1. Introduction

The linguistics program at MIT evolved in a highly unusual manner for an academic discipline with ancient beginnings. Essentially, the linguistics section was formed not so much to teach a body of knowledge as to contribute to it. From the beginning, research has been primary at MIT. In fact, the linguistics group was originally a research group, associated with the Research Laboratory of Electronics (RLE), with certain of its members also teaching language courses within the Department of Modern Languages. This dual affiliation still exists, with RLE providing the group with various administrative services, a portion of its funding, and physical facilities, but with the majority of the staff holding faculty appointments in what is now the Department of Foreign Literatures and Linguistics.

2. Background: To 1951

In an easily traceable chronology, the origins of the framework within which linguistics was to grow at MIT lie in the development of radar. Even before World War Two, there was increasing interest at the Institute in the new fields of communications and electromagnetic theory, with a concomitant increase in collaboration between physicists and electrical engineers. After the Radiation Laboratory was organized in 1940 to participate in the development of an airborne radar, scientists from still other fields joined the related projects. The impressive output of their combined efforts, both applied and theoretical, encouraged those involved to seek to extend the collaboration into peacetime. Thus, when in 1946 the Radiation Laboratory ceased to exist, a new Research Laboratory of Electronics was formed to absorb its programs. In order to resolve problems of funding and of title to incalculably valuable projects and resources which had been financed by the government, the support of basic research at RLE and other such facilities was given to agencies of the Armed Services. The terms of this sponsorship were such as to allow continuation of the open research climate that had proved so fruitful.^[1]

RLE was a new kind of laboratory. It existed independently of any department at MIT and still does, although laboratory workers are often affiliated with one of the teaching departments as well. The original staff consisted of physicists and electrical engineers, but quite quickly mathematicians, biologists, psychologists, and chemists, as well as linguists and others, were attracted by the communality of interests and the research climate that encouraged it.^[2]

From the start, much of the investigatory effort at RLE was devoted to communications problems of various sorts: in 1946, when the laboratory was formed, one of the four major research groups was labeled “Communications and Related Subjects.” It was probably Norbert Wiener who was most personally responsible for the development of the communication sciences at MIT. His cybernetics, along with Claude Shannon’s mathematical theory of communication, stimulated interested scientists from various fields to probe further and further into the implications of information theory and communications engineering.

In addition to and separate from the RLE communication sciences group there was at MIT an Acoustics Laboratory where work was being carried out on speech and phonetics. Out of this laboratory came the still very active Speech Analysis Group in RLE headed by Kenneth N. Stevens.

An early impetus for the formation of the linguistics group at MIT came from a number of speech communication conferences organized originally by William N. Locke, then Chairman of the Department of Modern Languages. The first meeting on speech analysis, chaired by Locke, was held on 2 December 1949. Marking the opening of formal discussion between linguists and engineers as well as others, the conference was attended by forty scientists, including Roman Jakobson of Harvard, commonly regarded as one of the first linguists to sense the potential of the new scientific developments, Gunnar Fant of the Royal Institute of Technology of Stockholm, who spent some time at MIT and was to coauthor the pioneering Preliminaries to Speech Analysis, and John Lotz of Columbia, who soon afterward gave a series of lectures at MIT on the problems of linguistics.

The interest generated by this meeting led to a more formal Speech Communications Conference, the first of many. It was held at MIT from 31 May to 3 June 1950 under the auspices of the Acoustical Society of America, the Carnegie Project on Scientific Aids to Learning at MIT, and the Psycho-Acoustic Laboratory at Harvard University. An interesting picture of the various approaches to speech communication at the time is afforded by the titles of the papers presented at that meeting:^[3]

• “Introduction: A Definition of Communication,” S. S. Stevens
• “Information Theory Point of View in Speech Communication,” R. M. Fano
• “Speech, Language, and Learning,” Norbert Wiener
• “Typology of Languages,” Paul Menzerath
• “Description of Language Design,” Martin Joos
• “Relation of Phonetics and Linguistics to Communication Theory,” Oliver Straus
• “Speech and Language,” John Lotz
• “Pathology in Speech Communication,” Ira J. Hirsh
• “Language Engineering,” George A. Miller
• “Communication Patterns in Task-Oriented Groups,” Alex Bavelas
• “Sonograph and Sound Mechanics,” Jean Dreyfus-Graf
• “Calculation of Vowel Resonances, and an Electrical Vocal Tract,” H. K. Dunn
• “An Apparatus for Speech Compression and Expansion and for Replaying Visible Speech Records,” F. Vilbig
• “Spectrum Analysis,” Franklin S. Cooper
• “Correlation Function Analysis,” L. G. Kraft
• “System-Function Analysis of Speech Sounds,” W. H. Huggins
• “Portrayal of Some Elementary Statistics of Speech Sounds,” S. H. Chang
• “Autocorrelation Analysis of Speech Sounds,” K. N. Stevens
• “Theory of Operation of the Cochlea: A Contribution to the Hydrodynamics of the Cochlea,” C. F. Ranke
• “Theory of the Acoustical Action of the Cochlea,” J. Zwislocki
• “Neurophysiology of the Auditory System,” Robert Galambos
• “Auditory Masking and Fatigue,” Walter A. Rosenblith
• “Binaural Localization and Masking,” W. E. Kock
• “Reversed Speech and Repetition Systems as Means of Phonetic Research,” W. Meyer-Eppler

As far as the incipient approach to linguistics was concerned, it was clear that interest at this point was directed toward a machine-type analysis of language.

3. The Beginnings: 1951-1955

At the Research Laboratory of Electronics, William Locke was directing a small group doing research on the acoustics of speech. The facilities and equipment were made available to interested scholars outside of MIT, one of whom was Morris Halle, then studying at Harvard. Born in Latvia, Halle specialized in Slavic languages in his Harvard Ph.D. studies but, as an associate of Roman Jakobson’s, had the deep concern for general properties of language that had characterized traditional linguistics and much of Jakobson’s work. In 1951 Halle officially joined the faculty of the Department of Modern Languages, where he taught Russian and German, and the staff of RLE, where he participated in phonetic research. It can be said that Morris Halle’s official arrival marked the actual beginning of the linguistics group at MIT, not only because he was its first member but also because it is generally acknowledged that he has been its chief architect and builder as well.

The character of the speech investigations carried out by Halle and others at RLE in the early fifties was distinctive in its linguistic orientation. Research emphasis was placed on the relationship between the behavior of the human vocal tract and its acoustical output, and the resulting studies provided much information on how the anatomical makeup of man’s vocal tract is utilized in producing the acoustical signal of speech. Examples of acoustical investigation of the period include measurements of the acoustical spectra of consonants, studies of the sounds of Russian, and an attempt to realize a complete automatic recognition scheme for English speech (Hughes (1960)). Speech perception also came under examination, leading to such results as evidence of the influence of context on the perception of stop consonants (Schatz (1954)).

Generally considered to be the most important work to come out of this period, however, because of its influence on the course of phonological study, was Preliminaries to Speech Analysis by Jakobson, Fant and Halle (1952). According to the Preface: (p. v):

This report proposes some questions to be discussed by specialists working on various aspects of speech communication. These questions concern the ultimate discrete components of language, their specific structure, their inventory in the languages of the world, their identification on the acoustical and perceptual levels and their articulatory prerequisites.

Preliminaries was thus an attempt to categorize the phonetic capabilities of the human vocal apparatus by positing a framework underlying the sound structure of all natural languages. Speech sounds are approached as complexes of properties, regarded for a variety of reasons as binary categorial markers. Termed “distinctive features,” these properties are the elements made use of by all languages in the formation of their sound systems.

This conception of sound structure represented an elaboration of ideas of Jakobson’s dating back as far as the 1920’s, a development made possible by the refined equipment newly available to phoneticians. At the same time, in its approach to phonology Preliminaries represented a departure from contemporary linguistic studies.

There was still another area of study which turned out to be a source of stimulation and manpower for the nascent linguistics group, namely, machine translation. Also carried out at RLE as part of the general program of investigation into communication problems but separate from speech research, work on the translation of languages by computer during the 1950’s attracted worldwide attention. The tie-up with linguistics was a direct one: in order to program a machine to translate one language into another, it is necessary to have a thorough grasp of the structure of both languages.

