Literary criticism — book review structure, voice, stance

The book review is the dominant prestige form of American literary journalism. It runs from 1200 to 4000 words in The New York Review of Books, The London Review of Books, Bookforum, The Atlantic, The New Yorker, and the Sunday New York Times Book Review. To write one well is to demonstrate a particular C2 competence: the ability to describe a literary object accurately, evaluate it with controlled confidence, and sustain a voice for several thousand words without lapsing into either summary or polemic.

Three skills define the form. The first is the balance between description and evaluation — neither pure plot summary nor pure opinion, but a controlled alternation between here is what the book does and here is what I think about what it does. The second is the critic’s stance — a coherent angle from which the book is read, neither neutral (an impossibility) nor merely partisan. The third is quotation craft — the selection and framing of short passages that earn their place by doing argumentative work, not by displaying the reviewer’s taste.

For Russian-speaking writers at C2, the form has a particular trap. Russian critical tradition — descended from Belinsky through Bakhtin through the Formalists — is comfortable with theoretical apparatus and extended argument from first principles. The American book review is, by convention, more empirical and more conversational: it begins with the book, not with a theoretical frame, and earns its abstractions by working through the text. The C2 challenge is to bring genuine analytical depth without retrofitting the review into a literary-theory paper.

Structure — the five-beat review

A standard American book review runs in five beats, with elastic word counts:

1. Frame (150-250w) — a hook that situates the book: in the author’s career, in a literary moment, in a critical question. Not plot summary.
2. Description (300-500w) — what the book is and does. The setting, the central story, the technique, the voice. Compressed; the reader cannot read the book before reading the review, so they need this.
3. First evaluation (300-500w) — what the book does well. Specific. With at least one quoted passage that demonstrates the claim.
4. Second evaluation or complication (300-500w) — what the book does less well, or what is more complicated than it first appears. The serious critic’s signature: even the praise has limits, and the criticism is specific.
5. Lift (150-300w) — the place of the book in the larger picture: the author’s project, the genre, the current literary moment. A short, resonant close, often slightly elevated.

Total target: 1200-2500 words.

Step-by-step craft

1. Decide your angle before you write

The review is not a report on the book; it is a reading of the book. Before drafting, write one sentence that names your angle: This is a novel about marriage that pretends to be a novel about real estate. This is the book the author has been trying to write for twenty years, and it succeeds in places where her earlier work failed. The novel’s strengths and its faults come from the same impulse — an unwillingness to leave any thought unstated. The angle is the spine; the review is the body that grows around it.

2. Read the book twice

The first reading is for the experience; the second is for the review. On the second pass, mark passages that demonstrate the book’s central technique, mark passages that fail or strain, and mark the structural turning points. Russian-speaking writers, in my experience, often skip the second reading and review from notes and memory. The American book review’s quality depends on the reviewer having held the whole book in mind, which requires a second pass.

3. Open with a frame, not a summary

The lede of the review is rarely X has written a book about Y. The frame is more oblique: a fact about the author’s career, an observation about the genre, a sentence about the literary moment, a particular detail from the book held up as emblematic. The frame buys the reader a reason to keep reading; the summary, in the lede position, gives them a reason to stop.

4. Compress the description

The descriptive section is where the reviewer earns trust. The reader needs enough of the book to follow the evaluation but not so much that the review becomes a Cliffs Notes substitute. The rule of thumb: describe in the present tense (the novel follows, the narrator returns to), name the technique not just the content (the third-person free indirect discourse keeps us close to Helen without ever letting her speak), and avoid retelling the plot. A reader who wants the plot can read the book.

5. Quote with restraint and purpose

Quoted passages should be short — usually one to three sentences — and should do something the reviewer’s prose cannot do. The quote demonstrates the prose, the technique, the voice, the failure. Frame every quote: introduce what the reader should notice, then quote, then comment. Russian critical tradition often quotes long passages and lets them speak; American convention is tighter. A 2000-word review contains perhaps four to seven short quotations.

6. Hold a position, but qualify it

The American book review tolerates strong evaluation if the evaluation is specific. This is a brilliant book is unhelpful; The novel succeeds because of its handling of time — its forty-year span feels compressed without ever feeling thin is a useful claim. Negative evaluation works the same way: The book fails is empty; The book’s last hundred pages cannot sustain the pressure the first three hundred have built is a claim the reader can test.

