Dungeon Contractor: Taxonomy


This discussion over dungeon construction became far too theoretical over practical. Where I had hoped to discuss nails and material weights I ended up talking gravitational waves. No, that's not the best analogy. I ended up talking zoning laws and building permits, the social side of dungeon construction. Truthfully, this is the side that no one wants to talk about, because it is the dreadful side. It is the bureaucratic side, the dull side, the side that is all regulation and not creative freedom.

And with this, wrote he, the theory would be finished. Mea Culpa.

This discussion over dungeon construction became far too theoretical over practical. Where I had hoped to discuss nails and material weights I ended up talking gravitational waves. No, that's not the best analogy. I ended up talking zoning laws and building permits, the social side of dungeon construction. Truthfully, this is the side that no one wants to talk about, because it is the dreadful side. It is the bureaucratic side, the dull side, the side that is all regulation and not creative freedom.

With this episode, we wrap up the politics. Next time is the mid-range transitional segment, and then we start talking how much earth to move. There does remain one final matter of obscure orchestration to deal with, and that is dungeon taxonomy.

If you have been paying attention, you will have noticed a sort of pattern to this discussion. There are complications that turn the issue of dungeon construction into a hefty test, and then always a cheat. Taxonomy is one of the cheats.

Taxonomy is the study of how to organize things, or at least the discipline that does the organization. There are many ways to organize things, and since we are discussing contracting let us discuss buildings. Say we took a small city, and set about organizing all the buildings in it. There would be many ways to do so, purpose of the building, style of the building, time constructed, and so on, and each listing with varying numbers of categories and subcategories serving a different sort of purpose well.

The taxiomatic guideline in dungeon design that is the most useful is "what is the dungeon supposed to do to the characters?" There are surprisingly few responses. The advantage of knowing the answer should be readily obvious. If the best dungeon is the one that fits the best with the overall adventure, but there are only a few types of fit, the matter becomes manageable for mere humanity.

To go with the construction analogy, there are an awful lot of ways to hold two things together: Screws, nails, bolts, and adhesives to name only a few. Each of these categories breaks down into many, many more, as a trip to the hardware store will tell you. But when a situation presents itself of how to join something, people first answer the question of what type should be used, and then move onto the more exacting matter of the specifically appropriate method.

So, let us to the list:

  1. The Bleeder: Arguably the bleeder is the original type of dungeon. It exists for the sole purpose of nastiness. The bleeder is a gauntlet fraught with danger, sometimes irrationally so. It is as much a test of endurance as any other quality, seeing how far the characters can push it before they get slapped down. It is also the perfect dungeon for character aggrandizement. The more you kill, the more you can take. Bleeders will always have a place. They are so rootsy and Old Skool that people will always nostalgize them. Furthermore they epitomize what a dungeon is supposed to be about: a deadly violent place. It is easy to fall in love with a bleeder, because what they aim to do can be done so well.
  2. The Organism: The organism aims for something approximating realism. It is an entity to be interacted with. What is important is how all the parts fit together and they will necessarily fit together in a nicely complex way. It survives in its own ways, the characters are only there to interface with the nominal goings on. For all I have railed against them, organisms are my favorite sort of dungeon. I like them because they remind the players the world does not necessarily revolve around their characters. Furthermore I like the strike team quality that such a dungeon typically engenders. Planning is the watchword, and complicated systems involve complicated responses. Of course, it is also the most demanding, in that if the systems are not up to speed the dungeon will not be. Similar ideas could be stated about the other two types, but, in organism, volume counts. It involves thinking and rethinking and overlaying levels of detail, many of which never get used.
  3. The Storybook: Most dungeons exist to serve the story, but sometimes the dungeons contain a story within them. The point is not that most dungeons do not have a story of some sort operating in them (even bleeders have a reason as to how they got that way), but that in some dungeons, the existence of the story is the reason behind the existence of the dungeon. Generally speaking, to get away with this the story has to be both obvious and simple, one loudly screaming singular storyline that drives the dungeon. The story is the reason to go from one room to another, and if not the real reason the meta one, as the players want conclusion as much as the characters want whatever their goals are. This easily stands to be the most powerful type of dungeon, but reliance on a story is a virtue and a vice. You can't turn The Sun Also Rises into a dungeon, and there are plenty of bad stories out there that make serviceable dungeons.

