Los truncamientos en la conversación coloquial: estudio de las huellas de formulación discursiva desde un modelo de unidades de lo oral

One of the principal characteristics of colloquial speech compared with other discourse genres is its spontaneity and the fact that it is constructed on the spot, without any prior planning (Beinhauer 1963[1929], Ochs 1979, Narbona 1992, Briz 1998, among others). The spontaneity of dialogue leaves numerous traces or signs in the resulting speech––ruptures, self-repairs, abrupt changes of construction, incomplete segments––for which we use the term speech-formulation phenomena, given that they manifest the processes of formulation underlying the extemporaneous construction of discourse. This thesis focuses on a specific speech-formulation phenomenon called morphosyntactic truncations. Morphosyntactic truncations can be defined as speech segments that are interrupted and that manifest some kind of morphosyntactic change with respect to the overall structure of the speech. We understand spontaneous colloquial discourse as a dynamic, unfixed form that is constructed and reconfigured in a non-linear fashion by speakers as they adapt to the constraints of the communicative and interactional situation. Contrary to perspectives such as the generativist, which relegates truncations and other speech-formulation phenomena to the status of mere “performance errors” (Chomsky 1965: 4, in Eklund [2004: 88]), we advocate their significance and legitimacy as an object of study. The study of speech-formulation phenomena provides a valuable lens through which to view discourse and, at the same time, reframes our understanding of discourse as a process rather than a product. The study of these phenomena enables us to unravel the processes underlying the construction and formulation of discourse, that is, the processes of syntax in progress or on-line syntax (Auer 2009, Linell 1982, López Serena 2017). Moreover, given that that the units of traditional syntax have proved unsuitable for studying certain phenomena of spoken, colloquial language (Narbona 1992, Hidalgo and Pérez 2004), we sustain that the appropriate theoretical and methodological framework for undertaking the analysis of truncations is that of the new models of units designed for segmenting spoken language (Pons 2014a). This thesis has two main aims: first, to design a typology and a theoretical characterization of speech-formulation phenomena and, more specifically, of truncated segments; second, to develop a proposal to integrate truncated units in the description of Spanish colloquial syntax within the framework of a system of conversational units: the model formulated by the Val.Es.Co. group (Briz and Grupo Val.Es.Co. 2003, 2014). In summary, this thesis seeks to contribute to the study of truncations, which are understood as a trace of the formulation of speech that enables us to discern the wider processes of speech construction. The thesis works with the theoretical framework of a system of units that can be implemented in order to accommodate these segments and thereby achieve a complete segmentation of colloquial conversation in Spanish. Based on the analysis of 19 colloquial conversations from the Corpus Val.Es.Co. 3.0, a method was designed to identify the different types of truncations and to analyse their structural, morphosyntactic, prosodic, and semantico-pragmatic features. Multivariate statistical analyses (multiple correspondence analysis, cluster analysis, and decision tree) complement the linguistic description of truncations by determining the statistical significance of the linguistic features analysed. These statistical analyses enable us to determine which of the linguistic features analysed are more or less significant for the characterization of different types of truncation. The results of this thesis comprise i) a classification of speech-formulation phenomena, ii) a classification of the different types of truncation and iii) a method to integrate truncations in the Val.Es.Co. system of conversational units. In conclusion, this research has enabled the development of our understanding of the processes involved in the planning and construction of colloquial speech, with the ultimate aim of advancing a theory about formulation and its phenomena. In this way, this thesis seeks to justify the importance of analysing truncated units, which are often considered mere syntactic residues, as well as to demonstrate the applicability of the syntactic theory of spoken language using the Val.Es.Co. system.