7. Lift the close

The final paragraph should leave the reader with a sentence to remember. The reliable move is to widen the frame: from this book to the author’s career, or to the genre, or to the literary moment. Avoid the summary close (In conclusion, the book has both strengths and weaknesses) — it is the structural equivalent of saying nothing.

8. Hold a particular stance

Every review has a stance — the angle from which the critic reads the book. The stance might be formal (interested in technique), thematic (interested in what the book is about), comparative (interested in how it sits relative to other books), or evaluative (interested in whether it succeeds). Hold one stance per review; reviews that try to be everything are usually nothing.

9. Trust your initial reaction, then test it

The first reaction to a book is often the right one but is rarely the final reaction. Trust the initial feeling that the book works or does not work; then test the reaction against specific evidence from the text. If the evidence supports the reaction, you have a review. If the evidence undermines it, you have something more interesting.

The stance — what it is and how to find it

The critic’s stance is the angle from which the book is read. It is the difference between here is a book and here is a book read in a particular way. Finding the stance is the central intellectual work of the review.

The formalist stance

The book is read for its technique — how it is constructed, what its prose does, how it manages time or perspective or voice. James Wood’s reviews are usually formalist. The stance produces reviews rich in specific quotation and technical analysis.

The thematic stance

The book is read for what it is about — its concerns, its arguments about the world, its engagement with current questions. Reviews in The New York Review of Books often take this stance, treating the book as an intervention in ongoing debates.

The comparative stance

The book is read against other books — by the same author, in the same genre, addressing the same questions. The comparative review situates the new book within a tradition; the critic’s authority comes from having read widely.

The evaluative stance

The book is read for whether it succeeds at what it is trying to do. The evaluative review is the most common form in newspaper book sections; it requires the critic to identify what the book is attempting before judging its success.

A skilled critic combines stances within a single review but foregrounds one. The reader should be able to summarize the critic’s angle in one sentence after finishing.

Full model text — 750-word annotated book review

The model below is a compressed review of an imagined novel, The Quiet Years, by an imagined author, Diana Frawley. The five beats are marked in brackets.

An Almost Quiet Book

[Frame] Diana Frawley has been writing the same book, in various costumes, for twenty years. The Quiet Years, her sixth novel, is the version she has been working toward — the one in which the costume finally falls away and the book that has been waiting underneath stands up. It is the best thing she has done. It is also, as one might expect from a novelist who has spent two decades approaching a single problem, a book whose strengths are inseparable from its limitations. To admire the book honestly is to admire the limitations too.

[Description] The Quiet Years tracks a marriage between two academics, Helen and Aaron Marsh, across the period from 1994, when they meet at the University of Wisconsin, to 2019, when their elder daughter takes a job in Berlin. The novel’s central technique is its handling of time. Frawley moves not by chapters but by years — sometimes one to a section, sometimes three or four, occasionally a single afternoon stretched across thirty pages. The third-person narration sits close to Helen for most of the book and shifts, twice, to Aaron, with a deliberate awkwardness the prose makes no effort to disguise. Plot, in the conventional sense, is thin. A great deal happens; almost nothing accumulates in a way that demands the reader’s attention until, suddenly, in the last sixty pages, everything that has been quietly accumulating presents itself as having mattered all along.

[First evaluation — what the novel does well] The book’s most considerable achievement is its prose at the sentence level. Frawley has always written well; here she writes at a higher pitch than she has reached before. Consider this passage, from page 174, in which Helen is observing her husband during what will turn out to be the central evening of the novel:

Aaron was carrying the empty bottle to the kitchen with the careful, slightly performed competence he reserved for after dinner, the competence of a man who has decided, for reasons he has not examined, that to take this particular bottle to the kitchen is the small thing he will do well tonight.

What the passage does — and what Frawley’s earlier novels could not do — is hold a quiet observation in a long sentence without strain. The sentence’s syntactic patience matches its subject’s emotional patience. The book is full of such sentences. A reader who has admired Anne Tyler or Stewart O’Nan and wished for a slightly stricter prose will find Frawley’s the right next book.

[Second evaluation — the complication] The same patience that produces the prose also produces the novel’s structural problem. The Quiet Years depends, for its accumulating effect, on the reader’s willingness to give Frawley two hundred and fifty pages before anything turns. That willingness, in 2026, is asking a great deal. The novel’s first half is full of fine observation and almost no propulsion. Even readers prepared to admire the project may find themselves checking, at page 130, for how much remains. Frawley’s editor has not pushed her to cut, and the book would be both better and more widely read if a hundred pages had been compressed into sixty. There is a version of this novel that does what it does in three hundred and twenty pages instead of four hundred and forty, and that version is the masterpiece. The version we have is something closer than Frawley has been but still a step away.

