From Frankfurt and Cairo to Damascus: Recent Models of the Umayyad Mosque Clock, The Umayyad Mosque Clock, Abdel Aziz al-Jaraki, Eilhard Wiedemann, Fritz Hauser, Fuat Sezgin, Donald Hill, Ridhwan al-Sa'ati, Banu Musa, Al-Jazari, Al-Khazini, history of clocks, Islamic technology
by Abdel Aziz al-Jaraki *
Table of contents
1. Setting the Stage
2. Clocks in Islamic Technology
3. Our study of Ridhwan al-Sa'ati's clock
4. The Components of the Damascus Clock
4.1 The Upper Level
4.2 The Lower Section
5. Critical Assessment of the Two Recent Models
5.1 Features of Islamic clocks
5.2 The Frankfurt Clock
5.3 The Cairo Clock
Figure 1: Dome of the clocks in the courtyard of the of Umayyad Mosque in Damascus. © Bernard Gagnon (31 March 2010). (Source).
About two years ago, a group of researchers in the field of Muslim Heritage in Cairo announced that they reconstructed a replica of the Umayyad mosque water clock. This replica was shown at a museum in Cairo, claiming that it is identical to the original Umayyad mosque clock, or what is known in heritage studies as the clock of Ridhwan al-Sa'ati (sa'at Ridhwan). These scholars claimed that their reconstructions were carried out according to the original Ridhwan al-Sa'ati's original manuscript.
Earlier, Professor Fuat Sezgin announced that he discovered the principles of that clock and built a sample of it in the eighties of the last century, which he displayed in his own museum in Frankfurt. He called this replica the clock of Ridhwan al-Sa'ati, meaning the Umayyad mosque clock. He also published photos of it in the German Arab World Magazine published in Arabic, as well as in the website of his institute.
These attempts, regardless of their authenticity and motives, come as a good step in explaining this device and its scientific significance, hence generating new interest and raising queries on the perceptions of these claims.
There were however, earlier serious attempts that enjoyed a different kind of credibility and followed scientific methodology in research. The first significant contributions were by two German academic scholars Eilhard Wiedemann and his partner Fritz Hauser. They published their work about this clock in 1915. Their work contained deep analytical study with engineering drawings to unravel, albeit with limited success, the mysteries of the Ridhwan al-Sa'ati manuscript. In 1984, Fuat Sezgin collected these essays and published them in German among the complete works of Wiedemann.
Later on, the English scholar and historian of Islamic technology, Donald R. Hill published the book Arabic Water Clocks with Arabic and English sections. The book was released in cooperation with the Heritage Institute of Aleppo University. In a chapter of the book, he spoke about the Umayyad mosque clock, where he reproduced some of the drawings of Wiedemann and Hauser.
The next contribution came from Sheikh Muhammad Ahmad Dahman who wrote about the Umayyad mosque clock, and edited the manuscript of Ridhwan al-Sa'ati, with some historical commentaries about the issue of the clocks in general, but without offering or adding anything new to what was already mentioned in the manuscript, nor in the geometric explanation or the mechanical illustration.
Figure 2: Floor plan and perspective drawing of Umayyad Mosque in Damascus, drawn by Saeed Arida (2003). © Nasser Rabbat, the Aga Khan Program for Islamic Architecture, MIT. (Source).
Prompted by these challenges, we collected everything that had been said or published about the Umayyad mosque clock, whether it was a description or a study, set out a strategy to unravel its mysteries in a scientific manner and without making assumptions that do not comply with a scientific reality or historical proof. In our study, we had to investigate the various mechanical works of Arab and Muslim scientists, from Banu Musa to Taqi al-Din ibn Ma'ruf.
