When Red Herrings Stink

I'm going to go out on a limb and say something that may be controversial, at least among writers: Readers should understand why a red herring (something that is said or happens in a novel or story that leads the reader to a false conclusion) was not the solution to the puzzle by the time the tale is over.

Until recently I didn't think this was a controversial opinion. I thought it was a standard approach to writing mysteries. Sure, I'd sometimes heard authors say before that they didn't need to explain by the end of their stories why Character X said Y because Y was a red herring, but I thought they were mistaken, and since I wasn't their teacher, it wasn't my place to correct their misguided notion. But recently I edited a story by an author I respect, someone who's a solid writer, and the issue arose. Since I was this person's editor, it was my job to say my piece.

I'm going to talk about the story, but I'm completely changing the names and plot so that you can't identify the author because who this person is doesn't matter. In the whodunit story, Princess Consuella tells Annie the Amateur Sleuth that murder suspect Bad Bad Leroy Brown lied about something, based on personal observation, and therefore, it seems, Leroy must be the killer. Princess Consuella was believable and seemed absolutely certain, so I suspect most readers would have finished that scene believing Leroy had indeed lied and thus must have been the killer. It's what I thought. Yet at the end of the story, I learned I'd been fooled. Leroy may be bad, but he never killed anyone--at least not in that story.

I raised the problem with the author--that no explanation of Princess Consuella's statement about Bad Bad Leroy Brown was provided by the story's end. Either Leroy did lie (which by the story's end didn't seem right, since we never learned any reason Leroy would have lied about the issue in question) or the princess had been wrong (but how could that have been true, since she had seen with her own eyes the thing she was certain Leroy lied about, and it wasn't the type of thing that could have been misunderstood, and she had no reason to lie, either). The reader would be left wondering how to reconcile this situation, so some  explanation should be provided, I said. The author pushed back, saying that no explanation was necessary since it was a red herring designed to fool the reader into thinking the wrong suspect was the killer. The reader learns who the actual killer is by the end, and that's what matters, the author said; we don't need to revisit the red herring. 

That response prompted me to do some research about red herrings. Had I been wrong all these years? Did red herrings, by their very nature, not require explanation? To my surprise, I found nothing addressing this issue. There are a lot of articles about crafting solid red herrings, but I found nothing addressing the idea that red herrings should be explained by a story's end, that the reader should be able to understand how she got fooled. Even now, some time later, I remain quite surprised, because if authors can toss in red herrings without eventually providing an explanation for them, it makes writing too easy. It feels like a cheat.

In the case of Bad Bad Leroy Brown, sure, he could have been lying for reasons the reader never learns, despite seeming to have no reason to lie. Alternately, Princess Consuella could have lied for reasons the reader never learns about or she could have been wrong, despite being so certain and giving the reader no reason to explain how she could have been so mistaken. It certainly would make life easy for authors if they could write red herrings that didn't have to be explained in the end, but I think it would leave readers with a bad taste in their mouths. That is why I believe such scenarios need to be resolved. Did Leroy lie and why? Or did the princess get it wrong and how could that be? Without an explanation, the red herring feels contrived. It makes me feel like the author was playing games with me. 

This is why I recommended the author use a little misdirection when the red herring was introduced. More specifically, I suggested that when the princess called Leroy a liar, the author should use the wiggle word "recall" in the dialogue. Notice the slight difference:

Scenario A: The princess slams her hand on the table, its sound echoing throughout the castle. "Bad Bad Leroy Brown is a liar! I was sitting right next to him in the dungeon cafe last week, and he didn't leave money for his meal on the table when he left. I wonder what else he's lying about. I bet he rips off restaurants throughout the kingdom all the time. He's a rip-off artist."

Scenario B: The princess slams her hand on the table, its sound echoing throughout the castle. "Bad Bad Leroy Brown is a liar! I was sitting right next to him in the dungeon cafe last week, and I don't recall him leaving money for his meal on the table when he left. I wonder what else he's lying about. I bet he rips off restaurants throughout the kingdom all the time. He's a rip-off artist."

In Scenario A, the reader ends the story shrugging, thinking Leroy (who has a reputation for honesty, despite his name) had no reason to lie when he said he paid for his lunch, yet the princess's adamant accusation against Leroy remains unexplained. (She too had no reason to lie and her certainty indicated she hadn't made a mistake.)  In Scenario B, the reader can go back and reread the language of the princess's accusation and think, "Oh. The author fooled me."

Here's why Scenario B works: Because (1) the reader has no reason to think the princess lied; (2) the princess seems so certain, so the reader will believe her account; and (3) the princess distracts the reader by slamming the table, muttering about what else Leroy might have lied about, and declaring that he's a rip-off artist, the reader easily could read right past the key words--the princess didn't recall Leroy leaving his payment. When the reader gets to the end of the story, she could flip back to reread the princess's accusation and think, "Oh! It was right there. She merely didn't remember it. I was distracted by her certainty. I was fooled fair and square." That's the way to make a red herring work. That's the way to make the reader feel satisfied rather than feeling played.

Alternately, the reader could learn by the story's end that Leroy did lie for reasons unrelated to the murder. If there was a good reason for his lie, especially something that worked well with the plot, then revealing both the lie and the reason for it could have elevated the story. It also could have left the reader feeling satisfied because, while she was fooled, she wasn't played for a fool. Distracting the reader into missing a key word is playing fair with the reader. In contrast, dropping a lie into the story to fool the reader without any ultimate explanation isn't playing fair,  not to me, at least.

