Well just deciding that something happens moves us away from what I think of as simulation. I think of simulation as starting with something external, right? Some process or event that we’re trying to replicate in some way.
That sounds right to me.
So to use the dragon as an example, how do we replicate a fictional creature? I expect you’d say we give it some thought and ask “what would happen if…” as you mentioned. But then there are all those many factors to consider. The dragon’s intellect and cunning, its ability to communicate and work with others. Its stamina for flight, its ego and other personality quirks… all those factors that need to be considered, how are they determined?
It seems to me that all of that is up to the GM, right? If the GM is the source of all those factors, then they’re determining the thing that’s being simulated. And that doesn’t seem to make a lot of sense to me.
So if you’ve decided your dragon is clever and cautious and can fly for extended periods with no problem, then you have your dragon attack the PCs from the air at a distance, never getting close enough for them to do much… you’re simulating.
If I imagine my dragon as cruel and megalomaniacal and overconfident with a desire to see its foes crushed in its claws, and so my dragon attacks from the air and then lands and closes with the PCs… I’m also simulating.
Are both of these true? If so, how do we keep it all straight? I mean, did I design my dragon that way for the purpose of including all the PCs in this scene? Or did I design it that way because it felt right? Is there tension there between simulation and narrative or gamist goals?
If the GM decides all the factors that need to be considered for simulation, then is it simulation?
If you go in with the premise that GM decisions cannot be counted as simulation, then you will (as I have said to others above) be left with no explanation. You've implicitly ruled out from first principles the possibility that "GM decides" can equate with simulation through the way in which they decide.
So lets start there, yes. GM decides. What's at issue is
how they decide. Above I said that I could not tell from one written example whether simulationism was being performed, and I queried what was meant by "extrapolates". The quick definition I find of extrapolates is - "extend the application of (a method or conclusion) to an unknown situation by assuming that existing trends will continue or similar methods will be applicable." Some are content to leave that as an "authorial decision" which seems to resist further scrutiny. Perhaps some sort of dualistic explanation involving a creative spirit comes into play.
Setting aside the possibility of a creative spirit that pops ideas into our heads at the moment we need them, I am going to offer the wild thesis that what GM decides is motivated. Above have been discussed the possibility of motivations that are non-simulationist and I agree that what GM says next can be motivated by a diversity of goals and concerns. That is, I agree that non-simulationist play is possible.
As to simulationist play, it relies on assuming, adopting and establishing referents and relationships. A model, if you will. This is a process as fluid and incomplete as thought. When the GM is asked to decide what the dragon will do (where dragons have been established as a referent) they deliberate upon its properties and relationships, and come to a verdict about how it will behave. It will behave as dragons do, and there shall be a behaviour distinctive of dragons (or at least, this kind of dragon) in the world, which will go on to be true at all places in times (barring impinging circumstances.)
Very often GM will not rely on their cognitive powers alone to achieve that. Game texts, lore, additional game design, often in incompletely externalised form. That is, a sketch of a mechanic for dragon behaviour might connect with a partial model of dragon behaviour GM has in mind. From time to time GM will be tasked to judge things outside their current model (which, it bears restating, is
necessarily and
pragmatically incomplete.) They do so with respect for the coherence of their model i.e. within the scope of their conferred expertise in the imaginary domain.
I can see two ways this can be misunderstood (and there are no doubt others.) One is on metaphysical grounds. As I alluded to above, under certain world views all of this is unnecessary and improbable. A creative spirit (or something equally inscrutable) really does pop ideas in our heads. Perhaps the spirit can be informed by frameworks and develop feelings about what sorts of ideas are good ones, but this must always fall short of any sort of dispositional model. Another is taking the differences between model representation on organic neurological structures and model representation on other architectures to rule out the former. Especially without making any allowance for the playful purpose (low-stakes) and constraints (comparatively low-investment) of the former. There might even be an assumption that all this has to play out in a highly self-conscious mode of thinking. That, too is mistaken.