[bowman] fundamental PROCESSES IN HISTORICAL SCIENCE 565 



of essential correctness in historical records be taken as 24/25 or as 

 99 /lOO respectively, instead of 49 /50, the acceptance of the two records 

 in agreement against a third, will rest on a probability of 26;1, or of 

 101:1, respectively. 



6. Apparent modesty of probable statements in historical narra- 

 tion. The prevailing resort to probable conclusions in historical 

 narration springs in part from a feeling in historians that they are more 

 modest in making hypothetical statements than categorical. This 

 apparent modesty is only superficial, and is due to the fact that un- 

 truthful braggarts are not hypothetical, but almost always categori- 

 cal, in their statements. Such braggarts, however, while they ex- 

 ternally (but not really) observe the 5th requisite of trustworthiness 

 by making no admittedly probable statements, openly violate all the 

 other four requisites, except occasionally the 1st. Only where the 

 entire five requisites are exemplified together, is a statement trust- 

 worthy; and the historian who first exemplifies faithfully the first 

 four, by doing his utmost to obtain and state clearly the information 

 desired by readers, with strict impartiality and without exaggeration, 

 and then observes also the 5th by cutting out rigorously all clever 

 guesses, all pet theories, and all favorite hypotheses, which he can 

 almost, and yet not quite, prove, is the self-abnegatory and truly 

 modest man; but the historian who includes such probabilities in his 

 narrative is fundamentally immodest, because he wishes readers to 

 accept, or at least to attach weight to, these guesses, without proof, 

 and therefore simply because they are his. He thereby exceeds his 

 evidence, and since the only ground for acceptance of his con- 

 clusions must be either in the evidence or in himself, his appeal to the 

 reader rests necessarily on his personal opinion and prestige. 



7. Contemporaneousness as a test of historical trustworthiness. 

 In the introduction to the present part of this paper (p. 494), it was 

 stated that two important features in any historical method are its 

 test for trustworthiness, and its treatment of conflicting statements in 

 trustworthy records where the circumstances of the discrepancy are 

 unknown; and that this test and this treatment consist, in the pre- 

 vailing method, of contemporaneousness and harmonization, both 

 of which are incorrect, probable processes. The test by contem- 

 poraneousness originates chiefly from the fact that all trustworthy 

 records must depend directly, or through trustworthy media (men or 

 previous records), on actual participants (and therefore, contempor- 

 aries) of the events in question; and thence the deduction is made 

 that contemporaneousness in itself, and its opportunities, are the 

 criterion of trustworthiness. This is an unscientific inference. At 

 Gottingen, in a test of a pre-arranged scene sprung unexpectedly upon 



