Miss Scarlet, With the Candlestick, In a Murder Mystery

One of the main rules of writing a mystery, present in almost every how-to book you’ll come across, is this: the reader must feel that he could’ve solved the mystery along with the protagonist, if he were as smart as she is. Now, this is very similar to another potential rule of mysteries: the reader must be able to solve the mystery along with the protagonist, if he is as smart as she is. The distinction between the two rules is slight — but important. Since this is a blog entry, I’m not going to pretend that I’m not spending the next thousand or so words discussing the difference. So let’s dive in!

What made me think of this right now, apart from some brief discussions with friends, was reading one of Jeffery Deaver’s Lincoln Rhyme thrillers. The Cold Moon, in particular. Now, it may have just been me, but I felt that the strings were a lot more visible in this installment than some of the previous ones, so to speak. In particular, I noticed that Deaver often ratchetted up the tension by doing something like this:

scene 1 – bad guy’s POV, ending on the note of “Ahahaha, I’m going to do something horrible to a sympathetic VICTIM, something you, the reader, know the good guys don’t suspect at all!”

scene 2 – good guys’ POV, discussing something the reader knows is unimportant while ignoring the clue that might lead them to the bad guy’s new plan, ending on a note of “I know what we can do! This other thing [that coincidentally will leave the bad guy open to do exactly what he was planning last scene]!”

scene 3 – victim’s POV, wandering deeper and deeper into the bad guy’s planned trap, until — surprise!!!! Right after we cut away from the good guys, super genius Lincoln Rhyme figured out the significance of that clue everyone was ignoring, and he sent the cops to arrive in time! Or the person we thought was a bad guy is actually a good guy with benevolent intentions! Or… nothing happens, and…

… scene 4 – bad guy’s POV, we discover that although it sounded like he was planning to get the victim in scene 1, he was actually planning something different elsewhere

I guess it’s clear from my tone that this set-up stopped working for me after the second or third time, but what’s maybe not so clear is that I wondered why. This isn’t something new for Deaver — such a sequence of events happens once or twice every case for Lincoln Rhyme. And up until now, I haven’t really given it a second thought. But suddenly, in the middle of this one, it felt cheap. What’s more, it felt retroactively cheap, like, hey, he’s been doing this all along in every single book. Why didn’t I notice before?

Part of it is that in other books, about half the time we get scene sequences like this one, and the other half of the time, in scene 3, the bad guy really does do what he was planning in scene 1. This makes everything feel less cheap because you know the author actually will go there and kill off some innocent character in a horrible way. In fact, the sense of disappointment sets in only once the author has done the “false alarm!” version two or three times in a row, and you finally realize that no one is actually going to die/get assaulted/have the s*** beaten out of them.

But the other part of it is this whole thing about mysteries having to feel fair. Look, we have another word for mysteries that actually are fair: puzzles. As Terry Pratchett has his character Commander Vimes point out, a lot of the “deduction” that takes place in fictional mysteries — when Sherlock Holmes realizes the significance of the dog not barking in the night-time, or Nero Wolfe asserts that a woman in distress would never say X, Y, or Z — would be extremely dubious in the real world. Sure, you can draw some conclusions about people based on the way they act or appear, but the truth of the matter is that sometimes people do things which aren’t normal for them because of strange circumstances. (Pratchett’s example, essentially: maybe some guy is wearing certain clothes because he happens to have been DIY-ing his deck that day, not because he’s a carpenter/painter etc.) The only way a reader can “really” follow along with the detective is by making some tacit assumptions about the world of the mystery based on the knowledge that this is, in fact, fiction.

Think of Encyclopedia Brown, a poster boy for actually-fair mysteries: yeah, good luck convicting a con artist on the basis that he mentioned a bunch of pigs looking up at his plane as he flew over but pigs can’t look up. Oh, you touched the car hood and it was hot, so you know the suspect must be lying about having spent the whole afternoon at home? Good for you. That is just one tiny mote of information, and to build your case, you’re gonna need a mountain.

