma tched by an expression with exactly the same content and structure
in the TL. What must be expressed is a problem as difficult as that of how it can
be expressed. Following SL norms may involve insignificant changes in the
overall meaning. On the other han d, deviations from typical TL patterns may
result in a translation that will sound “ foreign”. It is obvious that literary
translator s have to be competent in handling the SL patterns in order to
correctly render the message in the TL and to produce a TT wh ich will read
naturally and smoothly .

5.6.5. TRANSLATING SPECIALIZED COLLOCATIONS

With specialized translations, an essentially functionalist approach must be
adopted. The main criterion considered in using translation strategies and
operations is the function of the TT in the TLC, called skopos (Reiss and
Vermeer 1984, Vermeer 1996). In other words, when confronted with
specialized translations , the translator’s task is to ensure that the translated text
should “function” adequately in the target spec ific situational and cultural
context.
As regards the translation of ‘ready -made language’ (Heltai 2004: 55),
this is “a very important aspect of L 1 ↔ L 2 specialized translations” (Croitoru
and Dumitrașcu 2006b: 108). Different types of specialized texts contain
different amounts of ready -made language and the use of ready -made language
may also differ from one section of a text to another.
The most frequent ready -made units used in specialized texts are
collocations and certain discourse organizing lexic al phrases. Their great
number in such texts, whether technical, scientific, legal or beaurocratic, call for
a high degree of routinization . This is due to the fact that specialized texts give
descriptions of procedures, phenomena, processes, by frequently using
prefabricated /routine phrases. (see Appendix 4. )
If routinization is excessive, texts are perceived as pretentious and jargon –
like, whereas if they lack routine units, they are considered not to match the
register. According to Howarth (1996: 131 i n Croitoru and Du mitrașcu 2006b:
109), “it is partly by means of collocations, restricted in a technical sense, that
one recognizes register ”. Time is also one of the factors accounting for the use
of ready -made units in technical texts; in translating such texts a high
processing speed is necessary (sometimes, sight translation or interpretation is
needed; therefore, the process is performed under conditions of stress).
Overstepping the limits of routinization may lead to jargon and
translationese, the term jargon referrin g in this case to the specialized
terminology of a science or occupation, or to a certain style, typical of technical
language which contains many ready -made units. To put it in Heltai’s words,
“the use of ready -made language, however, always involves the risk of socio –
linguistic error: if the translator oversteps the limits or falls short of the expected
(and acceptable) degree of routinization, the target text may be perceived by
readers as jargon or translationese” (Heltai 2004: 59).
Translationese is similar to jargon (see 4.5.2. ) in that it is difficult to
comprehend, due to direct and indirect SL interference. An effect of the indirect
interference may be a higher number of routinized expressions, nominalizations,
and impersonal constructions. Accor ding to Heltai (2004: 59),

“indirect pragmatic influence may manifest itself in what has been called quasi –
correctness, i.e. the omission of optional translation shifts, especially
explicitation, while direct interference is manifested first of all in col locations
and other ready -made units”.
On the ot her hand, the odd collocations ‘created’ by the translator under
the ST influence will be interpreted as unusual collocations in the TL due to the
translator not knowing the register specific ready -made unit s in the TL. A
ready -made unit in the SL may not have a corresponding unit in the TL, and if
the translator gives it a literal translation, the TT may get a foreign flavour, and
will be worse than non -translated texts.
If translators of specialized texts have good phraseological competence,
they will handle the ready -made language effectively. If not, the result will be
the end -users’ negative attitude towards their translated texts which will be
nothing else, but some poorly written “original” texts.
In order to avoid such situations, a phraseological competence needs to
be developed in trainee translators. They have to know the ready -made phrases
used in various registers in both languages. They also need to know how to
match them and how to evaluate them from the sociolinguistic point of view.
Thus, the translators will be able to handle ready -made language, and will know
when and which ready -made units they can use.
Many problems in translating (fixed) lexical patterns used in specialized
texts are due to the monosemantic or po lysemantic nature of the domain –
specific terms making up such patterns.
Analysing the monosemy or polysemy of certain terms in English and
Italian, Scarpa (2008: 189) suggests the following possible situations that a
specialized tr anslator can be confronted with when there is no perfect semantic
correspondence between the concept and the term available in English and
Italian (Scarpa’s examples):

 Monosemy in the SL and polysemy in the TL
In English there are more terms used to denot e different concepts which
can only be translated into Italian by a unique term. For example, the
terms torque and moment used in physics to denote rotations and lever
systems, respectively, are both translated into Italian by one term, i.e.
momento .
 Polysemy in the SL and monosemy in the TL
In Italian there are more terms used to denote different concepts which
can only be translated into English by a unique term . For example, the
adjective standalone may be translated into Italian by standalone ,
indipe ndente or autonomo if attached to software (e.g. standalone
program  programma indipendente ), and by isolato if attached to solar
photovoltaic (e.g. standalone PV system  sistema fotovoltaico isolato ).

Similarly, doping is preserved in Italian if used w ith reference to sports,
but it is translated by drogaggio in the photovoltaic technology.
 Polysemy in the SL and polysemy in the TL
Both in English and Italian, there is a network of signifiers for one and
the same term, which do not correspond in the t wo languages. For
example, the Italian term montante has at least three equivalents in
English, depending on the concept they refer to, i.e. stud (vertical beam in
constructions), column (mechanical mounting), strut (uprights in
aeronautics). Moreover, stud corresponds to borchia, chiodo, perno,
bulletta etc, whereas strut is translated by puntone (in constructions) and
contropalo (in carpentry).
 Synonymy
There are variants within the same discipline, but only in one of the two
languages. For example, the terminals in physics correspond to the Italian
terms capi (electric circuit), morsetti (of a battery), and terminali (i.e. the
extreme connecting points of a conductor).

Sometimes m ention is also made to the use of compensation in
translating idioms. This strategy implies either omitting or playing down a
feature such as idiomacity at the point where i t occurs in the ST and introducing
it elsewhere in the TT. As pointed out by Baker (1992: 78), compensation “is
not restricted to idiomacity or fixed express ions and may be used to make up for
any loss of meaning, emotional force, or stylistic effect which may not be
possible to reproduce directly at a given point in the TT”.
Register also plays an important role in specialized translation .
Referring to the continuous register “elevations” , Scarpa (2008: 195 -196)
notices that such “elevations”, which are operated by the translator in order to
adapt the TT to the degree of formality and abstractness of technical and
scientific Italian , are particularly visible at the lexical and terminological
levels: e.g. to sell vs. collocare sul mercato and to borrow vs. contrarre un
prestito (in economics); hotels vs. strutture ricettive di tipo alberghiero and
land reform vs. riforma agraria o fondiaria (human geography); river blindness
vs. oncocercòsi or oncocerchias and scan vs. TAC o tomografia assiale compu –
terizzata (medicine).
As a conclusion , certain common translation strategies may improve the
functionality and formality of the SLT by increasing the informative val ue of
the text (referential functions and metalinguistics) and by neutralizing those
extremely informal approaches to the SLT (the phatic and expressive functions).
Moreover, the English scientific texts designed for students and for the large
audience oft en use an emotive tone and figurative language to facilitate the
understanding of new concepts.

5.6.5.1. On the translation of legal collocations

Legal translation is the translation of texts within the field of law, i.e. of legal
texts. The language us ed in such texts reflects the speakers’ keen consci ousness
of precedent , aspect which affects every choice of word, every turn of phrase.
For this reason, legal language tends to be quite conservative , and formulaic
in the sense that it tends to retain st andard phrases and formulas that are not
frequently used in everyday situations. Legal language is also highly formal,
this being the expression of the formality characterizing the legal process itself .
Moreover, legal language is definite , precise , and technical . Since
everyday language lacks the necessary consistency and precisi on, a large
number of words are used in legal language in strict accordance with the
definitions understood by all those concerned. The people having a legal
profession make caref ul distinctions be tween words such as action (a lawsuit,
not a movement), consideration (support for a promise, not kindness), execute
(to sign, not to kill), and party (a principal in a lawsuit, not a social gathering),
which have different meanings in t he legal discourse and in the everyday usage.
Special attention is also devoted to words which may seem interchangeable to
ordinary people in certain contexts. The difference be tween residence and
domicile , dictum and decision , privilege and right , may be of little consequence
in everyday language, but these distinctions are definitely crucial in legal
contexts.
Mention should also be made that there are specialized words and phrases
uniquely or nearly uniquely used in legal texts such as: tort, fee simpl e, and
novation . Archaic terms such as: herein , hereto , hereby , heretofore , whereas ,
whereby , and wherefore ; said and such (as adjectives) are also specific to legal
langauge and legal texts. Last, but not least, legal language makes use of loan
words and phrases from other languages terms derived from French (such as
estoppel , laches , and voir dire ) and Latin ( certiorari , habeas corpus , prima
facie , inter alia , mens rea , sub judice ).
Numerous difficulties in approaching legal texts arise from the fact tha t
law is a culture -dependent subject field. Every legal system is the result of a
complex socio -political background which is greatly influenced by history and
by the customs and habits of its people. Despite the fact that two legal systems
may have the sa me origins, their structure is hardly, if ever, identical. The
diversity of legal systems makes the research in the field of legal terminology
extremely challenging. Not few are the instances when certain legal concepts in
a given legal system lack appropr iate equivalents in other legal systems or
formal equivalents in different languages may be used with different meanings,
depending on the legal discipline to which the term is attached. It is the case
with the noun contract which has perfect formal equiva lents in both in
Romanian and Italian (i.e. contract and contratto , respectively), but which may

have different semantic interpretations, in the three languages , depending on the
context in which it is used.
The complexity of the legal discourse , as illust rated above, fully justifies
the necessity that translators should approach this type of specialized text
carefully. Translating legal texts requires attention to detail and sensitivity to
the consequences of subtle contextual changes because legal languag e differs
from everyday language to a large extent. Understanding legal language helps
translators to develop a kind of textual model, a sense of how language
functions in the legal discourse, whereas unfamiliarity or little familiarity of the
translators with these aspects may easily result in an inappropriate transfer of
the SLT in the TL.
Legal translation is regarded by some specialists in the field as one of
the most “challenging endeavours, combining the inventiveness of literary
translation with the terminological precision of technical translation” (Harvey
2002 in Š arcevic 2000: 1 ).
The general assumption is that translation of legal discourse should be,
like other areas of translation, reader -oriented. This explains why many linguists
consider it u seful and appropriate that theories of general translation should
apply to legal texts , as well. Mention should be made that there has been much
debate on whether functionalist theories such as Catford’s concept of situation
equivalence , Nida’s theory of formal correspondence and Vermeer’s skopos
theory should be extended to legal texts or not. This is due to the fact that the
typically recipient -oriented nature of these theories seems unacceptable in the
case of legal discourse, which is “subject to rigoro us rules of interpretation”
(Garzone 2000: 2).
Some specialists consider that the strategy of a legal translation very
much depends on the text type and on its function. However, considering the
special nature of legal texts, it has been generally agreed t hat substance must
always prevail over form in legal translation. Nevertheless, the issue of
applying literal or free translation to legal texts is very controversial.
For the sake of preserving the letter of law, legal translators have
traditionally been bound by the principle of fidelity to the ST. To put it
differently, the translators’ task was to literally translate legal texts,
reconstructing the form and, more important, the substance of the ST as closely
as possible.
Even though “translation of leg islation and other normative texts require
absolute literalness” (Didier 1990: 280, 285 in Šarcevic 2000: 3), there are
certain legal texts, for example judgments, which allow more freedom in
translation.
This view is not shared by Weisflog who distinguis hes between authentic
texts of national legislation and legislative texts translated ‘purely for

information purposes’ (Weisflog 1987: 195 in id. ibid) and states that legal
translators should use literal translation , regardless of text type and function.
Different from literal translation and free translation , the principle of
legal equivalence (Beaupré 1986: 179, Herbots 1987 in Garzone 2000: 5)
allows more flexibility in the process of legal translation. According to this
principle, “the translation of a legal text will seek to achieve identity in meaning
between original and translation, i.e. identity of propositional content, as well as
identity of legal effects” (Sager 1993: 180 in id. ibid.), pursuing, at the same
time, the objective of reflecting th e intents of the person or body which
produced the ST.
Although discussed from the perspective of legal texts, the above –
mentioned translation strategies are also valid in the case of shorter strings of
legal language such as: stock phrases and collocatio ns. In our case , special
attention will be devoted to the latter.
English collocations used in the legal discourse have various structures.
Depending on the elem ents making up such patterns, Dumitrașcu and Kania
(2010) suggest that frequently used legal co llocations might be grouped as
follows:
 NP [ N + N]: child abuse, court order, family assets, land certificate,
estate contract, assistance order, a.s.o.
 NP [N + N + N] : child assessment order, child safety order, child
support maintenance, community rehabi litation order, community service
order, a.s.o.
 NP [N (+N) + PP] : child of the family, child protection in divorce,
court of first instance, declaration in course of duty, evidence of identity,
notice of intended prosecution, objection in point of law, of fences against
international law and order, presumption of due execution, rectification
of will, trial without pleadings, abuse of process, a.s.o.
 NP [Adj. + N ( +N)]: domestic violence, capital punishment, criminal
court, equitable interests, inferior cour t, legal separation, matrimonial
proceedings (causes), primary evidence, unfavourable witness, legal aid
order, legal aid scheme, a.s.o.
 NP [Adj. + N + N/PP] : criminal bankruptcy order, fast track cases,
fiduciary duties of directors, special procedure ord ers, wrongful
interference with goods a.s.o.
 NP [Adj. + Adj. +N] : private international law, a.s.o.
 VP [V + N (P)/PP] : to administer justice, to meet a claim, to claim
damages, to abide by a decision, to abide by the law, a.s.o. (for further
examples of l egal collocations grouped according to topic and to key
word see Appendices 2 and 3)
In translating such collocations from English into Romanian and Italian,
translators may choose between the form -based strategy and the meaning

based strategy. The former will be preferred with legal collocations that are
formally and semantically equivalent . e.g. to be above the law  RO: a fi mai
presus de lege; IT: essere al di sopra della legge, to break the law  RO: a
încălca legea; IT: infrangere la legge, to make n ew laws  RO: a da legi noi;
IT: fare nuove leggi, to obey/ observe the law  RO: a respecta legea ; IT
rispettare/ osservare la legge, to repeal a law  RO: a abroga o lege; IT:
abrogare una legge). The latter, i.e. the MBS , will be used when the SL
combi nations and structures have to be translated on the basis of a conceptual
representation of their ST meaning (e.g. to take the law into one’s hands  RO:
a lua legea în propriile mâini ; IT: farsi giustizia da se’, to be outside the law 
RO: a fi în afara legii; IT: essere fuori dalla legalità/ legge, to enforce the law
 RO: a aplica legea ; IT: applicare la legge, fare rispettare la legge, to keep
inside/ within the law/to keep the law  RO: a rămâne în legalitate ; IT: restare
nella legalità).
Although the two translation strategies facilitate the transfer of English
legal collocations into Romanian and Italian, not all the TL structures are
perfect equivalents of the SL collocations. In certain situations, there are
obvious formal differences between th e three languages in contact, ( e.g. law
breaker  RO: infractor ; IT: trasgressore della legge, law court RO:
judecătorie/ tribunal IT: tribunale ), whereas in others the lack of total
equivalence results from formal and/or lexico -semantic differences (law school
 RO: facultate de drept ; IT: facoltà di legge, to lay down the law  RO: a
face legea; IT: stabilire la legge/ dettar legge, according to/ in accordance with
the law RO: conform legii IT: in base alla legge ). However, th ere are also
numerous instances in which there is almost total equivalence between English,
Romanian and Italian: in the name of the law  RO: în numele legii IT: in
nome della legge , the law is the same for everybody RO: legea este aceeași
pentru toată lumea ; IT: la legge è uguale per tutti.
An interesting aspect worth mentioning is that the English noun law, has
two equivalents both in Romanian ( lege, drept ) and in Italian (legge, diritto)
which are not interchangeable. This may be a problem for untrained translators,
because each of the two equivalents available in Romanian and Italian has to be
used in specific patterns. Lege and legge are however more frequently used (see
the examples above and Appendi x 6), whereas drept and diritto are selected
when referring to the subdomains falling within the larger domain of law: law
of contract  RO: drept contractual ; IT: diritto contrattuale, (international) law
of nations  RO: drept international; IT: diritto internazionale, law of the sea
 RO: drept maritim; IT: diritto maritimo. In addition, examples such as draft
bill  RO: propunere de lege; IT proposta di legge, and bill  RO: proiect de
lege; IT: progetto di legge prove that the noun bill, used as a term in legal
English ha s a meaning which is totally different from that specific to common
usage.

When comparing and contrasting English legal collocations with their
corresponding Romanian and Italian patterns, it may be easily noticed that
certain SL collocations may be trans lated by using several synonymous TL
equivalent structures: to abide by the law  RO: a se conforma / a respecta/ a se
supune unei legi; IT: rispettare/ osservare la legge, to be against the law  a fi
ilegal, a fi contrar legii; IT: essere illegal/ essere contrario alla legge/ andare
contro la legge, to be inside/ within the law RO: a fi in legalitate; IT: essere
nella legge/ legalità/ essere nei confini della legge, to go to law  RO: a apel a
la lege/ a merge pe căi legale; IT: ricorrere alla legge / procedere per vie legali,
to pass a law  RO: a aproba/ vota o lege ; IT: approvare/ varare una legge, to
remain within the law  RO: a rămâne în legalitate ; IT: rimanere nella legalità/
entro i co nfini della legge, to violate a law  RO: a încalca legea ș IT: andare
contro la legge / violare la legge/ contravvenire a una legge.
Last, but not least, reference should be made to culturally marked legal
collocations in the case of which, linguistic and d omain -specific constraints are
doubled by cultural constraints imposed by the specificity of the LCs brought
into contact. In such situations, it is of utmost importance that translators should
facilitate the transfer of the SL cultural realities to the TL C so as to make
information accessible to the end -users of their translations. For example,
collocations such as la legge Bacchelli or la legge delle 12 tavole can be
translated into English and Romanian by formally equivalent patterns, but they
will be ac cesible to the TL culturally (and legally) ignorant end-users only if
translators provide footnote explanations: la legge Bacchelli  Bacchelli law
 legea Bacchelli – law according to which exceptional individuals who find
themselves in a difficult econo mic situation receive a life annuity, la legge
delle 12 tavole  the law of the 12 tables  legea celor 12 table – the ancient
legislation that stood at the foundation of the Roman law.
To conclude , in translating legal texts, there is always intercultur al
communication. In addition, it should be kept in mind that the legal system is
“an institutional context where the exercise of power through the manipulation
of language is central” (Eades 2004: 304).
The frequent use of collocations in legal texts may create num erous
problems in translation. This is because certain domain -specific terms occurring
in such fixed lexical patterns often make legal collocations totally unintelligible
to translators unaware of their context conditioned meaning(s). Even if cer tain
translators may be unaware of the arbitrariness of collocations in non -technical
domains, such arbitrariness becomes obvious in specific sublanguages, in
general, and in the legal discourse , in particular. To put it differently, “it is
partly by means of collocations restricted in a technical sense that translators
recognize a certain register” (Howarth 1996: 131 in Hansen, Malmkjaer and
Gile 2004: 59), in our case, the legal one.

Moreover, not few are the instances when familiar words are used with a
completely different meaning in the legal discourse. Therefore, the cultural
mediators’ task is that of appropriately transferring the message contained in the
SL so that the TL speakers may get exactly the content encoded in the original
message , but in a form specific to the TLC. In other words, translation in the
legal setting has to be more than a straightforward mechanical process. Legal
translators are not supposed to merely transfer the original text from one
language into another. They must be able to “take linguistic decisions with a
legislative impact” (Correia 2003: 43).

5.6.5.2 . On the translati on of business collocations

Business English, as a type of domain -specific language or ‘ special language ’,
microlanguage ’, ‘technolect ’ (Garzone 2003: 23), covers a wide range of
factual and functional aspects. The former presuppose exploration of the ways
in which activity is organized in areas such as marketing, advertising, sales,
employment, banking, finance, computer and the internet, whereas the la tter
include the ‘essential language of negotiations’ and the ‘language for business
correspondence’(Campbell 2000: V).
Being a ‘contextual -functional variety of the language’ (Berruto 1980: 29
in Garzone 2003: 23), business English is usually defined in relation to the
professional, disciplinary or technical fields to which it pertains, i.e. to
marketing, advertising, sales, employment, banking, finance, computer and the
internet. The main characteristic of this type of functional -contextual language
variety is the “additional correspondence between signifiers and signifieds”
(Berruto 1974: 68 – 69 in Garzone 2003: 25), i.e. the use of specific lexicon
which makes business English less accessible for the non -specialists than
ordinary language. In other wor ds, business English qualifies as a ‘sub -code’.
Although it is not subject to special grammatical and phonological rules, this
domain -specific language has, like any ‘special language’, certain morpho –
syntactic features which are used with abnormal frequen cy, as well as
distinctive discoursal and pragmatic characteristics.
In her approach to economic terminologies, Opriț -Maftei (2018: 78 -79)
identifies the following charcteristics of business English:
 use of specific (semi -technical and highly technical) vocabulary;
 use of academic words;
 use of polysemous words (e.g. balance in accounting refers to the
differemce between the debit and credit sidesof an account, whereas
balance in banking denotes the amount available in an account for
withdrawal or use) ;
 use of nouns of Greek or Latin origin ;

 use of complex pharses (e.g. interest -bearing checking account );
 use of phrasal verbs and phrasal nouns;
 use of passive voice for clarity and indirectness;
 use of the first conditional to give instructions or issue war nings;
 use of metaphors;
 use of internationally recognizable abbreviations, acronyms and
initialisms;
 use of phrases and formulas employed in international business
correspondence.
However, the most important attribute of ‘microlanguages’, in general,
and of business English, in particular, is the lexicon specificity.
At this point mention should be made that the economic development at
international level in the last decades of the previous century and the ever
growing interest of businessmen in establi shing solid and profitable business
relationships with partners from other countries have highlighted the necessity
to facilitate effective communication between the various business partners.
Since communication among people belonging to different countri es cannot be
discussed outside the cultural specificity of the two systems brought into
contact, it becomes clear that the transfer of the message from one language to
another is hardly, if ever, a mere transfer of words ; the cultural element is
always pre sent when the representatives of different language systems interact.
The same holds valid for business communication. As highlighted by Garzone
(2003: 52),
“this is fully justified, as in no other workplace environment as international
business do subjec ts with a different cultural, linguistic and ethnic background
come into contact and have to face issues of mutual or common interest. More
often than not, people from different countries and organizations have to perform
problem -solving tasks concertedly, in situations where important decisions have
to be taken and a lot is at stake, in economic terms”.
In spite of the fact that intercultural differences may represent a problem
in purely interactional conversations, the impact of cultural diversity is much
more evident in transactional conversations, i.e. “in conversations aimed at
achieving some objective that is external to the communicative exchange”
(Garzone 2003: 52). Transactional conversations are assumed to ‘change the
world’, which means that any c ommunication error due to linguistic and cultural
problems may result in obvious and tangible damage, i.e. in failure to close a
deal or conclude a commercial agreement. Consequently , when communicating
with partners belonging to other LCs, businessmen mus t be fully aware of the
ways in which familiarity or lack of familiarity with the cultural specificity of
the languages in contact may affect communication and often business success.

