Abstract
As Covid-19 made its way to the shores of Aotearoa New Zealand in 2020, Kiwis took to Twitter to share their experiences and opinions regarding both the virus and government responses. In this paper, we examine a corpus of 1001 tweets to see just how Twitter users utilize different linguistic and politeness strategies when sharing directives conveying stance toward government Covid-19 measures. While there has been research into the use of directives in spoken and written language, there has been little exploration of how directives are used in the language of Social Media. Often considered difficult to classify even in more traditional language genres, Twitter corpora pose additional challenges. We propose a strategy to identify and classify directives using syntactic and pragmatic features, and use this strategy to identify linguistic patterns of both in relation to stance toward government Covid-19 measures. While we find that the most prototypical directive within our corpus is largely consistent with previous definitions, we also find that tweeters exhibit a striking amount of variation in directive strategy, emphasizing the need for a nuanced approach to directive identification and classification. We further note trends in the directive strategies utilized by tweeters expressing different stances toward government measures.