The Thousand and One Nights (Alf Layla wa-Layla) from the Earliest Known Sources: Arabic Text Edited with Introduction and Notes, vol. 3, Introduction and Indexes.

• Author: Hamori, Andras

The first two volumes of Muhsin Mahdi's great work on the Nights (Leiden, 1984) contained his critical edition of the Galland manuscript, and his arguments for a radically new view of the relations among manuscripts. The third and final volume begins with three chapters on "the history of the Arabic Nights from its arrival in Europe early in the eighteenth century through the publication of the last of the four editions during the first half of the nineteenth century," a history that includes, as already argued in part 2, some spectacular moments in the life of the Syrian family of manuscripts. There follow three interpretations, and two important indices: an extensive general index to the Arabic text of the Nights including both names and objects, and an index to the critical apparatus and description of manuscripts.

The historical chapters have, besides thorough scholarship, the quick plotting, ambiguous characters, and rich local color (much of it from personal letters and diaries) of a good detective story. Mahdi describes how Galland, taking the title literally, went about looking for a complete Thousand and One Nights to translate, and failing to find any such thing, "concocted one" over the years, supplementing his three-volume manuscript of 282 nights (acquired in 1701) at first with Sindbad, which he had translated before launching his Nights project, then with stories (Qamar al-Zaman and perhaps Ghanim, in Mahdi's view) from an unidentified manuscript in his possession, and finally with stories told to him by Hanna, a "Maronite from Aleppo" whom he met in 1709. By a coincidence worthy of the Nights the Galland manuscript (G), whose putative shortfall of nights set in train fraud after fraud, "includes what all knowledgeable students of the Nights have considered the original core of the work as we know it today," a core that Mahdi regards (p. 8) as a "collection of relatively homogeneous stories that betray the hand of a consummate storyteller," composed "during the thirteenth and fourteenth centuries." G, most likely dating from the fourteenth century, remains to this day the oldest and best witness to the "original core." (This reader's question raised in JAOS 107 [1987]: 182, as to whether the mention of an ashrafi dinar might not necessitate a later terminus post quem is answered by the picture of an ashrafi dinar dated 690/1291, printed opposite the title page.)

As the history of the Syrian family of manuscripts continues, the Bibliotheque du Roi becomes the scene of crafty goings-on. Enter Chavis (Shawish, fl. 1780s, in Paris) and Sabbagh (d. 1816, in Paris) whose manuscripts, despite hints of doubt, have been accepted as genuine (e.g., H. and S. Grotzfeld, Die Erzahlungen aus "Tausendundeiner Nacht" [Darmstadt 1984], 33-37). In vol. 2 of this work, Mahdi argued that collation shows these manuscripts to be forgeries, somewhat clumsy in the first instance, made with a wily brilliance in the second. Where the Chavis MS (Ch) goes beyond G, Chavis was either translating from Galland's French, (or, in the case of Zayn al-Asnam, from the French translation from the Turkish made by Petis de la Croix and passed off by Galland's publisher as Galland's) or transcribing stories contained in a manuscript available to him (now BN 3637). Sabbagh's manuscript (S), allegedly a copy of a Baghdad manuscript (that has never come to light), would represent a unique instance in the Syrian branch of a recension with a full count of 1001 nights. Features that to some scholars indicate the authenticity of the manuscript indicate, to Mahdi, Sabbagh's cleverness in covering his tracks. So for instance, for the Grotzfelds the fact that S copies the colophon of its exemplar excludes the possibility of its dependence on G (Die Erzahlungen, 36). Mahdi, as elegantly wily here as Sabbagh may have been, suggests that the fabricated colophon, "attesting to the name of the scribe and the date and place of copying, namely, Baghdad on 10 Jumada II, 1115/21 October 1703, was

meant to show...