Generative and Non-Linear Phonology

Durand Jacques, Jacques Durand

• Publisher: Routledge
• Year: 2014
• eBook ISBN: 9781317902263
• Topic: Languages & Linguistics
• Subtopic: Linguistics
• Index: Languages & Linguistics

Generative phonology is a developing field of linguistics, and is producing both rival interpretations and models. This book provides a clear and accessible evaluation of the debate. It provides a detailed overview of the main models, revealing that they are often complimentary rather than contradictory, and how these can be interconnect and be used together to explore the subject.

Chapter 1

Introduction

1.1 Scope of this book

In 1968, Chomsky and Halle’s monumental book The Sound Pattern of English, was published. While work in generative phonology stretched back to the late 1950s, The Sound Pattern of English (hereafter SPE) defined a paradigm that is often referred to as ‘standard’ or ‘classical generative phonology’. Since SPE offered inter alia a theory of the internal structure of sound-segments, a theory of levels and derivation, a theory of the linkup between syntax and phonology, much work since 1968 has been devoted to upholding or rejecting some fundamental thesis or other of this work.

Initially, the major body of assumptions made by Chomsky and Halle remained fairly stable and most researchers were content, as it were, to nibble at the SPE edifice. More recently, however, many of the SPE theses have been much more fundamentally questioned within a variety of frameworks including Autosegmental Phonology, Metrical Phonology, CV Phonology, Dependency Phonology, Three-dimensional Phonology, Grid-only Phonology, etc. All these models can be described as nonlinear by opposition to standard generative phonology where, as will be detailed in this book, a phonological representation is simply depicted as a linear arrangement of sound-segments (even though each segment is composed of simultaneously occurring features) and where phonological rules operate on such strings (they delete, insert or permute segments, or change their feature-values).

Despite undeniable differences in outlook between the current alternative frameworks, there are many lines of convergence between them and a fair measure of agreement on what a theory of phonology should account for. In particular, they all share the belief that phonological representations need to be much more articulated than traditionally assumed and that a number of phenomena (particularly, but not exclusively, stress and tone contours) cannot be appropriately accounted for if phonological representations are limited to string-like arrangements of segments and boundaries. Chapters 6, 7 and 8 are devoted to non-linear issues and it will be shown that the insights of various independent frameworks can be combined in ‘multidimensional’ representations. Whether the label ‘non-linear’ is always appropriate in connection with current proposals is not a question with which I will be concerned here (see J. M. Anderson 1987b). I shall use it as a cover term for a set of hypotheses concerning phonological structure which I will in turn subsume under the umbrella term ‘generative phonology’.

There are good reasons to retain the label ‘generative phonology’. Firstly, even if many current insights had been anticipated in other schools, the ‘non-linear’ models described here have clearly arisen from the practice of generative phonologists confronted with phenomena which were difficult to handle within the SPE notation. Secondly, portions of the standard theory (eg distinctive features) are often preserved in ‘non-linear’ accounts. Thirdly, the ‘non-linear’ frameworks do not exhaust the set of new proposals concerning phonological structure. Two important developments, presented in Chapter 5, Underspecification Theory and Lexical Phonology, address questions which are partially orthogonal to the linear-non-linear dimension. Finally, SPE-based analyses often form the backdrop to current work (cf eg Mohanan 1986; Kenstowicz and Rubach 1987) or are even recast, with minor changes, into new theoretical frameworks (cf Halle and Mohanan 1985 and Chs. 4–5). In other words, while there are many local changes, there is no sharp overall discontinuity with the past. In fact, many of the issues still debated in phonology – eg the role played by levels – go back to the practice of seminal thinkers such as Bloomfield or Sapir. Readers of recent technical articles certainly need to be as familiar with the conceptual and notational apparatus of SPE as with the tools devised within more recent frameworks.

I have therefore chosen to offer a broad view of the field and hope that this book will give the reader a good perspective on current developments while preserving a modicum of historical depth. I have presupposed very little knowledge of phonology on the part of the reader but assume a basic knowledge of phonetics of the sort which is available in Ladefoged’s A Course in Phonetics (1982). The rest of this chapter sketches an introduction to standard generative phonology by relating it to classical structuralist assumptions. However, no attempt is made at a real historical reconstruction and the reader is asked to exercise some charity in this connection. There are excellent discussions of past schools of phonology in S. R. Anderson (1985), Fischer-Jørgensen (1975), Lass (1984) and Sommerstein (1977), among others. Chapters 2 and 3 are devoted to the theory of distinctive features and related topics and sketch much of the background required to present recent technical developments. In Chapter 4, The derivational issue: aspects of the abstractness-concreteness debate’, I broach questions which, while not currently fashionable, confront any phonologist in his or her practice: What distance is allowed between underlying representations and surface representations? Should underlying representations be constrained? Should rules be ordered? While the framework of Natural Generative Phonology devised in the 1970s in reaction against SPE abstractness will be found wanting, it forces phonologists to think hard about their constructs and seems to me to deserve a place in the survey offered here. In addition, without a good understanding of the classical analysis of English segmental phonology presented in 4.2, frameworks such as Lexical Phonology would be difficult to present and discuss. Most of the chapters can be read in relative independence of one another but, of course, examples recur and there are many cross-references.

1.2 From classical phonemics to generative phonology

At the root of writings on sound structure from a classical or structuralist perspective is the assumption that, behind the diversity of sounds as physical units, there is in each language a small set of sounds whose function is to distinguish words from one another. For instance, the difference between the spoken realizations...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Preface
• List of Abbreviations
• 1 Introduction
• 2 The theory of Distinctive Features
• 3 Binarism, full and partial specification, markedness and gestures
• 4 The derivational issue: aspects of the abstractness-concreteness debate
• 5 Underspecification Theory and Lexical Phonology
• 6 Metrical structures
• 7 Autosegmental and Multidimensional Phonology
• 8 An outline of Dependency Phonology
• Appendix: Phonetic Symbols
• Bibliography
• Index