Theory of Flight


A Physical Description of Flight ©

David Anderson

Fermi National Accelerator Laboratory


Scott Eberhardt

Dept. of Aeronautics and Astronautics University of Washington
Seattle WA 91895-2400


To be found at;


Theory of Flight


In The Daily Telegraph, 3may01, p3, Robert Uhlig refers to a report by Dr. David Anderson in New Scientist, which I have not seen.  However, I have read the report at


Having been a technician in the RAF, and having done an engineering degree at Cambridge, I felt very foolish at having accepted the conventional (Bernoulli) theory of flight. Just by reading the brief report in The Daily Telegraph on 3may01, the conventional theory collapsed.


Three points arise.



The report “A Physical Description of Flight” over-gilds the lily, and obscures the key points which were clearer in The Daily Telegraph, although even there, they were not as clear as they could have been. The key points are;


  1. Lift is caused by imparting downwards momentum on air
  2. Drag is to be minimised in achieving process (a).



An opportunity is given to us to contemplate the two versions of Newton’s Second Law of Motion.


  1. F = ma is an inappropriate version of Newton’s Second Law.
  2. F = d(mv)/dt describes the basic process very well for our purpose.

Here we can compare and contrast two apparently identical mathematical statements of the law, and we realise that a mathematical statement is not what it appears. Otherwise, the two versions would be equally appropriate, which they are not.

In order to create lift for the aircraft, downwards momentum has to be imparted on some air. Acceleration is not a relevant factor.

I have for many years wondered about the lack of interest in the two versions of Newton’s Law. This discussion gives us much food for thought.




I had to study fluid dynamics to some degree at college, and notice that the Bernoulli theory is far from the conventional approach in that field, which is to take a “hands off”, distant, macroscopic approach on the grounds that fluid flow in detail is not well understood. I now realise that the Bernoulli approach flies in the face of that general principle in fluid mechanics.


Ivor Catt    25apr02



I shall restate 3 above. When I studied fluid mechanics in Cambridge in 1957, the reigning approach was to say that we did not know enough about the detail, so we should take a macroscopic approach. Thus, in the case of designing the shape of a boat’s hull, we would experiment on a model and separate out drag resulting from friction from drag resulting from turbulence. No one noticed that in the case of theory of flight, this approach was being flouted. Bernoulli is a microscopic approach, dealing with the local pressure at the surface of each part of the wing. In 3 above, in the tradition of the discipline “fluid mechanics”, I take the macroscopic view using Newton’s Second Law of Motion, distancing myself from the minutiae of the situation. The overall effect is lift, drag, and downwards momentum delivered to the air (plus the initial problem, which is the weight of the aircraft). Approach closer to the wing will result in confusion and unnecessary complexity. Now, what we have to do is deliver lift to the wing by giving downwards momentum to the passing air. In the process, we must try to minimise drag, and so minimise the required engine thrust.


I was amused by a friend, who reported that his pilot friend said that in order to fly upside down, the wing shape had to be changed by putting out the flaps. This was the pilot’s effort in extremis to save the Bernoulli theory. It took me some time to realise that under that theory, a plane could not fly the right way up with its flaps out.

Ivor Catt   24feb03


The extraordinarily shallow thinking that prevails in this context, even when billions of dollars are at stake, is demonstrated by the ICBM. In this case, the error goes right back to the Germans’ choice of Peenemunde, in a flat area near the sea.

Work done = force x distance.

Power = force x velocity.

In the case of a rocket, the objective is to deliver velocity to that part of the rocket that remains after the propelling fuel has been lost. The error seems to have been to think that work done = force x time. In fact, delivering a force to a stationary rocket is of no help, as we can see in the hours before launch, when the earth below the rocket delivered just that, and the rocket remained stationary. When the propellant is ignited below a stationary rocket, it largely takes over the job of the ground, at great expense.

The error in choosing Peenemunde was repeated when flat Florida was chosen.

Consider a rocket just after lift-off. With vertical lift-off, massive amounts of fuel are expended in holding the rocket more or less stationary just off the ground. This is what should have been avoided at all costs. Thus, the use of fuel to increase the velocity of a rocket should only be used when the rocket already has velocity. The launch of a rocket should take place in a mountainous area, and gravity used to give the rocket as much velocity as possible before take-off. This involves having the rocket descend from a high point down guiding rails, and then having its direction reversed by a curve in the rails, so that it leaves the ground at high speed in a vertical direction. Only thereafter should rocket fuel be brought into play, to be used more efficiently on a rocket which already has significant forward velocity.

Recap. What gives additional speed to a rocket is not force (provided by the rocket fuel), but power, which is force x velocity. Thus, force should only be applied after the rocket has achieved velocity, I suggest by the use of gravity.

Although it is possible that Florida and Texas were chosen for the US space programme because they were politically more powerful than the mountainous states, I suggest that their choice resulted from lack or thought rather than their greater political power.

Ivor Catt    10feb03