The continuum between lexical and grammatical evidentiality: a functional analysis of Spanish parecer

Abstract

This paper discusses different degrees of lexical/grammatical evidentiality. I show that the specific function of evidential parecer constructions is linked to the grammatical or lexical status of the verb. The data suggest that there is a correlation between the continuum from lexical to grammatical, on the one hand, and the modes of knowing expressed, on the other. It is shown that the higher the degree of grammaticalization of a parecer construction is, the less evidential values it can express. A series of functional-pragmatic tests indicate that the parece que construction involves a broader array of evidential types than both me/le parece que and parecer + infinitive. These two constructions can be considered the lexical and the grammatical endpoints of the continuum, respectively. The tests confirm the claim that me/le parece que is attached to the conceptualizer, whereas the grammaticalized parecer + infinitive construction conveys an inferential qualification that is restricted to the speaker. The concept of (inter)subjectivity is presented as a useful addition to source-evidentiality. Subjectivity refers to the speaker’s exclusive access to the evidence, whereas intersubjectivity is about shared access to the evidence. The intersubjective statements, here expressed by parece que, are considered to have broader support than the subjective ones conveyed by parecer + infinitive. They can be seen as more reliable and, hence, can explain why the Hispanic linguistic literature labels them as expressing “near-factivity”, while subjective statements, based on the speaker’s inference only, lead to a “non-factive” interpretation. Thus, the different distribution of the evidential readings with parecer gives a more delicate analysis of the different evidential values of lexical and grammatical expressions. *

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 30 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.