By that definition nothing can ever be diegetic. Watching a movie? It was recorded on a camera and then edited. Reading a book? There was an author, multiple drafts, and editor. No fiction whether it's movies, books, games, have no outside influence.
The camera is non-diegetic, right? As is the movie screen, the projector, and the auditorium.
Right. The events of writing the story aren't part of the story; nor is
being a member of the audience part of the story. That's all.
My proposition is that the map in the fiction is diegetic, whilst the prop representing it isn't.
I think this quickly gets into the metaphysical weeds. The prop qua token is not part of the fiction. But the map-type that it instantiates is also instantiated in the fiction (assuming it's an accurate depiction of what the characters in the fiction are looking at).
This would be analogous to diegetic sounds: the sound waves in the cinema are not the same as the sound waves coming from the (imaginary) character's (imaginary) radio. But the same audial type is instantiated in both cases: that's why we can say that the audience and the character
hear the same sound.
As Sorensen says "When the dragon reaches 0 HP and dies, it does not die because it reached 0 HP: it dies because it has suffered so much damage it cannot endure further. Your game’s abstractions are representative, not authoritative."
Well, this brings us back to the point that has been obvious since Gygax wrote his PHB and DMG: in D&D, hit points are frequently not representational. As I've posted already in this thread, they're just a clock. (Gygax does allows that for some monsters, which do get worn down by suffering many wounds, they are representational. But he is clear that that is not the case for all hp loss.)
The people who invented RQ, RM etc weren't
confused about the way hp worked in D&D. They wanted the process for resolving combat to be representational.
So according to you, the guy who wrote a manifesto on what simulationism is has restrictions so strict that any game that claims to be simulationist can't actually be simulationist?
(1) Some people's manifestoes call for the impossible. So if that's the case here, it would hardly be unique.
(2) I think the manifesto's principles can be realised, or approximately so, under certain conditions. As I posted, they are very demanding. I don't really see how it is going to work beyond map-and-key play; and map-and-key play is fairly tightly constrained in what it can deal with.
I mean, it's not as if the example I gave is some random hypothetical. I've got 1000s of hours of GMing experience with one of the most simulationist systems around (Rolemaster). I have a fairly good sense of what its limitations are.
I don't accept your reasoning because if I did then no TTRPG I can imagine could be considered simulationist and the word has no meaning.
I'm not sure how familiar you are with RPGs like RQ, RM, Classic Traveller, Bushido and the like. They all aspire to simulationism of the sort that Sorensen describes. Whatever their various limitations, they are completely different from D&D in the degree to which they achieve that aspiration.