The Supremacy of Context

In OSR circles people talk about “the Hickman Revolution” as a harbinger of TSR shifting away from the old school ethos in favour of producing a different style of gaming product – the tightly railroaded Dragonlance modules being the archetypal example of the sort of work that resulted.

Tracy and Laura Hickman self-published their early modules before TSR picked them up, and Tavis Allison of The Mule Abides actually found an original version of Pharoah in which the four guiding principles of their design work were enunciated as follows:

1. A player objective more worthwhile than simply pillaging and killing.
2. An intriguing story that is intricately woven into the play itself.
3. Dungeons with some sort of architectural sense.
4. An attainable and honourable end within one or two sessions of playing time.

The comments thread on David McGrogan’s post on the subject is worth a read because Tracy Hickman actually shows up in the comments to give his thoughts – namely, that he agrees that the four principles there are guiding principles of his and Laura’s work, and that Pharoah didn’t actually succeed very well because it’s basically just another dungeon crawl.

The four principles did, however, seem to meet a need in the market, one perhaps created by the influx of people into the hobby who came from a fantasy fiction fan background instead of having a grounding in wargaming. They certainly appear to have been guiding lights of the post-Gygax TSR – see, for instance, the TSR Code of Ethics, which seems to be so massively compatible with points 1 and 4 that it could have been specifically written with them in mind. Even TSR seemed to buy into the new orthodoxy of the mid-1980s and 1990s that dungeon crawls with no greater motivation on the part of the PCs than profit were, if not actually bad and wrong, then at least juvenile and unworthy.

However, the Hickmans hit the scene at a time when a lot of people were also bringing new ideas to the table, or shifting their own ideas accordingly. Gary Gygax himself was developing ideas about what James Maliszewski referred to as Gygaxian Naturalism, and in the early 1980s finally came around to putting out a World of Greyhawk boxed set describing his campaign setting – a type of product that, in the 1970s, he couldn’t see a need for. Just as the Hickmans were delivering a fat dose of story to the heart of D&D, others were intent on increasingly detailed worldbuilding and verisimilitude.

If you buy into one of the various threefold models of RPG theory, you could analyse this in very different ways. As per the Forge, this was an era when Simulationism of world or genre was king, with Gamism sidelined and Narrativism waiting the coming of Great King Ron to awaken it from its slumbers. Those whose definition of “story” is less straitjacketed than the Forge’s may prefer the rec.games.frp.advocacy threefold, which would interpret this as a time of divergence, with world-focused Simulationism and story-focused Dramatism forging different paths out of the Gamist motherland.

I see it a bit differently, though. What Gygax was doing with his naturalism and what the Hickmans were doing with their stories were both different approaches to tackling the same problem, a problem which White Wolf’s house style and the Forge’s brace of Narrativist story games would also confront: namely, the problem of providing a novel and different way of handling context in a tabletop RPG.

For the purposes of this article, I am defining “context” as being “information which prevents the fiction or some element of the fiction from being arbitrary”, where the “fiction” is defined along Dungeon World-inspired lines as “the imaginative construct that is the focus of play in RPGs and story games”.

Fiasco offers a good example of the impact of context. Based purely on game mechaics, each game of Fiasco unfolds in more or less the same fashion – people collaboratively jot down some notes, improv some scenes, and pass counters around. If you jotted down a transcript of a Fiasco game that focused solely on game mechanical processes and paid no attention to what people were actually saying, you wouldn’t see any difference between individual games, but actually the differences are enormous because Fiasco sessions distinguish themselves by the context you hang off the game mechanics rather than the counter-passing itself.

That doesn’t mean that game mechanics are divorced from context, mind (though people did tend to forget that back in the 1990s). For instance, in OD&D‘s core booklets all non-magical melee weapons are basically the same and do the same damage, so it makes no real difference whether you are using a rapier or a longsword. Subsequent revisions introduced in supplements brought in finer distinctions like variable weapon damage, which added context: suddenly, some weapons were more dangerous than others even before magic was brought into the picture.

But isn’t everything in an RPG contextualised to some degree? Not necessarily. Inspiration behind Ravenloft, one of the modules which really put the Hickmans on the map, was – as I’ve described previously – a dungeon adventure where Tracy Hickman and his fellow adventurers encountered a vampire, and Tract was thrown off enough that this potent creature seemed to have no backstory and purpose in life that when griping about it to Laura later they cooked up the idea for Ravenloft. From there the original Ravenloft module’s focus on Strahd’s backstory, and the later campaign setting’s emphasis on well-developed NPCs, were a natural reaction.

