Get the full ebook with Bonus Features for a Better Reading Experience on ebookgate.

com

Investigating Lexis Vocabulary Teaching ESP

Lexicography and Lexical Innovation 1st Edition
José Ramón Calvo-Ferrer

[Link]
vocabulary-teaching-esp-lexicography-and-lexical-
innovation-1st-edition-jose-ramon-calvo-ferrer/

OR CLICK HERE

DOWLOAD NOW

Download more ebook instantly today at [Link]

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

The Vocabulary Files C2 Advanced IELTS 7 0 8 0 9 0 1st

Edition Andrew Betsis

[Link]
ielts-7-0-8-0-9-0-1st-edition-andrew-betsis/

[Link]

AS pure mathematics C1 C2 3rd ed Edition Val Hanrahan

[Link]
edition-val-hanrahan/

[Link]

A Slow Death 83 Days of Radiation Sickness Nhk Tv Crew

[Link]
sickness-nhk-tv-crew/

[Link]

Warships in the War of the Pacific 1879 83 1st Edition

Angus Konstam

[Link]
pacific-1879-83-1st-edition-angus-konstam/

[Link]
C2 Re envisioned The Future of the Enterprise 1st Edition
Marius S. Vassiliou

[Link]
enterprise-1st-edition-marius-s-vassiliou/

[Link]

Longitudinal Developments in Vocabulary Knowledge and

Lexical Organization 1st Edition Brigitta Dóczi

[Link]
knowledge-and-lexical-organization-1st-edition-brigitta-doczi/

[Link]

Teaching and Learning Vocabulary Bringing Research to

Practice Elfrieda H. Hiebert

[Link]
bringing-research-to-practice-elfrieda-h-hiebert/

[Link]

The Lexical Approach The State of ELT and a Way Forward

Language Teaching Publications 1st Edition Michael Lewis

[Link]
and-a-way-forward-language-teaching-publications-1st-edition-michael-
lewis/
[Link]

Teaching Academic ESL Writing Practical Techniques in

Vocabulary and Grammar 1st Edition Eli Hinkel

[Link]
techniques-in-vocabulary-and-grammar-1st-edition-eli-hinkel/

[Link]
Investigating Lexis
Investigating Lexis

Vocabulary Teaching, ESP,

Lexicography
and Lexical Innovation

Edited by

José Ramón Calvo-Ferrer

and Miguel Ángel Campos Pardillos
Investigating Lexis:
Vocabulary Teaching, ESP, Lexicography and Lexical Innovation

Edited by José Ramón Calvo-Ferrer

and Miguel Ángel Campos Pardillos

This book first published 2014

Cambridge Scholars Publishing

Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Copyright © 2014 by José Ramón Calvo-Ferrer,

Miguel Ángel Campos Pardillos and contributors

All rights for this book reserved. No part of this book may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise, without
the prior permission of the copyright owner.

ISBN (10): 1-4438-6807-8

ISBN (13): 978-1-4438-6807-5
 TABLE OF CONTENTS

Foreword .................................................................................................... ix

Part I. Lexical Theory and Acquisition

Chapter One ................................................................................................. 3

“You Shall Know a Collocation by the Company It Keeps”:
Methodological Advances in Lexical-Constellation Analysis
Moisés Almela

Chapter Two .............................................................................................. 27

From EFL to CLIL: Teaching and Learning Vocabulary in the Primary
Classroom
María Tabuenca Cuevas and Gema Alcaraz Mármol

Part II. Legal Terminology

Chapter Three ............................................................................................ 41

Pronunciation Skills in Legal English for Interpreters:
English Latinisms and Cognates
Miguel Ángel Campos Pardillos

Chapter Four .............................................................................................. 57

The U.S. Supreme Court Cognitive Metaphors: Law, Deep Roots
and the Right Soil
Laura Vegara Fabregat

Chapter Five .............................................................................................. 79

A Comparative Study of Latinisms in the European Legislation:
Degrees of Syntactic Integration in English, Spanish and Greek
Document Versions
Rita Winiarska
vi Table of Contents

Part III. Dictionaries: Past and Present

Chapter Six ................................................................................................ 95

Online Arabic-English-Arabic Specialised Dictionaries
Reima Al-Jarf

Chapter Seven.......................................................................................... 103

General Remarks on the Challenges of Integrating Scientific
and Technical Words in General Dictionaries
Isabel Balteiro and José Ramón Calvo-Ferrer

Chapter Eight ........................................................................................... 123

Gentyll On-Line Glossaries: Professional Titles of Women and Men
in a Series of Fields of Activity
Mercedes Bengoechea, María Rosa Cabellos and José Simón

Chapter Nine............................................................................................ 133

The Dictionary of Richard Percyvall’s Bibliotheca Hispanica (1591):
Structure and Composition
David Carrascosa Cañego

Chapter Ten ............................................................................................. 145

First Anglicisms in the Spanish Press: Treatment Given by the Royal
Spanish Academy Dictionary
María Vázquez-Amador & M. Carmen Lario-de-Oñate

Part IV. New Challenges, New Approaches

Chapter Eleven ........................................................................................ 163

Swingvergüenzas A Contra Blues: A Study on Creative Code Mixing
in Spanish Music
Paula López Rúa

Chapter Twelve ....................................................................................... 181

New Challenges in the Translation of Terminology for Software
Applications
José Ramón Belda Medina
 Investigating Lexis vii

Chapter Thirteen ...................................................................................... 191

The Terminology of the Video Game Industry: A New Type
of Specialised Language
Ramón Méndez González

Chapter Fourteen ..................................................................................... 205

The Multiple Shades of Erotica: Translating Romantic
and Erotic Fiction into Spanish
Scheherezade Surià
 FOREWORD

This book is a collection of essays representing various aspects of

lexicography; in all cases, its authors have attempted to combine state-of-
the-art research with a user-friendly approach which makes it attractive for
readers worldwide. It is divided into four major sections: (i) Lexical theory
and acquisition, (ii) Legal terminology, (iii) Dictionaries, (iv) New
challenges.
The first section, which deals with basic theoretical issues regarding
vocabulary, consists of two studies. The first one, “You Shall Know a
Collocation by the Company It Keeps”: Methodological Advances in
Lexical-Constellation Analysis” studies collocations, offering a new
development within standard collocational analysis, that of lexical
constellations, which improves semantic description through the use of
clusters. The other paper in this section, “Vocabulary in Primary Language
Learning” revisits the first stages of vocabulary learning and acquisition
by analysing lexical choices in primary school textbooks, and the
implications such choices have upon primary school teaching.
The second section focuses on legal terminology, an area of specialized
vocabulary which has received comparatively less attention than others
(e.g. business), but the importance of which is progressively gaining
ground with the weight of transnational political bodies. The first paper,
“Pronunciation Skills in Legal English for Interpreters: English Latinisms
and Cognates” analyses a gap in interpreter training regarding the use of
adapted and unadapted words of Latin origin in legal English, and
emphasizes the need for a more thorough approach to specialized
vocabulary learning which also includes pronunciation. This is followed
by an innovative study of another almost uncharted territory: legal
metaphors, entitled “The US Supreme Court Cognitive Metaphors and
their Translation into Spanish: Law, Deep Roots and the Right Soil”. This
paper examines the conceptual framework underlying some common
metaphors in legal English, insofar as they shape legal reasoning and
become a challenge for translators. The third paper in this section, “A
Comparative Study of Latinisms in European Legislation. Degrees of
Syntactic Integration in English, Spanish and Greek Document Versions”,
deals with Latin expressions and how they have made their way into
 Investigating Lexis ix

English, Spanish and Greek syntax, while in most cases remaining clearly
recognizable as a sort of legal lingua franca.
The third part of the book offers new insights into dictionaries, a tool
which has had to face new challenges over the last decades and with the
advent of globalization and new technologies. The first chapter, “Online
Arabic-English-Arabic Specialized Dictionaries”, points out some of the
shortcomings of the dictionaries presently available for specialized translation
between English and Arabic, and makes a few suggestions for improvement in
terms of coverage, accuracy and inclusiveness. The second contribution,
“General Remarks on the Challenges of Integrating Scientific and Technical
Words in General Dictionaries” deals with a problem that lexicographers
have traditionally faced, and has received scant scholarly attention: i.e. to
which extent scientific terms should be included in a dictionaries intended
for the general public, and the amount of information they should offer.
The third paper, “Gentyll On-Line Glossaries: Professional Titles of
Women and Men in a Series of Fields of Activity”, is a poignant revelation
of how, in spite of the alleged gender neutrality and political correctness in
lexicographic practice, terminological banks continue to base human
denomination on the masculine term. For their part, the fourth and fifth
papers look at past lexicographic practice. The first one, entitled “The
Dictionary of Richard Percyvall’s Bibliotheca Hispanica (1591): Structure
and Composition”, provide insight into a 16th century grammar-cum-
dictionary for English learners of Spanish, analysing the criteria used, the
English translation suggested and the role of Latin in its creation process.
The second one, “First Anglicisms in the Spanish Press: Treatment Given
in The Royal Spanish Academy Dictionary” describes the practices by
domestic prescriptive authorities with regards to the admission of
Anglicisms in Spanish –a rather controversial issue, especially when the
self-declared role of such authority is to “cleanse” the language from
impurities.
The final section, entitled “New challenges, new approaches”, explores
areas of vocabulary absent so far from academic analysis, either because
they correspond to recent technological developments, or because scholars
might have feared to tread into controversial territory. In the first study,
“Swingvergüenzas A Contra Blues: A Study On Creative Code Mixing In
Spanish Music”, cases of linguistic creativity are analysed, with a number
of word-formation processes half-way between code-mixing and true
borrowing whose stylistic effectiveness makes them attractive to music
audiences. The second paper, “New Challenges in the Translation of
Language for Software Applications”, looks into the localisation of
software applications and the lexical problems such translation process
x Foreword

