Noam Chomsky (7 page)

Since the sentence features two subordinate clauses, we have three base or deep structure components from which we can assemble the surface structure sentence by applying various transformations – including two active–passive transformations. Informally the three base structures are (where items in square brackets indicate grammatical structure items):

Note the ‘zero’ item [o] in (1), which indicates the missing (or elided) agent, that is ‘someone’ who fired the man, or in the [passive] transformation ‘was fired by “someone”‘. Such empty/dummy categories or trace elements became important features of deep structures. Note also that the [s’] items are the embedded clauses or sentences.

While this is only the briefest glimpse of a very complex theory to generate sentences – in any language for that matter – it suffices here to point out that a powerful tool of linguistic analysis was launched. Not only did it successfully generate sentences, but it also generated an enthusiastic following with intense debates about the often minute technicalities that determine a set of phrase structure rules. The followers, though, were still confined to a small circle of
MIT
workers, but slowly and surely the circle widened despite the best efforts of the establishment detractors to have them suppressed.

It was at this time in the 1960s that another project on which Chomsky had been working with his old friend and colleague Morris Halle came to fruition. They began to co-write a monumental work that was eventually published as
The Sound Pattern of English
(1968). Close to 500 pages long, it spells out just about everything there is to know about the sounds of English. Among the book reviews on
Amazon.com
is this one:

This was posted in 2001, some 33 years after the first publication. Note also the reader’s lament that Chomsky was apparently lost to phonology after the event. That is not strictly true, as Chomsky and Halle to this day have always paid special attention to the phonological component of grammar. Indeed Chomsky in particular has always reserved a special interface for the phonological form (
PF
) of a sentence. Even more interestingly, he has always postulated that the conversion to
PF
is the final step in the generation of a sentence. This seems to go against the traditional idea that phonology (and phonetics) are at the beginning, followed by morphology, syntax and semantics. However, it seems common sense to maintain that, while all the rules of morphology, syntax and semantics are generated in the language capacity, which is situated in the brain, it is only the final step to convert all of this into a ‘spoken’ sentence that emanates from our mouth. For example, at the
PF
interface we can postulate that we ‘delete’ all the surplus elements from the sentence, such as empty categories and ‘null’ elements. There is even room for some pragmatics inasmuch as our speech-motor-system imposes certain constraints, such as having to take a breath every now and then, thus interrupting the flow of speech – which in its mental representation (sometimes called Logical Form or
LF
) has no such limits.

In 1970 the
MIT
Press launched
Linguistic Inquiry (
LI
)
under Samuel Jay Keyser. With Chomsky and the members of a ‘Who’s Who in Chomskyan linguistics’ on the editorial board, there was a mistaken assumption that it was a ready-made vehicle for dissemination, without any editorial constraint on Chomsky and his colleagues. In time
Linguistic Inquiry
became one of the best-known linguistics journals in the world.

As Chomsky’s books on linguistics began to sell well to a wider academic audience (as opposed to his difficulties in getting his political activist materials published), many an academic publisher began to sense a coup and anything written by Chomsky was snapped up. Later on this also applied to anything Chomsky said in public lectures on linguistics. Apart from the 1975 republication of his early work
The Logical Structure of Linguistic Theory
, there was never to be another book to equal the scope of
Sound Pattern of English
(1968) and the smaller
Aspects
of the Theory of Syntax
(1965). As his talks, interviews, lectures and articles were now becoming instant publications, there appeared a new type of collected works, often edited by a third person, such as C. P. Otero or Luigi Rizzi. Also, whenever and wherever Chomsky went, he typically delivered three types of talks or lectures: one on hard-core linguistics, one on language and philosophy, and one on political activism. Interviewers often quizzed him on all three topics and subsequent transcripts were cut up and reassembled in publications of one sort or another. One such groundbreaking publication was
Language and Responsibility: Based on Interviews with Mitsou Ronat
(1979).

Chomsky’s technical contributions to linguistics in the 1970s are much concerned with extending the Standard Theory, so much so that it became known as the Extended Standard Theory (
EST
). Inevitably, perhaps, it was succeeded by the Revised Extended Standard Theory (
REST
). Apart from the ever increasing number of strange acronyms, these developments grappled with the interface of the application of semantic rules, and a recognition that ‘deep’ structures might just be ‘shallow’. In other words, a sentence might be generated at one level of representation only. Further revisions gave rise to the notion that movement of sentence constituents leave behind traces, and that individual constituents can have intermediate categories. The latter became known as the x-bar theory, that is by virtue of the notation of placing one or more bars over category x. For example, a noun (
N
) phrase ‘the very fast car’ is a double-bar n, ‘very fast car’ is a single-bar n, and ‘car’ is
N
without a bar.
²⁵
This type of analysis allows for various types of expansions of sentences and phrases without having to invent new names for branch nodes. Another innovation was the lexical hypothesis that assigned a mini-grammar to lexical items (based on a feature matrix), which would allow for lexical selection rules to apply. For example, the English verb ‘bring’ selects two noun phrases, one as a theme (or patient/undergoer) and one as topic (or agent). In traditional grammar these noun phrases were commonly known as object and subject respectively. Note the reversal of the order in
REST
: the verb first selects for the theme (traditional object) and then later for the topic (traditionally the subject). This seems to be a touch of common sense that has been obscured by traditional grammar, which always begins with the subject. One can test such assumptions by some ad hoc self-investigation: what is the association one has with the verb ‘bring’? ‘Bring what?’ seems to be a natural response, and only later one asks for the ‘subject’, that is, ‘who brought it?’ Those who disagree might point out that, in English at least, one always starts with a subject as in ‘he brought the food’, hence the mental process of assembling this sentence should also start with the subject ‘he’. Not so, argue Chomsky and generative grammarians. First there is evidence from child language acquisition that in English children first acquire the verb-theme (for verbs that have the theme-topic choices) syntactic constituent before they advance to the fully fledged topic-verb-theme syntax. Research in neuroscience also suggests that the brain does not necessarily compute strings from left to right (as in parsing a sentence) but has the capacity to assemble (and comprehend) backwards and indeed in any reasonable order. Clear evidence is also presented by some verb-initial languages that place the so-called subject last (i.e. after verb and object). Latin is famous for not having any particular word order as the morphology clearly fixes the structure of the sentence: we know what the theme (object) and topic (subject) is by virtue of their word inflections. Linguists have long argued if there is a natural or universal order of basic syntactical constituents, and questions of left-branching or right-branching have exercised the minds of many generative linguists in particular. As we shall see, the Revised Extended Standard Theory (
REST
) was soon laid to rest, with new and exciting theories being developed by Chomsky and others in the 1980s, ‘90s and now.

