Drafting

Anne Lamott, in Bird by Bird (1994), named what she called "shitty first drafts" and made a case that the willingness to produce them is not a failure of craft but a prerequisite of it: "All good writers write them. This is how they end up with good second drafts and terrific third drafts." The shitty first draft exists to give you something to revise. A blank page cannot be revised.

What this means in practice: the first draft is not for the reader. It is not even for the future writer who will revise it. It is for the present writer who needs to get through it. Quality standards do not apply. The first draft is the writer figuring out what happens and why it matters. That figuring-out process is inherently messy; demanding cleanliness from it is like demanding a clean workshop in the middle of a build.

The further implication: a writer who has never produced a shitty first draft has never tested the limits of their working method. The writers who declare that they revise as they go — polishing each sentence before proceeding to the next — are either working on short-form material where this is feasible, or they are producing very slowly and risking significant stall when they reach genuinely difficult material. The shitty-first-draft permission structure is most important precisely when the material is hardest.

Most writers have an internalized evaluating voice that assesses work as it’s being produced. In revision, this voice is essential — it’s what identifies what’s working and what isn’t. In drafting, it is actively harmful. It applies revision standards to first-draft work, finds the work lacking by those standards (it is), and generates paralysis. The work stops because the writer cannot produce anything that passes this internal quality filter.

The inner critic’s fundamental error is temporal: it evaluates drafting as if it were finished prose. These are different cognitive tasks. The critic is not wrong that the sentence isn’t good enough — it isn’t, by design. The error is applying the "good enough" standard at the wrong moment.

Here’s what’s interesting: the inner critic is not a pathological intrusion. It’s a necessary capacity applied at the wrong time. The same voice that ruins drafts is what makes revision productive. The goal is not to destroy the inner critic but to schedule it correctly.

Write fast. Speed doesn’t leave room for evaluation. If you’re writing quickly enough that your fingers are barely keeping up with your thoughts, the critic can’t get in front of the prose. Speed isn’t the goal; sufficient speed to outrun evaluation is. Many writers find they need to be moving fast enough to feel slightly out of control.

Turn off the screen. Literally. Write in the dark so you can’t see what you’ve produced. The impulse to evaluate is partly a response to seeing the words; remove them from view. This technique, sometimes called "dark drafting," produces first-draft material without giving the editorial mode anything to latch onto. The lack of visual feedback is a feature.

Set a timer. Commit to a defined period of writing without stopping, evaluating, or revising. The Pomodoro technique (25 minutes of focused work, 5-minute break) works for some writers; longer sessions of 90 minutes or more work for others. The goal is structured removal of the critic’s opportunity to interrupt. The timer creates a container: for this period, writing is the only permitted activity.

Give yourself explicit permission. State it to yourself: this draft is allowed to be bad. The badness is not final. This is the exploration, not the building. The psychological mechanism of explicit permission is underrated. Many writers know abstractly that first drafts are supposed to be rough and still write as if they’re not allowed to be rough. Explicit, verbal self-permission — said aloud if necessary — can override the implicit standard.

Write notes-forward. If you can’t write the scene itself, write notes about what the scene needs to accomplish, what the characters are feeling, what the key exchange of dialogue will be. Then convert the notes into prose. The notes aren’t the draft, which removes their pressure; but once the notes exist, converting them to prose is easier than writing the scene from a blank page.

Some writers draft scenes out of sequence — writing the scenes they’re most excited about first, regardless of where they fall in the story, and filling in connecting tissue later. This maintains momentum and generates strong material in the places where enthusiasm is highest. The risk is continuity problems and orphaned scenes that turn out not to fit when the surrounding material is drafted. Out-of-sequence drafting produces an irregular quilt; sequential drafting produces a continuous thread.

Sequential drafting (beginning to end) maintains continuity and means that each scene is written with full context from everything that came before. The risk is that writers get stuck in scenes they find difficult and stall rather than skipping ahead.

Neither method is correct universally. Writers who find themselves repeatedly avoiding a specific scene are usually avoiding it for a reason worth examining — the scene may be in the wrong place, may be doing something the story doesn’t need, or may require more preparation than what precedes it. See Writer’s Block as Signal for the diagnostic framework.

The hybrid: draft mostly sequentially, but skip problem scenes. Write a placeholder — [INSERT: confrontation scene between Marcus and his brother] — and keep going. Return to the skipped scene once you know what follows it. Knowing what comes after a difficult scene often clarifies what the scene itself needs to do; many scenes are difficult to write before the draft because the writer doesn’t yet know what consequences they need to produce.

Some writers intentionally write a very rough discovery draft — more like extended notes, stream of consciousness exploration, scene sketches — before writing what they consider the actual first draft. The discovery draft is private, disposable, exploratory. The first draft is the actual attempt.

The discovery draft has a specific purpose: it allows the writer to fail completely, learn from the failure, and bring that knowledge to the real draft. A discovery draft of a novel might be 30,000 words of rough scene sketches that gets thrown away; the real first draft then proceeds from a position of understanding that the discovery process provided. The 30,000 discarded words are not waste. They are preparation.

Others try to produce something closer to final quality on the first pass. This works better for writers who have done more planning — see Plotting vs Discovery Writing — and know their material well enough that the drafting is more execution than discovery. When the outline is detailed enough, some of the discovery work has already happened; the draft can aim higher.

Neither approach is correct. The question is whether you’re the kind of writer who discovers by outlining (in which case, a detailed outline followed by a reasonably clean first draft) or whether you discover by drafting (in which case, a rough discovery draft followed by a real first draft, or a rough first draft followed by heavy structural revision).

The most effective drafting practice is daily or near-daily writing. Not because inspiration arrives on a schedule — it doesn’t — but because sustained projects require sustained momentum. A project worked on every day for six months proceeds differently than a project worked on in intensive weekend sprints separated by two-week gaps. Daily practice keeps the story’s texture and characters present in working memory; long gaps force reconstruction of what was already understood.

Daily word count targets create accountability without requiring inspiration. 500 words a day produces 180,000 words in a year — two drafts of a standard novel. The target should be low enough to be achievable on difficult days. On easy days, you’ll exceed it. On difficult days, you’ll hit it. The writing produced on difficult days is not noticeably worse than the writing produced on easy days — not in first draft, where the standard is existence rather than quality.

The general rule: don’t revise heavily during drafting. Light notes are fine — a comment in the margin saying "expand this scene" or "find better word" or "check timeline" — but stopping to revise a chapter before moving forward kills momentum and focuses attention backward rather than forward.

The exception: if you realize you’ve made a structural error that will invalidate everything subsequent, address it. Drafting twenty more pages on a foundation you know is wrong is not forward movement — it’s investment in material you’ll have to discard. The key word is know: suspicion that something is wrong is not the same as knowing it is. When you suspect a problem, make a note and keep drafting. When you know a problem, assess whether it affects everything that follows before deciding to fix it now or flag it for revision.

Most of what feels like knowing-a-problem while drafting is actually suspecting-a-problem. The perspective to evaluate that suspicion accurately is usually available in revision, not in the middle of the draft.