On Categorization, Rule Buckets, and Diegesis

16 Oct, 2025

Standing on the Shoulders of Giants

This discussion is going to center a couple of concepts that are useful to understand when talking shop about TRPGs.

1. Diegesis. This is the cornerstone of what I want to discuss. Diegesis and its adjectival form diegetic are the terms that come from video production to describe when an element, like audio, are recorded as a part of a scene rather than piped in afterward. For a focused discussion on diegesis and how it relates to tabletop, please read this post by Cavegirl.
2. RISS. I can’t do this any service by summarizing it, but it is a framework for talking about styles of game preparation and styles of games-at-the-table. I recommend familiarizing yourself with RISS.
3. GNS. I don’t personally care for Gamist vs Narrativist vs Simulationist as axes of discussing tabletop games, but there is a Wikipedia article. Part of the problem is that you can basically develop an opinion on both the GNS Theory and games as soon as you read the acronym. Still good to know that it exists and unfortunately this discussion might dip its toes into GNS territory if I’m not careful. You can read the high points here.
4. Rules. I’m probably going to borrow some ideas from Jay Dragon’s excellent treatise on Does Super Mario Bros 1985 Have Rules so you should read it and support her. Since originally drafting this post I have also read MIT Press's The Rule Book, which is very well worth your time to read.
5. Ontology. Ontology is the idea that the existence of one thing is necessary and sufficient to imply the existence of some other number of things with relationships to the first thing. The easiest example is that humans and dogs are unrelated things that exist, but the category of humans called Pet Owners necessarily implies the existence of another category of things called Pets, which sometimes include dogs.

Get On With It

There has been a lot of good thinking in this generation of Post Old School Renaissance games about the relationship between “story games” (driven largely by Apocalypse World and its descendants), the OSR, and the Free Kriegspiel Renaissance (FKR). The OSR and the modern Story Game are siblings descended from the reaction to 3.5e and its family of games that are today considered “Traditional” games, and arguably the old school style of play that fought its way to rebirth was itself really a revival in the Free Kriegspiel wargaming that predated Dungeons and Dragons.

So what?

Well people like categories, and I do too. But I think people talk about game categories in ways that I find annoying, so I want to see if the way I’m thinking about it resonates with anyone else. I’m going to try to set up some ontologies and then we can talk about how that may inform how we talk about games and categorize mechanics.

Webster’s Dictionary Defines Ontology As…

First, here is a sketch of the things that I think are implied when we talk about the roleplaying games that I have the most experience with. Roleplaying games definitely exist that forgo elements that are in this drawing, but what we are focusing on here is implied existence and implied relationships.

Now that everyone is looking at this and mad, we can set up some operating definitions. You’ll see that this sketch is roughly halved into objects that exist diegetically, or in the game fiction, and objects that exist non-diegetically, or outside of the game fiction. Our first ontology that we propose is that everything that exists on one side of this dividing line should either act on or be acted upon by something on the other side of the line. If an object exists on one side and does not have an interface with the other side, it probably does not need to be in the game.

• Players are the people that are participating in the game. One specific type of player is the Game Manager. Players directly interact with the Game State, they usually directly interact with the Rules, and at least some players will directly interact with the Game Prep.
• The Rulebook/System is whatever container you use for the Formal Rules of your game. This includes house rules, because it's really just the rules and procedures that the table has socially agreed to.
• Game Prep is whatever has been brought to the game to set the stage and facilitate play. This could be a module or it could be a half-baked thought, I’ve played fun RP sessions with both. Game prep is on the diegesis line because it physically exists outside of the game but it forms the foundation of the fiction of the game.
  • One aside I want to introduce here is that the System and the Prep are considered separately here because different families of games rely more heavily on one than the other. When we talk about games associated with the OSR movement and call them "Adventure Games", it's because the system mostly exists to provide a mechanical interface for exploring the prep/fiction/adventure/module. In games that apply more formal structure and heft to characters' abilities and relationships, the prep exists to provide a fictional realm for players to exercise their characters.
• Tools, Props, Metacurrencies, and Character Sheets are not necessarily separate objects, but these buckets reflect whatever physically or digitally is used to save, communicate, or advance the current game state, including random number generators like dice and cards.
• Player Characters and Non-Player Characters are diegetic objects which act in the Imagined Space based on the internal rules and/or prep work of the players
• The Imagined Space is roughly made up of a Setting with a Timeline and Logic. You can argue with me about this or you can be cool, that’s your choice.
• The Narrative is potentially a fraught term. For the purposes of term definition, I claim that the narrative of a game is the shared understanding and assembly of events that have taken place in the game world (as experienced at the table) and its impact on the relationships between player characters and non-player characters. I believe that it exists as a synthesis between the diegetic and the non-diegetic elements of play, specifically the Game State and the Imagined Space.

