Ain Shams University

Womens College for Arts

Science and Education

The Translation of Synonyms in the Holy Qur'an: A

Corpus-based Approach
A Thesis Submitted to:
The Department of English Language and Literature
Women's College for Arts, Science and Education
Ain Shams University in Fulfillment of the Requirements for the Degree of M.A

By
Summya Moammer Muhammed
Supervised by
Prof. Wafa Batran

Dr. NagwaYounis
Lecturer of Linguistics
English Language Department
Faculty of Education
Ain Shams University

Professor of Linguistics
Head of English Language
and Literature Department
Womens College for Arts,
Science and Education
Ain Shams University

2013

Chapter One: The Theoretical Framework of the Study

1 Introduction:
With increasing importance of corpus linguistics, as a branch of studying
language scientifically in a computer readable format, scholars begin to
use corpora in their researches and studies. Corpus (pl. corpora),
according to language science, is a body of written text or transcribed
speech serving as a basis for linguistic analysis and description. The term
corpus is the Latin word for body. Accordingly, a corpus is defined as
anybody of text (Kennedy, 1998, pp. 3-4). Among the primary values of
employing corpora are the opportunity to investigate large amounts of
data and conduct empirical research. Moreover, using many techniques
such as key words in context (KWIC), concordances, automatic
frequency counts of words and the like, is among the values of corpus
linguistics that facilitate the research process.
Corpus linguistics is a concept of debate by many linguists who define
it differently. In terms of what corpus linguistics is, not only have
various definitions been offered, but alternatives have been explicitly
addressed and rejected. These include that corpus linguistics is a tool, a
methodology, a method, a methodological approach, a discipline, a
theory,

theoretical

approach,

paradigm

(theoretical

or

methodological), or a combination of these.

Teubert (2005, p.4) states that corpus linguistics is not in itself a method:
many different methods are used in processing and analyzing corpus

-1-

data, it is rather an insistence on working only with real language data

taken from the discourse in a principled way and compiled into a corpus.
Kennedy (1998, p. 7) explains what corpus linguistics is. He points out
that corpus linguistics, in competition with other theories, would not be a
theory. Furthermore, it is not a branch of linguistics in the same sense
such as phonology, sociolinguistics or so on. Those branches describe/
explain some aspect of language use, while corpus linguistics is a
methodology rather than an aspect of language that requires explanation
or description. This methodology may be used in almost any area of
linguistics.
What characterizes corpus linguistics as an independent discipline within
general linguistics is not just its specific methodology or the particular
nature of its object of study, but its unique approach to the study of
language which is firmly based on the integration of four interdependent,
equally and important elements: data, description, theory, and
methodology. Their relationship can be expressed in terms of a continual
process involving corpus creation, discovery, hypothesis formation,
testing and evaluation. Corpus constitutes the raw linguistic data which is
the driving force of research. These data are assembled and recorded
according to explicit design criteria, based on a minimum set of principle.
They are then examined and processed- by the corpus linguist through
computerized tools and techniques (e.g. concordances, word frequency
lists, word frequency statistics, collocational profiles). The facts about
language discovered by this bottom-up approach are accumulated and
systematically organized in new descriptions of language behavior.
Corpus linguistics offers a set of methods that can be used in the
investigation of a large number of different research questions, as Leech
(in J. Svartvik, 1992, p. 107) defines it " a methodological basis for
-2-

pursing linguistic research". Also Leech (ibid, p.105) has noted about the
nature of corpus-based research that: "The focus of study is on
performance rather than competence and on observation of language in
use leading to theory rather than vice versa". Leech puts key features of
corpus linguistics:
-focus on linguistic performance rather than competence.
-focus on linguistic description rather than linguistic universal.
-focus on quantitative, as well as qualitative models of language.
-focus on more empiricist, rather than rationalist view of scientific
enquiry.
The major shift in methodology associated with corpus linguistics comes
not from theory, but rather from the use that corpora make possible.
For these advantages of corpus linguistics the researcher applies it in a
translation study. Recent years have especially witnessed a significant
growth of corpus-based translation studies that appeared in the beginning
of 1990s. Laviosa believed that "with the help of corpora, the translator
can easily find out the most suitable words and expressions" (Laviosa,
2002, P. 78). Thus, the current study aims to demonstrate the usefulness
of the corpus methodology in studying the translation of synonymous
words in the Holy Qur'an. This is because it is one of the problems that
face translators who are interested in Qur'anic translation. A translator
may find it difficult to make distinctions in his/her translation in general
among words that have nearly the same meaning. Baker (1992, P.22)
states this problem by mentioning that: "Target Language may make
more or fewer distinctions in meaning than the source language. What
one language regards as an important distinction in meaning another
language may not perceive as a relevant".
-3-

