The Skill List Project: Scene Beginnings and Endings

This is another post in The Skill List Project: an attempt to list all the skills involved in writing and selling fiction, particularly science fiction and fantasy. In the past few posts, we’ve been looking at writing scenes. We’ve talked about the flow of plot from scene to scene and the need for conflict, difficulties or uncertainty as the heart of a scene. This time around, we’re going to look at a scene’s beginning and end: the two structural foundations on which the rest of the scene depends.

“Late In, Early Out”

Hollywood has the adage, “Start every scene as late as possible and end as early as possible.” For example, if a scene is about Lois confronting Lex in his office, you don’t show Lois driving up to the building, parking her car, going past security, talking to a series of receptionists, entering Lex’s office, and going through the niceties of Lex saying, “May I offer you some coffee?” Instead, you cut straight to Lois, already in the office, asking her first confrontational question. Similarly, when the interview is over, you don’t show Lois walking out and going down to get her car; you simply cut to the next thing Lois does that’s meaningful to the plot.

Of course, books aren’t movies, and a novel-length work often needs to talk the reader through significant transitions. Even movies do a certain amount of scene-setting: instead of cutting directly to Lois in the office, a movie might taken a few seconds to set an emotional tone, showing how huge and impressive the office building is, and how small Lois is in comparison. This can heighten our sense that Lois is up against serious opposition, and that Lex may be too much for her. This ratchets up the tension, which is seldom a bad thing.

Still, movies usually minimize these intervals of transition and scene-setting; they give enough to orient the viewer and set an ambiance, then they go straight into action that’s pertinent to the plot. The scene ends when the plot-pertinent action is over…except for an ending-hook, which we’ll talk about shortly.

“Late In, Early Out” in Prose

As I’ve said, books aren’t movies, and screenwriting principles don’t translate directly to prose. Even so, novelists should be on guard against any tendency to produce mere filler…and that tendency often surfaces at the start and end of scenes.

How much transition and scene-setting will actually work for the reader? How much does the reader need, and how much will the reader enjoy? These questions require artistic judgment, and the answers depend on many factors including the general tone of the story (fast action? sprawling epic?), the narrator’s voice (is the viewpoint character engaging enough to spin out long descriptions, or should you get straight to the point?), and the context within the story (do you want to maintain the same pace as previous scenes, or is it time for a change-up?). There are no formulas, but you still have to ask yourself: when does “useful transition and scene-setting” become “counterproductive padding”? The answer is usually, “Several sentences less than what you put in your first draft.”

Figure out where your scene actually begins, and start there. The scene begins when the struggle begins. Struggles may begin subtly—in our Lois-and-Lex example, the struggle may begin as the building’s imposing architecture assaults Lois psychologically—but subtle or not, the reader must have a sense that something is happening right now and the story is moving forward. Otherwise, the reader may be tempted to “skip ahead to the good stuff” or even set the book down.

“Every Scene’s Ending Must Rise”

Last time, I mentioned a workshop that Nancy Kress and I taught at a convention. During that workshop, Nancy said, “Every scene’s ending must rise.” I took that to mean that every scene should end with a sense of, “There’s more to come, and it’s going to be juicy.” In other words, the ending should have a hook that makes readers keep reading.

Hooks can take many forms. As a clichéd but simple example, the Lois-Lex scene might end with Lex picking up a phone and saying, “That woman has become an inconvenience. Deal with her.” This does two things. First, it promises action to come: something bad will happen to Lois. Naturally, we want to find out what the bad thing is and whether Lois will be okay. But the hook does a second important thing: it assures us that the scene we just saw was important. The scene wasn’t just Lois asking pointed questions and Lex stonewalling her; the scene was about Lois getting under Lex’s skin, whether or not he showed it. If the hook isn’t there, how do we know that the conversation actually achieved anything? The hook tells us the conversation did have an effect. It assures us that the ball was moved forward and the scene wasn’t a waste of time.

So a good scene ending often does (at least) two things: it points toward the future, and also caps off the past. It provides evidence that the preceding scene went somewhere; it wasn’t just talk or meaningless action.

If you look at good scene endings, those two factors are always present, however subtly: some assurance that the scene really has shifted the ground, and a promise that as a result, important developments are on the way. This is especially vital in low-key scenes where readers might ask, “Well, what did that accomplish?” Many scenes don’t have fireworks or immediate repercussions; conversations, for example, may be so guarded that readers can’t tell if any forward progress was made. The scene ending must put such doubts to rest. It doesn’t have to be flashy—it might just be a thought in the viewpoint character’s mind, “He’s good at saying nothing; I’d better follow a different lead”—but one way or another, you must signal that the scene has had an effect.

