Evaluating an Arabic-English machine translated text: An Analytical and Pedagogical Approach

     کلیة التربیة

        کلیة معتمدة من الهیئة القومیة لضمان جودة التعلیم

        إدارة: البحوث والنشر العلمی ( المجلة العلمیة)

    =======

Evaluating an Arabic-English machine translated text:

 An Analytical and Pedagogical Approach

By

Dr. Marghany Mahmoud Marghany

Higher Institute for Specific Studies, Giza

}         المجلد الثانی والثلاثین– العدد الثالث– جزء ثانی - یولیو 2016م  {

http://www.aun.edu.eg/faculty_education/arabic

Abstract

The present study discussed the translation problems which Egyptian EFL undergraduates encountered when translating an Arabic source text into their target language by using machine translation. Data were collected based on an English machine translated text of an Arabic source text. A Descriptive analytical technique was employed for the purpose of data analysis. That is, the English machine translated text was analyzed in light of both the Arabic source text and the appropriate English translated text. Findings indicated various types of syntactic, lexical and grammatical problems affecting the learning of translation skills by Egyptian EFL undergraduates. Teaching implications suggested solutions for such problems.

Key words: Egyptian EFL undergraduates, machine translation, word-for-word translation, connotation, the Grammar-Translation method, and L2 learning process. 

Introduction

The of word-for-word translation spreads nowadays because of the available advanced technology to which Egyptian EFL learners have an easy access. The spread of such a type of translation is also attributed to the heavy use of machine translation by Egyptian EFL learners. There are various examples of word-for-word translation which resulted from using machine translation such as "the ears of Morocco" instead of "a call for Maghrib prayer", "university cities" instead of "students' hostel", etc. this type of translation can undoubtedly impede the process of teaching and learning translation as an academic course.  Even though the Arabic word 'Modon' can be literally translated into English as 'cities' which is certainly correct. However, this word connotation does not refer to parts of a country, but rather to a place where students can reside.   This sort of translation encroaches by all means the fundamental meaning of translation as being a process mainly concerned with conveying the exact meaning from one language to another (Oxford 1990; Bell 1993). Such a way of word-for-word translation stimulated the researcher to have further insights into the machine translation output, and identify its differences which are distinguished from a human English translated text. Furthermore, such differences have effects on how Egyptian EFL learners practically perceive the process of translation, and on which basis they choose an English equivalent for an Arabic word. Therefore, it was felt necessary to analyze the effect of using machine translation on helping Egyptian EFL undergraduates when they do translation into their English target language. It also addressed how translation can be methodologically viewed within the process of L2 learning. Cook (2007) explained that the field of second language acquisition (SLA) has a negative evaluation of translation as a L2 learning task because of its affiliation to the Grammar-Translation method. Therefore, translation is not considered as a fundamental means of L2 learning process. According to Cook, this is evident as several studies handled translation as a target per se; not as a means of L2 learning.    

Problem Statement

Advanced technology has facilitated the process of doing translation among Egyptian EFL learners in general. Egyptian EFL undergraduates in particular, who are studying translation as an academic course, can easily get an Arabic text translated into English. However, the reliability of its translation remains questionable. Therefore, this study examined the erroneous differences resulting from translating an Arabic source text into English by using machine translation. Consequently, the study handled the types of such erroneous differences whether syntactic, grammatical, or lexical. It also discussed the effect of using machine translation on the process of teaching translation to Egyptian EFL undergraduates who are majoring in English language and translation.

Aims

The present study aimed to identify the possible erroneous differences that may result from using English machine translation of an Arabic source text. It also intended to clarify how such erroneous differences can be misleading when the learning of translation skills by Egyptian EFL undergraduates are concerned. In other words, it aimed to highlight the negative effects of machine translation on the process of teaching translation to Egyptian EFL undergraduates. Finally, it drew implications for how these negative effects of machine translation can be overcome.

Questions

Therefore, the present study attempted to answer the three following questions:

1. Why cannot machine translation be reliable for learning the translation skills by Egyptian EFL undergraduates?

2.  How can the negative effects of machine translation on the EFL learning process be overcome?

3. Is translation a useful task for the purpose of L2 learning?

Significance

The study is generally significant as it helps judge the reliability of machine translation as one of the available options for Egyptian EFL undergraduates. Its significant is particularly related to machine translation effects on teaching translation as an academic course to Egyptian EFL undergraduates. It is also significant as it classified the potential syntactic, lexical and grammatical erroneous differences of using machine translation. The classification of such differences may help inspire programming solutions for such erroneous differences in the future. It is methodologically significant as it evaluated the machine translated output based on descriptive analysis rather than relying on a questionnaire-based task as other researchers like Kadhim, Habeeb, Sapar, Hussin, and Abdullah (2013) did. The study is also of a special significance as it sheds some light on the importance of translation role in the process of second or foreign language learning. Hence, it introduces academic translation courses as important components of the investigation of foreign or second language learning process.