entails and provides numerous examples to illustrate the major challenges

translators face when adapting these terms into Spanish by using different
lexical resources. The third contribution, “The Terminology of The Video
Games Market: A New Type of Specialized Language” deals with a
completely new product, whose innovativeness can only be equalled by its
financial weight as a thriving area, and whose terminology may be
challenging for those translators not familiar with the whole concept.
Finally, “The Multiple Shades of Erotica: Translating Romantic and Erotic
Fiction into Spanish” is a daring study of the lexical choices made by
translators faced with erotic content, focusing on one of the most recent
literary best-sellers whose explicit language regarding sexual organs and
practices forces the translator to open new paths in the literary vocabulary
in the target language.
 PART I

LEXICAL THEORY AND ACQUISITION

 CHAPTER ONE

“YOU SHALL KNOW A COLLOCATION

BY THE COMPANY IT KEEPS”:
METHODOLOGICAL ADVANCES
IN LEXICAL-CONSTELLATION ANALYSIS

MOISÉS ALMELA

Introduction
Traditionally, collocation has been conceived of as a bipartite structure.
In all the dominant approaches to collocation, the structure of this type of
word combinations is analysed into two parts  though not necessarily into
two words, because one of the two parts of a collocation can be a complex
item (García-Page 2011). In fact, as Martin (2008) remarks, the notion that
collocation is made up of two parts is one of the few points over which
most experts in the field are generally agreed.
This is not to deny that there are fundamental discrepancies concerning
the way the two parts are categorised. Thus, the distinction between base
and collocator in the literature on phraseology establishes a hierarchical
relation between an autonomous item and a dependent element (Hausmann
1979, 1990, 1998; Írsula Peña 1994; Liang 1991), while the distinction of
node and collocate in corpus linguistics is purely methodological and
distinguishes only between input and output in the process of collocation
extraction (Jones and Sinclair 1974; Krishnamurthy 2004; Phillips 1985;
Sinclair 1991). However, over and above these differences, there is a
common ground shared by all these approaches. All of them describe
collocation as a relation between two parts.
The idea of collocation as a bipartite structure has been called into
question by research conducted in the framework of the Lexical
Constellation model (hereinafter, LCM), developed by members of the
LACELL research group at the University of Murcia (Cantos & Sánchez,
2001; Almela 2011a; Almela et al. 2011a, 2011b). Central to the LCM
4 Chapter One

programme is an attempt to optimise the methods of semantic description

used in corpus-based lexicology. In pursuing this goal, previous LCM
research has concluded that the established dualism of node and collocate
is not suitable for capturing the complexity of collocational relations. The
main reason for this is that the strength of attraction between a node and a
collocate cannot be established independently of the effects produced by
other collocations of the same node. Besides, these effects have implications
for the analysis of meaning.
Hence, where the Firthian motto reads: “You shall know a word by the
company it keeps”, the LCM adds: “Collocations, too, shall be known by
the company they keep”. In fact, much of the company attributed to
individual words in collocational studies is in reality attributable to an
interplay of different collocational patterns. While collocational research
has mostly been concerned with investigating the effect of a word on
neighbouring words, LCM research has directed the attention towards the
effects that a collocation produces on neighbouring collocations. For an
overview of theoretical foundations of the LCM and potential applications
in the field of lexicography, the reader is referred to Almela et al. (2011b).
The main difference between the LCM and other approaches to the
phenomenon of interlocking collocation lies in the account of interactions
among bi-grams. Like lexical constellations, collocational networks
(Williams 1998, 2001; Alonso et al. 2008) and collocation chains (Alonso
Ramos and Wanner 2007) represent forms of co-occurrence patterning
where two or more collocations share one of their elements. With respect
to these approaches, the specific contribution of the LCM lies in the
development of a method for describing dependencies among different
collocates of a node. In addition to stating the fact that two or more
observed collocations have an element in common, the LCM apparatus is
suited to determine whether the presence of one of these collocations
increases or diminishes the probability of the other.
Further differences include the fact that collocational networks are
conceived to describe the vocabulary of specialised sublanguages, while
the LCM has so far been applied to the description of general English.
Also worth mentioning is the difference with respect to the theoretical
background research into collocation chains. The study of collocation
chains carried out by Alonso Ramos & Wanner (2007) is informed by
Meaning-Text Theory, whereas the LCM adopts a usage-based approach
to language.
The aim of this study is to present a methodological innovation in the
LCM. The new version presented in this paper introduces a step consisting
in the use of clustering techniques. This step is aimed at guiding the
 Methodological Advances in Lexical-Constellation Analysis 5

description of the semantic relations underlying the network of co-

collocations (i.e. collocations that are strengthened or co-activated by the
presence of other collocations of the same node).
The paper is formally structured as follows. The next section provides
a summary of the LCM methodological framework and explains the new
step introduced in this study. Then, section 3 will present the new
methodology at work. The collocational patterns analysed include
interactions of verbs and premodifiers in the lexical environment of the
noun decision. Finally, section 4 will discuss the findings and make
suggestions for further research.

The Method of Constellational Analysis

At present, the method applied in the LCM consists of three main
steps: (i) extraction of collocates, (ii) identification of inter-collocability
relations, (iii) semantic grouping and analysis of co-collocates. The first
step involves simply an extraction of statistically significant co-
occurrences (collocates) of a node word. At this point there is no
difference with respect to standard practices of collocation extraction in
corpus linguistics. The specificity of constellational analysis is introduced
in the second step. It is at this point where the distinction between the
categories of collocate and co-collocate is established.
LCM research has identified so far two main forms of co-collocation,
that is, of inter-collocational dependency (Almela et al. 2011b). The first
one is positive inter-collocability, and the second one is negative inter-
collocability.
Positive inter-collocability obtains in cases where one collocation
makes a contribution to the activation of another collocation of the same
node. For example, the probability that reject goods converges with faulty
goods is higher than the probability of reject co-occurring with goods.
This can be interpreted as an indication that the selection of one of these
collocations favours the selection of the other.
Negative inter-collocability obtains when the collocability of a node
and a collocate is restricted by the presence of other collocates of the same
node. For example, the probability of ship goods converging with faulty
goods is considerably lower than the probability of ship co-occurring with
goods. This indicates that the selection of one of these collocations repels
the selection of the other.
The technique for detecting cases of positive and negative co-
collocation is based on comparisons of conditional probabilities (Almela et
6 Chapter One

al. 2011b). The values compared in this phase of constellational are the
following ones:

x The probability that a collocate of the node is selected given as a fact

the co-occurrence of the node and another collocate: P(c1|n,c2), where
n stands for the node, and c1 and c2 represent two different collocates;
x The probability that the same collocate (c1) is selected given as a fact
the occurrence of the node: P(c1|n).

For the sake of brevity, we will refer here to the first value as P1, and
to the second one as P2. P1 can be described as a value of conditional
probability at the inter-collocational level, and P2 as a value of conditional
probability at the intra-collocational level. If P1 is higher than P2, we can
say that c1 is a positive co-collocate of c2 relative to n (Almela et al.
2011b). To render the terminology more symmetrical, we can further add
that c2 is a positive co-node of c1 relative to n. This formulation expresses
the fact that the collocation of c1 and n is made more probable (or
strengthened) by the presence of c2. An additional requirement for positive
inter-collocability is a frequency threshold. If the frequency of the 3-gram
(n,c1,c2) is lower than 2, the combination is excluded from being a
candidate for positive inter-collocability.
Negative inter-collocability obtains when P1 is lower than P2. In this
case we can say that c1 is a negative co-collocate of c2 relative to n, and
conversely, that c2 is a negative co-node of c1 relative to n. This
formulation expresses the fact that the collocation of the node with c2
diminishes the probability of finding the collocation (n,c1).
Like the distinction between node and collocate (see Introduction), the
distinction between co-node and co-collocate is purely methodological: it
depends on which collocational pattern has been used as input and which
one has been obtained as output. If we are investigating the contexts of the
pattern faulty goods, return will be obtained as a positive co-collocate, and
faulty will be the co-node of return, but if we decide to investigate the
contextual effects of return goods, it may turn out that we also obtain
faulty as a positive co-collocate of return. The same applies to negative
inter-collocabilitymore on this issue in Almela et al. (2011b). Therefore,
it is important to emphasise that inter-collocability can be mutual, although
it does not have to. It may operate in the two directions, from c1 to c2 and
vice versa, or only in one direction. From the fact that c1 is a positive or
negative co-collocate of c2 it does not necessarily follow that c2 must also
be a positive or negative co-collocate of c1. The reason for this is that
conditional probabilities are directional. The probability of finding a word
 Methodological Advances in Lexical-Constellation Analysis 7

or expression a in the context of another word or expression b can be

considerably higher or lower than that of finding b in the context of a.
In earlier versions of the method, the third step in constellational
analysis consists in the semantic grouping of co-collocates and the lexical
description of the resulting structures. Unlike the previous steps, this task
is not automatised and consequently depends more on intuition. Despite
this weakness, the step is absolutely necessary to account for a
fundamental aspect of constellations, namely, the underlying semantic
motivation of inter-collocational dependencies. Previous studies have
shown that words from the same conceptual domain will also tend to share
a substantial amount of their co-collocates. This suggests that lexical
constellations may well be interpreted as representing specific surface
realisations of more abstract semantic structures (Almela et al. 2011a).

