Giorgio Magri

I studied philosophy and then mathematics at the University of Milano. I got my PhD in Linguistics at MIT in 2009. Since 2012, I am a permanent researcher at the French CNRS, affiliated with the lab SFL in Paris. My curriculum vitae available here contains links to all my work (papers, manuscripts, handouts). 

My research is framed within generative linguistics. Its core idea is that, to understand languages, we need to understand the formal grammars that generate them: what are the mathematical properties of those grammars? what are the algorithmic implications of those properties? 

My current research focuses on quantitative probabilistic phonology. It has two goals. First, to explicitly formulate and document a number of mathematical generalizations on rates of applications of variable phonological processes. Second, to understand which grammatical models (within a constraint-based architecture) are consistent with or follow from these generalizations. Here is a paper that summarizes this research program. So far, I have focused on four generalizations:

• The Stochastic generalization, initially proposed by J. Bresnan (here is a paper with A. Anttila; we are currently extending this material into a monograph).

• The Constant Difference Generalization, initially proposed by W. Labov and more recently revived by B. Hayes and K. Zuraw (here is a representative paper).

• The Innocuous Concatenation Generalization (here is a paper that discusses this generalization in the categorical setting and here is a paper that describes it in the probabilistic setting).

• The Strict Domination Generalization, initially proposed (albeit in a much stronger form) by W. Labov (here is a short paper with E. Flemming).

My previous research focused on three topics:

• During graduate school, I worked on natural language semantics, trying to derive the phenomenology of individual-level predicates from independently motivated assumptions of the theory of scalar implicatures (here is a paper from that period). 

• Towards the end of graduate school, my research interests have shifted to sound patterns, trying to distill the implications of different theories of phonology for learnability and acquisition (here is a representative paper). 

• My interests in learnability led to some work on Bruce Tesar's theory of output-drivenness and its implications for modeling opacity in constraint-based phonology (here is a representative paper). 

I can be contacted by email at magrigrg@gmail.com. My office in Paris is room 134 at 59 rue Pouchet.