Noam Chomsky (8 page)

In his
GB
lectures Chomsky set forth a revolutionary departure from the previous model (
REST
), replacing a plethora of rules with simple but powerful principles that could account for much of the data. We have already heard of the x-bar theory, and now there was Move a (move alpha). The movement of constituents in transformations, especially in question formation (known as Wh-movement), had become so complex that at times it became impossible to follow the rules. If one could find an over-arching principle that accounted for all types of movement, then the day was saved. To do so Chomskyan linguists devised Move a, which allows for movement of any phrasal or lexical items as long as it involves substitution or adjunction. The technical details are still complex, as various other principles of
GB
intersect. Introducing now the Binding Theory, we have to have a look at anaphoric relations such as:

(examples based on Smith).
²⁹
It seems clear that the instances of ‘himself’ can only refer to John, while the instances of ‘him’ cannot refer to John but rather to anyone else, including Bill. In other words ‘himself’ is bound to an antecedent while ‘him’ is free (i.e. not bound). Consider then the sentence

where ‘him’ can indeed refer to John as well (note that we cannot replace ‘him’ with ‘himself’ in this context). Given that pronouns play a vital role in any language, this simple sounding principle of ‘binding’ accounts, in combination with other principles, for a whole host of phenomena that previously were difficult to capture in a unified way. Note also that ‘himself’ (in (d) above) cannot refer to Bill. This is due to a locality principle that says that ‘binding’ only applies to a local domain, that is it cannot jump across certain syntactic boundaries. The same applies to movements, that is Move a. Another crucial theory that evolved was the so-called Theta Theory (abbreviated as θ-theory, and derived from the ‘th’ in ‘thematic’). It is a clever reworking of what in traditional grammar is generally known as (in)transitivity or valency. As we know in traditional parlance, a verb can be either intransitive (have only a subject) or it can be transitive (have a subject and object), or even ditransitive (subject, direct and indirect object). In Chomsky’s new GB theory, which by now is really a collaborative effort involving many linguists, these thematic relationships between verb and arguments (noun phrases) are newly elaborated in terms of the verb selecting the number of nominal arguments it needs so as to generate a well-formed sentence. Since the terms ‘subject’ and ‘object’ have become difficult to define, especially across languages, there are new labels for them, namely topic (for subject) and ‘theme’ (for object) and ‘agent’ and ‘patient’, depending on the role these nominal arguments play. We know from English that the active-passive transformation of

leaves the cat as ‘agent’ and the mouse as ‘patient’, regardless of the cat being the ‘topic’ in (a) and the mouse being the ‘topic’ in (b). In Government and Binding theory it is stipulated that each nominal argument bears one, and only one, theta-role, hence the selection of ‘agent’ and ‘patient’ as labels that do not change even if their position in the sentence changes. English as an accusative language has many verbs that are thus labelled ‘accusative’ because they typically have an agent as a subject. Given this scenario it is interesting to note that English also has un-accusative verbs, such as ‘undergo’, where the subject is the patient as in:

Theta theory also accounts for so-called ergative languages, which have quite a different system of verbs, namely predominantly passive-like verbs whereby the patient is the subject and the agent is a kind of optional object. Take an example from the Polynesian language of Niue:

This is a canonical sentence in Niuean and the literal translation is best rendered in the English passive (Niuean does not have a passive construction). The agent in the above sentence,
‘he pusi’
(the cat), is marked with an ergative case marker and can be optionally deleted to yield:

Note that sentence (b) cannot mean ‘the mouse was eating’. This is due to the fact that the Niuean verb
‘kai’
(to eat) selects for theme (object) the agent, and for topic (subject) the patient. When the theme (object) is deleted optionally, the patient remains as topic (subject). If Niuean wants to express the equivalent of the English ‘the mouse is eating’, it will have to use a different verb that is intransitive and selects for agent as topic (subject). Parallel to accusative languages, ergative languages have some un-ergative verbs (or it can be argued that accusative languages have some ergative verbs, and ergative languages have some accusative verbs).

Finally we get to the ‘government’ of the Government and Binding theory. This again involves a clever reworking of what is known in traditional grammar as ‘case’. Latin, for example, is famous for its complex system of cases: nominative, vocative, accusative, genitive, dative and ablative. (English retains some in the pronominal system, such as ‘he’ (nominative) and ‘him’ (accusative). While in structural and descriptive grammars ‘case’ is a feature of nouns depending on what structural role they play,
GB
goes a step further and asserts that, since it is the verb that selects its nominal arguments, it must also be the verb that ‘governs’ case. In other words, the verb assigns case to nouns it governs. The details of this operation are too complex to elaborate here, but this opens up a whole new way of looking at it. For example, one can now say that nouns (or noun phrases) must be ‘licensed’ by a case node somewhere in the derivation. Another highly evocative explanation is that each new binary pair of syntactic objects ‘merges’ (naturally known as operation
MERGE
) into a constituent and all its syntactic objects must be ‘checked’ against the principles given in
GB
, such as having satisfied the requirements of the ‘government’ theory.

