| Home | Inventories | Courses | Reference | Pre-prints | E-publics | In the news |

Islamic manuscripts: E-publications

Jan Just Witkam, The oldest known dated Arabic manuscript on paper (dated Dhu al-Qa`da 252 (866 AD).


This is MS Leiden Or. 298. Arabic, paper, 241 ff., upright script (with application of ihmal), dated Dhu al-Qa`da 252 (f. 241b; 866 AD), and thereby probably the oldest dated Arabic manuscript on paper, bound in a full-leather standard Library binding.

The volume contains an incomplete copy of Gharib al-Hadith, by Abu `Ubayd al-Qasim b. Sallam al-Baghdadi (d. 223/837), GAL G I, 107. See Voorhoeve, Handlist, p. 95. The present MS has been used as MS No. 3 by Muhammad `Azim al-Din in his edition of the text: Gharib al-Hadith li-Abi `Ubayd al-Qasim b. Sallam al-Harawi. Hydarabad 1384-1387/1964-1967 (4 vols.). On p. xvi of vol. I, he gives a short note about the importance of the MS.

See on the reception of this work also Gérard Lecomte, `A propos de la résurgence des ouvrages d`Ibn Qutayba sur le Hadit aux VIe/XIIe et VIIe/XIIIe siècles. Les certificats de lecture du K. Garib al-Hadit et du K. Islah al-Galat fi Garib al-Hadit li-Abi `Ubayd al-Qasim b. Sallam`, in BEO 21 (1968), pp. 347-409.

See G.W.J. Drewes (a.o.), Levinus Warner and his Legacy (Leiden 1970), No. 57, pp. 75-76, with reproduction of a page of MS Leiden Or. 298. The leaves are in disorder and have been as long as the manuscript has been in Leiden. See for a reconstruction of the correct order of the leaves by M.J. de Goeje, see CCO 1725 (IV, p. 49):

- guz' 9: ff. 1-20.
- guz' 10: ff. 21, 160, 161, 22-27, 116, 117, 28-33, 162, 163, 214.
- guz' 11: ff.155-157, 217, 218, 110-115, 210-213, 215, 216, 158, 159,206.
- guz' 12: ff. 207-209, 187-203.
- guz' 13: ff. 204, 205, 89-104, 219, 220.
- guz' 14: ff. 221-240.
- guz' 15: ff. 105, 73-88, 106, 107.
- guz' 16: ff. 108, 109, 134, 118-133, 164, 165, 186.
- guz' 17: ff. 166-185.
- guz' 18: ff. 135-154.
- guz' 19: ff. 34-53.
- guz' 20: ff. 54-72, 241.

Images from the manuscript: f. 2b f. 7a f. 239b f. 241b

Additional impressions of the MS:

• Numerous marginal and interlinear notes. Substantial marginal note on f. 131a. Some of these may have been written by the Musahhih who mentions his name on f. 241b.
• On f. 107b is a Khutbat al-Nikâh, in an old hand, albeit much younger than the copyist's, possibly the same hand as of the owner's note on f. 241b, which is dated 750 AH.
• On f. 241b a collation and reading (Tashîh, Qirâ'a) note, done with the imâm Ibrahim b. [...] Muhammad b. Ahmad b. Muhammad b. Ishaq al-...i, dated Monday with 6 days left of Gumâda II 378 (988 AD).
• On f. 241b an owner's note by Manuchihr b. Khusraw b. [...], Madinat al-Salam Baghdad in the year 750 AH.
  

The manuscript is divided in agzâ' of 20 ff. each, with a title page and with an empty last page for each quire. These were apparently made for the purpose of recording protocols of reading sessions (Samâ`at). But there no notes or protocols concerning these reading sessions on the title-pages and the blank final pages, except for a few sets of owners' notes. Each quire begins with a basmala of its own.

