Character analysis is one of the most common assignments for All the Light We Cannot See, and one of the hardest to grade fairly. Students know the plot well because the characters are memorable, so their essays are full of accurate detail. The problem is that accurate detail is not the same as analysis, and many essays never move beyond describing what Marie-Laure and Werner do.

Grading these essays well starts with separating three layers of thinking. The first is description, which reports events and traits. The second is interpretation, which explains what a character's choices reveal, and the third is evaluation, which considers what the author wants the reader to conclude about that character. Most students land in the first layer, and a good rubric shows them how to climb.

It helps to give students a concrete example of each layer. Saying that Marie-Laure memorizes the model of Paris is description, while explaining that the memorization turns her father's love into a usable skill is interpretation. Showing that contrast in class or in written feedback makes the abstract expectation visible, and students often produce stronger paragraphs immediately afterward.

Assessing Development Over Time

Both protagonists change across the novel, and strong essays track that change with evidence from different points in the timeline. A student writing about Werner might contrast his early fascination with radio at the orphanage with his later discomfort at the training school, using specific moments to show the shift. Essays that cite only one part of the book usually present a flat character instead of a developing one.

• Evidence drawn from at least two distinct stages of the character's story
• Explanation of what causes the change rather than only stating that it happened
• Attention to how other characters, such as Etienne or Frederick, shape the protagonist
• Awareness of the author's technique, including how alternating chapters frame each character
• A conclusion that says what the character's journey suggests about the novel's larger ideas

Giving Feedback on Common Weaknesses

One frequent weakness is the character list, where a paragraph strings together traits such as brave, curious, and kind without any supporting scene. Feedback should ask the student to choose one trait and prove it with a moment from the text. Another weakness is moral simplification, especially with Werner, where students label him as either good or bad and ignore the ambiguity the author builds into his choices.

Feedback on these issues works best when it is phrased as a question that opens the analysis. Asking "What does Werner's silence in this scene suggest about how he sees his own responsibility?" invites the student to think, while a comment that says "needs more depth" invites nothing. Precise questions like these are also easier for AI tools to generate consistently when the rubric describes what depth looks like.

Comparing the Two Protagonists Fairly

Many assignments ask students to compare the two main characters, which adds another layer of assessment. Strong comparisons do not simply alternate between the characters but organize the essay around a shared idea, such as how each one experiences isolation or how each relies on a different sense to understand the world. That organizing idea should be visible in the thesis and carried through every paragraph.

Watch for essays that give one character three pages and the other one paragraph. Balance is a legitimate criterion, and it often reveals that the student understands one storyline much better than the other. A note pointing this out helps students see where their knowledge or evidence needs to be strengthened before the next draft.

Making Grading Manageable at Scale

Character essays tend to arrive in large batches at the end of a unit, and the temptation is to grade quickly and rely on general comments. Using a rubric with clear levels for description, interpretation, and evaluation gives structure to that speed. AI-assisted grading can apply those levels consistently and draft feedback that names the specific layer a student has reached.

Teachers can then focus on the papers that need the most human attention, such as essays with unusual arguments or students who have plateaued. The goal is to protect the time that matters most, which is the thoughtful conversation about a student's ideas. Everything else in the process can be streamlined without losing quality.