That makes me very, very curious.
Does universal expansion imply increasing distances between every atom in the universe, or only between, say, galaxies? I can't see how the second would work, but the first would be ridiculous as well - increased distances between atoms covalently bonded together? Shouldn't the chemical properties of compounds change as bond distances change? Or would the necessary bond distances for a reaction change at the same pace as the universe is expanding?
I originally arrived at the line of thinking via the measurement question - is anything in the universe at an unchanging distance from anything else? (According to expansion theories, anwyay.) If that's the case, would it not be possible to measure the wavelength of the light reaching relatively-fixed object A from its counterpart object B? If the red shift still occurs, then we know time is slowing down. If not, then expansion wins.
Expected problems for this measurement would include scale (if chemical bonds are the only distances in the universe that don't change, how do you measure the wavelength from one atom to the next?), and more scale (would the change be measurable on a scale smaller than intergalactic?).
Oh, and the original assumptions. Those might be a problem, too. (That something (anything) is fixed, and that my mental model is not orders of magnitude oversimplified).
You might not need two fixed point objects. An equivalent approach may be to find the approximate average of movement with a high degree of measurement over all objects and map this out. If the total movement is greater than zero, then everything would be expanding, less and things would be moving closer together, and at zero matter would be static.
Check out the Friedmann equations for average density of the universe.
The expansion of the universe is causing things which are not gravitationally (or stronger) bound to get further apart. So, your later scenario is closer to true, except that some galaxies are gravitationally bound to each other. For example, Andromeda is bound to our Milky Way and will get closer for the next 2½ billion years — resulting in a collision.
The AstronomyCast podcast (http://www.astronomycast.com/) has a fairly good explanation of the expansion of the universe in several episodes. Don't recall which episode, but I'd try ep. 5 and 6 (on the Big Bang) or 11 (Dark Energy). Actually, I'd recommend the whole show.
I originally arrived at the line of thinking via the measurement question - is anything in the universe at an unchanging distance from anything else? (According to expansion theories, anwyay.) If that's the case, would it not be possible to measure the wavelength of the light reaching relatively-fixed object A from its counterpart object B? If the red shift still occurs, then we know time is slowing down. If not, then expansion wins.
Expected problems for this measurement would include scale (if chemical bonds are the only distances in the universe that don't change, how do you measure the wavelength from one atom to the next?), and more scale (would the change be measurable on a scale smaller than intergalactic?). Oh, and the original assumptions. Those might be a problem, too. (That something (anything) is fixed, and that my mental model is not orders of magnitude oversimplified).