Mechanisms for Translating English Collocations into Arabic

Brashi (2005) conducted a study using collocational competence and collocational performance data among English-Arabic student and professional translators, with the aim being to increase understanding of the mechanisms underlying translating English collocations into Arabic. The study involved the use of Arabic collocations in translation by student and professional English-Arabic translators, and how this use of Arabic collocations was related to the translators understanding of Arabic collocations. The study provided a practical approach to the translation of collocations. Brashi concluded that translators should not create their own collocations when an equivalent target collocation exists, because the resulting collocation may be misinterpreted. In order to produce acceptable target language collocations, translators must completely separate the SLCs from the TL. This will help them retrieve equivalent TLCs from their memories or other sources. Translators should produce TLCs that sound natural and native-like when read by a monolingual native speaker. Furthermore Brashi refers to the importance of the translators’ collocational ability which enables them to select the most appropriate collocations. A near-perfect knowledge of SLCs and their equivalents in the TL is an essential requisite for, what this study terms, collocational performance in translation. This includes the awareness of the differences between the source and target languages in terms of the formation of collocations. Although the study focused on translating English verb + object and adjective + noun collocations into Arabic, translators can apply Brashi’s suggested model in the following ways: translating other types of collocation; translating from Arabic into English; and translating between any two language pairs.

Bahumaid (2006) investigated the procedures employed by translators in rendering collocations whose TL equivalents are unknown to them. The two part study involved four Arab university instructors who taught translation and had worked in the area. It comprised a translation test consisting of 30 sentences on different types of contextualized collocations, involving 15 English and 11 Arabic collocations, as well as 4 Arabic phrases. Some of the collocations were of the general type, and others were specific. The results indicated that translators resorted to several procedures in dealing with these kinds of collocations. The research also found that culture-bound and register-specific collocations created the greatest challenge in translation, while collocations with literal meanings were significantly easier to render. Also uncovered was the student translator’s use of strategies such as: giving the meaning of the collocations; using synonyms or near-synonyms; attempting literal translation; and avoiding the translation completely.

A research study was conducted by Dweik and Abu Shakra (2010) to investigate strategies used by MA translation students when they encountered difficulties in translating collocations from Arabic into English in religious texts. The study findings revealed a clear correlation between translation strategies used and student translation errors, as well as indicating the use by students of two major strategies: achievement strategies (also referred to as compensatory strategies); and the reduction strategy, which includes both the avoidance as well as the deletion strategies. They also concluded that, regardless of the implied message, in literal renditions of semantic collocations, the lack of attention paid by the translator to the translation would usually be a contributing factor. In other words,  most of the TL readers more possibly loss the entire connotations which may result in unsuccessful translation. On the other hand, in translating lexical collocations, Dweik and Abu Shakra suggested that translators shouldn’t resort just to either the strategy of synonymy or the strategy of generalization. The researchers stated that resorting to those two strategies is an indication of translators’ lack of awareness of collocational restrictions.

Although only recently collocation has become the focus on one particular linguistic study, it attracts an increasing interest from various linguists and is defined and classified in numerous ways. As emerges from this review, there is no exhaustive and uniform definition or categorization of collocation. Collocations are considered one of the most problematic and error-generating areas of in translation. The difficulty of translating collocations is highlighted by number of linguists (e.g., Newmark 1988; Baker 1992) and translators in the course of translating as well. However, collocation has been neglected in translation studies, especially from English into Arabic. Apart from some attempts made by a few writers on translation (Baker 1992; Ghazala 1995), the literature on collocations has been mainly confined to lexicology and applied linguistics, especially the domains of language pedagogy and second language acquisition (for example, Carter, 1987; Cowie 1978; Richards 1971). Adopting the corpus-based approach, this comparative translation study is an attempt to fill the gap, to some extent, in the literature of the translation of collocation and collocability. Research investigating collocations in translation has rarely, if not, adopted the corpus-based approach in dealing with data collection Bahumaid (2006). The study opts for parallel corpus to avoid limitations of test approach in terms of time-consuming and error-prone, aspects that may entail the loss of data productivity.

Furthermore, as was shown earlier in the introductory chapter, this study arises from the fact that, to the best of the researcher’s knowledge, there has been no previous research on translational strategies of English collocations into Arabic in the translation of literary works. In response to the issues in translating collocations, particularly from English into Arabic, empirical studies are required. These should provide the new insights that are needed.

