The Unguarded Use of Stylistic Features and the Arabic Discourse Marker ‘Wa’ in Media Translation of News Reports

This research paper is a process-oriented study that essentially focuses on the psychology (Mapping Theory) of the trainee translator by examining the use of some stylistic features and the interpretation of the Arabic discourse marker wa into English in news reports. The purpose of this study is to investigate the extent to which students of undergraduate translation courses at Princess Alia University College / Al-Balqaa' Applied University consult and are largely influenced by their SL native language (Arabic) in translating news reports into English (TL). The random sample of the study consists of (46) female students with similar linguistic, socio-cultural and educational backgrounds. Trainee students underwent a task of pre-test and post-test translation of a news report during the summer semester of the academic year 2013-2014. A three-step analytical model was applied in this study. The first step of the model involved segmenting the target text into paragraphs and sentences. The second step was the description of the functional relations that connect units of the target text at each level. The last step involved identifying the discourse markers at the boundaries of units. Obtained results revealed that students showed conspicuous bias towards Arabic (SL) in using some stylistic features and the discourse marker wa (and) as a safe strategy for linking sentences in English (TT), even though done improperly.


Introduction
Translation as a process is concerned with the psychology of the translator, or more aptly, it is concerned with trying to find out what happens in the mind of a translator when he/she is engaged in a translation activity.It is the branch of translation studies that S. Holmes presents in his influential paper The Name and Nature of Translation Studies that serves as the founding statement of translation as an independent and distinct discipline.In his paper, he crucially puts forward a framework that delineates what translation studies cover.Areas of research are divided into pure and applied studies.The branch of pure research further branches off into theoretical and descriptive studies (DTS), which has three possible foci: product-oriented, function-oriented and process oriented.Being a mentally-induced process, translation involves shifting focus constantly between microanalysis and macroanalysis of source text (ST) and target text (TT), i.e. the mind continually compares between the sense of individual utterances and the overall sense of the text as a whole, either consciously or unconsciously, forming a mental representation called the "intertext".As such, the "intertext" is viewed as a composite that consists of intertextual relations where it is located.One such relation is cohesion, which is obtained from connecting segments in texts, whether among sentences, paragraphs or portions within the same sentence by means of the employment of connectives referred to as discourse markers (DMs).
Languages differ noticeably as to the frequency their linguistic systems tolerate using discourse markers to reflect textual logical relations created.The Arabic language structurally and intrinsically relies heavily on the explication of textual relations by resorting to using a battery of discourse markers referred to as adawaat-u l-rabt ṭ or ḥuruuf al-9aṭf-, i.e. connective particles (Tahineh, 2011, p. 226).The structure of English, comparatively, allows the employment of cohesive devices, but to a lesser extent than Arabic (Newmark, 1982, p. 178).The judgment launched by a discrete translator is not unheedful of the type of text being processed.Texts vary in the linguistic choices and stylistic features they offer even within the same register, such as journalism.An opinion article or an editorial argues a point, while a news report leans towards being informative, in which case the chief function is simply to inform facts and describe situations.Different in orientation, a news report differs in its linguistic reflections of extralinguistic factors demonstrated in the economical use of connectives, terser (and /or) shorter sentences and punctuation among other things.

Problem Statement
Experience in teaching translation courses for Arab students majored in English language and literature seems to suggest that discourse markers present a hurdle to students when translating texts into English.It has been observed that translation trainees at these courses find it difficult to discern the proper usage of both Arabic and English discourse markers although the notion of forming a coherent and well-organized text might be quite accessible to most of them, if not all.The extent of the likeness and difference in terms of the peculiarity of the structure of both these two languages determines the dimension of the problem.Students also exhibit quite similar difficulty in realizing that there is a peculiarity for each text type even within the same register, namely journalism.Regrettably, translation trainee students seem to resort to consulting the same system of uniform strategies when dealing with journalistic sub-genres, like news reports; the product of their translation (TT) is stripped of its basic features that distinguish it as a news report.Ultimately, it becomes difficult to discern them from other sub-genres within the same register.A good translator is one who respects the craft of writing by accounting for the text's language, structures, content, as well as any other extralinguistic factors at work.Any violation of these maxims would certainly result in distorting the meaning, which is the prime priority of translation.

Questions of the Study
1) What concepts do Arab trainee students seem to hold regarding the use of cross-lingual discourse markers in general and the discourse marker wa (and) in particular?
2) What are the mechanisms that should be followed to preserve the stylistic features, which are distinctive of English news reports?
3) What are the implications for a more controlled employment of the discourse marker wa (and) in Arabic-English media translation of news reports?

Significance of the Study
Examining the way translation students perceive the functions of discourse markers in general and the discourse marker wa in particular owes its significance from the fact that it helps to: 1) provide guided application of theoretical principles and techniques learned 2) integrate theory and practice under stimulated translation situations 3) call the active players in the translation process, trainers and trainees alike, to be particularly attentive to the specificity that each text type possesses.Trainers in translation courses must sensitize trainees to the importance of preserving the stylistic features of text types

