Bottom-Up Worldbuilding

[Yora]

I've started worldbuilding for my campaigns 15 years ago, and I was always enticed by the glamour of continent-sized settings, with long histories and dozens of elaborately detailed cultures and prominent landmarks. Back when I was 22, it didn't bother me that such a thing might take me ten years to finish. It would absolutely be worth it. Starting at the top with a grand concept and working my way down to have everything fit seamlessly together would be absolutely worth it.
But looking back at it now, the use I got out of all of it for actual campaigns was very modest, and those campaigns would really not have been very different had I run them in any other generic setting. The Top-Down approach has not served me very well, and as I am getting older and wiser as a GM, even with all the experience I gained, I don't see a great setting that generates great campaigns coming from it even if I spend another 10 years chipping away at it. And by now, I really don't want to prepare for a campaign some years down the road. I want to have something that I can take out on the road in a month.
The big issue I have experienced with the Top-Down approach is that it takes very long to produce content that you can actually use in an adventure to have the players can interact with. Which is why I am coming around in my older years to seriously give Bottom-Up a real chance. Start with specific things that will be cool to put into the players' grubby fingers. And when you really look at it, almost none of the great expansive fantasy worlds started like that. Some big brand RPG settings being the exception, like Eberron and Golarion. But usually it's really just one pretty small region with its own local factions and locations, and only the vaguest hints of other countries beyond the borders of the local map.
I'd love to discuss and hear how other people have made good and bad experiences by starting the creation of a setting with just one contained location and expanding out from there. Either to keep it a modestly sized region or to build it out into a much larger world.

While discussing this question with someone earlier today, the idea of a setting's skeleton came up. In my years working with Top-Down approaches, my experience was that the content that I created was all about creating often quite fascinating silhouetes, but there was little in the way of working parts when you looked inside of them. General outlines, but little in the way of internal structure.
Thinking of a setting's skeleton seems a quite clever approach to me. Bottom-Up does not have to mean that you start with only one village and no plans for what lies beyond the surrounding hills, which you will deal with when the campaign gets there. The skeleton can be the conflicts of various general interest groups, broad environmental and economic presures on the local populations, and the dynamics that result from all that. And instead of the concerns snd long term plans of the great and powerfu, these would be things that are going on on the ground, with the people that the PCs would be directly interacting with face to face. If this structure is in place, then the specific details that are unique to each town, valley, or island can be attached to it when these places become relevant to the campaign that is being played. With the skeleton in place, new locations, people, and situations can be integrated into the existing material and feel like coherent expansions of what is already there.

One method that I used heavily when building Top-Down, but which I think is just as usefuk, if not even more so with Bottom-Up design, is stocking a big box of props: Cultures, monsters, magic, locations, conflicts, deities, environment types, and all that stuff. In my experience, simply making things up as you go, and going with whatever feels fitting at the moment, usually ends up creating something that looks just like what has been done a million times before. Things feel fitting precisely because everyone else has combined the same pieces in the same way countless times before. To make a new setting distinctive and coherent, it is useful to decide in advance what kind of setting elements you think would work with the innitial vision, and even more importantly which ones would not work with it but conflict with it and dilute or distract from a specific style idea. Not everything that goes into the prop box has to be used when fleshing out the setting as it develops. If some pieces never seem a good fit for all the places you work on, so be it. But when looking for ideas, it can really help to start rummaging through the prop box and see which pieces have not been used yet, or which have worked wonderfully in the past. And when you got an idea to use something that is not in the prop box, then use that to think again why you didn't put it in originally, and if it would really mesh well with the other pieces you already used. If you decide that something completely new should be used anyway, take a moment to add it to the prop box list. If you used a new monster or form of magic now, then you can use it again later in other situations and context to make it feel more like an integral part of the word and not just a one time oddity.

[brian 333]

+1.

+10, actually.

Knowing that the Kingdom of Honoria has a long-standing dispute with the Witch Queen of Malevilund is an important detail if the players are operating on the borders of those two nations. Not so important if they are headed the other way.

But, if in the brainstorming of the parts you need, the idea comes up, it is easy to turn into a tavern tale that adds historical depth to your world.

Don't discard ideas you have along the way, but stick to what your players will be dealing with and flesh out random stuff when the next adventure is ready. Having the framework, (stick figure,) allows you to grow in a direction that fleshes out your world over time.

[Mechalich]

There's a fundamental issue with world-building in that a planet is simply too big to function as a setting, but choices made at the planetary scale have vast influence at much small scales where a setting actually operates. This is, of course, true in the real world as well, in that many global traits of planet Earth have vast influence over smaller-scale life, and as those traits change over time (ex. climate cycles or continental drift) life has to change in response.

