Wang Chong Wang Xiaotong

W

Wang Chong FABRIZIO P REGADIO The Chinese philosopher Wang Chong was born in Shangyu (modern Zhejiang) in AD 27. According to his biographies, he came from a poor family and devoted most of his life to teaching, but also held minor posts in the state administration. Later he retired to compose the work by which we know him today, the Lunheng (Balanced Discussions). This extensive treatise in 85 chapters (one of which is lost) was completed in AD 82 or 83, about 15 years before the author’s death. Wang Chong lived about one century after Confucianism had emerged as imperial ideology. In this process, the “rationalist” and socially minded philosophy taught by Confucius had been integrated with cosmological doctrines extraneous to the letter of his teaching. The school of though: to which Wang Chong belonged propounded, on the contrary, a reading of the classic texts devoid of esoteric interpretations. In his work, Wang Chong analyzes, with a strongly skeptical and even iconoclastic spirit, ideas expounded by earlier thinkers and beliefs shared by the people of his time (e.g., the recourse to divination, the belief in ghosts, the search for physical immortality, and the idea of an individual spirit that persists after death). Wang’s typical procedure is to bring out contradictions in the anecdotes and accounts that he first quotes in full. Through an exemplary logical method, he often does not hesitate to take as true one detail that he has previously refuted, if this may serve to invalidate a different detail. Philosophically, Wang Chong maintained that all phenomena arise spontaneously, and are not expressions of Heaven’s will. Related to this was his opposition to the belief in prophecies and portents, through which Heaven was deemed to legitimate or censure rulers, and assent to or dissent from their policies. While Wang Chong rejected the blend of these forms of esoterism with Confucianism, he fully accepted the metaphysical and cosmogonical doctrines traditionally placed under the egis of Daoism. In this way, he anticipated some of the new developments in post-Han Confucianism.

References Chan Wing-tsit. A Source Book in Chinese Philosophy. Princeton: Princeton University Press, 1963. Forke, Alfred. Lun-heng. Philosophical Essays of Wang Ch’ung. Vol. 1. London: Luzac and Co., 1907; Vol. 2. Berlin: Georg Reimer, 1911. Fung Yu-lan. A History of Chinese Philosophy, Vol. 2: The Period of Classical Learning. Princeton: Princeton University Press, 1953. Hui, Huang, ed. Lunhengjiaoshi (A Critical and Annotated Edition of the Lunheng). Changsha: Shangwu Yinshuguan, 1938. Needham, Joseph. Science and Civilisation in China, Vol. 2: History of Scientific Thought. Cambridge: Cambridge University Press, 1956.

Wang Xiaotong A NG T IAN S E There is no record of Wang Xiaotong’s early life nor his year of death. We estimate that he flourished from the second half of the sixth century to the first half of the seventh century. The little we know of him is from a memorial he presented to Emperor Gaozu of the early Tang dynasty (AD 618–906) on the occasion of the submission of his mathematical text to the throne. The mathematical text he submitted was known as Jigu Suanjing (Continuation of Ancient Mathematics) and was subsequently selected as a prescribed text for imperial examinations in AD 656. In his memorial, which is now attached to his mathematical text, he mentioned that he had studied mathematics from a very young age. He studied the Jiuzhang Suanshu (Nine Chapters in Mathematical Art) thoroughly and had great admiration for Liu Hui’s in-depth commentary on the text. On account of his mathematical acumen, Wang Xiaotong was appointed as an instructor in the Department of Mathematics, and later as a deputy director in the Astronomical Bureau. In AD 623, together with Zu Xiaosun, an official of the board of Civil Office, he was appointed to re-examine the adequacy of the current calendar. Composed by Fu Renjun and promulgated for use since AD 619, the calendar had on several occasions

2200

Wang Xichan

been found to be losing accuracy in the predictions of solar and lunar eclipses. Based on the structure of the Kaihuang calendar composed by Zhang Bin of the previous Sui dynasty (AD 581–618), Wang Xiaotong criticized the adoption of the ding shuo method and the precession of equinoxes in the current calendar. The critique sparked a 3-year debate between Fu Renjun and Wang Xiaotong, culminating in a submission of a proposal for rectifications consisting of more than 30 errors to the Astronomer-Royal. This does not necessarily reflect Wang Xiaotong’s achievement in calendrical science. On the contrary, his view of adhering to the traditional model without taking into consideration the uneven apparent motion of the sun and the precession of equinoxes in calendrical calculations was a retrogressive one. His contribution lies in mathematics. From the late Han dynasty in the first century onward, Chinese mathematicians were familiar with quadratic equations and their solutions. But it was not until the appearance of the Jigu Suanjing that equations of the third degree were presented. They arose because of the special needs of engineers, architects, and surveyors of the Sui dynasty. There are 20 practical problems in the Jigu Suanjing consisting of a problem (No. 1) on calendrical calculation, six problems (Nos. 2–6 and 8) on engineering constructions, seven problems (Nos. 7, 9–14) on the volume of granaries, and six problems (Nos. 15–20) on right-angled triangles. The solutions of most of the problems involved equations of the third degree. For example, problem No. 15 says: “There is a right-angled triangle, the product of 1 and whose hypotenuse two sides of which is 706 50 is greater than the first side by 36 109 . Find the lengths of the three sides.” Wang’s solution amounts to the formation of the cubic equation as follows: S p2 x 3 þ x2 ¼ 0 2 2S where P is the product and S the surplus. As a matter of fact, most of the problems in the last three categories involved the use of the equation x3 þAx2 þ Bx ¼ C; where A, B, and C are positive numbers. Wang Xiaotong provided the rules for the arrangement of the equations in all these problems but did not explain the procedure for arriving at such equations. He also did not discuss the equations of higher degrees. Numerical equations of degrees higher than the third degree occurred first in the work of Qin Jiushao around AD 1245. See also: ▶Liu Hui and the Jiuzhang Suanshu, ▶Calendars, ▶Qin Jiushao