Theoretical Background
Translation is a process that consists in "the attempt" to replace one "written message" in the Source Language with another in some other language or languages (Newmark, 1982, p. 7).Underlying this definition is the assumption that this process involves some loss of meaning since languages differ essentially not in what they may (meaning) convey, but rather in what they must (structures, forms) convey.The problem of meaning and equivalence, as some linguists like Jakobson observe, is concerned with differences in the structures and terminology rather than in any inability of a language to render a message into the other language (Munday, 2001, p. 37).This provokes a dialectic tension for trainee students based on the claims and argument each language presents.The basic loss verges on a continuum from over-translation, increased details, to under-translation, increased generalization (Newmark, 1982, p. 7).
Relatively, this loss of meaning may be compensated for based on the interrelation of global and local elements constituting a text as prescribed by Peter Newmark (1982, p. 15), who would set out prescriptions for the translation exercise depending on the text type.An informative or expository text, for that matter, should focus on the target language in which the loss of meaning should be small, while the translation should show slightly greater length when compared to the original.
The classification encompasses all texts that adhere to the informative or expository function.Thus, there would be "considerable use of the third person, past tenses, multi-noun compounds and passives" in a scientific report (Newmark, 1982, p. 15).What is even more, some figures of thought expressed through unusual metaphors and comparisons in the ST should be "reduced to their sense if the text has a mainly informative function".The main focus of the informative function is to describe an "external situation, reality outside language, including reported ideas or theories" as the students in a translation task are required to do.These texts, typical informative texts, are concerned with "any topic of knowledge".Function-oriented, they are at cross purposes with texts that deal with literary subjects "as they often express value-judgments".They tend to lean towards "expressiveness" (Newmark, 1988, p. 41).The general makeup for an informative text is "standard: a textbook, a technical report, an article in a newspaper or a periodical, a scientific paper, a thesis, minutes or agenda of a meeting.One normally assumes a modern, non-regional, non-class, non-idiolectal style, with perhaps four points on a scale of language varieties: (1) a formal, non-emotive, technical style for academic papers, characterized in English by passives, present and perfect tenses, literal language, Latinized vocabulary, jargon, multi-noun compounds with 'empty' verbs, no metaphors " (Newmark, 1988, p. 40).Concerned with extralinguistic situations, informative texts consist of "third person sentences, non-emotive style, past tenses.Within this functional framework, narrative texts relating a sequence of events are "likely to be neater and closer to translate than description, which requires the mental perception of adjectives and images" (Newmark, 1988, p. 50).
The significance of a text's macro structure gains in momentum in translation and elsewhere, for according to the German theorist Nord (2001, p. 38) "text type classification sharpens the translator's awareness of linguistic markers of communicative function and functional translation units".In their proposed comprehensive model for translation, Hatim and Mason (1990, pp. 140-243) define text type as "a conceptual framework, which enables us to classify texts in terms of communicative intentions serving overall rhetorical purposes".Rhetorical purpose is defined as "the overall intention of a text producer, as instantiated by the function of a text, e.g. to narrate, to counter argue".According to this typology, texts are classified into three major forms; expository texts, argumentative texts and instructional texts.Under expository texts, three sub-types are identified: "conceptual" dealing with the analysis of concepts, "narrative" focusing on actions or events, and "descriptive" which describe objects and situations.Expository texts assume an informational role in which the focus is on providing a detached account or a description of the events or objects being presented.
The translator's ability to recognize a text as a "token" of the text type will enormously color his/her comprehension as well as rendition into the target language.He/She must respect the text by accounting for its coherence-forming forces like language, structures, content, elements, chunks and text type (Newmark, 1982, p. 6).The text type alerts the translator to consider the way in which the overall discourse relations (local and global, or micro and macro) evolve into a coherent whole.
Journalistic language not only encompasses lexical features, but is also characterized by distinctive "functional, lexical and structural features" (Abdelfattah, 1996, p. 129).In general, newspapers adhere to what is classified by linguists as the "expository" writing style in journalism.This text-typology is recognized by Hatim (1997) as criterial for identifying the global structure of texts.
In line with the journalistic profession of news report writing, its style is characterized as "simple, concise, clear, consistent, objective, easily digestible, impersonal, precise, up-to-date and with a sense of immediacy and to the point".The word choice should be common, with a preference of everyday words which are generally the most effective ones for telling a news story.If the writer opts for a word or a term that is "abstruse" or obviously unfamiliar to most readers, he/she should either define the expression or put it in a context that defines it.Sentences should be straight-forward and generally short.The most trusted vehicle to convey an idea is the simple declarative sentence.Normally, sentence order used is (1) the subject (2) the finite verb, and (3) indirect object and direct object.However, the writer has a license to deviate its order if he/she wants to emphasize an element otherwise (passive voice, movement of adverbial).An average sentence length is about twenty words.The average lead sentence, however, is necessarily longer; a desirable length is no more than twenty-five words.
Paragraphs tend to be brief too, often consisting of a few sentences.(Readers have increased attention if ideas are broken up into short, tight segments).The Interest of readers has to be maintained through many stylistic features, among them are short paragraphs and clearly expressed sentences.The choice of vocabulary and grammar will influence the type of sentences that should predominantly be narrative statements (rather than questions or exclamations).Among the canonized style guidelines idiosyncratic of newspaper reports as dictated by The Economist Style Guide (1997, pp.5-7) are: 1) Do not be too chatty.This particularly contrasts sharply with the Arabic newspapers style which is rife with redundant statements characteristic of repetition, 2) In general, be concise.Remember Voltaire's saying "The best way to be boring is to leave nothing out".
3) Do your best to be lucid.Simple sentences help.Keep complicated constructions and gimmicks to a minimum.
4) Long paragraphs, like long sentences, can confuse the reader.
The journalistic style of Modern Written Arabic is one style among the extensive range of discourse styles that Modern Standard Arabic language covers (Ryding, 2005, p. 8).This style uses clear, simple and straight to the point language distinctively far from the recourse of sophisticated expressions and the complexity of syntax.This style shows flexibility "in adopting new structural and stylistic features" to cope with the changes that its counterpart adopts in the English language (Abdelfattah, 1996, p. 130).Gully (1993, p. 20) points out that this style is "so readily open to change that it assimilates new types of expression and grammatical constructions with varying degrees of ease".
On a macro-contextual level, the word "connectivity" refers to the connection established between the writer, reader and the text, and the translator acts as a mediator in the transference process.In this connection, Holmes (1988, p. 96) has proposed a mental approach called the "mapping theory".He states: I have suggested that actually the translation process is a multi-level process; while we are translating sentences, we have a map of the original text in our minds and at the same time, a map of the kind of text we want to produce in the target language.Even as we translate serially, we have this structural concept so that each sentence in our translation is determined not only by the sentence in the original but by the two maps of the original text and of the translated text which we are carrying along as we translate.It follows, then, that the verbal processing, having been established as a process, undergoes simultaneous, interdependent and holistic processes constituting collectively one structure, a complex, that is.
At sentence boundaries, discourse markers are classified as follows: additive, contrastive, explanatory, inferential, sequential, alternative, exceptive, background, subjective, and interactive.At paragraph boundaries, discourse markers are found to serve three main functions: continuity, refocus and change of topic.

