Vous êtes sur la nouvelle plateforme d’Érudit. Bonne visite! Retour à l’ancien site

Saudi Students’ Translation Strategies in an Undergraduate Translator Training Program

  • Omar Atari

…plus d’informations

Corps de l’article


The growing interest in translation within the Arab world has manifested itself in the emergence of translator training programs, both at graduate and undergraduate levels in several Arab universities (e.g. Saudi Arabia, Morocco, Jordan). This newly-emerging trend has, as a matter of course, resulted in an upsurge of research into translation studies. The major thrust of this research has been product-oriented. Therefore, it exhibits the features of contrasting the target text with the source text (Henceforth TT and ST).

One major strand of these product-oriented contrastive studies attempted, and rightly so, to base its analysis on the interaction between micro-text and macro-text processing. Emery (1991:129-137), for instance, acknowledges that both macro-dimensions (discoursal, textual, pragmatic and semiotic) and micro-dimensions (syntactic, lexical) of the translation process are inextricably intertwined and interdependent. (Cf. Farghal, 1995; Menacere, 1992; Emery, 1989; Shunnaq, 1999; Aziz, 1985; Atari, 1994).

The second strand of these product-oriented research studies has taken up the issue of shifts in tenor and register in the target texts (Henceforth TT). For instance Farghal and Shunnaq (1995: 208) addressed the tenor-related problems in the translation of legal texts into Arabic. (Cf. Shaker 1995; Asfour, 2000; Farghal and Naji 2000; Ali, 1988; Sa’Adeddin, 2000, among others).

The Hatim and Mason model (1990) represents a third strand of analysis. It proceeds according to a clear vision of the linkage between product and process that seeks to capture the intricate dialectic underpinning of genre, discourse and text. Hence, they present a tripartite framework for the analysis of the communicative, pragmatic and semiotic dimensions of texts.

It should be noted here that this brief review of the relevant product-oriented studies with reference to English and Arabic is by no means exhaustive. It provides a preliminary account of the directions taken by researchers into translation studies that address the issues related to English and Arabic.

This review of published studies demonstrates the dirth of process-oriented research. To the best of my knowledge the only study that could be classified as a process-oriented piece of research is the one by Khanji et al. (2000:448-557) in which they classify the strategies used by Arabic-speaking translators’ strategies in simultaneous interpretation.

As translation teachers and researchers, we must be interested in process-oriented research that takes account of the mental processes and transformations embedded within the act of translation. So far, there has been a severe shortage of this type of research. This paper seeks to provide a counter weight to this imbalance. It examines Arab translator trainees’ strategies of translation through the use of the think-aloud technique.

This paper will describe the translation strategies employed by a sample of under-graduate Arabic-speaking translator trainees (i.e. Saudis) engaged in the act of translating from English into Arabic. Thus, the paper shall examine not only the strategies used frequently, but those strategies that are not used. In its conclusion, the paper will pursue the implications of our findings and offer recommendations for the teaching methods of translation.

Method of Data Collection: The Dialogue Protocol

As a reaction to the limitations of the product-oriented approach, the think-aloud protocols have been used to reveal aspects of the translation process. The use of think-aloud reveals both the strategies and the mental processes associated with them. (Cf. Krings 1986; Lörscher, 1986, 1991; Kussmall, 1995; Börsch, 1986).

Due to the danger of influencing the subjects’ responses during the monologue think-aloud (i.e. the subject is trying to inform the instructor/researcher what is going on inside the head while sometimes being prompted by the researcher), I have employed a dialogue between pairs of translator trainees. This approach groups translator trainees into pairs, each member of pair was asked to translate from English into Arabic individually. During the act of translation the individuals were encouraged to interact with each other. Their enquiries, comments, etc. were tape-recorded.

Framework: Translation Strategies

Lörscher (1996: 27-28) defines translation strategies as procedures that the subjects employ in order to solve translation problems. He adds:

“…accordingly, translation strategies have their starting-point in the realization of a problem by a subject, and their termination in a (possibly preliminary) solution to the problem or in the subject’s realization of the insolubility of the problem at the given point in time. Between the realization of a translation problem and the realization of its solution or insolubility, further verbal and /or mental activities can occur which can be interpreted as being strategy steps or elements of translation strategies.”