References Guo, Shuchun. Wang Xiaotong. Zhongguo Gudai Kexuejia Zhuanji (Biographies of Ancient Chinese Scientists). Ed. Du Shiran. Beijing: Kexue Chubanshe, 1992. 317–9. Mikami, Yoshio. The Development of Mathematics in China and Japan. 2nd ed. New York: Chelsea Publishing Company, 1974. Qian, Baozong ed. Suanjing Shi Shu (Ten Mathematical Classics). Beijing: Zhonghua Shuju, 1963. Ruan, Yuan (AD 1799). Chouren Zhuan (Biographies of Mathematicians and Astronomers). Vol. 1. Shanghai: Shangwu Yinshuguan, 1955.

Wang Xichan L IU D UN Wang Xichan (July 23, 1628–October 18, 1682), sometimes known by his literary name, Xiao’ an, was from Wujiang, Jiangsu province, China. When he was sixteen, the Ming Dynasty (1368–1644) collapsed and the door to social advancement through the imperial examination system was suddenly closed. After several unsuccessful suicide attempts, Wang abandoned his hopes for an official career and became one of the most distinguished Ming loyalists in his area. He seems to have survived by teaching literature, although his main interest was in astronomy, which he had studied on his own since his youth. Despite both poverty and illness, Wang made unremitting efforts to observe the heavens, calculate planetary positions, and write about astronomy. Beginning in the late Ming Dynasty, Western astronomy had been disseminated intermittently into China, and subsequently it was adopted by the Qing (1644– 1911) rulers. In 1646, a number of astronomical treatises, earlier written or translated by Western missionaries serving the Ming court, were published together by order of the Qing emperor under the general title Xi Yang Xin Fa Lin Shu (Astronomical Treatises of the New Methods of the West, 1646). Unfortunately, the “New Methods” introduced by the missionaries did not reflect the advanced achievements of modern astronomy in seventeenth-century Europe. Even worse, there were some defects and internal contradictions, especially in the parts dealing with the cosmological theory of planetary motions. In his Xiao An Xin Fa (New Method of Xiao’an, completed in 1633), Wang argued that all of the Western techniques could be reconciled with classical Chinese schemes and therefore could be used to revive the lost traditional astronomy of ancient China. By means of trigonometry, which was not used in traditional Chinese astronomy and mathematics, Wang created a

Water in India: Spiritual and technical aspects

series of methods to calculate ecliptic positions and predict planetary occultations. Forty years later, in his Wu Xing Xing Du Jie (On the Angular Motion of the Five Planets, completed in 1673), Wang criticized the contradictions in the Xi Yang Xin Fa Li Shu and proposed instead his own model of the planetary motions which differed from both the Aristotelian– Ptolemaic and the Tychonic geostatic models. In addition, he attributed the planetary motions to an attractive force radiating from the outermost moving sphere (i.e., the primum mobile). Wang also suggested that there were planets inside the orbit of Mercury which might account for the appearance of sunspots. All of these thoughts, along with some of his methods, were heuristic and had considerable influence on his successors. In general, Wang was one of a few pioneers who responded to Western science by conscientiously studying it; he was open to critical acceptance of its major ideas when suitably reinterpreted for use in seventeenthcentury China. Nevertheless, some of Wang’s arguments were clearly exaggerated. For instance, he assumed that Western astronomy had in fact originated in ancient China, but such claims were due to his radical nationalism and traditional reluctance to recognize any innovations from the Westf.

2201

Veda(s), Brāhman.a(s), Upanis.ad(s), Purān.a(s), and Smr.ti(s). Water is regarded as the primordial substance from which the universe came into being. Water occupies the highest place amongst the five basic elements of nature, called pañcamahābhūta. These are: ākāśa (ether, substratum, space), vāyu (air), teja or agni (radiation, energy, or fire), āpa (water), and pr.thivī (Earth). These five bhūta constitute the physical universe. Air is said to have been generated from space, fire from space, water from fire, and earth from water. Fire and water, which are said to pervade the entire universe, have a close nexus and are believed to possess procreative powers (Narayana, 1995). The tripartite nature of agni has been connected with the three forms of water – celestial, atmospheric, and terrestrial, called by different synonyms in Sanskrit so that they have characteristics and attributes responsible for different cosmic, atmospheric, or terrestrial actions. In the social and religious traditions and culture of India since Vedic times, water has enjoyed a unique status. Water is the single most important tool/mode for performing daily religious rituals or social ceremonies and a primary means for purification of body and soul in Hindu culture. From birth till death in a Hindu society, water remains an essential ingredient in performing all rituals.

