Unless the fifth force doesn't necessarily interact with everything, all the time.
An ordinary rock, for example, might happily go through its entire existence without ever interacting with magic. Ditto for a simple pine tree. But an Elf can't survive a day without it; a Dragon, not an hour.
(neat side effect: this also allows one, if desired, to slot our own real world into the game universe by simply saying that for whatever reason the force of magic never - or only very rarely - interacts with it)
So you are going to postulate a force that, in complete contrast to what we know of the other forces, doesn't interact with those "existing" forces unless it feels like it? Just so that "magic" can retain it's place as a "special snowflake"? Well, whatever floats your boat. If I absolutely have to pick an underlying "truth" I think I'll go with something more akin to the World of Darkness (Mage) ideas.
I don't necessarily want the exact same game each time but I do want them to be compatible as much as possible; such that if a character rolled up in campaign 1 somehow reaches the world on which campaign 4 is set it can - with a minimum of tweaking - fit right in.
Compatibility is nice, but it doesn't
have to be mechanical. I have used Universalis to generate background happenings for HârnMaster, for example; the same character ports from one system to another with vastly different statistics, but it's still the same character. The statistics map into how events resolve in the world under the mechanics in use; as long as the resolution
could conceivably arise under either system in use, the translation is good enough.
I've learned the easiest way to ensure a different game-play experience is to have at least some turnover of players between campaigns.
I would say it's easier to change system, but it's a fussy point; changing players works, too.
Then answer me this: in Keep on the Shadowfell, one of the Hobgoblins at or near the start of the second deck uses a staff in combat that does something like 4d6 electrical to whoever he can touch with it, once per round. This is an amazing item! So amazing that I had to limit its use to three times per day total, once my players got it.
Yet it's not listed in his possessions, nor in the treasure for the room, indicating in theory the party isn't supposed to get it even though they've just seen - and, most likely, felt - it used against them. Why?
Because the Hobgoblin
Warcaster (clue
#1 ) is casting a spell through an implement.
That 4d6 is pretty real to the character that has to eat it, regardless what edition it happens in!
I guess it's a persoanl thing, but I never see hit points as "real" to the character. They have no idea how many hit points they have, or how many they have lost; they just know they are hurt/surprised/off balance/winded/shocked/whatever.
Except there's no mention anywhere of him casting a spell; and even if it was a spell pre-cast "offstage" it'd need to be re-cast before the staff could work again - but the staff just keeps on working.
Clue
#1 is that the Hobgoblin is a Warcaster - and I'm going to take a wild guess that that is some sort of, you know, spell caster. Clue
#2 is that he has a "Quarterstaff" melee basic attack; I'm going to assume that this is the same staff, since I don't see it as sensible for him to wield two at once. Clue
#3 is that the "Shock Staff" attack is rechargable, meaning he cannot automatically cast/use it every turn - it's the monster equivalent of an "encounter" power, since "daily" and "encounter" are typically synonymous for monsters.
So where in the nine hells is the in-world consistency hiding in all this?
The world abides by the system, thus it is consistent. The
reasons behind that consistency are not any part of the realm of the
game system - they are part of the
game world, which is the province of the gaming group.
These "reasons behind the resolutions" are the "underlying meta-system" I have been talking about. They are not a part of the "game system", which just defines how resolutions are made. The game system is how the actual real players at the actual real table decide what happens in the imaginary game world. Why and how those resolutions happen in the imaginary game world is the province of imagination, not of game rules.
This is one thing 3e, for all its other failings, largely got right: everyone in the game world functions about the same in terms of how they are "mechanically" built. It's consistent - as it should and must be to preserve believability.
Consistency doesn't require that all things must be built the same way. Otherwise AEDU characters would be "consistent" where earlier edition characters are not (which, even as a lover of 4e, I think is a ridiculous notion!)
If that staff has the power to give out 4d6 electrical damage when a Hobgoblin wields it, then it either has the power to give out 4d6 electrical when I wield it, assuming I pick up the right end; or there's a bloody good reason (and "useable only by Hobgoblins" certainly qualifies) spelled out in the module where it appears as to why it does not.
The staff doesn't have the power to dish out electrical damage - the Hobgoblin Warcaster wielding the staff does, in exactly the same way as a Fighter wielding a sword has the power to dish out weapon damage, and the Wizard wielding a wand has the power to dish out force damage with a Magic Missile or a Cloud of Daggers.
If the staff itself has no magic to it and the cause for its 4d6 electrical is a spell the Hobgoblin cast, then that spell needs to be written up in the module where it appears as the module has just introduced it to the gameworld and either the DM, the players, ior both may want to use it again...and for consistency it needs to work the same next time as it does this time.
Why? Hobgoblin Warcaster is not a playeable class. If it became so, I would expect such a power to be at least notionally possible (though not necessarily written up). It wouldn't necessarily do the same damage, though, since damage is an abstract, system concept, not a world object concept. There is no "Law of Conservation of Hit Points", nor need there be.