Francie Prose dissects writing through by parts: Words, Sentences, Paragraphs, Narration, Character, Dialogue, Details, and Gesture. She devotes a whole chapter to Chekhov, which I found so interesting that I bought The Lady with the Dog and Other Stories, which is also available from Project Gutenberg.

Prose has taught writing for many years, and also has written many books. She presents what could be very dry material, extremely well. It does read like a textbook so it took me several months to finish it. There's a lot of material here, and it's extremely interesting to me.

I've always thought of writing and programming to be very similar, and this book demonstrates to me that they are. What I also learned is that I could never be a writer. I don't care about how I write for it to matter. My goal in writing is to convey what I think in a way that suits me.

Programming and writing are different at the detailed, technical level. Programmers are fortunate to have pretty simply rules for what will produce good code at the micro-level. They also have the ability to write a test, which allows them to massage their creations (again at the micro-level) to increase the quality of their designs. Writers do not have such a luxury. It's hard enough to solve the problem at the macro-level (plot, themes, and so on). Authors have no way to get feedback without asking somebody.

I don't think I could work that way. Changing what I do based upon another person's opinion is tough for me. I want concrete feedback based on data gathered by tests -- even when they are usability surveys. You can't do this with a book, except perhaps textbooks, which don't have to have a plot. Great authors have to learn to be great very quickly. Great programmers can learn how to be great coders over decades. Perhaps that's too harsh but that's the way it seems to me.

Fiction is also quite foreign to me. I'm too concrete. I like reading fiction, but I mostly read non-fiction, like this book, and even the fiction I read is often on the lighter side. Prose demonstrates what it takes to make credible fiction using texts that, for the most part, I've never read before. This, too, was fascinating. She also picked her authors from the past so that she could show what makes writing timelessness.

I recommend this book even for non-writers. I often underestimate what it takes to write anything. Prose reminded me that writing well is extremely hardwork.

[p4] What writers know is that, ultimately, we learn to write by practice, hard work, by repeated trial and error, success and failure, and from the books we admire. And so the book that follows represents an effort to recall my own education as a novelist and to help the passionate reader and would-be writer understand how a writer reads.

[p10] AFTER my novels began to be published, I started to teach, taking a succession of jobs as a visiting writer at a series of colleges and universities. Usually, I would teach one creative writing workshop each semester, together with a literature class entitled something like "The Modern Short Story" -- a course designed for undergraduates who weren't planning to major in literature or go on to graduate school and so would not be damaged by my inability to teach literary theory. Alternately, I would conduct a reading seminar for MFA students who wanted to be writers rather than scholars, which meant that it was all right for us to fritter away our time talking about books rather than politics or ideas.

[p61] Finally, before we leave the subject of sentences, let's return once more to Hemingway, and to the passage from his memoir of his youth in Paris, A Moveable Feast, in which he describes his working method and which subsequent generations of writers have taken as a form of implicit literary advice:

Sometimes when I was starting a new story and I could not get it going ... I would stand and look out over the roofs of Paris and think, "Do not worry. You have always written before and you will write nom All you have to do is write one true sentence. Write the truest sentence that you knom" So finally I would write one true sentence, and then go on from there. It was easy then because there was always one true sentence that I knew or had seen or had heard someone say. If I started to write elaborately, or like someone introducing or presenting something, I found that I could cut that scrollwork or ornament out and throw it away and start with the first true simple declarative sentence I had written.

For years, I've heard this passage about the one true sentence cited as a sort of credo. And I've nodded my head, not wanting to admit that I honestly had no idea what in the world Hemingway was talking about. What is a "true" sentence in this context-that is, the context of fiction? What makes Hemingway's advice so hard to follow is that he never quite explains what "true" means.

[p62] Perhaps it's wisest to assume that Hemingway, like countless others, was simply confusing truth with beauty. Possibly what he really meant was a beautiful sentence -- a concept that, as we have seen, is almost as hard to define as the one true sentence.

