The project investigates the empirical and formal properties of repair phenomena in syntax that can be characterized as "last resort" operations.

The project investigates the phenomenon of repairs in syntax that are characterizable as ''last resort'' operations (textbook examples for repairs are do-support in, e.g., English, or grammatical resumption in, e.g., Hebrew).

On the empirical side, the project seeks to provide a taxonomy of syntactic repairs as they can be found in different languages. Result of this inquiry will be a database summarizing different types of repairs and their properties.

On the theoretical side, the points of departure in one part of the project are a) the idea that repairs lend themselves to an optimality theoretic analysis, and b) that repairs sometimes exhibit opacity effects, which points to optimization being cyclic. a) poses a problem for purely cyclic theories (e.g., the Minimalist Program), b) for purely optimizing theories (such as parallel optimality theory). Accordingly, this part of the project focuses on the idea that all repairs in syntax involve cyclic optimization. Research questions to be
pursued are, among others, what further arguments for the cyclic nature of the optimization process can be given, and how large cyclic domains are generally supposed to be.

In another theoretical part of the project, the question will be pursued what an analysis of repairs within the Minimalist Program (without ranked and violable constraints) that preserves the core characteristics of the notion of a repair may look like.