Now. I’ve done something terrible here and I’ve created like 13 categories of things, each of which has some relationship to most of the others so if we wanted to enumerate all combinatoric relationships it would be like 78 interactions. Maybe that would be fun for some people to read, but I want to skip that step, zoom back out, and recontextualize why I wanted to write this post to begin with.

I Wrote this Discord Post In Between Meetings

Earlier today I asked whether there had been much discussion about the difference between game rules that directly interact with the fiction of the game, the narrative of the game, or the game itself. The examples I gave were:

• Fiction: Tool usage. If the game says that I have 50 feet of rope, my character can, in fiction, do something that the logic of the game says I could do with 50 feet of rope.
• Fiction: A combat ability that explicitly describes what the character is doing. Bad example but I hadn’t thought that hard about it yet.
• Narrative: Introduce an NPC. In Daggerheart, the mastermind rogue (or whatever it’s called) has a character ability that lets them create an NPC. In The Electrum Archives, Fixers can spontaneously have contacts in organizations. This is one of my favorite things in games, full disclosure.
• Narrative: GM Intrusions in Cypher, which let the GM hand out a resource in exchange for changing the flow of the scene.
• Game Itself: Rerolls, HP, dice/card manipulation. Things that directly let you fiddle with non-diegetic elements like tools and your character sheet with limited narrative or fictional contextualization, or in exchange for modifying the game state in some way

Let’s Refine That

I want to introduce one more quick working definition and then we will see what we can use it for:

Interaction. An Interaction is the way that one of our diegetic or non-diegetic objects affects another object. We might enumerate types of interactions later but let’s keep it vague for now.

So let’s walk through one of my examples and maybe one more test case and we will see how this bears out. One of my most-run modules this past year was Michael Madsen’s Caves of Cowardice. In one of the dungeon areas, the players find themselves on one ledge overlooking a chasm while some innocent bystander NPCs stand on another ledge about 30 feet away. We were playing His Majesty the Worm. One of my play groups had rope and iron spikes in their inventory, and the thief character could scale sheer cliffs, so they used the rope and spikes to create a tightrope in order to safely traverse the ledge. The interactions at work here are:

1. Players recognize that they have resources on their Character Sheets
2. Players desire to interact with Non-Player Characters, who were a part of the Game Prep (written into the module)
3. The thief’s Character Sheet establishes in-fiction Logic (he can climb the sheer walls of the chasm)
4. Players and Game Manager negotiate whether the rope and spikes are sufficient for an ad-hoc rope bridge in the Logic of the space, and the rope is sufficient for the physical geography (Setting) of the space
5. Challenge is resolved without engaging the Formal Rules because the Game State and the Imagined Space sufficiently resolved any uncertainty

My other play group who attempted this challenge took a more direct approach. They also had rope, so they decided to throw the bitter end of rope to the NPCs, asked them to tie a knot around their midsection, and then encouraged the NPCs to jump from their ledge and swing to the Player Characters’ ledge. I, as GM, didn’t feel that they had done enough to mitigate risk, so I made the safe traversal a “Test of Fate” (in-game method of resolving uncertainty). The players failed the test and the NPCs did not survive the encounter. The interactions at work here are:

1. Players recognize that they have resources on their Character Sheets
2. Players desire to interact with NPCs (etc, etc)
3. Players have sufficient rope to reach the NPCs but do not fully consider the Logic of the physical space
4. Game manager determines that NPCs can tie a rope (no DC15 dexterity check today) but is unsure how this may shake out, so the System’s method of action resolution is invoked to fill the gaps in the Logic
5. Players use Tools to execute the game mechanic and fail the test, and the Non-Player Characters exit the scene in more pieces than they started the scene.