3. Methodology

Overview:

This chapter highlights the use of the corpus-based approach as a method of data collection in this study. Some other aspects regarding the corpus will also be presented within this chapter: justification of the corpus selection, vis. Hemmingway’s novel The old man and the sea, general overview of the novel, and validation of Arabic translators choice of the novel. Then, this chapter deals with the selection of collocation in this study (e.g., its scope, selection process). After that, it describes the theoretical schemes of data categorization and the method of data analysis.

3-1- Data collection

3-1-1-Method

In this study, data were collected by using corpus-based approach, parallel corpora in particular. This has been done by selecting an English novel as the source text (ST) and its four translations into Arabic as the target text (TT). Partington (1998) states that a corpus-based approach has become an established tool for linguistic analysis, providing data that are significantly different from the other ways of data collection.

Data obtained from a corpus in English language with its four translations in Arabic  permit to explore the translational variation in dealing with collocation between languages of different linguistic and cultural system. Moreover, data collected by using this approach can be a source for comparative analysis of the four translators’ performance in terms of the strategic choices. Such comparative analysis may raise the external validity of the research and lead to results that may assist the apprentice translators to improve their decision-making strategies (Venuti 2004:182). Comparative studies will allow other researchers to check the validity of current thoughts about collocations already being made, and provide new and enriching linguistic research. It may also allow other researchers not only to check the validity of the basic claims being made in this study, but also to offer different interpretations of these results to enrich the linguistic literature. Furthermore, Baker (2006) states that parallel corpus provides translational behaviour of lexes and structures (collocations in this study) in terms of language pair relationships. According to Munday (2007), parallel corpora can permit to both quantitative and qualitative analysis.The combination of these methods of analysis can “compromise complementary interdisciplinary methodology that reveals different patterns of translational strategies which may otherwise pass unnoticed” (Munday 2007: 181). In addition, by selecting parallel corpora in this study, the researcher can, statistically compare ST with TTs in terms of the translational strategies and critically analyse the potential linguistic, cultural, and contextual factors implied in the selection of individual strategies. Besides, parallel corpora are consistent with the aims of the study as it provides the researcher with various examples of how certain lexical items or grammatical structures have been dealt with in translation and what strategies were employed to reach optimal translation.

3-1-1-The selection of the corpus ‘The old man and the sea’

Ernest Hemingway’s English novel The old man and the sea was selected on the basis of Hemmingway’s reputation. He is a well-known writer who has contributed a great deal to English literature and the English language through his mastery of flexible and simple English prose. Ernest Hemingway’s writing is among the most recognizable and influential prose of the twentieth century.

In its literary discourse The old man and the sea uses his prose as a ‘metaphor’ for man’s struggle against the forces of nature. The novel, only just over 100 pages long, relates the story of an out-of-luck old fisherman. In this novel, and using Hemingway’s unique style and expert deep-sea fishing knowledge, the narrator relates the thoughts of the protagonist throughout his journey. The story uses portrays practical fishing terms, detailed descriptions, and figurative language, but always through simple, direct language (Bonynge 1977:11), which was noted when he was awarded the Nobel Prize for The old man and the sea in 1954.

Regarding the linguistic style of this novel, it is noticed that, under the influence of his journalistic style, Hemingway, uses a minimum amount of words ( Graham 1989). However, these words imply maximum meaning and effect. Thus, its reputation, the simple, yet immensely powerful, pithy, and direct language, as well as the frequent figurative and metaphoric use of language in this novel inspire its selection for this study. The selection of this novel is motivated by Hemingway’s uncomplicated technique, the plain grammar and easily accessible language, as well as the economy of carefully-measured words. For example, the author frequently uses simple adjectives such as (big, great, small) as a style of understatement. However, these adjectives are used in different contexts referring to different literal and metaphoric meanings in order to influence the reader’s reaction. Such use of language led the critic Anthony (1953) to describe the novel as an incomparable text in which every word has a symbolic meaning and no word is superfluous. Furthermore, this novel is rich in terms of the culturally-bound, language-specific, register-based collocations that entail translators to resort to different strategies in order to produce an affective or appropriate translation.

3-1-2-The selection of translators

Unlike most translation research on collocations which depends on students’ outputs as an object of study, this study relies on professional translators so that more valid data can be obtained. There seems to be a consensus that data obtained from professional translations are more significant and central than from non-professional ones (Toury 1995).  The primacy of professional translations has been established within the framework of a norm-based theory of translation (see Toury 1995). Chesterman (1993:7) also confirms that data obtained from the translational performance of competent professional translators may provide a basis for translation studies.