The interdependence between business communication and culture has
favou red the emergence of various theoretical approaches to this issue. The
starting point in analyzing this interaction is represented by Geert Hofstede’
view on culture . He states that the term culture should be extended to
individual life experience, being t hus used to refer not only to national and
ethnic communities, but also to categories of people. He supports this statement
by saying that any national culture is made up of smaller cultures such as
organizational cultures and other sub-group cultures:
“National culture is that component of our mental programming which we share
with most of our compatriots as opposed to most other world citizens. Besides
our national component, our cultural programs contain components associated
with profession, regional b ackground, sex, age group and the organizations to
which we belong. National cultural programming leads to patterns of thinking,
feeling and acting” (Hofstede 1991, in Garzone 2003: 56).
Similar distinctions are adopted by Trompenaars and Hampden – Turner
(1993, 1997 in Garzone 2003: 56) who refer to the existence of three levels of
culture in the professional/organizational world, namely: the cultures of
national or regional societies , corporate or organizational cultures and
professional cultures. Moreov er, approaching culture from the perspective of
translation, Vermeer (1983, in Garzone 2003: 56) distinguishes between
paraculture , i.e. the culture of large national/ethnic groups , and diaculture ,
which corresponds to the sociological concept of sub -cultu re, characterizing
restricted social groups.
Taking this into consideration, it may be easily deduced that the issue of
translating business discourse is definitely very controversial, thus, greatly
debated on.
Like in the case of general texts, speciali sts have tried to find out whether
the translation of domain -specific texts, in our case business texts, presupposes
mere transcoding or transposition of sense . In other words, the dilemma has
been whether translators should consider word -for-word or meani ng-for-
meaning translation. This dichotomy corresponds to the already established
oppositions: literal vs. free translation, formal correspondence vs. dynamic
equivalence, semantic vs. communicative translation and overt vs. covert
translation, respectivel y (House 1997).
Modern translation theories suggest that an essentially functionalist
approach should be used in the case of domain –specific translations. This is due
to the fact that such an approach provides the “principle in the light of which a
choice can be made between different alternatives available” (Garzone 2003:
84). Consequently, the translator’s choices in this situation are made according
to the purpose of the overall translational action, i.e. the function the translated
text will have to ser ve in the target language culture.

As far as the translation of business – specific collocations is concerned,
it is generally agreed that, like in the case of other domain specific collocations,
the most appropriate and useful approach is the functionali st one, as it allows
translators to alternate between the FBS and the MBS .
Translators play an important role in the process of translation. The
extensive use of collocations in business texts requires that translators should
have very good phraseological competence, being thus, able to handle the
ready -made language effective ly. They have to know the ready -made phrases
used in various registers in both languages, as well as to match them and
evaluate them from the sociolinguistic point of view. The elemen ts influencing
the quality of a domain -specific translated text abounding in collocations are:
 the translator’s linguistic competence and cognitive experience for the
decoding and encoding phases of comprehension and production;
 the translator’s collocatio nal competence in the language for special
purposes (LSP);
 the translator’s capacity of giving accurate and adequate
representations in the TL, which depends on the properties of the
discourse and on his/her background knowledge.
Business English, as a wel l-represented type of ‘special’ language, makes
extensive use of collocations. Hill (2000: 51 in Andone and Soare 2002: 19)
suggests that the collocational patterns most often encountered in business
English are the following:
 Adj. + N : slack business lipsă de afacere, stagnare economică, slack
demand cerere slabă, infant industry  industrie nouă, future
deliverylivrare la termen, current deposit deposit curent, temporary
account cont temporar, limited liability  răspundere limitată, active
stock  acțiuni în plină tranzacție, a.s.o.
 N + N (+N ) (compound noun): bear market – piață în scădere, market
leader lider de piață, market research  studiu de piață, market
trends tendința pieții, exchange rate  rată de schimb, inflation level 
rată a infla ției, dividend reinvestment plan  plan de reinvestire a
dividentelor, a.s.o.
 V + A dj. + N: to make fair profits  a realiza beneficii normale, to
revise the original plan  a reanaliza planul inițial, to withdraw/
revoke a (an import/export) licence  a retrage o licență de import/
export, a.s.o.
 V + Adv.: to examine thoroughly  a analiza în detaliu, to increase
dramatically  a crește în mod dramatic, to decline/ fall/ drop slightly/
moderately/ substantially /sharply  a scădea puțin/ moderat/ în mod
substanțial/ puternic, a.s.o.
 Adv. + A dj.: deeply/greatly indebted  cu mari datorii, strictly
confidentia l  strict confidențial, badly damaged  grav avariat, a.s.o.

 V + P rep. + N : to withdraw from a market  a retrage de pe piață,
to run something for profit  a administra / a conduce în mod rentabil
/realizând un benficiu, to sell under price  a vinde sub preț, to exempt/
free from a tax  a scuti de plata taxelor, to take somebody into
partnership  a lua pe cineva ca asociat, a.s.o.
 V + N: to ban p romotion  a interzice reclama, to dominate the
market  a domina piața, to flood/ gut the market  a inunda piața, to
penetrate the market  a pătrunde pe piață, to charge a price a cere un
preț, to fix/ settle/ set a price  a fixa/ stabili un preț, a.s.o.
 N + V: profits fall/ decline/ drop/ shrink profiturile scad , profits
rise/ accrue/ amount  profiturile cresc , the business is booming 
afacerea înflorește, he market is booming  prețurile cresc vertiginos,
a.s.o
 N + V + Adv.: a product sells badly  un produs se vinde greu, a
product sells well  un produs se vinde bine, a.s.o.
 N + and + N: coin and bullion  rezerve metalice în monede și
lingouri.
Some of these collocational patterns ar e also referred to by Hollinger
(2005: 62 -80), who provides e xtensive lists of examples (selected examples):
 V + N/ Pron. (or PP): to sign/ conclude/ enter/ implement/ execute a
contract , to reach/ sign/ conclude an agreement , to witness an
agreement , to discharge/ fulfill/ meet an obligation , to raise/ enter/ file/
lodge/ make a claim , to accept/ buy/ contract/ deliver/ load/ order/ reject/
refuse/ sell/ ship/ supply/ unload/ warehouse goods , to recover/ collect/
settle/ reschedule debts , a.s.o.
 Adj. + N., or N (used attributively) + N: active/ bank/ blocked/
credit/ checking (Am. E)/ current/ deposit/ dollar/ frozen/ notice/
numbered/ time/ savings account , available/ capital/ circulating/ current/
fixed/ floating/ frozen/ hidden/ intangible/ liquid/ net/ net current/
permanent/ tangible/ wasting assets , finance/ holding/ insolvent/
insurance/ investment/ joint -stock/ parent/ private limited liability/ public
limited liability/ shipping/ transport/ transportation company , price cut/
pressure/ rise/ reduction/ war/ sensitivity , market force/forecast/ leader
/researc h/ trends, cost accountant/ centre/ effectiveness/ minimization,
a.s.o.
 V of incomplete predication/ link verb + Adj.: to go bankrupt/
broke/ bust, to be/ run/ come short (of stocks/ supplies/money, etc.)
 V + Adv.: to invest/ lose/ rely heavily, to incre ase/ rise/ climb
dramatically/slightly/ steeply/ substantially/ steadily/ sharply , a.s.o.
 Adv. + Adj.: severely undercapitalized , fiercely/ keenly competitive ,
a.s.o.

 P + N: below/ under/ above the line , at limit , a.s.o. (for examples of
collocations inclu ding the noun business used as a ‘node’ and as a
‘collocate’ see Appendix 5 ).
Considering the great variety of business collocations , mention may be
made that their translation from one language to another implies, on the one
hand, observing a series of li nguistic and domain specific constraints, and on the
other, finding the most appropriate translation strateg ies to transfer them in the
TLC.
Similarly to legal collocations, business English collocations often have
perfect equivalents both in Romanian and in Italian: law of supply and demand
 RO: legea cererii și a oferte ; IT: legge della domanda e dell’offerta ,
temporary account  RO: cont temporar ; IT: conto temporaneo , to dominate
the market  RO: a domina piața ; IT: dominare il mercato , exchange rate 
RO: rată/ curs de schimb  IT: tasso di cambio , to sign a contract  RO: a
semna un contract IT: firmare un contratto .
There are, of course, numerous situations when only partial equivalence
can be achieved in the process of translation: infant industr y RO: industrie
nouă ; IT: nuova industria , term delivery  RO: livrare la termen ; IT;
consegna a termine , active stock RO: acțiuni în plină tranzacție ; IT: azioni
attive .
It is interesting to notice that certain synonymous English collocations
used in bu siness have only one equivalent in Romanian and Italian: profits fall/
decline/ drop/ shrink  RO: profiturile scad; IT: i profiti calano; profits rise/
accrue/ amount  RO: profiturile cresc ; IT: i profiti aumentano . (for examples
of business collocations translated from English into Romanian and Italian see
Appendices 8 and 9).
Far from having covered all the aspects involved in translating business
collocations, this section has hopefully highlighted the diversity and complexity
of such domain specific l exical patterns and the possible approaches to be taken
into consideration in translation.

5.6.5.3 . On the t ranslati on of technical collocations

Technical texts are another illustration of how domain -specificity influence s
and conditions the translator s’ choices. Similarly , to legal and business texts,
the translation of technical texts depends on the translator s’ knowledge of the
linguistic characteristics of the text, as well as on their cultural and domain –
specific, in this case technical , knowledge .
The translating process involves two distinct stages: 1. the decoding the
original author 's message as a receptor, and 2. the encoding this message for the
TL receptors. Mention should be made that, both in decoding and encoding a

text, translators have to take into consideration the purpose of the text i.e. who
or what it was written for, and its social function.
As suggested by Croitoru (1996: 157), a technical translation is “based
on its practical utilization at the “social demand” level. Due to the informative
function of a technico -scientific text, i.e. factual, concrete and material/ and
laying focus on concrete descriptions, the act of rendering such a text into the
TL requires the necessary knowledge of denotation, unlike the translation of
litera ry discourse which requires the very good knowledge of connotation as
well” . This is due to the fact that one and the same lexical item is “a kind of a
mirror -word ” in technical texts, and “a kind of a window word ” in literary ones.
For example, a noun suc h as claw , has certain denotative meanings in the
technical discourse (gheară, dinte, clește de scos cuie (tech.), clichet (mach.), or
when associated with a belt fastener: agrafă de curea (mach.) it may also mean
ambreiaj, cuplaj cu dinți ( mach. ) when ass ociated with clutch and coupling),
whereas in the literary discourse, chiefly in poetry, its denotative meanings are
doubled by connotative ones, the semantic area of this word being remarkably
vast ( sfâșiere, zbatere ).
Technical texts exhibit various cha racteristic lexical features. There are
very few such texts with which the translator may experience little or no
difficulty as far as the technical vocabulary is concerned. Thus, in translating
technical texts, dictionaries are very use ful, but far from s ufficient. Moreover, a
large number of technical words belong to the so -called sub-technical
vocabulary . According to Cowan (1974: 391 in Croitoru 1996: 159), sub-
technical vocabulary includes “those context -independent words which occur
with high frequenc y across disciplines, e.g. inference , simulate , isolate ,
function , etc”. In other words, the lexical items included in the category of sub –
technical vocabulary have the same meaning in several scientific or technical
disciplines.
Another definition of sub-technical vocabulary is provided by Trimble.
In his opinion this term “refers to those words t hat have one or more ‘general’
English meanings and which in technical contexts take on extended meanings,
technical or specialized in some fashion” (Trimble 19 85:129 , in Croitoru 1996:
160). According to Kenne dy and Bolitho (1991: 58 , in id. ibid. ), sub-technical
vocabulary “consists of those words which are not specific to a subject
specialty, but which occur regularly in scientific and technical texts, e.g.
reflection, intense, accumulate, tendency, isolate and dense ”.
When confronted with sub -technical words, translators may know their
common meaning , but they may be confused when coming across them in a
domain -specific context. For example, the noun dog refer s to ‘a domestic
animal ’ in general contexts, whereas in technical usage it has totally different
meanings: (constr .) pârghie de fier, cârlig, scoabă, crampon; (l.) buturugă; nod;
ciot; (mach.) antrenor, clamă, clichet, gheară, dinte; camă, falcă, piedică,

opritor; (mach.) limitator; antrenor; falcă mandrinei; (met.) menghina de mâna;
(min.) clichet, catel la ambreiaj, scoabă, piedică în fața roților unui vagonet;
(nav.) mâner de poartă etanșă; (petr.) praf de grafit; (tele.) jack; (text.) clichet,
cârlig, sabot; clemă; catel; (tech.) deget, piedică, declic, trăgaci. Thus, a
technical translator who does not use technical (specialized) dictionaries, or
ignores the context in which a certain lexical item is used will most probably
make inappropriate translati ons which will sound strange or even funny to the
end-users.
As regards technical collocations made up of such words that allow
multiple interpretations in the common usage and in domain specific contexts,
they may be a puzzle for translators. The semantic content of such fixed lexical
patterns will be appropriately decoded if the context in which they are used is
taken into consideration.
The differences in collocational patterning between two languages
brought into contact reflect the preferences of spec ific language communities
for certain modes of expression and certain linguistic configurations (Baker
1992: 49). Most words have several senses and form different sets of collocates
for each sense. For example, the verb to run meaning ‘to manage’, colloca tes
with nouns like company , institution and business , whereas the same verb,
meaning ‘to operate, to provide’, will co -occur with nouns like service and
course . Thus, there is a strong relationship between the number of senses a
word has and its collocati onal range.
The more general a word is, the broader its collocational range, the more
specific it is, the more restricted its collocational range.
To conclude, when confronted with collocations and collocational
patterns, translators have to take into cons ideration numerous and various
aspects. As suggested by Margareta Ulrych, translators must be “aware of how
lexical boundaries separate words with related meanings and of how these
boundaries differ across languages” (Ulrych 1992: 279). In other words, the y
must have very good phraseological competence and always keep in mind the
fact that each language has specific collocational and colligational patterns.
When translating, one must also take into account the fact that words
interact with other words in a text to produce textuality.
“The cohesion and coherence of a text largely depends on the way lexical, as
well as grammatical, co -referential networks have been set up. The nature of
lexical and grammatical semantic relations in a text is closely related to meaning
and the stylistic effect the writer wishes to convey. Examining a text in terms of
its lexicon may reveal cohesive links that are not immediately noticeable, but
that may sign al recurrent underlying themes. Translators , therefore , need to
identi fy co -referential relations in the ST and the way they contribute to the
overall message of the text” Ulrych (1992: 279).

Their task will be to recreate the same message and effect in the TT by
means of the lexical patterns available in the TL.
As regards specialized translations , domain -specific knowledge is a
prerequisite of adequate comprehension of the “new” and “given” information.
The same assumption holds for any kind of special text, whether technical,
legal, or medical.
“A trained translator sho uld be able to translate very difficult scientific and
technical texts, but s/he should also have acquired more generalized
specializations, such as research skills, terminology management, and electronic
information sources” (Ionescu 2000: 204).
Last, but not least, as suggested by Nida (1996: 114),
“in order to evaluate the success of a translation, it is important to analyze not
only the ways in which monolingual people in the intended audience understand
and appreciate the quality of a translation, but also to have the judgement of
competent bilinguals with experience in creative translating and with sensitivity
to the purposes of the communication in both form and content”.
Moreover, si nce
“translation occurs between languages, between texts between writers and
readers, an between cultures, the complexity of these varied relationships
accounts for the fuzziness of much translation and the difficulty of reliably
pinning down any of the elusive ‘universals’ of translation. Thus, just as meaning
is ofte n a cline, so are texts hybrids composed of many different elements, and
cross -cultural communication can end up being as much about the hybrid ‘third
space’ in -between as about the discrete cultures the commission translation”
(Hatim and Munday 2006: 330) .

CONCLUDING REMARKS

The most relevant aspects approached in the present chapter are summarized
below.

1. Collocability is an aspect of utmost importance in translation because it is
specific to each and every language. In spite of the debates on the myth of the
‘perfect translation’, or of the ideal translator, not few are the instances when
collocability brings about translation problems, even in the case of translators
who have good lingui stic and cultural competences. Most problems occur with
non-native translators, who are always ‘caught’ because of their incorrect use of
collocations . In order to avoid such situations, a translator must have a very
good command of the two LCs brought into contact, i.e. a translator must know

his/her mother tongue very well and he/she must also be linguistically as
competent as a native speaker of the target language culture (TLC) .

2. Distinction is generally made between linguistic and cultural approaches to
translating collocations . This distinction is operatio nal in some situations, but
linguistic and cultural aspects h ave to intermingle so that equivalence between
the SL and TL collocational patterns may be achieved.

3. In the complex and intricate process of translation , the translators’ task is to
continu ally search and research, deconstruct and reconstruct the text, as their
world is one of dichotomies pertaining to the traditional areas of activity of
translators (technical, literary, religious translator, etc.), to modes of
translating (written, oral) a nd to the translator’s priorities or focus (literal vs.
free, form vs. content, formal vs. dynamic equivalence, semantic vs.
communicative translating, translator’s visibility vs. invisibility, domesticating
vs. foreignizing translation).

4. In translatin g collocations, formal and semantic aspects are given priority,
but most often their appropriate transfer in another LC is conditioned by
pragmatic, stylistic and cultural constraints.

5. The translator’s cultural awareness is essential in appropriately translating
collocations. The collocations “created” under the influence of the ST will be
interpreted as unusual collocations in the TL. The problem in this case is
represented by the end -users’ negative attitude towards such poorly written
“original” tex ts which will be identified by them as poor translations .
In order to avoid such situations, phraseological competence needs to be
developed in trainee translators. They have to be familiar with the lexical
patterns used in various registers in both langua ges and they also need to know
how to match and evaluate them from a socio -linguistic point of view.

6. General translations are usually contrasted with specialized translations, the
criterion being mainly the type of vocabulary used. Specialized transla tions are
mainly directed towards transferring information from one language into
another . Literary translations may be considered to have fulfilled their
objective if the TL reader has access, like the SL reader, to the author’s universe
of ideas and feel ings.

7. When confronted with literary texts , the translator’s task is far from being
easy. He has to decode the author’s universe, to render the text message
observing its denotations and connotations in the two languages in contact, and
to re-create th e SLT in such a way as to make the TL readers believe that they

have in front of their eyes not a translation, but the original text. In doing this,
translators have to identify, on the one hand, with the author’s universe in the
TL system and with the TL reader, on the other.

8. The frequent use of collocations in legal texts may create numerous problems
in translation. This is due to the fact that certain domain -specific terms
occurring in such fixed lexical patterns often make legal collocations totally
unintelligible to translators unaware of their context conditioned meaning(s).
Even if certain translators may be unaware of the arbitrariness of collocations in
non-technical domains, such arbitrariness becomes obvious in specific
sublanguages, in gene ral, and in the legal discourse , in particular.

FINAL REMARKS

Starting from the idea that language varies to a great extent across
cultural boundaries, and that lexical patterns, in general, and collocations , in
particular, are illustrative for the way in which linguistic and cultural constraints
may affect the non -native speakers’ fluency in a language, I thought it as
interesting as useful and challenging to make a comparative – contrastive
approach to English collocations in Romanian and Italian in order to see the
extent to which these three languages share linguistic and cultural patterns.
The starting point of my research was the theoretical background
regarding the various types of (fixed) lexical patterns in English and, where
possible, in Romanian and Italian. Given the large number of studies devoted to
such patterns in English, and the sometimes confusing approaches suggested by
the specialists in the field, I considered it necessary to methodologically
organize and present the formal and functional approaches to lexical patterns in
English in order to clarify the position of collocations in the larger class of
lexical patterns. This initial stage of my research represented a prolific source
for further dis cussions on lexical pattern typologies and t erminology
controversies in relevant English studies, and on how certain terms borrowed by
the Romanian studies on lexis, phraseology and semantics can be the source of
further terminological controversies.
I sel ected and methodologically presented aspects essential for the
psycholinguistic , sociolinguistic and cultural approaches to collocations in
English, and I provided a comprehensive description of the linguistic
approaches to such lexical patterns, as well a s of the possible definitions ,
classifications and of the features which distinguish collocations from other
(fixed) lexical patterns in English.
In the chapter devoted to the lexico -semantic approach to collocations , I
made reference to theoretical aspect s which, on the one hand, account for
possible formal and semantic changes in collocations ( denotative and
connotative meanings, sense relations ), and condition the interpretation of
larger stretches of texts in which collocations are used, on the other, ( lexical
cohesion , context and co-text).
Since collocations are agreed to occur in numerous types of discourse ,
and to be context -depend lexical patterns, a better understanding of the
restrictions imposed by such factors could only be achieved by enlargin g on the
stylistic approach to collocations.
As far as the aspects relevant for such an approach are concerned, I
believe that specific reference had to be made to the functional styles in
English, and to the various situations in which the use of colloca tions has to be

adjusted to the communicative situation, either according to the social status of
the speakers (membership to a speech or discourse community , register , sex,
age, etc.), or according to the type of discourse envisaged ( general , literary ,
specialized ).
Last, but not least, the validity of the linguistic and cultural specificity
characterizing collocations was accounted for in the chapter devoted to their
translation from English into Romanian and Italian. The theoretical background
includes g eneral translation theories adapted and applied to the translation of
collocations , and a number of useful (adapted) schemata meant to illustrate
specific strategies which can be used in the translation of lexical patterns in
general, and of collocations , in particular.
In spite of the fact that most of the initial assuptions were validated, and
most of the research aims were achieved, certain gaps in the Romanian and
Italian studies which have been devoted mostly to idiomatic lexical patterns ,
made it diff icult to provide a well-balanced comparative -contrastive
approach in the three languages under discussion.
In addition, t he reduced number of studies specifically devoted to
collocations in Romanian and Italian, and the fact that most of these studies
have English approaches as their starting point, affected the balance of my
compararative -contrastive research orientation . This situation also explains why
not all the examples of (fixed) lexical patterns in English are provided with both
their Romanian and Italian equivalents.
Another problem preventing me from ‘giving equal rights’ to the three
languages approached in my study was the limited usefulness of the dictionaries
available in these languages. Besides the two bilingual (English – Romanian)
nominal and verbal collocation dictionaries, and the English collocation
dictionaries which supported, up to a point, my research, the limited usefulness
of the phraseological dictionaries (of idioms , phrasal verbs , proverbs ), as
well as of the general , and specia lized dictionaries ( economics , business ,
marketing , law), (monolingual or bilingual) available in the three languages
made it difficult to find equivalents for all the collocations analyzed and
commented upon . Hence, the words and terms making up collocati ons had to
be looked up in more dictionaries in order to check the formal and semantic
validity of the equivalents, and to render the right meaning of the respective
collocations in the target language(s).
As regards specialized collocations , the lack of dictionaries updated to
the economic, social, political and cultural realities in both Romanian and
Italian resulted in the impossibility of providing translation equivalents in these
languages for all the collocations se lected .
Far from being complete, th is Multifaceted Approach to Collocations in
English, Romanian and Italian allows for a two -folded conclusion . On the one
hand, the ‘journey’ of English collocations in the Romanian and Italian

territories brings to the fore interesting aspects related to t he possibility of
transferring linguistic and cultural information across borders. On the other
hand, such a transfer is hardly an easy task for the language mediators or
translators, because the resources they have access to (studies, dictionaries,
releva nt corpora) are not equally consistent and relevant in all languages.
Referring to the former dimension of my conclusions , the aspects dealt
with in my study point to the fact that the theoretical and practical aspects
regarding English collocatio ns, in al l their complexity, may be transferred
across Romanian and Italian borders with the appropriate linguistic and cultural
changes that such a transfer implies.
The definitions and classifications of English collocations, as well as the
specific features dif ferentiating collocations from other more or less fixed and
semantically transparent or opaque lexical patterns are valid and applicable in
both Romanian and Italian.
Moreover, the lexico -semantic and stylistic aspects related to the use of
collocations a nd other lexical patterns are common to the three languages since
linguistic phenomena such as sense relations between words ( synonymy ,
antonymy , hyponymy ), lexical cohesion , context and co-text, functional
styles and text types are present in any language .
The significant differences existing between languages, which are
obvious especially at the levels of formal and cultural specificity, are best
visible in translation . This also holds valid in the case of the various types of
lexical patterns existing in any language.
As regards the patterns specific to English, their translation into
Romanian and Italian is not always problematic. The formal simplicity and
semantic transparency of certain lexical patterns in English , as well as the
existence of identica l or similar patterns in Romanian and Italian allow for a
rapid and easy transfer of the SL patterns in the TL(s). However, linguistic and
cultural specificity sometimes make things more difficult, especially if the
translator is not collocationally compet ent in the languages he approaches.
Hence, lexical patterns such as strong coffee or as easy as ABC may be easily
mistaken in translation ( strong coffee *cafea puternică  *caffé potente ; as
easy as ABC *ușor ca ABC *facile come ABC ) if the language mediator is
not familiar with the collocational restrictions governing the corresponding
Romanian and Italian patterns ( strong coffee  cafea tare  caffé forte ; as
easy as ABC ușor ca bună ziua  è un gioco da ragazzi ), respectively.
More problematic are those lexical patterns which, besides being
formally fixed , are also more or less semantically opaque . This is especially
the case of idioms , which are most ofte n contrasted with collocations in
relation to their semantic compositeness. If the meaning of a collocation may be
deduced from the sum of the meanings of its members, this situation does not
hold valid in the case of idioms . Consequently, the appropriate translation of

idioms is conditioned not only by the translat ors’ knowledge of the SL and
TL(s) formal restrictions, but also by their awareness regarding the semantic
content encoded in idioms . That is why an idiom such as to kick the bucket can
be approp riately translated into Romanian ( a da ortul popii ) and Italian ( tirare
le cuoia ) only by those translators who are idiomatically and culturally
competent in the languages brought into contact. Th is is also the situation with
proverbs and sayings. The tran slation of such culturally -marked patterns in
other languages is always conditioned by the translators’ collocational and
cultural competence in the SL and TL(s). Nevertheless, the existence of
culturally -shared patterns facilitates the transfer of proverb s and sayings in
other cultural spaces.
Last, but not least, the translation of domain -specific collocations used
in various types of specialized texts brings to the fore another problem that
translators have to face, i.e. the specialized meaning(s) o f domain – specific
terms (legal , economic , technical , etc.) making up such collocations. Seen
from this perspective, the approach to translating domain -specific collocations
is even more complex. This is due to the fact that the linguistic , collocational
and cultural competences of the translator has to be supported by solid
terminological competence in the specific domain approached.
As far as the latter dimension of my conclusion s is concerned, I should
say that it is centered on the problems encountered in my research and it
includes the following aspects:
 Lexicographers should envisage the creation of monolingual and
bilingual domain -specific collocation dictionaries , as well as the
updating of the bilingual domain -specific dictionaries available in
Romania by including newly – created collocations adapted to the
cultural, economic, political and social nowadays realities (economics,
banking and finances, business, marketing, management, public
administration, EU legislation, politics, etc.).
 The appr opriate translation of domain -specific collo cations can only
be made if translator s work with specialist s in the field or if theyr are
specialist s themselves .
 The translators specialized in a certain field should be trained to
identify and observe domain -specific collocations so as to ensure the
quality of the translation and to facilitate and optimize the translation
process.

