By Elizabeth S. Craig, @elizabethscraig
Mystery writers that I’ve met tend to fall into a couple of different groups—writers who have picked their killer before they start writing their story (or early in their draft) and those who decide by the end of the book who the killer will be.
I’m a fan of picking my murderer at the end of the book. The only problem is the fact that I’ve (fairly recently) started outlining. When I outline, I lay out the murderer…but I leave it open to change. For my Penguin books, however, the outline goes to my editor so that she has an opportunity to weigh in. I’ve changed the killer a couple of times for my editor and once I dramatically changed a character’s personality and manner in order to keep the character as the murderer.
My editor and I agree that we want the murderer to be a surprise to as many readers as possible. Although some readers always seem to guess the killer’s identity.
My considerations when choosing the murderer:
Will the readers guess this killer easily?
Does this choice make sense? Is the motive believable enough?
Are there enough clues pointing to this choice to keep the mystery fair?
Does this character play a big enough role in the book to make the mystery fair?
Even if the answers to one or more of these questions is no, I might venture ahead with my plan—but I’d make revisions to make it work. I can ramp up the motive or add an additional motive. I can tweak the character’s personality to make it seem as though he has a quick temper or is especially sensitive to criticism, or is overprotective of family members. I can add clues and distract from those clues. And I can add scenes to make the character play a larger, more vital role in the story.
I’ve usually already included one surprise earlier in the book—I murder the most likely suspect halfway through the book. Out of my 4 or 5 suspects, one ends up dying and another will be the murderer. I throw out red herrings to suggest other suspects as the murderer and work to make the red herrings fair instead of frustrating. I have the sleuth recognize them as false leads before going on too long.
Another (large) red herring can be made by misstating the true motive in the case. Maybe your sleuth has been focused on financial gain but the real motive is revenge. Or a crime of passion. Or misplaced self-defense. This can be a useful way to distract attention from your murderer.
Reasons I’ve changed a murderer for my editor:
I changed the murderer once because my editor felt the killer (the victim’s husband) was the most-likely choice. Sadly, in real life most people are killed by those closest to them.
I’ve changed a murderer once because my editor felt the character was too unlikeable…and that readers would guess she was the killer because they wanted her to be the killer.
It might sound difficult to have to change the killer (especially in a completed draft, which I’ve done on several occasions), but it isn’t. After all, all the suspects have equal motive, means, and opportunity…I just shift elements around so that the red herrings become clues. And, for that to happen, I need to drop the clues more carefully than I did the red herrings. So I might have a distraction occur in the scene right after placing the clue—maybe there’s an argument between two other characters. Maybe the sleuth immediately discovers a red herring that she thinks is much more important than the real clue she just stumbled across. Maybe there’s even another dead body. At any rate, switching the killer is a pretty easy fix.
If you write mysteries, have you ever changed the murderer’s identity? How do you select the killer for your books?
Image: MorgueFile: Pippalou
I always try to think WWAD (What would Agatha do?). She went with psychology, and that, I think, gives you some flexibility. To really work as a suspect, the most important characters are going to have to have a hidden layer to them. Several of them will certainly have secrets. So it isn’t all that hard to say “you know what, I’ve burned my steps here. The audience already knows enough to solve the case with this character and I’ve still got five chapters to go…. I guess it’ll turn out to be that one!”
I suppose, in that sense, I split the difference: I like to start the story with a good idea of what’s behind the crime. However, I am always alert to better turns and twists, and building up secondary characters who are in on the plot.
I also almost always use a double-crime. (At least in a full-length mystery.) That’s an Agatha Christie trick too. Have a mystery/suspense plot, with a crime conspiracy of some sort driving a lot of the action, but the murder itself is actually something else, maybe something more personal springing from the main story that the detectives have uncovered. (Say, there is a smuggling ring and one of the key members is murdered, but it turns out it’s his wife who killed him because she thought all his secrecy was due to an extramarital affair.)
I have started stories before I really knew what either of the plots were, but I’ve found it takes me years to finish a story when I start that way. I do a better job with a story if I have an idea of what that “front” crime is (which may also involve a murder) and then I can make up the other as I see the characters develop.
And yes, I really have to watch that I don’t fall into the same pattern on killers so that people won’t be able to guess because the killer always turns out to be the mean person or the most charming person. (Actually, Christie often did that – made the killer someone you liked.)
Camille–WWAD….ha! Love it! I need to stick that on my laptop with a sticky note. :)
You know, I haven’t done any crime conspiracy, although I always enjoyed those stories growing up (in film and in books). Tough to do with the series I’ve got now, but maybe in the future. I do a double-crime sometimes in a different way–I’ll have two different murderers with two different motives. Although that does tend to make for a more-lengthy book, usually. Depends on what kind of a deadline I’m working with.