Four Arabic translations of the novel The old man and the sea have been selected in this study. The use of four translations of the same ST, rather than one translation, was based on what Munday (2007: 46) proposes. The use of one-to-one parallel corpora results in a translation that only represents “one’s individual introspection, albeit contextually and contextually informed”. Therefore, my selection of translations is intended to collect more valid data and more comparable results. The selected translations are produced at different times and places by the following four professional Arabic translators:

1. Zyaad Zakrya (1980)
2. Muneer Balabeki (1992)
3. Fadhil Habeeb Muhesen (2001)
4. Samir Izzat  Nasser (2007)

Many factors have been taken into consideration in the selection of translators’ samples in this study. These factors are: the reputation, knowledge and experience of the translators, and their majoring in translating the literary text. Hatim & Munday (2004) confirm that a translator should be culturally and linguistically competent in languages to produce effective and adequate translations. The four translators under-study have long years of experience in translation. Their linguistic competence in Arabic and English can be proved by their being native speakers of Arabic, their qualifications in English language, their membership of the international and national union of translation as well as their remarkable contribution in the translation field.  It is recognised that their cultural awareness can be ascribed to their long translational experience of dealing with the English cultural concepts.

A further motivation in selecting these translators in particular is their experience in translating literary works. Their great contributions in translating literary works  indicate that they have exclusive combination of translation expertise and artistic knowledge. Considering that each one of them has translated more than 100 literary books (AOFT 2006) may refer to their long experience in dealing with the challenges of translating the literary language. Their extensive experience equips them with a variety of strategies or techniques in dealing with the various translational challenges including those of collocations.

3-3 The selection of collocations

As mentioned earlier in the literature review, collocation is a linguistic item which is used and classified into many different types such as lexical and grammatical collocation. This study focuses on lexical collocations only, adjective+ noun in particular (hereafter collocations and lexical collocations are used interchangeably). This selection was based on the following: First, as has been mentioned in the literature review, “grammatical collocation has received its fair share of emphasis in the work of grammarians and lexicographers” (Abu-Ssaydeh 1991:66); therefore, it seems that more attention in research should be given to lexical collocation. Second, the key challenges in translation are lexical rather than grammatical – i.e. collocations or idioms (Newmark 1988:32). The area of collocation within lexis, in particular, is significant in translation and forming a particular challenge for translators. The meaning association between the lexical components of a collocation is apparently arbitrary and non-predictable. For example, the adjectives rich and wealthy are synonymous; however, most native speakers of English are more likely to produce a collocation such as wealthy imagination not rich imagination. In Arabic, the noun xayaal (imagination) collocates with the adjective waasi3 (wide), or the adjective xiSb (fertile).

The interest of this research in the translation of (ANCs) arises mainly from their frequent occurrence in the selected novel. In addition, prior research shows that the most commonly used categories of collocations are adjective+ noun collocations (Kayo 2005). Accordingly, this paper will limit itself to the strategies used in translating ANCs only.

The sample selection of ANCs has been guided by the definition of the target population and decisions concerning the method of sampling it. Cruse’s (1986) notion of mutual expectancy between the collocates is adopted because it is mainly concerned with adjective + noun clusters. All ANCs extracted from the corpus meet Cruse’s basic criterion of a collocation- that is, there must be a syntagmatic association between adjective and noun. Firth (1968) found there was a syntagmatic relation between two words if they co-occur more frequently than expected in spoken or written language, and if their grammatical roles are different. Thus, each component of the selected collocations such as great size and great help habitually co-occurs with each other and semantically transparent. Moreover, each component of these collocations has different grammatical roles.

Although, it is theoretically taken for granted, according to Sinclair (1991), that collocations can consist of two or more words, only sequences of consecutive adjective and noun have been selected for this study. The selection of directly adjacent ANCs was based on practical applications, viz. consecutive words were found to be particularly common in the texts analysed in previous collocational research. Seratan (2011) confirms that the practical research of collocations deals almost exclusively with binary collocations, made up of only two words. Accordingly, collocations such as high cumulus clouds , great erect tail, and  wide lavender stripes were excluded from the scope of the study.

The list of syntagmaticly-associated and directly adjacent ANCs was exhaustive since there were many collocations that met these criteria in the selected corpus. Thus, the selection has been made to ANCs that are related to the size (length, width, height, and depth). The reason for this selection is twofold. First, the frequent use of such kind of collocations due to the nature of the theme in the selected novel. That is, the writer uses myriad descriptions for the main hero (the old man) and the fishing-related elements such as of the sea, fish, sharks, and cord. Second, many of the size-related collocations used in the novel are culture-bound, context-dependent and metaphorically used. So translators may resort to different strategies to overcome the translational challenges.