APPENDICES

APPENDIX 1 . GENERAL COLLOCATIONS: DO AND MAKE
COLLOCATIONAL PATTERNS

 do + NP: ~ the dishes, the hair ,the homew ork, the ironing, the laundry ,lunch,
the make -up, a movie, the nails, the tango, the twist, the washing, the waltz,
 do + NP (+/ – DET+ N ) + PP.: ~ everything for, nothing for, a favour to,
wonders for,
 do + PP + NP.: ~ for a living, with smth., witho ut smth.,

 make + N: ~ + advances, allegations, arrangements, arrests, assertions,
calls, changes, choices, claims, comments, compromises, concessions,
contributions, conversation, copies, corrections, cracks, cutbacks, cuts,
deductions, demands, discove ries, duplicates, economies, faces, friends, gains,
generalizations, headway, history, home, jokes, journeys, loans, modifications,
money, movies, observations, pledges, predictions, preparations, progress,
projections, purchases, reductions, reference, re forms, representations,
reservations, restitution, revisions, room, runs, savings, withdrawals.

 make + DET. + N:
 ~ + a + bid, bomb, booking, breakthrough, buck, bundle, call, case,
cast, change, check, commitment, comparison, complaint, confession,
contribution, cringe, cry, date, deal, debut, decision, dent, deposit,
diagnosis, distinction, documentary, donation, foray, fortune, gesture,
getaway, guess, judgement, jump, killing, laughingstock, list, living, look,
loss, mark, mess, mint, mistake, move , noise, payment, plea, point, profit,
promise, proposition, push, racket, recommendation, rendezvous, request,
resolution, response, sale, scapegoat, scene, selection, showing, sign,
sortie, speech, splash, square, statement, stink, stop, suggestion, swe ep,
trade, video, vow, waterproof, will, wish.
 ~ + an + allowance, allusion, analogy, apology, appeal, appearance,
appointment, approach, entrance, entry, exception, excuse, impression,
investment.
 ~ + the + adjustments, announcement, assumption, bed, brea k,
connection, grade, impossible, introductions, news,
papers/headlines/front page, presentations, remark, switch, transition,
trip, voyage.
 ~ + every : attempt, effort, endeavour.
 ~ + no : apology, attempt, claim, difference, distinction, effort, mention,
mistake, move, odds, reply, sense.

 make +/- smb./smth. + ADJ: available, bankrupt, certain, easy, happy,
homeless, ill, jealous, legal, mad, mandatory, merry, miserable, nauseous,
nervous, plain presentable, ready, redundant, suspicious, welcome.

 make + NP (+/ -DET + ADJ + N ): a clean break, a clean sweep, a common
cause, a concerted effort, a fast buck, a fresh start, a full/good/remarkable
recovery, a good match, a good/bad call, a huge profit, a left/right turn, a mental
note, a pit stop, a positive con tribution, a round trip, false representations, an
early start, an educated guess, an ideal gift, good copy, good progress, good
sense, good time, great/major/giant strides, serious accusations, the first move.

 make + NP (+/ – DET+ N ) + PP.: a bolt for s mth, a botch of smth., a break
for sth, a/the difference between/ to, a fool of smb, a fuss of smb/ smth, a fuss
about smth, a go of smth, a dart for, a dash for smth., a day/ night/ evening of
smth., a dive for, a hash of smth, a hole in smth, a muck of smth., a nonsense
of smth., a note of, a nuisance of oneself, a pass at smb., a play for smb./smth.,
a run for it, a show of , a spectacle of oneself, a start on smth., a success of
smth., advances to, allowances for smb, amends to sm b/for smth., alterations
to, an approach to, an ass of oneself, an enemy of smb, an example of smb, an
exhibition of oneself, an issue (out) of smth, an offer on/ for smth., capital out
of smth., friends with smb, fun of smb/ smth, game of smb, headway towards/
in/ with, inferences about/ from smth, inroads into/ on smth, love (to/ with smb),
make martyrs of, mincemeat of smb/ smth, make mischief between, make mock
of smb, motion with, no secret of smth, noise for smth., overtures to, peace w ith
smb., plans for, pots of money for, provisions for, reparation (to smb) for smth.,
threats against.

 make + NP (+/ -DET + ADJ + N ) + PP.: a beeline for smb/smth, a big deal/
thing of/out of/about smth., a deep impression on smb., a down payment on, a
drama out of smth., a fool of oneself, a good/bad fist of smth, a grab for/at
smth, a joke (out) of smth, a man (out) of smb, a meal (out) of smth, a mess of
(doing) smth, a mockery of smth., a monkey (out) of smb, encouraging noises
about smth., go od use of smth., hard work of smth, heavy weather of smth, heavy
work of smth,, the supreme/ultimate sacrifice.

 be made + NP/ PP: for life, (out) of smth., for each other,of/ from smth.,
of money, of sterner stuff, of stone.

 make + (NP – smb/smth. ) + bare infinitive: ask, accept , believe, escape,
exit, hope , ends meet, pay, save, say, sneeze, wish, wonder

APPENDIX 2 . LEGAL COLLOCATIONS GRO UPED
ACCORDING TO TOPIC

Collocations associated with the CRIMINAL : a convicted killer, a sex
offender, br utality and insanity matched by enormous cunning, committable
mental illness, competent/capable to stand a trial, grotesque killing, macabre
killing, not to leave a single clue, obscene parody, obscene phone caller, overtly
schizophrenic and potentially su icidal, sex offenders, sexual deviation, sexual
psychopath, to add appalling refinement to an attack, to be released from
mental institution, to be seized by an uncontrollable sexual urge, to break into
the apartment, to burgle an apartment, to commit a r obbery, to commit breaks,
to commit suicide, to cover leads, to deny assaulting young women, to fit the
profile, to gag a victim with something, to gain access to locked apartments, to
indecently assault young women, to leave a clue, to molest women, to ki ll
women, to ransack an apartment, to refuse to answer any questions, to remove
property, to stand trial, to strike at random.

Collocations associated with the VICTIM : a grossly exposed state, brutally
murdered, genuine/spurious suspicions/alarms/tips, t o be sexually assaulted, to
be sexually molested, to be strangled by human hands, to become alarmed, to be
obviously dead, to drop dead of a heart attack, to identify a murderer, to
identify the attacker, to identify the man responsible, to report prowlers /obscene
phone calls/letters, to testify against somebody.

Collocations associated with the POLICE: a host of experts, a partially
inaccurate psychological profile, a sign of forceable entry , a solution to the
atrocities, a total lack of clues, a whole n ew avenue of inquiry, administrative
anomaly, exhaustive inquiries, forensic medicine, gruesome detail, initial
suspicion, investigation team, minute detail, natural detection technique, police
report, possible candidates for the crime, scores of detectiv es, supernatural
detection technique, the capture of the strangler, the conviction of the strangler,
the established strangler type, the police mounted the greatest man -hunt, the
strangler dragnet, to administer lie -detector tests, to arrive on the sce ne within
minutes, to bring to the police headquarters, to build up a complex
psychological profile of the strangler, to cancel all police leave, to carry out
house -to-house inquiries, to chase criminals, to conduct a through
investigation, to conduct rout ine questioning, to confirm fears, to confirm
suspicions, to corroborate a confession, to deal with homicide, to deliberately
keep secret, to disclose facts, to establish a likeness of the attacker, to establish
the identity of the killer, to fight crime. to have ample evidence, to have no clue

to one’s actual identity, to hold a press conference, to hold identity parades, to
hunt criminals, to interrogate criminals/suspects, to lift prints, to look for clues,
to make no headway, to pick up possible candi dates/suspects for the crime, to
pursue criminals, to question at length, to question possible candidates for the
crime, to reassign detectives to homicide, to receive a compliant, to record a
confession, to reduce the investigation team, to search for clu es, to stop and
frisk suspicious persons, to take over the investigation, to toss a room/luggage,
to trace possible suspects, to wage a war on crime, to wind up a conference on
murders, uncorroborated testimony.

Collocations associated with the CRIMINAL C OURT: expert witnesses for
the prosecution/defense, further orders of the court, to accept the prosecution
argument, to await trial, to be committed to a hospital for criminally insane, to
be convicted by testimony, to convict of armed robbery/ breaking a nd entering/
theft/ assault/ sexual crimes, to convict the man responsible, to sentence to life
imprisonment.

APPENDIX 3. LEGAL COLLOCATIONS GROUPED
ACCORDING TO KEY WORD

ADMINISTRATION n. – administration action, administrati on bond,
administration of poison, administration order, administration pending suit,
current administration , incumbent administration , (in)efficient administration ,
outgoing administration .

CLAUSE n. – break clause , cesser clause , charging clause , equali ty clause ,
honour clause , interpretation clause , paramount clause/ bill of landing

CHILD n. – battered child , child abuse, child assessment order, child
destruction, child employee, child of the family, child of unmarried parents,
child protection in div orce, child in care, child in need, child support
maintenance, foster child , illegitimate child , natural child, protected child ,
welfare of the child .

COURT n. – to address the court , Admiralty Court , borough court , to
appear before the court , to bri ng somebody before the court , civil court ,
Commercial Court , Companies Court , contempt of court , county court , court
case, court hearing, court martial, Court of Appeal, Court of Arches, Court of
Chancery, Court of Chivalry, Court of Common Pleas, Court of Criminal
Appeal, Court of Ecclesiastical Causes Reserved, Court of Exchequer, court of
first instance, court of inquiry, Court of Justice of the European Communities
(European Court ), court of last resort, Court of Probate, Court of Queen’s
Bench, court of record, Court of session, court of summary jurisdiction, court
order, court proceedings, criminal court , election court , High Court of Justice,
higher court , court proceedings, inferior court , International Court of Justice,
magistrates’ court / police court , military court , naval court , patents county
court , Patents Court , payment into court , to preside over the court , prize court ,
rules of court , Rules of the Supreme Court , to settle something out of the court ,
superior court , to take somebody to court, Tolzey Court , youth Court .

DECLARATION n. – declaration against interest, declaration concerning
pedigree, declaration concerning public or general rights, declaration in
course of duty, declaration on intention, declaration of trust, false declarat ion,
to issue a declaration , joint declaration , to make a declaration , preliminary
declaration , previous declaration , to sign a declaration , unanimous
declaration, voluntary declaration .

EQUITABLE adj. – equitable assignment, equitable charge, equitable
easement, equitable estate, equitable estoppel, equitable execution, equitable
interests, equitable lease, equitable lien, equitable mortgage/charge, equitable
presumptions, equitable remedies, equitable rights, equitable waste.

ESTATE n. – estate agent, estate contract, estate duty, estate for years, estate
owner, estate rentcharge, estate subsisting, privity of estate

EVIDENCE n. – admissibility of evidence , admissible evidence , to admit
evidence advice on evidence , best -evidence rule, to call evidence ,
circumstantial evidence , evidence in rebuttal, evidence obtained illegally,
evidence of character, evidence of disposition, evidence of identity, evidence of
opinion, evidence of user, hearsay evidence , incriminating evidence , parol
evidence , parol eviden ce rule, presumptive evidence / prima facie evidence ,
primary evidence , real evidence , secondary evidence , to set much store by
evidence , similar -fact evidence , to withdraw evidence .

JUDGE n. – circuit judge , judge advocate, judge in his own case, presidin g
judge .

JUDICIAL adj. – judicial cognizance/ judicial notice, judicial dictum, judicial
discretion, judicial immunity, judicial precedent, judicial review, judicial
separation order, judicial trustee, judicial barrister.

LAW (S) n. – to abide by law, adjective law, to administrate law,
administrative law, air- force/ service law, to apply law, approximation of
laws, case law, civil law, common law, Community law, company law, conflict
of laws/ private international law, consumer law, contract law, to c ontravene
law, directly applicable law, directly effective law, to enforce law,
environmental law, foreign law, general principles of law, to implement law,
imposition of martial law, ineffective law, to infringe law, infringement of law,
internationa l law/law of nations, intertemporal law, just law, labour law, Law
Commission, Law Lords, law merchant, law of obligations, law officers of the
Crown, law of nations, law of the sea, Law Reform Committee, law reports, law
sittings, Law Society, military law, municipal law, natural law/positive law,
objection in point of law, oppressive law, private law, proper law of a contract,
property law, public law, public international law, powerful law, rules of law,
service law, statute law, substantive law, sus law, to suspend law, violation of
law, unenforceable law, unworkable law, welfare law .

LEGAL adj. – legal aid order, legal aid scheme, legal assignment, legal
easement, legal estate, legal fiction, legal fraud/ constructive fraud, legal lease,

legal memory, legal mortgage, legal person, legal rights, legal separation,
Legal Services Ombudsman, legal year.

LEGISLATION n. – to adopt legislation , to apply legislation , Community
legislation , delegated/ subordinate legislation , effective legislation , inadeq uate
legislation , to obstruct legislation , a package of legislation , strict legislation ,
subdelegated legislation to vote down legislation .

LIABILITY n. – absolute liability , to accept liability , to acknowledge liability ,
to admit liability , to assume liability , civil liability , contribution, employer’s
liability , to deny liability , full liability , to incur liability , limited liability ,
parents’ liability , products liability , strict liability .

MARRIAGE n. – breakdown of marriage , common -law marriage ,
consummation of marriage , jactitation of marriage , marriage articles,
marriage brokage contract, marriage by certificate, marriage by Registrar –
General’s licence, marriage by religious licence, marriage ceremony, marriage
settlement, nullity of marriage , restraint of marriage , sham marriage , voidable
marriage .

NOTICE n. – blight notice , default notice , enforcement notice , notice of
abandonment, notice of discontinuance, notice of dishonour, notice of intended
prosecution, notice of intention to defend, notice to produce, notice to quit,
notice to threat, prohibition notice , special notice , stop notice .

PROOF n. – burden of proof , proof beyond any reasonable doubt, proof of age,
proof of birth, proof of handwriting, proof of marriage.

TRIBUNAL n. – adminis trative tribunal , agricultural land tribunal , tribunal
of inquiry .
WILL n. – execution of will, interpretation of will, mutual wills , joint will,
nuncupative will, privileged will, proving a will, rectification of will,
republication of will, signature of will, tenancy at will.

WITNESS n. – adverse witness , compellable witness , hostile witness , key
witness , recall of witness , reliable witness , subscribing witness , trustworthy
witness , unfavourable witness , witness order, witness’ oath, witness warrant,
zealous witness .

APPENDIX 4. SPECIALIZED COLLOCATIONS INCLUDING
COMPUTER TERMS

NP [Det. + N]: my account info, my computer, my contact details, my
profiles, my status, my webcam.

NP [ADJ. +N]: analogue computer, comedy website, commercial website ,
company website, corrupted data, corrupted file, digital computer,
education(al) website, golf website, news website, official website, parallel
computer, personal computer, personal website, portable computer, powerful
computer, travel website, unoffic ial website.

NP  [N + N/PP]: block of text, computer desktop, computer literate,
computer virus, display image, home computer, home page, left mouse button,
notebook computer, program files, program icon, right mouse button, site
address, web browser, we b crawler, web design, web designer, web developer,
web development, web directory, web editor, web guide, web host, web index,
web link, web master, web network, web page, web publishing, web resource,
web server, web service, web tool, web traffic, web u ser, website address,
website URL, website designer, word document.

VP  [V + NP]: to access the website, to add files, to burn CDs or DVDs, to
browse/search the Web/a website, to check out/ look into/ look at/ visit a
website, to click on /follow a link , to click/release the right/left mouse button,
to click search, to click on a window, to close a document/window, to compress
files, to connect to the internet, to copy a document/file, to create a document/
file/webpage, to cut a document/file, to delet e/erase files/programs, to double
click on a program icon, to download files/ information, to drag the mouse, to
edit a document/ options, to eject/ remove/take out the disk, to explore the
internet, to format a disk/ document, to hide folders , to import documents, to
insert a disk, to interface the computer, to link the computer, to log on the
computer, to log off the computer, to move the cursor, to move and resize the
program window, to move a text/ a file/ a document, to network the computer,
to open a file, to organize files, to press a button, to print out a document, to
reboot the computer, to rename a document/file, to restart the computer, to
return to home page, to run a program/ the computer, to save a file/ document,
to scroll down the text, to select a file/program, to select an option, to share
documents/photos, to show offline messages, to shut down the computer, to sign
in/out/up, to start up/ power up the computer, to surf the web, to switch on the
computer, to switch off the computer, to t ype a document, to type in a website
address, to unzip documents/files, to wipe the disk, to zip documents/files.

APPENDIX 5. COLLOCATIONS INCLUDING THE NOUN BUSINESS
AS A ‘NODE’ AND AS A ‘COLLOCATE’

[N] + business: airline business , banking business , clothing business , cosmetic
business , core business , design business , distribution business , entertainment
business , export business , family business , fashion business , film business ,
fishing business , franchise business , freight business , funeral business ,
gardening business , holiday business , insurance business , internet business ,
manufacturing business , movie business , music business , pharmacy business ,
publishing business , racing business , telecommunication business , television
business , textile busines s, tobacco business , transport business , travel
business , tourism business , tourist business , whisky business .

business + [V] : business be amalgamated with, business boom, business close
down, business collapse, business decline, business do, business expand,
business fail, business flourish, business go bankrupt, business go well,
business grow, business pick up, business prosper, business take off.

business + [N] : business account, business acquaintance, business activity,
business acumen, business advisor, business affairs, business analyst, business
appointment, business arrangement, business aspect, business asset, business
assignment, business associate, business association, business banking,
business card, business circle, business class, busine ss client, business
climate, business colleague, business commitment, business communication,
business community, business concept, business conduct, business confidence,
business conglomerate, business connection, business consultant, business
contact, business context, business contract, business corporation, business
correspondence, business correspondent, business counseling, business crime,
business culture, business customer, business cycle, business deal, business
dealings, business degree, business delegation, business development, business
directory, business disaster, business discipline, business district, business
economics, business economist, business efficiency, business elite, business
empire, business end, business engagement, business enterprise, business
entity, business entrepreneur, business environment, business equipment,
business ethics, business executive, business expansion, business expense,
business expertise, business failure, business headquarters, business hours,
business initiative, business instinct, business interests, business investment,
business investor, business jet, business journal, business journalist, business
leader, business lease, business letter, business life, business link, business
lunch, business magazine , business management, business manager, business
matters, business meeting, business need, business news, business objective,
business operations, business opportunity, business outlook, business package,

business page, business park, business partner, business partnership, business
people, business performance, business person, business plan/planning,
business point of view, business position, business practice, business premises,
business presentation, business property, business proposition, business
publication, business purposes, business relationship, business reply mail,
business requirement, business return, business rival, business schedule,
business school, business secret, business section, business sector, business
seminar, business sense, sh ow business , business side, business skill, business
stationery, business strategy, business studies, business success, business suit,
business survival, business tenancy, business tenant, business tie, business
transaction, business travel, business traveller, business trip, business tycoon,
business user, business as usual, business venture, business visitor, business
vocabulary, business woman, business world, business worry.

[ADJ.] + business : awful business , bad business, big business , brisk busine ss,
catering business , computer business , core business , current business , daily
business , defunct business , dreadful business , efficient business , excellent
business , family business , flourishing business , funny business , good business ,
growing business , impo rtant business , international business ,
investment business , large business , local business , long -established business ,
lucrative business , mail order business , medium -sized business , official
business , personal business , poor business , pressing business , private
business , profitable business , prosperous business , retail business , routine
business , run -down business , slack business , slow business , sluggish business ,
small business , state -owned business , strange business , successful business ,
terrible business , thriving business, unfinished business , urgent business ,
wholesale business , etc.

[V] + business : attend to business , be in business , be back in business , build up
business , carry on (often law) business , close down business , conduct business ,
deal with business , discuss business , do business , drum up business , encourage
business , establish business , expand business , finance business , find business ,
generate business , get down to business , give up business , go into/ after/ out of
business , go about one’s business , grow business , handle business ,
have business , have no business , hunt for business , interfere in sb. else’s
business , join business , leave business , look for business , lose business , make
sth. business manage business , mind one’s own business , moder nize business ,
open business , own business , promote business , prop up business , put sb./ sth.
out of business , restructure business , run business , set up in business , start
business , streamline business , talk business , take over business , transact
business , tout for business , wind up business , work in business

[PREP.] + business : in business , on business

PHRASES: business and pleasure, business as usual, business or pleasure,
mix business with pleasure , a place of business doing sth/to do sth business ,
none of your business / no business of yours, any other business , make it one’s
business to do sth. a business , to stick one’s nose into sb. else’s business .