References Chen Meidong et al. eds. Wang Xi Chan Yan Jiu Wen Ji (Collected Papers on the Studies of Wan Xichan). Shijiazhuang: Heibei Science and Technology Press, 2000. Jiang, Xiaoyuan. Wang Xichan. Zhong Guo Gu Dai Ke Xue Jia Zhuan Ji (Biography of Ancient Chinese Scientists). Vol. 2. Ed. Shiran Du. Beijing: Science Press, 1993. 1005–15. Sivin, Nathan. Wang Hsi-Shan (Wang Xichan). Dictionary of Scientific Biography. Vol. 14. Ed. Charles C. Gillispie. New York: Charles Scribner’s Sons, 1976. 159–68. Wang, Xichan. Xiao An Xin Fa (New Methods of Xiao’an, completed in 1663). Shanghai: Commercial Press, 1936. Wang, Xichan. Wu Xing Xing Du Jie (On the Angular Motions of the Five Planets, completed in 1673). Shanghai: Commerical Press, 1939. Xi, Zezong. Shi Lun Wang Xi Chan De Tian Wen Gong Zuo (An Essay on the Astronomical Work of Wang Xichan). Ke Xue Shi Ji Kan 6 (1963): 53–65.

Water in India: Spiritual and Technical Aspects

Vedas and Their Chronology The content of the Vedas is astonishingly scientific although much of it remains to be interpreted correctly. Knowledge of the Vedas is synonymous with knowledge of the science and metaphysics of creation. The time periods of various Vedas are as follows:

R. gveda Sāma Veda Yajurveda

6500–3100 BCE 3100–2500 BCE 2500–2000 BCE

An organic chronological development in India from 6500 BCE has been suggested (Frawley 1994), taking into account both Vedic literature and recent archaeological findings:

6500–3100 BCE 3100–1900 BCE

Pre-Harappan (early Rigvedic period) Mature Harappan (period of four Vedas)

W The Myth of Aryan Invasion

K. N. S HARMA The subject of water has been treated spiritually, philosophically, cosmologically, medically, and poetically in the ancient Indian literature comprising the

For the past hundred years or so, Mortimer Wheeler’s hypothesis of “swashbuckling, horse-riding lightskinned Sanskrit-speaking people, called Aryans, coming to Northwest India from Central Asia or Central Europe in about 1500 BCE and razing to the ground the

2202


highly urbanised Harappan civilization,” was accepted. These so-called Aryans were supposed to have composed the Vedas over the centuries, spread Sanskrit, and built the Ganges civilization. According to this concocted theory, these “invading” Aryans destroyed the Harappan civilization which had been flourishing there for over a millennium and whose habitants were Dravidians. References in the R. gveda point to India’s being a land of mixed races. The R. gveda also states, “We pray to Indra to give glory by which the dasyu will become āryan.” This statement confirms that to be an āryan was not a matter of birth. While the word dasyu meant uncultured and illiterate, āryan meant noble, wellcultured people. It is also used in the context of addressing a gentleman or lady (āryaputra, āryakanya). Nowhere in the Vedic literature is this word used to denote race or language. This was a notion of Max Mueller who, in 1853, introduced the word ārya into the English language to refer to a particular race and language. When challenged, he refuted his own theory later in 1888. Jean-Francois Jarrige, Director of the National Museum of Asiatic Arts of Paris, the excavator of the famous proto-historic site of Mehrgarh (Pakistan) and member of the French Academy, carried out extensive work over the last 30 years in the Indian subcontinent

and showed that the excavations revealed artifacts from a much older civilization of the Sarasvatī era (Pushkarna, 1998). The civilization included metropoli like Mohenjodaro, Harappa, Ganveriwala (in Pakistan), and Dholavira and Rakhigarhi; towns like Lothal, Surkotda, Banawali, and Kalibangan, and villages like Kunal, in India. The excavations exposed not just these towns or cities, but also an earlier settlement beneath it, and an even earlier one further down. Figure 1 is a map indicating rivers in the time of Sarasvatī, and Fig. 2 shows the spread of Harappan and Sarasvatī civilization sites. According to archaeologist Bisht, before the mature Harappan stage, many regional cultures like Amri, Kot Dirji, Kalibangan, Dholavira, and Lothal had coalesced into the cultural umbrella of Harappa. They were bound by common economic compulsions and cultural ethos. The site of Dholavira is an excellent example of a Harappan city that tells the history of Early, Transition, Mature, Late, and Final Phases of the Indus-Sarasvatī (Harappan) civilization in India (ca. 3500–1700 BCE). The site spreads over an area of 100 ha. This compares well with the size of Harappa, Rakhigarhi, or Ganveriwala. All these evidences have proved the invasion theory to be unfounded, showing that the Aryans were the native Indians, having an unbroken chronology of civilization.

Water in India: Spiritual and Technical Aspects. Fig. 1 Course of erstwhile Sarasvatī River (Kalyanaraman, 1998).


