[bowman] fundamental PROCESSES IN HISTORICAL SCIENCE 581 



ditions governing their creation, are: — (1) The narrative is intended 

 essentially for users who have no access to the narrator's sources ex- 

 cept through the narrative itself; (2) the narrator's essential duty 

 is to formulate a trustworthy narrative on this basis, so that the user, 

 without further test and without further access to the sources, shall 

 have essentially correct information concerning the events in ques- 

 tion; and (3) the narrator, in fulfilling this essential duty, cannot 

 explain, or even cite systematically, the sources of his individual 

 statements, because the limitations of space and laws of perspective 

 in general and the particular purpose of the work as fixed by the users 

 for whom it is essentially intended will not allow him to do these two 

 things at once: he cannot make his narrative trustworthy, and at the 

 same time attempt to prove, in the narrative itself or by foot-notes, 

 that it is so. 



iii. What is the attitude of the prevailing method towards records 

 as such? Professor Fling, in his Source Book of Greek History 

 (preface, p. vi), states this attitude, as follows: "One of the main pur- 

 poses of this critical work is to make the pupil comprehend the uncer- 

 tainty and unreliability of much of our information upon Greek his- 

 tory, and that this is due to the character of the evidence with which 

 we are obliged to work. A further purpose is to bring out the idea that 

 in history the only 'authority' is the source, and that the writer of a 

 historical narrative cannot take refuge behind the dogma of infalli- 

 bility, but must prove all that he asserts by the citation of evidence." 



In connection with this attitude and statement, the following 

 three points should be noted: — 



1. The attitude contravenes the essential principle, and thus 

 defeats the very purpose, of all historical records and narratives as such. 

 The attitude demands of the recorder and narrator what he cannot 

 give. In some miraculous, superhuman way (for it is not a human 

 possibility), he must prove "all that he asserts by citation of evidence," 

 and yet continue his narrative: i.e., he must do two mutually exclusive 

 things at one and the same time. The inevitable result of this atti- 

 tude is given by Professor Fling unconsciously within the bounds of 

 his own statement. He names two purposes, and they are exactly 

 complementary. In the second we have an attitude that drops over- 

 board the material of historical science; and in the first we have the 

 "uncertainty" and "unreliability" — or in plainer words, the scepti- 

 cism — that must descend, not only upon Greek, but upon all, history, 

 as the natural and necessary result of this unscientific proceeding. 



2. In the above statement, the narrator who does not prove all 

 that he asserts by citation of evidence is taxed with taking refuge 

 behind the "dogma of infallibility," whereas the only "authority" in 