The first conference on machine translation was held at MIT in June 1952. At RLE Y. Bar-Hillel, for one, worked in this area. Also interested was William Locke, who later coedited a book on Machine Translation of Languages (Locke and Booth (1955)). Then, in 1953, Victor Yngve, a physicist who had concerned himself with the idea of translating languages by machine as early as 1949, arrived at MIT. He became director of the Mechanical Translation Research Project at RLE and, with Locke, edited the journal Machine Translation.

Although the course of the future at MIT lay in what was beginning to happen in machine translation and speech analysis research at RLE, there were also, beginning in 1950, two undergraduate linguistics courses. These were taught by Frederick Bodmer, author of The Loom of Language (1944) and a lecturer in the Department of Modern Languages until 1955. The 1950-1951 MIT catalog describes the courses as follows:

L71, L72. SOCIAL LINGUISTICS. Seminar to examine the social functions of language, the interrelations of language and culture; types of languages, primitive and modern; advantages and disadvantages of various systems of writing; the implications of illiteracy; international communication; language planning.

L73, L74. APPLIED SEMANTICS. Seminar to stimulate meaning-consciousness through the discussion of questions such as: What happens when we make or interpret a statement? Why do words get confused with things? What causes semantic blockages? How can they be removed? What are the verbal means of mobilizing emotions? How is language used by the politicians, the salesmen, etc.? What are the most commonly practiced tricks in verbal argument? When is clarity not enough?

In 1952, courses L71, 72 became L75, LANGUAGE AND SOCIETY. The course descriptions remained the same in both cases until 1955.

In their first two years the linguistics courses attracted about twenty students in all. But by 1952 the combined enrollment rose to about forty and remained there until 1961, by which time the offerings had been expanded somewhat.

Courses arising directly out of the communication research at RLE were also beginning to be offered in various departments. The MIT catalog for 1952-1953 lists the following psychology course taught by G. A. Miller:

14.77. PSYCHOLOGY OF COMMUNICATION. Survey of phonetic, acoustic, perceptual, statistical, developmental, behavioral, and social studies of the psychology of language and communication. Particular emphasis is placed upon the effects of verbal habits upon perception, learning, and thinking.

Then, in 1953, Morris Halle and Walter Rosenblith, a specialist in communications biophysics, began to teach a graduate course offered jointly by the Electrical Engineering Department and the Modern Language Department:

6.696. HEARING, SPEECH AND LANGUAGE. Discussion of properties of signals in audio communication systems in the light of characteristics of listeners and speakers. Auditory thresholds, masking, perception of complex acoustic stimuli. Production of natural and synthetic speech. Physical methods of speech analysis. Effects of filtering and clipping. Linguistic units of analysis. Statistical properties of speech signals. Intelligibility criteria for audio communication systems. Integration of experimental data on hearing and speech with recent developments in linguistics and information theory.

The direct descendants of these courses continue to be given by the corresponding departments at MIT today.

To summarize, by 1955 the situation at MIT is as follows. Morris Halle, William Locke, and Victor Yngve are on the faculty of the Modern Language Department and, at the same time, on the staff of RLE, where, as part of the group investigating various aspects of communication, they and others are working on speech analysis (Halle) and machine translation (Yngve). The general tone of the research is to bring to bear upon the study of language the results of the wartime advances in fields such as information theory, communications engineering, and acoustics, while at the same time making available to interested specialists in other areas the results of language study. This was directly reflected in the coursework that came out of RLE research, namely, Halle and Rosenblith’s HEARING, SPEECH, AND LANGUAGE, and Miller’s PSYCHOLOGY OF COMMUNICATION. Also being offered were two undergraduate courses in linguistics, taught by Bodmer. Finally, the most significant published work of the pre-1955 years was Preliminaries to Speech Analysis (Jakobson, Fant, and Halle (1952)), which was to critically affect the nature of phonological investigation for a great many linguists.

4. 1955-1960

Until 1955 the focus of linguistic study at MIT was on the sound structure of language, that is, phonology and morphology, with special emphasis on applied aspects in the form of speech analysis and machine translation. With the arrival of Noam Chomsky that year, however, more and more attention was to be devoted to syntax and to general linguistic theory.

Chomsky had received an M.A. and Ph.D. in linguistics from the University of Pennsylvania, where he studied under Zellig Harris. After four years as a Junior Fellow at Harvard, he came to MIT in 1955: he taught Scientific French and Scientific German, as well as the two general linguistics courses in the Department of Modern Languages and at the same time worked with the machine translation group at RLE. The research situation there appeared well-suited for Chomsky, who was already bringing to bear on linguistics timely developments in philosophy, mathematics, and psychology. In 1953 he had published an article on “Systems of Syntactic Analysis,” and The Logical Structure of Linguistic Theory, which formed the basis for Syntactic Structures, was completed in 1955. Chomsky’s central interests were already expressed in the Introduction to The Logical Structure of Linguistic Theory. First there was the development of

a unified approach to syntactic theory as a whole, including a theory of relations among sentences as an integral part. This latter subtheory… turns out to play a central role in the establishment of the fundamental units of syntax, i.e., in the procedures of constituent analysis… The central conclusion is that a new level of transformational analysis is needed…

Furthermore,

one cannot describe a linguistic system in any meaningful way without some conception of what is the nature of such a system, and what are the properties and purposes of a grammatical description. For this reason it is important to develop a precisely formulated and conceptually complete construction of linguistic theory, based on the clearest possible elementary notions… (Chomsky (1955, pp. 1-3)).

That is, he posited a theory of transformations which was conceived as part of a theory of grammar, “part of an attempt to construct a formalized general theory of linguistic structure and to explore the foundations of such a theory” (Chomsky (1957, p. 5)).

It could probably be agreed that in the two decades since Chomsky’s first publications, his work has stimulated a re-evaluation and ultimately a restatement of the fundamental problem of linguistics. First to feel the impact, quite naturally, was MIT. There the focus of attention began to shift away from the use of linguistic knowledge and methods in the solution of engineering problems. Instead, linguistic theory became the prime concern, and the complexities of syntax attracted more attention as their susceptibility to study appeared to increase.

The evolving theory began to be presented in a number of published works and at various professional meetings, and although there was no formal linguistics program until 1961, interested scholars began to come to MIT. The machine translation project, especially, aided by a Carnegie grant, was able to support the work of a considerable number of linguists during this period, including, among many others, not only Chomsky but Applegate, Matthews, Klima, Lees, Lukoff, and Viertel.

In 1956 Joseph Applegate, who like Chomsky had a Ph.D. in linguistics from the University of Pennsylvania, joined RLE and the Modern Languages Department where he was to relate ongoing work to the study of Berber languages. That same year G. Hubert Matthews, a Ph.D. from the University of Pennsylvania, arrived at RLE and in 1961 became part of the Modern Language faculty as well. With a background in mathematics, German literature, linguistics, and anthropology, he was to become particularly interested in formal languages and in studying American Indian languages in the light of new theoretical devices. Edward Klima, a Ph.D. candidate in linguistics at Harvard, joined the Modern Languages Department in 1957 and the following year became associated with the machine translation group as well. His interests were to center around comparative and historical linguistics, English, and later the study of language acquisition.

In 1959 Jerry Fodor became an instructor in the Department of Humanities. After then spending some time in 1960, along with Jerrold Katz, as a research assistant on the Transformation and Discourse Analysis Project at the University of Pennsylvania, he joined the staff of RLE in 1961. Since that time he has done much work in the philosophy of language, in language perception and acquisition, and with Katz, who also came to RLE in 1961, in the development of a semantic theory. In 1960 Roman Jakobson was officially appointed an Institute Professor in the Department of Modern Languages while still retaining his affiliation with Harvard. It should be pointed out, however, that Jakobson had not only been a Visiting Professor in 1957 but had, from the beginning of Morris Halle’s affiliation with MIT, been deeply interested in the direction of research there. Thus he maintained a close and continuous association with the MIT group long before his official appointment.

The published studies that came out of MIT during the late fifties represented the beginnings of what some have called a new “paradigm” of work in linguistics. Chomsky’s Syntactic Structures in 1957 was preceded not only by the formidable unpublished work on which it was based, i.e., Chomsky (1955), but also by Chomsky, Halle, and Lukoff’s “On Accent and Juncture in English” (1956), where it was shown that intricate stress patterns could be predicted from syntactic structure, and by Chomsky’s “Three Models for the Description of Language” (1956). In 1959 “On Certain Formal Properties of Grammars” appeared, and that same year (1959) saw the publication of what struck many as an explosive review by Chomsky of Skinner’s 1957 work, Verbal Behavior.

