Al-Jazarī’s Compendium of Ingenious Devices: A model of representing and communicating technical knowledge in a medieval Islamicate context

Constantin Canavas

p. 71-82

Résumés

Le Livre sur la Connaissance des Appareils Merveilleux d’Al-Jazarī : un mode de représentation et de communication du savoir technique dans le contexte musulman médiéval.
En 602 H./1206 après J.-C. al-Jazarī publia son Livre sur la Connaissance des Appareils Merveilleux à Diyarbakir en Anatolie. Le traité copieusement illustré décrit principalement des horloges et autres automates pneumatiques-hydrauliques-mécaniques. Dans sa préface l’auteur évoque comme motivation de publication le désir de son seigneur d’obtenir une documentation sur les appareils merveilleux mais fragiles qui pourrait survivre dans le temps. On peut arguer que le traité d’al-Jazarī’s balance entre une description pratique (comment faire fonctionner) et l’ambition de procurer un manuscrit regardé par le gouverneur Artuqid comme un objet prestigieux. Les riches illustrations des manuscrits allaient jouer un rôle très important dans la réception du traité.

In 602 H./1206 C.E. al-Jazarī published in Diyarbakir, Southeast Anatolia, his Compendium of Ingenious Devices. The richly illustrated treatise describes mainly clocks and other pneumatic-hydraulic-mechanic automata. In his foreword the author mentions as motivation for the publication the desire of his sovereign to have a documentation of the marvellous but fragile devices that would survive in time. The study argues that al-Jazarī’s treatise balances between his practical description of how-to-make-it-work and the fulfilment of the ambition of the sponsor of the manuscript who presumably regarded the content and the illustrations as an object of prestige. The copious illustrations played an important role in the later reception of the technical treatise.

Entrées d’index

Mots-clés : Al-Jazarī, automates, horloges hydrauliques, illustrations, manuscrits arabes

Keywords : Al-Jazarī, Automata, Hydraulic Clocks, Illustrations, Arabic Manuscripts


Texte intégral

Introduction

1In 602 H/1206. CE al-Jazarī published in Diyarbakir, Southeast Anatolia, his treatise Kitāb fī ma‘rifat al-ḥiyal al-handasiyya (Compendium/The Book of Knowledge of Ingenious Mechanical Devices). The treatise describes mainly elaborate constructions of clocks and other automata on a pneumatic-hydraulic-mechanic background. In his foreword the author mentions the motivation of publishing the treatise: His sovereign, the Artuqid prince Nāṣir al-Dīn Maḥmūd, preoccupied by the fragility of the marvellous devices, ordered a documentation that would survive in time. Thus, a major medieval Arabic technical treatise was composed – accompanied by rich illustrations, which were often reproduced in the subsequent centuries.

2The present study focuses on the publication and diffusion of al-Jazarī’s treatise as programme of representation and transmission of technical knowledge in an environment of medieval Islamicate societies since the 7th century H./13th century CE. The term Islamicate will be used following the considerations of Marshal Hodgson1 in order to encompass the vast spectrum of ethnic, religious and linguistic diversity which determines the historical, cultural and intellectual frame in which medieval science and technology was developed in the Middle East under Muslim rulers and under the dominant influence of Islamic culture.

3The study argues that al-Jazarī’s treatise balances between his practical description of how-to-make-it-work and the ambition of the sponsor or the future owner of the manuscript who regarded the content and the illustrations as objects of prestige. The text is characterised by a formalisation of representing technical knowledge. This formalism includes an overview, a functional description, a detailed description of the construction, and practical recommendations regarding pitfalls, failure modes and alternative construction materials for each automat. The present study explores the relationship between al-Jazarī’s treatise and the Arabic tradition of automata literature, and focuses on the practitioner’s perspective of al-Jazarī. Presumably the copious colourful illustrations were later perceived not as explanatory technical figures, but rather as precious examples of manuscript production – an effect which became a motive but also a limiting condition for the diffusion of the treatise as technical book.