Limitations

The study is limited to the analysis of machine translation of an Arabic source text and its effects on the learning of translation skills by EFL undergraduates in Egypt who are enrolled in the Departments of English Language and Translation. 

Literature review

            Due to the present study nature, the studies reviewed are of two types: studies on machine translation and studies methodologically discussed translation a means of L2 learning and teaching process. Examples of the first type are Kit and Wong (2008), Tripathi and Sarkhel (2010), and Kadhim, Habeeb, Sapar, Hussin, and Abdullah (2013). Meanwhile the second type of studies includes Garcia (1995), Liao (2006), and Machida (2011).

Kit and Wong (2008) examined how the technology-based machine translation is suitable for translating various L1-based legal texts into English. They also compared the outputs of various technologies of machine translation of those source texts. Those technologies are: Babel Fish, Google, PROMT, WorldLingo, Systran and SDL.  The source texts were written in different languages; European languages (Dutch, German, Portuguese, Spanish, Swedish), Middle East language (Arabic) and Asian languages (Chinese, Japanese, and Korean).  Kit and Wong explained that the extent of machine translated text is largely influenced by the source text quality. The incorrect machine translation can be attributed to grammatical errors (including spelling and punctuation errors) contained in the source text. In addition, legal texts are of special nature concerning word order and sentences length. Culture knowledge represents another main constraint for the legal texts machine translation since the machine translation tools are not well-equipped with various legal cultural types of knowledge. Kit and Wong accounted the inaccuracy of machine translation outputs based on the linguistic differences (including morphology, syntax, semantics) between those first languages and English; where the Dutch has the highest score of such linguistic differences followed by the Asian languages. However, the element of linguistic differences between Arabic and English has the least score among all examined first languages. In this regard, Kit and Wong (2008:316) said "Interestingly, linguistic diversity seems not powerful enough to prevent Arabic from receiving better evaluation scores over five languages: three Asian and two European". In this concern, Al-Salman (2004), and Zughoul and Abu-Alshaar (2005) pointed out that Arabic is a world important language whose written texts were experimentally translated by using machine translation tools since their early emergence. The Arabic language importance stems from its well-known morphological complexity.

Tripathi and Sarkhel (2010) discussed different machine translation approaches. They admitted that Google Translator and Babel Fish as online translation websites do not accurately provide the sought translation. In this regard, Yates (2006) proved that Babel Fish does not provide satisfactory solutions for the translation of source legal texts written in German and Spanish into English. However, these machine translation tools are considered as helping and guiding factors in the mainstream of doing translation. This point, the researcher believes, is the milestone which ESL/EFL learners should be aware of and it is the duty of those in charge of teaching translation academic courses to alert their students to. Tripathi and Sarkhel indicated that the investigation of machine translation mainly comes under the field of Computational Linguistics. It aims to convert the source text meaning into the translated one. It consists of two levels: 'Metaphrase' and 'Paraphrase'. Whereas the first level aims to literally translate the original text, the second one may dynamically look for the equivalents of the original text. They added that machine translation employs various 'methods including dictionary based machine translation', 'rule based machine translation'. The dictionary based method depends on electronic dictionaries feedback to provide equivalents to the source texts lexis. Therefore, this method might be helpful in translating phrases, but may fall short to translate full sentences. The rule based machine translation works on the morphology, syntax and semantics of the source and target languages. It relies on direct approach and transfer model. On the contrary to the direct approach, the transfer model relies on dictionaries belong to the source and target languages dictionaries and their rules of grammar; where the source text is converted into an abstract for which a target language equivalent is found.              

Kadhim, Habeeb, Sapar, Hussin, and Abdullah (2013), investigated the quality of machine translation using Google and Babylon tools to translate 40 headlines of Arabic news into English. The researchers used a questionnaire task administered to 40 professionals to evaluate the quality of the two machine transition outputs. Kadhim, et. al. concluded that in terms of clarity, Google and Babylon equally provided clear machine translation output (80%). However, Google was proven to be more accurate than Babylon in its automated translation output (77.5% compared to 75%). However in terms of style Babylon outperformed Google (72.5% compared to 70). They advocated they use of machine translation at different educational levels from schools right up to territory education due to its quick performance of translation that can save time and money for EFL learners. They need not check words up in the English dictionaries and seeking the help of professional translators bears high costs.

