Frequency Effects on Determining English Secondary Stress

This paper reviews and builds upon previous studies about how word frequency is relevant in determining English secondary stress in derived forms, using corpus-based research. The result of statistical research will incorporates into Optimality Theoretic Grammar. I propose the group-specific approach; if the frequency of the derived forms is high, the stress pattern of the words will be marked at the lexicon independently. If the frequency is low, however, the word tends to depend on base 'form to determine its secondary stress pattern. In the middle of the two, there exists some variation and phonology-priority constraint ranking. Nevertheless, this normalized approach cannot generalize all of English secondary stress patterns, as exceptions still remain in the end. The exceptions will be marked with lexical-specific indexations