Figure 1. Some lexical constellations of goods

For example, verbal collocates of goods that express a process of

TRANSPORTATION tend to function as positive co-collocates of
adjectives expressing DANGER. Meanwhile, adjectival collocates of
goods that describe some kind of FLAW in a product tend to function as
co-nodes of verbs that describe a decision of NON-ACCEPTANCE of the
goods. These two patterns of co-collocation are graphically represented in
Figure 1. The node is emphasised in bold type and capital letters, and each
line stands for a relationship of statistically significant co-occurrence
between the node and a collocate. The arrows represent relations of
8 Chapter One

positive inter-collocability, and the circles have been used to group

together words that are related within the same conceptual domain. Thus,
following the steps indicated in Almela et al. (2011a), the lexical
constellations displayed in Figure 1 can be turned to lexico-conceptual
constellations, as in Figure 2, where the link of positive inter-collocability
holds between semantic groups, and not just between individual items.

Figure 2. Some lexico-conceptual constellations of goods

The semantic motivation of inter-collocability paves the way to

interesting advances in the description of lexical structure. One of the
reasons why collocational studies have proven so fruitful to lexicology is
their capacity for identifying correlations of distributional classes and
semantic classes. For example, we know that nouns occurring as direct
objects of the verb face tend to share an aspect of their meaning. The list
includes difficulty, challenge, crisis, dilemma, threat, problem, etc. This
set of nouns constitutes both a distributional class and a semantic class. It
is a distributional class because they share a distributional feature: they are
statistically significant co-occurrences of the same item (face) and in the
same slot (direct object). At the same time, they represent a semantic class,
because they form a semantically coherent set. They all refer to concepts
related to DIFFICULTY.
To a large extent, the potential of collocational data for providing
semantically relevant information is owed to this alignment of
distributional classes and meaning groups. However, the standard
 Methodological Advances in Lexical-Constellation Analysis 9

techniques of collocational description had allowed us to detect such

correlations only at an elementary, coarse-grained level of analysis. It was
possible to identify semantic sets of collocates, but not semantic sets of co-
collocates. In this respect, constellational analysis supersedes the
traditional methods of collocation-based analysis of meaning. The
technique applied in the LCM makes it possible to identify correlations of
distributional and semantic classes at more than one level (i.e. not only
between nodes and collocates, but also between co-nodes and co-
collocates).
For example, the traditional techniques of corpus collocational analysis
determine that tough and difficult are collocates of decision. In the ukWaC
corpus (at SketchEngine) these words form statistically significant pairs
with decision within a 4:4 collocational span. The two pairs form
statistically significant co-occurrences with three different measures
(logDice, MI, T-score). What these techniques fail to tell us is that the
presence of tough and difficult in the context of decision is motivated not
by decision alone, but by interactions of the collocations tough/difficult
decision with other collocational patterns of decision. While there are
verbs that avoid co-collocations with tough/difficult decision for
example, in the ukWaC the frequency of occurrence of review, reconsider
or reverse in this context is zero there other verbal collocates of decision
that are strongly associated with this pattern of inter-collocability. Thus,
the verb face is a prominent co-node of tough and difficult. The probability
that tough is selected as premodifier of decision is increased by almost 40
times if the collocation converges with face. A similar difference is
obtained in the case of difficult.

P(tough|face,decision) = 38/304 = 0.125

P(tough|decision) = 962/302679 = 0.003
P(difficult|face,decision) = 80/304 = 0.263
P(difficult|decision) = 2034/302679 = 0.007

In the light of these data, it would be inaccurate to say that tough is a

collocate of decision without specifying that it is also a co-collocate of
face, because the probability that tough is selected as a premodifier of
decision increases with the presence of the verb face but decreases with
the presence of other verbal collocates of decision, such as review,
reconsider, or reverse, inter alia.
Ultimately, the method of lexical-constellation analysis arises from a
revision of the concept of lexical gravity. Mason (2000: 270) defined
lexical gravity as “the restriction a word imposes on the variability of its
10 Chapter One

context”, but as Cantos & Sánchez (2000) remarked, the trouble with this
concept is that, contrary to what collocational studies have often assumed,
the node does not exert an unlimited influence on its environment. The
restrictions imposed on the variability of the context of a word are shaped
by an interplay of several factors, including the attraction between
different collocational pairs (and not just between words). What is more
important for lexical studies is that the patterns of interaction among
collocates of a node are usually motivated by an underlying semantic
structure.
To continue with the example above, the fact that tough and difficult
are positive co-collocates of face is not unrelated to the fact that face
(again as a verb) collocates with nouns that share a same semantic feature
(DIFFICULTY) with the adjectives tough and difficult (e.g. face +
difficulty, challenge, crisis, dilemma, threat, problem, etc.).
At present, a major weakness of the LCM method for grouping co-
nodes and co-collocates into semantic sets is that it relies mainly on the
analyst’s intuition. The proposal we submit in this paper is intended to
minimise the subjective component in this task and strengthen the
commitment of the model to an objective methodology. One way to
achieve this goal is to base the semantic grouping of co-nodes on
classifications obtained from hierarchical cluster analysis. The
amalgamation of co-nodes according to clustering techniques can be used
as a guide to the manual amalgamation of co-collocates.

A Case Study: Lexical Constellations of Decision

In this section the analytical framework sketched out above is applied
to the description of lexical constellations formed around the noun
decision. The methodological decisions adopted in this case study are
similar to those applied in the analysis of lexical constellations of goods in
earlier research (Almela 2011a), except for the fact the present study
introduces the use of cluster analysis as previous step to the semantic
analysis of co-nodes and co-collocates.
The data and the examples have been extracted from the ukWaC corpus
(1,565,274,190 tokens), accessible at the SketchEngine query system. The
analysis is focused on capturing features of inter-collocability in verb +
noun and premodifier + noun collocations. Previous research has proven
successful in detecting cases of positive inter-collocability within this
grammatical framework (Almela 2011b; Almela et al. 2011a). Therefore,
all queries are syntactically restricted. We have taken into account only
occurrences of the noun phrase (i.e. the adjective-noun collocation) as a
 Methodological Advances in Lexical-Constellation Analysis 11

direct object of the verb in an active construction, or as the subject in a

passive construction (the connection between the two constructions is that
in both cases the collocation premodifier + decision performs the same
semantic role). The WordSketch function has been useful in limiting our
queries to the foregoing grammatical scheme. Nevertheless, manual
supervision was required in order to detect possible parsing errors.
As instructed in the previous section, the method of constellational
analysis applied in this study consists of four steps. Each of these steps is
described in detail in the subsections below (3.1 to 3.4).

First Step: Collocation Extraction

The first step was to extract a list of verbal and modifier collocates of
decision. Statistical significance was defined in terms of logDice  for an
explanation of the advantages of this measure, cf. Rychlý (2008). The
number of verbal collocates was limited to 25 due to limitations of space,
given that a dendrogram containing more than 25 variables in this case,
the variables are verbs would not fit in one page (see Figures 4 and 5 in
the next subsection). Table 1 shows the list of 25 top verbal collocates of
decision arranged in order of decreasing logDice score. The list of
adjectives was allowed to be larger because a higher number of cases does
not affect the size of the dendrogram but helps to make the analysis more
accurate. Therefore, the list of premodifiers was extended to include 50
items (Table 2). Again, the collocates in the table have been arranged in
order of decreasing logDice score.

