Postmortem: Twilight Opens a Door · 4:04am May 27th, 2021
I finished a short story. I feel good about it.
But I want to take a minute to go over what I set out to do versus the final product to sort through my method and identify what I think worked, what didn't, and what I can try to do different next time. This is a fairly long blog post, and it's mostly for my own benefit, but for anyone who's interested, all the juicy behind-the-scenes stuff is behind the break. There's also a sneak peek at what I have planned to work on next at the end.
I was reading Creating Character Arcs by K.M. Weiland and I got to this part:
Visualize a locked door separating the First Act from the Second Act. The First Plot Point is where the protagonist sticks his key in that door and unlocks it. And like Pandora's box, he ain't never going to get it shut again.
And this sparked an idea: What if the turning point is actually opening a door?
And that's the whole story of where Twilight Opens a Door came from. A gimmicky idea to experiment with just to practice the storytelling craft. The whole thing was very low-stakes for me, and gave me the freedom to just have fun.
What came next was a rough structural outline. I wanted to practice writing a proper archplot. What is an archplot? From Story by Robert McKee: it's a single, active protagonist in a consistent reality with linear time and causality, struggling through an external conflict that resolves in a closed ending. So it's basically the "standard" type of plot most stories aspire to. (This is in contrast to "miniplot" which involves passive and/or multiple protagonists, internal conflicts and open endings, and "antiplot" which is typified by coincidence, nonlinear time, and inconsistent realities. Most stories are never "pure" forms of any of these, but they form a triangular spectrum of storytelling frameworks.)
Twilight Opens a Door as I wrote it, doesn't really fit the archplot. Certainly nowhere near as well as I had wanted it to. The conflict is more internal than external, Twilight is mostly passive and reacting to coincidence, reality is completely warped, and the ending is... Well. I'll get to that. But overall, it's somewhere in the middle of the three types, archplot-adjacent.
So what happened? My original outline was 5 chapters that went something like this:
- Twilight at Golden Oaks stresses out over something out of place, finds a strange door, opens it, and gets pulled through by [something]
- Same as chapter 1 but more things are out of place, the strange door isn't there, the [something] that pulls Twilight through is behind the front door this time
- Princess Twilight in her castle, aware that something is going on, rampaging around opening every door until she gets pulled through by [something] and winds up in a surreal hallway where she meets Discord who reveals himself to have been behind the whole thing (more on that later). Twilight opens a door to escape.
- Back in Golden Oaks, Twilight gets "spit out" from the front door. Everything is exactly where it should be, but Twilight has a sneaking suspicion that it's still not the proper reality. She finds the strange door and opens it.
- Back where she started, Twilight has learned to chill out about misplaced items. The end.
So from the outset, I clearly failed at crafting a proper archplot. (TBH I think I have some sort of fundamental aversion to writing archplots. Not in a "I'm too cool to play by the rules" way, but more like I just have more fun playing in other parts of the story-space.) This really only occurs to me now I suppose, since after I decided to write an archplot, at no point did I stop to ask myself "does this fit the archplot?"
First thing to go was the [something] pulling Twilight through the doors. Instead I opted to end each chapter on just opening a door. The sharp ending with lack of closure is more likely to grab reader interest (cliffhangers, yo) and it's easier to write.
And yeah, Discord was originally going to be the antagonist of the story. Given that I had already abandoned a consistent reality, he seemed fitting, and I had a fun idea to give Discord some "meta" lines as an allusion to the Star Trek TNG episode Tapestry where Q gives Picard a chance to change his past as a means to save his life in the present. (It's a good episode, go watch it.) Ultimately Picard undoes the change, having learned that the life he lived was the only life he wanted and if that meant it ended there, then so be it. (He lives anyway, of course. It was 90's episodic tv after all.) The idea here was that Discord would be helping Twilight learn to not stress out so much.
Discord was also supposed to make a joke to Twilight about "you probably didn't even notice all the misplaced commas" which I had already set up for by deliberately putting a bunch of commas in the wrong places during Chapters 1 & 2. This was a bad idea from the start, I wouldn't do it again, but I have exactly zero desire to go back and fix them.
I wrote the first chapter and then sat on it for a month because I couldn't get myself started on chapter 2. So I went back to the outline and took a chainsaw to it. Chapter 2? It's repetitive and boring. Gone. Chapter 3? Okay, we can do the castle. New setting, toss Starlight in there for some variety. That's the new Chapter 2, end it on opening a door. Where? I dunno. Wherever Twilight ends up around the 2k word mark. New Chapter 3 is the surreal hallway with Discord. Cut the repetitive backtracking after that and new Chapter 4 is back to normal. The end.
I was still having trouble getting to work on chapter 2 at this point, but I felt better about it. Enough to go ahead and publish Chapter 1, mostly just to see how interested readers would be in it. Answer: not enough to get 10 votes (up or down), but enough to encourage me to keep going. (Thank you, everyone who read it.)
So where the hell did The Myth of Sisyphus come from and what happened to Discord?
