Metrical rules and the notion of ‘maximum’: a reply to Derek Attridge

Attridge (2003) is in effect an argument against generative metrics, focusing on the notion of ‘stress maximum’. He argues that ‘stress maximum’ is inadequate on its own terms, which I will show is incorrect when we use a fuller definition (based on the original definition from Halle and Keyser, 1966). His solution is to present metre as control directly over the phonetics of the text, but I will show that metre controls metrical elements (maxima) rather than phonetic elements (stressed syllables).