The Compendium of Ingenious Devices and its author

4Modern perception of technological achievements in the medieval Islamic world is marked by the name al-Jazarī and his treatise The Book of Knowledge of Ingenious Mechanical Devices, published at the beginning of the 13th century CE (beginning of the 7th century of the Islamic calendar). Although no archaeological evidence in connection with his detailed descriptions is available, the illustrations and the descriptions of spectacular devices from his book have become symbols of the subsequent interest in the field of medieval technology in the Islamic world.2 The present study focuses on the publication and diffusion of al-Jazarī’s treatise by considering it as a programme of representation and transmission of technical knowledge in an environment of medieval Islamicate societies since the 7th century H./13th century CE.

5The general context of the medieval Arabic book in its manuscript form cannot be analysed here.3 Even if books with technological subjects occupy a comparatively limited space in modern overviews of the medieval Arabic literature, al-Jazarī’s treatise is mentioned very frequently.

6The Book of the Knowledge (Compendium) of the Ingenious Mechanical Devices (Kitāb fī ma‛rifat al-ḥiyal al-handasiyya) by Badī῾ al-Zamān Ibn al-Razzāz al-Jazarī, was published–according to the information given in the colophon – in 602 H./1206 CE. The Arabic text was reproduced in a modern critical edition by A. Y. al-Hassan.4 An annotated English translation was published by D. R. Hill.5 In his introduction, al-Jazarī informs us that his sovereign, the Artukid prince of Amid (today Diyarbakir) Nāṣir al-Dīn Maḥmūd commissioned the treatise. It describes marvellous automata in the court of the Artukid ruler. In his introduction to the English translation of al-Jazarī’s treatise D. R. Hill mentions 11 extant manuscripts and some fragments (including separate illustrations) of two mutilated manuscripts scattered in several libraries and collections.6 Some years later, in 1981, Hill publishes an updated list including 14 manuscripts.7

7Very little is known about the person and the life al-Jazarī. We practically have no further information about him, unless his own comments in the introduction of his compendium in conjunction with the Artukid rulers, in whose service he claims that he has been for twenty-five years at the time of Saladin and his Ayyubid successors.8

Technological treatises in a medieval Islamicate context up to al-Jazarī

8Al-Jazarī constitutes a major reference point in the horizon of medieval technological treatises in an Islamicate context.9 One of the considerations of the present study is the grade and the way in which this horizon is reflected in al-Jazarī’s treatise. In his introduction to the treatise, al-Jazarī refers anonymously to earlier scholars and later craftsmen. In his introduction to the English translation of al-Jazarī’s treatise D. R. Hill recapitulates authors and Arabic treatises that touch or elaborate issues to be found in al-Jazarī’s work. According to Hill, the oldest ones should be the works attributed to Pseudo-Archimedes dating probably from the end of the 8th century CE. They include a famous hydraulic-mechanic clock – similar to the ones described by al-Jazarī, who explicitly mentions Archimedes.10

9An Arabic treatise with similar title and content as the book of al-Jazarī is the Kitāb al-ḥiyal (Book of Ingenious Devices) by Banū Mūsā (9th century CE).11 It describes some 100 “marvellous mechanical devices,” and is mentioned explicitly by al-Jazarī, although mostly with a critical tone.

10Further names are also mentioned in conjunction with specific devices (e.g. musical instruments).

11The extant manuscripts of al-Jazarī and of Banū Mūsā yield a challenge for a comparative study not only of the textual references, parallelisms etc., but also of the iconographical material. Such a comparative iconography should consider the hypothesis of iconographical cycles for the illustration programme in both treatises, with the possibility of a common provenance or of similarities (or differences) during the copying procedures. The comparison could also encompass the status of the authors in the social environment of the court. The brothers Banū Mūsā were active participants of the Abbasid caliphate court in 9th century CE Baghdad, whereas al-Jazarī served under several Artukid rulers in Amid (today Diyarbakir) in the periphery of the declining Abbasid caliphate, but in the core of the Seljuk realm at the end of the 12th /beginning of the 13th century CE. The differences in the social conditions might have influenced in different ways the emergence of both technological treatises – the subsequent copying, and the later transmission practice, however, might have brought both works closer together. Undoubtedly, both issues – the iconographical comparison and the social framing of transmission paths and strategies – deserve a much more thorough analysis than the one for which the present study wants to yield a possible and plausible outline.

12Hill presents in his introduction later authors of relevant Arabic treatises not explicitly mentioned by al-Jazarī: Al-Khuwārizmī, whose Mafātīḥ al-‘Ulūm (The Keys of the Sciences) was published in 975-976/991-992 CE, and Riḍwān al-Sa‘ātī, who describes a hydraulic-mechanic clock at the Bāb Jayrūn, Gate of Damascus Mosque (1203 CE).

13The question concerning the explicit or implicit influence of Greek-Hellenistic works translated into Arabic is more complicated. Arabic translations of the works of Philon (ca. 200 BCE) and Heron (1st century CE) on pneumatics, hydraulics and automata are mentioned in Arabic bibliographies12, but these authors are not explicitly mentioned by al-Jazarī. In fact, he only mentions Apollonius the carpenter, “the Indian” – probably a corruption of “the geometrician” in Arabic, an attribute of Apollonius of Perga, as D. R. Hill has suggested in his introduction to the translation of al-Jazarī’s treatise.13 It should be noted that pseudepigrapha in the transmission lines, as well as recompilations of manuscript folia during copying render the answering of questions concerning the influence lines of the translated Greek-Hellenistic works extremely difficult. F. D. Prager gives a typical example of such confusing transmission lines in his description of the transmission line of Philon’s Pneumatica.14 A whole chapter (Chapter 30) in Philon’s Arabic version in the edited compilation seems to be an adaptation of a similar device description from al-Jazarī’s treatise inserted posteriorly (probably during the 14th century CE) into the Arabic copy of Philon’s treatise.

14Similar difficulties arise also whenever illustrations from different books are compared in an effort to trace directions of influence among the different books and transmission lines among the different copies – an issue that goes beyond the scope of the present study and cannot be treated here.

Contents and structure of al-Jazarī’s book

15In his introduction al-Jazarī explains his motives, and refers to past scholars. He divides his treatise into six categories (naw‘). In Category I water-clocks and candle-clocks are treated in ten chapters – each chapter being dedicated to one device. Category II comprises vessels and pitchers suitable for drinking sessions. It is also divided in ten chapters. Category III is divided in ten chapters, and contains pitchers, basins and other devices for hand-washing – even one supposed to control automatically the amount of blood soaked during phlebotomy. The next Category IV treats fountains and musical automata in ten chapters. Category V treats machines for raising water in five chapters. The last Category VI comprises five chapters with miscellaneous constructions (such as doors, etc.).

16In analysing the programme of al-Jazarī when representing technical knowledge, it is significant to study the structure of the single chapters. Indeed, al-Jazarī’s formalisation of representing the technical devices – one chapter for each device – has a more or less fixed structure: at the beginning of the chapter he presents an overview of the device, both in text and illustration. There follows a functional description of the device as a whole, followed by detailed functional descriptions of parts of the device in several sections. Finally, al-Jazarī provides the reader with practical recommendations for each case (pitfalls, possible failure causes, alternative materials in case of functional failures, etc.).

17The textual descriptions are generally accompanied by illustrations explicitly mentioned and considered in the text. Illustrations encompass overview, as well as details of the devices treated. These elements will be exemplified in the following description of the 1st Castle Clock.