Garcia (1995) examined the translation of three Spanish source texts by advanced Spanish learners of English. The number of collected English versions varied among the three Spanish source texts. Whereas 97 English versions were collected for the first Spanish text, 76 English versions were collected for the second. In addition, 91 English versions were collected for the third Spanish source text. Garcia gave examples of wrong ways of Spanish-text translation into English. For instance, 'abandono' cannot be literally translated into English as 'abandon his suit'. Instead, 'took off his suit' is the proper way of translation the Spanish sentence 'abondono el traje que llevall puesto'. Some Spanish students could not differentiate between the meanings of 'course', 'class', and 'lesson' as they generalized the use of 'course' to translate the three Spanish equivalents into English. Other students were confused in using the English homophonous words like 'wondering' and 'wandering' while translating from Spanish into English. Garcia advocated the usefulness of integrating translation exercises into L2 classrooms.                      

Liao (2006) examined how 351 Taiwanese college students view translation as an element of L2 learning process and what learning strategies they apply when they use translation in order to learn English. Data were collected through three questionnaires and an interview of ten participants only. Although EFL learners heavily depend on translation as a L2 learning strategy in order to realize facilitate their understanding and production of their target language, SLA research paid little attention to how L2 learners themselves view the role of translation in L2 learning process. Liao used three sets of questionnaire for the purpose of data collection, namely the inventory for beliefs about translation, the inventory for translation as a learning strategy and the individual background questionnaire. Liao made use of descriptive statistics like frequencies, means and standard deviations to delineate a full description of his sample's beliefs about translation. Participants' responses to the three questionnaires were analyzed by factor analysis to determine how they varied in their responses. The effect of participants' variables on their beliefs about translation was described based on the Multivariate analysis of variance (MANOVA). Liao reported that participants positively viewed translation as it helps them learn different English language skills as 16 items out of the 24-item first questionnaire obtained the highest mean score whereas only four items obtained low mean score. In addition, the interview analysis revealed that most interviewees believe that translation help overcome memory constraints of language learning and lessen anxiety. On the other hand, they expressed fears that translation may mislead L2 learners whenever a word has various meaning or translation can hinder their ability to develop an English native like style. Liao found that variation in using translation as a learning strategy among his participants is related to their different proficiency levels in English.             

Machida (2011) examined the possibility of using a L1 input              in order to enhance chances of L2 learning. In other words, translation could be viewed as a positive aid to L2 learning. In doing so, translation can be methodologically integrated into the process of L2 learning. Translation as a L2 learning task helps improve ESL learners' understanding of L2 structural forms, particular meanings in specific contexts, and their L2 writing skills. In addition, ESL learners will be knowledgeable of L2 culture besides their own L1 culture in order to give L2 accurate message.

Machida asserted the importance of using translation as a L2 task and its effective impact on L2 learning. Machida accounted that ESL learners build their cognition based their L1 knowledge of which they can positively construct L2 structural forms and semantics. In other words, L1 has undeniable role in understanding L2 structure and meaning. Machida explained that historically translation is a means that helped L2 learners to understand the meaning and grammar rules of their target language. Therefore, in spite of the fact that L2 teaching methods has called for overlooking the use of a L1 in L2 teaching like the Direct Method, Audiolingualism, Natural Approach; it is time for restoring the L1 positive role in L2 learning and teaching processes. The Communicative Approach has fallen short to compensate for grammar rules learning as it mainly focuses on producing functions such as greeting and asking for directions. Thus, as Hinkel and Fotos (2002) indicated tasks of communication still need to produce more grammatically accurate outcomes. Machida explained that translation a L1 text into the target language is useful in sustaining their L2 learning opportunities. When doing translation from their L1, L2 learners try to identify the exact L2 equivalents. In the process of translation, Machida view ESL learners' L2 linguistic, cultural and cognitive knowledge as less than that of their L1. L2 translation tasks help linguistically ESL learners to develop their L2 vocabulary and non-linguistically help them develop their L2 'conceptual knowledge' and 'contextual knowledge'. In addition, translation helps sustain ESL learners' understanding of L2 texts comprehension, and L2 social and cultural norms.

Methodology

Data Collection

An Arabic source text was selected to be translated into English using machine translation (Appendix A). The English machine translated text (Appendix B) was then analyzed in comparison to the original Arabic source text. 

Data analysis      

The study depended on the descriptive and analytical technique in order to determine the types of possible erroneous difference resulting from using machine translation. This technique has been intensively used in several descriptive and analytical studies. The Arabic source text was divided into sentences so as its English machine translation could be well judged and classified.  Therefore, each Arabic sentence was displayed followed by its English machine translation, which is quoted between two inverted commas, for the purpose of clear and accurate analysis. Descriptive statistics was conducted to indicate the percentage and mean scores of the syntactic, lexical and grammatical erroneous differences. Findings were then related to the process of teaching translation to Egyptian EFL undergraduates in order to draw teaching implications.

Analysis of the English machine translated text

 The first Arabic sentence:

لا شک أن الصحافة الإلکترونیة أحدثت تطوراً کبیراً فی عالم الإعلام من خلال نشر ومتابعة الأخبار على مدار الساعة،

The English machine translation:

"There is no doubt that online journalism has made a significant development in the media world through the publication and follow the news around the clock".

Analysis:

-          On the contrary to the source text context, the word "journalism" could have a broader meaning which refers to the profession itself. Thus, the word 'press' is more appropriate to this context.

-          The translation of this subordinate clause "… through the publication and follow the news around the clock", could be modified as '… through publishing and following the news up around the clock'.

The second Arabic sentence:

فاستخدمت کل إمکانات الرسالة الإعلامیة، وخلقت علاقة حمیمیة بینها وبین والقارئ الذی یستطیع المشارکة بکتابة الخبر وإبداء الرأی،

The English machine translation:

" it used the all the potential of the media message, and created an intimate relationship between them and the reader who can participate left the news and opinion "

Analysis:

-            Grammatical erroneous differences are evident in the redundant use of the definite article and omission of plural mark 's' in "all the potential of the media message", which could be better translated as 'all potentials of the media mission'.

-            The use of "message" instead of 'mission' refers to a lexical erroneous difference as the latter provides a broader meaning than that given by using "message".

-            Certainly, the adjective "intimate" does not fit this context in describing the noun "relationship". Instead, it was appropriate to use the adjective 'close' to describe the online press relationship with the reader.

-            The Arabic prepositional phrase (بکتابة الخبر وإبداء  (الرأی which consists of a preposition and two noun phrases  was translated as "left the news and opinion" instead of 'through news writing and opinion expression'. The English machine translation refers here to both syntactic and lexical erroneous differences.

The third sentence:      

کما استطاعت استقطاب العدید من الشرائح خاصة أنها سریعة التأثیر والوصول          إلى القارئ.

The English machine translation:

"as attracting many private slides able they are quick impact and access to the reader".

Analysis:

-            Apparently "private slides" is a word-for-word translation which is inappropriate to this context. This refers to the existence of a lexical erroneous difference as the connotation of selected words does not convey the meaning contained in the source text. Instead, "private" should be translated as 'particularly' and "slides" as categories.

-            The use of the adjective "private" violates the original word class of the Arabic word"خاصة" which functions as an adverb in the source text. The English machine translation, in this case, refers to a syntactic erroneous difference.

-            The English machine translation used "access" as a noun instead of the adjective 'accessible' which describes the noun 'online press'.

-            In addition, the English machine translation of "they are quick impact and access to the reader" refers to both syntactic and lexical differences which contradict the source text. It should be translated as 'the online press was able to attract several categories particularly it has rapid effect and is accessible to the reader'.

-            The English machine translation "private slides able" also refers to a grammatical erroneous difference due to copula omission which should precede the adjective "able".

The fourth Arabic sentence:      

ونوضح فیما یلی أهم تعریفات الصحافة الإلکترونیة وممیزاتها والعوامل التی تساعد على نجاحها، بالإضافة إلى کیفیة التحریر فیها

 The English machine translation:

"And explain Here are the top online journalism definitions and advantages and the factors that help to its success, as well as how the editing".

Analysis:

-          A clumsy structure based on word-for-word translation. It refers to syntactic differences which contradict the source text "explain Here are" and  "definitions and advantages and the factors"

-          The use of the word "help" is a main manifestation of word-for-word translation. Instead, it should be translated as 'factors leading to its success'.

-          In addition to the syntactic difference existing in "as well as how the editing", a grammatical difference also exists by overlooking the passive form. That is, it should be translated as 'how its editing was conducted'.

