A narrator says a visit lasted only a few minutes. Later, another detail suggests that the afternoon had passed. Is the narrator concealing something, remembering badly, or speaking loosely? A discrepancy creates a question before it supplies an answer.
Calling a narrator unreliable can be useful, but it can also end the investigation too early. Try describing the particular limit on the account instead.
Separate knowledge from judgment
A character may report an event accurately while misunderstanding another person’s motive. They may also lack information the reader receives elsewhere. Mark whether your uncertainty concerns what happened, why it happened or how it should be judged.
For an invented example, a narrator sees a friend leave without speaking and concludes that the friendship is over. The departure is an observed action; the conclusion is an interpretation. The story may later confirm it, complicate it or leave it unresolved.
Collect a pattern, not one suspicious sentence
Write down a few moments where description and explanation pull apart. Look for repeated omissions, confident guesses or revisions to an earlier account. A single contradiction may have a local explanation, so resist making it carry the whole argument.
Include a moment where the narrator appears candid or self-correcting. Evidence that challenges your theory can help you make the theory more precise.
Ask what your access depends on
Do you ever see events outside the narrator’s presence? Are you reading a diary, a retrospective account or a voice speaking as events unfold? These questions clarify what kind of knowledge the narrative offers without requiring you to diagnose the character.
Purdue OWL’s guidance on close reading emphasises testing interpretations against details and the work as a whole. Use that approach to keep your suspicion open to revision.
Try a final sentence that names the limit: “The narrator notices actions carefully but treats other people’s motives as settled.” That gives a reading group something concrete to discuss. Our ending comparison explores how different versions can alter what an audience knows.
