4 SCIENTIFIC METHOD 



future, as when we deduce a particular eclipse ; in 

 the past, as when we deduce a particular animal from 

 fossil remains : in the present, as when we deduce that 

 this glass of water contains an innumerable number of 

 molecules of oxygen and hydrogen. It is not by in- 

 duction but by deduction that we know definite 

 particulars beyond experience in the future, in the 

 past, and even in the present. 



All inference then begins from, and proceeds beyond, 

 experience. Yet it is common among philosophers to re- 

 duce knowledge to mere experience, and define science as 

 systematic experience. System, yes : the very difference 

 between ordinary and scientific knowledge is that while 

 the former is sporadic, the latter is systematic. As 

 Aristotle said, every science selects a subject, such as 

 number in arithmetic, magnitudes in geometry, investi- 

 gates as many propositions of its subject as possible, 

 and aims at last in bringing the multitude of propositions 

 under a few principles, in order to convert its whole 

 knowledge of the subject into a system. Science then 

 is systematization. But it is not mere systematic ex- 

 perience, but also systematic inference from, and beyond, 

 experience. The science of the moon, for example, 

 must start from its visible face. But, as the moon 

 rotates only once in a revolution round the earth, it 

 shows us only one face, of which alone we have ex- 

 perience. How then do we know that it has another 

 side always turned from us, and in fact that it is a 

 sphere ? Not by experience, but by inference from 

 experience of other spheres. Science then is systematic 

 knowledge by experience and inference. 



Now scientific method is merely the way, or ways, of 

 using different orders of inference in investigating any 

 subject of science with a view to its system. Sometimes, 