2203

Water in India: Spiritual and Technical Aspects. Fig. 2 Harappan sites in Northwest India (Kalyanaraman, 1998).

Place of Water in Ancient Indian Literature Water has enjoyed a high status in the social and religious context of ancient Indian culture. It was an essential medium for performing daily religious rituals and social ceremonies and a primary means for purification of body and soul. Even after thousands of years, the rivers in India, especially Gangā (the Ganges) and Yamunā are considered divine and capable of purifying a sinful body with a few drops of water. Agni (fire) is said to be born from the waters. Both water and fire are said to possess procreative powers. Water is considered a mother while fire is seen as a prolific generator (Narayana, 1995). The place of water as a life-giving and life-sustaining element was very high. It was considered a cleanser of sins and regarded as a divine protector. It was addressed by various names – nectar, honey, ambrosia – in prayers. Stagnant water was considered unhygienic. The cosmic energy is the generator of the universe, the embryo of waters the leader of

humans, most virile defender of the human race, it remains ever illuminated by its own radiance and it provides sustenance for its beloved progeny (RV 3.1.12). Water verily is arka (essence). What was there as froth of water hardened and it became earth (the embryonic state of the Universe). Earth was formerly water upon the ocean of space (Atharvaveda AV 13.6). Water or the water element had many names in the Vedas and other Vedic literature. It has more than 100 synonyms. These are not synonyms in a strict sense, since they have been used to indicate their different forms/states and contexts. Some names for water are: Ambu, toyam, vāri, jalam, āpah., bhes.ajam, udakam, salila, madhu, ambha, ghr.tam, and ks.īram. For example, salila is a technical word in the Vedas which is different from āpah. as mentioned in a mantra from śatapatha brāhman.a (11.1.6.1) which says that āpah.

W

2204


were indeed salila earlier. Salila is the primordial state of the universe, when there is nothing manifest.

Nature’s Forces as Deities In Vedic cosmology, pr.thivī (earth) symbolizes the material base and dyāvā (heaven) symbolizes an unmanifested immortal source. Together and between them, they form the paryāvaran.a (environment). The seer praises Heaven and Earth (dyāvā pr.thivī) by saying, “You are surrounded, Heaven and Earth, by water; you are the asylum for water; imbued with water; the augmenters of water; vast and manifold; you are the first propitiated in the sacrifice; the pious (people) pray to you for happiness, that the sacrifice may be celebrated. May Heaven and Earth, the effusers of water, the milkers of water, dischargers of the functions of water, divinities, the promoters of sacrifice, the bestowers of wealth, of renown, of food, of male posterity, combine together.” RV 6.70.4–5. Water stands for all the elements, because it is really a combination of water, fire, and earth, according to the tripartite creation of the gross elements. Water is all pervading.

Waters Regarded as Having Medicinal Powers The Vedas also mention the medicinal qualities of water. R. gveda hails water as the reservoir of all curative medicines and of nectar. It invokes water which the cows drink and offers oblations to deities presiding over the flowing waters: O Water, which we have drunk, become refreshing in our body. May you be pleasant to us by driving away diseases and pains – O divine immortal waters (RV 63). The Atharvaveda describes various sources of waters and describes them as dispellers of diseases and as more healing than any other healer. The scriptures believed that waters avert pain that they are restorative and curing. Wherever waters fall on earth, plants grow. The hymns in Atharvaveda (6.23, 24, 57) hail water as possessing medicinal qualities. A hymn in Atharvaveda prays for waters to cure “incurable” diseases. The scriptures, Sam . hitā(s), also regard water as capable of alleviating pain. “O water which we have drunk, become refreshing in our body. Be pleasant to us by driving away diseases and pains.” The R. gveda (1.161.9) states, “There exists no better element other than water which is more beneficent to the living beings. Hence waters are supreme.” Varuna is a cosmic ruler as well as the deity that dwells in waters, presides over them and is prayed to for granting strength and virility from waters. Another hymn from the R. gveda says,

Ambrosia is in the waters, in the waters are medicinal herbs; therefore, divine (priest), be prompt in their praise. Soma has declared to me, ‘all medicaments, as well as Agni, the benefactor of the universe, are in the waters’: the waters contain all healing herbs. Waters, bring to perfection all disease-dispelling medicaments for (the good of) my body, so that I may long behold the Sun. Waters, take away whatever sin has been (found) in me, whether I have (knowingly) done wrong or have pronounced imprecations (against holy men) or (have spoken) untruth. I have this day entered into the waters – we have mingled with essence – Agni, abiding in the waters, approach, and fill me, thus (bathed), with vigour (RV 1.23. 19–23). According to Taittareyī Āran.yaka (7.3.2), Agni is an antecedent form and sun the later. Water is a compound and lightning is the joining element. Maitreyī Sam . hitā further divides water into three places – sky, earth, and mid-region. Vājasaneyī Sam . hitā describes the medicinal use of waters. “O water, which we have drunk, become refreshing in our belly. May you be pleasant to us by driving away diseases and pains.” The verses are recited while touching one’s navel after drinking liquid in a sacrificial procedure. Like Agni and Vāyu, āpah. (water) also serves as a purifying agent.

