Living by Fiction

by Annie Dillard

  There is something about plain writing which smacks of moral goodness. Interestingly, some writers turn to it more and more as they get older. There is a modesty to it. Paul Horgan uses three different plain prose styles in his Richard trilogy, a series of novels which take the form of autobiography. Henry Green uses plain prose in his autobiography. Graham Greene uses it in his autobiography. It is a mature prose. It honors the world. It is courteous. Its credo might be that of French entomologist J. Henri Fabre: “Lucidity is the sovereign politeness of the writer. I do my best to achieve it.” Part of its politeness to readers is based on respect; this prose credits readers with feeling and intelligence. It does not explain events in all their ramifications; it does not color a scene emotionally so that a reader knows what he should feel.

  This prose is humble. It does not call attention to itself but to the world. It is intimate with character; it is sympathetic and may be democratic. It submits to the world; it is honest. It praises the world by seeing it. It seems even tobelieve in the world it honors with so much careful attention. In the nineteenth century, readers liked their prose syntactically baroque and morally elevating. Each bit of world was a chip off the old sublime, and tended distressingly, in the prose which described it, to ascend to heaven before we got to know it.

  Nothing so fair, so pure, and at the same time so large, as a lake, perchance, lies on the surface of the earth. Sky water. It needs no fence. Nations come and go without defiling it. It is a mirror which no stone can crack, whose quicksilver will never wear off, whose gilding Nature continually repairs…

  —Thoreau,Walden

  Our reaction to such ebullience is of course to reverse it. We have modern tastes and like writing which is precise and uncluttered. We are agnostic or materialist and like writing fastened to the world of things. This plain prose represents literature’s new morality. It honors each thing one by one, without metaphor. No angelic systems need be dragged in by the hair to sprinkle upon objects a borrowed splendor. Instead, each of the world’s unique objects is the site of its own truth and goodness. Each thing is its own context for meaning. Its virtue is its stubborn uniqueness, in its resistance to generalization, or even in its resistance to our final knowledge of it. The most general trend we know is speciation.

  Plain prose can be polished to transparency without losing strength. At its best, its form follows its function so accurately that its very purity and hard-won simplicity excite our admiration almost in spite of themselves. It does not err on the side of exuberance. It does,in theory , win through to material “things as they are”—things seen without bias or motive. That can be its epistemological claim. And aesthetically it can claim control, purity, and the dignity of material essences. And it can claim the just precision of a tool, the spareness of bone, the clarity of light. Do not confuse these claims with the clichés of contemporary craftsmen in materials who labor to help you understand that a wooden spoon may have integrity, a wooden apple barrel may have dignity, a wooden bench simple functional beauty, so that you labor in turn to find some kindling and a match. There is nothing clichéd about clear prose yet. For all its virtues, fine writing may be a mere pyrotechnic display, dazzling and done. And plain writing is not a pyrotechnic display, but a lamp.

  If we call very opaque modernist prose a painted sphere, and plain prose a clear windowpane, then we will see that these are extremes; most literary prose belongs somewhere in the middle. Or if we call fancy experimental prose “poetry” and plain prose “science,” then again we will see that most prose falls somewhere in the middle. In this middle ground we have contemporaries like Borges and Ralph Ellison writing complex modernist fiction using straightforward prose. And we have writers like Tillie Olsen writing stories of intimacy and depth whose virtues are traditional, yet whose prose surfaces are nevertheless dislocated and modernist. Most frequently we have writers of traditional literature modified by this century’s concerns, who use a beautiful and strong literary prose modified by this century’s taste for flat and shifting surfaces. In England, for instance, a prose which is at once perfectly clean yet capable of stunning elegance still endears the writers of the last generation to mine: the prose of E. M. Forster and Joyce Cary on the elaborate side, and on the spare side, Henry Green and Anthony Powell.

