Projecting possessors: A Morphosyntactic Investigation of Nominal Possession in Tigrinya

Date
2013-05-28
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In this dissertation, I examine the grammatical expression of possession in Tigrinya, a lesser-studied Semitic language of Ethiopia and Eritrea. I show that possession in Tigrinya is encoded by two strategies, which differ in both structure and function: (i) PREDICATIONAL STRATEGY has the particle nay and is used for alienable possession; (ii) ARGUMENTAL STRATEGY or BARE POSSESSION has no nay and is used for inalienable possession. To account for such differences, I propose different treatments for both types of possession. For alienable or nay-marked possession, I consider two competing hypotheses that have been proposed for a similar element yå- in Amharic, a very closely related language: Either yå- is a genitive case-marker (Ouhalla 2004) or yå- is a LINKER (den Dikken 2007a). I argue that the two hypotheses are untenable because they fail to account for all the facts of Tigrinya. I develop an alternative proposal and claim that nay is a nominal copula and that its role is to introduce a predicational relation between the nay-marked predicate (e.g., possessor) and its subject (e.g., possessee). For inalienable or bare possession, I first compare them with similar constructions in Hebrew: Construct state nominals (CSNs). I show that Tigrinya bare possessive nominals (BPNs) are a type of CSN. I also consider two competing hypotheses previously proposed for Semitic CSNs: Head movement (SHM) (Ritter 1991) and snowballing phrasal movement (SPM) (Shlonsky 2004). I argue that SPM is both theoretically and empirically inadequate for Tigrinya and that HM should be recast in line with current Minimalist assumptions (Chomsky 1995b et seq.). I develop an alternative analysis arguing that both Tigrinya and Hebrew CSNs involve a non-standard head movement – head-to-Spec movement accompanied by Morphological Merger (Matushansky 2006) in their DP structure. Additionally, I claim that differences between Tigrinya and Hebrew arise due to unique properties of the categories N and D: While Hebrew Ns are inherently specified for a [DEF] feature, Tigrinya Ns are not. Also, while the [DEF] feature of the D head in Hebrew is inherently unvalued and strong ([_DEF*]) and causes the N to move to D, the [DEF] feature of the D head in Tigrinya is inherently valued ([±DEF]) and does not cause the noun to move. Finally, generalizing from the proposal of possession herein, the dissertation makes predictions about alienable and inalienable possession in natural languages in general. It proposes that cross-linguistic variation arises due to the availability of both an argumental and a predicative strategy for expressing alienable possession. Each of these strategies requires different functional categories and induces substantial differences in the syntactic structure.
Description
Keywords
Linguistics
Citation
Gebregziabher, K. (2013). Projecting possessors: A Morphosyntactic Investigation of Nominal Possession in Tigrinya (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/27453