PHILOSOPHY OF PHYSICS — MILNE 223 



of certain fallacies of thought ; the philosophy of physics considered 

 as a method is in fact the philosophy of Hume. Again, modern 

 logic, in the works of Whitehead and Russell, is the fearless eradica- 

 tion of things that cannot be observed. The theory of number, both 

 of finite number and of infinite number, is the strict application of 

 such a simple experimental process as that of counting objects. 



It is in the domain of microscopic phenomena that the successes 

 of modern physics have been most conspicuous. I should like now 

 to invite your attention to the application of the methods of physics 

 to the less fashionable domain of macroscopic or ordinary-scale 

 phenomena, and to discuss these in relation to certain large-scale 

 phenomena. It is often said that the concepts of space and time 

 may break clown in atomic phenomena. But it is no use saying 

 this unless we are clear as to what these concepts mean in larger- 

 scale phenomena, indeed until we are certain that they are only 

 concepts. For example, in order to be able to attach a meaning 

 to the uncertainty principle of Heisenberg we have to give form 

 to the notions of position and momentum, of energy and time, which 

 involve both kinematical and dynamical notions; and the principle 

 implies the transfer to atomic phenomena of notions derived from 

 ordinary experience. 



Now, there are two well-defined branches of physical science which 

 concern themselves with the macroscopic phenomena. One is the 

 theory of relativity; the other is thermodynamics. Both of these 

 bring us immediately to time. The one makes little of time ; the other 

 makes much. 



The one, relatively, appears to relegate time to the role of a co- 

 ordinate. It is often represented as claiming to show that time and 

 space, not separately real, are but aspects of a higher reality, " space- 

 time ", which is the true framework of events. The individual is 

 supposed to make his private choice of his resolution of this frame- 

 work into his own separate space and time, but no one choice is 

 to be preferred to any other. An individual observing two separated 

 distant events may describe one as preceding the other, while 

 another ma}'^ describe them as simultaneous. The time ordering of 

 the events thus depends on the particular observer, and has nothing 

 to do with the events themselves. As Jeans points out in the book 

 mentioned, this raises acute difficulties concerning the reality of 

 evolution. He mentions that it has been suggested that "the concept 

 of evolution in time may lose all meaning " so that we cannot speak 

 of the universe evolving as a pattern is woven on a loom. If " time 

 is merely a geometrical direction of our own choice in a continuum ", 

 the pattern is already spread out, and future events have the same 

 kind of reality as past ones. " Indeed, an inhabitant of a nebula 



