This article was made possible by the generous support of our patrons. Please consider joining them by supporting us on Patreon.
It used to be part of my job (self-employed) to advocate for the Big Model’s terms. I never chased people around the internet and corrected their usage, an impossible task, but when people came to my spaces, I took pains to make sure we all used the terms correctly.
In my spaces now, I prefer not to use the terms, because I don’t want to sit here and correct people’s usage all the time. They still come up, though. If we want to talk about rpgs with any reference to their history — rpgs in general today, and especially my and our games in particular — there’s no way to do it without reference to GNS and the Big Model.
So here’s a quick retrospective. This is my take today, 15–20 years later, on these topics.
Previously: Revisiting Task & Conflict Resolution
Briefly, the Terms
“GNS” stands for “gamism,” “narrativism,” and “simulationism.”
We discussed them a lot at the Forge, which was a webforum for indie ttrpg creators back in like 2001–2012. They were the core of a larger body of ideas called the “Big Model.”
Ron Edwards coined the term “narrativism.”
“Gamism” and “simulationism” were terms borrowed (and yep, misapplied) from earlier discussions on rec.game.frp.advocacy, a usenet discussion group in the mid and late 90s. I don’t know exactly who coined them or whether they were in use before then. With “dramatism,” a term the Big Model didn’t use, they were the core of “the RGFA Threefold,” a body of ideas that the Big Model thought to build on and/or replace.
Narrativism
I think that the best way to understand narrativism is as a big, long-running game jam.
At the Forge, we thought that narrativism was its own kind of game, it’s own kind of gameplay, but I don’t think that’s true anymore. What’s true is that narrativism is a thing that games and gameplay can do.
For example, I really like games where you make a commitment on incomplete information, get the rest of the information you need, and try to make the best of it, or at least see now how your commitment plays out. Roborally does this, Diplomacy does it, my board game The Abductinators does it, Burning Wheel’s combat system does it. A million games do it. A million other games don’t.
We could give this thing a name for easy reference (“guess & script,” say), but it isn’t its own separate kind of game, it’s just a thing that you can do in a game. A dynamic that games can include.
Same thing with narrativism. Here’s the dynamic that narrativism refers to:
- The PCs have vision, self-interests, best interests, passion, an ideological commitment: something they want and care about. Lajos Egri says “passionate.”
- Their passions put them in conflict with others — other PCs or other NPCs, it doesn’t matter. Their passions oppose others’, threaten others’ interests, provoke others into passionate reaction.
- Both the PCs and their counterparts are equipped to pursue their passions in conflict. Egri says “fit.” They’re physically equipped, emotionally equipped, morally equipped; they have skills, tools, initiative, stamina, followthrough, staying power.
- Nobody pre-plans how it’s going to turn out. The characters are passionate, conflicted, and fit; now turn them lose. Play to let them pursue their passions. Play to find out how far they go, how they escalate, who comes out on top, who compromises, what they win, what it costs, what they prioritize, what they abandon. The only way to know how it plays out, is to play it out!
That’s narrativism, nothing else.
At the Forge, we’d all played narrativist before the term existed, just by arranging our games — often accidentally or unintentionally — to play that way. We’d enjoyed those games and we generally wanted more of them.
So the narrativist movement, the jam, was like: “here’s a cool, fun thing that games can do. Let’s make games that do it! Let’s play them and talk about them!” “Yes, let’s!” So we did.
A lot of the accidental details of the games we made came to be associated with narrativism, wrongly.
For instance, narrativism requires the GM not to plan out a storyline in advance, but people have come to associate it with various forms of player empowerment beyond that. Player narration, crossing John Harper’s line, “director stance” or “writer’s room” play, whatever. Some narrativist games use those techniques but narrativism itself doesn’t require them or refer to them.
Anyway now, in 2025, I don’t think that narrativism is a kind of game anymore. I think it’s a fun thing that a game can do. But: it’s real, it’s a dynamic that definitely exists in some games and definitely doesn’t exist in others. We can recognize it when we see it, seek it out, and do it when we want to.
