PART III OF OUR SERIES ON GRABBING AND
HOLDING READERS EXAMINES THE
IMPORTANCE OF POWERFUL ENDINGS
September 1989
A while back a friend of mine was flying from Los Angeles to New York. He was in the first-class section, a luxury to which he is not much accustomed, and the chap seated beside him was some sort of yuppie businessman, on his way to or from some sort of hostile takeover. The little swine had a clear enough conscience to lose himself altogether in the inflight movie, a pleasure my friend was willing to forgo.
The yuppie laughed immoderately all through the film. When he unplugged his earphones even as they rolled the final credits, my friend asked him how he’d liked it.
“Not so great,” the young man said.
“But you laughed your head off,” my friend protested. “If you hadn’t been belted in you’d have fallen out of your seat.”
“Oh, I’m not saying it wasn’t funny,” the little shark replied. “There were some great laughs in the thing. But, you know, it just wasn’t a very good picture.”
Now this story might do little more than illustrate the perversity of the Young Undeservedly Prosperous but for the specific film involved. It was Burglar, the Whoopi Goldberg vehicle based (more or less) on a book called The Burglar in the Closet, a mystery novel written by, uh, me. And the chap seated beside the chortling little chiseler was my agent, the redoubtable Knox Burger.
And, worst of all, the damned whelp was right. Burglar was a million laughs, but it just wasn’t a very good movie. And virtually everyone who saw it reacted pretty much the way Knox’s seatmate did. They roared while they were in the theater, and then they told their friends not to bother going. This was true of the insider audiences; laughter was riotous at the large Manhattan house where I saw the film screened, and the very people who laughed the loudest then went home and wrote scathingly negative reviews. The reaction was the same at the theaters in suburban shopping malls. Everybody had a good time for 90 minutes and went out shaking the old head in disgust.
Why should a film—or fiction in any form—provoke this sort of contradictory response? How could audiences have such a good time with the picture while it was going on and respect it so little once it was over? In the particular case of Burglar, I think there are several answers. The gags were too easy, the characterizations were shallow, the relationships were too hard-edged—there were lots of things wrong with this movie, and most of them need not concern us here. But one factor that I’m sure contributed to the film’s failure to generate good word-of-mouth was the relative weakness of its ending. The ending was soft, and it left the audience unsatisfied.
“The first chapter sells the book,” Mickey Spillane has said. “The last chapter sells the next book.”
Very true. But there’s even more to it than that. The last chapter sells the next book by convincing the reader that the book he’s just finished was terrific. That doesn’t just make him a customer for your next effort, but it makes him a powerful salesman for what he’s just finished reading. The stronger your ending, the more likely he’ll be to recommend the book to his friends. It is word-of-mouth ultimately that creates bestsellers. Nothing else, no amount of advertising and publicity, can sustain a book that does not get touted by those who read it. And a book with an unsatisfying ending just cannot generate strong word-of-mouth on a broad scale.
“First you’ve got to get them on board. Then you’ve got to keep them on the boat for the duration of the voyage. And, finally, you’ve got to kill them at the end.”
That, as those of you without chronic short-term memory problems will recall, has been our premise in this space for the past three months. If, as we’re always being told, a work of fiction has a beginning, a middle, and an ending, it seems reasonable to suppose that a writer is variously challenged by each of these three components. His job in the beginning is to hook the reader into the story, while in the middle it is his task simply to keep the person reading.
At the end, he has to pay off all the book’s promises. He has to give the reader everything he signed on for—and more. A weak ending can kill a good book, and a really powerful ending can save a book with not that much else going for it.
During the past year I’ve read a pair of unusually well-written first novels, both of them suspense yarns, one set in Michigan, the other in the Florida Keys. Both books have generated a lot of favorable comment among mystery pros, no doubt because of the genuine excellence of their writing. Neither did as well with the public at large, and I think I know why. The ending of one was improbable, almost silly, while the other ended very inconclusively. I enjoyed both immensely while I was reading them, but ended feeling somehow cheated and unsatisfied.
I know I’ve hurt my own sales in the same fashion in at least one book. Ariel, a novel I published ten years or so ago, was a story of psychological suspense featuring a 12-year-old girl who may or may not be evil, and who may or may not have murdered her baby brother in his crib. And the ending is inconclusive. You don’t find out for sure what the girl is and what she did. A few reviewers liked the enigmatic ending, but more than a few did not, and I don’t blame them. It was vague because I was vague—I didn’t know what had happened. I would have greatly preferred a less uncertain ending if I could only have come up with one.
What makes an ending work?
Maybe the best way to answer that is to listen to a Beethoven symphony. By the time the last note of the coda has sounded at the end of the fourth movement, you damn well know it’s over. When that last ringing chord hits you, every musical question has been answered, every emotional issue has been resolved, and you don’t have to wait for the folks around you to start applauding in order to be certain the piece is done. If Ludwig van B. had set Ariel to music, there wouldn’t have been anything enigmatic about the ending, believe me.
It’s generally a good deal easier to write an ending with impact if you have that ending in mind from the onset. The more clearly you are able to perceive it as you go along, the more you can shape the various elements of the story so that the ending will resolve them in a satisfying fashion.
