An error analysis on the use ofcohesive devices in writing by freshmen majoring in English at thang long university

Vietnam national university, Hanoi
College of foreign languages
====***===
trần thị hải bình
An erro r anal ysi s on the use ofco hesi ve
devi ces
in writing by fre shm en majo ring in En gli sh
at thang long uni versi ty
phân tích lỗi trong cách sử dụng
phơng tiện liên kết văn bản trong bài viết
của sinh viên chuyên anh năm thứ nhất
tại trờng đại học thăng long
Course work
Field: Methodology
Supervisor: vũ thúy quỳnh, m.a
1
hanoi, December 2005
2
ACKNOWLEDGEMENTS
I would like to express my deepest thanks firstly to my supervisor, Mrs. Vũ Thuý Quỳnh,
M.A. who has enthusiastically helped and encouraged me to finish the research project.
Without her experienced guidance and valuable comments, my research would still be far
from finished. I am also indebted to her for her substantial contributions in proofreading
and help me make necessary changes.
My gratitude is also sent to all of my instructors in my M.A. courses at Post-Graduate
Studies, College of Foreign Languages, Vietnam National University, Hanoi. Their
precious and professional lectures and tutoring have helped me a great deal in
understanding profound concepts of the field in English teaching methodology while I
attended the courses.
Last but not least, I appreciate constant supports from my colleagues at Thang Long
University, my beloved family and my friends.

3
Table of contents
Acknowledgement
Table of contents
List of tables, charts and figures
Chapter One: Introduction
1.1. Reasons for choosing the topic
1.2. Objectives of the study
1.3. Scope of the study
1.4. Significance of the study
1.5. Methods of the study
1.6. Organization of the study
Chapter Two: Literature review
2.1. Factors affecting language learning
2.2. Errors analysis
2.3. The notion of errors in language learning
2.4. Errors vs. mistakes
2.5. Causes of errors in language learning
2.5.1. First language interference
2.5.2. Causes independent from first language
2.6. The concept of cohesion
2.7. Cohesive devices in writing
2.8. Types of cohesion
2.8.1. Grammatical cohesion
2.8.2. Lexical cohesion
2.9. Summary
Chapter Three: Research Methodology
3.1. Subjects
3.2. Instruments of data collection
3.3. Method of data analysis

i
ii
iv
1
1
2
2
2
3
3
5
5
6
6
7
8
8
9
10
11
11
12
17
18
19
19
20
20
4
3.4. Summary

Chapter Four: Presentation and Analysis of Data
4.1. Errors in the use of reference
4.1.1. Errors in the use of demonstrative reference
4.1.2. Errors in the use of personal reference
4.1.3. Errors in thes use of comparative reference
4.2. Errors in the use of conjunction
4.2.1. Errors in the use of adversative conjunction
4.2.2. Errors in the use of causal conjunction
4.2.3. Errors in the use of additive conjunction
4.3. Errors in the use of lexical cohesion
4.4. Summary
Chapter Five: Implications
Chapter Six: Conclusion
Bibliography
Appendices
21
22
23
23
29
30
32
33
34
36
36
38
39
44
5

LIST OF TABLES, CHARTS AND FIGURES
Table 2.1 Types of cohesion at linguistic level.
Table 2.2. Types of grammatical and lexical cohesion
Table 2.3. Personal reference
Table 2.4. Demonstrative reference.
Table 2.5. Comparative reference
Table 4.1. The number of errors in the use of cohesive devices
Table 4.2. Errors and their causes
Table 4.3. Errors in the use of demonstrative reference
Table 4.4. Errors in the use of the definite article.
Table 4.5. Errors in the omission of ‘the’
Chart 4.1. Sources of errors
Figure 2.1. Types of reference
Figure 2.2. The process of recognizing and identifying errors
6
CHAPTER ONE
INTRODUCTION
1.1. REASONS FOR CHOOSING THE TOPIC
Since Vietnam has opened its door to the rest of the world, more and more people with their
wish to join world trends have rushed to learn foreign languages, especially English. This
demand in language learning has brought about a great amount of positive changes in language
teaching in Vietnam. Language teachers have looked for and tried different methods and
techniques in teaching and learning in order to find the effective ones. The effectiveness of a
teaching method or technique is reflected in the learners’ language competence that is their
abilities to perform the four language skills: Reading, Listening, Writing and Speaking.
Amongst these skills the two productive skills, writing and speaking, are considered more
difficult than the others as the learners need to use the language to convey their messages
comprehensibly and accurately in real life communication. When a message is unsuccessfully
conveyed, the factor, which is most likely to be blamed for is errors in the use of the language.
It is natural in language teaching that learners make mistakes and errors when writing in

