Essay Review: Medieval Mechanical Devices: The Book of Knowledge of Ingenious Mechanical Devices: Kitāb fī maārifat Al-Ḥiyal Al-Handasiya by Ibn Al-Razzāz Al-Jazarī

Wiedemann E. Hauser F. , “Über die Uhren im Bereich der islamischen Kultur”, Nova acta (Abh. der Kaiserl. Leop.-Carol. Deutschen Akademie der Naturforscher) , Band C, Nr. 5 (Halle, 1915), hereafter abbreviated as WH1.

Three works on Islamic clocks that have appeared in the past few years have been omitted from Dr Hill's bibliography. The first is the study by Kennedy E. S. Ukashah W. , “The Chandelier Clock of Ibn Yūnis”, Isis, lx (1969), 543–5 (see also WH1, 18), in which the authors discuss the candle-clock described in a unique manuscript in the College of St Joseph, Beirut and attributed, falsely, I think, to the celebrated tenth century Egyptian astronomer Ibn Yūnus. al-Jazarī describes a different candle-clock which he attributes to an individual called Ibn Yūnus al-Aṣṭurlābī (WH-, 156) or Yūnus al-Aṣṭurlābī (Hill, p. 87). Two manuscripts of al-Jazarī's treatise which I have consulted (MSS Istanbul Topkapi A3350 and A3472) have respectively Yūnus al-Aṣṭurlābī and Yūsuf al-Aṣṭurlābī. The second is S. Tekeli's edition, translation, and commentary of the treatise on clocks entitled al-Kawākib al-dawrīya … by the sixteenth century Ottoman astronomer Taqī al-Dīn ibn Ma^crūf, published in Ankara in 1966 as “The clocks in [the] Ottoman Empire in [the] sixteenth century …” and reviewed by Olaf Pedersen in Centaurus, xvi (1972), 328. See also ref. 8 below on another such treatise by Taqī al-Dīn. The third is Derek J. de S. Price's paper “Mechanical water clocks of the fourteenth century in Fez, Morocco”, Proceedings of the Tenth International Congress of History of Science (Ithaca, 1962), 599–602. One could also mention Baillie G. H. , “The first public clock: Damascus, circa 1150”, The horological journal, lxxvi (1934), 108–14, based on Wiedemann–Hauser's translation of Riḍwān's treatise.

Suter H. , “Die Mathematiker und Astronomen der Araber und ihre Werke”, Abhandlungen zur Geschichte der mathematischen Wissenschaften, x (1900); Krause M. , “Stambuler Handschriften islamischer Mathematiker”, Quellen und Studien zur Geschichte der Mathematik, Astronomie, und Physik, Abt. B, iii (1936), 437–532; and Brockelmann C. , Geschichte der arabischen Litteratur (2 vols, 2nd ed., Leiden, 1943–49; Supplementbände, 3 vols, Leiden, 1937–42). See Krause , op. cit., 491, and Brockelmann , op. cit., vol. SI, 902 on al-Jazarī, and also ref. 4 below. Brockelmann lists a second work on clocks attributed to al-Jazarī on which it would be interesting to have more information, if only that it is yet another copy of al-Jazarī's major work.

It is a pleasure to thank the Director of the Topkapi Library for the generous facilities afforded me there in July 1975. Photographs of MSS Topkapi A3350 and A3472 are also preserved in the Egyptian National Library in Cairo, catalogued as 486 and 487 in the riyāḍa (= mathematics) collection. The first Topkapi manuscript is a very beautiful copy dated 863 Hijra (= 1459), and lacks the first page or two. The second Topkapi manuscript is probably the earliest extant copy of al-Jazarī's treatise (see the description by Stchoukine I. in Gazette des Beaux-Arts, xi (1934), 134–40) but the date of copying given in the colophon does not seem to relate to this particular manuscript. This colophon, which is not entirely clear to me, reads (pious phrases have been omitted): and translates as follows (cf. the colophon of the Bodleian manuscript translated in Hill, p. 206): “This copy was made from another copied from the handwriting of the author. The lettering and the passages substituted for certain original statements [al-abdāl?], as well as the drawings of the pictures of the [different] categories [of mechanical devices] were what he had marked as being accurate and had drawn in his own hand, may God Almighty have mercy on him. That was in the year 602. … The completion of his copy was at the end of Sha^cbān in the year six hundred and two [about 10 April 1206], by the hand of … Muḥammad ibn Yūsuf ibn ˓Uthmān al-Hiṣkafī …”. In the Bodleian manuscript (Hill, p. 206) it is stated that al-Jazarī completed his treatise on Jumādā II 4, 602 Hijra (= 16 January 1206). Now the copyist of the Topkapi manuscript invokes God's mercy on al-Jazarī, and this means that al-Jazarī was deceased. I am inclined to think that the end of Sha˓bān, 602, which was about three months after Jumādā II 4, 602, refers to a copy by al-Hiṣkafi from which our manuscript was made.

