Structures and Procedures
The goal of Proceduralism should not be a totalizing system of Procedures and Mechanics to cover every possible fictional situation—as is sometimes presented in systems with universal mechanics, but instead be to add, clarify, and explain the Procedures for specific aspects of play that are either focal points for a system or require greater control by the rules. [All Dead Generations: PROCEDURALISM]
Posing the Question
Here’s a troublesome question about structures, as they relate to RPGs: are they the same thing as procedures? Allow me to illustrate the confusion with this quote from Marcia of Traverse Fantasy.
Rolls for searching a room or forcing open a door are operations or rules; the dungeon crawl is a structure or procedure of play wherein these situations take place. [Theoretical & Practical Proceduralism - emphasis mine]
As we can see, she seems to use structure and procedure as synonyms. Of course, that may be because she’s not using “structures” under the same definition as, say, the Alexandrian. Still, we can find that Justin’s definition of structures is indeed present in this quote, if we break it into its micro and macro sizes. What he calls GM rulings would fall in the first part, of rolling a Search check; scenario structures occupy the second half, where dungeon-crawls are spelled out, since it is established that a dungeon-crawl is a scenario structure. Even accounting for different terminology, then, scenario structures seem to be equated to procedures.
So, the question remains: are structures just procedures?
Procedures, like any term in our (fortunately!) fractured hobby, have a lot of different definitions. But for most definitions, structures and procedures do share a number of similarities. They both:
- Are ways in which the play conversation takes a certain shape;
- Are ways of deciding on the recurring application of rules and mechanics;
- Are useful tools for approaching the design of our systems and adventures;
- Are prevalent in our games, even if they’re often being used only subconsciously;
- Are generally invisible to most groups using them.
It is looking a lot like scenario structures and procedures are the same thing. But then, a lot of definitions of procedures have a particular feature that I think helpfully sets them apart from what we’ve been trying to get at with structures.
Marcia says this about procedures:
A procedure is the diachronic (i.e. sequential, as opposed to in-place), ordering logic of interactions between agents and the other system elements. It is like a subroutine in a program, and often has the structure of a loop. It is diachronic because it is an order of operations, where individual operations can be prescribed specifically or can be slotted into a space. [Theoretical & Practical Proceduralism]
Mr. Prismatic Wasteland agrees, saying:
Procedures are a type of rule that provides the order of operation (which order typically repeats) for a game and works to structure play. [What Even Is a “Procedure”? — Prismatic Wasteland]
This focus on order of operations is, I think, the point with which we can make a meaningful distinction between scenario structures and procedures.
Structures vs. Procedures
Scenario structures don’t care about the order of operations themselves. Procedures, conversely, care about this kind of order a lot, because they’re invested in when to apply other mechanics or rules (and possibly even other procedures). Structures only really bother with order to the extent they have incorporated a procedure within themselves.
The structure of walking into a dungeon room and being presented with a pair of gerblins, a mysterious floating orb and a shelf full of books doesn’t care what order these things are presented to the players, only that they exist to be presented and are presented. There might exist practical advice on presenting one thing or the other first, but the scenario structure doesn’t inherently bother with that. It doesn’t even care if you’ve actually entered the room before interacting with the stuff in it (since you might hear the gerblins from the other side of the door, for instance).
What the structure cares about is that you have a room with stuff to present to the players. If they open a dungeon room and find only a metaphorical “Under Construction: your GM will have content for you here shortly!” sign and you’re left floundering mid-game, the dungeon-crawling structure looks at you disappointedly and says “I told you they would be entering rooms, why don’t you have the rooms with the stuff in them?”. It cares about knowing what players will be doing and guiding the GM towards having the right kind of established information for it.
Meanwhile, when using a dungeon exploration procedure that states that the order is “1. enter room; 2. interact with stuff in the room; 3. pick an exit”, if you come across that same “Under Construction” sign, the procedure is just checking its watch waiting for you to be done with 2 so it can get to 3. It never asked you to have a room with stuff in it. The existence of rooms is implied in the procedure, sure, but it doesn’t know anything about what these rooms should be like, or what should be in them.
