This paper investigates the evidential potential of verbal constructions with the Spanish verbs amenazar (to threaten), prometer (to promise), parecer (to seem), and resultar (to turn out) within a linguistic system that lacks a grammaticalized category of evidentiality. The study addresses the functional status of these constructions and their position within the system of knowledge source marking in contemporary Spanish. The primary objective is to identify and describe the evidential values of the specified verbal structures, as well as to clarify their role within the indirect evidentiality subsystem of the language. The empirical data for this research are drawn from contemporary Spanish print and electronic media, complemented by evidence from national Spanish language corpora. The methodology employed encompasses functional-semantic and pragmatic analysis, contextual interpretation, and diagnostic testing. The findings demonstrate that the verbal constructions under scrutiny undergo a process of grammaticalization, resulting in the loss of their predicative autonomy and their subsequent functioning as grammaticalized means of propositional modification. It has been established that these constructions predominantly express inferential evidentiality, while exhibiting variation in their degree of subjectivity and their capacity to appeal to an external source of knowledge. The novelty of this research lies in uncovering the functional heterogeneity of these markers within a unified model of indirect evidentiality. Consequently, the study concludes that the analyzed linguistic means constitute an autonomous segment within the evidentiality field of the Spanish language.