APPE NDIX 6 . COLLOCATIONS INCLUDING THE NOUNS LAW (EN),
LEGE / DREPT (RO) AND LEGGE / DIRITTO (IT) AS ‘NODES ’ AND
‘COLLOCATES ’

EN:

 NP [ (Adv.) + Adj. + law]: administrative law, civil law, common law,
corporate law, criminal law, (directly) applicable law, (directly) effective law,
domestic law, environmental law, foreig n law, ineffective law, international
law/law of nations, intertemporal law, just law, maritime law, military law,
moral law, municipal law, natural law/positive law, oppressive law, private
law, private international law, public law, public international law, powerful
law, statutory law, tort law, unenforceable law, unworkable law.
 NP [N + PP  P + NP – law] – approximation of laws, conflict of laws,
general principles of law, imposition of martial law, infringement of law,
objection in point of law, rules of law, violation of law.
 NP [(Det.) + law + PP]: proper law of a contract, law of nations, law of
obligations, law of the sea.
 NP [ N + law]: adjective law, admiralty law, air – force/ service law,
business law, case law, Community law, company law, con sumer law,
contract law, copyright law, family law, labour law, property law, reading
law (legal apprenticeship), service law, statute law, substantive law, sus law,
tax law, welfare law.
 NP [Adj . + N + law]: personal injury law.
 NP [law + N]: law clerk, Law Commission, law enforcer, law enforcement,
Law Lords, law merchant, law officers of the Crown, Law Reform Committee,
law reports, law sittings, Law Society law suits, law ways.
 VP [V + NP (Det) + law]: to administrate law, to apply law, to break the
law, to contravene law, to enforce the law, to implement law, to infringe law,
to obey/ observe the law, to pass a law, to repeal a law, to suspend law, to
violate a law.
 VP [V + PP  P + NP (law)]: to abide by law, to be above the law, to be
against the law, to go to law, to keep within/inside the law, to remain within
the law.

RO:

 NP [ lege + Adj] .: lege constituțională, lege permanentă.
 NP [ lege + NP]: lege a brevetelor , legea cererii și a ofertei.
 NP [ lege + PP]: lege de procedură, lege în vigoa re .

 NP [ N+ PP P + NP lege]: anexă la o lege, fraudă la lege.
 NP [drept + Adj]: drept administrativ, drept /cod civil, drept commercial,
drept contractual, drept cutumiar, drept internațional, drept internațional
privat, drept maritim, drept material, drept penal, drept procedural.
 VP [V + lege]: a aplica legea, a explica o lege, a interpreta o lege.

IT:

 NP [legge + Adj] : legge costituzionale, legge ordinaria, legge speciale,
legge eccezionale; legge delegata, legge finaziaria.
 NP [legge + N]: legge quadro, legge cornice, legge delega, legge stralcio.
 NP [N + PP P + NP legge] : a norma di legge , a termini di legge , abbozzo
di legge, addizioni alla legge , approvazione della legge, correzioni alla legge,
il potere della legge , in nome della legge, l'autorità della legge , la maestà della
legge, lo spirito della legge.
 NP [diritto + Adj ]: diritto contrattuale, diritto internazionale, diritto
maritime, diritto privato, diritto procesuale, diritto pubblico, diritto societario.
 VP [ V + legge] : abolire una legge, ricorrere alla legge, essere previsto
dalla legge.

APPENDIX 7 . (FIXED) LEXICAL PATTERNS IN TRANSLATION

7.1. COLLOCATIONS

 white wine  vin alb vino bianco ;
 a white night noapte albă  una notte in bianco ;
 a white pape   una pagina bianca ;
 the white of the egg  albusul oului  il bianco dell’uovo ;
 to show the white feather a ridica steagul alb;
 the white of the eye  albul ochiului il bianco dell’occhio ;
 turn pale/ white with fear a se face galben ca ceara/ca făclia de ceară/
lămâia/ turta de ceară, galben de frică, mânie/ spaimă  diventare bianco per
la paura ;
 to turn smb’s hair white /grey far venire i capelli bianchi ;
 the yellow of an egg  il giallo dell’uovo/ il rosso dell’uovo ;
 a vedea totul în roz vedere tutto rosa;
 red wine vin roșu  vino rosso ;
 to see red  a vedea roșu  vedere rosso;
 to be red with anger  essere rosso di rabbia ;
 to have blue blood a avea sânge albastr u avere il sangue blu;
 to be green with e nvy essere verde d’invidia;
 to give someone/get the green light a da undă verde ;
 grey matter materia cenușie  materia grigia ;
 grey sky  cer cenușiu  cielo grigio ;
 a black list  lista neagră lista nera;
 black magic  magie neagră magia nera;
 the black sheep of the family  oaia neagră a familiei  la pecora nera
della famiglia ;
 black despair  disperazione nera .

7.2. BINOMINALS

 the land flowing with milk and honey/ the never -never land  țara unde
umblă câinii cu covrigi în coadă/ unde curge lapte s i miere ;
 to live on bread and cheese  a se ține cu pâine și apă;
 to quarrel with one’s bread and butter  a se supăra ca văcarul pe sat;
 to fret and fume /to turn Turk  a se face foc (și pară de mânie) /a se face
Dunăre turbată ;
 the butcher the baker a nd the candle -stick-maker  Nea Stan si nea Bran

 If ifs and ands were pots and pans/ If my aunt had been a man, she’d have
been my uncle /  Dacă badea ar avea cosițe i s-ar zice lele/ dacă baba ar avea
roate i s -ar zice tramvai / dacă cocoșul ar face ouă i s-ar zice găină .

7.3. SIMILES

 as ugly as sin/as a sinner /as death /as a scarecrow  urât foc/ foamete / de
mama focului / ca ciuma /ca dracu’ /ca noaptea ;
 to grin like a Cheshire cat  a râde ca prostul ;
 as white as a sheet/ ghost /snow  alb ca varul/ zăpad a/ hârtia bianco
come un panno lavato;
 as red as a beet/ beetroot/(boiled) lobster/rose/ turkey -cock  roșu ca un rac
fiert/ sângele/ focul/ para foculu/ sfecla/ gotca  essere rosso come un
gambero/ peperone/ pomodoro;
 as black as thunder ink/coal  a fi negru ca cerneala/ catranul/ abanosul/
cărbunele/ fundul ceaunului/ noaptea/ păcatul/ corbul/ pana corbului/ smoala/
tăciunele  eseere nero come l’inchiostro/ carbone ;

7.4. IDIOMS

 to go over/to cross the Great Divide / to make a die of it (AmE)  a da ortul
popii / a-l lua aghiuță ;
 to live in a fool’s paradise  a-și inchipui că l -a apucat pe Dumnezeu de un
picior ;
 Thank Heaven !, Heaven be praised  Cerul să fie lăudat!
 Good heavens ! Holly Moses ! Good gracious! Great Scott ! Sfinte Sisoe
 at Jericho ! at th e ends of the earth  la dracu /dracu -n praznic//unde a
ințărcat dracul copiii/ unde si -a pierdut dracul potcoavele/ unde și-a spart
dracu’ opincile ;
 go to hell/blazes/ Jericho /the devil /the Dickens  lua- te-ar dracul /naiba ;
 he is in the seventh heaven  parcă a prins pe Dumnezeu de un picior ;
 he has the devil’s own luck  I-a pus Dumnezeu mâna în cap ;
 to observe the Sabbath  a ține duminicile ;
 to break the Sabbath  a nu ț ine duninicile ;
 like master , like man; like mother , like child ; like father , like son  Cum e
stăpânul , așa e si sluga / Cum e turcul și pistolul /Cum e sacul și petecul ;
 to live like a lord/prince  a trăi ca un boier/ a tră i ca un nabab44 /a trăi
regește ;
 the land of Nod  țara lui Moș Ene;

44 nabab (< French ‘ nabab’) – the Sultan’s superior officers in Iran and Pakistan.

 when the Ethiopian changes his skin, on St. Tib ’s eve / when the pigs begin to
fly  la calendele grecesti / la paștele cailor/ la moș așteaptă/ c ând or zbura
porcii ;
 to carry coals to Newcastle  a merge la vie cu strugurii -n batistă ;
 to take the French leave  a o șterge englezește / a șterge/sp ăla putina45;
 to walk Spanish  a umbla cu grijă .

7.5 PROVERBS AND CLICHÉS

 Don’t look a gift horse in the mouth.  Calul de dar nu se caută la dinți 
A caval donato non si guarda in bocca .
 He who sows wind/storm will reap whirlwind  Cine seamănă vânt, culege
furtună  Chi semina vento, raccoglie tempesta.
 An old ox makes a straight furrow .  Găina bătrână face supa bună 
Gallina vecchia fa buon brodo.
 A good beginning is half the battle .  Ziua bună se cunoaște de dimineață.
 Il buon giorno si conosce dal mat tino.
 Appetite comes with eating.  Pofta vine mâncând.  L’appetito vien
mangiando.
 Fish begins to stink at the head.  Peștele de la cap se împute.  Il pesce
puzza dalla testa.
 Hell is paved with good intentions.  Drumul spre iad e pavat de intenții
bune.  La via dell’inferno e lastricata di buone intenzioni.
 Lies have short legs.  Minciuna are picioare scurte.  Le bugie hanno le
gambe corte.
 Little is better than non e.  Mai bine mai puțin decât nimic.  Meglio poco
che niente.
 Better late than nev er(but better never late). Mai bine mai târziu decât
niciodată.  Meglio tardi che mai.
 There is no rule without exception. Nu există reguli fără excepție. Non
c’è regola senza eccezioni.
 There is no rose without thorns.  Nu există trandafiri fără spin i.  Non
c’è rosa senza spine.
 There is no smoke without fire.  Nu iese fum fără foc. Non c’è fumo
senza arrosto.
 Nothing is for free. Nimic nu e pe gratis. Non si fa niente per niente.
 Promise little and do much .  Promisiunea dată, datorie curată.  Ogni
promessa è debito.

45 putina (< Latin) – a type wooden barrel used for keeping cheese or pickles.

 A fault confessed is half redressed.  Greșeala recunoscută e pe juma’
iertată.  Peccato confessato è mezzo perdonato.
 When the cat’s away, the mice will play.  Când pisica nu -i acasă, șoarecii
joacă pe masă.  Quando il gatto non c’è i topi ballano.
 He laughs best who laughs last.  Cine râde la urmă, râde mai bine. 
Ride ben chi ride l’ultimo
 Two dogs strive for a bone and the third runs away with it.  Când doi se
ceartă, al treilea câștigă.  Tra due litigandi il terzo gode.
 All is well that ends well.  Totul e bine când se termină cu bine.  Tutto è
bene quel che finisce bene.
 All is not gold that glitters.  Nu tot ce zboară se mănâncă.  Non e’ oro
tutto quel che brilla.
 The devil is not so black/the lion is not so fierce a s he is painted.  Dracul
nu este chiar așa de negru cum se spune.  Il diavolo non è cosi’ brutto come
si dipinge.
 He that never climbed never fell.  Din greșeli se învață.  Sbagliando si
impara.

APPENDIX 8 . ENGLISH – ROMANIA N ‘BUSINESS’ COLLOCATIONS

business + [V]: business expands  afacerea se extinde; business is slow 
afacerea este înceată.

business + [N] : business address  adresa firmei, sediu; business advisory
council comisia de consultanță pentru afaceri; busines s agent (USA)  șeful
de sindicat la nivel local; business appointment  întâlnire de afaceri;
business bad debts  creanțe irecuperabile create în decursul activității firmei
(deductibile din profit); business brokers  intermediari de afaceri; business
call  vizită de afaceri/ făcută pentru a discuta cu cineva de afaceri; business
centre  centru comercial parte a unui oraș în care se află principalele bănci,
magazine și birouri; business circles  cercuri de afaceri; business class
clasa business; business college/school  colegiu /școală de studii economice;
business combinations  regrupări de intreprinderi; business connections 
relații de afaceri; business consulting  consultanță de afaceri; business
corporation  societate comercială, corporație; business correspondence 
corespondență de afaceri; business correspondent  corespondent de afaceri,
jurnalist care scrie articole despre știrile din domeniul afacerilor; business
counsellor  consultant de afaceri; business cycle  ciclu al afacerilor/ ciclu
economic; business damage  daune comerciale; business data processing 
informatică de gestiune; business day  zi de lucru; business efficiency
exhibition  expoziție pentru eficientizarea afacerilor; entity concept 
conceptul de entitate lucrativ ă; business establishment  intreprindere,
companie, firmă, societate comercială, asociație lucrativă; business
environment  mediu de afaceri; business ethics  etică în afaceri; business
expansion  extinderea activității; business expenses  cheltuieleile firmei,
cheltuieli cu firma; business expertise  profesionalism, abilitate în afaceri;
failure  eșec comercial, faliment; business for sale  fond de comerț de
vanzare; business games  jocuri de afaceri adesea pe calculator în care
indivizi sau grupur i concurează pe o piață ipotetică; business interruption
insurance  asigurare pentru o sumă fixă pentru situația întreruperii afacerii;
business judgement rule  tribunalele nu intervin în conducerea
intreprinderii; business law  cod comercial, legea soc ietăților; business
letter  scrisoare de afaceri; business liability  răspundere comercială
(pentru neîndeplinirea unor obligații asumate în activitatea comercială);
business loan  credit pentru activitatea curentă; business loans  pierderi
de exploata re, pierderi din activitatea de bază; business lunch  dejun de
afaceri la care se discută chestiuni legate de afaceri; business name 
denumirea firmei; business ownership  participație la capitalul unei firme;
business plan  plan de afaceri; business premises  sediu; business
portofolio analysis  analiza a portofoliului de afaceri; business premises

policy  polița de asigurare pentru sediu; business property  patrimoniul
societății; business prospects  afaceri posibile; business pubblications 
publicații economice; business quarters  sediu comercial al întreprinderii;
business receipts  veniturile societății; business relations  relații
comerciale; business risk  risc de firmă; business segment reporting 
raportare segmentală a activității ec onomice; business structure  structura
firmei, structura afacerii; business tenancy  proprietate deținută în scop
comercial; business transaction  comerț, tranzacție comercială; business
travel/ trip  călătorie de afaceri/ în interesul firmei; business unit  unitate
comercială; business year  an comercial/ financiar;

[V] + business : to be in business  a fi în afaceri, a fi implicat în activități
comerciale; to go into business  a intra în afaceri, a crea o firmă; to go out
of business  a ieși di n afaceri , a înceta activitatea comercială; on business
 în interes de servicu; business card  carte de vizită, business equipment
 echipament de birou; business hours  ore de lucru, program de lucru; big
business companii foarte mari; any other business  orice altă problemă;

PHRASES: business to business advertising  publicitate adresată
companiilor.

APPENDIX 9. ENGLISH – ITALIAN BUSINESS COLLOCATIONS

[N] + business : the cotton business  il commercio del cotone; family business
 impresa familiare, insurance business  ramo assicurazioni; line of
business ramo di attivita.

business + [N] : business activity  attivita commerciale; business address 
indirizzo d' ufficio; business administration  amministrazione aziendal e/
gestione aziendale, business administrator  amministratore aziendale;
business agent  agente di affari; business appointment  appuntamento
d'affari; business assistence  assistenza alle aziende; business associate 
collega d'affari, socio; busin ess association  impresa, azienda; business
barometer  barometro economico; business budget  budget aziendale; (
econ.) business call  visita d'affari; business card  biglietto di visita;
business cash  moneta commerciale; business centre  centro d egli affari/
centro commerciale; business class  classe imprenditoriale, uomini d' affari;
business class (transp.)  classe business ; business combination  fusione di
aziende; business concern  impresa, azienda; business connections 
relazioni d' affa ri;business consultant  consulente commerciale/ consulente
aziendale, commercialista; business correspondence  corrispondenza
commerciale; business correspondent  corrispondente commerciale;
business credit  crediti societari; business customs  consue tudini
commerciali; business cycle (econ, US)  ciclo economico; business day 
giomo lavorativo; business deal  operazione commerciale; business deposits
 depositi commerciali; business economics  economia aziendale; business
economist  aziendalista/ economista aziendale; business efficiency 
efficienza aziendale; business end  parte operativa di un oggetto (punta di
coltello, canna di fucile, etc. ), business enterprise  impresa/ azienda,
business entity  entita aziendale; business entity concept  concetto
dell'entita aziendale; business ethics  etica aziendale/ imprenditoriale;
business expenses  spese di esercizio; business failure  fallimento
aziendale, business finance  finanza delle imprese, finanza aziendale;
business fluctuations  fluttuazioni economiche; business forecasts 
previsioni commerciali/ aziendali; business formation  formazione di nuove
imprese; business game  gioco aziendale, business gift  doni aziendali;
business hours  orario di esercizio/ larvoro/ apertura; busin ess income 
reddito aziendale/ di impresa; business indicator indicatore commerciale;
business intelligence  informazioni commerciali; business interruption
insurance  assicurazione contro l'interuzzione di esercizio: business
interruption policy  polizza di assicurazione contro l'interruzione di
esercizio; business investment  investimento aziendale; business investor 
azienda investitrice; business leader  leader capitano d'industria; business

loan  mutuo aziendale; business judgement  giudizio dell'imprenditore;
business law  diritto commerciale; business letter lettera commerciale,
lettera di affari; business lunch  colazione di lavoro; business machines 
macchine per ufficio; business manager  dirigente aziendale; business
market  merc ato delle aziende; business marketing research  ricerca di
mercato aziendale; business mathematics  computiseria; business meeting 
riunione di affari; business motive  movente commerciale; business name
 ragione sociale/ ditta; business objectives  obbiettivi aziendali; business
organization/ impresa/ azienda; business panic  panico economico; business
park  zona industriale, zona commerciale; business performance indicator 
indicatore di risultati aziendali; business plan  piano aziendale; business
policy  politica aziendale; business portofolio analysis  analisi di un
portafoglio prodotti; business premises locali aziendali; business premises
policy  polizza di assicurazione dei locali aziendali; business process
redesign  riprogettazion e di processi aziendali; business process
reengineering  reingegnenzzazione dei processi aziendali, business rate 
imposta locale sul’ attivita economica; business recovery ripresa dell’
attivita commerciale, business relations  relazioni d’affari; business reply
card  cartolina con risposta pagata; business report  rendiconto
d'esercizio; business reputation  reputazione commerciale; business risk 
rischio commerciale; business reply service  servizio di risposta affrancata;
business saving  risparmio d'impresa; business school  scuola aziendale/
facolta di economia e commercio; business sector  settore commerciale/
delle imprese; business sense  senso degli affari; business services  servizi
aziendali; business services industry  industria dei servizi aziendali: business
simulation  simulazione di gestione; business statistics  statistiche
commerciali; business strategies  strategie aziendali; business structures 
strutture aziendale; business studies  studi di amministrazione az iendale,
economia e commercio; business survey  indagine congiunturale; business
taxation  imposizione sugli utili d’ impresa; business taxes  imposte sugli
utili d'impresa; business tv  televisione aziendale; business transaction 
operazione commerc iale ; business travel/ business trip  viaggio d'affari;
business unionism  sindacalismo aziendale, business unit  (a) unita
operativa;  (b) impresa, azienda; business volume  volume di scambi,
volume delle contrattazioni; business wealth  patrimon io aziendale ;
business world  mondo degli affari; business year  esercizio;

[V] + business: to be away on/for business  essere via per affari, per lavoro;
to be in business  essere in affari/ avere tutto pronto/ essere pronto
all'azione; to be abro ad on business  essere all'estero in viaggio d'affari; to
be out of business  non essere piu in affari; to carry on business  svolgere
l'attivita economica; to do business  fare affari; to do a profitable business

 svolge una remunerativa attivita; to establish a business  fondare un'
impresa,to get down to business (idiom) mettersi al lavoro, cominciare a
lavorare; to give up business  ritirarsi dagli affari; to go into business 
entrare in affari/ avviare un'attivita economica/ darsi agli affari; to go out of
business  cessare l’attivita, ritirarsi da un'attivita, fallire, fare fallimento; to
keep business relations with sb  intrattenere relazioni d'affari con qn; to own
a thriving retail business  possedere una prospera impresa di commercio a l
dettaglio;to put sb out of business  fare fallire qn; to retire from business 
ritirarsi dall'attivita; to set up in business  mettersi in affari; to set up a
business  fondare un'impresa/ aprire un’ impresa; to transact business 
trattare affari, to travel on business  viaggiare per affari, viaggiare per
lavoro;

business + [V]: business is beginning to slow down  l'attivita sta
cominciando a rallentare; business is contracting  l'attivita economica e in
contrazione; business is expanding l'att tita economica e in espansione;
business is at a stand still  il mercato e fermo, si fanno pochi affari ; business
is slack  il mercato e debole/gli affari languono;

[ADJ.] + business : a nasty business  una brutta faccenda/ un brutto affare;

PHRASES : any other business  varie ed eventuali; business as usual 
l'attivita prosegue regolarmente: ”siamo aperti”, business – to- business
advertising  pubblicita da azienda a azienda; business -to-business marketing
 marketing da azienda a azienda; business -to-consumer marketing 
marketing da azienda a consumatore; business before pleasure  prima il
dovere, poi il piacere; Now we are in business !  Adesso ci siamo!; We are
not in the business of….  il nostro scopo non e di…/ noi non siamo qui per' … ;
to get down to business  cominciare a lavorare: Let' s get down to business
 bando alle chiacchiere!; mettiamoci sotto!; to have no business doing smth.
 non toccare a qcno., fare a qcosa; to know one's business  saper fare il
proprio mestiere , like nobody's business  a piu non posso/ all’ impazzata/ da
non dirsi, to mean business  fare sul serio, to send sb. about his business 
dire a qcno. di farsi i fatti suoi/ mandare q a quel paese; to get down to the
business of doing sth  accinger si a fare qc; to make it one’s business to do
sth.  farsi un dovere di fare qc.; It's no business of mine ( none of my
business )  non è affar mio; la cosa non mi rigurarda: Mind your own
business  bada ai latti tuoi; non t'impicciare! business is busin ess  gli
affari sono affari.

GLOSSARY

anomalous collocations -lexical patterns regarded as problematic in terms of
lexico -grammar. This is mainly due to the fact that they cannot be decoded
purely compositionally, nor encoded freely. The class of anomalous
collocations includes ill-formed collocations (e.g. at all, by and large, of
course, stay put, and thank you ), cranberry collocations (e.g. in retrospect, kith
and kin, on behalf of someone/something, short shrift, and to and fro ). defective
colloca tions (e.g. at least, a foregone conclusion, in effect, beg the question )
and phraseological collocations (e.g. in action, into action; on show and on
display; and to a – degree and to a – extent ).

antonymy – the semantic relation of opposition or unrelat edness between words

binomials or binominal expressions – dyads or conjoined pairs, unrestricted as
to word class, but normally occurring in fixed order as ‘irrever sible binomials’

clichés – fixed lexical patterns which resemble ossified collocations; an
illustration of the fact that the mutual expectancy of lexemes has become fixed
in certain contexts leadi ng to loss of meaning

coherence – refer ring to a variety of “conceptual resources ensuring that
meanings are related discernibly”, and it “has the r ole of underlying these
surface phenomena” (Hatim and Munday 2006: 68)

cohesion – a lingusitc concept which “subsumes the diverse relations which
transparently hold among the words, phrases and sentences of a text” (Hatim
and Munday 2006: 68)

colligation – an association of a word looked upon as a unique lexical item
rather than as a member of its class (Tognini Bonelli 1996: 74) with
grammatical categories (Sinclair 1988), or with a particular position in a
sentence or text (Kennedy 1991, Hoey 1997)

collocate – a word which occurs in close proximity to a word under
investigation is called a collocate of it

collocation – the occurrence of two or more words within a short space of each
other in a text. The usual measure of proximity which is called the span is a
maximum of four words intervening. Each citation or concordance line
exempli fies a particular word or phrase. This word or phrase is called the node .

