Copyright © 1999 Massachusetts Institute of Technology
Speech is the most common medium by which language is transmitted. Phonetics is the science or study of the physical aspects of speech events. It has a long history. For centuries, phonetic descriptions of particular languages have been undertaken to preserve or reconstruct their pronunciation. Phonetics developed rapidly in the nineteenth century in connection with spelling reform, language (pronunciation) teaching, speech training for the deaf, stenographic shorthands, and the study of historical sound changes. Today phonetics is an interdisciplinary field combining aspects of linguistics, psychology (including perception and motor control), computer science, and engineering. However, in the United States especially, "phonetics" is commonly considered a subfield of linguistics, and "speech science" or "speech" is the more general term.
A common question among linguists and nonlinguists alike is, "What is the difference between phonetics and phonology?" One answer is that phonetics is concerned with actual physical properties that are measured or described with some precision, whereas phonology is concerned with (symbolic) categories. For example, the phonology of a language might describe and explain the allowed sequences of consonants in the language, and the phonetics of the language would describe and explain the physical properties of a given consonant in these different allowed sequences. Nonetheless, even though phonetics deals with physical properties, it is just as much concerned with linguistic knowledge as with behavior. It is not the case that phonology can be identified with "competence" and phonetics with "performance," for example.
Phonetics is usually divided into three areas: speech production, acoustics, and perception. Phoneticians want to understand not only how speech is produced, perceived, and acoustically structured, but how these mechanisms shape the sound systems of human languages. What is the range of speech sounds found in human languages? Why do languages prefer certain sounds and combinations of sounds? How does speech convey linguistic structure to listeners? These are some of the key questions in phonetics. Phoneticians stress that only if the array of sounds used across languages is studied can complete models of speech be developed.
Speech production is the basis of traditional phonetic transcription systems such as the International Phonetic Alphabet (IPA), as well as some phonological feature systems (Catford 1977; Ladefoged 1993; Laver 1994; see PHONOLOGY and DISTINCTIVE FEATURES). The components of speech production are the airstream mechanism (the process by which air flow for speech is initiated), phonation or voicing (production of a sound source by the vibrating vocal cords inside the larynx -- this is the most important sound source in speech), ARTICULATION (modification of the phonation sound source, and introduction of additional sound sources, by the movements of articulators), and the oro-nasal process (modification of the sound source by the flow of air through the nose). Speech sounds are traditionally described as combinations of these components. On the other hand, PROSODY (the suprasegmental variation of loudness, length, and pitch that makes some segments, or larger groupings of segments, more prominent or otherwise set off from others) is traditionally described not in speech- production terms but more in terms of the dimensions of amplitude, duration, and frequency (Lehiste 1970; see STRESS, LINGUISTICS; TONE).
Speech acoustics concerns the properties of speech transmitted from speaker to hearer. Speech sounds are usually described in terms of their prominent frequency components and the durations of intervals within each sound. The source-filter theory of speech production and acoustics (Fant 1960; Stevens, forthcoming) describes speech as the result of modifying acoustic sources by vocal-tract filter functions. The acoustic sources in speech include phonation (described above), noise produced in the larynx (such as for aspiration and breathiness), and noise produced by air flowing through a constriction anywhere in the vocal tract (such as for a fricative sound or after the release of a stop). Every speech sound must involve one or more such sources. The source(s) is then modified by the filter function of the vocal tract. The most important aspect of this filtering is that the airways of the vocal tract have particular resonances, called formants, which serve to enhance any corresponding frequencies in a source. The resonance frequencies depend on the size and shape of the airway, which in turn depend on the positions of all the articulators; thus, as the articulators move during speech, the formant frequencies are varied.
At the same time, phonetics can be divided into practical, experimental, and theoretical approaches. Practical phonetics concerns skills of aural transcription and oral production of speech sounds, usually in conjunction with a descriptive system like the IPA. Different levels of phonetic transcription are traditionally recognized. In a phonemic transcription, the only phonetic symbols used are those representing the phonemes, or basic sounds, of the language being transcribed. In an allophonic transcription, additional symbols are used, to represent more detailed variants (or allophones) of the phonemes. The more such detail included, the narrower the allophonic transcription.
