This view of the case (that a "ghost" may be a reflection of a dead man's dream) will become less difficult to understand if we ask ourselves what natural thing most resembles the common idea of a ghost. You are reading alone at night, let us say, the door opens and a human figure glides into the room. To you it pays no manner of attention; it does not answer if you speak; it may trifle with some object in the chamber and then steal quietly out again. _It is the House-maid walking in her Sleep_. This perfectly accountable appearance, in its aimlessness, its unconsciousness, its irresponsiveness, is undeniably just like the common notion of a ghost. Now, if ordinary ghosts are not of flesh and blood, like the sleep-walking house-maid, yet are as irresponsive, as unconscious, and as vaguely wandering as she, then (if the dead are somewhat) a ghost _may_ be a hallucination produced in the living by the _unconscious_ action of the mind of the dreaming dead. The conception is at least conceivable. If adopted, merely for argument's sake, it would first explain the purposeless behaviour of ghosts, and secondly, relieve people who see ghosts of the impression that they see "spirits". In the Scotch phrase the ghost obviously "is not all there," any more than the sleep walker is intellectually "all there". This incomplete, incoherent presence is just what might be expected if a dreaming disembodied mind could affect an embodied mind with a hallucination. But the good old-fashioned ghost stories are usually of another type. The robust and earnest ghosts of our ancestors "had their own purpose sun-clear before them," as Mr. Carlyle would have said. They knew what they wanted, asked for it, and saw that they got it. As a rule their bodies were unburied, and so they demanded sepulture; or they had committed a wrong, and wished to make restitution; or they had left debts which they were anxious to pay; or they had advice, or warnings, or threats to communicate; or they had been murdered, and were determined to bring their assassins to the gibbet. Why, we may ask, were the old ghost stories so different from the new? Well, first they were not all different. Again, probably only the more dramatic tales were as a rule recorded. Thirdly, many of the stories may have been either embellished--a fancied purpose being attributed to a purposeless ghost--or they may even have been invented to protect witnesses who gave information against murderers. Who could disobey a ghost? In any case the old ghost stories are much more dramatic than the new. To them we turn, beginning with the appearances of Mr. and Mrs. Furze at Spraiton, in Devonshire, in 1682. Our author is Mr. Richard Bovet, in his Pandaemonium, or the Devil's Cloister opened (1683). The motive of the late Mr. Furze was to have some small debts paid; his wife's spectre was influenced by a jealousy of Mr. Furze's spectre's relations with another lady.