It is normally presented with other words to the left and right and these are
called collocates (see span, node collocate)

componential analysis -a technique of semantic analysis examin ing the basic
meaning components of a word and allowing contrast with other t erms in the
same semantic field

connotations – subjective, second -order or peripheral mea nings, depend ing on
a relation between the word and the speaker/hearer, as well as on intertextual
norms

context – a term with two related meanings in modern linguistics .
a. In any continuous text, the words that come on either side of a word or
phrase sel ected for study constitute the context of that word or phrase. In
this sense, the context means the linguistic environment of any expression
under scrutiny. Sometimes, to distinguish this meaning from the other, the
term co-text is used;
b. The general, non -linguistic environment of any language activity can also
be called its context . Here, it means the socio -cultural background. Some
theories of language, (see Firth 1987), use context or ‘context of situation’,
to mean a level of language description where the limitless complexity of
the nonlinguistic environment is organized into linguistically relevant
categories.

context of situation – a term used to denote an institutionalized situation
characterized by specific participants and settings and either a spoken or written
mode or a combination of both.
Note : Some examples of contexts of situation are: a lecture, dining out at a
restaurant, Christmas festivities, etc. These are all institu tionalized contexts,
more precisely contexts of situation, as oppos ed to seeing a ghost, a random
context.

context -specific language use/discourse – terms used to refer to the ad hoc use of
language elicited by a given situation but not likely to occur again.
Note : Such pieces of language are ‘one -offs’ elicited by a specific contextual
variable such as the emotions or attitudes of the language -user.

conventional, conventionalized – terms used of expressions that conform to usage and
which, therefore, do not attract attention. Such expressions contrast with novel o nes
(e.g. a year ago vs. a grief ago or the cat swallows the canary vs. the canary swallows
the cat )

corpus – a collection of naturally -occurring language text, chosen to characterize a state
or variety of a language.

Note : In modern computational linguist ics, a corpus typically contains many
millions of words: this is because it is recognized that the creativity of natural
language leads to such immense variety of expression that it is difficult to
isolate the recurrent patterns that are the clues to the l exical structure of the
language.

co-text – words and phrases that are semantically and collocationally connected to the
expression under consideration and help d efine its meaning and function
Note : Typically, the co-text of an expression occurs within t he same utterance
or sentence, but co -text may also occur outside them as in preceding and
following utter ances or sentences. The co -text of a selected word or phrase
consists of the other words on either side of it. This is a more precise term than
conte xt or verbal context.

cultural concepts – abstract notions which map and construct the world -picture in a
culturally -specific way, their specificity being implemented at the cognitive level

cultural connotations – the interpretative relation existing bet ween linguistic signs and
symbols specific to any other cultural non -verbal code (stereotypes, prototypes, myths
and other such cultural entities)

cultural model – “a construction of reality that is created, shared and transmitted by
members o f a group” ( Bonvillain 2003: 3)

cultural semes -words and word -combinations denoting idioethnic realia
Note : They form part of the lexical meaning and reflect general knowledge
about the realia whether material, social or historical.

culture – “that complex whole wh ich includes knowledge, belief, art, morals, law,
custom and any other capabilities and habits acquired by man as a member of society”
(Tylor 1871 in Holli day, Hyde and Kullman 2006: 59)

denotation – a stylistically neutral and objective relation between a word and the
world, the basic or core meaning of the word, which is not deniable

discourse -language in use -naturally -occurr ing spoken or written language
Note : It is sometimes used as a very general form for language patterns above
the sentence. To so me people, discourse suggests the spoken form of the
language, and text the written form.

discourse communities – represented by people identify ing themselves as
members of a social group (family, neighbourhood, professional or ethn ic
affiliation, nation) and acquir ing common ways of viewing the world through
their interactions with o ther members of the same group
Note Common attitudes, beliefs and values are reflected in the way members
of the group use language.

epithet – “a stylistic device based on th e interplay of emotive and logical meaning in an
attributive word, phrase or even sentence used to characterize an object and pointing out
to the reader, and frequently imposing on him, some of the properties or features of the
object with the aim of givin g an individual perception and evaluation of these features
or properties” (Galperin 1977: 157)

euphemism – a word or phrase used to replace a word or expression by a con ventionally
more acceptable one

familiar collocations -combinations of words keep ing regular company with each other
Note : There are obvious overlaps here with types of fixed expression
categorized as stock phrase or metaphoric (e.g. vicious circle, innocent
bystander, amicable divorce ).

field – used in lexical semantics to refer to “th e particular activity, cultural feature,
social institution or topic for which a particular set of ideationally related lexical it ems
is often evolved or adapted. Each field has a specialized topic -related vocabulary, some
of which may turn up in other fie lds with different meanings ” (Carter 1998: 53)

fixed combinations – limited to formulae or fixed collocations that are semantically
transparent

fixed expression – a term adopted from Alexander (1978, 1979, also Carter 1987) , used
to denote different type s of phrasal lexemes , phraseological units , or multi -word
lexical items : frozen collocations, grammatically ill -formed collocations, proverbs,
routine formulae, sayings, similes

fixed phrases – variable phrases built round a slightly specialized meaning o f a word
that goes with a specific grammatical environm ent and in regular collocations;
combinations show ing co-selection and shared meaning (e.g. a piece / item of
information , a word of advice , the prospect of an agreement , a breath / drau ght/ gulp /
sniff / whiff of air , the erosion of confidence , a collapse / slump in demand , a stand
against the enemy , limitations on expenditure , a fall in the output , the rate of
unemployment , a (high) incidence of unemployment )

formulae – lexical patterns such as prover bs, slogans , quotations , gambits , and
closed -set turns (e.g. You have never had it so good and Shut your mouth ), which are in
no way anomalous with res pect to the language as a whole

fossilised or frozen metaphors -includ e ‘pure idioms’ such as skate on t hin ice or spill
the beans , which are anomalous only in the sense that they cann ot be manipulated
grammatically

free combinations – lexical patterns usually made up of a small number of words, each
of which is taken to hav e its usual independent meaning

Note: This implies that the meaning of the whole combination is derived by
general semantic combinatorial rules from the meanings of the words involved.

functional style – “a system of inter -related language means which serves a
definite aim in communicati on”, or, in other words, “the product of a certain
concrete task set by the sender of the m essage” (Galperin 1977: 32 – 33)

grammatical collocations – phrases consisting of a dominant word, which can
be a noun, an adjective, or a verb, and a preposition or a grammatical structure,
such as an infinitive or a claus e

harmonic phrase – “an element in the context of a modal verb that reinforce s
or disambiguates”
Note : The term harmonic is used “slightly more loosely, to cover all
combinations of a modal and a nother word or phrase which expresses the same
degree of modality” (Coates 1983: 45, in Croitoru 2009: 28 ).

hyponymy – 1 a paradigmatic relation of sense “which rests upon the
encapsulation in the hyponym of some syntagmatic modification of the sense of
the superordinate lexeme (Lyons 1977: 294 -295). 2 an inclusive sense relation
that exists between specific and general lexical items, the meaning of the
specific item being included in the meaning of the more general item. (Carter
1998: 20 -21)

ideational idioms – a type of idiom convey ing impressionistic representations
of aspects of the physical, social, and emotional w orlds of a language commu –
nity; a functional idiom

idiom – a group of two or more words which are chosen together in order to
produce a specific meaning or effect i n speech or writing
Note : The individual words which constitute idioms are not reliably
meaningful in themselves, because the whole idiom is required to produce the
meaning. Idioms overlap with collocations, because they both in volve the
selection of two or more words. At present, the line between them is not clear.
In principle, we call co -occurrences idioms if we interpret the co -occurrence as
giving a single unit of meaning. If we interpret the occurrence as the selection
of two related words, each of which keeps some meaning of its own, we call it
a collocation . Hence, hold talks, hold a meeting, hold an enquiry are
collocations, whereas hold sway, hold the whip hand are idioms.

idiom principle (IP) – one of the main princip les of the organization of
language according to which the choice of one word affects the choice of others
in its vicinity

Note : Collocation is one of the patterns of mutual choice, and idiom is another.

idiomaticity – used to denote indivisible lexical units whose components cannot
undergo any variations, or can var y only within definable limits

idiomatic – used to signify conformit y with the usage of a language

idioms of encoding – also termed ‘ phraseological peculiarities ’ or
‘phraseological idioms ’ (Makkai 1972: 56f.), involv ing collocational
preference restrictions

idioms of decoding – misleading ‘ lexical clusters ’ including examples such as
hot potato and fly off the handle

interpersonal idioms – a type of idiom which occurs in discourse with a
pragmatic function: greet ings, farewells, warnings, d isclaimers, etc.

jargon – used to denote words existing in almost every language and whose role
is to preserve secrecy within one or another social group

lexemic idioms – patterns considered p roblema tic from the lexico -grammatical
and semantic points of view, includ ing phrasal verbs , pure idioms such as
spill the beans, and opaque compounds such as forefinger and blackbird

lexical cohesion – “when two words in a text are semantically related in some
way”(Nunan (1993: 29)

lexical collocations – illustrations of “a relation of mutual expectancy or
habitual association” (Mc Arthur 1992, in Howard and Zé Amvela 2000: 114),
or of a partnership between ‘two “equal” lexical components, consisting of
nouns, adjectives, verbs, and adverbs and not normally containing prepos itions,
infinitives or clauses
Note : In lexical collocations , the occurrence of one word predicts the
possibility that another word will occur in the context, either in some syntactic
constr uction or across syntactic boundary.

lexical item – refer ring to single words, compoun ds, and multiword expres sions
Note : In the case of the last, only those items which are fixed expres sions, that
is, idioms, semi -idioms, literal idioms, and restricte d collocations qualify as
lexical items.

lexico -grammar -point ing to the arrangement of vocabulary in accordance with
the rules of gram mar so as to communicate coherent messages a t the micro –

level of discourse

lexis -the set of all wor d-forms existing in a language

lexical function – a general and abstract meaning, coup led with a deep syntactic
role, expressed in a variety of ways
Note : The use of lexical functions in the study of patterns has a great
contribution to the design of collocational diction aries, where a persistent
weakness is the failure of adequately specifying the semantic categories to
which the collocates belong.

lexical idioms – polymorphemic words or multi -word nouns, verbs, a .s.o.

logogens – used to refer to “tuned perceptual devic es that respond to sensory
and semantic input” (Morton 1979: 112 in Garman 1991: 279)

mental lexicon or semantic memory – used to refer to “the speakers’
encyclopedic knowledge of the world” (Gramley 2001: 7 -8)

node – used for the word whose lexical beh aviour is under examination in a
collocation

open -choice principle (OCP) – “a way of seeing language texts as the result of
a very large number of complex choices. At each point where a unit is
completed (word, phrase, clause), a large range of choices op ens up, and the
only restraint is grammat icalness” (Sinclair 1991: 109)

overlexicalization – a process implying the development of a specialized
lexicon and which is closely linked to the jargons developed by groups within a
society, but without being in opposi tion to socially dominant norms

phrasal lexemes – including phrases and idioms , i.e. lexical patterns which,
due to their semantic, lexico -grammatical, or pragmatic features which are
regarded as holistic units, ra ther than compositional strings

phrasal verbs – verb combinations in which each word contributes something,
semantically recognizabl e, to the meaning of the whole
Note : In some cases, it is mainly the verb, and in other cases it is mainly the
particle that is prevailing in stating the mea ning of the whole combination (e.g.
get / come along + with )

phrasemes – including frozen collocations or compounds where one element
has a me aning unique to the combination

phraseoloids – represented by restricted collocations where there is limited
paradigmatic variability

phraseological combinations – combinations with one component used in a
literal sense, and the other one used figuratively (e.g. meet a demand )

phraseological fusions (also called ‘ idioms’ ) -combinations which are
‘unmotivated’ (or semantically opaque) and, in general, structurally fixed (e.g.
spill the beans )
Note : They are indivisible and indecomposable phrases in whose general
meaning one cannot detect any connection with the words the phrase is made
up of.

phraseological unit – “a lexicalized , reproducible bilexemic or polylexemic
word group in common use, which has relative syntactic and semantic stability,
may be idiomatized, may carry connotations and may have an emphatic or
intensifying function in a text” (Gläser 1998: 125)
Note : Ioana Scherf (2006) defines the phraseologic unit (or frazeologism ) as
the fix word combination, whether idiomatic or not, which is made up of at
least two words, but functions as a single semantic unit in language.

phraseological unities – partiall y motivated patterns, whose meaning can be
seen as a metaphorical extension of some original neutral sense (e.g. blow off
steam – metaphorical extensi on from the technical meaning)
Note : Phraseological unities are stable phrases, which, like idioms, have a
meaning of their own, distinct from the meaning of the component elements,
although these are connected logically.

phrasicon -the whole inventory of idioms and phrases , both word -like and
sentence -like set expressions

pragmatemes – phrases which are tran sparent in meaning, but which are fixed in
the sense that by convention one wording is consistently chosen over other
possible alte rnatives in any given situation

professionalisms – words used in a definite trade, profession or occupation by
people connec ted by common inte rests both at work and at home

proverbs – patterns with an incongruity of meaning in the sense that the literal
meaning differs significantly from the context to which it refers

psycholinguistics – the study of normal and abnormal use o f language and

speech to gain a better understanding of how the human mind functions (Scovel
1998: 129)

pure idioms, semi -idioms, and literal idioms – categories referr ing to the
semantics of idioms
Note While pure idioms are completely non -literal, semi -idioms are only
partly so .

relexicalization – a process involv ing the creation of a dialectal semantics
which could be exemplified by a reversal of the normal meanings of words
Note Upright man and law used in criminal slang have totally different
meaning s: upright man might mean ‘leader of a gang of criminals’, whereas
law might mean ‘crime’.

qualia – a specific kind of semantic features which allow the speakers to derive
appropriate meanings of collocations compositionally

reference – the relatio nship hold ing between an expression and what that
expression stands for on particular occasions of its utterance

restricted collocation – a group of words co-occur ring with each other only in
limited ways ( e.g. shrug one's shoulders, addled eggs/brains )
Note: Restricted collocations include more fixed and closed partnerships
between lexical items (e.g. stark naked, pitch black, soft water, accept defeat )
and irreversible binominals (e.g. cash and carry, ups and downs, hit and miss,
assault and battery ).

sememic idioms – patterns problematic from a pragmatic or socio -cultural point
of view, being best represented by proverbs and formulaic greetings

semi -restricted collocations – lexical patterns in which the number of
substitutable items in different synta ctic slots is more determined (e.g. to fan a
riot/ discontent/disturbance/ hooliganism )

semi -transparent metaphors – patterns requiring some specialist knowledge in
order to be decoded successfully (e.g. grasp the nettle, on an even keel, the
pecking orde r, throw in the towel, under one’s belt )

sentence -like units – a type of units function ing pragmatically as sayings ,
catchphrases and conventional formulae (e.g. There’s no fool like an old fool,
The buck stops here, You don’t say so )

similes – patterns consisting of fossilized lexis, whose function is to emphasize
a feature or characteristic
Note : The tradition al structure is (as) + ADJ. + as + NG e.g. (as) clear as
crystal, dead as a doornail, as good as gold, as nice as pie, as right as rain,
(as) whi te as a sheet.

simple formulae – routine compositional strings which, besides being
syntagmatically fixed, have some special discoursal function or are iterative or
emphatic (e.g. alive and well, I'm sorry to say, not exactly, pick and choose, you
know )

slot-and-filler model – a model according to which the syntactic structures
forming a series of slots are filled wi th choices from the dictionary

sociolinguistic skills – including “the ability to perform various speech acts, the
ability to manage conver sational turns and topics, sensitivity to variation in
register and politeness and an understanding of how these aspects of language
vary according to social roles and settings” (Ranney 1992: 25)

sociolinguistics – the interdisciplinary study of the syst ematic correlative
variation in the structure of a language, on the one hand, and in the structure of
the community using it , on the other

span – the measurement, in words, of the co-text of a word selected for study
Note : A span of -4, +4 means that fou r words on either side of the node word
will be taken to be i ts relevant verbal environment

Speech Accommodation Theory (SAT ) (Giles 1973) – linguistic theory
according to which the speaker’s choice of a prestigious or a non -prestigious
speech style need not be the result of his/her social class position or the
formality/informality of the speaking context
Note : Rather, it could be mediated by interpersonal accommodation processes.

speech community – denoting all the people speak ing a single language and
sharing notions of what is same or di fferent in phonology or grammar
Note : This would include “any group of people wherever they might be and
however remote might be the possibility of their ever wanting to communicate
with each other, all using the same l anguage” (Spolsky 1998: 25).

speech network – concept developed by Milroy (1978, 1980 in Bonvillain
2003: 3) refer ring to the specific linkages of persons through shared varieties
and spe aking rules across communities
Note : The key element distinguishing speech communities from speech

networks is the actual interaction between members, which is present only in
the case of the latter.

stylistics – a discipline which “studies the expressive means of language taking
into account the modifications of meanings which various expressive means
undergo when they are used in different functional styles ” (Galperin 1977: 29)

synonymy – “a bilateral or symmetrical sense relation in which more than one
linguistic form can be said to have the same conceptual or proposit ional
meaning” (Carter 1998: 20 -21)

terms – lexical items used to denote new concepts that appear in the process of,
and as a result of, technical prog ress and development of science

text – used interchangeably with discourse
Note : Text is preferred for short pieces of language, extracts from a whole.

traditional combinations – including words that may combine only with
certain other words

transparent metaphors – institutionalized lexical patterns which may be
decoded successfully by means of the speake rs’ real -world knowledge (e.g.
alarm bells ring, behind someone's back, breathe life, into something, on
(some)one's doorstep, pack one’s bags )

unique (fixed, frozen) collocations – lexical patterns in which one (or
sometimes both) of the words cannot be replaced (e.g. addled eggs, to peel
onions )

unrestricted collocations -lexical items which are open to partnership with a
wide range of items (e.g. take a look/ a holiday/ a rest/ a letter/ time/ notice/ a
walk )

word -like units – a type of units which fu nction syntactically at or below the
level of the simple sentence (e.g. in the nick of time, to break one’s journey )

REFERENCES

1. Akhmanova, O. (ed.) 1974. Word -Combination: Theory and Method . Moscow:
Moscow University Press in A. P . Cowie (ed.) 1998 .
2. Aijmer, K. 1996. Conversational Routines in English: Convention and Creativity .
London: Longman.
3. Allerton, D. 1984. “Three (or four) l evels of word co -occurrence r estriction”
Lingua , 63, pp. 17 – 40.
4. Alexander, R. J. 1978/ 1979 . “Fixed expressions in Engl ish: A linguistic,
psycholinguistic, sociolin guistic and didactic study”, in Anglistik &
Englischunterricht , (Part 1), Vol. 6/1978, pp. 171 -188, (Part 2) Vol. 7/1979,
pp. 181 -202.
5. Alexander, R. J. 1984. “Fixed expressions in English: reference books and th e
teacher” English Language Teaching Journal 38/2: 127 -34.
6. Altenberg, B. 1998. “On the Phraseology of Spoken English: The Evidence of
Recurrent Word -Combinations” in A. P. Cowie (ed.) Phraseology. Theory,
Analysis, and Applications . Oxford: Clarendon Press , pp. 101 – 122.
7. Amosova, N.N. 1963. Osnovui anglijskoy frazeologii . Leningrad: U niversity Press
in R. Moon 1998 in A.P. Cowie (ed.) 1998 .
8. Anderman, G., and M. Rogers. (eds.) 1999. Word. Text. Translation . Clevedon,
Buffalo, Toronto, Sidney: Multilingual Ma tters Ltd.
9. Andone C. and G. Soare . 2002 . ‘Teaching Lexical Collocations to Business English
Students’ in the volume Abordări interculturale în teoria și practica limbilor .
Ploiești: Universitatea “Petrol și Gaze”, pp.18 -27.
10. Arnold, V. I .1959. Leksikologhia anglijskovo jazyka. Moskva in A.P. Cowie (ed.)
1998.
11. Arnold, V. I. 1973. The English Word . Moscow : Vuisshaya Shkol a in A.P. Cowie
(ed.) 1998.
12. Austin, J. L 1962. How to D o Things with Words . London: Oxford University
Press.
13. Avădanei, C. 2000 . Construcții idiomatice în limbile română și engleză . Iasi:
Editura Universității Alexandru Ioan Cuza.
14. Bahns, J., H. Burmeister, T. Vogel 1986. “The pragmatics of formulas in L2 learner
speech: use and development, Journal of Pragmatics , 10/6, pp. 693 -723.
15. Baker, M. 1992 . In Other Words. A Coursebook on Translation . London:
Routledge .
16. Ball, Christopher. 1987. “Lexis: The Vocabula ry of English” in W.F. Bolton and D.
Crystal (eds.) The English Language , London: Sphere Books, pp. 165 -197.
17. Blom, J. P. and Gumperz, J. 1972. “Social meaning in linguistic structures: Code –
switching in Norway”, in D. Hymes & J. Gumperz (eds.), Directions in
Sociolinguistics: The ethnography of communication , Holt, New York:
Rinehart and Winston, pp. 407 – 434.
18. Becker, J. D. 1975. The Phrasal Lexicon . (Artificial Intelligence Report, No. 28:
Report No. 3081), Cambridge, Mass: Bolt, Beranek, and Newman.
19. Beaug rande, R. de. 1991. Linguistic Theory: The d iscourse of fundamental w orks.
London: Longman.

20. Beaupre, M. 1986. Interpreting Bilingual Legislation. Toronto: Carswell.
21. Bejoint, H. 1994. Tradition and Innovation in Modern English Dictionaries .
Oxford: Oxford U niversity Press.
22. Bell, R. 1991. Translation and Translating: Theory and Practice . London and New
York: Longman.
23. Bens on, M. 1985. “Collocations and i dioms” in R. Ilson, (ed.) 1985. Dictionaries,
Lexicography and Language Learning . Oxford: Pergamon Press, p p. 61- 68.
24. Benson, M. 1989. “The Structure of Collocational Dictionary” in International
Journal of Lexicography . 2/I: 1 -14.
25. Benson, M., E. Benson and R. Ilson. 1986a. Lexicographic Description of English .
Netherlands: John Benjamins Publishing Company.
26. Benson, M., E. Benson and R. Ilson. 1986b. The BBI Combinatory Dictionary of
English: A Guide to Word Combinations. Amsterdam: John Benjamins.
27. Berrutto, G. 1974. La sociolinguistica . Bologna: Zanichelli.
28. Berrutto, G . 1980. La variabilita sociale della lingu a. Torino: Loescher.
29. Biber et al. 1998. Corpus Linguistics: Investigating Language, Structure and Use .
Cambridge: Cambridge University Press.
30. Bidu -Vrănceanu A., C. Călărașu, L. Ionescu Ruxăndoiu, M. Mancas, G. Pană
Dindelegan 2005. Dicționar de științe ale limbii . București: Editura Nemira.
31. Bidu -Vrănceanu, A. 2007. Lexicul specializat în mișcare. De la dicționare la tetxte .
București: Editura Universit ății din București.
32. Bloomfield, L. 1933 . Language . New York: Holt , in D. Kenny . 2000 and in N.
Bonvillain, 2003.
33. Bolander, M. 1989. “Prefabs, patterns and rules in interaction? Formulaic speech in
adult learners’ L2 Swedish” in K. Hyltenstam and L. Obler (eds).
Bilingualism across Life – Span . Cambridge: Cambridge University Press, pp.
73-86.
34. Bollinger, D. 1976. “Meaning and Memory” in Forum Linguisticum . I/I:1 -14.
35. Bolton, K., and D. Crystal . 1987. Lexis: The Vocabulary of English . Oxford:
Oxford Universi ty Press.
36. Bonvillain, N. 2003. Language and Communication. The Meaning of Messages.
New Jersey: Prentice Hall.
37. Brake, T., D. Medina -Walker and T. Walker. 1995. Doing business internationally:
The guide to cross -cultural success. Irwin: Burr Ridge.
38. Bresna n J. 1982. “The Passive in Lexical Theory” in J. Bresnan (ed.). The Mental
Representation of Grammatical Relations , Cambridge, Massachusetts: MIT
Press.
39. Bryne, D. 1971. The Attraction P aradigm . New York: Academic Press.
40. Cabré, M. T. 2000. “Sur la represent ation mentale: bases pour une tentati ve de
modelisation” in A. Bidu -Vrănceanu 2007. Lexicul specializat în mișcare. De
la dicționare la tetxte . București: Editura Universității din București.
41. Cacciari C. and S. Glucksberg 1995. “Imaging idiomatic expressions: Literal or
Figurative Meanings” in Evaraert, M., E -J., Van der Li nden, A. Schenk, and
R. Schreuder (eds.) 1995. Idioms: Structural and Psychological Perspectives ,
Hillsdale, New Jersey: Lawrence Erlbaum Associates, pp. 43 -56.