English. How to cope with and when to give feedback to these errors are vital in teaching
language as it may either result in motivation or discouragement in language learning. Some
teachers’ concern is directed to contrastive analyses of Vietnamese and English with the hope
to predict and prevent errors before they appear. This theory has been supported by Lado
(1957). However, Richards (1971) in his research found out that apart from the first language
interference, there were other causes which are products of intra-lingual analogies such as
overgeneralization, incomplete application of rules or ignorance of rule restrictions as well as
false concepts hypothesized. So far, there has been scarily any research on the causes of errors
in students’ writing in English in Vietnamese universities. Therefore, I am attempted to carry
out a research study applying error analysis in clarifying learners’ errors in the use of cohesive
devices in writing at Thang Long University as “cohesive devices are crucial in writing”
( Zamel,1983:1).
1.2. OBJECTIVES OF THE STUDY
7
The purpose of this study is to obtain quantitative data for the investigation the types and
causes of errors in the use of cohesive devices in writing by first year undergraduate students
majoring in English at Thang Long University. Thus, it aims to seek answers to the following
questions:
1. What are common errors in the use of cohesive devices in the students’ writing? and
2. What are the major causes of these errors?
The answers to these questions will serve as help in giving recommendations to reducing and
preventing the problems of coherence in students’ writing.
1.3. SCOPE OF THE STUDY
It is believed that different types of learners committed different types of errors. Also, the type
of errors are various according to different stages in learning process. Errors are made in both
of the productive skills: writing and speaking. Due to the limitation of time, the study is
confined itself to errors in the use of cohesive devices in the writing by Vietnamese pre-
intermediate students.
1.4. SIGNIFICANCE OF THE STUDY
Many studies on errors have been carried out in the field of teaching English in the world.

Researchers like Zamel (1983), Richard (1971) and Corder (1967) among others emphasized
the importance of errors in theory as well as in practice of foreign language learning and
teaching. According to Corder (1967), errors are traced to their sources are beneficial in
different ways. Firstly, they help language teachers know how much progress a learner has
made in the target language, in which language area he needs help and what sort of help he
needs. Secondly, they provide researchers with evidence in language learning process;
therefore, researchers through errors discover strategies applied in acquiring a language. Apart
from that, errors can serve as good feedback to learners for self-adjustment. Despite these
benefits, few studies on errors derived from Vietnamese learners have been made. For these
reasons, this study should be conducted to find out types of errors, specifically errors in the use
of cohesive devices in Vietnamese learners’ writing and what their causes are.
8
It is hoped that the findings of the research would be useful to teachers as well as learners of
English. Once the type and the causes of a particular error are properly found, teachers will
have a better understanding of students’ problem in using cohesive devices in writing and can
develop proper solutions.
1.5. METHODS OF THE STUDY
The subjects of the study are two classes of first year undergraduate students at Thang Long
University. They are at the age ranged from 18 to 19. Their major at the university is English.
Though they come from different areas in the country, they are considered at the same level of
English as they all learned English at high school and have passed the university entrance
examination on three subjects including English.
This is intended to be a quantitative research study using compositions as a technique of
eliciting data for the analysis, statistical counting as measurement of results. Students’ papers
were collected every week. Any errors in the use of cohesive devices were found and
classified according to the cohesion-category by Haliday and Hasan (1976). Then the
occurrence frequency of each error type was counted. The data and the list of the errors was
the source for the analysis.
1.6. ORGANIZATION OF THE STUDY
The study is composed of six chapters.

