A.T. said:
If the choice of reference pressure is so arbitrary, then so is the claim about which side is most responsible for the net force.
The two things have nothing to do with each other. No matter what you use as a reference, you can observe and quantify which side changes more -- it isn't affected by the choice of reference. That's what "arbitrary" means!
Let's say you have a parcel of still air. Call the pressure 14.7 psi. Call it zero. Call it -47. It doesn't matter: when a wing flies through it, the pressure of the air above the wing goes down and the pressure of the air below the wing goes up, both can be measured or calculated separately and the change in the upper surface is usually greater than the change for the lower surface.
Arbitrarily, let's plug in some numbers:
Let's call our reference pressure -47 psi.
Above the wing, the pressure is measured at -49 and below the wing, the pressure is measured at -46. So the pressure above the wing went down by 2psi and the pressure below it went up by 1psi. So the upper surface contributed 2/3 of the lift.
No matter what you pick for the reference, the answer remains the same.
Something else to consider here is that the wing itself is probably not air tight and certainly doesn't contain a vacuum. The completely separate aerodynamic forces on the top and bottom surfaces need to be considered separately when designing the structure of the wing. IE: even though you might think there is a pressure of 14psi pushing down on the top surface of the wing, the rivets holding the top surface on the wing are under tension, not compression.