Abstract
This paper investigates the emergence of lexicalized effects of word usage on word duration by looking at parallel changes in usage and duration over 130years in New Zealand English. Previous research has found that frequent words are shorter, informative words are longer, and words in utterance-final position are also longer. It has also been argued that some of these patterns are not simply online adjustments, but are incorporated into lexical representations. While these studies tend to focus on the synchronic aspects of such patterns, our corpus shows that word-usage patterns and word durations are not static over time. Many words change in duration and also change with respect to frequency, informativity and likelihood of occurring utterance-finally. Analysis of changing word durations over this time period shows substantial patterns of co-adaptation between word usage and word durations. Words that are increasing in frequency are becoming shorter. Words that are increasing/decreasing in informativity show a change in the same direction in duration (e.g. increasing informativity is associated with increasing duration). And words that are increasingly appearing utterance-finally are lengthening. These effects exist independently of the local effects of the predictors. For example, words that are increasing utterance-finally lengthen in all positions, including utterance-medially. We show that these results are compatible with a number of different views about lexical representations, but they cannot be explained without reference to a production-perception loop that allows speakers to update their representations dynamically on the basis of their experience.
Highlights
It is well-established that a number of usage factors affect word duration – including frequency, the word’s predictability in context, and the position of the word in relation to utterance boundaries
Using this unique data set, we are able to obtain a direct view of the accumulation of usage-based effects in lexical representations over time. These show up in the form of robust parallels between changes in word duration and usage. We suggest that these findings are compatible with a range of different views about lexical representations, but are difficult to explain without reference to the so-called production-perception loop (Pierrehumbert, 2001; Wedel, 2007)
Model does not control for the local effect of position in the utterance, which makes it difficult to tell whether this is a local or lexical effect. To separate these two effects, we look both at the local effect of utterance-final position and the lexical effect of typical position within the utterance
Summary
It is well-established that a number of usage factors affect word duration – including frequency, the word’s predictability in context, and the position of the word in relation to utterance boundaries. Bybee 2002; Jaeger and Buz to appear) They can manifest as context-dependent, local adjustments that apply online during speech production. We do not assign special theoretical significance to variation in word duration, and take it to be one of the many phonetic reflexes of more general processes of hypo- and hyper-articulation (cf Lindblom et al 1995) conditioned by usage factors Other examples of such reflexes include variation in segmental and syllabic duration, the peripherality of vowels and consonant deletion (Jurafsky et al, 2001; Bybee, 2002; Aylett and Turk, 2006; Cohen Priva, 2015)
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
