Eckard,
I'm a little surprised that nobody rose to your bait in debate on the argument of Cusanas, neutral centrality, Maxwell and the always ambiguous 'luminiferous aether'. All good points, especially with the number of unsolved mysteries of the classical age that have been subsumed by Quantum Mechanics.
I dug out an old (falling apart at the seams, actually) book by Isaac Assimov, still one of the most readable of introductory authors; and refreshed on his account of the state of the art in the mid to late 1800"s. The wave theory of light was prevalent but had many problems. While detected effects displayed a transverse wave signature, transverse waves were known to be limited to being conducted through solids or along liquid surfaces, while longitudinal waves could be conducted by materials in any state, whether solid, liquid or gaseous. So for energy to be transfered via a transverse wave required that an 'aether' be an extremely tenuous gaseous substance yet have a rigidity greater than steel. So, in todays' theoretical venue it is quite appropriate to ask if it requires a medium to transfer energy. And if I may make a pun; 'the medium is the message', that is to say that a quantity of inertially bound medium (energy) can be the discrete carrier of energy transfer. Though that itself raises the ontological question of Time and Space, just as does QM.
The problem with Time and Space being emergent from simple change of positions, conveniently ignores that there is no reference of distance traveled by any nondescript particle in that change of position. Yet QM arbitrarily conducts that analysis in a preconditioned co-ordinate system. And simply arguing causality only adds to that contradiction. So by default, the advantage goes to Time and Space being physically, if not materially, real.
Personally, I subscribe to a hypothesis that density varies in direct inverse proportion to velocity, acting progressively upon and from an upper density bound. This argues for a quality of Time being uniform, at least in the confines of a constant density region of an inertially bound quantity of energy. After all, if it is the same moment in time at every point in a constant density region, then there is no problem if that rest mass is truly at rest. But the least movement will require the density to relate to different points within its constant density boundary. Very much like putting a finger on a flake of cigarette ash on a page of paper, and smudging it with a swipe of the finger. We just don't know how fast time is going at one second per second, we can only calibrate it to our own reckoning of 'one second'.
So I have no problem with Time having qualities of both Classical and Relativistic properties. It is we whom complicate what would be simple for nature, by deconstructing it in any attempt to analyse. In reality, nature would not geometerix space into a cube or a sphere, but we can analyse the contrary properties of Space as a cube being the most efficient filler of space, like a Cartesian referrence system; and a sphere being the most efficient encapsulation of space. Nature would be doing both in one fell swoop. So Space would need Time, neither would exist alnoe.