anything to intervene in the scientific study of languethat is not derived from
langueitself (Milner’s first, third and fourth axioms express this principle,
which, for him, is key); of refusing to treat langueas a stable system (which
does not mean that linguistic phenomena are sheer chaos: Deleuze and Guattari
defend the idea that a language is a system of variations); and, finally, of
rejecting the homogeneity of the object in another way, by taking account of
the diversity of phenomena, the impossibility of defining an abstract,
a-historical langue, at the expense of variations of register, dialect, and
generations (as we have just seen in the case of English).
What Deleuze and Guattari bring out is the blind spot of most versions of
linguistic science (in this respect, Milner is an exception): they presuppose a
philosophy of language. Deleuze and Guattari at least possess the merit of
suggesting one, even if it remains in outline state. In the remainder of this
work, I shall attempt to develop it somewhat. This immediately poses a
question: is the philosophy of language that can be developed starting out
from Deleuze and Guattari a Marxist – or marxisant– philosophy?
Deleuze and Guattari’s relations with Marxism are controversial and, at
best, distant. Giving one’s magnum opusthe general title of ‘Capitalism and
Schizophrenia’ is not innocent: it implies a certain relationship to Marxism,
but also a deliberate shift, since Marxists are not in the habit of reflecting on
madness and desire. However, there is one thing that Deleuze and Guattari
did inherit from Marxism: critical vigour. It is insufficient to give us a Marxist
philosophy of language, but it will enable us to sketch it out. This critique
must be developed and its relevance established. To do this, I am going to
consider the most solidly established research programme in linguistics
worldwide, the one with which most mortals identify linguistics: Chomsky’s.
The importance of the Chomskyan research programme in the history of
linguistics, its dominant position (in a domain where theories proliferate:
each linguist has her own grammar), are not the only reasons for my choice.
First of all, there is its empirical importance – that is, its ability to account
for linguistic phenomena. (In this respect, the history of the research programme,
which is revolutionised by Chomsky every five years, is full of lessons: a sort
of involution has seen the model take its distance from the explanation of
grammatical phenomena, in order to concentrate on marginal phenomena
and ever greater levels of abstraction.) But, above all, there is the explicit
character of the model: Chomsky is one of the rare linguists who is fully
18 • Chapter Two