This course covers advanced methods and ideas in the logical analysis of language, with an emphasis on type-logical methods for the analysis of natural language syntax and semantics. The course has a 'capita selecta' format, focusing on various aspects of the connection between language and reasoning.

The 2019-2020 edition consists of three parts.
In the first part, we introduce categorial grammar logics, starting with Lambek's Syntactic Calculus. We show how the syntactic calculus can be extended with control operations that allow for restricted forms of reordering and/or restructuring. We discuss the 'proofs-as-programs' interpretation of syntactic derivations, comparing the set theoretic models of formal semantics with vector-based modelling as used in present day NLP.
In the second part, we situate typelogical grammars in the wider context of substructural logics. The focus in this part is on prooftheoretic methods (display calculi, decidability) and on soundness/completeness with respect to algebraic and Kripke-style relational semantics.
In the final part of the course, we introduce the machine learning perspective on the themes above. We show how from vector-based representations of word meanings one can compute the interpretation of larger phrases in a compositional way.