Only one creates a fiction where no explanation is possible.I don't see important differences between the two.
Playing my MHRP Fantasy game, it is not possible for the characters to learn how dragons fly, because there is no explanation.
Playing a regular MHRPG game, it is not possible for the characters to learn how "unstable molecules", "Pym particles" and "mutant abilities" work, because they are all nonsense.
In all these cases, if play turns its attention to these questions, the fiction yields, and can yield, no answer.
Whereas when attention in play turns to the history of the runes, answers are feasible (as is demonstrated by the fact that they were generated via the standard procedure of play). And can continue to be provided if attention continues to be paid.
Sure? You're now giving an example of contradiction of a known natural law, or at least generalisation of tendency, in the domain of human geography.in constructing a fictional world, we create both fictional laws (this spell requires a more skilled wizard than that spell) and fictional history (Melvaunt lies along the Moonsea). Both of these have opportunities for contradictions; e.g., the previously established climate of the Moonsea may not support the population of Melvaunt. In that case, the only solution for the players is to not look too closely about the details--to not worry about things like "how many people can this farmland support". That is a similar suspension of disbelief to not worrying about "how exactly does hyperspace work".
As I said, I don't see how this relates to the rune example, in which there was no actual, or threatened, contradiction of any known natural law or generalisation of tendency. Any more than, for example, the GM mentioning a NPC but not having decided yet what their name is, or who their parents are, or whether they sleep in a bed or on the floor, creates a need to suspend disbelief.