On the mitigating function of the Spanish evidential se ve que

A large part of the bibliography on evidential markers in Spanish points to the mitigation of commitment on what has been said as one of their defining features (García Negroni 2002; Ruiz Gurillo, 2005; Fuentes 2009; Martín Zorraquino, 2005; Kotwica 2013; Alonso-Almeida 2015; Marcos 2016). Nevertheless, it has also been said that the semantic core meaning of evidential constructions is a 'source of information' (Aikhenvald, 2004, Boye 2010, Squartini, 2008, Diewald and Smirnova 2010, Cornillie and Gras, 2015) and, therefore, the mitigation will only be considered as a pragmatic/contextual meaning of those evidential markers (Kotwica 2013, Estellés and Albelda 2014, Alonso-Almeida 2015). Through an extensive study of corpus, this paper aims to discuss the semantic and pragmatic values of the Spanish evidential construction se ve (que) ('apparently, seemingly') in order to pinpoint the role and incidence of mitigation in this construction. Se ve (que) (lit. 'it is seen that') is considered to be an evidential construction in terms of Traugott and Trousdale (2013), with a medium-high degree of syntactic fixation. There exists also the parenthetic variant se ve, which is used in medial and final positions within the utterance. I have analysed in the corpus all the evidential occurrences of se ve (que) that are constructionalized following the criteria established in Albelda (2016a). The corpus consists of a compilation of conversations of the corpus Ameresco (Albelda and Estellés online, esvaratenuacion.es) and of the corpus PRESEEA (Project for the Sociolinguistic Study of Spanish from Spain and America, preseea.linguas.net). In all, around 150 cases of evidential se ve (que) have been retrieved and analysed. Following Delbecque and Cornillie (2008), I distinguish the notions of speaker commitment and speaker involvement, and I apply the mitigating function to both of them. The results of the analysis show that these two notions operate in opposite directions in the evidential subtypes of se ve (que): (1) from reportative evidences to conjectural inferentials (Squartini 2008), there is a scale of lesser to greater speaker involvement; (2) among the inferentials, one can observe a scale of greater to lesser speaker commitment ranging from circumstantials to conjecturals. One of the conclusions derived from the previous results is that the mitigating meaning is not uniform in se ve (que); thus, we should consider it only as a pragmatic value but not a semantic one.