A manuscript arrives. The prose is technically accomplished — sentences varied, vocabulary precise, paragraphs well-shaped. And yet something is missing. The editor reading it can name the problem within a few pages, even if the writer cannot: the text has style, but it has no voice.
These are not synonyms. Treating them as such is one of the more persistent confusions in craft discussion, and it costs writers more than they realise.
What style actually is
Style is the set of observable, describable choices a writer makes at the level of language: sentence length and rhythm, syntactic complexity, diction range, punctuation habits, use of abstraction versus concrete detail, paragraph architecture. Style is, in principle, teachable and transferable. It can be analysed, catalogued, and imitated.
Walker Gibson’s 1966 study Tough, Sweet and Stuffy: An Essay on Modern American Prose Styles remains one of the more rigorous attempts to define prose style as a classifiable phenomenon. Gibson argued that American written prose could be mapped across three broad registers — the curt, first-person-centred “tough” style of journalism and hard-boiled fiction; the intimate “sweet” style of advertising; and the impersonal “stuffy” style of officialdom and academia. What made his taxonomy useful wasn’t its comprehensiveness but its precision: each style could be identified through specific, measurable surface features. Gibson was describing style as a set of technical choices, not as an expression of a self.
This matters because style is, to some extent, a choice. Writers pick up stylistic habits from what they read, from the genres they work in, from editors who have corrected them. Two writers working in the same register — spare literary fiction, say, or long-form magazine journalism — will often share stylistic markers without sharing anything deeper. Their sentences may look similar. They will not sound the same.
What voice actually is
Voice is harder to locate on the page because it is not primarily a set of surface features. It is closer to a sensibility — the particular angle from which a writer sees and the assumptions embedded in how they frame what they see. Voice shows up in which details a writer notices, which comparisons they reach for, which ideas they treat as requiring explanation and which they assume the reader already holds. It is present in the gap between what the text says and what it implies.
A 2023 literature review of research on authorial voice in writing, published in Social Sciences & Humanities Open, traced how the concept has been debated across fifty years of composition and linguistics scholarship. The review found that voice has been theorised as an individual property, a social construction, and more recently as something dialogic — shaped by the writer but also by the discourse communities they inhabit. What remained consistent across those frameworks was the core distinction: voice is not reducible to linguistic surface features in the way that style is.
In manuscripts we work on, this shows up in a specific way. A writer may have absorbed the stylistic conventions of a genre so thoroughly that their prose is indistinguishable, at the sentence level, from competent work in that genre. But if you read two pages, then set the manuscript down and try to describe the writer’s perspective — what they find important, what they find funny, what assumptions they carry about the reader — and you come up with nothing, that is a voice problem, not a style problem. The prose is correct. There is no one home.
Why they get conflated
The conflation is understandable. Voice does manifest through style. A writer whose voice is laconic will tend toward shorter sentences. A writer with a digressive, essayistic sensibility will lean toward subordination and parenthetical qualification. So stylistic analysis can sometimes approximate voice analysis, and this is part of why the terms blur.
Forensic linguists have had to draw the distinction sharply, because attribution depends on it. As EBSCO’s research overview on forensic linguistics and stylistics notes, authorship attribution analysis compares “syntax, word choice, vocabulary, punctuation, and other elements of written language” — but also something harder to specify: the persistent idiosyncratic choices that survive even when a writer is attempting to disguise their style. Those idiosyncrasies are the trace of voice operating below the level of conscious stylistic decision. A writer can change their style deliberately. They cannot easily change their voice, because they are often unaware of what, exactly, their voice consists of.
This is the core asymmetry: style is largely conscious; voice is largely not. Writers can choose to write in short, declarative sentences. They cannot choose the way they habitually structure an argument, the logic of their analogies, the things they find it necessary to say.
The imitation test
A practical way to see the distinction is to consider what happens when skilled writers imitate other skilled writers. A competent stylist can approximate Hemingway’s surface features — the parataxis, the dialogue punctuation, the avoidance of adverbs. The result will be stylistically recognisable as Hemingway-adjacent. It will not have Hemingway’s voice, which has nothing to do with his surface style and everything to do with his particular relationship to silence, omission, and what his sentences refuse to say.
Roland Barthes, in his 1967 essay “The Death of the Author,” argued that writing is “the destruction of every voice, of every point of origin” — that the moment language begins, origin and identity dissolve into the text. Editors encounter the opposite problem daily. Voice is not absent from text; it is persistently, stubbornly present even when the writer wishes it weren’t. Ghostwriters know this. So do editors who work on collaborative projects: the seams between voices are usually visible regardless of how carefully the style has been unified.
What editors actually do with this distinction
When we read a submission, the questions we are asking about style and voice are distinct. Style questions are: Is this prose clear? Are the sentences varied? Does the register suit the audience? Is the punctuation doing something intentional or just erratic? These are questions with broadly checkable answers.
Voice questions are different: Does this text have a perspective? Is there a mind behind these sentences, making choices, not just filling in the genre template? Does the prose carry any conviction — any sense that the writer has a particular relationship to this material, rather than a general competence with language?
Style problems are fixable by a good line editor. Voice problems are not. If a manuscript lacks voice, the editor cannot supply one. The most an editor can do is strip away the stylistic noise that is obscuring the voice that is already there — or tell the writer, honestly, that the work needs to be reconceived at a more fundamental level than editing can address.
This is why the distinction matters practically. Writers who focus only on style — on sentence-level correctness, on varying their syntax, on using the right kind of prose for the genre — can produce technically accomplished work that leaves no impression. The style serves no one because there is no voice for it to serve.
What it means for revision
Style is revised by working at the sentence level: reading aloud, cutting, varying rhythm, checking that the diction is consistent. Voice is not revised the same way. Voice is engaged by asking harder questions about the work: What does this piece actually think? What is the writer’s genuine relationship to the subject — not their performed relationship, but their actual one? Where does the prose go flat, and is the flatness concealing something the writer is avoiding?
The manuscripts that arrive with strong voice and weak style are, in our experience, easier to work with than the manuscripts that arrive with strong style and no voice. The first kind has something that can be clarified. The second kind is clear already, and the clarity reveals nothing in particular.
That is, in the end, what editors mean when they say a piece of writing has no voice. Not that it lacks personality or warmth or some quality of likability. They mean it has no perspective. It is competent language in the service of no discernible intelligence. Style without voice is prose that knows how to behave and has nothing to say.
