Most of you are probably familiar with the “Dramatism/Narrative” “Simulationism” and “Gamism” model (called “GNS” or “The Threefold Model”) begun on rec.games.frp.advocacy over 10 years ago, and further refined on the Forge. Though some people have little use for it, I've found it very helpful. I'm also impressed by the various “stances” players adopt when playing (“Immersive/Deep IC,” “Author/Director,” “Actor,” “Token/Strategist,” etc.), but this column will focus on GNS issues.

In the past, I've done a lot of diceless, even systemless games. These were always dramatist one-shots (or few-shots), and were a lot of fun. But I always felt that longer chronicles needed a fundamentally different approach.

One problem I had with older attempts at dramatist games is that they essentially amounted to “advice the GM to fudge results and make sure the story goes the way he wants it” grafted on to very straightforward, essentially old-school RPGs. White Wolf has made a lot of money on this, but ask DeliriousKitten about “Greg and the Giant Snake” sometime, if you want a gloriously bad example of a GM "roleplaying with himself." But even with the most well-intentioned GM, this crude form of dramatism often takes the edge off the danger to the PCs.

Recent advances in Dramatist technology, such as Trollbabe and Spirit of the Century, have revived my interest in the play style, but these games had not been published when I began work on Tarafore back in “the year of our lord, 2,000.”

So simulationism was (and to a large degree still is) my bag, my bias, my approach. The original ideas and drafts that became the Tarafore System were created for the sheer joy of creation. But as soon as I entered the alpha-testing and revision stage, I focused on what I wanted the game to do: simulationism without massive complexity.

The premise of my Tarafore System is that a system does not need to be complex or difficult in order to facilitate simulationist play. Simulationism, as I understand the term (and my understanding has been primarily shaped by the glory days of r.g.f.a.), means that only game-world considerations affect decisions. In other words, you don't make decisions based on what makes the best story, or what provides the best challenges to the players, but based on “what would happen” within the game world.

Now, this isn't to say that good drama and good challenges can't arise in a simulationism-driven game, but that they arise organically, based on the actions of the PCs and NPCs in the setting. And if it comes down to a choice between what makes a better story and what makes the most sense within the game world, “what really happens” wins.

The two best things about simulationism, in my opinion, are that 1) you can play deep in character and not worry about meta-concerns and 2) when something dramatic does happen – a prophecy coming true or a plot coming together – it has a massive “wow” factor because you know it happened in-world, with no string-pulling from the GM.

1) As a player, I tend to play deep-immersive, and script immunity, movie-style plots, and other such things drives me batty, as if my character was going mad as the world stopped following the normal rules of causality. It's been a huge source of frustration, and is one reason I don't enjoy playing much.

2) In a Tarafore setting chronicle, a group of NPCs called (unimaginatively) the “dark riders” were working for a chaos sorcerer known only as (equally unimaginatively) “the shaman.” Now the shaman has been haunting the characters, sending his three-eyed ravens to spy on them, and so on, for many sessions. This was good foreshadowing in a dramatist sense, but it was what he would do in-world. He was interested in more than one of the PCs, so he spied on them, and it took the PC's a while to discover that the ravens weren't normal birds (the fact that they constantly had ravens following them got their attention after a while, so they investigated, and one incredibly agile PC got close enough to see a raven's third eye).

One of the PC's was an exiled Nova Roman noble whose family had been destroyed in the current Emperor's rise to power. Brendan had based the character off “warlike Lucius” in Shakespeare's Titus Andronicus, and had even named him Lucius Senecus.

The dark riders first sent phantom versions of themselves and then zombie versions of themselves to harrass the PCs, before actually facing the party. When they did, instead of attacking, they offered to give Lucius the power to destroy the simpering worm of an emperor that had so cowardly assassinated his family – not only that, but they offered him the chance to put himself on the throne, or at least in a position of high leadership under a new emperor who owed Lucius everything. In essence, they offered to fulfill the character's main goal in life.

As soon as I finished with the in-character speech, Brendan's posture changed. He looked around the room, laughed that wonderful Chicago-Irish laugh of his, and said, “sorry, guys, but I've got to play in-character.” Then he turned to me and said “Yes.”

That, and the party-rending battle that followed, still stands out as the single most dramatic and powerful moment in any RPG I've ever played or GM'ed. And the later return and redemption of Lucius Senecus provided even more drama, again, in-character and unmanipulated.

