Greenhouse Dungeons
In this blog post, Luke Gearing used an opening line I liked. I’m stealing it.
Mouth Brood by Amanda Lee Franck is the one of the most personally inspirational RPG books I’ve ever read. The premise is straight outta Vandermeer: the players are sent to investigate an ecosystem that popped, all of a sudden, out of nowhere into a place it definitely shouldn’t be. There are, of course, organisms in this place. Some of them are extremely fucked up. Most of them are just living their lives. Your job is to collect samples of all of them.
Reading Mouth Brood almost a year ago now (after it was recommended to me by Luke, incidentally) was a brain-rewiring experience. It was the sequel to another such experience, the one I had a few years before when I finished Annihilation for the first time. It occurred to me again that it was in fact possible to represent the bizarre, horrible beauty of raw ecology in art.
Ever since that second brain wave, I’ve been slowly struggling through how I would tackle a similar premise. Over time, and with more reading, this line of reasoning has evolved (metastasized?) into an entire way of thinking in my mind. A different way to approach RPG design, concerned with that strange feeling one gets when scuba diving, or examining the underside of a log you’ve just flipped over, or watching flies dance in the sunbeams that stream between fronds in the greenhouse.
I have not yet completed or published any works in this vein, though I am approaching the time when I can. This essay is a lab report from an experiment still in-progress. It will need to be updated. It is also, notably, an exercise in setting design goals that are artistic rather than incentive-based. In this way, it is more for myself than anyone else. It is quite likely that the kind of game I am describing seems nightmarishly uninteresting to you. This is the nature of personal obsessions, as well as experiments.
Simulationism (?)
When I started to think about ecology in games, the first place my mind went was to simulations, that is, probability tools of a greater-than-average complexity designed to take inputs from the referee and adjudicate outcomes. On the surface, it seemed to me the most obvious way to represent changes and basic functions of non-existent ecosystems. It evokes images of a living system that explodes under its own power, fully formed, each time the book is opened, independent of both the writer and the referee.
However, the more I think about it the more wrongheaded this seems. Mouth Brood, for instance, completely avoids straightforward simulation. Instead, it reveals the ecological matrix with a variation of Troika! “mien” tables—when encountered, every organism is in the process of demonstrating one of its relationships to the creatures around it. If any modeling of the system is necessary, it is left up to the referee. That self-generating system I imagined might just not be suited for RPGs as a medium. After all, any computation is traditionally done manually, and ecosystems are notoriously difficult to compute. Finally, it occurs to me that the main draw of simulation is to remove control from both myself and the referee in service of immersion, but the participants who have the most to gain from this, the players, are perhaps unlikely to notice. Scratch that—they are quite likely to notice if their referee has to spend an inordinate amount of time muttering to herself as she repeatedly references her calculations from behind the screen every time a twig is stepped on.
All of that being said, some simulation may still benefit us. Rules do indeed elide, but when dealing with systems as complex as your average tidal pool there is a great deal of context which can only be accessed with a certain degree of elision.
As an exercise, let’s try to imagine what that might look like.
After defining their ecological relations to each other in the text, give every organism a Population Value between 1 and 100. When an organism dies as a result of the players’ actions, subtract one from this value. Add 1d6 to the values of the organism that the dead creature feeds on, or otherwise tends to harm. Consider populating these changes throughout the entire bestiary. Next, phrase room encounters in terms of possibilities: where you would otherwise station a specific encounter, instead list the organisms that might appear there. When the players enter, the referee rolls d100 for each organism listed—they only appear if the result is lower than their Population Value.
Is this practical? Functional? Fun? Further research is necessary. Even exercising restraint, it’s certainly more crunchy than most dungeons ever get just for figuring out what’s in a given room. However, it does enable us to model things like extinction and the fragility of ecological balance that previously would have to be improvised by the ref on the fly, if at all. I can certainly see this working, potentially very well, for a game in which the players are tasked with population control. Another interesting outcome is that it codifies every death not only as a action within the ecosystem, but also as a choice. Do you kill this scary predator that is threatening you, knowing that it could potentially lead to more frequent encounters with worse things down the line?
Intentionally Unintentional
Something commonly said about the ecological survival video game Rain World is that it “does not care about the player.” What people generally mean by this is that, in contrast to many other video games, the world and “enemies” are specifically designed to give the impression that they are not designed to impart a specific experience. They are not, the game seems to be saying, simply obstacles in your path, nor are they thoughtfully-placed boons to speed your progress along. Instead they are simply organisms, with their own goals and behaviors, and their impact upon the player’s progress is as unmotivated as a fox biting the neck of a mouse.
