See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/263106706

Reference in English–Arabic translation

Article in Babel Revue internationale de la traduction / International Journal of Translation · December 2007
DOI: 10.1075/babel.53.4.05ehr

CITATIONS READS

2 2,023

1 author:

Martin Ehrensvärd
University of Copenhagen
8 PUBLICATIONS 102 CITATIONS

SEE PROFILE

All content following this page was uploaded by Martin Ehrensvärd on 02 June 2021.

The user has requested enhancement of the downloaded file.

 Reference in English–Arabic translation

Martin Ehrensvärd
University of Aarhus

1. Introduction

The subject of this article is the explicitness of reference in English–Arabic trans-

lation. The research published here is inspired by an article from 1993 by Yowell
Y. Aziz, ‘Explicit and Implicit Reference in Arabic–English Translation’, Babel 39,
129–150. Aziz shows that remarkable differences obtain in the explicitness of vari-
ous referring expressions when an Arabic text is compared to its English transla-
tion and that overall, the English text is the more explicit. Aziz does not, howev-
er, compare any English texts to their Arabic translations. In what follows, this is
what I will do.

2. Terminology and methodology

In order to facilitate a comparison with Aziz’ results, I have adopted his termin-
ology and methodology. However, I have simplified it somewhat since a less fine-
grained methodology is sufficient for my purpose here.
The term reference denotes the relationship between an object in the outside
world (real or fictional) and the utterance of it, as, e.g. , my use of ‘this article’
above: by use of this expression I refer to the actual article that exists in the out-
side world. There are three kinds of referring expressions: Proper nouns, noun
phrases (henceforth: NPs), and pronouns. These can be divided into subtypes.
Proper nouns are either names or titles. NPs can be definite or indefinite, and pro-
nouns are mostly personal pronouns but can also be e.g. demonstratives (give me
that). Definite NPs can be further divided according to the means by which they
are rendered definite: by possessives (his car), by demonstratives (this car), or by
the definite article (the car).
Reference is explicit if an expression fulfils two conditions: (1) it is detailed
and definite and hence serves to identify the referent unambiguously and (2) it

Babel 53: 1 (2007), –. issn 0521–9744 / e-issn 1569–9668

© Fédération des Traducteurs (fit) Revue Babel
 Martin Ehrensvärd

r­ efers directly and not through another expression, e.g. a pronoun. The explicit-
ness of a referring expression is determined by the extent to which it fulfils these
conditions. The most explicit referring expressions are names because they refer
uniquely and directly to the referent. They are followed by titles which refer slight-
ly more obliquely to their referent.
NPs refer less explicitly than proper nouns since they do not refer uniquely to
their referent but depend to a larger extent on the context of utterance. They do,
however, refer fairly directly. Definite NPs are more explicit than indefinite ones
since they contain much more information by which to identify the referent.
Within the subtypes of definite expressions there are also differences in explic-
itness: Both possessive NPs (Peter’s car/his car) and demonstrative NPs (this car)
are more explicit than NPs with definite article (the car) since the possessive and
demonstrative pronouns contain more information than the definite article: The
definite article only shows that the referent is to be found in the context of utter-
ance and does not in itself provide further information, but the possessive con-
tains information as to the relation of the NP, and the demonstrative contains spa-
tial information. Another subtype of definite expressions involves elliptical head
NPs (Peter’s car is bigger than mine). These are less explicit than full head NPs (Pe-
ter’s car is bigger than my car).
Pronouns have only implicit reference since they, being proforms, do not fulfil
the second condition.
The corpus investigated comprises a novel of 174 pages, Sula by Toni Morrison
from 1973 (I have used the edition published in London by Pan Books in 1991),
and its translation into modern standard Arabic (Sula, translated by Amal Mansur,
Amman: Al-Ahlīya li–l-Našr wa-l-Tawzī, 1995, 141 pages). The translation was
published after Morrison’s 1993 win of the Nobel Prize and is hence expected to
be a good one. It has the further advantage of being quite literal.

3. The texts

I will present the results of the comparison of the referring expressions in the Eng-
lish and Arabic Sula-text according to the different patterns found under the two
main headings: NPs and pronouns. Note that any pattern which occurs less than
five times in the texts is left out, and note that only cases where a difference in ex-
plicitness is involved will be noted. The English original will serve as basis for the
comparison, and each pattern will be illustrated by an example.
 Reference in English–Arabic translation 

4. NPs

NPs are of two types: indefinite and definite.

