After a busy week that limited my writing, I'm ready to revisit the topic of mixing the megadungeon with horror. Last time I talked about it, I was pointing out how the current megadungeon, the Black City, has evolved into more of a zoo than I had hoped (Help, My Megadungeon is a Zoo) - now I'm ready to turn to the question whether mixing the megadungeon structure and a strong horror theme is even viable.
Let's take as given that the megadungeon is a good structure for a campaign. There are plenty of folks that have tried to run one, unsuccessfully, and written off the structure as dull and unviable. That hasn't been our experience; however, the defense of the megadungeon format is a different post. I've got other problems to tackle here. The megadungeon is a good way to manage the Sandbox Triangle. The triangle comes from the non-gaming disciplines of project management and product development; if you increase the features of something, like software, either the time gets extended or you have to throw more resources on the project. Newsflash: your campaign is also a product. Furthermore, in sandbox terms, the creator is always balancing the freedom the players have, with the size of the sandbox, and the level of detail - particularly during inception. The dungeon, any dungeon, is a constrained structure that limits freedom logically, allowing the DM to generate greater depth of detail or a broader scope for the same time effort.
There is an appropriate discussion to be had about the merits of a gigantic dungeon with minimalist descriptions, versus the small dungeon with baroque detail. Both emphasize different corners of the triangle. We'll come back to that point.
So now I turn my attention to two exercises - listing the characteristics of a good megadungeon, listing the characteristics of good horror scenarios, and then taking a look and see where there are and are not synergies between the two.
The characteristics of a megadungeon: many large levels, highways in and out to various levels, themed levels and sub levels, interesting set piece locations, multidimensional challenges (including puzzles, tricks, traps, and combat challenges), lots of empty space for exploration, deeper levels = deeper danger, resource management, factions and NPCs and varied threats, and a steady flow of information to the players that allows planning and strategy - typically through patrons, rumors, and treasure maps. Megadungeons as written products are frequently sparsely described, to support more scope and larger levels.
Horror scenarios usually involve some kind of twist or trick to bring home the horror - I listed a bunch of them a while ago, here: Horror in D&D. Everyone has their favorite techniques, and I always learn something when folks share. Horror scenarios usually have a strong contrast between the mundane world and the nmonsters; monsters are rare, dangerous, terrifying, and unknown. Horror requires atmosphere, a build up of tension, a reveal. NPCs antagonists verge on the gothic or grotesque. There are a wide range of effects you can achieve, creating stress, tension, uneasiness, the creeps, shock, mounting dread, sometimes even scaring the players themselves. Horror scenarios are baroquely described, to support the depth of atmosphere and detail; the locales are frequently small or limited.
So how do the demands of the two structures match up? One of the largest issues I see is the difference in scale and detail - the sparseness of description in a large megadungeon vs the baroque detail of a small horror site. Next is the need for a varied set of threats in the megadungeon, versus the rarity and danger and uncanny monsters in a horror scenario. Then there's the paradigm of dungeon level = danger level, and GP = XP. This implies balance and the ability to plan risk vs reward, which frequently has no place in horror. Finally, in most horror games, the player characters have a hero-complex and are motivated by philanthropic goals or save-the-world morality; D&D characters are free to be heroic, but the game doesn't dictate it, and actually rewards a bit of callous treasure grabbing.
I have some ideas on how to reconcile some of these problems. The megadungeon could still be large and sparse, for instance, reserving the horror bits for detailed and intricate set-pieces scattered throughout. The unique and terrifying monsters can be reserved for the same set pieces.
Avoiding the megadungeon zoo and the banality of monsters in the rest of the encounters is a major problem. I could take a cue from the Black City and make many of the opponents mundane threats like other explorers. I'm not sure that could carry a large megadungeon on its own. But I feel like the answer is in this direction - use lots of human opponents, of many different stripes, and perhaps even use recurring monsters that fit the theme and become mundane over time, to fill the role of combat threats. The monsters in the set pieces are there to bring the weird and the horror and the atmosphere.
I did broach the question of Call of Cthulhu versus D&D style play some months ago (LOTFP vs Call of Cthulhu), and I'm quite confident a D&D style exploration game with the GP = XP paradigm works better for a sandbox than COC's heroic style, which compels characters to follow a plotted path. I tried a COC sandbox a year back and it left me cold.
My fixation with the megadungeon structure is why I keep coming back to it over other project ideas, like that Spanish Main sandbox or the Colonial hex crawl. An easier horror campaign could certainly be built by creating an early modern sandbox area and populating it with smaller locales made by converting Call of Cthulhu adventures to a D&D format, or using them for inspiration to make something period-appropriate. After 30 years of short Call of Cthulhu adventures in my collection, there are tons and tons of haunted sites, sleeping monsters, nefarious wizards and cults that could fill out even a large sandbox. That option is always available, if I end up dissatisfied with the horror themed megadungeon inquiry.
I think when I return next, it will be looking at a few of my Junkyard megadungeon ideas, like Harror Home Manor or the Benighted City of Lichtstadt, and seeing how they'd fare against the challenges I've presented here.