Coming to a Bad End

by John Robert Marlow

Post image for Coming to a Bad End

Coming to a Bad End: Rabbit-Hats, Cliffbangers, and Other Cheats
by John Robert Marlow

Few things in life are worse than a bad story. One of them is a good story with a bad ending. At least with the bad story, it’s pretty clear what you’re dealing with, often in the first few chapters. So you really can’t blame the writer when, despite numerous warning signs, you slog all the way to page 347 before throwing in the towel. In a sense, you knew what you were getting into.

Not so with a good story. There you are, swept along by the narrative, engrossed in fabulous dialogue from characters so real it seems they’ll step off the page—when something goes so terribly wrong that every fiber of your being shrieks: That’s not the way it’s supposed to be!

Somehow, the author veered off course and, in the end, you’re left feeling disappointed, cheated, even angry. A number of authorial missteps can lead to this dark place, but the major boo-boos fall into several broad categories…


A rabbit-hat ending is one where some wildly unlikely occurrence happens at just the right time, and in just the right place, to turn a dire (or hopeless) situation into a happy ending. The bankrupt protagonist is about to hang himself—when he suddenly inherits a fortune, or wins the lottery. The hit man has Our Hero in his sights—but slips on a banana peel, and falls in front of a speeding train. And so on.

Such events are about as likely as—but less believable than—magically pulling a white rabbit out of your hat. It destroys all credibility, is often unintentionally comedic, and always makes the author seem too lazy or unimaginative to construct a plot that stands on its own merits—instead of resorting to slight-of-hand and parlor tricks. Aristotle complained about this sort of thing two thousand years ago, but many new writers have yet to see the memo.

Despite all of that, you can sometimes get away with this sort of thing when writing comedy—because in this situation, you want the reader to laugh at the sheer improbability of it all. (For a full treatment of coincidence, see What a Coincidence! The Use and (Mostly) Misuse of Coincidence).)


We’ve all seen these. Engrossed in a book or movie, we’re hanging off the edge of our seat. Wondering, perhaps, how Our Hero can possibly extricate himself from the most impossible situation of all when, suddenly, the protagonist wakes up—and we learn it was all “just a dream.”

Most of us react with something approaching disgust. Oddly enough, the better the story has been (until now), the stronger this feeling is. Why? Because everything we just experienced, the whole roaring roller coaster of human emotions…never happened. It was all for nothing. Useless. Pointless. Unfulfilling—for us and for the character.

It should come as no surprise, then, that our own readers and viewers will react in precisely the same way. Assuming we make it past the agents and editors, that is. And yet, beginning writers continue to make this same mistake.

A similar dynamic is at work when, instead of finding it was all a dream—we find instead that the main character is batso. Again, the whole thing didn’t really happen, except in the deranged mind of our protagonist.

Exceptions do exist: either of these scenarios can be successfully pulled off. The problem is that every new writer thinks his story is the exception—and almost every new writer is wrong. Most seasoned pros who think this are also wrong; that’s how we know what it feels like to encounter one of these tales.

One writer who wasn’t wrong authored the screenplay for Identity—a low-budget but brilliant example of the batso protagonist.

Still, any time you’re considering either of these moves, reconsider ten times over—because your story is unlikely to be that rare exception.


Once in a while, an author will get so wrapped up in his own fictional world that he will unknowingly drive the story over the double yellow line, so to speak. The result is a head-on smashup with reader expectations. The most frequent examples involve children and pets.

The Fly (1986) and The Fly II provide an illustrative example. In the first film, Seth Brundle unintentionally transforms himself into a sort of monster. In the sequel, something similar happens to a dog—and this time, the action is intentional. The first film works. But it’s hard to say how well the second works, because that scene is so disturbing that it’s hard to think of anything else—and, in fact, hard to watch the movie at all. That one scene is so far over the line that it destroys the effect of the film as a whole.

A big part of the reason the first film works and the second doesn’t is this: what happens to Seth in The Fly is—however unintentionally—his own fault. He meddled with things he shouldn’t have, and he made a mistake. In the second film, the poor doggie is an innocent victim of a malevolent scientist. A great many audience members (and readers) cannot bear to see bad things happen to children or to animals—even when adults are considered fair game.

That’s one of the few taboos we have left, and you ignore it—and other line-crossing maneuvers—at your peril. Stray too far over the line and, as far as your readers are concerned—the story ends right there.

