220 THE LIMITATIONS OF SCIENCE 



as subjective, and as debatable as the postulates of 

 religion and philosophy. And they also lack the sim- 

 plicity and generality which have made the postulates 

 of geometry universally acceptable and have given to 

 them the appearance of finality. And from these meta- 

 physical postulates of natural science, the scientific 

 conclusions follow which somehow impress us as being 

 positive knowledge not subject to human opinion. 



Since most of our scientific postulates and the 

 hypotheses which are developed from them are not 

 questions of objective facts, they must have an influ- 

 ence which may be helpful, harmful, or indifferent. 

 And it should be the chief duty of the historian and 

 the critic to trace this influence. Instead of this 

 critical spirit, it is only too common for men of science 

 to insist that all hypotheses are meritorious. This 

 opinion is so widespread, that it is frequently stated 

 that even if a hypothesis be entirely false, yet it may 

 be useful. The reason for this lack of discrimination 

 lies probably in the fact that science has gradually 

 created two worlds connected by only a slender thread ; 

 one the experimental world of our sensations, and the 

 other a fictitious world of our imagination. What- 

 ever the prevailing hypothesis may be, the experi- 

 mentalist continues his work, confident in the belief 

 that if his results agree with a hypothesis they will be 

 taken as a confirmation of it; and if they disagree, then 

 he may be sure that the hypothesis will be modified so 



