Endings are a problem for a lot of writers on a lot of levels. People have trouble making them convincing, trouble making them dramatic, trouble keeping them from dragging out or being too abrupt. One of the problems that seems surprisingly common is picking out the ending at all – in other words, some writers just sail right on past the appropriate endpoint without realizing they have done so. This can result in a number of problems – gigantic million-word tomes that weren’t meant to be million-word tomes and don’t work as million-word tomes, plot that wander off into the weeds and never really reemerge, and extremely grumpy writers who complain that they don’t understand what their problem is, but something’s wrong and they can’t stop doing it.
In my experience, there are several different reasons why this sort of thing happens, and the first step, as always, is diagnosis.
The first type of writer (first chronologically, as in, the first sort I ran into) usually has a main character whom they find tremendously appealing. They’re usually planning to write a huge multi-volume saga about their protagonist (sometimes, they’ve already done so). In essence, they want to write a detailed biography of the character’s life, career, and adventures. None of this is a bad thing; the problem is that they get so involved that they can only see their character’s life as a whole, birth to death. It never occurs to them that they’ll have to consciously break it up into coherent episodes…so they don’t.
A slightly different type of writer is…perhaps “easily distracted” is the best description. They start off with a Plan, but somewhere in mid-book, when they are trying to fill up all those pages between the beginning and the climax, they get intrigued by a secondary character, or a chunk falls out of a subplot and leaves them with new and fascinating vistas. The original plot-problem gets lost – yes, yes, we’ll take care of putting the True King back on his throne in Iceland just as soon as we straighten out all the politics of these desert nomads that are just so interesting… Essentially, the main characters get partway through one story and then get sidetracked into a different one. And this process can repeat several times (the desert nomad politics were getting too complicated or the writer was getting bored with them, and then they captured a spy and had to go chasing all the way to China to find out who was behind that plot, and we’re off into a third story without ever resolving either of the first two).
Finally, there are writers who have grown up on complex, braided novels with multiple subplots of varying lengths. In this sort of novel, the author has to start introducing new subplots just a little before the previous set gets wound up (because most of the subplots don’t last the whole length of the novel). The trouble here tends to be that the writers are too good at doing multiple complex plot braids. They can and do keep introducing new plot-arcs, sliding them smoothly in between the events that wrap up the previous plot arc, but they don’t know how to make them come out even. So the braid gets longer and longer and just keeps growing, with no obvious end point in sight.
The general fix for all of these is the same: at some point, the writer has to decide what the central plot-problem is, and then they have to check back periodically to make sure they are headed toward solving it. If something more interesting comes along in mid-book, that’s fine, but they have to stop for a minute and consider how they are going to make the shift. By the time the writer is in the middle, they’ve usually given the reader a pretty good notion of what the central plot-problem is, and they can’t break that promise and still keep most of their readers. What they can do is twist it and expand it – you thought the problem was getting the One Ring to Rivendell and handing it off, but no, the real problem is getting it to Mount Doom and destroying it and Sauron for good. (The real problem is not, however, a digression into dwarven politics when they discover survivors in the Mines of Moria, or haring off on a quest for the Entwives. It’s actually still the same problem the story started with – getting rid of the ring – it’s just become more complicated than anybody thought at the start.)
But whatever the new plot is, it still has a clear endpoint. When the One Ring goes into Mount Doom, that plot problem has been solved. When Aragorn is crowned and marries Arwen, their plot thread is done. When Frodo departs Middle Earth, the emotional wind-up is over. There are still plenty of new stories to be told (that’s a lot of what is in the Appendices), but the story that was being told in The Lord of the Rings is over.
A story that has one main character has one central plot-problem – there may be a lot of things that are important to that character, but there’s only one that is the most important. And yes, people’s notions of what is most important can change over the course of an adventure, but if that’s what’s going on, then that is the central plot line. And yes, some writers can sit down and do that sort of change without thinking very far ahead…but if you are having trouble overshooting your endings, then you are not one of those writers, and you will have to decide what the central plot-problem is, and what it means to say that it is solved or resolved.
If the core of the problem is not being able to separate episodes, figuring out what the central problem of the book is should go a long way toward letting the writer see where the end is. If the navy commander’s central problem is that he’s never been recognized for his achievements, then the end of the book comes when the Admiral shakes his hand and says, “Good job; we’re giving you a promotion for winning that battle.” It ends there even if the writer knows perfectly well that the “promotion” has all sorts of problems and political strings attached, because those problems are new, they’re not the one this story is about.
If the core problem is getting distracted by shiny new plot ideas that could happen to the protagonist, once again, you start by deciding early on what the central problem of this book is. Every couple of chapters, you step back a little and check to make sure you are still heading toward the resolution of that problem. You don’t have to have a hard-and-fast ending in mind (many writers do, but it’s not a requirement), but you do have to know what “over and done with” means. You might have thought the hero and the villain would have a confrontation in a movie theater, but instead it happens on a beach – it isn’t the details that are important, it’s the fact that they have a confrontation and settle things once and for all.
If you’re a braider, you need to know what your central plot thread is, and then you have to pay attention to getting all your subplots to come out more-or-less even at about the same time as the central thread wraps up…and that means that at some point, you stop introducing new subplots and plot arcs. OK, if it really is going to be a long series, you may want to plant a few things for later, like the existence of the heroine’s estranged black sheep brother or ex-fiancé, but planting things is a lot lower-key and more subtle than introducing a subplot. Most of the time, what you are planting for the next book has little to do with this one, and it should ideally read that way – as minor background detail that the reader doesn’t need to remember, not as an intriguing mystery that they’ll want solved now. Also, whatever you are planting should, if possible, not make its first appearance in the last four or five chapters of the book. Not if you are going for subtle, anyway.
If you’re braiding several different characters’ plots…well, each of them has a central problem that is most important to them; your challenge is to see that all three or four of the characters resolve their central problem at more or less the same time. If one will take much longer to finish with than the others…maybe what you have is a main character and some subplots, rather than a true ensemble cast with braided plotlines.