(Wa) in Arabic
Discourse markers in Arabic are regarded as a prevalent feature.Neither a sentence nor a clause within a text is absent from being coordinated or subordinated by means of a discourse marker.The tendency of using connectors or discourse markers in Arabic plays a crucial role in the cohesion of the Arabic text and contrasts sharply with the condensed written English (Ryding, 2005, p. 407).Textual coherence is also sustained through strings of paragraphs connected together by connectors at the beginning of sentences.Discourse markers serve functions at different textual levels: phrase, clause, sentence, and paragraph.They are divided into simple and operational connectives depending on the influence they exert on the following sentence, clause or phrase.Of these simple connectives in MSA that possess a special status is wa 'and'.Wa is unarguably of the highest frequency of all discourse markers (almost 50 percent of all Arabic connectives) and occurs at all levels of text, to "signal an additive relationship" (Al-Batal, 1990, p. 245).
In a comparative study on the functions of wa and its English formal counterpart, and, Fareh (1998) identifies and discusses nine semantic relations that and serves, namely, consequence, sequence, contrast, simultaneity, concession, condition, addition, explanation and comment.Significant of this study is the conclusion that the relation between Arabic wa and (and) is definitely not a one-to-one correspondence.Wa shares some of these semantic relations with and in which case it is acceptable to provide it as a valid linguistic substitute.However, the Arabic wa is employed to express logical relations that hold among various elements in the text where and would certainly not be used in the process of translation.Based on the classification of Arab grammarians and rhetoricians, wa performs functions to express relations of resumption, as well as adverbial wa, swear wa, wa meaning by/ along, wa for choice, redundant wa, wa to introduce praise or admiration and wa to introduce threat/ underestimation/ contempt.See appendix B.

Review of Literature
The language of media in general and journals in particular has long been attracting linguists and researchers of linguistic interest.This comes as no surprise as the mass media present audience with a wide range of linguistic varieties more than any other field of communication.For example, the newspaper includes an array of sub-genres like news reports, editorials, articles, reviews, letters, captions, headlines, advertisements, sports news and crossword puzzles, to name only a few.As such, language is an intrinsic part of the content conveyed by different media channels.
By its interdisciplinary character, translation is at the direct intersection with stylistics (Newmark, 1982, p. 17) and register.As defined by Richards et al. (1985, pp. 277-278), style is a variation in a person's speech or writing.The language of the media, especially the press, presents a wide range of variety to be an area worthy of investigation.Crystal (1987, p. 388) argues that there are common superficial similarities shared by newspaper styles arising out from the nature of the press in spite of the fact that stylistic preferences vary a great deal.According to him, newspapers and magazines do their best to excel not only in the quality of analysis and presentation of news, but also in the quality of writing.Among the characteristics that commonly represent the language of the press are the guidelines issued by The Economist Style Guide (1997, pp.5-7) that require the use of short words, unique metaphors, the active voice, an everyday word instead of a foreign phrase, use of everyday language, avoiding redundancy, avoiding being didactic, avoiding the overdo of contracted forms, conciseness, lucidity, and most importantly using short paragraphs and short sentences.News reports rank among the simplest forms of sub-genres that the press media includes (Harris & Spark, 1966).In spite of their lucidity, translation trainee students still confuse the proper use of such stylistic features.
Despite its richness, little has been written on investigating the inescapable influence that the SL exerts on trainee translators in the field of journalistic translation.Even more scarce has the discussion been with reference to sub-journalistic genres and text types like news reports.Munday (2001, p. 11) comments on the scarce research carried out in the area of descriptive translation studies (DTS) "Despite some later work on think-aloud protocols (where recordings are made of translators' verbalization of the translation process as they translate), this is an area of research which has still not yet been systematically analyzed".Holmes (1988, p. 96) has proposed a mental approach called the "mapping theory" describing translation as a process in which the translator's verbal processing of the text undergoes simultaneous, interdependent and holistic processes constituting collectively one structure, a complex, that is.During the process of translation, the translator's mind acts like a processing unit that supervises multi-level activities of reading, processing and writing, and most importantly projecting the map of the product, namely the TT.
In the same vein, Hönig (1991, p. 78) maintains that while translating the source text (ST), the translator "moves out" the text of its natural context and deals instead with its mental projection on his/her mind.According to him, a translator applies two kinds of processing, "the controlled and the uncontrolled "processing, in which the latter takes place as the translator applies semantic patterns based on the perception of his/her experience.As processed patterns, discourse markers are essential tools that function to connect textual segments that ultimately achieve the text's coherent value.Processed in the mind of the translator like other frames, they take place either at the controlled or the uncontrolled workspace, that is, "the conscious or unconscious workspace".Siepmann (2005) states that discourse markers are language strings that function by signaling the coherence among linguistic units in order to facilitate the receiver's processing task.
Documented in the literature regarding the difference between Arabic and English is the noticeable disparity in the use of discourse markers.The interest of this type of connectivity in its relation to translation is discussed by Baker (1991) in her coursebook on translation with a contrastive approach of cohesive devices in Arabic and English.
While English prefers to present information in small chunks to mark semantic relations between paragraphs and sentences, Arabic, on the other hand, tends to present the information in noticeably large chunks (Baker, 1991, p. 192).In addition, English relies on "a highly developed punctuation system" into which Arabic has been recently introduced," this is partly because punctuation and paragraphing are a relatively recent development in Arabic" (ibid, p. 193).Original Arabic texts, as Baker observes, "do not normally display these features, "which are associated with translated Arabic texts".These features are: "short sentences, a varied array of conjunctions, and absence of the typical conjunctions (wa, fa, and a few other particles).Arabic discourse markers that occur sentence-initially to connect sentences, because they are semantically and grammatically optional, are loosely attached to their host sentences.In addition, they have a cross-sentential function, "linguistic clues", that serves to guide the reader's interpretation to the meaningfulness of the text (Kohlani, 2010, p. 1).Halliday and Hassan (1976), throughout their thorough investigation of the devices that contribute to building the text cohesion, were able to identify five cohesive devices (reference, substitution, ellipsis, conjunction and lexical cohesion).A conjunction is classified as located on the "borderline of the grammatical and the lexical levels since some conjunctives can be interpreted grammatically whereas others involve lexical choices" (Halliday & Hassan, 1976, p. 303).
The most frequently used discourse markers that prove to have a static record in currency in Arabic discourse are wa "and" (almost 50% of all markers) (Al- Batal, 1990, p. 245;Ryding 2005, p. 409) and fa "then".Problems identifying the functions of discourse markers are due to the multiplicity of the functions of these two markers as one marker may signal multiple functions between sentences (Fareh, 1998, p. 304).The functions of and in the English language have been discussed by many linguists (Halliday & Hassan, 1976, pp. 226-273;Dijk, 1977, p. 58;Stubbs, 1985, pp. 77-80;de Beaugrande & Dressler 1981, pp. 71-81;Quirk et al., 1986, pp. 930-934;McCarthy, 1991, pp. 48-49).They agree that and is used to signal relations of: consequence, sequence, contrast, concession, condition, addition, explanation, similarity and simultaneity.Linguists in the Arabic discourse followed a different route in identifying the functions of discourse markers.They paid little attention to their role as text-building devices and focused rather on their functions from a structural perspective.Structurally, discourse markers, the so-called adawat "particles", in Arabic written discourse were classified into classes in accordance with the syntactic properties they possess (Fareh, 1998;Abdel Hameed, 1965;Anees, 1966;Ansari 1958;Hamad 1984).

