A New Scenario of Machine Translation: Dynamic Contextual Effects with Diverse Paratextual Application

Abstract

Not many people want to read an incomprehensible machine-created translation (MT) even when it is accessed for free on the web. Thus, pre-editing is done to improve the readability of the MT output. However, syntactic and lexical simplification of pre-editing does not explicate the implicit meanings of culture-specific words and messages. To compensate for the simplification-resulted flaw, this paper proposes the use of some paratexts such as notes and pictures to create a better contextual effect in the MT text. The pragmatic function of paratexts can be justified by framing it within the framework of Gutt’s relevance-theoretic concepts. Breaking free from the unified contextual effect, this paper supports dynamic contextual effect and illustrates how weak, fair and strong contextual effects are created by comparing two sets of three English MT texts that consist of an MT without paratexts, an MT with added notes and an MT with added notes and pictures. The pragmatic significance of this research shows that paratexts play a useful role in MT-enabled communication and that the application of paratextual aids enhances MT audience’s comprehension based on the dynamic contextual effect created in the new MT scenario.