The problem in a translational context is how to become aware of these

subtle meanings if they are intuitively registered over time and used
unconsciously. Some synonymous words may be translated by some
translators as having the same meaning, while the context may be
understood to show differences, however slight they are.
It is widely held that meaning is not ascribable to words in isolation.
Rather, meaning is established on the basis of words and the context in
which they occur. This view has been developed within different
linguistic frameworks. One such framework is corpus linguistics. In this
framework, it is claimed that language exhibits systematic patterns of cooccurring words to an extent not yet fully recognized (Dam-Jensen. &
Koring, 2008, PP. 205-206).
Thus, through the advantages of a corpus-based approach there is the
possibility of making use not only of the material devices (the corpora
themselves, the list of co-occurrences, and the statistical measures), but
also of conceptual tools developed in corpus linguistics. The researcher is
going

to employ the concept of "semantic prosody", one of the key

notions of corpus studies, which goes back to John Sinclair's work on

collocations. Semantic prosody applies to relations between words that
involve evaluative meaning. On this assumption, certain words
prototypically combine with words with either a negative, positive or
otherwise value-laden semantic profile by which they acquire a
connotational colouring (Partington, 1998, P.68).
In terms of studying synonyms in the Holy Qur'an this study will only be
concerned with synonyms at the word level. Furthermore, this study is
restricted to the seven translations offered at the website.
-4-

This research is divided into four chapters. The first chapter tackles the
theoretical background of the study and how we use the selected corpus
in translation studies and in lexicographical research. The second chapter
gives a review of literature about the subject. Chapter three analyzes the
data in question. Chapter four sums up the study.
1.2 The Aims of the research:
The present study aims to investigate one of the problems that hinder a
translator in Qura'nic translation. It is the translation of synonymous
words that have nearly the same meaning. Moreover, the study aims to
exploit corpora in Qur'anic translation. The ultimate goal is to know
which strategy has been adopted by each translator. This study aims to
answer these questions:
1. How could the researcher exploit corpora in Qur'anic translation
analysis?
2. How could the researcher analyze the meaning of a word through
its collocations and semantic prosody?
3. Were they translators accurate in their translation of those
synonyms, i.e do the translators deffrentiate between synonymous
words?
4. What are the strategies that the translators followed or may follow
to overcome the problem of translating synonyms?
1.3 Importance of the Study:
This study provides objective and scientific analysis of the translation of
synonymy in the Holy Qur'an. Also this study examines a number of
translations on the websites in question to show the extent to which the
-5-

various translators were aware of synonyms as a linguistic phenomenon.

This may help professionals, translators and scholars who are interested
in Qur'anic translation.
Furthermore, this research contributes to Muslims task to convey the
message of the Qur'an to non-Arabic speakers as clear as possible.
1.4 Limitations of the Study
The researcher selects Qur'anic Corpus as a tool for analyzing a sample
of synonyms in the Holy Qur'an that are selected randomly, i.e. without
specific standards and conditions; this is done to achieve results with
more reliable judgments. This proves that the methods used could be
applied to any synonymous words without specific standard. In this study
we are concerned with synonyms at the word level. Furthermore, this
study is restricted to the seven translations offered at the website in
question. They are of these translators according to the arrangement of
the website: Sahih International 1997, Pikthall 1930, Yusuf Ali 1934,
Shakir 1999, Muhammed Sarwar 1981, Muhsin Khan 1996 and Arberry
1955.
Among the problems that encounter the researcher is the lack of
references that count all the synonyms in the Holy Qu'ran. For the
disagreement concerning the definition of synonymy there are no specific
standards for determining all the synonyms in the Holy Qu'ran. So one of
the first references that the researcher adopts in her study as a base is

(Synonymy in the Holy Qur'an

between Theory and Practice (Al-Monjed,1997). It contains three

hundred and fifty synonyms, but this study will select randomly from
them fifty words.
-6-

