The Parable of Smoke at the Caves of Chaos

In module B2 “Keep on the Borderlands” for early edition D&D, there is an encounter area that the locals call “The Caves of Chaos” where much of the adventuring is assumed to take place (there are other small opportunities in the keep itself, but it's mostly up to the DM to generate action there). The titular caves are set in the walls of a vale, and each series is inhabited by members of the typical adversarial “monster” humanoid species, one with goblins, one with orcs, etc.

So imagine you are a young GM, and the party comes to these caves, and they quite cleverly think… “What if we just smoke the monster's out of the cave and wait here at the mouth of the cave to ambush them?” They say that they intend to do this. As the DM of the game… do you let it work?

This situation happened to me in a play of that module, and I have heard it repeated at random by at least three other DMs who have run the module. And in many cases, an issue occcurs. The DM, figuring that the interesting part of the module is having the characters encounter the denizens of the caves room by room… since that's how the encounters are presented… will try to figure out how this method of smoking them out will not work. He'll note that smoke rises, so that any fire at a cave entrance will not go into the cave, but instead go straight up. Or some other of the many plausible rationales one can come up with why this will not work.

The player, wanting his clever ploy to work, will note that there must be a little wind, and, if so, at least some of the cave entrances must be aligned so that the smoke will in fact be blown in by the wind. Or some other of the many plausible rationales one can come up with why this must work. They use a cape to blow smoke into the entrance. Or they just build the fire in the entrance.

Then the GM counters that the caves are slightly inclined into the sides of the valley, and so the smoke just rolls back up the caves, and back outdoors again. The module doesn't say this, but… it doesn't say that this is NOT true either. It's not implausible on the face of it.

Then the player counters with something about how smoke from certain trees is heavier than air, and so will sink down into the caves. And the DM will counter with the fact that those trees are not found in this area. Etcetera, etcetera, etcetera.

And everybody quits playing, because everybody feels that the other side is, essentially, trying to cheat. Again, this happened to me, and at least three other DMs that I've heard tell very much the same story. This is not a “parable” in the sense of it being fictional, games collapse because of this a non-trivial amount of time. If not in this case in this module, in other cases playing other adventures.

But what's going on here? Well, players and GMs at many gaming tables are informed that, in order to maintain “immersion” or some other such principle, that we only talk about the in-game situation in a way that talks about the plausibility of in-game action. The issue with this is that in many cases the obvious answer to the question is not at all clear, or, rather, there are a LOT of plausible ways to proceed. In the example, both the player and the DM are presenting arguments about how it's plausible for the plan to work or not work that are at least prima fascia plausible. There's no way for us to teleport to the fantasy world in question, and give it a try to test which is “actually” right. The fact that we're playing a TTRPG allows us to try things that are not possible in a CRPG or MMORPG… no game presentation of this module would have an option for smoking the denizens of the caves out. This is the advantage of TTRPGs, but unlike computer RPGs, we don't have an unbiased arbiter of what you can and cannot do.

This is why we have “resolution systems” in TTRPGs. To determine what happens in situations where two or more plausible things may happen, and clearly somebody participating in the game thinks one way would be fun, and another participant thinks another way would be interesting (see The Lumpley Principle).

The problem with the early editions of D&D which were used to play this scenario are that there are various resolution systems, but there are no generic resolution systems. Generally there is only the system that determines who prevails in a “Combat” situation, and other special rules about things like thieves climbing walls (but note, not any way to determine if anybody else can climb something). So using these systems, there's no moment where we can say “Let's roll for it.” Instead you get the GM being final arbiter of these things, and he'll put his foot down at some point, and the clever player feels like their fun is being stomped on. (Note that one of the Basic editions, I think Mentzer, does have a generalized resolution of “roll under your stat,” which could be used in this case to fix the problem, at least on the surface.)

Because what is really going on is a difference of opinion as to what would be fun here between the participants. Nobody is really “cheating,” they're using the authority the game gives them to argue that their case is the most plausible, which, again, nobody can win (often escalating into battles of ego to see who will give in to the other's brilliant intellect). Even when somebody does give up, it's because they're intuiting that the other person doesn't like where they're going.

The way resolve these things, where the resolution system cannot, or will only do so in a dissatisfactory way, is for the players to cease talking about the in-game situation, and talk about what they want as the real people engaging in the activity. The DM should say, “The module is really all about you meeting the encounters room by room, wouldn't that be more fun?” And the player should say, “Well I really like the idea of how TTRPGs are flexible enough to allow out of the box solutions, can't you let it work and we'll see how it goes? After all, we'll still be going in room by room to clear out the tunnels once we've ambushed the denizens… and we'll still have to fight them as they come out. ”

And so on and so forth until everybody is satisfied with how play is going to go. Arguments to plausibility simply will never satisfy either side in a case like this, and each participant feels that the other is trying to get away with something. When all they're doing is trying to make the game more fun. So instead, drop back from the fictional world, and discuss what you want as participants in the game. At that level of discussion, these things can be worked out in most cases.

Rarely it'll turn out that the participants' play priorities are so disparate that it's just not going to work for everyone to play together. But if that's the case, then you can fold the game peacefully understanding that nobody is wrong, you just need to find people to play with that have those same proirities. More often, however, a compromise of some sort is reached (“Let's say that the wind works to drive smoke into this one cave here, but it won't work for the others, cool?”), or one side just realizes that the other side's way of playing might just be as fun as what they envisioned. Or some other positive outcome.


See also The Parable of the Burning of The Village of Hommlet for a slightly different example, where the players seem to clearly be within their rights to do some sort of activity in the game (in fact the module may seem to inform the players that they should do the unfun thing), but that activity is not fun at all for one or more participants. See also My Guy Syndrome where players do things that are meant intentionally to be unfun for others, using the excuse “It's what My Guy would do here.” As if there weren't many other plausible things that their guy might do in those cases. Using a plausibility (or perhaps verisimilitude) argument to enable them to act in an anti-social manner towards one or more of the other participants. An example being the Parable of the Thief Who Stole from the Paladin. See also Edgelord.