Although a good portion of these studies, by virtue of their mathematical approach, seemed inaccessible to many linguists, the theories they represented could not long go unnoticed in a small field where ordinarily everyone read everything and knew everyone. Of great help was the more reachable review of Syntactic Structures in Language (1957) written by Robert Lees, a member of the machine translation project and later a student of Chomsky and Halle’s. Then in 1958 the third of a series of conferences held in Texas on the subject of Problems of Linguistic Analysis in English provided an opportunity for Chomsky to personally present, explicate, and defend his work. Out of this conference came the well-known article on “A Transformational Approach to Syntax” (1962). In 1958 five representatives of MIT presented papers at the annual meeting of the Linguistic Society of America – Applegate, Halle, Lees, Matthews, and Lenneberg (a psychologist doing work related to the MIT approach to language). The following year, 1959, Halle’s Sound Pattern of Russian was published. And in 1960 Halle, Jakobson, and Chomsky were on the program committee of a symposium held in New York on the Structure of Language and its Mathematical Aspects. Chomsky’s “On the Notion ‘Rule of Grammar'” (1961) was presented at this meeting, as was Halle’s paper “On the Role of Simplicity in Linguistic Descriptions” (1961).

By 1960, then, the “MIT school,” as it had come to be known, had undoubtedly emerged as a force to be reckoned with in linguistics, though whether for better or worse was still very much an open question for a good many workers in the field.

An understanding of the early history of MIT linguistics involves explicit recognition of the fact that it was not developed in an intellectual or academic void. By itself it could be interpreted as a synthesis of the seventeenth- and eighteenth-century interest in universal and philosophical grammar and the scientific advances of a new age. However, between the classical period and the late fifties linguistics had taken on a different form, and it was the existing science that the new theories and theorists had to react to. The profoundly different approach that transformationalists espoused was offered to a well-established field, one that had attained a highly refined stage in many areas. In some respects the new approach was seen as growing out of the accomplishments of structural linguistics, particularly with regard to the emphasis on precision of formulation. But in still other, deeper ways it was regarded as a reaction against what were judged to be serious shortcomings in the linguistic science of the fifties. Inevitably, then, a substantial part of the work of this period dealt not only with the revised theoretical and methodological principles themselves, but also with what were interpreted, on the basis of these principles, as weaknesses in modern structuralism. As a result a portion of the writing and speaking took on a certain polemic tone that was to have some effect on work in the field for several years.

Returning now to the increasing exposure MIT ideas were receiving by the end of the fifties, we may note the accompanying increase in interest in them, particularly on the part of scholars attracted to the math/engineering and language/linguistics combination. By 1959 two Ph.D. theses were being completed under the auspices of the Electrical Engineering Department by students working with Chomsky and Halle: one dealt with speech analysis (Hughes (1960)) and the other was Lees’s well-known syntactic work, published the following year (1960) as The Grammar of English Nominalizations. The years 1959 and 1960 saw, in addition, two undergraduate linguistics theses and a Master’s thesis in speech analysis, again carried out under the supervision of Chomsky and/or Halle. Also studying with them on one basis or another were a linguistics student from Columbia, a speech student from Boston University, and a Slavic specialist from Harvard. Still another group of students pursuing academic degrees at different levels and with different specialties were associated with the machine translation project. They included linguists, logicians, mathematicians, and engineers, and they were affiliated with, for example, Harvard, UCLA, Columbia, the University of Texas, Brown, and MIT.

During the years under examination, the actual course offerings in linguistics were somewhat extended. Such expansion was limited, however, by the absence of a degree-granting program, as well as by the research-centered orientation of a still new subject area that was just developing teachable subject matter.

The first change evident in the catalog was in the course description of one of the original undergraduate linguistics courses. In 1955, the year Chomsky took over these courses, the catalog remains unchanged with respect to L75, LANGUAGE AND SOCIETY, although the actual course content undoubtedly underwent some modification. The other course, however, formerly APPLIED SEMANTICS, appears with a new title, a new number, and a new description:

L78. LINGUISTIC STRUCTURE. Problems and methods of the descriptive study of language. Study of form in language: definition and determination of linguistic elements; techniques of analysis; mathematical models for linguistic behavior (communication theory, logical syntax); construction of an objective and abstract theory of formal linguistic structure; systematic application of this theory to the description of English structure. Study of meaning in language: relation of linguistic form and meaning; various approaches to a theory of meaning; contemporary philosophies of language and the influence of symbolic logic; extensional and intensional structure. Comparison of the logic of natural languages with that of special purpose language systems (mathematics, science).

The next change appears in the 1957-1958 catalog, which lists a new undergraduate course in phonology taught by Morris Halle:

L76. PROBLEMS OF PHONOLOGY. Aims and principles of a scientific description of the structure of a spoken language. The mechanism and the acoustics of speech. The phonemic system of a language. The pair test. Distinctive features as a descriptive framework. The phoneme. The interrelation between phonology and other linguistic levels. Some problems of historical phonology. Practical exercises in preparing a phonemic analysis of a language with the help of a native informant.

In 1959-1960 the catalog description of Chomsky’s course on LANGUAGE AND SOCIETY was modified, and the following year Halle took over the teaching of this introductory course:

L75. LANGUAGE AND SOCIETY. Introduction to the scientific study of human language. Outline of syntactic theory; brief survey of English syntax. Elementary data on production and physical nature of speech. The phoneme and distinctive features. Phonemic systems. Morphology: review of some morphological devices. Languages of the world, their historical relationships and structural classification. Speech and writing. Language, culture, and cognition.

Included in the same catalog were two new listings which marked the beginning at MIT of official credit for participation in guided research in linguistics:

L71. SPECIAL PROBLEMS IN LANGUAGE AND LITERATURE.

L72. SPECIAL PROBLEMS IN LANGUAGE AND LITERATURE. Advanced work in literature or linguistics for unusually well-prepared students. Offered by members of staff in their special fields.

In the following year, 1960-1961, a graduate course taught by Yngve was added to cover mathematical aspects:

L79. LANGUAGE, SYMBOLIC PROCESSES, AND COMPUTERS. Symbols, their storage and manipulation: formal systems of symbols, natural languages, computer programming languages, codes, notation; systems of mathematics, logic, physics, etc.; relation of symbols and symbol systems to the physical means of representation and manipulation; relation of symbols and symbol systems to the subject matter represented and its systematization; symbolic processes, induction, deduction, reasoning; translation and translatability.

To summarize, by 1960 the MIT “school” of linguistics had gained a considerable reputation. Interest on the part of graduate students at the Institute and elsewhere was growing, and a few new courses were set up to teach the new subject matter. Involved in MIT research projects were from eight to twelve graduate students from other universities, and additional requests for graduate study in linguistics were being received at a growing rate.

Not surprisingly, then, a proposal for a graduate program in linguistics at MIT was prepared by a committee consisting of Roger Brown (a social psychologist with an interest in the psychology of language), Noam Chomsky, Morris Halle, Roman Jakobson, William Locke, Walter Rosenblith, Jerome Wiesner (now President of MIT and at that time Director of RLE and a leader in the development of communication sciences), and Victor Yngve. The completed proposal was submitted by the Department of Modern Languages on 1 July 1960. The recommendation made was that a graduate program in linguistics be introduced at the Institute in the fall of 1961, under the Department of Modern Languages, leading to the degree of Doctor of Philosophy in Linguistics. A preference was expressed for at first accepting only candidates who had already proven their competence in one of the two main areas with which the group was concerned, language-linguistics or mathematics-logic-computers.

Our plan is to emphasize the areas in which we have strength and in which training is badly needed by young linguists all over the world. It is not the intent of the Department to staff what might be called a “classical” linguistics department. Thus, we do not expect to have subjects in many of the conventional areas such as the history of particular languages and their development or methods of field work. Students would be expected to master these on their own, or to take subjects in other institutions, such as summer institutes of linguistics.