42. Cacciari, C and P. Tabossi 1988. “The Comprehension of Idioms” in Journal of
Memory and Languag e. 27, 668 -683.
43. Carter, R. 1987. Vocabulary . London: Allen and Unwin.
44. Carter, R. 1998. Vocabulary. Applied Linguistic Perspectives . London and New
York: Routledge.
45. Catford, J. C. 1965. A Linguistic Theory of Translation: A n Essay on Applied
Linguistics. London: Oxford University Press. in D. Kenny. 2000.
46. Čermak, F. 1994. “Idiomatics” in P.A. Luelsdorff (ed). Prague School of Structural
and Functional Linguistics ”. Amsterdam and Philadelphia: John Benjamins,
pp. 185 -95
47. Chafe, W. L. 1968. “Idiomaticity as an anomaly in the Chomskyan paradigm” in
Foundations of Language 4, pp. 109 –127.
48. Chatto and Windus 1929. The Slang Dictionary . London in L., Levițchi 1970.
49. Chernuisheva, I. I. 1964. Die Phraseologie der gegenwärtigen deutschen Sprache .
Moscow : Vuisshaya Shk ola in A. P Cowie (ed) 1998.
50. Chesterman , A. 1997. Memes of Translation. The Spread of Ideas in Translation
Theory . Amsterdam/Philadelphia: Benjamins.
51. Chițoran, D. 1973a. “Colocație și sens. O contribuție la studiul sensului elementelor
lexicale” in Studi i si cercetări lingvisitce . XVI, 5, pp. 655 -667.
52. Chițoran, D. 1973b. Elements of English Structural Semantics . București: Editura
Didactică și Pedagogică.
53. Chomsky, N. 1965. Aspects of the Theory of Syntax. Cambridge: MA: MIT Press.
in D. Kenny, 2000.
54. Coate s, J. 1983. The Semantics of Modal Auxiliaries . London and Canberra: Croom
Helm Ltd.
55. Cohen, J. 1979. “The s emantics of metaphor” in A. Ortony (ed.) Metaphor and
Thought . Cambridge: Cambridge University Press.
56. Colțun, G. 2000. Frazeologia limbii române . Chi șinău: Editura Arc.
57. Constantinescu, I. 1976. “Preliminarii la o analiză contrastivă a colocațiilor” in
Studii si cercetări lingvisitce . XXVII, 3, pp . 267 -278.
58. Cook, G. 2003. Applied Linguistics Oxford: Oxford University Press.
59. Cop, M. 1990. “The Function s of Collocations in the Dictionary” in Magay, T., J.
Zigàny (eds.) Buda Lex ’88. Proceedings – Papers from the 3rd International
Euralex Congress. Budapest, 4 -9 September 1988. Budapest: Akadèmiai.
60. Correia, R. 2003. “Translation of EU Legal Texts” in A. T ossi (ed.) Crossing
Barriers and Bridging Cultures. The Challenges of Multilingual Translation
for the European Union . Clevendon: Multilingual Matters.
61. Coșeriu, E. 2000. Lecții de lingvistică generală , Chișinău: Editura ARC.
62. Coulmas, F. 1979. “On the sociolinguistic relevance of routine formulae”. Journal
of Pragmatics , 3, pp. 239 -66.
63. Coulmas, F. (ed). 1981. Conversational Routine . The Hague: Mouton.
64. Coulthard, M. 1985. An Introduction to Discourse A nalysis . Harlow: Longman.
65. Cozma, M. 2006. Translating Legal -Administrative Discourse . Timișoara: Editura
Universității de Vest.
66. Cowan, R.J. 1974. Lexical and Syntactical Research for the Design of EFL Re ading
Materials . TESOL, 8. 1978.

67. Cowie A. P. 1981. “The treatment of collocations and idioms in learner
dictionaries” Applied Linguistics 2/3 223 -35.
68. Cowie, A.P. 1986. “Collocational Dictionaries – a Comparative View” in M.
Murphy (ed.) Proceedings of the Fourth Joint Anglo -Soviet Seminar on
English Studies. London: The British Council. pp. 61 -9.
69. Cowie, A. P. 1988. “Stable and Creative Aspects of Vocabulary Use” in R. Carter,
and M. J. Mc. Carthy (eds.). Vocabulary and Language Teaching . London:
Longman, pp .126-37.
70. Cowie, A. P. 1992. “Multiword lexical units and communicative language
teaching” in P., Arnaud and H., Bejoint (eds.) Vocabulary and Applied
Linguistics . Basingstoke: Macmillan.
71. Cowie, A. P. 1994. “Phraseology” in R. E, Asher and J. Simpson (eds. ). The
Encyclopedia of Language and Linguistics . Oxford: Pergamon Press.
72. Cowie, A. P. (ed) 1998. Phraseology. Theory, Analysis, and Applications . Oxford:
Clarendon Press.
73. Cowie, A. P. 2002. “Examples and Collocations in the French Dictionnaire de la
Langue ” in M.H. Corréard (ed). Lexicography and Natural Language
Processing . UK. Euralex.
74. Cowie, A. P. and P. Howarth. 1996. “Phraseological Competence and Written
Proficiency” in G. M. Blue and R. Mitchell (eds.) Language and Education .
British Studies in Appli ed Linguistics II, Clevendon: Multilingual Matters.
pp 80 -93.
75. Cowie, A. P. and R. Mackin, (eds.) 1975. Oxford Dictionary of Current Idiomatic
English . Vol. 1. London: Oxford University Press.
76. Cowie, A. P., R. Mackin, and I. R., McCaig (eds.) 1983. Oxford D ictionary of
Current Idiomatic English . Vol. 2. Oxford: Oxford University Press.
77. Croitoru, E. 1996. Interpretation and Translation . Galați: Porto Franco .
78. Croitoru E. 2004 . English through Translations. Interpretation and Translation –
Oriented Text Analysi s. Editura Fundației Universitare “Dunărea de Jos”
Galați.
79. Croitoru E. 2005. Traps in Translating Literary Texts. în the volume of yhe
Intenational Conference Tradition, Modernity and Post -Modernity . Iași:
Universitas XXI, pp. 406 -416.
80. Croitoru, E. 2009. “ From Collocations to Harmonic Phrases in Translation” in A.
G. Romedea (ed) Review of Interstud Interdisciplinary Studies of
Contemporary Discursive Forms. Bacău: Alma Mater, pp. 25 -32.
81. Croitoru, E. and A. M., Dumitrașcu. 2006a. “Collocations and Colligati ons in
Translating and Interpreting” in Annales Universitas Apulensis , Series
Philologica, tom 2, Alba Iulia: Editura Universității „1 decembrie 1918”. pp.
293-300.
82. Croitoru, E. and A. M., Dumitrașcu. 2006b. “Collocations and Colligations in the
Specialize d Discourse” in the volume Specialised Discourse. Theory and
Practice . Galați: Editura Europlus, pp.103 -113.
83. Cruse, D. A. 1986. Lexical Semantics New York: Cambridge University Press.
84. Crystal, D. 1992. An Encyclopedic Dictionary of Language and Languages .
Oxford: Blackwell.

85. Crystal, D. 1996. The Cambridge Encyclopedia of the English Language .
Cambridge University Press.
86. Crystal, D. and W. F. Bolton (eds.) 1993 . The English L anguage . London: Penguin
Books.
87. Dardano, M. and P. Trifone. 1995. Grammatica ital iana. Bologna: Zanichelli
Editore.
88. Dardano, M. and P. Trifone. 2006. Grammatic a italiana con nozioni di linguistica.
Bologna: Zanichelli Editore.
89. De Mendoza Ibáňez, F.J.R, 2004. Annual review of Cognitive Linguistics. Vol. 2,
Amsterdam/Philadelphia: John Benjamins Publishing.
90. Deveci, T. 2004. “Why and how to teach collocations” in English Teaching Forum .
April 2004.
91. Dirk Geeraerts 1995. “Specializa tion and reinterpretation of idioms” in Evaraert,
M., E -J., Van der Linden, A. Schenk, and R. Schreuder (eds.) Idioms:
Structural and Psychological Perspectives , Hillsdale, New Jersey: Lawrence
Erlbaum Associates, pp. 57 – 74.
92. Dollerup, C. 2006. Basics of T ranslation Studies . Iași: Institutul European.
93. Drazdauskiene, M. L. 1981. “On stereotypes in conversation, their meaning and
significance” in F. Coulmas. Conversational Routine . The Hague: Mouton.
94. Drew, P. and E. Holt 1995. “Idiomatic expressions and their role in the organ ization
of topic transition in conversation”, in Evaraert, M., E -J., Van der Linden, A.
Schenk, and R. Schreuder (eds.) 1995. Idioms: Structural and Psychological
Perspectives , Hillsdale, New Jersey: Lawrence Erlbaum Associates, pp. 117 –
32.
95. Dumitrașcu, A. M. 2006. “Translating Metaphors in Newspaper Headlines” in
Discurs specializat. Teorie și practică , Galați: Editura Europlus, pp.114 -123.
96. Dumitrașcu, A. M. 2007. “ Patterns of Idiomacity” in the online Journal of
Language and Linguistic Studies , vol. 3, nr. 1 /2007, pp. 1 -10.
97. Dumitrașcu, A. M. 2008a. “On the Semantics of Computer Collocations”, in Caiet
de semiotică, nr. 19, Editura Universității de Ve st, pp. 51 -59.
98. Dumitrașcu, A. M. 2008b. “ Equivalence and Non -Equivalence in Translating Do
and Make Colloca tions into Romanian and Italian”, in volumul Comunicare
interculturală și literatură , no 4. Galați: Editura Europlus. pp. 132 -140.
99. Dumitrașcu, A. M. 2009a. “On teaching and learning English collocations” in L.
Frențiu (ed) Romanian Journal of English Studi es. Timișoara: Editura
Universității de Vest, pp. 387 -396.
100. Dumitrașcu, A. M. 2009b. “On the stylistics of collocations” in Mélanges
francophones. Formes textuelles de la communication. De la production à la
réception , fasc. XXIII, no. 3. vol III, 1, pp. 79 -88.
101. Dumitrașcu , A. M. 2009c “Collocationally -marked cultural i dentity” in Identity,
Alterity, Hybridity , Galați University Press, pp. 461 -475.
102. Dumitrașcu, A. M. 2009d. “On translating collocational patterns shared by general,
literary and specialized cont exts”, in The Annals of Dunărea de Jos
University of Galați, Fasc. XXIV, year II, no. I. 2 Common Lexis.
Specialized Lexis . Galați: Europlus, pp. 360 -367.

103. Dumitrașcu, A. M. and R.E, R. Kania 2008. ” On some Criminal Collocations” in
Studii de traducere: Re trospectiv ă și Perspective , vol 2, Galați: Galați
Universit y Press, pp.105 – 117.
104. Dumitraș cu, A. M. and R.E, R. Kania 2010 . ”Is law the same for everybody?”, in
Challenges in Translation , H. Parlog, L. Frențiu and L. Fraț ilă (eds .), West
University of , Timi șoara, 2010, pp. 135 -152.
105. Dumitrescu, F. 1958. Locuțiunile verbale în limba română . Bucureș ti: Editura
Academiei.
106. Dumitriu C. 1977. Gramatica limbii române explicată. Morfologia . IașiȘ Junimea.
107. Dumistrăcel, S. 1980. Lexic românesc. Cuvinte, metafore, expr esii. București:
Editura Științifică și Enciclopedică.
108. Eades, D. 2004. ‘Beyond difference and d omination? Intercultural communication
in legal contexts’ in F. Kiesling and S.C.B., Paulston (eds.) 2005.
Intercultural Discourse and Communication. Essential R eadings . UK:
Blackwell Publishing.
109. Evaraert, M., E -J., Van der Linden, A. Schenk, and R. Schreuder (eds.) 1995.
Idioms: Structural and Psychological Perspectives , Hillsdale, New Jersey:
Lawrence Erlbaum Associates.
110. Fennel,B., C. Herndl and C. Miller 1987. “Mapping Discourse communities”
College Composition and Communication , 152, pp. 80-89.
111. Fernando, C. 1996. Idioms and Idiomacity . Oxford: Oxford University Press.
112. Fernando, C. and R.H. Flavell. 1981. On idiom: Critical views ad perspectives .
Exeter: Univer sity of Exeter Press.
113. Fillmore, C. J., P. Kay, and M. C. O’Connor. 1988. “Regularity and idiomacity in
grammatical constructions: the case of let alone ” in Language , 64, 3: pp.
501-38.
114. Firth, J. R 1957. Papers in Linguistics 1934 -1951. London, New York and Toronto:
Oxford UP in D. Kenny, 2000. Lexis and Creativity in Translation. A corpus –
Based Translation Study . Amsterdam: St. Jerome Publishing.
115. Firth, J. R 1968a. Selected Papers of J.R. Firth 1952 -1959. edited by F. R Palmer,
London and Harlow: Longmans, Green and Co Ltd in D. Kenny. 2000 Lexis
and Creativity in Translation. A corpus -Based Translation Study.
Amsterdam: St. Jerome Publishing.
116. Firth, J.R. 1968b. “A synopsis of linguistic theory”. in F. Palmer (ed.) Selected
Papers of J. R. Firth 1952 -1959 . London: Longman, pp. 168 -205.
117. Fontenelle, T. 1998. “Discovering Significant Lexical Functions in Dictionary
Entries” in A.P. Cowie (ed) Phraseology. Theory, Analysis, and Applications .
Oxford: Clarendon Press. pp. 189 – 207.
118. Forster, K. I. 1976. “Accessin g the mental lexicon” in R. J. Wales and E. Walker
(eds) New Approaches to Language M echanisms . Amsterdam: North –
Holland.
119. Francesconi, A. 2008. I falsi amici. Un confronto contrastivo spagnolo -italiano ,
Chieti : Solfanelli .
120. Francis, G. 1993 “A corpus -driven approach to grammar – principles, methods and
examples” in M. Baker, G. Francis, and E., Tognini -Bonelli, (eds.) Text and

Technology: In Honour of John Sinclair . Amsterdam: Benjamins, pp. 137 –
156.
121. Francis, G. 1995 “Corpus -driven grammar and its relevance to the learning of
English in a cross -cultural situation” in A. Pakir (ed.) English in education:
Multicultural perspectives . Singapore: Unipress.
122. Fraser, B. 1970. “Idioms within a transformational grammar” in Foundations of
Language . 6, pp. 22 -42.
123. Frățilă, L. 2006. On Language and Ecology . Timișoara: Editura Universității de
Vest.
124. Gairns, R. and S. Redman 1991. Working with Words . A Guide to Teaching and
Learning Vocabulary . 6th edn. Cambridge: Cambridge University Press.
125. Galperin, I. R. 1977. Stylistic s. Moscow Higher School.
126. Garman, M. 1991. Psycholinguistics . Cambridge: Cambridge University Press.
127. Garzone, G. 2000. Legal translation and functionalist approaches: A Contradiction
in Terms? http:/ /www.tradulex.org/Actes2000/Garzone . pdf
128. Garzone, G. 2003. Domain -Specific English and Language Mediation in
Professional and Institutional Settings . Milano: Archipelago Edizioni.
129. Gazdar, G., E. Klein, G. K., Pullum, and I., Sag 1985. Generalized Phrase
Structure Grammar . Oxford: Basil Blackwell.
130. Geeraerts, D. 1989. “Types of meaning in idioms” in M. Everaert and E -J van der
Linden (eds.) Proceedings of the First Tilburg Workshop on I dioms , Tilburg:
ITK.
131. Geerarerts, D. 1995. “Types of Meaning in Idioms” in M. Evaraert and E. van der
Linden (eds.) Proceedings of the First Tilburg Workshop on Idioms . ITK
Tilburg.
132. Geeraerts, D., S. Grondelaers, P., Bakema 1994. “The structure of lexical variation.
Meaning, naming and context”in Cognitve Linguistic Reasearch , vol. 5,
Berlin: Mouton de Grutyer.
133. Gernsbacher, M. A. and M. Schlesinger , 1997. “The proposed role of suppression
in simultaneous interpretation”. in Interpreting 2 (1-2), pp. 119 -140.
134. Gibbs, R. 1986. “Skating on thin ice: Literal meaning and understanding idioms in
conversation” in Discourse Processes , 7, pp. 17 -30.
135. Gibbs R. 1994. The Poetics of the Mind . Cambridge: Cambridge University Press.
136. Gibbs, R. W. 1995. “Idiomacity and human cognition” in Evaraert, M., E -J., Van
der Linden, A. Schenk, and R. Schreu der (eds.) 1995. Idioms: Structural and
Psychological Perspectives , Hillsdale, New Jersey: Lawrence Erlbaum
Associates, pp. 97 – 116.
137. Giles, H. 1973. “Accent mobility: A model and some data.” Anthropological
Linguistics, 15. pp. 87–105.
138. Giles, H and N. Coup land. 1991. Language : Contexts and Consequences . Milton
Keynes: Open University Press.
139. Ginzburg R. S., S.S., Khidekel, G. Y., Knyazeva and A. A. Sankin, 1979. A Course
on Modern English Lexicology . 2nd edn. Moscow: Vuisshayal.
140. Gläser R. 1986a. “A Plea f or Phraseo -Stylistics” in D. Kastovsky and A. Szwedek
(eds.) Linguistics across Historical and Geographical Boundaries. In Honour

of Jacek Fisiak , and Linguistic Theory and Historical Linguistics . Berlin:
Mouton de Gruyter, pp. 41 -52.
141. Gläser, R. 1986b. Phraseologie der englischen Sprache . Leipzig: VEB Verlag
Enzyklopädie.
142. Gläser R., 1988. “The Grading of Idiomacity as a Presupposition for a Taxonomy
of Idioms” in W. Hüllen and R. Schulze (eds.) Understanding the Lexicon:
Meaning, Sense and Word Knowledge in Lexical Semantics . Tübingen: Max
Niemeyer, pp. 264 -279.
143. Gläser, R. 1998. “The Stylistic Potential of Phraseological Units in the light of
Genre Analysis” in A. P. Cowie (ed) Phraseology. Theory, Analysis, and
Applications . Oxford: Clarendon Press, pp. 125- 143.
144. Glyyn, D. 2004. ‘Constructions at the crossroads’. in De Mendo za Ibáňez, F. J. R,
2004. Annual review of Cognitive Linguistics, volume 2, Amsterdam/
Philadelphia: John Benjamins Publishing.
145. Godman, A. and R. Veltman. 1990. “Language development and the translation of
scientific texts”. Babel, 36, 4, pp. 193 -211.
146. Gramley, S. 2001. The Vocabulary of World English . London: Arnold.
147. Granger, S. 1998. “Prefabricated patterns in Advanced EFL Writing: Collocations
and Formulae” in A.P., Cowie (ed). Phraseology. Theory, Analysis, and
Applications . Oxford: Clarendon Press. p p 145-160.
148. Granger, S. and Meunier F. (eds.) 2009 . Phraseology. An Interdisciplinary
Perspective. Amsterdam/ Philadelphia : John Benjamins Publishing Company.
149. Granger, S. and M. Paquot 2009. “ Disintangling the phraseological web ” in
Granger, S. and F. Meuni er (eds.) Phraseology. An Interdisciplinary
Perspective. Amsterdam/ Philadelphia : John Benjamins Publishing
Company, pp. 27- 49.
150. Greenbaum, S. 1970. Verb -Intensifier Collocations in English . Mouton: Hague.
151. Greenough, J. B. and C. L. Kitteridge 1929. Words and T heir Ways in English
Speech . New York in L., Levițchi 1970.
152. Gross, M. 1981. “Les bases empririques de la notion de prédicat sémantique” in
Langage , 63: pp. 7 -52.
153. Gupta, A. and Ferguson, J. (1992) ‘Beyond ‘Culture’: Space, Identity, and the
Politics of Difference’. Cultural Anthropology 7 (1), pp. 6-23
154. Gutwinsky, W. 1976. Cohesion in Literary T exts. The Hague: Mouton in R. Carter
1998.
155. Halliday, MAK. 1961. “Categories of the Theory of Grammar”. Word 17/3” pp.
241-292.
156. Halliday MAK. 1966. “Lexis as a linguistic level” in C. E. Bazel, et al (eds.) In
Memory of J. R. Firth . London: Longman , pp. 148 -162
157. Halliday, M.A.K. 1978a. “Antilanguages” in Language as Social Semiotic. London:
Edward Arnold, pp. 164 -182.
158. Halliday, M.A.K. 1978b. Language as Social Semiotic . London: Edward Arnold.
159. Halliday, M.A.K. 1985. Spoken and Written Language . Oxford University Press.
160. Halliday, MAK 1985. An Introduction to Functional Grammar . London: Edward
Arnold.

161. Halliday, M.A.K. 1994. An Introduction to Functional Grammar . London: Edward
Arnold.
162. Halliday, M.A.K. 2002. Linguistic Studies of Text and Discourse. London:
Continuum.
163. Halliday, M.A.K. and R. Hasan. 1976. Cohesion in English . London: Longman.
164. Halliday, M.A. K., and R. Hasan . 1989. Language, Context and Text: Aspects of
Language in a Social -Semiotic Perspective . Oxf ord: Oxford University Press.
165. Hanks , P. 1987. “Definitions and Explanations” in J. Sinclair (ed.) Looking Up : An
Account of the Cobuild Project in Lexical Computing . London: Harper
Collins. pp. 116 -136.
166. Hansen G., K. Malmkjaer and D. Gile 2004. Claims, C hanges and Challenges in
Translation Studies . Amsterdam, Philadelphia: Benjamins.
167. Harris, Z. 1991. A Theory of Language and Information . Oxford: Oxford University
Press.
168. Hasan, R. 1987/1996. “The grammarian’s dream: lexis as most delicate grammar”
in M.A. K. Halliday and R. Fawcett (eds.) New Developments in Systemic
Linguistics , vol. 1 Theory and Descriptions . London: Pinter (1987); and in R.
Hasan, C. Cloran, D. Butt and G. Williams (eds) Ways of S aying: Ways of
Meaning . London: Cassel, pp. 73 -103.
169. Hatim , B. and L. I. Mason. 1992. Discourse and the Translator. London and New
York: Longman.
170. Hatim, B. and J. Munday. 2006. Translation. An Advanced Resource Book. London
and New York: Routledge
171. Hausmann, F. J. 1979. “Un dictionnaire des collocations est -il pos sible?” in
Travaux de linguistique et littérature 17 (1), pp. 187 -195
172. Hausmann, F. J. 1988. ”O diccionario de colocacións. Criterios de organización” in
X. Ferro Ruibal (coord.), Actas do I Coloquio Galego de Fraseoloxía ,
Santiago de Compostela: Centro Ra món Piñeiro, pp. 63 -82.
173. Hawkes , T. 1984. Metaphor . London: Routledge.
174. Healey, A. 1968. “English Idioms” in KIVUNG (Journal of the Linguistic Society
of the Un iversity of Papua New Guinea) , pp. 71 -108.
175. Heltai, P. 2004. “Ready -made language and translation” , in G. Hansen, K.
Malmkjaer and D. Gile (eds.) Claims, Changes and Challenges in
Translation Studies , Amsterdam /Philadelphia: John Benjamins Publishing
Company, pp. 51 -71.
176. Herbots, J. H. 1987. “La traduction juridique. Une point de vue belge” in Le cahie r
de droit . no. 28. pp. 813 -844.
177. Hervey S. and I. Higgins. 1992. Thinking Translation . London/New York:
Routledge.
178. Hervey, S., I. Higgins and L. M. Haywood. 1995. Thinking Spanish Translation .
London: Routledge.
179. Hezberg, B. 1986 . “The politics of discour se communities”, paper presented at the
CCC Convention, New Orleans: La March.
180. Hill, J. 2000. ‘Revising priorities: from grammatical failure to collocational
success’ in Lewis, M. (ed.) Teaching Collocation. Further Developments in
the Lexical Approach . Boston: Thomson Heinle, pp. 47 -69.