It’s a dynamic that exists in all of my games from Dogs in the Vineyard (2004) to Apocalypse World (2010). This is because I made them for the jam!
I’m not really doing the jam anymore, but I still enjoy making and playing games that work that way sometimes. It’s a dynamic that still appears in some of my games, even while I’m happily exploring others.
Not Narrativism
To deal with the next couple of ideas, I think we need a working example of a game dynamic that isn’t narrativist. Let’s use the one in Murderous Ghosts:
- The character’s a cool and likeable stand-in for the players, not necessarily passionate and driven. Once the game’s underway, they have no vision or motivation but escape.
- The character’s not caught in a conflict of interests, exactly, or not necessarily. They’re in a dangerous situation that’s over their head, trying to survive.
- They’re not “fit.” If they escalate against the ghosts they’ll lose, we all know this. We aren’t playing to find out who they are, what they’ll sacrifice, where they draw the line, what it’ll cost them. We’re playing to find out only if they can get out of here alive.
If we entertain Egri’s theory of stories, playing Murderous Ghosts doesn’t give us one. And indeed, playing Murderous Ghosts is fun and exciting in its own right, but it’s not satisfying in the same way a compelling story game is. People who go into it expecting a literary or even a folk ghost story are often pretty disappointed by it. What you get instead is a quick, tense little nightmare of simple action, shallow or nonexistent character, unclear motivations, and unanswered mysteries, with (usually) an anticlimactic death cutting it abruptly off.
So: a good example of not a narrativist dynamic.
Murderous Ghosts threw off some of my narrativist colleagues pretty bad. It was my first non-narrativist game, arguably; certainly my first since 2001. It had some of the hallmarks of a narrativist game — an escalating situation, no pre-planning — and the genre trappings we narrativists had come to love. It even says PbtA on the back cover!
But they just couldn’t get it to play narrativist. They tried and tried.
They couldn’t figure out how I’d so badly bungled the design and how poorly I suddenly understood the brief!
I don’t have any grudge about this, I mention it just to reinforce: a good example of a non-narrativist dynamic.
Gamism & Simulationism
We thought that narrativism represented it’s own whole kind of rpg, which meant that there must be other kinds of rpgs too. Gamism & simulationism were our notions for what those other kinds of games must be.
Gamist rpgs would be games you play to win, achieve some kind of victory, or show your mettle as a player, like in normal games, I guess, like Twilight Imperium, Dodgeball, or Tetris. Simulationist rpgs would be games you play to examine your subject matter in depth, whether your subject matter is a fictional world, a fictional character, or the simulative model of your rules.
Something like that. Because gamism and simulationism never had their own long-running design & play jams like narrativism did, they were never anything but notional. Nobody ever laid down a good workable spec.
It’s tempting to look at Murderous Ghosts’ non-narrativist workings and say, “if it’s not narrativist, then aha, it must be gamist or simulationist. Which?”
But you and I could put our heads together and come up with interesting game dynamics all day long, limited only by our inventiveness as creators. Narrativism is exactly one of them. We’re supposed to divide the rest between gamism and simulationism? Why?
No, I don’t think that we should expect gamism and simulationism to mean anything at all, just because narrativism does.
That is — the gamism and simulationism of the Big Model. The terms gamism and simulationism were originally coined on rec.games.frp.advocacy back in the 90s, along with a term the Big Model didn’t adopt, dramatism. My vote would be to return them whole and outright. In the RGFA Threefold, they mean pretty much what you think they mean. We could forget that the Big Model ever seized on them:
Whenever you see the idea that gamism, simulationism, and “narrativism” (that is, dramatism) are spectrums, or that you might make a gamist decision in a sea of simulationist decisions, or that a game’s mechanics or subsystems can be broken down into their gamist, simulationist, or “narrativist” (dramatist) components, those ideas hearken back to the RGFA Threefold.
Incompatible Creative Agendas
The Big Model took the line that narrativism, gamism, and simulationism are mutually incompatible. When you play, you’re doing at most one of them, never two, never all three.
