I have been in the business of rejection since I was 21 years old. Before I graduated from college,I commuted one day a week from New Haven to New York, where I worked as an unpaid intern for the economics and Bibles editor at Oxford University Press. Mostly what I did was type rejection letters.

It was a good, and typical, introduction to the world of publishing.

After graduation, resisting the siren call of working on a dude ranch in Wyoming, I went to work full time for another Oxford editor. I typed more letters spurning hopeful authors. I had to learn to decipher my boss's scrawl at the top of a manuscript and I had to learn the code.

"Reject!" meant just send the form letter. You would thank those writers for submitting, say that "after careful consideration" their project wasn't appropriate for our list, and then thank them again for thinking of us and wish them good luck.

"Nice reject!" meant give the same basic message but tart it up a little, to soften the blow: That meant adding phrases like "while your writing is a pleasure to read …" or, "As much as I would like to be able to publish your fine book …"

There were always writers who made the work of rejection easy for us. They were the ones who hadn't done their research. They got the "We don't publish (whatever)" letter. Fill in the genre blank: memoir, Festschrift, textbooks, fiction. Sometimes, if she was feeling generous, my boss would list (in tiny chicken scratch) a number of other publishers for the author to try.

And there were real rejection letters, where she took time to explain to authors why she wasn't going to be making an offer to publish their book. Either the argument was too polemical or unfair, the research seemed synthetic, or, most often, she couldn't imagine a market for the book. Rarely was a project turned down because the writing was bad. (Plenty of badly written books were published.)

When I began working for an executive editor, I entered a fiefdom and was no longer responsible for typing since my boss had a secretary. Now my job was to read manuscripts and write reports giving him reasons to reject. That was a lot harder.

First you have to figure out what the project is, and then try to decide whether it's worth publishing. Sometimes the author wrote a good introduction that set out clearly the purpose of the book. As a reader, you could crib from that. Other times, you might slog through hundreds of pages and still not know what the argument was.

Most editors don't need to read more than the first 50 pages before they know if they are going to publish a manuscript. Those first 50 pages, needless to say, should be stellar.

Once I felt that I understood what the author was trying to do, and where the work fit into the scholarly terrain, deciding whether or not to recommend publication was an even greater burden. After all, these were manuscripts that had been toiled over for years by professors at the top of their fields who taught in some of the best universities in the country. Was this breaking new ground or was it just filling another post hole? Would the book get trade review attention? Would it "back-list" and end up selling for years in the course adoption market? Would the author's second book be the one we really wanted to publish? I was a fresh-out-of-college English major. What did I know?

One of the things I knew was how the manuscript had come in. Had it been recommended by another Oxford author? Had it scored a dissertation prize? Had my boss, the executive editor, specifically asked to see it?

Although most editors, when approached at a conference, will agree to look at anything ("Your shopping list? Sounds fabulous! Send it to me!") there's a difference between seeking out work and saying "yes" to get rid of a persistent hopeful. The provenance of a manuscript, I'm sorry to say, matters. It usually doesn't hurt to be unconnected, but it sure can help to have powerful friends lobbying for you.

When I saw copies of the executive editor's rejection letters I was shocked. Even when he was the only one who had read a manuscript, the editor almost always said that he had "discussed it with his colleagues." Or, if I had read the manuscript first, he would say that he had "brought it up in an editorial meeting" where, although he would have liked to pursue the project, he couldn't get the support of others.

It was a sympathetic move. No one likes to be the heavy, and he understood that authors want to know that their work has been given serious consideration. It was a lesson I took with me when I became an editor.

At most presses, especially good ones that get lots of submissions, writing rejection letters is a mainstay of an editor's job. It's never fun. Often, even getting to the point where you are certain you are going to reject a manuscript is a painful process. You want to say yes. You read to be able to say yes. It's rare that you look at something and, think, "No way." Frequently, even though some part of you knows you are never going to publish it, you think, "Maybe." So you let it sit for a while, let it mellow like a just-opened bottle of wine.

Those manuscripts can hang around long enough to turn into vinegar. Then you see them, look at the date they were logged in, and panic. Laden with guilt about keeping the manuscript for too long, the nice rejection letter you would have written if you weren't so pressed for time gets turned into a formulaic "No thanks."

One of the things you generally don't want to do as an editor is enter into a dialogue with an author you're not going to publish. There is a temptation to give advice. It's what editors do. If you cut this, move that, go back to the archives, then it might work.

But giving that kind of advice can be misleading. You don't want to offer too much help if you don't want to hear from the author again. Most editors are polite. They hedge on their positions, saying things like, "I'm sure that other publishers will come to a different decision," but if they don't offer suggestions, authors shouldn't write back — or worse, call — to ask for specifics.

Occasionally, an intemperate author will fire back a response. Generally that is not a good idea. Lick your wounds for a while, and then figure out where else you can send the manuscript. Just because an editor rejected your first book doesn't mean that she wouldn't be interested in your next. Don't burn your bridges. Unless you're Norman Maclean. An editor at Knopf wrote to inquire about his second book. The first, A River Runs Through It, was a phenomenal success after being turned down by every house in New York and eventually picked up by the University of Chicago Press.

Maclean replied, "The dream of every rejected author must be to see, like sugar plums dancing in his head, please-can't-we-see-your-next-manuscript letters standing in piles on his desk, all coming from publishing companies that rejected his previous manuscript, especially from the more pompous of the fatted cows grazing contentedly in the publishing field." He concludes his published letter, "If the situation ever arose when Alfred A. Knopf was the only publishing house remaining in the world and I was the sole remaining author, that would mark the end of the world of books."

You can get away with that if you're a gazillion years old, you're already being published by a loyal press, or you're from Montana.

Surely that Knopf editor was a little stung, but most of us have had the experience of seeing a book we've turned down go on to great critical or commercial success. Most don't lose much sleep over it; generally we know exactly why we chose not to be the one to offer the author a contract and would most likely make the same decision again. Sometimes, however, we make mistakes. Turning down your book, for example, could have been a mistake.

After I finished rejecting authors, I went into the business of rejecting applicants to Duke University. Reading a few pages of their application materials was a breeze after slogging through hundreds of pages of dense academic prose. But the rejecting part was similar. Not a lot of fun, and often the cause of much mental back-and-forth (although in admissions, a committee had the final say.)

But I had learned an important lesson from those years in publishing.

Given the high quality of many submissions, the decision to reject most of them is as much a matter of taste as anything, and there are plenty of other editors (and admissions officers) with tastes different from my own. I knew that what I was saying to an author wasn't usually, "I don't think you should be published." It was simpler than that. It was, "You will probably be published, but I'm not going to be the one to do it." When I wished them success with their manuscripts, I generally meant it.

Rachel Toor is an assistant professor of creative writing at the Inland Northwest Center for Writers, in Spokane, the M.F.A. program of Eastern Washington University. Her Web site is www.racheltoor.com and she welcomes comments and questions at careers@chronicle.com. For an archive of her previous columns, see http://chronicle.com/jobs/news/archives/columns/page_proof