44 THE AIM AND ACHIEVEMENTS OF SCIENTIFIC METHOD. 



points are distributed in the absolute theory, or else to supple- 

 ment the existing distribution of these by a supply of 

 " possible material points" which are indistinguishable from 

 the real spatial points which it is sought to exclude. If we 

 are to avoid both horns of the dilemma we must fall back upon 

 that severance of the worlds of thought and perception which 

 has already engaged our attention. The truths of geometry 

 will then hold only of conceptual space, and their approximate 

 validity in the perceptual world is really *> matter of lucky 

 coincidence, however it may have come about.* 



As in other cases to be considered in the sequel the 

 attitude which a thinker will take up on such a question as 

 this will depend very largely upon his intellectual tempera- 

 ment. If he shares with Professor James and his fellow 

 ^Humanists a constitutional distrust of " all noble, clean-cut, 

 fixed, eternal, rational, temple-like systems " of thought he will 

 probably hold by the relational theory of space, and glory in 

 the name of empiricist which is hurled at him as a reproach. 

 If, on the other -hand, his faith in the organic unity of all 

 truth is sufficiently robust to push logical arguments daunt- 

 lessly to their ultimate conclusions even when they seem to 

 the Humanist " oddly personal and artificial," or even " bureau- 

 cratic and professional iti an absurd degree,"! he will be an 

 absolutist. Such a thinker will prefer to leave the difficulties 

 presented by the psychological analysis of space perception to 

 be reconciled with his main convictions by some theory of 

 error. 



In the article to which reference has just been made, 

 Mr. Eussell disposes of the question of the status of moments 



* Our argument is not affected by the consideration that, after all, 

 our space may be really of a non-Euclidean pattern, so that it is only 

 approximately represented by the usual homaloidal model. This would 

 mean that we fail to discriminate correctly the actual spatial relations 

 which are presented to us. It could not negative the arguments in favour 

 of the Objectivity of points. 



+ James, Humanism and Truth, p. 467. 