While linguistics will form the core of our curriculum, we are in a particularly favorable position to capitalize on developments in neighboring fields. The interdepartmental tradition of the Institute makes it possible for linguistics to thrive here in a new and fruitful way. Active interest in the complex problems of language has been shown by members of several departments, and the work in linguistics as outlined here is an essential part of the program of the Center of Communication Sciences. Much work in acoustics, communication sciences, logic, mathematics, neurology, physiology, psychiatry, and psychology is pertinent to linguistics. We hope that some of our students, as well as students from the related fields, will wish to do serious work in the border areas…

(“Graduate Work in Linguistics at the Institute: A Proposal from the Department of Modern Languages” (1960)).

The program was visualized as beginning modestly, with a gradual increase over the first three years to about fifteen students. It was hoped that needed fellowships would come in large part from the National Science Foundation, which had been supporting the research programs in linguistics, and from the National Institutes of Health.

The subjects with immediate application to linguistics being offered at the time were:

6.696. HEARING, SPEECH, AND LANGUAGE. (Graduate) (Described previously.) Rosenblith, Halle

21.76. SYMBOLIC LOGIC. Introduction to the principles of deductive logic. The major results concerning the structure of mathematics. The place of modern logic in philosophy; its influence on the formulation of philosophical problems and on the search for methods for their solution. Chomsky

14.77. PSYCHOLOGY OF LANGUAGE AND COMMUNICATION. Statistical and experimental studies of the psychology of languages. The effects of verbal habits on perception, learning, and communication. Howes

6.574. STATISTICAL THEORY OF INFORMATION. (Graduate) Introduction for graduate students to the quantitative study of communication processes. Definition of a measure of information and study of its properties; representation of messages; efficient coding, transmission in the presence of random disturbances, channel capacity; coding to increase the reliability of transmission. Extension of the theory to continuous channels; signal detection in the presence of disturbances. Fano

L75. LANGUAGE AND SOCIETY. (Described previously.) Halle

L76. PROBLEMS OF PHONOLOGY. (Described previously.) Halle

L78. LINGUISTIC STRUCTURE. (Described previously.) Chomsky

L79. LANGUAGE, SYMBOLIC PROCESSES, AND COMPUTERS. (Described previously.) Yngve

Recommended for addition, ultimately, should the program be approved, were the following subjects:

• Scientific Study of Language
• History of Indo-European Languages
• Structural Analysis of English
• Structural Analysis of Russian
• Problems of Machine Translation
• Seminar on Non-Indo-European Languages
• Seminar on General Linguistics
• Philosophy of Language (Dept. of Humanities)
• Language and Literature (Dept. of Humanities)
• Language Learning and Disturbances (Dept. of Psychology)

According to Institute rules and regulations, the authorization of a department to grant a new graduate degree had to come from the MIT Corporation, on the basis of a favorable recommendation from the President on administrative aspects of the proposed program and a recommendation from the MIT Faculty on its academic aspects. The Faculty based its recommendation on the report of the Committee on Graduate School Policy. And this Committee, in turn, generally based its recommendation on the report of a small ad hoc committee set up expressly for the purpose of examining and appraising the academic resources within and available to the department seeking to offer the degree.

The MIT Administration made the decision that the field of linguistics justified support at MIT at a level consistent with its offering the Ph.D. degree. To set the remainder of the necessary machinery in motion, an Ad Hoc Committee on the Linguistics Doctorate was formed to evaluate the adequacy of MIT’s resources for offering the proposed program, and a meeting of the committee was held at MIT on 5 October 1960. The members were:

Professor Patrick Hurley (Chairman of the Committee)
Executive Officer
Department of Geology and Geophysics
MIT

Dr. Bernard Bloch
Agassiz Professor of Oriental Languages and Literature
Yale University

Dean Peter Elder
Graduate School of Arts and Sciences
Harvard University

Dr. Einar I. Haugen
Visiting Professor
Department of German
Harvard University
and
Thompson Professor of Scandinavian Languages
University of Wisconsin

Professor Donald Marquis
School of Industrial Management
MIT

Professor Robert Solow
Department of Economics
MIT

Also present at the meeting as invited guests were Harold Hazen, Dean of the MIT Graduate School, John Burchard, Dean of Humanities, William Locke, Roman Jakobson, Noam Chomsky, Morris Halle, and Victor Yngve.

The recommendation of the committee was unanimously favorable, and shortly afterward the Graduate Policy Committee met and accepted the report. The following motion was then submitted to and accepted by the meeting of the Faculty:

That the Faculty recommend to the Corporation that it authorize the Department of Modern Languages to recommend the award of the Degree of Doctor of Philosophy in the field of Linguistics.

The MIT Corporation approved the motion in December 1960.

5. The Degree-Granting Program

5.1. The Flavor of the Program

In September 1961 the first officially registered graduate students in linguistics entered MIT. The program they became part of had, it is apparent, a distinctive development. Indeed, it is because the years before its official inception were so critical to the ultimate form and shape of the program that they have been examined so closely here.

The linguistics program developed as a natural outgrowth of the research activities of the MIT linguists, and an attempt is still made to maintain a close tie between teaching and research. Graduate students have always been expected to play an active role in research, that is, to work on real problems, from an early point. The academic program, therefore, was organized so as to encourage students to participate in the ongoing research of the group. This fact should be borne in mind as the different facets of the MIT program are surveyed, for without it the picture is fragmentary. The course descriptions found in the MIT catalog, for example, are only partially revealing. They do not show that the advanced courses particularly, but to some extent the introductory courses as well, take up whatever relevant study the instructor is engaged in at the moment, and that the subject matter of most of the published work to come out of MIT appeared first in one class or another and was shaped by the resulting discussion and criticism. As one of the effects of this classroom approach, there was and still is an unusual amount of “sitting in” on courses by both faculty and students.

A related factor was the absence, during the early years of the program, of an extensive body of immediately relevant literature. Available to the students of the early sixties at MIT were the classical studies and an impressive array of work out of the structuralist school, but very little that dealt directly with transformational generative grammar. The field was simply too new.

Halle and Chomsky deliberately sought to avoid traditionally run courses in which large numbers of students were lectured to on the known works in the field. There was a time, however, when such courses would not have been possible anyway, for there were too few works – known or unknown. Gradually, of course, this changed, but for many years the motto at MIT still remained “If it can be read, it needn’t be taught.” To a certain extent this holds today.

Beginning students, especially those with a traditional academic background, often found the lack of something concrete to study disconcerting. The advice was simply to hang around a lot and let all the talk seep in; eventually it would start to come together. Sometimes this did not happen; more often it did.

Although no particular background is required of program applicants, eligible students in mathematics or science have generally been preferred over those in humanities or social science, all other things being equal. Students who are uncomfortable with scientific constructs and methods may be uncomfortable at MIT, and it is generally more difficult for them to develop a working familiarity with the theory. This is not to say that there is no place for non-science majors at MIT. On the contrary, many are accepted and many do extremely well. Indeed, of the MIT alumni presently on the linguistics faculty, not one had a background in science.

At the beginning the special nature of the MIT program entailed a certain problem for its survival, namely, the finding of qualified staff. A heavy responsibility was thus laid on Halle and Chomsky, and then Klima, in the first few years for the group had to train its own personnel. Klima and Matthews, as we have noted, spent several years at MIT before the program officially began. Paul Postal, with a Ph.D. from the Yale Department of Anthropology, was a research associate for two years before joining the linguistics faculty in 1963. And both Jerrold Katz and Jerry Fodor also put in their RLE years before becoming Assistant Professors on the humanities staff in 1963. Later, Kenneth Hale in 1966 was an exception. But the current faculty includes, in addition to Halle, Chomsky, and Hale, four graduates of the MIT program – Paul Kiparsky, who entered MIT in 1962, and John Ross, James Harris, and David Perlmutter, who became graduate students in 1964.

A further characteristic of study at MIT has been the availability of the faculty for discussion and guidance. It is expected that students will begin serious independent research in the second year of the program, and it is also expected that they will work closely with one or another of the staff as their research progresses. Indeed, it is hoped that they will feel free to discuss any point with any of the instructors at any time: this is necessary if the hang-around-and-let-it-seep-in approach is to work. As the field has grown and pressures on staff members have increased, this approach has been somewhat modified and the easy and open faculty availability of the early days has undergone a certain constriction; nevertheless, there is a constant attempt to preserve as much of it as possible.