Experimental phonetics is based on the use of laboratory equipment. Laboratory techniques (see Hardcastle and Laver 1997) are generally needed to understand exactly how some sound is produced and to detail its acoustic and/or perceptually relevant properties. When experimental phonetic methods are used to answer questions of interest to phonologists, it is sometimes called "laboratory phonology".
Certain acoustic measurements of speech sounds have become common, especially since the advent of the sound spectrograph, which produces visual displays (spectrograms) of frequency and intensity over time. The most common frequency measurements are the frequencies of vowel formants, of the formant transitions between consonants and vowels, and of the fundamental frequency of phonation. Changes in the source, and the durations of intervals with different sources, are also important in speech; for example the duration between the release of a stop and the onset of voicing (an interval filled with aspiration) is called voice onset time (VOT). There is now a wide range of the world's languages receiving detailed experimental descriptions in terms of these and other measures (e.g., Ladefoged and Maddieson 1996), although of course there remain many languages with unusual sounds whose production, acoustics, and/or perception are not well understood. Furthermore, manipulation of such measures to provide a range of artificial speech stimuli is an important tool of SPEECH PERCEPTION, in determining which acoustic properties matter most to listeners. Experimental phonetics also contributes to the development of speech technology (see SPEECH SYNTHESIS and SPEECH RECOGNITION IN MACHINES).
Theoretical phonetics is concerned not only with theories of speech production, acoustics, and perception but also with theories to explain why languages have the sounds, grammatical structures, and historical sound changes that they do, and theories to describe the interrelationship of the more abstract patterns of phonology and the physical forms of speech sounds. Phoneticians look for recurring patterns of variation in sounds (see PHONOLOGICAL RULES AND PROCESSES), and then try to understand why they should occur. Phonetic constraints on phonology may be proposed as part of either a reductionist program (see REDUCTIONISM), in which phonology is reduced to phonetics, or an interface program, in which phonetics and phonology are usually recognized as separate components of a grammar.
Catford, J. C. (1977). Fundamental Problems in Phonetics. Bloomington: Indiana University Press.
Fant, G. (1960). Acoustic Theory of Speech Production. The Hague: Mouton.
Hardcastle, W. J., and J. Laver. (1997). The Handbook of Phonetic Sciences. Oxford: Blackwell.
Ladefoged, P. (1993). A Course in Phonetics. 3rd ed. Fort Worth, TX and Orlando, FL: Harcourt Brace Jovanovich.
Ladefoged, P., and I. Maddieson. (1996). The Sounds of the World's Languages. Oxford: Blackwell.
Laver, J. (1994). Principles of Phonetics. Cambridge: Cambridge University Press.
Lehiste, I. (1970). Suprasegmentals. Cambridge, MA: MIT Press.
Stevens, K. N. (Forthcoming). Acoustic Phonetics. Cambridge, MA: MIT Press.
Asher, R. E., and E. J. A. Henderson, Eds. (1981). Towards a History of Phonetics. Edinburgh: Edinburgh University Press.
Connell, B., and A. Arvaniti, Eds. (1995). Phonology and Phonetic Evidence: Papers in Laboratory Phonology 4. Cambridge: Cambridge University Press.
Denes, P. B., and E. N. Pinson. (1993). The Speech Chain: The Physics and Biology of Spoken Language. 2nd ed. New York: W. H. Freeman and Company.
Docherty, G. J., and D. R. Ladd, Eds. (1992). Gesture, Segment, Prosody: Papers in Laboratory Phonology 2. Cambridge: Cambridge University Press.
Keating, P. A., Ed. (1994). Phonological Structure and Phonetic Form: Papers in Laboratory Phonology 3. Cambridge: Cambridge University Press.
Kingston, J., and M. E. Beckman, Eds. (1990). Papers in Laboratory Phonology 1: Between the Grammar and Physics of Speech. Cambridge: Cambridge University Press.
Ladefoged, P. (1996). Elements of Acoustic Phonetics. 2nd ed. Chicago: University of Chicago Press .
Copyright © 1999 Massachusetts Institute of Technology