Ok, why is the speed of light a constant? The answer is that we say that light travels at constant speed because it’s an observation that works with all the currently excepted theories.
We are stuck saying that light travels at a constant speed because that’s an observation that fits in a coherent theory with all sorts of other detailed observations of diverse phenomena.
I am slowly coming to the point, which is hard to take in emotionally, that that’s all the justification we have for any scientific theory.
Your distinction between "description" and "explanation" is interesting. It seems like "explanation" is very much what you describe in your second sentence: describing how an observation is a consequence of a broader array of forces from which other, diverse phenomena also emerge.
So you can explain a theory in much the same way by embedding it in a yet broader basis that explains multiple theories. For example, the theory of magnetism was long ago explained as a special case of electromagnetism, and now of the electroweak force. In the same way, gravity could be explained by quantum field theory or string theory, but more work needs to be done to do so.
Your feeling of disappointment might come from the issue of "intuition," which can make explanations feel satisfying.
Many explanations for everyday things feel intuitive; they "make sense" because they fit patterns that we're familiar with. Many physics theories are deeply unintuitive unless you spend time with the detailed math. That doesn't mean the explanations aren't explanations; it just means that they don't feel satisfying (without intense studying).
Eistein's original formulation of SR required two postulates: that the laws of physics are indistinguishable between any two inertial reference frames, and that the speed of light is constant in all inertial reference frames. Both of these are experimentally motivated.
However, that's not the only way to obtain SR. You can, for example, assume Maxwell's equations (or any of the higher-level electrodynamic field formulations) and postulate #1, and obtain the invariant speed of light as the only valid solution for self-propagating EM waves in free space. I rather prefer this formulation, as the EM field formulation feels more "fundamental" than a fixed speed of light.
You can also derive SR from postulate #1 and some properties of space: isotropy, homogeneity, and memorylessness.
You've got two problems, though:
1) We don't yet have a consistent explanation for what gravity is. General Relativity is good, but breaks down at small distances.
2) At some point, physics becomes math, and at some point after that there are no explanations. For example, why does 1=1? Basically, the answer is "by definition".
It's a perfectly reasonable question to ask. As we learn more about science we are able to unravel layers of mystery from physical phenomena and explain observations that were previously taken to be axioms.
Although the statement "mass curves spacetime" is true, I don't think it is simple enough to be considered an axiom. We'll have to dig deeper. Who knows, maybe the OP's son will do it.
I believe speed of light in vacuum is invariant because universe has maximum speed and light can't go faster. If it wasn't invariant then in some cases it would have to.
Then again why universe has speed limit? Does it have anything to do with causality?
Any scientist who can't explain to an eight-year old what he is doing is a charlatan.
—Kurt Vonnegut, Cat's Cradle
Because the hyperbolic metric Einstein created allows us to define how much space-time can change the amount of mass(and vice versa!). And this only works if the speed of light is constant—that's actually the God-like conceptual leap Einstein took when realizing the Michelson-Morley experiment implied much bigger things about space, time, mass and gravity, and not just mere light speed.
Also, it keeps several other super-important constraints in place too. That's the problem with gravity—it's more like an infinite Jenga puzzle and anything you add to the explanation must not contradict dozens of other old physics experiments about things having nothing to do with gravity. We can't throw out all the other stuff we do know after all. Like the constancy of the speed of light in any reference frame. And the conservation of energy.
And Maxwell's electrodynamic equations. And a bunch of other specific experiments I don't even know much about.
I don't know how much of thermodynamics has been "relativized", but gravity can't contradict the laws of thermodynamics either. It's actually comforting that the world of thermodynamics is so tiny, cold, deals with such a large number of "particles"(ensembles), and is built from insanely complex mathematical machinery that we probably have no hope of ever devising an experiment that could say anything meaningful about gravity in terms of thermodynamics. (I so hope that I'm very wrong here.)
However, Gravity is famously in contradiction with Quantum Mechanics, though I haven't yet progressed far enough in either to be able to say specifically what the contradiction is. I suspect it's just the lack of a math "bridge" between Relativity, which is built up from calculus, differential forms, tensors, Rienmannian geometry, the calculus of variations and weird ways QM uses linear algebra, eigenvalues, infinite bounded operators and insane algebraic concepts like von Neumann algebras to setup infinite matrices of probabilities that map back to the periodic, bounded solutions of the Schrödinger wave equation that tell us what the energy levels of some physics experiment must be.
Oh and the model of gravity has to be able to fit in well with any new stuff we discover too. Being able to pull off all of that is why I place Einstein in the pantheon of geniuses! He's kind of like a one-armed man who holds off an entire army of swordsmen using only a pocket knife.