28 PROCEEDINGS OF THE CANADIAN INSTITUTE. 



of our nature which, has been more unsatisfactorily treated. For a 

 full exposition of the manifestations of habit, or a clear understanding 

 of the purposes which they serve, we must look away from professed 

 works upon the subject. 



It may not be uninteresting to inquire into the causes of this 

 deficiency of treatment. They may probably be reduced to three : — 



(a) The failure to bring the subject under discussion, a defect 

 common to many works on psychology, may have been in some cases 

 the result of mere ovei'sight. 



(b) It may have been from despair of ability to remove difficulties 

 in the way of explanation ; for some of the questions involved have 

 appeared to distinguished writers inexplicable. For example. Dr. 

 Reid says : " I see no reason to think that we shall ever be able to 

 assign the physical cause either of instinct, or of the power of habit;" 

 and Dugald Stewart, in the second volume of his works, speaking of 

 the fact that while in some cases the muscles, under the law of habit, 

 are increased in strength through repeated exercise, they yet become 

 more and more obedient to the will, says : " This is a fact of which 

 it is probable that philosophy will never be able to give any 

 explanation." 



(c) But the failure of adequate treatment in cases where the 

 phenomena of habit have not been overlooked is mainly owing to 

 the fact that certain important phases have entirely escaped detection, 

 which, if their significance had been recognized, would probably have 

 shown the inadequacy of the methods that had previously been 

 adopted in explanation. Thinkers have a common tendency to forget 

 (pai'ticularly when dealing with mental phenomena), that while a 

 science remains stationary at a certain poiut, and until its advance- 

 ment beyond this point is made possible through the acquisition of a 

 more comprehensively detailed knoioledge of particulars, of con- 

 comitant phenomena and conditions, the ordinary methods of classifi- 

 cation and generalization, while their validity remains untouched, 

 may yet prove in the strictest sense useless and inapplicable, and the 

 facts may show themselves amenable only to a higher method. 



It is proposed in the jiresent paper to maintain that this is pre- 

 cisely the case in regard to the phenomena nnder discussion. By the 

 bringing in of particular instances it will be shown that the ordinary 

 definitions and explanations which have been given are inadequate. 

 It is not maintained that the ordinary scientific methods v[i\xs.tforever 