I feel the trick is to devise the planetary-scale changes before trying everything else, since they impact everything if significant changes are made from the Earth baseline to things like gravity, or oxygen level, or which sapient species are present. However, once that's done the world-builder needs to drill down to an operational zone that can be handled in a reasonable time frame. This can be done both from a top-down or bottom-up perspective, but bottom up naturally maintains such limits while top-down has to be more conscious of them, but it's necessary regardless.

The trick is to keep to a reasonable regional size for the setting and to limit the number of things in the 'prop box' for that region, which should be considerably less than the whole world. For example, even the largest pre-industrial Empires topped out at a few million square miles (Rome, for example, peaked at 2.3 million) and those Empires held nothing like the whole world's diversity in terms of biodiversity or cultural diversity, or even technological diversity. Earth has ~150 million sq miles of land area. A fantasy setting that is in a highly arable area probably shouldn't grapple with more than 1% of that (settings that are primarily steppe, desert, tundra, or jungle can be larger because the population density will be much lower).

[Yora]

The more I think about it, the more I see that Bottom-Up is a quite different idea form Inside-Out, though in practice it makes sense to combine them and do both. In many places you see people recommend to just start with a town and a dungeon and begin playing, with more stuff being added as the party travels and explores. That Inside-Out worldbuilding. But most of the time, my perception of these suggestions is that it is assumed the world is going to be "Generic D&D Land". You simply create specific locations for an environment you already know.

But Bottom-Up can also mean that you start with foundations for the whole setting, which provides structure and a framework for all the things you later put on top of it. In a Top-Down approach, you generally start with a big picture of landscapes, populations, and perhaps top-level politics, diplomacy, and conflict. And as you work your way down towards specific locations, you have to start thinking increasingly about how those broad stroke ideas can translate into adventure content, often for new, unknown 1st level PCs or something of that kind. That's where I always had a huge problem.
Bottom-Up means starting with establishing structures that are needed to support certain kinds of adventures that are desirable in the campaign. An aspect of what I described as the bones of the setting. When I decided I want to start worldbuilding back from square one yesterday, my choice was to start thinking about low-level adventures for no-name treasure hunters first, and from that exploring what kind of environments, cultures, technologies, magic, and monsters actually will contribute to that.

My first choice was that I want to make a setting that is tailor made for Classic Dungeon Crawling play. This means the setting has to have lots of small and medium dungeons that are inhabited by monsters and filled with gold and silver and magic items. Regardless of the landscapes and cultures I will be going with, this is already the first thing established about the setting at the most fundamental level: Dungeons, monsters, money, magic. I also want to go with small civilization as a purely personal aesthetic preference.
Now I've been long very critical of the old cliche "Back in the day everything was great, but now it all sucks and will only get worse forever". I get an allergic reaction to Lost Golden Ages. But if that's what it takes to have dungeons, monsters, money, magic work believably, then that's what the setting needs to be. Fortunately, someone suggested that the age of riches and magic was actually awful and the people today are glad it's over and that they live in small, fractured, and isolated communities again. This can take on some libertarian or hippie undertones, but I like that much better overall.
From this, I got the idea that the backstory of the setting is that one time, a warlord managed to conquer a majority of the known world under his rule. This produced tremendous wealth, but only for him and the nobles loyal to him, and the common people had to suffer severely for it. The emperor's mages had the resources to create large quantities of magic items, but those only served to secure the power of the nobles and subjugate the masses. So now we got a lot of gold and a lot of magic items, but nobody today thinks fondly of the time that produced them.
And that means most people have some cultural reservations about going to look for them. Even if you don't consider the treasures to be cursed, it doesn't look good on you if you start getting power and influence with the tools of the hated ancient tyrants. That also provides a reason why there's still plenty of the stuff in the ground to be dug up.
And finally, there's the question of how the Emperor was overthrown and his Empire fell apart. Since the nobles had hoarded all the magic to keep the peasants in line, a regular rebellion seems unlikely, and those usually just produce a new tyrant at the end. Instead, I am taking a note from history from famous conquerors like Alexander and Toyotomi Hideyoshi, and as soon as the emperor was dead, his generals started fighting each other and turning their provinces into competing kingdoms that simply exhausted themselves against each other. In this case until there was nothing left. So now I got a bit of history as well, pretty much for free,

This all seems like a quite exciting setting to me, but there's not actually any kind of specifics yet. Not one map, one landscape, one town, one dungeon, one NPC, one monster. It's pure foundation. Bare bones with no meat on them. As Bottom as it can get.

[brian 333]

Again, +1

Begin with a seed and let it grow.

The region from which the adventurers come may not be representative of the whole world. For example, there may be a democratic society of dwarves who keep the barbarians of the fallen empire out because they are uncouth and stinky, or an elven tyrant who excludes them because they are disruptive influences.

[Yora]

I had been thinking for a while about how I want to incorporate gods into my new setting, and do it in a way that makes them more meaningful and involved than I've usually see in RPG settings.
And then I remembered to go back to the initial idea to make a setting that is build around a specific type of play. Before asking how to make the gods interesting, the better question to ask when working Bottom-Up is how the gods can be included in the gameplay?