It was always understood that to maintain the research emphasis, the informal classroom approach, and the give and take necessary to both, it was imperative to restrict the size of the group. Seven students were accepted the first year, ten the next, ten the year after, and so it went. Since the program is designed to be completed in four years, the entire enrollment of full-time students rarely exceeded thirty-five or so. This number was augmented by post-Doctoral scholars in the program, numbering five or six a year. Today, despite the ascendancy of the MIT school and the expansion of graduate linguistic study in general, limiting the size of the group is still considered necessary to maintain the quality and flavor of the program. Thus, although applicants now total about eighty each year, the number of admissions remains at ten or eleven.

The current program has undergone changes since the early sixties, but by and large they have been changes of degree rather than substance. For example, a substantial literature now exists, and so does a widespread interest in linguistics, even on the undergraduate level. Thus students often arrive at MIT knowing more about transformational generative grammar than advanced theorists knew not so many years ago. At the same time, the enthusiasm that accompanies genuine discovery has passed into the questioning and criticism that arise when a theory has gained wide acceptance and is being pushed to its limits. As a result, although students are better prepared in some ways to deal with what they find at MIT, they are still not offered a Body of Truth to learn.

There are now some classes, more things to read, more people to talk to. But there is probably also more pressure and, as a result of alternative ways of grappling with the theory, more dissension. There are more introductory courses and they are a good deal more introductory than they ever were, but the opportunities for casual faculty-student contact have narrowed. On balance, it could probably be said that certain focuses have shifted and certain extremes have been blunted, but the particular flavor of the MIT linguistics program appears in general to persist.

No survey of this program would be complete without mention of the special role Morris Halle has played in its development. Others have given significantly of their time and their efforts, as well as their brainpower. But it is generally agreed that Halle, intellectually, spiritually, and practically, has been the strongest single force in building the group and in guiding and guarding its growth.

5.2. Official Aspects of the Program

In order to be accepted as a full-time student in the linguistics program, it is necessary to satisfy Institute requirements for admission to the Graduate School and to have done well in previous academic work. However, beyond competence (i.e., three years of formal study) in two languages in addition to English, there are no explicit prerequisites in terms of courses or fields of specialty. Rather, applicants should demonstrate evidence of their ability to effectively engage in serious study of complex subject matter. This may take the form of academic work of very high quality, original research, challenging work experience, in-depth mastery of a language or language group, and so on.

Once a student has been accepted in the linguistics program, he is normally required to take a “core” of linguistics courses consisting of the following (the courses will be described in a later section):

• INTRODUCTION TO LINGUISTICS, I: SYNTAX
• INTRODUCTION TO LINGUISTICS, II: PHONOLOGY
• INTRODUCTION TO LINGUISTICS, III: THEORY OF GRAMMAR
• MATHEMATICAL BACKGROUNDS FOR COMMUNICATION SCIENCES
• LINGUISTIC STRUCTURE
• LINGUISTIC CHANGE
• ADVANCED PHONOLOGY
• SURVEY OF GENERAL LINGUISTICS
• TOPICS IN THE GRAMMAR OF A NON-INDO-EUROPEAN LANGUAGE

A full-time student is expected to take the equivalent of four subjects for credit each semester.

The degree candidate must also complete a program of studies in a minor field, the choice of which is made by the student, subject to the approval of the department. The minor program ordinarily involves three one-semester graduate subjects, selected so as to form a meaningful whole. Likely fields for the completion of minor studies include electrical engineering, mathematics, philosophy, psychology, anthropology, literary theory, and advanced work in some group of languages. Work done before entering MIT is sometimes accepted as satisfying all or part of the minor requirement. Students may also take advantage of the cross-registration agreement with Harvard to complete their minor.

In the middle of the second year of full-time study, students normally take a general examination given by the linguistics section. Originally this was a written examination testing the following areas – general linguistics (problems of grammatical and phonemic structure), linguistic theories and methods, practical linguistic analysis, the relation between linguistics and adjacent fields, and questions on the student’s special field. Later the last two topics were changed to historical linguistics and the structure of English. Still later (1970), the examination was replaced by the required preparation of two original in-depth studies, one syntactic and the other phonological. A board of three faculty members then administers an oral examination on the papers submitted by the student and related topics. Passage or failure is determined by the linguistics faculty, and revision of one or both of the papers and/or re-examination may be requested.

Finally, there is the research leading to a dissertation project, the completion of the dissertation, and the passing of an oral thesis defense.

From time to time, at the discretion of the department, special students are admitted who wish to study in the program without being degree candidates. Also admitted are post-Doctoral fellows who demonstrate a serious interest in the study of linguistics as undertaken at MIT. Such appointments are normally made for a period of one year.

5.3. Substance of the Program

1961-1962

Although the original proposal by the linguistics group called for the admission of from three to five students each year at first, the program actually started with seven regular students, one of whom later left, and three special or part-time students. The regular students included a psychologist, a chemist, a philosopher, an electrical engineer, and several mathematicians.^[4] All are at present actively involved at various institutions.

The linguistics section of the newly formalized Department of Modern Languages and Linguistics (MIT course of study number 23) began with a staff of six: Jakobson, Halle, Chomsky, Klima, Matthews, and Yngve (who by then was officially affiliated with the Department of Electrical Engineering but maintained his contact with the linguistics group). Furthermore, a significant shift was taking place: both interest and funding were now settling on the linguistics program itself, and researchers who were becoming increasingly concerned with linguistics in its theoretical aspects no longer had to operate through machine translation or other projects. It now became possible for the linguistics group directly to support the work of interested scholars such as S. Jay Keyser, Carlotta Smith, and a good many others.

Eleven linguistics courses were offered in the first year of the program, five of which (marked with asterisks) were new:

*23.701. CRUCIAL PROBLEMS IN LINGUISTICS. Word: form and meaning, grammatical and lexical analysis. Word classes and morphological categories. … Jakobson

23.71. SPECIAL PROBLEMS IN LANGUAGE AND LITERATURE.

23.72. SPECIAL PROBLEMS IN LANGUAGE AND LITERATURE. Advanced work in literature or linguistics for unusually well-prepared students. Offered by members of staff in their special fields.

*23.731. STRUCTURE OF ENGLISH. Syntax: detailed study of the grammatical structure of sentences in contemporary English. Special reference to formal correlates of traditional grammatical statements about English syntax, such as those by Jespersen, Poutsma, and Curme. Consideration of current descriptions of English syntax, such as those by Harris, Hill, and Fries. Klima

*23.732. STRUCTURE OF ENGLISH. Morphology and phonology. Continuation of 23.731. The interrelation of syntactic, morphological, and phonological phenomena in the English language. The phonological system and the structure of morphemes. Special reference to current interpretations by leading structuralists. The dialectology and historical phonology of English. Klima

23.751 LANGUAGE AND SOCIETY. Introduction to the scientific study of human language. Outline of syntactic theory; brief survey of English syntax. Elementary data on production and physical nature of speech. The phoneme and distinctive features. Phonemic systems. Morphology: review of some morphological devices. Languages of the world, their historical relationships and structural classification. Speech and writing. Language, culture, and cognition. Halle

23.761. PROBLEMS OF PHONOLOGY. Aims and principles of a scientific description of the structure of a spoken language. The mechanism and the acoustics of speech. The phonemic system of a language. The pair test. Distinctive features as a descriptive framework. The phoneme. The interrelation between phonology and other linguistic levels. Some problems of historical phonology. Practical exercises in preparing a phonemic analysis of a language with the help of a native informant. Halle

*23.781. SURVEY OF GENERAL LINGUISTICS. Critical study of a series of major works on the nature of language. Readings in linguistics, philosophy, and psychology. Chomsky

23.782. LINGUISTIC STRUCTURE. Problems and methods of the descriptive study of language. Study of form in language: definition and determination of linguistic elements; mathematical models for linguistic behavior; construction of an objective and abstract theory of formal linguistic structure; systematic application of this theory to the description of English structure. Study of meaning in language: relation of linguistic form and meaning; various approaches to a theory of meaning; contemporary philosophy of language and the influence of symbolic logic; extensional and intensional structure. Comparison of the logic of natural languages with that of special purpose language systems (i.e., mathematics, science). Chomsky

23.791 LANGUAGE, SYMBOLIC PROCESSES, AND COMPUTERS. Symbols, their storage and manipulation: formal systems of symbols, natural languages, computer programming languages, codes, notation systems of mathematics, logic, physics, etc.; relation of symbols and symbol systems to the physical means of representation and manipulation; relation of symbols and symbol systems to the subject matter represented and its systematization; symbolic processes, induction, deduction, reasoning; translation and translatability. Yngve

