In the talk I will propose a variant of Boersma's Gradual Learning Algorithm for Stochastic Optimality Theory. While in the original version the learner is always (or tries to become) a speaker, I assume that the learner is both speaker and hearer. This learning theory is applied to the OT system from Aissen (2000), which was developed to explain the typology of differential case marking. It can be shown that the constraint sub-hierarchies that Aissen assumes to be universal follow from the statistical patterns of language use that have been uncovered in several corpus studies, if one adopts the bidirectional learning approach.

Not all case marking patterns are learnable by the Bidirectional Gradual Learning Algorithm (BiGLA), and some patterns are easier to learn than others. If learning with limited resources is repeated over several generations, one can distinguish stable and instable language types, and certain tendencies for language change emerge. In the second part of the talk I will present and discuss some experimental findings on the basis of this evolutionary approach.