Page:Popular Science Monthly Volume 27.djvu/835

person's credit, or would lead a customer to doubt whether his banker was solvent; but in such cases all that is possible is a guess more or less judicious, and a guess, however judicious, is a totally different thing from settled rational belief. As regards all detailed matters of fact, I think there is a time, greater or less, during which the evidence connected with them may be collected, examined, and recorded. If this is done, a judgment can be formed on the truth of allegations respecting them at any distance of time. Such judgments are rarely absolute; they ought always or nearly always to be tempered by some degree of doubt, but I do not think they need be affected by lapse of time. If, however, this opportunity is lost, if no complete examination is made at the time of an incident, or if being made it is not properly or fully recorded, clouds of darkness which can never be dispelled settle down upon it almost immediately. All that remains behind is an indistinct outline which can never be filled up. Under certain conditions rare occurrences are quite as probable as common ones. The main condition of the probability of such an event is that the rare occurrence should, from its nature and from the circumstances under which it occurs, be capable of being observed, and that the evidence of it should be recorded in the manner which I have already described. If a moa were caught alive and publicly exhibited for money, or if the body of a sea-serpent were to be cut up upon the coast and duly examined by competent naturalists, the existence of moas and sea-serpents could be proved beyond all reasonable doubt. The reason why their existence is disbelieved or doubted is not that they are seen, if at all, so seldom, but because in each particular instance they are seen, if at all, in such an unsatisfactory way that it is doubtful whether they ever were seen. There are innumerable ghost-stories in circulation, but as far as I know no instance has ever yet been even alleged to exist in which the existence of a ghost has been properly authenticated as readily and as conclusively as that of any other being whatever. Stories of the interference of unseen agents stand upon exactly the same footing, speaking generally. Isolated instances occur in all ages and countries, but the common characteristic of them all is to be unauthenticated. Ten cases distinctly proved under the conditions referred to. . . would do more to settle the question of the existence of miracles as a class than innumerable cases depending on assertions which were not properly examined when they were originally made, and which can now never be examined. On the other hand, what reason can possibly be suggested why the action of an invisible person upon matter should not be ascertained just as clearly as the action of a visible person? The restoration of a dead body to life might, if it occurred, be proved as conclusively and as notoriously as the death of a living person, or the birth of a child. If such events formed a real class to which new occurrences might be assigned, a large number of instances of those occurrences would be, so to speak, upon record, established beyond all