Hypothesis of the Study
Translation students are largely influenced by their native language, namely Arabic, in translating stylistic features and the Arabic discourse marker wa (and) in English news reports, as they believe it is necessary to reproduce the same Arabic forms of style and wa (and) when translating from Arabic (ST) into English (TT).

Method
The study draws on the analytical model employed by Kammensjö (2005) and Kohlani (2010) as a pattern in segmenting the text into parts, though with slight modifications to better serve the purpose of the study.The model consists of three steps: segmenting the texts into units, describing the functional relations among these units, and identifying the discourse markers at unit boundaries (Kohlani, 2010, p. 12).The first step proceeds in a top-down analysis with segmenting the uppermost unit of structure analysis, the text, into two levels of text structure, the paragraph and the sentence.This course of segmentation has as a goal determining the unit boundaries at which discourse markers, like wa, occur.The second step in this model is to identify functional relations between paragraphs and sentences.At this point, the two steps make it helpful to identify discourse markers at sentence and paragraph boundaries.Based on that description, it proceeds to describe the function of discourse markers as located in their environment in these texts using text-type theory.Both of these relation-based frameworks represent the text structuring relations from the perspective of the text-producer, taking intended communicative goals as the means for interpreting these relations.Therefore, these descriptive tools provide a semantic / pragmatic description of the environment in which discourse markers occur.

Instrument
The instrument employed in this study is a translation test of a local news report published in Al-Rai daily newspaper ‫اﻷردﻧﻴﺔ‬ ‫اﻟﺮأي‬ ‫ﺻﺤﻴﻔﺔ‬ on February 19, 2006.The newspaper is reputed for its high circulation, distribution and coverage of local, national and international events.This reflects its importance as a wide-ranging and accessible newspaper to the average Jordanian reader.Moreover, the article that students are required to translate is of the journalistic variety, simple and straightforward in its presentation as compared to other varieties like editorials or opinion articles which employ the argumentative and evaluative style.The nature of the argument in opinion articles involves ample and frequent use of various discourse markers (Kohlani, 2010).The style of the news report is direct and keen on presenting facts and events in a lucid and concise manner.This in itself helps eliminate the degree of ambiguity in the sentences of the text that students are tasked to translate.Furthermore, it helps in the consistency of the result of the analysis; the researcher configures the mental scheme and mapping processed in the mind of the translator to better understand the issue.
Students were pre-and post-tested.The pre-test sample constituted the data for the control group without having received any training or theoretical orientation whatsoever at the initiation of the course.The post-test sample formed the experimental group after receiving the necessary theorization on problems and a by -way of suggested strategies to overcome them.The instructor spent six weeks in the course introducing theories and relevant exercises on a daily basis.The main thrust of the exercises was to identify the problems encountered and to reinforce the theory learned in this course.The practice stage comprised a multi-level host of exercises (above word level, sentences, texts from English to Arabic and vice versa) in which the students were tasked these exercises to improve their mastery of the theories learned.The students were expected to implement the techniques when appropriate.To this effect, the exercises aimed at preparing the students to cope with the problems encountered during the process of translation.They also provided them with guided supervision and experience in bilingual translation.

Population of the Study
Forty six undergraduate (third and fourth year) female students at Princess Alia University College/Al-Balqa'a Applied University in Jordan majoring in English Language and Literature were randomly selected during the summer course of the academic year 2013/2014.Originally, this section, the only one offered during the summer course, included 54 students of which 8 students were excluded because of the incompletion of the translation task required.Students have completed most of their linguistics and literature requirements for the B.A. program.These courses include beginning writing, advanced writing, grammar, language skills, syntax, semantics, short story, drama, novel, etc., after which they are allowed to be enrolled in this translation course.Most of the students share approximately the same linguistic, sociocultural and educational background.Since they live in an Arabic-speaking community in which English is taught at public schools as a foreign language, they are expected to share the same attitudes and conceptions of the translation process as a whole.

Five Sample Individual Tables
In an answer to the questions raised by this study, an examination of the eight individual tables has revealed the following results of patterns on students' manipulation of some textual stylistic features and the translation of the Arabic discourse marker (wa).
Tables 1-5 show the stylistic choices students opted for when they were subject to the pre-test and post-test translation exercise.The choices (patterns) are identified to centre on segmenting chunks into paragraphs, sentence boundaries, punctuation, sentence length and handling the discourse marker wa.Results indicated that in translating the newspaper news report from Arabic into English, students used Arabic stylistic features and text-building devices such as the DM wa as obtrusive upon the TT linguistic manifestations of the text's sense.They seem to be immensely influenced by the way their language chunks and segments experience in terms proper to its environment.Students coerce these linguistic and stylistic preferences that their source language has on the codes and structures of organizing the TT.(1b) Two short sentences separated by full stop with 1(a) 54.3 5.
(1b) Comma between 1 ( a+b) 6.5 (Overall) No sentence breakers at all or rarely so, among sentences 6.5

Discussion
The source text that the translation task presents is an instance of expository text in journalistic writing and consists of two paragraphs.A model translation of the ST is provided by the researcher (See Appendix A) against which the students' areas of difficulties will be identified.As illustrated in the TT, the number of boundaries among global structural organization units (i.e., paragraphs) is obviously larger.The TT distinguishes six segments on the large scale of the text, namely, six paragraphs as compared to only two in the ST.

