Tone Categories Research Articles

Modal Types and Ut, Re, Mi Tonalities: Tonal Coherence in Sacred Vocal Polyphony from about 1500

An evaluation of Pietro Aron's understanding of "mode in polyphony" (Trattato della Natura et Cognitione di tutti gli tuoni [1525]) coupled with an analytical examination of Josquin's motet oeuvre leads to a new theory of tonal coherence in sacred vocal polyphony from about 1500. This theory provides a designation for, and identifies the markers of, the tonalities represented in the motets of Josquin; it is based on a nomenclature derived from the technical vocabulary of modal theory and the exigencies of a musical tradition rooted in plainchant, hexachordal manipulation of the gamut, and counterpoint. Three main tonal categories are defined and identified by hexachordal function of final as Ut, Re, and Mi tonalities. Comments by Glarean (Dodecachordon [1547]) support the essentially practical basis of these tonal designations. Each of these tonalities consists of, and is defined by, a distinct collection of "modal types," melodic-contrapuntal paradigms of tonal coherence that are identifiable in a variety of structural guises, from simple contrapuntal contexts to more abstract frameworks. Analytical examples from Josquin's motets illustrate procedural similarities in works belonging to different modal categories and structural distinctions within the same category, as well as demonstrating why diverse analytical approaches have seemed successful with certain works but inappropriate to others.

How effective are long term mean and standard deviation as normalisation parameters for tonal fundamental frequency?

An attempt is described to ascertain how well the F0 of the isolation tones of a variety of Chinese can be normalised using the mean and standard deviation from seven speakers' long term F0 distributions. Six of the seven speakers' cumulative mean and standard deviation stabilised after approximately 20 s and 10 s, respectively, of voiced speech. Little difference was found between speakers' long term and isolation tone standard deviation, and their long term mean was on average 10% lower than their isolation mean. Their F0 range and standard deviation were also much greater than comparable measures from non-tonal English. A z-score normalisation of the isolation tone F0 with long term parameters resulted in a three-fold reduction in the original between-speaker variance. This degree of reduction was enough to resolve the tonal categories, and the remaining between-speaker variance was comparable to raw within-speaker variance: but the reduction was only half as good as normalisation with isolation tone parameters, and is therefore deemed unsuccessful. Because of the demonstrated relationship between long term and isolation values, it is speculated that the long term approach still has a potential use in comparing varieties where isolation tone normalisation parameters are not equatable.

Perceptual dependency of pitch and duration in categorization of pure tones

To explore possible auditory bases of perceptual categorization, 62 naive listeners were asked to assign tones into one of four classes: “low (pitch)‐short (duration),” “low‐long,” “high‐short,” and “high‐long.” The stimuli were sets of 81 pure tones obtained by factorial combination of nine values of frequency with nine values of duration. In one condition, frequency ranged from 0.25 to 1.0 kHz in ≅ 100‐Hz steps, and duration from 100 to 180 ms in 10‐ms steps; in another condition the ranges were 0.536 to 1.0 kHz in 58‐Hz steps and 100 to 356 ms in 32 ms steps, respectively. The tones were presented at a rate of 1/3 s and the SPL was 60 dB. A single stimulus, absolute judgement task was used and only the very initial performance was studied (first 10 rain). The 50% “low‐high” pitch boundary was measured as a function of tonal duration, and the 50% “short‐long” duration boundary measured as a function of tonal frequency. Listeners were able to consistently categorize the tones into the four classes from the first presentation of the set of tones. For both stimulus conditions, the pitch boundary was placed around the arithmetic mean of the frequency range (X̄ = 0.61 or 0.77 kHz) irrespective of tone duration. The duration boundary was generally placed below the arithmetic mean of the duration range X̄d for frequencies below X̄f, but above X̄d for frequencies higher than X̄f. That is, the higher the pitch the shorter the perceived duration. The results suggest that in auditory categorization with naive listeners, pitch can be judged independently of duration, but the duration judgment is not independent of pitch. [Supported by NRC Post‐doctoral Fellowship to the author.]

Tone effects on consonants

Although consonant effects on tone are well-known, the possibility of tone effects on consonants has been largely neglected. This paper assembles together evidence of three kinds showing that tones can affect the nature of adjacent consonants. Diachronic changes, synchronic rules and allophone distributions, and instrumentally-verified phonetic differences which depend on different tone contexts are discussed in turn. A partial typology of the effects is constructed. It is shown that a correlation of consonant type and tonal category may, in some instances, be ambiguously interpreted as either due to a consonant effect on tone or a tone effect on consonants. A final section discusses how far the phonetic evidence helps to explain the phonologized synchronic and diachronic effects.

Noncategorical perception of tone categories in Thai

Arguments continue over categorical perception of phonetic segments versus continuous perception. In tone languages, phonemic tones are characterized principally by F0 contours. Theories of categoricity would predict continuous perception of tones; i.e., there out to be no peaks in the discrimination of variants at category boundaries, more as in experiments with isolated synthetic vowels than with synthetic stop consonants. This was the outcome of earlier work [A. S. Abramson, J. Acoust. Soc. Am. 33, 842 (1961)], but it seems to have been contradicted by recent work on Mandarin [S. W. Chan, C-K. Chuang and W. S-Y. Wang, J. Acoust. Soc. Am. 58, S119 (A) (1975)]. The present study involves a considerably larger number of subjects, 34 native speakers of Thai, a language with five phonemic tones. Sixteen flat F0 variants synthesized on a syllable of the type [kha:] were sorted into the three “static” high, mid and low tones with considerable overlap. Discrimination tests yielded a high level of discrimination across the continuum with no effects of boundaries between categories, thus implying noncategorical perception of tone categories.

Measuring larynx movement in standard Thia using the cricothyrometer.

Vertical movement of the larynx during connected speech was investigated in Standard Thai - a language that has five phonologically contrastive tones. The effects of pitch, consonant phonation-type, vowel quality, tonal categories, and position in the utterance were determined. Utterance position was found to be the factor most clearly associated with variations in larynx movement. The occurrence of an overall rise-fall pattern of larynx movement distributed over utterances of varying pitch patterns leads to the conclusion that larynx height is not a principal factor regulating pitch in connected speech. Implications for models of pitch production, distinctive features for tone, theories on the historical development of tone and the role of larynx movements in the production of consonants and vowels are discussed.

Some effects of stimulus intensity on response of auditory nerve fibers in the squirrel monkey.

Some effects of stimulus intensity on response of auditory nerve fibers in the squirrel monkey.

Functional architecture in cat primary auditory cortex: columnar organization and organization according to depth.

Functional architecture in cat primary auditory cortex: columnar organization and organization according to depth.