Home » David Farland, Fiction Fundamentals, Writing Advice » When to Stop Polishing a Manuscript

When to Stop Polishing a Manuscript

10 February 2015

From author Dave Farland:

Many new writers don’t know when to stop polishing a manuscript and move on to the next. Part of the reason for that might have to do with Ernest Hemingway.

Many years ago, a writer asked Hemingway, “How many times should I rewrite a manuscript?” Now, Hemingway hated dumb questions, so he answered “Oh, at least 60.”

He loved doing that to writers. On one occasion, a writer asked him what kind of chair he preferred to sit in, as if perhaps the brand of furniture that an author had planted his butt on might somehow confer literary genius.

Hemingway answered, “I don’t sit when I write, I stand.” And a generation or writers began to write standing up. The problem with that is that you can go to any one of Hemingway’s old homes or offices, and see the chairs that he sat on.

On another occasion, a writer asked him how long she should wait between drafts when revising, so that she would be able to look at her story “cold.” He suggested that it should be two years.

Think about it. If Hemingway did sixty drafts of a novel and waited two years between each draft, he would have never finished a single book. Don’t listen to bad advice, even when it comes from a genius.

Back when I first began writing, I used an old typewriter. I didn’t like it. I had to really bang the keys hard, it was noisy, certain keys didn’t work well, and the type was uneven. Because of this, doing rewrites was difficult. I’d type out a draft, make extensive corrections on the page with a pencil, and then try to type out a perfectly clean copy.

Using that system, it would have been foolish to repeat the process sixty times. Because of this, in the 1920s and 30s, a professional writer would typically try to learn to write a finished copy in a single draft. It was simpler to write out a nice outline in longhand, and then thoughtfully type out one clean draft, than to retype a piece over and over.

. . . .

Of course with the development of computers, revising became quite easy. My first computer would allow me to put only 2 pages of text on a disk, but by the late 1980s I was able to get first a whole chapter, and then with the addition of a hard drive, an entire novel in a single file. It wasn’t until then that rewriting became so easy that it became problematic.

You see, as an editor I’m looking for stories that have some originality, that carry an author’s own voice, his odd quirks. But when a new writer begins showing a manuscript around to members of her workshop and polishing it further and further, eventually the author tends to lose her own distinct voice. The result is, that the story can become less interesting to me as an editor with every draft.

So the question is, how many revisions does a novel or short story really need?

. . . .

As I rewrite, I try to avoid changing both the voices of my characters and my own narrative voice. Rather than polishing away the differences between voices, I think it’s better to look for ways to heighten the unique characters in the tale.

In fact, on one of my last rewrites, I do what I call a “voice edit,” where I go through key characters person by person to make sure that their voices are consistent.

Link to the rest at David Farland and thanks to Cora for the tip.

Here’s a link to Dave Farland’s books.

David Farland, Fiction Fundamentals, Writing Advice

26 Comments to “When to Stop Polishing a Manuscript”

  1. Alcoholics, depressives and men who kill themselves are not who I choose as my role models. When you get all three in one then it’s time to not listen to that person, or whatever “lessons” they may have.

    • You know, I was going to ignore this, because it’s your usual Dougie Downer crap, but to disrespect another person because they were mentally ill just ticked me off. Hemingway’s issues weren’t personal failings, he was sick.

      I have people in my family going back generations who suffered from depression and alcoholism, and it did not make them less of a human being.

      No one suggested you model your life by following in his footsteps, but one might think twice about dismissing advice from a writer who produced such classic work in spite of his health.

      • I personally think it’s a matter of background.

        Meaning… Let’s take your last paragraph (“No one suggested… but one might think twice…”). If you take his advice to the letter, then… 60 revisions, standing up, every two years.

        When the person is sort of mainstream (sort of), those little differences are relatively simple to get (and even so…). When the person is off-chart (genius, depressive, any and all…) it can become very very difficult to see what’s good advice and what’s dangerous and one of the reasons he’s having a bad time. Or a quirk, a patch, that works for him precisely because he’s a genius.

        So, in general, I would pass and not listen to his recommendations. If I had the background or the knowledge to properly check those? Then, sure. But it can be a ton of work. In the time I manage to understand a twentieth of hs mindset I can understand twenty writers. Unless he’s family, not worth it.

        Take care.

    • Agreed that Hemingway isn’t a role model for life — even if you could imitate mental illness, who would want to? — but he definitely is a role model for art. I appreciate his books much more as I get older.

  2. Many moons ago I was talking to a writer I admired not just for her stories, but because she was so prolific. I asked her how many drafts she wrote. She said two. Then she said she went through a long period where she couldn’t finish anything and realized she was getting stuck on draft 3 or 4. She figured out that it was because on draft number 3 she lost faith in her story and her ability. So she wrote the first draft as fast as she could and if she still liked it, she cleaned it up in draft #2 and that was it. She sent it off to her editor and went to work on something else. If the editor wanted revisions, she did the revisions, but other than that, the story was done.

    This struck me as sensible. It certainly worked for her.

    • I like this a lot.

    • This is what I do. After revising once it’s time for a professional to take a look at it. It is a far better use of your time than revising the same manuscript over and over again.

      There are many ways to become more skilled at something. Working with, and learning from, an editor is far more efficient than going over the same manuscript yourself a dozen times until you hate it.

      • And after five or six books, you don’t even need a professional to help at the back end any more. Just a proofreader.