  The species of prose blur. Writers of course vary their modes from book to book. Sometimes even within a book—a book of short stories—we find radically different kinds of excellent prose bent to radically different and excellent ends. I am thinking here of Richard Selzer’sRituals of Surgery , ofGetting into Death by New York writer Thomas M. Disch, and ofBeasts of the Southern Wild by Carolina writer Doris Betts. Every mode is an option now. It is anybody’s ball game.

  Let me add this. You know how a puppy, when you point off in one direction for him, looks at your hand. It is hard to train him not to. The modernist arts in this century have gone to a great deal of trouble tountrain us readers, to force us to look at the hand. Contemporary modernist fine prose says, Look at my hand. Plain prose says, Look over there. But these are matters of emphasis. So long as words refer, the literary arts will continue to do two things at once, just as all representational painting does two things at once. They point to the world with a hand.

  I think the very finest works of art do both things at once and well. Cézanne’s still lifes and landscapes, for instance, depict. They push the paint surface into a modified simulation of deep space; the meadow tilts back toward the mountain. At the same time they pull deep space up toward the surface of the picture plane; the mountain looms flat against the canvas. The paintings’ greatness depends on this spatial tension. Just so do artifice and sincerity meet and balance in a great work of art. We teeter at the edge of the artists’ representations, affected by their depths and at the same time admiring their effects. Look at Shakespeare. Who could say where the greater power ofKing Lear resides? Do we enter it as an emotional world of enlarged sympathies wherein we lose ourselves? Or do we admire it more as an artifice of theater and language?

  Finally, it is interesting to note Robbe-Grillet’s peculiar notion (which I think is accurate) that a writer thinks of a future novel first as “a way of writing.” The narrative, he says, “what will happen in the book [,] comes afterward, as though secreted by the style itself.” This is interesting because it stresses again the primacy, for the modernist, of notions of surface treatment and handling. One does notchoose a prose, or a handling of paint, as a fitting tool for a given task, the way one chooses a 5/16 wrench to loosen a 5/16 bolt. Rather—and rather creepily—the prose “secretes” the book. The narrative is a side effect of the prose, as our vision is a side effect of our seeing. Prose is a kind of cognitive tool which secretes its objects—as though a set of tools were to create the very engines it could enter, as though a wielded wrench, like a waved soap bubble wand, were to emit a trail of fitted bolts in its wake.

  PART THREE

  Does the World Have Meaning?

  CHAPTER 8

  The Hope of the Race

  Octavio Paz points out that since the breakdown of religion and metaphysics, “we have criticism instead of ideas, methods instead of systems. Our only idea, in the proper sense of the term, is Criticism.” The Myth of Criticism, he repeats, is “the only modern idea.”

  Depressing as the thought is, the myth of criticism is a workable myth, as powerful as any. Criticism is a kind of modern focusing of the religious impulse, the hope of the race: the faith that something has meaning, and we may apprehend it. Are we people given to know the meaning of something, even if it is only the tiny textual world fashioned by our neighbor? (If so, the mind will never run out of meaningful objects.) Art has meaning, which criticism discerns. This is a cheerful state of affairs, actually, but it obtains only if we grant that art actually does mean (a topic I will tackle presently) and also if we grant that criticism can know art.

  May we entertain these hopes?

  May We Discover Meaning?


  Bear with me, please, for a few difficult points before the shooting starts. Readers who are not interested in the complex internal problems of literary criticism should skip directly to (Peirce); there, this chapter, and the book as a whole, begin to take off at last.