If you can’t devise such a signal, take it as a danger sign. Every scene should advance the story; at the end of a scene, you should be able to say what it actually accomplished and how it contributed to the story as a whole. What has changed between when the scene started and when it ended? You, the writer, should always be able to answer that question…maybe not in the very first draft if you’re still just exploring the story, but definitely before the final draft is done.

And the reader should also feel that the scene accomplished some purpose. Readers may not be aware of a scene’s full implications—it’s lovely when a scene contains under-the-radar bombshells that will detonate later—but whatever your long-term plans might be, you have to think short-term too. You can’t let your readers think, “That scene was a waste of time—nothing happened.” The scene’s ending is where you can (crudely or subtly) indicate how the scene moved everything forward.

(P.S. Negative accomplishments still count. If Lois’s interview with Lex goes nowhere, no problem. You can just end the scene with Lois deciding, “All right I’ve tried the civilized approach and it didn’t work. Time to get primeval.” That’s perfectly fine. Lois may have failed to achieve her goals, but the scene still advanced the action: it persuaded Lois to change tactics and to try more underhanded measures. Lois’s failure is the story’s win.)

This Scene Ending Too

Time for me to end this scene, having accomplished as much as I wanted about the structure of scenes. Next time, we’ll move on to actual content…and I think we’ll begin with the skill of writing description. In the meantime, what are your thoughts on scene beginnings and ends?

Filed under For Novelists, learning to write, writing process. You can also use to trackback.

There are 8 comments. Get the RSS feed for comments on this entry.

  1. 1. Traci Loudin

    Great post, thanks! I especially liked the part about every scene’s ending rising. Reassures me about the way I’ve been writing some of my recent scenes. As a reader, I get tired of every scene ending in a cliffhanger. It’s exhausting, and it makes me think, “What’s the point of even having chapters?” A chapter or scene should conclude, but it should also promise more. Sometimes, that may mean a cliffhanger, but it doesn’t have to be.

  2. 2. Mary

    Sometimes two scenes abut each other so closely that it makes sense to give a sentence or two transition rather than chop the scene, which may make the gap seem larger.

    Especially if they’re short. Too many short scenes, the story comes across as choppy.

  3. 3. Rocky

    Just found this great site and this article series last night, and now caught up with all of it.. I am struggling to work my story into something that I enjoy writing, rather than just put in the things I feel i ought to write about.

    It might be useful to see an example of what an outline looks like in a future article here.

  4. 4. James Alan Gardner

    An outline can be as much or as little as you need. There’s also a difference between an author’s working outline (the stuff you figure out in advance so you can write the story) and the sales outline (the pitch that you give to an editor in order to sell the story).

    Your working outline can be as terse as you like — it’s your reminders to yourself. I have the Penguin Classics version of “Bleak House” by Charles Dickens, and in the back, it reproduces Dickens’ own notes for the novel. Some chapters only have a single line. Others have longer entries, but very cryptic unless you know what’s going on. For example: “Miss Flite’s friends? — Her birds? Yes slightly. The birds. Not the friends” Dickens knew what he meant, and that’s all that mattered. It was never intended for other people to see.

    Sales outlines have to be more intelligible, since you want an editor to read the pitch and become excited. I’ll point you toward “I have this nifty idea…now what do I do with it?” edited by Mike Resnick ( It contains a number of outlines that various SF writers used to sell well-known novels. Well worth having a look at. —Jim

  5. 5. Rocky

    I’ll check that out, thanks.


  1. The “Skill List Project” for fiction writers | thoughtsignals
  2. The Skill List Project: Rewriting at SF Novelists
  3. The Skill List Project: Index at SF Novelists

Author Information

James Alan Gardner

James Alan Gardner got his M.Math from the University of Waterloo with a thesis on black holes...and then he immediately started writing science fiction instead. He's been a finalist for the Hugo and Nebula awards, and has won the Theodore Sturgeon Memorial Award as well as the Aurora award (twice). He's published seven novels (beginning with "Expendable"), plus a short story collection and (for street cred) a Lara Croft book. He cares deeply about words and sentences, and is working his way up to paragraphs. Visit site.



Browse our archives: