Investigations in a simplified bracketed grid approach to metrical structure

In this dissertation, I examine the fundamental mechanisms and assumptions of the Simplified Bracketed Grid Theory (Idsardi 1992) in two ways: first, by comparing it with Parametric Metrical Theory (Hayes 1995), and second, by implementing it in the analysis of several case studies in stress assignment and syllabification. Throughout these investigations, I argue for Simplified Bracketed Grid Theory as the superior theory of metrical structure. In Simplified Bracketed Grid Theory (SBG), metrical grids are constructed derivationally through the application of rules which can insert or delete brackets and grid marks. This can be contrasted with Parametric Metrical Theory (PMT), which builds up metrical grids using binary feet as metrical building blocks. In Chapters 2 and 3 of the dissertation, I draw out the implications of this fundamental difference between these two theories of metrical phonology, paying attention in particular to the range of attested stress systems which each theory can account for and the complexity of the resulting accounts. Three areas of comparison are examined in detail: foot parsing, the inventory of metrical feet, and the status of the foot as a prosodic unit (as evidenced by foot extrametricality). In every case, I argue that SBG provides a level of internal consistency and simplicity which cannot be found in PMT. A more concrete picture of the strengths of the SBG theory is presented by means of multiple case studies of stress assignment and syllabification (Arabic dialects in Chapter 4, and the Algonquian language Passamaquoddy in Chapter 5). In addition, the SBG formalism is extended in two ways: first, to apply to issues of syllabification, and second, to deal with stress systems which refer to more than two categories of syllable weight. The flexibility of SBG in being able to account not only for complex stress systems, but also for areas of phonology outside of stress assignment (i.e., syllabification) is taken to be an argument in its favor.