The contexts and the co-text of the source collocation in the original text have been considered as additional inspection in the inclusion or exclusion of collocations for this study. For example, considering the context of the adjective ‘great’ in collocations such as the great negro, a great help, and great manager in the original text determines the inclusion of the first two collocations and exclusion of the third one as they refer to the size and value respectively. Similarly, the context of the collocation big sea turtle in the original text determines that the adjective ‘big’ describes the turtle not the sea. That means the collocation is excluded based on the criteria of the directly adjacent ANCs.

Applying the above-mentioned criteria resulted in the selection of 156 ANCs at the first stage. Then, these collocations have been validated by three native speakers of English and two linguists which led to reducing them to 100 collocations. The intuition of the NSs served as a filter (i.e., to be sure that the selected items are acceptable collocations) because “deciding whether or not a collocation is acceptable constitutes part of the intuitive knowledge of the native speaker of the language” (Buhameed, 2006: 135). The consultation with two linguists was done to confirm that the selected collocations meet Cruse’s basic criterion of collocation (i.e. syntagmatic relationship) and to ensure that each collocation consisted of adjective and noun. This led to collocations such as “much faith” and “little chance” to be excluded from this study because the first components (much, little) are considered as determinatives by most scholars in spite of being considered adjectives by traditional grammarians (Huddleston 1995: 116).

Table (1) lists the 100 ANCs that have been extracted from the novel. The researcher took into the account the repetition of some collocations and recognised the strategies used in translating each of them (the number in brackets in front of some collocations refers to their repetitions in the original text). This is so because lexical items, especially in the literary text, acquire different connotations and shades of meaning in different contexts which require using different strategies when being translated.

Table (1) ANCs extracted from the novel

1. deep wrinkles
2. full length (2)*
3. the whole boat
4. a long time (4)
5. big fish (7)
6. great fish (12)
7. long way
8. great drives
9. the longest ball
10. a great mistake
11. the big Leagues (2)
12. a short time
13. the high capes
14. great occurrences
15. the great well
16. the deepest holes
17. the deep wells
18. small tuna (3)
19. deepest lines
20. big pencil
21. big dolphin (2)
22. small line (3)
23. the slight bulge
24. a big school of dolphin
25. little chance
26. the tiny fish
27. the small shade
28. great value
29. big drum
30. small tunas
31. a great help
32. deep prisms
33. the high sun
34. the great weight (2)
35. a huge fish
36. grand leagues
37. the great hill
38. a big reserve of line
39. a lesser depth
40. small bird (3)
41. small breeze
42. full piece
43. great size
44. small boat
45. thin feathers
46. the biggest fish
47. small sea
48. little line
49. higher level
50. Light brisa
51. Less eye
52. the great birds
53. the great negro
54. high chairs
55. a medium-sized hook
56. tremendous effort
57. the greater speed
58. a great island
59. the big line
60. the big cord
61. the whole bonito
62. great dignity
63. his great speed
64. small lurch
65. a great canyon of clouds
66. a great bursting
67. a faint slackening of the pressure
68. very big circle
69. great temptation
70. a long island
71. huge bulk
72. huge pectorals
73. their whole bodies
74. great tail (3)
75. his great length
76. small drink of water
77. the double rope
78. bigger skiff
79. great strangeness
80. huge jaws
81. little hope
82. the biggest dentuso
83. great thing (2)
84. great handicap
85. great hunger
86. The big shark
87. The short club-(2)
88. the wide trail
89. broad head
90. The great sea
91. Little harbour
92. The little patch
93. A huge tail
94. The long backbone
95. A vast school of porpoise
96. The small  shrimps
97. double lip
98. Long jumps
99. The big eyes
100. The great trunkbacks

*The number in brackets in front of some collocations refers to their repetitions in the original text.

3-3 Categorization of translational strategies

The categorization scheme used for classifying the strategies employed in translating collocations in this study is based primarily on Buhameed’s (2006) study results mentioned in section (2.4). However, this categorization scheme is expanded to accommodate the strategies observed in this study. Additional categories added to Buhameed’s categorization scheme include collocational equivalent, descriptive equivalent, encapsulation and mistranslation. The researcher also added a new strategy of missing adjective to accommodate the outcomes. In particular, it was observed that translators in some instances tend to delete the first component of ANCs (i.e. adjective) and translate the noun only (Missing adjective).

