Coreference in English and German A Theoretical Framework and Its Application in A Study of Court Decisions 1st Edition Sanna Engell

Coreference in English and German A
Theoretical Framework and Its

Application in a Study of Court
Decisions 1st Edition Sanna Engell
Visit to download the full and correct content document:
https://ebookmeta.com/product/coreference-in-english-and-german-a-theoretical-fram
ework-and-its-application-in-a-study-of-court-decisions-1st-edition-sanna-engell/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Soviet Agriculture in Perspective A Study of its

Successes and Failures 1st Edition Erich Strauss
https://ebookmeta.com/product/soviet-agriculture-in-perspective-
a-study-of-its-successes-and-failures-1st-edition-erich-strauss/
Understanding the Investor A Maltese Study of Risk and

Behavior in Financial Investment Decisions 1st Edition
Antonietta Bonello
https://ebookmeta.com/product/understanding-the-investor-a-
maltese-study-of-risk-and-behavior-in-financial-investment-
decisions-1st-edition-antonietta-bonello/
A Theoretical Framework for Language Education and

Teaching 1st Edition Paolo E. Balboni
https://ebookmeta.com/product/a-theoretical-framework-for-
language-education-and-teaching-1st-edition-paolo-e-balboni/
Science of the Soul in Ibn Sina's Pointers and

Reminders: A Philological Study (English and Arabic
Edition) Michael A. Rapoport
https://ebookmeta.com/product/science-of-the-soul-in-ibn-sinas-
pointers-and-reminders-a-philological-study-english-and-arabic-
edition-michael-a-rapoport/
Property and its Forms in Classical German Philosophy
1st Edition David James
https://ebookmeta.com/product/property-and-its-forms-in-
classical-german-philosophy-1st-edition-david-james/
Vitamin D Basic Research and its Clinical Application

A. W. Norman
https://ebookmeta.com/product/vitamin-d-basic-research-and-its-
clinical-application-a-w-norman/
External Communication In Social Media During

Asymmetric Conflicts A Theoretical Model And Empirical
Case Study Of The Conflict In Israel And Palestine 1st
Edition Bernd Hirschberger
https://ebookmeta.com/product/external-communication-in-social-
media-during-asymmetric-conflicts-a-theoretical-model-and-
empirical-case-study-of-the-conflict-in-israel-and-palestine-1st-
edition-bernd-hirschberger/
Book of Seven Seals: The Peculiarity of Revelation, Its

Manuscripts, Attestation, and Transmission (English and
German Edition) Thomas J. Kraus
https://ebookmeta.com/product/book-of-seven-seals-the-
peculiarity-of-revelation-its-manuscripts-attestation-and-
transmission-english-and-german-edition-thomas-j-kraus/
Pacific Islanders under German Rule A Study in the

Meaning of Colonial Resistance 1st Edition Peter
Hempenstall
https://ebookmeta.com/product/pacific-islanders-under-german-
rule-a-study-in-the-meaning-of-colonial-resistance-1st-edition-
peter-hempenstall/
Sanna Engell
Coreference in English and German

A Theoretical Framework and its Application
in a Study of Court Decisions
λογος
Sanna Engell
Coreference in English and German

A Theoretical Framework and its Application
in a Study of Court Decisions
Zugl.: Bonn, Univ., Diss., 2015
Bibliografische Information der Deutschen Nationalbibliothek
Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der

Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind
im Internet über http://dnb.d-nb.de abrufbar.
c Copyright Logos Verlag Berlin GmbH 2017

Alle Rechte vorbehalten.
ISBN 978-3-8325-4339-6
Logos Verlag Berlin GmbH

Comeniushof, Gubener Str. 47,
10243 Berlin
Tel.: +49 (0)30 42 85 10 90
Fax: +49 (0)30 42 85 10 92
INTERNET: http://www.logos-verlag.de
For my parents
Acknowledgements
This book would not have been written without the support of many
people. First of all, I would like to thank my thesis supervisor, Professor
Dr. Jürgen Esser. The present study has benefited greatly from his
feedback, suggestions and advice. He introduced me to many fascinating
research areas of linguistics and provided an excellent research
environment. I would also like to thank Professor Dr. Klaus P. Schneider,
second supervisor of my thesis, for valuable suggestions and comments
on the text.
Furthermore, I would like to express my thanks to Professor Dr. Rolf
Kreyer and my fellow doctoral candidates, Vera Benninghoven, Laura
Göttmann, Sebastian Patt, Julia Sosnizka and Sharmila Vaz. We had
many insightful discussions and it has been very interesting and helpful to
exchange ideas and read each other’s theses. Any remaining infelicities in
this book are, of course, my responsibility alone.
A big thank you goes to my parents, who have supported me through
my entire life. I dedicate this book to them.
Finally, I thank my sisters and my friends for their interest and en-
couragement and for being there for me.
Table of contents
1 Introduction...........................................................................................1
2 The concept of coreference ..................................................................7

2.1 A definition of reference .................................................................7
2.2 Types of reference .........................................................................13
2.3 A definition of coreference............................................................17
2.4 Types of coreference .....................................................................20
2.5 Syntactic categories in nominal coreference chains......................29
3 Coreference chain elements in English and German:

devices and functions..........................................................................33
3.1 Devices for anaphoric coreference ................................................33
3.1.1 Main devices for anaphoric coreference .............................33
3.1.1.1 Pronouns.................................................................33
3.1.1.2 Determiners............................................................39
3.1.1.3 Full noun phrases ...................................................40
3.1.2 Supplementing devices for anaphoric coreference .............52
3.2 Main devices for cataphoric coreference ......................................58
3.2.1 Pronouns..............................................................................58
3.2.2 Possessive determiners........................................................62
3.2.3 Full noun phrases ................................................................62
3.3 Devices for non-directional coreference .......................................66
3.3.1 Main devices........................................................................66
3.3.1.1 Pronouns.................................................................66
3.3.1.2 Full noun phrases ...................................................66
3.3.2 Supplementing devices........................................................77
3.4 Functions of coreference chain elements in English and
German ..........................................................................................81
3.4.1 Functions of anaphoric and non-directional chain
elements...............................................................................81
3.4.1.1 Level 1: creation of coreference ............................81
viii Table of contents
3.4.1.2 Level 2: reduction, clarity of reference and

specification ...........................................................84
3.4.1.3 Level 3: stylistic variation, evaluation and
emphasis.................................................................90
3.4.2 Functions of cataphoric chain elements ..............................94
3.4.2.1 Level 1: creation of coreference ............................94
3.4.2.2 Level 2: textual organization and creation of
suspense .................................................................96
3.5 Factors influencing the choice of coreferential devices................99
3.5.1 Language-specific features..................................................99
3.5.2 Accessibility of the referent ..............................................103
3.5.3 Genre or sub-genre of the text...........................................109
3.5.4 Stylistic preferences of the text producer..........................115
4 Related fields of research .................................................................117

4.1 Contrastive linguistics .................................................................117
4.1.1 Overview ...........................................................................117
4.1.2 Particularly relevant studies ..............................................119
4.2 Language and law........................................................................121
4.2.1 Overview ...........................................................................121
4.2.2 Particularly relevant studies ..............................................124
5 The Appellate Committee of the House of Lords and the

Bundesgerichtshof in the legal systems of the United Kingdom
and Germany.....................................................................................131
5.1 Legal traditions: common law and civil law ...............................131
5.2 The Appellate Committee of the House of Lords and the
Bundesgerichtshof .......................................................................136
6 Methodology of the study of coreference in the court

decisions .............................................................................................145
6.1 Compilation of the corpus ...........................................................145
6.2 Steps in the analysis of the court decisions .................................154
6.2.1 Identifying coreference chains ..........................................154
Table of contents ix
6.2.2 Types of coreference .........................................................161

6.2.3 Syntactic categories of the chain elements .......................165
6.2.4 Devices for coreference.....................................................165
6.2.5 Functions of the coreference chain elements ....................171
6.2.6 Factors influencing the choice of coreferential devices....172
7 Results and discussion ......................................................................177

7.1 Frequencies and lengths of coreference chains ...........................177
7.2 Types of coreference ...................................................................185
7.3 Syntactic categories of the chain elements..................................191
7.4 Devices for coreference ...............................................................196
7.4.1 Devices for anaphoric coreference....................................196
7.4.1.1 Main anaphoric devices .......................................196
7.4.1.2 Supplementing anaphoric devices .......................205
7.4.2 Main devices for cataphoric coreference ..........................208
7.4.3 Devices for non-directional coreference ...........................215
7.4.3.1 Main non-directional devices...............................215
7.4.3.2 Supplementing non-directional devices...............220
7.5 Functions of the coreference chain elements ..............................225
7.5.1 Functions of the anaphoric chain elements .......................225
7.5.2 Functions of the cataphoric chain elements ......................232
7.5.3 Functions of the non-directional chain elements ..............236
7.6 Factors influencing the choice of coreferential devices..............241
7.6.1 Language-specific features................................................241
7.6.2 Accessibility of the referent ..............................................245
7.6.3 Genre or sub-genre of the texts .........................................249
7.6.4 Stylistic preferences of the text producer..........................253
8 Summary and conclusions ...............................................................255
References..............................................................................................269
1 Introduction
Establishing coreference is one of the most important means to connect
linguistic items with other linguistic items in the same text (cf. Brinker
2010: 37; Kunz 2010: 1). Coreference can be found very frequently in
most texts and coreference chains often extend across long passages of a
text or sometimes across an entire text (cf. Kunz 2010: 1). Studying
coreference is essential for investigating how texts are produced and un-
derstood. Therefore, coreference has frequently been a research topic in
textlinguistics and is also studied in other branches of linguistics, such as
computational linguistics or psycholinguistics (cf. e.g. Halliday/Hasan
1976: 3, chapters 2 and 6; de Beaugrande/Dressler 1981: 54-66; van
Deemter/Kibble 2000; Schmitt 2003; Brinker 2010: 26-44; Stede 2012:
chapter 3; Warren 2013: 203f.). The findings of linguistic studies of
coreference can be applied, for example, in machine translation or text
summarization (cf. e.g. Leass/Schwall 1991; Azzam et al. 1999).
The present study is a textlinguistic study of coreference. In this study,
coreference is understood as the relation between two or more linguistic
items in a text which refer to the same referent in the extralinguistic
world (cf. e.g. Halliday/Hasan 1976: 3; Esser 2009: 34f.; Brinker 2010:
26). The following example illustrates this. It is taken from a short story
by Hemingway and is provided by Schubert (2012: 34).
(1) Nick was hungry. He did not believe he had ever been hungrier.
(Hemingway 1980: 459)
In the above example, Nick, He and he refer to the same referent in the
extralinguistic world, namely to the person Nick. The three linguistic
items are thus in a relation of coreference and are considered as elements
of a ‘coreference chain’ (cf. Halliday/Hasan 1989: 84, who use the term
‘identity chain’, and Biber et al. 1999: 234; Azzam et al. 2000: 77; Kunz
2010: 1, 10).
While there are several studies of aspects of coreference in English or
in German, there exist so far only few contrastive studies of coreference
2 Chapter 1
in these two languages (for a notable exception cf. Kunz 2010). The pre-
sent study intends to contribute to closing this gap by providing an in-
depth study of coreference in English and German in general and in Eng-
lish and German court decisions in particular. The study aims to answer
the following research questions. First, it examines which linguistic de-
vices can be used to establish coreference in English and in German. Sec-
ond, it investigates which functions the elements of coreference chains
fulfil in the two languages. Third, it analyzes which factors influence the
choice of devices for coreference in English and German.
In order to answer these research questions, the present study com-
bines a theoretical part and an empirical part. The theoretical part presents
a new framework of the linguistic devices for coreference in English and
German, their functions and the factors which influence their use. This
framework is then put to the test in a study of a 34,000 word corpus of
British and German court decisions. One of the reasons for choosing the
sub-genre of court decisions1 for the corpus study was that the linguistic
features of court decisions have been shown to be relatively strongly in-
fluenced by the legal system(s) and the legal culture(s) of their country of
origin (cf. Wetter 1960: 32f.; Lashöfer 1992: 2, 34-41, 73-80, 119-126). It
was therefore assumed to be particularly interesting to study how these
genre-related extralinguistic factors may influence the choice of devices
for coreference in court decisions.
The theoretical part of the present study begins with chapter 2. Since
‘reference’ and ‘coreference’ are two related concepts, section 2.1 pro-
vides a definition of the concept of reference. The next section 2.2 distin-
guishes different types of reference, such as ‘generic’, ‘non-specific’ and
‘specific’ reference. Section 2.3 contains a detailed definition of corefer-
ence. The following section 2.4 discusses different types of coreference.
Two types of coreference have already been firmly established in the lin-
guistic literature: ‘anaphoric’ and ‘cataphoric’ coreference (cf. e.g. Halli-
1
In the present study, court decisions are considered as a sub-genre of the genre
‘legal texts’ (cf. section 3.5.3).
Introduction 3
day/Hasan 1976: 14, 17; Biber et al. 1999: 234, 348; Esser 2009: 35, 49f.).
In addition to these two types of coreference, the present study suggests a
third type of coreference, which is called ‘non-directional coreference’
(for brief remarks on concepts related to non-directional coreference cf.
Martin 1992: 145f.; van Deemter/Kibble 2000: 629f.). Section 2.5 then
explains which syntactic categories can be part of coreference chains in
English and German.
Next, sections 3.1 to 3.3 present the different linguistic devices which
can establish anaphoric, cataphoric and non-directional coreference in
English and German. Based on previous research (cf. e.g. Halliday/Hasan
1976: 37f., 278; Hoey 1991: 53, 70; Brinker 2010: 26-29, 39), the present
study develops its own comprehensive framework of devices for corefer-
ence in English and German. Section 3.4 then explains the different func-
tions of elements of coreference chains in English and German, for ex-
ample the functions ‘reduction’ or ‘specification’. The present study
suggests a new framework of functions, based on previous research (cf.
e.g. de Beaugrande/Dressler 1981: 60f.; Linke/Nussbaumer 2000: 312-
314; Esser 2009: 50, 123f.; Schubert 2012: 32f.). The final section of
chapter 3, section 3.5, explains which factors can influence the choice of
devices for coreference in an English or German text. Four factors are
distinguished: language-specific features, the accessibility of the referent,
the genre or sub-genre of the text and the stylistic preferences of the text
producer (cf. e.g. Ariel 1988: 84; Biber et al. 1999: 237f.; Swanson 2003:
183f.; Kunz 2010: 19, 22).
After the detailed theoretical discussion of coreference in English and
German in chapters 2 and 3, chapter 4 provides an overview of further
fields of research which are related to the present study. Since the present
study compares how coreference is established in English and German,
section 4.1 gives an overview of the research field ‘contrastive linguis-
tics’. In addition, section 4.1 summarizes the results of studies in contras-
tive linguistics which are particularly relevant for the present study. Next,
since the empirical part of the study analyzes coreference in court deci-
4 Chapter 1
sions, section 4.2 provides an overview of the research field ‘language

