COMPARISON OF EMPIRICAL AND EXPLANATORY METHODS 107 



lowed by uplift and by renewed erosion consequent thereon, belong under 

 geography. 



When it is perceived that there are two kinds of geographical equip- 

 ment possible, and that both kinds are in use, it is evident that both must 

 ])e learned by all who wish to understand current geographical literature. 

 But when it comes to writing a description of an observed landscape the 

 writer should, unless he is writing so carelessly that he combines the two 

 equipments in a semi-conscious or accidental manner, consistently use 

 either one equipment or the other. He has therefore to determine which 

 one of the two equipments he will employ, and he will most wisely choose 

 the one that gives the greatest aid in his practical work. Hence he ought 

 to make conscious experiment with both methods, and his experiments 

 ought, for the best results, to be made in duplicate — that is, several differ- 

 ent landscapes ought to be treated in each of the two methods — and the 

 resulting descriptions should be impartially compared. At the present 

 stage in the evolution of geographical science nothing is more important 

 in the education of a young geographer than intentional, impartial ex- 

 periment with both these contrasted methods of description. 



But it often happens that inborn temperament and habit of thought as 

 determined by education are more influential than impartial judgment in 

 making choice between the two methods. The empiricist objects to ex- 

 planatory concepts because of their theoretical nature, their frequent 

 complication, and their possible error; he prefers concepts of a simpler 

 kind, based directly on observed forms and independent of all theory, 

 because they are then — as he thinks — so easily conceived and so safe, or 

 because they are so generally understood, and perhaps still more because 

 they were taught to him years before. On the other hand, the rationalist 

 finds empirical concepts insufficient in providing accurate and intelligible 

 descriptions ; he prefers explanatory concepts because, by the very reason 

 of their complexity, they better represent the complexity of actual land 

 forms and of increasing geographical knowledge. He frankly recognizes 

 that explanatory concepts are relatively complicated, and he well knows 

 that they must be made familiar by careful study before they can be 

 easily and effectively used ; he recognizes that if they are uncertain they 

 must be used only in a tentative way until they are safely established; 

 but he is willing to give time to establishing them and to familiarizing 

 himself with them, because they prove to be so extremely helpful in his 

 work. 



The empiricist is apt to treat explanatory concepts as if they were mere 

 fancies ; he insists that he does not wish to go beyond matters of fact. 

 The rationalist recognizes that while explanatory concepts are trulv the 



