Rhetorical analysis has a specific failure mode the other two AP Lang essays don't. A student reads a passage, spots three techniques, names all three correctly, and still walks away with 2 out of 4 on evidence and commentary. The names were right. Nothing explained what the techniques were doing to the reader.
What this essay actually asks you to do
You're given a passage, usually a speech, letter, or piece of nonfiction, and asked to analyze how the writer builds an argument through rhetorical choices. The rubric is the same shape as the argument essay: thesis (1 point), evidence and commentary (0–4 points), sophistication (1 point). The difference is what counts as evidence. Instead of outside examples, your evidence is the text itself, and instead of arguing a position, you're explaining how someone else already argued theirs.
That distinction matters more than it sounds like it should. Argument essays reward you for building a case. Rhetorical analysis rewards you for noticing a case someone already built and explaining how the pieces work.
The device-naming trap
Here's what a 1-point evidence paragraph looks like:
"The author uses repetition when she says 'we must act, we must speak, we must not be silent.' She also uses a metaphor comparing silence to a disease. These rhetorical devices make her argument more persuasive."
Every claim in that paragraph is true. The devices are correctly named. And it earns almost nothing, because naming a device isn't the same as explaining what the device does in this specific passage, for this specific audience, in service of this specific claim.
Here's the same evidence with real commentary:
"The triple repetition of 'we must' compresses three separate demands into a single rhythmic unit, so that by the third repetition the audience experiences the sentence as a chant rather than a list of asks, a shift that makes refusal feel like breaking with a group rather than declining a request. The disease metaphor for silence works the same lever from a different angle: comparing inaction to something that spreads without treatment reframes staying quiet as actively dangerous rather than passively neutral, which raises the emotional cost of doing nothing."
Notice what changed. The second version never just names "repetition" or "metaphor" and stops. It explains the specific mechanism (compression into rhythm, reframing inaction as danger) and ties that mechanism to what the author needed the audience to feel at that exact moment.
The question that separates 1-point commentary from 4-point commentary
Ask it after every device you name: what does this specific choice do to this specific audience, right here, that a different choice wouldn't have done?
If a student swapped the metaphor for a different one and the analysis paragraph would read almost identically, the analysis hasn't actually engaged with why the author picked that metaphor over another. Strong commentary is disposable if you change the evidence. Weak commentary survives any swap because it was never really about the specific words on the page.
Building the thesis around the author's rhetorical situation
A strong rhetorical analysis thesis names the author's central strategy, not just the topic. Weak: "The author uses several rhetorical strategies to convince her audience." That sentence could sit on top of any passage assigned in any year. Strong: "By repeatedly reframing a specific policy debate as a matter of collective survival rather than political preference, the author pressures her audience to treat disagreement as a moral failure rather than a legitimate position." That names the actual strategy this specific author is running, which gives the body paragraphs somewhere real to go.
Earning the sophistication point here
Sophistication on rhetorical analysis usually comes from explaining how multiple techniques work together rather than as a list of separate, unrelated choices. If the repetition and the disease metaphor are both doing the same underlying work (making inaction feel dangerous and refusal feel like betrayal), say so directly. That's a genuinely more sophisticated read than treating each device as its own isolated point, and it's also just true to how real writing actually works: authors rarely deploy one technique in a vacuum.
What published research on automated essay scoring shows
- Stahl et al. (BEA 2024): Scoring systems that generate explicit rubric-grounded reasoning before assigning a score align better with human raters than systems judging surface fluency, the same gap between naming a device and explaining its effect that separates weak rhetorical analysis from strong.
- Doewes et al. (EDM 2023): QWK measures how consistently a rubric gets applied across raters or models; it says nothing about whether a single essay's score is correct in isolation.
- College Board AP English Language rubric: Evidence and commentary is worth four of six points specifically because the rubric rewards explaining a rhetorical choice's effect, not identifying that the choice exists.
If you want to see how a rubric-aligned score compares to an actual AP reader, FRQuick's published benchmark on human-graded AP Lang essays lands within one point of the human score most of the time (methodology).
Before you submit
Go back through every device you named and cross out any sentence that only identifies it. What's left after the crossing-out is your actual commentary; if there's not much left, that's the gap to close before you turn the essay in. Then check whether your thesis names the author's actual strategy or could sit on top of a different passage entirely.
Once that holds up, grade your rhetorical analysis essay for free and see whether the commentary actually explains effect, or just restates identification in longer sentences.
FRQuick is not affiliated with the College Board or Advanced Placement. AP is a registered trademark of the College Board.