Chapter one gives reasons for choosing the topic, objectives and methods of the study. It also
narrows the scope of the study and briefly presents an overall out-line of the research study.
Chapter two reviews the literature related to the study. Firstly, it presents the factors affecting
language learning which is divided into two types: external and internal. Secondly, error
analysis and errors in language learning are discussed. The literature related to errors is given;
it includes the notion of errors in language leaning, the distinction between errors and
mistakes, main causes of errors. Lastly, cohesion in writing is mentioned, it consists of the
concept of cohesion, cohesive devices and types of cohesion.
9
Chapter three describes in detail the research methodology which comprises the information of
the subjects, instruments of data collection and methods of data analysis.
Chapter four presents the statistical results and the analysis of the data. The statistical results
are shown in the tables which are the basement to determine the causes of each type of errors
in the use of cohesive devices in writing.
Chapter five named Implication with the recommendations for correcting errors in the use of
cohesive devices in writing, suggestions for teaching in order to prevent and eliminate these
errors.
Chapter six closes the study with a conclusion which gives a summary of the whole study and
provides suggestions for further studies.
10
CHAPTER TWO
LITERATURE REVIEW
In this chapter, the literature related to errors and cohesive devices in writing is discussed in
order to provide the study with the sufficient theory background.
2.1. FACTORS AFFECTING LANGUAGE LEARNING
On the basis of language learning process theories, it is clear that language learning bear a lot
of influences and the factors affecting language learning are categorized into two types:
external factors and internal factors.
External factors include: the first language, language environment and the formal teaching.
First of all, the mother tongue may be more or less helpful for the learners of a new language

as they have already learned how to do with that language. Universal features in languages can
assist learners to learn a new language. On the basis of behavior psychology, Lado (1957) and
Fries (1965) suggested two types of transfers: transfers are positive when the first language
and the target language share similar features, negative when there are different features in the
two languages. Language environment is also of great significance to success in learning a
foreign language. Lastly, the role of formal teaching was asserted in the research by
Littlewood (1980) when he proposed that certain techniques or methods proved to be relevant
in particular groups of learners.
Dulay et al. (1982) named two internal factors: the filter and the monitor. According to
Dulay, language learners do not acquire what is exposed to them, but select what they find
suitable, relevant and interesting. Motivation, as he defined, is understood as “incentive, the
need or the desire to learn the second language” (Dulay et al., 1982:47), if motivation is low,
failure is likely reported. “The monitor is the part of the learners’ internal system that appears
to be responsible for conscious linguistic processing” (Dulay et al., 1982:58). It appears when
learners try to learn or to apply a linguistic rule or structure or when he is given tasks requiring
grammatical judgments.
Foreign language learning are influenced both outside from teaching and learning
environment, and inside from what and how learners process the language.
11
2.2. ERROR ANALYSIS
In 1970s and 80s, a large number of papers on error analysis were published. Subsequently, a
more positive attitude towards errors has emerged. In the past, errors were deemed and errors
now are viewed as natural and important part of learning process because they can yield
information about learning language. This positive attitude towards errors is especially
important in the wake of the Communicative Language Learning and Teaching. Many
researches on errors in second language learning have been done by several scholars like
Corder (1967), Richard (1992) and Selinker (1992). Error Analysis is the identification,
description and explanation of errors either in its spoken or written form. Following Corder
(1967), Choon (1993) gives some suggestions on carrying out an error analysis research.
According to her, one has to identify the errors first, then the errors are classified according to