    • This is me. The third draft usually kills it. If the story survives a third draft, it is not recognizable as draft 1 at all.

    • Very sensible, and may finally explain to my satisfaction why I can’t bear to do any more than two drafts.

  3. This really resonates with me, especially since I just submitted a short 9K piece this morning. My usual MO is to write a terrible first draft just to have something to work with, improve it greatly in the second draft, then do multiple refining passes until I am absolutely sick of looking at it. By then I’ve lost all perspective on whether it’s good, bad, or even coherent.

    This time it took fewer passes and I felt like I had a pretty good thing up until the end. Hopefully this means I’m improving as a writer. This morning I hit my limit and decided it was time to send it off. We’ll see what the editor has to say.

  4. I’m a big proponent of fewer ‘drafts’. Rewriting several times works for some folks and I am not denying that, but I agree with the writer Jaye is referring to. If I go over something too many times I get so sick of it I don’t care anymore and that leads to more problems than I started with.

    I am not a fast writer, but I only do 2 drafts…

    I write 1 draft really, but I review the previous day’s work everyday and fix all the dumb stuff I did. It helps me to get a running start again to reread what I wrote too, so it’s killing 2 birds with one stone.

    When I’m done, I run through looking for the usual offenders like there vs. they’re and checking everyone’s eye color to be sure it didn’t change mid story. When that’s done it does to my editor.

    If he is ok with it, it’s done. Of course he’s never been ok with it… 🙂 I look at his suggestions, fix the stuff I agree with and live with the rest.

    If he ever reads something and comes back with ‘this is a mess and needs tons of work’, chances are (if I agree) I’m better off tossing it and working on something else.

    Crap is crap and you can’t polish a t***.

    • I keep a running list of “things to change” for the second draft. Widows and orphans. Setups and payoffs.

      Second draft: I go back and change all that.

      Then I let it sit for a week or two and do a final edit for dialogue and typos.

  5. I’m not a “one and done” guy, but I’ve decided that a first draft and two full revision passes are enough for me to catch most mistakes and craft my prose to the point where I feel that it’s fully in my voice.

    For my first novel, I actually rewrote each *chapter* twice before going onto the next one, so I feel like I’ve come a long way.

    That said, I did sell that book to Pyr, so from my POV there’s value in spending more time crafting when you’re starting out.

  6. I have always wondered why one needs discrete drafts that get their own names. There are lots of ways of producing a book.

    If it works, wonderful. But is there some reason to presume it is necessary?

    • It’s the old secret handshake myth. X drafts = Chosen for SUCCE$$!!

      I’ll have to steal Hemingway’s line. I’ve told newbies I write 25 drafts. The one story I’ve sold to a publisher was a single draft; two if you count the editing pass.

    • Even for pre-computer days, numbered drafts seems an odd concept. Aren’t there always scenes which flow almost perfectly onto the page, while others require multiple passes to wrestle them into shape?

      Maybe the “numbered drafts” concept comes from writing or typing a new complete clean manuscript from the previous copy, which has many scenes unchanged and many others polished or re-shaped or re-purposed multiple times. So that wouldn’t be a “second draft” (for instance), rather a second clean copy.

  7. Mr. Farland either has a very bad memory or he used several very inefficient word processors back when he started with computers.

    Single-sided 3.5-inch disks from 1983 could hold around 60,000 words. Double-sided disks, starting in 1984, could easily hold a large novel of 125,000 words along with its associated, albeit simple, formatting instructions to print those words out onto 500 typewritten pages. In 1986 came the 1.44 MB version, the most popular, that could hold 1,000 pages or 250,000 words. The standard for the various machines (whether general purpose personal computers or dedicated word processors) was to have two disk drives: one that held both the operating system and the word processing program, the second for the data file(s).

    It never required waiting until the late ’80’s for rewriting to become “so easy that it became problematic.” A hard drive was never needed for writing—and rewriting—a book. They were merely more convenient for holding tons of files.

    • Isn’t it funny how technology dictates art? It’s doubtful that Dark Side of the Moon could’ve been made today because there are just too many damn choices to make in the studio. Limitations really boost creativity.

      U2 also has complained about the same thing. ProTools plants doubt.

    • God Bless the real floppy disk.

      AppleWriter was available in 1980. It worked fine. IBM followed with a WP when it introduced the PC in 1981.

    • Stop! I’m having flashbacks of gold letters on a blue background…

      Word perfect? I can’t remember if it was that or MS Works.

    • Single-sided 3.5-inch disks from 1983 could hold around 60,000 words.

      As Terrence O’Brien implies, those weren’t real floppy disks. The 3.5-inch disk actually came rather late in the evolution of the floppy, and had a much higher data density than the 5.25-inch disks that came before it.

      I still have an ancient TRS-80 Model I kicking around. The 5.25-inch floppies on that machine stored a grand total of about 90 kilobytes per disk, minus directory files and other overhead. If you only had one drive, you had to have the entire operating system on every disk (since the OS had to be mounted at all times in order to run; it would not fit in RAM). You also had to have your word processor on the same disk. A blank 90K disk could store close to 15,000 words – but if you only had one drive, you could only store a few pages on a system disk.

      Other 8-bit computers had similar limitations. It may be that this is what Mr. Farland is recalling. In any case, your incredulity is purely a function of your time horizon: you don’t remember back far enough.

Sorry, the comment form is closed at this time.