Reflection of Explicitation in Scientific Translation: Neural Machine Translation vs. Human Post-Editing
Year: 2024
Type: article
Abstract: The present paper reports on the findings of an empirical comparative study on the extent to which explicitation is employed in the translation of a scientific text as conducted by Google Neural Machine Translation (GNMT) vs its post-edited (PE) version. A recent report released in English by the World Meteorological Organization in September 2023 was selected as the source text for the present study. The purpose of the study is to reveal how domain-specific acronyms and technical terms are lexically expanded (explicitated) in a GNMT output compared to its post-edited (PE) version as performed by a team of professional translators at a translation service provider in Amman-Jordan. Explicitation in translation can be obligatory or optional. The type of explicitation investigated in the present study is optional, pragmatic explicitation. The results show that GNMT has its limitations in dealing with scientific terms and acronyms in translating scientific texts from English into Arabic. In contrast, human post-editing -which used a raw GNMT output in the initial stage- explicitated domain- specific terms and acronyms in its post-editing version producing a text with a higher level of readability and naturalness for domain expert readers and non-expert readers.
Author: Ogareet Y. Khoury
Institution: Al-Ahliyya Amman University
Language: English
Subfield: Language and Linguistics
Field: Arts and Humanities
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W4402100932