1. make 6. reverse 11. challenge 16. await 21. confirm

2. reach 7. appeal 12. announce 17. uphold 22. defend
3. influence 8. reconsider 13. review 18. regret 23. affect
4. inform 9. justify 14. welcome 19. implement 24. issue
5. take 10. overturn 15. defer 20. delay 25. explain

Table 1. Top collocates of goods (grammatical relation: “object_of”)

12 Chapter One

1. informed 11. recent 21. rational 31. quick 41. big

2. final 12. key 22. initial 32. correct 42. bad
3. tough 13. planning 23. wrong 33. buying 43. hard
4. conscious 14. wise 24. political 34. ultimate 44. tribunal
5. right 15. important 25. crucial 35. deliberate 45. subsequent
6. strategic 16. judicial 26. sensible 36. formal 46. tactical
7. unanimous 17. controversial 27. clinical 37. ethical 47. momentous
8. difficult 18. own 28. collective 38. future 48. early
9. purchasing 19. policy 29. court 39. brave 49. funding
10. investment 20. original 30. major 40. executive 50. majority

Table 2. Top collocates of goods (grammatical relation: “modifies”)

Second Step: Identification of Co-Collocates

In a second step, the values of conditional probabilities were calculated
and compared following the procedure explained in section 2. Table 3
shows a sample of the data used for identifying co-collocates of one of the
verbs, overturn. The first column is a list of potential co-collocates. The
next column indicates the raw frequency of the whole combination (verb,
premodifier, noun) in the corpus (for instance, the frequency of overturn
an earlier decision). The third column provides the raw frequency data of
the collocational pair formed by the noun (decision) and each of the
collocates listed in the left-most column (for instance, the frequency of the
collocation original decision in the corpus is 721).

f(v,m,n) f(m,n) P(m|v,n) P(m|n)

original 23 721 5.00% 0.24%
early 5 669 2.83% 0.22%
tribunal 13 189 1.09% 0.06%
court 2 653 1.09% 0.22%
initial 4 477 0.87% 0.16%
executive 5 207 0.43% 0.07%

Table 3. Frequency and probability data for calculating co-collocates

of overturn

Along with the frequency of the noun, which is not shown in the table
because it remains constant (302679) in all the rows, the frequency data in
the second and in the third columns are necessary in order to calculate the
values of conditional probabilities in the remaining columns. The fourth
column returns the value of P(m|v,n), that is, the probability that the
 Methodological Advances in Lexical-Constellation Analysis 13

premodifier occurs given as a fact the occurrence of the verb + noun

collocation; the second one specifies the probability that the premodifier
occurs given the selection of the noun, which is formally expressed as
P(m|n). Hence, these two columns return the values labelled as P1 and P2
in section 2.

Figure 3. Comparing intra- and inter-collocational conditional

probabilities from Table 3

The order of the rows is determined by the difference between the

values of the last two columns. Thus, the premodifier at the top of the list
is the best candidate for positive co-collocation. Original is the
premodifier with the highest difference between the value of P1 (inter-
collocational conditional probability) and of P2 (intra-collocational
conditional probability). The difference between these two values is
graphically represented in Figure 3. The top line represents P1, and the
bottom line represents P2. The figure shows that all the premodifiers
analysed in this table are positive co-collocates of overturn relative to the
noun decision, and that original is the most prominent one.

Third step: cluster analysis

The third step in the new version of the model is cluster analysis. Using
SPSS 19 and selecting square Euclidean distance as a measure, the verbs
were grouped into clusters. In the settings, the verbs were entered as
variables, and the premodifiers as cases. The reason for not doing the
opposite can be explained with reference to figures 4 and 5. The
dendrogram in Figure 4 shows the results from clustering the 25 top
14 Chapter One

logDice premodifiers of decision according to their distribution in

collocations of decision with its top 50 logDice verbs (always within the
syntactic pattern specified at the beginning of this section). This clustering
is much less discriminatory than the one displayed in the dendrogram of
Figure 5, which shows the clustering of 25 top logDice verbal collocates of
decision according to their distribution in collocations of decision with its
top 50 logDice premodifiers. This is the main reason why in this study it
was decided to treat the verbal collocates as potential co-nodes and the
premodifiers as potential co-collocates, rather than the other way round.
The results displayed in the dendrogram of Figure 5 suggest the
existence of three clusters of verbs according to their distribution in
lexicogrammatical contexts of the type ‘verb + premodifier + decision’:

x First cluster: {overturn uphold confirm reverse reconsider review

defend regret appeal challenge welcome}
x Second cluster: {influence inform affect justify implement}
x Third cluster: {defer delay announce await reach issue take explain
make}

The first cluster is semantically more homogeneous than the other two.
All the verbs in the first cluster express some aspect of a REACTION to a
decision. In three verbs from the cluster, the reaction is described in terms
of an axiological value, i.e. a judgment of APPROVAL or DISAPPROVAL:
defend, challenge, welcome. In the remaining verbs from the cluster, the
reaction described implies a process of REVISION. This subframe is
organised around a temporal sequence. One of the verbs (appeal) refers to
the initial stages, when the revision is demanded but still not undertaken,
let alone completed. Two other verbs, reconsider and review, denote the
process of undertaking the revision itself. Finally, the event denoted by
four other verbs implies that the revision has already been completed.
Different verbs denote different outcomes of the process of revision:
confirm and uphold express an action by which the earlier decision is
affirmed, while reverse and overturn express an action by which the
earlier decision is altered or cancelled.
 Methodological Advances in Lexical-Constellation Analysis 15

Figure 4. Cluster analysis of 25 premodifiers

More differences among the verbs from this cluster arise from their
degree of contextual specialisation. Uphold, overturn and appeal are
highly characteristic of the legal jargon, while the remaining verbs are more
versatile in this respect. However, these are minor differences which do not
alter the conclusion that the verbs from this cluster form a consistent
semantic set structured around the central notion of REACTION to a prior
decision. Therefore, there are good reasons to predict that the verbs from this
cluster will tend to co-activate collocations describing decisions made in the
past. The answer to this prediction will be given in 3.4.
16 Chapter One

Figure 5. Cluster analysis of 25 verbal co-nodes

The second cluster is smaller, as it contains only five verbs. Three of

these verbs have clearly related meanings: influence, inform and affect
express ways of exerting some form of INFLUENCE on a decision. Thus,
INFLUENCE can be deemed to constitute the central conceptual domain
of the cluster. Nevertheless, it has to be conceded that the other two verbs,
justify and implement, are less directly connected to this domain.
Considering that the domain INFLUENCE conveys a semantic dimension
of PROSPECT (i.e. projection of possible events or state of affairs in the
future), it is reasonable to expect that verbs from this cluster will tend to
function as co-nodes of premodifiers referring to FUTURE EVENTS. The
 Methodological Advances in Lexical-Constellation Analysis 17

answer will be given in section 3.4, when the whole process of

constellational analysis is completed.
Lastly, the third cluster is less homogeneous than the other two. It
contains two verbs of COMMUNICATION (announce, issue) and two
verbs expressing POSTPONEMENT (defer, delay), plus three support
verbs (make, take, reach) that as such are not associated with any specific
conceptual domain. An indication of their status as support verbs in this
context is the fact that the collocations take/make/reach a decision are
equivalent to the simple verb decide, as far as denotation is concerned. The
semantic relation expressed by these verbs corresponds to the lexical
function Oper1 in Meaning-Text Theory (see Melþuk 1998). One member
of the cluster, await, cannot be pigeonholed into any of the aforementioned
categories.

Fourth step: co-node and co-collocate grouping

In the fourth and last step the clusters obtained from the previous step
are used as guidelines for the classification and description of inter-
collocability relations. Put differently, the amalgamation of co-nodes
according to clustering techniques in the third step is used now as a
template for the manual amalgamation of co-collocates.
Following this procedure, the clusters of verbs obtained from the third
step should be used now as a starting point for merging and dividing
premodifiers into groups. The expectation was that co-nodes (in this case:
verbs) that are grouped together in the same cluster would tend to select
similar groups of co-collocates (in this case: premodifiers). If distributional
classes are closely aligned with semantic classes, as is normally the case,
we can predict that verbs from the same cluster will tend to activate
premodifiers from related conceptual domains. Although the combination
of verbs and premodifiers does not form part of any established
grammatical typology of collocation, previous LCM research has shown
that these two categories can interact and form semantic patterns (Almela
2011b; Almela et al. 2011b). Therefore, the classification of premodifiers
according to clusters of verbs should lead, in principle, to coherent results.
Table 4 contains the lists of positive co-collocates of verbs
amalgamated in the first cluster (negative inter-collocability has not been
taken into consideration in this empirical study). Within each list the
premodifiers have been arranged in order of decreasing prominence, that
is, of decreasing difference between the values of inter-collocational and
intra-collocational conditional probabilities (the same criterion has been
applied in tables 5 and 6 for verbs from the second and the third cluster).
18 Chapter One

The results displayed in Table 4 are highly consistent and conform to the
initial expectations that cluster analysis is an adequate starting point for
grouping co-collocates under specific groups of co-nodes. Thus, the first
pair of verbs to be joined together by a cluster, overturn and uphold, are
also those that share exactly the same pattern of positive co-collocation.
The lists of premodifiers associated with these verbs are identical. Even
the order in which the premodifiers are arranged is exactly the same in
both lists.

verbs (co-nodes) premodifiers (co-collocates)

original, early, tribunal, court, initial, executive,
overturn
controversial
original, early, tribunal, court, initial, executive,
uphold
controversial
confirm original, early, controversial, funding, initial
reverse early, original, court, initial, controversial, funding
reconsider early, original
original, recent, own, early, initial, controversial, tribunal,
review
formal, funding
defend original, controversial, difficult
regret early
appeal court, tribunal, planning, original, bad, early
challenge tribunal, court, planning, executive, bad, original
welcome recent, tribunal, court, unanimous