A lot of people had a lot of trouble with this. “Absurd!” they said. “I do two or all three of them all the time!”
Now anymore, I don’t think that gamism and simulationism mean anything, so saying that they’re mutually incompatible doesn’t mean anything either.
But look at the dynamic that narrativism represents. Compare it with the dynamic from Murderous Ghosts. For a single character, in a single situation, in a single game? They actually are mutually incompatible. Your character can’t really be passionate and not passionate, locked in conflict and not locked in conflict, fit and unfit at the same time. The GM can’t both pre-plan a storyline and not pre-plan one.
Say that you, the player, think that we’re playing narrativist, so you’ve got your passionate, fit character all ready to go, eager to escalate conflicts, draw lines, find out who you are, what you stand for, and how far you’ll go. But I, the GM, am going for a different dynamic in the game, so I’ve pre-planned a storyline with a big boss NPC you can’t defeat until the 3rd act. I think you can see where that’ll be frustrating for both of us.
So I can see where the Big Model was coming from.
The Big Model called it “creative agenda clash,” and said that it can ruin games. I think it’s just regular old creative differences — which can ruin games.
Thanks For Reading
As always, it’s my pleasure to answer questions. If you have any, please ask!
PS: Some Other Named Game Dynamics
Here’s my always-favorite, Sandra Snan’s Blorb Principles.
Here’s a manifesto by Sam Sorensen: New Simulationism. In it, he describes an interesting dynamic that rpgs can have.
Here’s a piece by Eero Tuovinen: Observations on GNS Simulationism. In it, he describes, not one, not two, but five interesting dynamics that rpgs can have. (In support of one of my points above, note that they aren’t compatible with each other, but GNS simulationism is nevertheless supposed to include them all.)
Any of these could, cultivated by motivated fans, form the basis of a long-running design- and/or play jam. Maybe some of them are already doing it. Maybe Blorb!
Christian Griffen says:
Yeah, totally on board with this regarding narrativism as a dynamic. And we got some awesome games out of this jam!
I feel like the way I understood Gamism (which may not be how others interpreted it) was also a dynamic, one that crossed into the player domain. There’s a fictional challenge, and tools to address it, and the fitness that’s being measured is the player’s ability to creatively address the challenge. I did try and start a jam about this with Beast Hunters, but it turns out not that many people were excited about that dynamic, at least not _all the time_. Like, it’s fun here and there, but as the core of a game that players do throughout, it’s actually pretty exhausting. So now I think it’s best sprinkled in here and there, in certain context.
Matt W says:
Wait, isn’t Gamism qua “the fitness that’s being measured is the player’s ability to creatively address the challenge” the entirety of the OSR??
Paul T. says:
I think this is fairly, but not entirely, accurate. I see much of/most OSR people as quite obviously Gamist, in the Big Model sense, but there are also variations and groups which are much more varied in their approach. Arguably, the NSR folks are shifting away from obvious Gamism towards something else (perhaps Simulationism, in the Big Model, perhaps not).
It’s fairly interesting to observe, actually! But I have found that embracing a Gamist lens of old-school play has really improved my games (as well as my participation as a player in those of others), so it’s at least strongly in the ballpark of whatever-is-actually-going-on.
Like, assuming that old school play is Gamist, in my experience, will reliably at least get you from “no fun” or 50% fun to 90% fun.
It’s certainly interesting to think about!
Levi Kornelsen says:
Saying that ANYTHING is the entirety of the OSR is always a mistake.
But it’s in that stew.
Christian Griffen says:
From what I’ve seen of the OSR — and I would love to get recommendations for counter examples! — when it is about addressing challenges it’s either mechanical (e.g. character builds), or it encourages going _outside_ the mechanics for creative solutions (e.g., combat is so deadly that we solve problems in free negotiation with the GM). I’m not knocking that approach, I just haven’t seen the same kind of attempt at making mechanics that _bake in_ the creative approaches, as we’ve seen Nar games bake in thematic tension and decisions.