And a lot of the joy of it is that it was totally unexpected and unplanned. It happened because the bad guys made the right offer – which is exactly what they'd do. They'd shown their muscle to prove what they could do, and then offered a Faustian bargain to get what they really wanted. They just happened to make an offer Lucius couldn't refuse, and then the fun began.

When Simulationism Starts

In my understanding of Simulationism, it actually starts when play begins. Character generation can support or undermine simulationist play, but it's not a part of it.

For example, a simulationist game doesn't have to (and indeed, there are reasons why it shouldn't) include random character creation in which the player makes few decisions, or even one in which social birth status and inherent traits are determined randomly, but skills and professions are determined by the player based on those random rolls.

In my experience, the more simulationist the game is going to be, and the more deep in-character immersive the players will be, the more strictly the GM needs to vet the characters – not so much in power level as in relationships. The party won't stick together because of the “PC glow,” so they need to have other reasons to be close.

I'll give you two examples from my Tarafore setting, one a successful chronicle, and the other my only truly unsuccessful chronicle since inaugurating the Tarafore system. I was GM in both cases, and learned a great deal by comparing the two games in retrospect.

In the successful chronicle, we started with three players. Two of the characters were middle-aged nobles (Anneas Fiorentino and Santiago Peccavi) from somewhat unfriendly families. They'd worked together for years to maintain the fragile peace in the city of Cassa, and were old friends, despite their families' distrust. The third character was an assassin working for the Fiorentinos. So the earliest sessions involved two old friends and an employee working together toward goals that were already important to them: easy as pie.

The fourth character came in a month or so later, when school started back up. She played a NYC cop who'd been shot during a bank robbery, and awakened on Tarafore (the passage from Earth to Tarafore is death. Tarafore isn't the afterlife by any means – it's just that some people who die get a second chance on a new world). He (the character was a man) was a bit of a celebrity by virtue of being a “Child of Earth,” and proved his worth easily through his martial, investigative, and leadership skills. He fell in with the group easily once he reached Cassa – basically, everyone was interested in him, and he found that the two nobles had ideals similar to his.

The fifth character, Ansa, was a Sibrith con-woman who was searching for her little brother and trying to carve a place out for herself in Cassa. We had a little trouble integrating Ansa, as two of the others perceived that she was insufficiently serious in relation to the troubles they faced. However, she proved her worth and worked with them to preserve the city. Her brother was among the children they rescued, and she even managed to create a position as the Sibrith ambassador to Cassa. As a culture, the Sibrith value chutzpah and initiative so much that they rewarded her for inventing the position by making it official and giving her a salary (this was entirely in-character, in fact, it was the absolute apotheosis of the Sibrith as I had imagined them – Bravo, Dannie!).

One of the original pair of nobles died in the conflict, and the player replaced him with his son, Julio. Of course, there was no trouble bringing Julio into the group – he came back from the University in Sturm to bury his father, Santiago, and finish his unfinished business. Once again the characters made life easy on me.

Anyway, it worked incredibly well because the characters all had inherent reasons to work together. I think introducing them a few at at time helped a lot, too. The nature of the conflict, which threatened their city, made things easier, as it gave them a strong in-character reason to pull together.

Okay, enough about the good. Let's look at the bad and the ugly:

The unsuccessful chronicle had the deck stacked against it from the beginning. It was the first time I'd GM'ed a game via chat program (we'd all moved to different states, so we couldn't play face to face anymore), it was the first time I'd had seven players (I know it was stupid, but I couldn't tell them “no” - they are my friends! From now on, I'll probably limit the number of players, and just apologize profusely to whoever gets left out), and, worst of all: I didn't vet the PC's properly.

First, none of the PC's knew each other before the game began. That was just plain stupid of me. DUMB da-DUMB DUMB DUUUUUMMMB! Second, while they all had interesting agendas, none of the agendas really fit well together. Needless to say, they had little reason to work together, and trying to keep any kind of unity, structure, or sanity to the game was like herding cats.

The chronicle had some great moments and some awesome characters, but ultimately, the lack of cohesion proved incredibly frustrating to me, I didn't have fun, we didn't finish the chronicle (it just sort of trailed off), and, overall, I consider it a failure.

