Tobin describes the methodological connection between phonological theory and phonetics by way of a comprehensive and insightful survey of phonology’s controversial role in twentieth-century linguistics. He reviews the work of Saussure, Jakobson, Troubetzkoy, Martinet, Zipf, and Diver, among others, and discusses issues in distributional phonology through analyses of English, Italian, Latin, Hebrew, and Yiddish. Using his theory to explain various functional and pathological speech disorders, Tobin examines a wide range of deviant speech processes in aphasia, the speech of the hearing-impaired, and other syndromes of organic origin. Phonology as Human Behavior provides a unique set of principles connecting the phylogeny, ontogeny, and pathology of sound systems in human language.
|Publisher:||Duke University Press|
|Series:||The Roman Jakobson series in linguistics and poetics|
|Sold by:||Barnes & Noble|
|File size:||880 KB|
About the Author
Yishai Tobin is Professor of Linguistics Ben-Gurion University of the Negev, Israel.
Read an Excerpt
Phonology as Human Behavior
Theoretical Implications and Clinical Applications
By Yishai Tobin
Duke University PressCopyright © 1997 Duke University Press
All rights reserved.
Phonetics versus Phonology
The Prague School and Beyond
The autonomy of linguistics and the use of the comparative method were suited to the task of describing and cataloguing the units and patterns in language. But in the 20th century linguists have started to ask why these units and patterns exist, i.e., they seek explanations for, not just descriptions of linguistic structure. We have reached the point where the continued re-working of the patterns in language no longer gives us answers to our questions.
— J. J. Ohala (1983c: 232)
Linguists have for many decades recognized that not all possible sequences of types of sounds occur in the morphemes of any single language. These restrictions were called "phonotactic" or "distributional" constraints by the classical structuralists and "morpheme structure" rules or conditions by generative phonologists. Although these constraints may originally have been conceived as merely descriptive statements one could make about the sound pattern of the language, there is sufficient evidence now that native speakers are also aware of them, that is, that many of the constraints noted by linguists are psychologically real.
— J. J. Ohala and M. Ohala (1986: 239)
Theoretical and Methodological Assumptions
In this chapter, I will briefly examine the historical development of the distinction between phonetics and phonology as presented by the structuralist forerunners of the theory of phonology as human behavior: Ferdinand de Saussure; the Prague school communication-oriented approach; and André Martinet and the functional school approach. In particular, I will focus on how each approach specifically deals with phonetics and phonology from the point of view of the interaction between the communication (i.e., the establishment of oppositions) and the human (i.e., the cognition, perception, and behavior of human beings) factors underlying its definition of language. I have selected these particular Saussurian, sign-oriented, or semiotic approaches to compare and contrast because they all share a similar definition of language within a general European structuralist framework.
Most of post-Saussurian twentieth-century humanistic thought (notably including semiotics and linguistics) as well as the social, behavioral, or cognitive sciences (e.g., sociology, psychology, and anthropology) ultimately can be defined and described as belonging to this larger structuralist paradigm. Yet, within this general structuralist framework as originally outlined in Saussure ( 1966:16), where linguistics was to be part of a larger science of semiology (later to be known as semiotics), linguistics (as opposed to semiotics) has often been considered to be the better-defined discipline and has even been viewed as the most developed branch within the larger structuralist paradigm. For example, in his brilliant and provocative introduction to Propp (1984: xx) (reviewed by Tobin and Simms 1988), Anatoly Liberman states: "In all discussions of structuralism, linguistics occupies a prominent place. Literary scholars, sociologists, and anthropologists constantly accuse their opponents of underestimating or overestimating the achievements of modern linguistics. Such arguments create the impression that linguistic structuralism is something well-defined, which is wrong. Practically all European and American linguistics after World War I has been structuralist."
The underlying assumption of my research is that every linguistic theory is the direct result of a specific set of theoretical axioms that is related to how the linguist defines language, defines a linguistic problem, determines the source, kind, and amount of data to be selected and analyzed, chooses a methodology to select and analyze the data, and compares and contrasts the analyses in light of all the above. These five criteria basically serve to describe how and what a particular linguist or school of linguistics views as the goals of linguistic research.
From a Saussurian, sign-oriented, or semiotic point of view, language may be defined as a system of systems that is composed of various subsystems (revolving around the notion of the linguistic sign) that are organized internally and systematically related to each other and that is used by human beings to communicate. Theoretically, this semiotic definition of language implies the dichotomy between langue and parole, langue being an abstract code composed of signals and meanings and their paradigmatic, or associative, and syntagmatic relations, a complex code that is shared by a community of speakers, and parole being the concrete and seemingly chaotic realization of this complex abstract code — exploited by individual speakers — to communicate specific discourse messages in different linguistic and situational contexts. The primary task of the linguist is to postulate the abstract code or system of systems of a language in order to explain the nonrandom distribution of the linguistic forms of that language in linguistic and situational discourse contexts. Methodologically, this definition of language implies a respect for and reliance on actual or real (as opposed to contrived or solely introspective) data culled from discourse and a commitment to deal with the human factor (i.e., the cognitive, perceptual linguistic and nonlinguistic behavior of human beings) as it is relevant to communication in different linguistic and situational contexts.
