Fuzzy Limits: Researching Discourse in the Internet with Corpora

Abstract

Internet has provided us with an amount of linguistic data without precedents. For those who research discourse and communication, it is an unexpected gift with a huge potential. However, this gift comes with important challenges we have to face. First, large corpora make us to use quantitative methods in fields where we were used to qualitative approaches. In order to change it, new strategies are being developed, such as the Corpus Assisted Discourse Studies (Baker et al. Discourse Soc 19(3):273–305, 2008; Partington et al. Patterns and meanings in discourse. John Benjamins Publishing Company, Amsterdam, 2013). Secondly, traditional units of analysis need to be redefined. Communication through Internet has its own characteristics, and some of them do not fit in previous definitions. There are two main reasons for this regarding discourse analysis. On the one hand, current interactions are multimedia. Video, image, and sound are not necessarily subordinated to text in Internet, and researchers “need to look beyond language to better understand how people communicate and interact in digital environments” (Jewitt. Multimodal analysis. In: Georgakopoulou S (ed) The Routledge handbook of language and digital communication. Routledge, London, 2016). Recent approaches, such as Multimodal Critical Discourse Studies (Machin. Crit Discourse Stud 10:347, 2013), move in this direction. On the other hand, limits have become fuzzy. Interactions in Internet work in new ways, even when we call them conversations or chats (Alcántara-Plá. Estudios de Lingüística del Español 35(1):214–233, 2014). If we study them with our current units of analysis, these “conversations” will seem fragmentary and unstructured. In this chapter, we describe these new challenges and the solutions that have been adopted so far, drawing attention to the major problems that still remain unsolved.