An example: The 1st Castle Clock

18The first chapter of Category I treats in ten sections a water-clock (binkām), designed to indicate “solar hours” – in fact, twelve equal daylight and twelve equal night divisions. In the first section he refers to previous clocks attributed to “the excellent Archimedes” and one predecessor (al-Qajān?) concerning the divisions of a semicircle or a circle corresponding to the Zodiac. The textual description of three versions of the Zodiac representation is accompanied by three explanatory illustrations. The intention of al-Jazarī and of his predecessors is, obviously, to correlate the measuring of the solar hours with astronomical phenomena, as those observed and calculated by means of astrolabes.15 After criticising these models of Archimedes as incorrect, al-Jazarī describes the appearance of the clock comparing it with a house (or castle), and giving approximate dimensions (“about twice the height of a man”). He then describes the outside appearance of the castle-like construction with a semicircle at the top comprising six visible Zodiac signs – the other six invisible signs being on the semicircle underneath the horizon. Beyond the gate there are several small doors (or windows) with their leaves, which open and mark the passing of the equal “hours” of day (12) and night (12). Further, he describes the visible effects of other devices marking the passing of two, three or six “hours,” the most spectacular perhaps being the five artificial musicians – two drummers, two trumpeters and a cymbalist – situated in front of the main gate, who play their instruments for a while at the sixth, ninth and twelfth “hour,” as well as the movements of two artificial falcons and the falling of balls from their beaks.16 The general description is completed by an overview coloured illustration, which, in the case of the manuscript Topkapı Serai No 3606 in Istanbul, is on a folio which has been taken away from the original manuscript and is nowadays part of the Goloubew Collection at the Museum of Fine Arts in Boston (plate vi).

19The second section is dedicated to the functional description of two major components: the water reservoir (which is the source of the motive power of the whole device), as well as the requisite water equipment comprising the float chamber, the float and the tap. The translator of the treatise, D. R. Hill, has also meticulously reconstructed the Castle Clock (and several other devices as well which are described in al-Jazarī’s book) on the basis of al-Jazarī’s descriptions and drawings. Summarising his experience he comments: “The text exemplifies admirably the main characteristics of al-Jazarī’s descriptive abilities: a careful description of the processes of manufacturing, emphasising the really significant points, and a clear presentation of the functioning of the individual items, both singly and in relationship to one another.”17

20The third section describes the design and the construction of the flow regulator (dastūr) in conjunction with the current position of the sun in the Zodiac circle. This conjunction is crucial in order that the device allows for different flow times for defining the varying elapse of time for the twelve day and the twelve night equal divisions (“hours”) at the different seasons of the solar year.18

21In the fourth section al-Jazarī describes the assembly and the positioning of the components already described in the previous sections (water reservoir, float chamber, float, tap, and flow controller).19

22The fifth section explains how the duration of the water-flow corresponding to each “hour” in daylight and in the night varies during the year.20 In the next five sections al-Jazarī describes in detail the construction of the clock building, as well as the mechanisms for imparting motion to the moving parts (e.g. figures of the musicians). He provides with precise details of the materials used and with practical recommendations concerning the methods of assembling and fitting parts together. At several places he indicates what the operator of the plant should do in order to start, synchronise, or adjust the whole device. At the end he gives suggestions for coating and colouring the components of the device.21

Variations

23Once he has established an operating mechanism, al-Jazarī suggests variations of the devices in several chapters inside each Category. Taking as example Category I (hydraulic clocks), we have a Castle Clock with 5 musicians in the first chapter, whereas in the second chapter a Castle Clock with 7 musicians (drummers and cymbalists) is treated. Chapter 7 describes the Candle-Clock of the Swordsman from which the passage of the constant hours can be told – like in the case of the Castle Clocks.

24The kind of detail variations on the basis of an initial frame pattern corresponds to similar exposition patterns of former treatises on ingenious mechanical devices – alike whether Greek/Hellenistic (i.e. in the Pneumatics of Philon and Heron) or Arabic (i.e. in the book of Banū Mūsā).

25Beyond such variations, al-Jazarī invokes the possibility that future craftsmen make their own decisions and choices, e.g. regarding the figures appearing mechanically according to the elapsing time.22

Rich illustrations and their tragic history

26Most of the 14 extant manuscripts of al-Jazarī’s treatise listed by D. R. Hill in 198123 are richly illustrated – both with drawings arranged with text of the same folio, as well as with folio-size colour illustrations (fig. 1). One extant manuscript, BN Arabe 5101 (Bibliothèque Nationale, Paris), is in fact without illustrations, however, the spaces for illustrations have been reserved in the process of copying.24

