This is the second fundamental law of thermodynamics,[2] the law of the degradation of energy. For energy has a sort of gravitation of its own. It always wants to run down hill. Heat seeks its level as well as water. If we lay a hot plate, say, a temperature of 100°, on or under a plate at zero, the heat will spread to the cold plate until both are at 50°, disregarding radiation losses. And when they have come to the same temperature, it is impossible to get out of them any further heat movement. "You cannot run the mill with the water that's gone by." You have to have a fall of temperature to run any kind of heat engine. Every machine, every chemical and physical process, every living being, is leaking energy all the time, that is, transforming it into unavailable forms. That is the way we get our living. The sun is dissipating its heat energy throughout space at a great rate. Our allies, the plants, manage to catch a tiny bit of it and store it in starch and oil, but we eat these and send the energy on its way as heat again. The whole universe, regarded as a big machine, is running down like a clock and, it seems, must ultimately come to a stop, unless, indeed, there is a self-winding attachment hidden away in it somewhere, or somebody outside of it all to wind it up occasionally.
This, however, is one of those questions which Ostwald calls "pseudo-problems" and from which he would free us by applying the energetic philosophy. His test is the following: "Suppose the problem solved and assume any one of all possible answers to be correct, we can then investigate what effect this would have on our conduct. If it produces no effect, the problem is thereby indicated to be a pseudo-problem." He takes for example the following:
Did the world have a beginning in time or has it existed from all eternity? By the way of experiment we will assume that it has existed since eternity, and will ask what would change in our conduct by this knowledge? I find, at least for myself, that nothing would change by this knowledge, and just as little if we assume that there was a beginning in time. Hence I must say that even if I positively learn in some way which of the two possibilities is correct, it would be a matter of perfect indifference to me, and this being the case, we have here a pseudo-problem. The significance of this procedure is apparent from the answer to the question as to what we call "correct" or "true." The answer was that which enables us to make accurate predictions. Something that does not allow us to make any prediction whatever is essentially of no interest to us in any way, and there is no need of being concerned about it.--"The Modern Theory of Energetics" (_Monist_, 1907).