Many diverse models of linguistic analysis, both quantitative and qualitative, have been developed to describe, interpret, and explain concrete, individual linguistic phenomena (parole) according to general, communal, and abstract theoretical tenets (langue) in various structuralist guises and under quite different names. The notion of a general structuralist model (following a basic semiotic orientation) may be viewed as being a theoretical and methodological bridge between the abstract and the concrete levels of human linguistic phenomena.
Indeed, one may safely say that a model has been established (a model basically derived from phonology within linguistics) that has then been extended to all of what Culler (1976: 10?11) defines as the "fundamental problems of what the French call the 'human sciences': the disciplines that deal with the world of meaningful objects and actions (as opposed to physical objects and events themselves) ... based on the sign and on sign-systems [that] pave the way for a general study of the ways in which human experience is organized. In other words: a search for an underlying meaningful system to explain what may appear to be disparate, and even chaotic, superficial phenomena through the use of meaningful signs."
The structuralist model for linguistics is illustrated schematically in figure 1.1. Very often, this methodological model has presupposed an analogy between the model itself and the phenomena it purports to explain. This is not surprising when we consider the fact that it is the analyst's definition of language that determines which linguistic phenomena are chosen as being important and relevant to the theory and the analysis. Therefore, the model is predetermined by the theoretical units found in the analyst's definition of language.
Ferdinand de Saussure ( 1966: 8) first maintained that one of the chief tasks of linguistics was to define itself. He further claimed that the objects of study in language are not given in advance, as are the objects of study in other sciences. In short, for linguistics, it is the definition of language espoused by a theory that actually creates the object of study.
The structural model for linguistics that was later extended to other fields and disciplines was originally based on the sound systems of language, namely, the dichotomy between the abstract systems of hypothetical sound units postulated by the linguist and the concrete sounds of language produced and perceived by members of a speech community, which traditionally has been studied under the rubrics of phonology and phonetics, respectively.
More specifically, one may view the differences between phonetics and phonology in the following way. Phonetics is the description of what sounds occur where and which features (acoustic and articulatory) they are composed of (i.e., how they are articulated and perceived) — the "what," "how," and "where" of the realized sound system of (a) language:
what + how + where = description.
Phonology is a postulation and classification of the abstract units of the sound system of (a) language (e.g., the notions of phonemes and/or distinctive features etc.) as well as an explanation of the phonotactic distribution of sounds (both on the paradigmatic level, within different classes, and on the syntagmatic level) — "why" different sounds occur and do not occur in specific phonetic environments and/or in collocations with other sounds:
why = explanation.
Schematically, one may place phonetics and phonology within the structuralist framework as illustrated in figure 1.2.
The theoretical and methodological distinctions between (a possibly autonomous?) phonetics and (a possibly autonomous?) phonology or the extent to which phonetics may or may not contribute to phonology (or possibly vice versa?) has not always been clearly defined in linguistics and may even be a point of contention between different linguistic approaches. This chapter is based on four fundamental theoretical and methodological tenets:
1. The concept of the phoneme composed of distinctive articulatory and acoustic features — sounds that make communicative oppositions — reflects the abstract level of paradigmatic sound systems in langue and underlies the communication factor.
2. The concept of allophones composed of secondary or nondistinctive features — sounds that do not make communicative oppositions — reflects the concrete nonrandom syntagmatic distribution of the variations within a single and specific phoneme and may be related to the human factor.
3. In order to explain why the abstract phonemes of (a) language may also distribute — like allophones — in a nonrandom manner, we must appeal directly to the human factor in addition to merely classifying the phonemes and their distinctive features that underlie the communication factor.
4. It is only through the incorporation of the human factor as a means to explain the nonrandom distributions of sounds (beyond the traditional communicative notions of phonemes/allophones and distinctive and secondary features) that a structural analysis can achieve both abstraction (via the postulated units of analysis) and explanatory generality (the ability to explain why these different sound units function and distribute as they do in realized speech on the largest scope possible). Therefore, in this chapter, I will trace the development of the interaction between the communication-oriented approach of Ferdinand de Saussure and the Prague school and the potential (and often unrealized role) of the human factor as found in the work of André Martinet and William Diver as a means to better understand the differences between phonetics and phonology.
The Historical Development of Phonetics versus Phonology
Ferdinand de Saussure: The Notion of System
In their usually accepted senses today, the terms phonetics and phonology were basically reversed by Saussure. What today is generally called phonetics (i.e., the physiological and acoustic features of sounds and their classification according to those features) was referred to as phonology by Saussure. On the other hand, Saussure viewed phonetics as a historical science, a study of the diachronic development or evolution of sounds, that is, analyzing speech events and changes over time. Therefore, what he labeledphonetics (and not phonology) belonged to his definition of linguistics as the scientific study of language. Saussure's notion of phonology, however, as consisting of the study of the articulatory mechanism and various perceptual implications of the various speech sounds, all of which do not change, was, therefore, viewed merely as an auxiliary discipline belonging exclusively to the realm of speaking. Despite the auxiliary nature of phonology, however, Saussure noted that it served as a necessary tool for dealing with the written forms of language. Linguistic evidence furnished by written forms is valid only when interpreted, thus requiring an understanding of the phonological system of each language, that is, a description of its sounds. Each language, of course, according to Saussure, operates with a fixed number of well-differentiated phonemes.
