Noam Chomsky (4 page)

Any summary of the very private life of the Chomskys must inevitably note the many contradictions that arise from the clashes between the private and public spheres of their lives. To date it appears that they have handled it very well and been able to maintain the separation between the private and the public, notwithstanding the current effort to shed light on the former. As pointed out at the beginning of this chapter, the best we can learn from Chomsky’s private life is that it is based on a supreme belief in a humanistic education, as passed on from parents to children, as continued in a few good schools and universities, and above all on autodidactism.

Linguistics, or the study of language, is by some considered an odd science. The essential tool for the study of language is language itself. It is a paradox. However, this has not deterred mankind from investigating and speculating since time immemorial. Indeed it is probably one of the oldest sciences known to mankind and, as Chomsky maintains, it is a science like any other. Doing linguistics is, as Chomsky puts it with good humour, doing what any other natural scientist does: looking for the key under the lamppost, like a drunk, because that’s where the light is.
¹

One of the earliest known grammarians and searchers for a key was an Indian linguist named Panini, who, in the fifth century BC devised the rules for Sanskrit morphology. In ancient Greece the study of language became closely associated with philosophy: the paradox of ‘all Cretans are liars’ reverberates to this day. When classical Greek and Latin became the languages of scholarship throughout much of Europe, there came with it an intense preoccupation with the grammars of these two languages. The ‘grammar schools’ in the English education system survive to this day, albeit in much reduced form. Not surprisingly, practically all the terminology to do with grammar comes from study of the classics. Greek and Latin grammars were the baseline for all other languages. Grammars of English, French, German, Hebrew, Russian, Arabic, Chinese and of other literate languages became incorporated in a more general study of language then called philology. The history of these languages became of great interest and historical changes were investigated, such as the Great Vowel Shift in Germanic languages. The French Port-Royal grammarians of the seventeenth century instigated another aspect: that of the connection between language and thought. Following in part the French philosopher Descartes, they proposed that grammatical categories and structures were analogous to logical patterns of thought, hence mental constructs. Chomsky was to incorporate these ideas into his Cartesian Linguistics.
²

In the meantime, though, the philologist tradition with the emphasis on the history of languages gave way to the synchronic study of languages. While the nineteenth century was still mainly concerned with historical linguistics, the twentieth century began to focus on living languages as they were spoken. This meant all languages of the world, the majority of which had no literate traditions. The investigation of speech also meant a refocusing on phonetics and phonology, the study of the sound systems of a language. The theory of a ‘phoneme’ as the psychological representation of an actual speech sound (phonetic form) became widely accepted.
³
It means that a speaker recognizes a variety of closely related speech sounds as one significant sound that imparts meaning when forming a word.

The English words ‘mat’ and ‘mad’, for example, are distinguished only by the phonemes of/t/ and /d/ (it is a convention to place phonemes between slashes). However, the native English-language speaker/listener can accommodate quite a number of different phonetic realizations of the /t/, such as the plain [t] and the aspirated [t
^h
] (it is a convention to put phonetic sounds between square brackets). In Hindi, however, these two sounds are phonemes, /t/ and /t
^h
/, which impart different meaning in a word that has otherwise the exact same sounds. As such, different languages have different sets of phonemes, which allow for various phonetic realizations. The set of actual phonetic realizations in all languages of the world is captured in the set known as the International Phonetic Alphabet (
IPA
). Note that in American English the word ‘math’ as distinguished from ‘mat’ determines that there are two phonemes, namely the /t/ and the /th/. The German language, for example, has neither the phoneme /th/ nor any sound that in English is realized as /th/.

This idea that speakers have acquired a narrow set of phonemes from the much wider set of all possible speech sounds is in itself revolutionary, for it points to the further theory that an actual language might be just the narrow realization of a much wider (or deeper) system that applies to all languages – an idea also seized on by Chomsky, minus the proviso, however, that any such underlying systems should have a ‘psychological reality’. His position has always been that linguists are seeking real systems, systems that should have reality, just like other parts of biology. Since the domain can be regarded as part of psychology, that means they should have psychological reality, which is just reality in the domain of psychology. There is a huge literature claiming that something ‘more’ is needed to establish psychological reality, like information about processing. That is just mysticism, in Chomsky’s view. Of course, one wants theories of language (or chemistry, etc.) to be verified wherever they can be tested, but nothing more than that is involved. Information about, say, processing does not give some mystical insight into ‘reality’ that evidence from other psychological experiments fails to give (and informant work, with oneself or another native speaker, is just a kind of psychological experiment, which could be made as precise as one likes, if there is any need).

The birth of modern theoretical linguistics is generally attributed to the Swiss linguist Ferdinand de Saussure (1857–1913), who set the scene by distinguishing between
langue and parole
.
⁴
The former refers to the internal system and structure of language
per se
, while the latter refers to the use of language. Again this is quite a revolutionary concept inasmuch as previously it was commonly thought that language use (as communication) determines the structure of language. Indeed, the idea is still alive and well today and serves as an antidote to Chomskyan linguistics. Saussure made, as did Chomsky later on, the principled distinction on the grounds that all people seem to have the same language capacity, but that different people seem to make use of this language capacity in vastly different ways. According to this theory, the language capacity comes first and language use second. The latter has only a minor influence on the former. One should note though that there is a difference in how Saussure and Chomsky conceive of
langue:
for Saussure
langue
was a social concept (it is societies that have a
langue)
, while for Chomsky the corresponding concept is one of individual psychology.