The type of play I want to build for is classic dungeon crawling, so ideally some two thirds of playtime are spend exploring underground ruins. How can the gods be an element of that exploration? If you approach it like this, it becomes easy: Put a lot of depictions of the gods and their symbols and iconography on the walls, doors, floors, and pedestals throughout the dungeons. Not just as decorations, but to give the environment additional context information about who has build the complex and for what purpose. And to step it up another level, make this context information useful hint to deal with the obstacles that are encountered in the dungeon. A trickle of murky water coming from an old spout under the image of the god of death will look very unappealing, while the same drips of dirty water coming from the image of the goddess of healing might look like a chance of salvation for a mortally poisoned party member. A great door with symbols of the forge god can promise a hoard of enchanted weapons if the door can be opened.
The players don't need to know all the gods and their typical symbols. But make them prominent enough and players might make notes and sketches to consult with a priest or sage when they go back to the town. That's added interactivity that will reinforce the presence of the gods in the players' minds.

Another big way in which "religion" plays a big part in a typical dungeon crawling campaign is going to the temple in town between adventures to have the various ailments treated that characters have suffered in the dungeon. In some versions of Dungeons & Dragons, all clerics have access to the same spells, and even when clerics of different gods have different specializations, the basic spells to treat wounds and curses are generally universal and available to any cleric of sufficient level. The result of this is that it really doesn't matter what god a temple belongs to, and there's no reason to interact with the specific deity and faith in any way.
But instead of making the priests of the temples clerics with the whole cleric spell list, the temple could instead be build around a sacred shrine or relic that imparts only a few and very specific powers on the priests. Not every temple is the same, and to have specific conditions remove or to acquire protective item against specific threats, the players first need to identify the kind of deity whose priests could help them and then find a temple that is within their reach. Now again, the differences between the gods and priests become important, and it is practically useful for the players to learn the differences.

[Catullus64]

I had been thinking for a while about how I want to incorporate gods into my new setting, and do it in a way that makes them more meaningful and involved than I've usually see in RPG settings.
And then I remembered to go back to the initial idea to make a setting that is build around a specific type of play. Before asking how to make the gods interesting, the better question to ask when working Bottom-Up is how the gods can be included in the gameplay?

My approach is emphasis on local & small scale gods. Whenever I design a small region to be an adventure setting (which frankly constitutes most of my worldbuilding activity), I always make deities for prominent natural features nearby: rivers, woods, and mountains especially. Larger towns & villages might have patron gods that literally dwell in the local shrine. Give them a few personality traits (this god always speaks in alliterative verse, this one enjoys offerings of wine, but please, for all our sakes, don't give him too much) and specific magic powers (this god is a shapeshifter, this god heals the sick, this one makes illusions), and think about them mostly in terms of their relationships to the villagers. How do the villagers propitiate them? How do they benefit the villagers in return? How do they express their displeasure? Create them like you would NPCs, not like static setting elements.

I also come up with gods of minor domains that adventurers are likely to invoke. The big gods of storms, civilization, death, rebirth, etc, aren't too interested in your individual problems, but the gods of crossroads, locked doors, campfires, or taverns? Adventurers are going to have business with them all the time.

Big, important gods have big, important temples, and their religion is officiated by big, important people. The common folk certainly honor their existence and revere them, and might join in rites related to them on important festival days and in times of crisis, but their day-to-day religion isn't generally going to revolve around the sort of cosmic, setting-wide deities that fantasy RPGs like to focus on. Therefore I think that it's ok to leave your large-scale deities as broad-strokes archetypes with loosely understood myths and characteristics.

[brian 333]

Braethandra, Goddess of hearth, home, and cooking, revered in every village home, is propetiated by offering care and comfort to a housecat.

Depicted as an older woman in fables and winter tales, she has a magic broom that she talks to and it seems to understand and obey her.

In the trade towns she is patron of herbalists, midwives, and alchemists.

In the city she is an aged crone who is mother of the chief deity. She is patron of magic, alchemy, poisons, and political schemeing behind the throne.

The players know the familiar hearth goddess and are shocked to learn Grammaw is a badass in the city!

Dimemol is patron of a clear pond with no bottom. Clerics with water breathing spells visit his underwater temple, but mostly pray in its aboveground mirror.

Wine and fruit are offered, and the health of the river his spring feeds is domain. All in the region revere him and pray to him for regular rain.

200 miles away the locals pray to another god for rain. They wouldn't know the name Dimemol if you told them.


My formula has always been to let the players of clerics design their gods and only add my own deities as I need them. It's fun to watch the players of two clerics hash out how or if their deities relate and interact. Since I have always loathed DMPC, err, deities taking over campaigns, I have always had them act exclusively through characters with access to divine magic.