Paragraph Boundaries
Paragraphs, which represent a textual unit somewhere below the text and above the sentence, help structure the text's organizational plan.A unit of text structure, paragraphs manifest a "conceptual unity" whether in content or function (Kohlani, 2010, p. 165).It is this conceptual unity that acts as a means to identify paragraph boundaries.If two textual entities diverge in topic and subsequently, in functions, boundaries between them will be drawn as two paragraphs.Table 1A shows a preference of students to follow the ST paragraphing segmentation of 43.5% over any other means of marking paragraph boundaries.Only 6.5% of students showed a sensitivity to paragraph delineation as that which appears in the TT proposed by the researcher.While 8.7% of students followed their own paragraphing system, 34.8% considered the text as one stretch of language without necessity to draw boundaries among the various segments; the text is translated as consisting of one whole paragraph.
The basis on which paragraph boundaries are determined take into consideration what most trainee translators fail to observe, first and foremost, the "cognitive configurations" manifested throughout the elements of the text (Callow, 1998, p. 210).Students' consideration for their concept about paragraphing is being motivated by "mechanical aspects of the writing process" (Hatim, 1997, p. 58) rather than any concern for the conceptual unity holding among the discourse segments in the text.
In the Source Text, sentences (1a) and (1b) constitute one structural paragraph in isolation of sentence (1c).This is due to the fact that sentence (1a) represents an exposition to the topic and sentence (1b) provides continuity to the event described in the introductory sentence.The two sentences, therefore, convey one idea and serve one function, "i.e.introducing the text and thereby providing a context within which the text is to be understood" (Kohlani, 2010, p. 167).Brown and Yule (1986) observe, "If the text displays other signals of the writer's intention to set this stretch of discourse as one paragraph, then this confirms that the writer is marking a topic-shift at this point"(99).Topic-shifts, or as referred to by Khalil (2000, p. 134), "thematic break", are the grounds on which paragraph boundaries are marked.Yet, in the printed text, sentence (1c) is imbedded in the supposed structural unity that both (1a) and (1b) form, and all three sentences are displayed as one independent orthographic paragraph.
Albeit it is evident that there is a topic shift at the onset of sentence [1c], students prefer relying on the orthographic conventions for marking paragraph boundaries in Arabic.This reflects that they are heavily influenced by their native language, disregarding the stylistic conventions of writing in English.What also indicates discontinuity of topic is, as Van Dijk (1982, p. 181) maintains, the shift of "participants, time, circumstances and global event or action.At this point, he observes, "one may assume that there is beginning of a new episode".Just as thematic continuity is an indication for paragraph boundaries, so does the fulfillment of the function that a paragraph is meant to serve.Students showed their lack of understanding of what makes a paragraph by acting upon their misconceptions.
Nor does the experimental group show a more consistent result in relation to the way they handled chunking paragraphs.Except for 17.4% of students who treated the text as one paragraph, the other categories related to the number of paragraphs show a rather drastic result; 58.7% followed the number of paragraphs in the ST, 2.2% the TT paragraphing and 21.7% designed their own paragraphing delineation.
For students, what marks the boundaries of a paragraph is that which appeals to the eye.Therefore, unity of sense is ignored in the interest of layout printing conventions such as spacing and indentation.In other words, they pay little or no regard for the structural paragraph for the sake of the orthographic paragraph.Thus, the ST in the translation task relies in segmenting paragraphs genuinely on specific grounded printing conventions in news report craftsmanship, rather than on any consideration for the conceptual unity holding the various elements of the text.The English translation (TT), thereby, presents a closer approximation to the concept of paragraph designation by relying on a structural rather than on formal basis.What renders the ST paragraphing system lacking is that it is structured in a way that invites for disruption in the flow of ideas.This may result in marking paragraph boundaries "in the middle of a structural paragraph or putting together several paragraphs as an indentation unit" (Longcare 1979, p. 115).
The immense emphasis on determining the paragraph boundaries in the ST stems from the fact that it is at the onset of sentences that wa is more likely to appear.
Wa in Arabic performs no less a vital role at the global level in connecting paragraphs in the text and is used most frequently in prefacing sentences and paragraphs.As a text-building device in Arabic discourse, the continuity of thought in the ST is preserved and marked via the inception of the second paragraph by the connective wa, ‫وﺗﺒﻌﺎ‬ ‫اﻻﺗﻔﺎﻗﻴﺔ‬ ‫.ﻟﻨﺺ‬ Linguists identify the functions associated with wa when occurring at the onset of sentences and paragraphs as addition or resumption "to preserve sentence continuity".Wa is the device that makes the continuity of the same idea possible despite the existence of a paragraph boundary.Wa appears between two chunks of texts or when the division "is not intended as a complete break" in the ongoing idea (Cantarino, 1975, p. 19).Wa, as Al-Batal (1990, p. 246) remarks, indicates that "the argument or discussion is still ongoing with no major breaks".1B shows that while the students' minds were continually comparing between the ST paragraphs and the means to connect paragraphs in the TT, 17.4% of the resulting mental representation they formed used and to connect this paragraph to the previous one in the TT.Translingually, this textual function of wa is not usually expected to exist in other languages, such as English, and, therefore, is often omitted in translation (Khalil, 2000, p. 142).Post-test results also show that there is an increase of 23.9% in connecting paragraphs with and.This, of course, stands at odds with English, which rarely uses and as a sentence or a paragraph opener.As a result, in Arabic-English translation, many of the examples of wa should be replaced by no connectives, or by a connective other than 'and' (Fareh, 1998, p. 309).

