An entropy-based study of simplification in interpreting : a constrained language perspective

Abstract

Drawing on information entropy theory and using wordform entropy and part-of-speech entropy as key indicators, this study analyzes the differences in lexical and syntactic complexity between interpreted English, English as a second language, and native English. In terms of wordform entropy, the findings reveal that interpreted language does not differ significantly from second language output, with both exhibiting simplification tendencies compared to native language. Similarly, for part-of-speech entropy, interpreted and second language exhibit no significant differences, and both are significantly lower than native language. These findings suggest that as constrained languages, interpreted and second languages share common traits and exhibit simplification in lexical choice and syntax. By introducing the concept of “constrained language,” this study argues that both translation processes and second language use face similar challenges: bilingual activation during language contact and heightened cognitive load from real-time processing, which together result in shared linguistic features such as simplification and explicitation. This research not only expands the study of translated language as a “third code” but also reveals that the inherent features of translated language (e.g., simplification) are not merely translation universals but rather reflect cognitive processing mechanisms shared with second language production. These findings provide both theoretical and empirical support for understanding the relationship between translated language and other forms of contact language.