Evaluation of Attitudinal Lexis in Twitter Political #Hashtags: A Corpus-Based Appraisal Analysis

Abstract

This research is a corpus-based analysis of a trending topic marked by the dominating hashtag #MuslimBan to investigate the influential role of evaluative language in social media. It combines the Appraisal theory of Martin and White (2005) as rooted in the systemic functional grammar of Halliday & Matthiessen (2004) and corpus linguistics to analyze a mini-corpus of 4082 tweets posted on January 28th, 2017 using AntConc for Word List, Keyword List and Concordance tool. The main corpus is tested against a reference corpus of news articles from New York Times newspaper. The aim of the study is to explore how a stance is created on social media discourse at a lexicogrammatical level and how this helps unite the microbloggers. Findings revealed that the main corpus abounds in much negative appreciation and judgment of the trigger (the MuslimBan order). It also showed that most evaluation falls in the Attitude category of Martin and White (2005) as emotional release is usually expressed via the frequent use of adjectives, whereas physical release of anger is usually expressed in using action verbs. The lexical analysis also revealed that the interpersonal function is more vigorous in the main corpus and that social media is more influential in spreading slogans and communicating stances than news articles.