Output-Driven Maps

1.1 Output-Orientedness: Maps and Generalizations A long-standing issue of interest in phonology is the extent to which phonological maps from underlying forms to phonetic forms (in the terms of generative phonology) can be characterized in terms of conditions on the output forms (Chomsky 1964, Kiparsky 1971, Kiparsky 1973, Kisseberth 1970, for a broader overview of the issue see McCarthy 2007a). The intuition might be expressed like this: a map will by default preserve the input in the output, and any disparities between the input and the output are motivated by restrictions on the output form. I will call this intuitive property output-orientedness. While the statement “all restrictions motivating input-output disparities are output restrictions” may seem intuitive and concrete, formalizing it in a satisfactory way is not so straightforward. Any direct formalization requires defining exactly what kind of thing an output restriction is, and what exactly it means for a restriction (or set of restrictions) to “motivate” an input-output disparity (motivations tend to be murky things). Furthermore, it is not immediately obvious how to give such a characterization without delving deeply into the mechanics of a particular theory.