DMs are wary of giving power to players, partially because few DMs can prepare for or react to players with too much freedom and partially because players are guaranteed to abuse it. This is why wishes are granted so rarely, why campaigns tend to end when PCs ascend to godhood or gain their own kingdoms, and (in part) why low-level campaigns are so much more common than high-level campaigns. I’m a little less concerned about this than most, I think, and a lot of that is the social contract we have; my players want to play in the game, so anything that dramatically disrupts the game violates their own interests. But it’s also because I like seeing what players do with power. Per the above, I’ve granted wishes, I’ve made PCs into gods, and I’ve only spent more time on low-level campaigns because my players haven’t requested high-level ones. But my capstone is probably the time I game my party immeasurable, universe-rending power, but didn’t tell them they had it, just to see how long it would take them to figure it out and what they would do once they did.
Let me tell you about the principle of explosion.
By the middle of The Great Tower of Oldechi I was getting outside of the typical D&D realm; the players had made it through fields and tundra and mountains and such, and it was time for them to deal with a cemetery dimension and an abandoned, futuristic factory. Floor 13 was the Black Labyrinth, a maze-like series of smallish caves that connected to each other. The first thing the players had to do was solve a puzzle to create a map of the floor, and that puzzle was a word association node game.
I love this type of puzzle. The most famous example I can give you is probably the Funny Farm puzzle, but scattered throughout the Internet you’ll find several similar games. I wrote my own small puzzle with D&D terms and campaign information, like linking “floors” to “The Rolling Plains” (the name of Floor 1) to “planes” to “elementals” and so on. Because this puzzle was the map of the floor, each node’s name said something about its contents. The “darkness” area was pitch-black, the “Astral Plane” node let the players float around, and the “minotaurs” node was full of minotaurs to fight. Of course, the players spent a lot of time trying to find the safest way to a node named “gold” or “treasure” only to find a bank offering reasonable loans. I wish I could tell you exactly what it was, but that puzzle doesn’t exist in the place I uploaded it online. Let that be a lesson: cloud storage is trash; don’t use it for anything you care about.
I designed the puzzle before I filled each node with things, and I hit a bit of a wall when I then had to populate a node named “explosions”. It bridged other nodes so perfectly I couldn’t remove it, but I didn’t want it to just be a violent hellscape. My players would have expected that, and I’m nothing if not subversive. Instead I Googled for other things named “explosion” that weren’t actual fire and fury.
I stumbled upon the principle of explosion. To put it simply enough that I understand it, it’s a principle of logic that says if we assume two contradictory things are true, literally everything is true and the concept of truth is meaningless. Wikipedia explains it better than I can, and xkcd does not. I took this “anything can be true” idea and thought “What about a room where everything the players say becomes true? That seems fun and weird, but incredibly dangerous. It’s as likely to collapse the entire campaign as it is to be an entertaining encounter. I shouldn’t do it.” So I did it.
In the explosions room, the answer to every question was “yes”. The players didn’t know this; they only knew the room was named “explosions”. The first interaction went like this.
PC 1: I enter the room.
DM: Okay.
PC 1: Do I die?
DM: Yes.
Things went bad in a hurry. I only dropped the player to 0 hit points, for obvious reasons, and the players lingered outside the room trying to figure it out.
PC 2: I make an Arcana check. Does the room detect as magic?
DM: Yes.
PC 2: I throw a dagger into the room. What happens to it?
DM: The dagger is in the room. (For anything that couldn’t be answered with a “yes” or “no”, I did whatever felt consistent with previous answers.)
PC 2: I cast ray of frost into the room. What happens?
DM: The ray turns solid and falls to the ground. (For anything where consistency was irrelevant, I did whatever was funny.)
PC 2: Huh.
PC 3: I stick my hand into the room. Am I alright?
DM: Yes.
PC 3: I enter the room slowly.
DM: …
PC 3: Do I take any damage?
DM: Yes. (I rolled some level-appropriate amount.)
PC 3: But after the damage, I’m fine?
DM: Yes.
PC 1: I’ll enter the room again. Do I die?
DM: Yes.
It took a legitimate half-hour for the players to figure out the room has specific rules beyond “kill the tank”. Even then they phrased their questions very carefully, not fully confident they had the answer, and they spent five minutes teasing out their actual goal, an object they needed to touch to signal their progress across the nodes.
PC 2: I look around. Do I see the object we need?
DM: Yes.
PC 1: I touch it. Does it glow?
DM: Yes. You’ve completed this room.
PC 4: Do I see a huge pile of gold?
DM: No. Because you’ve completed the room, the magic fades.
PC 4: (Turns to PC 1) You couldn’t wait one minute?
PC 2: So the answer to every question was “yes”?
DM: Yes.
PC 1: What if we’d asked if there was a staircase to the top floor of the tower?
DM: The campaign would have become much shorter.
This wasn’t snark, this was the actual answer*. The party then would have had to face an L32 boss with L13 PCs, but at least they would have felt clever. I probably would have let them jump back to Floor 13. Probably. Similarly, if they asked to gain ten levels, they would have, even if it made the next few floors very boring. The same went for gaining new equipment, or killing another character, or even switching the campaign to another edition. All answers were “yes”. But the players didn’t destroy anything, because that didn’t interest them. They didn’t want to exploit godlike power dropped in their laps. They wanted to solve this puzzle, then move to the next room, eventually beat the floor, and continue through the tower. They signed up to play in a campaign, and when they got the option to ruin the campaign, they decided against it.
I can’t top the explosion room and I don’t intend to try. My players will likely never have the sort of limitless power they had in that session. Though, now that I think about it, the pantheon campaign is a cute idea…
* — Upon further examination it might be both.