These are the primary forms. They fit within that whole psychological expression of gaming styles known as GNS. The logic behind that is the take home topic for the week. There are plentiful other forms, but take these three as their primary colors. Some ones of note are the Rubik, the bleeder-organism where the puzzles are tantamount and the place is more passively hostile, the...Horror, a bleeder-storybook where all the death serves to evoke a nasty mood, and the Show, the organism-storybook where the players are more or less caught in someone else's dungeon story.

The nuances are fun to play with, but they are not the real utility of the matter. The utility comes in getting to recognize the different forms and when they come about. The utility comes in being able to recognize what you need at any given time. If you can identify that what the adventure is calling for is a bleeder with a strong storybook opening, it makes design much easier because you know the direction to aim.

It makes plagiarism easier as well. If you can identify what it is about someone else's dungeon that makes it tick, it is easier to strip and retrofit. While I doubt that we shall see the day when all dungeons are rated in these terms in the magazines, it could be done.

Stay tuned for next time, when we get in a violent fistfight with the owner.

I don't even get the concept of the dungeon, and I always argue with my gamemaster about this. So what if it is in the name of the game? There are plenty of other ways to strengthen the party, and drive the story line. I don't want to see dozens of stone rooms and hallways covered by a fine layer of dust, and an occasional pentagram or artifact. I play these games to escape my mundane existence, so dazzle me with locales of granduer and amazement, man! Give me moss covered ruins on top of the tallest mountain, clouds as far as the eye can see; give me spires that reach towards the heavens, I WANT A FRAGGIN SHARED DRUG TRIP!

Mmmm, chips.

Possibly the best article of the series so far.

I like the nomenclature, nice simple tool. As I was reading I was noding along, remembering some of the dungeons I've played and what category they fell in.

To keep pushing your taxonomy, some of the best dungeons consist of three or four servings: ie: The intro is a Story book: the heroes meet the sole survivor of the last party that went in and got slaughtered by bugbears. As they enter it turns into a Bleeder: The bugbears have gotten reinforcements and set some traps. Then to hook 'em up, their is a bit of Story book (how it got that way and what might happen next if the PC's don't intervene). It becomes an Organism: the heroes have to figure out what's going on and how to stop it.

Throw in a few Bleeder areas, some Storybook (the carrot in front of the PC's to keep them guessing) and voilà! The classic one track dungeons work well too, for a while.

I think though that the GM plays a major part in her/his telling of the story and game mastering. A great storyteller but bad tactician will mostly master storybooks but won't make good bleeders.

I'm just rambling the heat wave is getting to me, not enough caf.

And the shared drug trip is a must for anygame (the nice things about RPG's is the lack of any serious side effects and the fact that you can stop your trip on command).

Gotta get caf.

Cthulhu Matata.

Well, Aforseille, the term "dungeon" doesn't necessarily refer to a series of caves or rooms. The same classifications can belong to keeps, forests, cities, deserts, mountain passes, or anything you please. The actual setting is irrelevant; the area will be designed either to challenge the players, to draw them along a story, or to make them interact with a living, breathing environment.

Whoops--you killed the dragon.. that's gonna attract the kobolds hiding in the hills. They'll trap the entrance while you're looking for its lair so they can loot it before you get the chance! Did I mention the bugbear mercenaries they can afford now in their war with the local goblins? But then the hard-pressed goblins will summon a demon and make a desperate deal for its help in a climactic battle--but the battleground includes the innocent village that was, until recently, terrorized by Mister Dragon.

See how easy that was?

Ok, I know what you're saying. I even said that the story line can be driven using OTHER ways. And despite the fact that the "dungeon" can be something that isn't a series of caves or rooms, it usually is.

Aforseille - Read the earlier articles in the series.

Aubri - Read the earlier articles in the series.

