[478] Ibid. vol. i. pp. 69, 374, &c.; vol. ii. p. 216, &c.
[479] Rollof's Saga, c. 35, &c.
[480] It is peculiarly unfortunate that such as maintain this opinion are obliged to take their first step from the Moorish provinces in Spain, without one intermediate resting place, to Armorica or Bretagne, the province in France from them most remote, not more in situation than in the manners, habits, and language of its Welsh inhabitants, which are allowed to have been derived from this island, as must have been their traditions, songs, and fables; being doubtless all of Celtic original. See p. [3] of the Dissertation on the Origin of Romantic Fiction in Europe, prefixed to Mr. Tho. Warton's History of English Poetry, vol. i. 1774, 4to. If any pen could have supported this darling hypothesis of Dr. Warburton that of this ingenious critic would have effected it. But under the general term Oriental, he seems to consider the ancient inhabitants of the north and the south of Asia, as having all the same manners, traditions, and fables; and because the secluded people of Arabia took the lead under the religion and empire of Mahomet, therefore everything must be derived from them to the Northern Asiatics in the remotest ages, &c. With as much reason under the word Occidental, we might represent the early traditions and fables of the north and south of Europe to have been the same; and that the Gothic mythology of Scandinavia, the Druidic or Celtic of Gaul and Britain, differed not from the classic of Greece and Rome.
There is not room here for a full examination of the minuter arguments, or rather slight coincidences, by which our agreeable dissertator endeavours to maintain and defend this favourite opinion of Dr. W., who has been himself so completely confuted by Mr. Tyrwhitt. (See his notes on Love's Labour Lost, &c.) But some of his positions it will be sufficient to mention: such as the referring the Gog and Magog, which our old Christian bards might have had from Scripture, to the Jaguiouge and Magiouge of the Arabians and Persians, &c. (p. [13]). That "we may venture to affirm that this (Geoffrey of Monmouth's) Chronicle, supposed to contain the ideas of the Welsh bards, entirely consists of Arabian inventions" (p. [13]). And that, "as Geoffrey's history is the grand repository of the acts of Arthur, so a fabulous history ascribed to Turpin is the groundwork of all the chimerical legends which have been related concerning the conquests of Charlemagne and his twelve peers. Its subject is the expulsion of the Saracens from Spain, and it is filled with fictions evidently congenial to those which characterize Geoffrey's History" (p. [17]). That is, as he afterwards expresses it, "lavishly decorated by the Arabian fablers" (p. [58]). We should hardly have expected that the Arabian fablers would have been lavish in decorating a history of their enemy: but what is singular, as an instance and proof of this Arabian origin of the fictions of Turpin, a passage is quoted from his fourth chapter, which I shall beg leave to offer, as affording decisive evidence, that they could not possibly be derived from a Mahometan source. Sc. "The Christians under Charlemagne are said to have found in Spain a golden idol, or image of Mahomet, as high as a bird can fly—it was framed by Mahomet himself of the purest metal, who, by his knowledge in necromancy, had sealed up within it a legion of diabolical spirits. It held in its hand a prodigious club; and the Saracens had a prophetic tradition, that this club should fall from the hand of the image in that year when a certain king should be born in France, &c." (vid. p. [18], note.)
[481] The little narrative songs on Morisco subjects, which the Spaniards have at present in great abundance, and which they call peculiarly romances, (see vol. i. book iii. no. xvi. &c.), have nothing in common with their proper romances (or histories) of chivalry, which they call Historias de Cavallerias; these are evidently imitations of the French, and shew a great ignorance of Moorish manners: and with regard to the Morisco, or song romances, they do not seem of very great antiquity; few of them appear, from their subjects, much earlier than the reduction of Granada, in the fifteenth century: from which period, I believe, may be plainly traced among the Spanish writers, a more perfect knowledge of Moorish customs, &c.
[482] See Northern Antiquities, passim.
[483] Ibid.
[484] Saxon Gram. p. 152, 153. Mallet, North. Antiq. vol. i. p. 321.
[485] See a translation of this poem, among Five pieces of Runic Poetry, printed for Dodsley, 1764, 8vo.
[486] Vid. Mallet, Northern Antiquities, passim.