Bradel Bookwork – Scene Selection · 9:16am Jan 25th, 2014
This one's gonna be a little rough around the edges, folks. Let me just say that up front.
Here's the deal. "Three Nights" is a hot mess. Just how hot? Well, let's see.
(1) It's a Hearth's Warming story, and it's still not done a month after Christmas. At this point I'm just hoping we've still got polar vortices and snow running around for as long as it takes for me to finish. And since I'm in Southern California, land of perpetual 70° weather, I have no practical reference for winter scenes anymore.
(2) It's got some of the worst chapter structure ever. First chapter: 1000 words. Second chapter: 4000 words. Third chapter: 10,000 words. Be nice if I could get a little balance in there, but no, I've basically written two separate prologues for one 10,000-word story.
(3) There is, I suspect, a distinct lack of narrative cohesion. I have some ideas about how to address this, but the whole thing needs a throughline (or possibly more than one) to make it hang together correctly.
(4) It includes a truckload of OC's, one of whom is a major character. Along with the standard cast of ponies one needs for stories such as these.
(5) Oh yeah, and I wrote a Hearth's Warming carol. Yes, the bloody story even includes a real, live musical number. And commissioned artwork which I'm dying to show you because I'm in love with it. But what's the bit I can't get done? The important bit: the writing.
Now, don't get me wrong here. Hot mess at present or not, I think this is my second strongest piece of work after "Bell, Book & Candle". But it's challenging me to play around with things I don't think I've really done a whole lot with before. And one of the big ones here, the reason I'm up late writing a blog post, is choosing when to start and stop scenes. This is the big issue I'm worried about in editing—whether I'm making smart choices about my scenes or whether I'm going to need to go back and do a good bit of rework.
Before we go any further tonight, let me lay down a framework for this discussion.
When I talk about scenes, I'm using a textbook definition here. A scene is a story unit presenting continuous action without skips in time or location. We have, I think, a natural tendency to think of events as occuring in scenes. When we think about how a story is structured, we think about a sequence of scenes. When we think about our own lives, we often think about them as sequences of disconnected (but related) discrete events.
So let's say you're writing a story. You've got a checklist of plot points you need to make happen. Putting aside character and setting for a moment—those are relatively easy to work into the story once you've got a framework, I think—how do you go about deciding on the scenes that your story needs? And in particular, how do you know where those scenes start and end?
The best advice I've ever heard about scene selection is actually advice about crafting stories as a whole: start as close to the end as you possibly can. Put another way, don't waste your reader's time by making him or her read about events that fundamentally aren't important to the plot of the story.
This is an issue I've been seeing pop up in my EQD pre-reading lately, too. We usually just peg it as "there's a problem with pacing", but it's often easy to be more detailed than that. The problem with pacing is that an author hasn't chosen his or her scenes well. They start too early and meander for a while before anything actually happens, or they overstay their welcome after the goals of the scene have been accomplished and run on too long. If I see 3000 words of meandering character interaction and vague description before I hit a plot, it's a good bet that I'll be rejecting that story. That's not to say a story absolutely must have some action in the first few thousand words—of course not. But it should have something to hook the reader, and one of the best ways to do that is to get the reader asking questions, wanting to know more about what's being expressed on the page. Action can do that. Subtle mysteries, things that feel out of place in an otherwise conventional narrative, can also do that.
Now, I'll begrudgingly make allowances that sometimes it's good to toss up a scene that doesn't seem to do a lot of plot work, if you can get enough out of it. Here, I'm referring to the whole issue of multitasking your words again. I've been reading the preview chapters for Brandon Sanderson's upcoming novel Words of Radience lately, and his first chapter is this in spades. He's got two characters traveling on a boat, and as some fantasy afficionados are liable to tell you, boat travel is usually the death of good fiction. For that matter, travel in general is a poor topic for writing. Getting from Point A to Point B is the sort of thing that's usually best left for a section break. But you can do it, and do it well.
So, Chapter 1 of Words of Radiance. We've got two characters on a boat, and they're dyed-in-the-wool academic types. This is a recipe for disaster. Academic characters usually aren't very active to begin with (cf. active characters are interesting characters), and now you're going to throw them into one of the most boring situations that can arise in fantasy fiction? How do you fix this problem?
Cue the appearance of a rare giant sea creature, a bit like a humpback whale. Now our academic characters have something to do: engage in hands-on research. And just like that, we're elaborating on both setting and characterization, and making our characters active-thus-attractive even while they're stuck on the boat. Plus, of course, we've now managed toa ctually write a scene about travel on a boat, which means we can check off miserably mundane plot item that needs to be dealt with but that no one wants to read about.
When does the scene start? When the giant sea creature appears. When does the scene end? As soon as we've mined the scenario for characterization, setting, and activity. We don't really care about the sea creature, so we're free to move on as soon as the scene has hit its highlights.
I kind of wish I had something deeper to say about picking scenes besides, "Start them as close to the end as you can, and stop them as soon as they're reasonably wrapped up"—but that really is the key piece of advice I keep coming back to. Ideally, by the time you get around to writing, you ought to have your story mapped out well enough that you know the major beats you need to hit and you can break them down into scenes, so the major work (for me, at least) comes in knowing how to bridge those scenes.
