Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.

Date of Award


Access Type

Campus Access

Document type


Degree Name

Doctor of Philosophy (PhD)

Degree Program


First Advisor

Kyle Johnson

Second Advisor

Rajesh Bhatt

Third Advisor

Luiz Amaral

Subject Categories



This dissertation explores a longstanding challenge in work on gapping through the empirical lens of gapping in Farsi (the Tehrani variant of Modern Persian). While gapping has much in common with more uncontroversial elliptical constructions such as VPE and sluicing, it also differs from ellipsis in ways that accounts combining TP or CP coordination and constituent deletion fail to explain. Johnson (1994/2004, 2009), and extensions of Johnson's work in Coppock (2001) and Lin (2000,2001), among others, propose a solution to this dilemma for gapping in English that relies on LOW COORDINATION and, consequently, on the assumption that VP, or υP, is the verbal constituent targeted in gapping. I argue that the same kind of solution cannot be extended to Farsi because there is not evidence of low coordination structures in gapping and, crucially, VPE does not exist independently in the langauge. It is shown that while gapping in Farsi exhibits the same core characteristics that distinguish it from ellipsis in English and other languages (restriction to coordinations, forward only directionality), it does not exhibit those that motivate the low-coordination structure. I propose an alternativeMOVEMENT PLUS DELETION account that addresses the first set of differences while retaining (i) TP, as opposed to VP, coordination and (ii) ellipsis as the source of the gap. This is achieved by designating the coordinate head itself the syntactic licensor, and by requiring that the gapped TP must move to a specifier position of the coordinate head in order for ellipsis to be licensed via an agreement relationship between it and the licensing head. The necessity of movement in establishing the ellipsis-licensing agreement relationship is argued to account for the contrast between the acceptability of gapping in embedded clauses and the ungrammaticality of gapping inside islands. Thus this account, while it derives the gap from ellipsis, also has affinities with accounts, particularly that of Johnson (1994/2004), which seek to explain the distinctive properties of gapping by invoking movement. In addition to contributing a study of gapping in Farsi, in which gapping has not been previously studied, to the cross-linguistic data on gapping, a major empirical contribution of this work is the finding that the restriction against embedded gapping observed in English is likely not a definitional property of gapping across languages.