As I pointed out in the article on “Reference Systems” published in the Winter 1977-78 issue of Reciprocity, the representation of the physical universe in a three-dimensional spatial coordinate system is not fully in agreement with reality. This system cannot represent some of the properties that do exist, such as motion in time. whereas it portrays some properties of the universe that actually do not exist, such as the directions of scalar motions.
Gravitation, which is purely a scalar motion, simply reduces the scalar magnitude of the distance between the gravitating object A and any space-time location B. There is nothing in this phenomenon itself that gives it a direction in the context of the fixed reference system; that is, nature does not distinguish between a scalar motion of A relative to B, and a similar motion of B relative to A. It follows that when this motion is viewed in the context of the fixed reference system, where the motion must have a direction, this direction is imputed by chance. The motion as seen in the reference system will therefore be divided equally between motion of A toward B and motion of B toward A, even where, as in the case of gravitation action on radiation, all motion originates at the gravitating object A. This issue does not arise where A and B are both masses, as in that case there is a symmetrical distribution of the motion but it has a bearing on any case where the motion is asymmetrical.
The same effect can be seen in the induction of electric charges. The motion due to the charge, like the gravitational motion, is scalar, and even though it originates with object X, the motion as seen in the fixed reference system is divided equally between motion of X toward (or away from) Y and a similar motion of Y relative to X. The vibrational motion of X then becomes vibration of both X and Y.