Max Gladstone is a member of Tiara Club, in that he was one of the nominees for the John W. Campbell Best New Writer last year, along with me, Mur Lafferty, Zen Cho, and Stina Leicht. He’s also the author of the Craft Sequence novels and now? Also the creator of a whompingly enjoyable interactive fiction app based on his storytelling universe. This seemed like a good place for Max to come and talk about the lessons learned while writing this epic interactive endeavor. Now hold still, and let him fill you with MAD CRAZY WISDOM.
I don’t outline as a rule. Or, I didn’t. But recently I had to learn.
About a year ago, Interactive Fiction moguls Choice of Games asked if I’d like to write a game set in the weird world of my legal thriller fantasy novels Three Parts Dead and Two Serpents Rise. I said yes, because yes is what you say in these situations. Cool opportunity? Paid for writing? Lich-king lawyers? Sign me up. And I wrote a game, which was awesome: Choice of the Deathless hit app stores just before Christmas (that link’ll take you to the appropriate store for your device). It’s out there now, and while I’m biased reviews indicate it’s excellent and I think you should play it. (Okay, plugging done.)
I put that bit at the beginning so you know this story ends happily. Because what follows is a tale of a writer struggling with the practical magic of plots and outlines, and it’s not always pretty.
ALSO HERE IS AN IMPORTANT DISCLAIMER. This is a personal narrative about how I grew through this one writing project. Please don’t mistake this for a “How You Must Write” story—everyone who writes comes from a different place, and each writer has her own goals. But if your history is anything like mine, the journey I’m describing here might be helpful to you. Then again, it might just be funny.
Long as I can remember, I’ve been a pantser: with character, goal, and milieu in hand, I venture forth into Story Jungle and hack a way through to the El Dorado of a finished draft. (Though since this is a first draft we’re talking about, El Dorado often looks more like El Pyrite and needs an awful lot of basic plumbing before I can invite friends over. How much? I’ve written 160,000 word drafts, added 20,000 words in revision, and ended up with a 100,000 word final manuscript after I cut all the bad stuff. Not pretty, but it works. Sort of.)
I’d outlined in the past out of curiosity and necessity, but each time I found the story’s rhythm only when the outline shattered and new characters snuck out from behind the arras. Fine, I thought, that’s how I’ll use outlines from now on: like bamboo scaffolding, strong, flexible, and easy to destroy when necessary.
Those were the days before I sat down to write a game. And when I say “write,” I don’t mean “write dialogue for”—Choice of Games are choose-your-path novels with a stat screen, taking some gaming DNA from Bioware RPGs and some from the Lone Wolf books of yore and legend. Choice of Games provides the scripting language; I provided the code.
On the one hand, awesome. My first computer was an Apple II+; I spent days coding adventure games for that box, using the simplest BASIC anyone has ever used. “Upgrading” from the II+ to something with a modern GUI felt wonderful, but also as if someone had cut off my legs—I couldn’t program this weird new device, and when I tried to learn, the coding books I found spent a lot more time trying to teach me how to alphabetize a CD collection (Me: “I’m fucking ten! I don’t have any CDs and if I did why would I want to alphabetize them?”) than how to write a game. (And yes, eventually I learned how to code simple stuff, though I never recaptured that initial “I can do anything with just what came in the box” feeling.) So, ChoiceScript in hand, I felt that old joy: I can tell your machine what to do, and I’m going to tell it to tell you a story.
On the other hand—terror. As it says on the tin, ChoiceScript games are all about Choices. Fun! At first. Until I started work and realized that in practice, each story beat needed around eight possible outcomes, and a success and failure path for each potential answer. A beat that took 250 words in a traditional short story might need 2,000 words to offer a semblance of true interactivity.
And so the cost / benefit of pantserhood tipped sharply toward ‘cost.’ It’s trivial to cut and rewrite 2,000 words of story because that scene needs to do something else. Cutting 16,000 words at one stroke gives writers strokes in turn.
So I had to think through the story before I started. The usual “write what comes and fix it later” approach wouldn’t work. But I couldn’t outline. Oh no! That’s for plotters. I, on the other hand, like any kid who learned how to program from antique books on BASIC at his local public library, would draw a flow chart.
I started out with a flowchart of choice-diamonds. What choices did the player face? What paths could she walk? Would success here bypass an entire branch of the story later on? How would other characters respond to her decisions?
