Writing a campaign setting...

[Grinner]

Convention holds that to create a setting, you start by drawing a map. After doing that, you define the political boundaries and then describe the cultures that reside within. While doing this, you give consideration to what sorts of creatures inhabit the lands and how they act. Slowly, you build on these things, expanding on the technology, religion, politics, and customs. If you've managed to make it that far, you might then add a few plot hooks and call it finished.

I have to wonder if that's the best way of doing things though...What's recently occurred to me is more or less going in reverse order. Know what you're aiming for, yes, but start out by defining plot hooks. Plot hooks are the real meat of any setting. They're what GMs latch onto. The details help a knowledgeable GM set the stage, so to speak, but they're not very useful on their own.

It's also worth noting that you really don't even need to know what you're talking about. You can just make up some names and fill in the details later.

What are your thoughts? How do you normally go about worldbuilding?

[BWR]

On the few occasions I've tried to do this rather than use prexisting campaigns, I've throught about all sorts of other stuff than geography. What is it in flavor and execution that makes this different from other settings?

The first one I tried, for D&D, was based on the idea that there are many (hundreds? thousands? more?) of Prime Material worlds that exist. Even though the Blood War mostly ignores the Prime, at some point one of these worlds is going to be the site of a major incursion for whatever reason. What would a world look like where the good ol' 'infinite' summoning the tanar'ri and baatezu can pull off actually happened? They came and fought and killed eachother and the survivors mostly left. What was left was a ruined world where the survivors were holed up in a few enclaves protected by powerful casters. The outside was a blasted wasteland, full of wandering fiends who didn't bother to go back home. The City of Fortir was to be the starting point of the campaign, a strict theocracy ruled by the Council of Clerics, whose iron will and power had kept the city safe from the ravages of the fiends. A highly lawful and somewhat narrow-minded place always on the lookout for fiendish incursions. Games would have focused on local politics, resource savenging and later fiend hunting.
Geography was an afterthought.

[Eldan]

Don't start with maps. Start with ideas.

1. What makes this world different from other worlds? Why would anyone use this world instead of a published one?
2. What is the world about? What themes are important in its background?
3. What kind of campaign would I run in it? Genres? Focus? Time period?

Then think about politics. I think that's more important than geography.

Start with number three. Define a list of things players should be able to do. If you want them to explore wilderness, you need wilderness and things to find there. So far, so obvious. For political games, you need opposing factions, probably at least a handful of them. And so on.

[Corneel]

The Giant has written a nice series of articles about worldbuilding. And the first article (http://www.giantitp.com/articles/YPgbz2j3PckGjjviJU5.html) might be especially of interest to you.
The articles can be found by clicking "Gaming" in the sidebar.

[Grinner]

The Giant has written a nice series of articles about worldbuilding. And the first article (http://www.giantitp.com/articles/YPgbz2j3PckGjjviJU5.html) might be especially of interest to you.
The articles can be found by clicking "Gaming" in the sidebar.

Those are actually what I was drawing from in the first paragraph of the OP, in addition to a couple other things (DMGs, gaming blogs, etc). Eldan had a better description of the process, I think, but my point still stands. So far, all of the advice I've seen begins with defining things, but I'm wondering if it wouldn't be better to start by defining events and extrapolating the physical aspects from there.

For example, let's say the town's water supply has been poisoned. From here, we can ask ourselves a few things: "What's the name of the town?", "Who poisoned the water?", and "Why did they poison the water?" I'd wait to answer the first question, but the second offers many tantalizing possibilities. The pack of bandits camped on the nearby mountain could have something to do with it, or perhaps someone is trying to exact some terrible revenge on the town. If it's the bandits, then the town probably has something valuable to them in its possession, and they've proven more than capable of defending it by force of arms. If the cause is revenge, then the town may have some dark secret in its past. Maybe they lynched somebody, and the ghost has come back to haunt them.

Also, I think I'll call the town Hatchettsville, after its founder, Tucker Hatchett.

So there you go. Where I normally have trouble generating these sort of things, I've just organically invented a remote town besieged by bandits and populated by martially-adept* people who wrongly killed somebody at some point in the near past.

Clearly, the best way to explore this method is to experiment further, but I'd like to know if anyone has any insight to offer.