Banu Musa bin Shakir in their famous Kitab al-Hiyal (The Book of Ingenious Mechanical Devices) written probably in the second half of the 9th century in Baghdad, and then the book of Al-Jazari Al-Jami' Bayn al-'Ilm wa al-'Amal al-Nafi' fi Sina'at al-Hiyal (A Compendium on the Theory and Practice of the Mechanical Arts), completed in 1206 CE in Diyarbakir, South-East modern Turkey. We find the next contribution in Kitab Mizan al-Hikma (The Book of the Balance of Wisdom) completed by Abu al-Fath Abdurrahman al-Khazini in 1121-22 CE. In the middle of the 16th century, Taqi al-Din ibn Ma'ruf wrote the original book Al-Turuq al-Sanniyya fi al-Alat al-Ruhaniyya (The Sublime Methods of Spiritual Machines).
With proper understanding of the early mechanical treatments by the Muslim scholars, we were able to develop an appreciation of various devices and tools used and hence enable us to finally recreate the clock.
Here, we find it necessary to simplify the subject of the above mentioned device and to describe it according to the manuscript made by one of the most prominent scholars who worked in running this device in the year 600 of Hijra / 1202 CE, that is Fakhr ul-Din Ridhwan bin Mohammad bin Ali al-Sa'ati al-Khurasani al-Dimashqi, who served as minister in the Ayyubid State.
The manuscript of his treatise in which he described the clock is located in the library Forschungsbibliothek in Gotha, Germany. A second manuscript is extant. It was copied from the original by the Egyptian engineer Baylak Abdullah al-Qabajaqi, some 56 lunar years after the first manuscript was written. A third manuscript is a copy made in Egypt in the time of Ahmad Taymur Basha; it was found in the Egyptian National Library. All three manuscripts are totally identical.
The Umayyad mosque clock was located on two levels (as described by Ibn Jubayr) on the right side of the eastern door of the Umayyad Mosque. It consisted in three basic parts. From the description mentioned in the original manuscript, we can describe the clock components as follows:
4.1 The upper level
The upper level contains:
The external visible part overlooking the passers-by is a large wood and copper board of approximate dimensions 240 x 240 centimetres. It contains the basic functions of the clock, which are the following (from bottom to top):
a) Brass doors, each door rotates around axes, on the surface of each of them there is a number that appears to the viewer after an hour has passed, and those, like the daytime hours, are twelve doors.
b) Below each door there is an indicator or pointer with a crescent on it which slides with the passage of time to show the parts of an hour.
c) Above each door there is a copper dome, each dome rises after an hour has passed, with the rotation of the door.
d) At the sides of the copper doors there are two falcons (Bazan) with spread wings, made of copper, each throwing copper balls into a vase to generate ringing sound at hourly intervals.
e) In the middle of the upper section there is a semi-circular disc, called ‘the night circle' (Da'irat al-Layl), which has 12 equi-spaced circular openings. These openings allow light, from a lamp behind the disc, through as the clock turns indicating time at night.
f) Within the lit night circle there is another circle onto which a horoscope (zodiac) is drawn. This gives specific information on the change of the seasons.
g) At the centre of the former circle there is an arrow, which revolves with the passage of the daylight hours, to indicate the angular movement of the sun from sunrise to sunset.
These circles are covered in half-form by a wooden plank having side columns called Samt al-Ufuq (The Line of the Horizon).
The back of the upper section contains the mechanical control system that uses ropes and pullies to convert the movement coming from the lower house into different other moving parts of the clock.
4.2 The lower section
The lower section houses the engine that generates the movements and transmits them by ropes and pullies to the parts in the upper part. The engine works by a float in a water tank (bankan). Upon draining the water from the tank, through an orifice at the bottom, the float moves down assisted by gravity, pulling a rope over a pulley which causes the movement of all other parts. The float movement is controlled by the speed with which the water surface moves down. This movement is regulated by a control valve attached to the orifice.
In the year 2000, the author of this article has contacted Professor Fuat Sezgin to alert him to the numerous anomalies in the model of the clock which was constructed and exhibited in his museum and that from the description and photos, it seems that this model has little to do with the Umayyad Mosque Clock. The author invited Professor Sezgin to Damascus so that he could witness a more representative model which we constructed of the real clock. Unfortunately, Professor Sezgin refused to admit that his model did not match the drawings of Eilhard Wiedemann and Fritz Hauser, neither it follows the description of the clock in the original manuscript of Ridhwan al-Sa'ati.