and law’. Furthermore, section 4.2 discusses previous studies in the field
‘language and law’ which are closely related to the present study.
As the empirical part of the present study focuses on coreference in
British and German court decisions, chapter 5 briefly describes the legal
systems of the United Kingdom and Germany. The legal systems of the
two countries are based on two different ‘legal traditions’, ‘common law’
and ‘civil law’ (cf. Dickson 2005: 2; Merryman/Pérez-Perdomo 2007: 1;
White/Willock 2007: 162; Slapper/Kelly 2012: 80). These legal traditions
are explained in section 5.1.
The court decisions analyzed in the present study are decisions of the
British Appellate Committee of the House of Lords and of the German
Bundesgerichtshof. Section 5.2 thus explains the position of the Appellate
Committee of the House of Lords in the British court system and the po-
sition of the Bundesgerichtshof in the court system of Germany. One of
the reasons for choosing decisions of the two courts was that the courts
had a similar position in the court systems of the United Kingdom and
Germany. Furthermore, both courts are, or were, final courts of appeal.2
Several previous studies have also focused on the linguistic features of
the decisions of final courts of appeal (cf. e.g. Wetter 1960; Lashöfer
1992; Waters 1997; Blom-Cooper 2009). The results of the present study
could thus be more easily compared with the results of these studies. This
comparison of results can be found in chapter 7.
Chapter 6 marks the beginning of the empirical part of the present
study. The chapter explains the methodology of the empirical study of
coreference in the court decisions. Section 6.1 describes how the corpus
of court decisions was compiled. Section 6.2 then lists the different steps
that were taken in the analysis of coreference in the corpus. First, in order
to gain a general impression of the coreference chains in the court deci-
sions, the following features were analyzed: the numbers and average
2
The Appellate Committee of the House of Lords was abolished in 2009 and
replaced by the Supreme Court of the United Kingdom (cf. section 4.2.2 below).
Introduction 5
lengths of the coreference chains, the frequencies of the different types of

coreference and the syntactic categories of the coreference chain elements
(cf. sections 6.2.1 to 6.2.3). Then, the framework of coreference devel-
oped in the theoretical part was tested by analyzing the devices for
coreference, the functions of the coreference chain elements and the fac-
tors which influenced the choice of coreferential devices in the court de-
cisions (cf. sections 6.2.4 to 6.2.6).
Chapter 7 presents the results of the empirical study and discusses
them. It shows that the descriptive framework of the present study was
successfully applied to the court decisions. In addition, it points out the
similarities and differences between the ways in which coreference was
established in the English sub-corpus and the German sub-corpus of the
present study.
Chapter 8 sums up the main findings of the present study. Further-
more, the chapter includes suggestions for further research.
2 The concept of coreference
2.1 A definition of reference
The present study is a study of coreference in English and German in
general, and in British and German court decisions in particular. Before
the concept of coreference will be explained in section 2.3, the related
notion of reference will be discussed in the present section and different
types of reference will be described in section 2.2.
The concept of reference has been dealt with in different disciplines,
such as logic, semantics, pragmatics or textlinguistics (cf. e.g. Halli-
day/Hasan 1976: chapter 2; Vater 2012: 1). There exist several different
definitions of reference. The definition of reference in the present study is
based on the approaches to reference in semantics and textlinguistics.
Furthermore, the study adopts a part of the terminology used in the prag-
matic approach to reference. We will first consider the way in which ref-
erence is usually defined in semantics. This will be illustrated by a defini-
tion by Lyons, who discusses the concept of reference from a semantic
point of view. Lyons (1977: 174) writes:
The term ‘reference’, as we shall define it below, has to do with the
relationship which holds between an expression and what that expres-
sion stands for on particular occasions of its utterance.
According to Lyons, reference can thus be described as the relation be-
tween a linguistic expression and the entity in the extralinguistic world to
which the expression refers. Lyons (1977: 177) calls this entity ‘referent’.
In the present study, ‘entity’ is used with a very general meaning that in-
cludes, for example, people, concrete objects or abstract concepts. It
should be noted that Lyons’ definition of reference can be traced back to
the writings by Frege (1975 [1892]: 41) and other philosophers who also
discussed the relation between a linguistic item and the entity to which
the item refers (for an overview cf. e.g. Wunderlich 1974: 236-255; Ab-
bott 2010: chapter 2). Lyons (1977: 176) stresses that reference is an ut-
8 Chapter 2
terance-bound concept: only linguistic expressions that are used in spe-

cific utterances can have referents in the extralinguistic world whereas
lexemes, which are more abstract and do not occur in actual utterances,
do not have referents.
Reference from a pragmatic point of view has been described, for ex-
ample, by Searle (1969). According to Searle (1969: 22), uttering some-
thing in a language means to perform ‘speech acts’. Searle distinguishes
between different types of speech acts. We will not discuss all types of
speech acts in detail here, but will instead focus on the ‘propositional
speech act’. Searle (1969: 23f.) explains that the propositional speech act
consists of a ‘speech act of reference’ and ‘speech act of predication’.
When performing a speech act of reference, the speaker or writer uses a
linguistic item to refer to an entity in the extralinguistic world. When the
speaker or writer performs a speech act of predication he or she says or
writes something about the entity to which he or she refers. In the present
study, the term ‘text producer’ will be used as a synonym for ‘speaker or
writer’ and the term ‘text receiver’ as a synonym for ‘listener or reader’.
The following example illustrates the speech acts of reference and predi-
cation.
(1) Sam smokes habitually. (Searle 1969: 22)
In the above example, the text producer uses the noun phrase Sam to refer
to a person. This is the act of reference. Furthermore, the text producer
uses smokes habitually to say something about the person Sam. This is
the act of predication. Searle (1969: 28) emphasizes that linguistic items
do not refer themselves. Instead, it is the text producer who uses a lin-
guistic item to refer to an entity in the extralinguistic world. Searle (1969:
28) writes:
To say that an expression refers (predicates, asserts, etc.) in my ter-
minology is either senseless or is shorthand for saying that the expres-
sion is used by speakers to refer (predicate, assert, etc.); this is a
shorthand that I shall frequently employ.
The concept of coreference 9
For reasons of brevity, other authors have followed Searle’s approach and
use the expression “a linguistic item refers” as a short form for “the text
producer uses a linguistic item to refer” (cf. e.g. Lyons 1977: 177; Chur
1993: 9). The present study will also use this short form.
Halliday/Hasan (1976: chapter 2) discuss the concept of reference
from a textlinguistic point of view. They suggest the following definition
of reference:
There are certain items in every language which have the property of
reference, in the specific sense in which we are using the term here;
that is to say, instead of being interpreted semantically in their own
right, they make reference to something else for their interpretation.
(Halliday/Hasan 1976: 31)
Halliday/Hasan (1976: 37) thus view reference as a feature of certain lin-
guistic items, such as personal pronouns, demonstrative pronouns or ad-
verbs like there and then. These items can only be correctly interpreted
by a listener or reader if they (i) refer to something that is present at the
moment of the utterance or (ii) refer to another item in the same text. The
difference between (i) and (ii) can be illustrated by the following two ex-
amples.1
(2) For he’s a jolly good fellow
And so say all of us. (Halliday/Hasan 1976: 32)
(3) John has moved to a new house. He had it built last year. (Halli-
day/Hasan 1976: 55)
In example (2), he and us can be interpreted correctly if they refer to peo-
ple who are present when the utterance is made. By contrast, in (3) the
interpretations of He and it become clear since He refers back to John and
it to a new house (cf. Halliday/Hasan 1976: 32, 54f.). Example (3) thus
1
In examples (2) and (3), the word-forms that are most important for the discussion
have been marked by broken underlining. The same will be done in many of the fol-
lowing examples in the present study.
10 Chapter 2
illustrates that linguistic items can refer to other linguistic items within
the same text.
When using the term ‘text’, it should be noted that there exist many
different definitions of ‘text’ in linguistics (cf. e.g. Halliday/Hasan 1976:
1; Vater 2001: 21; Schubert 2012: 27f.; Esser 2009: 9). For the purpose of
the present study, ‘text’ is understood as a prototype category which has
more and less typical features (cf. Vater 2001: 21; Esser 2009: 9). Typi-
cally, a text is assumed to be longer than a sentence; however, one-word
utterances may also count as texts. Both instances of spoken and written
language are accepted as texts (cf. Esser 2009: 9, 23; Schubert 2012: 27).
Returning to the topic of reference, the present study uses a concept
of reference which includes aspects of both Lyons’ (1977: 174) and Hal-
liday/Hasan’s (1976: 31) definitions. ‘Reference’ in the present study is
understood as a superordinate term for ‘endophoric’ and ‘exophoric’ ref-
erence. These terms have been suggested by Halliday/Hasan (1976: 33)
and have been elaborated, for example, by Halliday (2014: 624-626). We
will now discuss endophoric and exophoric reference and then suggest a
definition of ‘reference’.
In the present study, ‘exophoric reference’ is defined as the relation
between a linguistic item and whatever it refers to in the extralinguistic
world. This definition of exophoric reference is thus very similar to Ly-
ons’ (1977: 174) concept of reference (cf. also Esser 2009: 35; Halliday
2014: 624f.). The term ‘endophoric reference’ is used in the present study
to describe a relation between two or more linguistic items in a text that
can be explained as follows: one or more items refer backwards or for-
wards to another item in the text. The referring items cannot be correctly
interpreted on their own but only due to the linguistic item to which they
refer (cf. Halliday 2014: 625). This concept of endophoric reference has
been illustrated in example (3) above, where He referred to John and it to
a new house. The definition of endophoric reference in the present study
thus draws on aspects of Halliday/Hasan’s (1976: 31) definition of refer-
ence, which has been explained above. The difference between endo-
phoric and exophoric reference can be shown with the help of the follow-
ing example.
(4) Wash and core six cooking apples. Put them into a fireproof dish.
In example (4), them refers back to six cooking apples. Thus, them and six
cooking apples are in a relation of endophoric reference. At the same time,
both them and six cooking apples refer exophorically to something in the
extralinguistic world, namely to six cooking apples. Endophoric and exo-
phoric reference in example (4) can be visualized as in Figure 2.1 below,
which is taken from Esser (2009: 35) in a slightly adapted version.
Extralinguistic world
Six cooking apples
Exophoric Exophoric
reference reference
… six cooking apples … … them …
Endophoric reference
Figure 2.1: Endophoric and exophoric reference
With regard to exophoric reference, previous studies have discussed