Sentence Boundaries
Another level of structural organization, the sentence enjoys a special status across languages as it is considered the basic unit in the structure of the text.Identifying sentence boundaries in Arabic, however, presents students with a difficulty because sentences sometimes defy a systematic clear-cut delineation.Hypothetically, the paragraph boundaries can be marked more easily, based on the ability to recognize them on conceptual and functional grounds.The means for the identification of sentence boundaries is language-specific (Kohlani, 2010, p. 184).
The criteria set for identifying sentence boundaries in Arabic are based on syntactic and semantic grounds: "structural independency" and its "capability of communicating a complete thought" (Kohlani, 2010, p. 191).A sentence is structurally independent when, as Allerton (1969, p. 30) puts it, it exhibits "freedom of occurrence relative to its neighbors".Coordinated clauses or sentences in Arabic constitute independent predicative structures, which indicate the development of one single idea (Cantarino, 1975).The immense emphasis on determining the sentence length and boundaries stems from the fact that it is at the onset of sentences that wa is more likely to appear.Although clauses connected by wa are structurally independent of one another, semantically, they develop one idea.Side by side with structural independence, for a unit to be considered a sentence, it should give a complete transcript of thought (Chafé, 1979).
Typically, sentences occurring after the introductory sentence in expository texts "are often initiated with wa-(and) and/or another connective expression" (Ryding, 2005, p. 409).This is considered as a token of good style in Arabic.English style of writing, on the other hand, warns against the initiation of and at the onset of sentences and, therefore, as a rule, is ruled out in translation from Arabic into English.This stylistic feature is especially grounded and common in journalistic and expository prose.As Ryding (2005, p. 419) points out, "this process of using a starting formula to introduce a sentence is especially common in journalistic and expository writing and gives it what Johnstone refers to as certain formulaicity".
Wa signals the continuity of the flow of ideas that any connected propositions imply in the texts "without implying any closer, more logical relationships" (Cantarino, 1975, p. 11).The problem of dealing with connectives across languages stems from the fact that each connective signals various relations or, conversely so, a relation is realized by resorting to various connectives.This can also add to the burden of the translator who attempts the transference of meaning across languages.The relation between the Arabic wa and and is definitely not a one-to-one correspondence.Wa shares some of these semantic relations with and in which case it is acceptable to provide it as a valid linguistic substitute.However, the Arabic wa is employed to express logical relations that hold among various elements in the text where and would certainly not be provided in the process of translation. .The continuity of sense is maintained through the explication of relations among the sentences in the text by the overt resort to the connective wa.Both these two paragraphs employ wa to connect clauses and phrases in the text, while, at the same time, other wa connectives convey the meaning of addition.However, some instances show that wa is loosely attached to connect sentences without being bound to any function that communicates meaning except for that of redundancy.The whole text in Arabic seems to be informed and structured by a network of relations based on coordinating sentences and linking the two paragraphs by means of redundant wa that loosely connects textual segments in order to maintain coherency among the propositions of the text.(1b) Additive marker (also, moreover) at the beginning of sentence 1(b) 2.2 6.
(1b) Comma between 1 ( a+b) 6.5 Results in Table 2 consistently show that the students' choice for a given linguistic code is decided by what is called for by their SL with little or no regard to the nuances worth their attention in the TT.The boundaries between sentences (1a) and (1b) are handled differently by students into six variant categories.Under Arabic text influence, 30.4% of the respondents opened sentence (1b) with a comma and and as an equivalent to wa in the ST.Obviously, this pattern, which dropped tangibly to 15.2% in the post-test, echoes the barriers separating the sentence in the Arabic text.Another pattern of 4.3% ensued from the students' attempt to replace coherent ties in the ST with equally coherent ones in the TT by a full stop to separate sentence (1a) and ( 1b).Yet, even in the abnormally rising post-test results of 15.2%, subjects felt that it is mandatory to reproduce wa into an equivalent and.The third pattern reflects an awareness of the nuances between ST and TT text-building devices.Less than half 45.7% of the subjects, rising to over half 54.3% in the post-test, drew that distinction between the two universes of discourse, Arabic and English, thereby consciously separating sentences (1a) and ( 1b) by means of a full stop.Students in the resulting fourth pattern showed sensitivity to render wa at the onset of sentence (1b) into and and tried instead to replace it with another connective that gives the meaning of addition.In the pre-test 4.3% of students, as against 2.2% in the post-test, provided equivalents to convey the meaning of addition rather than and, such as moreover, also, though improperly.Just on the other point of the extreme, 6.5% of the subjects deemed that both these sentences, once translated, should not be subject to the marking process, and consequently, presented them as one long sentence coordinated by means of and, as the nearest possible equivalent to wa .(1c) Initial sentential and 1(c) 2.2 2.
(1c) Arabic or English text influence (full stop at end of sentence) 87.0 4.
(1c) Additive marker other than and (also, moreover) at the beginning of sentence 1(c) 10.9 Normally, translators need to be heedful to the fact that "English needs fewer connectives" (Newmark, 1982, p. 178) than is the case with other languages.Similarly, as evident in Table 3, students' attempts to draw boundaries between sentences (1b) and (1c) reveal varying types of difficulty.One such type (4.3% pre-test against 2.2% post-test) is to preface sentence (1c) with and in an overt imitation of the sentence in the ST.Other resulting types reveal also a source of difficulty such as separating the two sentences by a comma by (2.2% pre-test and post-test), additive marker other than and (also, moreover) at the beginning of sentence 1(c) by 8% pre-test rising noticeably to 10.9% post-test, to name but a few.Furthermore, 78.3% of subjects responded positively to the boundaries between (1b) and (1c), which indicates that the influence exerted here is in favor of the TT rather than the ST, especially in the post-test results of 87%.
The difficulty is also accountable on the unsystematic use of punctuation marks in Arabic.Arabic writers include punctuation marks carelessly, or as Ghazala (2004) puts it, "poorly and haphazardly, by way of decoration" (230).The source of confusion for students in identifying sentence boundaries when translating from Arabic into English is ascribed to the unreliable punctuation system in Arabic as compared to the systematic, disciplined use of it in English.In Arabic, as Kohlani (2010, p. 185) explains, "punctuation is considered as innovation under the influence of European languages and was not introduced to the language until the first decade of the 19th century".The high superfluous inclusion of full stops and commas in Arabic is merely for decorative reasons because, as Holes (1995, p. 204) argues, "Arabic has its own native system of textual chunking which relies on coordinating and subordinating conjunctions which perform the dual role of signaling formally the beginnings and endings of sentence groups, and indicating the nature of the logical and functional relationships between them".He observes that this transition is true of expository texts which rely on Western punctuation norms instead of any utilization of its old norms.
At some points, discourse markers in Arabic, such as the connective wa (and), function in texts as punctuation marks function in English texts.Much to the burden of a translator who has a responsibility to maintain the coherent thread running through the TT, these connective words are not always translatable as they sometimes perform strictly grammatical functions rather than adding semantic content.These connective words are indispensable for the authenticity of a text in Arabic.Moreover, the subjects' decision on whether to drop or retain the punctuation marks appears unconscious.The findings showed the unconscious and haphazard overuse of commas and full stops more, rather than less.This confirms that even in translating into English, which attaches a high importance on judicious and careful use of punctuation marks, students do not internalize it in their minds as an essential aspect of discourse analysis that signals a semantic relationship between sentences and clauses.Results reflected a lag behind in terms of accuracy in marking sentence breakers of 13.0% pre-test against 4.3% post-test for inappropriate use of full stop, 34.8% against 19.6% for improper use of commas, while 15.2% against 6.5% showed no discern, or rarely so, to sentence breakers.The results show the students' tendency to chunk information in larger than necessary discourse segments.
Another obstacle for students' inability to decide on sentence boundaries is the length of the Arabic sentence as compared to the English sentence.In an Arabic text, it is possible that the whole paragraph consists of one long sentence separating its chunks by means of commas.This is rather an extravagant practice in English and is often rejected in favor of the use of obviously shorter sentences.
This forces students to think that just as Arabic language relies highly on coordinating the various elements in the text as an organizational strategy, so does English.The inconsistency of using the punctuation system in Arabic drives students to deal with the English sentence as the Arabic one.The practice becomes more plausible for them to confuse the stylistic features they find in English with those existing in their own language, being Arabic, in light of the fact that the Arabic sentence is naturally longer than the English one.As stated before, a whole paragraph in Arabic may consist of one extraordinarily long sentence separating chunks by inconsistent use of commas.This entices students to wrongly conceive of the possibility that they may also apply the extraordinary long sentence in Arabic and the haphazard, careless use of punctuation marks to English language, too.In translation from Arabic into English, the unit of discourse should always be "as small as possible and as large as is necessary" (Newmark, 1982, pp. 15-17).He argues that, "the shorter the translation, the better it is likely to be".
Side by side with the more general function of continuity and smooth transition among the discourse elements in the text, there exists a more specific function of addition.As an additive discourse marker, wa signals adding new information to previous ideas.It functions as an additive marker within sentences "to link clauses, phrases, and words" Ryding (2005: 409) .English, on the other hand, uses a comma to separate each item.Yet, results revealed a reverse pattern to what is theorized here.Over half of the study subjects (58.7% pre-test) deem it obligatory to reproduce the same number of conjunctive (wa)s within the internal structure of sentence (1c), as against only 56.5% in the post-test results .Just as the sentence in the ST includes three (wa)s, so does the TT.What corroborates the assumption that students still feel obliged to consult the codes that their Source Language dictates is their attempt to coerce the Arabic sentence structure over the English sentence structure; that is, they cling to reproduce the same basic word-order in Modern Written Arabic, which is considered to be VSO, unlike English in which the basic pattern is SVO.English grammar necessitates that the subject is in "pre-verbal position" (Givón, 1983, p. 29).Over 17% of the students in the control group (17.4%) once pre-tested opened their sentences with a verb, such as, Signed Al-Hussein Ben Talal University and Jordan Badia Research … as compared to 15.2% in the experimental group in the post-test.
The care with which translators should handle the nuances in the TT's linguistic system, including cohesive devices, cannot be overemphasized.As a rule, cohesive devices such as discourse markers are always under the jeopardy of over-translation and over-explication.Therefore, they are often deliberately omitted as they often have a "phatic" character in order to maintain the readers' interest (Newmark, 1988, p. 59).