[Lift] Frawley’s project, considered across the six novels, has been to write a domestic realism without the apparatus that domestic realism usually borrows from genre — without crisis, without revelation, without the convenient deaths and discoveries that organize a more obviously plotted book. The Quiet Years is the most serious version of that project she has produced. If she has not yet written a great novel, she has now written a book that makes clear what the great novel, when it comes, will look like. The wait will be worth it. It has, in places, already been.

Three subgenres of the American book review

The book review in American magazines divides into three recognizable subgenres. A C2 critic should know each and recognize which one a given assignment calls for.

The single-book review

The most common form: 1200 to 2500 words on a single new book. The structure is the five-beat review described above. Most newspaper book sections and most magazine reviews fall here.

The omnibus review

The omnibus or essay-review covers two to five related books, treating them as a group. The New York Review of Books and The London Review of Books publish these regularly. The structure is more elastic; the critic uses the books as occasions to develop a larger argument about a genre, a theme, or a period. Length runs 3000 to 8000 words.

The career review

The career review treats an author’s body of work, often occasioned by a new book but using the book as a way into a larger assessment. The New Yorker and The Atlantic publish these on major writers. The structure follows the author’s development; the new book is positioned within the larger career arc. Length runs 4000 to 10000 words.

Each subgenre uses the five-beat structure adapted to its scope. A C2 critic should be able to produce the single-book review at standard length and recognize when an omnibus or career review is the form the material calls for.

How critics build authority

A critic’s authority is the central currency of literary journalism. The reader trusts the critic’s evaluation only to the extent that the critic has demonstrated qualities that justify trust. Four practices build authority over a career.

Specific reading

A critic who can quote the book accurately, situate it within the author’s body of work, and notice technical features other readers have missed earns specific authority on that book. Generic praise or generic criticism builds no authority.

Calibrated stance

The critic who praises everything has no praise to give that anyone takes seriously. The critic who dismisses everything is no longer read. The credible critic praises specifically when praise is earned and criticizes specifically when criticism is earned; the calibration over time builds trust.

Visible reading

The critic should read books the reader has not read and should make the reader want to read them. The visible reading — the evidence that the critic has lived with the book carefully — is what distinguishes the review from a sales pitch.

Engagement with the field

A critic operates within a critical tradition. Naming other critics, engaging their positions, taking part in ongoing debates about authors and genres — these practices situate the critic within a community of readers and writers.

The first paragraph of a review — five reliable patterns

The opening paragraph of a book review carries the same weight as the lede of a feature. Five patterns recur in American magazine practice.

Pattern 1 — the author’s career

Diana Frawley has been writing the same book, in various costumes, for twenty years. The book is situated within the author’s larger project; the review becomes an act of placement.

Pattern 2 — the literary moment

At a moment when American fiction has, by general agreement, abandoned the social novel, X arrives to insist that… The book is positioned within current literary debates.

Pattern 3 — the controversy or contradiction

The book has been called both X’s masterpiece and X’s failure. Both readings have some basis. The truer reading lies between them. The opening positions the review as an intervention.

Pattern 4 — the particular detail

On page 174 of X’s new novel, in a passage that runs across two paragraphs, the prose does something I have not seen it do before. The opening anchors in a specific moment from the book itself.

Pattern 5 — the question

What is a novel about a marriage actually about? Diana Frawley’s sixth book is the most considered answer she has yet produced. The opening poses the question the review will explore.

Each pattern earns the reader’s first 100 to 200 words. The choice depends on the book and the critic’s angle.

Common pitfalls

Plot-summary review

A review that retells the plot for 800 words and then evaluates in 200 is not a review; it is a synopsis with an opinion attached. Compress the description; expand the evaluation.

Unspecific praise or blame

A brilliant novel / a disappointing novel — empty until earned by specific claim. Every evaluative sentence should be followed by an explanation the reader can test.

Theoretical apparatus front and center

Importing literary theory as the frame — Derrida’s notion of the trace can help us understand… — is the academic move, not the magazine move. The American book review uses theory glancingly if at all, and only when it actually clarifies the book.

