بِسْــــــــــــــــــــــمِ اﷲِالرَّحْمَنِ الرَّحِيم
By: Donald R. Hill & Ahmad Y. al Hassan
The history of engineering in Islam is a very wide subject indeed, and it is not easy to do it justice in a single article. The intention is to present the reader with a fair and accurate picture of the scope and significance of Muslim achievements in this field, and to indicate how Muslim engineers served the needs of society and how, in a number of instances, their work was of importance in the development of modern engineering. To realize this intention in the space available some omissions are inevitable. In the first place, and without entering into a lengthy discussion about the precise meaning of the term, ‘engineering’ is taken here as a concept that involves some degree of complexity.
In civil engineering, small structures such as dwelling houses and short-span bridges have in general been disregarded. In the mechanical field military machines are not discussed and devices that require the frequent and repeated use of the human hand have been omitted. There is therefore no direct discussion of hand tools, personal weapons or textile machinery. An exception is made in the case of instruments – surveying and astronomical – because of the mathematical skills required for their construction and use.
Irrigation and water supply
For the sake of convenience it is necessary to divide the subject of civil engineering in the Muslim world into several sections, but in fact much of the subject is broadly contained within the field of irrigation and water supply. Dams are used to impound and divert irrigation water, bridges to cross canals and surveying techniques to align and level canals and qanats. Water-raising machines, which are discussed under mechanical engineering, are of course an integral part of hydraulic engineering schemes. In this section, therefore, we shall confine our attention to the principal irrigation systems, and to the means for transporting water to the fields and to urban communities.
There are four different types of irrigation. Basin irrigation, which was the method used in Egypt from ancient times until quite recently, consists of levelling large plots of land adjacent to a river or a canal, each plot being surrounded by dykes. When the water in the river reaches a certain level the dykes are breached, allowing the water to inundate the plots. It remains there until the fertile sediment settles, whereupon the surplus is drained back into the watercourse. Perennial irrigation is a method of watering crops regularly throughout the growing season by leading the water into small channels which form a matrix over the field. Water from the main artery – a river, major canal orqanat – is diverted into supply canals, then into smaller irrigation canals, and so on to the fields. Terrace irrigation is a method used in hilly country and consists of the formation of a series of terraces stepped down a hillside. Irrigation is by means of stored rainfall, wells, springs and occasionally qanats. Wadi irrigation depends upon sporadic rainstorms in otherwise arid lands. It consists of impounding the storm water behind dams and using this water to irrigate the fields adjacent to the watercourses.
The famous dam at Ma’rib in the Yemen was the focal point of such a system. Following its original construction in the eighth century B.C.E. it was successively raised, not to impound water for long periods but to raise the wadi floods to increasingly higher levels in order to irrigate more and more land by means of a canal system which used the wadi itself as a drain. The final breakdown of the dam is thought to have occurred about a quarter-century before Muhammad’s birth. From the second century B.C.E. until the beginning of the first century C.E. the Nabateans in southern Palestine and Jordan developed a thriving agriculture based upon wadi irrigation. Whereas irrigation in the Yemen depended upon a single large dam, the Nabateans built thousands of little barriers sited across one wadi after another in order to divert or capture the one or two weeks of runoff occurring each year.
All these methods of irrigation originated in antiquity and it cannot be said that any radically new techniques have been added to the repertoire of Egyptian and Mesopotamian engineers. It could scarcely be otherwise: the basic problem of impounding the water, conducting it to the fields and finally draining the surplus remains as it has always been. Irrigation, and particularly perennial irrigation, however, is a branch of civil engineering which has always demanded a high degree of technical and administrative skills. The construction of dams, canals and qanats, matters to do with water flow and control, and elaborate surveying problems; all present themselves uncompromisingly and demand the attention of experts. From one area to another, there will always be differences in hydraulic conditions, climate, soil and terrain, so that the engineers must apply their knowledge and experience to produce the best system for a given set of conditions.
It is sometimes said that large cities are one of the main characteristics of Islamic civilization, and it is of course true that great cities such as Baghdad, Cairo and Cordoba, with their flourishing economic, commercial and intellectual life, were a major component of that civilization. It hardly needs emphasizing, however, that life in these large urban centres would have been impossible without the support of a thriving agriculture. Many Muslim cities were founded after the advent of Islam – e.g. Baghdad, Basra and Shiraz –- and we therefore find that the efforts of the engineers were taxed to the utmost either in extending existing systems or in creating completely new ones.
