it is practical. it is nonsense unless you speak the language. it does not cease to exist when you cease to believe in it.
sure, why not? well, is that how people use them?
theoreticians and experimentalists work in both paradigms. we call them academics and artists, clerks and barristers; scientists and performers, researchers and engineers.
want to develop a specialized notation for the topology of spell lattices with a well-defined interior product? want to record a treatise on a reclusive bardic school to preserve its art for posterity? theoretical magic.
gonna generate a unique chimera, because there aren’t enough of those already? whatever the means, if you’re going to terrorize the nearby village with it, it also becomes practical divine work.
[arcane and tech are pretty rote here. how about solipsism - you hatch a chicken’s egg under a toad, get (disappointingly) an owlbear, and isn’t it a pity a rampaging owlbear took most of the hens last month. or paragon - you reshape the egg’s developmental fate with a touch and a tailored viral payload. don’t take a martial approach, unless ‘live taxidermy’ sounds exciting.]
remember that each of these is a distinct fluff, preferably not a distinct physics. say we have a flaming sword. how is that possible?
arcane: because it isn't.
bardic: because the story says so.
divine: because you wished for and were granted it.
invocation: because you paid for and were granted it.
solipsistic: because it is.
paragon: because the flame is in your nature.
martial: because your sword is just that good.
theoretic: because it is a weak conduit to the elemental plane of fire, or because of the fuel-fuller along its length, or so forth.
but every fluff is an approximation to (an observation of) the underlying mechanism, which depends on the cosmology.
in a reality where there are no gods to grant spells, the audience’s expectations grant clerics power instead. in a reality where lightning comes from the elemental plane of air, magnetism may be due to the plane of earth, and electromagnets represent an exceptional confluence rather than an immediate consequence. or the elements do just spontaneously pair like that.
if your approximation isn’t very good - that is, your predictions go awry fairly often, - then maybe you’re just the first to spout one.
(protip: it’s hard to refute nonsense if there’s enough of it at once. however, if your audience is not captive, they will tend to have left.)
