Date of Completion


Embargo Period



evidentiality, focus, discourse, tense, mirativity, semantics, syntax, Southern Aymara, Spanish

Major Advisor

Jon Gajewski

Associate Advisor

Željko Bošković

Associate Advisor

Magdalena Kaufmann

Associate Advisor

Stefan Kaufmann

Field of Study



Doctor of Philosophy

Open Access

Open Access


This dissertation addresses a number of phenomena in the semantics and syntax-semantics of evidentiality focusing on Southern Aymara within a cross-linguistic perspective. Traditionally, the literature has identified three evidentials in this language: the direct evidential enclitic =wa, the indirect evidential suffix -tay, and the reportative evidential free morpheme siwa. This dissertation concentrates on four issues in connection to these markers. The first issue I address is the association of evidentiality and focus in the same marker, namely, the direct evidential =wa. This is a topic that has been previously acknowledged (Muysken 1995; Faller 2002), but has remained unaccounted for. More generally, =wa is likened to focus particles, such as English even, in an approach that makes explicit how an evidential meaning fits with a focus meaning. Then I analyze the discourse contrasts in sentences involving direct evidentiality. By showing that there is a contrast between sentences with and without =wa, I make the novel observation that sentences with and without =wa differ with regard to who is held responsible for the evidence that is available, and discuss the consequences of this contrast, which is tied to, e.g., storytelling. I also discuss indirect evidentiality and mirativity. Adopting a comparative approach contrasting Quechua and Southern Aymara, two Andean languages that are typologically nearly identical, I discuss different analyses that account for indirect evidentiality and argue that an analysis that introduces a learning time (i.e., a time at which the speaker acquires relevant evidence) is more appropriate. I extend the analysis to account for cases involving mirativity. Finally, I focus on clauses with more than one evidential in Aymara. I provide an analysis that involves syntactic and semantic considerations that make explicit under what conditions evidentials can co-occur, making explicit how different evidential meanings can be combined in a single clause, and extend the analysis to other languages for which this phenomenon has been reported.