Few books make narrative voice as unavoidable as Ljudi govore, whose title alone tells readers to listen to who is talking. Students usually notice this, but noticing is not the same as analyzing. Many essays stop at saying that the book contains many voices, which is an observation rather than an argument, and teachers need a way to push students past that first step.

The first thing to assess is whether the student can tell the speakers apart and explain how the text lets us do so. That might involve vocabulary, rhythm, the things a speaker chooses to mention, or what a speaker refuses to explain. An essay that treats every voice as interchangeable is missing the most basic feature of the book, and your feedback should say so directly.

The second thing to assess is whether the student connects voice to meaning. It is one thing to say that a speaker sounds weary or matter-of-fact, and another to argue that this tone changes how the reader weighs the events described. Strong essays make that second move consistently, while weaker ones leave the tone as a label attached to a quotation.

Separate Description From Interpretation

A useful grading habit is to mark each sentence in a body paragraph as either describing or interpreting. Description tells us what a speaker says or how they sound, while interpretation explains what that choice does for the book as a whole. If a paragraph is mostly description, the student likely needs a prompt such as "so what does this voice make us understand that a neutral narrator would not?"

• Does the essay name specific speakers or moments rather than speaking of voices in general?
• Does it point to concrete language features such as repetition, plain diction, or abrupt endings?
• Does it explain the effect of hearing events secondhand through ordinary people?
• Does it consider what the voices leave unsaid?
• Does the argument about voice shape the whole essay rather than one paragraph?

Watch for Confusing the Speaker With the Author

A common error in voice essays is attributing a speaker's views directly to Petrović, as if every statement were the author's own belief. Teachers can address this by asking students to distinguish between what a character says and what the book as a whole shows. A single sentence of feedback, such as "whose opinion is this, and how do we know the book endorses it?", often corrects the habit.

The error is more than a technicality, because it flattens the layered effect that makes the book interesting. If students believe each voice is the author speaking, they cannot explore irony, contrast, or the gaps between what people say and what the reader senses. Treating this confusion as a priority comment pays off across the entire essay.

Give Credit for Honest Uncertainty

Students working with a demanding, unconventional text will sometimes write that they are unsure what a passage means or who is speaking. A rubric that penalizes every hint of uncertainty pushes them toward confident-sounding nonsense. Instead, reward essays that acknowledge ambiguity and then reason carefully through the possibilities, since that is how experienced readers handle difficult prose.

You can build this into your criteria by describing the top level as one where the student "reasons through ambiguous moments rather than avoiding them." That wording tells students that grappling with difficulty is valued. It also helps graders, who otherwise may mistake a bold but unsupported claim for confidence worth rewarding.

Make Feedback Faster Without Making It Generic

Voice essays tend to share a few predictable weaknesses, which means a bank of well-written comments can cover most of what you need to say. Pair each comment with a short revision task, for example asking the student to rewrite one paragraph so that every sentence interprets rather than describes. Concrete tasks like this are far more useful than a general encouragement to dig deeper.

AI-assisted grading can help here by flagging where an essay relies on description and by drafting feedback tied to your voice-analysis criteria. Teachers remain the final judges, but a first pass that highlights patterns across a whole class can show you which skill to reteach before the next assignment. That turns grading from a one-way verdict into information that improves instruction.