458 INTRODUCTION TO PHILOSOPHY OF SCIENCE 



Logical Sum of all Relations, like Omniscience, like Beauty 

 Absolute — so, too, Eternity or Fate, Unconditioned Free- 

 dom or Self-determination, Perfect Justice, Universal Har- 

 mony, the Goodness of God, Felicity Divine, and many other 

 supreme ideals and supreme perfections of rational experi- 

 ence and thought, are all of them forms of Being absolute, 

 constituting an Overwork!, a realm Superrational." * 



Unfortunately Keyser has not told us any of the details of 

 this Overworld. To debate its existence is vain. It is not 

 only demanded by the rational, but required also by our 

 aspirations. "In some sense, whatsoever quickens, lures 

 and sustains, exists." 2 As to its nature, we are left more 

 or less completely in the dark. Since it is superrational we 

 are not entitled to ask that it be described. All that we can 

 know is that it differs from the rational in the same way that 

 a perfect lever differs from actual levers, and a circle differs 

 from the series of polygons. But there seems to be no princi- 

 ple according to which the nature of this difference can be 

 ascertained. Hence we cannot conclude, for example, from 

 the nature of human knowledge to the nature of Omniscience. 

 But we can argue that Omniscience need not be the same as 

 finite knowledge, hence the fact that human knowledge im- 

 plies a realm of the unknown within which it operates is in 

 no way incompatible with true Omniscience. Beyond this we 

 cannot go in the penetration of the Overworld. For the 

 rest we must remain satisfied with the fact that it sheds 

 "a mystic radiance like the 'obscure clarity that falls from 

 the stars.'" 3 



The most interesting feature of Keyser's position is the 

 way in which he endeavors to follow out the hints which the 

 realm of obvious data offers of something beyond itself. 

 Phenomenalism insists that the given should be limited 

 strictly to that which is clearly given. But what is one to do 

 with hints? Is the fact that a datum suggests something be- 

 yond itself part of that datum? The way in which a series of 

 elements suggests its own limit is quite compelling. But 



i Ibid., p. 73. 2 Ibid., p. 75. 3 Ibid., p. 74. 



