La structure actantielle des discours de haine dans les plateformes participatives en ligne

The aim of this article is to show the interest of analysing online hate speech using an actantial approach. Our premise is that definitions of online hate speech are mainly based on the identification of & quot;hateful & quot; keywords. However, in order to understand and analyse hate speech, this method is not sufficient, for three reasons: firstly, because several online information media introduce moderation practices to censor insults and swearwords; secondly, because a good part of disqualifying, discriminating, etc. speech is based on implicit, euphemistic and unwritten language; thirdly, because this approach does not inform us about the narratives, the thought patterns and the underlying reasoning -even if fallacious -that underlie hateful speech. In order to decipher these narratives, it is necessary to go beyond the lexical level, or even beyond the sentence, to look at the commentary as a complex unit of meaning. Our contribution will focus on a specific case, that of hate speech against migrants. We will show that the enunciators of hate speech intervene in the narrative construction in two ways: they identify third-party actants and thus construct micro-narratives of hate; they directly assume an actantial role within a specific scenographic device.