Table 4. Co-collocates associated with verbs from the first cluster

More generally, we can observe that the most repeated premodifiers in

this cluster are related to two main conceptual domains. The first one
includes premodifiers that are used to describe PAST EVENTS: original,
early, initial, recent. All the verbs from this cluster are associated with at
least one of the premodifiers from this list, and many verbs are associated
with three of them. This is the case of overturn, uphold, confirm, and
reverse, all of which exhibit positive co-collocation with early, original
and initial. Notice also that review is associated with the four premodifiers
from this group.
The second conceptual domain that is commonly found though to a
lesser extent among co-collocates associated with the first cluster is
structured around the notion LEGAL SYSTEM, more specifically in
connection with the setting COURT OF LAW: tribunal and court (used as
premodifiers) are positive co-collocates of five verbs from this cluster
 Methodological Advances in Lexical-Constellation Analysis 19

(overturn, uphold, appeal, challenge, welcome). Besides, court occurs in

the list for reverse, and tribunal in the list for review.

verbs (co-nodes) premodifiers (co-collocates)

policy, future, political, purchasing, investment, buying,
influence subsequent, planning, strategic, key, major, clinical,
judicial, funding, initial
future, policy, investment, purchasing, funding, clinical,
inform
planning, buying, strategic, subsequent, ethical
investment, purchasing, future, policy, planning, buying,
affect
funding, major
justify political, funding, policy, investment, clinical
implement policy, strategic, tribunal, key, court

Table 5. Co-collocates associated with verbs from the second cluster

The discrimination between the first and the second cluster of verbs is
convincing. By comparing tables 4 and 5 we can arrive at the following
conclusions: firstly, the two clusters are highly coherent internally; and
secondly, there is a noticeable semantic contrast between them, which
speaks for the discriminatory power of the technique applied. While the
prevailing conceptual domain in co-collocates associated with the first
cluster is PAST EVENT, the dominant domain in the second cluster is
exactly the opposite, that is, FUTURE EVENT. All the verbs from this
cluster are co-nodes of premodifiers that in some or other way describe an
action of planning or a projection of future events. This is obvious in the
case of premodifiers such as future, planning, and investment. The same
semantic dimension is less obvious though still present in premodifiers
such as strategic, policy, political. These three words relate to an action of
planning, and the notion PLANNING in turn contains a reference to
FUTURE EVENTS. In a different sense, the word subsequent is also
related to the notion of FUTURE EVENT, because it describes a temporal
sequence.
Another conceptual domain that is repeated in the second cluster is the
TRANSACTION frame. Buying, purchasing and investment (used as
premodifiers) are perfect examples of this category  note that investment
is related simultaneously to the two dominant conceptual domains in the
cluster, since it combines the semantic features PLANNING and
TRANSACTION. Funding is also related to this domain because, after all,
it also makes reference to a transfer of capital.
20 Chapter One

verbs (co-nodes) premodifiers (co-collocates)

defer final, funding, investment, hard, big
delay final, planning, funding, purchasing, policy, key
announce final, funding, majority, formal, unanimous, major
await final, initial, court
final, own, unanimous, informed, correct, collective,
reach
planning, majority, sensible, formal, initial, rational
issue final, formal, recent
tough, final, strategic, difficult, major, brave, key, policy,
hard, deliberate, important, political, formal, executive,
take
conscious, momentous, crucial, sensible, early, collective,
ultimate, right, controversial, tactical
explain (none)
informed, right, own, final, conscious, important, wrong,
purchasing, quick, correct, difficult, investment, rational,
make strategic, sensible, wise, bad, crucial, key, major, ethical,
tough, buying, momentous, executive, big, clinical, initial,
tactical, deliberate, collective, ultimate

Table 6. Co-collocates associated with verbs from the third cluster

Unfortunately, the third cluster is slightly more heterogeneous than the

other two (see Table 6). Besides, the discrimination between this cluster
and the previous ones is less convincing than that between the first and the
second one. Put briefly, the problem is that the third cluster contains many
items that were found to characterise elements from the first cluster or
from the second cluster (e.g. early, initial, planning, funding, investment,
policy, political).
In part, this can be explained by the fact that the third cluster contains
three ‘support’ verbs: take, make and reach. Support verbs are largely
delexicalised and semantically underspecified. This also implies that their
collocational behaviour is highly versatile  that is, they occur in a wide
range of lexical contexts. This remark concerns their relations not only to
collocates but also to co-collocates. Compared to the other verbs in the
cluster (the purely ‘lexical’ verbs), the lists of positive co-collocates
attributed to the support verbs particularly to make and take, which are
more frequent than reach is significantly larger and more heterogeneous.
 Methodological Advances in Lexical-Constellation Analysis 21

Figure 6. Positive inter-collocability relations among collocates of

decision

Yet, in spite of the presence of support verbs, there are some features
that lend a certain degree of cohesiveness to the co-collocates derived from
the last cluster. The most noticeable feature is that the first six verbs to be
joined together in the cluster (defer, delay, announce, await, reach, issue)
have the adjective final as their most prominent positive co-collocate.
Moreover, there seem to be two dominant conceptual domains that are
specific to this cluster. The first of them encompasses adjectives that
describe the RATIONALITY of a decision: informed, rational, sensible,
deliberate, conscious. The second group includes adjectives that depict a
COLLECTIVITY as the agent of the decision: unanimous, collective,
majority. The three words from this group are positive co-collocates of
reach.
The end result of the whole process is sketched in Figure 6. This figure
offers a general picture of the most systematic patterns of co-collocation
22 Chapter One

obtained from the semantic analysis of the three clusters and of the
corresponding lists of co-collocates. Like in figures 1 and 2, the lines stand
for statistically significant co-occurrences of the node, and the arrows
symbolise links of positive inter-collocability. These arrows are oriented
from co-nodes to co-collocates.

Concluding remarks
The main objective of this study has been to present and put into
practice a new version of the method of constellational analysis. The
investigation was motivated by the suggestion that cluster analysis might
provide a suitable basis for the classification and analysis of co-collocates.
The evolution can be summarised as follows: in earlier versions of the
model the analyst had to classify the co-collocates with the only guidance
of his or her intuition; in the new version, the classification of co-
collocates is assisted by the cluster solution obtained for their co-nodes.
The theoretical underpinning of this strategy resides in the principle that
distributional classes tend to be closely aligned with semantic classes, and
consequently, that verbs from the same cluster will tend to activate
premodifiers from related conceptual domains.
In general, the results obtained are satisfactory, because the analysis of
co-nodes that were amalgamated into clusters in the third step of the
methodology dovetail with those obtained from the analysis of co-
collocates in the fourth and last step. This consistency observed between
the results obtained from the two steps can be summarised in two points.
Firstly, it has been observed that the least homogeneous group of verbs
(those from the third cluster) is also attributed the least homogenous group
of co-collocates. It is not difficult to explain why the support verbs make
and take admit a much broader range of positive co-collocates than the
fully ‘lexical’ verbs of the list. The reason is that make and take are high-
frequency items and that they are delexicalised and semantically
underspecified.
Secondly, there is a remarkably coherent relation between the
conceptual domains assigned to specific groups of co-nodes and the
conceptual domains that prevail in the corresponding groups of co-
collocates. Thus, verbs related to the REACTION domain (first cluster)
favour the co-activation of premodifiers related to the PAST EVENT
domain. Likewise, verbs that are related to the INFLUENCE domain tend
to favour the selection of premodifiers describing FUTURE EVENTS or
PLANS. These conclusions reinforce the idea that cluster analysis is a
suitable starting point for classifying co-collocates into semantic sets.
 Methodological Advances in Lexical-Constellation Analysis 23

Nevertheless, the method put forward in this study does not entail a
suppression of the analyst’s intuition. The method is intended to impose a
stricter control on the use the subjective component, but not to eliminate it.
The analyst’s intuition is still needed at specific points of the analysis, for
instance in cases where it is necessary to divide each group of co-
collocates into smaller semantic subsets. This point is illustrated by the
way the case study has been carried out. A detailed analysis of the
dominant conceptual domains in each cluster would not have been
possible without recourse to intuition. However, the role of intuitive
judgements has been limited to a relatively fine-grained level of
description and, crucially, this task has been preceded and informed by a
coarser-grained classification that has been carried out automatically and
on more objective grounds. In this sense, the process applied in this study
can be characterised as semi-automatic. The clustering technique applied
in this study supplies us with reliable coarse-grained divisions to get
started, but later, when it comes to drawing finer-grained delimitations
among conceptual domains, the intervention of the human analyst appears
to be unavoidable.
One of the main limitations with which the present study must be
confronted is that the analysis of clusters was performed with 25 variables.
Adding more variables to the analysis could have served to render the
cluster solution more discriminatory and solve difficulties such as those
encountered in third cluster of verbal collocates of decision, whose co-
collocates overlap with some items derived from the first and the second
cluster. Thus, a suggestion for future research is that the same empirical
case study be replicated with a larger number of verbs used as variables.
Another limitation that future research can resolve is the fact that this
study has been applied only to a subset of the collocational field of
decision. The study has taken into account only interactions between
verbal collocates and modifiers of the head noun in the THEME
constituent. The reason for having circumscribed the case study to a
specified subset of collocational patterns was simple: as a pilot study, it
was advisable to test the model on a small-scale set of data before applying
it at larger scale. Now that the preliminary results have lent credibility to
the model, our next steps should lead to the study of interactions among
collocates in a broader range of syntactic patterns around the nodefor
instance, interactions between verbal collocates and modifiers in the
AGENT noun phrase, or between predicative adjectives and the subject
head noun, to quote only some of the possibilities that can be explored in
later studies. It is reasonable to expect that, with a higher number of
variables and a broader range of contextual patterns, the replication of this
24 Chapter One

study in forthcoming LCM research will lead to more accurate results and
more revealing findings.

