Aspect splits without ergativity: Agreement asymmetries in Neo-Aramaic

This paper looks at two different aspect splits in Neo-Aramaic languages that are unusual in that they do not involve any ergativity. Instead, these splits are characterized by agreement reversal, a pattern in which the function of agreement markers switches between aspects, though the alignment of agreement remains consistently nominative-accusative. Some Neo-Aramaic languages have complete agreement reversal, affecting both subject and object agreement (Khan 2002, 2008; Coghill 2003). In addition to this, we describe a different system, found in Senaya, which we call partial agreement reversal. In Senaya, the reversal only affects the marker of the perfective subject, which marks objects in the imperfective. We show that a unifying property of the systems that we discuss is that there is additional agreement potential in the imperfective. We develop an account in which these splits arise because of an aspectual predicate in the imperfective that introduces an additional φ-probe. This proposal provides support for the view that aspect splits are the result of an additional predicate in nonperfective aspects (Laka 2006; Coon 2010; Coon and Preminger 2012), because it allows for the apparently disparate phenomena of split ergativity and agreement reversal to be given a unified treatment.