Applying Linguistics: Analyzing the Differences Between Human and Machine Translation of Selected Texts
Year: 2016
Type: article
Abstract: Machine translation systems experience a trade-off between efficiency and efficacy: while quicker and easier than human translation, machine translations are less accurate and comprehensible. This study analyzes two sets of English to Spanish translations to compare the differences between human and machine translation and evaluate the performance of machine translation. Google Translate—the best-performing machine translator—represents machine translation as a whole in producing English to Spanish translations of the given text. The primary investigator—a non-native speaker and graduating student of Spanish Language, Literature & Culture at Syracuse University—serves as the benchmark for the average non-native speaker of Spanish at the advanced-mid level of proficiency. Findings are divided into four groups: lexical findings, sentential findings, phrasal findings, and stylistic findings. From these findings, the study is able to establish that machine translation struggles to decipher context, which results in errors in word choice and lack of proper technical language; displays inconsistency in its errors, and lacks the capability to learn. Such individual findings led to the overall conclusion that Google Translate—and thus machine translation—performs below the level of an average non-native speaker of Spanish.
Author: Nicole Howell
Language: English
Subfield: Language and Linguistics
Field: Arts and Humanities
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W2514264285