Conclusion
This study has attempted to examine the influence that the SL of the trainee translators exerts on the TT.A salient indicator to diagnose this mode of interference is through the examination of the interaction between discoursal and stylistic features with register interlingually.In order to account for this problem in translation, the study takes Holmes' mental approach, the "The Mapping Theory", as its base.
Data obtained from the study on the conscious handling of the Arabic discourse marker wa (and) and some stylistic features when translated into English in newspaper news reports have revealed that SL and hence ST text-building devices interfere with TT linguistic manifestations of the text's sense.Translators, whether they are professionals or trainees, need to eliminate any kind of interference, because no matter how plausible it is, it is always taken to be a mistranslation.Their translation should sound impervious to interference, which abound in poorly written and sometimes inaccurate TTs.They need to use common sense to determine when the TT diverges from the norms existing in the ST.Their translation should show the value of theory and the force of practice.His/her thorough knowledge of both SL and TL should be backed by empirical evidence to determine the text's grammatical, syntactic, semantic and discourse oddities to be eliminated.It follows, then, that the problem facing the trainee translator is one of choice, for accuracy and for adaptation.In their information-processing role, the translators should undermine any choice, which renders their translation anything but neutral, to the extent possible.What is equally shocking and exasperating in the study is that some post-test results showed that there is no real advance from the patterns in the pre-test.Trainee students are shocked when they come to realize that personal value judgment does play a role in the choices they take in translation.Unless exercising conscious and deliberate regard, their choice of multi-level text-building devices, local and global, as well as stylistic features, will be greatly influenced by the linguistic norms of the SL.In accordance with Holmes Mapping Theory, the translators should consider solving some particular problems of the SL text, proceeding in a phased manner.They should group them under general heads to handle, such as the title, the structure (paragraphing and sentence connectives), shifts, metaphors, cultural words, translationese, proper names, neologisms, untranslatable words, ambiguity and level of language while simultaneously projecting a map of the TT to be produced.