27Unfortunately, several of these illustrations do not accompany the original manuscripts any more. Long before scholarship in the field of the history of science and technology in Islamicate context directed a focus on the treatise of al-Jazarī, the richly illustrated manuscripts attracted the interest of art collectors. The result is that several folia with colour illustrations were removed from the original manuscripts to become parts of private or public art collections. This is the case e.g. of the illustration of the 1st Castle Clock in plate vi, originally part of the manuscript Hagia Sophia, 3606 in Sulemaniye U. Kütüphanesi, Istanbul, which is nowadays part of the Goloubew Collection at the Museum of Fine Arts in Boston (fig. 1). This manuscript has lost almost half of its main illustrations (23 out of 50); at least 15 of them could be traced in several collections.25 D. R. Hill reproduces in his English translation photographs of 32 of such folia with miniatures, which found their way into various private and public collections in Europe and in the USA. “These miniatures,” remarks Hill, “are well known to those concerned with Islamic Art, and a considerable literature has grown up around them.”26

Transmission and recension of the iconographic programme

28The colour illustrations of several manuscripts of al-Jazarī’s treatise have been commented by several art historians since the beginning of the 20th century. Especially those illustrations related to the overview of the devices (e.g. the monumental clocks) can be found in all compendia of Islamic art in general and of Islamic book illustrations in particular (e.g. by Richard Ettinghausen, Oleg Grabar and Robert Hillenbrand). The production of richly illustrated manuscripts of al-Jazarī’s treatise in the 13th century CE coincides with what Hillebrand called “the first golden age of Islamic book painting.”27 The production of book illustrations seems to explode in palaces of local rulers for reasons that are most probably related to ambitions and strategies of political representation and figurations of power – including ideological goals in re-writing genealogical stories, such as in the case of the Persian Shahnameh.28 Thus, the illustrations of al-Jazarī’s treatise should be considered in this political context.

29The aspect of representation of prestige and power through the monumental illustrations of several clocks (e.g. the elephant clock and the Castle Clock) distinguishes al-Jazarī’s manuscripts from those by the brothers Banū Mūsā. Both treatises include technical descriptions accompanied by explanatory drawings which may be traced back to models proposed by several authors of the Hellenistic period (Philon of Byzantium) and the Late Antiquity (e.g. Heron of Alexandria); however, even the technical drawings in al-Jazarī’s treatise are additionally characterised by a decorative as well as an “entertaining” ambition, as Müller-Wiener claims.29 A more detailed comparative study of the transmission of both iconographic programmes, that of al-Jazarī and that of the brothers Banū Mūsā, reveals however a more complex relation. In a study of the transmission of the iconography of the Banū Mūsā treatise as a commenting technique the author has traced later iconographic elements of explicitly decorative character which in some cases would contradict the textual information and imply a manifold (i.e. not simply functional) interaction between the distinct code systems, i.e. the textual and the iconographic. “The relative independence of iconographical codes, especially when the represented objects are taken from the contemporary everyday life (beakers, jars, pitchers, fountains), introduce a distance to the purely technical-functional status.”30

30A further claim is that the brothers Banū Mūsā have a more intellectual ambition than the practitioner (muhandis/engineer) al-Jazarī. Actually, al-Jazarī criticises Banū Mūsā in that several of their ingenious devices simply can’t work. On the other hand, many descriptions by Banū Mūsā can be read as exemplifications of the understanding of their time concerning the physical concept of the void as well as paths of coping with complexity of physical models. These aspects have been elaborated elsewhere by the author.31

31The “blurred borders between technology and sheer abstract concepts” (such as that of the “void”), as well as between “technology and art” in the illustrated manuscripts of al-Jazarī’s transmission have been recently traced from the perspective of history of science and technology by G. Saliba32. From the perspective of an art historian, C.-D. Haase considers al-Jazarī’s treatise as a recension of earlier Arabic versions on automata under the title The Book of Knowledge of Ingenious Mechanical Devices. He elaborates on the self-understanding of the author-muhandis in the 12th-13th century CE in the Islamicate context pointing out that “The creative process involved in the development of automata and their fantastical functions, consisting of aesthetic assignments and various mechanical operations that are put into effect by engineering experience, corresponds to the classical image of the artist as ‘poietes’.”33

What is the value of al-Jazarī’s treatise as a reference technical book?

