How to read an editorial letter and a marked-up manuscript
An editorial letter is an argument about a book, delivered to the one person who can act on it. Authors open them badly, reading for the verdict rather than for the reasoning, and the reasoning is the part that transfers to the next chapter and the next book. What follows describes what a letter should contain, how to sort what it says, and where an author's authority begins.
What is an editorial letter supposed to contain?
An editorial letter is a developmental instrument. The EFA describes developmental editors as dealing with content, organization, and genre considerations, and as producing a manuscript evaluation or an editorial letter that sets out large-scale issues.1 That is the scope: the shape of the book rather than its sentences.
A letter that does its job covers five things. It identifies what the manuscript does well in specific terms, names the central problem once, and traces that problem to the places in the text where it shows. It then sets out the options for fixing it with what each will cost, and says what falls outside the editor's remit.
Not every letter follows that order and none of it is mandated. What is reasonable to expect is that the letter argues rather than asserts, so an author who disagrees can identify the point of disagreement.
What is working, and why does that belong in the letter?
An author revising a book needs to know what not to break, and only the letter can tell them. A letter that lists only problems invites a revision that removes the book's strengths along with its faults, and the author has no way to tell which is which.
Specificity is what makes this useful rather than decorative. A sentence saying the dialogue is strong tells an author nothing. A note saying that the arguments between two characters carry the exposition without pausing for it identifies a technique the author can keep using deliberately.
The professional standards support this indirectly, requiring changes to be made without altering intended meaning and asking editors to explain changes as appropriate.2 Knowing which effects were intended is part of that, and the letter is where an editor demonstrates they understood them.
What is the central problem, and why is it named once?
Most manuscripts have one problem with many symptoms, and listing the symptoms separately produces a letter of thirty items an author cannot prioritize. Take a book whose viewpoint character has no stake in the outcome. It will generate complaints about pace, about the middle, about the ending, and about the reader's attachment to the cast, and all four are the same problem.
Naming it once, then showing where it surfaces, gives an author a single thing to solve. Naming it thirty times gives them thirty repairs that will not fix anything.
Naming it once also protects the author from over-revision. Symptoms addressed one at a time get patched, the book gets longer, and the cause remains where it was. Authors who have revised a manuscript five times without it improving are usually working symptom by symptom.
Which comments must be dealt with, and which are optional?
A good letter separates these explicitly, and an author is entitled to ask for the separation if it is missing. The distinction is not between important and unimportant but between findings about the book and preferences about the writing.
Editors Canada's standards ask editors to consider the implications of time, cost, production processes, and the intended audience when suggesting changes.2 That is a requirement to think about what a change will cost the author, which is what makes a priority ordering possible at all.
An author working to a deadline should say so before the letter is written. A letter that assumes six months of revision is a different document from one that assumes three weeks, and only the editor knows which they wrote.
If I can only do three things before submitting, will the letter say which three?
It will if the author asks, and it is worth asking. A ranked short list is a different piece of work from a full letter, and it is the version most useful to an author with a submission window.
The three are usually not the three largest problems. They are the three whose repair is visible in the opening chapters, because that is what an agent or acquiring editor reads before deciding.
An editor cannot promise that the changes will alter the outcome, and Writer Beware is direct that no reputable editor claims a book has huge commercial potential or is likely to become a best seller.3 A ranked list is a judgment about the manuscript, not a prediction about the market.
What is a stet, and is it final?
Stet is the instruction to leave the original as it stood, and in an author's hands it is final. An author writing stet against a change has rejected it, and no further justification is owed.
The useful question is whether an author wants the reason anyway. Some do, because a change rejected on page 40 will be proposed again on page 300 unless the underlying preference is recorded, and the reason is what makes recording it possible.
The standards ask for queries that are clear, tactful, succinct, and actionable, and for editors to explain changes as appropriate.2 An editor who explains a change once, then accepts a stet without arguing, is meeting both halves of that. Arguing a second time is a failure of the first explanation rather than persistence.
Do I get a second pass after I revise?
Only if the agreement says so. Nothing external sets the number of passes, and Editors Canada's standards describe skills and practices rather than a count of readings.2
The reason it matters is that revision introduces new text. A chapter rewritten after a copyedit has never been edited, and it sits in a file where every other chapter has been, which makes the difference hard for anyone to see later.
Where a second pass is wanted, agreeing it in advance costs less than adding it afterward. An editor holding time for it schedules differently. A manuscript returning months later, changed heavily, may be quoted as new work rather than as a revisit, which is a reasonable position rather than an opportunistic one.
Am I entitled to ask for more or less intervention?
Asking is entirely legitimate, and the request is more useful before the pass than after. Depth of intervention is the single largest variable in how a returned manuscript looks, and two competent editors working at different depths produce very different files from the same chapter.
The standards require an editor to work within the scope of the project and the stage of editing, which makes the level a matter of agreement rather than of the editor's preference.2 An author who wants light touch is defining scope, not restricting the editor's competence.
Saying so partway through is also legitimate. An author who receives the first two chapters and finds the pass heavier than expected should say so immediately, because the alternative is paying for a whole book edited at the wrong depth.
What is outside the editor's remit, and who should look at that instead?
A letter should say this rather than leaving it implied. Legal review, specialist factual verification, market judgment, and design are all outside what an editor supplies, and each has someone else who does it.
The standards make this an obligation at the outset, asking editors to confirm before starting that they have the skills, training, and experience the work needs.2 The corollary is naming what they do not.
We flag terms and claims we cannot verify rather than correcting them, and we say plainly where a question belongs with a lawyer or a subject specialist. A letter that states professional facts and then disclaims professional opinion is contradicting itself, and an author is entitled to notice. Every document a client sends us is treated in confidence, including whatever an author says in reply.
Should an author accept or fight the copyedit?
As a policy, neither one works, and both are common. Accepting everything without reading produces a book the author has not fully chosen. Rejecting on instinct produces a book that keeps the errors the author is used to seeing.
The workable approach is to sort rather than to decide globally. Mechanical corrections can be accepted in bulk. Changes to meaning, voice, rhythm, or a deliberate choice deserve individual attention, and those are the ones a good editor has queried rather than made silently.
Where an author finds themselves rejecting the same category of change repeatedly, that is information worth sending back rather than acting on alone. One message settling it puts the decision on the style sheet and ends the disagreement for the rest of the book.
References
- Editorial Freelancers Association, Editorial Service Definitions. https://www.the-efa.org/editorial-services-definitions/ ↩
- Editors Canada, Professional Editorial Standards 2024: A. The Fundamentals of Editing. https://editors.ca/publications/professional-editorial-standards/fundamentals-editing/ ↩
- SFWA, Writer Beware: Editors and Editing, updated December 14, 2025. https://sfwa.org/other-resources/writer-beware/editors/ ↩