So that's my advice about red herrings. If you're going to use them,  make sure they're explained by the end so they don't seem contrived. Otherwise, you're taking an easy way out and you're not playing fair with the reader. Just like fish that sits out too long, that approach stinks.

I welcome your comments on this issue. And if I'm wrong and there are tons of articles addressing this subject and I need to brush up on my research skills, please share that information too.

***

In other news, here's a little BSP: I recently had a new short story published. "An Inconvenient Sleuth" appears in issue eight of Black Cat Mystery Magazine. In this whodunit, Kendra Silver, Dogwood Valley's celebrated amateur sleuth, is murdered. Who saw that coming? Certainly not anyone who thought  Kendra was invincible because she led a cozy life in a cozy town. But now that someone has killed Kendra, her best friend, Whitney, feels compelled to help the police unmask the culprit.  

Black Cat Mystery Magazine is available in trade paperback and ebooks from all the usual sources. You also can buy it directly from the publisher, Wildside Press, by clicking here.

Hiding the Ball

by Steve Liskow

If you read or write police procedurals, you probably know far too much about fingerprint patterns, blood spatter,

decay rates, DNA matching, ballistics and digital technology. Modern law enforcement relies on forensic evidence to solve crimes, and it works, which is all to our benefit. But it reduces the human (read, "character") factor in modern stories. I can't avoid them altogether, but I try to rely on them as little as possible.

Why, you ask. OK, I'm not a Luddite (although I do write my early notes with a fountain pen) but...

Readers want to participate in your mystery. The stories from the Golden Age--back before you and I were even born--required that the sleuth share his or her discoveries with the reader so we could figure out the solution (or, more typically, NOT) along with him. That's why so many of the classic stories of Agatha Christie, Nero Wolf and their peers had a sidekick as the narrator so he didn't have to give the sleuth's thoughts away. It also explains why those stories are so convoluted and complicated. The writers did what attorneys now call "hiding the ball," burying the real clues in mountains of red herrings, lying witnesses, contradictory information and complicated maps, not necessarily drawn to scale.

The Ellery Queen series featured the "Challenge to the Reader" near the end of the book, stating that at that point the reader had ALL the necessary information to arrive at the "One Logical Solution." It was a daunting challenge that I think I met only once or twice. Agatha Christie said she did her plotting while doing household chores. I'd like to see the banquets she prepared to come up with some of Poirot's herculean feats.

If you withhold the clues and pull them out at the end like a rabbit out of a top hat, readers accuse you of cheating. I still remember an Ellery Queen novel that solved the murder of a twin brother by revealing at the end that there were actually triplets. Tacky, tacky, tacky.

You need to put the information out there where readers can see it, but without making it too obvious.

Magicians accompany their sleight of hand with distractions: stage patter, light and smoke and mirrors, scantily clad assistants, and anything else that will make you look that way instead of at them while they palm the card or switch the glasses. And that's how you do it in mysteries, too.

There are a few standard tricks we all use over and over because they work.

The first is the "hiding the ball" trick I mentioned above. If you describe a parking lot with twelve red Toyotas, nobody will notice one with a dented fender or an out-of-state license plate. The B side of this is establishing a pattern, then breaking it. Often, the sleuth believes that oddball is a different culprit and not part of the same case, but he finally figures out that it's the only one that matters and the others were decoys.

You can also give people information in what retailers used to call a "bait and switch." Stores would advertise an item at a low price, then tell customers that item was already sold out and try to sell them a more expensive version. You can give readers information about a person or event, then tweak it later so it points somewhere else. The classic police procedural The Laughing Policeman hinges on a witness identifying an automobile parked at a scene...then years later realizing that a different car looks a lot like it. Oops.

You can give information and later show that the witness who mentioned it was lying. The trick here is to plant a reason for the witness to lie early in the story and leave the connection until later on. If the reader sees the reason with no context, he'll overlook it until you make it important again when you pull the bunny out of the derby. This is one of my favorites.

I also like to focus on a fact or circumstance that's irrelevant and keep coming back to it. Later in the story, your detective can figure out that it's meaningless...OR realize that he's look at it from the wrong angle. My recent story "Look What They've Done to My Song, Ma" has PI Woody Guthrie and his musician companion Megan Traine trying to clean up a music file so they can identify the voice that's talking underneath the singer. It's not until late in the story that Guthrie realizes the voice doesn't matter--at least, not the way he thought it did, because the speaker isn't the person everyone assumed it was. That story appears in the July/August issue of Alfred Hitchcock's Mystery Magazine, along with stories by known accomplices John Floyd and O'Neil De Noux.

The "how could he know that?" clue gets lots of use, too. Someone makes a comment, suggestion, or observation and the sleuth doesn't realize until later that he couldn't have known the murder weapon or condition of the body unless he was there. If you haven't used this one at least once, raise your hand.

Its second cousin is the condition or event that didn't happen, the old case of the dog that didn't bark in the night, first used in "Silver Blaze," an early Sherlock Holmes story. Its fraternal twin, which makes a handy red herring in the age of technology, is a missing computer file. Since it's missing, we don't know what's on it...or if it's even important. The sleuth can spend pages or even entire chapters worrying about that missing file folder or computer. In The Kids Are All Right, I had two murder victims whose computers were found with the respective hard drives removed. The implication was that missing files would implicate the killer. But can we really be sure?

Now, what's your favorite way to deal off the bottom?