32Considering the question of the reception of al-Jazarī’s treatise as a reference technical book in a posterior Islamicate context, the evidence does not allow yet a consistent answer. As already mentioned, the treatise has been laboriously copied since its first publishing in 1206 CE. A major feature of these copies is the set of colour illustrations – apparently a criterion of the high prestige attributed to the copied treatise. Among the 14 extant copies listed by Hill there is a Mughal copy of the 17th century CE, as well a Persian translation from as recently as 1874. However, there is no mention of the author, al-Jazarī, in standard Arabic biographical works. Besides, no reference to al-Jazarī’s treatise could be traced yet in later Arabic, Persian or Ottoman books. Modern literature on medieval Arabic engineering elaborates on al-Jazarī, without indicating further explicit continuities in the East.34 Whether the study of Mughal literature or of Chinese sources from the Yuan period could yield indications of explicit or implicit reception is still an open question. The Islamicate context of South and East Asia lies beyond the scope of the present study.

33Undoubtedly, the book of al-Jazarī enables us to assess the state of the art of mechanics in the Middle East at the transition from the 12th to the 13th century CE. As L. White Jr. remarks in his foreword to the English translation by D. R. Hill, some technological inventions attributed to European authors during the Renaissance (like e. g. Leonardo da Vinci) have already been described by al-Jazarī.35 Possible transmission paths towards the West, however, though quite plausible, still remain untraced.

Conclusion

34In assessing the state of the art of mechanics in the Middle East at the transition from the 12th to the 13th century CE through the treatise of al-Jazarī the modern scholar is confronted with a spectrum of ambitions of the actors involved in the production and the reproduction of the treatise. The explicit ambition of the author is the description of how-to-make-the-devices-work. On the other hand, the ambition of the alleged sponsor, the Artukid prince of Amid, and the subsequent owners of the copied manuscripts seems to be associated rather with the manuscript book itself, and – especially – with the rich illustrations which render the manuscript a prestige material object. As suggested above, the comparatively large number of the extant illustrated manuscripts of the treatise should be associated to the latter ambition.

35Beyond the horizon of technology, but still with respect to the content, the treatise implies affinity to two further Arabic literary genres: the ‛ağā’ib texts and the Books of Gifts and Rarities (Kitāb al-hadāyā wa’l-tuḥaf). The ‛ağā’ib literature treats descriptions concerning marvels of the world36 – mainly natural wonders and curiosities, but also artificial marvels, i.e. marvellous machines like the devices of al-Jazarī. However, no cross-references between ‛ağā’ib texts and al-Jazarī’s treatise have been traced. The Books of Gifts and Rarities are “light” readings with amusing and unusual anecdotes on precious gifts and collection objects,37 including marvellous machines like the devices of al-Jazarī. The most known example under this title presumably dates back to the 11th century CE, however, later books and references (e.g. by al-Maqrizī who died in 1441 CE) are well known too. No traceable evidence of a possible impact of al-Jazarī’s treatise on this book production is reported.

36Al-Jazarī’s treatise experienced a revival in the “orientalistic” perception of mechanics and art history in an Islamicate context at the end of the 19th/beginning of the 20th century. At that period, E. Wiedemann and Fr. Hauser. published German translations of singular devices, including descriptions and commentaries on the functionality of the depicted devices.38

37The dark side of the impact related to the fascination initiated by al-Jazarī’s illustrated manuscripts has been depicted above. This fascination continues in current intellectual debates. In a recent study, George Saliba refers to al-Jazarī’s clocks as “machines that represent more than one reality all at once”, as examples of transcending “blurred borders between technology and sheer abstract concepts” by introducing “a whole new dimension of musical production as time passes.”39 If such concepts and perceptions can be expected in a medieval Islamicate context, future revealing of evidence regarding al-Jazarī’s cross-over impact can come from quite unexpected fields.

Image

Plate. Front view of the 1st Castle Clock (Museum of Fine Arts, Boston, Goloubew Collection 14.533; folio from a mutilated manuscript dated 755 H./1354 CE, the remains of which are in the library of Sulemaniye U. Kütüphanesi, Istanbul, Hagia Sophia, 3606; dimensions 39.37 x 27.62 cm; ink, opaque watercolour and gold on paper).