Mysteries where you actually can solve the crime based on the information in the story necessarily come with a set of restrictions on the world of the story. People who read Encyclopedia Brown or Five Minute Mysteries go into the experience with the knowledge that they’re really playing a game of Where’s Waldo, except with knowledge or logic instead of images. Find the single contradiction or mistake — knowing that there’s only one and that it’s key to the solution of the mystery. Know exactly which aspects of reality it’s OK to disregard because the story hinges on a single trick.

Encyclopedia Brown is a great example of showing how the conventions of certain genres (and knowledge of those conventions) affect how the reader experiences the “solubility” of the mystery. If you picked up an Encyclopedia Brown collection without knowing or being able to figure out how stories like that work, you’d quickly get frustrated trying to pick out the one fact the author decided is relevant from what might seem to you to be a myriad of other good reasons for suspecting certain characters: “What about physical evidence? Why isn’t it more likely that the guy just made a mistake in what he said and the other suspect with the motive and opportunity is still the culprit?” etc.

Clues in a mystery have to do double-duty: they have to work logically within the world of the story, and they have to send genre-coded signals to the reader about what to expect. In “The Adventure of the Speckled Band“, the bell pull and the mention of the “gipsies” both make sense in terms of the story — the bell pull serves a purpose that only the bell pull might do, and it makes sense that Victorian gentlemen of Holmes and Watson’s time and upbringing might consider those of certain backgrounds to be suspicious. However, the presence of the gipsies and the description of the bell pull also signal the reader in certain ways. The way in which the gipsies are mentioned hints to any reader who’s familiar with mysteries that they will not be the eventual culprits — such a solution would be too simple and easy. Equally, Watson and Holmes’s lingering on the bell pull, noting puzzling things about it without actually explaning any, shows that it’ll be important to the real solution of the mystery. In fact, speaking as someone who’s written even just the few mysteries that I have, one of the hardest problems I face as a writer is making clues just prominent enough that the reader will remember I mentioned them but just inconspicuous enough that it’s not like I’m bashing him or her over the head with a sledgehammer labelled, “A Clue!”

And I think it’s this second purpose — this needing clues to fit into the genre paradigm — that strongly contributes to a reader’s sense of “I could’ve solved this myself, if I’d been as smart as Sherlock.” Going back to the Lincoln Rhyme example from above, all the logical clues are in place. We really do know as much as Rhyme knows at any point in time — more, in fact, because we’ve been inside the bad guy’s head. And, heck, in the book I was describing above, the reader can predict what’s going to happen, based solely on the fact that the last-minute twists start to follow a pattern (“Aha, he’s doing that thing again where he wants me to think that woman will get murdered, but she definitely won’t because the killer’s actually trying for something else… aaaaand, nope, she didn’t.”)

But I still felt that sense of “unfair, unsolvable!” that I would’ve got if I read a Sherlock Holmes story where at the last minute, Holmes pulls out a heretofore unmentioned fingerprint and explains that although Watson mentioned nothing about it, he found it and matched it with one of the suspects around page 5. When sometimes the victim dies and sometimes the set-up is a bluff, the fake tension is still just as much a cheat for the reader — it violates the major convention of mysteries is that the author should never hide a detail the reader would notice if he or she were physically present in the world of the story, be it by cutting away from a scene before the important stuff happens, by doctoring a character’s thoughts in an artificial way so they don’t mention the one crucial detail, or by neglecting to describe certain aspects of the setting. But this way the reader doesn’t come away with the sense that the author’s wussing out by trying to cultivate a sense of threat in a story where there is none. Part of what’s expected in forensic thrillers is that some — not all, but at least one — sympathetic victims are going to get it.

So maybe this is why I feel that that golden rules of books and courses and articles really translates into something most would-be authors, myself included, find most difficult of all: read extensively in the genre you want to write. Because the conventions of stories are something you pick up through experiencing stories as a reader. Discussing them as a writer or as a critic helps me articulate why things work or don’t work in precise, technical terms, which is helpful when I’m trying to figure out what’s wrong with my story and how I can fix it. But knowing that something’s wrong in the first place is a matter of experience with how different stories make me think and feel.