which linguistic units can refer to something in the extralinguistic world
(cf. e.g. Vater 2012: 64 for an overview). While it was sometimes as-
sumed that only noun phrases refer exophorically, today many authors
12 Chapter 2
suggest that other types of phrases and clauses or clause complexes2 have
referents in the extralinguistic world as well (cf. e.g. Jackendoff 1983:
52f.; Chur 1993: 8; Vater 2001: 90). In the present study, it is thus as-
sumed that clauses, clause complexes and different types of phrases can
refer exophorically.
When discussing the referents of clauses or clause complexes, previ-
ous studies have suggested that the extralinguistic referent of a clause or
clause complex is its content, i.e. for example the situation(s), event(s) or
fact(s) that it describes (cf. Halliday/Hasan 1976: 52; Lambrecht 1994: 53,
74f.; Esser 2009: 40; Vater 2012: 65, 68). Different terms have been used
for the referents of clauses or clause complexes (cf. e.g. Lambrecht 1994:
53, 74f.; Vater 2005: 76). In this study, the term ‘proposition’ will be
used (cf. Lambrecht 1994: 53, 74f.; Esser 2009: 40 and cf. Searle 1969:
29-33 for the use of the term ‘proposition’ in speech act theory). It will
thus be assumed that in cases of coreference between noun phrases and
clauses or clause complexes, the shared referents are the propositions of
the clauses or clause complexes.
Based on the distinction between exophoric and endophoric reference,
in the present study ‘reference’ is defined as a superordinate term for (i)
the relation between a linguistic item and its referent in the extralinguistic
world and (ii) the relation between a linguistic item and another linguistic
item to which it refers. As has been explained above, the referent of a lin-
guistic item can be, for example, an entity or a proposition.
In previous discussions of reference, a further important topic has
been the question where the referents of linguistic items are located.
Many authors have suggested that these referents can be found in a ‘con-
ceptual’ world and not in the real world (cf. Jackendoff 1983: 29;
Linke/Nussbaumer 2000: 306; Vater 2001: 89). One reason for this sug-
gestion is that it is possible to refer to entities that do not exist in reality,
2
In the present study, the term ‘clause complex’ is used to refer to a syntactic unit
which consists of at least one independent clause and which may include further
clauses which are linked to the independent clause by means of coordinating con-
junctions or subordination (cf. Esser 2009: 25f.; Halliday 2014: 9, 436).
such as unicorns, or to abstract concepts or events in the future (cf.

Linke/Nussbaumer 2000: 306; Vater 2001: 89; Kunz 2010: 31). As for
entities that exist in the real world it has been assumed that linguistic
items do not refer directly to these entities, but to the mental concepts that
the text producer has of them (cf. Jackendoff 1983: 29;
Linke/Nussbaumer 2000: 306). In the present study, the view that linguis-
tic items refer to a conceptual world is accepted; however, the term ‘ex-
tralinguistic world’ is used instead of ‘conceptual world’.
2.2 Types of reference

In section 2.1, two different types of reference have already been de-
scribed: exophoric and endophoric reference. Exophoric reference was
defined as the relation between a linguistic item and its referent in the ex-
tralinguistic world. Endophoric reference was defined as the relation be-
tween a linguistic item and another linguistic item in the same text to
which it refers. Both exophoric reference and endophoric reference have
further sub-types of reference. When discussing the different types of ex-
ophoric reference, the present study will focus on the exophoric reference
of noun phrases. The reason for this is that the study analyzes nominal
coreference and thus focuses on noun phrases (cf. section 2.4 below).
In the present study, the following types of exophoric reference are
distinguished: ‘specific reference’, ‘non-specific reference’ and ‘generic
reference’. Specific reference is established when a linguistic item refers
to a specific entity in the extralinguistic world (cf. e.g. Quirk et al. 1985:
265; Biber et al. 1999: 234, 260). Here are two examples.
(5) The lambs were eating quietly. (Quirk et al. 1985: 247)
(6) I’ve just bought a melon. (Quirk et al. 1985: 274)
In example (5), the noun phrase The lambs refers to a specific group of
lambs. In example (6), I refers to a specific person and a melon to a spe-
cific melon. The two examples show that specific reference can be estab-
14 Chapter 2
lished by noun phrases with a definite article, an indefinite article or ones

without an article (cf. Quirk et al. 1985: 265, 274f.).
In contrast to specific reference, non-specific reference is established
when a linguistic item refers to a non-specific entity in the extralinguistic
world (cf. e.g. Biber et al. 1999: 260; Radden/Dirven 2007: 94-96). The
following two examples illustrate that.
(7) Sorry, if I drink a milkshake now, it will upset my stomach. (Rad-
den/Dirven 2007: 96)
(8) I’m looking for a millionaire, she says, but I don’t see many around.
(corpus example from Biber et al. 1999: 2603)
In example (7) above, the noun phrase a milkshake does not refer to a
specific milkshake in the extralinguistic world. Instead, it refers to any
milkshake. Similarly, in example (8) the noun phrase a millionaire does
not refer to a specific millionaire, but to any millionaire. Non-specific
reference can be established by noun phrases which include an indefinite
article (cf. Biber et al. 1999: 260). As Radden/Dirven (2007: 96) explain,
non-specific reference usually occurs in constructions which contain ne-
gations, imperatives, conditionals or verbs of want, need or desire.
A third type of exophoric reference to be discussed here is called ‘ge-
neric reference’. Generic reference is established when a linguistic item
refers to a class of entities as a whole (cf. e.g. Lyons 1977: 194; Quirk et
al. 1985: 281; Biber et al. 1999: 265). Three examples are provided below.
(9) Bull terriers make excellent watchdogs. (Quirk et al. 1985: 281)
(10) The tiger is becoming almost extinct. (Quirk et al. 1985: 282)
(11) A lion is a friendly beast. (Lyons 1977: 194)
3
In the present study, all corpus examples cited from Biber et al. 1999 are examples
taken from the Longman Spoken and Written English Corpus (cf. Biber et al. 1999:
xxvi, 5).
In example (9), the noun phrase Bull terriers does not refer to specific
bull terriers, but to the class of bull terriers as a whole. The noun phrase
thus has generic reference. Similarly, in example (10) the noun phrase
The tiger refers to the whole class of tigers and in example (11) the noun
phrase A lion refers to the class of lions as a whole. Examples (9) to (11)
show that generic reference can be established by noun phrases with the
zero article, a definite article or an indefinite article (cf. Quirk et al. 1985:
281; Biber et al. 1999: 265f.). It should be noted that generic reference
can also be established by noun phrases which have a non-count noun as
their head and which do not include a determiner (cf. Quirk et al. 1985:
275, 282, 286f.; Biber et al. 1999: 265; Lyons 1999: 180f.). The follow-
ing example illustrates this.
(12) Hydrogen is lighter than oxygen. (Quirk et al. 1985: 282)
In the above example, the two noun phrases Hydrogen and oxygen both
have a non-count noun as head and do not contain a determiner. Both
noun phrases can be interpreted as having generic reference (cf. Quirk et
al. 1985: 282). The reason why the noun phrases can be assumed to have
generic reference is that each of them refers to a gas – hydrogen or oxy-
gen – in general and not to a specific instance of this gas. Lyons (1999:
181f.) explains such cases as follows:
Just as we can make a statement about ostriches in general, that is,
about the class of entities satisfying the description ostrich, so we can
about the substance matching the description butter, or the quality
matching the description sincerity – as opposed to specific quantities
or instantiations.
Thus, although strictly speaking noun phrases headed by non-count nouns
do not refer to classes of entities, they can still be considered as having
generic reference. The present study follows this approach and assumes
that noun phrases headed by non-count nouns can have generic reference.
16 Chapter 2
Having discussed three different types of exophoric reference, we

now turn to two types of endophoric reference: anaphoric and cataphoric
reference. Anaphoric reference is established when a linguistic item refers
backwards to another linguistic item in the text (cf. Halliday/Hasan 1976:
33; Esser 2009: 49; Brinker 2010: 31). An example of this type of refer-
ence is provided below.
(13) Davis opened a door. ‘Here’s your room. I’m afraid it’s a bit un-
tidy.’ (corpus example from Sinclair 1990: 384)
In this example, the personal pronoun it refers back to the noun phrase
your room. Since the direction of the reference within the text is back-
wards, example (13) illustrates anaphoric reference.
‘Cataphoric reference’ is used as a term for forward-pointing refer-
ence within a text (cf. Halliday/Hasan 1976: 33; Esser 2009: 49; Brinker
2010: 31). An example of cataphoric reference is given below.
(14) I would never have believed it. They’ve accepted the whole scheme.
In this example, the pronoun it refers forwards to the second clause
(They’ve accepted the whole scheme). The pronoun it and the second
clause are thus in a relation of cataphoric reference.
As has been shown in the present section, we can distinguish between
different types of exophoric and endophoric reference. The distinctions
made in the present study are summarized in Figure 2.2 on the next page.
4
In the present study, all corpus examples which are cited from Sinclair 1990 are
examples taken from the Birmingham corpora of texts (cf. Sinclair 1990: x).
Reference
Exophoric reference Endophoric reference
Specific Non-specific Generic Anaphoric Cataphoric

reference reference reference reference reference
Figure 2.2: Types of reference
It should be noted that some authors make more detailed distinctions be-
tween different types of exophoric reference (cf. e.g. Lyons 1977: 178-
197; Radden/Dirven 2007: 89-112). However, the distinctions as shown
in Figure 2.2 are sufficient for the purpose of the present study.
2.3 A definition of coreference

