This morning, there was a bit of a literary Twitter debate about Chekhov’s Gun. My own contribution to this was, as usual, a dumb pun:
(A “Gundam,” if you don’t know, is a giant robot.)
Like many internet craft fights, much of the Chekhov’s Gun debate turned on differing interpretations of the principle. (The worst for this is “kill your darlings” which people take to mean everything from “cut whatever sounds too clever” to “delete all of your favorite parts of the story” to “trim whatever is unnecessary.”) In the case of Chekhov’s Gun—roughly “if there is a gun in the first act, it has to go off in the third act”—it seems to be interpreted in three different ways.
Every single element of the story should have a narrative point and there should be no irrelevant details.
“Remember to foreshadow.” If something big is going to happen, you need to lay the groundwork for it. This interpretation basically inverts the phrasing: “If a gun is going off in the third act, remember to put it in the first act.”
Don’t make promises to the reader that you don’t follow through on. You can’t raise the stakes in a dramatic way, then drop the stakes-raising element.
The third is the best and most useful interpretation in my view. The writer Isaac Butler has a good thread on this:
In terms of what Chekhov is reported to have actually said, different phrasings support various interpretations. Chekhov did apparently once say: “Remove everything that has no relevance to the story. If you say in the first chapter that there is a rifle hanging on the wall, in the second or third chapter it absolutely must go off." But another time he said, “One must never place a loaded rifle on the stage if it isn't going to go off. It's wrong to make promises you don't mean to keep.”
Regardless of Chekhov’s different iterations, the third interpretation is the most useful principle in my view. I also think it is what he meant. Chekhov used “gun” for a reason. A gun is a dramatic element that implies dramatic consequences. It’s not Chekhov’s Sunglasses or Chekhov’s Hamburger. Chekhov didn’t say: “don’t have a can of soda in the first act that someone doesn’t drink by the third act” or “if you mention a salt shaker, someone better get thirsty by the end of the story.” It’s gun. A weapon that kills. If you see a deadly weapon in a short story, you expect it to be used… for good or ill.
Following through on the promises that a work of fiction implicitly makes with the reader is a very useful guiding principle. The opening of the story typically sets up the expectations of the story. The genre it will be in, the tone, the stakes, and so on. If you switch these things halfway through the story—having a story about a giant asteroid crashing to earth turn into a story about finding a lost cat, say, with no mention of the planet-ending rock again—the reader probably will get annoyed.
This doesn’t mean that stories have to be overly neat and tidy. And the gun doesn’t have to “go off” in a literal way. Perhaps the gun jams, or perhaps the police rush in and arrest the protagonist for possession of an illegal firearm. (Although the police actually cracking down on guns in America perhaps shifts this story to the fantasy genre…)
In any event, you want to have the dramatic events feel “earned” while also being original. There’s another craft principle that says an ending should feel “surprising yet inevitable.” You could expand this basically anything in the story. Subvert my expectations, please, but be clever about it.
But let me get back to my joke pun: Chekhov’s Gundam. Maybe we could derive a real principle from this. Following our third interpretation of Chekhov’s Gun, Chekhov’s Gundam says that the genre expectations you set up at the opening of the story need to be followed through by the rest of the story. If you evoke the tropes of a haunted house tale, for example, but nothing spooky happens. If you conjure the tropes of a noir plot, but then there’s no murder, mystery, or investigation. Etc.
This is a really common problem when writers change the level of realism in a story. For example, a story opening and seeming to be 100% the real world then suddenly, without warning, turning magical in the last third. Or conversely a story seems magical or surrealist at first, but is actually 100% real after the first few misleading pages.
Genres are, in a sense, a pact with the reader. The tropes that are evoked at the start are a promise of the pleasures to come and the ground to be explored. You don’t want to promise your reader pleasures you can’t deliver on. Even if you have other pleasures to offer, who wants a bait and switch?
This principle is similar, though different, than the SFF concept of “squid on the mantelpiece,” which is about avoid a mishmash between plot stakes and character stakes as I understand it. If you have an alien invasion wrecking the city, the story shouldn’t be about someone failing at their job interview. As the Turkey City Lexicon says: It’s hard to properly dramatize, say, the domestic effects of Dad’s bank overdraft when a giant writhing kraken is levelling the city. This mismatch between the conventional dramatic proprieties and SF’s extreme, grotesque, or visionary thematics is known as the “squid on the mantelpiece.”
Chekhov’s Gundam, as I’m defining it, is about following through on the genre promises that you make to the reader. This means both the plot stakes, but also other genre expectations. Spookiness in a horror story, a mystery plot in a mystery story, etc.
As with Chekhov’s Gun, Chekhov’s Gundam doesn’t mean you have to be cliché or that you can’t subvert expectations in a clever way that still delivers. There are lots of great haunted house stories that end with an explanation that isn’t a ghost. Perhaps it was a human hiding in the walls, an alien, rats, a character going mad, or something else. But those stories still deliver on the promises—weird noises at night, scares, etc.—of the genre even as they subvert it.
Basically, give the reader the pleasures you are promising them. Or they might not read you again.