Game Ideas Unlimited: Dungeons
September 5, 2003 in Articles

I haven’t written a piece specifically about dungeon adventures in a long time. Certainly many of the ideas have been useful for that sort of play setting, and some sprang from it, and sometimes an article would mention how to apply its ideas to dungeon play; but not since An Amusing Dungeon, the second article in this series, have I addressed a subject specifically about dungeons. Even then, the idea presented reached beyond dungeon play to scenario design; it was only mostly about dungeons.
Perhaps as I return to something that in my mind is really about dungeon play it will again prove to reach beyond that narrow confine into something more.
I want to talk about the random encounter, the wandering monsters of our early play. These came up in Encounters, where we talked about having random encounters with people the characters know. This time we’re going back to the more traditional encounter, with a wandering monster.
It’s an idea which has generally fallen into disfavor. Most gamers today, players and referees, prefer all encounters to be fixed. Yet I have found much value in the wanderer concept, which can be applied in many ways, if the right questions are asked. For example, in Tristan’s Labyrinth (which I wrote for Multiverser: The First Book of Worlds) all of the encounters are in a sense “random”. The dungeon model there is a seemingly unending maze of corridors, and there are no fixed places and no fixed encounters. It is assumed that as the player characters wander (or even if they remain stationary), so, too, the residents of these caves are also wandering, and at a random time paths may cross such that the player character and the hungry creature encounter each other. Where that happens is completely irrelevant to the scenario; the dice determine when it happens next, and what is encountered, and with that decided the question of where finds the response, wherever the character is at the moment the encounter occurs. The complaint commonly raised against such random encounters is that they derail the real adventure; that, though, assumes that there is another purpose to the adventure, and in Tristan’s Labyrinth there really isn’t one. It is entirely about surviving in a strange environment, and killing monsters.
Tristan’s Labyrinth maintains the randomness of such encounters in two senses: they occur at random locations (determined by random intervals of time in the game world), and they involve randomly selected creatures (the randomness determines sliding up and down a scale of relative difficulty in the encounter). An as yet untitled scenario for The Third Book of Worlds approaches scenario design in a different fashion which is yet very like this. The encounters in part of that world are sequenced in the order in which they will occur, but the timing and position in the setting at which they will happen is left to the judgment of the referee. In that sense, all the encounters are wandering ones, but they are sequenced such that the adventure unfolds dramatically through them.
All of this is connected to an early realization I had about wandering monsters and the lairs in which they wander. Often they don’t make sense in themselves. I came swiftly to identifying what I designated closed versus open dungeons, and tended to do far more design in the closed variety over the years.
The difference between the two ought to be explained. In an open dungeon, the roll of the dice can indicate any creature from a long list, which usually has been composed based on the strength of the creature (or creature group) rather than on any particular sense of it belonging where it is found. In my first dungeon, I on more than one occasion rolled the dice and produced a monster in a location where it would have been so completely cut off from anything that it must have been here all along, yet that too was impossible because it could not have survived here more than a short time. I also became aware that my players had as one of their objectives completely clearing the finite dungeon of all evil and dangerous creatures, one way or another. The open dungeon meant that there was no way to do so. Monsters seemed to appear by spontaneous generation, not by any logical means of arrival at the specified location.
Gradually as I completed that first dungeon, I capped the wandering monsters to a very limited number, all of which I listed and detailed as prepared encounters, and I tried to give connections to more and more of them to explain what they were doing here. In essence, I turned it into a closed dungeon, one which contained such creatures as had place within it and did not replace these when they were eliminated.
Most of my work thereafter took the closed form, in more and more sensible ways. If you were in an area with dragons, you might encounter a young dragon, or an otyugh or other creature that’s able to feed off the wastes and scraps of the monsters without becoming their victim. If you’re in an area in which gnolls, orcs, and hobgoblins each have colonies of their own, you’re highly likely to encounter residents of those colonies involved in activities that carry them outside their dwellings. In some areas, there’s a good chance of encountering random vermin that could tunnel in through the walls or hide in the debris, but in other areas these are extremely rare, because they don’t make sense.
It took a long time to get there. One small dungeon I created contained exactly twelve wandering monster entries, each of which had a defined course of wandering through the dungeon, resulting in rather complex tables to determine which creatures could be encountered near which rooms–necessary in a compound in which some areas were not accessible without passing through inhabited areas or locked doors. Yet even this had one of the significant features of closed dungeon design. When you destroyed the monster, it was destroyed. If your creatures are residents of a nearby lair, they’re part of the number there, and destroying them means they’re not there later. Only creatures who can come in from the outside can be replaced if killed. Further, some thought is given to how many of such creatures could actually be in the area as defined before it becomes unreasonable. You don’t necessarily have to know where each one makes its place of rest, but you do have to keep them limited to few enough that they could each have such a place. That’s good closed dungeon design.
I don’t mean that open design is unworkable. I have seen dungeons in which open design made perfect sense. In one adventure, my character was sixteen miles under ground when he entered a vast cavern in which the river flowing through it provided water for the irrigation of the fields of mushrooms and fungi which spread through the darkness, and we traveled several miles across these hidden cultivated expanses of agriculture before reaching the narrows through which that river poured into a rapid and beyond to a lake vast enough that it hosted two cities on its shores. If your dungeon is connected to that sort of vast landscape of kingdoms and nations, a stray carrion crawler or curious minotaur could quite easily wander into an area you thought had been made safe in previous adventures. He could even take a room.
Today I don’t create as much detail in advance; but then, I don’t run as many dungeons as I once did, and I don’t put the time into designing them. My scenarios are driven by many things other than maps and encounters. Yet I find this distinction between open and closed dungeons serves me in good stead. It helps to recognize which you have when you move into a new area. When the player character boards the pirate ship, you need to decide how large the ship is and how many pirates it carries, because it’s a closed scenario and must make sense as one. On the other hand, if he boards the abandoned space station which has been infested with alien monsters, it might be a large enough chunk of real estate with enough hiding places in it that there is no getting every last one, unless you’ve brought high-tech extermination gear. If it’s not a space station but a planetside spaceport, you’ve definitely got the vastness of the planet to provide more encounters. So whatever type of dungeon you’ve built, consider what’s in it, and whether anything else can get there.
Next week, something different.
—–
M. Joseph Young is co-author of Multiverser and Vice President for Development at Valdron Inc. His many contributions to online literature are indexed for convenience, and he looks forward to discussing these things by e-mail or on our Gaming Outpost forums.