JSUHEME OF INTERPRETATION. 429 fhe utmost which we accomplish by means of the semi-his Un-ical theory, even in its most successful applications, is, that after leaving out from the mythical narrative all that is miracu- lous or high-colored or extravagant, we arrive at a series of credi- ble incidents incidents which may, perhaps, have really occur red, and against which no intrinsic presumption can be raised. This is exactly the character of a well-written modern novel (as, for example, several among the compositions of Defoe), the whole story of which is such as may well have occurred in real life : it is plausible fiction, and nothing beyond. To raise plausible fic- tion up to the superior dignity of truth, some positive testimony or positive ground of inference must be shown ; even the highest measure of intrinsic probability is not alone sufficient. A man who tells us that, on the day of the battle of Plataca, rain fell on the spot of ground where the city of New York now stands, will neither deserve nor obtain credit, because he can have had no means of positive knowledge ; though the statement is not in the slightest degree improbable. On the other hand, statements in themselves very improbable may well deserve belief, provided they be supported by sufficient positive evidence ; thus the canal dug by order of Xerxes across the promontory of Mount Athos, and the sailing of the Persian fleet through it, is a fact which 1 believe, because it is well-attested notwithstanding its remark- able improbability, which so far misled Juvenal as to induce him to single out the narrative as a glaring example of Grecian men- dacity. 1 Again, many critics have observed that the general tale of the Trojan war (apart from the superhuman agencies) is not more improbable than that of the Crusades, which every one ad- mits to be an historical fact. But (even if we grant this position, which is only true to a small extent), it is not sufficient to show an analogy between the two cases in respect to negative presump- tions alone ; the analogy ought to be shown to hold between them Kcview, October 1846), pointed out that those remarks required modification, and that the idea of allegory in reference to the construction f the mytlitt was altogether inadmissible. 1 Juvenal, Sat. x. 174 : " Creditur olim Velificatus Athos, et quantum Graecia mendax Audet in historic! ," etc.
