FOR THE PAST 20 years Noam Chomsky has been at the center of one of the great, quiet intellectual adventures of the 20th century -- a massive cross-disciplinary effort by mathematicians, psychologists, biologists, linguists and many others to understand the nature of language and communication.

They have turned linguistics into an intensely intellectual activity, perhaps the most precocious of the behavioral sciences, with its own arcane vocabulary, abstractions, formulas, procedures. Much of this is as inaccessible to the untrained as the workings of astrobiology.

Almost from the time he joined the Massachusetts Institute of Technology in 1955, Chomsky has helped transform linguistics from a field controlled by behaviorists -- those who watched and reported on the preformance of speech -- to a territory dominated by scientists and philosophers of an idealjstic and cognitive persuasion.

Chomsky virtually revolutionized contemporary linguistics when he reached all the way back to Plato and Descartes for an old idea that he clothed in new grammatical equations and definitions. In Syntactic Structures (1957) Chomsky presented and seemed to demonstrate the proposition that every human being has an innate ability to acquire language, and this ability to learn language is called into use when one hears, at the right age, language for the first time. He also offered a concept -- it came to be known as "generative" or "transformational-generative" grammar -- which made it possible to predict ("generate") the sentence combinations in a language and to describe their structure.

During the ensuing years Chomsky told us in a number of books -- Current Issues in Linguistic Theory (1964), Language and Mind (1968), Problems of Knowledge and Freedom (1972) -- that his transformational-generative approach could explore the entire system of surface and deep structural relationships in language, that linguistics must subsume both phonology and semantics, and "that there apparently are deep-seated and rather abstract principles of a very general nature that determine the form and interpretation of sentences." Chomsky's implication was that perhaps we all use language in somewhat the same way that we all see color or hear sound. Thus, language can provide a window to the mind, which itself cannot be observed, and may instruct us in how choices, concepts, fantasies are made.

Even those social scientists who believe Chomsky has gone awry or too far with his deductive and mentalistic grammar, acknowledge that Chomsky is the most celebrated linguist of the past two decades and a scholar of world renown. His work has propelled linguistics into new fields and revolutionized tionized its goals, techniques, and range of knowledge.

But, ironically, it took the Vietnam war and Chomsky's passionately idealistic stand against U.S. actions in Southeast Asia to bring him to public attention. He joined Norman Mailer, Dwight Macdonald, Robert Lowell, and others to march on the Pentagon in October 1967. In speeches, essays, and in the many books he wrote on language, Chomsky waged his intellectual war against "The War."

Language and Responsibility brings together both sides of Noam Chomsky in a lengthy dialogue with the French linguist Mitsou Ronat, who shares both Chomsky's philosophy of language and conceptions of "American imperialist ideology." Originally published by Ronat in French, it was revised somewhat by Chomsky during translation into English. The purpose of the book apparently is to present Chomsky's post-Vietnam, post-Watergate views of U.S. society and politics, as well as give a contemporary summary of what Chomsky's generative grammar does and what modern linguistic study is searching for.

The political part of the discussion is clear enough, and familiar. There's a "very effective kind of ideological codntrol" in the U.S., Chomsky says, managed by politicians, media lords, capitalists, and mainstream American intellectuals. This "system of thought control" restricts how we perceive ourselves, the alternatives we can imagine, our understanding of the rest of the world, and, most importantly, it prevents any major ideological changes from taking place in the U.S.

Why our culture operates as it does is an immensely complicated subject. Chomsky oversimplifies it, having reached his conclusions during a period of extreme stress. He accounts for the homogeneity of American political thought as a byproduct of our lacking a historic Left, the Communist scare of the late '40s, and the goal of U.S. foreign policy "to create a global economic order that conforms to the needs of the U.S. economy and its masters."

We're asleep, says Chomsky: "In comparison with Cointelpro [the F.B.I.'s program begun in the '50s to disrupt the U.S. Communist Party] and related government actions in the 1960s, Watergate was a tea party.... It was the improper choice of targets, not improper acts, that led to Nixon's downfall." He "made a mistake in his choice of enemies: he had on his list the chairman of IBM, senior government advisers, distinguished pundits of the press, highly placed supporters of the Democratic Party. He attacked the Washington Post , a major capitalist enterprise."

"In the United States," he continues, "political discourse and debate has often been less diversified even than in certain Fascist countries, Franco Spain, for example."

Chomsky is simply wrong on some counts. The student revolts of the '60s were not, as he says, merely efforts "to open up the universities and free them from outside control." They were part of the hysteria, anarchy and general breakdown of authority at that time. Liberal intellectuals in the '50s did not fault Joseph McCarthy, as he says, "on the grounds that his methods were not the right ones for ridding the country of real communists." Many of them faulted McCarthy (and Nixon) because his witchhunt against fewer than 50,000 members of the U.S. Communist Party was mad political sport, contrary to Jefferson's and Madison's Constitution, and diverted us from finding a modus vivendi with the Soviet Union.

But Chomsky also speaks truths. We do not have significant Marxist or Socialist journalism in the U.S., and they might well enliven debate. Capitalism, albeit modified and socialized, is our way of economic life, and we're indoctrinated to it. Indeed, Solzhenitsyn is a fanatic, and we do enjoy the way Soviet dissidents confront moral issues while we ignore moral issues at home.

To find the rewards in the eight chapters of Language and Responsibility that deal with linguistics requires an active familiarity with Chomsky's theories and the vocabulary and equations he uses to describe them. Chomsky is speaking to another linguist, Mitsou Ronat and the dialogue is for the initiated -- those who understand that "deep structure," "universal grammar," "rewrite rules," and "the principle of recoverability of deletion" define a Chomskyan world where all who speak meet as equals.

In these conversations which took place in 1976, Chomsky is looking far into the future of his science. He believes that "human nature is not as yet within the range of science. Up to the present, it has escaped the reach of scientific inquiry; but I believe that in specific domains such as the study of language, we can begin to formulate a significant concept of 'human nature,' in its intellectual and cognitive aspects."

As he and Madame Ronat examine the domain of their science and the available concepts for exploring it (most of them are Chomsky's discoveries), one has the sense of something important taking place, and that to bring it to public understanding will require explication by someone other than Noam Chomsky. Meanwhile, he has grown more tentative in his claims, more speculative in his visions.