Explanation based on Bernoulli effect requires longer path of air taking on top than on the bottom of the airfoil to create speed/pressure difference. With symmetrical airfoil both paths are the same regardless of the angle of attack. So when you mention AoA you implicitly lead to the explanation that lift, in majority, is not based on the Bernoulli effect.

I've read excellent article debunking the Bernoulli effect and lift many years ago, I'm not sure I can find it again...

Explanations based on the Bernoulli effect are trying to explain a speed differential by pretending that two particles that were separated on the leading-edge of an airfoil, to then travel one above the airfoil, one below, would then rejoin at the trailing edge of the airfoil. And so, if you were to change the upper-camber of the airfoil, the flow on the upper part would need to accelerate to be able to join the trailing edge at the same time. And that would create a lower pressure, therefore lift.

The nonsensical part of this model is that a particle on an upper streamline has anything to do with a particle on a lower streamline and that it is trying to keep up with it. Not so of course.

But the lift created by a pressure difference due to a locally faster flow still holds.

> So when you mention AoA you implicitly lead to the explanation that lift, in majority, is not based on the Bernoulli effect.

For a NACA 0012, you'll need an AoA, to have a faster flow on the upper part of your airfoil, as it it symmetric. Other airfoils are perfectly fine creating lift at 0 AoA.