categories such as: semantic errors (wrong words, wrong forms, etc.), grammatical errors
(tense, preposition, etc.), global errors and local errors. She suggested that “the system of
classifying errors should be flexible” (Choon, 1993: 2). The last step is determining how much
they deviate from the target language norm, to what extent they affect communication. Error
Analysis can help language teachers manner the specific and common language problems
students have so that he or she can know what should be focused more in a syllabus. Choon
(1993) advised teachers to conduct Error Analysis at the beginning of the course when the
items have not been fully learnt and remedy these first.
By classifying errors that learners made, researchers could learn a great deal about the second
language acquisition process by inferring the strategies that the learners were adopting. For
learners themselves, errors are ‘indispensable’ since the making of errors can be regarded as a
device the learner uses in order to learn (Selinker, 1992:150).
2.3. THE NOTION OF ERRORS IN FOREIGN LANGUAGE LEARNING
On the basis of theory, Behaviorists view errors as symptom of ineffective teaching or as
evidence of failure. They also view it as being due to largely to the first language interference.
When errors occur they are remedied by a bombardment of correct forms which can be
achieved by the intensive drilling or over teaching. On the other hand, Mentalists , who
following cognitive principles, suggested that learners process the new data in his mind and
come up with a set of rules that produce new patterns in the target language. Consequently,
12
errors are inevitable; in fact, they even become a part in learning process and developing
competence. Errors are not regarded as a sign of failure, but evidence that the learner is
working toward the correct rules. The attitude of Mentalists is positive toward errors in
language learning, it removes the anxiety caused by the behaviorist in classroom.
Richards et al. (1974) believed that both children learning the first language, and children and
adults learning foreign languages likely to produce errors of following types:
i) The omission of grammatical morphemes
ii) The double marking of a given semantic feature
iii) The over generalized application of irregular rules
iv) The use of one form for several required

v) The wrong word ordering
2.4. ERRORS VS. MISTAKES
The distinction between “errors” and “mistakes” has been given by many linguists though it is
impossible to indicate any sharp differentiation. According to Klassen (1991), the term “error”
is used to refer to a form of structure that a native speaker deems unacceptable because of the
lack of language competence. Chomsky (1965) initiated the distinction when he suggested that
there were two types of errors: one resulting from verbal performance factors, the other from
inadequate language competence. Later, Corder (1967) named the former mistakes and the
later error. Mistakes are said to be unsystematic in nature and correctable when attention is
drawn to its producers. Errors, on the other hand, refer to any systematic deviations from the
rules of the target language system. In short, errors are caused by lack of knowledge about the
target language or by incorrect hypothesis about it; mistakes are caused by temporary lapses of
memory, confusion, and carelessness and so on. If we are uncertain whether one of the
learners has made an error or a mistake, the crucial test must be: can he correct himself when
challenged? if he can, probably it is a mistake; if not, it is an error.
2.5. Causes of errors in foreign language learning
13
There are a number of reasons for how learners make errors; they take root from both social
factors and cognitive factors (Myles, 2002). Basically, two types of causes are classified: (1)
first language interference and (2) causes independent of the first language interference.
2.5.1. First language interference
The notion of first language interference is understood as negative transfer from the first
language to the target language, it is the way of learning new habits is hindered by previously
learnt ones. Lado (1957) claims that “errors are originated in the learners’ disposition to
transfer forms and meanings, and the distribution of forms and meanings of their native
language and culture to the foreign language and culture” (1957:1). Myles (2002) considers
transfer an important cognitive factor related to writing errors .The study of transfer involves
the study of errors (negative transfer), facilitation (positive transfer), avoidance of the target
language forms, and their over-use (Ellis, 1994). Corder (1967) observed language learners
make hypotheses about the language they are learning, tried to compare it with their native

language, then came to the conclusion that errors in foreign language reflected the first
language’s features. Later in 1978 he recasts interference as learners’ reliance on the first
language as their strategy of communication, which means learners use literal translation as a
learning strategy to overcome their ignorance.
Why do language learners apply their native language in second language acquisition? The
answer lies in four major factors. Firstly, it is the performance pressure. In class room setting,
the learners may be forced to perform tasks they do not want or their linguistic competence fail
to meet; therefore, they fall back on the language most familiar to them that is their mother
tongue. Windowson (1990) realized that when learners write under pressure, they may rely on
systematic resources from their native language for the achievement and synthesis of meaning.
Secondly, the limited foreign language environment also contributes to errors in language
learning. The lack of natural linguistic inputs with native speakers results in learners’ recourse
on their language. Moreover, language tasks assigned for the learners have a significance
affect on their verbal production. Among these tasks, translation is said to “increase the
foreign language learners’ reliance on first language structures” (Dulay et al., 1982:110).
Lastly, Dulay et al. (1982) considered the monitor as “an important factor associated with the
learner’s use of foreign language acquisition” (1982:110). Learners tend to think in the first
language and attempt to put the idea in the target language. Thus, the first language
14
interference takes place because of four factors: performance pressure, limited language
environment, manner of eliciting verbal performance and the monitor use.
Myles (2002) defined the above four factors as social factors affecting writing in foreign
language. These factors are closely related to learners’ attitudes, motivations and goals.
“Research based on direct and indirect measures generally shows that learners with positive
attitudes, motivation, concrete goals will have attitude reinforced if they experience success.
Likewise, learners’ negative attitudes may be strengthened by lack of success or by failure”
(2002: 2). He concluded that learners’ attitudes, motivations and goals can explain why some
foreign language writers perform better than others.
French (1958) when looking for common errors in English wrote:
The fact that the errors are common indicates that they have a common cause. That

