Quicksand Sandbox: What are we going to do tonight, Brain?

One thing I am constantly struggling with as a GM is making up my mind what kind of game mode I actually want my campaigns to run in. The linear plotted adventure went out the window years ago, but since then it’s ben an endless back and forth between enthusiasm and disdain for sandboxes and dungeoncrawls, social games and exploration games, ongoing campaigns and episodic one-shots. Which I think ultimately comes down to a  disconnect between the kind of narratives I am dreaming of and the realities of running a game with other people. I want Conan, Star Wars, and Indiana Jones, but these are all tales written by a creator who controls the thoughts and actions of all the characters and has full control over the past, present, and future of every scene at the same time. You can not replicate a book or a movie exactly in an RPG. You can only work towards running a game that will look like just as great a story in hindsight.

Of the many possible open-ended game modes to chose from, the two I know I am not interested in are hexcrawls and megadungeons. Which happen to be by far the most popular, or at the very least the ones that have most been written about. But the discourse about these two modes has led to the articulation of a valuable and important concept: Default Goals and Default Actions.

In a game that has a group come together at regular or irregular intervals and ask the immortal question “What are we going to do tonight, Brain?”, there needs to be a default goal. If nobody has any special plan, then the whole group should be in agreement to default to one standard activity. Otherwise they just keep akwardly sitting around in confusion and are likely to start wrecking things to get any response to their presence from the game world. In the case of the hexcrawl and the megadungeon, these default actions are exploring new hexes and going to the dungeon respectively. Or simply “explore”. But as Matt Colville pointed out quite correctly I believe, “explore” is not a good goal. Exploration is walking around blindly and waiting for something interesting to fall into your path. It’s still waiting for the world and the GM to give them a task to deal with. Without understanding why, I think this is really the issue that always had me feel very uncomfortable about the thought of running a hexcrawl or megadungeon. It just doesn’t seem to have the potential for the kind of narratives I want my games to produce.

My work on the Ancient Lands setting began as an attempt to create a more realistic portrayal of tribal societies in a fantasy world that wasn’t as distorted as the nonsense you get about life in “warrior cultures” in movies and books. But while I think that it’s a fascinating subject and some elements of this will greatly help me making the Ancient Lands feel like an actual world, I have come to really appreciate minimizing exposition and player buy-in. Instead relying strongly on familiar archetypes to allow players to correctly guess what is what in this fictional world. However as part of it, I found one solution to letting PCs go on adventures and being indispensible for the survival of the clan, which was to define PCs as hunters for magical artifacts that can help defend the clan against their enemies and hostile spirits. Somehow this got stuck in my mind as the paradigm that all adventures in the Ancient Lands have to be treasure hunts and that all PCs have to be treasure hunters. After all, treasure hunting is what makes characters progress in B/X, so it seems to be a perfect match, right?

But in hindsight I really just handicapped myself with this approach. Without the addition of “return it to your clan to defend it against attacks”, the default goal of “find treasure” is just as hollow as “explore”. But while reading Kevin Crawford’s excellent Spears of the Dawn, I finally came to the realization that default goal does not have to be the only goal. It’s the goal that you can always go pursuing if you don’t have anything else planned right now. I believe the key to a successful sandbox campaign is to make it a hybrid campaign of exploration/treasure hunt and player-initiated story adventures. On their own, neither can stand by itself. At least not in a way that I want to run it. Pure exploration is aimless. And sending the players to do whatever they want in a world they know nothing about is a recipe for getting them stuck in the quicksand of unlimited options.

A much more appealing approach to sandboxes is “come for the plunder, stay for the people”. The treasure hunt is a device to get players to start interacting with the world in an easy to grasp and straightforward way so that they get opportunities to form connections with the setting and the NPCs and get dragged into local conflicts. The platonic ideal for player initiated adventures always seems to me best represented by the classic Kurosawa movie Yojimbo. There is no quest giver and barely even a hook to get the hero into this adventure. He is just passing through a village when he sees that local gangs are making trouble. Even though the locals tell him to just be on his way, he is intrigued and stays to see what happens next. At this point he has no plan and not even a clear goal and completely plays it by ear, but once he has established some connections to the village it very quickly grows into a complex web of cunning deception and daring swashbuckling that simply is a blast to behold. This is what I believe player-initiated adventures in a sandbox should be like.

But in an RPG, walking down the street until the players run into something that grabs their curiosity is not feasible. If they currently have nothing to do, they need a default goal to fall back on that will keep them entertained until they find something more interesting to do instead. Putting the limitation on character creation that all PCs have to have a drive to look for magical wonders in ancient places seems like a perfect solution for a wilderness sandbox.