*Or maybe they're not so good with weapons, but instead have a shaman looking out for them or something.

Edit: Also, the reason Tucker had the surname Hatchett was due to his phenomenal skill with the axe. He initially retreated to the mountains to live quietly, but a number of would-be heroes followed him to study beneath him. From there, merchants gathered to service the heroes, and some enterprising lass set up an inn. Pretty soon, a whole town had sprung up around Mr. Hatchett's abode.

Edit 2: Mr. Hatchett originally won his glory as a footsoldier for the king while fighting in the Killing Grounds. It's said he singlehandedly slew two hundred men in the defense of a wounded comrade. In reality it was twenty or so. Still an impressive figure, though.

I could go on to detail the king, his kingdom, the war, and so on and so forth, but I think I'll stop here.

Edit 3: Really truly lastly, the Killing Grounds are a blasted wasteland, the result of trench warfare, and they're analogous to the no-mans land of WW1. That might imply that guns are thing in this setting, though...

[Bacchanalian]

I've been building a world as I go. I posted about it a while back, but the basic progression. Note that the characters started at level 4 or 5 (I can't remember) because my wife is controlling all of them as we're playing a one on one campaign at home to acquaint her with tabletop RPGing.

Livestock is disappearing from farms around a local town and there's rumor of a rewards, so the party travels there to investigate. While staying in town, one of their party is kidnapped in the night and disappears. They find his trail by asking around, and follow it into the woods beyond the farms outside of town. They come upon a character (Duskblade, friendly to them) fighting with an evil Dragon Shaman. He serves a young green dragon who orchestrated the kidnapping to bring the party to him so he could ransom their friend for all of their magical items and gold.

From here as we played on I started thinking of why. Beyond the basic "dragons are greedy", I wanted a bigger and better hook. So I decided some bigger, nastier dragon was extorting magic items from the smaller, younger dragons of the region. Why beyond the greed? Because he wasn't just a dragon, he was a Great Wyrm Red that had turned himself into a Dracolich for fear of death, and in his drive for immortality decided that he could elevate himself to the level of a deity if he were to gather enough magic items in one place and perform some spell or ritual. So he's massing an army of followers who serve him, hoping to be elevated somehow or bestowed with divine favor when he becomes a god, and these followers have launched a massive campaign of tracking down magical items. They will eventually become contraband and extremely rare/expensive, with patrols of goblins led by ogres of various classes as the leaders. These will serve not only the head honcho dragon, but the regional younger subordinate dragons who in turn serve the red out of fear of him killing them or driving them out.

As she makes decisions ("hey, this dragon corpse has to be worth something, let's go to town and get a cart for it and ask around while we're there" insert small band of goblins and ogres trying to gather the corpse when they arrive back, insert local blacksmith directing them to a nearby town where he apprenticed for a man that can work the dragon bones into weapons, insert that town having just been raided days before by an army of goblins and ogres looting magic items but leaving other treasures behind oddly), I develop the story one step ahead of her.

Eventually I'm going to need to flesh out the world a lot more, but for the first 4 or 5 sessions I've winged it and it's resulted in an awesome hook that has me drooling. Once she figures out what's in store I think she'll be even more excited than she already is too.

[sktarq]

Honestly I start with a single question.
"What kind of stories to I want to tell in this world?"
Everything else comes from that. From there it is about picking moods, themes, and genres.
Then what system supports that easiest with notation that the to-be players may or may-not be familiar with it.
That gives you a skeleton to pick up ideas about plot hook ideas. What kind of plots do you want to play with. Then pick things that will drive lots of those kinds of plothook hooks. National structures or geographical ones both can do this.

[brian 333]

My first campaign began inside a dungeon. The only map I needed at the time was the dungeon map, and the only background information I needed was the basic premise proposed in the PHB, that adventurers were part of a semi-medieval society based on magic as opposed to technology.

I didn't draw the map for the starter town until several weeks into the quest. By then the players had completed the first dungeon, (it took them several tries,) met an innkeeper, a general merchant, and a smith, so the village needed those in it when I drew the map they needed for a counter-attack by the goblins they had angered by attacking a camp. (It was a random encounter.)