Four years later, in an interview with a journalist in Dubai, we gave preliminary information regarding our model of the clock and its external design. Then, in 2006, we published an article in Arabic in the Syrian Journal Al-Ma'rifa (No. 518, November 2006) which included more details and information concerning this device. In addition, a short study in English was published in www.MuslimHeritage.com, the academic portal specialising in the spread of knowledge about science, technology and medicine within Muslim culture.
There is a clear need to clarify some facts about this topic and to ward off confusion by earlier claims.
5.1. Features of Islamic clocks
Many pages will be required to explain the mechanics of Islamic water clocks. However, a quick summary gleaned from the water clocks (22 in number) since that of Harun al-Rashid up to the Clocks of the city of Fez, leads us to give the general principles which they all shared:
- Generating movement from a simple water engine;
- Machines that cause different activities;
- Different movements that indicate the time.
These movements and activities differ according to the importance, the size and the location of the clock, from a small box-clock, to middle-sized clocks for the palaces, to bigger sized clocks for the cities, which were located in important centres.
The clock of Ridhwan al-Sa'ati was placed near the Umayyad mosque close to the Green House (al-dar al-khadra') at the government house. It was maintained and managed directly by a government minister.
In terms of timing, some of these clocks indicate day timing only (from sunrise to sunset) and some of them indicate both the day and night timing.
The timing varies between the division of the time of the day into 24 equal parts, like our present timing, and this system was called ‘the flat hour' (al sa'at al-mustawiya) i.e. the equal hour. In Muslims societies, this timing system received lower priority in relation to the other timing system, that which divides the daylight into 12 equal parts, and the night into 12 equal parts. In this timing system, the daytime hours in the longest day of summer measured 76 minutes, while the night hours of the same day measured 44 minutes of our present timing system. Such an hour was called ‘the temporal hour' (al-sa'a al-zamaniyya). This timing system was favoured over the equal hour system, because:
- It directly links to the daily working activities of society, which mainly take place during the daylight. Hence, the day from sunrise to sunset is divided into 12 parts.
- It helps in setting the prayer times, like salat al-Dhuhr (midday prayer) at the 6th hour steadily throughout the year and the salat al-Maghrib (sunset prayer) at the 12th .
- It is a result of the partition of the solar time (sundial), which operates from sunrise to sunset giving similar time each day throughout the year.
One of the important features of the Umayyad mosque clock is its ability to show both time systems automatically. It does so by intelligent use of fluid mechanics and control devices. Another feature of interest is that the clock has eight external activities showing the time in different ways. This fascinating nature of the clock matches the Umayyad mosque as an architectural edifice of civilisation.
We now discuss the features of the models reconstructed in Frankfurt or Cairo, and examine the claims of their makers.
First of all, every working model among those reconstructed is not necessarily a genuine replica reconstruction of the original Umayyad mosque clock. The important thing, however, is the matching of the specifications and the operating method of the clock. Most water clocks rely on the presence of the doors that open and two birds that throw the balls and the possibility of the existence of a circle that is lit and shows the time at night.
5.2. The Frankfurt Clock
By scrutinising what was published previously about the Frankfurt clock and comparing it with the manuscript description of the original clock of Ridhwan al-Sa'ati, we find the following:
- The overall l total size is smaller than the dimensions of the Umayyad mosque clock as in the last chapter of Ridhwan al-Sa'ati manuscript.
- It does not show a separate external water engine as described by the manuscript of the original clock.
- It does not have a separate external water regulator which manages the clock (called the Tray).
- The inner workings have been covered from behind by a wooden board.
- The upper circle - the night circle which is lit – located at the upper level of the clock, in the middle, is unlike that of the original clock but more resembles that of Al-Jazari Water Clock.
- Half of the upper circle at the Umayyad Mosque Clock is hidden by a piece of wood which consists of side columns called Samt al-ufuq (the line of the horizon) which indicates the sunrise and sunset. This is missing in the Frankfurt museum clock.
- The two falcons that throw balls onto vases in the Frankfurt museum clock are installed differently from the way they are installed in the Umayyad mosque clock.