In the meantime, the 1970s also proved to be a fertile period for Chomsky’s more philosophical enterprises. Having already discussed his claim on contemporary philosophy via his exchanges with the influential Harvard philosopher Quine, there was a flood of philosophical works, including an enlarged version of
Language and Mind
(1972) and the classic
Reflections on Language
(1975). The latter is a compilation of essays and lectures given at McMaster University in Ontario, in which he asks the time-honoured question ‘Why study language?’, and sets off to provide some intriguing answers. Neil Smith, an eminent British linguist who wrote a book on Chomsky’s ideas and ideals, notes that this volume is important in elaborating on the mind’s ‘modularity’.
²⁶
He credits Chomsky with providing ‘a wealth of evidence that the language faculty does indeed constitute a separate module, akin in many respects to any other organ of the body.’
²⁷
While neuroscience has not yet provided conclusive evidence to support such a claim, there are increasingly sophisticated experiments with results that point in this direction. For example, researchers from the Wellcome Department of Imaging Neuroscience in London have reported that children who grow up bilingual have a denser brain structure than monolingual children in parts of the cortex that are said to be responsible for fluency of speech.
²⁸
As such, empirical science is confirming the rationalist deductions made by the likes of Chomsky. Indeed Chomsky makes the point in
Reflections on Language
that the tasks of empirical science and that of rationalist philosophy go hand in hand. All great scientists have been philosophers to a degree, and many a great philosopher has started out as a scientist/mathematician, Bertrand Russell being a notable example.

The early 1970s also saw a remarkable meeting between Chomsky and Foucault on Dutch television. Foucault as the modernist gay philosopher par excellence and Chomsky as the linguist cum straight philosopher would appear to be as different as chalk and cheese. There are some deep ironies involved. Chomsky as the supreme rationalist is deeply indebted to the French philosopher and scientist Descartes, so much so that Chomsky’s whole enterprise is sometimes referred to as
Cartesian Linguistics
– indeed the title of a volume he published in 1966. There are other French antecedents, such as the Port-Royal school of linguistics and Ferdinand de Saussure, who pre-figured the essential Chomskyan dichotomy of
langue
versus
parole
. With such a pedigree it comes as a suprise that Chomsky has an almost absolute disdain for postmodern deconstructionist French philosophy, which has language at its core. There may be several explanations for this state of affairs. One is that post-modern French philosophy makes a strong claim on ‘political critique’, thus giving it the status of political activism. Marxist analysis, which underlies much of contemporary French philosophy and political activism, is anathema to Chomsky. Suffice to say here that Chomsky decries all obfuscation in political activism, calling for plain language and plain common sense when dealing with the lives of ordinary working people. According to Chomsky, the playful and highly idiosyncratic political discourse of French intellectuals, which is understood only in the salons of the elitist Left, is a betrayal of the working classes.

A discussion between Michel Foucault and Noam Chomsky on Dutch TV in 1971, moderated by Fons Elders.

By the 1980s some of Chomsky’s many students had become leading linguists in their own right and they could pursue Chomsky’s linguistics programme with further expansions, revisions – even innovations. And Chomsky himself continued with a flurry of radical innovations. First to appear were the
Lectures on Government and Binding: The Pisa Lectures
(1981), often referred to by the acronym
GB
. The Italian linguist Luigi Rizzi, who had been an associate professor of linguistics at
MIT
, arranged Chomsky’s visit and his lectures in Italy. Rizzi had done for Italian what Richard Kayne had done for French. Both linguists had shown that generative grammar does indeed apply to languages other than English – a suspicion sometimes raised, since hitherto much data had been based on English. Chomsky dismissed such suspicions as myths and pointed to his own early work on Hebrew, to G. H. Matthews’s
Hidatsa Syntax
(1965), to Robert Lee’s
MIT
dissertation, which was partly on Turkish, and most significantly to the arrival of Ken Hale at
MIT
in the mid-1960s, who elevated
MIT
to a world centre of Australian and American Indian languages – all within the generative framework.