|Title:||'An Integrated Approach to the Syntax-Phonology Interface: A Cross-Linguistic Study'||Add Dissertation|
|Author:||Yusra Yahya||Update Dissertation|
|Email:||click here to access email|
|Degree Awarded:||'English and Foreign Languages University, Hyderabad' , 'Center for Linguistics and Contemporary English'|
|Linguistic Subfield(s):||'Linguistic Theories'|
|Abstract:||This dissertation attempts to achieve two main goals. The first is to explore, within the context of Chomsky’s Phase Theory and multiple spellout (MSO) approach, a more general interface account integrating cross-linguistic variations regarding the ways in which syntactic structures can be defined (mapped) onto phonological form at the major phonological phrase level. The second aim is to translate (most of) these ways into a new (spellout-based) OT grammar. One new thesis I propose in this dissertation is that a phase head, mainly v*, belongs to its domain for spellout purposes contra Chomsky (2000, 2001 and subsequent work). However, whether the phase head undergoes interpretation along with its complement or not is something that depends on the spell out instruction that this phase head itself may instantiate. Moreover, this work proposes an extension on Hale and Selkirk (1987) and Truckenbrodt (1995). Within the framework of Truckenbrodt (1995), my extension can be read as the following: just as Wrap-XP is defined in relation to VP as a lexical maximal projection, another form of Wrap call it Wrap-XP’ applies to IP as a functional maximal projection. This extension falls as a natural result once we assume the new theory of phases and MSO. Wrap can now apply to VP and IP, which are nothing but the complements of the phase heads v *and C respectively. This extension helps us to account for the larger (major) phonological phrase domains observed in languages as Xhosa (Zerbian, 2004). Moreover, if this extension is on the right track, then we perhaps have now the answer to handle the question why is it that many linguists tend to assume that only CP is a phase (in some language)? How this extension is derivationally implemented is a simple issue that builds on the role of phase heads though this time as transfer hinders, rather than transfer initiators. I, accordingly, propose that a phase head C with a strong wrapping spellout instruction can, in the case where PIC2 is the relevant, result in hindering, rather than initiating, spellout on a lower v* phase with the consequence that we have only one instance of spellout applying to the domain of C.
- Elements liable to wrapping effects:
ii. IP (my proposal)
In this work, I also define, develop and defend a new spellout mechanism. This mechanism builds on the role of phase heads as both the loci of spellout features and the transfer triggers via either Phase Impenetrability Condition 1 (PIC1) and/or Phase Impenetrability Condition 2 (PIC2). The assumption here is that phase heads, mainly v*, can regulate the spellout process by deciding both the kind of spellout applying and the timing of spellout.
i.Phase heads as the loci of parametric variations: The type of spellout applying
ii.Phase heads as spellout triggers: The timing of spellout relevant
Moreover, it is proposed in this work that some forms of XP movement are not motivated by an EPP feature of a strong phase head mainly v*, but they are rather motivated by a last resort strategy to accomplish the spellout instruction of this phase head. Last, but not least, a new spellout-based OT grammar is defined by translating the newly assumed spellout functions into spellout constraints.