mechanics, we are at once puzzled by the fact that most mechanical theories make assumptions about the forces at work in the world, and that all of them predict coming facts. But forces form no part of the experience or of the mere description of motion. And the future is not yet given to be described. How then does all this agree with the definition in question? Very well indeed. For those who assume forces to explain given motions, always assume just those forces that will directly explain, not any description at random of the motions given in experience, but the simplest possible description. Any motion being relative, never for our experience absolute, we can assume at pleasure any point in the world as the origin or point of reference that shall be regarded as at rest, and so we can get an infinite number of descriptions of any given motions. We can make any object in the world move at any desired speed or in any desired direction, simply by altering the origin to which we shall choose to refer its motion in our description thereof. But all these possible descriptions are not equally useful for the purposes of the science. Some one of them is the simplest for all the motions of the system in question; and this we regard as best expressing the actual natural truth in the matter. The assumption of just such forces as woidd explain this simplest system of motions as described, satisfies us. We say, these forces are the real ones at work. But still we know that the forces assumed only express in another form the fact that the description in question is the simplest. Is this, however, really all that
