*ABA in Multidimensional Paradigms: A MAX/DEP-based account

Abstract

The last decade and a half has witnessed intensive research into *ABA universals—generalizations such as “If a nominative and the corresponding dative have the same exponent, then the corresponding accusative has that exponent, too” (Caha 2009; Smith et al. 2019). Most existing work on these universals has only focused on one ‘paradigm column’ at a time, by checking a given paradigm’s nominative singular, accusative singular, and dative singular, for example, with no heed to whether any of the relevant exponents would also show up in that paradigm’s nominative plural, accusative plural, or dative plural. However, some recent literature has pointed out that inspecting full paradigms is crucial to our understanding of *ABA, because some classic accounts that derive *ABA column-internally turn out to also make predictions as to what may or may not happen across columns, and those predictions are often incorrect (cf., among others, Christopoulos & Zompì 2022). In this dissertation, I review those incorrect predictions and replace them with a novel generalization specifically concerning *ABA-like effects in multidimensional paradigms. I then set out to derive this generalization by setting up an exponent-selection system wherein exponents may both be underspecified and be overspecified with respect to their exponenda, with each of these departures from a perfect match being penalized but not necessarily fatal. In particular, I explicitly implement this intuition in optimality-theoretic terms, via a strict-domination ranking of violable Max and Dep constraints (cf. in particular Ackema & Neeleman 2005; Wolf 2008; Müller 2020), and I show that the resulting system, while restrictive enough to derive the desired generalization, is also powerful enough to afford a natural account of some notoriously unnatural (‘morphomic’) exponent distributions in the inflection of Germanic pronouns and Romance verbs.