There really are no exceptions to this, outside of twisted tales aimed at small markets.


Stories are about things that happen to people. So when nothing really happens at the end—things just sort of peter out, or keep going the same way they’ve been going all along, with characters who are unchanged by the journey they’ve taken—readers feel cheated. “What’s the point?” they ask.

Conflict requires resolution. Two dogs, one bone. That’s story. One dog gets the bone, one doesn’t. That’s resolution. Sure, there are open-ended stories, and some of them work, and some of those that work succeed commercially. But something of great importance is resolved, and we very seldom end where we began.

There are stories that break this rule. Most of them you’ll never see, because the manuscripts and screenplays are sitting on someone’s closet shelf or hard drive. A Simple Plan (novel and film) might seem to be a rule-breaker but, really, it’s not; great changes have taken place, and both the main character and his wife have been profoundly changed.

Memento, on the other hand, is a rule-breaker—but the particulars of this example (scenes presented in reverse order; lead character with no short-term memory; circular plot) are so unique to this particular story that it’s hard to see them applying to anything else.

Very few commercially successful stories lack definitive resolutions. And so—absent some astonishingly good reason to break with tradition—you should strive to provide one.


For the most part, heroes are heroes because—eventually—they succeed. Or, as Eddie Dodd says in True Believer: “Don’t give me that liberal yuppie bull**** about a good fight… A good fight is one you win!

Not every protagonist wins, of course, but the overwhelming majority of commercially successful protagonists do triumph in the end. Even in those instances where they fail to achieve what they set out to do, they very often wind up with something of greater value.

Wall Street is one example of this: when Bud realizes that his blind ambition is about to wreck thousands of lives—including his father’s—he abandons his quest for material wealth, turns on his erstwhile mentor, and averts catastrophe. He loses what wealth he’s gained, loses his girlfriend, even loses his freedom (at least temporarily)—but he gains the self-respect that comes with being an honest man. It sounds corny but, when done well (as it is here)—it works.


We’re all familiar with the “cliffhanger,” where a scene ends on a major revelation, or with an important character’s fate left dangling. This technique originated with early silent film serials—where the hero would literally be left hanging from the edge of a cliff at the end. To find out what happened to him, moviegoers would have to return the next week. Modern tv series frequently employ this same technique just before commercial breaks and episode endings—albeit in a somewhat less literal sense.

The cliffbanger, on the other hand, is what you get when the hero falls to his death, banging into the ground at the bottom. (Ouch.) From a reader / audience perspective, this is even worse than the dream / nutjob ending: here, the character they’ve most strongly identified with…dies (or does something heinous—as in The Mist, where a man shoots his whole family to save them from a monster that never comes). What kind of ending is that?

One that angers your audience, that’s what kind. Here again, the trials and tribulations along the way become pointless—and the more readers like the character, the more they’ll dislike you for doing this. So don’t.

Which isn’t to say that protagonists can never die: 300, American Beauty, Gladiator, and many other stories prove that. But note that with 300, we know going in that everyone dies; with American Beauty, we know in the first two minutes that Lester is going to die; with both American Beauty and Gladiator, we are at least somewhat comforted by the assurance that the character lives on—somewhere else—after death.

A dead protagonist must make sense, must (in retrospect) seem inevitable, and should serve a purpose. (The 300, for example, changed the course of civilization.) Still, this is dangerous terrain, and alternative routes should be strongly considered.


This is the kind of story where approaching doom is certain, but the day is saved when—for no apparent reason—one of the characters suddenly undergoes a radical personality shift, doing something (or failing to do something) that is completely inconsistent with his or her previous actions.

The Bad Guy who’s been trying to kill the Good Guy for the last hundred pages sprouts a conscience. The meticulous planner overlooks the obvious, with drastic consequences. The abusive husband turns gentle. The evil corporate CEO donates his fortune to charity and takes up residence in a monastery. And so on. Point being, there’s no previous setup, and the out-of-character action (whatever it is) proves crucial to the plot’s outcome.

Such devices are just that: devices—artificial, out of place, and unbelievable. Ultimately, they do as much damage as the situations they’re intended to resolve.

There are no exceptions. When turnarounds work, it’s because they’ve been set up earlier. Darth Vader’s turn in Jedi—perhaps the single biggest character turn in cinematic history—works beautifully, because it’s been set up for three movies. We don’t see it coming—but when it arrives, we understand, and everything clicks into place. It makes sense; it feels right.