While there exist several different definitions of reference (cf. section 2.1
above), many authors agree on a definition of ‘coreference’ (cf. e.g. Hal-
liday/Hasan 1976: 3; Esser 2009: 34f.; Brinker 2010: 26). Following the
approaches of Halliday/Hasan (1976: 3), Esser (2009: 34f.) and Brinker
(2010: 26), in the present study coreference is defined as the relation be-
tween two or more linguistic items in a text that have the same referent in
the extralinguistic world. The following example of coreference is taken
from a short story by Hemingway and is provided by Schubert (2012: 34).
(Hemingway 1980: 459; repeated from chapter 1)
In example (15), the noun phrases Nick, He and he exophorically refer to
the same person, i.e. they have the same referent. We can thus say that
Nick, He and he are ‘coreferential’ linguistic items (cf. e.g. Halli-
day/Hasan 1976: 3; Quirk et al. 1985: 347; Swanson 2003: 28). Further-
more, if coreference exists between two or more linguistic items in a text,
these items are considered to form a ‘coreference chain’ in the present
18 Chapter 2
study (cf. Halliday/Hasan 1989: 84, who use the term ‘identity chain’,
and Biber et al. 1999: 234; Azzam et al. 2000: 77; Kunz 2010: 1, 10). The
linguistic items which form coreference chains are called ‘coreference
chain elements’. In example (15) above, Nick, He and he are considered
as elements of a coreference chain.
Establishing coreference in a text serves an important function: it con-
tributes to connecting different clauses or clause complexes in a text with
one another (cf. Halliday/Hasan 1976: 3). Thus, the concept of corefer-
ence is closely related to the concepts of ‘cohesion’ and ‘coherence’. The
term ‘cohesion’ refers to the overt grammatical and lexical relations be-
tween the linguistic items in a text, both within a clause or clause com-
plex and across the boundaries of clauses or clause complexes (cf. de
Beaugrande/Dressler 1981: 3, 48-50; Bußmann 2002: 352; Esser 2009: 14;
Schubert 2012: 31f.). Cohesion is thus concerned with the actual word-
forms in a text (cf. Esser 2009: 14). It can be established, for example, by
a pronoun which refers back to a previous noun phrase in the same text
(cf. de Beaugrande/Dressler 1981: 49; Bußmann 2002: 352; Esser 2009:
14). Cohesion can thus be found in example (15) above since the pro-
nouns He and he refer to the previous noun phrase Nick.
The term ‘coherence’ has been used to describe the semantic relations
between the linguistic items in a text. These semantic relations do not
have to be overtly marked (cf. de Beaugrande/Dressler 1981: 4; Bußmann
2002: 351; Esser 2009: 15; Schubert 2012: 20). Coherence is based, for
example, on temporal relations and relations of causality (cf. de Beau-
grande/Dressler 1981: 4f.; Esser 2009: 15). Example (16) illustrates the
concept of coherence.
(16) The policeman held up his hand. The car stopped. (Quirk et al. 1985:
1428)
The two clauses in (16) are not explicitly connected by the use of pro-
nouns or other linguistic devices of cohesion. However, because of his or
her world knowledge the text receiver understands that the behaviour of
the policeman (holding up his hand) probably was the reason why the car
stopped (cf. Esser 2009: 11). Since the two clauses in (16) are linked by a
causal relation, we say that there is coherence in example (16). When dis-
cussing cohesion and coherence, it should be noted that some authors
have defined these concepts in a different way than it has been suggested
here (cf. e.g. Halliday/Hasan 1976: vii, 4-7, 23; Brinker 2010: 17).
While cohesion and coreference are closely linked, it is important to
point out that cohesion does not always imply coreference (cf. Esser 2009:
45; Schubert 2012: 37). An example of cohesion without coreference is
given below.
(17) These biscuits are stale. – Get some fresh ones. (Halliday/Hasan
1976: 92)
In example (17), there is a cohesive relation between the two clauses be-
cause some fresh ones refers back to These biscuits. As has been ex-
plained in section 2.2, this type of backwards reference within the text is
called endophoric or, more specifically, anaphoric reference. However, in
(17) there is no coreference between the noun phrases These biscuits and
some fresh ones. The semantic relation between the two noun phrases can
be called ‘co-classification’ (cf. Halliday/Hasan 1989: 74). ‘Co-
classification’ has been defined as follows:
In this type of meaning relation, the things, processes, or circum-
stances to which A and B refer belong to an identical class, but each
end of the cohesive tie refers to a distinct member of this class. (Hal-
liday/Hasan 1989: 74)
In example (17) above, both These biscuits and some fresh ones refer to a
class that could be called ‘biscuits’. However, each of the noun phrases
refers to different biscuits and thus to different members of the class.
Another example of cohesion without coreference is provided by
Schubert (2012: 49). The example is taken from a novel by David Lodge.
20 Chapter 2
(18) Morris went down to the sixth floor, crossed the landing and trav-
elled up to the ninth, […]. (Lodge 1978: 226)
In example (18), there is a cohesive relation between down and up since
the two adverbs are antonyms and the use of antonyms is considered as
one of the many devices that can establish cohesion in a text (cf. Esser
2009: 43; Schubert 2012: 49). However, the two adverb phrases are not
related by coreference but instead by ‘co-extension’. ‘Co-extension’ is
used as a term for the relation between linguistic elements that refer to
something within the same “general field of meaning” (Halliday/Hasan
1989: 74). It can be argued that in example (18) down and up both refer to
something within the general field of meaning called ‘directions of move-
ment’.
2.4 Types of coreference

The present study focuses on a type of coreference in English and Ger-
man that is called ‘nominal coreference’ (cf. Kunz 2010: 1). ‘Nominal
coreference’ is here used as a term for coreference between two or more
linguistic items of which at least one is a noun phrase. The other items in
a nominal coreference chain do not have to be noun phrases, however. If
two or more linguistic items in a text are in a relation of coreference and
if at least one of them is a noun phrase, the items are considered as ele-
ments of a ‘nominal coreference chain’ in the present study. For the sake
of brevity, we will speak of ‘coreference’ instead of ‘nominal corefer-
ence’ in most cases in the following sections and chapters of the present
study, except for in cases where this might lead to ambiguities.
We can distinguish three different types of nominal coreference:
‘anaphoric’, ‘cataphoric’ and ‘non-directional’ coreference. The concepts
of anaphoric and cataphoric coreference are well established in linguistic
research (cf. e.g. Halliday/Hasan 1976: 14, 17; Biber et al. 1999: 234, 348;
Esser 2009: 35, 49f.; Kunz 2010: 11). In addition to these two types of
coreference, the present study suggests the third type of non-directional
coreference (for brief remarks on concepts related to this type of corefer-

ence cf. Martin 1992: 145f.; van Deemter/Kibble 2000: 629f.).
In the present study, the concept of anaphoric coreference is defined
as follows. Anaphoric coreference is established if a linguistic item refers
backwards to a previous linguistic item in the same text and both of the
linguistic items have the same referent (cf. Halliday/Hasan 1976: 14;
Biber et al. 1999: 234; Esser 2009: 35). The second linguistic item thus
depends on the first linguistic item for its interpretation. That is, the text
receiver can only correctly interpret the second item if he or she has heard
or read the first item. An example of anaphoric coreference is provided
below.
(19) Students can opt for a specialized learning track focused on hospital-
ity, engineering, health or other sectors. They connect with employ-
ers through internships. (Foroohar 2014: 32)
In this example, Students and They both refer to the same referent in the
extralinguistic world. Since They refers backwards to Students, example
(19) contains an instance of anaphoric coreference. Examples (3), (4), (13)
and (15) in sections 2.1 to 2.3 above also contain anaphoric coreference.
As for anaphoric coreference chains, one point that has been dis-
cussed in the literature is to which elements of the coreference chain an
anaphoric chain element refers back if the coreference chain consists of
more than two chain elements (cf. e.g. Halliday/Hasan 1976: 15; Emmott
1989: 64-75; Hoey 1991: 72; Martin 1992: 141). The following example
serves as an illustration. It is repeated from section 2.3.
(Hemingway 1980: 459; example provided by Schubert 2012: 34)
In the above example, Nick, He and he are elements of the same corefer-
ence chain. The second chain element, the pronoun He, refers anaphori-
cally to Nick. For the third chain element, the pronoun he, it seems more
difficult to determine to which of the other two chain elements it refers: it
22 Chapter 2
might be argued that the pronoun he refers back only to the first chain
element, Nick. By contrast, it might also be assumed that the pronoun he
refers back only to the previous chain element, He, or that he refers back
both to Nick and He (cf. Halliday/Hasan 1976: 15; Emmott 1989: 64-75;
Hoey 1991: 72; Martin 1992: 141). The present study assumes that each
anaphoric coreference chain element refers back to all the previous ele-
ments of the same coreference chain since it is in a relation of coreference
with all of them. Thus, it is suggested that in example (20) he refers back
to both He and Nick (cf. Hoey 1991: 72).
In contrast to anaphoric coreference, cataphoric coreference is estab-
lished if a linguistic item refers forwards to a linguistic item in the same
text and both of the linguistic items have the same referent (cf. Halli-
day/Hasan 1976: 17; Biber et al. 1999: 348; Esser 2009: 49f.). Here is an
example.
(21) This should interest you, if you’re still keen on boxing. The world
heavyweight championship is going to be held in Chicago next June,
so you should be able to watch it live. (Quirk et al. 1985: 1462)
In example (21), the demonstrative pronoun This and the second clause
complex both refer to the same proposition and thus have the same refer-
ent. Since the pronoun This refers forwards to the second clause complex,
example (21) contains an instance of cataphoric coreference. In the pre-
sent study, the element of a coreference chain which follows a cataphoric
element of the same coreference chain is called the ‘postcedent’ of the
coreference chain (cf. e.g. Bache 2000: 174; Schwarz 2000: 52; Vater
2012: 92). Usually, cataphoric coreference occurs less frequently in a text
than anaphoric coreference (cf. de Beaugrande/Dressler 1981: 60f.; Quirk
et al. 1985: 347, 351).
In section 2.2 above, the concepts of anaphoric and cataphoric refer-
ence have been described. It should be noted that anaphoric reference
does not always imply anaphoric coreference and that cataphoric refer-
ence does not always imply cataphoric coreference. The following exam-
ple contains an instance of anaphoric reference, but anaphoric coreference

is not established in the example.
(22) They didn’t stop at the nearest building, but drove on to a bigger one
with many more vehicles outside. (British National Corpus CEU
3012; example provided by Esser 2009: 41)
In this example, a bigger one refers back to the nearest building, which
means that there is anaphoric reference within the text. However, corefer-
ence is not established in example (22) because the two noun phrases the
nearest building and a bigger one each refer to a different building. The
two noun phrases therefore refer to two different members of the class of
buildings. Instead of coreference, example (22) thus illustrates ‘co-
classification’, i.e. two linguistic items which refer to two different mem-
bers of the same class (cf. section 2.3 above).
Cataphoric reference without cataphoric coreference can be found in
cases in which cataphoric reference is established only by the definite ar-
ticle the. Some authors have suggested that the definite article the can be
used to refer forwards to a noun phrase postmodification (cf. Halli-
day/Hasan 1976: 72; Quirk et al. 1985: 268; Biber 1999: 264; Esser 2009:
49; Halliday 2014: 625). This type of cataphoric reference has been called
‘structural’ cataphoric reference (cf. Halliday 2014: 625). Here is an ex-
ample.
(23) The patterns of industrial development in the United States are too
varied to be categorized easily. (corpus example from Biber 1999:
264)
Biber et al. (1999: 264) suggest that in the above example the definite ar-
ticle The refers forwards to the postmodifying prepositional phrase of in-
dustrial development in the United States. Thus, example (23) contains
cataphoric reference, but it does not contain cataphoric coreference.
As a further remark on anaphoric and cataphoric coreference, it
should be noted that both anaphoric and cataphoric coreference are based
24 Chapter 2
on endophoric reference since anaphoric and cataphoric coreference chain

elements refer forwards or backwards within in the text. In addition, ana-
phoric and cataphoric coreference are also based on exophoric reference
because the elements of an anaphoric or cataphoric coreference chain re-
fer exophorically to the same referent in the extralinguistic world.
Having discussed anaphoric and cataphoric coreference, we will now
turn to non-directional coreference. By contrast to anaphoric and cata-
phoric coreference, non-directional coreference is not based on endo-
phoric reference, i.e. it does not include linguistic items which refer for-
wards or backwards within the text. Instead, non-directional coreference
is only based on exophoric reference. Non-directional coreference is es-
tablished if two or more linguistic items share the same referent but the
linguistic items do not refer forwards or backwards to each other (for
brief remarks on related concepts cf. Martin 1992: 145f.; van Deem-
ter/Kibble 2000: 629f.). Since the linguistic items do not refer in any di-
rection within the text, this type of coreference is called ‘non-directional’
in the present study. Linguistic items that are in a relation of non-
directional coreference can be successfully interpreted without taking into
account other linguistic items in the same text. An example of non-
directional coreference is provided below.
(24) One big disadvantage is that turtles do not have gills and so cannot
survive underwater indefinitely, like fish, but must come up for air
regularly. However, turtles can remain submerged for much longer
than, say, people can. (Browder 2004b: 1370)
In example (24), coreference is established between the two instances of
the noun phrase turtles since the two noun phrases have the same referent
in the extralinguistic world: both noun phrases refer to the class of turtles
as a whole. The first instance of turtles does not refer forwards or back-
wards in the text. The second instance of turtles also does not refer ana-
phorically or cataphorically. Thus, the two instances of turtles are not in a
relation of anaphoric or cataphoric coreference, but in a relation of non-