Notes de bas de page

1 Hodgson M., The Venture of Islam: Conscience and History in a World Civilization, vol. 1, Chicago & London, 1974.

2 Recent repercussion of this phenomenon is the importance accorded to the reconstruction of al-Jazarī’s automata in controversially debated exhibitions intended for the large public, such as “1001 Inventions” documented in Al-Hassani, S. T. S. (dir.), 1001 Inventions: The Enduring Legacy of Muslim Civilization, Washington DC, National Geographic, 2012 (3rd ed.), as well as the exhibition in the Istanbul Museum for the History of Science and Technology in Islam documented in the overview catalogue published by Fuat Sezgin in 2010, p. 32, 107, 109, 127, 136-137. A central initiative in both exhibition projects consists in reconstructing and exhibiting devices described (and occasionally depicted) in extant Islamicate manuscripts – a controversial issue when no archaeological evidence supports the reconstruction. Al-Jazarī’s devices constitute a significant contribution to this endeavour; they are present in several thematic sections of the exhibitions and the catalogue editions.

3 For an overview on Arabic manuscript cultures see Déroche F., Le livre manuscrit arabe. Préludes à une histoire, Paris, Bibliothèque nationale de France, 2004; Déroche F. & Rossi V. S., I manoscritti in caratteri arabi, Roma, Viella, 2012.

4 Al-Hassan A. Y. (dir.), Kitāb fī ma‘rifat al-ḥiyal al-handasiyya. “The Book of Knowledge of Ingenious Mechanical Devices” by al-Jazarī, Aleppo, Institute for the History of Arabic Science, 1981.

5 Hill D. R., The Book of Knowledge of Ingenious Mechanical Devices (Kitāb al-ḥiyal). An annotated translation of al-Jazarī’s book on machines, Dordrecht, Reidel, 1974.

6 Ibid., p. 3-6.

7 Hill D. R., Arabic Water-Clocks, Aleppo, Institute for the History of Arabic Science, 1981, p. 89-91.

8 Hill D. R., The Book of Knowledge…, op. cit., p. 3; Hill D. R., “Al-Jazarī”, in Selin H. (dir.), Encyclopaedia of the History of Science, Technology and Medicine in Non-Western Cultures, Dordrecht, Kluver Academic Publishers, 1987, p. 472-473; see also the article on al-Jazarī by D. R. Hill in the Encyclopaedia of Islam (2nd ed.), Leiden, Brill, 2004, suppl., p. 266b-267a.

9 For an overview of medieval technology in Islamicate context see Al-Hassan A. Y. & Hill D. R., Islamic technology. An illustrated history, Cambridge, Cambridge University Press, 1986/1992.

10 Hill D. R., The Book of Knowledge…, op. cit., p 8-11.

11 Al-Hassan A. Y. (dir.), Kitāb al-ḥiyal. “The Book of Ingenious Devices” by the Banū (sons of) Mūsā bin Shākir, Aleppo, Institute for the History of Arabic Science, 1981. An English translation was published by Hill D. R. : “The Book of Ingenious Devices” (Kitāb al-ḥiyal) by The Banū (sons of) Mūsā bin Shākir, Dordrecht, Reidel, 1979. Parts of both texts, that of al-Jazarī as well that of Banū Mūsā, were eclectically chosen, translated into German, commented and published by E. Wiedemann at the beginning of the 20th century. See also Hauser Fr., Über das kitāb al-ḥijal (Das Werk über die sinnreichen Anordnungen) der Benū Mūsā, Erlangen, Max Mencke, 1922.

12 Heron’s Pneumatica were translated into Arabic by Qustā bin Lūqā in 864 CE. For the Arabic transmission of Heron see the critical edition by Schmidt W., Herons von Alexandria Druckwerke und Automatentheater, vol. I., Leipzig, Teubner, 1899 (repr. Stuttgart, 1976). For the Arabic transmission of Philon see the annotated and richly commented English translations of both Latin and Arabic versions by Prager, F. D., Philo of Byzantium: Pneumatica, Wiesbaden, Reichert, 1974. A critical edition of the Arabic text of Philon’s Pneumatica was published by Carra de Vaux B., Le livre des appareils pneumatiques et des machines hydrauliques, par Philon de Byzance, édité d’après les versions arabes d’Oxford et de Constantinople et traduit en français. Notices et extraits des manuscrits de la Bibliothèque Nationale et autres bibliothèques, Paris, 1903, 38, p. 27-235.

