CERTAINTY REQUIRED FOR ACTION. 19- 



stract^ or applied) science certainty is neither fre- 

 quent nor necessary. But in philosophy, which is 

 the science of life, we require from our theory 

 practical certainty in addition to its theoretic probab- 

 ility, and as we must act, we must act often on 

 very slight probabilities. While science, therefore, 

 must remain conscious of all sorts of improbable 

 and barely possible theories, seeing that they may 

 suggest fruitful experiments and so enlarge the 

 bounds of knowledge, philosophy, when it has once 

 decided on the right solution, must sternly and 

 rigorously put aside all its rivals, even though its 

 choice was originally arrived at by a very slender 

 preponderance. It must act and act without waver- 

 ing and without hesitation, as soon as its initial 

 inquiry has been concluded, nor allow itself to be 

 easily dismayed by difficulties or deterred from 

 following Its principles to their consistent conclus- 

 ions. Philosophy, at all events, cannot serve both 

 God and Mammon. Any inconsistency and any 

 hesitation is bound to be false, whatever theory of 

 life is true. Such a thing, therefore, as a provisional 

 theory of life would be absurd. How different Is 

 the course of merely theoretic science : upon all 

 disputable points, it may, nay must, keep any number 

 of provisional hypotheses before Its eyes, and must 

 be slow to decide in favour of one or the other ; it 

 must be for ever doubting and testing, and, if con- 

 venient, may even adopt conflicting explanations In 

 different branches of Its Inquiries, and trust to fresh 



^ Such as e.g. geometry. As its subject-matter is ideal Space 

 and not the Real at all, all its assertions must be certain and 

 necessary. But the necessity of mathematics is simply an ex- 

 ample of the necessity possessed by all thought as thought [cf. 



ch. iii. § 15]. 