The fifth Arabic sentence:

تعریف الصحافة الإلکترونیة: استوقفت ظاهرة الصحافة الإلکترونیة الکثیر من الباحثین والدارسین، فتابعوها بالرصد والتحلیل،

The English machine translation:

"Definition of online journalism: stopped press the phenomenon of electronic lot of researchers and scholars, Fattabaoha monitoring             and analysis".

Analysis:

-            This English machine translation reveals an existence of syntactic erroneous difference in "stopped press the phenomenon of electronic lot of researchers and scholars". It should be translated as 'many researchers and scholars have deliberated the online press phenomenon'.

-            The verb "stopped" is inappropriate as well and should be translated instead as 'deliberated'. This refers to a lexical erroneous difference from the original text.

-            The English machine translation could not decode the Arabic verb phrase  تابعوها , which should be translated as 'they followed it up'. In this context, the English machine translation covers the three types of erroneous differences (syntactic, lexical and grammatical).

-            Another syntactic erroneous difference is the machine translation inability to translate the Arabic prepositional phrase بالرصد والتحلیل , as it omitted the preposition 'through'.     

 The sixth Arabic sentence:

وکانت نتیجة ذلک ظهور الکثیر من التعریفات الخاصة بها

The English machine translation:

"and the result was the emergence of a lot of their                  own tariffs,"

Analysis:

-          Clearly, the English machine translation using the word 'tariffs' instead of the word 'definitions' is a self-evident proof of its inability to determine the exact connotation of a word and hence providing another word which certainly violates the main meaning of the Arabic source text.

-          In addition, the use of "their own tariffs" instead of 'their related definitions' clearly indicates the huge difference between the machine translated sentence and the one existing in the Arabic source.  

The seventh Arabic sentence:

فقد عرفها البعض بأنها " نوع من الاتصال بین البشر، یتم عبر الفضاء الإلکترونی - الإنترنت وشبکات المعلومات والاتصالات الأخرى - تستخدم فیه فنون وآلیات ومهارات العمل فی الصحافة المطبوعة،

The English machine translation:

 "has some known as a kind of communication between human beings, it is across cyberspace - Internet and information networks and other communications - used the Arts and mechanisms and skills to work in the print media,"

Analysis:

-            The proper translation 'some defined it as' was replaced by the English machine translation "has some known as" in an indication of a syntactic erroneous difference.

-            Furthermore, the misuse of the verb "know" instead of 'identify' refers to lexical erroneous difference.

-            Another lexical erroneous difference can be found as well in the use of "human beings" instead of 'humans'.

-            The omission of the prepositional phrase 'in which' refers to a syntactic difference.

-            The use of the prepositional phrase "to work in" instead of 'necessary for working in' indicates both grammatical and syntactic erroneous difference.

The eighth Arabic sentence:       

مضافاً إلیها مهارات وآلیات تقنیات المعلومات التی تناسب استخدام الفضاء الإلکترونی کوسیط أو وسیلة اتصال، بما فی ذلک استخدام النص والصوت والصورة والمستویات المختلفة من التفاعل المتلقى  .

The English machine translation:

"plus the skills and mechanisms of information technologies that fit the use of cyberspace as a mediator or means of communication, including the use of text, audio, image and different levels of interaction with the receiver"

Analysis:

-       Lexical differences are apparent in using "plus", "audio" and "receiver" instead of 'in addition', 'sound' and 'the audience' respectively. Such differences may affect and change the source               text meaning.  

Results and Discussion

Table 1:

Descriptive statistics of Types of Erroneous Differences Resulted from Machine Translation

Table 1indicates that word selection represents the highest features of differences between the English machine translated text (12.08%) and the Arabic source text. The type of syntactic differences comes in the second place based on its percentage (8.73%). In addition, grammar represents the least aspect of differences between the machine translated text and the source one (4.7%). This indicates the machine translation failure to select the exact words that are consistent with the source text. This is attributed to the fact that machine translation cannot define the accurate connotation of certain words contained in the source text. It is a privilege which human mind can successfully fulfill and the advanced technology is still short of realizing. Therefore, it is not surprising that the syntactic type of differences comes in the second place after word selection. The ability of properly putting translated sentences in order requires cognitive skills that can only and fully be restricted to human mind. Displacement of words in a violation of proper word order greatly affects the original intended meaning of the source text that may lead to producing a meaning-distorted target text which syntactically contradicts the source text. In addition, the least type of differences between the source and machine translated texts is grammar. This indicates the advancement of technology used in the machine translation that can solve problems of grammar errors even though its percentage (4.7). This percentage is not relatively low if it is considered per se; it shows there are grammatical lapses which the machine translation can account for, solve or correct. The identification of the three lexical, syntactic and grammatical types of differences between the source text and the machine translation output can negatively affect the way EFL undergraduates do translation from Arabic into English and hence may have negative effects on their L2 learning process.     