Acknowledgements
This study presents preliminary results from a research project funded
by the Spanish Ministry of Education (Ref. FFI2012-38724). We are
grateful for this financial support.

Bibliography
Almela, M. (2011a). Improving corpus-driven methods of semantic
analysis: a case study of the collocational profile of incidence. English
Studies 92 (1), 84-99.
—. (2011b). The case for verb-adjective collocations: corpus-based
analysis and lexicographical treatment. Revista de Lingüística y
Lenguas Aplicadas 6, 39-51.
Almela, M., Cantos, P. & Sánchez, A. (2011a). From collocation to
meaning: revising corpus-based techniques of lexical semantic
analysis. In I. Balteiro (Ed.), New Approaches to Specialised English
Lexicology and Lexicography. Newcastle upon Tyne: Cambridge
Scholars Press, 57-72.
Almela, M., Cantos, P. & Sánchez, A. (2011b). Towards a dynamic
combinatorial dictionary: a proposal for introducing interactions
between collocations in an electronic dictionary of English word
combinations. In I. Kosem & K. Kosem (Eds.), Electronic
Lexicography in the 21st Century: New Applications for New Users.
Proceedings of eLex 2011, Bled, 10-12 November 2011. Ljubljana:
Trojina, 1-11.
Alonso A., Millon, A. & Williams, G. (2011). Collocational networks and
their application to an E-Advanced Learner’s Dictionary of Verbs in
Science (DicSci). In I. Kosem & K. Kosem (Eds.), Electronic
Lexicography in the 21st Century: New Applications for New Users.
Proceedings of eLex 2011, Bled, 10-12 November 2011. Ljubljana:
Trojina, 12-22.
Alonso Ramos, M. & Wanner, L. (2007). Collocation chains: how to deal
with them? In K. Gerdes K., T. Reuther & L. Wanner (Eds.),
Proceedings of the Third International Conference on Meaning-Text
Theory, Wiener Slawistischer Almanach, Sonderband 69. Munich: 11-
20.
 Methodological Advances in Lexical-Constellation Analysis 25

Cantos, P. & Sánchez, A. (2001). Lexical constellations: what collocates

fail to tell. International Journal of Corpus Linguistics 6 (2), 199-228.
García-Page, M. (2011). Collocations complexes (application à
l’espagnol). Lingvisticae Investigationes 34 (1), 68-111.
Hausmann, F. (1979). Un dictionnaire des collocations est-il possible?.
Travaux de linguistique et de literature 17 (1), 187-195.
—. (1990). Le dictionnaire de collocations. In F. Hausmann, O.
Reichmann, H.E. Wiegand & L. Zgusta (Eds.), Wörterbücher,
Dictionaries, Dictionnaires: Ein internationals Handbuch zur
Lexicographie / An International Encyclopedia of Lexicography /
Encyclopédie internationale de lexicographie, Vol 2. Berlin: Walter de
Gruyter, 1010-1019.
—. (1998). O diccionario de colocacións. Criterios de organización. In
Actas do I Coloquio Galego de Fraseoloxía (Centro Ramón Piñeiro,
Santiago de Compostela, 1997), 63-81.
Írsula Peña, J. (1994). Substantiv-Verb-Kollokationen. Kontrastive
Untersuchungen Deutsch-Spanisch. Frankfurt a. M.: Peter Lang.
Jones, S., Sinclair, J. (1974). English lexical collocations. A study in
computational linguistics. Cahiers de Lexicologie 24: 15-61.
Krishnamurthy, R. (Ed.) (2004). English Collocation Studies. The OSTI
Report. London: Continuum.
Liang, S. Q. (1991). À propos du dictionnaire français-chinois des
collocations françaises. Cahiers de Lexicologie 59 (2), 151-167.
Martin, W. (2008). A unified approach to semantic frames and
collocational patterns. In S. Granger & F. Meunier (Eds.),
Phraseology. An Interdisciplinary Perspective. Amsterdam: John
Benjamins, 51-65.
Mason, O. (2000). Parameters of collocation: the word in the centre of
gravity. In J.M. Kirk (Ed.), Corpora Galore. Analyses and Techniques
in Describing English. Amsterdam: Rodopi, 267-280.
Mel’þuk, I. (1998). Collocations and lexical functions. In A.P. Cowie
(Ed.), Phraseology. Theory, Analyses and Applications. Oxford:
Clarendon Press, 23-53.
Phillips, M. (1985). Aspects of Text Structure. An Investigation of the
Lexical Organisation of Text. Amsterdam: North Holland.
Rychlý, P. (2008). A lexicographer-friendly association score. In P. Sojka
& A. Horák (Eds.), Proceedings of Recent Advances in Slavonic
Natural Language Processing, RASLAN 2008. Brno: Masaryk
University, 6-9.
Sinclair, J. (1991). Corpus, Concordance, Collocation. Oxford: Oxford
University Press.
26 Chapter One

Williams, G. (1998). Collocational networks: interlocking patterns of lexis

in a corpus of plant biology research articles. International Journal of
Corpus Linguistics 3 (1), 151-171.
—. (2001). Mediating between lexis and text: collocational networks in
specialised corpora. ASp (online) 31-33, 2-12.
 CHAPTER TWO

FROM EFL TO CLIL:

TEACHING AND LEARNING VOCABULARY
IN THE PRIMARY CLASSROOM

MARÍA TABUENCA CUEVAS

AND GEMA ALCARAZ MÁRMOL

Introduction: From EFL to CLIL in Primary Education

The focus on foreign language learning in primary education in Spain
is currently going through many changes. Until recently, English as a
foreign language (EFL) was relegated to one subject taught throughout the
curriculum between two to three hours a week depending on the grade.
Language learning at this level has traditionally been based on the
development of the four skills: speaking, listening, reading and writing
with the use of textbooks in the classroom. According to Nation (2001) a
minimum of 2000 words would be necessary to be able to communicate at
a basic level in a foreign language. At the end of six years of classroom
instruction, students were expected to have acquired sufficient receptive
vocabulary and productive vocabulary knowledge in English. Receptive
vocabulary corresponds to the words a student can understand in contrast
to productive vocabulary which are the words that a student can produce.
This vocabulary can also be grouped into structural vocabulary
(prepositions, auxiliary verbs, etc.) and lexical vocabulary, (less frequent
words) which have more weight of meaning.
Nonetheless, two studies by Jiménez and Moreno (2005), Jiménez and
Terrazas (2008) have shown that primary school students in many cases
only reached a level of 1000 words at the end of sixth grade. These results
bring to the forefront the little attention that is paid to vocabulary in the
classroom reflecting the perspective put forth by Milton (2009: 1) that
“outside the arena of specialist vocabulary studies there seems to be a
28 Chapter Two

long-standing idea that words are just words and that learning words is
unsystematic”.
Current changes within the last five years, that include the introduction
of bilingual Spanish/English programmes in many schools, have shifted
the focus from language learning to learning the content of curricular
subjects using English. To this end Content and Language Integrated
Learning (CLIL) methodology is being used in the classroom. This type of
methodology can be a double-edged sword. On the one hand, students are
expected to improve their skills in a second language. On the other hand,
the change from L1 (Spanish) to L2 (English) as the language of
instruction may increase the degree of difficulty in the acquisition of
content. This added difficulty may lead to the student’s frustration and
lack of motivation, which largely affects the learning process which makes
L2 learning so important in CLIL. In general terms, within the CLIL
methodology, the L2 is expected to be learned implicitly, as it is the
content of the subject which is the main thread of the lesson (Ellis, 1994:
24, cf. Milton, 2009). Vocabulary is acquired by means of reading texts,
listening to the teachers’ explanations and communicating with the teacher
and other students in the L2. This does not mean that explicit reference to
the L2 should be made. In fact, it is recommended to pay attention to
target words or structures that can be especially important for the content
to be taught, especially at low levels (Coyle et al., 2010). Thus, the lack of
emphasis on vocabulary in language acquisition is once more made
evident.
Vocabulary is essential in language learning, and it is especially
important in a CLIL context, where the language is not the object to be
learned but the vehicle through which learning is expected to be
transmitted. Nonetheless, there is no consensus and no focus on the
methodology for vocabulary acquisition, one of the essential stepping
stones in acquiring language proficiency as vocabulary is seen as
“unchallenging as a pedagogical or an academic issue” (Milton, 2009: 1).

Word Frequency
Research has shown that there are different factors that affect how
students acquire vocabulary and one of these is word frequency. The
number of times a student has been exposed to a word has been the basis
for many studies which have tried to ascertain how many times it is
necessary to see a word before it is acquired. Mason & Krashen (1997)
and Grabe & Stoller (1997) argue that exposure to language leads
inevitably to language learning; this is in sharp contrast to studies by
 From EFL to CLIL 29

Coady (1997) who shows that this is not the case. However, the literature
in the field shows there is no consensus on the number of times a word
needs to appear before it is assimilated. The table below summarises some
of the results of studies in the field.

