Representational properties common to phonological and orthographic output systems

Linguistic and psycholinguistic research over the last decade has provided a wide array of evidence indicating that the mental representations of spoken word forms are highly structured objects. One prominent aspect of the enriched notion of phonological structure is that the representation of segmental length (quantity) is separate from that of segmental identity; and another is that segmental sequences are organized into prosodic constituents. This paper discusses data from acquired dysgraphia which suggest that certain attributes of phonological forms also hold of orthographic forms as well - including the skeletal representation of quantity and the grouping of segments into syllabic constituents. E.g., the geminate letters ee and tt in orthographic forms like career and butter are represented and processed as single graphemic objects that are associated with two CV units on a tier that encodes quantity. It is argued that such properties of phonological and orthographic form reflect general conditions imposed on lexical representation, and not modality-specific representational constraints.