Does this mean that you have to have the whole book outlined, in your mind or on paper, before you write it? As one who almost never uses an outline, I’m hardly inclined to advance such an argument. It is possible, however, to know your ending without knowing just how you’re going to reach it.
Several novelists, most recently E. L. Doctorow, have likened the writing of a novel to driving at night. You can see only as far as your headlight beams reach, but you can drive clear across country that way.
Very true, and I’ve written any number of books in just that fashion. But I’ve been most successful when, while I could not see past the range of my headlights, I nevertheless knew my ultimate destination in advance. If I just hop in the car with no goal in mind, I may have an enjoyable journey, but I run the risk of not getting anywhere, or not even really knowing when the trip is over. (In point of fact I travel that way all the time in real life, but it doesn’t work as well in fiction.)
Dorothy Salisbury Davis, who does very well indeed with beginnings and middles as well as endings, has said that she can’t comfortably write a mystery novel unless she knows from the onset who did it. She may change her mind in the course of the book, she may wind up hanging the murder on someone other than her initial choice, but she always has a solution in mind even as she constructs the problem.
I haven’t always done this, but I certainly have an easier time when I do. It seems to me, too, that a substantial portion of the books I’ve abandoned over the years have been ones for which I did not have a strong ending in mind from the beginning. I ran out of gas on those books not specifically because I wound up painting myself into a corner or wandering in an insoluble maze but simply because each book sort of wobbled to a halt. I think it may have been the lack of a concrete destination in the form of a foreseen ending that brought this about.
The most satisfactory endings resolve everything. Like that Beethoven coda we just heard, they answer questions we never even thought to ask.
Most of my books are mystery novels, concerned with a crime and its solution. Find the murderer and you’ve found the ending. Mysteries, however, are frequently concerned with more than crime and punishment, and sometimes an ending has to do more than name a perpetrator and clap the cuffs on him.
Eight Million Ways to Die is a good case in point. The book begins with the murder of a call girl, and my detective, Matthew Scudder, is hired by her pimp to find out who killed her. The stakes are raised when two other prostitutes die, one an apparent suicide. And, finally, Scudder brings the killer to justice. He does so by making himself a stalking horse, a move that almost fails when the killer waits in Scudder’s hotel room with a machete. But Scudder and justice prevail, and the bad guy gets what’s coming to him, and the ending is dramatically satisfying.
But the string of murders is not all that the book is about. It’s also about life and death in New York, and it’s very much about Scudder’s attempt to come to terms with his alcoholism. He struggles to stay sober as he chases the killer through the city’s terrible streets, and the book follows him in and out of ginmills and detox wards and AA meetings. After the book has seemingly ended, after the killer has been found out and dealt with and the solution explained to his client, there is a final chapter in which Scudder is brought face to face with his own illness and has to confront himself or back down.
The first ending, the unmasking and apprehension of the killer, is dramatically effective but not everything it might be. Because of the story itself, the killer is not someone we have met before. (Hollywood can’t bear this sort of thing, and in the film version the killer is the sneering villain we’ve met early on.) But the second ending more than makes up for it. A considerable number of people have told me, in person or through the mails, how much impact the ending had for them. Many of them have assured me that they cried, that they were moved to tears.
And that is what an ending ought to do. It ought to move a reader. It need not move him to tears—although that doesn’t hurt. But it ought to leave him knowing that he’s been in a fight and that the fight is over. You don’t have to leave him feeling happy—although a downbeat ending is usually hard to bring off effectively. But you do have to leave him feeling complete. He may finish wondering what will happen to the characters afterward, and that’s all right, as long as you leave him feeling that the issues raised in this part of their story are resolved.
Not every successful book has an ending that works in this sense. Some people break the rules and seem to get away with it. The example that comes first to mind for me is John Le Carré, who has made an occasional habit of endings that I can only assume are intentionally obscure. Both The Spy Who Came In from the Cold and A Small Town in Germany have ambiguous last pages; you have to read them over a second or third time in order to be certain just what is taking place. The author’s writing is so clear elsewhere that it is puzzling that his ending should be so murky. I can’t seriously argue that this weakness, if that’s what it is, has hurt Le Carré with readers or critics. He’s doing just fine, and for all I know maybe I’m the only person who finds his endings opaque.
Any questions? Yes, Rachel?
Why “kill them at the end,” sir? Why such a violent image?
I don’t know, Rachel. I’ve asked myself the same question, and originally looked around for a way to paraphrase Donald Westlake’s original observation that triggered this series of columns. But I can’t find an alternative that works as well.
Comedians, and performers in general, use that metaphor. “I killed them in Keokuk,” the vaudevillian would say. “I knocked them dead. I beat their brains in. I slaughtered them.”
I guess the implication is that the audience—and in our case the reader—is overpowered by the material. It overwhelms him, and killing is the ultimate way of being overwhelmed because it is undeniably final. What you may be objecting to, Rachel, is the implication of hostility between the comic and his audience, the writer and his reader. If you’re trying to kill your readers, doesn’t that mean that you hate them?
No, not in this case, not when they pick up the book hoping to be killed in just this fashion. Even if you continue to dislike the metaphor, I’d urge you to strive for fictional endings that seem to fit it. Because this kind of metaphoric death is anything but final. Unless you kill them at the end, they won’t keep coming back for more.