common root is not to be found in a wide variety of languages exhibiting innumerable
differences in syntax, accidence and idiom. Explanation does not lies in cross-
association and instinctive translation of the mother tongue, but in the usages of
English itself; for these usages provide the only factor which is common to all regions,
all students and all methods (1958: 7)
2.5.2. Causes independent from the first language
Causes independent of the first language include: overgeneralization, false concepts
hypothesized, incomplete application of rules, cross association, and fossilization.
Overgeneralization: According to Jakobovist (1969), overgeneralization is the application of
previous available strategies in new situations. Richard (1974), Jain (1969) and Littlewood
(1980) defined the term “intra-lingual interference”. Littlewood (1980) suggests that
overgeneralization and transfer have the same strategy; the difference is the employment of
knowledge of the foreign language in the former and of the first language in the latter.
False conceptualization: Learners’ faulty understanding of distinctions of target language
items leads to false conceptualization, Richard (1971) blames poor presentation or presentation
based on contrastive approach for the confusion such as the use of verbs “come / go”, “was /
is”, of past and present markers.
15
Incomplete application of rules: Richard (1971) noted down two factors leading to
incomplete application of rules as the use of question in classroom as elicitation techniques
and learners’ interest in communication which helps them to achieve efficient communication
without a mastery of the target language rules.
Cross association: The notion of cross association is proposed by George (1972). It is
different from overgeneralization in the way that interference does not come from the prior
learning items, but from the adverse direction. George (1972) wrote “cross-association is the
phenomenon of mutual interference between partially learned items, neither being inhibited
but one or both being affected by the other” (1972:153).
Fossilization: “Fossilization is referred to as a phenomenon that takes place as a learner
internalizes an incorrect form” (Brown et al., 1987: 186). This is believed to exist in
adolescents and adults’ pronunciation, and also manifests in some syntactic structures or

vocabulary a learner uses. Three factors contribute to this phenomenon: mother tongue
influence, communication needs, and teachers’ feedback.
Sometimes it is difficult to decide exactly which process is applied in a certain error.
Littlewood (1980:29) concluded that many processes might operate simultaneously and
reinforce each other in causing the learners to produce errors.
2.6. THE CONCEPT OF COHESION
Texts, sequences of sentences or utterances which seem to hang together, contain what are
called text-forming devices. These devices are words or phrases which enable speakers or
writers to establish relationships across sentence or utterance boundaries, and help to tie
sentences in a text together. According to Yule (1996), a text is usually considered to have a
certain structure which depends on factors quite different from those required in the structure
of a single sentence; some factors are described in terms of cohesion, or the ties and
connections which exist within a text. Renkema (1993) considers cohesion as the connection
which results when the interpretation of a textual element dependant on another element in the
text. In short, that texts cohere or stick together, “have texture and this is what distinguishes it
from something that is not a text” due to the help of cohesive devices. Schiffrin (1978) defines
cohesive devices as “clues used by speakers and hearers to find the meanings which underlie
16
surface utterance” (1978:9). Halliday and Hasan (1976) regard a text ‘as a semantic unit: a unit
not of form but of meaning’ (1976: 2). They describe cohesion as a semantic concept that
refers to relation of meaning existing within a text, not as a structural unit. Therefore, their use
of the term cohesion refers specifically to non-structure text forming relations and it often
occurs where the interpretation of some element in the discourse is dependent on that of
another. Their focus is on the cohesive ties between sentences because they are the only source
of textual, while within the sentence there are structural relations as well.
2.7. COHESIVE DEVICES IN WRITING
In writing, cohesive devices are crucial for they turn separate clauses, sentences, and
paragraphs into connected prose, signaling the relationships between ideas, and making
obvious and visible the writer’s “line of thought” (Boadhead and Berlin, 1981:306).
Researchers have pointed out that these ties are an important property of writing quality (Witte

