A Critical Appraisal of Baker's Model of Equivalence with a Special Focus on Machine vs. Human English into Arabic Translation: Harry Potter Extracts Case Study

In the translation process, there is a relationship between the source text and target text, commonly known among researchers as ?equivalence?. This article critically analyses the concept of equivalence proposed by notable translation theorists, with particular attention paid to the applicability of Baker's 1 model of equivalence to the quality of the Arabic machine translation provided by a web-based platform (i. e. Google Translate) of several Harry Potter2 extracts compared to their Arabic human translations. The concept of equivalence between the source and the target texts was a central concept within the discipline of translation in the 1960s and 1970s. An examination of the literature reveals that translation scholars have attempted to differently theorise equivalence, along with its applications within the translation process. An influential model proposed by Baker1 comprises four different levels: word level, above word level, grammatical level, and register level. This article critically analyses the machine translations of five extracts randomly selected from ?Harry Potter and the Philosopher's Stone? 2 (i. e. two extracts at the register level, two extracts above word level, and one extract at the grammatical level), and compares them with their counterparts in the human translation. Ultimately, the quality of the machine translation and the human translation are compared and evaluated in accordance with Baker's 1 model.