Keeping the Furniture
A companion to Reject Narrative Structure.
There is a pressure that shows up during play — the small persistent tug to bend what you're doing toward a beat, to make this moment land, to give the scene a shape it hasn't earned yet. It is real, most people have felt it without naming it, and it costs you the only part of the evening you actually get. You are at a table with your friends for three hours. Every unit of attention spent shaping is a unit not spent noticing.
And the deeper premise is stronger than it looks written down: if making shape out of experience is compulsive — if you genuinely cannot help it — then doing it on purpose is redundant labor. You are pushing something that was already moving. The instinct is reliable enough that you can stop feeding it and lose nothing.
That's the argument, and it holds. What follows isn't a rebuttal. It's the part that comes after you accept it.
The compulsion runs in one head at a time. Four people leave the table with four accounts of what happened, and they don't match — different protagonist, different turning point, different thing that mattered. Nobody's lying. Each of you narrativized faithfully, from where you sat.
So "the story takes care of itself" is true, and it takes care of itself four separate times.
Whatever makes those converge is not storytelling. It's coordination — whose turn it is to speak, when a scene has ended, what the group agrees just happened, who gets to matter next. That machinery was living inside the thing being rejected, and when you throw out the arc it goes out with it, quietly, and you don't notice until the evening is over and the quiet player hasn't said anything since the first hour.
Kill your inner storyteller. Keep the furniture he was sitting on.
Nobody at the table narrates. You answer a phone, you argue about a color, you haul a box.
Then you're in the parking lot and somebody says the purple thing and four people laugh at the same object, and there it is. Not built during. Assembled after, by everyone at once, out of what actually happened.
This is worth trusting because you can watch it happen everywhere else. Films get their shape in the cut, not on the bus. Records get assembled after the fact from sessions that weren't a plan. The material is genuine; the shape is later, and it belongs to whoever's holding it.
Which means the advice really is stop trying. Not stop caring. The editor shows up on their own, and they do better work when you haven't been trying to help.
A two-minute sketch about a man being moved desk to desk needs no plot and takes none. Stretch that to ninety minutes and a subplot grows on it, and the subplot is the worst part.
Twelve episodes stand up plotless. Two hundred cannot — they grow a romance, an acquisition, a departure, an arc, whether or not anyone at the start wanted one. Not because the makers lost their nerve. Because length demands it, and refusing only makes the demand louder.
So the advice has a shelf life, and it's short. One evening is exactly where it's free. A campaign will accrete a spine no matter what anyone says at session zero, and the honest choice is whether you get to see it coming.
Which makes the time-box the load-bearing part. Not a convenience. The reason any of this works.
Take a game that has three acts printed into it — setup, escalation, climax.
Delete those three words. Play it. Nothing breaks. Nobody at the table was ever told they were in act two, and the scenes don't care what the section header called them.
Now delete the scene boundary instead. Everything falls over. The refresh, the once-per-scene ability, the resource that recharges, the thing you only get once per session — all of it was counting against that line. The count was doing the work the whole time, and the three-act vocabulary was a label sitting on top of it, taking credit.
This is the distinction the essay turns on, and it survives outside games. Any methodology that promises to abolish structure will be found, on inspection, to have kept a cadence and renamed the phases. The labels reattach the moment you stop looking. Let them. They're free, and they're not the load.
Prep structure is not felt structure. What the group encounters is a room, some people who want incompatible things, and a clock.
"Put characters somewhere and say go" fails in a featureless room and works in a good one, and the difference is engineering.
A classroom works because the same people are forced to keep meeting. A shopping street works because there's a reason to walk down it daily and somebody's always at the counter. A hostile ship works because the pressure is that it will kill you.
That last one isn't incidental. Consequence is what makes presence weigh anything. Strip out the stakes and being here now is just being here — pleasant, weightless, over. The reason nobody tells stories about the sabbatical is that nothing was pressing on it.
So the design question was never what happens. It's what keeps generating. Build the engine, then genuinely don't plan the output.
Remove structure, put nothing in its place, and resolution doesn't become free. It becomes social. Which means it becomes status: the loudest person, the most senior, the one everyone's known longest, the one with the best voice.
Dice are not simulation. They're the impartial refusal — the thing that can tell the highest-status person at the table no without anyone having to be brave. Randomizers, oracles, timers, a clock nobody controls: same job. They put refusal somewhere other than a person's discretion.
Before you strip any piece of structure, check it against one question: after this, where does no live?
If the answer is "with whoever's most confident," put something back.
The founding move of the argument is one player, at a table, quietly running a different game inside the one everybody else showed up for.
That's not a small thing, and the problem isn't permission — it might have been wonderful, and the others might have loved it. The problem is that it was decided by one person about an evening belonging to several.
Everything above is like this. You can drop the arc for yourself in about a second. You cannot generate turn order for other people. The one thing that genuinely requires the group is the agreement about how the group will work, and it can't be arrived at during play by someone acting alone, however good their instincts.
So: have the conversation first. Not a contract. Five minutes and a shared decision. Tonight there's no plot. We're going to find out what the place does.
That's the whole ask, and it's the only part of this that isn't optional.
Say the costs out loud, because a manifesto that only lists gains is selling something.
Mystery needs asymmetry. A frame everyone can see relocates surprise from concealment to combination — the twist has to come from collision now, not from something withheld. That's a real trade.
Nothing accumulating means nothing at risk. Fine across forty minutes, where the immediate goal carries the weight. Hollow across forty hours.
And this is all theory, which is the cheapest thing at any table. One evening will tell you more than the argument will.
The form that gave us that phrase abolished the script and then wrote more rules than anyone: who, what, where, find the game, heighten, edit. Nobody in that room believes structure and spontaneity are opposed. They know exactly which structure they removed and exactly which one they kept.
That's the whole position. Drop the arc. Keep the turn order.
Then go be there.
Written in reply to Brendan Albano's Reject Narrative Structure (Pony Press), with gratitude. Share it, copy it, disagree in the margins.