The aim of the present work is to describe the semantics and the discursive functions from a general cognitivist point of view of the usage of the Present Perfect in the spoken Spanish of Chile, Paraguay and Uruguay. It is argued that cross-linguistic values often ascribed to perfect, such as continuity, current relevance and recency to the speech time –ST– do not offer a consistent view of the actual usage. It is assumed that a basic meaning of the perfect operates in the studied dialects and is retrievable in all tokens, which differs significantly from the current descriptions of the perfect of “general” Spanish. The results show that the ST might very well be an inference of the basic meaning of the Perfect but it is not an intrinsic component of the Perfect’s semantics. Based mainly on Dahl & Hedin (2000), as well as on Langacker (1987), the revitalizing of the concepts type and token reference are suggested as key principles for identifying the respective domains of the Spanish Present Perfect and the Simple Past in the studied area. The Perfect, through type reference, makes an assertion of a situation as a representation of the class-type of the verbal semantics. The Simple Past, however, through token reference conceptualizes the situation as having explicit or implicit anchoring in the chronological axis of time. Three main kinds of contexts occur typically with the Perfect in the samples: detemporalized ascertainment, summary (in a broad sense of the word) and aspectual complexity. Summary scanning (Langacker, 1987), i.e. the schematic and holistic detemporalized conceptualization of the development of a given situation, is claimed to be used by informants for discursive purposes, granting a greater rhetorical weight to the Perfect. The results founded in this thesis indicate that the perfect tenses in Spanish have followed (and are following) different developmental paths that are not necessarily restricted to the same sequences and mode of grammaticalization.