Some of the problems involved me getting used to the peculiarities of online vs. face to face gaming, but I don't think it was the central issue. During the course of this chronicle, when we had too many players missing to run the game, I ran a very successful Halloween survival-horror one-off. In that game, the PC's were required to be old friends, preferably college friends. I spent nearly an hour of game time letting the PC's hang out (the players chat in character) at the cabin before starting the NPC actions into motion. This made the characters feel like old friends, and made the ensuing drama (which resulted in two unplanned PC on PC killings and some emotionally traumatic moments) much more powerful. The one off was wholly successful, though perhaps a bit too emotionally intense/traumatic for some of my players' tastes.

While the sheer number of characters in the Tarafore chronicle made things more difficult, the core problem was that the pc's didn't have strong in-character ties, and the simulationist nature of the game and deep-in-character methods of the players didn't allow the “PC glow” to make them work together. Oh, well: live and learn. I won't make that mistake again in a simulationist-heavy group . . . and hopefully, neither will you. Trust me, it's not a fun mistake to make.

How the Tarafore System Supports Simulationism

A system can either support or hinder a given style of play, as Ron Edwards' excellent essay “System Matters” points out (and my experience bears out). A game with such strongly gamist (in fact, wargame) roots as D&D can make it hard to run a simulationist game, because of the gamist assumptions built into the fabric of the system. That's not to say “D&D sucks,” just that it does some things better than it does others.

With this in mind, I tried to keep my Tarafore System as simulationist as possible. I avoided point-based character creation, not only because I think it encourages min-maxing, but because the balance issues are much more of a gamist concern than a simulationist one.

Additionally, it's incredibly hard to write a point-buy system in which there are not “right” and “wrong” ways to build a character. For example, if you want a bunch of Dexterity-based skills in GURPS, you'll pay a lot more if you get an average Dexterity and buy the skills up to a high level than you will if you buy a high Dexterity and get lower-level skills. You'll end up with the same effective skill level, but you'll save a bunch of points. Why should a less powerful (lower DEX) character be more expensive than a more powerful one? Well, because it's nearly impossible to write a point-based system without these flaws and break points.

With subjective character creation, you get the character you want, and there's no min-maxing or accidentally paying too much for your character's skills involved.

I made the Tarafore System's combat system as realistic as I could (without making it insanely complex), focusing on providing realistic results: that is, wounds had three components: impairment, bleeding, and whether they stop the target. I handled these as realistically as I could, given the data and systems available to me. The end result is that even a powerful character has to be careful, because a head shot can be instantly fatal, and a bad enough body shot can put him in danger of death.

I did a little research on real world poisons, and modeled them as best I could using the extended test sub-system. The example in Column 3: This Is a Test shows how this works. Infected wounds and other diseases work the same way, though the length of the test will vary.

Perhaps the most important pro-simulationist feature of the Tarafore system is its utter lack of meta-mechanics (such as spending “Karma” for a reroll). All of mechanics are from the PC's perspective, rolling for things that can be perceived in the game world, such as trying to climb a tree, resisting a toxin, or trying to hit, intimidate, or seduce another character.

Final Thoughts

I don't want to give the impression that I'm a simulationist snob, or that I disdain gamism and dramatism. Now that I've hammered out my simulationist magnum opus, I've become much more interested in Dramatism (thanks, Ron Edwards!!).

I'm even becoming more interested in Gamism in RPGs, and I'm hoping to eventually understand it to the point of being able to write a gamist RPG that allows really effective player choice, immediate feedback/reward for actions, and player challenge in the simplest package possible. My lofty goal will also be to allow social, psychological, and other “combats” to be as interesting and challenging as physical battles. That's a big-ticket order, and I'm probably 5+ years away from being able to pull it off. But what the heck – this is a hobby, and I've got no deadlines to meet. The joy is in the journey, as they say.

But my Tarafore System was conceived and forged in Simulationism, and I think it does a good job of it. I don't believe the massive book keeping and accounting involved in point-based systems like GURPS and Hero are at all necessary for a simulationist-friendly game. Nor do I think that a complex game system is necessary – in fact, I think that much of GURPS's complexity comes from lingering gamist elements that underpinned the beginnings of our hobby.

I don't want to sound like I'm bashing GURPS – I don't recall it being marketed as “the ultimate simulationist game,” though I've heard players refer to it that way. GURPS is what it is, and many people get a lot of good use out of it, and have for many years. I just want to challenge the assumption that a GURPS-like level of complexity is a prerequisite of simulationist friendly gaming.