131 



SCIENCE 



[jSr. 8. Vol. XXXVII. No. 943 



ness, emulation, kindness, reasoning power 

 and the like — each of which was a faculty 

 or essence whose nature and degree of 

 strength at any one time could be measured 

 once for all. They had not even begun to 

 abandon the notion that men were classi- 

 fiable mentally into a rather small number 

 of rather distinct types. 



The early practise of diagnosis amongst 

 psychologists, so far as there was any, con- 

 sisted in inferring the condition of the 

 individual's memory, imagination, power 

 of attention, suggestibility and the like, 

 from some facts in his behavior in which 

 these magic powers " expressed " them- 

 selves, and in defining him as of this or 

 that "type" on the basis of his divergence 

 from the average in one or the other direc- 

 tion. Workers in schools who had pro- 

 gressed so far as to know of these prac- 

 tises either accepted their validity or dis- 

 trusted them without knowing why they 

 did so. The great majority of teachers con- 

 tinued to judge intellect, character and 

 skill by the traditional means used and 

 slowly improved by generation after gen- 

 eration since man could think and speak. 



The active experimentation with objec- 

 tive "tests" of mental traits during the 

 last score of years has shown that "good- 

 ness" of memory in the sense of a uniform 

 power to hold all that is acquired, closeness 

 of concentration in the sense of a uniform 

 power to resist at will distractions of every 

 variety and other similar general excellen- 

 cies or defects, were myths. The measure- 

 ments of correlations of the last decade 

 have shown that "types" of attentive- 

 ness, of imagery and the like, and of intel- 

 lect or character as a whole, either do not 

 exist at all or are so complicated by inter- 

 mediate conditions as to be of no service to 

 thought or practise. 



The experimentation with tests and the 

 measurements of correlations began, how- 



ever, with faith in the dogmas which they 

 eventually disproved. The great majority 

 of workers in the early days of tests as- 

 sumed that certain formal general powers 

 of mind existed, the strength of each being 

 easily diagnosed by any one of its mani- 

 festations. Learning ten digits or non- 

 sense-syllables' was a test of "memory." 

 He in whom the words violin, whistle, bell 

 and brook aroused images of sound was 

 thereby classified apart from him in whom 

 they aroused images of sight. "Endur- 

 ance" was measured by the ergograph, or 

 by addition, or by the rate of tapping. 

 Idiocy, imbecility and feeble-mindedness 

 were three real entities, with symptoms 

 awaiting our discovery. Even to-day much 

 of this expectation that the intellectual 

 and moral condition of an individual can 

 be adequately described in adjectives, and 

 manifests itself by the clear presence or ab- 

 sence of symptoms in the way that measles 

 and smallpox do, remains. 



Experimentation began to cure itself of 

 these traditions as soon as it began to test 

 the tests themselves. As fast as it aban- 

 doned the habits of assuming a priori what 

 a certain fact in behavior signified concern- 

 ing the individual's constitution, and 

 sought instead to discover by exacts 

 that is, quantitative — observation just what 

 constitutional features it actually did go 

 with, the older easy but false diagnosis was 

 exposed and the basis for present and fu- 

 ture achievement was laid. 



Such studies of symptoms and tests are 

 roughly of two sorts, those in which two 

 distinct groups are compared in respect to 

 the trait in question and in respect to the 

 various possible sjmiptoms of it, and those 

 in which a long series of groups, each dif- 

 fering but little from the next in order, are 

 so compared. In both cases certain prin- 

 ciples of method are very desirable, almost 

 necessary, because of the continuous grada- 