4.1 Indefinite NPs

In the corpus, there are examples of two types of relevant Arabic correspondences
to English indefinite NPs.
(1) Indefinite NP: Definite article NP
(2) Indefinite NP: Possessive pronoun NP

4.1.1 Indefinite NP: Definite article NP

(1) Men in khaki work clothes will pry loose the slats of Reba’s Grill (p. 3)
wa-sa-yantaziu riğālun bi-malābisa l-amali l-khākīyati alwāḥa maṭami
rībā
(And men in the khaki work clothes will tear away the slats of Reba’s Res-
taurant, p. 9)
This pattern is by far the most common in the corpus with 847 occurrences (cf. 48
examples of English definite article NPs rendered into Arabic as NPs without art-
icle: see 4.2.1). The greater part of the English indefinite NPs translated as definite
are non-specific and have no article, as in (1), but there are a number of examples
of specific English NPs with indefinite article as in (2):
(2) But the yell had brought a male nurse (p. 9)
lākin al-ṣarkhata ğāat bi-l-mumarriḍ
(But the yell brought the male nurse, p. 13)
Fifty-seven of the examples of the pattern are due to the Arabic usage of the defin-
ite article with uncountables, as in (3):
(3) He’d . . . let the tenor’s voice dress him in silk (p. 4)
wa . . . yadau ṣawta l-muġannī yudaththiruhu bi-l-ḥarīr
(And . . . he’d let the singer’s voice cover him with the silk, p. 10)
This third type has a bearing on Arabic as such, and it is natural that Arabic uses
the definite article more than English since Arabic has less article options than
English in that it lacks the opposition indefinite article vs. zero article. It does not
relate to our question here of the explicitness of referring expressions involved in
translation, so these 57 examples must be left out of consideration.
However, in a large part of the 790 other examples of this pattern, it would
have been acceptable Arabic to use referring expressions without the definite art-
 Martin Ehrensvärd

icle, and the fact that the translator chooses to do so anyway points to a preference
for a greater explicitness.

4.1.2 Indefinite NP: Possessive pronoun NP

(4) . . . the streets of Medallion were hot and dusty with progress (p. 5–6)
wa-kānat šawāriu mīdālyūna ḥārrata wa-muġabbarata ma numūwihā
(And the streets of Medallion were hot and dusty with its progress, p. 11)
88 times an English indefinite NP is translated as a possessive pronoun NP (there
are no examples of English possessive pronoun NPs translated as indefinite NPs).
This shift shows greater explicitness in the target language (henceforth: TL) text.

4.2 Definite NPs

In the corpus, there are examples of six types of relevant Arabic correspondences
to English definite NPs.
(1) Definite article NP: Indefinite NP
(2) Definite article NP: Possessive pronoun NP
(3) Definite article NP: Demonstrative NP
(4) Demonstrative NP: Definite article NP
(5) Elliptical head NP: Full head NP
(6) Possessive pronoun NP: Definite article NP

4.2.1 Definite article NP: Indefinite NP

(5) Returning to his cot he took the blanket (p. 13)
c
āda ilā sarīrihi wa-akhadha baṭṭānīya
(He returned to his bed and took a blanket, p. 16)
There are 48 examples of this shift in the corpus (cf. 4.1.1). Some of the examples
involve superlatives as in (6), and in that case Arabic cannot use the definite art-
icle:
(6) . . . but without the least hope in the world (p. 55)
. . . lākin dūna adnā amalin fī l-ālam
(But without least hope in the world, p. 48)
In these cases, the indefinite phrase is logically definite and only formally indef-
inite. But apart from these, there are a number of examples where the translator
could have used the definite article (as in [5]), but did not, hence making the TL
text less explicit. These serve as a limited amount of counterexamples to the great
number of examples of the mirror image of the pattern in 4.1.1.
 Reference in English–Arabic translation 

4.2.2 Definite article NP: Possessive pronoun NP

(7) When the slave completed the work (p. 5)
indamā akmala l-abdu amalih
(When the slave completed his work, p. 10)
There are 64 examples of this shift (cf. 51 examples of English possessive pronoun
NPs translated as definite article NPs: see 4.2.6). This shift shows greater explicit-
ness in the TL text, but note that the counterexamples of the mirror image of the
pattern also are fairly numerous.

4.2.3 Definite article NP: Demonstrative NP

(8) . . . the word the nurse . . . had called him (p. 10)
tilka l-kalimatu llatī daāhu bi-hā l-mumarriḍ
(This word the nurse had called him, p. 14)
There are 27 examples of this shift (cf. 24 examples of English demonstrative NPs
translated as definite article NPs: see 4.2.4). This shift would show greater explicit-
ness in the TL text if it were not for the mirror image of this pattern, the examples
of which are almost as numerous.