Reviewing the book the author did not write

The classic critical failure: complaining that X did not write a different book than the one X wrote. Review the book in front of you, on its own terms; then assess how well it achieves what it is trying to achieve.

The unearned ending

And so this book reminds us that life is complicated. The lift must come from the specific reading, not from a generic reflection.

The middle of the review — keeping attention

A 2000-word review presents an attention problem in its middle. The reader has read the opening; the closing is still distant. Three techniques help hold attention through the middle paragraphs.

Compare to another book

A specific comparison to another book — one the reader might have read — refreshes attention. Where Frawley’s earlier work resembled Anne Tyler at her most patient, the new novel is closer to the late Stewart O’Nan… The comparison gives the reader a reference point and lets the critic position the book in a tradition.

Quote a specific passage

A well-chosen passage breaks the rhythm of the critic’s prose and provides physical evidence for the claims being made. Use a passage in the middle of the review as much as at the beginning.

Shift register briefly

After several paragraphs of measured analytical prose, a single sentence of personal reaction (I read this novel twice; the second time, I read it with a pencil) refreshes the reader. The shift should be brief; sustained personal voice in a review tilts the genre toward memoir.

Connectors and phrases bank

• Framing the review: X has been writing this book, in various forms, for years; X’s new novel sits in a long American tradition of…; The book arrives at a moment when…
• Describing technique: The novel’s central device is…; The third-person narration sits close to X without ever letting her speak; The prose’s signature is its handling of time / silence / refusal…
• Praising specifically: What the novel does that her earlier work could not is…; The book’s most considerable achievement is…; The sentences carry their length without strain because…
• Criticizing specifically: The same patience that produces the prose also produces the structural problem…; The novel cannot sustain across four hundred pages the pressure it builds in the first hundred…; What we miss is not plot but consequence…
• Quoting frames: Consider this passage, from page X…; Here is the moment, midway through the novel, when…; The book’s central sentence — and there is a central sentence — reads as follows…
• Lifting the close: X’s project, considered across the six novels, has been…; If she has not yet written the great American novel of marriage, she has now written…; The wait will be worth it. It has, in places, already been.

The quotation as evidence — selection and framing

Quoted passages in a review are the critic’s evidence. A claim about a novel’s prose must be supported by an actual sentence from the novel; a claim about its handling of time must be supported by an actual passage in which time is handled. The selection and framing of quotation is itself a craft skill.

Choose passages that demonstrate one thing

A long passage that demonstrates several things is harder for the reader to assess than a short passage that demonstrates one. Choose the one-sentence quote that shows the prose’s signature; choose the three-sentence quote that shows how the technique works; resist the temptation to quote four pages.

Frame before, comment after

Every quote should be framed by a sentence that tells the reader what to notice (The novel’s central technique is its handling of time. Consider this passage, from page 174…) and followed by a sentence that interprets it (What the passage does is hold a quiet observation in a long sentence without strain.). Quote-naked passages do not work in reviews; they read as filler.

Vary quote length

A review with five quotes all of the same length reads as monotonous. Mix short (a phrase) with medium (a sentence) with longer (a paragraph). The variety produces cadence; the cadence holds attention.

Verify accuracy

Quote against the published text, not from memory. A misquoted passage destroys the critic’s authority. American magazine reviews are not fact-checked at the level features are, but accuracy on quotation is the critic’s own obligation.

The challenge of the negative review

Negative reviews are harder to write well than positive reviews and harder to publish. The American magazine culture, particularly in the past decade, has tilted toward positive coverage; the negative review has become correspondingly more difficult and more valued when it is well done.

The fair negative review

A fair negative review meets the book on its own terms. It acknowledges what the book is trying to do, evaluates whether it succeeds at what it is trying to do, and identifies the specific failures. It does not dismiss the book without engagement; it does not attack the author personally; it does not project a different book onto the one in front of the critic.

When to be negative

Negative reviews are most useful when a much-praised book is over-praised, when an important writer’s new work fails in ways that earlier work did not, or when a problematic book is being received without critical attention. They are less useful as gestures of contrarianism or as career-building displays.

The ethics of the negative review

A negative review can damage an author’s career, reduce a book’s readership, and affect the lives of real people. The critic should be willing to defend the review to the author, should base every claim on specific evidence from the book, and should avoid the temptation to be entertaining at the book’s expense.