*23.729. MECHANICAL TRANSLATION AND LANGUAGE PROCESSING. A survey of research in the area of language processing by machine. Successes and inadequacies of current techniques. Remaining obstacles to adequate mechanical translation. Morphology; syntax; lexicography; semantics; automatic analysis and synthesis of sentences. Structure of English, German, and other languages. Yngve

These courses were, as we have noted, supplemented by a number of related courses in other departments. One change that took place was in the graduate course offered by Electrical Engineering and taught by Halle and Rosenblith since 1953. It became an undergraduate course handled by Halle and Kenneth N. Stevens, but it was still available to interested linguistics students:

6.36 SPEECH COMMUNICATION. For well-qualified students with a serious interest in speech communication. Survey of structural properties of natural languages with special emphasis on the sound pattern. Acoustical theory of speech production; anatomical considerations; the vocal tract as a generator of acoustical signals; acoustical correlates of vocal-tract configurations, both stationary and dynamic. Perception of speech: the auditory capabilities of man; the role of set in perception; analysis by synthesis. Mechanical recognition and generation of speech. Speech transmission and compression: the concept of intelligibility; effects of distortion on intelligibility. Stevens, Halle

It should be noted that the official beginning of the graduate linguistics program marked the official end of undergraduate linguistics courses at MIT for almost a decade. Although qualified undergraduates could and did enroll in several of the courses, there was not to be another specifically designed undergraduate introductory course until 1970.

Reviewing the first-year courses, it is immediately evident that the linguistics offerings had been expanded but not altered in any radical way. The aim and the scope of the teaching program remained entirely consistent with those of the research program. The mathematical-computer and machine translation aspects of the field were covered, as was the other applied form, speech analysis; but there was a greater number of courses relating to theoretical aspects. English was the only language covered in any comprehensive way, and it was treated in fairly great depth and breadth. In short, as the original proposal had stated, there was no attempt to duplicate linguistics programs elsewhere.

It might be noted that although this was undoubtedly deliberate, it might not have been possible without the presence of Harvard some blocks away, together with a long-standing exchange program that permitted full-time graduate students from one university to take as much as half their work at the other.^[5]

The linguistics group remained subsumed under the retitled Department of Modern Languages and Linguistics, which in 1969 was renamed the Department of Foreign Literatures and Linguistics. Its ties to the Research Laboratory of Electronics were also maintained, and in 1961 there were a number of related projects being carried out there.

1962-1963

Ten new graduate students were admitted in the fall of 1962, the second year of the program’s existence, bringing to sixteen the total enrolled for a degree. This time there were linguists in the group as well as mathematicians, a computer specialist, a biologist, and a philosopher-psychologist. One failed to complete the program and a second went into medicine after finishing; the rest remain active in the field.

There were a number of staff additions and new courses that year. The noted Indo-Europeanist Jerzy Kuryłowicz of the University of Krakow accepted an appointment as Visiting Professor. G. Hubert Matthews, who had been teaching German, initiated a course in TYPOLOGY OF GRAMMARS:

23.783 TYPOLOGY OF GRAMMARS. Study of the grammars of languages belonging to divergent families with a view to reaching a sharper characterization of universal features of grammatical structure, with the majority of examples to be drawn from non-Indo-European languages.

Barbara Hall (Partee), then a second-year graduate student in the program, began to teach a new course designed to convey to non-science students the mathematical concepts and tools needed to work with linguistics at MIT:

23.771. MATHEMATICAL BACKGROUNDS FOR COMMUNICATION SCIENCES. Fundamentals of discrete mathematics for non-mathematicians. Introduction to various topics in set theory, probability and information theory, logic, foundations, theory of automata and computability.

There were four additional new courses:

23.702. SEMINAR IN LINGUISTICS. Special problems in linguistics. Permission of instructor required.

23.734. STRUCTURE OF RUSSIAN. Survey of the syntax, morphology, and phonology of contemporary literary Russian, with emphasis on the application of modern theory to classical problems of Slavic linguistics. Consideration of the relevant dialectological and historical evidence. (Adequate knowledge of Russian required.) Halle, Klima

23.772. MATHEMATICAL MODELS IN LINGUISTICS. Study of mathematical models that relate to linguistic competence and performance. Formal properties of grammatical systems. Grammars of natural and artificial languages. Models of the speaker and hearer that incorporate representations of his knowledge of the language. Probabilistic models for the user. Stress on inter-connections of theory of grammar and theory of automata, and implications for the study of human cognitive processes wherever possible. Chomsky

23.785. LINGUISTIC CHANGE. Review of instances of change in phonology, morphology, and syntax. Examples drawn primarily from histories of Indo-European languages with special emphasis on English and Russian. An attempt to develop a general theory of linguistic change. Halle, Klima

Chomsky took over the teaching of the first semester of STRUCTURE OF ENGLISH to free Klima to participate in the above new courses. The description of Jakobson’s course was modified:

23.701. CRUCIAL PROBLEMS IN LINGUISTICS. Sound and meaning: a study of the various types of relationships between phonetic and semantic aspects of language.

Thus, in 1962-1963 the full roster of linguistics courses was as follows:

• CRUCIAL PROBLEMS IN LINGUISTICS, Jakobson
• SEMINAR IN LINGUISTICS, Staff
• SPECIAL PROBLEMS IN LANGUAGE AND LITERATURE, Staff
• STRUCTURE OF ENGLISH I: SYNTAX, Chomsky
• STRUCTURE OF ENGLISH II: MORPHOLOGY AND PHONOLOGY, Klima
• STRUCTURE OF RUSSIAN, Halle, Klima
• LANGUAGE AND SOCIETY (Introduction), Halle
• LINGUISTIC STRUCTURE, Chomsky
• PROBLEMS OF PHONOLOGY, Halle
• MATHEMATICAL BACKGROUNDS FOR COMMUNICATION SCIENCES, Hall
• MATHEMATICAL MODELS IN LINGUISTICS, Chomsky
• SURVEY OF GENERAL LINGUISTICS, Chomsky
• TYPOLOGY OF GRAMMARS, Matthews
• LINGUISTIC CHANGE, Halle, Klima
• LANGUAGE, SYMBOLIC PROCESSES, AND COMPUTERS, Yngve
• MECHANICAL TRANSLATION AND LANGUAGE PROCESSING, Yngve

(The last two courses listed were offered jointly by Modern Languages and Electrical Engineering.)

An important occurrence for the MIT linguistics group in 1962 was the meeting there of the Ninth International Congress of Linguists. (Chomsky’s Current Issues in Linguistic Theory (1964) was presented at that Congress in an early form.)

1963-1964

The year 1963 again saw the admission of ten new students, with what was becoming the typical diversity of backgrounds. Accordingly there were further additions to the staff and to the roster of courses. Matthews, formerly a research associate, became an Associate Professor, and Paul Postal, who had been associated with RLE, became an Assistant Professor in the department. Moreover, Jerrold Katz and Jerry Fodor, who had also been at RLE, joined the staff of the Humanities Department.

