Structure, shape, and where a book begins
Structure is the part of a manuscript its author can least easily see, because the author already knows what is coming on the next page. Very little about it is settled by rule, and the little that is settled comes from proposal guidelines and style manuals rather than from craft advice. These questions gather what can be said about shape from published sources, and what an editor can report without pretending to know how the book should have been written.
Should I outline first or find the shape by writing?
Nonfiction settles this by the calendar rather than by preference. A proposal is judged before the book exists, and both Princeton University Press and the University of California Press ask for a chapter outline as part of it. Princeton wants planned chapter titles with a paragraph describing the contents of each.1 The University of California Press wants a detailed outline showing what each chapter contributes to the logic of the argument.2 An author selling nonfiction on proposal has produced an outline whether or not the method suited them.
Fiction carries no such requirement, and nothing published settles which method produces better novels. Claims in both directions are common and none of them rests on evidence we could check.
What can be checked is the finished manuscript. An editor can reconstruct the outline a book actually has, as opposed to the one it was meant to have, and hand it back. Authors who drafted without a plan often find that the reconstruction is the first clear view of the book.
Does the book have parts, and do the parts earn their titles?
Parts are a real structural level with formatting rules attached. The Chicago Manual of Style treats the format of chapter titles and the titles of other parts as a matter of manuscript preparation, alongside the format of subheads.3 A book with parts is making a claim that its chapters fall into groups.
A part title earns its place when the chapters underneath it share something a reader can name after finishing them. Where the only thing the chapters share is their position in the sequence, the division is decorative, and numbering them Part One and Part Two makes that visible rather than hiding it.
The checkable half of this is consistency. Part titles, chapter titles, running heads, and the table of contents have to agree with one another, and they frequently do not after a book has been reorganized. That is copy editing work rather than a judgment about the shape.
Is the timeline linear, and if not, can a reader follow it?
A nonlinear timeline creates a consistency problem before it creates an artistic one. Copy editing covers checking mechanical consistency and factual accuracy, and it produces the style sheet on which a book's own facts are recorded.4 Ages, dates, seasons, elapsed time, and the order of events all belong there.
What breaks in a reordered book is usually the record rather than the reader's patience. A character is thirty-four in one chapter and thirty-one in a chapter set two years later, or a Tuesday becomes a Thursday. The error survives because the author reads the chapters in writing order rather than in reading order.
Whether the order can be followed is a separate question. An editor answers it as a reader, by naming the point at which a scene could no longer be placed in time. Some books want that uncertainty and keep it, which is why the finding is reported rather than corrected.
Where does the book actually begin, as opposed to where I started writing?
Reordering and cutting are both part of structural editing, and applied to an opening they mean proposing a later chapter as the first one.4 The finding is common enough to expect.
A structural report will often locate the start at the point where the situation changes. What precedes it is then treated as material the author needed in order to write the book, rather than material a reader needs in order to read it. That earlier material is rarely wasted. It usually returns in pieces, later, where a reader has a reason to want it.
The decision stays with the author, and there are books whose slow opening is the point. What an editor owes in that case is the reason for the finding rather than the finding alone, so that the author can weigh it against what the book is trying to do.
Do I have a middle, or a beginning followed by an ending?
In nonfiction this shows up in the chapter list. The University of California Press asks what each chapter contributes to the logic of the argument, and a list where four chapters contribute the same thing has a middle made of repetition rather than progression.2
In fiction the same fault appears as scenes that restate a situation instead of changing it. The characters are in the same position at the end of the scene as at the start, and the scene exists to keep the reader company between the two events that matter.
The test an editor applies is mechanical enough to be fair. Each scene or chapter is summarized in one line describing what changed. Where a run of lines describes nothing changing, the middle is the problem, and the run is where it lives.
Is the ending earned?
Whether an ending is earned is a reader's judgment, and it is the one an author cannot make about their own book. Nobody publishes a test for it. An editor claiming a rule here is describing a preference and calling it a standard.
What can be described without pretending is the relationship between the ending and what came before. An editor can list the elements the book set up and say which the ending uses, which it drops, and which it resolves by introducing something new. That list is evidence, and an author can disagree with the conclusion drawn from it while still finding the list useful.
Endings also fail for reasons that are not about structure at all, including pace, a shift in voice, or a last chapter written months after the rest. Those are worth separating out, because each has a different remedy.
Is this scene necessary?
The question works better in the plural. A single scene examined on its own almost always justifies itself, because it contains something the author likes and nothing in view contradicts it. Set beside the scene forty pages later that does the same work, it becomes answerable.
Structural editing covers revising, reordering, cutting, and expanding, and duplication is the finding that most often triggers the first three.4 What an editor supplies is the pairing, with both passages quoted, rather than a verdict on either.
Which one survives is less obvious than it looks. The better-written scene is sometimes the one in the wrong place, and cutting the weaker one leaves the book with its best writing stranded where it does the least work.
What does it mean when the book seems to end twice?
The usual finding is that one of the two endings is a resolution and the other is an epilogue that has not been labeled as one. Labeling can be the entire fix, because a reader who knows they have reached an epilogue reads it with different expectations and stops asking when the story will finish.
The less comfortable finding is that the first ending is the book's ending and the second is the opening of a different book. That happens when an author writes past the finish because the material is still interesting. It is one of the few structural findings where the remedy is to keep the pages for a later book rather than to repair them.
An editor reports which of the two the manuscript looks like and why. The author decides, because that decision depends on whether there is a second book.
What if two scenes do the same work?
Duplication is one of the few structural findings that can be demonstrated rather than argued. The two passages go side by side in the editorial letter, and the overlap is visible without anyone having to be persuaded of it.
It is also the finding authors most often disagree with, and the disagreement is frequently right. Two scenes can cover the same ground for different purposes: one establishes a fact and the other shows a character's reaction to it a hundred pages later, and only the first is redundant.
We mark the pairs, quote them, and say what each appears to accomplish. Cutting is not done silently on that basis, because a decision that changes what a book contains belongs with the person whose name is on it. Every document a client sends us is treated in confidence.
References
- Princeton University Press, Proposal Guidelines. https://press.princeton.edu/node/593 ↩
- University of California Press, Book Proposal Guidelines. https://ucpress.edu/resources/book-proposal-guidelines ↩
- The Chicago Manual of Style, 18th edition, chapter 2, Manuscript Preparation, Manuscript Editing, and Proofreading. University of Chicago Press. https://www.chicagomanualofstyle.org/book/ed18/part1/ch02/toc.html ↩
- Editors Canada, Definitions of Editorial Skills, June 2019. https://editors.ca/hire-an-editor/skills/ ↩