Negation, whose universality is unanimously confirmed in the existing literature on human language ([9,25,20,18,17], among others), is a unique property of human language: "Negative utterances are a core feature of every system of human communication and of no system of animal communication" (Horn and Kato, 2000: 1). Linguists ([21,8,17], among others) usually divide negation into two types: explicit negation and implicit negation.
Similarly, there have been several studies which deal, among other things, with the linguistics of negation in Arabic, where negation is also divided into explicit and implicit negation [6,24,4,5,23]. Al-Makhzumi (2005: 265), for example, defines negation as "a linguistic category which is opposed to affirmation and intended to disprove or deny the truth value of a proposition". While explicit Arabic negation employs negative particles, implicit negation uses grammatical devices such as interrogatives and conditionals.
2. Review of Literature
While English verbal negation is uniform in nature, as it only employs the negative particle not in such negation, its Arabic counterpart is highly diversified. In terms of translation, there are only very few studies on the translation of negation. Apostolatu and Apostolatu (2012) deal with literary translation of English negation into Romanian. Dendane and Dendane (2012) refer to the one-to-many correspondence between the English particle not and the many counterparts in both standard and vernacular Arabic. Li (2017) points out the difficulty Chinese EFL learners face when expressing adverbial negation by too due to its Chinese counterpart, which functions as an intensifier.
A similar mishap may occur in English-into-Arabic translation. Farghal and Almanna (2015: 27) briefly examine negation while discussing syntactic features in translation. It is argued that the negation embedded in too can pose a challenge because it requires a translation procedure that recovers negation in Arabic, whether explicitly or implicitly.
1. I think you've been too busy to notice where I have been.
'a unn 'ana-ka kunta ma uulan jiddan li-tulaaḥi a 'ayna 'anaa
think (I) that-you be-you busy very to-notice where I
"I think you were too busy to notice where I am".
Al-Ghazalli (2013) discusses the translation of Arabic implicit negation in a poorly-written paper. Apart from the poor quality of this study, Al-Ghazali unjustifiably argues for unpacking Quranic implicit negation in rhetorical questions. According to his misconception of rhetorical questions, the Quranic verse hal yastawii al-'a'maa wa-l-baṣiir [Q equal the-blind and-the-sighted] is erroneously rendered as a generic rather than a rhetorical question by Quran translators as in Yousef Ali's Can the blind be held equal to the seeing (p. 135) and M. Pickthal's Are the blind and the seer equal (p. 133). Therefore, he claims, implicit negation should be made explicit as in Are the blind and the one who sees equal? Definitely, this is not true (p. 139).
3. Research Method
This is an empirical study based on the extraction of ample textual data involving English adverbial/determiner negation along with their target Arabic counterparts from existing works and their translations. The textual data (270 examples) features two types of markers of adverbial negation (too and hardly/scarcely/barely). The study provides both a quantitative and a qualitative analysis of the data. The sources of the textual data include five series of Harry Potter (HP, henceforth) by J. K. Rowling, namely Harry Potter and the Sorcerer's Stone (1), Harry Potter and the Prisoner of Askaban (2), Harry Potter and the Goblet of Fire (3), Harry Potter and the Order Phoenix (4), and Harry and the Deathly Hallows (5). The first two are translated by Sahar Mahmoud, the third and fourth by Rajaa Abdullah, and the fifth by Ahmed Mohammed (see references for complete information). The textual data sources also include the blue Flower (BF) by P. Fizgerald, 1996 (translated by Ali Suleiman 2015), The Fault in our Stars (FS) by J. Green, 2012 (translated by Baseel Intwan 2015), The Help (TH) by Stockett, 2009 (translated by Hassan Al-Bustani 2010)), The Future: Six Drives of Global Change (GC) by Al Gore, 2013 (translated by Adnan Gergeos 2015), and the Making of Economic Society (ES) by R. Heilbroner 1962 (translated by Rashid Al-Barrawi, 1976).
In terms of the type of negative marker, the English corpus is distributed as shown in Table 1 below:
Apart from the more generic negation by negative markers such as not, no and never, adverbial negation is mainly performed by the negative adverbs too, hardly, scarcely, and barely. These negative adverbs equip the utterance with an inherent negative orientation that needs to be accounted for when translating into other languages, Arabic in our case. The following discussion sheds light on the procedures Arabic translators resort to when dealing with this subtle area of negation which does not formally exist in Arabic.
Negation by too
The negative adverbial too, which is the most frequent in the data (100 occurrences), calls for a variety of translation procedures which may or may not involve the use of generic negation in Arabic. The data shows that Arabic negation by a generic negative particle is recovered in 47 out of 100 cases, while implicit negation procedures are employed in the remaining 53 cases. As can be noted, the discrepancy in choosing between the procedure of generic negation and that of implicit negation when rendering too negation is insignificant.