  Certain contemporary critics, who might be called quasi-structuralists, hold that both texts and criticism are locked in “the prisonhouse of language” (the phrase is Nietzsche’s), so that while criticism fails to refer meaningfully to anything, that is all right, because so does art. This view licenses the critic to perform all sorts of aerialist feats. It was with great good humor that Neil H. Hertz once provisionally titled an essay “Wordsworth’s Influence on Milton.” Of course, these critics deny the notion of influence and the notion of theme. All the universe is language, and language is self-referential. Critic Jeffrey Mehlman, inRevolution and Repetition , took a word which Freud used in a specific, quirky, and fully elaborated sense (unheimlich—uncanny), and “inscribed” Freud’s meaning for that word back into an unrelated text by Marx. By this curious device, Mehlman produced, as one might expect, an ingenious and novel reading of the Marx text. (It is as if we were to take Psalm 121—“I will lift up mine eyes unto the hills, from whence cometh my help”—and read “the hills” as Ché Guevara might use the term.) If we complain that such artificial readings are merely playful or ingenious, that they are irresponsible because they do notrefer to anything actual, these critics can reply that all mental effort and verbal artifact is artificial, playful, ingenious, and irresponsible.

  This is an extreme position. Let us step down a notch to another position, which holds that while art may be meaningful within itself, the art object is always unknowable. Criticism must always try to know a text on its own terms; but it will always fail. Criticism cannot know its object. There is no guaranteed thread of connection between any interpretation and any text; so criticism is a particularly fanciful and baroque form of skywriting. Formalist critics, for instance, may hold that a text is wholly sealed. Its locked lights bat back and forth inside their can. Therefore, as Harold Bloom said once: “There are no interpretations but only misinterpretations, and so all criticism is prose poetry.” In this context, “prose poetry” signifies a system coherent but detached, whose reference never overflows its pages and never meets any external object—world or text. And so criticism is blind and dreaming.

  But this, too, is an extreme position which the embattled intellect cannot long defend. For it is simply too absurd to approach an object which you (and you alone) must define as unapproachable, and then to produce a careful interpretation of it which you must judge meaningless. Even the most depressive critic among us does not doubt, I think, that his interpretation is worth a little something. Why else would he teach it, write it, and defend it? And he does not really doubt, I think, that the text at hand leaks light all over creation. It may be reductive to say thatOthello is about jealousy or that “The Emperor of Ice Cream” is about a funeral, but it is more reductive to say that these texts do not concern these things at all, nor do they touch anything whatever that we can name; and it is a mad exaggeration to say that the words of a text are runic, like so many dots of paint. We can interpret texts because texts use a shared language which refers, however clumsily, to a shared world. We may neverexhaust the meaning of a text, or our knowledge even of its textual surface; but to acknowledge that we can never know all is not to decide that we can know nothing.

  Further, any kind of critical interpretation may be sound and useful without being airtight or guaranteed. Must we say that we can never interpret? May we suggest that the crow and the dust of snow and the hemlock tree in Frost’s little poem “Dust of Snow” are reminders of death? Or could we just as reliably hold that the crow and the snow and the hemlock are symbols of homosexual desire, or Yankee food taboos? Skepticism and the relativism it breeds are, in this case and as usual, nonsense. The fact that there are cases difficult to interpret within a given cultural context does not alter the fact that there are clearer cases. Criticism accumulates an ordered pile of sound work behind it just as physics does. (It could even be that criticism is on firmer ground than physics—because cultural phenomena occur on an accessible middle ground, and human fabrications fit human understanding.) The concepts of class, marketplace, and culture, for instance, are interpretative notions which have assumed the status of common “hard” knowledge, just as Freud’s interpretative concepts have, and those of Malthus.

  To judge among interpretations and methods, we must resort either to the authority of the author’s stated intentions, if any, or to common sense, or to a consensus among educated people of goodwill. None of these satisfies anyone, but there is nothing else. Any effort to make of criticism an exact science necessarily limits its materials to only those things which can be known for certain—such as the number of times a given word occurs in an established text—and therefore evades criticism’s interpretative function almost entirely. (Of course, criticism also contributes to knowledge high heaps of more or less undisputed data—data historical, biographical, and bibliographical. Such studies constitute a great part of criticism’s business, but they are not at issue here except insofar as they may enlighten interpretations of texts.)