Matt W says:
To the extent there’s anything at all consistent in the OSR, it’s a hatred of character builds. Addressing challenge through character builds is essentially the other branch of Gamism from OSR.
Back on story-games.com (RIP), we had some great discussions about Technical Agenda as a thing separate from Creative Agenda. In fact, those discussions are part of what led to Sandra’s formulation of Blorb and her other “nonsense” words.
For example, a lot of gamers who want Narr play, dislike the feeling of stepping outside their characters necessitated by a lot of early story-games that used Flags and various metacurrencies. (You might almost call Technical Agenda the same thing as *Stance* Agenda, though it’s not a 100% overlap.) So PbtA games might work for them, but Burning Wheel wouldn’t.
With Gamism, we see the same thing with “addressing the challenge via direct negotiation of the game’s fiction” versus “addressing the challenge via the game mechanics.”
Gre says:
Echoing the above I would 100% say the OSR has been another long running play and design jam where people explored and developed a type of play. It was just more focused on play and less academical and much more decentralised and consequently we never got a uniform definition. OSR is probably not even a good name (no surprise) and there are games lumped under that umbrella that do something different. But there is certainly a body of work and a well established “thing games can do”.
I was never ver active on the Forge, but I was absorbing and learning from you all a lot, so the GNS is very much the foundation of my understanding of rpgs, but I agree the not-narrativist terms were poorly developed and all of them terribly missapplied outside the space and prefer to find other ways to talk about it. Principles, agendas, player goals, etc. work much better and can be applied on a game-by-game basis.
As a final thought I think there’s another movement/jam that has been happening for a long time, it’s just that it’s happening outside the spotlight and very few people in that jam talk about it in ways that could be construed as a working, replicable model.
ndpaoletta says:
I’m interested in the idea that simulationism doesn’t mean anything in the Big Model. I agree that we never got to a satisfactory wrangling, but there was a fruitful conversation coming out of Chris Lehrich’s bricolage stuff that has really stuck with me. Honestly, the PbtA lineage is a pretty simulationist jam to me, in terms of what “genre emulation” actually is and how many games (including mine!) construct something new with the existing pieces by assigning new meaning to them.
But, looking back at some of those threads (AND I DID), Chris was already treating bricolage as something outside the Big Model and not particular to Creative Agenda.
The other thing that sticks with me is that the whole Big Model was attempting to talk about color, and we ended up getting stuck with the boring stuff (creative agenda) and never quite got there.
Vincent Baker says:
Yeah!
Nat says:
I’ve always thought that a big part of the problem with getting a coherent understanding of “Simulationism” in the GNS sense is that it defines as “one kind of play” what I’ve always seen as 2 not-terribly-closely-related styles of RPG play or maybe priorities in RPGing. IME — not just at the table, but in discussions online and in meatspace over decades — folks who are /really/ into a game that gives them the feel of, say, a comicbook supers world and folks who are /really/ into, say, an authentically historical world are pursuing two different things, don’t see each other as being part of the same style of gameplay, and don’t tend to overlap much. It’s why, when we were inventing what became the Threefold, those two got split into separate conceptual playstyles. Threefold Simulationism was conceived to cover the desire for “realism” and fidelity to “how it would ‘really’ happen”. The folks who wanted genre emulation, at least in that discussion, didn’t think of themselves as “simulating” in the same way, and tended to clash with the self-described Simulationists as much as or more than those “storytellers” clashed when trying to play with self-described Gamists.
But, of course, the Threefold had it’s own category that probably is better thought of as being at least 2 distinct playstyles lumped together: Narrativism was the outcome of folks realizing that there were people and groups who were prioritizing something over /both/ Gamism /and/ Simulationism. And it lumped together at least two distinct desires: the desire to emulate a particular genre, setting, and/or story structure; and the desire to collaboratively create an engaging, satisfying story. IME, those two things /do/ frequently go together at the table, but they sometimes also clash: a GNS-Narrativist style story is definitionally unplanned, “play to find out” and all that, so the person who wants their game to be a satisfying story in that sense can easily run into creative differences with the person who wants the game to follow genre tropes for structure and outcome. The former person might find it stifling or uninteresting to follow genre conventions /too/ closely, because it cuts down on the “finding out”—where’s the interest, where’s the fun, if you know how it’s going to turn out? The latter person might find it frustrating or random to leave everything up to some combo of randomizers and differing ideas of what a character would do—how is it a satisfying story if it doesn’t fit the expected structure?