directional coreference.
In the present study, a linguistic item which is the first element of a
coreference chain and which is followed by an anaphoric or a non-
directional element of the same coreference chain is called the ‘antece-
dent’ of the coreference chain (cf. e.g. Bache 2000: 174; Vater 2012:
90f.). Thus, in example (24) above the antecedent of the coreference
chain is the first instance of the noun phrase turtles.
It should be pointed out that the two coreference chain elements in
example (24) each have generic reference: instead of referring to specific
turtles, the two instances of turtles refer to the class of turtles as a whole
(cf. section 2.2 above for a definition of generic reference). Like example
(24) above, example (25) below also includes noun phrases with generic
reference which establish non-directional coreference. Since example (25)
contains two different chains of non-directional coreference, indices are
used to distinguish between the two chains (cf. e.g. Stede 2007: 56 for a
similar use of indices).
(25) Over time all rock types1 can convert into other forms and this has
often been termed the rock cycle. Igneous and sedimentary rocks2
can become metamorphic rocks under pressure and heat. All rock
types1 can erode to form the layers of sediment that can eventually
become sedimentary rocks2, […]. (Krom 2008: 36)
Example (25) contains the two noun phrases all rock types and All rock
types. These noun phrases are in a relation of non-directional coreference
since they have the same referent and do not refer forwards or backwards
to each other. The two noun phrases have generic reference because they
do not refer to specific rock types, but to the class of rock types as a
whole. Furthermore, example (25) contains two instances of the noun
phrase sedimentary rocks. These two noun phrases also form a chain of
non-directional coreference. The two instances of sedimentary rocks have
26 Chapter 2
generic reference since they refer to the class of sedimentary rocks in

general.
When searching for examples of non-directional coreference, it was
found that this type of coreference is often established by noun phrases
with generic reference. However, the present study assumes that non-
directional coreference can also be created by noun phrases with non-
specific or specific reference. This will be shown in examples (26) and
(27). Example (26) contains two noun phrases with non-specific reference
which form a non-directional coreference chain. In example (26), the
capital letters A and B, which indicate that there are two different speak-
ers, have been added by the author of the present study.
(26) A: Would you like a cup of tea?
B: No, thank you. I don’t fancy tea so early in the morning.
A: So do have something else. Have a milkshake instead.
B: Sorry, if I drink a milkshake now, it will upset my stomach.
(Radden/Dirven 2007: 96)
In the above example, a relation of non-directional coreference is created
between the two instances of the noun phrase a milkshake. The two in-
stances of a milkshake in example (26) do not refer to a specific milk-
shake, but instead to any milkshake. Therefore, they have non-specific
reference.
The present study assumes that noun phrases with specific reference
can also establish non-directional coreference, for example noun phrases
which refer to specific dates, lengths, weights or amounts of money and
which do not refer forwards or backwards. Example (27) illustrates this.
(27) With reference to the letter of 5 May 1992 and in accordance with
Article 15 of the Convention for the Protection of Human Rights
and Fundamental Freedoms, I have the honour to inform you that
under legislative decree no. 225 of 9 March 1993, the state of emer-
gency was lifted from the town of Elazig and its district; however,
under the legislative decree no. 226 of 9 March 1993 the state of
emergency was declared on the town of Bitlis and its district. (Mur-
doch 2002: 118)
In example (27), coreference is created by the two instances of the noun
phrase 9 March 1993 because both noun phrases refer to the same date.
Since the two noun phrases do not refer forwards or backwards in the text,
the relation between them is interpreted here as a relation of non-
directional coreference. The above discussion of example (27) further-
more shows that the present study assumes that coreference can also be
established by noun phrases which occur as postmodifications of other
noun phrases since both instances of 9 March 1993 are noun phrase post-
modifications.
The three different types of coreference which are distinguished in the
present study are summarized in Figure 2.3 below. As has been explained
above, anaphoric and cataphoric coreference are based on endophoric and
exophoric reference. By contrast, non-directional coreference is only
based on exophoric reference.
Coreference
Anaphoric Cataphoric Non-directional

coreference coreference coreference
Based on Based on
endophoric and exophoric reference exophoric reference
Figure 2.3: Types of coreference
In some cases, coreference is established by the use of proper names,

such as John. In these cases, it seems slightly difficult to determine
whether the proper names establish anaphoric or non-directional corefer-
ence. The following example illustrates this.
28 Chapter 2
(28) Presently the Rabbit came up to the door, and tried to open it; but, as
the door opened inwards, and Alice’s elbow was pressed hard
against it, that attempt proved a failure. Alice heard it say to itself
“Then I’ll go round and get in at the window.” (Carroll 1998 [1865]:
34)
In this example, coreference is established between the two noun phrases
Alice’s and Alice. The noun phrase Alice does not refer backwards as
clearly as, for example, a personal pronoun such as she. Nonetheless, the
present study suggests that the noun phrase Alice in example (28) estab-
lishes anaphoric coreference with the noun phrase Alice’s. The reason for
this suggestion is that the name Alice can be used to refer to many differ-
ent people in the extralinguistic world. Thus, when the text receiver hears
or reads example (28), he or she must keep in mind that the noun phrase
Alice in (28) refers backwards to the noun phrase Alice’s and that the two
noun phrases refer to the same person. Therefore, the present study as-
sumes that proper names establish anaphoric coreference (cf. Halli-
day/Hasan 1976: 19; Martin 1992: 99).
As a further point, it should be noted that anaphoric and non-
directional coreference can occur within the same coreference chain. This
is shown in the next example.
(29) Catfish are not built for speed; they have no reason to swim fast.
Most of the animals that catfish eat are slow moving, […]. (Browder
2004a: 300)
In this example, coreference is established between the following noun
phrases: Catfish, they and catfish. The pronoun they establishes anaphoric
coreference because it refers backwards to the noun phrase Catfish. By
contrast, the noun phrase catfish in the second orthographic sentence es-
tablishes non-directional coreference since it does not refer forwards or
backwards in the text.
In addition to distinguishing between anaphoric, cataphoric and non-
directional coreference, a distinction can also be made between corefer-
ence within a clause complex 5 and coreference across clause complex

boundaries (cf. e.g. Halliday/Hasan 1976: 7f.; Schubert 2012: 31, 33f.).
Two examples are shown below. Example (31) has already been quoted
as example (19) above and is repeated here for convenience.
(30) David Cameron said he is working hard to achieve ‘massive struc-
tural reforms’. (Wintour 20126)
(31) Students can opt for a specialized learning track focused on hospital-
ity, engineering, health or other sectors. They connect with employ-
ers through internships. (Foroohar 2014: 32)
In example (30) above, David Cameron and he are elements of the same
coreference chain. In this example, coreference is established within a
clause complex. By contrast, in example (31) coreference is created
across clause complex boundaries. The three types of coreference (ana-
phoric, cataphoric and non-directional) can each be established within a
clause complex or across clause complex boundaries. For anaphoric
coreference, this has been shown in examples (30) and (31) above. For
examples of cataphoric and non-directional coreference within a clause
complex and across clause complexes cf. sections 3.2 and 3.3 below.
2.5 Syntactic categories in nominal coreference chains

As has been noted in section 2.1, many authors assume that not only noun
phrases can refer to something in the extralinguistic world, but also other
phrases and clauses or clause complexes (cf. e.g. Vater 2001: 90). As a
consequence, coreference can be found not only between noun phrases,
but also between other phrases, clauses or clause complexes (cf. e.g. Hal-
liday/Hasan 1976: 52f., 74f.; Esser 2009: 40f., 45f.).
As has been explained in section 2.4, the present study focuses on a
type of coreference in English and German that is called ‘nominal
5
For a definition of the term ‘clause complex’ cf. section 2.1 above.
6
The article by Wintour (2012) is an online source which does not provide page
numbers.
30 Chapter 2
coreference’, i.e. coreference between two or more linguistic items of

