Temporal Phenomena in Natural
Phonology – Bridging the Gap between
Discrete Phonology and Dynamic Phonetics |
|
Geoffrey Schwartz
(Poznań) |
Decades of research in experimental phonetics and speech perception
have shown that speech is a dynamic process. Speech production is based on
continuous movements of the articulators over time. At the same time listeners
rely heavily on the dynamic nature of acoustic transitions in perceiving
linguistic contrasts. In fact, the lack of both linearity and acoustic-phonetic
invariance in the speech signal has cast doubt on the practice in phonology of
mapping discrete units (phonemes made up of static features) onto the dynamic
speech signal. While listeners clearly
perceive the sounds of speech categorically, and speakers certainly place
speech sounds into different categories of discrete segments, phonological
descriptions rarely consider the dynamic nature of speech.
Though phonological theories occasionally employ “temporal”
descriptions (to distinguish long from short vowels, affricates from stops,
etc.), their features are usually static. Thus, most accounts of phonology seem
unable to incorporate dynamic phenomena inherent in the sequences of segments
they describe. Indeed, the degree and duration of many features can vary,
sometimes regardless of phonetic context. This variation can be the key to
explaining curious synchronic and diachronic phenomena.
The framework of Natural Phonology (NP), developed by David Stampe
and Patricia Donegan, offers the possibility of incorporating speech dynamics
into a phonological theory based on discrete units. Thanks to both context-free
and context-dependent processes and implicational conditions, the framework can
show how dynamic phenomena of speech work on a segmental level. In other words,
in NP one can make statements about the degree and duration of various phonetic
features. Thus, formalized versions of conditions like “all things being equal,
X is longer than Y” are often employed. This study will investigate a number of
phonetic phenomena in an attempt to build on the inventory of processes and
implicational conditions of Natural Phonology. Both context-free and context-dependent
phenomena will be examined instrumentally, with an eye toward providing an
enriched perspective on synchronic and diachronic processes.
Phonetic studies will include some or all of the following:
1. Another look at final devoicing
In NP, obstruent devoicing is a universal process, one that is more
likely in word-final position. In some languages, the process neutralizes a
contrast, while in other languages it occurs, but the voicing contrast is
maintained. Phonetic studies have shown that voicing contrasts and
neutralizations are essentially temporal in nature, so NP is equipped to
explain final devoicing in all its incarnations.
2. Contrasting hushing affricates in Polish; Secondary
palatalization in Russian
The distinction between palato-alveolar and alveolo-palatal
affricates in Polish has long baffled non-natives trying to learn the language
(and natives trying to spell the language).
At the same time students of Russian have to learn sets of consonants
paired for palatalization. We shall see that the perception and production of
both contrasts depends largely on temporal differences in frication noise.
Incorporating these phenomena into NP can have interesting synchronic and
diachronic implications.
3. Palatalization of velars
This common diachronic change has received much attention, but only
recently have people considered the possibility that its roots lie in the
perception of temporal phenomena such as noise bursts, VOT, and formant
transitions. These features can display context-free as well as
context-dependent effects. NP has the flexibility to include both.
4. Miscellaneous
Numerous other problems in phonology, including spirantization,
intervocalic voicing, distinctive tones, the phonological status of glides, and
intonation, can be understood in terms of speech dynamics. NP offers a
framework that allows us to model this dynamism while retaining discrete units
of analysis.
Selected References
Cleary, Mirada &
Pisoni, David, 1998. Speech
perception and spoken word recognition:
Research and Theory. Research on Spoken Language Processing
22: 3-46.
Donegan, Patricia, J. 2002. Phonological processes and phonetic
rules.
Donegan, Patricia.J., 1979. On the Natural Phonology of Vowels.
Ph.D. Dissertation,
Ohio State University. Ohio State University Working Papers in
Linguistics 23. Republished 1985, New York: Garland Publishing Company.
Lass, Norman J., ed. 1996. Principles of Experimental Phonetics.
St. Louis: Moseby-Year.
Schwartz, Geoffrey, Forthcoming. Palatalization in Common Slavic: an
experimental
approach. Folia Linguistica Historica.
Schwartz, Geoffrey, Forthcoming. Temporal Phenomena in Polish
Hushing Affricates.
Poznań Studies in Comtemporary Linguistics.
Schwartz, Geoffrey, 2001. Speech Perception, Language Change, and
the Slavic
Palatalizations. Folia Linguistica
HistoricaXXI, 1-2, 277-300.
Stampe, David, 1973. A Dissertation on Natural Phonology. Ph.D
dissertation, University
of Chicago. Republished with annotations, 1979, New York: Garland Publishing Company.