181. Hockett, C. F. 1958. A Course in Modern Linguistics . New York: Macmillan. in R.
Moon, 1998a.
182. Hoey, M. 1997. “From concordance to text structure: new uses for computer
corpora”. in P. Melia and B. Lewandowska -Tomaszczyk (eds .). Proceedings
of the Practical Application of Linguistic Corpora Conference , Poland: Lodz
University Press, pp. 2 -23.
183. Hoey, M. 2005. Lexical Priming . A New Theory of Words and Meaning . London
and New York: Routledge.
184. Hofstede, G. 1984. Culture’s C onseque nces: International Differences in Work –
Related V alues. Newbury Park, CA: Sage.
185. Hofstede, G. 1991. Cultures and Organizations. Software of the Mind. London:
McGraw -Hill.
186. Holliday, A . 1999. “Small cultures.” in Applied Linguistics, 20 (2), pp. 237 -264.
187. Holliday, A., M. Hyde, J. and M. Kullman. 200 4. Intercultural Communication. An
Advanced Resource Book . New York: Routledge.
188. Hollinger, A. 2005. A Lexicological Approach to Financial and Business English .
București: Editura Universitară.
189. House, J. 1977. A Mo del fo r Translation Quality Assessment. Tubingen: Gunter Narr.
190. Hunston, S. and G. Francis 2000. Pattern Grammar. A Corpus -driven Approach
to the Lexical Grammar of English. Amsterdam: John Benjamins.
191. Howard, J. 1988. Words and their Meaning . London and N ew York: Longman.
192. Howard, J. and E. Zé Amvela 2000. Words, Meanings and Vocabulary. An
Introduction to Modern English Lexicology . London and New York: Cassell.
193. Howarth, P. A. 1996a. Phraseology in English Academic Writing . Tubingen: Max
Niemeyer.
194. Howarth, P. A. 1996b. “Phraseology in English Academic Writing: Some
Implications for Language Learning and Dictionary Making”.
Lexicographica , series maior 75, Tübingen: Max Niemeyer.
195. Hristea, T. 1984. Sinteze de limba română . București: Editura Albatros.
196. Hulban, H. 2002. Syntheses in English Lexicology and Semantics . Iaș i: Editura
Spanda.
197. Hunston, S. and G. Francis 2000. Pattern Grammar. A Corpus -Driven Approach to
the Lexical G rammar of English . Amsterdam/ Philadelphia: John Benjamins
Publishing.
198. Hunston, S., G. Francis and E. Manning 1997. “Grammar and vocabulary. Showing
the connections” English Language Teaching Journal vol. 51. no. 3, pp. 208 –
216 in Carter , R. 1998 .
199. Hymes, D. 1974. Foundations of Sociolinguistics . Philadelphia: University of
Pennsylvani a Pres s in N. Bonvillain, 2003. Language and Communication.
The Meaning of Messages. New Jersey: Prentice Hall.
200. Ionescu, D. C. 2000. Translation Theory and Practice , București: Editura Universal
Dalsi.
201. Iordan, I. 1975. Stilistica limbii române , București: Editura Științifică.
202. Janssen, A.J.M.T. 2003. “Monosemy vs. Polysemy” in H. Cuyckens, R. Dirven and
J. R. Taylor (eds.). Cognitive Approaches to Lexical Semantics . Berlin and
New York: Mouton de Gruyter, pp. 93 – 122.

203. Jantunen, J.H. 2001. “Synonymity and lexical simplification in translations: a
corpus -based approach”. in Across Languages and Cultures , 2(1), 97 -112, in
J.H. Jantunen, “Untypical Patterns in Translation ”, in A. M auranen and P.
Kujamäki (eds.) 2004. Translation Universals. Amsterdam / Philadelphia:
John Benjamins Publishing Company.
204. Jantunen, J.H. 2004. “Untypical Patterns in Translation” , in A. Mauranen and P.
Kujamäki (eds.), Translation Universals . Ams terdam/ Philadelphia: John
Benjamins Publishing Company.
205. Katamba , F. 1993. Morphology . England: Palgrave.
206. Katan, D. 2004. Translating Cultures. An Introduction for Translators, Interpreters
and Mediators . UK: St. Jerome Publishing.
207. Katz, J. J. 1973. “Compo sitionality, idiomacity and lexical substitution” in R.
Moon, 1998a. Fixed Expressions and Idioms in English. A Corpus -Based
Approach . Oxford: Clarendon Press.
208. Katz, J. J. and P. M. Postal 1963. “Semantic interpretation of idioms and sentences
containing t hem” MIT Research Laboratory of Electronics Quarterly
Progress Report , 70, pp. 275 -82. in R. Moon, 1998a. Fixed Expressions and
Idioms in English. A Corpus -Based Approach . Oxford: Clarendon Press.
209. Kavić, A. 2005. “N -of-N collocations describing groups of a nimals: An internet –
based analysis” in H., Pârlog (ed) British and American Studies . Timișoara:
Editura Universității de Vest, pp. 337 -348.
210. Kennedy, G. 1991. “Between and through: the company they keep and the
functions they serve”. in K. Aijmer and B. Al tenberg (eds.) English Corpus
Linguistics. Studies on Honour of Jan Svartvik . London /New York:
Longman, pp. 95 -110.
211. Kennedy, C. and R. Bolitho 1991. English for Specific Purposes . Essential
Language Teaching Series, London: Macmillan.
212. Kenny, D. 2000. Lexis and Creativity in Translation. A corpus -Based Translation
Study . Manchester/ Northampton :St. Jerome Publishing .
213. Kiesling Scott, F. and C. B. Paulston (eds.). 2005. Intercultural Discourse and
Communication. Essential Readings . UK: Blackwell Publishing .
214. Kjellmer, G. 1987. “Aspects of English Collocations”. in W. Meijs (ed.), Corpus
Linguistics and Beyond. Proceedings of the Seventh International Conference
on English Language Research on Computerized Corpora . Amsterdam:
Rodopi pp. 133 -140.
215. Kjellmer, G. 1991. “A mint of phrases” in K. Aijmer and B. Altenberg (eds.).
English Corpus Linguistics . London: Longman , pp. 111 -27
216. Klappenbach, R. 1968. “Probleme de Phraseologie” in Wissenschaftliche Zeitschrift
der Karl -Max-Universität , 17/5. pp.221 – 227.
217. Kramsch , C. 1998. Language and Culture. Oxford: Oxford University Press.
218. Krashen, S. D. 1981. Second Language Acquisition and Second Language
Learning . Oxford: Pergamon Press.
219. Krashen, S. D. and R. Scarcella 1978. “On routines and patterns in language
acquisition and performance” Language , 28/2, pp. 283 -300.

220. Kroeber, A. and C ., Kluckhohn 1952 . Cultures. A Critical Review of Concepts and
Definitions. Peaboy Museum papers, vol.47, 1. Cambridge Mass: Harvard
University.
221. Kunin, A. V. 1955. English -Russian Phraseologic al Dictionary . Moscow.
222. Labov, W. 1972. Sociolinguistic Patterns . Pennsylvania: University of
Pennsylvania Press in N. Bonvillain, 2003. Language and Communication.
The Meaning of Messages. New Jersey: Prentice Hall.
223. Lamb, S. M. 1962. Outline of Stratifi cational Grammar . Berkeley: ASUC. in R. ,
Moon 1998.
224. Langacker, R. W. 1987. Foundations of Cognitive G rammar: Theoretical
Prerequisites. Stanford, CA: Stanford University Press.
225. Lakoff, G. and M. Johnson 1980. Metaphors We Live By. London: University of
Chicago Press.
226. Lakoff, G. and M. Johnson 1982. Metaphor and Communication . L.A.U.T., Trier.
227. Leech, G. 1981. Semantics . 2nd ed., Harmondsworth, Penguin.
228. Leech, G. 1987. Meaning and the English Verb . London: Longman.
229. Leech, G. 1990. Semantics. An Introductio n to the Study of Meaning . Penguin
Books.
230. Leech, G. and M. H. Short 1981. Style in Fiction . London and New York:
Longman.
231. Lenci, A, Montemagni, S. Pirrelli, V. 2005 . Testo e c omputer. Elementi di
linguistica computazionale , Roma: Carocci Editore.
232. Lo Casci o, V. 1995. ‘ Semantica lessicale e i criteri di collocazione nei dizionari
bilingui’ , in T. De Mauro and V. Lo Cascio (eds.) 1997. Lessico e
grammatica. Teorie linguistiche e applicazioni lessicografiche. Roma:
Bulzoni.
233. Levițchi, L. 1970. Limba engleză con temporană. Lexicologie București: Editura
Didactică și Pedagogică.
234. Levițchi, L. 1993. Manualul traducătorului de limba engleză. București: Editura
Teora.
235. Levinson, S. C. 1983. Pragmatics . Cambridge: Cambridge University Press.
236. Lewis, M. 1993. The Lexical Approach. Hove: Language Teaching Publications.
237. Lewis, M. (ed.) 2000. Teaching Collocation . USA: Thomson.
238. Lewis, M. 2002a. The Lexical Approach. The State of ELT and a Way Forward .
USA: Thomson Heinle.
239. Lewis, M. 2002b. Implementing the Lexical Approach. Putting Theory into
Practice . USA: Thomson Heinle.
240. Lipka, L. 1974. “Probleme der Analyse englischer Idioms aus struktureller und
generativer Sicht” in Linguistik und Didaktic, 20. pp. 274 -85.
241. Lodge, D. 1977. The Modes of Writing. Metaphor, Metonymy and the Typology of
Modern L iterature, London: Arnold.
242. Lombardo, L., L., Haarman, J. Morleyand and C. Taylor 1999. Massed Medias.
Linguistic Tools for Interpreting the Media Discourse. Milan: Edizioni
Universitarie di Lettere Economia Diritto.
243. Lonsdale, D. 1997. “Modelling cognition in SI: Methodological issues”. in
Interpreting 2 (1-2), pp. 91 -117.

244. Lyons, J. 1977. Semantics. Cambridge: Cambridge University Press.
245. Lyons, J. 1981. Language and Linguistics. An Introduction . Cambridge, Cambridge
University Press .
246. Lyons, J. 1995. Linguistic Semantics . An Introduction. Cambridge: Cambridge
University Press.
247. Makkai, A. 1972. Idiom Structure in English . The Hague: Mouton.
248. Malinowski, B. 1938. ‘The Anthropology of Changing African Cultures. In
Methods and Study of Cul ture Contact in Africa’. in L. P. Mair, (ed.)
International Institute of African Languages and Cultures. Memorandum
XV. Oxford: Oxford University Press.
249. Malmkjaer, K. (ed). 1991. The Linguistic Encyclopedia . London: Routledge.
250. Malmkjaer, K. 2005. Linguisti cs and the Language of Translation . Edinburgh:
Edinburgh University Press.
251. Marinucci, M. 1996. La lingua italiana . Torino: Bruno Mondatori edizioni
Scolastiche.
252. Marslen –Wilson, W. D and A. Welsh 1978. “Processing interactions and lexical
access during wor d recognition in continuous speech” Cognitive Psychology .
10: pp.29-63.
253. Martin, J. R. 1992. English Text: System and S tructure . Amsterdam: John
Benjamins Publishing Company.
254. Mauranen, A. 2000. “Strange strings in translated language: a study in corpora”, in
M. Olohan (ed.), Intercultural Faultiness. Research Models in Translation
Studies 1. Textual and Cognitive Aspects . Manchester / Northampton: St.
Jerome Publishing, pp. 119 -141.
255. Mauranen, A. 2004. Corpora, Universals and Interference , in Translating
Universals , A. Mauranen and P. Kujamaki (eds.), Amsterdam, Philadelphia:
John Benjamins Publishing Company, pp. 65 -83.
256. McCarthy, M. 1984. “Some pragmatic features of written text”, manuscript,
English Language Research, University of Birmingham in R., Carter. 1998.
Vocabulary. Applied Linguistic Perspectives . London and New York:
Routledge.
257. McCarthy, M. J. and R. Carter. 1994. Language as Discourse : Perspectives for
Language Teaching . London: Longman.
258. McIntosh, A. 1961. ‘Patterns and Ranges’ in Language . Vol. 37, No. 3.
259. McIntosh, A 1966. “Patterns and ranges” in A. McIntosh and MAK Halliday
Patterns of Language. Papers in General Descriptive and Applied
Linguistics. London: Longman, pp. 182 -199.
260. McMordie, W. 1972. English Idioms . London: Oxford University Pre ss.
261. Mel’čuk, I. A. 1988. “Semantic Description of Lexical Units in an Explanatory
Combinatorial Dictionary: Basic Principles and Heuristic Criteria” in
International Journal of Lexicography . I/3: pp. 165 -88.
262. Melčuk, I. 1995. “Phrasemes in language and p hraseology in linguistics” in Idioms:
Structural and Psychological Perspectives . Evaraert et al. (eds.) pp. 167 -232,
Hillsdale, New Jersey: Lawrence Erlbaum Associates.

263. Mel’čuk, I. A. 1998. “Collocations and Lexical Functions” in A. P., Cowie (ed)
Phraseol ogy. Theory, Analysis, and Applications . Oxford: Clarendon Press.
pp. 23 -53.
264. Mel’čuk, I. A. and N.V. Pertsov. 1987. Surface Syntax of English, a Formal M odel
in the Meaning -Text Theory Amsterdam, Philadelphia: Benjamins
Publishing.
265. Mel’čuk, I. A. and L. W arner 1994. “Lexical co -occurrence and lexical Inheritance.
emotion l exemes in German: a lexicographic case study” Lexikos 4, Afrilex –
Reek/ Series 4. Stellenbosch/ Republik Südafrika. pp. 88 -160.
266. Merilă, I. 2014. Constructing Otherness in Salman Rushdie’s Novels . Germany:
Lambert Academic Publishing.
267. Meyer, D. E. and R. W. Schvaneveldt. 1971. “Facilitation in recognizing pairs of
words: evidence of dependence between retrieval operations”. Journal of
Experimental Psychology . 90: pp. 227 -234.
268. Milroy, L. 1 980. Language and Social Networks . Oxford: Blackwell in N.
Bonvillain, 2003. Language and Communication. The Meaning of Messages.
New Jersey: Prentice Hall.
269. Milroy L. and J. Milroy 1978. “Belfast: Change and variation in an urban
vernacular” in P. Trudg ill (ed.) Sociolinguistic Patterns in British English .
London: Edward Arnold in N. Bonvillain, 2003. Language and
Communication. The Meaning of Messages. New Jersey: Prentice Hall.
270. Milroy L. and J. Milroy 1992. ‘Social network and social class: Toward an
intergrated sociolinguistic model ’. Language and Society 21, pp. 1 -26, in N.
Bonvillain, 2003. Language and Communication. The Meaning of Messages.
New Jersey: Prentice Hall and in R. Carter 1998.
271. Mitchell, T. F. 1971. “Linguistics ‘going -ons’: on collo cations and other lexical
matters arising on the syntagmatic record” in Archivum Linguisticum , II (ns),
pp. 35 -69.
272. Mitchell, T. F. 1975. “Linguistic ‘goings -on’: collocations and other lexical matters
arising on the syntagmatic record” in T. F. Mitchel (ed .) Principles of
Firthians Linguistics . 99-136, London: Longman.
273. Moon, R. 1992a. ”Textual aspects of fixed expressions in learners’ dictionaries” in
P. Arnaud and H. Bejoint (eds.) Vocabulary and Applied Linguistics .
London: Macmillan, pp. 13 -27.
274. Moon, R. 1992b. “ There is no reason in roasting the eggs : a consideration on fixed
expressions in native speakers dictionaries” in H. Tommolam K.Varantola, T.
Salmi -Tolonen and J. Schopp (eds). Euralex’ 92 Proceedings , II, Tampere:
University of Tampere, pp. 493 -502.
275. Moon, R. 1994. ‘The analysis of fixed expressions in text’ in M. Coulthard, (ed.)
Advances in Written Text Analysis . London: Routledge.
276. Moon, R. 1998a. Fixed Expressions and Idioms in English. A Corpus -Based
Approach . Oxford: Clarendon Press.
277. Moon, R . 1998b. “Frequencies and Forms of Phrasal Lexemes in English” in A. P.
Cowie (ed) Phraseology. Theory, Analysis, and Applications . Oxford:
Clarendon Press. pp. 79 -100.

278. Morton, J. 1969. “Interaction of information in word recognition” in Psychological
review 76: pp. 165-178.
279. Morton, J. 1979. “Word recognition” in J. Morton and J. C Marshall (eds.)
Psycholinguistic series 2: structures and processes . London: Elek.
280. Morton, J. 1980. “The logogen model and orthographic structure” in U. Frith (ed)
Cognitive P rocesse s in S pelling . London: Academic Press.
281. Munday, J. 2001. Introducing Translation Studies . Theories and Applications .
London and New York: Routledge.
282. Munteanu, C. 2007. Sinonimia frazeologică în limba română din perspectiva
lingvisticii integrale . Pite ști: Editura Independența Economică.
283. Munteanu, C. (ed.) 2008. Discursul repetat. Iași: Institutul European.
284. Musacchio, M.T. 2002 “ I tecnicismi collaterali ”, in M. Magris, M.T. Musacchio ,
L. Rega and F. Scarpa (eds.) Manuale di terminologia. Aspetti teoric i,
metodologici e applicativi , Hoepli: Milano.
285. Musacchio, M.T. and G. Palumbo. 2008. “Shades of grey: a corpus -driven analysis
of LSP phraseology for translation purposes”, in C. Taylor Torsello (ed.)
Corpora for University Level Language Teach ers. Bern/ Berlin/Frankfurt:
Lang.
286. Nabokov, V 1964. Notes on Prosody . Bolingen Series.
287. Nattinger, J. 1988. “Some current trends in vocabulary teaching” in R. Carter, and
M. McCarthy. Vocabulary and Language Teaching . London: Longman, pp.
62-82.
288. Nattinger, J.R. and J .S. DeCarrico, 1989. “Lexical phrases, speech acts and
teaching conversation” in P. Nation and R. Carter (eds.) AILA Review 6:
Vocabulary Acquisition. Amsterdam: AILA. pp. 118 – 139.
289. Nattinger, J.R. and J.S. DeCarrico. 1992. Lexical Phrases and Language Tea ching .
Oxford: Oxford University Press.
290. Neagu I. V. 2012. Black Humour: A Stylistic Approach . Germany: Lambert
Academic Publishing.
291. Neagu, M. 2005. Cognitive Linguistics. An Introduction . Bucure ști: Editura
Didactică și Pedagogică R.A.
292. Newmark, P. 1985. ‘The Translation of Metaphor’ in W. Paprotte and R. Dirven
(eds.) The Ubiquity of Metaphor. Amsterdam, Benjamins.
293. Newmark, P. 1988. A Textbook of Translation . New York: Prentice Hall.
294. Nicolas, T. 1995. “Semantics of Idiom modification” in Evaraert, M., E -J., Van der
Linden, A. Schenk, and R. Schreuder (eds.) 1995. Idioms: Structural and
Psychological Perspectives , Hillsdale, New Jersey: Lawrence Erlbaum
Associates. pp. 233 -252.
295. Nida, E. A. 1964 ‘Principles of correspondence’ in L. Venuti, (ed.) (2000) The
Translation Studies Reader .
296. Nida, E. A. 1996. The Sociolinguistics of Interlingual Communication . Bruxelles:
Editions du Hazard.
297. Nida E. and C. Taber. 1969. The Theory and Practice of Translat ion. Brill: Leiden.
298. Niemeier, S. and R. Dirven (eds.) 2000. Evidence for Linguistic Relativity.
Amsterdam, Philadelphia: John Benjamins.

299. Nord , C. 1997a. Translation as a Purposeful Activity. Functionalist Approaches
Explained . Manchester: St. Jerome.
300. Nord , C. 1997b. “A functional typology of translation»”, in A. Trosborg (ed.) Text
Typology and Translation . Amsterdam/Philadelphia: Benjamins, pp. 43 -66.
301. Nowattny, W. 1962: The Language Poets U se, London: The Athlone Press.
302. Nunan, D . and R. Carter (eds.). 199 3. Introducing Discourse Analysis . Penguin
English.
303. Omazić, M 2009. “Processing of idioms and idiom modifications” in Granger, S.
and F. Meunier (eds.) Phraseology. An Interdisciplinary P erspective.
Amsterdam/ Philadelphia : John Benjamins Publishing Compan y, pp.67 -79.
304. Ogden, C. K. and L.A. Richards. 1923. The Meaning of Meaning . New York:
Harcourt, Brace & Company in M., Zdrenghea, 1977. Introduction to
Semantics . Cluj – Napoca: Universitatea “Babeș -Bolyai”.
305. Opriț -Maftei C. 2018. English Borrowings in Rom anian; Economic Terminologies .
Galați: Europlus.
306. Palmer, H. E. 1936. “The Art of Vocabulary Layout” in Bulletin of the Institute for
Research in English Teaching 121: 1 -8; pp. 14 -19.
307. Palmer, F. R. 1981. Semantics . 2nd edn. Cambridge: Cambridge University P ress.
308. Paprotte, W. and R. Dirven (eds.). 1985. The Ubiquity of Metaphor. Amsterdam:
Benjamins.
309. Pawley, A. 1985. “On Speech Formulas and Linguistic Competence” in Lenguas
modernas , Chile 12: pp. 84 -104.
310. Pawley , A. 1986. “Lexicalization” in D. Tannen, an d J. E. Alatis (eds.) Languages
and Linguistics: The Interdependence of Theory, Data and Application .
Georgetown University Round Table on Languages and Linguistics 1985.
Washington DC: Georgetown University Press, pp. 98 -120.
311. Pawley, A. and F. H., Syder 1 983. “Two Puzzles for Linguistic Theory: Nativelike
Selection and Nativelike Fluency” in J. C. Richards and R.W. Schmidt (eds.)
Language and Communication . London: Longman, pp. 191 -225.
312. Pârlog, H. (ed) British and American Studies . Timișoara: Editura Unive rsității de
Vest.
313. Pârlog, H., P. Brânzeu, A -C., Pârlog 2009. Translating the Body . Iași: Institutul
European.
314. Peters, A. M. 1983. The Units of Language Acquisition . Cambridge: Cambridge
University Press.
315. Pinker, S. 1995. The Language Instinct . London: Pen guin.
316. Pustejovski, J., S. Bergler and P. Anick 1993 “Lexical Semantic Techniques for
Corpus Analysis” in Computational Linguistics . 19, pp. 331 -358.
317. Quemada, B. 1968. Les dictionnaires du français moderne , 1539 -1863: étude sur
leur histoire, leurs types, et leurs méthodes . Paris: Didier.
318. Radford, A. 1988. Transformational Grammar . Cambridge: Cambridge University
Press.
319. Ranney , S. 1992. ‘Learning a New Script: An Exploration of sociolinguistic
competence’. in Applied Linguistics. Vol. 3, Oxford: Oxford Univ ersity
Press.