4.2.4 Demonstrative NP: Definite article NP

(9) He ran . . . across this field (p. 8)
rakaḍa . . . abra maydāna l-maraka
(He ran across the battlefield, p. 12)
There are 24 examples of this shift (cf. 27 examples of English definite article NPs
translated as demonstrative NPs: see 4.2.3). These serve as counterexamples to the
mirror image of the pattern in 4.2.3, and since they are almost as numerous, no
conclusions can be drawn from them.

4.2.5 Elliptical head NP: Full head NP

(10) . . . Nel told her [her name] and then asked, “What’s yours?” “Mine’s Ro-
chelle . . . ” (p. 26)
. . . akhbarathā nīl bi-ismihā thumma saalathā: “wa-anti?” “ismī rūšīl . . .”
(Nel told her her name then she asked her: “And you?” “My name is Ro-
chelle”, p. 27)
There are nine examples of this shift (there are no examples of English full head
NPs translated as elliptical head NPs). This shows greater explicitness in the TL
text.
 Martin Ehrensvärd

4.2.6 Possessive pronoun NP: Definite article NP

(11) . . . of the man breathing music in and out of his harmonica (p. 4)
. . . li–l-rağuli lladhi yatanaffasu mūsīqā min al-hārmūnīka
(Of the man who breathed music from the harmonica, p. 10)
There are 51 examples of this shift (cf. 64 examples of English definite article NPs
translated as possessive pronoun NPs: see 4.2.2). These serve as counterexamples
to the mirror image of the pattern in 4.2.2, but being not quite as numerous, the
combined weight of these two patterns shows a slightly greater explicitness in the
TL text.

5. Pronouns

In the corpus, there are examples of four types of relevant Arabic correspondences
to English pronouns.
(1) Pronoun: Proper noun
(2) Pronoun: Definite article NP
(3) Pronoun: Possessive pronoun NP

Note that I have left out of consideration the cases (there are fifteen altogether)
that involve the English pronoun one, as in (12):
(12) That one (p. 65)
dhālika l-ṭifl
(That child, p. 56)
Here the translator has no option other than translating with an NP since Arabic
does not have a pronoun which is used in a similar way.

5.1 Pronoun: Proper noun

(13) Blasted and permanently astonished by the events of 1917, he had re-
turned to Medallion . . . (p. 7)
maṭūban wa-madhhūlan bi-istimrārin bi-sababi aḥdātha ām 1917 kāna
šādrāku qad āda ilā mīdālyūn . . .
(Ruined and permanently alarmed because of the events of the year 1917,
Shadrack had returned to Medallion, p. 12)
There are seven examples of this shift (against two counterexamples). This shows
greater explicitness in the TL text.
 Reference in English–Arabic translation 

5.2 Pronoun: Definite article NP

(14) Then Reverend Deal took it up (p. 16)

thumma akhadha l-qissu dīl zimāma l-umūr
(Then the Reverend Deal took up the reins of power, p. 18)
There are 42 examples of this shift (there are no counterexamples). Ten of the ex-
amples involve the weather as in (15) or other expressions where English uses an
impersonal expression which must be rendered as a definite article NP in Arabic.
(15) . . . later, when it got cooler . . . (p. 75)
. . . fīmā bad ḥīna yabrudu l-ğaww . . .
(Later when the weather got cooler, p. 64)
These must be taken out of consideration since the translator does not have the op-
tion of rendering them as pronouns. Another seven examples must be subtracted
because they involve al-amr, ‘the case’, which again is necessary in Arabic in cer-
tain contexts and does not leave the translator with an alternative. But 25 examples
still remain where rendering the pronoun with a pronoun would have been ac-
ceptable and where the translator chose not to. These show greater explicitness in
the TL text.

5.3 Pronoun: Possessive pronoun NP

(16) Hating BoyBoy, she could get on with it (p. 36)

bi-karāhiyati būybūya sa-tastaṭīu muwāṣalata ḥayātihā
(With hate of BoyBoy she would manage the continuation of her life, p. 35)
There are five examples of this shift (there are no counterexamples). This shows
greater explicitness in the TL text.

6. Summary and conclusion

In each of the patterns, when compared to its mirror image, the TL text is more
explicit than the source language (henceforth: SL) text. In some instances, the gap
is too narrow to draw any conclusion (cf. 4.2.3 and 4.2.4 with their 27 and 24 ex-
amples respectively), but in other instances, the gap is very wide (cf. 4.1.1 and 4.2.1
with their 790 and 48 examples respectively).
In Aziz’ survey, as mentioned, the TL was English, and it was the TL text which
was the more explicit with regard to reference. In my survey the TL was Arabic,
 Martin Ehrensvärd

and it was again the TL text which was the more explicit with regard to reference.
This makes intuitive sense: One might expect the English and Arabic speaking au-
diences not to be so familiar with each others culture, and hence the need for re-
ferring expressions to be more explicit in order to facilitate the understanding of
them. This leads to the conclusion that TL texts, at least when the culture gap is
wide, are more explicit in their use of referring expressions than the SL texts. This
conclusion must remain tentative so long as a very limited number of texts have
been surveyed in this way.