Rivers as Goddesses The rivers and river waters have been treated with great reverence since ancient times. Traditionally, rivers such as Gangā, Yamunā, or Narmadā are worshipped as goddesses. Every morning and evening, on the banks of the Gangā at Haridwār, there is daily “Gangā worship” with lighted lamps in the presence of thousands of devotees, traditional holy music, and chants of mantras. The Sarasvatī River was one of the largest rivers in ancient India before 3000 BCE and drained the Sutlej and the Yamunā. By the end of Harappan culture, the Sarasvatī went dry, bringing about the end of Harappan civilization around 1900 BCE. The Vedic texts are replete with references to the river Sarasvatī and the seas.

River Sarasvatı¯: Myth or Reality? Whereas the famous River Gangā is mentioned only once in the R. gveda, the River Sarasvatī is mentioned at least sixty times. Sarasvatī is now a dry river, but it once flowed all the way from the Himalayas to the ocean across the desert of Rajasthan. Research by Dr Wakankar has verified that the river Sarasvatī changed course at least four times before going completely dry around 1900 BCE. The latest satellite


Water in India: Spiritual and Technical Aspects. Fig. 3 Archaeological sites along erstwhile Sarasvatī River.

data combined with field archaeological studies have shown that the R.gvedic Sarasvatī had stopped being a perennial river long before 3000 BCE (Giri). Numerous archaeological sites have also been located along the course of this river, thereby confirming Vedic accounts. The Sarasvatī is now dated long before 3000 BCE. This means that the R. gveda described the geography of North India long before 3000 BCE. This shows that the R. gveda must have been in existence no later than 3500 BCE. Tritium (hydrogen isotope) analysis of deep water samples taken by Bhabha Atomic Research Centre (BARC) has provided a broad spectrum dating for the waters of the Sarasvatī river now revealed as groundwater sanctuaries and aquifers. The waters range from 4,000 to 8,000 years BP (Kalyanaraman). For over 2,000 years (6000–4000 BCE), the Sarasvatī flowed from Bandarpunch massif (Sarasvatī-Rupin glacier confluence at Naitwar in western Garhwal). Figure 3 shows the location of archaeological sites along Sarasvatī river. A remote sensing study of the India desert reveals numerous signatures of palaeochannels. The LANDSAT imagery highlights the course of the River Sarasvatī in Punjab, Haryana, and Rajasthan. The digital enhancement studies of IRS-1C data (1995), combined with radar imagery from European remote sensing satellites ERS I/2, have identified subsurface features and reorganized the palaeochannels beneath the sands of the Thar desert in northwestern India (Paik, 2000).

Ancient Weather Science In the early R.gvedic period, the performance of Yajnas (sacrifices) as described in Vedic and Brahmanic

2205

literature has some scientific basis. These rituals were performed to ensure timely and adequate rainfall for abundant availability of food, thriving of animal and plant life and for overall human prosperity (Sharma, 2002). The Yajnas are very closely related to the evolution of the universe, the solar system, human procreation, occurrence of the seasons, rainfall and life on earth. Several places in the Vedic literature mention that Yajnas produce rain which in turn produces food. The subject of meteorology was dealt with by several ancient sages, including Nārad, Kashyap, Garg, Parāshar, Vasisht.ha, Druhin, Brihaspati, Devala, Vajra, Sahadev, and Rishiputra. Asht. ādhyāyi (500 BC) by Pān. ini is perhaps one of the earliest post-Vedic works giving information on rainfall measurement and droughts. The Arthashāstra (fourth century BC) of Kaut. ilya gives information on rainfall distribution in the country as well as the methods of its measurement. Several of the old treatises in the form of Purān.as, such as Matsya Purān.a, Vāyu Purān.a, or Vishn.u Purān.a, dealt with the role of the sun in rainfall. Water science, which was initially based on observations, kept being refined and in the post-Vedic period from the third century BC to the sixth century AD, many of the earlier hypotheses came to be perfected. Kaut. ilya even gave an estimate of the average rainfall in various parts of India. In Varāhamihira’s time, the units of measurement of rainfall were the pala, adhaka, and dron.a. The raingauge was round with a diameter of one hasta (18 in., 46 cm) and had marks of pala. When it was full, it indicated one adhak of rainfall. One dron.a means 2–1/2 in. (6.4 cm) of rain. Accordingly to Parashar, the height and diameter of the raingauge should be 8 angulas (6 in., 15 cm) and 20 angulas (15 in., 38 cm), respectively, and when it is filled to the brim it measures one adhaka. Parashar also gives a method of measuring rainfall on the ground. If it measures four cubits on the ground it amounts to one dron.a. Ramanathan (1993) described the significance of rituals and yagnas (sacrifices) with their cosmic and scientific interpretations in great details. Observations and measurements of wind, clouds, lightning, thunder, rain, solid precipitation, atmospheric optical phenomena, and agricultural meteorology were presented. Just as the water vapours are carried higher in the form of clouds and are condensed in the presence of cold air existing in the sky, similarly one can reach the height of spiritual progress and can get strengthened due to restraining the breath through Yogic exercise (YV 23.26). The uniform water passes upwards and downwards in the course of days, clouds give joy to the earth; fires rejoice the heavens (RV 1.164.51).