320. Riffaterre, M. 1964. “The Stylistic Function” in Proceedings of the 9th International
Congress of Linguistics , The Hague.
321. Roberts, M. H. 1944. The Science of Idiom: A Method of Inquiry into the Cognitive
Design of Language . Publication of the Modern Languages Association of
America 69: pp 291 -306.
322. Robinson, P. 1991. ESP Today. A Practitioner’s Guide . UK: Prentice Hall
International Ltd.
323. Rosch, E. H. 1973. ‘On the internal structure of perceptual and semantic
categories’. in T.E., Moore Cogniti ve Development and the Acquisition of
Language . New York: Academic Press.
324. Rosch , E. H. 1975. ‘Cognitive representation of semantic categories’. in Journal of
Experimental Philosophy . General, 104 (3), pp. 192 -233.
325. Rosch, E. H. and C.B. Mervis 1975. Family Resemblance Studies in the Internal
Structure of Categories . in Cognitive Psychology . no. 7, pp 573 -605.
326. Sadock, J. M. 1972. “Speech act idioms” in Papers from the Eighth Regional
Meeting of the Chicago Linguistic Society : pp. 329 -339.
327. Sadock, J. M. 1974. Toward a Linguistic Theory of Speech Acts . New York:
Academic Press.
328. Sankoff, G. 1972. “Language use in multilingual societies: Some Alternative
Approaches”, in J.P. Pride and J. Holmes (eds.) Sociolinguistics , Middlessex:
Penguin, pp. 33 -51.
329. Šarcevic, S. 1997. New Approach to Legal Translation . The Netherlands: The
Hague, Kluwer Law International.
330. Šarcevic, S. 2000. Legal Translation and Translation Theory : A Receiver -Oriented
Approach , http://www. tradulex.org/Actes2000/sarcevic.pdf
331. Saussure, F. 1998 . Curs de lingvistică generală . Iași: Polirom.
332. Saville -Troike, M. 1986. The Ethnography of Communication: An I ntroduction .
Oxford: Blackwell.
333. Scarpa , F. 2002. “‘Closer and closer apart’? Specialized translation in a cognitive
perspec tive” in A. Riccard i (ed.) Translation Studies. Perspectives on an
Emerging Discipline . Cambridge: Cambridge University Press, pp. 133 -149.
334. Scarpa , F. 2008. La traduzione specializzata. Un approccio didattico professionale ,
2nd ed , Milano: Hoepli.
335. Schäffner , C. and B. Adab 1 997. “Translation as Intercultural Communication –
Contact as Conflict” in M. Snell – Hornby, Z. Jettmarova and K. Kaindl (eds.)
1997. Translation as Intercultural Communication. Selected Papers from the
ESR Congress – Prague 1995, Amsterdam/ Philadelphia: John Benjamins,
pp. 325 – 337.
336. Schenk, A. 1995. ‘The syntactic behavior of idioms” in M. Everaert, E -J. van der
Linden, A. Schenk, and R. Schreuder (eds.) Idioms: Structural and
Psychological Perspectives , Lawrence Erlbaum Associates, pp. 253 –272.
337. Scherf, I. 2006. Expresii fraze ologice în limbile germană și română. Studiu
comparativ. București: Editura Didactică și Pedagogică RA.
338. Schiffrin, D. 1987. Discourse Markers . Cambridge: Cambridge University Press.

339. Schroten, J. 2000. ‘Equivalence and mismatch of semantic features: Coll ocations in
English, Spanish and Dutch’ in S. Niemeier, R. Dirven (eds.) 2000. Evidence
for Linguistic Relativity. Amsterdam, Philadelphia: John Benjamins.
340. Scollon, R., and S. Wong Scollon. 2001. Intercultural Communication. A Discourse
Approach. UK: Black well Publishing.
341. Scovel, T. 1998. Psycholinguistics. Oxford: Oxford University Press.
342. Searle, J. R. 1975. “Indirect Speech Acts” in P. Cole and J. L. Morgan (eds.).
Speech Acts. Syntax and Semantics . vol 3. New York: Academic Press, pp.
59-82.
343. Serianni, L . 1989 ’ Lingua medica e lessicografia specialistica nel primo Ottocento ’
in ID. Saggi di storia linguistica italiana , Napoli: Morano.
344. Serianni, L. and A. Castelvecchi, (colab.) 2004. Grammatica italiana . Italiano
comune e lingua letterarria . Torino: Utet Libreria.
345. Seymour, C. 1967. Stylistics: Quantitative and Qualitative . v.1.
346. Sinclair, J. 1966. “Beginning the study of lexis” in C. E. Bazel, J.C. Catford, MAK
Halliday and R. H. Robbins (eds) In memory of J. R. Firth . London:
Longmans, Green and Co Ltd.
347. Sinclair, J. 1987 a. “Collocation: a progress report” in Steele and Threadgold (ed.),
Language Topics , pp. 319 -331.
348. Sinclair, J. M. 1987 b. Looking up: An Account of the Cobuild Project in Lexical
Computing . London: Harper Collins.
349. Sinclair, J. 1991. Corpus , Concordance, Collocation . Oxford : Oxford University
Press.
350. Sinclair, J. 1992. “Trust the text” in M. Davies and L. Ravelli (eds.) Advances in
Systemic Linguistics London: Pinter, pp. 5 -19.
351. Sinclair, J. 1996. “The search for units of meaning”. Textus 9: pp.75 -106.
352. Sinclair, J. 1998. “The lexical item”. in E. Weigand (ed.). Contrastive Lexical
Semantics . Amsterdam/ Philadelphia: John Benjamins, pp. 1 -24.
353. Sinclair, J. M., S., Jones and R. Daley. 2004a. English Collocation Studies: The
OSTI Report . London and New York: Continuum.
354. Sinclair, J. 2004b. Trust the Text. Language, Corpus and Discourse .London and
New York: Routledge.
355. Slama -Cazacu, T. 1957. Relațiile dintre gândire și limbaj în ontogeneză. București:
Editura Academiei Republicii Populare Române.
356. Smith, L. P. 1925, Words and Idioms . London: Constable.
357. Snell -Hornby, M. 1988. Translation Studies: An Integrated Approach . Amsterdam:
John Benjamins .
358. Snell – Hornby M., Z. Jettmarova. And K. Kaindl (eds.) 1997. Translation as
Intercultural Communication. Selected Papers from the ESR Congress –
Prague 1995,.Amsterdam/ Philadelphia: John Benjamins.
359. Șorcaru, D. 2000. “Stylistic imprints as marks of iden tity” in H. Hulban (ed.) Style
in Language Discourses and Literature IV , Perspectives of the English
Language Series Iași: Tehnopress , pp. 21 -33.
360. Șorcaru, D. 2004. “Ways of pleating stylistic functions in Virginia Woolf’s To the
Lighthouse ” in H. Hulban, ( ed.) Style in Language, Discourses and Literature

II, Perspectives of the English Language Series 4, Iași: Editura Lumen. pp.
281- 308.
361. Spencer, J. and M. Gregory 1965. “An Approach to the Study of Style” in N.
Enkvist, J. Spencer and M. Gregory Linguistic s and Style . London: Oxford
University Press.
362. Spencer -Oatey H. 2000. Culturally Speaking. Managing Rapport through Talk
across Cultures . London and New York: Continuum.
363. Spillner, B. 1994. “Terminologies et connotations” in Francais scientifique , pp.53
364. Spolsky, B. 1998. Sociolinguistics. Oxford: Oxford University Press.
365. Stock, O., A. Ortony and J. Slack 1993. “Building Castles in the Air. Some
Computational and Theoretical Issues in Idiom Comprehension” in C.
Cacciari and P. Tabossi (eds.) Idioms. Processi ng, Structure and
Interpretation , Hillsdale, New Jersey: Lawrence Erlbaum Associates.
366. Strässler, J. 1982. Idioms in English: A Pragmatic Analysis . Tubingen: Verlag.
367. Stubbs, M. 1983. Discourse Analysis: The Sociolinguistic Analysis of Natural
Language . Oxfo rd: Blackwell.
368. Stubbs, M. 2002. Words and Phrases. Corpus Studies of Lexical Semantics. UK:
Blackwell Publishing.
369. Superceanu, R. 1998. The Rhetoric of Scientific Articles. A Genre Study . Timișoara:
Editura Orizonturi Universitare.
370. Svensen, B. 1993. Practical Lexicography . Oxford: Oxford University Press.
371. Swales , J. 1990. Genre Analysis: English in Academic and Research Settings .
Cambridge: Cambridge University Press.
372. Swinney, D. A. & A. Cutler. 1979. “The access and processing of idiomatic
expressions” in Journal of Verbal Learning and Verbal Behavior , 18.5, pp.
523-534.
373. Tabossi , P. and F. Zardon 1995. “The Activation of Idiomatic Meaning” in
Evaraert, M., E -J., Van der Linden, A. Schenk, and R. Schreuder (eds.) 1995.
Idioms: Structural and Psychological Perspectives , Hillsdale, New Jersey:
Lawrence Erlbaum Associates, pp. 273 – 282.
374. Tannen, D. 1989. Talking Voices: Repetition, Dialogue and Imagery in
Conversational Discourse . Cambridge: Ca mbridge University Press.
375. Taylor, J. 2002. Cognitive Grammar. New York: Oxford University Press.
376. Teliya, V., N. Bragina, E. Oparina and I. Sandomirskaya 1998. “Phraseology as
language of Culture: Its Role in the Representation of a Collective Mentality ”
in A. P., Cowie (ed) Phraseology. Theory, Analysis, and Applications .
Oxford: Clarendon Press. pp. 55 -75.
377. Tognini Bonelli, E. 1996. Corpus Theory and Practice . TWC.
378. Toolan, M. 1998 . Language in Literature: An Introduction to Stylistics . London and
New Y ork: Arnold Publishers.
379. Torsello, C. T. 1992. English in Discourse. A Course for Language Specialists. II.
Padova: Cleup Editore.
380. Toury G. 1995. Descriptive Tra nslation Studies and Beyond. Amsterdam/
Philadelphia: Benjamins.

381. Toury G. 1999. “A handful of paragraphs on ‘translation’ and ‘norms’” in C.
Schäffner (ed.). Translation and Norms . Clevedon: Multilingual Matters, pp.
9-31.
382. Trifone, P. and M. Palermo 2002. Grammatica italiana di base . Bologna:
Zanichelli Editore S.p.A.
383. Trimble, L. 1985. English for Science and Technology . Cambridge: Cambridge
University Press.
384. Trompenaars, F. and C. Hampden – Turner 1993. The Seven Cultures of
Capitalism . London: Piatkus.
385. Trompenaars, F. and C. Hampden – Turner 1997. Riding the Waves of Culture .
Understanding Cultu ral Diversity in Business. London: Nicholas Brealey
Publishing.
386. Tylor, E. 1871. Primitive Culture . London: John Murray, in Holliday et al. 2004.
387. Ulmann, S. 1973. Meaning and Style . Oxford: Basil Blackwell.
388. Ulrych, M. 1992. Translating Texts. From Theory to Practice . Genoa: Cideb
Editrice.
389. Van Roey , J. 1990. French – English Contrastive Lexicology: An Introduction.
Louvain -la-Neuve: Peeters.
390. Veisbergs, A. 1997. ‘The Contextual Use of Idioms, Wordplay and Translation’. in
D. Delabastita. Traductio – Essays on Punning and Translation. UK: St.
Jerome Publishing.
391. Venutti, L. 1995. The translator’s Invisibility : A History of Translation . London
and New York : Routledge
392. Verdonk, P. 2002 ( 2008 ). Stylistics. Oxford: Oxford University Press.
393. Verhaar, J. W. M 1966. Proceedings of the 9th International Congress of Linguists ,
The Hague. in I. R. Galperin 1977.
394. Vermeer, H. J. 1983. Aufsätze zur Translationsthoerie . Heidelberg in G. Garzone
2003.
395. Vermeer , H. 1996. A Skopos Theory of Translation (Some Arguments For and
Against ), Heidelberg: TEXTconTEXT.
396. Vinay, J. P. and J. Darbelnet 1995. Comparative Stylistics of French and English:
A Methodology for Translation . (translated by J. C. Sager and M. J. Hamel).
Amsterdam / Philadelphia: John Benjamins.
397. Vinogradov, V. V. 1947. “Ob osnovnuikh tipakh frazeologicheskikh edinits v
russkom yazuike” in A. A. Shakhmatov, 1864 -1920. Sbornik statey i
materialov , Moscow: Nauka pp. 339 -64 in A.P. Cowie (ed.) 1998.
398. Wagner, E., S. Bech and J. M. Martinez. 2002. Translating for the Euro pean Union
Institutions . Manchester, UK and Northampton MA: St. Jerome Publishing.
399. Wales, K. 1991. A Dictionary of Stylistics . Longman: London and New York.
400. Weigand, E. (ed.) 1998a. Contrastive Lexical Semantics. Amsterdam/ Philadelphia:
John Benjamins.
401. Weigand , E. 1998b.“Contrastive Lexical Semantics” in E. Weigand (ed.)
Contrastive Lexical Semantics, Amsterdam/ Philadelphia: John Benjamins,
pp. 25 -44.
402. Weinert, R. 1995. “The role of formulaic language in second language acquisition”
in Applied Linguistics . 16, pp. 180 -205.

403. Weinreich, U. 1966. “On the semantic structure of English”, in J. H. Greenberg
(ed.), Universals of language . 2nd Edition. Cambridge, MA: MIT Press, pp.
142–217.
404. Weinreich, U. 1969. “Problems in the Analysis of Idioms” in J. Puhvel (ed.)
Substance and Structure of Language . Berkeley and Los Angeles: University
of California Press, pp. 23 -81.
405. Weisberger, J. L. 1929. Muttersprache und Geistesbildung . Göttinge n:
Vandenhoeck & Ruprecht 64 in A.P. Cowie (ed.) 1998.
406. Wentworth, H. and S. B. Flex ner 1960. Dictionary of American Slang. New York:
Thomas Y. Crowell.
407. Werlich, E. 1976. Text Grammar of English . Heidelberg: Quelle & Meyer.
408. Widdowson , H. G. 1995. “Discourse Analysis. A Critical View” in Language and
Literature. 4, 3: 157 -72.
409. Widdowson, H. G. 1996 (2000). Linguistics. Oxford: Oxford University Press.
410. Widdowson, H. G. 1997. Stylistics and the Teaching of Literature . Longman:
London.
411. Widdowson, H. G. 2004. Text, Context, Pretext. Critical Issues in Discourse
Analysis . UK: Blackwell Publishi ng.
412. Wikberg, K. 2009. “Phrasal similes in the BNC” in Granger, S. and F. Meunier
(eds.) Phraseology. An interdisciplinary perspective. Amsterdam/
Philadelphia : John Benjamins Publishing Company, pp. 127 – 142.
413. Wittgenstein, L. 1972. Philosophical Investig ations . Oxford: Basil Blackwood .
414. Wood, M. M. 1981. A Definition of Idiom , MA thesis: University of Birmingham
(published in 1986) in R., Moon. 1998a. Fixed Expressions and Idioms in
English. A Corpus -Based Approach . Oxford: Clarendon Press.
415. Woods, N. 2006. Describing Discourse . A Practical Guide to Discourse Analysis.
London: Hodder Arnold.
416. Woolard, G. 2000. ‘Collocation – encouraging learner independence’ in L.,
Michael (ed.) Teaching Collocation . USA: Thomson.
417. Wray, A. 1999. ‘Formulaic language in lear ners and native speakers’ in Language
Teaching 32 (4), pp. 213 -231.
418. Yule, G. 1996 (2007). Pragmatics. Oxford: Oxford University Press.
419. Zdrenghea, M. 1977. Introduction to Semantics . Cluj-Napoca: Universitatea
“Babeș -Bolyai”.
420. Zgusta, L. 1967. “Multiword l exical units”, Word , 23, pp. 325 -39 in R., Moon.
1998a. Fixed Expressions and Idioms in English. A Corpus -Based Approach .
Oxford: Clarendon Press.
421. Zgusta, L. 1971. Manual of Lexicography . Janua linguarum, Series maior 39. The
Hague: Mouton.
422. Zugun, P. 200 0. Lexicología limbii române . Iași: Editura Tehnopress.

CORPUS

1. *** A Guide to Idioms . 2000. Boston: Thomson.
2. *** A Guide to Phrasal Verbs . 2000. Boston: Thomson.
3. *** English for Business Administration . 2002. Bucharest: Cavallioti Publishers
and the British Council.
4. ***Lo Zi ngarelli 2003 – Vocabolario della lingua italiana 2003.Bologna:
Zanichelli Editore.
5. *** Longman Dictionary of Contemporary English (LDCE) 2003. England:
Pearson Educational.
6. *** Longman Dictionary of English Language and Culture (LDELC) 2001.
England: Pear son Educational.
7. *** Webster's New Encyclopedic Dictionary (WNED), 2000.
8. *** Dizionari Garzanti inglese -italiano, italiano -inglese , 2001. Italia:Garzanti
Editore.
9. *** Dizionario di sinonimi e contrari generici, specifici, analoghi, inverse . 2001.
Italia: G arzanti editore.
10. *** Oxford Advanced Learner's Dictionary (OALD), 1995. Oxford: Oxford
University Press.
11. *** The Oxford Talking Dictionary (OTD) (virtual version).
12. ***Random House Webster's (virtual version).
13. Aioane, M. and A. Pintilie 2005. Dicționarul p rincipalelor verbe italiene și
construcțiile lor specifice . București: Editura Polirom.
14. Bantaș, A. 2002. Dictionar frazeologic englez -român, București: Editura Teora.
15. Bidu -Vrănceanu, A. 2001. Lexic științific interdisciplinar . București: Editura
Universi tății din București.
16. Bidu -Vrănceanu, A. 2000. Lexic comun, lexic specializat . București: Editura
Universității din București.
17. Campbell, B. 2000. English for Business , USA: Thomson
18. Costescu. E. 1979. Dicționar frazeologic român – italian . Bucureșt i: Editur a
Științtifică și E nciclopedică.
19. Cowie, A. P, R. Mackin and I. R. McGaig 1993. Oxford Dictionary of English
Idioms . Oxford: Oxford University Press.
20. Dima, G., L. Tomescu and E. Vasilescu 2000. English – Romanian Dictionary for
Mechanical and Metallurgica l Engineering . vol. 1 (A -K). București: Editura
Științifică F.M.R.
21. Dima, G., L. Tomescu and E. Vasilescu 2000. English – Romanian Dictionary for
Mechanical and Metallurgical Engineering . vol. 2 (L -X). București: Editura
Științifică F.M.R.
22. Dixon, R. J. 2004 . Essential Idioms in English. Phrasal Verbs and Collocations .
USA: Longman.
23. Duncan, F. H. and M. Cârcotă – Dumitriu 2000. English – Romanian Dictionary of
Accounting, Economic and Financial Terms , București: Editura R.A.I.,
Imprimeria Coresi.
24. Flower, J. 20 02. Phrasal Verbs Organizer . Boston: Thomson Heinle.

25. Guazzotti, P. and M. F. Oddera 2006. Il grande dizionario dei proverbi italiani .
Bologna: Zanichelli Editore.
26. Hill, J., and M. Lewis (eds.). 2002. LTP Dictionary of Selected Collocations .
USA: Thomson .
27. Hulban, H. 2002. 111 Verbs and their Phrases. Iași: Editura Spanda.
28. Istrate, M., D. I. Frențiu, G. C. Frențiu 2008. Dicționar juridic român –italian și
italian – român . București: Editura Hamangiu.
29. Ivanovic, A, and P. H. Collin 2006. Dicționar de marketing englez – român .
București: Editura Niculescu .
30. Knowles, E. 2006. Oxford Dictionary of Phrase and Fable . Oxford: Oxford
University Press.
31. Lefter, V. 1974. Dicționar de proverbe englez -român . Casa ScânteiiȘ București.
32. Martin, A. E. 2001 (ed.) A Dictionary of Law . New York: Oxford University Press.
33. Megalo, C. Perillo, M. and M. Giova 2008. Testi e linguaggi. Antologia , Napoli :
Fratelli Ferraro Editori.
34. Neagu, M, and D. Șarpe, 1998. Dicționar de termeni economici englez – român .
Galați: Editura Alma.
35. Pârlog, H. and M. Teleagă 1999. Dicționar englez – român de colocații nominale .
Timișoara: Editura Mirton.
36. Pârlog, H. and M. Teleagă 1999. Dicționar englez – român de colocații verbale .
Iași: Editura Polirom.
37. Picchi, F. 2006. Dizionario economico e commerciale inglese – italiano . Bologna:
Zanichelli, Il Sol e 24 ore.
38. Picchi, F. 2006. Dizionario economico e commerciale italiano -inglese . Bologna:
Zanichelli, Il Sole 24 ore.
39. Pop, L. 2003. Dicționar englez -român de afaceri . Iași: Editura Polirom.
40. Ragazzini, G. and Biagi, A. 2006. Dizionario inglese -italiano. Concise Bologna:
Zanichelli.
41. Ragazzini, G. and Biagi, A. 2006. Dizionario italiano – inglese. Concise Bologna:
Zanichelli.
42. Robbins, S., 2004. Business Vocabulary in Practice . Great Britain: Harper Collins
Publishers.
43. Seely, J. 2003. Law in Everyday Life . Oxfo rd: Oxford University Press.
44. Seidl, J and W. McMordie 1988. English Idioms . Oxford: Oxford University Press.
45. Smyth , F. 2003. Crimes and Victims . England: Abbeydale Press.
46. Thorne, T. 1990. Dictionary of Contemporary Slang. New York: Random House.
47. Tiberii, P. 2012. Dizionario delle collocazioni. Le combinazioni delle parole in
italiano , Milano: Zanichelli Editori .
48. Tomici , M. 2009. Dicționar frazeologic al limbii române. București: Saeculum
Vizual.
49. Urzì F. 2009. Dizionario delle Combinazioni Lessicali , Luss emburgo: Edizioni
Convivium.
50. Voroniuc, A. 2006. Dicționar englez – român de ter meni economici și juridici.
Iași: Institutul European.
51. Voroniuc, A. 2006. Dicționar român – englez de termeni economici și juridici . Iași:
Institutul European.

52. Willis, D. 1990. The Lexical Syllabus. A New Approach to Language Teaching .
London and Glasgow: Collins ELT.
53. Woolard, G. 2005. Key Words for Fluency . Pre-Intermediate Collocation Practice .
USA: Thomson Heinle.
54. Woolard, G. 2005. Key Words for Fluency . Intermediate Co llocation Practice .
USA: Thomson Heinle.
55. Woolard, G. 2004. Key Words for Fluency . Upper -Intermediate Collocation
Practice . USA: Thomson Heinle.
56. Wright, D. (ed.) 1986. English Romantic Verse England: Penguin Books.
57. Wright, J. 2002. Idioms organizer . Boston: Thomson Heinle.