Implications and Future Directions
The broader implication of the findings is as follows: Trainee translators from Arabic to English face significant problems in the adequate translation of some discourse markers, most prominent of which is wa, and other stylistic features in newspaper expository texts, i.e., news reports.These problems may survive until advanced stages of language learning.The persistence of such errors in translating the Arabic discourse marker wa, in addition to rather long sentences and injudicious use of punctuation marks when translating into English showed that they are related to the trainee students' language system features.Treatments on the subject have tangibly fallen short on the discussion of methods, approaches or practical examples.On the whole, they make no attempt to distinguish what goes on in the mind of a trainee translator when dealing with types or quality of texts.
Research into trainee translators' problems with discourse markers in general and text-type stylistic features in particular would benefit from a corpus-based research that contrasts the use of discourse markers across varied text types in Arabic and English, indispensable of other stylistic features pertaining to each text-type.Increased research and teaching is required here.The finds in the current study also suggest that research into the economical use of discourse markers in the English learners with and without these markers and at two levels of text structure (the sentence and the paragraph) is worth investigation.It may reveal negative transfer effects worthy of pedagogic intervention.It would be enriching to compare and contrast similar data from a text-type across Arabic and English languages, regarding their frequency and distribution.Emphasis on the value that Holmes' "Mapping Theory" possesses in regulating the mental process a translator exercises consciously and deliberately should not be undermined throughout linguistic transference research.
The drawback explicit in the findings of the performance of the undergraduate students of translation courses calls for the need to reconsider instruction methodology, approaches and efforts in this regard to explicate the fine differences that each discourse universe draws in forming its experience with language and how to switch swiftly and safely between any two or more discourse universes.

Table 2 .
Patterns of "mapping out" sentence boundaries, punctuation and translation of Arabic (wa) in sentences 1Full stop + and { at the beginning of 1(bAdditive marker (also, moreover) at the beginning of sentence 1(b) 4.3 6. (1b) No full stop + sentential and {linking1(a)+1(b Arabic text influence {comma initial and} 15.2 (1b) Full stop+ and {at the beginning of 1(b)} 4.3 (1b) Full stop+ and {at the beginning of1(b)} 15.2 (1b) Two short sentences separated by full stop with 1(a) 45.7 (1b) Two short sentences separated by full stop with 1(a) 54.3 (1b) Additive marker (also, moreover) at the beginning of sentence 1(b) 4.3 (1b) Additive marker (also, moreover) at the beginning of sentence 1(b) 2.2 (1b) No full stop + sentential and {linking1(a)+1(b)} 6.5 (1b) No full stop + sentential and {linking 1(a)+1(bArabic or English text influence (full stop at end of sentence) 78.3 (1c) Arabic or English text influence (full stop at end of sentence) 87.0 (1c) The same number of conjunctive ands (3 was in Arabic, 3 ands in English between phrases) 58.7 (1c) The same number of conjunctive ands (3 was in Arabic, 3 ands in English between phrases) (f) linked by and as one sentence} 10.9 (1f) Arabic text influence {and at beginning of 1(f) {1(e) +1(f) linked by and as one sentence} 13.0 (1f) Arabic text influence {comma at the end of sentence 1(e) + and at beginning of 1(f)} 26.1 (1f) Arabic text influence {comma at the end of sentence 1(e) + and at beginning of 1(f)} 15.2 (1f) Arabic text influence {full stop at the end of sentence 1(e) + and at beginning of 1(f)} 21.7 (1f) Arabic text influence {full stop at the end of sentence 1(e) + and at beginning of 1(f)} 15.2 (1f) Comma + initial and {sentential Arabic text influence {comma at the end of sentence1(e) + and at beginning of 1(f)} 6.5 (1f) Comma + initial and {sentential Arabic text influence {comma at the end of sentence1(e) + and at beginning of 1(f)} 15.2

Table (
1A).Patterns of "mapping out" paragraph boundaries Arabic text influence(initial and /and ) at the beginning of sentence ‫و{‬ ‫ﺗﺒﻌﺎ‬ ‫ﻟﻨﺺ‬ ‫}اﻻﺗﻔﺎﻗﻴﺔ‬

Table 2 .
Patterns of "mapping out" sentence boundaries, punctuation and translation of Arabic (wa) in sentences 1[a], 1[b]

Table 3 .
Patterns of "mapping out" sentence boundaries, punctuation and translation of Arabic (wa) in sentence 1[c] (1c)Additive marker other than and (also, moreover) at the beginning of sentence 1(c) 10.9 Table 4. Patterns of "mapping out" sentence boundaries, punctuation and translation of Arabic (wa) in sentence 1[e]

Table
Locally, the ST includes five instances of wa as statement openers at sentence boundaries, namely ‫ﻋﻦ‬ ‫اﻟﻤﺬآﺮة‬ ‫وﻗﻊ‬ ‫و‬

Table 3 .
Patterns of "mapping out" sentence boundaries, punctuation and translation of Arabic (wa) in sentence 1[c] . More specifically, Arabic uses wa in lists to list items, as evident in the multiple wa's

Table B2 .
Examples of the Various Functions of the Arabic wa / English and Function Example Consequence I felt a severe headache and I went to see a doctor Sequence John peeled off the orange and I ate it Contrast John is an introvert and Mary is an extrovert Simultaneity I am eating and my brother is reading Concession She studied hard and she failed Condition Help me solve the problem and I will reward you Addition She is pretty and she usually puts on attractive clothes Explanation We are left with one choice and that is to fight Comment Tempted by the five-hundred dollar prize, John drank 10 cans of pepsi and that can be dangerous Resumption Kaana Sulayman al halabi yamshi fi:shaari wa kaanat yadaahu fi: jaybih Suleiman Al Halabi was once walking along a street with his hand in his pocket Manner daxala Zaydun wa huwa yabtasim Zaid came in smiling Oath Wa Allahi la?usaa idannak I swear that I will help you Adverbial (by/ along) sirtu wa aljbal I walked by the mountain Option Kul ma shi?ta min inabin wa mawzin wa burtuqaal Eat whatever you want: grapes, bananas or oranges Redundance ma ra?aytu ahadan ?illaa wa huwa yabtasim Everyone I saw was smiling Praise/admiration zaarani ?axuuka wa ayyu rajul Your brother visited me.What a man he is!Threat / underestimation …..wa man huwa Ahmad?……who is Ahmad, anyway?
Arabic text influence(initial and / and at the beginning of sentence ‫و{‬ ‫ﺗﺒ‬ ‫ﻌﺎ‬ ‫ﻟﻨﺺ‬ ‫}اﻻﺗﻔﺎﻗﻴﺔ‬ 17.4 (1e) Arabic text influence(initial and / and at the beginning of sentence{‫و‬ ‫ﺗﺒﻌﺎ‬ ‫ﻟﻨﺺ‬ ‫}اﻻﺗﻔﺎﻗﻴﺔ‬