I then converted that flowchart into skeletal code, logic and command without dialogue or description. After a few scenes I felt comfortable enough to write in skeleton-code first, sans flowchart: your Determination needs to be this high for you to win your fistfight with the demon. Staying awake all night reviewing these wards will impose the following penalty to your Sleep stat, which will in turn make everything a little harder. In what ways might your law-wizard attempt to depose a goddess, or betray her firm, or romance a fellow junior associate?
After another couple chapters, I’d written enough skeleton code that I could content myself with a short list of notes before I started coding: a line for each choice, a few bullet points for potential consequences, clear scene breaks, and on to the next chapter.
I couldn’t deny it any longer: I was outlining. And it felt awesome.
What had changed? Necessity, for one thing, but on reflection the outlines I wrote for Choice of the Deathless (and the outlines I’m using for my next book) differed in one huge way from my earlier attempts. Before, my outlines were bullet lists of scenes:
Joey and Sarah meet Dragon Lord Vorthax.
Vorthax eats Jimmy; Sarah escapes on rocket sled.
Joey wakes up inside Vorthax’s stomach.
Cloud-dwellers save Sarah, nurse her back to health.
Joey talks with Vorthax.
Cloud-dwellers attack Vorthax.
That’s okay as it goes—I have a list of setpieces, and I can think of cool stuff to write for each. What’s missing?
Choice. Or, to put it another way: drama. That list elides any moment when Jimmy, Sarah, or Vorthax consider their options, stand torn between gut-desire and heart-need, sacrifice one possibility for the sake of another. In fact, I’ve written the negative space of drama: all of the cool and/or heartbreaking stuff that happens because of choices. But I haven’t written the choices themselves. They’re an afterthought to the action.
As a result, the scenes in my outline have a sort of loose cause-and-effect relationship but nowhere to go. The outlines I learned to write for Choice of the Deathless were that negative space. The cool results, those weren’t nearly as important as the dynamics of the choice. Let’s add some of that stuff to the outline above:
(Sarah and Joey need to save their Mom, who’s been taken captive by bandits. Sarah decides they’ll go to Dragon Lord Vorthax, even though they know he has a nasty and carnivorous reputation.)
Joey and Sarah meet Dragon Lord Vorthax.
(Vorthax asks both of them why they’ve come. Joey speaks up first: he wants power to save Mom. Vorthax asks what he’s willing to trade for it. He says, everything.)
Vorthax eats Jimmy. (Sarah runs first, then stops, almost goes back to save her brother; decides she can’t.) Sarah escapes on rocket sled.
Joey wakes up inside Vorthax’s stomach. (What does he do next? Jump around inside Vorthax’s stomach & start pulling out wires until the Dragon talks to him.)
Cloud-dwellers save Sarah, nurse her back to health. (Should Sarah ask the cloud-dwellers to try and save her mother, or Joey?)
Joey talks with Vorthax. (Vorthax has swallowed the boy to teach him dragon-magic, but this will involve Joey transforming into something not quite human.)
Cloud-dwellers attack Vorthax. (Does Joey kill the Cloud-dwellers with his new power? Does Sarah kill Joey to save her new friends?)
Silly, sure, over-the-top, limited in a lot of ways, but also propulsive; each scene has a dramatic purpose. Thinking in terms of choices rather than setpieces felt like turning my brain inside-out at first, but over time I grew comfortable with the process. As a result, I wrote better, and faster. The game’s web of decisions seemed less a danger and more an opportunity—I could hide neat worldbuilding down little-traveled pathways, sneak humor into secret moments, build consistent characters who seemed villains when set up one way and beleaguered heroes in another context, all because I knew basically where I was going and how I wanted to get there.
I’m working on my next book now, taking the outline road with this new approach—building around choices, conflicts, and costs rather than action set-pieces. Once the dramatic framework’s in place, I think—I hope—everything else will grow in between. The cool stuff will come, because it will have a reason to be there.
Of course, writing this next book I’ll probably learn a whole bunch that will make all I’ve written here seem quaint. But that’s the process, isn’t it? We’re always learning. Or we should be. Sometimes, we learn from people; most of the time, we learn by doing, thoughtfully and with conviction. There’s no hope in waiting to write until we’re already perfect writers.
Heck, when I started writing Choice of the Deathless I couldn’t outline my way out of a paper bag.