The addition of Katz and Fodor to the MIT teaching staff made possible two undergraduate courses which, though given by the Humanities Department, were directly relevant to linguistics:

21.87. PHILOSOPHY OF LANGUAGE. Current philosophical approaches to questions about language in general and meaning in particular. Study of authors representative of the Positivist movement and of Ordinary Language Philosophy. Particular attention to such topics as the “performative” analysis of speech and the implications of the philosophy of language for philosophical psychology. Fodor

21.88. SEMANTICS. Characterization of the form of a semantic theory of a natural language. Basic consideration of (1) phenomena falling within the domain of a semantic theory and (2) descriptive goals of a semantic theory, the kind of theory employed in pursuit of these goals, and the empirical and methodological requirements upon an adequate semantic theory. Special attention to the role of a semantic theory in the solution of philosophical problems. (Additional requirement: thorough knowledge of material in Chomsky’s Syntactic Structures.) Fodor, Katz

Added directly to the linguistics offerings were the following two courses:

23.703. LINGUISTICS AND POETICS. Prosodic patterns in word-and-sentence phonology; their utilization in metrics; prosodic and metrical typology. Metrical role of inherent distinctive features. Morphological and syntactic constituents of verse. Jakobson

23.736. STRUCTURE OF GERMAN. Survey of the syntax, morphology, and phonology of contemporary literary German, with emphasis on the application of modern theory to classical problems of Germanic linguistics. Consideration of the relevant historical evidence. (Adequate knowledge of German required.) Klima

And the descriptions of STRUCTURE OF ENGLISH and LINGUISTIC STRUCTURE were reworded as follows:

23.731. STRUCTURE OF ENGLISH. Introduction to the grammatical devices of English. The interrelation of syntactic, morphological, and phonological phenomena in the English language. The phonological system and the structure of phonemes. Special reference to current interpretations by leading structuralists. Chomsky

23.732. STRUCTURE OF ENGLISH. Detailed study of the grammatical structure of sentences in contemporary English. Special reference to formal correlates of traditional grammatical statements about English syntax, such as those by Jespersen, Poutsma, and Curme. Klima

23.752. LINGUISTIC STRUCTURE. The empirical requirements for a theory of linguistic structure that will encompass and account for linguistic competence and the ability to use and acquire it. Study of aspects of English syntax, semantics, and phonology insofar as they furnish critical tests for such a theory. Systematic presentation of a theory of linguistic structure, its abstract properties and range of empirical adequacy. Explanatory models in linguistics; their implications concerning the nature of cognitive processes. Chomsky

It is clear that at the close of the academic year 1963-1964, with twenty-six students, six of whom had completed three years and were deeply into their dissertations, with a growing number of special students and post-doctoral scholars, with a staff of six, plus three members of other departments, offering a roster of more than twenty courses, and with RLE research activities continuing, MIT had a full-scale linguistics program in operation.

In applying for an NIH training grant in 1964, the group described the ultimate questions toward which their research was directed in the following terms:^[6]

What do the artifacts of communication and in particular language tell us about the natural capabilities of man?; or what inferences can be made concerning the intrinsic capabilities of the organism that masters and utilizes the systems of expression and communication? …

The research strategy for seeking answers to these questions was then reviewed:

For each language we attempt to construct abstractly a device (called technically a generative grammar of this language), which specifies the structural information about sentences that is available in principle to the speakers of the language. … A generative grammar provides an explicit account of one primary aspect of the state of the perceiving organism… Thus language provides a unique opportunity for precise formulation and fruitful empirical study of some of the central questions of perceptual psychology.

…our training and research programs have a strongly interdisciplinary flavor, and consequently most of the faculty have close contacts with members of other departments and are frequently involved in joint research projects with members of other departments at MIT and Harvard. Most of our students have similar interests, or acquire them during their graduate training, and we attempt to encourage their participation in such work…as fully as possible.

The major research interests of the staff were sketched as follows:

Theory of Grammar (Chomsky, Jakobson, Katz, Postal)
The primary goal of linguistic investigation is to construct grammars for particular languages and, on the basis of these, to determine the general properties of any human language. A grammar represents in a precise way the ability of the speaker to understand an arbitrary sentence of his language and to produce sentences (in general, new sentences) on the appropriate occasion. Thus a grammar pairs phonetically represented signals with semantic interpretations; it generates an infinite set of such pairs. … The theory of grammar is concerned with the overall structure of grammar, the interrelations of these subdomains (syntax, phonology, and semantics), the question of evaluating grammars and determining their empirical adequacy. The problem of evaluating grammars, which is central to the theory of grammar, is essentially a fundamental aspect of the problem of accounting for the ability of the child to learn a language. …

Formal Languages (Chomsky, Halle, Matthews)
When the properties of a generative grammar are formulated with sufficient precision, they can be studied abstractly, apart from any question of empirical interpretation. This purely mathematical study of grammars…has been quite fruitful. For one thing, it has been shown that some simple types of grammars relate very closely to systems that have an independent, automata-theoretic interest, as well as to familiar concepts of logic and even classical mathematics that on the surface appear quite independent. Furthermore, properties of these formal systems…have on several occasions suggested empirical consequences concerning actual languages. Ultimately, one hopes to be able to study the mathematical structure of generative systems of a type that are adequate for natural language. …

Syntax (Chomsky, Klima, Matthews, Postal)
The syntactic component of a grammar generates an infinite set of structures which support phonological and semantic interpretation. … Our work in syntax divides naturally into two subparts, theoretical and applied. In theoretical syntax we are concerned to formulate the general properties of this recursive system…in as explicit and narrow a fashion as possible, on the basis of the data concerning syntactic structures of given languages which is provided by the work in applied syntax.

Phonology (Chomsky, Halle, Jakobson, Matthews, Postal)
The phonological component of a grammar is an input-output system which converts surface structures, provided by the syntactic component, into phonetic representations. The problems of phonology are to determine the character of the rules that constitute this system and the manner of their functioning, and to determine the elementary units (the universal system of phonetic distinctive features) and general laws regarding these units that delimit the set of possible phonetic representations. This work is naturally heavily dependent on detailed phonological and phonetic investigation of particular languages. A particularly important aspect of current work concerns the syntactic determinants of phonetic structure. …

Semantics (Fodor, Katz, Postal)
The semantic component of a grammar assigns semantic interpretations to the deep structures provided by the syntactic component. Research into the structure of the semantic component is quite recent, but highly promising. … Thus, the theory of semantics is concerned to discover a general system of semantic representation…and to determine the nature of the “projection rules” that assign semantic interpretations to larger units on the basis of the semantic interpretations of their parts. …

Historical Linguistics (Halle, Jakobson, Klima, Matthews)
Historical linguistics is the name given to the study of the changes that grammars of languages undergo in the course of time. Since grammars are ordered sets of rules, historical linguistics is primarily interested in the changes that affect these rules. It seems that the most important mechanism of linguistic change is the addition of a single rule to the grammar. … As different rules are added to different dialects the dialects will differ more and more. … A surprising result of our studies appears to be that…radical differences are not found; instead, languages conserve their grammars intact over many generations…and change only slowly as the number of rules added increases. Instances where radical differences between the grammars of two generations can be established are rare. …

Language Perception and Acquisition (Fodor, Menyuk^[7])

The linguistic theories being studied in this program strongly suggest that the process of language learning involves a theory-building mechanism available to human infants and depends in only a minor manner on repetitious and rote reinforcement. …

Models of Visual Perception (Eden,^[8] Chomsky, Fodor)
A consistent interpretation of the orientation and depth judgments in optical illusions, as well as a description of human performance in the psychophysics of space, can be made by constructing an abstract (generative) theory for the geometry of the visual perceptual world. A visual impression is “recognized” when the postulates adopted by the observer generate a precept well-formed according to the rules of the theory.

Handwriting (Eden, Halle)
In a manner analogous to the techniques of phonology, cursive writing in a given script can be characterized by a finite sequence of discrete symbols (strokes). These strokes may be thought to represent sub-routines or algorithms for generating an appropriate segment of the written text. Such descriptions have been successfully applied to several styles of writing. Moreover, the technique has been extended so as to program a device to “read” handwritten English words. …

1964-1965

Linguistics applicants in 1964 totaled over forty; eleven were selected to become part of the program. Permanent-staff expansion came to a temporary halt that year. Aside from the addition of two Visiting Professors – Jerzy Kuryłowicz (for whom this marked a return) and Peter Colaclides – the only change was the replacement of William Locke as Chairman of the Department of Modern Languages and Linguistics by William F. Bottiglia, who taught French and Italian literature in the department. Chomsky was on leave, working at the Center for Cognitive Studies at Harvard, and Postal took over the teaching of STRUCTURE OF ENGLISH I, LINGUISTIC STRUCTURE, and SURVEY I.

