284
the immense succession of fictions provided by himself and his successors) definitely establishes the fact that no trace whatever exists of any considerable body of præ-Mohammedan or non-Arabic fiction in the extant texts of the Thousand and One Nights. He points out that the language of the collection is in no respect classical, containing many words in common and modern (as opposed to literary) use, that it is generally of a character to be referred to the decadence of Arabic literature and that all the tales, even when dealing with events supposed to have occurred and persons to have dwelt in Persia, India and other non-Arabic countries and in præ-Mohammedan epochs, invariably, with the naïvest anachronism, confine themselves to depicting the inhabitants, manners and customs of such cities as Baghdad and Cairo and are throughout impregnated with the strongest and most zealous spirit of Mohammedanism, and especially that the men and manners described are almost exclusively those of the epoch of the Abbaside Khalifs. Galland himself, in his preface, attributes the work to an unknown Arabic author: and the Sheikh Ahmed Shirwani, editor of the unfinished Calcutta edition of 1814–18, states in his introduction his belief that the author was a Syrian Arab, who wrote in a simple conversational style, which was not of the purest. Von Hammer himself allows that the first complete version could not have been finished till the beginning of the eleventh century, and it therefore could not have been known either to Mesoudi or the author of the Fihrist. Relying upon these and other arguments of equal weight, M. de Sacy concludes that the book was