My first group came from tabletop wargaming, so we were into the tactical aspect of the game more than the social aspect. Our first ventures into the campaign hosted by our local gaming store lead us to explore politics, social interaction, and actual roleplay, as opposed to simply playing a set of skills and stats in a new kind of wargame.

Over time we ventured farther from home, and started new characters from other regions, and now my home-campaign world has grown from a single piece of graph paper with a dungeon map, (it is the Tomb of Lord Brytol, and within it is The Greensword of Brytol, relic of a noble house that died with the old lord.) It now fills three plastic storage bins with about five thousand hand-written pages including character sheets, maps, world-building details, records of adventures, and dungeons, dungeons, dungeons. (My guys never lost their roots in wargaming, and the tactical aspects of the game were always high on the list of important facets of the game for us.)

So: if your goal is to play a game with friends, focus only on what is needed for that campaign. Extra details constrain your group to a particular set of responses to any situation. For example, if your campaign begins in the Dungeon of Death near the village of Boot Hill, which lies in Necrotopia, the players can safely assume they will be dealing with undead pretty much their whole campaign. If the Dungeon of Death lies in an unspecified region with the town of Baker as their fallback position, then Baker could lie in Necrotopia, or it could lie in the County of DuLac, where the local Lord Lance plans to hold a joust for the entertainment of the masses, or it could lie in the Border Barony of Bellous, where ogres and giants have begun to raid, or it could...

In other words, more detail equals less flexibility on the part of the DM. This flexibility is especially important when beginning a new campaign with new players because it allows the DM to tailor the night's events with the tastes of the player group in mind. My group liked tactics, so our events were uniformly tactical in nature at first, and I was able to branch out from there, exploring each player's tastes over time. My first city, (the City of Stivic, Star of the North, so named because it was the only place on the player's hand-drawn map to that date which was drawn with a star for its location,) was begun for the sole purpose of finding a squalid frontier city which was suitable for the location of a thieves' guild, which one of my PC's belonged to and which was hoped to be a useful resource by that player.

The ability to tailor a campaign to the tastes of the players is critical with a new group of players. Old hands at RPG's know the tropes and are happy to explore convoluted storylines involving several competing factions, but this is simply too much information for a new player. He shouldn't feel obligated to do weeks worth of homework before he can feel like he's a contributing member of the group. He should jump in right away with enough information to do what needs to be done, but with nothing more. From there he will be drawn to discovery, and as you learn what questions he asks, you can fill in the details the player finds important.

Edit: The benefits to this system of campaign world generation are many, but one I overlooked that is most important: It allows the players to feel they had a hand in shaping the world, giving them a feeling of ownership you won't find in a stock campaign setting with all the details fleshed out from the beginning.

[Eldan]

I write campaign settings by starting with an idea. As an example, my little darling, Etherworld.

It started with a discussion on another forum. It was tangentially related to setting building. Someone asked a question along the lines of "If there's adventurers with teleport and invisibility and so on in the world, why are there even classic fortresses with stone walls?"
Someone else, a bit further down, replied: "Well, if you keep going down those lines, everyone would live in genesis'd demiplanes and only come out with astral projection."

So, I ran with that. A world where everyone lives in Genesis demiplane fortresses.

Of course, it changed a lot since that first idea. But I started by giving it a better reason "Mage war followed by global-scale demon invasion followed by exodus to demiplane fortresses", then started detailing some of those demiplanes.

Then, of course, I needed a reason for people to go adventuring. Which became "How about people go out into the Aether for trade and exploration?" But of course, the Aether is damn empty, so it became more an analogue of space or the open ocean instead of an exploration goal, so I added ships and everything went sort of "Age of Sail"-shaped, from there.

[Gogolski]

I like to involve players in the way I flesh out my world after I'm done with a rough draft that settles some regions or world features, the creatures running around on it and the mayor political/cultural interactions between these.

I'm just beginning to write up a world. I started with some features/regions where I think players can have an interesting adventure. These are mostly geological:

=>I wanted to re-use the kilometers long/wide creatures I created for a world that didn't see play after I created it. They have a big city on their back or inside them and they slowly wander around the world.
=>I wanted to have places where players could fall to their death, so I have some regions that deal with enormous heights:
* A tree (quite over the top tree)
* A cliff (quite over the top cliff)
* A weird balloony-surface thing high up in the clouds (weird)
=>I wanted to have water/underwater stuff, so I created a few floating city-states.

