Changing Preferences

This is an exploratory document for a new research line in logical semantics which is emerging from several current developments in computer science. Standard logic employs ''at' unstructured sets of statements for its theories and unstructured classes of models for its semantic universes. Nowadays, however, there is an incipient literature on structured universes of models as well as structured theories, both employing 'preference relations' of some sort. The purpose of this brief report is (1) to propose a more systematic framework for this trend, while also connecting it up with some historical predecessors, (2) to design some new logical systems bringing preferences out explicitly, thereby highlighting the theoretical properties of 'preferential reasoning' while raising some new kinds of technical question for further research, and (3) to link up with some current computational ideas (emerging also in the eld of linguistics), by bringing in the general dynamic logic of procedures as well as logical systems providing suitable ne-structure of information states.