The Sandbox Revisited

While the notion of “sandbox” style play has become widely accepted as one of the most salient aspects of old school gaming, I must confess I’m not entirely comfortable with the way the term is often used. Before I go further, however, I want to make it clear that, generally speaking, I agree with the larger set of ideas that the term “sandbox” typically designates. That said, I believe that there are several conceptual problems inherent in the idea of sandbox play as it is conventionally understood.

Let us begin with a definition. As I understand it, “sandbox” play refers to a style of gaming in which events are not pre-scripted or pre-determined. Players are free, and even encouraged, to interact with the game world as if it were a real world, to be able go where they wish and do what they want in any given situation. As such, a sandbox campaign is often contrasted with the so-called “story-driven” campaign, a campaign in which characters are expected to follow a predetermined sequence of events leading up to some final, predetermined ending or climax. It is in this contradistinction to “story” that my problem with the idea of sandbox play lies.

The problem is that “story” is seldom clearly defined in this schema. I think too often many DMs feel that, in order to run an old school (sandbox) style game, they must completely eschew anything that even remotely smacks of a narrative arc or sequential telos in their campaigns. Any element that suggests that the DM may have anticipated or even planned where the action might go is to be rejected or, at least, regarded with real skepticism. All too often, this leads to a situation in which the players just kind of wander around in the game world, looking for things to do, while the DM hesitates to give them something to do, because “giving” implies a purpose, and purpose implies that some future outcome is anticipated, and in this we have the beginnings of a story. In sandbox play, the story must emerge retrospectively; it can’t be planned in advance. But in my experience (I started gaming in 1978) D and D adventures even those from, back in the day, were often very mission-oriented. They assumed that the PCs would be likely to follow some sort of anticipated pathway as part of the adventure. After all, does not a mission imply a potential story?

Consider for example the case of Gary Gygax’s classic D series of modules for TSR. These are often held up as quintessential example of old school, golden age style game play and adventuring (a claim that is in my opinion entirely accurate by the way). But these modules, D1-3, most definitely do have an implied story arc to them. The modules presume that the PCs, having defeated the Giants on the surface, are planning to on track down the ultimate source of the threat, the drow, in order to learn about them and perhaps visit them with some measure of retribution. My point is that a strong sense of narrative is suggested at the very outset of the adventure. Gygax, in other words, did not just give us an environment filled with lots of cool stuff and say “OK just let the PCs wander around in this underworld and make their own adventure.” No, the whole thing begins with a very a very clearly delineated sense of purpose and direction, a mission. My guess is that most people who played this module pretty much followed this implied narrative trajectory, at least the first time through.

But here’s the rub. What makes these modules old school is not the presence or absence of story elements in and of themselves, but the fact that the PCs don’t have to follow the story if they don’t want to. The module allows for the possibility that the PCs might just decide to wander around and explore, or that they might interact with the story in any way they see fit. Heck, they might even decide to try and cut a deal with the drow. But this is not to say that the modules don’t have strong story elements built into them. The conceptual problem with most definitions of sandbox play is that, we typically have to completely reject the idea of story in order to define what we mean by “sandbox.” I don’t think there is anything wrong having strong story elements in a game as long as the story is not pre-scripted and the players are not forced to do anything they don’t want to do. The idea of sandbox play too often is taken to mean that the players must drive all of the action and that the DM should react entirely to them, having as little influence on the sequence of events that comprise the game as possible. I don’t think this approach necessarily leads to better gaming, nor do I think, as I hope my earlier example illustrates, it reflects the style of play characteristic of D and D’s golden age. Moreover, this conception of sandbox gaming is apt to impose a prescriptive limitation on the DM’s imagination.

It makes more sense to me to define the problem, not in terms of the type of world that the characters inhabit (ie. sandbox), but from the characters’ perspective itself. Several years ago, the writer of the Burning Void (now Errant Dreams) Blog offered up an excellent series of essays on the concept of “free will” in role playing games as opposed to the idea of “railroading.” I believe that, even though she is not speaking specifically of old school gaming, her essays offer a more fruitful and useful way of conceiving this style of play than that suggested by the “sandbox.” According to her, what really matters is whether or not the players are free to choose their own responses to what the DM gives them. In this schema, the DM doesn’t have to worry about whether he or she is including too much narrative, actual or implied, in the game. The question becomes are the characters free to do as they see fit and is the DM prepared to go in different directions (OK, that’s two questions). By the same token, DMs who don’t want to include any implied story elements in the game are free to not do so.

Reconceived in this way, the idea of the sandbox feels less prescriptive. It’s less about “though shall not have story” and more about “hey, do whatever you want, just make sure your players can too.”