Moreover, a modified sub-scheme is proposed mainly based on Ghazala’s (2004) study. As sub-categorization process went on the researcher also added a new strategy (i.e., shifting adjective into noun) to accommodate the outcomes. The new strategy is considered as part of the sub-scheme.

Tables (2) and (3) show the categorization scheme and sub-scheme used for analysing the translational strategies obtained in this study. Below is a definition of all these strategies:

1. Literal translation:  A word-for-word translation of a source language collocation;
2. Synonym or near-synonym: Resorting to use a substitute item for both or one of the collocation items if these items share certain semantic properties;
3. Avoidance: This cases where the translators opted for avoiding the entire collocation;
4. Collocational equivalent:  The translation of a collocational SL cluster is translated by a TL collocational cluster  (Newmark 1988);
5. Descriptive equivalent: The meaning of collocation is explained in several words (Newmark 1988);
6. Encapsulation: An expression in one language could be represented by “a single lexeme with roughly the same meaning in another language” (Lyons, 1978:262);
7. Mistranslation: Producing inaccurate or ineffective translation of the source language collocation which led the meaning of the source collocation to be distorted;
8. Missing adjective: Rendering the English adjective- noun collocation into only a noun in Arabic. That is, translators rendered the source language collocation incompletely.

Table (2): The categorization scheme  of translational strategies

Strategies  Source

– Literal translation

– Synonym- or near-synonym

–  Avoidance              Buhameed (2006)

– Collocational equivalent

– Descriptive equivalent  Newmark (1988)

– Encapsulation Lyons (1979)

– Mistranslation Brashi (2008)

– Missing adjective

Below is a definition of the four additional strategies used in the sub-scheme:

1. Classifier: Explaining unknown cultural expression in simple, general terms, demonstrating its class, type, category, etc. (Ghazala 2004);
2. Transcription:  The transference of the English word into Arabic alphabetic exactly as pronounced;
3. Footnote: Giving long and additional details about the source collocation;
4. Shifting the adjective into noun: Changing the adjective in the SL into a noun in the TL. The use of this strategy results in target collocation which consists of two nouns.

Table (3): The categorization sub-scheme of translational strategies adopted in this study

Strategies Source

• Classifier
• Transcription
• Footnote  Ghazala (2004)
• Shifting adjective into noun

According to Buhamaid (2006) the categorization of the translational strategies is a matter of more or less in most cases. What is categorized as a strategy of literal translation, for example, by someone, might be categorized as strategy of synonym by another. Thus, the researcher in this study has applied an accuracy check of the employed strategies. He has consulted Arabic linguist, specializing in translation, as well as Arabic postgraduate student majoring in translation in order to achieve optimal and less subjective categorization of translational strategies.

Following the data categorization and organization, the researcher used the Statistical Package for the Social Sciences (SPSS) 5.0 for Windows software package to analyse the data. Charts and tabulations including frequency distribution of the various strategies were used for the analysis to compare translators’ performance. In addition, the Yates Corrected Chi-square (χ2) test was used to determine whether there were considerable differences between the four translators in selecting the translational strategies. That is, this test was used to further ascertain the relations between strategy use and translators.

4- Data analysis and Findings

Overview:

This chapter presents a description of the procedures used in analysing data. Then, it discusses the results of this analysis. The purpose of this process is to recognize, illustrate, and exemplify all translational strategies used by the four translators in dealing with the selected items of ANCs. It also presents the frequencies of these strategies and points out the differences and similarities between the translations of the four translators.

4-1-Data analysis

All Arabic examples that refer to the strategies used by four translators have been back translated. The four translators and the utilized strategies were coded as follows:

