Then it comes down to what is meant by "vacuum." It might be necessary to add a term to the equations and that will disrupt their apparent canonical simplicity. That's the way it goes.
I think it goes even deeper than that. The speed of light is assumed to be constant on the basis that any observer anywhere who measures electrical and magnetic properties of a vacuum will get the same numbers. E.g., the electric field due to a unit of electrical charge is the same here as it is halfway across the universe.
If those properties can change from place to place, then the speed of light can likewise change, and an observer can detect his motion through a vacuum by monitoring its value. Who's to say they don't change?