The questions an editor asks while editing
Queries in a margin can read like criticism when they are usually the opposite: an editor asking rather than changing is declining to guess about somebody else's book. Editors Canada's Professional Editorial Standards set an explicit requirement here, calling for judgment about when to query and when to resolve a problem without consultation, and for queries that are clear, tactful, succinct, and actionable.1 This page explains what the common ones mean.
Why does an editor ask "is this what you meant here?" instead of changing it?
Because the alternative is a guess that would be invisible once made. A sentence with an unclear referent can usually be repaired two ways, and the two say different things. Choosing one silently means an author accepts a meaning they never chose.
The standards frame this as a duty rather than a courtesy, requiring that changes be made without altering intended meaning.1 An editor who cannot determine the intended meaning has no way to satisfy that requirement except by asking.
Communicating with authors is treated as a distinct part of the editing process, and the Chicago Manual of Style gives it its own place in the chapter on manuscript editing.2 The query is also cheaper than the alternative for the author. Answering a margin note takes seconds. Discovering in proofs that a sentence now says something else takes a re-read of the chapter, and sometimes it is not discovered at all.
What happens when a sentence has two possible readings?
Both readings are quoted back, so the author is choosing between things they can see rather than being asked an abstract question. A good query of this kind reads like a menu.
Ambiguity of this sort is rarely deliberate, and where it is, the author says so and it stays. Legal, medical, and technical writing is the exception in the other direction, since a sentence with two readings in an instruction is a fault regardless of intent.
The one thing an editor should not do is pick the more likely reading and move on. Likelihood is not knowledge, and the author is available. Where a manuscript produces dozens of these, the finding worth reporting is the habit rather than each instance, since a construction the author favors will keep producing them.
What happens when one paragraph contradicts another?
The two passages are quoted side by side with their page numbers, and the query asks which stands. Neither is corrected until the author answers, because an editor cannot tell which was written later or which the book depends on.
Contradictions cluster in revised manuscripts. A change made in chapter three has consequences in chapter eleven, and the consequences are the part that gets missed, because the author was thinking about chapter three.
Where the same contradiction appears in several places, it is reported once as a pattern rather than queried at every occurrence. That is what the requirement for actionable queries amounts to in practice.1 Forty notes saying the same thing are harder to act on than one note listing forty page numbers.
What does an editor do with a passage they cannot follow?
Say so, plainly, and ask the author to explain it in ordinary terms. An editor who cannot follow a passage is evidence about the passage, since editors read more slowly and more attentively than the eventual reader.
The useful form of this query names what broke rather than reporting confusion. A sentence where the subject changes halfway, a paragraph where three unnamed pronouns compete, or an argument with a step missing are all specific failures, and each has a different repair.
What an editor should not do is rewrite the passage into something coherent and hope. The rewrite would be the editor's argument rather than the author's, and once accepted it would be untraceable. Authors who accept every change without reading them are especially exposed to this, which is a reason for editors to be careful rather than a reason to blame authors.
What happens when a character is thirty-four in one chapter and thirty-one in another?
It comes back as a query with both page numbers, because only the author knows which age the book needs. The arithmetic is the editor's contribution: an age stated in chapter two, an interval stated in chapter six, and an age stated in chapter nine can be checked against each other.
Copy editing covers this as a matter of definition, since it is editing to ensure correctness, accuracy, consistency, and completeness.3 The EFA's own description of copyediting includes checking cross-references alongside spelling, grammar, usage, and punctuation.4 Continuity of this kind is where most of that work goes in a novel.
Once the author answers, the answer is recorded on the style sheet with the other facts of the book. A later pass then works from the record rather than raising the question again, which is most of the value of keeping one.
What happens when a name is spelled two ways?
The variants are collected with counts, which usually settles the matter before the author has to think about it. A name appearing two hundred times one way and eleven times another is not a real disagreement, and the eleven are corrected on the author's confirmation.
Where the split is even, or where both spellings appear deliberately, the query matters more. Characters who change their names, transliterations that vary by period, and place names with historical forms are all legitimate reasons to carry two spellings.
Sending a list of preferred spellings before the edit removes almost all of these queries in advance. That is the cheapest way an author can buy back an editor's attention for something harder, and it takes about ten minutes to compile.
What happens when an event is Tuesday in one chapter and Thursday in another?
Dates get the same treatment as ages, for the same reason. Days, dates, seasons, elapsed time, and the order of events form a chronology that can be extracted from the text and checked against itself, and the extraction is mechanical work rather than interpretation.
Books that move between timelines produce most of these findings. An author writing scenes out of order is working from a chronology held in their head, and the head version is more consistent than the manuscript.
An editor can build the chronology as a separate document and hand it back with the manuscript. That costs time and it is worth asking for on any book where the timeline carries weight. It is also the document a sequel will need.
What happens when three terms are used for the same thing?
An editor asks whether the terms should be standardized, because the answer is not automatically yes. Technical and academic writing usually wants one term used consistently, since a reader has to know that two words mean two things. Fiction often wants variety, and a character who calls a place something the narrator does not may be doing it for a reason.
Copy editing includes checking mechanical consistency, so the pattern will be found either way.3 What is at issue is what to do about it.
The decision goes on the style sheet as a rule rather than as a series of individual fixes. A term introduced in a later revision is then handled the same way without a new conversation, and a proofreader inherits the same rule.
Where does this figure come from, and why does the editor want to know?
A number in a book is a claim, and a claim without a source cannot be checked by anyone downstream. Copy editing includes checking factual accuracy, and a figure with no origin is the point at which that check stops.3
Numbers also drift. A figure typed as a round number to be refined later, a percentage remembered from a talk, and a statistic copied from a source that had itself copied it are all common. None of them looks provisional on the page.
An editor who cannot verify a figure flags it rather than removing it. Whether it stays is the author's decision, and where it came from is information only the author holds. A figure that cannot be sourced at all is usually better cut than defended.
Is this quotation accurate, and can I point the editor at the source?
Quotations are checked against sources where sources are supplied, and flagged where they are not. Transcription drift is the usual finding. A word changes, a comma appears, or an ellipsis covers more than it seems to, and none of it is deliberate.
A quotation that has traveled through several books before reaching this one is the highest-risk kind. Each transcription is an opportunity for a small change, and the version in circulation is sometimes not the version that was written.
Supplying sources with the manuscript, rather than after the queries arrive, is faster for everyone involved. It also lets an editor check the surrounding context. That is where quotations most often turn out to mean something other than what they appear to mean in isolation.
Is this person's title correct?
Titles change and books take years, which is the whole of the problem. A person described by the office they held when a chapter was drafted may have left it before publication, and the sentence still reads perfectly well.
An editor checks what can be checked against the record and queries the rest. Where a title was correct at the time of the events described, the fix is usually a word rather than a correction, since a former officeholder can be identified as such.
The related check is spelling and form of name, which people are entitled to have right. That belongs on the style sheet with everything else, and every document a client sends us is treated in confidence, working lists included.
References
- Editors Canada, Professional Editorial Standards 2024: A. The Fundamentals of Editing. https://editors.ca/publications/professional-editorial-standards/fundamentals-editing/ ↩
- 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/ ↩
- Editorial Freelancers Association, Editorial Service Definitions. https://www.the-efa.org/editorial-services-definitions/ ↩