In 2D airfoil diagrams, first the parcels approach horizontally. Then they are mysteriously accelerated upwards. Then the airfoil accelerates them downwards. Then behind the airfoil, they're mysteriously accelerated upwards again, so they depart horizontally, for zero net acceleration applied to each parcel.

As a whole, air around the wing ISN'T accelerating downwards.

As long as we ignore the "mysterious" part, or try to sweep it under the rug, our explanations won't satisfy.

A complete explanation must include the ground surface ...and then we discover that we're actually explaining a kind of "venturi effect," where the ground is a part of the system, and those mysterious accelerations before and behind the airfoil are caused by interaction with the ground.

This is a "big AHA!" situation: a 2D airfoil can only explain ground-effect flight, where the Newtonian force-pair is located between ground and the airfoil, so no net downward acceleration of air exists. (In other words, if we erase the ground from our diagram, we've just violated Newton's 3rd law.)

A 2D airfoil diagram without the ground surface ...that's a sort of "inertial drive," where the upward force on the airfoil has no corresponding down-force (it does not give a net downward acceleration to the air as a whole.) Easy to fix. Just add the ground surface back in. And then confront the fact that this is a venturi, and not an explanation of flight.