[p64] I asked a friend, a poet who also writes essays and memoirs, if he had any thoughts about the paragraph. He said he thought of the paragraph as a form, like a poetic form, perhaps a bit like a stanza. Then he added something that I myself have noticed. He said that when he was writing an essay, there came a point at which he knew what his first few paragraphs would be. That was the point at which the essay organized itself in his mind and fell, as if with a series of clicks, more or less into place.

But how, precisely, can we tell when these clicks are supposed to occur? Once again, it seems easier to learn by example than by abstraction, by reading Babel's fiction to see how his ideas about electrical storms and rhythm operate in practice.

[p95] He touches Sheilah's hand. The children have their aunt now, and he and Sheilah have each other. Everything works out, somehow or other. Let Agnes have the start of the day. Let Agnes think it was invented for her. Who wants to be alone in the universei' No, begin at the beginning: Peter lost Agnes. Agnes says to herself somewhere, Peter is lost.

This is hardly what one would calling choosing a point of view and sticking to it, which is yet more proof that any set of "rules" offers only the loosest of guidelines. Even so, regardless of how blithely and confidently it breaks the rules and expands the third-person form, the Gallant story is in the "third person." That is, it employs third-person pronouns -- he did this, she said that -- rather than the "I" form.

[p250] All of which brings up yet another reason to read. Literature is an endless source of courage and confirmation. The reader and beginning writer can count on being heartened by all the brave and original works that have been written without the slightest regard for how strange or risky they were, or for what the writer's mother might have thought when she read them.

Often, when I teach, I like to draw up a reading list composed entirely of masterpieces that, for one reason or another, might have been thoroughly trashed by the more conventional newspaper review or the writing workshop. Much of the work I've mentioned so far in this book might run afoul of some of today's amateur or professional critics. And actually, many things that we ourselves consider indispensable for a work of fiction may turn out, the more we read, to be superfluous. If the culture sets up a series of rules that the writer is instructed to observe, reading will show you how these rules have been ignored in the past, and the happy outcome. So let me repeat, once more: literature not only breaks the rules, but makes us realize that there are none.

[p264] It does seem like quite a price to pay for the freedom to sit in your room and think about metaphors and paragraph breaks. But Babel's crime and his punishment had something to do with the fact that art implies a kind of freedom, the freedom of choice, of possibility, of the individual imagination. Which is why dictators-and big corporations-tend not to like art and [p265] artists, except those of a highly predictable and malleable sort. If art demanded Babel's life, we can certainly handle whatever inconvenience or effort it seems to require from us.

Isaac Bashevis Singer once said, "If Tolstoy lived across the street, I wouldn't go meet him." And you know what he means. The work is the work, what exists on the page is what matters, and we need not have tea with the writer in order to understand and love the writing. But whether or not we can understand and love Tolstoy's work without meeting him, there is much that is heartening about his life. To read his biography is to watch a writer destroying the printer's plates of Anna Karenina because he wanted to make some last-minute revisions, and one who had started out imagining the novel as something more in the manner of a sermon against an adulterous woman. The less admirable parts of his biography -- the long, nightmarish marriage, the selfish ideologue he became, the cruel (to his family) way he chose to die-also have a strangely liberating aspect: how orderly and thoughtful our own lives seem, by comparison.

[p268] Recently, a friend told me that her fears and concerns about the current state of the world were making it hard for her to write. I e-mailed her a copy of Herbert's poem and suggested it might help her problem, perhaps just a little.

A few hours later, she called back. "But that is the problem," she said. "He's talking about a rose. But how do you know if you've created a rose -- or just a weed?"

She's right. That is the problem. So one final reason for reading is to confront this problem of roses versus weeds in the company of geniuses, and with the pleasure of looking at the roses that have actually been produced, against all odds. If we want to write, it makes sense to read -- and to read like a writer. If we wanted to grow roses, we would want to visit rose gardens and try to see them the way that a rose gardener would.