Given the whole binary nature of derivations, there is another important process that has become part of the key of
GB
. This is the idea that the parameters of a language are fixed by making choices between binary features. This is a powerful explanatory tool that allows us to make quite fundamental decisions, for example as to what language we are speaking. A child growing up in Niue will have as input sufficient data that will decide whether or not Niuean is an accusative or ergative language. Once this fundamental parameter is fixed, the child will go on to fix further parameters until the whole set of language principles is applied and specific structures have been acquired. This unified approach has been labelled by Chomsky as ‘principles and parameters’ and constitutes one of the great advances in linguistics today.
GB
was further refined by Chomsky with his publication of
Barriers
(1986), a slim but highly technical
Linguistic Inquiry
monograph. In it he discusses possible ‘barriers’ to government and movement, noting that one might ‘expect that one barrier suffices to block government, whereas more than one barrier inhibits movement, perhaps in a graded manner.’
³⁰
As can be expected, it is a highly technical treatise that has baffled many a postgraduate student and fully fledged linguist alike, but as it already points in the direction of a further development, it is worthwhile pointing out one of the key issues covered. Non-lexical categories now also become syntactical projections as part of the x-bar theory. While we are familiar with
NP
(Noun Phrase),
VP
(Verb Phrase),
PP
(Prepositional Phrase) and the like, which open up lexical slots to be filled, we now have the more abstract Complementizer Phrase (
CP
) and
IP
, where I stands for the category of ‘inflection’, which includes tense, modals and agreement (case). In other words, when we start off with a verb at the beginning of a derivation, the verb projects various lexical categories and these non-lexical categories. Leaving aside the
CP
, the idea of the
IP
being a projection that governs various constituents along the tree of derivation is very interesting. It is perhaps the common sense realization that tense is a concept imposed on a sentence as a whole rather than being narrowly associated with the verb alone. The binary principle also gives rise to another common sense observation often obscured by traditional grammars, namely, as far as tense is concerned, there is the initial bifurcation of only past and non-past. The old trifurcation of past-present-future is thus an obfuscation that has plagued language learners for a long time. Many of today’s English-as-a-second-language textbooks still begin their lessons with the ‘present tense’ as the basis for all tenses available in English. Any serious reflection will show that the ‘present tense’ is in fact a highly specialized tense that evades easy definition. Beginning with the simple past tense is the natural way to go: an insight thanks to
GB
, if not to common sense.

A flurry of publications in the 1980s confirmed Chomsky as a pre-eminent philosopher, naturally in the domain of language. Having already noted that he has never engaged in any sort of political philosophy, especially of the French contemporary sort, we must keep to the strict demarcation of philosophy and political activism for reasons also stated above. Naturally even a philosophy of language impinges on the actions of mankind, and Chomsky for one is always ready to point out that our language capacity is the ultimate tool to change the world and make it a better place. Still, as he points out in his seminal collection of philosophical essays
Rules and Representations
(1980), ‘the study of acquisition of knowledge or of the interpretation of experience through the use of acquired knowledge still leaves open the question of the causation of behaviour, and more broadly, our ability to choose and decide what we will do.’
³¹
What is becoming a recurrent theme in Chomsky’s system of ideas and ideals is also highlighted again in
Rules and Representations
, namely the biological basis of language capacities. Trying his best to delimit the emerging idea of a specific science to be known as bio-linguistics, he resents being co-opted into the latest trend of the time, called cognitive linguistics, in which language is part of a wide-ranging cognitive apparatus – also acknowledged to have a biological basis. He says that ‘one must deplore the common tendency to insist that the mechanisms of language must be special cases of “generalized learning strategies” or general cognitive mechanisms of some sort.’
³²

By the end of the 1980s Chomsky had become a global phenomenon. The first major bibliography of his work, compiled by two Indian authors in 1984, listed more than 180 publications by him and double that number for publications about him.
³³
A 2005 search on Amazon Books listed more than 600 items by, with or about Chomsky. He has become the most cited living author of our time, and he is among the top ten authors of all time. His voice is heard in academia beyond linguistics and philosophy: from computer science to neuroscience, from anthropology to education, mathematics and literary criticism. If we include Chomsky’s political activism then the boundaries become quite blurred, and it comes as no surprise that Chomsky is increasingly seen as enemy number one by those who inhabit that wide sphere of reactionary discourse and action.

‘Chomsky and Predicate’ comic strip.

Chomsky’s critics in linguistics and philosophy have mainly been focused on the latter field (for example Chomsky vs. Quine), even though some commentators labelled arguments in the former as the ‘linguistics wars’.
³⁴
Chomsky saw the whole excitement merely as a healthy debate in linguistics. He uses much more forceful language in the realm of philosophy, as we have seen already. While we will allude to some of his more irrational critics in the field of political activism, there was one so-called affair, starting in late 1979 and stretching into the 1980s, that seemingly began as an academic exercise. A French academic named Faurisson was suspended on the grounds that his university would not defend him from violence after he privately published some monographs on gas chambers; he was later brought to trial for ‘falsification of history’ (a crime in France). Chomsky, together with 500 others, signed a petition in favour of freedom of expression to be applied to Faurisson. To put it more precisely, Chomsky pointed out that he rejects the Nazi–Stalinist doctrine that the state has the right to determine Historical Truth and punish deviation from it. This, and only this, was the issue at stake. In what was an early sign of ‘political correctness’ gone wrong, there was outrage from a confused Left and Right, all vilifying Chomsky as some sort of criminal. Chomsky protested that in no way did he subscribe to Faurisson’s alleged anti-Semitism and Holocaust denial; he merely supported his right to freedom of expression. Chomsky’s dictum is that one must engage with the issues and not with the man or woman professing them. Had more Germans engaged with the issues raised in Hitler’s
Mein Kampf
, Hitler might not have come as far as he did. One cannot tackle an idea by removing the person holding it. It is thus somewhat surprising – Neil Smith in his book on Chomsky makes the same point
³⁵
– that in Barsky’s otherwise sympathetic biography of Chomsky it is noted that ‘the Faurisson affair does tend to throw some of Chomsky’s character flaws into relief’.
³⁶