Linguistic Theory in America

281 Pages·1986·16.805 MB·English

by  Frederick J. Newmeyer (Auth.)

LINGUISTIC THEORY
IN AMERICA
Second Edition
Frederick J. Newmeyer
DEPARTMENT OF LINGUISTICS
UNIVERSITY OF WASHINGTON
SEATTLE, WASHINGTON
ACADEMIC PRESS, INC.
Harcourt Brace Jovanovich, Publishers
San Diego New York Berkeley Boston
London Sydney Tokyo Toronto
COPYRIGHT © 1986 BY ACADEMIC PRESS. INC
ALL RIGHTS RESERVED.
NO PART OF THIS PUBLICATION MAY BE REPRODUCED OR
TRANSMITTED IN ANY FORM OR BY ANY MEANS. ELECTRONIC
OR MECHANICAL. INCLUDING PHOTOCOPY, RECORDING. OR
ANY INFORMATION STORAGE AND RETRIEVAL SYSTEM. WITHOUT
PERMISSION IN WRITING FROM THE PUBLISHER.
ACADEMIC PRESS, INC.
1250 Sixth Avenue, San Diego, California 92101
United Kingdom Edition published bx
ACADEMIC PRESS INC. (LONDON) LTD.
24-28 Oval Road, London NW1 7DX
Library of Congress Cataloging in Publication Data
Newmeyer, Frederick J.
Linguistic theory in America.
Bibliography: p.
Includes indexes.
1. Generative grammar-History. 2. Linguistics-
United States-History. I. Title.
P158.N4 1986 415 86-10802
ISBN 0-12-517151-X (hardcover) (alk. paper)
ISBN 0-12-517152-8 (paperback) (alk. paper)
PRINTED IN THE UNITED STATES OF AMERICA
88 89 90 91 92 10 9 8 7 6 5 4 3 2
Preface to the Second Edition
Though I had no way of knowing it at the time, I wrote the first edition of
Linguistic Theory in America during the only major lull in syntactic research be
tween the mid 1950s and the present. As I was preparing the book in the late 1970s,
it had become unquestionably clear that generative semantics had collapsed, yet no
other worked-out alternative to the Extended Standard Theory had presented itself
as a pole of opposition to it. At the same time, the EST itself, which was bogged
down in its "Filters and Control" phase, seemed too unappealing to too many
linguists to be able to benefit to any great extent from its lack of competition. This
lull afforded me a golden opportunity to undertake a quarter-century retrospective
without having to worry that the contents of the next issue of Linguistic Inquiry
might render obsolete my overview of current work in the field.
But within months after the appearance of the first edition, the theories of govern
ment-binding, generalized phrase structure grammar, and lexical-functional gram
mar were introduced. The resultant explosion in work devoted to syntactic theory
exceeded anything the field had ever seen, surpassing even the late 1960s boom that
followed the publication of Aspects of the Theory of Syntax. More has been pub
lished in syntactic theory and analysis in the past half-dozen years than the previous
two dozen.
I think that the dust has settled sufficiently on the new frameworks to make the
time right for a second edition providing an account of their origins and develop
ment; I have therefore devoted most of the last chapter to outlining these exciting
recent trends in syntax. I have restructured the other chapters as well, most impor
tantly by adding a section on morphology and greatly expanding the discussion of
work in pragmatics. In order to keep the book to roughly the same length, I
ix
X PREFACE TO THE SECOND EDITION
truncated the possibly overly detailed account of the rise of abstract syntax and
generative semantics in the late 1960s.
I would like to acknowledge the helpful advice I received from a number of
persons, in particular Ron Amundson, A very Andrews, Noam Chomsky, Donald
Foss, John Fought, Morris Halle, Laurence Horn, James McCawley, Donna Jo
Napoli, Ivan Sag, Karl Teeter, and Alice ter Meulen. I owe a special debt of
gratitude to Timothy Sto well, Pauline Jacobson, and Annie Zaenen, who read and
commented on portions of the new material.
Preface to the First Edition
In this book I attempt to document the origins, birth, and development
of the theory of transformational generative grammar. Despite the fact that
it is now 25 years since Noam Chomsky completed his Logical Structure
of Linguistic Theory, what I have written is, to my knowledge, the only
work that combines a comprehensive account of the forging of modern
linguistic theory with a detailed elaboration and explanation of its devel
opment. I have not intended the book to be an introduction to linguistics,
and it cannot substitute for one. A minimal understanding of modern lin
guistic theory is presupposed throughout. Fortunately, there is no dearth of
texts in print whose purpose is to acquaint the reader with fundamentals of
theory. To my mind, the best is Neil Smith and Deirdre Wilson's lucid
Modern Linguistics: The Results of Chomsky's Revolution, which I can
recommend without reservation to the beginning student and advanced
scholar alike.
To keep this book to a manageable size, I have had to slight develop
ments in generative phonology. In fact, there is no discussion of develop
ments in phonology since the early 1960s. My omission is compensated for
in part by the existence of Alan Sommersteins Modern Phonology, which
covers the recent history of phonology in some detail.
