Analyzing Grammatical Errors in the Translations by Bing Translator from Indonesian Texts into French

Abstract

This study examined a few narrative texts accessed online from kompas.com and translated from Indonesian into French with the help of Bing Translator, which is a multilingual machine translation cloud service of Microsoft Corporation. This study used a qualitative approach to analyze the translated texts, with the view to identify the pre-determined phenomena, that of finding grammatical errors. A total of 18 texts were sampled with three different themes, of murder, theft, and accidents (fait-divers texts). The data analysis of the translated texts revealed 60 grammatical translation errors, namely 26 errors related to the theme of murder, 16 errors to the theme of theft, and 18 errors to the theme of accidents. The errors encountered included concordance errors of verbs, noun pluralization, English mixing, indirect speech, lack of prepositions, lack of accord, imparfait du décor/de description, locution prépositive, conjunction errors, word class, and verb construction. The frequency of errors was caused by sentences in Indonesian as the source language being too long, resulting in an ambiguous translation in the automatic translation system into the target language, namely French. The study is the evidence of the inevitability of error making in the use of a machine translator and the need of a human translator for effective translator.