and Faigleiy, 1981). In fact, these ties may be essential for preserving author’s meaning
(Raimes,1979). While native speakers of English generally learn to use these cohesive
elements as they do other aspects of language, English language learners seem to have great
difficulties in mastering them. Bacha and Hanania (1980) found that learners have many
problems with cohesive devices in writing.
2.8. TYPES OF COHESION
Halliday and Hasan (1976) give the most comprehensive description analysis of cohesive
devices five major types of cohesive ties: reference, substitution, ellipsis and conjunction and
lexical ties. The first four types are grouped as grammatical cohesion and the later is lexical
cohesion.
17
Table 2.1: Types of cohesion at linguistic level
Linguistic level at which “phoric” relation is
established
Type of cohesion
Semantic
Grammatical
Lexicogrammatical
Lexical
Reference
Substitution and Ellipsis
Lexical cohesion
( Halliday and Hasan, 1976: 318)
Conjunction is believed on the borderline of the two. However, it is better to put it in the group
of grammatical cohesion as it is mainly grammatical with a lexical component inside.
Types of cohesion in each group are given out in details as follows:
Table 2.2: Types of Grammatical and Lexical Cohesion
GRAMMATICAL COHESION LEXICAL COHESION
Reference
• Exphoric Reference

• Endophoric Reference
o Personal
o Demostrative
o Comparative
Substitution
• Nominal Substitution
• Verbal Substitution
• Clausal Substitution
Ellipsis
• Nominal Ellipsis
• Verbal Ellipsis
• Clausal Ellipsis
Conjunction
• Additive
• Adversative
• Causal
• Temporal
• Others
Reiteration
• Same word/Repetition
• Synonymy/ Near-synonym
• Superordinate
• General words
Collocation
(Adapted from Haliday and Hasan, 1976)
2.8.1. Grammatical cohesion
Referential cohesion
According to Haliday and Hasan (1976), there is referential cohesion in every language, they
are “certain items which have the property of reference (…), instead of being interpreted
semantically in their own right, they make reference to something else for their

18
interpretations” (1976: 31). In English these items are personals, demonstratives and
comparatives.
By contrasting Exophora, or Exophoric reference with Endophoric as a general name for
reference within the text, Haliday and Hasan make the distinction between situational and
textual reference clear.
Reference:
[ situational] [textual]
exophora endophora
[to preceding text] [to following text]
anaphora cataphora
Figure 2. 1: Types of reference
(Source: Haliday and Hasan, 1976: 33)
Exophora is situational reference refering to a thing as identified in the context of situation and
Endophora is textual reference refering to a thing as indentified in the surrounding text. As
general rule, reference items may be exophoric or endophoric, if endophoric, they may be
anaphoric or cataphoric.
Anaphoric and cataphoric reference indicate two different ways in which reference items can
function within a text. Anaphoric reference points the reader or listener ‘backwards’ to a
previously mentioned entity, process or state of affairs. In the following example, the
underlined words are anaphoric reference.
Example:
- The schoolmaster was leaving the village, and everybody seemed sorry. The miller
lended him the small tilted cart and horse to carry his goods.
Cataphoric reference points the reader or listeners forward _ it draws us further into the text in
order to identify the elements to which the reference items refer. They in the example is
cataphoric reference as readers have to read on, and are given their identities in the second
sentence.
Example:
19

