Teachers who assign Yevgeny Zamyatin's We often discover that the novel produces essays unlike anything else on the syllabus. Students arrive with strong opinions about surveillance, conformity, and personal freedom, but those opinions rarely come with the textual evidence needed to support them. Grading becomes difficult because the book is strange, fragmented, and written as a series of numbered journal entries that resist simple summary. A clear grading approach helps teachers separate genuine analysis from confident but unsupported claims.

Start by deciding what the essay is actually assessing before you read a single paper. In a unit on We, the most common targets are thesis quality, use of textual evidence, understanding of the One State's logic, and control of the narrator's shifting perspective. Each of these deserves its own rubric row so that a student who writes beautifully but ignores the text can see exactly where points were lost. Without that separation, comments tend to blur together and students cannot tell what to fix.

The novel's structure creates specific grading traps that are worth anticipating. D-503 narrates in the first person and slowly changes his mind, so students who quote his early confidence as proof of the author's view will misread the book. Others treat the One State as simply evil without noticing that D-503 genuinely believes it is rational and beautiful at the start. Marking these misreadings consistently, rather than case by case, keeps your feedback fair across a stack of one hundred papers.

Build a Rubric Around the Novel's Strangest Features

A workable rubric for We usually has four or five rows, each with descriptors that name what strong work looks like in this particular book. For evidence, a top-level descriptor might require students to cite specific records, such as the early entries about the Table of Hours or the later ones about the Great Operation. For analysis, it might require that the student explain why D-503's mathematical vocabulary matters rather than simply noting that he likes numbers. Specific descriptors like these make scoring faster and make written comments easier to write.

• Thesis: makes a debatable claim about the One State, D-503, or the novel's argument about freedom
• Evidence: uses specific moments from the records rather than general plot summary
• Analysis: explains how word choice, math imagery, or narration shapes meaning
• Context: connects the novel to its 1920s setting without turning the essay into a history report
• Control: organizes ideas clearly, integrates quotations smoothly, and follows citation expectations

Score Evidence and Analysis Separately

Many essays on We quote the right passage and then stop, leaving the quotation to speak for itself. A grader who gives full credit for the quote hides the real weakness, which is the missing explanation. Separate scoring lets you reward the student who found a good moment, such as D-503's discomfort with the imaginary square root of negative one, while still asking what that discomfort reveals about his mind. Students improve fastest when they see that finding evidence and explaining it are two different skills.

It also helps to keep a short list of the quotation habits you see most often. Common problems include dropping a quotation with no lead-in, quoting an entire paragraph when one phrase would do, and attributing translated wording to Zamyatin without acknowledging the translator. Because several English translations exist, remind students to cite the edition they actually read. A quick note in your rubric about edition-specific wording saves you from arguing over whether a quotation is accurate.

Handle the Narrator With Care

The journal format means that every claim in the novel passes through D-503's changing mind. Strong essays treat him as a character whose reliability shifts, noting where his tone moves from rehearsed certainty to confusion as I-330 enters his life. Weaker essays flatten him into either a villain or a victim. In your comments, ask students to name which stage of his thinking a quotation belongs to, since a line from the first records means something very different from one near the end.

Grading this skill well requires a shared reference point, such as a short timeline of D-503's development across the forty records. Some teachers hand out a one-page version of this timeline before the essay is due so that students can place their evidence accurately. Others keep it as a private scoring aid for themselves or for colleagues who share the grading load. Either approach reduces the chance that two graders reward the same paragraph differently.

Make Feedback Fast and Consistent at Scale

A single class set of essays on We can take a teacher many hours, and the quality of comments usually drops as the stack shrinks. The tenth paper tends to receive a thoughtful paragraph of feedback, while the ninetieth receives a score and a few words. One way to protect quality is to write a bank of comments tied to rubric rows and the novel's recurring problems, then customize each one with a specific detail from the student's own paper. This keeps the feedback personal enough to be useful without requiring a fresh essay of commentary every time.

AI grading tools can support this process by applying a teacher's rubric to each essay and drafting feedback that references what the student actually wrote. The teacher remains the final judge, reviewing scores and adjusting comments where the tool misses nuance, such as a clever but unconventional reading of the Green Wall. Used this way, the technology shortens the repetitive part of grading and leaves more time for the conversations about interpretation that make a novel like We worth teaching. The result is faster turnaround and feedback that students can actually use for revision.