Vincent Baker says:
Yep!
> the person who wants their game to be a satisfying
> story in that sense [play to find out] can easily run into
> creative differences with the person who wants the
> game to follow genre tropes for structure and outcome
I think we see this same tension in the world of PbtA games too.
TrooperSJP says:
I have always found the Threefold model more resonant with me, and I think I’m going to go back to that terminology…of course, it was made by people who were primarily simulationists, and that is where my heart lies, so there is probably a reason why it works better for me.
That said, people don’t know the Threefold model at all, the GNS had better dissemination so that is always going to be there as people’s go to definitions. But there is this fascinating morphing of GNS that has taken place in the hands of a younger generation that never read the original Forge manifestos. And that is what I wanted to touch on.
People are using the GNS terms now, and making games now, in ways that don’t reflect much of what people might have meant back then–i.e. how Vincent is defining Narrativism. What Vincent is describing as Narrativism (make a character with passions and then play to find out what happens), is for me, from a Threefold point of view, Character Simulationism. You set up the simulation, and then you see where it goes…and importantly, going back to John Kim’s discussion of simulationism: “A central feature of Simulationism is that it rejects closely identifying role-playing with either fiction writing or other types of games. This is important, because role-playing is often viewed and judged by external standards. For example, an RPG might be judged poor if the events it produces would not make a good book or movie of that type. Conversely, an RPG might be judged poor if it isn’t fairly balance like a board game or card game.” (from Darkshire.net)
Narrativism, how people use that term now, has really changed quite a bit from how Vincent is describing it…to being about reproducing satisfying narrative structures that more closely align with the tropes and flow of traditional Western story structures. Hence you get the GUMSHOE system removing all rolls to find clues because, “If you think about it, this is how the source materials we base our mystery scenarios on handle clues. You don’t see the forensic techies on CSI failing to successfully use their lab equipment, or Sherlock Holmes stymied and unable to move forward because he blew his Zoology roll. You don’t see this because, in a story, failure to gain information is rarely more interesting than getting it. New information opens up new narrative possibilities, new choices and actions for the characters. Failure to get information is a null result that takes you nowhere.
In a fictional procedural, whether it’s a mystery novel or an episode of a cop show, the emphasis isn’t on finding the clues in the first place. When it really matters, you may get a paragraph telling you how difficult the search was, or a montage of a CSI team tossing an apartment. But the action really starts after the clues are gathered. Investigative scenarios are not about finding clues, they’re about interpreting the clues you do find.” (From Ashen Stars).
And so many Narrativst games have become about reproducing the beats of the 3/5-Act Structure, or reproducing the tropes of traditional fiction. The player agency that is often built mechanically into such games are often not about PC agency to find out what happens, but Player meta-agency to move the activity closer towards a traditional narrative…not playing to find out what happens, but writing a satisfying story. That is very different from what Vincent describes.
If I’m asked to run something in the “Narrativist Gen 2, let’s writer’s room a story” way, I can do so. I’ve worked intentionally to be able to GM for those Narrativist Gen2 players or the Gamist players. They will have a good time. It isn’t my preferred style, but I GM Narrativist Gen 2/Story games surprisingly often, and we will get that Western narrative structure in there and the players get to write their characters as much as they want!
But my heart beats for a good deep character simulation. And that particular desire is often…misunderstood under GNS, both Gen 1 and Gen 2. That said, having purposefully put myself in the position to GM in all of these styles, I can see those moments where GNS styles can clash. GNS is often about game design, whereas the Threefold model was often understood to be about GM decision making during a session and perhaps other parts of the game contract…about goals. And thinking about it in terms of GM and also Player decision making? There are many moments when the is no conflict between those styles and everyone is fine. And there are definitely moments where there are going to clash.