On three minor points of notation used by Dr Hill. Rather than ‘constant’ hours and ‘solar’ hours (p. 6), why not ‘equinoctial’ and ‘seasonal’ hours? The medieval Arabic alphabetical (abjad) notation for numbers (p. 7) has been discussed by Irani R. A. K. , “Arabic numeral forms”, Centaurus, iv (1955), 1–12. A system of transliteration for the letters in medieval Arabic diagrams (p. 240) has been advocated by Hermelink H. Kennedy E. S. , “Transcription of Arabic letters in geometrical figures”, Journal of the American Oriental Society, lxxxii (1962), 204.

The following notes on two manuscripts that I have examined recently may be of use in future studies. Firstly, the manuscript numbered A3474 in the Topkapi Library appears to be another copy of the treatise by the Banū Mūsā that was not previously known to exist (cf. Hill , p. 9). The manuscript, which can be dated to the thirteenth or fourteenth century, lacks the first page or two and the last few pages. Descriptions and carefully drawn diagrams are given for a set of hydraulic devices numbered from 2 to 90. I cannot be certain that this is indeed a copy of the Banū Mūsā treatise since I am not familiar with any of the other manuscripts of their work. Secondly, Dr Hill (p. 11) mentions MS Gotha Forschungsbibliothek A1348, dated 961 Hijra (= 1554), as the only known copy of the treatise on clocks by Riḍwān. MS Istanbul Köprülü I.949 is another copy of Ridwān's treatise copied by the Egyptian scholar Bilīk ibn ˓Abd Allāh al-Qipjāqī in Cairo in 658 Hijra (= 1260) from the author's original manuscript. The existence of the Köprülü manuscript was not known to Wiedemann–Hauser but was noted by Krause (op. cit., 490), and this copy is also listed by Brockelmann (op. cit., vol. SI, 866). Two sets of photographs of this manuscript are preserved in the Egyptian National Library in Cairo, catalogued as no. 890 in the mīqāt (= astronomy) collection and no. 488 in the riyāḍa (= mathematics) collection. The text and diagrams are carefully copied, and al-Qipjāqī adds a comment of his own about Riḍwān that differs little from Hill's comment about him made seven centuries later, namely, that Riḍwān “may have been skillful with clocks, but he was not much good at writing Arabic”.

The introduction to WH1 constitutes a mine of information and references on this subject, which has yet to be properly explored.

I am not aware of any studies on the section on mechanical devices in the thirteenth century Andalusian Libros del Saber, but should not be surprised to learn of such a study by Wiedemann and/or Hauser. Another work on mechanical devices that has not been studied yet is a second treatise by Taqī al-Din (see ref. 2 above) entitled al-Ṭuruq al-sanīya fī l-ālāt al-rūḥānīya. A manuscript of this exists in the Library of Kandilli Observatory near Istanbul, and a set of photographs of an autograph copy dated 1552 from a private collection in Egypt is preserved in the Egyptian National Library in Cairo, catalogued as no. K3845. It is perhaps worth mentioning that the Egyptian astronomer Ibn Abī l-Fatḥ al-Ṣūfi, who copied the Bodleian manuscript of al-Jazarī, also wrote a short treatise on clocks (cf. WH1, 10), extant in MSS Cairo (Egyptian National Library) mīqāt 1169, 7 and Istanbul Fatih 5397, 4. Of considerably greater importance is a manuscript preserved in the Biblioteca Medicea Laurenziana in Florence, which may be regarded as a new major primary source for the history of Islamic technology. This manuscript, numbered 152 (formerly, 282), is catalogued as anonymi tractatus de mechanicis, but its significance has not yet been indicated in the modern literature on Islamic science. The work that concerns us here is entitled Kitāb al-asrār fī natā'ij al-afkār, “The book of secrets about the results of thoughts”, and occurs amid other mathematical treatises attributed to Abū ^cAbd Allāh known as Ibn Mu^cādh, who worked in Cordova in the eleventh century. (On Ibn Mu^cādh see the article ‘al-Jayyānī’ in the Dictionary of scientific biography.) The manuscript was copied in a clear Maghribi hand in the year 664 Hijra (= 1266), and is carefully illustrated. It is now unfortunately in bad shape, several of the one hundred pages being torn. There are thirty mechanical devices described and illustrated in the text. The thirtyfirst device is a horizontal sundial, and the treatise ends with statements on the solar meridian altitude at Cordova throughout the year and on the qibla or direction of Mecca at Cordova. A detailed study of this work would be worthwhile.