Imagine a group of friends sits down to play a tabletop RPG.
They picked a Dungeon Master ahead of time, to plan out the adventure and generally be the arbiter of what occurs and what’s allowed.
The remaining friends put together their characters, choosing types (such as warrior or wizard), stats (such as how intelligent their character is as opposed to how strong or nimble), names, species, and so on.
There are rules to these games, but they are fairly flexible, to allow for creativity on the part of the Dungeon Master as well as the players.
Suppose that after playing a few times, some of the players get tired of it, and want to switch to a different RPG. A space adventure, say. Neither the DM, nor the rest of the players, want to give up on what they’ve done so far, though. So they strike a compromise—their characters in their current game will play an in-game version of the space RPG, and accrue experience points based on how well they do.
At first this takes up about a fifth of their gameplay. But gradually, they spend more and more time on the subgame. What’s more, they create more subgames, of many different genres. Some are so completely unlike the one they’re playing as to be hardly comparable—focusing on boring domestic scenarios, for instance. Or working together to solve puzzle games.
At what point can they be said to ever play the original game at all? What if 80 percent of their gameplay takes place in subgames? But now, what if much of that 20 percent was used determining which subgame to play or creating new ones? At what point does the original game vanish entirely, as an entity?
The original game is formally higher up on the hierarchy than the subgames. The DM could decide to have a dragon attack while their characters’ attention is caught up playing house in a subgame. But to the extent that it’s hard to get a group of friends together who will regularly commit their time to a common game like this, the DM can’t just do whatever he wants. If people think he’s being unfair or aren’t having any fun, they can walk away.

If enough people do this, the game will simply be dead.
In short, the DM is constrained in as much as he wants to avoid killing the game entirely.
I ask again: at what point is it absurd to refer to the original game at all?
Previous Posts in This Thread: