On the measurement of the phonetic quality of vowels [antikvár]

ON THE MEASUREMENT OF THE PHONETIC QUALITY OF VOWELS Is a Universal Phonetic Standard Possible? KÁLMÁN BOLLA Department of Phonetics, Loránd Eötvös University, Budapest The phonetic description and the phonetic transcription (i.e. the selection of the appropriate phonetic symbols for the transcription) of the innumerable speech sounds which can be produced by the humán articula-tory organs and occur in natural languages is still not an easy task in spite of the several systems of description and transcription that have been pro-posed. Different authors often describe the same sound diíferently, and use different phonetic symbols to transcribe it, which falsely suggests that two different sound qualities are referred to. On the other hand, it alsó happens that the sama symbol is used to transcribe two different sounds. Is the description of sound quality so subjective, then ? Is it possible that the development of science and technology has not produced a phonetic standard, a frame of reference on which objective decisions can be based? How is the phonetic quality of speech sounds determined in practice ? — When reading phonetic literature one interprets the phonetic symbols of sounds, i. e. infers the articulatory and acoustic characteristics of the sounds from the graphic symbols. There is nothing wrong with this method if the choice and interpretation of the symbols is based on uniform and universally valid criteria. We know that this is not so. Just think how difficult it is to determine the quality of the Hungárián rounded half-open velar vowel if it is transcribed by the symbol [a]. — It is alsó a wide-spread practice to refer determine to a certain sound quality indirectly by comparing it to another sound occurring in other languages. However, the comparison can only give an inaccurate and approximate idea of the actual sound quality. For instance, if we determine the quality of the Finnish open [e] sound with reference to the vowels occurring in the Hungárián word mese ('tale'), our description of the sound will differ from its actual quality by two or three degrees of openness. — Sometimes we have to determine the quality on the basis of the articulatory description, the articulatory features of the sound, e. g. when we have to produce accurately the velar sounds [ux] and [y] known from Hungárián historical phonetics. However, the knowledge