Saussure laid the foundation for the distinction between phonology and phonetics (as established by his successors in the Prague school) by establishing the fundamental structuralist dichotomy of the abstract level of langue and the concrete level of parole. Furthermore, he laid the groundwork for the teleological, functional communication orientation in phonology as well as linking this teleological, functional communication approach to the role of the human factor. For Saussure, the description of sounds is directly linked with the articulatory act. The articulatory act is based on two elements that are constant, necessary, and sufficient for the production of sound: expiration and oral articulation (augmented by superimposed vibrations of the larynx and/or nasal resonances). Thus, Saussurian phonology consisted of the classification of sounds according to their oral articulation and degree of aperture and the organs involved in producing them:
1. occlusives (labials, dentals, gutterals);
2. fricatives or spirants (labiodentals, dentals, palatals, gutterals);
4. liquids (lateral articulations [l], vibrant articulations [r]); and
5. vowels (subclassified by degree of opening).
Saussure also recognized the role of the distribution of sounds and the need to study sounds as part of a spoken chain taking the reciprocal relations of sounds into account as part of the study of phonology.
Despite the fact that Saussure predated the sound spectrograph by approximately four decades (and hence the Jakobsonian notion of acoustic distinctive features), his delimitation of sounds within the spoken chain is implicitly based on auditory impressions. The phoneme, for Saussure, was inherently the sum of the auditory impressions as well as the articulatory movements and mechanisms. If the Saussurian sign was the invariable pairing of signal and meaning, the Saussurian phoneme was the fundamental pairing of the acoustic and the articulatory aspects of the signal. Furthermore, for Saussure, oppositions were everything. Therefore, the importance of analyzing phonemes was for the oppositions that they created in speech and not only the process through which these oppositions were obtained. Thus, we may view Prague school phonology as a natural extension of Saussurian phonology.
The Prague School: The Teleological-Functional-Communication Approach
The Prague school firmly distinguished phonology from phonetics. This distinction implied a theoretical model based on the dichotomy between langue and parole and a questioning of the Saussurian dichotomy between synchrony and diachrony on the sound level. Furthermore, it brought the teleological-functional notion of communication to the forefront of phonological analysis, making it the raison d'être of phonology.
Le Roy (1967: 64?65) sums up the role of the Prague school in distinguishing the difference between phonetics and phonology in the following way:
Phonetics which was formerly a science of observation — the ancient Indians achieved astonishing results in this field — and which studies the sounds of human speech from the articulatory and acoustic points of view, has been much improved since experimental methods were first applied to it at the end of the last century. From 1926 onwards a group of linguists from the Linguistic Circle of Prague, arguing from the fact that phonetics, since it started using apparatus and to be studied in the laboratories, has made enormous progress but at the same time has moved further and further away from linguistics, devised a completely different method for studying the sounds of language. At the first International Conference of Linguists at the Hague in 1928, R. Jakobson, S. Karcevskij and N. S. Trubetzkoy presented their famous Proposition 22 which marks the birth of the new discipline of phonology. The proposition dealt with the "significant differences" that characterize the elements of every phonological system and of the "phonological correlations" that are constituted by series of binary oppositions.
Starting from the principle that any science, to be valid, must be general, the members of this school wished to reintegrate phonetics into the framework of linguistics by rescuing it from the examination of the purely individual facts of parole and applying it to the more general study of langue, with particular emphasis on the relationship between the sound (or the complex of sounds) and its meaning. One might also say that phonetics studies what is actually pronounced, something that varies from one individual to another and even within the usage of one individual, whilst phonology studies what one is conscious of pronouncing or of hearing pronounced, and this is a constant within a given linguistic community.
Excerpted from Phonology as Human Behavior by Yishai Tobin. Copyright © 1997 Duke University Press. Excerpted by permission of Duke University Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of Contents
ContentsList of Figures and Tables,
I: Phonetics and Phonology: A Historical Overview,
1 Phonetics versus Phonology,
2 Phonology as Human Behavior,
II: Phonology as Human Behavior across Languages,
3 The Italian and Latin Connections,
4 The Hebrew and Yiddish Connections,
III: Phonology as Human Behavior. Panchronic, Pedagogical, and Textual Applications,
5 Panchronic Applications in Hebrew Phonology,
6 Pedagogical and Textual Applications,
IV: Phonology as Human Behavior: Developmental and Clinical Applications,
7 Developmental Phonology and Functional Clinical Applications,
8 Clinical Applications to Organic Disorders,
V: Phonology as Human Behavior: Audiology and Aphasia,
9 Audiology, Hearing Impairment, and Cochlear Implants,
Appendix 1: Phonology as Human Behavior: Take-Home Assignment,