Procedures and Conversational Turns
The only order game structures might care about, inherently, is the order of “establishment of fictional situation > characters’ intended actions in response to fictional situation > resolution of characters’ intent, changing the fictional situation”. This is a well-defined order, but it is not even properly enforced by the structures, either. They know this order exists because they want to help you fill every step of this order in a satisfying manner. But if you ask “can we go back to establishing something about the fiction after a player started declaring a character’s intent?”, for instance, the structures have no answer.
A procedure dealing with “fiction > intent > resolution” conversational order, on the other hand, wants a clear answer for that same question. It might say “you can”, or “you can’t”, or “yes, if…” or “no, unless…” – the answer itself can be a lot of things, but the procedure wants to have an answer. It doesn’t want people going back and forth on its order willy-nilly, it has an opinion on when they should move forward or backward. But it doesn’t, by itself, have an opinion on what fills each step of this sequence.
Procedures also often care about taking turns. They want to know when one or more players will next come to a decision point, because that’s when they’re likely to trigger some mechanic or a new procedure. That is also why so many scenario structures have procedures embedded in them: it is a quick way of ensuring you are giving these decision points to players. And when I say players here, that includes the GM: a procedure can also make sure that they have the opportunity to make the fiction react to the choices of the other players, giving them their own "turn" - think of how, say, an overloaded encounter die neatly bundles the dungeon's responses to the PCs exploration turn.
But if a scenario structure hasn’t adopted a procedure, it is fine without clear turns. A player can hog the spotlight in a turnless dungeon-crawl, proposing and resolving 100 actions before anyone else gets a go, and the dungeon-crawl structure will just shrug: it is satisfied that a player is engaged with the dungeon. It becomes the GMs sole responsibility to manage the spotlight. A procedure is more likely to step in with the GM to say “Stop, you’ve had your go, now it’s this other player’s turn”.
Structures and Fiction
Procedures are fairly blind to fiction. They themselves don’t know if their steps are creating something meaningful and coherent for the current fictional situation. Their output only makes sense for the fiction if a designer has meticulously curated and placed the procedure to fit a fictional time and place (probably by having a clear structure in place), or if a GM interprets their output and creatively frames them until they fit the current time and place of the fiction. The procedure for random encounters can be perfectly happy plopping down a 20 ft. high giant in a 5 ft. tall corridor. If you followed the proper steps to achieve that result, then it believes there should be a giant there. You can decide that this doesn’t make sense and change the output somehow so the fiction remains coherent, but that’s your work, not the procedure’s.
At most, a procedure can be “taught” to care about certain elements of the fiction (e.g. “roll again if the generated encounter doesn’t fit the space the PCs are currently in”), or it can expressly disclose decisions to the humans at the table (e.g. “if the result doesn’t make sense, do whatever seems most reasonable”). But it won’t care about the fiction on its own. Not so with scenario structures. In fact, it is typically a scenario structure that ties a procedure to some type of fictional context, that figures out why and how the procedure applies to the situation.
Structures don’t know everything about the fiction, but they know very important bits. A classic dungeon-crawl structure knows there exists a series of rooms connected to each other, and that these rooms need interactive stuff in them. The node-based investigation structure knows that a fictional mystery must have left behind clues, and that these clues point to different places and people that may have more clues.
They can still get a lot of things wrong about the fiction (particularly when they're not well suited for it), but they know about the overall shape of the elements that will go into the fiction. They know about these elements because these form the base of information players need in order to make appropriate choices, and they are the aspects of the fiction most likely to be changed by those choices.
Conclusion
So, structures are all about the what and the how of the decisions players (GM included) will make – procedures are much more about the when.
A structure may well be largely defined by one big procedure – that is the case with Whelan’s Flux Space or Blade in the Dark’s Downtime – so it is natural that sometimes the boundaries between them get blurry. But we can still say that in Blade’s Downtime, the bit that cares that the entanglements and the downtime activities happen in that order before the next score is the procedure. The structure is the bit that cares that the entanglements create tough choices for the players, and that those choices will feed into the choices of what downtime activities to pursue and which score to attempt next.
In summation: the way I see it, it's useful to distinguish between these two similar concepts. Procedures are about creating an order of operations for a game conversation to follow; structures are about the content and form of the conversation. Because structures care about how the content of the conversation should be presented, they can (and often do) have procedures as an integral part of themselves.
Procedures and structures have some similar goals, and as a result they frequently work together. But I’ve argued here, hopefully with some success, that we can draw a significant distinction between them.