Lord Brougham was awake, he did so reason, and was accordingly shocked.
So vivid was the apparition that he tumbled out of the bath and fainted. It is only some time after this, when writing up his diary, that he has no doubt that he had fallen asleep. Preconceived theories about apparitions now come up in his mind and get him into trouble; he must _explain_ his vision.
Now for the explanation. Lord Brougham finds, on returning to Scotland, that his former friend is dead, and that the time of his death corresponded with the time at which he had seen his apparition in Sweden, December 19th.
"Singular coincidence!" That is Lord Brougham's explanation; and that is the usual explanation; but it is ill-considered--it is weak--it does not cover the ground.
Lord Brougham had but two theories from which to choose: namely, Chance and Supernaturalism; and of the two horns of the dilemma he chose the easier one.
Let us, however, place ourselves, for the moment, on his ground, namely, that (1) It was a dream; and (2) dreams are so numerous that it is not surprising that some of them coincide with contemporaneous events.
Evidently the more numerous the coincidences, or the dreams which correspond to contemporaneous events, the weaker becomes the theory of _chance_ coincidences. Supposing, then, Lord Brougham's case to have been unique, that not another similar case was known to have occurred, then we should have no particular hesitation in assigning it to the category of chance coincidences; but even then it would be out of the order of _usual_ coincidences both in interest and the number of separate points involved; it would excite special interest, but the reference of it to chance would not be considered unreasonable: if, however, three or four such cases had been reported and discussed in a generation, thoughtful people would begin to inquire if there might not be some relation of sequence, or possibly of cause and effect; but when hundreds of cases have been reported, because they have been systematically sought for--veridical dreams connected with the moment of the death of the agent, with fainting, with trance, with moments of supreme excitement, or of extreme danger, so many different conditions in which by careful observation it is found that such hallucinations and symbols relating to actual contemporaneous occurrences originate and are telepathically transmitted--the matter is then quite removed from the category of chance coincidences, and any attempt to force these cases there to-day denotes either ignorance of established facts or inability to appreciate logical reasoning or even mathematical demonstration. This is all upon the supposition that the case in question was a dream. On the other hand, now place the case where it really belongs as a _waking_ or Borderland _vision_--an event in a class a hundred-fold less numerous than dreams--and in which class corresponding events are at least tenfold _more numerous_, and we see how conspicuously weak is the coincidence theory.