Split ergativity based on nominal type

This paper argues that split ergativity based on nominal type is a morphological phenomenon, not a syntactic one. We use three tests to identify the source of this type of split ergativity as morphological syncretism: (i) case agreement, (ii) syntactic ergativity, (iii) coordination. We illustrate the complex patterns of attested splits, demonstrating that analyses positing a single dichotomy (e.g. between first and second person pronouns versus all other nominals) are insufficient. A morphological syncretism analysis is provided whereby ergative case is deleted in featurally-marked contexts.