September 2017 saw the 150th anniversary of Alexander Melville Bell’s vowel model, with its unique innovation of tongue locations between front and back, what he actually called the location of a configurative aperture between the tongue and the palate, his postulated limit of the anterior cavity with its typical vowel resonance (A. M.Bell, 1867, Visible Speech, Methuen, London: 15-16, 71). Since then, the terms tongue height and backness have followed his usage, with some modification of terminology (for example, D. Jones, 1932, Outline of English Phonetics, Teubner, Leipzig; IPA, 1999, Handbook of the International Phonetic Association, Cambridge University Press), although it has been known for more than half a century that the alleged tongue activity would be irrelevant for tuning the vocal tract resonances for vowels (Chiba & Kajiyama, 1941, The Vowel, its Nature and Structure, reprinted in 1958 by the Phonetic Society of Japan, Tokyo; Fant, 1960, The Acoustic Theory of Speech Production, Mouton, the Hague; Fant, 1965, Formants and cavities, in Proc. of the 5th Int. Congress of Phonetic Sciences 120-141, Basel; Fant, 1980, The relation between area functions and the acoustic signal, Phonetica 37:55-86; Stevens & House, 1961, An acoustical theory of vowel production and some of its implications, Journal of Speech and Hearing Research 4:303-320; Stevens, 1972, The quantal nature of speech, in David & Denes, Human Communication, a Unified View 51-66; Stevens, 1989, On the quantal nature of speech, Journal of Phonetics 17:3-45).
Bell’s innovations were assumptions. He had no evidence because there was no method for observing or recording tongue positions, nor was there any adequate acoustic theory. He followed the established single resonance theory, associating one vowel resonance with the mouth cavity. Roudet (1911, La classification des voyelles de M. Sweet, Revue de Phonétique 1:347-356) had pointed out that both height and backness would determine the volume of the mouth cavity and hence the frequency of the vowel tone, implying they were mutually compensating with respect to the single resonance theory, rather than independent parameters. Bell never suggested what tongue height might actually do acoustically, nor did anyone else except Bell’s son Graham when he launched the double-resonance theory in (1879, Vowel theories, American Journal of Otolaryngology 1, reprinted in his Mechanism of Speech 1907 or later), and that was apparently still controversial in the 1920s.
D. Jones (1917, Experimental phonetics and its utility to the linguist, Nature 100:96-98) had made a set of X-rayed profiles for the cardinal vowels, and from the outset he withheld the images for all but [i a ɑ u], the corner vowels of his new cardinal vowel diagram. Ladefoged (undated, http://www.linguistics.ucla.edu/people/ladefoge/PLcareer.pdf) recalled his first visit to Daniel Jones in the 1950s, who explained the unpublished images “People would have found them too confusing”. Wood (1982, X-ray and Model Studies of Vowel Articulation, Working Papers 23:9-17, Dept. Linguistics, Lund University) examined 38 sets of X-rayed vowel profiles published between 1905 and 1975, from 15 different languages, and found 8 out of 22 instances of half-open [ɔ] at the same height as open [a~ɑ] while 6 were lower. Only 8 out of the 22 [ɔ] were where Bell and Jones expected them, higher than [a~ɑ]. Furthermore, 6 out of 34 instances of half-close [o] were the same height as [a~ɑ], and 6 were lower. Had Jones seen similar anomalous tongue heights in his undisclosed profiles? Such apparent randomness in X-rayed tongue height would certainly have been confusing to anyone believing in the tongue positions prescribed by the Bell model.
Bearing in mind then that height and backness are irrelevant for tuning the resonances of the vocal tract, how are they to be interpreted? There is a well-known similarity between formant diagrams and judgements of tongue height and backness (Essner, 1947, Recherches sur la structure des voyelles orales, Archives Néerlandaises de Phonétique Expérimentale 20:40-77; Joos, 1948, Acoustic Phonetics, Language Monograph 23, supplement to Language 24), and that is how these tongue features are understood here, without implying any causality. “More front” can be seen as a synonym for “higher F2”, and “more open” for “higher F1”, without asking why (Essner explained it by invoking the double resonance theory, while Joos’ explanation, misinterpreting mental acoustic analysis as articulatory analysis, is hardly open to investigation).
Welcome to SWPhonetics 