Ellipsis plays a key role in the understanding of the relationship between what is uttered and what is understood. In elliptical utterances, less is said than is understood. How is it that a mere fragment of a sentence can be understood as something more? This dissertation explores the syntax of one type of elliptical phenomena, Stripping. A speaker, Alexi, might utter (1), to which Tim might reply with a Stripping continuation, as in (1a). Tim's reply, apparently just a "no" along with a single word correction, is nonetheless understood as if it were a complete sentence, that in (1b). Does this interpretation arise as a result of some process which does not rely on syntactic structure within the ellipsis site, or is there hidden syntactic structure associated with the remnant "Ardbeg"? (1) Alexi: James likes "Glenlivet". 1a. Tim: No, "Ardbeg". 1b. Tim: No, James likes "Ardbeg".. I argue in this dissertation that Stripping remnants are indeed associated with silent syntactic structure, and that this structure is "isomorphic" to the structure associated with the antecedent. Evidence for this analysis comes from two empirical domains: Islands and Binding Condition C Effects. In Chapter 2, I present the results of four large scale acceptability judgment experiments which reveal that Stripping is insensitive to island constraints. That is, examples like (2b) are grammatical, contrary to various claims in the literature. I argue in this dissertation that Stripping remnants are indeed associated with silent syntactic structure, and that this structure is isomorphic to the structure associated with the antecedent. Evidence for this analysis comes from two empirical domains: Islands and Binding Condition C Effects. In Chapter 2, I present the results of four large scale acceptability judgment experiments which reveal that Stripping is insensitive to island constraints. That is, examples like (2b) are grammatical, contrary to various claims in the literature. 2a. Tim met [subscript island] the researcher who likes "Glenlivet"]. 2b. No, "Ardbeg". The acceptability of these examples leads us to examine just how much structure is contained within the ellipsis site, if any. If there is no structure within the ellipsis site (3a), or if the structure is non-isomorphic to the antecedent (3b), we would expect such examples to be acceptable, as they contain no island. If the ellipsis site contains structure isormorphic to the antecedent, such examples would only be expected to be acceptable if it were possible to repair island constraints through ellipsis. 3a. No, "Ardbeg". 3b. No, "Ardbeg" it was e [line through it was e.] 3c. No, "Ardbeg" Tim met [subscript island] the researcher who likes e [line through Tim met [subscript island] the researcher who likes e]. These three approaches are evaluated in Chapter 3 on the basis of two large scale plausibility judgment experiments. These experiments concern examples like (4), which were found to be as implausible under the given indexations as typical Binding Condition C violations. This is to be expected only there is structure within the ellipsis that contains material isomorphic to the antecedent, including the island and material outside it. If the ellipsis site contains no hidden material (4a), or some non-isomorphic material (4b), the pronoun of the antecedent would not be recovered within the ellipsis site. If the full isomorphic structure were recovered, that pronoun would appear within the ellipsis site, yielding a Binding Condition C violation. The implausibility of (4) reflects this violation. 4 a. He[subscript i] met the researcher who likes Mary. 4b. *No, James[subscript i].. 5a. No, "James[subscript i.] 5b. No, James [subscript i] it was e [line through it was e.] 5c. No, "James [subscript i] he[subscript i] met [[subscript island] the researcher who likes e [line through [subscript island] the researcher who likes e]. In Chapter 4, I argue that the sum of this evidence implicates that, even in instances of island-violating Stripping, the ellipsis site contains isomorphic structure. In turn, I argue that various alternative mechanisms by which the remnant might escape the ellipsis site, without also incurring an island violation, are inadequate, and instead argue that the remnant overtly moves out of the island to escape the ellipsis site. The ill-effects of this island-violating movement, I argue, take the form of an unlinearizable structure, a structure, and the violation associated with it, which is subsequently elided. Thus, the island insensitivity of Stripping is due to island repair. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]