Windtunnel tests show otherwise. Not the best photo, but it gets the point across. Photo Link. Also Youtube Link. As far as wings curved, to prevent turbulence is one of the reasons. Keeping the flow laminar vice turbulent is key to achieving a high L/D ratio, essentially giving good fuel economy. This was key for the P-51 airfoil and to the success of that aircraft, amongst other design choices. Furthermore, sometimes turbulence will help delay the onset of the stall by delaying airstream seperation, see vortex generators on top of wings, mostly in small aircraft looking for extremely low takeoff and landing speeds. Wings also have camber to generate more lift at a lower AoA, preventing the wing from needed excessive AoA at lower airspeeds to generate the same amount of lift. I believe it also improves L/D ratio, but I'd have to look up a few airfoils in my trusty Abbott and Von Doenhoff 'Theory of Wing Sections' to be sure. Symmetrical airfoils work just fine; mostly used by aerobatic aircraft since they spend quite a bit of time in inverted flight compared to most aircraft, something that positively cambered airfoils can do, just not do as well as symmetrical airfoils. While yes, Newtons Third Law basically tells us that air getting shoved downwards creates a force, and thus the reaction force is the wing getting shoved upwards is valid, especially for hovering helicopter theory, there's more to that to an airfoil. There's also Bernoulli's principle, where fluids that experience an increase in velocity results in a decrease in pressure. If you put static ports along the length, top and bottom, of an airfoil, and sum the static pressures (using the vector perpendicular to the port on the airfoil), you can accurate measure the total lift of an airfoil.