Teaching Implications and Conclusion

As for the first question of why cannot machine translation be reliable for learning the translation skills by Egyptian EFL undergraduates? The researcher concluded that the highest effect of machine translation on the learning of translation skills by Egyptian EFL undergraduates is visibly apparent on their choice of the L2 appropriate equivalents to the source text. In addition, the syntactic differences between the Arabic and English languages make it difficult to consider machine translation as a reliable source for the Egyptian EFL undergraduates when translating from Arabic into English. Even though grammar represents the least aspect of difference between the machine translated text and the source text due to advanced technology used in the machine translation tools, machine translation cannot be the solely source of solving grammatical problems of translation. Their solution should mainly rely on the learners' grammar knowledge. These findings are compatible with those of Garcia (1995),  Hinkel and Fotos (2002), Al-Salman (2004), Yates (2006), Kit and Wong (2008), Tripathi and Sarkhel (2010), and Kadhim, et. al. (2013).          

The second question how can the negative effects of machine translation on the EFL learning process be overcome? To overcome the negative effects resulting from machine translation particularly on the process of teaching translation to EFL Egyptian undergraduates and generally on their L2 learning process, the researcher suggests that translation should be taught at earlier educational stages. That is, instead of only introducing translation exercises to first secondary school students, simple translation exercises can be introduced to first preparatory school students. This suggestion entails taking the factors of students' age and English level into consideration. Therefore, the suggested translation exercises at that educational stage should be simple ones that help students understand the differences between English words of relatively similar meanings such as chair and seat; plant and tree; home and house, etc. In addition, they help students differentiate the various meanings of the same word like 'book' as a noun and a verb. The suggested translation exercises can take the form of multiple choice exercises or true/false exercises. For examples, a preparatory school student can be given an English sentence like "I booked a ticket yesterday' and then asked to choose one of two pictures 'A' or 'B' that depicts the sentence meaning. Picture 'A' is that of a book while picture 'B' shows a boy booking a ticket. Such exercises can early make students aware of the connotation of words and hence they will be able to select the exact word which they are looking for among other choices. The same exercise can be used in the form of true/false exercises as the student will be given a sentence and a picture and will be asked to say whether the sentence is 'true' or 'false' based on the given picture.

House (2015) and Richards and Renandyds (2002) stated that modern language teaching methods denounced the role of the Grammar-Translation method. However, this method cannot be avoided when teaching translation as an academic course to Egyptian EFL undergraduates. Therefore, instead of fully abandoning that fruitful method it can be updated and revised to serve well the purposes of translation teaching at an EFL learning setting. This method can simply help EFL undergraduates fully understand the exact meanings of different L2 words. Hence, they will be able to determine the exact word that can fit in well with the words included the Arabic source text when translating it into English. Thus, even if they tend to seek the help of machine translation they will be able to detect the provided lexical differences. This method can help EFL learners as well to be aware of the different syntactic forms of the target language sentences including the noun phrase and verb phrase forms. According to Celce-Murica (2007) and Colina (2002) this awareness is useful in guiding them to correct whichever machine translated sentences that differ from the Arabic source text. Furthermore, the role of the Grammar-Translation method in learning the target language grammar rules is undeniable. It bears good results of L2 grammar mastery and accuracy. In addition, learning English language skills does not contradict with learning translation because translation encompasses knowledge of grammar, structure, vocabulary. In this sense, translation can support the process of TEFL            in Egypt.  

The third question is translation a useful task for the purpose of L2 learning? Translation per se is a useful L2 learning task as it helps develop the writing, grammatical skills of L2 learners. For instance, Malmkjaer (2010), Campbell (2013), Colina (2003), and Gonzalez Davies (2004) called for the use of translation activities in L2 classrooms because it sustains L2 learners' development of the four language skills and thus strengthens the process of L2 acquisition. For instance, translation is intricately entwined with the development of L2 reading skills (Zare 2012; Kern 2008; and Prince 2012). As such L2 learners are indulged in a mental translation process while reading a L2 text in order to comprehend the L2 new vocabulary based on their L1 synonyms. In this regard, Beeby (2004), and Liao (2006) concluded that translation is considered as a helpful factor that contributes to the success of L2 learning as it sustains the cognitive factors of L2 learning process like comprehension and memorization. Similarly, Khansir (2012) explained that Corder (1981) refuted the claims against L1 use in L2 learning on the basis of L1 negative transfer. Corder considered translation as a compensation factor for the obstacles hindering L2 learning process. Corder called for reconsidering the term of L1 interference which can be modified into a communicative strategy adopted by EFL learners to facilitate their L2 learning.             

