During the dozen years I worked in scholarly publishing, I wrote scads of in-house manuscript reports, commissioned hundreds of outside readers, and read thousands of published book reviews. Although I would have denied it at the time, I never appreciated how hard it was to write those reviews.

Now I'm on the other side of the desk. A while back, I was asked by an old publishing colleague at a university press to read a manuscript for him. I had just written a book review for a major newspaper, which was about the hardest work I've ever done. Forget about getting the thinking right — the review itself had to be crafted in a smart, fair, and writerly way. After all, more than a million people would be seeing it.

With the manuscript report, I knew my readership was essentially one person. I was doing it for the author. I had no personal connection to her, so it was easy to slip into moments of pique while reading something that was somewhat less than perfect.

But as I was reading the manuscript and, later, writing the critique, I kept one thought in mind: If I didn't offer a detailed and substantive assessment, no one would. The author's editor might give her some general comments, but doing more than that, I knew, was not his job.

It was a lesson that had been drummed into me one night while I was working late in my office at Oxford University Press. The then-president was prowling the halls. He asked me what I was doing. "Editing a manuscript," I said, full of myself and my ability, as a twentysomething, to help full professors with their work. He got all stern and brow-furrowy and said, "Your job is to acquire books, not to edit them."

University presses have largely abdicated the job of substantive editing to outside readers, who write reports evaluating the quality of the work and offering both major and minor editing suggestions. In some ways, it's a remarkable system. Editors develop a stable of readers they know they can trust, who will be fair, rigorous, and prompt with their reports. For pennies, and because we are members of a community of scholars, we do editing work that is hard and, for the most part, invisible.

Timeliness is an important issue. Ionce sent a manuscript by a junior scholar to a well-established academic. Too many months later, I asked Professor Big when I could expect his report. He put me off with arrogance and false promises. For a variety of complicated reasons, instead of sending the manuscript to someone else, I had to wait for the academic star to pass judgment. After about 10 months, he sent in the review. It was positive but cursory. The author's tenure case was, at that point, shaky.

So here's an unabashed plea to readers from a recovering editor: Please don't agree to take on a manuscript if you are not going to be able to do it in the requested time. Be realistic.

To those of you — yes, you — with other people's manuscripts sitting on your desk, get to them soon, or be honest and tell the press that you can't do it. The months can pass quickly for you, but they drag by for the authors, especially those whose tenure and promotion may, in part, be dependent on their books getting good reports from you and other outside readers.

Many people — mostly editors at university presses — are concerned about how much departments rely on those reports in evaluating a scholar's career. They view that process as the intellectual outsourcing of promotion and tenure. Lindsay Waters, executive editor for the humanities at Harvard University Press, published a broadside about this in 2004, Enemies of Promise: Publishing, Perishing, and the Eclipse of Scholarship. His work was cited last year by the Modern Language Association in its own report on the issue: "As [Waters] points out, this process of external review serves to obviate the process of internal review: Departmental committees behave as if they cannot or should not determine the value of their junior colleagues' work unless university presses deemed sufficiently prestigious have determined the value of that scholarship for them."

The external-review process itself poses problems. It's a lot creakier than it looks. As a wee assistant, I was often charged with finding outside readers for a manuscript. Sometimes, despite my fumbling, I got excellent and appropriate ones. Other times the person who read the manuscript was recommended by someone who was recommended by someone else who couldn't do it. Reports from such third-choice scholars turned into the official, anonymous readers' reports collected by Oxford University Press that often went into an academic's tenure file.

Readers can be generous with both praise and criticism. A handful of times, they asked that I share their identity with the author and pass along an offer to be in touch directly to discuss the manuscript. A few readers were terrified that their cover would be blown and the authors would find out who had done them in.

More often than I would have liked, a reader took an author to task for neglecting to include references to the (anonymous) reader's work.

Presses generally send along a list of questions to be answered. Some readers will follow that list to the letter. But others know the task of evaluation is not that easy. It's not just a matter of "Does this work make a significant contribution to the field?" but, "What kind of a contribution does it make? To what will it be compared?"

That is, as I said, hard work. And the pay is peanuts. I had to keep reminding myself, as I read the manuscript for my editor friend, that it wasn't my job to fix the problems but simply to point them out. It didn't seem simple.

Agreeing to read a manuscript for a press is a weighty responsibility. I love that scholars are generally, as a community, generous to their colleagues. We do, after all, tend to write mostly for one another. Authors are often asked to suggest appropriate readers. To be able to have some giant in your field whose work you know and respect  — but whom you would not have the gumption to approach yourself — comment on and evaluate your work before publication is a gift. It's also an honor to be asked to write a review, and a privilege to get a jump on the slow publication process and read a book in your field in manuscript form.

The system tends to work well as long as both sides, publishers and readers, take their jobs seriously. Editors should know better than to send out manuscripts that are hopeless. Readers must keep in mind, among all the other important issues, that every extra week they take is a week out of someone's life. Departments need to sort out how they use the university-press review process for their own purposes.

Yet in spite of the problems with peer review, the many times that it works well should make all of us feel good about what we do, whether we are authors, publishers, or readers.

Rachel Toor is an assistant professor of creative writing at Eastern Washington University. For an archive of her previous columns, see http://chronicle.com/jobs/news/archives/columns/page_proof