This is, of course, a well-maintained illusion. I followed Rain World’s dev cycle over Tigsource. I have seen firsthand how much thought went into the design. The developers cared very much about player experience, but the genius of Rain World lies in their ability to obscure that.
When looking to TTRPGs, we again have a medium problem. The player experience is the whole of the game—it can’t be contrasted against an inflexible system of code like Rain World does. Additionally, trying to communicate to the players that we don’t care about their progress has to be done through a mediator, which makes the delicate balance of intentionality and obscurantism more difficult if not impossible. It also starts to sound like a bit too much protestation.
The solution? We must reject the question. Unlike video game developers, the player experience is truly out of our hands—we just need to lean into that. We need to dare to make things that seem hostile and odd and maybe even a little bit boring. In doing so, we are essentially daring our referee readers, and the players by proxy, to find the fun. We don’t have the luxury of smuggling in intentionality into the world, because we’re not building the world. We’re inspiring someone else to do it. So let’s just leave it out. Give them wild, purposeless things, and see what they make of it.
Against Monsters
In the vein of the above, it might be helpful to disabuse ourselves, at least in this context, from “instrumental” thinking—the kind of language and design that positions the world as a thing primarily in service of a user experience. We will use the monster as an example.
Mouth Brood is primarily a bestiary. It has twenty entries. It has exactly one monster.
Monsters are very good. I love monsters very dearly. But here’s the thing—nature has few monsters (you can make a very strong argument that it has only ever produced one). Not just because predators are rare, and big, scary predators even rarer, but simply because animals are typically not interested in violence for violence’s sake. They are, strictly speaking, not “interested” in anything much at all. They are animals. They follow their own, obscure biological prerogatives, usually having to do with feeding, mating, and escaping harm (sometimes, all three at once—here’s to you, male anglerfish). The result are creatures that, when you observe them in the wild, are always distinctly un-monstrous.
By no means am I suggesting that greenhouse dungeons should not have dangerous creatures in them. I think that it would probably be best to have quite a few. But what kind of things can we discover if we leave behind the artificial framework of “enemy,” and represent the inhabitants of our greenhouse for what they are? What kind of strange beings live on yonder ridge, who run fingers through each other’s hair and cry at the sound of thunder?
Connectionmaxxing
There is an oft-repeated phrase in the popular discourse around ecology: the idea that “everything is connected to everything else.” As I understand it, this is not precisely true. It would be more correct to say that anything could be connected to anything else, and indeed often is.
This is in sharp contrast to much of mainstream RPG design. I read many dungeons which seem to be a series of hermetically sealed compartments, through which only the players wander freely. Creatures act out vignettes within their given rooms, but rarely if ever do they leave them of their own accord, nor do the outcomes of their little mini-stories have much of an impact on their surrounds. There is absolutely nothing wrong with this style of design—on the contrary, it is perfectly unobtrusive. It gets out of the way of play. But living things have a habit of being perennially in the way—they go where they’re not supposed to go, do what they’re not expected to do, and always, always impact their environment.
As we saw in the section on simulations, mucking around with random encounters might get us part of the way there. There’s a bit there where I suggest that you propagate every shift of the Population Value throughout the entire ecosystem. I don’t yet know how practical that would be, but it would certainly aid in representing the chaotic downstream effects of ecological change.
This method can also be carried through into the design of the dungeon itself. Luke’s massive hexcrawl Wolves Upon the Coast has this fantastic quality—things connect to other things, especially within regions. Help one faction here, and you’re likely to piss off another just across the way. If you come across something interesting, poke about a bit and you’re likely to find someone who’s either looking for it or has been affected by it. Everything that has moved has also left a wake.
Within a much smaller, self-contained greenhouse, it may be possible to push this writing technique even further. Here’s the water source for the cavern—you hold the whole ecosystem in your hands. Here’s a species hounded by predators—if brought somewhere else, it will become invasive. Here’s an immense piece of carrion—it draws scavengers from far around, emptying the surrounding areas.
You can see how this web of influence is how we might begin to add back in some of the players’ power that was lost when we so rudely de-centered their experience earlier. It is a very different kind of power than the sort conferred by leveling up or learning a new spell. It is the power of transdiegetic learning, the kind of power that can only be found by applying inductive and deductive reasoning to one’s surroundings. It is the application of those uniquely human traits which give us such dangerous mastery over the ecosystems which taught them to us.
Evolution
This is a crucial, perhaps the most crucial component of representing ecology: how things change and adapt to change over time. Unfortunately, even I am not so foolish as to aspire to model real evolution with RPG math. Nor would it necessarily be productive to simply write evolutionary change into the text—such sections would be unable to account for the biospheric changes that would inevitably occur during play. Luckily, we are blessed with our referee, and her imagination. We must leave this to her, as we do all things.