Again, not a flaw in the reasoning - it's a language shortcoming.

You have two options:
A) Either the hub is hanging from spokes who don't show a significant increase in tension between loaded and unloaded axle.
b) or the hub is "standing" on the reduced pre-tension of a tensile-only construction element.

Your choice. Clumsy as it is I prefer option b.

Measuring spoke tensions with a Park tensiometer on a bike sitting on the ground loaded with nothing but itself I can't determine by measurements alone which is the bottom spoke, that gets lost in the noise. Enlisting my brother to sit on the bike I can measure a decrease in bottom spoke tension, but any increase in the other spokes pretty much gets lost beyond the resolution and repeatibility of the tensiometer.

So there you have it - Of course the wheel isn't standing on the spokes, in the manner we usually think of as "standing".
But given a reasonably decent measuring device, the only change between loaded and unloaded axle is a reduction of the tension previously registered to the spokes that occupy the space in a rather narrow sector between the hub and the ground.

Better yet - ignore the whole issue of "standing" or "hanging". Every spoke tries to pull the hub towards the rim, and the consequences of an axle load is that the spoke that pass between the hub and the ground for a short moment will pull less.