In my first pass at writing Chapter 1, I just used a placeholder [book title] by [author] for the out of place book Twilight freaks out over. While I was laying awake in bed one night, I had a random thought about ponifying the name of some author to fill that in, and I came up with turning Albert Camus into A. Camel. (And by this point if you aren't aware, The Myth of Sisyphus by Albert Camus is a real book.)
As I continued working on Chapter 2, and the more of a deal I made about that book in the story, the more I realized I should probably make it actually important to the story. So when I got to writing Chapter 3, I had a dilemma: keep Discord and use him to incorporate the ideas of Absurdism, or put A. Camel into the story in Discord's place. (I did toy with the option of having both of them, but could never reconcile how much of either or how it would help the story to do so.) Ultimately, I decided to run with Albert.
Unfortunately, I found myself completely at a loss for how to write him. At the same time, I realized that if I were going to do it this way, I should maybe, you know, actually read The Myth of Sisyphus myself. So I got a copy and read the whole thing in the week between publishing chapters 2 and 3. It's not a long book, only about 120~ pages with a reasonably sized font. Still, it's translated from French and is an essay about existentialist philosophy, so I could really only absorb about 30 pages at a time without my eyes glossing over. (Not because it was a bad read, but because it's a very intellectually demanding essay; Camus wants to make you think. Also Camus references the work of a lot of other writers/philosophers I'm not familiar with, so I had to make a lot of inferences about what he's saying.)
I am very glad I read the book. The wikipedia articles and youtube videos about it and about Absurdism in general do not do it justice. And in reading it for myself, I found the voice for Albert in my story. (If anyone reading this has also read Camus's work, I'd like to hear your opinion of how well I did at that.)
The challenge I had from this point was finding a way to fit the ideas from a 120 page essay written by a man who was awarded the Nobel prize in literature into a ~2k word chapter of a MLP fic and to do so while making a good faith effort to maintain an E rating. Because the essay deadass starts by asking "should you kill yourself?" (Short answer: No. Long answer: Nooooooooooooooooooooooooooooooooooo!)
I think I managed it okay. Chapter 3 probably would have benefitted from another 5-700 words to further develop the themes and ideas from Absurdism, and to question how alicorn immortality factors into The Absurd when mortality is kind of a big deal in Camus's essay, but I felt that I had gotten the important things into it already and I'm impatient when it comes to hitting the publish button.
That impatience is something else I had wanted to work on with this story. I had originally planned out a month to write the whole thing start to finish, spend a week or two editing, and then publish one chapter a week once it was already done. The problem I have with self-imposed deadlines is I know that the person who sets them is full of shit.
When I got to writing Chapter 3, and had decided on Albert over Discord, my first thought was to just end the story there. This didn't really feel right to me though when I tried skipping ahead to write the ending first; it was overly-sentimental and unsatisfying. But while I was still in the process of reading The Myth of Sisyphus myself, I remember waking up one morning with the thought that the ending would reveal the whole thing to have been taking place inside Twilight's head as she'd used a spell to absorb the book's contents in lieu of actually reading it. It was all a dream. And thus Chapter 4 as it appears gave me the anchor I needed to figure out how to proceed on Chapter 3.
It feels like maybe it was a bit of a cop-out to do it that way, but at the same time, I think the way I presented what the doors represented in Chapter 3, it could also be read that Chapter 4 is the dream. [insert Inception sound effect here] Either way, it felt like the most satisfying ending I could come up with, given where the story had already ended up. If I were to do it all again, I'd start by reading Camus first, write the whole story start to finish, edit, then publish. As noted in my previous blog, I typically have no idea what a story is really about until I've already written most of it, and this is probably the single biggest problem I've faced when it comes to writing.
So what now?
I have three story ideas I want to work on:
- Starlight and Sunburst have misadventures in creating artificial intelligence (sci/fi comedy miniplot I intend to write and publish piecemeal because there really is no over-arching plot I want to give the story, just some silly stuff based on the work of real-life AI safety researchers)
- FoE-adjacent wasteland archplot horror story about a changeling queen (rated M for violence and gore) (And I want to really push myself to do an archplot because it's good practice.)
- FoE-adjacent wasteland miniplot in the style of diary excerpts to give some backstory for Mum's Diner
All of these I want to keep fairly short. Like 10-20k words each at most. I'm incredibly lazy about putting words on the page, and if I try to do anything longer than that, it'll be a repeat of Sweet Nothings which took 2 or 3 years and is honestly a mess.
I want to start on the AI story first because I just need a break from wasteland ponies. That AU has dominated my imagination for way too long, and it feels good to reconnect with wholesome pony roots. But I do want to get back to my wasteland stories because they hold a personal significance to me that I'm not done exploring yet. So what I think I'll do is write the first 2 or 3 chapters of the AI story and publish them together, then continue updating that story irregularly while I start developing the changeling horror story. That one I will try to be as disciplined as possible about getting the whole thing done before I start publishing it so I can get a complete and cohesive story out of it.
The diary story is something I'll work on in the background, just filling in bits and pieces as it comes to me, then after the changeling story is complete, I'll figure out how to turn it into a coherent narrative and go from there.