W

2206


Water in India: Spiritual and Technical Aspects. Fig. 4 Major river basins of India.

Water in India: Spiritual and Technical Aspects. Table 1 Basinwise surface water potential of India (cubic km or BCM/year) Sl. No. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.

Name of the river basin Indus (up to Border) a) Ganga b) Brahmaputra, Barak & Others Godavari Krishna Cauvery Pennar East Flowing Rivers Between Mahanadi & Pennar East Flowing Rivers Between Pennar and Kanyakumari Mahanadi Brahmani & Baitarni Subernarekha Sabarmati Mahi West Flowing Rivers of Kutch, Sabarmati including Luni Narmada Tapi West Flowing Rivers from Tapi to Tadri West Flowing Rivers from Tadri to Kanyakumari Area of Inland drainage in Rajasthan desert Minor River Basins Drainage into Bangladesh & Myanmar Total

Average annual potential in river 73.31 525.02 585.60 110.54 78.12 21.36 6.32 22.52 16.46 66.88 28.48 12.37 3.81 11.02 15.10 45.64 14.88 87.41 113.53 NEG. 31.00 1869.35

Water management and reservoirs in India and Sri Lanka

O all learned people, fully realise your conduct towards different objects of the universe, know ye the electricity that maintains all beautiful objects, the sun, the invisible matter brought into creation, the invigorating vital airs, thus ye become the utilisers of all objects. Homage to him who knows the science of clouds, and to him who knows the science of electricity (Yajur Veda). These hymns give insight into the cloud-seeding phenomenon and occurrence of rains. Various activities occurring in the firmament were well observed and formulated in scientific terms.

Festivals Related to Water The focal point of Hindu social, cultural, and religious rituals has been water. Indian mythology attaches a lot of importance to the bath (snāna) which is mandatory for participation in any important religious occasion. A dip in holy rivers is considered an essential part of Hindu culture, especially on specific occasions such as the solar and lunar eclipses or occasions specified on the basis of specific planetary configurations, which are considered to have a cosmobiological effect on the human body. In a cycle of 12 years, a great Indian festival takes place which is known as Mahā Kumbha or Great Kumbha, where millions assemble and have a dip in the waters of the sacred rivers. Every third year, a smaller water festival called Ardha Kumbha (half Kumbha) is also held, where people congregate at specified places at the banks of the holy rivers (Sharma, 1998). In the mythological scriptures it says that the elixir of life, ambrosia (amr.t), that emerged from the churning of the ocean by the gods and demons, splashed out of the pitcher and fell to earth at four places: Haridwār, Prayāg, Ujjain, and Nāsik. These are located on the banks of the River Gangā, at the confluence of the rivers Gangā, Yamunā, and Sarasvatī, on the banks of Shiprā, and on the banks of Godāvarī. A tussle ensued among the gods and the demons for 12 days. During this period the moon did not let the amr.t fall from the pitcher, the sun did not let the pitcher crack, Jupiter protected it from demons, and Saturn saved it from being whisked away. Thus, these four planets, which were responsible for saving the pitcher of ambrosia (amr.t kumbha), have become an integral part of the Kumbha. Various astronomical conjugations during Kumbha represent various stages of the solar cycle which have a direct influence on human beings and the biosphere. The holding of Kumbha at an interval of 12 years is symbolic of the need for purifying the body by sublimating the inherent vices of the 12 sense organs – five organs of action, five organs of perception, and the mind and the intellect – thereby arousing the six psychic centres or chakra separated

2207

from each other at distance of 12 angula (finger widths) for attaining the amr.t kumbha or pitcher of ambrosia (Dixit, 2001). Water festivals are also celebrated in several other ways in North and South India. It is customary to take a dip in the holy waters of rivers on various auspicious occasions.

References Br.hadaran.yaka Upanis.ad. Ed. Swami Śivananda. P.O. Shivanandanagar, UP, India: The Divine Life Society, 1985. Dixit, Pankaj. The Mahakumbh: Its Sacred Significance. The Speaking Tree. Times of India. 2001. Frawley, David. The Myth of Aryan Invasion of India, Published by Voice of India. 2/18 Ansari Road, New Delhi, India, 1994. Giri, Svami B. V. Aryan Invasion. ▶http://www.gosai.com/ chaitanya/saranagati/html/vedic-upanisads/aryan-invasion. html. Kalyanaraman, S. R. gveda and Sarasvatī-Sindhu CivilizationDates of the Sarasvatī Sindhu Civilization (CA. 3100–1400 BC): ▶http:www.hindunet.org/hindu_history/sarasvatī/html/ rvssc.htm August 1998. Narayanan, Sampat, ed. Vedic, Buddhist and Jain Traditions. Vol. 2. New Delhi, India: IGNCA, 1995. Paik, Saswati. Saraswatī – Where Lies the Mystery!. Noida, Up, India: CSDMS, 2000. Pushkarna, Vijaya. Looking Beyond Indus Valley. ▶http:// www.appiusforum.com/week.indus.html July 1998 . Ramanathan, A. S. Weather Science in Ancient India. Jaipur, India: Rajasthan Patrika Limited, 1993. R. gveda Sam . hitā. Parts 1–4. Ed. Ravi Prakash Arya and K. L. Joshi. Delhi, India: Parimal Prakashan, 1997. Sama Veda Sam . hitā. 1st ed. Ed. Ravi Prakash Arya. Delhi, India: Parimal Prakashan, 1996. Sharma, K. N. Water – The Fulcrum of Ancient Indian SocioReligious Traditions. Proceedings of International Conference on Water Resources at the Beginning of the 21st Century, UNESCO, Paris, 3–6 June 1998. ---. Status of Water in Ancient Indian Literature and Mythology. Second International Conference of IWHA, Bergen, Norway, 2002. Yajur Veda Sam . hitā. 2nd ed. Ed. Ravi Prakash Arya. Delhi, India: Parimal Prakashan, 1999.