From the start of the Islamic Caliphate Irrigation works and water distribution were prominent among the State’s achievements. When al-Basra was established during Umar’s period, he started simultaneously building some canals for conveying water and for irrigation. Two important canals were built linking al-Basra with the Tigris River. These were al-Ubulla River and the Ma`qil River. Basra obtained the necessary drinking water, and the two canals were the basis for the agricultural development for the whole Basra region.
‘Umar also devised the policy of cultivating barren lands by assigning such lands to those who undertook to cultivate them. This policy, which continued during the Umayyad period also, resulted in the cultivation of large areas of barren lands through the construction of irrigation canals by the State and by individuals.
The various governors who were appointed by the Umayyads constructed several works to prevent the formation of new swamps and to dry old ones, through the building of dams that regulated the flow of water. We find in the original Arabic sources details about the irrigation works which were constructed in Iraq and in Syria in the regions of al-Basra, al-Kufa, Wasit, al-Bata’ih, al-Raqqa and several other areas.
When the Abbasids assumed power in the second/eighth century they followed the same policy. They expanded greatly the existing irrigation system, mainly to cater for the needs of the new city of Baghdad, whose population at its zenith was about 1 500 000. The network of canals between the Euphrates and the Tigris was extended, the great Nahrawan canal to the east of the Tigris was lengthened and two new systems on the rivers ‘Uzaim and Diyala were added.
Although there was some irrigation in Spain in Roman and Visigothic times, the large systems along the River Quadalquivir and in the province of Valencia were Muslim achievements. The rulers of al-Andalus and many of their followers were of Syrian origin, and the climate, terrain and hydraulic conditions in parts of southern Spain resemble those of Syria. It is hardly surprising, therefore, that the irrigation methods – technical and administrative – in Valencia, for example, closely resemble the methods applied in the Ghuta of Damascus.
There were many other irrigation systems in the Muslim world, ranging from the great canal networks of Egypt and Iraq, down to village fields watered from one or two wells. One of the largest systems was centred on the city of Marw in Khurasan on the River Murghab, which provided the irrigation water for extensive farmlands. In the fourth/tenth century the superintendent of irrigation at Marw was said to have had more power than the prefect of the city, and to have supervised a workforce of 10 000 men. Greatly surpassing this, however, was the land of Sughd (Bilad al-Sughd ) – now part ofUzbekistan. The mainstay of its fertility was the Sughd River, now called the Zarafshan, which flowed through the great cities of Samarkand and Bukhara. At the height of its prosperity in the third/ninth and fourth/tenth centuries this land was rich and fertile beyond compare, its agriculture supported by a vast network of canals extending for many miles around the two cities.
Given the large numbers of men required to construct, maintain and control the large irrigation systems, it is hardly surprising that most of the enterprises were under State control, although it was not unusual for work to be let out to subcontractors.There are several Arabic treatises which tell us a good deal about the methods used for surveying and some of them discuss the excavation of new canals and methods for maintaining existing ones.
We shall discuss land surveying in a separate section, but a section on quantity surveying in a treatise written in Iraq in the fifth/eleventh century is worth mentioning, since it also provides us with information about irrigation works in general. Instructions are given for calculating the quantities of earth to be excavated from canals of given lengths, widths and depths and for converting these quantities into manpower requirements. The canal banks were reinforced with bundles of reeds, and the man-hours required for preparing and placing the bundles are given.
For excavation, the number of diggers (called ‘spades’) was first calculated, and to these were added the numbers of carriers to each spade, which depended on the distance the spoil had to be carried. Overheads for ancillary workers and supervision were then added. There was a set price for each task, so in the end a Bill of Quantities was produced which would provide the estimate for the cost of the works and serve as a guide for the recruitment of labour. Or, if the project was let out to subcontract, the Bill of Quantities would be the main document for awarding the contract and for the subsequent measurements and payments. Quantity surveying methods have not therefore changed materially over the centuries. From this treatise, and elsewhere, we get a picture of a highly organized State enterprise, with an army of bureaucrats, engineers and surveyors, controlling a very large labour force, whose productivity and rates of pay were closely specified.
It is not usually easy to separate irrigation from water supply because both systems were derived from the same hydraulic works. Thus a dam would provide for both the town supply and the needs of the farmers, with one main feeder channel going to the irrigation system and another into the town. Or a canal would be led out from the main feeder canal into the urban centre. It was collected in a reservoir inside, or just outside, the city walls, and was conducted from there through pipes or open channels to the baths, fountains, houses for ritual ablutions, private and public buildings, and gardens. A particularly impressive example of artificial storage reservoirs can still be seen just outside the city of Qayrawan. Two large linked cisterns for receiving the waters of the Wadi Merj al-Lil when it was in flood were completed in the year 248/862-3. Although they appear to be circular, both are actually polygonal, the larger having a diameter of just under 130 m, the smaller one a diameter of 37.4 m. The smaller receives the waters of the wadi and acts as a settling tank; a circular duct several metres above its base connects it to the larger cistern, which has a depth of about 8 m. On leaving the larger cistern, the water is decanted a second time into two oblong covered cisterns.
