Revise means, to most students, fix the sentences. Reread the draft, tighten the wording, cut the adverbs, replace the weak verb, check the commas. That is editing, and it is worth doing, but it is not revision. Revision is seeing again: deciding whether the paper is the right shape.
The shape of a paper is the order of its claims. And the order of a paper's claims is, in a document, almost impossible to change. Moving one section means cutting and pasting, and then every transition sentence is wrong, and the introduction promises a sequence that no longer happens, and after forty minutes the paper is worse and you put it back. So nobody changes the order, and "revision" stays at the sentence level, and the paper keeps the shape it had at midnight on the first draft.
Cards fix this. Not by making prose easier to move — by separating the order from the prose entirely, so that the order can be tried, several ways, in minutes, before any prose is committed to it.
This is the step in the method that does the most for the paper and gets skipped most often. It should not be skipped, and here is how to do it.
What an arrangement is
An arrangement is an order of the cards. That is all. The cards do not change when the order does; the same claim card with the same evidence can be first in one arrangement and last in another.
On the table, an arrangement is the row you laid out. Photograph it — with your phone, from above — before you touch it. That photograph is arrangement one. Now you can move things, and if you make it worse, the photograph tells you where everything was.
Keeping the old arrangement is what makes trying a new one free. Without it, every move is a risk; with it, every move is an experiment.
The five orders worth trying
There are more than five, but these five produce different papers from the same cards, and one of them is usually better than the one you started with.
1. The order you wrote them in. This is arrangement one, and it is the order in which you thought of the claims — which is the order the reading happened to present them, plus the order your doubts arrived. It is almost never the order a reader needs. Keep it as the baseline.
2. Strongest claim last. Order the claims by how much evidence sits under each, and build from the thinnest to the thickest. The paper gains momentum; the reader's doubt is answered by the time it forms. Many papers want this order and most drafts have the opposite — the strongest claim first, because it was the one the writer was surest of, and then a long decline.
3. Strongest claim first. The reverse, and right for a different kind of paper: when the thesis is surprising and the reader will not keep reading unless you show early that you can support it. Lead with the best card. The weaker claims then read as and also, which is fine, because the reader is already on your side.
4. The objection first. Begin with the strongest counterpoint card — the case against the thesis, stated fairly — and let the claims answer it in turn. This is the shape of most good arguments in print, and almost no student drafts, because it feels like giving the other side the first word. It is. That is why it works: the reader who was going to raise the objection has already heard it, from you.
5. Chronological, or its opposite. If the claims are about events, try them in the order the events happened — and then in reverse, starting from the outcome and working back to the cause. The reverse order is a detective story and reads like one. The forward order is a history. They are different papers.
Try each. Read the claim titles aloud in the new order — just the titles, thirty seconds. You will hear, immediately, whether the sequence builds or wanders. Most of the five will be worse than arrangement one. One will not.
What to listen for when you read the order aloud
Does each claim make the next one easier to believe? A good order is one where the reader arrives at claim four already half-convinced of it, because claims one to three prepared the ground. If claim four would land just as well anywhere, it is not connected to its neighbors, and the order is a list.
Where does the reader first want to object? Read as a skeptic. At the point where you would say but what about—, that is where the counterpoint card goes, and the claim that answers it should follow soon after. If the objection arises at claim two and is answered at claim six, the reader spends four claims not listening.
Where does the background go? Background is the material that is true and argues nothing — the context, the dates, the who-was-who. Drafts front-load it, because it feels like the beginning. Readers skip it, because nothing is at stake yet. Try putting each piece of background immediately before the first claim that needs it, and nowhere else. The introduction gets shorter and the paper starts sooner.
What can go? In a new order, some card will turn out to be doing nothing. It was there because it came after the card before it. Now it is between two cards it has nothing to say to. Take it out of the row and see whether anything is missing. Usually nothing is.
Compare, do not choose in the abstract
Keep two arrangements — the photograph and the new row — and compare them claim by claim. Where do they differ? Usually in two or three places. Those two or three places are the revision decisions, and they are now concrete: does the objection go first or third? is a question you can answer by reading both ways aloud. Is the paper well-organized? is not a question anyone can answer.
When the new arrangement wins, photograph it too. Do not throw away the old one. Next week you may want the objection back where it was, and the photograph knows.
Then, and only then, the prose
Once the order is chosen, the transitions can be written — one bridge sentence at each junction, saying why this claim follows from the last. Transitions written before the order is settled are the first casualty of moving anything, which is why the cut-and-paste revision fails. Transitions written after are nearly free.
The introduction is written last, again, because now you know what order you are introducing.
If you have already drafted in the old order, the draft is not wasted. The paragraphs under each claim are still the paragraphs; they move with the claim. What has to be rewritten is the sentence at the top of each, and the introduction. That is an evening, not a week.
What this does to a paper
The paper you get from arrangement one is the paper the reading gave you. The paper you get from trying five orders and keeping the best is the paper you made from the reading, and a reader can tell the difference, though they will not be able to say what it is. What it is, is that the order is doing work.
Doing this in software
Software that works from cards will usually let you keep several orders of the same cards at once — an arrangement is its own object, with a name — and switch between them, or see two side by side with the differences marked. Moving a card in one order does not move it in another, and the draft is compiled from whichever order you choose. This is the one thing software does that the table cannot: the photograph is live, and the comparison is exact. Nothing about the method changes. You still have to read each order aloud and decide.
Everything here works with a pack of paper cards. Leyden & Beam does the same steps on screen, keeps every order you try, and never writes a word for you.
Start your 14-day trial