  Say, then, that although criticism may produce solid and useful interpretations of texts, it can never test its methods or guarantee its findings. Is criticism then prose poetry? What can we say of criticism’s many and excellent interpretations—J. Hillis Miller onBleak House , Harold Bloom on “Byzantium,” Northrop Frye on “Jerusalem”—that these readings are true? Or that they are, by consensus, probable, workable, and fruitful? We can say the latter. And that is good enough.

  For on what ground do the other branches of knowledge stand? After all, there is no epistemological guarantee betweenany subject andany object. The world itself may be sealed. If we are talking about knowledge (holding interpretation at bay for the moment), criticism is on no swampier ground than any other branch of knowledge. It could be, even, that texts are a great dealmore accessible to knowledge than other objects. At least we do not dispute that textsexist . Even when general debate stretches to the point where we doubt (or feign to doubt) that the world out there exists, any of it, we seldom if ever find our epistemological panic focused on the issue of texts. To vary a Woody Allen joke: It could be that the universe has no existence independent of our perceptions, that the universe is a fabrication of our dreaming senses—in which case I definitely overpaid for my copy ofLord Jim .

  At any rate, even if the great world exists, there are ample reasons to deny that we can know it. The abyss between any subject and any object and the utter lack of any guaranteed relationship between them confronts anyone who thinks about anything. All possible knowledge, from the identification of species to the size of your foot, is necessarily interpretative. Inhibitors in our neurons edit the garbled impressions of our meager senses before they reach our programmed brains. All language-using endeavor is culturally purblind. So all our data are at best biased. At worst, we can know nothing. Profound epistemological skepticism and the blithe relativism which accompanies it are always intellectual options, or at least roles, and for anyone who assumes them,all the branches of knowledge are prose poetry.

  But as C. S. Peirce wrote, perhaps sentimentally: “Let us not pretend to doubt in philosophy what we do not doubt in our hearts.” We may well doubt that all things can be known or understood, but we do not really doubt thatsome things can be known and understood—or else we would neither argue nor teach our children. It is always instructive to ask a relativist how he raises his children.

  Since we agree that some things can be known and understood, our human endeavor is to extend the boundaries of sense and meaning; it is to shift phenomena one by one out of the nonsense heap and arrange them in ordered piles about us. If you argue that this endeavor yields only a human kind of sense, and that our interpretations yield only human meanings, not absolute meaning
s, you will be required to propose a definition of meaning that isnot , first and last, meaning for people.

  If, as even the early skeptic Arcesilaus granted, we can obtain at best probable knowledge, that in itself is no mean feat. We have come a long way on probable knowledge.

  For Einstein, the final mystery of the universe was that we are able to know it. (“One may say that the eternal mystery of the world is its comprehensibility.”) Our little human prose poems somehow connect to the mechanism. Einstein must have felt very queasy when the 1919 solar eclipse demonstrated that light does indeed bend in a gravitational field—a demonstration that corroborated general relativity and with it curved, inbounded space. He must have been dumbfounded, must have wanted to issue a telegraphed disclaimer to the universe:I WAS JUST KIDDING . For although we often feel pleased by the connections our minds make, we rarely, I think, feel they are absolutely true. (What if Chekhov went to heaven and God clapped him on the shoulder and said, “YOU WERE CORRECT”?)

  Consider a tangential phenomenon. Opinion polls collect opinions, which are human judgments and interpretations. Opinion polls do not ask us to mark statements “true” or “false,” but “agree” or “disagree.” It could be that poll-takers would have a hard time finding any educated adult who would mark any interpretative statement “true.” I would readily mark a number of statements “agree” which I would hesitate or refuse to mark “true”—even statements with which I may passionately concur. Do I agree with the Sermon on the Mount? (Yes.) Is it true? (?) Do I agree with this editorial about the presidential candidate? (Yes.) Is it true? (?)