Can PCs die in the first scene to an NPC who isn’t important?
What happens if the PCs fail to find a clue?
What difficulty level do you give for a lock?
How do you decide what your character will do?
All of these questions have values and expectations and world views baked into them that can lead to different outcomes based on if you are swinging Gamist, Dramatist, or Simulationist. I play differently as a player if I’m asked to play in a Narrativist Gen2 Story game, than if I’m in a Simulationist Game or Gamist Game. I make decisions as a GM differently based on if I’m swinging Gamist, Dramatist, or Simulationist.
To go back to your metaphor about jamming.
GNS always felt to me as being created for designers to help them compose.
The Threefold Model always felt to me as being created for GM/Players to help the improvise.
Elsie says:
This definition of narrativism sounds a lot like what’s called ‘player agency’ nowadays.
Vincent Baker says:
Yep!
Arthur says:
Interesting point there about creative agendas and creative differences! I think the key differences here compared to the Big Model dogma are that:
– Not all differences in creative desires automatically lead to “creative differences”; if I’m doing my thing and you’re doing your thing, and our things are either orthogonal enough that they don’t really connect or are interacting in a way we find mutually interesting, then that’s great and we should keep doing that.
– Not all similarities in creative desires automatically avoid creative differences. Two people can want to play a game about passionate conflicts in theory but fail to agree on what type of conflict they find interesting in practice. One person’s dangerous drift into borderline pre-planning can be another person’s important guardrail to ensure that the players and their counterparts have that quality of “fitness”. Differences in taste can in fact be more of a problem when tastes are very closely aligned than when they are very different.
– A game’s design can lend itself to a particular creative desire or set of such, but it can’t create harmony at the table. Coherence does not come from the designer but comes from the participants sitting down and agreeing on a particular interpretation of the game text; the game text can make this easier by explaining itself better, but it can’t salvage a situation where participants arrive at intended creative directions which are simultaneously wholly compatible with the text but also wholly incompatible with each other’s enjoyment of the experience, as will be possible with all but the most restrictive designs.
Vincent Baker says:
Yes, yes and yes!
Frederik Jensen says:
It’s much easier to form a group identity if you define someone to exclude. I honestly think that played a part in why we had G&S in the model while everyone was into N.
Good post.
Teori-inblick #209 – ropeblogi says:
[…] Revisiting GNS av Vincent Baker berättar om författarens nuvarande syn på teorien. https://lumpley.games/2025/04/07/revisiting-gns/ […]
Shiny TTRPG links #220 - SMOOSH says:
[…] lumpley games gives us Revisiting GNS […]
la persona que escribe esta carta no es la que lees – piedrapapeld20 says:
[…] Vincent E. Baker (autor de Apocalypse World, Dogs in the vineyard y pope fundacional del PbtA, etc) revisita la GNS, uno de los términos que más polémica han levantado y que él trata con una visión cercana a […]
Spike Spiegel says:
Relatively new to the Forge material, but I did read the archived essays, and this blog post sparked a question, forgive me if I got it all wrong:
“The characters are passionate, conflicted, and fit; now turn them lose.”
Isn’t the expectation that the “turn them loose” part would produce a story with literary value called “El Dorado” in Big Model terms…?
Then, does the “characters are passionate, conflicted, and fit” preliminary requirement allows players to achieve El Dorado?
Or does “turn them loose” in fact implies “turn them loose, but equip the players with tools and techniques (potentially including meta-currencies, and a significant dose of author stance facilitators) they can use to ensure the production of a valuable story”?
Thank you!
Vincent Baker says:
Yep, you’ve got it.
According to the Big Model, the road to good stories in roleplaying is to create passionate, conflicted, and fit characters, then to turn them and the players loose with rules that let them pursue, escalate, and resolve their conflicts.