Recovering generic negation
Examining the recovery of generic negation, it is noted that there are three main procedures adopted by Arabic translators: negation by simple or complex structures (18 cases/38.29%), unpacking negation by coordination (17 cases/36.17%), and negation by indicating degree of attribute (12 cases/25.53%).
To start with negation by employing simple or complex structures (which is the most frequent in this category/38.29%), it is noted that translators may succeed in capturing too negation using this procedure, as can be noted in the examples below:
2. I was too late to save the girl. (HP/2)
lam 'astati' 'inqa -l-fataati fi-l-waqti-l-munaasib
not be able saving-the-girl-in-the-time-suitable
"I couldn't save the girl at the right time".
As can be observed, in (2) the translator has managed to properly recover Arabic generic negation by nominalizing the verb save (saving) and the verb drive away (driving away) in the English infinitive clauses. The English complex structure may be maintained in Arabic by employing the subjunctive case (which corresponds to the infinitive in English). For example, (2) above can be rewritten as lam 'astati' 'an 'unqi a-l-fataata fi-l-waqti-l-munaasib [not be able that save(I)-the-girl in-the-suitable-time] to mean the same thing, i.e. 'I couldn't save the girl at the right time'.
There are few cases (3 out of 18 instances), however, when the translator's recovery of Arabic generic negation does not convey the nuance of adverbial negation properly, as can be witnessed in (3) below:
3.... but Harry was too used to this to care. (HP/3).... wa-laakina haarii lam yahatam bi-haa aa
and-but Harry not care with-this
"... but Harry didn't care about this".
As can be seen, while the translator has managed to convey the general meaning in (3) by recovering Arabic generic negation, he has failed to do justice to the subtlety of the negation expressed by too. In (3), the negation in the Arabic utterance corresponds to English generic negation by not, viz.... but Harry didn't care about this. Hence, the subtle focus of English negation is lost.
To capture the inherent full force of English negation by too here, the Arabic rendering in (3) may be rephrased as follows:
4.... wa-laakinna haarii lam yahtam li-ta'awwudi-hi 'alaa haa aa
and-but Harry not care because-used-he on this
"... but Harry didn't care because he was used to this".
Unpacking too negation by a coordinate Arabic structure featuring explicit negation comes second in frequency at 36.17% (all from HP series). It proves to be a workable procedure, as can be observed in the following example:
5. Harry was too deeply asleep to hear her. (HP/4)
haarii kaana ɣaariqan fii nawmi-hi fa-lam yasma'-haa
Harry was sinking in sleep-his so-not hear-her
"Harry was deeply asleep, so he didn't hear her".
The translator in (5) has successfully managed to capture the focus of too negation despite the fact that she resorts to Arabic coordinate structure. In fact, the use of coordinate structure for unpacking the meaning of too negation has proved workable in all the cases in which this procedure is employed.
The third procedure (which indicates the degree of the attribute in question) accounts for 25.53% of the examples in this category. Semantically, it corresponds to awkwardly rephrasing too negation by using the phrase to the extent that with negation by not in English, viz. John was too short to touch the ceiling may awkwardly be rephrased as John was short to the extent that he couldn't touch the ceiling. In Arabic, this procedure proves very useful for rendering too negation. The following example is illustrative:
6. Professor Trelawney seemed too tipsy to have recognized Harry. (HP/5)
badat al-'ustaa atu triilawnii maxmuuratun li-darajati 'anna-ha lam ta'rif haarii seemed the-professor Trelawney drunk to-extent that-she not know Harry "Professor Trelawney was drunk to the extent that she didn't recognize Harry".
Employing implicit negation
Apart from a mixed bag of erroneous translations/under-translations (16 instances/ 30.19%), there are three main procedures here for capturing the meaning of too negation: employing the comparative form (16 instances/28.30%), employing negative verbs (11/20.75%), and indicating degree of attribute (10/18.87%).
The procedure using the Arabic comparative form emerges as very useful for handling too negation. The comparative Arabic forms `af'al min [more of an attribute than] and `akbar maṣdar (verbal noun) min [more of verbal noun than] here capture the nuance that the force of X's attribute goes beyond the capability of Act Y, e.g. `a kaa min `an yuxda' [cleverer than that (he) be deceived] and `akθara akaa'an min `an yuxda' [(has) more cleverness than that (he) be deceived] which both idiomatically translate into Ali is too clever to deceive. Following is an example from the corpus:
7.... because it was too long to memorize. ((FS)
... li'anna-ha 'aṭwalu min 'an tuḥfa a ɣayban
because-it longer from that be learned by heart
“... because it was longer than it could be memorized”.
