84 THE GREEK SCHOOL PHILOSOPHY. 



any art or science whatever ; we must collect the facts, and the things 

 to which the facts happen, in each subject, and provide as large a sup- 

 ply of these as possible." He then proceeds to say that " we are not 

 to look at once at all this collected mass, but to consider small and 

 definite portions" ..." And thus it is the office of observation to 

 supply principles in each subject ; for instance, astronomical observa- 

 tion supplies the principles of astronomical science. For the phenom- 

 ena being properly assumed, the astronomical demonstrations were 

 from these discovered. And the same applies to every art and 

 science. So that if we take the facts (ra virdp%ovra) belonging to 

 each subject, it is our task to mark out clearly the course of the 

 demonstrations. For if in our natural history (Kara rfjv loroptav] we 

 have omitted nothing of the facts and properties which belong to the 

 subject, we shall learn what we can demonstrate and what we cannot." 

 These facts, ra vrrdpxovra, he, at other times, includes in the term 

 sensation. Thus, he says, 12 " It is obvious that if any sensation is 

 wanting, there must be also some knowledge wanting which we are 

 thus prevented from having, since we arrive at knowledge either by 

 induction or by demonstration. Demonstration proceeds from univer- 

 sal propositions, Induction from particulars. But we cannot have 

 universal theoretical propositions except from induction ; and we can- 

 not make inductions without havin^ 1 sensation : for sensation has to do 



O ' 



with particulars." 



In another place,' 3 after stating that principles must be prior to, and 

 better known than conclusions, he distinguishes such principles into 

 absolutely prior, and prior relative to us: "The prior principles, rela- 

 tive to us, are those which are nearer to the sensation ; but the princi- 

 ples absolutely prior are those which are more remote from the sensa- 

 tion. The most general principles are the more remote, the more par- 

 ticular are nearer. The general principles which are necessary to 

 knowledge are axioms." 



We may add to these passages, that in which he gives an account 

 of the way in which Leucippus was led to the doctrine of atoms. 

 After describing the opinions of some earlier philosophers, he says, 14 

 " Thus, proceeding in violation of sensation, and disregarding it, be- 

 cause, as they held, they must follow reason, some came to the conclu- 

 sion that the universe was one, and infinite, and at rest. As it 

 appeared, however, that though this ought to be by reasoning, it 



12 Anal. Post. i. 13. 13 Ib. i. 2. 1J DC Gen. et Cor. i. 8. 