- They pressed round him in ragged fashion to take their money. Andy, Dave, Phil,
Stephen, Bob.
Haliday and Hasan (1976) classify three sub-types of referential cohesion: personal,
demonstrative and comparative. These various devices enable the writer or speaker to make
multiple references to people and things within a text.
Personal references are reference by means of function in the speech situation, through
catergory of person. These items are expressed through pronouns and determiners. They serve
to identify individuals and objects that are named at some other points in the text.
Table 2. 3 : Personal reference
Semantic catergory Existential Possessive
Grammatical function Head Modifier
Class noun (pronoun) Determiner
I me
you
we us
he him
she her
they them
it
one
mine
yours
ours
his
hers
theirs
[its]
my
your
our

his
her
their
its
one’s
(Source: Haliday and Hasan, 1976: 38)
Demonstrative references are references by means of location, on a scale of proximity,
expressed through determiners and adverbs. These items can represent a single word or
phrase, or much longer chunks of text _ ranging across several paragraphs or even several
pages.
Table 2. 4: Demonstrative reference
Semantic catergory Selective Non-selective
Grammatical function Modifier/Head Adjunct Modifier
Class Determiner Adverb determiner
this these
that those
Here [now]
There then
The
(Source: Haliday and Hasan, 1976: 38)
Comparative references are indirect references by means of identity or similarity, expressed
through adjectives and adverbs and serve to compare items within a text.
Table 2. 5 : Comparative reference
20
Grammatical function Modifier:
Deictic/Epithet
Submodifier/Adjunct
Class Adjective Adverb
same identical equal
similar additional

other different else
identically
similarly likewise
so such
differently otherwise
better, more etc
[comparative adjectives
and quantifiers]
so more less equally
(Source: Haliday and Hasan, 1976: 39)
Substitution
Substitution is the replacement of one item by another, as indicated in the Figure 1, the
distinction between substitution and reference is that substitution is the relation in the wording
rather than in the meaning.
There are three types of substitution_ nominal, verbal and clausal. They are the words, which
can only be interpreted in relation to what has gone before. Haliday and Hasan (1976) give out
the following list of the items that occur as substitutes:
Nominal: one, ones; same
Verbal: do
Clausal: so, not
The following underlined words are examples of substitution:
Examples:
- There are some new books on the shelf. These ones have been given by my uncle.
→ Nominal substitution
- So do you.→ Verbal substitution
- I think so. → Clausal substitution
Ellipsis
Ellipsis occurs when some essential structural element is omitted from a sentence or clause
and can only be recovered by referring to an element in the preceding text. “Ellipsis can be
21

interpreted as that form of substitution in which the item is replaced by nothing” (Haliday and
Hasan, 1976: 88). Consider the following discourse fragment illustrates for the point:
Example:
- Mary: I prefer the green
It is impossible to tell from the utterance what Mary prefers: the green dress, hat, or bag if
what is said before is not known (for example: Sylvia: I like the blue hat). Therefore, the green
is a elliptical nominal group
As with substitution, there are three types of ellipsis: nominal, verbal and clausal. In the
following examples, the ellipsis, which have been left out and marked by (0).
Example:
- My kids are very naughty. Both (0) are too small. → Nominal ellipsis
-A: Have you been working?
B: Yes, I have (0) → Verbal ellipsis
- A: Tom is staying for dinner!
B: Is he? He didn’t tell me (0). → Clausal ellipsis
Conjunction
Conjunction differs from reference, substitution and ellipsis in that it is not a device for
reminding the reader of previously mentioned entities, actions and states of affairs. In other
words, it is not an anaphoric relation. It is a cohesive device because it signals relationships
that can only be understood through reference to other parts of the text. Reference, substitution
and ellipsis are clearly grammatical as they involve closed systems presenting simple options
of presence or absence, and systems such as those of person, number, and proximity and
degree of comparison.
The cohesion of conjunction can be interpreted in terms of either experiential function of
language that is the relation between the meanings in the sense of representations of content,
our experience of external reality or the interpersonal function of language which is known as
the relation between meanings in the sense of representations of speaker’s own idea about the
situation. Haliday and Hasan (1976) named these relations: external and internal respectably.
22
The two types of conjunctive relation can be exploited whenever conjunction is used as a mean