which at least one is a noun phrase. A nominal coreference chain thus al-
ways includes at least one noun phrase, but it can also include chain ele-
ments which belong to different syntactic categories. We will now take a
closer look at the different syntactic categories of the elements of nominal
coreference chains in English and German.
Firstly, a nominal coreference chain can be formed by two or more
noun phrases. This has been illustrated in example (15) in section 2.3
above, which is repeated below as example (32) for convenience.
(Hemingway 1980: 459; example provided by Schubert 2012: 34)
In example (32), the coreference chain consists of three noun phrases
(Nick, He and he). A German example of two coreferential noun phrases
is provided below.
(33) Hallo, sagte Frau Brücker, als sei ich erst gestern dagewesen. Ein-
mal wie immer?
Sie hantierte an einer großen gußeisernen Pfanne. (Timm 2000: 7)
In addition to noun phrases, nominal coreference chains can also in-
clude determiners. In the present study, ‘determiner’ is understood as a
term for closed-class items which precede the head of a noun phrase (cf.
Quirk et al. 1985: 253). If an item such as these occurs as a noun phrase
head, for example in Pick these up! (cf. Halliday/Hasan 1976: 58), it is
classified as a pronoun and not as a determiner in the present study (cf.
Quirk et al. 1985: 372). It should also be noted that items such as our or
unser are considered here as possessive determiners and not as pronouns
(cf. e.g. Jackson 1982: 64). Furthermore, it is assumed that determiners,
as closed-class items, do not belong to the premodification of a noun
phrase (cf. e.g. Quirk et al. 1985: 1238f.).
As has been mentioned above, the present study holds that clauses,
clause complexes and different types of phrases have referents in the ex-
tralinguistic world (cf. e.g. Jackendoff 1983: 52f.; Chur 1993: 8; Vater
2001: 90). Furthermore, the present study assumes that determiners can-
not form phrases by themselves but that they occur as part of noun
phrases (cf. e.g. Quirk et al. 1985: 253, 1238f.). Since determiners do not
form phrases by themselves, it might be argued that they do not have ref-
erents in the extralinguistic world (cf. Vater 2001: 90). However, in sev-
eral textlinguistic studies possessive determiners have been marked as
elements of coreference chains (cf. e.g. Halliday/Hasan 1976: 54f.; Stede
2007: 56; Brinker 2010: 32). Therefore, possessive determiners are ac-
cepted as parts of coreference chains in the present study and it is as-
sumed that they can have extralinguistic referents. Two examples of
coreference chains that include determiners are given below.
(34) ‘Thanks,’ said Brody. He hung up, turned out the light in his office,
and walked out to his car. (corpus example from Sinclair 2005: 3867)
(35) Die Musiker ziehen amüsiert ihre Augenbrauen hoch, die Frauen
halten sich vor Staunen die Hände vor den Mund. (Lerchenmüller
2009: 1)
In example (34), coreference is established between the noun phrase
Brody, the pronoun He and the two instances of the possessive determiner
his. In example (35), the noun phrase Die Musiker and the possessive de-
terminer ihre are elements of the same coreference chain.
In addition to noun phrases and determiners, clauses or clause com-
plexes can also be part of a coreference chain (cf. Halliday/Hasan 1976:
52f., 66-70; Quirk et al. 1985: 1461-1463; Esser 2009: 40, Brinker 2010:
31). Example (36) illustrates this.
(36) We went to the opera last night. That was our first outing for months.
7
In the present study, all corpus examples which are cited from Sinclair 2005 are
examples taken from the Bank of English Corpus (cf. Sinclair 2005: xii).
32 Chapter 2
In the above example, the demonstrative pronoun That refers back to, and
is coreferential with, the clause We went to the opera last night. The fol-
lowing example is taken from a German newspaper article and is pro-
vided by Brinker (2010: 31).
(37) Als die Kinder die Macht ergriffen, gingen die Eltern in Deckung.
Luftballons flogen über eingezogene Köpfe. Mobiliar polterte über
das Parkett. Der Fußboden bebte unter stampfendem Toben. Im
Souterrain rieselte Kalk. Triumphgeschrei aus heiseren Kehlen hall-
te hinaus in den Grunewald. 33 Kinder (zwischen zwei und zwölf
Jahren) hatten die Freiheit entdeckt.
Das war gegen 18 Uhr am ersten Tag einer außergewöhnlichen Wo-
che. (Siebenschön 1972: 39)
In example (37), the demonstrative pronoun Das is in a relation of
coreference with all the clauses that precede it. The pronoun thus refers
back to relatively long passage of text (cf. Brinker 2010: 31). It should be
noted that Halliday/Hasan (1976: 52f., 66-70) use the term ‘extended ref-
erence’ to describe examples such as (36) and (37) since a linguistic item
refers back to an extended passage of text.
In addition to noun phrases, determiners and clauses or clause com-
plexes, adjective phrases can also be part of nominal coreference chains
(cf. Ross 1969: 356f.; Quirk et al. 1985: 349). This is shown by the fol-
lowing two examples.
(38) Jack is clever, but he doesn’t look it. (James 1980: 7; the example is
an adapted version of an example from Ross 1969: 357)
(39) Hans ist klug, aber seine Söhne sind es nicht. (James 1980: 7; the
example is an adapted version of an example from Ross 1969: 356)
The present study assumes that in example (38) the pronoun it is in a rela-
tion of coreference with the adjective phrase clever. Similarly, it is held
that in example (39) coreference is established between the adjective
phrase klug and the pronoun es.
3 Coreference chain elements in English and German:
devices and functions
3.1 Devices for anaphoric coreference
3.1.1 Main devices for anaphoric coreference
3.1.1.1 Pronouns
Both in English and in German, there exist various linguistic devices that
can establish nominal coreference, for example personal pronouns. The
present study calls them ‘devices for coreference’ or ‘coreferential de-
vices’. In this study, linguistic items are analyzed as devices for corefer-
ence if they establish coreference in a text. This does not mean that these
linguistic items can only be used to establish coreference. For example, as
has been shown in section 2.5, demonstrative pronouns can create
coreference. However, they can also be used to refer only exophorically
(cf. e.g. Halliday/Hasan 1976: 58). For instance, in an utterance such as
Pick these up! the speaker may use the demonstrative pronoun these to
refer to something that is present in the situational context. In the above
utterance, the demonstrative pronoun thus refers only exophorically, does
not create coreference and cannot be analyzed as a device for coreference
(cf. Halliday/Hasan 1976: 58).
The present study assumes that, with the exception of antecedents and
postcedents, each element of a coreference chain can be analyzed as a de-
vice for coreference. Furthermore, the present study makes a distinction
between ‘main devices’ and ‘supplementing devices’ for nominal corefer-
ence in English and German. While main devices can establish corefer-
ence without being combined with another device, supplementing devices
can only establish coreference if they are combined with a main device.
The present study assumes that, with the exception of antecedents and
postcedents, each coreference chain element can be analyzed as one main
device and one supplementing device. Moreover, the present study dis-
tinguishes between devices for anaphoric, cataphoric and non-directional
Another random document with
no related content on Scribd:
The youth turned and shuffled from the room, wiping his nose upon
his sleeve, and Sir Richard's attention came back to Blake.
"Thou are not ill-favored, fellow," he said. "'Tis a pity that thou beest
not of noble blood, for thy mien appearth not like that of one
lowborn."
"I never considered myself lowborn," said Blake, with a grin.
"Thy father, now—was he not at least a sir knight?"
Blake was thinking quickly now. He was far from being able as yet to
so much as hazard a guess that might explain his host's archaic
costume and language, but he was sure that the man was in
earnest, whether sane or not, and were he not sane it seemed
doubly wise to humor him.
"Yes, indeed," he replied, "my father is a thirty second degree Mason
and a Knight Templar."
"Sblud! I knew it," cried Sir Richard.
"And so am I," added Blake, when he realized the happy effect his
statement had produced.
"Ah, I knew it! I knew it!" cried Sir Richard. "Thy bearing proclaimed
thy noble blood; but why didst thou seek to deceive me? And so thou
art one of the Poor Knights of Christ and of the Temple of Solomon
who guard the way of the pilgrims to the Holy Land! This explaineth
thy poor raiment and glorifies it."
Blake was mystified by the allusion, as the picture always suggested
by a reference to Knights Templar was of waving white plumes,
gorgeous aprons and glittering swords. He did not know that in the
days of their origin they were clothed in any old garments that the
charity of others might bequeath them.
At this moment Michel returned bearing a wooden trencher
containing cold mutton and several pieces of simnel bread and
carrying in one hand a flagon of wine. These he set upon the table
before Blake and going to a cupboard fetched two metal goblets into
which he decanted a portion of the contents of the flagon.
Sir Richard arose and taking one of the goblets raised it before him
on a level with his head.
"Hal, Sir James!" he cried, "and welcome to Nimmr and the Valley of
the Sepulcher!"
"Here's looking at you!" replied Blake.
"A quaint saying," remarked Sir Richard. "Methinks the ways of
England must be changed since the days of Richard the Lion
Hearted when my noble ancestor set forth upon the great crusade in
the company of his king. Here's looking at you! Ods bodikins! I must
not let that from my memory. Here's looking at you! Just wait thou 'til
some fair knight doth drink my health—I shall lay him flat with that!
"But, stay! Here, Michel, fetch yon stool for Sir James, and eat, sir
knight. Thou must be passing hungry."
"I'll tell the world I am," replied Blake, feelingly, as he sat down on
the stool that Michel brought. There were no knives or forks, but
there were fingers and these Blake used to advantage while his host
sat smiling happily at him from across the rude table.
"Thou art better than a minstrel for pleasure," cried Sir Richard. "I'll
tell the world I am! Ho, ho! Thou wilt be a gift from heaven in the
castle of the prince. I'll tell the world I am!"
When Blake had satisfied his hunger, Sir Richard ordered Michel to
prepare horses. "We ride down to the castle, Sir James," he
explained. "No longer art thou my prisoner, but my friend and guest.
That I should have received thee so scurvily shalt ever be to my
discredit."
Mounted upon prancing chargers and followed at a respectful
distance by Michel, the two rode down the winding mountain road.
Sir Richard now carried his shield and lance, a pennon fluttering
bravely in the wind from just below the tip of the latter, the sun
glancing from the metal of his hauberk, a smile upon his brave face
as he chatted with his erstwhile prisoner. To Blake he seemed a
gorgeous picture ridden from out the pages of a story book. Yet,
belying his martial appearance, there was a childlike simplicity about
the man that won Blake's liking from the first, for there was that
about him that made it impossible for one to conceive him as the
perpetrator of a dishonorable act.
His ready acceptance of Blake's statements about himself bespoke a
credulity that seemed incompatible with the high intelligence
reflected by his noble countenance, and the American preferred to
attribute it to a combination of unsophistication and an innate
integrity which could not conceive of perfidy in others.
As the road rounded the shoulder of a hill, Blake saw another
barbican barring the way and, beyond, the towers and battlements of
an ancient castle. At a command from Sir Richard the warders of the
gate opened to them and the three rode through into the ballium.
This space between the outer and inner walls appeared unkept and
neglected. Several old trees flourished within it and beneath the
shade of one of these, close to the outer gateway, lolled several
men-at-arms, two of whom were engaged in a game that resembled
draughts.
At the foot of the inner wall was a wide moat, the waters of which
reflected the gray stones of the wall and the ancient vines that,
growing upon its inner side, topped it to form a leafy coping that
occasionally hung low upon the outer side.
Directly opposite the barbican was the great gateway in the inner
wall and here a drawbridge spanned the moat and a heavy portcullis
barred the way into the great court of the castle; but at a word from
Sir Richard the gate lifted and, clattering across the drawbridge, they
rode within.
Before Blake's astonished eyes loomed a mighty castle of rough
hewn stone, while to the right and left, within the great court, spread
broad gardens not illy kept, in which were gathered a company of
men and women who might have just stepped from Arthur's court.
At sight of Sir Richard and his companion the nearer members of the
company regarded Blake with interest and evident surprise. Several
called greetings and questions to Sir Richard as the two men
dismounted and turned their horses over to Michel.
"Ho, Richard!" cried one. "What bringest thou—a Saracen?"
"Nay," replied Richard. "A fair sir knight who would do his devoir to
the prince. Where be he?"
"Yonder," and they pointed toward the far end of the court where a
large company was assembled.
"Come, Sir James!" directed Richard, and led him down the
courtyard, the knights and ladies following closely, asking questions,
commenting with a frankness that brought a flush to Blake's face.
The women openly praised his features and his carriage while the
men, perhaps prompted by jealousy, made unflattering remarks
about his soiled and torn apparel and its, to them, ridiculous cut; and
indeed the contrast was great between their gorgeous dalmaticas of
villosa or cyclas, their close-fitting tights, their colored caps and
Blake's drab shirt, whip-cord breeches and cordovan boots, now
soiled, torn and scratched.
The women were quite as richly dressed as the men, wearing
clinging mantles of rich stuff, their hair and shoulders covered with
dainty wimples of various colors and often elaborately embroidered.
None of these men, nor any of those in the assemblage they were
approaching, wore armor, but Blake had seen an armored knight at
the outer gateway and another at the inner and he judged that only
when engaged in military duties did they wear this heavy and
uncomfortable dress.
When they reached the party at the end of the court Sir Richard
elbowed his way among them to the center of the group where stood
a tall man of imposing appearance, chatting with those about him. As
Sir Richard and Blake halted before him the company fell silent.
"My lord prince," said Richard, bowing, "I bring thee Sir James, a
worthy Knight Templar who hath come under the protection of God
through the lines of the enemy to the gates of Nimmr."
The tall man eyed Blake searchingly and he had not the appearance
of great credulity.
"Thou sayest that thou comest from the Temple of Solomon in the
Kingdom of Jerusalem?" he demanded.
"Sir Richard must have misunderstood me," replied Blake.
"Then thou art no Knight Templar?"
"Yes, but I am not from Jerusalem."
"Perchance he is one of those doughty sir knights that guard the
pilgrims' way to the Holy Land," suggested a young woman standing
near the prince.
Blake glanced quickly at the speaker and as their eyes met, hers fell,
but not before he had seen that they were very beautiful eyes set in
an equally beautiful, oval face.
"More like it haps he be a Saracen spy sent among us by the sultan,"
snapped a dark man who stood beside the girl.
The latter raised her eyes to the prince. "He looketh not like a
Saracen, my father," she said.
"What knowest thou of the appearance of a Saracen, child?"
demanded the prince. "Hast seen so many?" The party laughed and
the girl pouted.
"Verily an' I hast seen full as many a Saracen as has Sir Malud or
thyself, my lord prince," she snapped, haughtily. "Let Sir Malud
describe a Saracen."
The dark young man flushed angrily. "At least," he said, "my lord
prince, I knowest an English knight when I seest one, an' if here be
an English knight then Sir Malud be a Saracen!"
"Enough," said the prince and then, turning to Blake: "If thou art not
from Jerusalem where art thou from?"
"New York," replied the American.
"Ha," whispered Sir Malud to the girl, "didst I not tell you?"
"Tell me what—that he is from New York? Where is that?" she
demanded.
"Some stronghold of the infidel," asserted Malud.
"New York?" repeated the prince. "Be that in the Holy Land?"
"It is sometimes called New Jerusalem," explained Blake.
"And thou comest to Nimmr through the lines of the enemy? Tell me,
sir knight, had they many men-at-arms? And how were their forces
disposed? Be they close upon the Valley of the Sepulcher? Thinkest
thou they plan an early attack? Come, tell me all—thou canst be of
great service."
"I have come for days through the forest and seen no living man,"
said Blake. "No enemy surrounds you."
"What?" cried the prince.
"Didst I not tell thee?" demanded Malud. "He is an enemy spy. He
wouldst lead us into the belief that we are safe that the forces of the
sultan may find us off our guard and take Nimmr and the Valley."
"Ods blud! Methinks thou beest right, Sir Malud," cried the prince.
"No enemy indeed! Why else then hast the knights of Nimmr lain
here seven and a half centuries if there be no horde of infidels
surrounding our stronghold?"
"Search me," said Blake.
"Eh, what?" demanded the prince.
"He hath a quaint manner of speech, my lord prince," explained
Richard, "but I do not think him an enemy of England. Myself will
vouch for him an' you will take him into your service, my lord prince."
"Wouldst enter my service, sir?" demanded the prince.
Blake glanced at Sir Malud and looked dubious—then his eyes
wandered to those of the girl. "I'll tell the world I would!" he said.
CHAPTER X
The Return of Ulala

