Lucky are those writers who find their forms, and find them early enough to have time to say in them all that it is given them to say. I think of myself among the lucky in discovering the form known as "the familiar essay" in my late thirties.

When I began writing such essays, I have to report, I didn't think of them as familiar essays, or really as anything other than that journalist's all-purpose word, "pieces."

When I began writing such essays, I have to report, I didn't think of them as anything other than that journalist's all-purpose word, "pieces." I first wrote in the form in 1974, when I began editing the American Scholar, the quarterly journal of Phi Beta Kappa. As the magazine's editor, I had decided that it might be useful to have a member of the magazine's editorial board write an essay on a subject that happened to interest him at the moment. I wrote the first such essay; and when no one stepped in to write the second, I wrote it, too; and then I wrote the third and fourth -- and the hook was firmly in. I found I looked forward to writing one of these essays every three months; more than looked forward to it, I loved it. I had found my form.

Among the contemporary essayists I admired, George Orwell and James Baldwin ranked high -- James Baldwin before he had gone off the shallow end, became entirely politicized, and traded in his cool, high analytical style for a pot of message -- but never thought to imitate either. Among earlier precursors, I not merely admired but loved three other essayists: Max Beerbohm, who was a great stylist, an ironist of the highest power, and one of that small circle of writers I think of as laughing skeptics; H.L. Mencken, another laughing skeptic, for his high energy, his nice sense of absurdity, and Fieldsian (W. C. Fieldsian, that is) use of language; and, finally, jumping back a full century, William Hazlitt, who wrote with power and passion and conveyed a sense of the endless richness of life. Emerson I thought a great gasbag, and Charles Lamb I felt to be a bit too cute. I had not yet read Montaigne, the Shakespeare of the essay, at any rate not straight through and not with the care his one book, the masterpiece he called simply Essais, deserves.

I felt no "anxiety of influence," to use the critic Harold Bloom's overblown term, in connection with any of these writers. I took -- stole? -- from each what I found useful to the formation of my own style. Style is finally a way of seeing the world, and anyone who has any independence of mind is likely to see it in his or her own exceptional way. To write essays one needs, at a minimum, two items: a subject and a point of view. I found no dearth of subjects for essays, but my point of view, though largely formed at the outset, grew more complex, or so I would like to think. I would also like to think that the actual writing of essays contributed to this in a serious way.

Writing an essay has for me tended to be an act of self-discovery. By this I mean that I come to the writing of an essay not knowing what I think of the subject, except in an inchoate way; or if I believe I know what I think, the writing itself leads me into aspects of the subject whose importance or even existence I hadn't earlier recognized. An element of improvisation is entailed in essay writing, or at least it is for me. Like a jazz musician or a stand-up comic -- but with the added luxury of later revision -- the essayist is under an obligation to create the illusion of spontaneity. There are several ways of going about this, I suppose, but I myself have found that the safest is to try to be -- of all things -- truly spontaneous: to let 'er rip while writing.

Self-discovery is very different from self-therapy. Writing may have vast therapeutic effects, yet I, for one, find myself not much interested in them. Many major-league neurotics, as is well known, have been good, even great writers -- Proust and Kafka co-captain the first team here -- but neurosis is, somehow, better worked out through fiction and poetry. In the essay, sanity and all its lovely symptoms are what is wanted: balance, proportion, common sense, egotism held well under wraps. The essayist best sets up shop as an ordinary man or woman, only with more style, more lucidity, greater gifts of formulation.

A reader once wrote to me, after reading a collection of my essays, that they may be "familiar" but that they weren't all that "personal." I took this as a compliment. I keep the personally messy out of my essays: no side references to family members, no talk of sex or marriage, no heavy-breathing soulfulness. Quick to be bored, I fear boring in my essays, and there is no quicker way of doing so, in my view, than enforcing an unwanted intimacy on one's readers.

Few things please me more as a writer than to have an intelligent reader tell me that he finds that, in one or another of my essays, I have formulated his thoughts better than he has himself been able to do. Another pleasing compliment is to be told by a reader that, after finishing one of my books, he feels he knows me. But the people who tell me this, I usually discover, are, like myself, not people who wish to establish intimacy on the first, or even 15th, meeting with someone.

W.H. Auden said that he asked two questions of every poem he read: "The first is technical. `Here is a verbal contraption. How does it work?' The second is, in the broadest sense, moral: `What kind of guy inhabits this poem?' " The same questions can be asked of any essay, but the second question is the more lively one, and on its answer is likely to result the success of the essay. Too great depression, preciosity, strident complaint, and any other of a number of allied qualities mar, and are more likely to sink, an essay.

So can too stark confession. The one literary outing in which Hazlitt badly sputters is "Liber Amoris," his soppy account of his unrequited adoration of a lower-middle-class and quite sensible young woman named Sarah Walker, who recognized him as a fantast in romantic relationships. Like Montaigne, whom Hazlitt much admired because in him (as Hazlitt wrote) there is "no attempt at imposition or concealment, no juggling tricks or solemn mouthing, no laboured attempts at proving himself himself always in the right and every body else in the wrong," Hazlitt, great writer though he was in so many other respects, showed himself to be without gifts of introspection. But the larger problem was confession, which cannot be done gracefully. In literature as in life, the best advice is for confession to be done in three steps: be blunt, be brief, be gone.

Montaigne it was who invented the use of the "I" in essay writing -- invented and mastered it in one stroke. The trick, his writing demonstrates, is to write about yourself without the least suggestion of Copernican Complex -- that is, without conveying the feeling that the world revolves around you. ("But enough about me," runs an old joke about the egotism of writers. "What do you think of my new book?") The "I" ought never to be loom larger than the subject, but instead to serve as the means through which the subject is approached. In the course of doing so, if the planets are properly aligned and the gods kind, an interesting character may emerge along the way, and it might, mirabile dictu, belong to you, the lucky essayist.