Evolving interactional practices of emoji in text messages

In this article, we examine the usage of emoji in the 88milSMS corpus. After differentiating between emoji and emoticons, we situate the context, indicate general statistics and mention press interest. Next, we address linguistic issues: are emoji used more often in addition (either redundantly or necessarily, sometimes as “softeners” (adoucisseurs, Détrie & Verine 2015) or for lexical replacement, denoting a reference/referential function (Referenzfunktion, Dürscheid & Siever 2017)? Concerning emoji insertion positioning, which is the most popular and what does this mean? Other researchers refer to “the emoji code” (Danesi 2016; Evans 2017), and emoji classifications have been proposed, including references to syntactic, semantic (Barbieri, Ronzano & Saggion 2016), semiotic, phatic and emotive/sentiment (Novak et al. 2015) levels. Are these satisfactory or do we need to redefine levels, contexts and potential ambiguity? Part-ofspeech tagging (POS) and NLP software are then used to annotate SMS containing emoji within 88milSMS in order to investigate the immediate grammatical environment. This allows us to conduct contextual analysis relating to syntactic linguistic functions of emoji. Finally, results from two questionnaires are explored: 1. sociolinguistic factors (age, gender) of the SMS donors having used emoji in 88milSMS; 2. Comparison of SMS emoji usage with other instant messaging applications and social networks via a user-orientated questionnaire (Rascol 20171).