The reader may be puzzled by the "in America" in the title. Why
should the discussion of a scientific theory be constrained by a national
boundary? In part this is a reflection of the immaturity of our science: It is
a sad commentary on the state of the field that we can still talk realistically
of "American linguistics," "French linguistics," and "Soviet linguistics."
XI
xii PREFACE TO THE FIRST EDITION
But more positively, my choice of title was dictated by the fact that trans
formational grammar HAS begun to internationalize; wkin America" was
necessary lest I appear to be slighting recent work of theoretical importance
in Japanese, Finnish, Arabic, and other languages unknown to me.
Since there is no such thing as totally unbiased historiography, it would
be Utopian to imagine that an author could be free from background as
sumptions or beliefs that might color his or her perception of events. As a
PARTICIPANT (however noncentral) in the history I describe, I might be
particularly open to charges of bias. However, I feel that my participation
has given me a real advantage: It has permitted me an inside view of the
field that would be denied to the more displaced historian. I hope that the
reader will find this to be to the book's advantage.
For those who may be interested in my background, I have been in
linguistics since 1965, the year I received a B.A. in geology from the
University of Rochester. Fortunately, a senior year course taught by the
late William A. Coates, called 'The Languages of the World," so intrigued
me that I gave up any dreams I may have harbored of a career in petroleum
engineering (or worse). I received a Master's in linguistics from Rochester
the following year and would have stayed there longer had I not attended
Chomsky's lectures at the LSA Institute at UCLA in the summer of 1966.
They convinced me to transfer as soon as possible to a transformational
grammar-oriented department. In 1967 I was admitted to the Ph.D. program
at the University of Illinois, where I studied syntax with Robert B. Lees,
Arnold Zwicky, and Michael Geis, and phonology with Theodore Lightner.
My last year of graduate work was spent as a guest at MIT. After receiving
a Ph.D. from Illinois in 1969, I was hired by the University of Washington,
where I have been teaching ever since, with the exception of leave time
spent at the University of Edinburgh, Wayne State University, and the
University of London.
My earliest theoretical commitment was to generative semantics, and
I contributed several uninfluential publications in defense of a deep-struc
tureless model of grammar (1970, 1971, 1975). By 1972 or 1973, I began to
have serious reservations about the direction generative semantics was
taking. Since then, I have identified myself loosely as an interpretivist,
without committing myself to any one particular model in that framework.
Some who know me as a Marxist may be surprised and, perhaps,
disappointed that there is no obvious "Marxist analysis" given to the events
I describe. For this I make no apology. There is simply no evidence that
language structure (outside of limited aspects of the lexicon) is, in the
Marxist sense, a superstructural phenomenon. Even if it were, however, it
seems inconceivable that events taking place in such a short period of time
and involving so few, so sociologically homogeneous, participants could
lend themselves to such an analysis. Those interested in my views on the
relationship between Marxist theory and linguistic theory may find useful
Preface to the First Edition xiii
my review of Rossi-Landi's Linguistics and Economics, published in Lan
guage in 1977.
A final word on chronology. Throughout the text, I cite books and
articles by the year of their first publication, not by the year that they were
written. Since publishing delays are uneven, this has resulted in some cases
in replies appearing to be older than the publications to which they are
addressed and in other possibly confusing aspects of dating. When delays
are extreme or when the time lag is significant in some respect, the year of
writing (in parentheses) appears after the year of publication in the reference
list.
Chapter 1
The State of American Linguistics
in the Mid 1950s
1.1. A PERIOD OF OPTIMISM
If American linguistics was in a state of crisis in the mid 1950s, few of its practi
tioners seemed aware of it. Einar Haugen ( 1951 ), in his overview of the field, wrote
that "American linguistics is today in a more flourishing state than at any time since
the founding of the Republic" (p. 211). Commentators boasted of the "great
progress" (Hall 1951:101), "far reaching advances" (Allen 1958:v), and "defini
tive results" (Gleason 1955:11) achieved by linguistics, which was "compared in
method with field physics, quantum mechanics, discrete mathematics, and Gestalt
psychology" (Whitehall 1951:v). Even Kenneth Pike (1958), often critical of many
of the assumptions of the dominant American linguistic theorists, felt moved to
write that "theirs [Harris's and Bloch's] is an attempt to reduce language to a
formal analysis of great simplicity, elegance, and mathematical rigor, and they have
come astonishingly close to succeeding" (p. 204).
More than self-congratulation was going on. The psychologist John B. Carroll
(1953) wrote that linguistics was the most advanced of all the social sciences, with a
close resemblance to physics and chemistry. And Claude Lévi-Strauss, probably the
world's foremost anthropologist, compared the discovery that language consists of
phonemes and morphemes to the Newtonian revolution in physics (Lévi-Strauss
1953:350-351).
