Mat 19, 1911] 



SCIENCE 



773 



" There is no doubt but that ISTapoleon is the 

 most eminent man who has ever lived, yet it 

 should give us pause to think that this Titan 

 of Anarchy stands first in the thoughts of 

 most men." In the passage just quoted we 

 have one extremely eulogistic phrase " most 

 eminent man " counterbalanced by another 

 phrase of extreme disparagement " Titan of 

 Anarchy." A similar array of favorable and 

 unfavorable expressions can be found in any 

 impartial biography of Napoleon. It is this 

 peculiar blending in one man of different ex- 

 tremes which has given Napoleon and many 

 other men a great share of their celebrity; in 

 such cases the ratio of the numbers of ad- 

 jectives of praise against those of dispraise 

 fails to give a true answer to the question, 

 which man of a given group of men is the 

 most eminent or historically most important. 



The space method and reference frequency 

 methods of estimating fame are not open to 

 the objections which have been raised against 

 the adjective method. The historian in dis- 

 cussing, for example, the respective influence 

 of Euripides and Sophocles upon human af- 

 fairs must necessarily devote more space and 

 make more references to Euripides since his 

 influence in this respect was much the greater, 

 yet in doing this he need not necessarily em- 

 ploy any adjectives of direct praise or dis- 

 praise. 



The space method and reference frequency 

 methods are also more free from the errors of 

 personal equation than the adjective method. 

 In the sentence " Csesar was ambitious " one 

 person might regard ambitious as a term of 

 praise and another of dispraise, yet these two 

 persons would agree perfectly as to the num- 

 ber of lines in a biographical sketch of Osesar 

 or as to the number of times Caesar was re- 

 ferred to in an index. 



In the selection of a method for estimating 

 historical values it would seem then necessary 

 first of all to dissociate the question of merit 

 from that of fame, and the questions of excel- 

 lence in particular directions from the broader 

 questions of historical importance. Eor esti- 

 mating merit and excellence in particular 

 qualities, which is perhaps the chief concern 



of the critic, the adjective method proposed by 

 Dr. Woods may possess certain advantages. 

 But for estimating fame and historical im- 

 portance, which is the chief concern of the 

 " historiometrician," the adjective method 

 would seem far inferior to the space and ref- 

 erence frequency methods. 



As to the exactness of historiometry as a 

 science, may we not say what Huxley once 

 said of another science, the most exact of all. 

 It " grinds your stuff of any degree of fineness, 

 but nevertheless what you get out depends on 

 what you put in." 



C. A. Browne 



New York City 



dr. woods's application of the histometric 



METHOD 



The paper by Dr. F. A. Woods, published 

 in Science, April 14, giving the results of his 

 metrical investigation of the biographies of 

 eminent Americans is one of great interest. 

 Both in method and results it opens fields of 

 investigation of the highest sociological value. 

 He has proved the reliability of his figures by 

 reaching approximately the same results, for 

 the state of Massachusetts and the other thir- 

 teen original states, when using different sets 

 of data ; and while the variation in the results 

 indicate what would be considered in physics 

 as a large probable error, yet they are really 

 small considering the method used and the 

 number of observations. 



If the wide range shown thus by the dif- 

 ferent states in their production of eminent 

 persons per thousand of their white population 

 can not be explained by environment it is 

 evident that the arguments for the dominance 

 of hereditary ability will be strongly supported. 

 On the other hand, if it can be explained by a 

 high coefficient of skew correlation with one 

 or more series of quantities expressing any 

 antecedent social condition it leaves just so 

 much less for heredity to explain. Thanks to 

 the work of Galton and others, heredity is 

 already mathematically expressed by the cor- 

 relation of the characters of individuals in 

 successive generations. And perhaps for that 

 reason the tendency now is to exaggerate the 



