A few more thoughts.
How much lift is created by the upper surface and how much by the lower one?
It varies, as suggested by others. When I find and post a better manometer bank video showing the pressure variations with angle of attack, you will get a good intuitive picture of how the pressures (and lift) vary with wing alpha. Unfortunately, while we olde phartes grew up with wind tunnels hooked up to such banks (which were great for visualisation), these days we have you beaut electronic gadgets which measure the pressures directly. Great for engineering calculations but not so great for visualisation and intuitive understanding.
Megan's initial graphic is a direct picture of a snapshot of the manometer bank story at a particular alpha value in the wind tunnel and you will see lots of similar pictures in older texts. Keep in mind that it is talking about gauge pressure which is the difference between actual pressure and the background ambient pressure. So when we are talking about a "negative" pressure, we are talking about pressures a bit below ambient. When we are talking about "positive" pressures, we are talking about pressures a bit above ambient. His second graphic is how the first usually is shown in modern texts and is more useful for the engineering folk. The first (ie a time history of the manometer display) probably is better for the student pilot's understanding of what is going on ...
At flight school we were told that 2/3rds of a wing's lift came from the top surface 'sucking' the wing upwards - owing to the air speeding up over the top lowering its pressure relative to the air underneath the wing.
Proportions will depend on alpha and the particular wing section profile. In general, the "top" surface (ie whichever side is pointing to the heavens) will carry the greater burden on pressure delta from ambient. Better to think in terms of both surfaces providing pressure stuff and the forces which arise from those pressures. What is left when you average these out by doing some sums results in the net lift left over which is the useful bit for keeping the aircraft up, rather than going down.
It is very important to keep in mind that we can't generate forces directly in fluids - fine with solids - but doesn't work with fluids. For example, if you punch a solid brick wall, hard, it hurts real bad. The brick wall is a solid and you certainly can generate forces directly by interacting with (punching) it. However, if you try to punch a fluid, say, the water in a large bucket, or the air around you, the fluid just moves out of the way and doesn't do much anything. For fluids, to generate forces, you have to have pressure differences or gradients which then can apply forces to the bulk of the fluid - much the same level of difficulty as herding cats.
You can look at this from both sides of the table. If you see the effect of a force in a fluid, eg the fluid accelerates, turns or whatever, then there MUST be a pressure gradient somewhere there in the mix which is generating the force which, in turn, is causing that acceleration, turning, or whatever it is you are watching happen. This latter consideration is a very important part of explaining what air is doing as it meanders around an aerofoil. Because it is generally so poorly explained, most students are left confused with a thought that it is all a bit of magic going on. Not at all - for the speeds we are dealing with, it can be explained in terms of Newtonian mechanics - that discipline has stood the test of time for quite a while so it probably is pretty much OK for figuring out physical things. The only "difficulty" arises because so many people forget (or never knew) that we have to think pressure gradients, rather than directly in terms of forces, to figure out what is going on in fluids.
But when looking at a propellor in flight school
I was told oh no, a propellor works by pushing air behind it
Typical lack of understanding in the flight instructor fraternity. The propeller is much the same as a wing (hence, "fling wing" for helicopters - just ask Megan about that - apart from being a nice bloke, he has spent most of his adult life assisting helicopters to repel the earth in various rotary operations). Think more in terms that the propeller is accelerating a mass of air from one side to the other. Just like with a wing, it is not one side or the other doing the work - it is the mean, or net, of both sides each doing their bit to help. Again, Professor Newton had the basics of the story for the lower speed environment (after all, he was a Professor of Mathematics at the Univ of Cambridge).
And more recently I have seen it asserted that an aircraft flies by its wings pushing down the same weight of air as the aircraft weighs, and certainly a helicopter's blades do push air down.
That all sounds about right. Prof Newton, once again.
Tdracer, being another one of us engineers, tells the same sort of story.
Andycba's links are useful stuff for reading.
At the end of the day, Coanda really was talking about other stuff. His observation about fluid jets following surface profiles is fine and can be extended to the distributed flow over a lifting surface. But you can, probably more easily and with less appeal to black magic, explain that in terms of Newtonian mechanics, pressure gradients and the resulting forces generated on fluid bodies, allied with a consideration of boundary layer pressure gradient degradation leading to flow separation from the surface.
Bernoulli's theorem. Yep, works fine for the constraints inherent in the theorem, which are then conveniently ignored in the typical (wide of the mark) aviation student textbook. This is just a statement of conservation of energy and is very useful for figuring out the sums relating pressures and speeds. Has very little (directly) to do with lift stuff.
The engineering and physics story is that of the circulation model. This started off back in the mid-1800s, and was sorted out for wings independently by Lanchester (early 1900s in England) and Prandtl (war years in Germany). Kutta and Jukowski, again independently, fixed up a problem which caused some academic head scratching. The resulting model has been used over the past century to figure out values for lift by the aerodynamics fraternity. This model is able to be demonstrated experimentally and gives results which are so close to what is measured that the difference isn't really worth worrying about. Trailing vortices are a part of this story.
For me, I think the easiest way to explain lift is to start with the circulation model and bring in other stuff as necessary to fluff out the story as required along the way for the student .... Certainly, my theory students don't seem to have all that much difficulty getting their heads around the stuff.