THE STABILITY OF TRUTH. 651 



tlie calculations of Lord Kelvin and others, that a molecule is 

 as small in a drop of water as a marble in comparison with the 

 earth, then we may not look for these creatures. If we can not 

 find them, we do not know that they exist. If we do not Tcnoiu 

 that they exist, shall we " believe " that they do ? Is it not better, 

 as Emerson suggests, that we should not " pretend to know and 

 believe what we do not really know and believe'' ? 



It may be that the existence of life in a world once lifeless 

 renders spontaneous generation a " logical necessity." But the 

 " logical necessity " exists in our minds, not in Nature. Science 

 knows no " logical necessity," for the simple reason that we are 

 never able to compass all the possibilities in any given case. 



If we are to apply philosophic tests to the theories of reincar- 

 nation, we may find them equally eligible as articles of belief. 

 They are plausible, to some minds at least; they have logical 

 continuity. They are satisfying to the human heart, at least this 

 is claimed by their advocates. Their chief fault is that they can 

 be brought to no test of science and have no basis in inductive 

 knowledge. In other words, their only reality is that of the 

 vapors of dreamland. If plausibility and acceptability serve as 

 sufficient foundations for belief, then belief itself is a frail and 

 transient thiug, no more worthy of respect than prejudice, from 

 which indeed it could not be distinguished. Some such idea as 

 this seems to be present in the mind of Mr. Gladstone. In a 

 recent article, quoting in part the language of the honest Bishop 

 Butler, he ascribes to certain doctrines "a degree of credibility 

 sufficient for purposes of religion, and even a high degree of 

 probability." In other words, religion, which deals with human 

 hopes and fears, has less need of certainty than science, which is 

 ultimately concerned with human action. 



Haeckel makes the same distinction clearly enough. He uses 

 the term " belief " for " hypotheses or conjectures of more or less 

 probability " by which " the gaps empirical investigation must 

 leave in science are filled up. . . . These," he says, " we can not 

 indeed for a time establish on a secure basis, and yet we may make 

 use of them in the way of explaining phenomena, in so far as they 

 are not inconsistent with the rational knowledge of Nature. Such 

 rational hypotheses," he says, "are scientific articles of faith." 

 It is not clear, however, that so large a name as faith need be 

 taken for working hypotheses confessedly uncertain or transient. 

 The word " make-believe," used by Huxley in some such connec- 

 tion, might well be applied to hypothetical " articles of faith," 

 until given a basis by scientific induction. But it seems to me 

 that it is not necessary for the man of science to say " I believe," 

 in addition to " I know." He should put off the livery of science 

 when he enters the service of the Delphian oracles. 



