Cohesion in Human and Machine Translation | ||||
| CDELT Occasional Papers in the Development of English Education | ||||
| Article 5, Volume 76, Issue 1, October 2021, Page 101-133 PDF (1.41 MB) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/opde.2021.226278 | ||||
 View on SCiNiTO | ||||
| Author | ||||
| Mohamed Tohamy | ||||
| Abstract | ||||
| Machine translation technology has consistently improved to the extent that it is frequently compared with human translation performance. This study reports a simple comparison of two translations of an Arabic literary text: a short story written by the Egyptian writer Yousef Idris. One translation is human made and the other is a machine translation using Google Translate. The study focuses on the use of some cohesive devices in the two translated texts. Two types of grammatical cohesion were selected for analysis, namely, reference and conjunction. The findings of the study show that cohesive properties vary according to the translation method involved, wither it is human or machine translation. The results obtained are indicative of the limitations of machine translation to produce a reliable translation of a literary work. Unlike the human translator whose use of cohesive devices is balanced and fairly distributed to the text, Google Translate uses such devices excessively, which makes the machine translated text incoherent, loose and ambiguous | ||||
| Keywords | ||||
| Machine translation; human translation; Google Translate; Cohesive devices | ||||
|
Statistics Article View: 358 PDF Download: 310 |
||||