Abstract

The use of referring expressions (proper nouns, noun phrases [NPs] and pronouns) in a target
language [TL] text may differ from the source language [SL] text with regard to how explicit they
are: One may have a pronoun (less explicit) where the other has a proper noun (more explicit),
or one may have an indefinite NP (less explicit) where the other has a definite NP (more expli-
cit). But are these differences in any way systematic?
In an article in Babel, Yowell Y. Aziz shows that remarkable differences obtain in the explicit-
ness of various referring expressions when an Arabic text is compared to its English translation
and that overall, the English text is the more explicit. Now, does this mean that English generally
is more explicit in its use of referring expressions, or is the difference due, rather, to the need of
the TL text to be more explicit in order to facilitate the understanding of the text?
The last option would make good sense. The SL text is written for an audience who knows the
language and hence is expected to be more or less familiar with the culture. Therefore the author
may not need to be as explicit as he or she would be if writing for an audience not so familiar
with the culture.
As a preliminary test of this option, the article compares the explicitness of the referring ex-
pressions in an American novel, Sula by Toni Morrison, and its Arabic translation, and in this
way the article applies Aziz’ method, but instead of going from Arabic to English like Aziz, it
goes the other way around. The TL text again turns out to be more explicit than the SL text (in
fact, much more explicit), thus raising the likelihood that translations (at least across wide cul-
ture gaps) tend to be more explicit in their use of referring expressions than their sources are.

Résumé

L’utilisation d’expressions de référence (noms propres, syntagmes nominaux et pronoms) dans

un texte dans une langue cible peut différer du texte dans la langue source en ce qui concerne
leur caractère explicite : l’un peut utiliser un pronom (moins explicite) là où l’autre utilise un
nom propre (plus explicite) ou l’un peut avoir un syntagme nominal indéfini (moins explicite)
là où l’autre utilise un syntagme nominal défini (plus explicite). Mais ces différences sont-elles
systématiques ?
Dans un article dans Babel, Yowell Y. Aziz montre que l’on arrive à des différences notables
dans le caractère explicite de différentes expressions de référence lorsque l’on compare un texte
 Reference in English–Arabic translation 

arabe à sa traduction anglaise et qu’en général, le texte anglais est le plus explicite. Mais cela si-
gnifie-t-il que l’anglais est généralement plus explicite dans son utilisation d’expressions de réfé-
rence ou la différence réside-t-elle plutôt dans le besoin du texte dans la langue cible d’être plus
explicite afin d’en faciliter la compréhension ?
La dernière option semble judicieuse. Le texte dans la langue source est écrit pour un public
qui connaît la langue et qui est donc censé être plus ou moins familiarisé avec la culture. C’est
pourquoi l’auteur peut ne pas avoir besoin d’être aussi explicite que s’il écrivait pour un public
moins familiarisé avec la culture.
En guise de test préliminaire de cette option, l’article compare le caractère explicite des ex-
pressions de référence dans un roman américain, Sula de Toni Morrison, et dans sa traduction
arabe. Ce faisant, nous appliquons la méthode d’Aziz, mais au lieu de partir de l’arabe vers l’an-
glais, comme Aziz, nous procédons de manière inverse. Le texte en langue cible est à nouveau
plus explicite que le texte en langue source (en fait, beaucoup plus explicite), ce qui augmente la
probabilité que les traductions (du moins dans le cas d’écarts culturels larges) tendent à être plus
explicites que leurs sources en ce qui concerne l’utilisation d’expressions de référence.

About the Author

Martin Ehrensvärd, Ph.D. in Semitic languages (University of Aarhus, Denmark), has been
teaching classical Hebrew at the University of Aarhus from 1990 to 2003. In 2004 he taught
Arabic there, and he is currently a member of the academic staff at the University of Copen-
hagen. He has published five articles in refereed journals on classical Hebrew syntax and one on
Islamic mysticism, and he has translated Greek and Arabic mystical texts into Danish.
Address: Faculty of Theology, University of Copenhagen, Købmagergade 44–46, Copenhagen
1150 K, Denmark. E-mail: [email protected]

View publication stats