The Performance of Machine Translation System and Human Translation with Political texts” (2013), Reem Bazzal, Maha Atwi,Nour Al Onies, Inaam Kotiesh and Ghina Khamis

Abstract In their text “The Performance of Machine Translation System and Human Translation with Political texts” (2013), Reem Bazzal, Maha Atwi,Nour Al Onies, Inaam Kotiesh and Ghina Khamis intend to discuss the proper use of available online machine translation technologies and human in political translation and provide an objective view of its suitability and reliability. They examined the services of such technologies through the use of google translate translation and of the human translation andevaluated their work from four different linguistic aspects: syntax, morphology, semantics and comprehension. Atwi, Bazzal, Khamis, Kotiesh and Onies concluded that although machine translation systems are useful and largely utilized in economic, journalistic and literal fields, they cannot be used as a resource for translation when it comes to political texts mainly because of the incompatibility in semantics and pragmatics for they give improperly defined words andgrammatically wrong sentences. This paper is set out to verify this belief. Commentary To illustrate more, practical text has been submitted and it was translated by the Machine Translation which is google translate and by a human translation. The text they chose العولمة is made up of 250 words in its English version. It is therefore a reliable text set for examining the translation process and the linguistic rules translation ofpolitical text by MT systems and to produce substantial statistical analysis for their performance. After translating the text,the results were the following: Semantics: When it comes to semantics, the meaning or the interpretation of a word or sentence we noticed that unfortunately, machine translation translated some words and expressions literally and gave different meanings and translation too. For example Source text: Human Machine العولمة globalization العولمة مقابلات equivalents interviews تعميم generating circular heye عالمي worldwide the whole أجاز allow to leave use توجيهاواحدا one direction directed guidance مجمعاللغةThe Academy ofThe Arabic توسيعspreadexpanding لعربيةفيالقاهرة The Arabic Language Language Academy in Cairo In Cairo مصطلح term word توجيها sharing guidance كوكبة Planetary constellation Syntax: Concerning the word order, the human translation showed impressive results in giving proper sentences that follows the grammatical rules: Subject – Verb- Object ( in translation to English) and in the use of modifiers and determiners while when it came to machine translation. The sentences were weak, lengthy and didn’t follow the grammatical rules. • Source Text: و قد قرر مجمع اللغة العربية في القاهرة إجازة إستعمال كلمة العولمة Human Translation:The Academy of the Arabic Language in Cairo had decided to allow the use of the word globalization Machine Translation: and has decided the Arabic Language Academy in Cairo leave use the word globalization. • Source Text: الحقانهمنالصعببمكان،إعطاءتعريفدقيقللعولمة بسببكثرةالتعريفاتبشأنها Human Translation:It is quite difficult to give a precise definition of “Globalization” because it has so many definitions Machine Translation:Right it's very difficult, to give a definition precise of globalization because of the large number of definitions which Morphology: Concerningmorphology, machine translation has a problem when it comes to translation related to Inflectional morphemes, the exact number, gender of the subject and the exact verb. When translating from Arabic to English, the agreement between grammatically linked items was often missing which led to some sentences that can’t be understood. While human translation has proved its efficiency in giving the exact verb tense number, possession, and comparison. Examples: • Source Text: الإنكليزية, و التي تعني تعميم الشيء و توسيع دائرته ليشمل الكل العولمة هي إحدى المقابلات العربية لكلمة globalization Human Translation: “globalization is one of the Arabic equivalents of the English word “Globalization” that means generating and spread a thing Machine Translation: Globalization is one of the interviews; the Arabic word for the English and globalization means which circular Heye and expanding his circle to include the whole. Source Text: وقدقررمجمعاللغةالعربيةفيالقاهرةإجازةإستعمالكلمةالعولمة، بمعنى جعل الشيء عالما واحد، موجها توجيها واحدا في إطار حضارة واحدة و لذلك قد تسمى الكونية او الكوكبة Human Translation: The Academy of the Arabic Language in Cairo had decided to allow the use of the word “العولمة” that signify making something on an international level. Thereupon, globalization is an expression that connotes the idea of turning the world to a one single world, heading to a one direction of sharing one civilization so it might be called Cosmic or Planetary concept. Machine Translation: and has decided the Arabic Language Academy in Cairo leave use of the word globalization, in the sense to make global thing. if the term globalization means making the world on world, one direct guidance in the context of one civilization may therefore called cosmic or constellation. Comprehension: Differences العولمةهياحدىالمقابلاتالعربيةلكلمة (globalization) الانكليزية The machine translation result:Globalization is one of the interviews, the Arabic world for the English and globalization While human translation:Globalization • The original text:تعنيتعميمالشيءوتوسيعدائرتهلتشملالكل Machine translation:Which means circular Heye and expanding his circle to include the whole Human translation: That means generating and developing a thing to become as a unit. • The original text: العولمةاذامصطلحيعنيجعلالعالمعالماواحدا Machine translation:If the term Globalization means making the world one world. Human Translation:Globalization is a term that means turning the world to one unit. Analysis From here, we can find that the human translation is different than the machine translation; for the human translator has a mind and can comprehend and know how to make his article coherent and free of errors. The machine translated documents in the political field are incoherent and lack the sense of comprehension and grammatical coherence. There are many differences between human translation and machine translation, in which errors regarding morphology, syntax, semantics, mechanics, coherence and comprehension are being found. The word choice in machine translation is very weak and the meaning is sometimes vague unlike the human translation which possesses an excellent word choice and uses exact terms that makes the meaning clear. The word order in machine translation is unorganized which makes the sentence barely readable whereas in human translation the word order is well-organized which makes the sentence very clear. Moreover, machine translation has weak grammar such as in sentence structure. . So in machine translation a lot of grammar mistakes are made while a human translator avoids the grammatical mistakes. In human translation the original text is 90% compatible with the target text whereas in machine translation it is 50% compatible, the text might be explicable but barely readable. Recommendation Political translation is a challenging one; it requires highly specialized terminology which can't be comprehended by the translator easily. Specialized translators in the political field also face many problems in translating these documents for what they need from paying attention to every little letter, since any change in the translation may change the whole context; hence most of the translators depend on the machine to find out the specialized terminology. Although the machine helps translators translate their documents using the most accurate terms, they can't depend on it completely since it doesn't comprehend the grammatical, structural and comprehensive errors, so if the translator pays attention to these errors, the document will be a high-quality one, it will contain the most accurate terms, finished in less time, earning high revenue, and will be a reliable one. Conclusion It was conspicuous that the machine translated documents are incoherent and full of grammatical and structural errors, but the reader is on the safe side for the terms are accurate and the sentences can be easily comprehended. Although some decoys can be found in the machine translated documents because of the structural changes, these are no more than 40% of the whole document, hence here comes the role of the human translator to detect the errors and correct them. And whenever the translator do his job in a way using his translating skills with the help of the machine, his job will be achieved perfectly without any errors or changes in the meaning. References: Atwi.Maha, Bazzal. Reem, Khamis. Ghina, Kotiesh. Inaam and Onies. Nour. (2014). “The Performance of Machine Translation System and Human Translation with Political texts”. Lebanese International University. Issue 12