Corpus studies of emotions
In this last meeting of the Spring semester of Travelling Emotions, we will have an external guest invited by Multiling, followed by another presentation on a corpus-based study by Anne Golden:
1615 -1715 Ng Bee Chin. "The Semantics of Chinese Emotion – A Corpus Study"
1725- 1800 Anne Golden. "Emotions negotiated in L2 texts. A corpus study on written production by adult learners in a Norwegian test".
The invited lecturer:
Ng Bee Chin works mainly in the area of bilingualism and multilingualism with a focus on the impact of language contact on individuals and the community they live in. Her research approach is to explore both cognitive and social aspects of language acquisition and use. Currently, she is working on language identity, attitudes and use and language and emotion in multilinguals. She also works in the area of language as a source of intangible heritage with collaborators in art and design studies. She currently works in the Division of Linguistics and Multilingual Studies in Nanyang Technological University she is also the Associate Dean of Graduate Studies in the College of Humanities and Social Sciences.
Abstract of her talk:
This study aims to explore language specificity in the semantic organisation and distribution of emotion words in Mandarin Chinese. While prior studies have made anecdotal references to the prevalence of use of verbs in some languages (e.g. Russian and Mandarin Chinese) for expressing emotion words compared to English, this has not been supported by empirical evidence involving a comprehensive study of the emotion words vocabulary. To date, despite a proliferation of cross linguistic studies of emotion words, a database of a corpus of emotion words across languages is absent. A more acute problem in the field is the lack of comparable ways of identifying emotion words.
Using the framework proposed by Pavelenko (2008), emotion words in Mandarin Chinese are extracted and sorted into three semantic categories; emotion words, emotion laden words and emotion related words. Each word in the corpus is also tagged for frequency of use, valency, intensity and parts of speech. Each emotion word was also tagged for the broad emotion category (e.g. pride, shame, guilt, anger, joy, disgust etc.) it is a member of. Corpus data analysis method was then employed to study the patterns of the data. Not surprisingly, consistent with other reports on Mandarin Chinese in other domains (e.g. acquisition), it was found that verbs occupied the biggest percentage in both emotion words and emotion-related words categories. An analysis of the valence and intensity of emotion words shows crosslinguistic divergence from other studies reported. The varying distribution and behaviour of emotion words in the three categories identified by Pavlenko lends support to her proposed categorisation of the emotion lexicon. The study also represents a significant attempt at providing a working template for the identification of emotion words in emotion research.