One of the very useful things that the OSR has done is to enunciate the “gonzo” style of gaming, where you don’t worry about genre boundaries and just throw in stuff you think is cool and it’s more important to be offbeat and interesting than internally consistent. Of course, you can work hard on giving gonzo additions to your setting a well-rounded context, but you don’t have to. And even away from the gonzo style, there’s a fair amount of old school play (particularly the sort of thing that rec.games.frp.advocacy or the Forge would consider Gamist) where context matters less than interesting challenges. That treasure chest in the necromancer’s tomb which hasn’t been opened for a century and yet somehow has a live viper inside ready to attack anyone who tries to open it is a prime example of an instance where context has gone out of the window, and how much that sort of thing bugs you is a good measure of how important context in gaming is to you.

I would say, in fact, that more or less every major experiment or novel movement in RPG or storygame design has involved on some level a new look at context. The “Gygaxian naturalism” and extremely detailed campaign worlds of the 1980s sought to provide context through creating a strong sense of place so that players felt that they were exploring a living, breathing world. Dragonlance and other Hickman-inspired materials tried to create context via story, with the backplot framing the action and the plot of the linear adventures presented feeling more meaningful and rewarding to those who enjoyed that sort of thing as a result. White Wolf’s runaway success can be attributed in part to the way they adeptly used splats to create a social context for World of Darkness games. With Forgey-type narrativist stuff, the authorial power and narrative authority sharing comes down to sharing the responsibility of contextualising things. More recently, we have seen movements in gaming which question the very primacy of context; a lot of what people complained about as being “dissociated mechanics” in 4E D&D constituted game mechanics which didn’t pay much heed to context (just exactly how do you “trip” a gelatinous cube?), whilst the OSR’s bid to resurrect neglected playstyles includes an openness to styles where context is less important.

To get egregiously Ron Edwardsy for a moment, you can even break down different ways of applying context to a game in a way similar to his Story Before/Story Now/Story After. Context Before would be where you figure out the context of in-character elements before play – as you would when writing a campaign setting or preparing a linear adventure. Context Now would be when you figure out the context of things during play itself – the setup phase of Fiasco being a prime example of this. Context After would be where you have something happen by random chance or on a whim and then come up with a context after the fact – for instance, you roll up a wandering monster for the PCs to encounter, and then you figure out what that bugbear was doing in the Snow Prince’s dungeons later. (Or you don’t bother at all, in which case it’s Context Never.)

I would say that most games will include a mixture of different approaches to context in actual play, and would go so far as to say that purism here is not wanted. Pure, undiluted Context Before would require you to contextualised everything in advance of play, and constrain you from including anything you haven’t figured out a niche for entirely. Pure Context After would be a calvacade of improvised randomness that would be extremely difficult for players to meaningfully interact with because nobody would know the significance of anything until after the fact. Pure Context Now would force the referee or group as a whole to do all their worldbuilding on the fly, with all the inconsistencies that would inevitably introduced by doing all that in one go without a chance to take a wide view and see how it all fits together.

That said, some games will focus more on one type of contextualisation than others. Your classic White Wolfy plot-heavy high-prep chronicle is going to mostly involve Context Before, for instance. And Fiasco kicks off with a bit of Context Before (selecting the prewritten playset for the game or writing your own), then has a segment mostly focusing on Context Now (the setup phase), then shifts back to a play mode that mostly leans on Context Before again (since the scenes you play out will generally look to the playset and your setup for inspiration).

Perhaps one of the reasons some Forge games feel much more traditional than they are sometimes made out to be comes to the way they handle context. Dogs In the Vineyard, for instance, is a resolutely Context Before sort of game which provides a strong prewritten setting and encourages the referee to cook up a scenario beforehand and then unleash the players on it. Equally, some Forge games fall flat for me in part because they don’t pay enough attention to context to feel satisfying – for instance, Donjon has a game mechanic where if you succeed at a skill roll you get to make up what happens, to the extent that if you search for a secret door and succeed you get to declare that a secret door exists, regardless of whether there was one before or whether it makes any sense for it to be there, and there’s no game mechanical requirement for players inventing such details to properly contextualise them. At extremes, you have Forge games like My Life With Master which people claim provide Story Now, but are so laden down with designer-tailored context that the story you get out of them will more or less always be the same sort of story – and whilst that is a deliberate part of the design, I don’t see an awful lot of point in Story Now if you know beforehand what general direction the story is going to take because of the constraints you have channeled it down.

No RPG comes entirely without context, though an extremely setting-agnostic generic RPG could come close. But it seems to me that a drive to add more context to games has existed ever since the publication of D&D – after all, Empire of the Petal Throne is basically an OD&D hack with a substantially more developed setting. If you really dig verisimilitude and exploring a rich campaign world, context is obviously important, and that is what published settings seek to provide. If you prize realism, lapses in context will bug you, and systems that went to great lengths to provide realistic outcomes were thus a trend in the 1980s and still crop up today. If you are all about telling a story, regardless of whether that boils down to narrow Forgey Narrativism or some broader meaning, your story will make no sense without context.

