Buckle up. This is a weird one.

One thing that makes it hard to talk about “play” as a unified thing is that it’s not a unified thing. There are a host of different activities that fall under the umbrella, and some of them are radically different enough to defy generalization. One of these vectors is the level of engagement with the system.

Ok, roll with me a moment: the biggest difference between a game and a story is uncertainy. Excepting very edge cases, the story might have uncertainty within its fiction, but that’s a sham – it will be the same story each time you read it. On the other hand, a game can be expected to be different each time it’s played. If it’s not, then it’s a script, not a game.

Now, given that, where does this uncertainty come from? The obvious sources seem to be “dice” and “other people”, and that’s true enough, but I might go a step further and say that the difference is found in decisions, which are made either by people (who are highly variable) or by systems (which include an element of randomness to introduce variability). This is a very simple endpoint that glosses over the many, many ways that variability can be introduced, used and managed, but the heart of it is decisions.

However, a game is not just a collection of decisions. There is something that makes those decisions (individually or collectively) enjoyable to a player, otherwise they would not be playing.

There’s no one good answer to this. I could say something like “engagement”, “investment” or “stakes”, but that would just be a tautology – for players to care about decisions they need to care about decisions. Not super helpful in and of itself. From a certain perspective, one might even argue that a primary act of play is the creation of that investment in decision.

But knowing that there is something is valuable, since it then lets you look at specific games and specific people and ask why they care about that decision, and it’s often possible to find that out. There are a host of well known motivations – Challenge, fiero, empathy, drama – stuff like that, and there are ways to drive towards those if you have reason to think they’re the desired values.

Ok, so given all that, we have three interesting data points:

1) Games are made of uncertainty resolved by decisions.

2) Those decisions can be implemented in many ways

3) Those decisions can be valued in many ways.

Now, why is this interesting? Because if we accept the premise that making decisions is the defining activity of the game (and it’s cool if you don’t, but I’ll be riding this train for a bit) then it shows us where the nails that connect system to play are (or should be). That is: your play is full of decisions, how does you game’s design interact with that?

If that seems abstract, then consider the slightly more concrete question of “What decisions are being made in play without the game system, and is that gap a problem?” Consider the classic example of “we played all night and never touched the dice once!” This is one of those odd contradictions of game design because the play experience was great, but clearly there is some sort of disconnect because (implicitly) this was an experience the rules as written couldn’t provide. There’s an instinct to consider this as a flag for a bad design, but that seems excessive. It might indicate a mismatch – they might be trying to drive a hammer with with a screwdriver – but that is a different sort of problem.

But where this gets useful is that during that “no dice” session, decisions were still being made. They were just being made by a human rather than a system. This has risks, but in this case it turned out pretty well, and the question becomes whether there’s anything we can learn to replicate the success (and, ideally, fold those learnings into the system in order to automate it).

This is fine as far as it goes, but what’s important to note is that the reality is that while the dice may not have been engaged, this does not mean that the system was not. This is because while the system may offer the means of making decisions, it is also the container for many of the things which influence decisions.

Again, that’s pretty abstract, so let me make it concrete. If you are playing a very skilled thief and you come to a locked door, this is a moment of uncertainty – will you be able to get past the door? (The game may ask a different question, but let’s stick with this for now). The default system almost certainly has some means whereby I could roll some dice, compare it to something, and get a yes or no answers. That is one way to make the decision. Alternately, the GM may simply think “You are a very skilled thief, of course you can pick it, you succeed and move onto the more interesting thing.” This is also a way to make a decision.

But wait, you might think: Have we just stripped the situation of uncertainty? If nothing else, we have stripped it of uncertainty for the GM.

This is, I think, a really interesting philosophical question with curious implications. If you have a sense that uncertainty needs to be fair or evenly distributed in some way then having unevenly distributed uncertainty is a no go. If you shrug and note that it’s still uncertain for the players, then it’s no big deal. But this is such an essential difference that it has really profound impact in design and play.

However, it’s also a bit of a sidebar. The interesting thing in the thief example is that the system contains the information that your character is an accomplished thief in some way. It might be a high skill rank or many levels or whatever, but something in the game reflects and communicates that idea. That informed the GMs decision – should the fighter have come to the same decision, they would not have gotten a pass – so the system is still engaged, even if it’s not engaged in the ‘right’ way.

The result of this long, twisty route for me is to really look at this idea of there being different ways to engage the system in play and consider it as a feature rather than a bug. The premise of this would be that just as there are different kinds of investment and different kinds of decisions, it may be desirable that there be different means of engagement for the system.

This is hardly an unprecedented idea. Turn it a little bit in one direction and it’s the same argument for games having different subsystems for different types of action. But most existing examples proceed on a single axis: granularity. That is, they may have a simple procedure for doing one thing, but a more complicated one for doing something more specific. For example, a game might have a general resolution mechanic for most action, but a more detailed system for combat.

But the possibility which intrigues me is that this could be done on drastically different axes. Go back to that list of things that invest people in decisions and you could have different tools for engagement that support different modes with the expectation that they could change up on the fly.

(As an aside, I think the ability to implicitly do this is one of the appealing elements of very light, very interpretive games. 3d6 in Risus is a very concrete things, but the number of ways to use and interpret it is vast. This is harder as the system picks up more parts).

This is all crazily abstract, so let me make it a bit more concrete with a Fate Hack.

When engaging with Fate in a normal way, these are the things that aspects do at my table:

• At a cost, it might grant a bonus to a roll
• At a cost, it might be able to turn an opportunity into a resource or other outcome
• At a cost, it might remove an impediment
• At no cost, it might create or impede an opportunity
• For a payment, it might impose a penalty
• For a payment, it might remove an opportunity
• For a payment, it might become an impediment
  

In this case, “In a normal way” generally means “Declare an action, move around some fate points, roll some dice”, which is to say, directly engaging the system.

However, let’s say I want to capture that “we never touched the dice” feeling in a Fate game. I would simply narrate events, answer questions, and play NPCs until we come to a point of uncertainty. In many situations, I will look at it, look at the skills and aspects in play, and resolve the uncertainty with that information. I would probably also be mindful of how “pushing” an aspect might change the outcome, and use those possibilities as options for fate point spends to nudge things along. I’d need to come up with a good rule of thumb to handle situations where no “right” answer suggests itself, but that can be as simple as “Say yes or ask a question”.

I am, at this point, playing a RADICALLY different game than an “engaged” game of Fate. And yet, in practice, I can move pretty seamlessly between these two modes because they both draw on similar source material (the character sheet), they differ only in how the decisions are being made.

Now, I think this is a good thing. Tastes vary, but this is my jam. However, the question it now leaves me asking myself is how to write the game in such a way that this is the intent, not just a happy outcome. I don’t have an answer for that yet, but I feel like this rambling has brought me closer.