Duration, vowel quality, and the rhythmic pattern of English

Abstract

Languages with binary stress systems frequently tolerate a stress lapse over the final two syllables, but almost none tolerate a word-initial stress lapse. Lunden (to appear) argues that this lapse asymmetry can be explained by the presence of word-level final lengthening, which can then create the perception of prominence alternation in languages that use duration as stress correlate. The results of a production and a perception study with English speakers are presented which compare /ɑ/s that occur under stress lapse to /ɑ/s in non-stress-lapse positions. While word-final unstressed /ɑ/ is always longer than non-final unstressed /ɑ/, it is significantly longer when immediately following an unstressed syllable. Similarly, unstressed word-final /ɑ/ has a higher F1 and lower F2 than non-final unstressed /ɑ/, but word-finally this less-reduced vowel is closer to a full vowel when the final syllable is part of a stress lapse. The perception study finds that these differences have perceptual consequences that can lead to a perceived continued rhythm in stress lapse. The phonetic differences explain why a word-final unstressed vowel can be perceived as relatively strong when following an unstressed syllable but as relatively weak when following a stressed syllable.

SW ou WS para o acento primário em português brasileiro: as pistas de um sistema não transparente (SW or WS for the primary accent in Brazilian Portuguese: the clues of a non-transparent system)

“…Os estudos investigados (DELATTRE, 1969;CRYSTAL & HOUSE, 1988;van SANTEN, 1992;SWINGLEY, 2019;LUNDEN, 2017;WARNER & CUTLER, 2017) não apontam uma distinção desse quadro vocálico conforme a posição silábica. O estudo de Delattre (1969), por exemplo, que investigou a redução vocálica em quatro línguas -a saber, inglês, francês, alemão e espanhol -, contrastando a alternância vocálica em pares de palavras como com'peting e compe'tition, constatou que a redução vocálica, decorrente da atonicidade em posição medial, ocorre de forma mais efetiva em inglês do que nas outras línguas.…”

Word-final strength and weakness

“…All three levels of final lengthening were first documented by Oller (1973) for English. More recently, final lengthening at the word level has been demonstrated by Hogoboom for Norwegian (Lunden 2013) and English (Lunden 2017). We will focus on word-level final lengthening, leaving open the question of whether some processes motivated by final lengthening are in fact due to domainfinal generalization, where a higher level of final lengthening is responsible for word-final effects (e.g.…”

The status of word-final phonetic phenomena

“…Final lengthening is the phonetic phenomenon of the elongation of sounds at the right edge of words. Final lengthening, much like phonetic final devoicing, is found at all prosodic levels in varying amounts (Crystal and House 1988and Lunden 2017for English, Johnson and Martin 2001for Creek, Lunden 2006for Norweigan, Nakai et al 2009. In Figure 1, final lengthening and final devoicing are present on the final syllable of the nonce word "basádafa" that was read in a sentence frame in which it was not in focus.…”