of creating text as the line between the two is not always a clear cut.
Basically, there are four main types of conjunction: temporality, causality, addition and
adversity.
Adversative conjunctions such as however, on the other hand, etc. are adversative because
the information in the following sentence of a text moderates or qualifies the information in
the preceding.
Additive conjunctions signal the presentation of addition information such as and, moreover,
in addition to, etc.
Temporal conjunctions such as first, then, after that, etc. express the relationships which exit
when the events in a text are related in terms of the timing of their occurrence
Causal conjunctions interpret the relationship between the cause and consequence such as
because, because of, for, etc.
2.8.2. Lexical cohesion
Lexical cohesion occurs when two words in a text are semantically related in some way, in
other words they are related in terms of their meaning. There are two major categories of
lexical cohesion: reiteration and collocation.
Reiteration includes repetition, synonym, super-ordinate, and general words. The role of
reiterations in the text is referring back to the previously mentioned entity; thus, they fulfill a
similar semantic function to cohesive reference.
Collocation can cause major problems for discourse analysis because it includes all those items
in a text that are semantically related in some cases. This type of lexical cohesion, according to
Haliday and Hasan (1976), ‘is achieved through the association of lexical items that regularly
co-occur’ (1976: 284). It can be implied that collocation is expressed through open class items.
Any pair of lexical items which are in some way associated with each other in the language is
capable to bring about cohesion. The cohesive effect of these pair does not depend much on
any systematic semantic relationship as on the tendency to share the same lexical environment.
23
This effect even builds long cohesive chains across sentence boundaries. Therefore, there is no
limit to these items; this means it is difficult to establish sets of regularly co-occurring words
and phrases.

2.9. SUMMARY
This chapter has presented briefly the literature, which is relevant for the study. According to
Lado (1957), Fries (1965), Littlewood (1980) and Dulay et al. (1982), foreign language
learning are influenced both outside from learning and teaching environment, and inside from
what and how learners process the language. Affected by these factors, learners’ errors are
inevitable. It is error analysis which helps to turn these errors to the benefit of learning and
teaching foreign language. This has been proved by many studies on errors by Coder (1967),
Richard (1992), Selinker (1992) and Choon (2002). When analyzing errors, it is necessary to
distinguish mistakes and errors. The former are caused by the lack of knowledge about the
target language and the latter by temporally lapses of memory, confusion or carelessness.
Causes of errors are also presented in two main categories: First language interference and
causes independent from the first language. The focus of the study is on errors in the use of
cohesive devices so the concept of cohesion, the importance of cohesion in writing and the
comprehensive description analysis of cohesive devices by Haliday and Hasan (1976) are
included in this chapter.
24
CHAPTER THREE
RESEARCH METHODOLOGY
This chapter describes the subjects, instruments of data collection and methods of data analysis
of the study in detail.
3.1. SUBJECTS
The subjects of the study involved in this study were two classes of 42 first year undergraduate
students majoring in English at Thang Long University. There were 85 first year
undergraduate students majoring in English at the University. These 85 students were
randomly divided into four classes at the beginning of the academic year. In terms of the
length of learning time, they all had spent two and a half months taking part in courses for
students majoring in English at the university apart from three years of learning English at
high school. Thus, they were likely at the same level in English as they had passed the
entrance examination to the university and this examination was on three subjects including
English. Their English level was equal to the pre-intermediate. Also, the first year students

were chosen as subjects of the study because they had not learnt cohesive devices
systematically until last semester of the first academic year. Though they had already gained
some knowledge of English during school time, they did not have full understanding and be
able to use cohesive devices correctly in writing as in high school the focus of learning English
is more on grammar than on writing. In the situation that two teachers were responsible for
writing lessons in these fours classes and each of them taught had lessons in two classes every
week, two classes which were taught by one teacher were chosen; this factor helped to make
sure that the teaching and leaning condition was likely similar in two classes. The result of the
research is hoped to help teachers have focuses in the following semester when cohesive
devices are taught in grammar lesson. In general, the subjects were homogeneous in their age
as they all were first year undergraduate students with their age ranged from 18 to 19. At
present, they all share the same learning conditions and have learnt in the same cultural
background.
25