Author Word repetitions

Kachroo (1962) Up to seven
Saragi, Nation & Meister (1978) Ten occurrences
Bunker (1988) At least five
Horst et al. (1998) At least eight
Hulstijn, Hollander & Gradius (1996) Two or three encounters
Two encounters, but six exposures are
Rott (1999)
necessary for the effect to be significant
Reyes (1999) No less than nine occurrences are needed
Nation (1990 Five to sixteen exposures
Waring & Takaki (2003) Ten encounters
Piagada & Schmitt (2006) Twenty occurrences
Webb (2007) Ten encounters

Table 1. Number of word repetitions necessary for language

acquisition (Alcaraz-Mármol, 2010)

This number of word repetitions must not be confused with the idea
that students will automatically integrate new vocabulary idea, or that
vocabulary acquisition is not based on explicit instruction (Harris and
Snow, 2004). Research has shown that exposure alone to vocabulary is not
a guarantee for the acquisition of vocabulary. Studies by Alcaraz-Mármol
(2010: 189) have clearly demonstrated that “intensive and extensive
repetition together with some explicit attention to vocabulary” have shown
better results in retention. At this point it is necessary to discuss what type
of vocabulary students are exposed to in the primary classroom.

Vocabulary in textbooks
As previously mentioned, textbooks for English language are the
primary source for language for students. Recent studies have shown that
the type of vocabulary varies greatly in different textbooks resulting in
uneven exposure to vocabulary and word types (Scholdfield 1991; Scott
2005). Textbooks do not seem to follow a specific pattern in the way
vocabulary is introduced, and no rational is found behind the way
vocabulary is presented. Accordingly, the teaching of vocabulary is dealt
with arbitrarily.
30 Chapter Two

Mancebo (2005) has discovered that textbooks in Spain in general do

not follow the guidelines of the European Framework of Reference
regarding word frequency and word selection, albeit these guidelines are
general. Jiménez and Mancebo (2008) have shown that the focus is on
grammatical terms which generally follow the pattern of nouns, verbs,
adjectives and adverbs in decreasing repetition. They have also compared
textbooks and have found that specific vocabulary changes in different
textbooks, thus stating that there is a clear “lack of systematic criteria in
English L2 textbooks regarding the number of words and the type of
vocabulary contained in them” (p.162). They have called for parameters to
specify what kind of vocabulary students should know at the end of each
educational stage.
Thus, the selection of appropriate vocabulary and its arrangement is
one of the main issues in L2 vocabulary teaching. To this respect we
should have realistic expectations about what learners can learn, what they
cannot learn and what they need to learn. Frequency is the basis of a
pioneering attempt to establish solid criteria in the selection of vocabulary:
the Carnegie Report, which was part of the Vocabulary Control Movement
that arose towards the second half of the 20th century. Although it is
considered a reference, some problems can be found. Some scholars warn
about relying exclusively on frequency for vocabulary selection. One of
the most noteworthy weaknesses of the Carnegie Report is the modest
pedagogical usefulness of frequency lists (Richards, 1974). According to
McCarthy (1990) some of the most frequent words are “empty words”,
devoid of semantic meaning, such as articles or pronouns. Another fact to
consider is that language is in continuous change: new terms appear and
some terms change their frequency of use.
Therefore, it seems that L2 vocabulary teaching should not just be
exclusively based on frequency, as “the single word-frequency list alone is
not sufficient and must be supplemented by psychological considerations”
(McCarthy, 2001). In fact, Schmitt states that “word frequency alone is not
enough as there are many words which all learners […] are almost sure to
need even though they do not occur very high up on frequency lists”
(2000: 270).
In this line, authors such as Allen (1983) and White (1988) claim that a
vocabulary syllabus should be designed on the bases of what the student
needs. We need to identify the vocabulary that best adapts to the learners’
communicative requirements. White (1988) highlights the criteria of
availability, opportunism and centers of interest as the three aspects under
which school syllabi should be designed. That means that the learners’
 From EFL to CLIL 31

communicative needs should be taken into consideration as the first thing

in materials design.
Consequently, another issue which should be discussed is that most
textbooks for EFL in primary education have a set of routine activities
designed for vocabulary acquisition. In general there is little variety in the
types of exercises which include: gap fill, mix and match, and multiple
choice exercises. These exercises provide vocabulary practice once in that
specific activity. In contrast, in order for vocabulary to be acquired Stahl
(2005), among others, points out that repetition (not repetitive activities)
and multiple exposures to vocabulary are necessary. Others like Anderson
& Nagy (1991) point out the need for direct instruction of vocabulary for a
specific text. This leads to the need for vocabulary tasks to be restructured
as necessary (Kamil, 2004) for vocabulary acquisition. The real objective
as described by Stahl & Kapinus (2001) must be the following: that
knowing a word does not refer only to knowing the definition but also the
relationship with other words, and the word functions in different contexts.

Receptive knowledge and productive knowledge

In language learning skills are often not compensated; this is also the
case in vocabulary acquisition. Lexical competence consists of receptive
(R) and productive (P) knowledge. As Nation states: “receptive knowledge
involves perceiving the form of a word while listening or reading and
retrieving its meaning. Productive knowledge involves wanting to express
a meaning through speaking or writing and retrieving and producing the
appropriate spoken and written word form” (2001: 24). Passive knowledge
of lexis is often wider and includes the ability of comprehending core
meanings of lexis. According to Laufer & Nation (1995), productive
knowledge is more complex, as productive knowledge requires
familiarisation with a type of information that the learner has to be able to
recall, whereas receptive knowledge only requires partial information of
that concept.
In fact, we have to bear in mind that, especially at low levels, a new L2
word hardly has any relationship with other L2 words, as L2 words are not
highly integrated, given that their links are not very strong. On the
contrary, connections between L1 words are much stronger, providing the
speaker with the ability to “integrate information across and within
sentences and to generate accurate syntactic and semantic inferences about
words” (Nassaji, 2006: 398).
What seems clear from the research is the importance of vocabulary
knowledge in being able to understand and communicate in a foreign
32 Chapter Two

language. However, we should not think that receptive knowledge of

vocabulary comes completely before productive vocabulary learning. It
seems more appropriate to consider them two overlapping areas within the
learner’s general cognitive system (Melka, 1997). Scholars claim that they
may complement each other (Griffin & Harley 1996; Waring 1997), far
from the idea that these two dimensions of knowledge constitute an
ordered linear continuum.

Implications for CLIL

The importance of vocabulary acquisition in CLIL cannot be
underestimated. All interaction in the classroom is done in the foreign
language. Studies have clearly shown that the number of words a student is
familiar with is essential to following and understanding a lesson. To be
able to follow classes students need to have assimilated between 2000 and
3000 of the most common words of a language (Schmitt, 2000). The end
goal of language acquisition in spoken discourse, as pointed out by Nation
(2006), would be that 6,000-7,000 word families are necessary to
understand spoken text and that 8,000 to 9,000 word families are needed to
understand a written text. Thus, students would be expected to know what
a word means, how to transform it, and when to use it. It stands to reason
that as students are learning content and language is the vehicle, there must
be a special focus on specific vocabulary regarding content. How is this to
be done simultaneously? Nation (2001) has pointed out that knowing a
word involves a complex procedure shown in the table below.
Each section of the table is divided into three parts as well as indicating
productive and receptive knowledge. The first section, Form, includes
knowing what a word looks like and sounds like. In addition, it includes
the knowledge of word parts for word family creation. The second part,
Meaning, which refers to the ability of link the L2 word to the most
accurate word in the L1 maintaining the right connotations. The third
section of the table is often the most complicated for language learners as
it deals with how the language is used. If the end goal in CLIL is for
students to perform like native speakers, this would mean that students
need “to learn thousands of words … to discover which words can be
combined and which cannot …” (Milton, 2009: 2).
CLIL provides the opportunity for students to see vocabulary in
meaningful situations and with the classroom. This is closely related to the
Involvement Load Hypothesis (Laufer & Hulstijn, 2001), which suggests
that the higher the degree of involvement and cognitive effort during word
processing the better for acquisition. Recent studies seem to confirm the
 From EFL to CLIL 33

hypothesis. Keating (2008) observed that beginners obtained better

retention results if they completed a sentence writing exercise than if they
just filled blanks. In the same line, Lu (2013) found out that at the same
time blank-filling was more beneficial for acquisition than just reading.
This hypothesis is consistent upon the components of need, search and
evaluation. The first one points to the external or internal learner’s need to
know the meaning of a word. The second one leads the learner to search
for the meaning of a word in alternative ways, for instance asking the
teacher or looking up in the dictionary. As for the third aspect, evaluation
is a constant in CLIL. Content is transmitted in the L2 and the L2 is the
vehicular language in the classroom, so that the learner is permanently –
consciously or unconsciously – evaluating one’s own performance. These
aspects have to be normally promoted in EFL through different activities.
Nonetheless, they are present in CLIL due to the very nature of this
methodology itself.

R What does the word sound like?