And, of course, the Hickman Revolution was built on context. For PCs to have goals that feel more important than the simple amassing of riches and power, those motivations need to be properly contextualised, otherwise they’ll be non sequiturs. Having a strong story integrated into play means that the narrative provides a context for play; making a dungeon that makes architectural sense requires contextualising the various parts of the dungeon and restraining yourself from designing bits purely for the sake of throwing a cool challenge at the PC without answering the question of why those parts of the dungeon were built in the first place. And the honourable or otherwise conclusion of an adventure cannot be assessed as such if shorn of context.

With so many different tastes and motivations for gaming hinging on amping up context, it may be no surprise that more freewheeling, context-agnostic styles of play ended up neglected until the OSR came along. In addition, the provision of context helps to keep game designers busy and allows you to provide support for a game line – and in particular, support which exists within the existing system framework, rather than extending it in a potentially unsustainable manner. One criticism I’ve often heard about prestige classes in D&D 3.X is that whilst they were originally presented as having very specific contextual requirements (like being a member of a particular secret society), too often both referees and designers were sloppy about that – likewise, the worst excesses of the 3.X character optimisation quagmire are unleashed when people mash character options together without consideration of whether any of the weird choices they are making make any IC sense.

From the mid-1980s to mid-1990s, TSR seemed to base its D&D releases more around the development of context than anything else. For instance, the 2E Legends & Lore supplement got deeper into the cultural context of the pantheons it presented than Deities & Demigods ever did, and the 2E Book of Artifacts doesn’t really change the game mechanical underpinnings of Artifacts but greatly expanded on the terse backgrounds they had been given in 1E. The entire 2E Monstrous Compendium format is an exercise in contextualising the monsters in question, and the major emphasis on campaign settings rather speaks for itself. Even kits, arguably the only really significant addition to the 2E rules in its early supplement line that didn’t have some precedent in 1E, seem to be rooted in providing a bit more background contexts for PCs than the vanilla character classes do. It wouldn’t be until the Player’s Option books emerged that TSR seemed to be trying to actually experiment with the D&D system again in a context-light system-for-system’s-sake sort of way.

Using the idea of context also helps me understand why so many Forge games fall flat to me. Part of it is that whilst many Forge games are good at sharing out narrative responsibilities, they can overlook the provision of context as one of those responsibilities, and you end up with stories where sometimes stuff happens because a participant thought it would be cool at the time, rather than because it was contextually appropriate for it to happen, and quite often something which does not feel contextually appropriate will also not feel dramatically appropriate, which kind of undermines the narrativist story agenda. In addition, Forge games often use gimmicky (and usually rather dissociated) dice mechanics which can sometimes be a bit blind to context. Whereas task resolution is firmly contextually rooted, conflict resolution procedures which pay attention to context can be a bit more variable. For instance, the way escalation from talking to fisticuffs to shooting works in Dogs In the Vineyard seems to implicitly assume that this is an appropriate progression of escalating conflict in all contexts, which is a big fat dose of (citation needed).

In that light it is probably no surprise that the most successful indie games have been stuff based on FATE, wherein your Aspects are intrinsically contextual in the way they work game mechanically, and Apocalypse World, where the idea that you look to the fiction first and then apply the rules in a way that makes sense rather than looking to the rules first and generate the fiction from them is a fundamental axiom of the game.

Of course, context is not an automatic ticket to successful product design. Annoyingly linear adventures which don’t provide any scope for unexpected outcomes or player creativity are often rigidly rooted in a firmly contextualised setting and backstory. Vornheim by Zak Sabbath is a great example of an OSR supplement based primarily on various random tables and charts that leave you as referee to work out how the results you get out of it fit together, but still manages to provide an evocative tone and atmosphere despite being low on context.

Still, it strikes me that if you see the work of a game designer as providing prepackaged context for people’s home games – especially when it comes down to writing supplementary material – that could help you construct products with broad appeal. In particular, asking yourself the question “do I need to do more to contextualise this stuff, or can I stop there?” may be a good technique to avoid overwriting. Overwrite your story and you end up with an excessively linear adventure or metaplot, but provide just enough details so that the referee knows what is going on and why and you can trust them to take it on there. Overwrite your setting material and you can end up burying your customer in minutiae, but provide just enough detail that the referee understands how the significant parts of the locale operate in broad terms and you’re good. Overwrite an NPC description and you end up with a character that becomes a burden to read up on and tricky to portray “properly” – give just enough to let the referee know who they are, where they’re coming from, what they do and why they do it and that’s all you need to give them.

You can accuse the Hickmans of overwriting their material – especially when it comes to the Dragonlance modules – but you can’t say they didn’t hit on something people didn’t want.