The study of language has become something of a free-fire zone in recent years, with everyone who has the slightest interest in the subject eager to take a few pot shots at such age-old questions as "What is meaning?" Some public interest and amusement has been spurred by the intramural squabbling of Chomskians and neo Chomskians, generative semanticists and Modified Standard Theoreticians; and the narrow, doctrinaire character of those disputes has spurred people with a broader interest in language to join the fray. Walker Percy, a novelist (The Moviegoer, Love in the Ruins), sides with those who consider contemporary linguistics as just so much debris that needs to be swept aside before a really perceptive examination of language can begin.

To place Percy in an anti-Chomskian camp or an anti structuralist camp would be misleading, however; the 20 years' worth of essays newly collected in The Message in the Bottle are nothing if not iconoclastic. Percy styles himself a solitary thinker, sitting for from the inhospitable blinding turmoll of the university. The first essay in this volume is subtitled "How I Discovered the Delta Factor Sitting at My Desk One Summer Day in Louisiana in the 1950's Thinking About an Event in the Life of Helen Keller on Another Summer Day in Alabama in 1887," and you get the feeling that's how he does his best work, meditating in his Covington, Louisiana home on the state of languages and the world. If his antipathy toward the major centers of learning seems unscholarly and unprofessional, it's meant to.

The professionals have made a botch of it, in Percy's view: the transformational grammarians, bickering over esoteric aspects of an arcane theory, have lost sight of the forest for the trees. And they've lost sight of language, too--the proper concerns of linguists have been parceled out to specialists in psychology, anthropology, theology and so forth. Percy feels he is the man to bring order to all this mess:

As a non-psychologist, a non-anthropologist, a non-theologian, a non-ethologist--as in fact nothing more than a novelist--I qualify through my ignorance as a terrestrial Martian. Since I am only a novelist, a somewhat estranged and bewildered person whose business it is to see things and people as if he had never seen them before, it is possible for me not only to observe scientists observing people as data--in short to take a Martian view.

When he is not posing as a Martian, he compares himself to the boy who sees that the Emperor has no clothes--the Emperor, of course, being a contemporary linguistics garbed in imaginary transformational rules and fanciful levels of "deep structure."

But crediting himself with the "Martian view" doesn't get Percy very far toward debunking the Chomskians, and adds nothing but a gratingly false modesty to his broad assertions about the state of linguistics.

Unless the Martian is very much mistaken and it is here that he does resemble somewhat the boy who noticed something wrong with the king--it appears to him that while the prevailing behaviorist theory has been dismantled, no other theory has been advanced to take its place, this in spite of all the talk by transformationalists about "explanatory models."

But Percy's claim to a perspective that "can commend itself ... more by reason of its ignorance than its knowledge" is more than an attack on the scientists' narrow focus; it questions the fitness of an analytical, scientific method to the task of comprehending language. Not quite a full-fledged mystic, Percy nevertheless doubts not only the possibility, but also the ultimate worth of understanding. On the one hand he glories in the achievement, unique to the human race, of making the associative leap from the group of sounds in balloon to the real balloon; yet at the same time he views this as a self-deception and a tragedy. After all, the word balloon is not a balloon, and our confusion of the two is the cause of what he refers to variously (these essays were written at different times) as alienation or existential despair.

That first crucial grasp of language (which Percy dubs, for personal reasons, the "Delta Factor") is the object of "a mild twenty-year obsession" on Percy's part. It separates man from beast; it gives him a unique tool for understanding his condition. Percy associates it with another obsession of his, the sort of inexplicable, poetic joy that everyone experiences from time to time--this he calls the "Helen Keller phenomenon," because it is the way Helen Keller felt when, later than most people, she first connected the word water with the actual item. But he also blames language for twentieth-century man's estrangement from the natural world, by naming a thing. Percy feels, we subdue it, we reduce it to less than it is.

Beneath his loathing for the professional linguists lies the fact that they too aim to demystify something that Percy would like to see as great and irreducible. He believes that there is something about language that all the linguists, psychologists and logicians have missed--something that cannot be broken down into morphemes and phonemes, syntax and lexis, stimulus and response.

"But wait," says the Martian. "What about the actual event of language? The central phenomenon? What happens when people talk, when one person names something or says a sentence about something and another person understands him?"

It is a question that has no answer, because Percy will be satisfied with none. He doesn't want the acoustician or neurobiologist to tell him what happens physically, and he does not want metaphysical explanations from philosophers or algorithmic ones from grammarians.

While part of Percy wants to see language as magical and ineffable, another part of him would like to out-analyze the scientists. The Message in the Bottle slowly progresses from essays "toward" various theories of meaning and what-not to a final manifesto that, he says, only professional linguists will read: "A Theory of Language." This is the essay that boasts ignorance as one of its chief virtues. It begins with "the Discovery That an Explanatory Theory Does Not Presently Exist," and concludes with, if not a theory, at least a suggestion of where to look. Percy's occasionally cloying personal style becomes suffocating in the essay, because his foray into serious linguistic analysis is so confidently off the mark. He believes firmly and sincerely that the naive simplicity of his arguments have been overlooked by linguists for years, tangled as they are in their convoluted transformational web.

Percy's ignorance, mitigated only by a spotty familiarity with a wide range of writing on language, is no asset. The meat of his hinted-at theory is a rediscovery of Charles Peirce's linguistic philosophy. Percy says Peirce has been completely overlooked by modern linguists. His ignorance of the literature hurts him here--Henning Andersen, for one, has used a Peircean model of language acquisition in his work in generative phonology--and it hurts him everywhere he tries to argue on the linguists own turf. Percy adopts the jargon of the professionals and scholars without the sense, and ultimately he can only argue from his own intuitions about language.

The great paradox of modern linguistics is that while human judgments about the nature of language must be fundamentally introspective our intuitions are often misleading or simply wrong Percy is right when he says that closeness to language can be a bar to understanding it, that we think and talk about language in the very medium we are trying to study. Linguists cannot isolate their specimens and examine them under the objective lens of a microscope; instead they have tried desperately to replace intuition with a rigorous scientific method. Percy believes his death blow to the Chomskians is his assertion that, by their own admission, the transformational model is not actually the mechanism we use to create sentences, and that such artifices as deep structure don't correspond to any of our intuitions about language.

Our intuition is that sentences like Max is easy to please and Max is eager to please are basically the same construction, differing only in meaning--and that's the way two hundred years of English grammars described them. In fact, the structures of those sentences are very different, and one way to describe that difference is in terms of a notion of deep structure. It bothers Percy that the transformational grammar is not somehow built into our nerves and synapses, actually generating sentences while we talk, but that is the paradox: The model reflects a knowledge we do have about language, artificial though it may be. It is a grammar stitched together from imaginary rules and devices but, for all that, it is no less palpable and no less real.