Oh, it would kill me when Christie’s killer was someone I particularly liked! And she did it all the time! But those were the stories where I didn’t guess the murderer. A good technique, for sure.
I think most mysteries have some kind of double-crime going on, but it’s often less obvious than a crime conspiracy. I used that because it was what was in my mind. But a quieter version might be murders that cover up the motive — a long ago murder or an affair or cheating at something
The example I use is that there was a murder long ago, and now some murders that seem connected to it. It turns out to be the Vicar! Except than you realize he couldn’t have done the long ago crime… that was really done by his wife, and he committed the current crimes to protect her.
So when I say “crime conspiracy” that can just be another crime — by the same person or not. (But I do like when it is something ongoing and dramatic.)
I guess if there are clues pointing at everyone, then changing the killer isn’t that difficult.
Alex–A piece of cake, actually. Probably the easiest revision to make.
Elizabeth – That’s a really interesting question. I can see in both of those situations why you would have changed who the killer is. For what it’s worth, when I plan a story I start with the victim. And most of the time there’s more than one person who could have had a motive. So at the beginning of my planning I don’t always know who the killer is. I do choose fairly soon though so that I can build up the plot logically. I did change the killer once, but it’s a good thing I did. If I’d left it as it was the whole story would have been too implausible because my first choice of killer simply didn’t have a powerful enough motive to go as far as murder. Sometimes making a big change like that makes a book better.
You and I plan stories the same way. The most important person in the mystery is the victim, for sure.
Isn’t amazing how a major change can really save a book?
Hiding your murderer: there’s the straightforward approach, which is used by the TV show Castle in EVERY episode: First, drag the most likely suspect (husband) in front of the reader, then alibi him out but he gives a clue to the next most likely suspect, etc. until (at least on Castle) the only other character with a speaking part is the guilty one (and the “least likely”).
Or, if you want it to be the husband, do the above but in the end a clue arises (or they re-examine a clue from from early on from a different perspective) that brings them back to the husband.
Michael–Oh, that’s too funny. I’ve only seen Castle a couple of times…not enough to see the pattern. Yes, has to be a character with a speaking part, for sure! :)
With the instance where I changed the murderer so it wouldn’t be the husband…yes, I was bummed over that for a while. I really, really wanted the husband to do it, and I felt I’d deflected enough attention from him so that he wouldn’t be the obvious murderer. But my editor was pretty adamant. When I started looking for Plan B, I did find an alternative that I ended up liking better than the husband…so it worked out well in the end. But I was prepared to say, “Hey, I’ve added some more obfuscation here to make it so that the husband is a surprise.” I didn’t have to go there, though…a good thing.
Elizabeth, I’ve never written a murder mystery, but I love learning how writers do it. My favorite line from this post? “I murder the most likely suspect halfway through the book.” Only in writer-land can this statement sound normal! It cracks me up.
Julie–Ha! Yes, and the sad thing is that I don’t think a thing about saying it. I used that wording once when speaking to a group of readers for a panel I was sitting on and was flabbergasted when laughter burst out. I had to stop speaking and totally lost my place. :)
Wow. I never thought of the evolution in this fashion.
I’m personally concerned about the consequences of a killing – frequently offstage despite my more base nature – and how it changes the story.
Now, I’m going to have to think about how changing the source of harm improves/changes my original story.
You keep helping like this and I’ll end up having to drop “aspiring” from the letterhead. It’ll be all your fault!
Jack–Oh good! I’m glad it was helpful. Sometimes I worry because I feel that the way I approach so many aspects of story is from a commercial viewpoint….it’s really permeating everything I do. I really admire Camille’s way of looking at things (first comment above) because she is very character-driven in her approaches. And that ends up being commercially successful, too.
Elizabeth– I admire good mystery writers for their ability to keep readers guessing, and throwing that guesswork into doubt. And I share Julie M’s appreciation of your device–killing off everyone’s number 1 suspect halfway through the story. But either because I’m lazy, or just not clever enough, I’ve had to develop other strategies as a mystery/thriller writer. They mostly have to do with creating interest, not in terms of whodunit, but in “what happens next?” or “who will the killer choose as his next victim?”
Barry–I think that’s a great approach, too. In fact, my editor recently had to remind me that my subplots–to the readers–matter just as much as my mystery. So I’ve got minor storylines (well, again, they’re minor to *me*, but my editor tells me they’re absolutely vital to hooking readers) with…a missing cat. But the protagonist really cares about the cat. What’s going to happen to it? She canvasses the neighborhood looking for the cat (providing an excellent opportunity for my sleuth to question suspects), etc. So even the smaller elements in our story provide chances for suspense or tension. Very good point.
You’re so good in your thoroughness. I make a list at the start and give ALL my suspects a viable motive and try to trickle that evidence, but I think all your checks are good.
Hart—I know how thorough you are…you’re doing fine!