At the intersection of digital identities and new language and social practice online is the concept of searchable talk (ST). ST describes the process of tagging discourse in a social networking service (SNS) with a hashtag (#), allowing it to be searchable by others. Although originating in Twitter, ST has expanded into other SNS, and is used therein not only to mark language-based posts, but also multimodal posts and images. While scholars have elucidated the structure and function of ST, their studies have primarily examined ST within language-based posts; few have researched ST with respect to images and other types of multimodal environments. In addition, ST has primarily been explored in its SNS of origin, Twitter. This project directly addresses these gaps by adopting a social semiotic approach to ST in three SNS with very different technological affordances, Twitter, Tumblr, and Pinterest. Through a multimodal discourse analysis (Kress, 2009) combining both linguistic and other visual methods, I ask how visual and linguistic choices operate semiotically across SNS environments with different affordances and constraints. Specifically, I uncover the multiple meanings of Beyoncé across a data set of 300 tweets, posts, and pins composed from entering #Beyoncé in the search engine of each SNS. I argue that 13 meaning-based identity categories emerge for Beyoncé, and link these meanings to their visual and linguistic expressions. I then compare these findings across modes and across platforms. Ultimately, I assert that this cross-platform approach elucidates Beyoncé as a cultural object subject to reinterpretation where #Beyoncé means much more than just “Beyoncé.” That is, when considering its multiple roles and meanings, #Beyoncé becomes a site of visual and linguistic indexicality in a process of entextualization. In this process, it is SNS users’ reinterpretations – linguistically and visually – that realize racist, sexist, and hegemonic Discourses, as well as those of emancipation and resistance.

Notre travail de recherche s’intéresse à l’usage des technologies dans les interactions professionnelles, et en particulier dans des centres d’appel. Nos analyses s’appuient sur trois champs disciplinaires que sont l’Ethnométhodologie, l’Analyse Conversationnelle et les Workplace Studies. Notre étude cherche à contribuer à la réflexion sur des interactions médiées par les technologies en milieu professionnel afin de rendre compte des pratiques émergentes des participants et comprendre l’organisation séquentielle complexe des interactions entre conseiller/opérateur et patient/client s’appuyant sur l’utilisation de ressources technologiques. La thèse s’articule en une partie introductive et trois parties analytiques. La première étudie les modifications de cadre participatif en tenant compte du dispositif technologique comme point d’ancrage de l’activité des participants. Nous avons distingué deux configurations: i) soit le dispositif est ajusté par l’opérateur ; ii) soit l’opérateur s’ajuste au dispositif. La modification du cadre participatif peut être initiée de façon verbale ou non verbale, ou par l’un ou l’autre des participants.Dans une seconde partie, nous analysons l’intégration de l’écran comme artefact interactionnel dans l’activité des participants. Nous avons remarqué qu’ils rapportent à l’oral des informations écrites sur un écran, en employant des verbes introductifs du type « il dit que », que nous retrouvons dans les travaux sur le discours rapporté à l’oral. Nous avons voulu montrer le lien entre le cours d’action dans lequel sont engagés les participants et l’émergence de ces discours rapportés où la référence aux messages écrits peut transformer les écrans et les systèmes informatiques en « agents interactionnels ». La troisième partie se focalise sur un type d’appel où un client s’adresse au service pour résoudre un problème, et après vérification par l’opératrice, elle lui notifie un état a-problématique de son compte. Nous avons remarqué que l’activité de diagnostic opérée ici par l’opératrice dépend étroitement des informations du compte client indiquées sur l’écran. Ce sont ces données qui permettent à l’opératrice d’établir le diagnostic et de notifier l’état a-problématique du compte. / Our research focuses on the use of technology in interactions at work, particularly in the context of call centres. Our analyses draw on three theoretical domains: Ethnomethodology, Conversation Analysis and Workplace Studies. Our research seeks to contribute to current investigations on interactions mediated by technology in the workplace to reflect emerging practices of participants and to understand the complex sequential organization of interactions between councillor/operator and patient/user, based on the use of technological resources.The thesis comprises an introduction part and three analytical parts. The first part examines changes in participation framework taking into account the technological device as an anchor for the participant’s activity. For this, we distinguished two different configurations: i) the device is adjusted by the operator, or ii) the operator adjusts the device. The adjustment of the participation framework may be initiated through verbal or multimodal way, or by one or the other participant.In the second part, we analyze the integration of the screen as an interactional artefact in the participant’s activity. We noticed that they report oral information’s displayed on a screen, using introductory verbs such as "he said", which we found in the studies on reported speech in spoken interactions. We wanted to show the link between the course of action in which participants are engaged and the emergence of reported speech when referring to written messages that can transform the screens and the computer systems in “interactional agents”.The third part focuses on one type of call where a user call the service to solve a problem, and after verification by the operator, she notifies a status of his account. We noticed that the diagnosis activity reported by the operator is closely related to the user’s account information shown on the screen. These are data that allow the operator to diagnose and report the non-problematic status of the account.