Page:Mind (New Series) Volume 8.djvu/88

74 A. F. RAVENSHEAE : seemingly no easier to answer than the original one. That there may, nevertheless, be some advantage in thus splitting up the main question in the hope of some one or more of the subordinates ones being answerable is obvious. For unfavourable answers to the questions as to sincerity or opportunity condemn the testimony without more ado. A further advantage, however, becomes immediately apparent when we begin to take into account the subject of Corro- boration. We may often with the aid of corroboration of one kind or another draw conclusions as to the component questions far more certainly than we could as to the whole question ; and by an aggregate of testimony and other cor- roboration may in effect ensure that all the conditions have been satisfied that a single witness would conform to if perfect. Corroboration. The conditions of trustworthiness having thus been briefly enumerated, we are in a position to look more closely at the subject of Corroboration. This may take the form of concurrence or conflict of testimony,, and confirmation by other evidence, which may either be internal or external. The various kinds of corroboration will be reviewed in the order named, beginning with con- currence of testimony. As we have already seen, it is necessary in estimating the value of testimony to consider whether there is any want, of sincerity or anything likely to breed unconscious bias in the mind of the assertor. To satisfy the conditions of logical proof on these points in the case of a single witness might be extremely difficult or quite impossible ; but if the circumstances and interests of the individual witness, where there are several, are sufficiently varied, the difficulty dis- appears. If indeed there is known conflict of interest between the witnesses with regard to the matter asserted, and they nevertheless agree, we may safely infer that in- sincerity has not been operative with the bulk of the assertors. Since we are given a conflict of interest but a concurrence in statement, most of the assertors we are assured must have spoken against interest ; whence bias or insincerity cannot have been operative in their several cases. Bias and insincerity may therefore be eliminated by a concurrence of persons of sufficiently varied interests. In a less degree concurrence of testimony throws light on the extent to which others of the conditions of reliability are conformed to. We have seen that the value of testimony is largely dependent upon the sufficiency of skill or experi- ence on the part of the assertor, on his powers of observation