Drawing on cognitive-functional theories, pragmatics and conversation analysis, this dissertation offers anintegrated account of speech participant profiling in Spanish spoken interaction. Adopting an onomasiological perspective, I include various devices in my analysis, such as person deixis, devices that are linked to person deixis (such as vocatives, quantifiers) and indeterminate constructions. Through thecontrastive study of three types of multi-person interaction (informal conversations, TV-debates and parliamentary debates), I offer an integrated account of speech participant profiling in interaction, and I show how speech participant profiling contributes to an intralinguistic (as opposed to sociological) definition of genres. The notions of subjectivityand intersubjectivity (Benveniste 1966, Lyons 1982, Traugott 2003) are crucial in this analysis. The first level of analysis (and which has been privileged in the literature) is the identification orreferentiality. The discourse-driven approach I adopt reveals that forms that at first sight have a vague reference, such as indeterminate constructions, often receive a more specific reading in view of the scope restriction established at utterance- or discourse-level. Conversely, (plural) deictic forms, despite being given in context, present a certain degree of conceptual vagueness (though, here as well, scope restriction may operate). Given the referential overlap that arises between different constructions at discourse-level, I argue that the choice for a specific person reference device is then not (only) a matter of identificationbut also, and foremost, a matter of conceptualisation and speech participant profiling. At a second level, the analysis of the predicates with which the different constructions pattern shows that the choice of the construction is not indifferent for the meaning the predicate conveys at discourse level.More specifically, singular addresser and addressee deictic forms are privileged loci for the expression of explicit (inter)subjective meanings. The conceptual vagueness of 1st person plural deictics and of indeterminate constructions, by contrast, makes them particularly useful for theexpression of deontic meanings. The establishment of a contextual identity (Kerbrat-Orecchioni 2005) is achieved both at individual and at group-level. A third analytical layer is the effect of speech participant profiling on interaction structure. Both at turn-level and at discourse-level, speech participant profiling crucially influences turn-distribution and participant ratification (Goffman 1979). Deictic forms are good candidates for turn-taking and -giving. Also some non-deictic forms, such as vocatives, are crucial in this respect and, moreover, significantlycontribute to participant ratification, i.e. signalling that a participant is being addressed. Indeterminate constructions, by contrast, need amore specific anchoring and are bad candidates for turn-initial position. The notion of (inter)subjectivity has proven useful to account for the choices speakers make as to speechparticipant profiling and for the distinction between different genres.Whereas the three genres (whether planned or unplanned) clearly share the characteristics of interactive discourse in their frequency of deictics (if compared to non-interactive or written genres), the qualitative analysis reveals crucial differences between them. In line with research on other language, (inter)subjectivity is found to be very significant in informal conversation. Explicit stance-taking or stance-requesting, bycontrast, is most frequent in the TV-debate. When the speaker explicitly operates as a group representative, as inparliamentary debate, this is reflected in the use of plural forms. They then appear in contexts where in other genres a singular form appears,e.g. with cognition verbs. In conclusion, I have shown that it is profitable to combine different analytical layers and to look beyond mere person deixis in order to fully account for speech participant profiling in interaction. Moreover, it is now clear that the use of speech participant profiling plays a crucial rule in the configuration of different interaction genres. ReferencesBenveniste, Emile. 1966. De la subjectivité dans le langage. Problèmes de linguistique générale. Paris: Editions Gallimard.Goffman, Erving. 1979. Footing. Semiotica 25(1-2): 1-29.Kerbrat-Orecchioni, Catherine. 2005. Le discours en interaction. Paris:Armand Colin.Lyons, John. 1982. Deixis and subjectivity: Loquor, ergo sum? Speech, place and action. Studies indeixis and related topics. eds. R. J. Jarvella, and W. Klein, 101-24. Chichester: John Wiley & sons ltd.Traugott, Elizabeth Closs. 2003. From subjectification to intersubjectification. Motives for language change. ed. Raymond Hickey, 124-39. Cambridge: Cambridge University Press.