In many textbooks, such as Anderson (Fundamentals of Aerodynamics), Bertin (Aerodynamics for Engineers), and Houghton (Aerodynamics for Engineering Students), the authors present the idea that there is a starting vortex which is formed. This starting vortex and the bound vortex that form (due to Kelvin's Theorem) co-exist with lift (as Mclean puts it in Understanding Aerodynamics). However, I am unsure on why the starting vortex forms on the upper surface. Why does it not form on the lower surface and have clockwise rotation and lead to counter-clockwise bounded circulation at the airfoil?
While I know that this is not experimentally observed nor does it make intuitive sense, I would like to know why the starting vortex does not form on the bottom.
To shed more light, here are some instances of where authors state that the starting vortex starts on the upper surface, but never explain why it doesn't start on the lower one: (taken from Bertin): "At the instant of starting, the flow is a potential flow without circulation, and the streamlines are as shown in Figs. 6.3 a, with a stagnation point occurring on the rear upper surface"
(taken from Understanding Aerodynamics) "It can be shown, based on starting the flow from rest, that in the absence of viscosity, the nonlifting flow pattern of Figure 7.1.3a is the one that would occur... "
So how can it be shown (mathematically) or using strong rigorous theoretical arguments that the stagnation point cannot be at the lower surface before circulation?