544 THE ROYAL SOCIETY OF CANADA 



tion, probability is only a matter of favorable against unfavorable 

 chance, and therefore all its results, correct or incorrect, are accident- 

 ally obtained. Such conclusions, historical or other, even though 

 correct, are condemned by the 2nd methodic principle as unscientific, 

 because science must supply assured knowledge and this cannot 

 spring from accidents. 



iii. // a correct and an incorrect process coincide in part, correct 

 results obtained by the application of the incorrect process within the 

 coincident part are only accidental and afford no ground for accepting as 

 correct those results which are obtained by the application of the incorrect 

 process within the non- coincident part. This principle, as applied to 

 probability in historical science, may be illustrated by the correct, 

 as opposed to the prevailing incorrect, method of determining histori- 

 cal trustworthiness. The process of judging the trustworthiness of a 

 record by its exemplification of the five requisites of trustworthiness is 

 a correct process, because a record, in so far as it exemplifies these 

 five requisites, is necessarily correct. On the other hand, the prevail- 

 ing practice of judging the trustworthiness of a record by its con- 

 temporaneousness with the events narrated is an incorrect process 

 based on probability, because contemporaneous records are not 

 necessarily correct, and the only reason for expecting that they will 

 be so is that a contemporaneous writer presumably will have more 

 opportunities for informing himself concerning the events narrated 

 than a later writer, and for that reason it is held that there is more 

 chance or probability that a contemporaneous record will be correct. 

 These two correct and incorrect processes respectively coincide in 

 part, because some contemporaneous records ( = the coincident group) 

 exemplify the five requisites and will therefore be adjudged trustworthy 

 under either process, while other contemporaneous records ( = the 

 non-coincident group) do not exemplify the requisites. The fact that, 

 under the incorrect process, any record in the coincident group is 

 rightly adjudged as trustworthy, is only an accidentally correct result 

 from that process, and affords no ground for accepting any of its 

 conclusions respecting the non-coincident group, e.g., that a contem- 

 poraneous record which does not exemplify the five requisites is yet 

 trustworthy because contemporaneous, or that a non-contemporane- 

 ous record which does exemplify those requisites is untrustworthy 

 because not contemporaneous. 



iv. Where, under proper application, a process considered to have 

 been established by experience as correct fails in any instance to produce 

 a correct result, the process is thereby scientifically condemned and ought 

 to be abandoned unless the failure can be differentiated, i.e., unless a con- 

 trollable cause of it can be located and the process so adjusted as to prevent 