Water Management and Reservoirs in India and Sri Lanka A NDREW M. B AUER , K ATHLEEN D. M ORRISON Water storage and distribution technologies have played an important role in the histories of southern India and Sri Lanka. Given the variability in rainfall and the relatively dry conditions over much of the region, it would have been difficult for southern Asian agriculture, diet, and

W

2208


cuisine – particularly the heavy emphasis placed on rice – to have taken the forms they did without the historical development of water management techniques. Moreover, debates about the efficiency, sustainability, and equitability of modern “big dam” projects versus “traditional” methods of providing much-needed domestic and agricultural water continue to dominate Indian politics today. Here, we review historical forms of water management in South India and Sri Lanka, paying particular attention to ancient reservoir systems. South India and Sri Lanka are dominated by a monsoonal climate, whereby the southwest (or advancing) monsoon generally brings rains between the months of June and October, and the northeast (retreating) monsoon in November and December. In addition to monsoon strength, a variety of topographic factors relate to the distribution and concentration of rainfall across the region. Parts of the southwestern and western coasts of India receive between 3,000 and 3,200 mm of rainfall a year due to the orographic (related to, or caused by, physical geography, such as mountains or sloping terrain) effects of the Western Ghats – South India’s most pronounced north–south ranging mountain chain. On the eastern side of the Ghats, however, a rain shadow is created that markedly reduces the amount of precipitation in the central areas of the southern peninsula. For example, Bangalore – the capital city of the modern state of Karnataka – receives an average of around 850 mm of rainfall per year, while Hyderabad – the capital of Andhra Pradesh – receives an average of only 700 mm of rainfall. Indeed, much of the area within the orographic rain shadow of the Western Ghats can be considered a semiarid climate, with average rainfall levels falling as low as 400 mm (India Meteorological Department 1981). Sri Lanka’s topography creates a similar pattern in which the southwestern coastal areas receive the bulk of the advancing monsoon and the northern and eastern coasts – largely in the rain shadow of Sri Lanka’s central highlands – receive precipitation from the shorter, retreating monsoon. However, it is important to note that actual rainfall in all parts of the Indian peninsula and Sri Lanka can show significant intra- and interannual variability as a result of being dependent on the relative strength of the monsoon. South Indian and Sri Lankan reservoirs include a range of facilities constructed for the purposes of collecting and storing water, generally for agricultural production. These consist of artificial embankments built across paths of gravity water flow, whether intermittent streams, rivers, or simply slopes that might carry runoff after a monsoon rain. Reservoirs may or may not involve excavation of a basin to contain this water, but they are all storage or storage/distribution devices built on a relatively large-scale and meant to contain water behind an embankment or dam, rather than within its major

construction. In this sense, we make a distinction between cisterns (which collect and store water within a rock-cut or other constructed facility), wells (which tap the water table), reservoirs, and tanks. The term “tank” is widespread in the South Asian literature, indiscriminately used to describe almost any water-holding feature, although the term most frequently refers either to reservoirs or to temple tanks – large masonry structures that hold water for ritual ablutions and other functions associated with temple worship. Temple tanks often derive their water from the water table. As such, temple tanks and reservoirs are wholly different in construction, morphology, and operation, similar only in their capacities as water-holding devices and in certain parallels of meaning and symbolism (Morrison in press).

Early Forms of Water Management in South India and Sri Lanka The earliest culturally significant water catchment features in South India were not constructed reservoirs (in the terminology outlined above), but seasonal pooling basins, or cisterns, which developed naturally from the differential weathering of bedrock. In geomorphological terms, these are known variably as gnammas, rock pools, or weathering pits, and are considered to be characteristic features of residual hills and inselbergs – isolated hills composed of resistant rocks (e.g., granite or gneiss) that express pronounced topographic relief from a surrounding plain – throughout the heavily weathered terrain of the tropics and subtropics (Thomas 1994; see also Porembski and Barthlott 2000). In South Asia, such basins occur on the granitic gneiss hills that characterize much of the central and southern portions of the Indian peninsula, as well as parts of Sri Lanka (e.g., Fernando 1976). Indeed, because granite and gneiss are particularly impermeable rock types, the bare hills of South India generate large volumes of runoff water during the heavy monsoon months that collects in such depressions (Fig. 1). Water retaining rock pools appear to have taken on cultural significance as early as the Iron Age (1000– 500 BCE) in several regions of South India. During this period, mortuary and ritual sites were often marked by the construction of megalith monuments, and a clear cultural association between such ritual constructions and seasonal water basins can be established. Large concentrations of elaborately constructed megaliths – ranging from dolmen cists, stone circles, rock cairns, platform enclosures, stone spirals, and more – appear to have been deliberately placed adjacent to water basins in hilltop locations. Perhaps the most striking example of such an association occurs at the site of Hire Benkal in northern Karnataka, where hundreds of megaliths are found near a broad shallow water basin that likely began as a rock pool and was subsequently