I don't have a map yet, but when I thought these up, it immediately became clear that economics hold most of them together or give them a reason for existing:

The cliff (and a bit of ocean) divides the world into parts where some resources are found on one side and lacking on the other and vice versa. The drifting city-states are the most important ports in the world and the large beasts carry/swim/burrow some stuff from one end to another.

I have some big outlines for the world, now I'm working on demographics and determining relationships among races (some o them newly invented, some yet to be invented, others as per source books)

When the rough draft is done, the player will come in. I always require backgrounds from players. New world means starting at Lv1. I can then tailor in some of the stuff my players care about and they will recognize it or at least feel very comfortable with how certain stuff plays out.

Players are important!

[the_david]

I like Bacchanalian's explanation of building a sandbox setting. You could go a step further by letting your players set it all up. Just tell your players to build that character they've always wanted and then you use the background to build a campaign around that.

My standard example would be Thar Thunderhammer, a dwarf who dualwields the biggest weapons he can find. (Originally for 4e he would wield 2 craghammers, in 3.5 he'd wield 2 dwarven waraxes, and I'd have to change his name to Bor Bliksembijl.) Thar was a dwarven mercanary who tried to raise enough money for the bride price of Lit Lightninghammer. As he returned from his failed quest, he found the dwarven stronghold in ruins. His clan and the lightninghammer clan where gone, and monsters wandered around in the dwarven mines. The 2 dwarven warhammer relics where gone too. Thar swore he'd find the clans, the hammers and that he'd destroy every last monster responsible for this atrocity.
So what we have in this example are 2 missing clans, an abandoned dwarven stronghold and some monsters that pose a threat. In this case, the GM can fill in the blancs for the monster, which would be Thar's first favored enemy. (Balrogs, bugbears, dragons, drow, duergar, goblins, hobgoblins, illithids, necromancers, orcs, etc.) At least 3 adventure locations are given. The dwarven stronghold that is overrun by monsters, and the unknown location of those monsters that got away with the clan's relics, as well as the unknown location of the 2 clans. Now this is a nice adventure on its own. Add 3 more characters into the mix with their own backgrounds and you've got an almost complete campaign.

[brian 333]

The general consensus here appears to be: let your players help flesh out the campaign world.

I suppose this wouldn't apply if you are creating the next Eberron for commercial sale, but if your goal is a fun game night with friends, this is my best recommendation for the way to go.

[S@tanicoaldo]

I start by the cosmology. Who or what created the universe? How it was created and why?

Who came after that? Who created the material planes? Why? How?

How the races where created? And magic? And the gods? And the other planes?

How where the first cities? The first civilization? The first wars?

And only after all that I do the map.

[the_david]

The general consensus here appears to be: let your players help flesh out the campaign world.

I suppose this wouldn't apply if you are creating the next Eberron for commercial sale, but if your goal is a fun game night with friends, this is my best recommendation for the way to go.Yes and no. You can either build your setting around the stories the players come up with, or you could come up with a story yourself and go from there. In both cases you are telling a story, but it's either the story your players want to hear, or the story that you want to tell. Those 2 methods seem to work best for me, as you'll only need to make up whatever the story needs.

[Gogolski]

I start by the cosmology. Who or what created the universe? How it was created and why?

Who came after that? Who created the material planes? Why? How?

How the races where created? And magic? And the gods? And the other planes?
...
In my previous post I was talking about my rough sketch for my world. The pantheon and the cosmology is what I'll wrap around that rough sketch after I get input from my player and their characters...

It appears that the economy drives quite a lot of what is found in my first draft. That is what I'll focus on right now.

I like it a lot that when you start writing up stuff, some aspects just run to the forefront and scream that they must be dealt with first.

It's just that these first ideas looked cool and immediately present you with some ripples and wrinkles that must be ironed out first before they continue to be cool. After that first ironing, you get so much more than some cool ideas. You have an entire workable idea. A smooth base to build your other stuff on.

