When looking closely at the plots of many murder mysteries one can see where a great deal of time was lost in there being mistakes in interpretation of the evidence which were made toward the start of an investigation. They can be explicit, like thinking that a clue belongs to one person when it actually belongs to someone else, but it can also be much bigger in scope—mistaking a murder for gain as murder for revenge, for example. It’s possible for the detective to spend the first half a book (or more!) laboring under this kind of mistake. It can be a useful way to spend time, and can also be the setup for the big reveal at the end which shows the detective to be brilliant.

However, it can, especially upon close inspection, easily seem a bit far fetched for the brilliant detective to get locked into an incorrect interpretation. Often the reason why people begin with one interpretation is trivial—it can be as little as someone making an off-hand suggestion, or even just someone assuming it. And, to be fair, it is the job of the brilliant detective to question all of the things that ordinary people take for granted. That said, even brilliant detectives are human. Human beings need some sort of interpretive framework to operate within, even if only held provisionally, and that framework will dictate what is and what is not conceivable. As long as the current evidence keeps the current framework plausible, it is reasonable for even the most brilliant of detectives to work within it. That is, until it stops working. That’s what’s being described by the phrase, “once you eliminate the impossible whatever remains, however improbable, must be the case.”

The other issue that comes up with early mistakes in a murder mystery is that, if it’s not handled very well, it can easily diminish the re-reading value of the story. Now, I know that not everyone values re-reading as much as I do, but I tend to operate on the principle that if a book is not worth re-reading ten times, it’s not worth reading once. (Obviously, the only way to find out is to read it once, and there are plenty of exceptions.) So how do you make the first three quarters of a book a waste of the detective’s time without making it a waste of the re-reader’s time?

The answer, I think, is to make it not be a waste of the detective’s time, but that’s not obvious until you arrive at the full solution.

That is, during the early part of the book when the detective is laboring under a misinterpretation of the crime he must still be collecting clues that will help in the solution without understanding how they are. This is a very tricky balance; it can get frustrating when all of the clues point one way and the detective won’t even consider it because of one small mistake. It’s best, if at all possible, for most of the clues discovered early to work within the framework of the mistake but to work better within the framework of the truth. A really good example of this which comes to mind is in the Hercule Poirot novel Five Little Pigs (spoilers ahead).

There is a piece of evidence given which seems to fit in under the prevailing interpretation that Caroline murdered Amyas, but it works better under the theory that she didn’t. That is the wiping of the bottle. Until Poirot introduces the interpretation that Caroline was protecting her little sister, who she believed killed Amyas, her wiping of the bottle (as well as other actions) seem to indicate her guilt. Poirot’s collecting of this information did prove useful.

I think that this forms the ideal, though of course like most ideals it is difficult to achieve.

Incidentally, what we have said above would seem to suggest that red herrings are a problem for re-readability, and yet it is obvious that red herrings can work well in a detective story. Certainly, there is a place for red herrings, but I think that they must be used like a spice—too many of them can hurt a story. But how to use them?

I think that this principle we have laid out for early misconceptions in a detective story also points to the best way to handle red herrings. A red herring which is just a waste of time will not be satisfying on re-reading—unless it serves some other purpose. That, I think, is the key. The best red herrings will improve somebody’s life when they are cleared up. Lovers held apart can come together, a missing item of value can be restored to its owner, somebody in danger can be made safe—there are many options, the key thing that makes this a satisfying red herring is that it is a bonus. That in addition to solving the central mystery and putting things right that were put wrong through the misuse of reason, some other problem was set right too.

Hard to achieve, certainly, but I think worth striving for.