Disclaimer: I am not a physicist (nor did I stay at a Holiday Inn Express last night,) though I do have a degree in engineering and am generally a physical sciences geek.
The only way Ringer's original diagram could be correct, with the initial path of the ball perfectly perpendicular to the club face, would be if the club face was a frictionless surface, which of course it isn't. The diagram
assumes
that the initial direction is determined solely by the club face angle.
The initial flight of the golf ball at impact is a question of forces. There are forces perpendicular to the club face, but there are also forces parallel to the club face (the ones that make the ball spin.) Both contribute to the direction in which the ball travels.
Attachment 512
This diagram is going to take some explaining.
In figure A obviously the circle represents the ball and the triangular wedge is my representation of a golf club. It doesn't matter if your looking from the side or from the top. (Yes, physics really does work the same in all directions.) The arrow inside the wedge indicates the direction of travel, i.e. the swing path.
At impact the club face imparts force to the ball, which for the sake of analysis, I have broken down into forces normal and tangential to the club face (that's perpendicular and parallel to most people.) Vector 1 represents the normal force and vector 2 represents the tangential force.
In figure B I have removed the "club", leaving only the forces it imparts to the ball. I have also introduced two new vectors. These don't represent forces
per se/I], they're just conceptual fictions. Vectors 3 and 4 are equal in magnitude to the tangential vector (#2), but they point in opposite directions and run through the center of gravity of the ball. They could be forces, but they'd just cancel each other out.
In figure C I have combined vectors 1 and 4 to produce a single translational force acting on the center of gravity of the ball. I have also combined vectors 2 and 3 into a rotational torque on the ball. I can do this because they are of equal magnitude but opposite in direction with an offset in their lines of action.
The initial direction of flight will be in the direction of vector T, while the spin of the ball will be determined my the magnitude of the torque R.
Diagrams D, E and F show the same analysis but with a different swing path. Note the change in initial direction, despite the identical club face angle.
All of this assumes of course that the ball is a rigid body (which in the context of impact, it isn't) and that it is also a free body (which if it's sitting on the ground and we're swinging down on it, it isn't.) The exact proportions of the normal and tangential forces imparted to the ball will depend on club head speed, compression of the ball, friction between the ball and the club face, etc.
Conclusion: For the time being, I'm siding with the "swing path" people. The initial path of the ball is determined to some extent by both the swing path and the face angle, in combination with the surface characteristics of the club and ball. For the most part we're talking about deviations in swing path and face angle of a few degrees from the target line. In that range, I think the swing path is going to be the primary determinant, particularly with lesser lofted clubs.
I know there are at least a couple of books on the physics of golf,
The Physics of Golf
by Theodore Joergensen and
Newton on the Tee
by John Zumerchik. Has anyone read either of these, and if so, do they have anything to say about what goes on at impact?