 Suggestions for further research:

The researcher suggests the following topics for conducting               further research:

1. Investigation of the cultural, grammatical, and syntactic problems facing the improvement of machine translation performance.

2.  Examination of different ways in which translation can be exploited as a L2 teaching means.

3. Reviving and reviewing the position of Grammar-Translation method as an indispensible foreign language learning and teaching method.

4. Investigation of EFL learners' variables like age, proficiency levels, learning strategies that affect their L2 translation skills.

5. The examination of how EFL learners perceive the syntactic, lexical and grammatical differences between their L1 and L2 and its impact on their L2 learning progress.

6. Examination of the adopted learning strategies by Egyptian EFL learners when studying translation from and into English.

7. Investigation of the mental translation process which Egyptian EFL learners go through while learning L2 new vocabulary.

References:

-       Al-Salman, S. (2004). The effectiveness of machine translation. International Journal of Arabic-English Studies, 5, PP145-160.

-       Beeby, L. A. (2004). Language learning for translators. In Kirsten Malmkjar (eds.), Translation in undergraduate degree programmes. Amsterdam/Philadelphia: John-Benjamins Publishing Company.

-       Bell, R. T. (1993). Translation and translating: Theory and practice. London and New York: Longman.      

-       Campbell, S. (2013). Translation into the second language. New      York: Routledge.

-       Celce-Murcia, M. (2007). Towards more context and discourse in grammar instruction. TESL-EJ, 11, (2), 1- 5. 

-       Colina, S. (2002). Second language acquisition, language teaching and translation studies. The Translator, 8, 1-24. 

-       Colina, S. (2003). Translation teaching: From research to the classroom: A handbook for teachers. Boston:               McGraw Hill. 

-       Cook, G. (2007). A thing of the future: translation in language learning. International Journal of Applied Linguistics, 17, 3, 396-401.

-       Garcia, R. A. V. (1995). A new approach to the use of translation in the teaching of L2. Revista Alicantina de Estudios Ingleses, 8, PP239-252. 

-       Gonzalez, D. M. (2004). Multiple voices in the translation classroom. Amsterdam: John Benjamins.  

-       Hinkel, E. and Fotos, S. (2002). From theory to practice: A teacher's view. In E. Hinkel and S. Fotos (eds.,) New perspective on grammar teaching in second language classrooms, Mahwah. NJ: Lawrence Erlhbaum Associates, 1-15.

-       House, J. (2015). Translation quality assessment: Past and Present. New York: Routledge.   

-       Kadhim, K.A., Habeeb, L.S.,  Sapar, A.A., Hussin, Z., and Abdullah, M.M.R.T.L. (2013). An evaluation of online machine translation of Arabic into English news headlines: Implications on students' learning purposes. The Turkish Online Journal of Educational Technology, Vol., 12, 2, PP 39-50.

-       Kern, R. (2008). Making connections through texts in language teaching. S Language Teaching, 41 (3), 367-387.

-       Khansir, A. L. (2012). Error analysis and second language acquisition. Theory and Practice in Language Studies, 2 (5),              1027-1032.   

-       Kit, C. and Wong, T. M. (2008). Comparative evaluation of online machine translation systems with legal texts. Law Library Journal, Vol., 100, 2, PP 299-321.      

-       Liao, P. (2006). EFL learners' beliefs about and strategy use of translation in English learning. Regional Language Centre Journal, Vol., 37, 2, PP 191-215. 

-       Machida, S. (2011). Translation in teaching a foreign (second language: A methodological perspective. Journal of Language Teaching and Research, Vol. 2, no., 4, PP 740-746.

-       Malmkjaer, K. (2010). Language learning and translation. In Yves Gambier and Luc Van Doorslaer (eds). Handbook of Translation Studies. Vol. 1. Amsterdam and Philadelphia: Benjamins. 185-190.     

-       Prince, P. (2012). Towards an instructional programme for L2 vocabulary: Can a story help? Language Learning and Technology, vol. 16 (3), 103-120.   

-       Richards, J. C. and Renandyds, W.A. (2002). Methodology                     in language teaching. Cambridge: Cambridge University Press.

