A Corpus-Driven Approach on Learning Near Synonyms of Pain in Indonesian

Abstract

Pain is human-universal since it is experienced by people across the world. However, since it is related to personal feelings, different people may feel it in a different way and rely on language to communicate. This paper presents a cross-cultural comparison of the study of the emotion of pain in Indonesian by examining the usage of two near-synonyms: sakit and nyeri. This study aims to provide a new insight for L2 learners of Indonesian regarding the study of emotion. A corpus-driven method by using the usage-feature analysis (Glynn, 2010b) is employed to test the hypothesis on the semasiological structure of pain from Indonesian dictionary. The corpus data of Indonesian News 2020 with a total of 15,206,710 tokens were extracted from the Leipzig Corpora Data Collection of Indonesian (Goldhahn et al., 2012). A total of 400 examples of sakit and nyeri were extracted from the corpus data using AntConc version 4.1.2 (Laurence, 2022) for manual annotation. The manual coding of the lexemes was conducted based on cross-linguistic dimensions of pain proposed by Wierzbicka (2016). After manual annotation, two statistical analyses were conducted in R (R Core Team, 2022), namely Binary Correspondence Analysis (Glynn, 2014) and Binomial Regression Analysis (Levshina, 2015). The result of exploratory analysis shows that sakit and nyeri can be distinguished by bodily focus and intensity. However, the confirmatory analysis confirms bodily focus as the significant predictor. It means nyeri is strongly associated with pain on the part of body relative to sakit. The finding of the current study may have an implication for the possibility of combining cross-cultural competence with L2 vocabulary learning by making use of corpora in L2 learning design.