The foot: a unified entity for both metrical and segmental phenomena

The foot is the prosodic entity typically used to explain metrical phenomena like stress assignment and pitch accentuation. This paper attempts to widen the explanatory role of feet to include not only suprasegmental effects but also segmental behaviour. Segmental properties such as static distribution patterns and dynamic alternation processes are usually explained by referring to another prosodic entity, the syllable, which is often viewed as a subcategory within the foot. However, in many phonological descriptions these two prosodic units, the foot and the syllable, are interdependent: for example, the presence/absence of stress in a given foot structure goes hand in hand with certain segmental patterns; and melodically strong segments can serve to demarcate the edges of metrical (including foot) domains. In view of this, it is natural to question why two apparently similar units, which lie adjacent on the prosodic hierarchy, should operate independently and thus serve to explain quite distinct aspects of phonological behaviour. For the sake of descriptive unity, we propose that metrical and segmental properties be captured by referring to just a single organizing unit. And here we present arguments for assuming that this single unit should be the foot rather than the syllable. For several reasons a foot-based approach is to be preferred over a syllable-based one. The first reason makes appeal to structural stability: at certain levels of analysis (e.g. postlexical) syllable structure may be subject to reorganization, whereas metrical structure usually remains intact. The second refers to structural organization: as a structural entity the syllable continues to attract controversy, with different theories still unable to agree on basic notions such as the nature of sub-syllabic structure (e.g. the status of the coda) and the nature of syllable formation (e.g. dependency-based or template-based). Some theoretical approaches even take the extreme view that syllables are unnecessary as independent, formal units of organization (Harris 1994, Raimy 2000, Hale and Reiss 2008). By contrast, the foot is rarely associated with issues or controversies of this kind. On this basis, we argue that the foot should be retained as a core unit of phonological description, while the syllable should be abandoned in view of its questionable status. In fact, syllable-based explanations can often be recast as foot-based explanations. For example, consonant lenition in trochaic stress systems may be explained as a weakening process in foot-internal/final (dependent) positions, rather than as a characteristic of coda or syllable-final positions. Vowel reduction …