The second procedure utilizes negative verbs/verbals to relay the meaning of too negation. This is a familiar procedure in English as well as in Arabic to express negation implicitly rather than explicitly. For example, 'the act of denying doing something' implies 'the act of not admitting doing it'. Consequently, this procedure constitutes an important option when translating negation in general and too negation in particular. The example below is illustrative:
8. Anna dies or becomes too ill to continue writing it. (FS)
'aana maatat 'aw balaɣat min-al-maraḍi ḥaddan ḥaala bayna-haa
Anna died or reached from-the-illness degree prevented between-her
"Anna died or reached a degree of illness that prevented her from continuing writing".
The implicit negation in (8) is achieved by the use of the negative verbs ḥaala [prevented] and yafuuqu [exceed], which both imply propositions employing explicit negation in Arabic.
The third procedure employs the degree of the relevant attribute as a marker of implicit negation by using the degree formulas min + maṣdar [verbal noun], e.g. from the-smallness [i.e. too small] and verb/adjective + 'alaa [on], e.g. (grow) old on [i.e. too old]. Observe the example below:
9. Fines for violations were too small to be effective,... (ES)
fa-l-ɣaraamaat-u 'an-il-moxaalafaati kaanat min-aṣ-ṣ-iɣari biḥayθu
so-the-fines from-the-violations were from-the-smallness so
"The fines for violations were so small that they lost their effect".
Finally, we have the mixed bag, which includes mistranslations/under-translations that account for 16/53 instances (a full 30.19%) in the cases of implicit negation. The example below is illustrative:
10. It is still far too cold to undress at night. (BF)
laayazaalu-ṭ-ṭaqsu baaridan jiddan li-xal'i-θ- θiyaabi fi-l-layli
still-the-weather cold very to-take off-the-clothes in-the-night
"The weather is still very [too] cold to take off clothes at night",
11. maazaala-ṭ-ṭaqsu-l-baaridu yamna'u-naa min xal'i malaabisin-aa fi-l-layli still-the-weather-the-cold prevent-us from taking off clothes-our in-the-night
"The cold weather still prevents us from taking our clothes off at night".
To close this section on translating too negation, Table 2 presents the results reported in the discussion.
Negation by hardly/scarcely/barely
The sample of these negative adverbs (which is extracted from BF, FS, GC, and ES, to the exclusion of HP) includes 34 instances of hardly, 19 of scarcely and 11 of barely, coming to 65 instances. These adverbs share the fact that they communicate a negative orientation when used in English sentences. That is why they are often interchangeable albeit they may be sensitive to normality conditions, i.e. one may sound natural in one context, while another may not. For example, John was barely 17 when he joined college is natural, whereas John was scarcely 17 when he joined college is not. In terms of translation, the focus is on relaying the negative orientation which is shared by all of them.
The data shows three main procedures the translators have employed in rendering the negation by these adverbs: explicit negation (28 instances/43.75%), vernacular بالكاد (20 instances/31.25%), and implicit negation (12 instances/18.75%). The remaining cases go for mistranslation (3 instances) and omission (1 instance), together 6.25%.
Arabic negation by explicit negative particles emerges as the most common translation procedure for rendering the -ly negative adverbs, which clearly indicates the translators' awareness of their negative orientation. However, the coding of this orientation in Arabic seems to be a challenging task. In fact, more than half (16 out of the 28 cases/57.14%) involves under-translating this adverbial negation by rendering it into what corresponds to negation by not in English. Consequently, the subtle nuance of this type of negation is lost in translation.
Let us start with cases where the adverbial negation is accounted for properly in Arabic (12/28 instances/42.85%). The following example is illustrative:
12.... this little kid who could barely walk... (FS)... haa aa-ṭ-ṭiflu-ṣ-ṣaɣiiru 'alla ii laa yakadu yam ii...
this-the-kid-the-little who not hardly walk
"... this little kid who can hardly walk..."
The translator of (12) has successfully employed a negated yakadu, viz. laa yakaadu [not hardly], which exactly captures the meaning of the -ly negation in (12).
However, -ly adverbial negation does not seem as straightforward as (12) may suggest. While capturing the notion of negation in general, almost 58% of the Arabic renderings (16/28 instances) fail to account for the nuance inherent in -ly negation. Instead, this kind of negation is erroneously relayed as Arabic negation that corresponds to English negation by not, thus amounting to serious under-translations. Following is an illustrative example:
13. The children of large families hardly ever learn to talk to themselves aloud,... (BF) lam yata'allam 'awlaadu-l-'aa'ilaati-l-kabiirati 'an yataḥaddaθuu 'ilaa 'anfusihim not learned children-the-families-the-big that talk to themselves bi-ṣawtin masmuu-in... with-voice audible
"The children of large families did not learn to talk to themselves in an audible voice..."
