Phonological weight

Abstract

Grammars frequently categorize syllables for prosodic purposes, treating one class as heavier (e.g., more stress‐attracting) than another. While such categorization is usually dichotomous, complex and gradient scales are also attested, with various organizational criteria. This article reviews the range of phenomena that invoke weight distinctions and introduces some current debates concerning weight, touching on topics such as the syllable versus interval as the domain of weight, rich scalarity, process and position specificities, the role of onsets, the phonetic basis of categorization, and the mora.

Reducing language to rhythm: Amazonian Bora drummed language exploits speech rhythm for long-distance communication

“…The rhythmic organization of speech is complex and an area of extensive experimental research and lively theoretical debate [ 13 ]. Currently debated issues include what should be identified as the carrier of rhythmic beat in spoken languages [ 14 ], whether vowel-to-vowel intervals represent basic units of speech rhythm [ 15 ] and what is the role of rhythmic structures in language comprehension [ 16 ]. Against this background, we ask here which aspects of rhythmic structure are selected in conventionalized drummed speech to optimally contribute to its intelligibility.…”

“…Rhythmic units associated with syllables also play an important role in grammar in the sense that many languages categorize syllables, treating one class as heavier than another, for purposes such as stress placement and poetic rhyme [ 15 ]. Such weight distinctions usually, but not always, correlate with phonetic duration [ 44 ].…”

“…It has recently been proposed that the units of speech rhythm that encode weight for the purpose of stress placement might be vowel-to-vowel intervals (delimited by the beginning of each vowel, see further discussion and illustration below) rather than syllables, so far with mixed results [ 15 , 47 , 48 ]. In any case, weight distinctions of vowel-to-vowel intervals also correlate with phonetic duration [ 48 ].…”

Weight effects on stress: lexicon and grammar

“…We can conclude that this effect, albeit present, is not strong enough to impact stress propensity once antepenultimate stress in LLL words is taken into account. This fine-grained weight effect can only be accounted for if we assume a gradient notion of weight: both heavy syllables with obstruent and sonorant codas are treated as light overall, but the former are 'lighter' than the latter-see Ryan (2016) for an overview of gradient weight.…”

“…Assuming that English and Portuguese differ in terms of their prosodic representation (i.e., presence/absence of the foot), stress in these two languages is the result of weight and footing in English, and weight and domain parameters (AF) in Portuguese. In both languages, it seems that only a probabilistic grammar (e.g., MaxEnt Grammar; Goldwater and Johnson 2003, Wilson 2006, Hayes and Wilson 2008 can accurately account for the patterns observed, given the number of (sub-regular) exceptions found in these languages (see Ryan (2011Ryan ( , 2016 for English). In such a grammar, words that match the so-called 'regular' patterns are more likely to occur, but deviations from these patterns are also possible, given the probabilistic nature of the grammar in question.…”