1.5 Synonyms
One of the main goals of this study is to check synonymy of a given pair
of items. This study uses the corpus-based analysis and the computer
technology that could be helpful to identify easily the relative frequency
of words throughout the whole corpus of the Holy Qur'an. Subsequently,
the researcher can explore the collocates of words and further isolate the
various meanings or senses a word has. Such research might show that
near synonymous words are used in different ways.
Synonymy is understood as a gradual cline along which they may be
located different degrees of synonymy: near, cognitive and absolute.
However, there is a widely held opinion among semanticists that strict or
absolute synonymy is rare in human languages (Cruse, 1986). This
research explores the collocation behavior and semantic prosody of near
synonyms from a cross-linguistic perspective. By near synonyms the
research means lexical pairs that have very similar cognitive or
denotational meanings, but which may differ in collocational or prosodic
behavior. Near synonyms can differ in collocation, they are not
collocationally interchangeable (Partington, 1998, p. 77). They also differ
in semantic prosodies: e.g. tickle is negative, whereas flexible is positive
(McEnery, & Xiao, 2006, p.108).
The study argues that the Holy Qur'an never has two words that mean
nearly the same thing, and are used in the same range of grammatical and
lexical patterns. In sum, the present research shows that supposedly
synonymous words are not equivalent in their meaning, when their actual
patterns of use are analyzed empirically. Rather, corpus-based analysis
can be used in general to show how each synonym has its own preferred
-7-

collocates, different +preferred senses, and different distributions across

registers, for example big-large-great. There are certain distinct patterns
discernable in the use of these seemingly synonymous words. Big is
most often used to refer to physical size, while large is used more to
refer to quantities and amounts. Great is also used for amounts (Biber,
2004, p. 50).
1.6 Methodology
This research follows certain steps with the aid of two websites to
analyze the data in question.
1.6.1 Methodological Steps
Corpus-based analysis of synonyms and their translations in the Holy
Qur'an is conducted by certain steps. The methodology that the current
research adopts is based on two steps. The first is concerned with dealing
with large quantities of data. At the beginning, there is the question of
how to deal with the large amount of data obtained. In respect of
choosing corpus for this study's purpose concerning genre, it is Qur'anic,
but there are no special needs regarding mode or time of the texts
included in the corpus, because the Holy Qur'an has no specific time. If
using a bottom-up approach on a very large electronic corpus, this means
that either a vast number of concordances have to be analyzed or some
way of pre-selection has to be established. The procedure proposed by
Charteris-Black (2004) consists of two steps: a qualitative study of the
search

item,

which

comprises

identification,

interpretation

and

explanation, followed by a quantitative concordance analysis ( Oster,

2010, 730).

-8-

The second is concerned with retrieving and analyzing the data. We start
from a lexical unit in question. This procedure yields data in the form of
co-occurrences. Since out of the range of possibilities offered by corpusanalysis software (including word lists, statistics of word frequencies,
concordances and keywords in context), the most useful tool for semantic
analysis seems to be that of extracting a list of co-occurrences for a given
search word. This is the (semi) automatic part of the analysis, although
this does not mean that there are no decisions to be taken in this phase of
the process when looking at the resulting combinations of the search
word and its co-occurrences. Then we employ the key notions of analysis
that we adopt in our research: semantic prosody and collocation.
Together with collocation profile (i.e., lexical realization) and semantic
prosody we bind words tightly into their contexts and into linguistic
convention, forming extended units of meaning (Sinclair, 1996) .
There are some features afforded by most corpora which are useful to
rationalize the search process, among them there are regarding the search
word (node), and searching by lemmas, which makes it possible to find
different forms of the lexical unit in one search. This might not be
relevant in our study, because the result of the research for synonyms will
automatically include the occurrences of other forms (ibid, 734).
Another feature is limiting the search by the part of speech (POS) of the
search word. Collocations may differ considerably between the noun and
the verb. This is especially important in contrastive studies, as in most
languages verb and noun do not coincide in their form.
Other features are regarding co-occurrences such as differentiating the
search with respect to the POS of the co-occurring word. This is useful to
-9-

facilitate the subsequent phase of classifying the co-occurrences (ibid,

735).
1.6.2 On-line Websites Used in this Research
This study use this methodology using two websites that include Qur'anic
corpus:
1.6.2.1 http://corpus.quran.com/

The Qur'anic Arabic Corpus (http//quran.corpus.com) is an on-line

annotated linguistic resource with multiple layers of annotation including
morphological segmentation, part-of-speech tagging, and syntactic
analysis using dependency grammar and semantic ontology. The
motivation behind this work is to produce a resource that enables further
analysis of the Qur'an. It consists of 77, 430 words that form a distinct
genre difficult to compare to other texts of Arabic. Processing Qu'ranic
Arabic is a unique challenge from a computational point of view, since
the vocabulary and spelling differ from Modern Standard Arabic. It is
fully diacriticied. The Qur'an contains 6,236 numbered verses (ayat) and
is divided into 114 chapters (Dukes, & Habash, 2010, pp. 30-36)).
The popular online site of the Qur'anic Arabic Corpus has been used by
numerous Qur'anic scholars, language researchers and students of Arabic,
many of whom have shown particular interest in the morphological
annotation.
The

