Speaker commitment: Back to the speaker. Evidence from Spanish alternations

Abstract

This paper proposes an alternative cognitive account of the notion of speaker commitment in terms of speaker involvement and processing. The focus will be on the role of the speaker as conceptualizer. Invoking conceptualizer-related processing instead of speaker commitment has the advantage of avoiding reliance on non-speakerrelated dimensions to determine degrees of speaker commitment for introducing some propositional content. Our theoretical claim is based on two case studies from Spanish. First, canonical direct que ‘that’-clauses and oblique de que ‘of that’-clauses present an occasional switch to the alternate oblique and non-oblique construction, known as dequeísmo and queísmo, respectively. Dequeísmo has hitherto been related to notions such as doubt, hearsay, or distancing, i.e., to weak speaker commitment. Context analysis, however, shows that this approach is descriptively inadequate and that the phenomenon can best be accounted for in terms of speaker involvement: the speaker-conceptualizer is highly involved in selective information retrieval. Queísmo, by contrast, minimizes stage-managing, thus yielding low speaker involvement. In both cases, the relative strength of the speaker’s commitment is to be inferred on other grounds. Second, the Spanish modals poder, deber and tener que have been described in terms of weak, intermediate and strong speaker commitment. Yet, it will be shown that speaker involvement in downplaying the force structure decreases from poder to tener que. The more the deontic background can be subjectified the more the speaker is involved in the subjective construal. Here, weak commitment thus correlates with strong speaker involvement, and vice versa.

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 — 49 citations exist upstream.

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