-       Tripathi, S. and Sarkhel, J.K. (2010). Approaches to machine translation. Annals of Library and Information Studies, 57.   PP 388-393.

-       Yates, S. (2006). Scaling the tower of Babel Fish: An analysis of the Machine translation of legal information. Law Library Journal, 98, PP 481-491.

-       Zare, P. (2012). Language learning strategies among EFL/ESL learners: A review of literature. International Journal of Humanities and Social Science, 2 (5), 162-169.

-       Zughoul, M. and Abu-Alshaar, A. (2005). English/Arabic/English machine translation: A historical perspective. Translators' Journal, 50 (3), 1022-1041.     

Appendix A: An Arabic Source Text

لا شک أن الصحافة الإلکترونیة أحدثت تطوراً کبیراً فی عالم الإعلام من خلال نشر ومتابعة الأخبار على مدار الساعة، فاستخدمت کل إمکانات الرسالة الإعلامیة، وخلقت علاقة حمیمیة بینها وبین والقارئ الذی یستطیع المشارکة بکتابة الخبر وإبداء الرأی، کما استطاعت استقطاب العدید من الشرائح خاصة أنها سریعة التأثیر والوصول إلى القارئ. ونوضح فیما یلی أهم تعریفات الصحافة الإلکترونیة وممیزاتها والعوامل التی تساعد على نجاحها، بالإضافة إلى کیفیة التحریر فیها. تعریف الصحافة الإلکترونیة: استوقفت ظاهرة الصحافة الإلکترونیة الکثیر من الباحثین والدارسین، فتابعوها بالرصد والتحلیل، وکانت نتیجة ذلک ظهور الکثیر من التعریفات الخاصة بها، فقد عرفها البعض بأنها " نوع من الاتصال بین البشر، یتم عبر الفضاء الإلکترونی - الإنترنت وشبکات المعلومات والاتصالات الأخرى - تستخدم فیه فنون وآلیات ومهارات العمل فی الصحافة المطبوعة، مضافاً إلیها مهارات وآلیات تقنیات المعلومات التی تناسب استخدام الفضاء الإلکترونی کوسیط أو وسیلة اتصال، بما فی ذلک استخدام النص والصوت والصورة والمستویات المختلفة من التفاعل مع المتلقی.

Appendix B: An English Machine Translation Text

https://translate.google.com/

There is no doubt that online journalism has made a significant development in the media world through the publication and follow the news around the clock, it used the all the potential of the media message, and created an intimate relationship between them and the reader who can participate left the news and opinion, as attracting many private slides able they are quick impact and access to the reader. And explain Here are the top online journalism definitions and advantages and the factors that help to its success, as well as how the editing. Definition of online journalism: stopped press the phenomenon of electronic lot of researchers and scholars, Fattabaoha monitoring and analysis, and the result was the emergence of a lot of their own tariffs, has some known as "a kind of communication between human beings, it is across cyberspace - Internet and information networks and other communications - used the Arts and mechanisms and skills to wo rk in the print media, plus the skills and mechanisms of information technologies that fit the use of cyberspace as a mediator or means of communication, including the use of text, audio, image and different levels of interaction with the receiver.

ملخص الدراسة :

تقییم الترجمة الألیة لنص من العربیة إلى الإنجلیزیة: منهج تحلیلى تربوى

ناقشت الدراسة الحالیة مشاکل الترجمة التى یواجهها الطلاب المصریین الجامعیین الدراسیین للإنجلیزیة کلغة أجنبیة عندما یترجمون نصا أصلیا عربیا إلى لغتهم المستهدف تعلمها بإستخدام الترجمة الألیة، تم تجمیع البیانات البحثیة إستنادا على نص أصلى باللغة العربیة تمت ترجمته إلى اللغة الإنجلیزیة بإستخدام الترجمة الألیة، و لغرض تحلیل البیانات البحثیة أعتمدت الدراسة على الأسلوب التحلیلى الوصفى حیث تم تحلیل النص الناتج عن الترجمة الألیة باللغة الإنجلیزیة فى ضوء النص العربى الأصلى و الترجمة الإنجلیزیة الصحیحة لذلک النص، و أشارت نتائج الدراسة إلى أنواع عدیدة من المشکلات المتعلقة بتراکیب الجمل و الکلمات و القواعد النحویة التى تؤثر على تعلم مهارات الترجمة من جانب الطلاب المصریین الدراسیین للإنجلیزیة کلغة أجنبیة بالمرحلة الجامعیة، و قد أقترحت المضامین التدریسیة للدراسة حلولا لتلک المشکلات  .