Qur'anic

Arabic

Corpus

is

available

on

line

at

http://corpus.quran.com. The website includes a search feature which

allows keyword to be used in either Arabic or English. A word-by-word
contextual English translation is displayed along with a phonetic

- 10 -

transcription developed using an algorithm driven by diacritics and

annotations.
As it is shown on figure (1), it is the home page of the website. Through
the tools that are available on the left of the page, the researcher
identifies the search word to be analyzed through Qur'anic dictionary.
Through English translation tool which contains seven translations of the
Holy Qur'an that facilitate the comparison process among these
translations, the researcher investigates how translators deal with the
problem of translating synonyms.

Figure (1).

- 11 -

1.6.2.2 http://arabiccorpus.byu.edu/

This website provides a word in the Holy Qur'an with their

collocations and strings along the Holy Qur'an. It provides a word
with words before and after it as well as the number of occurrences of
the frequent words. The researcher uses this website less than the
previous one, because it provides the researcher just with collocates
and words before/after that we need in our analysis. It is as shown in

figure (2).
Figure (2).
Present methodology can be summarized by the following steps:
1-Selecting a sample of synonymous words that are to be searched
2-Selecting the Surah that contains the synonym (chapter) of the Holy
Qur'an.
2-Determaining the verse that is to be classified.
3-Subdividing the verse into features (keywords).

- 12 -

4- Finding the recurrences of the keywords (features) in other Surahs

(chapters).
5-For each verse extracted from the previous step the researcher studies
what is the subject this verse is talking about.
6-Writing out such observations to interpret the semantic prosody for
such words.
7-Investagating the translations of such words through the comparable
translations offered in the websites.
1.7 Theoretical Framework
1.7.1 Introduction
The study adopts a corpus-based approach of translation studies that
finds its roots in Corpus Linguistics. It is considered one of the branches
of linguistics that studies language on the basis of corpora, i.e., "bodies"
of texts assembled in a principled way (Baker, & Saldanha, 2009, p.50).
This methodology has tools that let users access and display the
information contained within a corpus in a variety of useful ways. For
example, it provides a word frequency list, which allows users to
discover how many different words are in a corpus and how often each
appears. Also, it offers lemmas of a word, i.e., all forms of the word
(Bowker, 2002, pp. 46-48). Alignment techniques on machine-readable
corpus are designed to provide explicit links between source sentences
and target sentences or source words and target words. Also,
concordances can show contexts which might help differentiate the usage
of alternate translations or near synonyms.
- 13 -

Firth declares that "You shall know a word by the company it keeps" (in
Palmer, 2004, p. 275). For example, different adjectives that are used to
describe a good-looking man and a looking woman are a beautiful
woman and a handsome man, but rarely of a beautiful man or a
handsome woman. This study applies semantic-prosody and collocation
as methods in analyzing the meaning of a word. Sometimes the
collocates of a word form a semantic class which can be characterized in
terms of attitudinal meaning.
1.7.2 Methods used in the Study
The meaning of lexical units must be determined with respect to two
linguistic dimensions, the paradigmatic and the syntagmatic.

The

paradigmatic dimension reflects how the senses of the words in a

language are related to each other, and crosslinguistically, to the senses
distinguished in other languages.

Monolingually, these relations are

typically described in terms of such relations as synonymy, antonymy,

hyponymy (Cruse, 1986, p.84).The syntagmatic dimension relates words
to the linguistic context, lexically, semantically, and grammatically.
Syntagmatic phenomena are typically described in terms of lexical cooccurrence (collocation), semantic preference (e.g. case roles, selection
restrictions, semantic prosody) and syntactic function (e.g. syntactic
dependency or valency) in attempt to establish languageindependent or
universal categories (e.g. frame semantics) against which the vocabulary
of different languages can be compared. They are closely related and
difficult to separate in practice. The meaning of a lexical item (its
paradigmatic status) can only be determined on the basis of the contexts
in which it occurs (its syntagmatic status) (Altenberg, & Granger, 2002,
p. 27).
- 14 -