That's...false? Salt gets spread on existing ice to melt it all the time. Where does the absorbed water go in this model? This assertion is so confusing.
As someone who grew up in a cold climate, the way I recall it is that the salt lowers the freezing point of water by forming a brine, allowing it to run off the road and/or be evaporated by the sun.
Sure, brine has a freezing point.
> a great deal of the benefit of salt comes from friction as cars drive over it, crushing pieces of ice and forming that slush we all know and love, even at temperatures where it won't actually melt the ice.
What do you think slush is? It has liquid in it, and that's not molten salt.
You may have never tossed a handful of rock salt on steps that have a quarter-inch of ice on them and watch holes actually form in the ice within a half hour or so.
I think the person is just saying that pre-emptively salting the road (before snow/freezing rain actually lands) provides some additional level of safety because the road might spend less time with a frozen coating (as opposed to salting only after the surface is frozen over).