The question of the manufacture of the paper is somewhat confusing. Joseph von Karabacek, Arab paper, 1887. Translated by Don Baker and Suzy Dittmar. Additional notes by Don Baker. London [ca. 1991], p. 53, says 'To produce papers with both surfaces smooth and suitable for writing, two sheets were stuck together', for which he refers to ample evidence in the Vienna collection. Van Koningsveld (see below) copies this opinion. Documentary evidence for the doubling of paper is given by von Karabacek on the basis of a passage in al-Muqaddasi, Ahsan al-Taqasim fi Ma`rifat al-Taqalim. However, when one reads this passage in the edition by M.J. de Goeje (first edition, Leiden 1877), p. 100, lines 5-6, it becomes clear that von Karabacek has misinterpreted al-Muqaddasi's passage refers which to the pasting of papers (said in a general way, meaning either pasting together or into scrolls, but not specifically meant as pasting two sheets on top of one another) and to the double pasting of boards. The entire quotation is nothing more than as follows hereafter, and is taken out of a context concerning the use of starch, not the manufacture of paper: wa-bil-Yaman yalzaqûna al-Durûg wa-yabattinûna al-Dafâtir bil-Nashâ'). 

Jonathan M. Bloom, Paper before print. The history and impact of paper in the Islamic world. New Haven & London 2001, pp. 58-60, equally disagrees with the theory that old paper consists of two leaves pasted together. He says (pp. 58, 60): 'The tendency for the pages to split is actually a result of delamination, a condition seen in many early papers, such as the Vatican manuscript. When the pulp was not sufficiently beaten, the outer layers of the cellulose fibers did not detach and form physical and chemical bonds with adjacent microfibrils, and the resulting paper has weak internal cohesion. This condition was exacerbated when the paper was given a hard surface with the application of size. The weaker interior splits easily in two, revealing a rough, woolly, and feltlike inner surface.' After having studied several of the wormholes in the paper I can personally confirm that the cross-section of the paper shows that the substance does not consist of two layers, pasted onto one another.

Bloom (op. cit., p. 58) mentions an Arabic manuscript on paper which is dated 848 AD, and which is kept in Municipal Library of Alexandria. Earlier reports on an older MS in Alexandria have been circulating for a number of years. However, François Déroche, 'Les manuscrits arabes datés du IIIe/IXe siècle', in Revue des Études Islamiques 55-57 (1987-1989), pp. 343-379, mentions the present MS as the oldest Arabic manuscript on paper (list on p. 346), and nowhere does he refer to a MS in Alexandria.
See for other details on the MS, and for an edition and annotated translation of the Hadith Umm Zar` (ff. 12b-13b) on the basis of the present MS: M.J. de Goeje, 'Beschreibung einer alten Handschrift von Abû `Ubaid's Garîb al-hadît' in: ZDMG 18 (1864), pp. 781-807, with illustration of the upper half of f. 23b. De Goeje has also made a list of paleographical peculiarities on the script as employed in the present MS. See also Goed gezien, p. 36, with illustration of ff. 135b-136a.
P.S. van Koningsveld has given a description of the present MS in Levinus Warner and his legacy. Three centuries Legatum Warnerianum in the Leiden University Library. Catalogue of the commemorative exhibition held in the Bibliotheca Thysiana from April 27th till May 15th 1970. Leiden 1970, pp. 75-76, parts of which are quoted here: 'The oldest dated Arabic manuscript on paper in Europe. The manuscript, which is dated 252 H (= A.D. 866) contains a big part of the famous work Gharibu 'l-Hadith by Abu `Ubayd al-Qasim ibn-Sallam (837 A.D.). The Arabs adopted the art of paper-making from the Chinese in the eighth century (about 751 A.D.). This paper was probably made in one of the paper-mills of Samarkand or Bagdad. An older manuscript on paper is perhaps the Greek Codex Vatican. Graec. 2200. Apart from the text, which gives clear explanations on the unusual words in the traditions of Muhammad and a number of his Companions, the codex is of extreme importance for the knowledge of the development of Arabic script. [...] (follows a quotation from de Goeje's article). Literature: CCO IV, 49; D. A. Felix: 'What is the oldest dated paper in Europe?', in: Papiergeschichte, Dezember 1952; [...]; M. A. Thompson: An Introduction to Greek and Latin palaeography, p. 35 and facs. 52, Oxford 1912; B. Moritz, Arabic palaeography. A collection of Arabic texts from the first century of the Hidjra till the year 1000. Cairo 1905, plates pl. 119-120.'


Colophon: Originally published on 20 June 2002 in the (now obsolete) site of the former Oriental department in Leiden University Library. Reviewed and republished on January 6, 2007. Last update: January 11, 2007.