On the universality of intonational phrases: a cross-linguistic interrater study

Abstract

This study is concerned with the identifiability of intonational phrase boundaries across familiar and unfamiliar languages. Four annotators segmented a corpus of more than three hours of spontaneous speech into intonational phrases. The corpus included narratives in their native German, but also in three languages of Indonesia unknown to them. The results show significant agreement across the whole corpus, as well as for each subcorpus. We discuss the interpretation of these results, including the hypothesis that it makes sense to distinguish between phonetic and phonological intonational phrases, and that the former are a universal characteristic of speech, allowing listeners to segment speech into intonational phrase-sized units even in unknown languages.

Using Rapid Prosody Transcription to probe little-known prosodic systems: The case of Papuan Malay

“…This is probably less of a problem in the case of pauses, but is certainly relevant in the case of subtler rhythmic or melodic changes. Note that unfamiliarity with the language of the recordings is equivocal with regard to complicating the annotation task, as already hinted at above and further discussed and exemplified in Himmelmann et al (2018). It complicates the task, because it makes it more difficult to identify words and the kind of prosodic events one expects to hear.…”

“…This contribution explores the usefulness in this regard of the Rapid Prosody Transcription (RPT) method, whereby naïve (i.e., linguistically untrained) listeners rate audio recordings of spoken language for prosodic prominences and boundaries. This approach was developed by Jennifer Cole and colleagues (e.g., Cole et al, 2010a;Cole et al, 2010b; see Cole & Shattuck-Hufnagel, 2016, p. 7-13 for a concise introduction and further references), following up on earlier work in the tradition of the perceptionoriented approach to intonation developed in Eindhoven as summarized in 't Hart et al (1990); see also Himmelmann et al (2018), p. 210-213 for further details and references. The present article follows up on a suggestion by Cole and Shattuck-Hufnagel (2016, p. 11) that "RPT can even be used to explore prosody, from the perspective of the listener, in languages for which the prosodic phonology has not yet been worked out, and such data may be then used as the basis for developing more articulated grammatical models.…”

Sequences of Intonation Units form a ~1 Hz rhythm

“…The sample includes well-studied languages from the Eurasian macro area 73 -languages that are spoken by large or relatively large speech communities (English, Russian and Hebrew), 74 as well much lesser-known and -studied languages, spoken in the Indonesian-governed part of Papua by 75 smaller speech communities. Importantly, our results generalize across this linguistic diversity, despite the 76 substantial differences in socio-cultural settings and all aspects of grammar, including other prosodic 77 characteristics (Himmelmann et al, 2018). In contrast to previous research, we estimate the temporal 78 structure of IUs using direct time measurements rather than estimations based on word count or syllable 79 length (Box 2).…”

“…This has been demonstrated both by means of qualitative 63 discourse analysis (e.g., Chafe, 1987Chafe, , 1994Ono & Thompson, 1995), and by quantifying the average 64 amount of content items per IU. Specifically, the amount of content items per IU was found to be very 65 similar across languages albeit strikingly different grammatical profiles (Himmelmann et al, 2018). Another 66 example for the shared role of IUs in different languages pertains to the way speakers plan their (speech) 67 4 actions.…”

“…We retrieved the respective audiofiles from the DoBeS archive. The Eastern Indonesia data in 324 Himmelmann et al (2018) included an additional language that we did not attempt to analyze because it 325 included only two recordings. Additionally, we could not retrieve the audiofile of one of the original 6 326 narratives from the Yali consensus set, so only 5 Yali recordings were analyzed.…”