We are all familiar with the upward force experienced by an airplane as it travels down the runway and through the air. But it is not entirely clear if we know the cause of this lift. Let's distinguish three tests for knowing the cause of some phenomenon:
(1) We can bring the phenomenon about with regularity and in a wide variety of circumstances.
(2) We have a scientific model which allows us to predict that the phenomenon will occur in these circumstances.
(3) We have a scientific model which includes accurate representations of the fundamental physical processes responsible for the phenomenon in these circumstances and this model allows us to predict that the phenomenon will occur in these circumstances.
(1) and (2) seem to me to be inadequate unless we adopt a non-standard account of causation. If I understand McCabe correctly, then he is arguing that we lack (3). This is because we believe that particle-to-particle interactions are the fundamental physical processes responsible for the generation of lift, but none of the models that we can work with accurately represent these processes.
As McCabe and Flatow explain, the common explanation in terms of a difference in pressure, known as Bernoulli's principle, fails. What is less clear is how the textbook explanation in terms of circulation relates to particle-to-particle interactions and how tenuous this relation can be consistent with a claim to know the cause.