Displaying a wide range of variation, case-agreement systems across Kurdish dialects have featured extensively in typological descriptions and theoretical analyses of diverse perspectives. On the theoretical side, both syntactic and morphological approaches have been invoked to account for the emerging systems of case-agreement in Kurdish. The non-nominative alignment, displaying (pseudo-)ergative properties, has been of particular interest to generative linguists, as it presents challenges to the current theoretical conceptualizations of case and agreement. The nature of the nominative case and oblique (aka ergative) case and how they relate to agreement are the recurrent themes of investigation. While both syntactic and morphological accounts have attempted to capture the case-agreement systems in terms of the assumptions made by the respective theories they invoke, neither has, so far, been able to provide a comprehensive account of the whole range of phenomena under investigation. The main reason why previous analyses have failed to account for the full range of phenomena or appealed to stipulative explanations of an ad hoc nature, I believe, lies in the non-exhaustive nature of the set of empirical data being studied. A challenge facing analyses of lesser-studied languages is the omnipresent uncertainty of the extent to which the full range of relevant data has been taken into account and analyzed. Previous analyses of case-agreement in Kurdish are no exception. In this talk, I will try to present new data sets that have either gone unobserved or underrated in the relevant literature. Refining the empirical patterns of case-agreement in Kurdish, I will argue that oblique-nominative and oblique-oblique types, represented by Sorani/ Hawrami and Ardalani, respectively, result from an identical underlying syntactic derivation, with the difference in the case-agreement patterns relegated to PF. To be more specific, I will illustrate that both case-agreement types start out as oblique-oblique in syntax. The discrepancy in the two alignment types ensues from the isomorphic morphological instantiation in the oblique-oblique type and the non-isomorphic morphological instantiation in the oblique-nominative type. The non-isomorphism of case-agreement in syntax and morphology in the oblique-nominative type will be argued to follow from an independent property of the language's morphology.
