Short Summary of the Workshop

The aim of this workshop is to investigate how Statistical Physics can help elucidating two long-standing linguistic questions: how does language change proceed in time, and what triggers syntactic change, in particular what is the role of prosody in syntatic change.

Empirical data will be provided by the recent history of the pronominal clitic system of Portuguese. The Tycho Brahe Parsed Corpus of Historical Portuguese (http://www.imw.usp.br/~tycho) will be intensively exploited for this purpose.

The theoretical frameworks of the discussion are the Generative Grammar approach to language and the Thermodynamical Formalism from Statistical Physics. The former includes the Minimalist version of the Theory of Principles and Parameters and Optimality Theory. The latter has been used in recent years as a basis for models of pattern identification, and for understanding critical phenomena in complex systems. In the present discussion, the Thermodynamical Formalism provides a model for the interaction between prosody and syntax in language acquisition.

The topics to be discussed during the workshop include:

• the interface between prosody and syntax in the general arquitecture of grammar.

• the implementation of a model for the interaction between syntax and prosody through the Thermodynamical Formalism.

• the relationship between acoustic data and phonological descriptions.

• the notions of Energy-Harmony in Optimality Theory as a tool to define stress patterns.
• combinatorial problems in the research of optimal rhythmic patterns, and the development of the Sotaq processor.
• parameters involved in the change which took place from Classical to Modern European Portuguese.
• language changes and generic bifurcations of dynamical systems, relations with genetic evolution.
• linguistic and computational issues related to the construction of historical linguistic corpora, including the implementation of tagging algorithms to morphologically rich languages.
• the design of syntactic parsers for non-rigid word order languages.

• statistical issues in linguistic description.