In some stories, that's quite easy. "A Filly's Guide to Not Making Headlines" has exactly three scenes, and they're pretty well split apart both in terms of location and in terms of chronology. So you can consider each one separately and pare it down to what you need without too much difficulty. "The Amazingly Awesome Adventures of Tank the Tortoise", similarly, breaks into a series of nice, discrete scenes. Chapters 1 and 2 of "Three Nights" are both single-scene affairs, and are fairly easy to construct.
Building scenes is harder, though, when you've got a long string of semi-continuous action. If you relate the whole thing, you're going to wind up having to describe a fair amount of walking around, or at least eliding through it repeatedly. This is the situation with Chapter 3. It takes place over six hours, during which time a number of things happen—but there are three or four distinct sections to that six-hour timespan. So do you narrate the whole six hours, just trying to ease your way through the boring bits? Or do you chop those six hours into multiple scenes you'll then recombine, and if so, where do you do the chopping?
It's not a terribly easy question, at least not for me, and it seems to rely heavily on figuring out what really matters in the outline you build for yourself. What does the reader absolutely need to know for plot purposes? How much value does the interstitial material have for setting the tone of your story or helping you give a fuller picture of character and setting. And at the end of the day, you just cut everything that doesn't seem to be serving some purpose, and hope you've cut enough but not too much.
One thing I found while writing Height was first person narrative is really helpful in this respect. The decision of what to include and what not is somewhat offloaded to the character (Rainbow really wasn't that interested in the mice, so she skipps over practically the Fluttershy's entire conversation with them). Leaving unimportant details out is also a very believable thing to do because that's what you do when your telling someone your story—whether its how your day has been or your entire life story.
Perhaps you could ask "if I were the POV character telling the story, what bits would I be skipping?"
1756213
Nice. I was really tempted to comment on "Height" here, and I was definitely thinking about it as I wrote this because it's similar in the sense of following a character for an extended timespan, but whether because it's late and my brain is fuzzy or because I'm just not great at picking up on some things, I hadn't really pieced together your point about using the first person voice and how it can play into this. But now that you mention it, yes, that seems very sensible.
(As a total aside from this, have you seen the episode summary for Simple Ways and are you looking forward to it anywhere near as much as I am? Yay, "Muddy Hole"!)
Brother, I wish I knew, but PBT has a good start.
I've been worrying about scene breaks quite a lot lately. I'm pretty good at in-early-out-late (mostly because I'm a pretty lazy writer), but that tends to make the writing feel choppy. That's fine if you're interleaving two sequences (i.e. Amazingly Awesome Adventures), but often it can be problematic.
Once I was paying attention to it, I began to notice that many authors tend to zoom out in the narrative -- still giving a series of events, but on a more abstract level -- rather than just end the scenes and skip chunks of time. Not always, of course. But still pretty often. If you've got a long period of time in which relevant events are rather sparse -- but which you can't just leap over, because there are relevant events -- zooming out like this is ideal.
The more of these I read, the more I'm certain I don't actually know how to write and never did.
Honestly, when it came to scenes I just... wrote them. I mean, okay, yes, scenes tend to have a purpose to them. "Princesses" opens the way it does because I wanted an action hook, I wanted to set up a joke (the big dramatic news sound innocuous), and I wanted some space to give Spinny some characterization. But aside from that... I just... y'know. Write. It seems to work out... maybe?
If I had to systematize, I'd say that the basic scene checklist is to ask if the scene is setup, payoff, or characterization. Ideally the scene is all three. Something is set up--there's a mystery, a bit of setting detail, &c, &c, there's a payoff--a joke is sprung, an action scene happens, a plot point is revealed, and there's characterization--the actions of characters, main or secondary, serves not only for setup/payoff but also to illuminate how they work. You can have a scene with only one thing, say, but you have to use those sparingly. And any scene--or more germane to the question, fragment of a scene--without any of those things needs to be cut[1].
[1] Which, technically, would mean that my "Let's talk about trade relations" scene from ACC should have probably been cut. Well, I never said I was any good at this.
1756213
In fact, first person almost seems to have a whole additional expressive layer in what the character does or doesn't see, note, or react to.
Nicely spotted.
1756366
Someone actually asked me how I was choosing when to switch perspectives in my 'live writing' last week. I had to think about it, because I had been doing it completely by instinct (this keeps happening to me).
Anyway, the answer ended up being pretty specific to comedy writing. It turned out I was switching every time there was a punchline--the switch is an easy way to keep from having to continuously one-up my own jokes as the scene progresses.
This is the important part. Perhaps you should have started your post here.
1756350 Yes, skipping chunks of time is risky. Can you take notes on what you find? Cormac McCarthy gives the impression that he never skips small chunks of time, but of course he does. It's some kind of magic trick.
Ending a scene when the POV character thinks they're still doing something interesting is also bad.
1756366
If only you'd had a pre-reader capable of pointing such things out!
You, specifically, like to throw in possibly extraneous world-building or backstory, or things that are part of some other story you're writing in your head using the same characters. And you know perfectly well that you're doing it. I think you need to write a novel once a year and flush all that stuff out.
1761403
It's both disturbing and gratifying to have someone figure you out so completely.
But, yes, that's pretty much exactly what I do.