13 Hill D. R., The Book of Knowledge…, op. cit., p. 12.

14 Prager F. D., op. cit., p. 126.

15 Hill D. R., The Book of Knowledge…, op. cit., p. 241.

16 Ibid., p. 17-19.

17 Ibid., p. 242.

18 Ibid., p. 21-23.

19 Ibid., p. 23-24.

20 Ibid., p. 24-25.

21 Ibid., p. 26-41.

22 Ibid., p. 18.

23 Hill D. R., Arabic Water-Clocks, op. cit., p. 89-91.

24 Hill D. R., The Book of Knowledge…, op. cit., p. 5.

25 Hill D. R., Arabic Water-Clocks, op. cit., p. 89.

26 Hill D. R., The Book of Knowledge…, op. cit., p. 5.

27 Hillenbrand R., Islamic Art and Architecture, London, Thames & Hudson, 1999, p. 125.

28 Müller-Wiener M., Die Kunst der islamischen Welt, Stuttgart, Reclam, 2012, p. 240 ff.

29 Müller-Wiener M., “Vom irdischen Paradies zum höfischen Theater – Islamische Automaten und mechanische Konstruktionen des 9. bis 13. Jahrhunderts”, in EOTHEN, Jahresheft der Gesellschaft der Freunde Islamischer Kunst und Kultur IV, 2007, p. 160.

30 Canavas C., “Commenting Arabic technological treatises in illustrated manuscripts”, Proceedings of the 3rd International Conference of the Manuscript Center, Biblioteca Alexandrina, Alexandria (Egypt), March 2006. Alexandria (Egypt), Manuscript Centre, Bibliotheca Alexandrina 2009, p. 110.

31 Canavas C., “‘Stealing the water’ and commenting the void: Philosophical and technological crossroads between Iraq and Syria”, Proceedings of the Scientific Symposium “Historical and Scientific Relations between Iraq and Syria”, University of Aleppo, 23-24 April 2008, Aleppo, Syria, Aleppo University Publications, Institute for the History of Arabic Science 2009, p. 433-441; Canavas C., “From Philon of Byzantium to al-Ğazarī: Shifting of perspective in dealing with complexity”, in Proceedings of the 9th Symposium on the History of Arabic Science, 28-30 October 2008, Damascus, Aleppo, Syria, Aleppo University Publications, Institute for the History of Arabic Science 2009, p. 37-47.

32 Saliba G., “Blurred Edges – At the Intersection of Science, Culture, and Art”, in Zielinski, S. & Fürlus, E. (dir.), Variantology 4, Köln, W. König, 2010, p. 356-361.

33 Haase C.-P., “Modest Variations – Theoretical Tradition and Practical Innovation in the Mechanical Arts from Antiquity to the Arab Middle Ages”, in Zielinski, S. & Fürlus, E. (eds.), Variantology 4, Köln, W. König, 2010, p. 213.

34 Hill D. R., A History of Engineering in Classical and Medieval Times, London, 1984/1996.

35 Ibid., p. xiv.

36 See the article on ‘ajā’ib by Dubler, C. E. in EI2.

37 Al-ḥijjāwī al-Qaddūmī, Ghāda (transl. & annotated), Book of Gifts and Rarities (Kitāb al-hadāyā wa’l-tuḥaf), Cambridge (MA), Harvard University Press, 1996, p. 20.

38 E. Wiedemann published a series of articles on al-Jazarī’s treatise in the period 1876-1928.

39 Saliba G., “Blurred Edges – At the Intersection of Science, Culture, and Art”, in Zielinski, S. & Fürlus, E. (dir.), Variantology 4, Köln, W. König, 2010, p. 358.


Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.