Translator Code number

Translator 1 T1

Translator 2 T2

Translator 3 T3

Translator 4 T4

Strategy  Coding

Literal translation

LT

Collocational equivalent

CE

Synonym- or near-synonym

S

Missing adjective MA

Avoidance              A

Mistranslation MT

Encapsulation E

Classifier C

Transcription   TR

Shifting adjective into noun SH

Footnote  FN

After categorization, data were organized in separated tables, according to the selected collocation and the translation strategies that were used by each translator (Appendix 1).The first stage of data analysis included the process of identifying and categorizing the translational strategies according to the categorization schemes described in the methodology chapter (tables 1 and 2). The second stage was to investigate the frequency of occurrence of translational strategies. This has been done by computerizing the strategies that are used by each translator in translating all ANCs using the Statistical Package for the Social Sciences (SPSS) 5.0 for Windows software package. This package was effective in identifying the overall frequency of use of each strategy by all translators, in addition to its frequency by each translator. In the third stage, the differences and similarities between the four translators in terms of the use of translational strategies were identified. The comparison between the four translators had quantitative and qualitative aspects. The first was related to the frequency of occurrence of translational strategies and was carried out using the Yates Corrected Chi-square (χ2) test in the SPSS.  This test was used to determine whether there were considerable differences between the four translators in selecting the translational strategies. That is, Chi-square test was used to further ascertain the relations between strategy use and translators. The four translators represented the independent variables in this study while the dependant variables were represented by the strategies used by these translators. The conventional level of 0.05, which is more often used to determine significance in educational research, was used as the statistical level of significance for all analyses. The conventional level of 0.05 indicates that there is a 5 precent chance that the results of the research are due to chance, rather than an actual relationship among the variables in the study. Accordingly, if (χ2) value was less than 0.05, the null hypothesis (i.e. there is no relationship between translators and strategy use), was rejected. This may  lead the researcher to conclude that factors other than chance such as cultural, contextual, or text type might affect translators to use different strategies  in order to reach acceptable translations (see table 4 below). Furthermore, the cross tabulation calculation has been used in order to facilitate the comparison process between the four translators in terms of using strategies.

Table (4): Yates corrected Chi-square results

Strategy (translators) Chi-square Value  DF Critical value (CV) Significance

LT      (T1, T2, T3, T4) 43.013 21 32.67 YES

CE     (T1, T2, T3, T4) 30.985 21 32.67 NO

S         (T1, T2, T3, T4) 34.600 21 32.67 YES

MA    (T1, T2, T3, T4) 32.667 21 32.67 YES

A        (T1, T2, T3, T4) 8.931 21 32.67 NO

DE     (T1, T2, T3, T4) 34.857 21 32.67 YES

MT    (T1, T2, T3, T4) 10.889 21 32.67 NO

E        (T1, T2, T3, T4) 6.294 21 32.67 NO

LT  (T1,T2,T3)  (T4) 18.641 7 14.07 YES

M   (T1,T2,T4)   (T3) 28.053 7 14.07 YES

DE  (T1,T2,T3)   (T4) 32.190 7 14.07 YES

Null Hypothesis (H0):                 No difference in strategies use between T1, T2, T3, T4

Alterative Hypothesis (H1):        There is a difference in strategies use between T1, T2, T3, T4

X= 0.05         Level of significance

The qualitative aspect shows whether the four translators had resorted to different or similar translational strategies in dealing with the various challenges of translating ANCs.

4-2- Findings

Translators have resorted to the following translational strategies in dealing with the challenges of translating ANCs: literal translation (LT), collocational equivalent (CE), synonym- or near-synonym (S), avoidance (A), missing adjective (MA), mistranslation (MT), descriptive equivalent (DE), and encapsulation.  Table (5) and Figure (1) below present these strategies in terms of their frequencies (Fr.) and percentages (%).Table (5) and Figure (1) show that strategies such as LT, S, and, CE were more frequently used by translators. On the other hand, strategies such as MT and E were less frequently used in this study. The strategy of E only constitutes 3.1 % of the translation strategies identified in this study.  Comparatively speaking, it has been noticed that there is some variation between translators in terms extent of usage such strategies. For example, T1 most frequently resorted to LT with (72). However, T4 restored to this strategy (16) times only. The strategy of MA is used (19) times by T2, but it is never used by T3.

The variation between translators is clearly noticed in terms of the type of used strategies. T1 and T4 resorted to all strategies. However, T2 did not use the strategy of DE and MT. The strategy of MT is used by T1and T2 just once. However,  it is used (8) times by T3 and T4. Table (5) also shows that there is a significant difference between translators themselves in terms of their strategy preferences. It is found that T1 is mostly resorted to the strategy of LT (72) times, while he used the strategy MA and MT just once. This can be seen in the case of T2 who frequently used the S (54) times and never used the strategy of DE and E. However, the strategy of MT was found being equally used by T1 and T2 at percentage of (5.5%). Similarly, T3 and T4 have equally resorted to this strategy at 44.4%) (see Figure 1).