Saussure also introduced the semantic dichotomy of the
signifiant
and the
signifié
. For example the English word ‘cat’ (as a
signifiant)
refers to a certain feline animal (the
signifié)
we know so well. There seems to be only an arbitrary connection between the form of the word (the sound or the written form) and what it stands for. A language unrelated to English may have a totally different sounding word for ‘cat’. While Chomsky always thought of this problem as uninteresting (or even trivial), it was the French philosophers of the modern era who made much of it. Indeed Chomsky points to Aristotle, who long ago made the observation that language involves sound–meaning connections, which are, of course, arbitrary.

Early twentieth-century linguistics, however, still proceeded along the lines of the Prague School, which focused on phonology. Roman Jakobson (1898–1983), one of the founders of that school,
⁵
was to become a friend of Chomsky, having first met him at Harvard University in 1951. Linguistics as the new kid on the science block took another twist in the meantime. On the one hand linguistics entered the domain of logical positivism via Bertrand Russell, who devised the principles of mathematics on the logic of language (or the language of logic), while on the other American linguistics had became heavily influenced by anthropology via Franz Boas (1858–1942), Edward Sapir (1884–1939) and Benjamin Whorf (1897–1941), to name the key players of the time. The Sapir-Whorf hypothesis that different languages generate different world views reverberates to this day. A kind of reverse history was repeated later when the French anthropologist Claude Lévi-Strauss
(b
. 1908), who was influenced by Jakobsonian linguistics in the 1950s and ‘60s, had a huge influence on European structural linguistics.
⁶
He emphasized ‘structure’ as the glue that holds together societies, for example as a kinship system. Language as such becomes a static structure that is comparable to the structure of a building. The task of linguistics was then to describe the various structures of the world’s languages.
⁷
An extension of this theory was brought about via psychology. It was assumed that language structures are learnt behaviour. The notable proponent of language as behaviour was the American B. F. Skinner (1904–1990).
⁸
Chomsky was to repudiate his theory famously in his review of Skinner’s work in 1959.
⁹

How did Chomsky become a linguist in the first place? As the story goes, more by coincidence than by design.
¹⁰
In the first instance one might point to his father, a noted Hebrew scholar. In retrospect Chomsky must have benefited from his father’s professional interest in language, even though Dr William Chomsky was more of a language teacher who also wrote a Hebrew grammar. Language teachers and language learners should not be confused with linguists, who study language
per se
. But Chomsky’s father also did scholarly work on medieval and historical Hebrew grammar, with which Noam was familiar as a child before he ever heard of linguistics. When he was only about twelve he read his father’s drafts of a study of a thirteenth-century Hebrew grammar, and later, while a Hebrew language teacher during his student days, no doubt used his father’s textbooks, such as
How to Teach Hebrew in the Elementary Grades
(1946). Growing up with English and Hebrew (the latter a second language), Chomsky also learned classical Arabic and basic French and German at university. None of these language skills, though, predetermined him to become a linguist.

Having drifted without a plan at the University of Pennsylvania, supporting his studies by teaching Hebrew at his father’s school, Chomsky nearly dropped out. The turning-point, as he acknowledged in many of his early books on linguistics, came in 1947 when he met Zellig Harris in a political circle (actually unconnected with the university). As luck would have it, Professor Harris had the distinction of establishing the first department of linguistics at any university in the
US
, but it was not this that won over Chomsky to the study of linguistics within their first couple of meetings. In the first instance Harris impressed Chomsky by his politics, which more or less matched his stance at the time. As such Chomsky also listened to Harris’s advice on his nearly stalled academic studies. He recommended courses in mathematics and philosophy, and mentioned in passing that he could also drop in on his lectures on linguistics. Chomsky responded with interest.

The group that met with Harris comprised a handful of graduate students. They did their academic work over a beer at the pub or at Harris’s apartment, all in an unstructured way, but that still earned the students the necessary credits to gain their degrees. This freedom to work and learn was then as rare as it is today, and many a student would have been unable to respond, having been conditioned otherwise by an earlier public school straitjacket. For Chomsky, of course, this was a most welcome continuation of his early unorthodox, Deweyite schooling.

In time Chomsky became immersed in Harris’s linguistics. As he read Harris’s 1947 draft
of Methods in Structural Linguistics
(1951) he became truly caught up in the field of linguistics (Chomsky is acknowledged in the book’s preface as having proofread the text).
¹¹
At that stage Chomsky took for granted that procedural analysis along Harris’s lines was the whole story. It was generally assumed at the time that the field was basically finished, apart from applying the methods to as many languages as possible. His courses with Harris were almost entirely an extension of the methods to longer discourses, since there was nothing more to say about the theory of sentences and their parts.

In 1948, when Chomsky was casting about for an honours’ thesis, Harris suggested that he work on Hebrew. Chomsky found an Israeli informant, followed the field methods procedures, then started applying Harris’s methods.

It was probably at this stage that the Chomskyan revolution started, for Chomsky started to question the methodology. He knew the answers to the questions, so why was he asking the informant? And the methods weren’t yielding what he knew to be true of the language. There was no way, for example, that these methods might yield the basic root-vowel pattern structure of Semitic. So Chomsky started on what he took to be a private hobby, influenced in part by what he knew about historical Semitic linguistics, and also what just seemed to make sense.

The result was the first (1949) version
of Morphophonemics of Modern Hebrew
, with a rudimentary generative syntax, and a detailed morphophonemics, with a carefully worked out evaluation procedure and an effort to show that the grammar was a ‘relative maximum’ in terms of that measure: that is, any reordering of rules would make it worse. The rule ordering was quite deep, about thirty or so. That was heresy in structural linguistics. From then on, Chomsky was working on his own. He never discussed his research. The only faculty member who looked at the thesis was Hoenigswald, out of a sense of responsibility. Harris had no interest in it. Still, Chomsky also continued to work along the lines of Harris’s methods, but by 1953, after years spent trying, and failing, to refine the methods so that they would work, he decided to forget the whole business and concentrate on generative grammar – his private hobby.