I ran into an intriguing editing problem recently. A client had a character who was disguising the fact that she was a woman. What made it tricky was that she was the narrator of a number of scenes, so we had to construct those scenes so as to mislead both the other characters and the readers. You’d be surprised how often gender gets mentioned in interior monologue. The client and I got a lot of practice at avoiding personal pronouns without looking like we were deliberately avoiding personal pronouns.
This started me thinking about the unreliable narrator, a technique that’s shown up in a couple of my favorite books. If, as we discussed last month, writing techniques should be treated as tools, what do you use the unreliable narrator for?
SPOILER ALERT: If you’ve never read Agatha Christie’s The Murder of Roger Ackroyd, go do it now. I’ll wait.
Okay, welcome back.
As you’ve just seen, unreliable narrators are a great way to set up a surprise ending. According to a lot of commentators, Christie’s revelation that the narrator was actually the murderer was a milestone of crime drama and may well have been her masterpiece. She took advantage of the fact that readers naturally assume the narrator is telling them the truth. Writers can also give readers a shock at the end because readers assume that the narrator is alive (Ambrose Bierce’s “An Occurrence at Owl Creek Bridge”), or sane (Chuck Palahniuk’s Fight Club).
To pull off this kind of surprise without alienating your readers, the narrator has to tell the truth, at least technically. One reason the unreliable narrator is hard to write is that you’re using your readers’ assumptions to slip stuff past them that would otherwise raise red flags – like the way my client’s character never uses personal pronouns in interior monologue. Christie’s book works because, at the end, Dr. Sheppard reveals how he described the actual murder without revealing that he was the one committing it. For readers, subtle misdirection is a terrific trick, but outright lying is an insult.
In Death’s Jest Book, Reginald Hill makes use of an unreliable narrator to create dramatic tension. If you haven’t read it, Detective Pascoe receives letters throughout the book from a brilliant ex-convict named Franny Roote, whom Pascoe had once helped convict and who has since been proven innocent. The letters are apparently simple, grateful reports of the literary research Roote is up to now that he’s out of prison. But Pascoe is convinced they are mocking, carefully constructed hints at Roote’s crimes, past and present. Unless Pascoe’s just being paranoid. His growing desperation to decipher who Roote is and what he’s up to is one of the things that drives the story forward. I won’t tell you how that one resolves. Go read it — it’s brilliant.
The most satisfying plot twists not only surprise your readers and change the future course of the story. They also rewrite what your readers have already read. When readers discover that my client’s character has been a woman all along, it rearranges the motives of a few other characters and the meaning of one or two earlier scenes. When Pascoe eventually discovers who Roote really is, it finally lets readers understand what the letters they’ve been reading all along really mean. And the ending of The Murder of Roger Ackroyd essentially rewrites the book from the beginning. That is one reason it’s a masterpiece.
In The Documents in the Case, Dorothy Sayers does even more with this narrative ambiguity over who a character really is. The novel consists of letters, evidence reports, news clippings, and notes by various characters that make up the case file for a possible murder. The various documents paint Margaret Harrison, one of the central characters, as either a cold, heartless adulteress who manipulated her lover into killing her husband or a misguided and misunderstood innocent who accidentally started something that spun out of control. Who she is depends on which narrator you believe, and the tension helps drive the story forward. But in a truly audacious move, Sayers doesn’t resolve the question at the end. She does reveal how the murder was committed and by whom, but readers are left wondering who Margaret Harrison really was. Some of us are wondering still.
Unreliable narrators are most useful in mysteries, where slipping clues past readers is the point – you’ll notice all three examples so far have been mysteries. But in The Catcher in the Rye, J. D. Salinger uses the unreliable narrator for something else entirely — to create empathy between readers and Holden Caulfield. Holden is unreliable not because he is misleading readers but because he is unaware of the depth of his own alienation and how it affects what he sees. Readers feel for him because they know him better than he knows himself.
Regardless of what you use it for, the unreliable narrator puts distance between what your readers are reading and the “reality” happening behind the narrative. Since readers are always aware at some level that fiction isn’t real – disbelief is suspended, not eliminated – giving them this glimpse behind the curtain lets them sink more deeply into the world of your story. It’s tricky to maintain suspension of disbelief at two different levels. But if you can pull it off, it makes for powerful storytelling.
So what are your favorite examples of unreliable narrator? Have you ever used an unreliable narrator yourself? If so, tell us about it.
A lot of people feel Agatha Christie cheated with The Murder of Roger Ackroyd. Do you think so? Or do you know of a case of unreliable narrator that went wrong?