The AP English Language and Composition rhetorical-analysis essay asks candidates to read a non-fiction passage — usually 600 to 800 words — and explain how the author's choices build an argument. It is the first of three free-response essays on the exam, sitting in Section II after the multiple-choice block. Scoring is analytic: two College Board readers each award 1 to 4 points on four rows, so the ceiling is 8 per essay and roughly 24 points across the three FRQs combined. Understanding exactly what those rows measure is the single highest-leverage move a candidate can make before May.
Most students entering the exam treat the rhetorical-analysis essay as a literary close reading. They reach for tone, mood, figurative language, and imagery as if the passage were a poem. The College Board course description, however, frames the task as argument analysis: writers are read for what they are doing with an audience, not for what they are being in isolation. In my experience, the candidates who miss a row are not the ones who fail to spot devices; they are the ones who name devices without connecting them to a claim. The reader's pencil hovers whenever a paragraph says "King uses anaphora" and stops there.
A second common misread is treating SOAPSTone — Speaker, Occasion, Audience, Purpose, Subject, Tone — as a checklist. SOAPSTone is a triage tool, not an essay structure. If a student writes six body paragraphs, one per SOAPSTone letter, the essay will read as a list and the thesis will evaporate. The College Board's official sample responses model a thesis that picks one or two rhetorical features and traces their effect across the passage; the SOAPSTone elements surface only when they help that trace.
The 2020 and 2019 scoring guidelines, still cited in the current Course and Exam Description, all return to the same four rows. A useful working name for them is thesis, evidence, commentary, and sophistication — but the actual rubric language is more specific.
| Row | What readers ask | 4-point signal | 1-point signal |
|---|---|---|---|
| Thesis / claim | Is there a defensible reading of how the argument works? | Names 2+ rhetorical choices and the effect on the audience | Restates the prompt or restates the topic |
| Evidence | Does the writer cite specific words, phrases, structural moves? | Quotes or paraphrases with line-level precision | General references ("early in the passage") or no references |
| Commentary | Does the writer explain why each piece of evidence does work? | Connects choice to audience, purpose, or claim | Stops at identification |
| Sophistication | Does the essay demonstrate control of nuance, complexity, or style? | A warranted qualification, a reader-aware aside, or a controlled stylistic choice | Binary language, no awareness of audience's possible pushback |
The trap is reading these rows as a hierarchy where evidence outweighs commentary. They are independent: a 4 in evidence and a 2 in commentary still averages down. In practice, evidence is the easier row to teach, so weak students over-invest there; the 5-rated sample responses on the College Board site show commentary taking up roughly 60 to 70 percent of the body-paragraph word count.
Trap one: Speaker as biography. Candidates write "Lin-Manuel Miranda, a famous playwright, uses…" and lose 90 seconds to material that earns no rubric credit. Speaker means the persona constructed inside the text — the voice, the role, the relationship to audience. A student who can re-state that the speaker is "a child describing a parent" is using SOAPSTone; a student who tells the reader Miranda's birth year is not.
Trap two: Audience as "the reader." "The audience feels persuaded" is a near-empty claim. The rubric rewards a specified audience — voters in a midterm, a school board, readers of an op-ed page. When a student says "ordinary Americans," the reader's mental flag goes up only if the essay then traces a feature (a statistic, a second-person address) that is tuned to that group. Otherwise, the claim reads as filler.
Trap three: Tone as a single adjective. "The tone is hopeful" scores a 1. "The tone shifts from indignation to measured concession in paragraph three" scores a 4. The 2019 sample response on the College Board site tracks tone across the passage rather than labelling it once. For most candidates I tutor, the upgrade from a 3 to a 4 on the sophistication row comes almost entirely from giving tone a trajectory instead of a label.
Imagine the first sentence of an essay: "In her 1963 letter from Birmingham Jail, Martin Luther King Jr. uses pathos effectively." Now compare it with: "King's letter recasts the moderate white clergyman as a reader who has already conceded the moral ground, and the rhetorical work of paragraphs 25 to 32 is to keep that concession from collapsing." Both sentences quote the prompt's source. Only the second names a rhetorical move — recasting the addressee — and predicts where the evidence will sit. The reader has 90 to 120 seconds to score the essay; a thesis that previews the paragraph map is the single fastest way to keep that reader with you through 800 words of body.
The College Board allots 40 minutes per rhetorical-analysis essay, with 15 minutes of reading time shared across all three FRQs at the start of Section II. A reliable split inside the 40 minutes looks like this:
If the second body paragraph runs past 11 minutes, the student should cut the planned third body paragraph down to a paragraph-and-a-half rather than rush the conclusion. Readers are explicitly trained to weigh the conclusion less heavily than the body, but a missing conclusion reads as a missing essay. Time-on-task variation matters: candidates who hold themselves to a hard 11-minute paragraph cap routinely land within 90 seconds of the timer, while those who "write until it feels right" typically overrun by 4 to 5 minutes and rush the last page.
The single highest-yield exercise is to take a released passage, write a timed essay, and then score yourself against the four rows before reading any other source. A scoring grid on a single page of paper — one row per rubric item, two columns (claim / evidence) — surfaces the real weak row within two practice cycles. Most candidates discover that they are evidence-strong and commentary-light, which is the inverse of what the 5-rated sample responses show. In a one-to-one programme, I would re-mark three of the student's timed essays and isolate the commentary deficit; the fix is rarely more evidence, and almost always more verbs that name rhetorical action.
Two official sources are worth sitting with: the 2019 rhetorical-analysis sample set and the 2020 updated scoring guidelines. They disagree on tone, not on the four rows, and that disagreement is itself useful training for the Warranted qualification slot of the sophistication row. Reading the released reader commentary teaches the student what the marking scale actually rewards, which is rarely the same thing as what a classroom teacher rewards in a formative essay.
AP English Language's rhetorical-analysis essay rewards a small set of moves executed consistently: a two-feature thesis, evidence tied to line-level language, commentary that names the rhetorical action, and a sophistication move that handles a likely counter-reading. Candidates who practise against the four-row rubric, not against a generic literary-analysis checklist, tend to find the second point on the commentary row within two or three timed cycles. The next step is a planned set of three timed essays on non-fiction passages, each followed by a self-score on the four rows.