Tracing Machine and Human Translation Errors in Some Literary Texts with Some Implications for EFL Translators

ABSTRACT

This research study aims at drawing a comparison between some internet emerging applications used for machine translation (MT) and a human translation (HT) to two of Alphonse Daudet's short stories "The Siege of Berlin" and "The Bad Zouave." The automatic translation has been carried out by four MT online applications (i.e. Translate Dict, Yandex, Mem-Source, and Reverso) that have come to light in the wake of COVID-19 breakout, whereas the HT was carried out by Hassouna in 2018. The results revealed that MT and HT made some errors related to (a) polysemy, (b) homonymy, (c) syntactic ambiguities, (d) fuzzy hedges, (e) synonyms, (f) metaphors and symbols. The results also showed that Yandex has dealt with polysemy much better than HT in "The Siege of Berlin," but the opposite has been noticed in "The Bad Zouave." Another crucial result is that HT has excelled all MT systems in homonymy and syntactic ambiguities in the two literary texts. A final result is that both MT and HT have dealt with fuzzy hedges at similar rates with little supremacy on the part of Reverso, whereas MemSource and Translate Dict have dealt with synonyms in the two literary texts much better than HT. The study concluded that EFL learners should be aware of the fact that in spite of the advantageousness of MT systems, their inadequacies should not be overlooked and handled with post-editing.