[BOWMAN] DISCREPANCY IN TRUSTWORTHY RECORDS 169 
and requires no especial treatment, except that, to whatever extent 
the circumstances of the original discrepancy remain operative, to that 
extent also may the original discrepancy itself be reproduced in subse- 
quent historical writings. Thus among the British, French and Prus- 
sian nations who shared in the battle of June 18, 1815, this contest was 
named by the British, after the village where Wellington had his 
head-quarters and dated his despatch of June 19 to Lord Bathurst, the 
Battle of Waterloo; by the the French, after a village in the rear of 
British centre, the Battle of Mont St. Jean; and by the Prussians, 
after a farm and house in the centre of the French line, the Battle of 
Belle Alliance: and while such a divided usage prevails in their respec- 
tive countries, the same divided usage may prevail in their respec- 
tive histories; and the resulting discrepancy is not, and will not, be re- 
cognized as such, so long as the attendant circumstances remain known. 
(b) WHERE THE DiscREPANCY Lies BETWEEN RECORDS OF ESSEN- 
TIALLY UNEQUAL TRUSTWORTHINESS AND THE ATTENDANT CIRCUM- 
STANCES ARE UNKNOWN. 
In the 26 cases the discrepancies lay between persons of essentially 
equal trustworthiness both in general and with respect to the particular 
statements made; hence no feature in these cases affects in any way 
the ordinary rule of historical method that where a discrepancy lies 
between a trustworthy and an untrustworthy record, or between records 
of essentially unequal trustworthiness, the conflicting statement of the 
inferior record is to be rejected in favor of the superior. Moreover, 
in those cases where two such records are essentially and equally trust- 
worthy in general, if it can be shown that either of them is of essentially 
inferior trustworthiness either with respect to the particular statement 
which conflicts or with respect to the particular field of facts to which 
such statement belongs, then also the conflicting statement of such 
record is to be rejected in favor of the contrary record and statement; 
and in the treatment of every case of discrepancy the first scientific 
requirement is to establish, if possible, in favor of either of the conflict- 
ing records, such an essentially superior trustworthiness, either in 
general or with respect to the particular statements which conflict, 
and to accept and reject one or other of them accordingly. 