Lörscher’s model consists of the following:

Table 1

-> Voir la liste des tableaux

In addition to these original elements of translation strategies, Lörscher (1996:28) proposes other potential elements which consist of:

-> Voir la liste des tableaux

Krings (1986:267) offers the following eleven features for a model of translation strategies:

Table 2

-> Voir la liste des tableaux

Gerloff (1986:253) offers the following model which comprises features similar to Krings’ and Lörscher’s. Her categories combine ST comprehension strategies and TT production strategies. These are:

Table 3

-> Voir la liste des tableaux

In brief, the aforementioned models of translation strategies are quite comparable with each other. They all attend to ST reading comprehension strategies and production strategies despite the variance in their use of terminology. For purposes of my study, after, preliminary examination of my subjects’ dialogue protocols, I chose to use Gerloff’s model (1986:274-262) with a few refinements of her terminology. Specifically, I chose the label “Monitoring of ST segment” instead of her “General Search and Selection” strategy. In addition, I used the category “Monitoring of TT segments” instead of her “Editing.”

Results / Observations

1. General Observations

I would like to point out here that the subjects sometimes tend to employ a variety of strategy types while handling the same ST or TT segment. Specifically, within the context of solving one translation problem, subjects tend to employ more than one strategy – often with a cluster of strategies. When a translation problem is being identified, subjects employ strategies of ST processing for comprehension such as re-reading, repeating pronunciation of a problematic ST segment, giving alternatives or tentative meanings for it in either the ST or TT. Thus, they infer a tentative selection of meaning choices and do immediate self-corrections. Then and almost simultaneously, they check the proposed TT segment (i.e. product) by constructing explanatory context either on the basis of personal experience, world knowledge or sentence context. They tend to go back to a ST segment in the preceding sentence or phrase to ascertain the appropriateness of their proposed TT segment. Yet, they quite often leave the proposed TT segment unchecked in order to move forward to handle a new ST segment thinking it will provide more insights into what they have already proposed.

The course of the translation process is actually spiral, comprising a tripartite hierarchical structure using three strategies: ST comprehension strategy leading to a prospective TT segment that is checked and re-checked through the employment of a TT monitoring strategy based on inference and reasoning. Lörscher (1986:287) refers to this non-linear, discontinuous translation process course as:

“The translational process of problem solving which manifests itself largely as a retrospective-prospective process can thus be compared to a chain of spirals. Although it generally proceeds in a prospective way, what proceeds is not of a purely linear kind. Rather it can be seen as a chain of loops with both retrospective and prospective elements.”

2. Specific Observations

In the following section of the paper I report on three elements: the number and type of strategies used, the most frequently used strategies and their substrategies and third, the least frequently used strategies and their substrategies. These results direct us as teachers of translation to the relevance of the strategy types that are neglected, in the hope that we start to pay special attention to them in our classroom translation teaching. The strategies most frequently used, on the other hand, will be looked at as another pointer to their relevance or irrelevance of ST reading comprehension and TT production (i.e. the two primary phases of a translation act).

Chart 1

Saudi Student’s Use of Translation Strategies

Saudi Student’s Use of Translation Strategies

-> Voir la liste des figures

Table 1

Saudi Students’ Use of Translation Strategies: Frequence of Use of Strategy Type and Substrategies

Saudi Students’ Use of Translation Strategies: Frequence of Use of Strategy Type and Substrategies

-> Voir la liste des tableaux

Table 1

Saudi Students’ Use of Translation Strategies: Frequence of Use of Strategy Type and Substrategies

Saudi Students’ Use of Translation Strategies: Frequence of Use of Strategy Type and Substrategies

-> Voir la liste des tableaux

As for the first element, it is obvious that the subjects in consideration did use all strategies of both the ST reading comprehension and the TT production. Specifically, the subjects in this study did employ all six strategy types proposed by Gerloff (1986). Also, the strategies they employed cover more or less all strategy types proposed by scholars such as Lörscher (1992, 1996); Krings (1986); Seguinot (1989); among others).