In Back to the Future, George McFly—who’s submitted to Biff’s abuse and humiliation for years—suddenly lashes out, with life-changing consequences. But again, we understand: he’s loosing a rage that’s been building for years. Like Darth, he was finally confronted with the one situation capable of effecting massive change. Because we know the character, we get it.

Bladerunner’s Roy Batty is another story. In the version of the film without voice-over narration, his final scene with Deckard is so massively inconsistent with his previous actions as to be totally incomprehensible. In the cut with the voice-over, it works.


The Princess Bride is not devoured by Rodents of Unusual Size. Rocky is not killed in the ring. Sam and Annie do not miss each other at the top of the Empire State Building on Valentine’s Day. And, like these tales—your story does not disappoint.

Or your sales figures will.

Author John Robert Marlow is available for professional editing, development, and consultation. If you’d like help taking your work to the next level, contact John here.



copyright © by John Robert Marlow
all rights reserved

{ 6 comments… read them below or add one }

Diana Lynn Tibert November 19, 2009 at 5:59 am

Phew! The hero can live. I’ve read several writing articles where the writer insists the hero must die at the end to make a great story. That seemed drastic to me and downright cruel. And too predictable. It’s good to read an article that supports my views. Thanks.

Eric C November 29, 2009 at 5:31 pm

Great website. I just found it from your guest post on the Urban Muse. Some thoughts:

Rabbit hats, or as I’ve commonly heard deus ex machina’s, are always unacceptable. I totally agree. Same with personality transplants. Bad Dream/Nutjob endings are also unacceptable, unless you’ve been playing with audience expectations, like in Life on Mars.

On cliffbangers and failure, what happened to tragedy? This used to be a staple of literature. I think these endings need to be set up properly, but I find many movies today almost get boring, because I have no worry that our hero will live and save the day. I’d rather watch a movie where that isn’t assured. Your example of The Mist, well, that ending stuck with me for days.

More on this, I feel like writers resort to the artistic endings mainly because they feel they can’t make artistic statements in the body of their work so they have to make them at the end. Not sure if this is a good or bad thing.

Finally, non-endings work if they thematically wrap up the work. I’m thinking of Hemingway, or The Grapes of Wrath.

John Robert Marlow November 29, 2009 at 9:34 pm

Hi Eric.

Sure, these things can sometimes be made to work, and I cite a few examples of stories that manage to pull it off. Commercially speaking, though, you’re fighting the tide, and are likely to face requests (in the publishing world) or demands (in the film world) for change. Worse, prospective buyers might simply pass on the project instead.

In the majority of cases where I see these things attempted by unpublished / unproduced writers, they not only don’t work—they’re completely inappropriate, and lack any kind of setup.

When they are attempted (successfully or otherwise) in print or film—which is rare—the writer is almost invariably well established. The Mist, for example, was (as near as I can tell) Stephen King’s ninety-ninth published work. In short, the things we’re talking about are risky moves for anyone—and often fatal for first-timers.

John Robert Marlow November 30, 2009 at 3:25 am

Hi Diana.

Absolutely. In 2008, Ryan Condal sold his spec script Galahad for $500,000. The script is extraordinarily well written, but the hero—Galahad—dies in the end. In an interview I’ll publish on my Lonely Keyboard site early next year, he says: “That’s the first thing the studio changed.”

Even if your storytelling abilities are outstanding enough to generate interest in a dead-hero tale, you will likely be asked—or told—to bring your dead hero back to life.

And it makes sequels so much easier.

Nora Weston November 30, 2009 at 9:58 am


Hello! Congrats on your success and thanks for the wonderful tips. I found you through Teresa Frohock. I especially hate a “dead-hero-tale.” I know many people who don’t mind that, but since a book takes a considerable amount of time to read…I like to feel content, rather than sad at the end of it.

Take care,

Nora Weston

Eric November 30, 2009 at 10:20 am

I guess I would just respond that, looking particularly at failure, that this is why so many people complain about Hollywood and that Hollywood doesn’t make good films anymore. While I don’t think that is 100% true, I do think this attitude makes films boring and predictable.

I guess I would make a distinction between endings that are technically indefensible, such as the dream, rabbit hat, personality change, and endings that are difficult/artistic, such as failure and non-endings.

Leave a Comment

Previous post:

Next post: