Conflictual Interactions in CMC: The case of Twitter contexts

The study aims to explore the nature and developmental mechanisms of conflictual interactions in one social media context where conflict is prevalent, Twitter, by investigating the role of linguistic strategies and linguistic politeness in such interactions under the constraints and reshaping of Twitter algorithms. It examines the kinds of conflictual linguistic strategies Twitter users employ to perform conflictual interactions, their distribution and how these strategies drive the development of conflictual Twitter interactions. It also delves into the role of linguistic politeness in these strategies and in conflictual Twitter interactions. The analysis is conducted based on a corpus of conflictual Twitter interactions. The in-depth textual analysis specifically designed corpus combines discourse analysis and Brown and Levinson’s politeness theory (1987/1978). It focuses on how speakers design turns, how speakers’ turns then trigger the next turns and finally how these turns are ordered chronologically to accomplish conflictual interactions with varying degrees of intensity. Given the sensitive nature of this analysis, the study employs two approaches: first, it pays close attention to the recipient’s interpretation of the (previous) speaker’s turn in order to gauge interlocutors’ meanings, as suggested by discourse analysis. Second, two follow-up surveys are conducted to complement the textual analysis. The surveys focus on Twitter users’ identification of conflictual linguistic strategies and their understanding of linguistic politeness, respectively. They assess the main findings of this study and provide additional information about language use that is not easily available from the textual analysis alone. Six main findings emerge from the analysis: first, like face-to-face conversations, participants’ roles in conflictual Twitter interactions are also constantly changing, alternating between being recipients and speakers, to issue the next turn. By successively carrying out this role shift, Twitter users cooperatively achieve conflictual interactions. Conflictual interactions continually evolve by users taking turns posting new turns in the same way as the current speaker. Second, conflictual interactions typically arise on Twitter when Twitter users disagree with other users’ viewpoints or detest other users. As the interaction progresses, its intensity then gradually escalates and suddenly drops to zero at some point when one of the participants withdraws from the interaction. Third, in such interactions, the current speaker objects to the previous speaker’s point of view using disagreements and/or personal attacks. Fourth, the literal function of Twitter users’ language use may differ from its interactional function: not all posts taken as a public attack by the recipient constitute personal attacks from a structural point of view; not all posts containing personal attacks are responded to with attacks. Fifth, when the recipient believes that they are being attacked and determines to attack the (previous) speaker, in return employing personal attacks, they tend to normalise personal attacks as an appropriate and effective strategy and to use attacks of a more aggressive nature to ensure that they are heard and that their anger is registered. Finally, almost all of the total posts (including both disagreements and attacks) in the corpus include linguistic politeness which functions to modify (i.e., mitigate or intensify) face-threatening acts. Although linguistic politeness is mostly used to mitigate the possible threat to the recipient’s face, more than one third of the total posts in the corpus including those involving linguistic politeness features still pose serious damage to the recipient’s face and are taken as an overt attack by the recipient.