With the advancement of translation technology,translators have transformed into Translation Service Providers(TSPs)who are probably spending more time on revision and revision-related activities such as post-editing.The urgent market demand has encouraged more research on translation revision and post-editing competences.However,few empirical studies of those two competences have been conducted.Most of them rely on product data and the results of previous studies are mixed.Therefore,it is necessary to conduct an empirical research to further investigate the difference between translation revision and post-editing competences.Moreover,neural machine translation(NMT)might make it possible to apply MTPE into literary translation.Previous comparative studies between MTPE and human translation mainly focus on non-literary translation.But there is a concern whether the quality of post-edited literary NMT will exceed that of revised literary HT.Therefore,the present study has adopted research methods like questionnaire,key logging and retrospective protocol so as to investigate the differences of translation trainees’ behaviors in revision and post-editing of literary translations and explore the application prospect of MTPE in literary translation.The research aims to answer the following questions:(1)How do participants rectify literary texts in revision and post-editing?(2)Compared with revision,what is post-editing effort spent by participants in post-editing of literary texts? Specifically,how much temporal,technical and cognitive effort do translation trainees spend in translation revision and NMT post-editing respectively?(3)Compared with revised text,what is final translation quality of post-edited literary texts? Specifically,what are the adequacy and fluency of revised and post-edited literary texts?The present study has collected questionnaire data from 48 respondents and invited 21 among them to involve in translation revision and post-editing tasks.The results showed that:(1)From the perspective of intervention behaviors,the number of changes made by translation trainees in NMT post-editing is one and half of that in translation revision.Translation trainees obtained higher intervention quality in post-editing than in translation revision.(2)As for post-editing effort,translation trainees indeed spent much more temporal,technical and cognitive effort in post-editing than in translation revision.(3)When it comes to translation quality,the average adequacy and fluency of revised HT and post-edited NMT are close to each other and there is no significant difference.To the end,the present study will provide some suggestions for related empirical research in terms of methodology.The findings of the present study will benefit for the translation training. |