Numa was hungry. For three days and three nights he had hunted
but always the prey had eluded him. Perhaps Numa was growing
old. Not so keen were his scent and his vision, not so swift his
charges, nor well timed the spring that heretofore had brought down
the quarry. So quick the food of Numa that a fraction of a second, a
hair's breadth, might mark the difference between a full belly and
starvation.
Perhaps Numa was growing old, yet he still was a mighty engine of
destruction, and now the pangs of hunger had increased his ferocity
many-fold, stimulated his cunning, emboldened him to take great
risks that his belly might be filled. It was a nervous, irascible,
ferocious Numa that crouched beside the trail. His up-pricked ears,
his intent and blazing eyes, his quivering nostrils, the gently moving
tail-tip, evidenced his awareness of another presence.
Down the wind to the nostrils of Numa the lion came the man-scent.
Four days ago, his belly full, Numa had doubtless slunk away at the
first indication of the presence of man, but today is another day and
another Numa.
Zeyd, three days upon the back track from the menzil of the sheik
Ibn Jad, thought of Ateja, of far Guad, congratulated himself upon
the good fortune that had thus far smiled upon his escape and flight.
His mare moved slowly along the jungle trail, unurged, for the way
was long; and just ahead a beast of prey waited in ambush.
But Numa's were not the only ears to hear, nor his nostrils the only
nostrils to scent the coming of the man-thing—another beast
crouched near, unknown to Numa.
Overanxious, fearful of being cheated of his meat, Numa made a
false move. Down the trail came the mare. She must pass within a
yard of Numa, but Numa could not wait. Before she was within the
radius of his spring he charged, voicing a horrid roar. Terrified the
mare reared and, rearing, tried to turn and bolt. Overbalanced, she
toppled backward and fell, and in falling unhorsed Zeyd; but in the
instant she was up and flying back along the trail, leaving her master
in the path of the charging lion.
Horrified, the man saw the snarling face, the bared fangs almost
upon him. Then he saw something else—something equally awe-
inspiring—a naked giant who leaped from a swaying branch full upon
the back of the great cat. He saw a bronzed arm encircle the neck of
the beast of prey as the lion was borne to earth by the weight and
impact of the man's body. He saw a heavy knife flashing in the air,
striking home again and again as the frenzied lion threw itself about
in futile effort to dislodge the thing upon its back. He heard the roars
and the growls of el-adrea, and mingled with them were growls and
snarls that turned his blood cold, for he saw that they came from the
lips of the man-beast.
Then Numa went limp and the giant arose and stood above the
carcass. He placed one foot upon it and, raising his face toward the
heavens, voiced a hideous scream that froze the marrow in the
bones of the Beduin—a scream that few men have heard: the victory
cry of the bull ape.
It was then that Zeyd recognized his saviour and shuddered again as
he saw that it was Tarzan of the Apes. The ape-man looked down at
him.
"Thou art from the menzil of Ibn Jad," he said.
"I am but a poor man," replied Zeyd. "I but followed where my
sheykh led. Hold it not against Zeyd sheykh of the jungle, that he be
in thy beled. Spare my poor life I pray thee and may Allah bless
thee."
"I have no wish to harm thee, Beduwy," replied Tarzan. "What wrong
hath been done in my country is the fault of Ibn Jad alone. Is he
close by?"
"Wellah nay, he be many marches from here."
"Where art thy companions?" demanded the ape-man.
"I have none."
"Thou art alone?"
"Billah, yes."
Tarzan frowned. "Think well Beduwy before lying to Tarzan," he
snapped.
"By Ullah, I speak the truth! I am alone."
"And why?"
"Fahd did plot against me to make it appear that I had tried to take
the life of Ibn Jad, which, before Allah, is a lie that stinketh to
heaven, and I was to be shot; but Ateja, the daughter of the sheykh,
cut my bonds in the night and I escaped."
"What is thy name?"
"Zeyd."
"Whither goest thou—to thine own country?"
"Yes, to beled el-Guad, a Beny Salem fendy of el-Harb."
"Thou canst not, alone, survive the perils of the way," Tarzan warned
him.
"Of that I be fearful, but death were certain had I not escaped the
wrath of Ibn Jad."
For a moment Tarzan was silent in thought. "Great must be the love
of Ateja, the daughter of the sheik, and great her belief in you," he
said.
"Wellah, yes, great is our love and, too, she knew that I would not
slay her father, whom she loves."
Tarzan nodded. "I believe thee and shall help thee. Thou canst not
go on alone. I shall take thee to the nearest village and there the
chief will furnish you with warriors who will take you to the next
village, and thus from village to village you will be escorted to the
Soudan."
"May Allah ever watch over and guard thee!" exclaimed Zeyd.
"Tell me," said Tarzan as the two moved along the jungle trail in the
direction of the nearest village which lay two marches to the south of
them, "tell me what Ibn Jad doth in this country. It is not true that he
came for ivory alone. Am I not right?"
"Wellah yes, Sheykh Tarzan," admitted Zeyd. "Ibn Jad came for
treasure, but not for ivory."
"What, then?"
"In el-Habash lies the treasure city of Nimmr," explained Zeyd. "This
Ibn Jad was told by a learned sahar. So great is the wealth of Nimmr
that a thousand camels could carry away not a tenth part of it. It
consists of gold and jewels and—a woman."
"A woman?"
"Yes, a woman of such wondrous beauty that in the north she alone
would bring a price that would make Ibn Jad rich beyond dreams.
Surely thou must have heard of Nimmr."
"Sometimes the Gallas speak of it," said Tarzan, "but always I
thought it of no more reality than the other places of their legends.
And Ibn Jad undertook this long and dangerous journey on no more
than the word of a magician?"
"What could be better than the word of a learned sahar?" demanded
Zeyd.
Tarzan of the Apes shrugged.
During the two days that it took them to reach the village Tarzan
learned of the white man who had come to the camp of Ibn Jad, but
from Zeyd's description of him he was not positive whether it was
Blake or Stimbol.
As Tarzan travelled south with Zeyd, Ibn Jad trekked northward into
el-Habash, and Fahd plotted with Tollog, and Stimbol plotted with
Fahd, while Fejjuan the Galla slave waited patiently for the moment
of his delivery from bondage, and Ateja mourned for Zeyd.
"As a boy thou wert raised in this country, Fejjuan," she said one day
to the Galla slave. "Tell me, dost thou think Zeyd could make his way
alone to el-Guad?"
"Billah, nay," replied the black. "Doubtless he be dead by now."
The girl stifled a sob.
"Fejjuan mourns with thee, Ateja," said the black, "for Zeyd was a
kindly man. Would that Allah had spared your lover and taken him
who was guilty."
"What do you mean?" asked Ateja. "Knowest thou, Fejjuan, who
fired the shot at Ibn Jad, my father? It was not Zeyd! Tell me it was
not Zeyd! But thy words tell me that, which I well knew before. Zeyd
could not have sought the life of my father!"
"Nor did he," replied Fejjuan.
"Tell me what you know of this thing."
"And you will not tell another who told you?" he asked. "It would go
hard with me if one I am thinking of knew that I had seen what I did
see."
"I swear by Allah that I wilt not betray you, Fejjuan," cried the girl.
"Tell me, what didst thou see?"
"I did not see who fired the shot at thy father, Ateja," replied the
black, "but something else I saw before the shot was fired."
"Yes, what was it?"
"I saw Fahd creep into the beyt of Zeyd and come out again bearing
Zeyd's matchlock. That I saw."
"I knew it! I knew it!" cried the girl.
"But Ibn Jad will not believe if you tell him."
"I know; but now that I am convinced perhaps I shall find a way to
have Fahd's blood for the blood of Zeyd," cried the girl, bitterly.
For days Ibn Jad skirted the mountains behind which he thought lay
the fabled city of Nimmr as he searched for an entrance which he
hoped to find without having recourse to the natives whose haunts
he had sedulously avoided lest through them opposition to his
venture might develop.
The country was sparsely settled, which rendered it easy for the
'Aarab to avoid coming into close contact with the natives, though it
was impossible that the Gallas were ignorant of their presence. If
however the blacks were willing to leave them alone, Ibn Jad had no
intention of molesting them unless he found that it would be
impossible to carry his project to a successful issue without their
assistance, in which event he was equally ready to approach them
with false promises or ruthless cruelty, whichever seemed the more
likely to better serve his purpose.
As the days passed Ibn Jad waxed increasingly impatient, for,
search as he would, he could locate no pass across the mountains,
nor any entrance to the fabled valley wherein lay the treasure city of
Nimmr.
"Billah!" he exclaimed one day, "there be a City of Nimmr and there
be an entrance to it, and, by Allah, I will find it! Summon the Habash,
Tollog! From them or through them we shall have a clew in one way
or another."
When Tollog had fetched the Galla slaves to the beyt of Ibn Jad, the
old sheik questioned them but there was none who had definite
knowledge of the trail leading to Nimmr.
"Then, by Allah," exclaimed Ibn Jad, "we shall have it from the native
Habash!"
"They be mighty warriors, O brother," cried Tollog, "and we be far
within their country. Should we anger them and they set upon us it
might fare ill with us."
"We be Bedauwy," said Ibn Jad proudly, "and we be armed with
muskets. What could their simple spears and arrows avail against
us?"
"But they be many and we be few," insisted Tollog.
"We shall not fight unless we be driven to it," said Ibn Jad. "First we
shall seek, by friendly overtures, to win their confidence and cajole
the secret from them.
"Fejjuan!" he exclaimed, turning to the great black. "Thou art a
Habashy. I have heard thee say that thou well rememberest the days
of thy childhood in the hut of thy father and the story of Nimmr was
no new story to you. Go, then, and seek out thy people. Make friends
with them. Tell them that the great Sheykh Ibn Jad comes among
them in friendliness and that he hath gifts for their chiefs. Tell them
also that he would visit the city of Nimmr, and if they will lead him
there he will reward them well."
"I but await thy commands," said Fejjuan, elated at this opportunity
to do what he had long dreamed of doing. "When shall I set forth?"
"Prepare thyself tonight and when dawn comes depart," replied the
sheik.
And so it was that Fejjuan, the Galla slave, set forth early the
following morning from the menzil of Ibn Jad, sheik of the fendy el-
Guad, to search for a village of his own people.
By noon he had come upon a well-worn trail leading toward the west,
and this he followed boldly, guessing that he would best disarm
suspicion thus than by attempting to approach a Galla village by
stealth. Also he well knew that there was little likelihood that he could
accomplish the latter in any event. Fejjuan was no fool. He knew that
it might be difficult to convince the Gallas that he was of their blood,
for there was against him not alone his 'Aarab garments and
weapons but the fact that he would be able to speak the Galla
tongue but lamely after all these years.
That he was a brave man was evidenced by the fact that he well
knew the suspicious and warlike qualities of his people and their
inborn hatred of the 'Aarab and yet gladly embraced this opportunity
to go amongst them.
How close he had approached a village Fejjuan did not know. There
were neither sounds nor odors to enlighten him when there suddenly
appeared in the trail ahead of him three husky Galla warriors and
behind him he heard others, though he did not turn.
Instantly Fejjuan raised his hands in sign of peace and at the same
time he smiled.
"What are you doing in the Galla country?" demanded one of the
warriors.
"I am seeking the house of my father," replied Fejjuan.
"The house of your father is not in the country of the Gallas," growled
the warrior. "You are one of these who come to rob us of our sons
and daughters."
"No," replied Fejjuan. "I am a Galla."
"If you were a Galla you would speak the language of the Gallas
better. We understand you, but you do not speak as a Galla speaks."
"That is because I was stolen away when I was a child and have
lived among the Bedauwy since, speaking only their tongue."
"What is your name?"
"The Bedauwy call me Fejjuan, but my Galla name was Ulala."
"Do you think he speaks the truth?" demanded one of the blacks of a
companion. "When I was a child I had a brother whose name was
Ulala."
"Where is he?" asked the other warrior.
"We do not know. Perhaps simba the lion devoured him. Perhaps the
desert people took him. Who knows?"
"Perhaps he speaks the truth," said the second warrior. "Perhaps he
is your brother. Ask him his father's name."
"What was your father's name?" demanded the first warrior.
"Naliny," replied Fejjuan.
At this reply the Galla warriors became excited and whispered
among themselves for several seconds. Then the first warrior turned
again to Fejjuan.
"Did you have a brother?" he demanded.
"Yes," replied Fejjuan.
"What was his name?"
"Tabo," answered Fejjuan without hesitation.
The warrior who had questioned him leaped into the air with a wild
shout.
"It is Ulala!" he cried. "It is my brother. I am Tabo, Ulala. Do you not
remember me?"
"Tabo!" cried Fejjuan. "No, I would not know you, for you were a little
boy when I was stolen away and now you are a great warrior. Where
are our father and mother? Are they alive? Are they well?"
"They are alive and well, Ulala," replied Tabo. "Today they are in the
village of the chief, for there is a great council because of the
presence of some desert people in our country. Came you with
them?"
"Yes, I am a slave to the desert people," replied Fejjuan. "Is it far to
the village of the chief? I would see my mother and my father and,
too, I would talk with the chief about the desert people who have
come to the country of the Gallas."
"Come, brother!" cried Tabo. "We are not far from the village of the
chief. Ah, my brother, that I should see you again whom we thought
to be dead all these years! Great will be the joy of our father and
mother.
"But, tell me, have the desert people turned you against your own
people? You have lived with them many years. Perhaps you have
taken a wife among them. Are you sure that you do not love them
better than you love those whom you have not seen for many
years?"
"I do not love the Bedauwy," replied Fejjuan, "nor have I taken a wife
among them. Always in my heart has been the hope of returning to
the mountains of my own country, to the house of my father. I love
my own people, Tabo. Never again shall I leave them."
"The desert people have been unkind to you—they have treated you
with cruelty?" demanded Tabo.
"Nay, on the contrary they have treated me well," replied Fejjuan. "I
do not hate them, but neither do I love them. They are not of my own
blood. I am a slave among them."
As they talked the party moved along the trail toward the village
while two of the warriors ran ahead to carry the glad tidings to the
father and mother of the long missing Ulala. And so it was that when
they came within sight of the village they were met by a great crowd
of laughing, shouting Gallas, and in the forerank were the father and
mother of Fejjuan, their eyes blinded by the tears of love and joy that
welled at sight of this long gone child.
After the greetings were over, and every man, woman and child in
the company must crowd close and touch the returned wanderer,
Tabo conducted Fejjuan into the village and the presence of the
chief.
Batando was an old man. He had been chief when Ulala was stolen
away. He was inclined to be skeptical, fearing a ruse of the desert
people, and he asked many questions of Fejjuan concerning matters
that he might hold in his memory from the days of his childhood. He
asked him about the house of his father and the names of his
playmates and other intimate things that an impostor might not know,
and when he had done he arose and took Fejjuan in his arms and
rubbed his cheek against the cheek of the prodigal.
"You are indeed Ulala," he cried. "Welcome back to the land of your
people. Tell me now what the desert people do here. Have they
come for slaves?"
"The desert people will always take slaves when they can get them,
but Ibn Jad has not come first for slaves, but for treasure."
"Ai! what treasure?" demanded Batando.
"He has heard of the treasure city of Nimmr," replied Fejjuan. "It is a
way into the valley where lies Nimmr that he seeks. For this he sent
me to find Gallas who would lead him to Nimmr. He will make gifts
and he promises rich rewards when he shall have wrested the
treasure from Nimmr."
"Are these true words?" asked Batando.
"There is no truth in the beards of the desert dwellers," replied
Fejjuan.
"And if he does not find the treasure of Nimmr perhaps he will try to
find treasure and slaves in the Galla country to repay the expense of
the long journey he has undertaken from the desert country?" asked
Batando.
"Batando speaks out of the great wisdom of many years," replied
Fejjuan.
"What does he know of Nimmr?" asked the old chief.
"Naught other than what an old medicine man of the 'Aarab told
him," replied Fejjuan. "He said to Ibn Jad that great treasure lay
hoarded in the City of Nimmr and that there was a beautiful woman
who would bring a great price in the far north."
"Nothing more he told him?" demanded Batando. "Did he not tell him
of the difficulties of entering the forbidden valley?"
"Nay."
"Then we can guide him to the entrance to the valley," said Batando,
smiling slyly.
CHAPTER XI
Sir James
As Tarzan and Zeyd journeyed toward the village in which the ape-
man purposed to enlist an escort for the Arab upon the first stage of
his return journey toward his desert home, the Beduin had time to
meditate much upon many matters, and having come to trust and
respect his savage guide he at last unbosomed himself to Tarzan.
"Great Sheykh of the Jungle," he said one day, "by thy kindness thou
hast won the undying loyalty of Zeyd who begs that thou wilt grant
him one more favor."
"And what is that?" asked the ape-man.
"Ateja, whom I love, remains here in the savage country in constant
danger so long as Fahd be near her. I dare not now return to the
menzil of Ibn Jad even could I find it, but later, when the heat of Ibn
Jad's anger will have had time to cool, then I might come again
among them and convince him of my innocence, and be near Ateja
and protect her from Fahd."
"What, then, would you do?" demanded Tarzan.
"I would remain in the village to which you are taking me until Ibn
Jad returns this way toward el-Guad. It is the only chance that I have
to see Ateja again in this life, as I could not cross the Soudan alone
and on foot should you compel me to leave your country now."
"You are right," replied the ape-man. "You shall remain here six
months. If Ibn Jad has not returned in that time I shall leave word
that you be sent to my home. From there I can find a way to return
you in safety to your own country."
"May the blessings of Allah be upon thee!" cried Zeyd.
And when they came at last to the village Tarzan received the
promise of the chief to keep Zeyd until Ibn Jad returned.
After he had left the village again the ape-man headed north, for he
was concerned over the report that Zeyd had given him of the
presence of a European prisoner among the 'Aarab. That Stimbol,
whom he had sent eastward toward the coast, should be so far north
and west as Zeyd had reported appeared inconceivable, and so it
seemed more probable that the prisoner was young Blake, for whom
Tarzan had conceived a liking. Of course the prisoner might not be
either Stimbol or Blake, but who ever he was Tarzan could not
readily brook the idea of a white man being permitted to remain a
prisoner of the Beduins.
But Tarzan was in no hurry, for Zeyd had told him that the prisoner
was to be held for ransom. He would have a look about for Blake's
camp first and then follow up the spoor of the Arabs. His progress,
therefore, was leisurely. On the second day he met the apes of Toyat
and for two days he hunted with them, renewing his acquaintance
with Gayat and Zutho, listening to the gossip of the tribe, often
playing with the balus.
Leaving them, he loafed on through the jungle, stopping once for half
a day to bait Numa where he lay upon a fresh kill, until the earth
trembled to the thunderous roars of the maddened king of beasts as
the ape-man taunted and annoyed him.
Sloughed was the thin veneer of civilization that was Lord Greystoke;
back to the primitive, back to the savage beast the ape-man reverted
as naturally, as simply, as once changes from one suit to another. It
was only in his beloved jungle, surrounded by its savage denizens,
that Tarzan of the Apes was truly Tarzan, for always in the presence
of civilized men there was a certain restraint that was the outcome of
that inherent suspicion that creatures of the wild ever feel for man.
Tired of throwing ripe fruit at Numa, Tarzan swung away through the
middle terraces of the forest, lay up for the night far away and in the
morning, scenting Bara the deer, made a kill and fed. Lazy, he slept
again, until the breaking of twigs and the rustle of down tramped
grasses awoke him.
He sniffed the air with sensitive nostrils and listened with ears that
could hear an ant walk, and then he smiled. Tantor was coming.
For half a day he lolled on the huge back, listening to Manu the
Monkey chattering and scolding among the trees. Then he moved on
again.
A day or two later he came upon a large band of monkeys. They
seemed much excited and at sight of him they all commenced to
jabber and chatter.
"Greetings, Manu!" cried the ape-man. "I am Tarzan, Tarzan of the
Apes. What happens in the jungle?"
"Gomangani! Gomangani!" cried one.
"Strange Gomangani!" cried another.
"Gomangani with thunder sticks!" chattered a third.
"Where?" asked the ape-man.
"There! There!" they shouted in chorus, pointing toward the
northeast.
"Many sleeps away?" asked Tarzan.
"Close! Close!" the monkeys answered.
"There is one Tarmangani with them?"
"No, only Gomangani. With their thunder sticks they kill little Manu
and eat him. Bad Gomangani!"
"Tarzan will talk with them," said the ape-man.
"They will kill Tarzan with their thunder sticks and eat him,"
prophesied a gray-beard.
The ape-man laughed and swung off through the trees in the
direction Manu had indicated. He had not gone far when the scent
spoor of blacks came faintly to his nostrils and this spoor he followed
until presently he could hear their voices in the distance.
Silently, warily Tarzan came through the trees, noiseless as the
shadows that kept him company, until he stood upon a swaying limb
directly above a camp of negroes.
Instantly Tarzan recognized the safari of the young American, Blake,
and a second later he dropped to the ground before the astonished
eyes of the blacks. Some of them would have run, but others
recognized him.
"It is Big Bwana!" they cried. "It is Tarzan of the Apes!"
"Where is your head man?" demanded Tarzan.
A stalwart negro approached him. "I am head man," he said.
"Where is your master?"
"He is gone, many days," replied the black.
"Where?"
"We do not know. He hunted with a single askar. There was a great
storm. Neither of them ever returned. We searched the jungle for