This finding should not surprise us, given the subjects’ academic background. These are all graduating students (i.e. enrolled in the last semester of a five-year translator training program at King Saud University, College of Languages and Translation). In particular, these subjects have had extensive translator training in a whole range of fields: Social Sciences, Islamic Studies, Scientific, Technical, Legal and Commercial fields. Moreover, the subjects have had several linguistics courses, namely, Semantics, Text Typology and General Linguistics. We may presume, therefore, that these subjects must have become more or less sensitized to L2 reading comprehension strategies and translation-specific strategies during their translator training courses, albeit in an imbalanced fashion, as will become apparent in the subsequent discussion of these findings.

The second finding is more revealing in the sense that it shows not only the most frequently used strategies but also the most frequently used substrategies of each, respectively. The most frequently used strategies of problem-solving have been the following: Monitoring of ST Segments, Monitoring of TT Segments.

This finding shows that ST reading comprehension and TT production seem to be the foci of these subjects’ translation endeavors. It is only natural that most of the subjects’ endeavors are centered on the interpretation of ST and the production of TT. More importantly, though, is the variance in employing certain substrategies within these categories, to the neglect of the other substrategies in the same overarching strategy.

The preceding chart and table (1) show that within the most frequently used strategy of ST monitoring, two or three substrategies were used by the six subjects whereas the others were not. The substrategies most frequently used were: Repetition of linguistic units at level of the word, or morpheme levels, re-reading ST segments, constructing tentative meanings of ST segments in ST language and/or in TT language. Within this strategy (i.e. Monitoring of ST segments) the six subjects did not use the other substrategies: giving synonyms, comparing the two languages, the use of fillers or skipping words.

The strategies used most frequently within the category of Monitoring of TT strategy were: immediate correction before writing and congruity assessment whereby the subject checks to see if the translation makes sense, maybe before writing the product or after. Unused substrategies within this same strategy (i.e. Monitoring of TT Segments) were: punctuation check and product quality assessment.

It is obvious that the subjects’ tendency to dwell on the word, morpheme, phrase and to a lesser extent a whole sentence through the repetition of these mini-text segments, re-reading them and giving tentative meanings in SL and TL, reflects the extent of the difficulty they have with bottom-up, language-based text processing strategies.

The findings listed below corroborate the preceding observation concerning strategies used least frequently and their substrategies (i.e. macro-text knowledge-based strategies. These are:

  1. Inferencing and Reasoning >

    • Use of text structure

    • Use of world knowledge

    • Use of personal experience

  2. Storage and Retrieval

    • Memory search

  3. Text Contextualization

    • Use of paragraph context

    • Use of larger context

    • Reference to author intent

All of the above-named strategies and their substrategies, that either have not been employed or used minimally, constitute the top-down, knowledge-based strategies for text processing. This observation provides further evidence that the subjects are entangled in text processing that is predominantly language-based and “bottom up.” Hence, there is an apparent imbalance in their overall strategy use. The balance is tilted towards one set of strategy types at the expense of another equally-important set of strategies. Finally, I must note with some surprise the complete absence of employment of memory search. I posit two reasons for this: the subjects’ overwhelming entanglement with the language-based translational problems did not leave enough room for reflection. Secondly, the research design itself may have had an impact. The use of the dialogue think-aloud method of data collection encouraged the paired translators to maximize their collaborative interaction at the expense of private reflection and memory search.


Based on the preceding discussion and analysis of data, it is evident that the translator trainees in this study dwelt on language-based, bottom-up processing of the ST. They tended to extract smaller units of the SLT segments and focus their attention on them in order to render the TL text. They chose single words and at best short phrases.

The major problem for these translator trainees is obviously of a local kind, especially lexical transfer; this seems to be a reflection of an inadequate level of competence in SL, especially. Furthermore, and based on the findings that these trainees’ translations are dominated by the lexicon and to a certain extent by minimal syntactic units, they do not check the utterances produced in TL with regard to their stylistic and text type adequacy.

These conclusions point to the following: classroom teaching has to incorporate exercises that will confront the translator trainees with the compelling need to utilize the paragraph context, the text structure and their readily acquired previous knowledge. This could be achieved if translator trainers supplement their translation assignments (i.e. SL texts) with a battery of questions that will prompt their trainees to utilize the larger contexts. In other words, translation teachers should transcend their traditional classroom feedback on their students’ translations to incorporate the text design, text structure, paragraph structure, and semantic/pragmatic relations that constitute the intricate web of relations underlying whole chunks of utterances, instead of single sentences in isolation.

Parties annexes