for ancient literature? It has just now been shown that the tragedians recognised that a curse or a deadly sin led to the resuscitation of the body; and yet they make lack of burial and violent death lead rather to the re-appearance of a ghost. Why then this discrimination between the effects produced by causes all of which in more recent popular belief produce the same effect? My answer is that popular belief in antiquity was the same as popular belief now in respect of all the causes, but that literary propriety forbade more than a mere verbal reference to so gross a superstition as bodily resuscitation. When a dead man was required in literature to re-appear, he was conventionally pourtrayed as a ghost, not as a walking corpse; and the convention was, I think, right and necessary.
For let it be granted for a moment that the popular belief of to-day dates from the earliest times, and that then as now the revenant was popularly pictured as a monster 'swollen and distended all over so that the joints can scarcely be bent; the skin being stretched like the parchment of a drum, and when struck giving out the same sound.' Could even Homer have re-animated the dead Patroclus, with this unearthly ghastliness added to his wounds and to his mangling by the chariot, and have brought him to Achilles in the darkness of the night, without exciting in his breast horror instead of pity and loathing for love? Euripides again was greatly daring when he assigned the prologue of a tragedy to Polydorus' ghost; but even he could not have restrained the unquenchable mirth of his audience, if his play had opened with a soliloquy by an agitated corpse. Epic and dramatic propriety must have demanded some refinement of so grossly material a conception. The canons of drama, we know, would not allow the enactment of a murder on the stage before the eyes of the spectators; would it then have been compatible with the restraint of Greek art to represent the murdered body as a revenant? Aeschylus himself, the lover of weird misbegotten shapes, would have recoiled from such an enterprise. But those same canons did permit a verbal description of the murder; and similarly the tragedians permitted themselves to refer, in imprecations and suchlike, to the horror of bodily resuscitation. The case then stands thus. References are, as we have seen, made by the tragedians to the possibility of men becoming reve-*