- The vases receiving the balls do not contain the mirror that is mentioned at the description of the Umayyad mosque clock. Unlike the original clock, the two vases are placed separately from the body of the clock. Hence they are not connected to the ‘house of balls' within the main body.
- The general structure of the columns, the doors and the night circle of the Frankfurt museum clock do not follow their description in the manuscript.
Notwithstanding these variations, there is a bigger problem with the Frankfurt clock in that there are many features missing but which are found in the original clock. Examples of these are:
- There are no domes over the gates
- There is no indicator or pointer that slides under the doors to show parts of the hour.
- There is no circle of horoscopes (zodiac) that indicate change of seasons.
- There is no arrow that rotates with the passage of daylight hours to indicate the angle of the sun. .
- There are no basements for the vases, and hence no conduits to guide the balls back to the house of balls.
5.3. The Cairo clock
If we move to the Cairo Museum model, we find that there is a great resemblance between this model and the Frankfurt Museum model, giving the impression that they have the same maker. There is however, one difference. The Cairo model has one dimension enlarged to 240 cm. This is somewhat strange as this dimension only appeared in a prior publication by the present author in newspapers and on the internet. Unfortunately, the maker of the Cairo model made a few unjustified changes such as reducing the number of the night circular openings from 12 to 5 and modifying the form of the birds to ones which have no resemblance to the originals.
Because of the very close similarity of the Cairo model to that of Frankfurt, all the anomalies identified earlier apply to it as well.
It is saddening to note that these two models do not do justice to the magnificence of the original clock which at the time matched the grandeur of the Great Umayyad Mosque.
[1.] Wiedemann, Eilhard & Hauser, Fritz, "Über die Uhren im Bereich der islamischen Kultur", Nova Acta: Abhandlungen der Kaiserlich Leopoldinisch-Carolinischen Deutschen Akademie der Naturforscher, 100 (1915), Heft 5 – reprinted in: Gesammelte Schriften zur arabisch-islamischen Wissenschaftsgeschichte (Frankfurt am Main: Institut für Geschichte der Arabisch-Islamischen Wissenschaften an der Johann Wolfgang Goethe-Universität, 1984 [= Veröffentlichungen des Institutes für Geschichte der Arabisch-Islamischen Wissenschaften, Reihe B, Band 1]), vol. III, pp. 1211-1482.
[2.] Eilhard Wiedemann, Gesammelte Schriften zur arabisch-islamischen Wissenschaftsgeschichte. Gesammelt und bearb. von Dorothea Girke. Frankfurt: Institut für Geschichte der Arabish-Islamischen Wissenchaften, 1984, 3 vols.
[3.] Donald R. Hill, Arabic Water clocks. Al-Sa'at al-ma'iyah al-'arabiyah. Aleppo: Aleppo University, Institute for the History of Arabic Science, "Sources & studies in the history of Arabic-Islamic science, history of technology series", N° 4.
[4.] Ridhwan b. Muhammad al-Sa'ati, 'Ilm al-Sa'at wa 'l-'Amal biha, edited by Muhammad Ahmad Dahman. Damascus: Maktab al-dirasat al-islamiyya, 1981.
[5.] On Ridhwan al-Sa'ati's life and work, see Abdel Aziz al-Jaraki,When Ridhwan al-Sa'ati Anteceded Big Ben by More than Six Centuries (published on 11 April 2007) and Moustafa Mawaldi, Ridhwan al-Sa'ati: A Biographical Outline (published on 29 June 2008.
[6.] See Abdel Aziz al-Jaraki,When Ridhwan al-Sa'ati Anteceded Big Ben by More than Six Centuries (published on 11 April 2007)
[7.]See Abdel Aziz al-Jaraki,When Ridhwan al-Sa'ati Anteceded Big Ben by More than Six Centuries (published on 11 April 2007).
~ End ~
* Abdel Aziz al-Jaraki is a professional architect in Damascus, with expertise in the Islamic tradition of mechanics. He is a consultant for the Foundation for Science, Technology and Civilisation (FSTC), UK.