Table (5): Frequencies & percentages of eight strategies used in translating Adjective-noun collocations

Translation strategies

Translator        LT       CE       S     MA      A      DE   MT      E

Fr. % Fr. % Fr. % Fr. % Fr. % Fr. %              Fr.              %              Fr.              %              Total per translator

T 1 72 45.5 28 21.3 20 17.3 1 2.7 4 13.7 2 7.1              1              5.5              5              29.4              133

T 2 40 25.3 16 12.2 54 46.9 19 52.7 3 10.3 0 0              1              5.5              0              0              133

T 3 30 18.9 59 45.0 12 10.4 0 0 13 44.8 6 21.4              8              44.4              5              29.4              133

T4 16 10.1 28 21.3 29 25.2 16 44.4 9 31.0 20 71.4              8              44.4              7              41.1              133

Total per  strategy  158 29.6 131 24.6 115 21.6 36 6.7 29 5.4              28              5.2              18              3.3              17              3.1              532

Total of strategies

532

Fr. = the frequencies of the strategies

Figure (1): The four translators and strategy use

A close examination also reveals that translators combined some of the above-mentioned eight strategies with additional four (Classifier, Transcription, Shifting Adjective into noun, and Footnote). ( see Table 6)

Table (6) shows that translators are frequently resorted to the strategies of C (17) times and TR (16) times.  However, the strategy of FN represented the less frequent strategy (4) times. It is also found that there is a little difference between translators in resorting to the strategy of C. Nevertheless, a variation can be found between translators in using the strategy of TR.

Table (6): Frequencies & percentages of strategies in the sub-scheme

Translator                                                           strategies

C TR SH FN Total per translator

Fr. % Fr. % Fr. % Fr. %

T1 4 23.5 6 37.5 5 27.7 0 0 15

T2 3 17.6 6 37.5 3 16.6 0 0 12

T3 5 29.4 2 12.5 5 27.7 2 50 14

T4 5 29.4 2 12.5 5 27.7 2 50 14

Total per strategy  17 30.90 16 29.09 18 32.7 4 7.2 55

Total of all strategies                                                           55

Below, all strategies will be illustrated in detail by using examples taken from the translators’ renderings with special reference to the similarities and differences between the four translators in terms of strategies use.

1- Literal translation

As has been seen in Table (5), this strategy emerges as the most conspicuous strategy accounting for 158 (29.6 %) of the whole ANCs’ translations. However, resorting to this strategy among the four translators is dissimilar. It accounts for 72 (45.5%) by T1, 40 (25.3%) by T2, 30 (18.9%) by T3, and 16 (10.1%) by T4 of all LT strategies. The use of this strategy is significantly different among the four translators (see table 4).

Below are examples from the data of this translation strategy

SL collocation LT provided

Big fish سمكة كبيره ( big fish)

Great fish سمكة عظيمه (great fish)

High chairs كراسي عاليه (high chairs)

High sun شمس عاليه (high sun)

Small shade ظل صغير (small shade)

Although the collocations of big fish, great fish, and high chairs seems to be acceptable collocations in Arabic, the adjectives عاليه  (high) and  صغير (small) do not sound natural or native-like to be used with  شمس (sun) and ظل  (shade) respectively in Arabic language. Assuming word-for-word correspondence between English and Arabic on the part of translators most likely lead to an unacceptable collocation in Arabic language. This assumption may lead to a negative transfer which results in non- native-like or unnatural translations. It seems more acceptable in Arabic language to use مشرق شمس (bright sun) and خافت ظل (faint shade) for (high sun) and (small shade) respectively.

2- Collocational Equivalent  (CE)

It is noticed that translators replaced some of SL collocations with TL ones. This strategy is the second most used strategy. Overall, it accounts for 131 (24.6%) of the whole strategies. Table (4) shows that the use of this strategy was statistically significantly different among T3 and T2. It most frequently used by T3 accounting for 59 (45.0%) and less frequently used by T2 accounting for 16 (12.2%) of its total number. Both T1 and T4 used this strategy similarly with 28 (21.3%) of its total number.

Below are examples from the data representing the translation strategy of Collocational Equivalent.

Source language collocation  CE provided

Great mistake   غلطة شنيعه (awful mistake)

Small breeze    نسيم لطيف   (Gentle breeze)

The greater speed سرعة فائقه  (supreme speed)

Little hope أمل ضعيف   (weak hope)

Great hunger جوع مدقع (miserable hunger)

3- Synonym- or near-synonym (S)

This strategy scored (21.6%) with the frequency of 115 times of all translations. As can be seen in table (5), T2 has resorted to this strategy 54 times, (46.9%) of his strategic choices.  This means that he was noticeably different in the use of this strategy compared with T1, T3, T4. This strategy accounts for 20 (17.3%) with T1, 12 (10.4%) with T3, and 29 (25.2) with T4 of their overall strategies.  The use of this strategy was also statistically significantly different among the four translators (see table 4).