There was a widespread feeling among American linguists in the 1950s that the
fundamental problems of linguistic analysis had been solved and that all that was
left was to fill in the details. The basic theoretical-methodological statements of
Bloch's "A Set of Postulates for Phonemic Analysis" (1948) and Harris's Methods
in Structural Linguistics (1951) seemed to render any more basic theoretical work
1
2 1. THE STATE OF AMERICAN LINGUISTICS IN THE MID 1950S
unnecessary. In fact, many linguists felt that the procedures had been so well
worked out that computers could take over the drudgery of linguistic analysis. The
time was near at hand when all one would have to do would be to punch the data
into the computer and out would come the grammar!
There was also a feeling that computers could solve another traditional linguistic
problem—translation. The idea of machine translation had been first suggested (in a
memorandum by Warren Weaver) only in 1949. By 1955, such translation work
was going on in three countries at half-a-dozen institutions. These six years were
enough to convert the skeptics, as William N. Locke put it in an enthusiastic review
article written in that year (Locke 1955:109).
Other postwar scientific developments seemed to be especially promising for
linguistics. A new field called "information theory" proposed methods of measur
ing the efficiency of communication channels in terms of information and redundan
cy. Shannon and Weaver (1949) in their pioneering study of information theory
pointed out the possible linguistic implications of the theory:
The concept of the information to be associated with a source leads directly, as we have seen, to a
study of the statistical structure of language; and this study reveals about the English language, as an
example, information which seems surely significant to students of every phase of language and
communication, (p. 117)
Shannon and Weaver's ideas were enthusiastically received by a large number of
linguists; prominent among these was Charles Hockett (1953, 1955), who set out to
apply the results of information theory in the construction of a Markov-process
model of human language.
Progress in acoustic phonetics also contributed to the general optimism. The
spectrograph, first made public in 1945, had replaced the inconvenient oscilloscope
as the most important tool for linguists in the physical recording of speech sounds.
There was a general feeling that spectrograms would help decide between compet
ing phonemicizations of a given language, the existence of which had posed a
perennial problem.
Finally, the synthesis of linguistics and psychology was adding a new dimension
to the study of language. The behaviorist theory of psychology developed by Clark
Hull (see Hull 1943) provided the linguists of the period with the theoretical appa
ratus they needed to link their approach to linguistic description with theories of
language acquisition, speech perception, and communication. By the early 1950s,
an interdisciplinary field of psycholinguistics had emerged, with important seminars
being held at Cornell in 1951 and Indiana in 1953. The progress reports from the
latter seminar were published in the first volume of papers dealing with language
and psychology, Osgood and Sebeok (1954), a work hailed by one reviewer as "a
scientific event of great importance" (Olmsted 1955:59). Among many other top
ics, the book dealt with issues such as the psychological verification of the phoneme
and the psychological criteria that would help decide between competing linguistic
analyses. The book itself inspired work conferences on many related subjects,
which themselves resulted in influential publications on content analysis (Pool
1.2. Structural Linguistics 3
1959), stylistics (Sebeok 1960), aphasia (Osgood and Miron 1963), and language
universals (Greenberg 1963).
Linguistics seemed so successful that it was being consciously imitated by the
social sciences.1 The anthropologist A. L. Kroeber (1952:124) asked "What is the
cultural equivalent of the phoneme?" and Kenneth Pike had an answer: the "behav-
ioreme." Pike (1954) was in the process of constructing a comprehensive theory in
which "verbal and non-verbal activity is a unified whole, and theory and meth
odology should be organized to treat it as such" (p. 2). Not only was it a cliché
"that what mathematics already is for the physical sciences, linguistics can be for
the social sciences" (LaBarre 1958:74; Le Page 1964:1), but many even held the
view that "as no science can go beyond mathematics, no criticism can go beyond its
linguistics" (Whitehall 195l:v). In this period the "linguistic method" was being
applied to the study of kinesics, folkloric texts, the analysis of the political content
of agitational leaflets, and much more.
1.2. STRUCTURAL LINGUISTICS
1.2.1. The Saussurean Heritage
The linguistics practiced in the United States in the 1950s, along with that in
much of Europe, owed an intellectual debt to the great Swiss linguist Ferdinand de
Saussure (1857-1913). Saussure's lecture notes, published posthumously as the
Cours de Linguistique Générale, represent a turning point in the history of lin
guistics. The central principle of the Cours is that a well-defined subpart of lan
guage can be abstracted from the totality of speech. This subpart Saussure called
"langue," which he contrasted with "parole," or "speech." Langue represents
the abstract system of structural relationships inherent in language, relationships
that are held in common by all members of a speech community. Parole, on the
other hand, represents the individual act of speaking, which is never performed
exactly the same way twice. Saussure compared language to a symphony. Langue
represents the unvarying score, parole the actual performance, no two of which are
alike.
Since in the Saussurean view, langue forms a coherent structural system, any
approach to language devoted to explicating the internal workings of this system has
come to be known as "structural linguistics." A structuralist description of a
language has typically taken the form of an inventory of the linguistic elements of
the language under analysis along with a statement of the positions in which these
elements occur. The point of such a rather taxonomic approach to langue was made
explicit in the Cours:
It would be interesting from a practical viewpoint to begin with units, to determine what they are and
to account for their diversity by classifying them. . . . Next we would have to classify the subunits,
•This point is argued at length in Greenberg (1973).