Common Russian-speaker writing mistakes

1. Theoretical opening — The novel can be read through the lens of… In Russian critical tradition this is conventional and respected; in American magazine reviewing it reads as academic transposition. Open with the book and your reading of it, not with a theoretical frame.
2. Excessive plot summary — Russian reviews often run 60-70 percent description; American magazine reviews target 25-35 percent. Compress the what to expand the how and whether.
3. The phrase the author — The author shows…, The author uses… American reviews prefer the author’s name, possessive constructions, or the work itself as agent: Frawley shows…, The novel uses…, the prose insists…. The author across a 2000-word review reads as student writing.
4. Calque on no doubt — No doubt, this is a significant novel. This is a calque on Russian несомненно used as an evaluation booster. The English idiom no doubt in this position reads as concessive (preparing a but); use clearly, unmistakably, or simply commit to the claim without the hedge.
5. Long unframed quotation — Russian critical tradition admits long quoted passages; American magazine convention frames every quote (introduce, quote, comment) and trims most quotes to one to three sentences.
6. Sentimental literary-cultural appeals — In the great tradition of Russian literature… and similar pan-cultural appeals do not work in American reviews even of Russian books. Specificity beats lineage.
7. The schoolbook summary close — In conclusion, the novel has both strengths and weaknesses. American reviews do not summarize themselves at the close; they lift to a wider frame or a resonant sentence.

Critics to read closely

The most efficient way to internalize the form is to read its practitioners with attention. Four contemporary American critics, in particular, repay close study.

James Wood

Wood’s New Yorker reviews and his book How Fiction Works are the most influential single body of criticism of the past two decades. Pay attention to how he uses short quoted passages to demonstrate technical claims; the move is precise and replicable.

Parul Sehgal

Sehgal, formerly at The New York Times and now at The New Yorker, brings an unusual range and a careful evaluative calibration. Pay attention to how her reviews establish stance early without overstating.

A. O. Scott

Scott’s reviews and his book Better Living Through Criticism demonstrate the craft of writing about books and films for a broad audience without losing analytical depth. Pay attention to his openings — almost never plot summary, often a frame from the writer’s career or the cultural moment.

Christian Lorentzen

Lorentzen, writing for Bookforum, Harper’s, and Granta, exemplifies the willingness to deliver substantive negative criticism with fairness and specificity. Pay attention to how he engages a book he disagrees with on the book’s own terms.

Reading any one of these critics closely, paragraph by paragraph, will teach more about American literary criticism than any abstract description can.

Notes on revising a review

The first draft of a review almost always over-summarizes and under-evaluates. The revision tilts the balance.

Cut summary, expand evaluation

Identify every paragraph that describes the book’s content rather than the book’s effect. Cut the descriptive paragraph to half its length; use the recovered space for evaluation. The reader can read the book for the content; they read the review for the critic’s reading.

Sharpen each evaluative claim

Every sentence that contains an evaluation should be testable. The novel succeeds is not testable. The novel succeeds in its handling of time, sustaining a forty-year span without losing density is testable. Replace the vague with the specific in every revision pass.

Strengthen the lift

The closing paragraph is the last thing the reader reads. Revise it more carefully than any other paragraph. A weak close drains a strong review of authority; a strong close lifts even a workmanlike review into something the reader remembers.

A working sequence for the review

For the C2 critic drafting a review, the following sequence is reliable.

1. Read the book once for the experience.
2. Set the book aside for two days.
3. Read the book a second time with the review in mind; mark passages.
4. Write one sentence that states your angle.
5. Draft the description section, in the present tense, naming the book’s central technique.
6. Draft the first evaluation, with one supporting quotation.
7. Draft the complication, with one supporting quotation.
8. Draft the frame and the lift; these often emerge together.
9. Read the full draft aloud once.
10. Revise for cadence and cut by 15 percent.

The full sequence typically takes between eight and twenty hours for a 2000-word review of a 350-page book. Reviews produced more quickly than this almost always show their speed; reviews produced more slowly often lose their angle.

Summary

• The five-beat review: frame, description, first evaluation, complication, lift.
• Open on a frame, not a summary; the frame names your reading.
• Description is compressed; technique matters more than content.
• Every evaluative claim is specific and demonstrable; quote with restraint and purpose.
• The lift widens the frame at the close — to the author’s career, the genre, the moment.
• Russian-speaking critics should especially watch the theoretical opening and the long unframed quotation.

Next lesson: Legal-style prose — IRAC, legal memoranda, brief writing, legal English conventions.