Publications
Publications
Papers, conference proceedings, thesis, and other published materials can be found in my Academia.edu profile.
For conference slides (listed in my CV), feel free to reach me at mcalderoncor[removethis]@umass.edu.
Spanish Adjectival Conjuncts: When Agreement Resolves Semantic Ambiguity. (PhD General Paper)
Abstract
When conjoined, certain Spanish adjectives show mismatched phi-features with respect to the nominal head they modify, creating two mismatched concord or agreement patterns, besides the standard fully-matched structure. Each of these three patterns triggers a specific meaning: gender, but not number-agreeing constructions trigger a distributive reading. No-gender nor number agreement (default agreement) triggers a subatomic reading, while fully-matching agreement constructions are ambiguous between these two. We analyze this contrast from a syntax-semantics perspective, proposing that defective adjectives project valued, interpretable number features, and their interaction with non-boolean conjunction determines phi-feature syntactic valuing and semantic interpretation. While the literature on Romance Languages has provided purely syntactic or morphological explanations for only-GENDER constructions, we offer a semantically-curated analysis of both linguistic levels. Besides, our proposal sees only-GENDER in the light of their relationship with standard-ambiguous and default-subatomic construction, a three-way pattern that, to the best of our knowledge, had not been described in the literature. We argue that the ambiguity reflects where conjunction operates within the NP and whether the number features are interpretable or not. We’ll show that the same difference explains the contrast between the two mismatched structures.
Working on an experimental follow-up with Özge Bakay.
A related puzzle: The consequences of atomicity, kinds, and number in number agreement.
Passive constructions in San Pablo Güilá Zapotec (PhD General Paper)
Abstract
The lack of passive constructions is a well-accepted assumption for Zapotec languages (see various authors in Operstein & Sonnenschein, 2015); from morphology, two empirical arguments have served this conclusion: (I) transitive (also reported: high transitive, active) verbs are arguably derived from their intransitive (also reported: low transitive, inactive) counterparts (e.g. dǐldtǐld ‘cause.INTR/TR’; jàk s-jàk ‘cure.INTR/TR’), and (II) the lack of productive passive morphology. This paper takes a step back from morphology and provides syntactic evidence from San Pablo Güilá Zapotec (GZ), revealing the presence of a passive voice construction, showing that, despite being ‘morphologically more basic’, intransitive verbs can encode passive syntax. Moreover, we show that the GZ passive is typologically canonical (Legate, 2021): agents are semantically present but syntactically demoted, patients are promoted to a subject position, and passive verbs, u-syncretic with unacussatives, are morphologically marked as they are different from their active counterparts.
WSCLA 27 handout Proceedings coming soon!
Presenting at the SSILA meeting in January 2026
An experimental follow-up: Does Agent-First preference guide incremental interpretation in Güilá Zapotec?