By way of illustration, in (13) the translator obliterates the subtle nuance of the negation in hardly by opting for explicit Arabic negation by lam [not]followed by the main lexical verb yata'allam [learn], which back-translates into English negation by not, viz. 'The children of large families did not learn to talk to themselves...'. To capture the negation inherent in hardly, one may need to employ a negated yakaadu viz. laa yakaadu 'awlaadu-l-'aa'ilaati-l-kabiirati yata'allamuuna-t-tataḥadduθa 'ilaa 'anfusihim bi-ṣawtin masmuu-in... [not hardly children-the-families-the-large learn-the-talking to themselves with-voice audible] "The children of large families hardly learn to talk to themselves in an audible voice...".
The second procedure for rendering -ly negation is the employment of the vernacular negative adverb bilkaad, which is derived from the standard verb yakaadu. The question is whether it is appropriate to use a vernacular form when it is possible to utilize the standard negated yakaadu. What is surprising here the absence of this vernacular form in Harry Potter where the informal register may sanction it as well as the frequency of using it in the other works - it accounts for 31.25% of the -ly data (20/64 instances). By way of illustration, the example in (14),
14. I hardly know you, Augustus Waters. (FS)
'anaa bilkaad 'a'rifu-ka yaa 'aɣusṭus wuutarz
I hardly know(I)-you oh Augustus Waters
"I hardly know you Augustus Waters",
can readily be rephrased naturally in standard Arabic using the negated yakaadu as in (15) below.
15. 'anaa laa 'akaadu 'a'rifu-ka yaa 'aɣusṭus wuutarz
I not hardly(I) know(I)-you oh Augustus Waters
"I hardly know you Augustus Waters".
Next we have the procedure of implicit negation which accounts for 18.75% in the –ly negation data (12/64 instances). They mainly employ paucity or difficulty expressions in an attempt to capture the negative nuance inherent in -ly negation. Consider the example below:
16. Auguste nowadays scarcely ever went out at all,... (BF)
fii miθli haa ihi-l-'ayyaami kaana min-n-naadiri bi-nasbati li-'uuɣast
in like these-the-days was from-the-rarity as-regards-Auguste
'an ta haba xaariji-l-manzili that go(she) outside-the-house
In (16), the translator successfully employs a paucity expression min-al-naadiri [from the rarity] to capture -ly negation. One should note that a negated yakaadu can be readily used for that purpose, viz. fii miθli haa ihi-l-'ayyaami laa takaadu 'uuɣast taxruju min-al-manzili [in like these days Auguste not hardly goes out from-the-house] "In these days Auguste hardly ever goes out".
To close this section, let us examine one mistranslation (out of 3) which are involving –ly negation in (17):
17. This hardly seems like a particularly exciting subject for historical scrutiny.
yakaadu haa aa 'a baha bi-mawḍuu'in muθiirin bi-wajhin xaṣṣin
hardly this like with-subject exiting with-face particular
"This almost seems like a particularly exciting subject for historical research".
To explain, the translator in (17) wrongly uses the affirmed rather than the negated yakaadu, which is an approximating rather a negating marker, i.e. here it communicates the message that 'X is almost Y'. To capture -ly negation, the translation should read laa yakaadu haa aa yabduu abiihan li-mawḍuu'in muθiirin bi-wajhin xaṣṣin li-baḥθi-l-taariixiyyi [not hardly this seems like with-subject exciting with-face particular with-the-research-historical] "This hardly seems like a particularly exciting subject for historical research...".
To conclude, Table 3 below presents the results reported in this section on the translation of -ly negation:
Overall, the findings reported in Tables 2-3 support the predictive title of this study that non-generic adverbial negation is a problematic area for Arabic translators. Together, the Tables indicate that 56 out of the 164 examples comprising the textual data (about 34.2%) involve translational problems including erroneous renderings, under-translations, mistranslations, and inappropriate use of vernacular forms. This constitutes solid evidence that the pragmatics of this type of English negation is a challenging task in translation activity. Arabic translators, both professionals and more so student translators, need to be alerted to the fact that this type of negation, which formally does not exist in Arabic (and probably in several other languages), calls, in the first place, for recovering generic negation and, in the second place, for appropriately investing implicit negation.
* This research was supported by Kuwait University, Research Grant No. [AE03/16]