Coreference in English and German A Theoretical Framework and Its Application in A Study of Court Decisions 1st Edition Sanna Engell

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Coreference in English and German A Theoretical Framework and Its Application in A Study of Court Decisions 1st Edition Sanna Engell

Uploaded by

Copyright:

Available Formats

Coreference in English and German A

Theoretical Framework and Its

Soviet Agriculture in Perspective A Study of its

Understanding the Investor A Maltese Study of Risk and

A Theoretical Framework for Language Education and

Science of the Soul in Ibn Sina's Pointers and

Vitamin D Basic Research and its Clinical Application

External Communication In Social Media During

Book of Seven Seals: The Peculiarity of Revelation, Its

Pacific Islanders under German Rule A Study in the

Coreference in English and German

Coreference in English and German

Bibliografische Information der Deutschen Nationalbibliothek

Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der

c Copyright Logos Verlag Berlin GmbH 2017

Logos Verlag Berlin GmbH

2 The concept of coreference ..................................................................7

3 Coreference chain elements in English and German:

3.4.1.2 Level 2: reduction, clarity of reference and

4 Related fields of research .................................................................117

5 The Appellate Committee of the House of Lords and the

6 Methodology of the study of coreference in the court

6.2.2 Types of coreference .........................................................161

7 Results and discussion ......................................................................177

8 Summary and conclusions ...............................................................255

sions, section 4.2 provides an overview of the research field ‘language

lengths of the coreference chains, the frequencies of the different types of

terance-bound concept: only linguistic expressions that are used in spe-

Six cooking apples

… six cooking apples … … them …

Figure 2.1: Endophoric and exophoric reference

With regard to exophoric reference, previous studies have discussed

such as unicorns, or to abstract concepts or events in the future (cf.

2.2 Types of reference

lished by noun phrases with a definite article, an indefinite article or ones

Having discussed three different types of exophoric reference, we

Exophoric reference Endophoric reference

Specific Non-specific Generic Anaphoric Cataphoric

Figure 2.2: Types of reference

2.3 A definition of coreference

2.4 Types of coreference

coreference (for brief remarks on concepts related to this type of corefer-

ple contains an instance of anaphoric reference, but anaphoric coreference

on endophoric reference since anaphoric and cataphoric coreference chain

relation of anaphoric or cataphoric coreference, but in a relation of non-

generic reference since they refer to the class of sedimentary rocks in

Anaphoric Cataphoric Non-directional

Figure 2.3: Types of coreference

In some cases, coreference is established by the use of proper names,

ence within a clause complex 5 and coreference across clause complex

2.5 Syntactic categories in nominal coreference chains

coreference’, i.e. coreference between two or more linguistic items of

The Return of Ulala

You might also like