One of the most effective methods for providing water in regions without perennial streams is the qanat, an almost horizontal underground conduit that conducts water from an aquifer to the place where it is needed. The technique probably originated in northern Iran in the eighth century B.C.E. It was in widespread use in the Muslim world in the medieval period and up to modern times. Indeed, recent estimates have shown that 75 per cent of all water used in Iran at the present time comes from qanats and that their total length exceeds 100 000 miles. The city of Teheran alone has thirty-six qanats, all originating in the foothills of the Alburz 8 to 16 miles away, with a measured flow of 6.6 million gallons a day in spring and never below 3.3 million in the autumn.Outside Iran, qanats are still in use in parts of the Arab world, notably in the south-east of the Arabian Peninsula and in North Africa.
The qanat system was used by the Umayyad and the Abbasid caliphs. The Caliph Al-Mutawakkil (847-866) constructed a qanat system for the supply of water to his new palace at Samarra. Recent excavations there showed that the water was obtained from ground water of the upper Tigris and conveyed to Samarra in qanatconduits totaling 300 miles in length.
Al-Karaji’s Inbat al-miyah al-khafiyya (The Bringing Out of Hidden Waters) is a technical treatise written about 1000 C.E. which gives good details on the finding of the water level, instruments for surveying, construction of the conduits, their lining, protection against decay, and their cleaning and maintenance.
The construction of qanats is in the hands of experts (muqanni) and the secrets of the profession are largely handed down by word of mouth from father to son. The termination of the qanat, either farmland to be irrigated or a community to be provided with potable water, or both, will be known in advance, as will the general location of likely aquifers. One of the main skills of the muqanni lies in determining, by examining the alluvial fans for traces of seepage and hardly noticeable changes in vegetation, precisely where the trial well is to be dug. When the excavators reach the impermeable layer the well is left for a few days while the muqanni estimates the potential yield of the well by hoisting up measured quantities of water and at the same time observing any fall in the water level. If necessary, further wells are sunk to ensure that genuine groundwater has been reached; the shaft with the highest yield is chosen as the ‘mother well’. The next step is for the surveyor – or senior muqanni – to determine the route, gradient and precise outlet of the qanat. The route will be selected according to considerations of terrain and, in some cases, questions of ownership.
To start the survey, a long rope is let down into the mother well until it touches the surface of the water, and a mark is made on it at ground level. The surveyor then selects a spot on the route 30 to 50 yards from the mother well for the first ventilation shaft. A staff is held on this spot by a labourer, and the surveyor measures the fall with a level. Nowadays a modern surveying level may be used but in earlier times one of the instruments described in the section below on surveying was used. A second mark is made on the rope coinciding with the measurement on the staff; the distance from this mark to the lower end of the rope will be the depth of the first ventilation shaft. He continues to level in this way along the route, marking the rope at the location of each shaft, until he reaches the end of the rope. He has then reached a point on the ground at the same level as the surface of the water in the mother well. For the mouth of theqanat he now chooses a place below this level, but higher than the fields, and divides the drop from the level point to the mouth by the number of proposed ventilation shafts and adds this amount to the previously surveyed depth of each shaft. In this way be determines the gradient of the conduit, which is usually from 1 : 1000 to 1 : 500.
After completion of the survey, a number of guide shafts, about 300 yards apart, are driven under the supervision of the surveyor. Then the rope with the marked length of each vertical shaft is handed to the muqanni, who now begins to work with his assistants by driving the conduit into the alluvial fan, starting at the mouth. At first the conduit is an open channel, but it soon becomes a tunnel. Another team sinks ventilation shafts ahead of the tunnellers, and labourers haul the soil up to the surface through these shafts. Two oil lamps are kept burning on the floor of the conduit; sighting along these, the muqanni keeps the tunnel in alignment, and they also serve as a warning of poor air, since they go out before there is a danger of a man suffocating. As the work nears the mother well great care has to be taken in case the muqanni misjudges the distance and strikes the full well, in which event he might be swept away by the sudden flow.
It can be seen, therefore, that the construction of qanats is a special example of the difficult and dangerous profession of mining (see Figure 1). It may be considered as one of the most successful of man’s inventions, since it has been in continuous use for over 2500 years.
Fig. 1 – The qanat