The subject offerings, on the other hand, did undergo a number of changes. First, the list of courses was increased by three:

23.754 FIELD METHODS. Survey of means for eliciting linguistically relevant data about a language from a native speaker; for obtaining correct phonetic representations, the basic phonological system, and basic grammatical structure. Suggestions for dealing with informants, elicitation, and data organization. Matthews, Postal

23.756 INTRODUCTION TO INDO-EUROPEAN ACCENTUATION. Discussion of the accentual systems of the major Indo-European language groups and inquiry into their historical evolution. Kuryłowicz

APPENDIX

Doctoral Dissertations Accepted by the MIT Department to January 1974

1965

• Foley, James, Spanish Morphology.
• Fraser, Bruce, An Examination of the Verb-Particle Construction in English.
• Gruber, Jeffrey, Studies in Lexical Relations.
• Hall, Barbara, Subject and Object in Modern English.
• Kiparsky, R. Paul V., Phonological Change.
• Kuroda, Sige-Yuki, Generative Grammatical Studies in the Japanese Language.
• Langendoen, Terence, Modern British Linguistics.
• Lightner, Theodore, Segmental Phonology of Modern Standard Russian.
• McCawley, James, Accentual System of Standard Japanese.
• Petrick, Stanley, A Recognition Procedure for Transformational Grammars.
• Rosenbaum, Peter, Grammar of English Predicate Complement Constructions.
• Schane, Sanford, Phonological and Morphological Structure of French.
• Zwicky, Arnold, Topics in Sanskrit Phonology.

1966

• Lieberman, Philip, Intonation, Perception, and Language.

1967

• Bever, Thomas, Leonard Bloomfield and the Phonology of the Menomini Language
• Chapin, Paul, On the Syntax of Word-Derivation in English.
• Fidelholtz, James, Micmac Morphophonemics.
• Harris, James W., Spanish Phonology.
• Ross, John R., Constraints on Variables in Syntax

1968

• Bedell, George, Kokugaku Grammatical Theory.
• Dougherty, Ray C., A Transformational Grammar of Coordinate Conjoined Structures.
• Naro, Anthony, The History of Portuguese Passives and Impersonals.
• Perlmutter, David M., Deep and Surface Constraints in Syntax.

1969

• Anderson, Stephen, West Scandinavian Vowel Systems and the Ordering of Phonological Rules.
• Jackendoff, Ray S., Some Rules of Semantic Interpretation for English.
• Kayne, Richard S., The Transformational Cycle in French Syntax.
• Stanley, Richard, The phonology of the Navaho Verb.
• Woo, Nancy, Prosody and Phonology.

1970

• Akmajian, Adrian, Aspects of the Grammar of Focus in English.
• Brame, Michael, Arabic Phonology: Implications for Phonological Theory and Historical Semitic.
• Dell, François, Les Règles Phonologiques Tardives et La Morphologie Derivationelle du Français (Topics in French Phonology and Derivational Morphology).
• Emonds, Joseph, Root and Structure-Preserving Transformations.
• Fodor, Janet Dean, Linguistic Description of Opaque Contexts.
• Geis, Michael, Adverbial Subordinate Clauses in English.
• Kimball, John, Categories of Meaning.

1971

• Caplan, David, Probe Tests and Sentence Perception.
• Culicover, Peter, Syntactic and Semantic Investigations.
• Fischer, Susan D., The Acquisition of Verb-Particle and Dative Constructions.
• Helke, Michael, The Grammar of English Reflexives.
• Kimball, John, Categories of Meaning.
• Nakau, Minoru, Sentential Complementation in Japanese.
• Williams, George, Networks of Anaphora: An Essay in the Syntax of Pronominalization.

1972

• Bresnan, Joan W., Theory of Complementation in English Syntax.
• De Rijk, Rudolphus, Studies in Basque Syntax: Relative Clauses.
• Howard, Irwin, A Directional Theory of Rule Application in Phonology.
• Jenkins, Lyle, Modality in English Syntax.
• Lasnik, Howard B., Analyses of Negation in English.
• Pope, Emily A. Norwood, Questions and Answers in English.
• Selkirk, Elisabeth O., The Phrase Phonology of English and French.
• Wasow, Thomas A., Anaphoric Relations in English.

1973-1974 (Some titles approximate)

• Bowers, John S., Grammatical Relations.
• Erteschik, Naomi, On the Nature of Island Constraints.
• Higgins, F.R., The Pseudo-Cleft Construction in English.
• Kornfeld, The Perception of Syntactic Boundaries.
• Leben, William R., Supra-Segmental Phonology.
• Maling, Joan M., The Theory of Classical Arabic Metrics.
• Permesly, Sylvia Schwartz, Some Aspects of Presupposition in Generative Grammar.
• Vergnaud, Jean-Roger, Syntax of French Relative Clauses.
• Wilson, Dierdre, On Presuppositions.

NOTES

References

• Bodmer, F.
  
  • 1944  The Loom of Language, ed. by L. Hogben (New York: Norton).
• Chomsky, N.
  
  • 1953  “Systems of Syntactic Analysis,” Journal of Symbolic Logic.
  • 1955  The Logical Structure of Linguistic Theory, mimeographed, MIT
  • 1956  “Three Models for the Description of Language,” I.R.E. Transactions on Information Theory, IT-2; reprinted in Readings in Mathematical Psychology, II, ed. by R. D. Luce, R. Bush, and E. Galanter (1965) (New York: Wiley).
  • 1957  Syntactic Structures (The Hague: Mouton).
  • 1959a “On Certain Formal Properties of Grammars,” Information and Control 2; reprinted in Readings in Mathematical Psychology, II, ed. by D. Luce, R. Bush, and E. Galanter (1965) (New York: Wiley).
  • 1959b Review of Verbal Behavior by Skinner, Language 35; reprinted in Fodor and Katz (1964).
  • 1961  “On the Notion ‘Rule of Grammar’,” Proceedings of the Twelfth Symposium in Applied Mathematics, ed. by R. Jakobson (Providence, R.I.: American Mathematical Society); reprinted in Fodor and Katz (1964).
  • 1962  “A Transformational Approach to Syntax,” Proceedings of the Third Texas Conference on Problems of Linguistic Analysis in English, 1958, ed. by A. A. Hill (Austin: The University of Texas); reprinted in Fodor and Katz (1964).
  • 1964  Current Issues in Linguistic Theory (The Hague: Mouton); reprinted, in a slightly earlier version, in Fodor and Katz (1964); originally presented, in a somewhat different version, as a paper in the session “The Logical Basis of Linguistic Theory” and appears under this title in Proceedings of the Ninth International Congress of Linguists, ed. by H. Lunt (1964) (The Hague: Mouton).
  • 1965  Aspects of the Theory of Syntax (Cambridge, Mass.: MIT Press).
  • 1966  Cartesian Linguistics (Cambridge, Mass.: MIT Press).
• Chomsky, N. and M. Halle
  • 1968  The Sound Pattern of English (New York: Harper & Row).
• Chomsky, N., M. Halle, and F. Lukoff
  • 1956  “On Accent and Juncture in English,” For Roman Jakobson, ed. by M. Halle, H. Lunt, and H. MacLean (The Hague: Mouton).
• Fodor, J. A. and J. J. Katz, eds.
  • 1964  The Structure of Language: Readings in the Philosophy of Language (Englewood Cliffs, N.J.: Prentice-Hall).
• Schatz, C.D.
  • 1954  “The Role of Context in the Perception of Stops,” Language 30.
• Skinner, B. F.
  • 1957  Verbal Behavior (New York: Appleton-Century-Crofts).
• Halle, M.
  • 1959  The Sound Pattern of Russian (The Hague: Mouton).
  • 1961  “On the Role of Simplicity in Linguistic Description,” Proceedings of the Twelfth Symposium in Applied Mathematics, ed. by R. Jakobson (Providence, R.I.: American Mathematical Society).
• Hughes, George W.
  • 1960  On the Recognition of Speech by Machine, MIT Ph.D. dissertation.
• Jakobson, R., C.G.M. Fant, and M. Halle
  • 1952  Preliminaries to Speech Analysis (Cambridge, Mass.: MIT Press).
• Katz, J. J.
  • 1966  The Philosophy of Language (New York: Harper & Row).
• Katz, J. J. and P. Postal
  • 1964  An Integrated Theory of Linguistic Descriptions (Cambridge, Mass.: MIT Press).
• Lees, R.
  • 1960  The Grammar of English Nominalizations [Indiana University Research Center in Anthropology, Folklore, and Linguistics, 12] (Bloomington: Indiana University Press).
• Locke, W. N. and Booth
  • 1955  Machine Translation of Languages (Cambridge, Mass.: MIT Press; New York: Wiley).
• Matthews, G. H.
  • 1964  Hidatsa Syntax (The Hague: Mouton).
• Postal, P.
  • 1964  Constituent Structure: A Study of Contemporary Models of Syntactic Description [Indiana University Research Center in Anthropology, Folklore, and Linguistics, 30] (Bloomington: Indiana University Press).