Spoken
P How is the word pronounced?
R What does the word look like?
Written
Form P How is the word written and spelled?
R What parts are recognisable in this word?
Word parts P What word parts are needed to express the
meaning?
R What meaning does this word form signal?
Form and
meaning P What word form can be used to express this
meaning?
Meaning
Concepts and R What is included in the concept?
referents P What items can the concept refer to?
R What other words does this make us think of?
Associations
P What other words could we use instead of this one?
Grammatical R In what patterns does the word occur?
functions P In what patterns must we use this word?
R What words or types of words occur with this one?
Collocations P What words or types of words must we use with
Use
this one?
Constraints on R Where, when, and how often would we expect to
use (register, meet this word?
frequency) P Where, when, and how often can we use this word?

Table 2. Description of “what is involved in knowing a word”, from

Nation (2001: 27)
34 Chapter Two

Pedagogical Implications
Many of the same difficulties regarding vocabulary acquisition in EFL
in primary education are currently still part of the problem with CLIL
methodologies. One of the hurdles is the identification of what vocabulary
needs to be taught and learned. It is necessary to highlight that this is not
an exclusive problem of the Spanish Primary Education system, but a
much wider issue. Language learning levels are currently outlined by the
Common European Framework of Reference of Languages (2001) in
which the four skills are described and graded by level from A1 to C2.
Nevertheless, within the framework, there is an absence of specific
vocabulary guidelines for these levels. This is a key issue which should be
addressed (Milton, 2009; Mancebo, 2005).
Material in the classroom and classroom language needs to take into
consideration both receptive and productive knowledge (Merikivi &
Pietilä, 2014). It is essential for students to be able to quickly acquire the
minimum number of words to be able to follow classes and thus be able to
consciously become aware that learning vocabulary is not only based on
learning meaning but also on form and use. This awareness would enable
students to understand the complexity of vocabulary and would help in the
acquisition of word families.
Classroom activities need to be adapted to incorporate a greater
involvement load. It is clear that students neither assimilate vocabulary by
simply seeing it, nor by doing activities that do not promote involvement.
Research has shown that students need to be actively engaged in the
acquisition of vocabulary and aware of form, meaning and use if they are
to become proficient in the language.
With all this in mind, it is important to be aware of the changes that
underlie the shift from EFL to CLIL, where most L2 learning occurs
implicitly, and where, at the same time, L2 communicative skills need to
be developed in order to follow lessons. That is why CLIL methodologies
need to be developed which specifically help to promote vocabulary
learning as vocabulary the key (Wilkins, 1972), and this implies
developing extensive word learning strategies. Therefore, we propose
some pedagogical tips to bear in mind, so that the input received by the
learners triggers the acquisition of target vocabulary:
Make the most of the multiple focus provided by CLIL: promote
reflection on the process of vocabulary learning using involvement load
activities.
 From EFL to CLIL 35

x Once the expected silent period is overcome, students should

communicate more than the teacher to work on not only receptive
vocabulary but also on productive vocabulary.
x Promote meaning negotiation.
x Work on word forms to create word families which in turn promotes
greater vocabulary acquisition.
x The power of word association: Use semantic fields in explanations of
concepts and promote the construction of semantic fields in learners.
x Make students aware of their growth in vocabulary size.

Bibliography
Alcaraz-Mármol, G. (2010). The role of frequency in the foreign language
classroom. Malaysian Journal of ELT research 6, 167-194.
Allen, V. (1983). Techniques in teaching vocabulary. New York: OUP.
Anderson, R. & Nagy, W. (1991). Word meanings. In R. Barr, M. Kamil,
P. Mosenthal & P.D. Pearson, (Eds.), Handbook of Reading Research,
Vol. 2. New York: Longman, 690-724.
Coady, J. (1997). L2 vocabulary acquisition. A synthesis of the research.
In J. Coady & T. Huckin (Eds.), Second language vocabulary
acquisition. New York: Cambridge University Press, 273-290.
Council of Europe. (2001). Common European Framework of Reference
for Languages: Learning, Teaching, Assessment. Cambridge: CUP.
Coyle, D., March, D. & Hood, P. (2010). CLIL (Content and Language
Integrated Learning). Cambridge: CUP.
Ellis, R. (1994). The Study of Second Language Acquisition. Oxford: OUP.
Grabe, W. & Stoller, F. L. (1997). Content-based instruction: Research
foundations. In T. A. Snow & D. T. Brinton (Eds.), The content-based
classroom: Perspectives on integrating language and content. White
Plains, NY: Longman, 5-21.
Griffin, J.F. and Harley, T.A. (1996). List learning of second language
vocabulary. Applied Psycholinguistics 17 (2), 443-460.
Harris, V. & Snow, D. (2004). Classic Pathfinder: Doing it for
themselves:focus on learning strategies and vocabulary building.
London: CILT.
Jiménez Catalán, R. M. & Mancebo, R. (2008). Vocabulary input in EFL
textbooks. Revista Española de Lingüística Aplicada 21, 147-166.
Jiménez Catalán, R. M. & Moreno Espinosa, S. (2005). Using Lex30 to
measure the L2 productive vocabulary of Spanish primary learners of
EFL. Vial 13 (2), 27-44.
36 Chapter Two

Jiménez Catalán, R. M. & Terrazas, M. (2008). The Receptive Vocabulary

of English Foreign Language Young Learners. IJES 2 (2): 201-15.
Kamil, M. L. (2004). Vocabulary and comprehension instruction:
Summary and implications of the National Reading Panel findings. In
P. McCardle & V. Chhabra (Eds.), The voice of evidence in reading
research. Baltimore, MD: Paul H. Brookes.
Keating, G.D. (2008). Task effectiveness and word learning in a second
language: the Involvement Load Hypothesis on trial. Language
Teaching Research 12, 365-386.
Laufer, B. & Hulstijn, J. (2001). Incidental vocabulary acquisition in a
second language: the construct of task-induced involvement. Applied
Linguistics 22 (1), 1-26.
Laufer, B. & Nation, P. (1995). Vocabulary size and use: lexical richness
in L2 written production. Applied Linguistics 16(3), 307-322.
Lu, M. (2013). Effects of four vocabulary exercises on facilitating learning
vocabulary meaning, form and use. TESOL Quaterly 47, 167-178.
Mancebo, R. (2005). ¿Hasta qué punto los libros de texto de primaria se
adecuan a las directrices de enseñanza de vocabulario del marco
común?. Paper presented at the commemoration of the European Day
of Languages. University of Valladolid.
Mason, B. & Krashen, S. (1997). Extensive reading in English as a foreign
language. System 25 (1), 91-102.
McCarthy, M. (1990). Vocabulary. Oxford: OUP.
—. (2001). What is an advanced level vocabulary? SELL 8(3), 149-163.
Melka, F. (1997). Receptive vs. productive aspects of vocabulary. In N.
Schmitt & M. McCarthy (Eds.), Vocabulary: Description, acquisition
and pedagogy. Cambridge: CUP, 84-102.
Merikivi, R. & Pietilä, P. (2014). Vocabulary in CLIL and in Mainstream
Education. Journal of Language Teaching and Research 5(3), 487-
497.
Milton, J. (2009). Measuring second language vocabulary acquisition.
Bristol: Multilingual Matters.
Nassaji, H. (2006). The relationship between depth of vocabulary
knowledge and L2 learners’ lexical inferencing strategy use and
success. The Modern Language Journal 90 (3), 387-401.
Nation, P.S.N. (2001). Learning Vocabulary in Another Language.
Cambridge: CUP.
—. (2006). How large a vocabulary is needed for reading and listening?
The Canadian Modern Language Review 63, 59-81.
Richards, J.C. (1974). Word lists: problems and prospects. RELC Journal
5(2), 69-84.
 From EFL to CLIL 37

Schmitt, N. (2000). Vocabulary in language teaching. Cambridge: CUP.

Scholfield, P. (1991). Vocabulary rate in coursebooks: Living with an
unstable lexical economy. Proceedings of 5th Symposium on the
Description and/or Comparison of English and Greek. Thessaloniki:
Aristotle University, 11-32.
Scott, J.A. (2005). Creating opportunities to acquire new word meanings
from text. In E.H. Hiebert & M. Kamil (Eds.), Teaching and learning
vocabulary: Bringing research to practice. Mahwah, NJ: Lawrence
Erlbaum, 69-91.
Stahl, S. (2005). Four problems with teaching word meanings (and what to
do to make vocabulary an integral part of instruction). In E. H. Hiebert
& M. L. Kamil (Eds.), Teaching and learning vocabulary: Bringing
research to practice. Mahwah, NJ: Lawrence Erlbaum, 95-114.
Stahl, S. A. & Kapinus, B. (2001). Word power: What every educator
needs to know about teaching vocabulary. Washington, D.C.: National
Education Association.
Waring, R. (1997). A study of receptive and productive learning from
word cards. Studies in Foreign Language and Literature 21(1), 94-114.
White, R. (1988). The ELT Curriculum: design, innovation and
management. Oxford: Blackwell.
Wilkins, D.A. (1972). Linguistics in language teaching. London: Arnold.
 PART II

LEGAL TERMINOLOGY
Random documents with unrelated
content Scribd suggests to you:
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
back
Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about books and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

[Link]