Craw Hammerfist said:
Why, thank you. That is the most cogent explanation for the sim-gamist issue I have heard. It does, however, raise a problem with point of view. You are assuming that a fireball would catch things on fire. If all of the players have that same assumption, then you can go forward in the sim experience. I don't assume a fireball would do more than scorch the walls. I worked in the oil field years ago and saw a handfull of small explosions. Those that were natural gas related typically did not catch things on fire. The fire was hot, but it was over fast. Most of the damage was concussive. An exploding gasoline can catches all kinds of things on fire, however because it throws fuel all over. What if fireball is more akin to the former? Does that mean 1e was gamist for me, but 3e is a sim?
No. To extend your discussion a little, a true high explosive like TNT or PETN almost never catches anything on fire directly because it is over even quicker than a fuel air explosion like natural gas. All of the damage is concussion and fragmentation. But think about that for a second. If it is the intention of the rules to portray fireball as an explosion, shouldn't most of the damage be concussive rather than heat? If we change 'fireball' to 'explosion', should a good portion of the damage be 'bludgeoning'? But in both editions, most of the damage was described as being of type 'fire', and in fact in 1st edition the 'fireball' was explicitly not a high explosive explosion since it was described as exerting negligible pressure on the surroundings - it expanded and engulfed, but it didn't 'blow up' things and create shrappnel.
Ok, then, suppose its still the intention of the designers for it to be literally an expanding ball of flame and not an explosion, and that the heat while damaging is so brief that it doesn't cause secondary fires. But there is a limit to that in simulationist thinking. As your experience indicates, it's quite alright to have a fire that only scorches the walls and the furniture (and a characters armor) without starting a secondary fire because those things are particularly inflamable, but in the rules as written in 3rd edition a fireball going off in a room of gasoline soaked straw doesn't start secondary fires. I don't think anyone has a reasonable expectation that an open flame would do lethal damage to thier person, but not set gasoline soaked straw on fire. A set of rules written with some simulationist intent would almost never say, 'fireball does no damage to objects', or 'fireball doesn't start secondary fires', even if it wanted to avoid the mechanical complexity of secondary fires. Rather, it would say something to the effect of, 'Fireball may start fires in inflamable objects', a phrase you actually here somewhat regularly in the description of fire spells in the first edition PH. A more sophisticated system might define what 'inflamable objects' were (probably with some sort of materials list) and what the likelihood of thier catching on fire would be for a given quantity of flame.
Granting that the fireball is just a single example, I'm sure that there are any number of instances where the natural consequences of a spell or action do not get tracked in D&D. However, for any given scenario, "what is happening in my head" is different, to varying degrees, from "what is happening in your head." How does this get rectified in a sim based game? (or is that the game itself?)
People who care about this sort of thing look for a game system that has a reasonable level of suspension of disbelief. Then the referee tends to fiddle with (house rule) the areas that don't quite jibe with his understanding, possibly with the input of the players. After that, it pretty much depends on the implicit social contract of the group. Some groups then percieve the literal reading of the rules to be the description of the 'physics' of the game world. Assuming that the rules are fairly solid, that will probably hold until a situation comes up which produces really illogical results. Afterwhich, usually with the input of the players, the referee will do some more fiddling with the rules to try to 'improve' the system. Other groups would give the referee latitude to override the rules in the middle of the game where they didn't seem to fit the groups logical expectation of event.
In my first edition games, you typically say a combination of both sorts of resolution going on. It works pretty well so long as your gamist players are willing to confine the overcoming of challenges to in game means. If you have a gamist motivated player (he wants to 'win') who percieves working the metagame to be easier than winning within the rules, that is, he percieves manipulating the DM into changing the rules to produce favorable interpretations of any of his actions is easier than overcoming the challenges under the current rules, then it can be a problem. To a lesser extent, it can be a problem where the setting the referee wants to simulate doesn't correspond to the setting in which a narrativist wants to conduct his story. You typically see this problem when the narrativist player wants to play a character that does dazzling stunts, and the referee wants to simulate 'gritty realism', or conversely (and I've been in this situation), you want to play a 'gritty' realistic character and the game system encourages over the top cardboard technicolor characters who do three impossible things every day before breakfast.
Is simulationism v gamism a purely subjective viewpoint?
Since in GNS, we are dealing with a description of human motivation, its all somewhat subjective. It's hard to climb in peoples heads and see why they are really doing something. It's hard to climb in your own head and see that.
But there are clear cases of 'gamism' vs. 'simulationism' that can be demonstrated. One of the most obvious is if in a game session, the players are asked to play chess or sodoko or solve a riddle. In purist simulationism, the players themselves are part of the metagame and don't really exist. Therefore, in a game from a purely simulationist perspective, puzzles within the game should never be solved by player skill, but rather by character skill. The thing theoretically being simulated is the character solving the puzzle, and this should be done according to the characters skill in 'Knowledge (enigmas)' and not the players. There is never a case were asking the players to solve a riddle is more simulationist than gamist.
Incidently, I believe that a certain amount of simulation ought to infuse a good game system (not to be confused with the belief that a certain amount of realism ought infuse a good game system), but that's where I draw the line. Ultimately, the game is about player enjoyment. Pure simulationism tends to end up reducing the role of the players (player and referee alike) in the game to the extent that it becomes unfun and purely mechanical. You might as well let the computer play the PC's too and just watch the game.
You can see this in other contexts too. Games like SimCity are very far out on the end of the 'simulationist' perspective. In fact, prefers people to refer to his games as 'digital toys' rather than as 'games', because there isn't alot of 'gamist' elements to them. His games are the thought experiment made real, more than they are games.