How to be impolite with emojis: A corpus analysis of Vietnamese social media posts

This study addresses a critical gap in the existing literature by investigating the manifestation of impoliteness through the use of emojis within the online Vietnamese community on social media. The research is guided by three central questions: (1) How do Vietnamese Facebook users use emojis in their posts and comments? (2) How do Vietnamese Facebook users perceive impolite behaviors in cyberspace? and (3). What strategies do Vietnamese speakers employ to express impoliteness with emojis on social media? Quantitative and qualitative analyses were performed on a corpus of posts and comments on a Facebook showbiz confession page. Results show that facial emojis, particularly those forming homogeneous sequences, are preferred, with laughter-related emojis prominently featured. Additionally, emotive particles, together with expletives, frequently co-occur with emojis, compensating for absent extralinguistic cues in computer-mediated communication. By administering checks using dictionaries, mutual information scores, collocation visualizations, and cosine similarity, a nuanced understanding of impoliteness in CMC was achieved. Religious influences, particularly from Buddhism, were found to play a significant role in shaping Vietnamese impoliteness perception, exemplified by terms such as vô duyên and sân si. A coding scheme informed by findings from the second research question on a sample of 100 first posts and comments in the main corpus was used. The study further substantiates the hypothesis that Vietnamese speakers predominantly employ implicational impoliteness strategies, particularly through multimodal mismatches facilitated by emojis. Conventionalized formulas featuring emojis were infrequent, suggesting a preference for more dynamic and context-specific impoliteness expressions. This research contributes to the refinement of impoliteness theoretical and methodological approaches, as well as providing a foundation for further studies in online discourse and natural language processing.