The following are some examples of the use of this strategy

Source language collocation  S provided

Great fish سمكة كبيره (big fish)

Little line    خط قصير (short line)

Big shark  قرش ضخم   (huge shark)

Short club هراوة صغيره  ( small bat)

Great help مساعده كبيره  (big help)

4- Missing Adjective (MA)

It is noticed that translators produced only part of the meaning of the source language collocation which led to the production of an incomplete message in some cases. This strategy is used only once by T1 and is not used by T4. However, it accounts for 19 (52.7%) with T2 and 16 (44.4%) with T3 of their strategic choices. (see table 5 and Figure 1)

Below are examples from the data representing this strategy

Source language collocation  MA provided

Small bird  طير ( bird)

Great negro    زنجي  (negro)

great island جزيره  (island)

Huge jaws فكين  (two jaws)

Big cord حبل (cord)

5- Avoidance (A)

It represented the fifth option that translators generally resorted to, accounting for 29 frequencies (5.4%) of the whole ANCs’ translations. However, avoidance was resorted to in only a few cases by T1 with 4 (13.7%) and T2 with 3 (10.3%). Differently, it scored a higher frequency of 13 (44.8%) by T3 and 9 (31.0%) for T4 of their total strategies.

6- Descriptive equivalent (DE)

This strategy is identified in this study where the translators resorted to using several words to explain the source language collocation instead of providing an exact collocation in the target language. However, it seems that the use of more words in the TL compared with the words in the ST may not make comprehension easy nor facilitate production of the target text because it sometimes lead to complication of sentence structure in TL. Descriptive equivalent comprised only 28 (5.2%) in the data, almost twice as many by T4 as by T1, T2, and T3 altogether (see Table 5).

Below are examples from the data representing the translation outcome of Descriptive Equivalent

Source language collocation  DE provided

Great temptation  اشياء كثيره اغرته  ( a lot of thing that tempt him)

Great trunkbacks   السلاحف العظيمة ذات الظهور الصندوقية   (the great turtles that have backs similar to trunk)

Short time  برهه من الزمن  (short period of time)

high capes الرؤوس البريه العاليه الداخله في البحر (the high land capes which entered in the sea)

Big leagues  المباريات الكبيره للبيسبول (Big matches of Baseball)

7- Mistranslating (MT)

This strategy accounted for 18 (3.3%) of the translators’ total renderings.  As can be seen in table (5), there is a remarkable difference between T1 and T2 on the one hand and T3 and T4 on other hand in the use of this strategy. It accounted for 1 (5.5%) and 8 (44.4%) for the two groups respectively.

The following examples illustrate the way meanings were distorted as a result of the translators’ mistranslation into Arabic of some of ANCs under study.

Source language collocation  MT  provided

Small bird   عصفور  ( sparrow)

Longer day يوم لاينتهي  (endless day)

His great dignity عظمتها  ( her greatness)

The great drives الابطال الراسخين ( the firm heroes(

Great bursting  وثبة هائله (huge jump)

8- Encapsulation (E)

This strategy was the least employed one in the whole translations accounting for 17 (3.1%). Recognizing encapsulation as one of the translation strategies that was  used in translating ANCs  into Arabic by three of the professional translators in this study may be inconsistent with what (Brashi 2006) assumes: it is not possible to encapsulate an English adjective + noun collocation into just a single word in Arabic. It is quite clear from table (5) that three of translators resorted to this strategy with roughly the same rate, 5 (29.4%) for T1 and T3 and 7(41.1%) for T4. However, it is noted that T3 did not resort to this strategy in his translations.

Examples of Encapsulation are presented below.

Source language collocation  E  provided

Big school of dolphin سرب من الدلافين  (group of dolphin)

Great canyon of clouds   سلسلة من الغيوم  (series of clouds)

A vast school of porpoise  سرب من الخنازير  ( group of porpoise )

Small drink of water جرعة  من الماء   (little of water(

Having a cursory look at figure (1), some points can be raised regarding translators performance of strategies use in general.

1- Comparing with the other strategies, LT, S, and CE represented the higher rate of the general translations of T1, T2,T3 respectively.

2- All strategies used by T4 with roughly approximate rate.

3- Strategies of DE and E were not used by T2 in all translations. Likewise, T3 did not resort to strategy of MA in his translations.

4- In all translations, strategies of LT (158), M (131), and S(115) had higher rate of use than the other strategies.