Part of what I want to highlight here is that, as far as control groups go, this is a pretty clean case. Same system, same module, but different players tackling a problem and the final Narrative is quite different! So this brings us to a thesis I’ve been developing today:

Narrative is the synthesis of the Game State interacting with the Imagined Space in response to the actions of the Players (to include the GM) over time.

Okay But for Real Now, So What

As I’ve been noodling on this today, there have been a few things I’ve wanted to address and I don’t know if I’ve satisfactorily gotten to any of them. The first one is that I think people with different perspectives have really strong feelings about what it means for players to impact the narrative of a game. As someone who has recently come to love the OSR and the thinking around it, I know that I am inclined to think that players have exceptional narrative control in sandboxy adventure games.

Players, piloting their Player Characters, have significant agency in how they pass time and maneuver in a setting, and ideally they have a clear understanding of the internal logic. They have a lot of agency in the Imagined Space. Where they tend to have limited control inputs is in the Game State. Their character sheets typically don’t have a lot of information (or “buttons”) and they don’t have much in the way of metacurrencies that can be used to abstractly manipulate the game universe. Thus, the Narrative truly will follow their actions! But no rewriting reality for them.

In comparison, let’s think about a classic “storygame”, like Scum and Villainy (a Forged in the Dark game in the vein of Firefly or Cowboy Bebop). These games tend to be based around Playbooks, which are a type of Character Sheet which spells out the ways that player characters can act in a scene, theoretically limiting (or at least enumerating) the ways that characters can directly act in the imagined space. FITD games also use flashback mechanics, which let players and the game manager negotiate metacurrency (Stress) costs to directly edit the timeline and setting of the imagined space. Triangle Agency similarly allows players to negotiate means of editing the timeline, setting and logic of the imagined space with a combination of metacurrency (Chaos) and character sheet abilities. This sort of game then allows input to the Narrative of the game by player and GM interaction with both the game state and the imagined space. It’s probably worth debating which games mediate this narrative influence through the character avatar vs which explicitly ignore the character avatar, but it’s not a conversation I feel personally invested in tonight.

If that’s the case, what makes either of these different from traditional games like 3.5e D&D? I think this tugs at a different axis – fundamentally what I want to develop here is a more cohesive language for talking about what people feel like they mean when they discuss the relationship of rules and players to a game feeling gamey, or a game feeling like storytelling, or a game feeling like a puzzle that just needs to be solved. But if I really have to say something, I feel like what makes traditional games feel so traditional is that the players interact with the imagined space as mediated by their character sheets while still mainly being constrained to the logic of the imagined space. They rarely provide tools for the players to do things like create NPCs other than through the specific, dreaded form of Game Prep known as the backstory.

If we consider 4e D&D or its inheritors (Draw Steel, Lancer, ICON) with this framework, we see a game that fully foregrounds the Game State and centers the change in Game State in the advancement of the Narrative – player characters execute abilities on their character sheet as they maneuver an abstracted space as minis on a game mat without requiring as much cognitive load in the Imagined Space. That’s not to say that people playing Lancer aren’t imagining their dope mecha doing cool Gundam Wing shit – the Narrative is still happening in an Imagined Space, but they are not required to consider the internal setting logic of a sword attack that makes an X pattern and they can focus on more important, cool shit.

Tabletop RPGs are a Land of Contrasts

I’d like to hear what you think, and if I see anyone online trying to call an entire game Narrativist or Simulationist, I’ll give you a thumbs down as loudly as I can. Part of what I want to get at, and may focus on in another post, is that it is worth thinking about how a game’s formal rules mediate interactions between players, their characters, and the narrative (as defined here). If you say that you feel disempowered to influence the narrative of your game, is that due to the formal rules, or is that due to the informal rules of the table? Let’s talk (as long as you are nice to me.)