Types of Nods. The polysemy of a social signal Isabella Poggi, Francesca D’Errico, Laura Vincze Roma Tre University Via Milazzo, 11 00185 - Roma E-mail: poggi@uniroma3.it ; fderrico@uniroma3.it ; laura.vincze@gmail.com The work analyses the head nod, a down-up movement of the head, as a polysemic social signal, that is, a signal with a number of different meanings which all share some common semantic element. Based on the analysis of 100 nods drawn from the SSPNet corpus of TV political debates, a typology of nods is presented that distinguishes Speaker’s, Interlocutor’s and Third Listener’s nods, with their subtypes (confirmation, agreement, approval, submission and permission, greeting and thanks, backchannel giving and backchannel request, emphasis, ironic agreement, literal and rhetoric question, and others). For each nod the analysis specifies: 1. characteristic features of how it is produced, among which main direction, amplitude, velocity and number of repetitions; 2. cues in other modalities, like direction and duration of gaze; 3. conversational context in which the nod typically occurs. For the Interlocutor’s or Third Listener’s nod, the preceding speech act is relevant: yes/no answer or information for a nod of confirmation, expression of opinion for one of agreement, prosocial action for greetings and thanks; for the Speaker’s nods, instead, their meanings are mainly distinguished by accompanying signals. Introduction When people discuss or argue, they may express their agreement by signals in various modalities – words, gesture, intonation, face, gaze, posture, head movements. They may look at you while smiling, clap their hands, tell you “bravo!” or simply nod. Nodding, that is, moving head slightly up and then down, is a polysemic signal, since it may have various meanings. It has been studied in the context of other head movements (Hadar et al., 1985; Kendon, 2002; Mc Clave, 2000; Heylen, 2005; Cerrato, 2005; 2007). In this work we investigate the communicative functions of nodding and its use in debates. Based on the analysis of a corpus, we propose a typology of nods and specify the aspects that distinguish its different types in face-to-face interaction. 1. Polysemy and body signals A typical aspect of signals, often analyzed in verbal language, is ambiguity: a signal (a perceivable stimulus produced by a Sender to convey some meaning) may correspond to more than one meaning. Sometimes the meanings of a signal have no relation with each other – like in homophony, e.g. bear and bare sound the same but convey very different meanings; but sometimes the different meanings share some semantic element – like in polysemy, e.g. ring as the square for a boxing match, and as a round jewel on your finger. The meanings of a polysemic signal are its “range of polysemy”. Like any kind of ambiguity (see figure-ground ambiguity in perception, or syntactic ambiguity of sentences), the polysemy of a word in a verbal language is generally overcome by considering context: I can tell which ring you are meaning now if you are talking of a boxing match. In polysemy there is something in common between the two or more meanings, but one or the other is “triggered” by the context. Take yes, which is an information when the previous context is a yes/no question, a permission if it follows a permission request. In both cases the present speaker B by saying yes confirms some belief hypothesized by the previous speaker A; but in the former context the belief concerns some information that A hypothesizes as true, so yes confirms it is true; in the latter, the belief hypothesized is that B does like/want A to do action X. So, the polysemic signal provides a single piece of meaning – B’s confirming some belief – while context adds another piece of meaning – whether the belief confirmed concerns some information or B’s willingness to let A do some action. This is generally how things go with polysemy in verbal languages, where a word must be uttered or written in precisely that way to be that word (i.e., to have that meaning). But a hand gesture or a head gesture generally mix up with simultaneous signals in other modalities, and this does make a difference; for example, in backchannel it is not the same if I nod while smiling or not (Bevacqua 2009). Moreover, nonverbal signals can be produced in a number of ways, depending on whether the movement is relaxed or tense, single or repeated; in the gesture “come here” (hand palm down with fingers flapping downward), it is not the same if I make it with a soft fluid movement or with hectic jerks (Poggi & Pelachaud, 2008). So what counts as context for words is something coming before or after the word itself, while for a smile or a nod, that are generally produced simultaneously with other signals, what counts as context may be the signals in other modalities. Consequently, if we want to disentangle the different meanings of the nod – its rich polysemy – but at the same time to understand what is common and what is different in them, we must 1. find out all the possible meanings of the nod, its “range of polysemy”; 2570