Sammy Q - Thanks, but you've slightly misread the notion. A bleeder or organism can have a solid plot hook like an organism or storybook can have a bitchin' fight at the end. But, yes, the best of the best are generally a mixture.

-J.S. Majer,
Desperado in the Culture Wars,
Cosmic Bandito Emeritus

Ok, I read the articles. Ok. That doesn't explain why 99.9% of gamers still have the same basic view of the "dungeon".

I read you JS but,

What I was refering to are multiple dungeoned adventures, or multiple layered dungeons like The Night Below boxed set. Which had many sites that "behaved" differently, but I guess it really depends on the GM too.

But what I think you meant was that a rubic-type dungeon can have varying degrees in its ingredients depending on the dungeon areas without switching from type to type.

Here are a few known dungeons and how they fit into your taxonomy:

Bleeders: Tomb of horrors, White plume Mountain, Lost Mines of Tsocjanth and the infamously bad Expedition to the Barrier's peak.

Organisms: The Floating Citadel in "Flame's Revenge" Dungeon magazine #9 I think. Milton's Foly in the Freeport Campaign.

Storybook: Well…. I'm stuck. I've played some but they were homemade rather than off the shelves.

Rubic: Dragon Mountain, The Night Below (the main dungeon), The temple of Elemental Evil.

Horror: The ultimate: Castle Ravenloft, Castke Forlorn.

We use to classify modules, not dungeons, using a simpler taxonomy: The Investigation, the Indy (Idiana Jones like adventure), the Cthulhu (hopeless), the Mouse trap and the Night of the Living dead (Mindless hack and slash).

Seems like some are classifying by plot type, and some are classifying by genre, while others are using game structure. These don't really mix.

Castle Ravenloft is a dungeon structure set in the gothic horror genre.

The following link is the best set of game plots I have ever seen:


Aforseille, the problem is inexperience and lack of talent. There aren't many good writers, and there aren't many good DMs.

Yeah, although I need to add that there will be disagreements over matters of what gets called as what. Taxonomy's like that.

There's an issue of Dungeon out there (10, I believe), that sums up my paradigms of the things. And my problem is that I don't own many off-the-shelfs outside of a lot of friggn' Dungeon mags. And I'm going to be going into all of this, breaking each division down into it's little components, but...

Tomb of Horrors - Rubik. Way more rubik. Deadly as sin, but something with that many picture puzzles can only be one thing.

Dragon Mountain - Ah, now here's one of note. It's supposed to be the Organism to end all organisms, a devilishly complex political system that the players come charging into. It ends up reading as a super-bleeder, albeit a trap heavy one. The Organic qualities are not fleshed out enough for it to work that way. It's a bleeder in a pure sense, a pure and total test of character endurance, as opposed to the more typical smash things up motif.


I'd call Dragon Mountain an example of bad Plot Coupons in action.

A "plot coupon" structure is basically the Donkey Kong approach to storytelling, where the plot exists only to get from A to B to C, without any overall story developing. A similar structure exists in the awful Speaker in Dreams from the official 3e adventures.

Plot Coupons drag players by the nose, without offering much choice and without inciting any overall curiousity. Usually, encounter A points only to B, which points to C, where the whole thing ends and everything is revealed (if there is anything to reveal). In Dragon Mountain, there isn't really much of a secret - nothing that you didn't know from the first kobold encounter. After a half a year playing the same kinds of encounters, without any new story elements being revealed, it's got to get boring.

A good example of an organism, at least as we played it, was The Sunless Citadel. This tight little gem has ongoing hostilities between a kobold and goblin tribe on one level, and a greater threat deeper. The deeper you delve into the dungeon, the more you know about the overall history of the place and what happened there, as well as what the main Big Bad is doing. But before you get there, you must negotiate or fight your way through the ongoing skirmishes on the top level. Our own party had a great time negotiating passage to and fro with the kobald leader, who I played as very shrewd, lawful, opportunistic, and hateful. She used the PCs to help her cause against the goblins.

Speaker in Dreams: barf barf barf! What a crappy adventure.

Dragon Mountain is meant to be an organism, although I have only read through it and I don't know how the DM I played it with (years ago) changed it to make it into a good adventure at the time.