Optimality theory: Roots and aims

In generative linguistics, Optimality Theory is a recent development. Currently, its most fruitful applications (although by no means the only ones) are in areas of phonology and the phonology-morphology interface (especially so in Prosodic Morphology). This contribution describes the aim of Optimality Theory, and the way it reaches those aims, concentrating on its constraint-based character. Constraints are assumed to be universal, but a distinct difference with other theories of a similar kind is that constraints are violable: they have an evaluative task with respect to phonological representations, and the status of the output of the phonological component vis-a-vis the constraints is that the output is a collection of ‘best’ forms, and not necessarily of ‘perfect’ forms. A second aim of this contribution is to reconstruct the roots of Optimality Theory within the generative tradition. They can be found in the 1970's notion of conspiracy, and in 1980's attempts away from Chomsky & Halle type rule-based analyses in subtheories such as autosegmental and metrical phonology, and the repair strategy approach.