Previously Unpublished Works, Fascicle 5

the public generally. The distinction is one which dates from his own lifetime; it was universally known, and amply accounts for the attribution of such a work as The Secret of Secrets original: "Secretum Secretorum" to Aristotle; no student in the Middle Ages could be blamed for receiving it as possibly genuine, however different from the books he met with in the schools.

Had then the book any claim to a Greek origin? I think not. Greek ideas, Greek commonplaces have been caught up into its text, Greek treatises have been incorporated with it, but the texture itself of the original work is oriental, not western. I believe it to have had its origin in the interaction between Persian and Syriac ideas which took place in the seventh to ninth centuries of our era, and probably at the same time as the Alexander legends studied by Dr. Budge.

It will perhaps be convenient to have a general summary of the history of the text, before going into particulars. No Syriac text has yet been found, though there is every probability that it has existed. The Arabic is found in two forms which I name provisionally Eastern and Western. Both obviously contain additions, but fortunately in the Western form the additions are nearly all at the end, while in the Eastern they are incorporated in the body of the work. By striking out these additions we get a fair idea of the text of the work, which can be most readily studied in Dr. Gaster's translation of the Hebrew version of the Western form, paragraphs 1 through 79 omitting 48. Though the list of discourses in the Eastern form is 10 and the Western 8 (or 7), the titles are the same and each is really in 10 discourses, though the Western calls some of them "gates."

A Latin translation of a part of the Western expanded text was made early in the twelfth century, but the whole text of the Eastern form was translated in the thirteenth century by Philip of Tripoli.

No Latin text corresponds in order or content to any Arabic text we know, and all the Latin texts we have are of the same general form. The tenth discourse of the Arabic is cut up, a great part omitted, the remainder displaced, part being added to the ninth, the rest to the second book. A new tenth book for the Latin is formed by taking the section on physiognomy The practice of assessing a person's character or personality from their outer appearance, especially the face. from the Arabic second book and making it a complete book. It seems possible that the Vulgate The common or standard version of a text. text thus formed is not the original Latin of Philip, but one edited by authority. The reasons are (1) Philip speaks in his introduction of geomancy A method of divination that interprets markings on the ground or the patterns formed by tossed handfuls of soil, rocks, or sand.. This (the calculation of the numerical values of