(Sometimes it doesn't work out that way and you better break up the wrinkly stuff and put it on the backburner to be rehashed later on...) Kill you babies...

Everybody who has ever had a few shots a world building probably knows that there is no sure or true way to building a world. The fun is in the creative process. I have dabbled in creating boardgames and getting playable prototypes on the table, long ago I also played in a band and making a song work is quite a similar experience and I think it's similar for most creative processes: It's fun, very fun and very satisfying (if it gets somewhere), but it's also work in a certain way. ...and it teaches you to kill your your babies... (if it doesn't work) That might be the hardest part to master sometimes...

[brian 333]

I wouldn't say kill your babies, (I like babies, it's teenagers I can't stand,) I would say learn to be a ruthless editor. Think G.R.R. Martin ruthless. If something was a great idea at the time but is just a bad fit, or if it was ill conceived and never panned out, then cut it. Keep the good parts, build on them, and get rid of the rest.

[darklink_shadow]

I usually start with a single hook I want to have at the end of the day, and reserve engineer the history and place that would provide such a hook, and then look at the world I built and find all the other plot hooks that would innate exist within that setting. And then, usually, I abandon my original plot hook for the more interesting plot hooks I uncovered along the way.

But not always. I think a big think to think about is that if X happened, why did it happen? Because of Y. Well, why did Y happen? Because of Z. Etc, until you get to basic primitive things.

I also think it's important to pay very close attention to economy and religion, because all actions can find their roots in one of those two things.

[Yora]

Honestly I start with a single question.
"What kind of stories to I want to tell in this world?"
Everything else comes from that. From there it is about picking moods, themes, and genres.

I think that should always be the starting point. But not just unconsciously and implicit, but something you really do some conscious work on. As a second step, I would expand that question to "What kind of characters are supposed to be the protagonists/PCs for these stories?". And then you can start looking for elements that create a world that has an environment that can create such characters and supports the kind of activities they are doing. A group of scavangers in a rich and prosperous kingdom doesn't make sense, to make up an extreme example.

Most books, movies, and videogames do perfectly well without having a world map even though they can include quite substential worldbuilding, so you can easily see how little a role it usually plays for most adventure stories.

[CGDG]

The Giant's article is very helpful, I find. As you've proposed, I tend to work form the end. I generally start with what I believe will be the most important; it doesn't always stay the most important, however, but it allows you to build your foundation based on the themes and characteristics you want to be the most predominant in your world.

[DoomHat]

Here's how I do it.
Ask each of your players for one thing they want to see included. It could be anything, a creature, a technology, an genre, anything.
-Then, look at what you've got and think about the implications.

For instance, I remember one time I got; Super-villains, Space Opera, and Sparkly Vampires.

Sparklpires was a troll pick, but I can work with it. It's something the group could easily get into hating, so I' made them the primary villains of the setting. I took an evil bishonen stereotype angle. They're the shiny dark lords and of the evil yet fabulous empire that paints itself as a utopia.

The players will be branded as criminally insane rogues, that way if other players want to actually be heroes they can, or they can live up to the image the empire paints of them. Further more, all the player will have access to superhuman powers. Ta-da, supervillains!

Then, it takes place in outer space, so lets throw in some modified versions of things I think are cool from space stories. There's something equivalent to Babylon 5, struggling to remain neutral as the Shining Night Empire continues cracking down ever harder on dissidents.

Super-villains imply super-heroes like smoke implies fire. So the vampires are sending appropriately flamboyant super-powered goons at our players. Each planet has a team of magic girls or super sentai guarding it that the players must defeat in order to survive or possibly conquer if they're so inclined.


It was a ton of fun and each player was invested because they each got something they wanted, guaranteeing immediate investment.

[EisenKreutzer]

Don't start with maps. Start with ideas.

1. What makes this world different from other worlds? Why would anyone use this world instead of a published one?
2. What is the world about? What themes are important in its background?
3. What kind of campaign would I run in it? Genres? Focus? Time period?

Then think about politics. I think that's more important than geography.

Start with number three. Define a list of things players should be able to do. If you want them to explore wilderness, you need wilderness and things to find there. So far, so obvious. For political games, you need opposing factions, probably at least a handful of them. And so on.

This is awesome advice.