Water Management and Reservoirs in India and Sri Lanka. Fig. 1 A rock pool at VMS 579 – an Iron Age occupational site in the Koppal district of northern Karnataka. Notice the artifact debris in the foreground (photo by Andrew M. Bauer).

Water Management and Reservoirs in India and Sri Lanka. Fig. 2 The site of Hire Benkal, showing dolmen cists on the quarried banks of a basin feature that likely began as a natural rock pool (photo by Andrew M. Bauer).

expanded by quarrying activities for the construction of monuments (Fig. 2). Additional sites also demonstrate associations between water and culturally significant ritual places. For example, a brick platform structure enclosed by granite boulders at Bandibassapa Camp, also a large megalithic complex in northern Karnataka, appears to have been situated adjacent to a rock pool that was later modified. Moreover, Gordon and Allchin (1955) reported 80 megaliths at a site near Bilebhavi where they identified two “tanks” (cisterns), one of which was “lined with stone slabs.” They also recorded a similar construction on a hilltop megalith site near Koppal (Gordon and Allchin 1955: 99). The association between megaliths

2209

and water basins has also been noticed in Tamil Nadu and Sri Lanka (cf. Seneviratne 1984; Myrdal-Runebjer 1996), suggesting a ritual importance to water, and possibly a sacred dimension to early water management throughout much of the region. Although rock pools likely served as the earliest water retention features in the region, it is clear that during the Iron Age humans began deliberately expanding pooling basins and creating them through activities of quarrying, excavation, and the construction of embankments, or bunds. This is evidenced not only at megalithic ritual sites, such as Hire Benkal, but also at settlement sites, where both cobble lined basins and constructed bunds are present. At the Iron Age habitation site of Kadebakele (northern Karnataka), for example, inhabitants modified the drainage pattern on top of a granitic hill to form a water catchment basin. Excavations in this reservoir show that it collected and held water for only part of the year, partially a consequence of the reservoir’s relatively small catchment area of .027 km2. The facility also experienced major drying episodes as well as significant siltation. Nevertheless, it certainly provided much-needed water to local residents at certain times. It is difficult to say to what extent these early water retention features may have supported cultivation. Most are quite small and lack provisions for water distribution necessary for large-scale agriculture. Moreover, they are often perched atop high hills with little cultivable land. However, some early reservoirs do occur within natural drainage ways, occasionally in association with check dams and at the base of topographic features where seasonal water could potentially be pooled for crop production. In fact, Devaraj et al. (1995: 66) report “interlaced, hydraulically laminated” deposits behind a “rammed” earthen construction near the base of a granite outcrop at Watgal – a Neolithic (3000–1000 BCE) to Iron Age (1000–500 BCE) period settlement site in northern Karnataka.1 It is difficult to characterize the entire range of variability in form and function of early water management constructions without more systematic work. Indeed, few regions have been systematically studied, and the patterns described above may not hold true across the entire region. Nevertheless, it is clear that water retention techniques began to be practiced in a variety of settings during the Iron Age (see also Allchin 1954). In addition, the development of water management technology during this period generally coincided with the introduction of new cultigens – including rice cultivation – suggesting that water retaining features became increasingly important to 1 However, it is important to note that the authors do not identify this feature as a reservoir, or attribute it with any water retaining “function.”

W

2210


agricultural production by the end of the first millennium BCE. In the first millennium AD there is stronger evidence for the construction of larger reservoirs that were used to meet the hydrological requirements of cultivation. A series of reservoir walls have been reported in the environs surrounding Sanchi – a well-known Buddhist monastic site in west-central Madhya Pradesh spanning the third century BCE to the twelfth century AD. These reservoir features are built of earthen embankments reinforced by stone masonry, with dams reaching nearly 6 m in height and expanding across drainage valleys, in some cases exceeding 1 km in length. Moreover, catchment areas range from 0.74 to 17 km2 (Shaw and Sutcliffe 2001), potentially damming considerably more water than the Iron Age reservoirs discussed above. Artifact associations, as well as the proximity of the embankments to the site of Sanchi, have allowed scholars to suggest that several of these features were constructed as early as the second century BCE, while others were built throughout the first millennium AD (Marshall 1940: 13; Shaw and Sutcliffe 2001). Although dating features such as reservoir walls remain problematic without direct geochronological assessment (e.g., radiocarbon, optically stimulated luminescence, etc.) or inscriptional data, textual references from the Early Historic Period suggest that reservoir construction was prevalent in some parts of India, and in South Asia more generally, during this time. For example, Chakravarti’s (1998) analysis of the Arthaśāstra – an economic and political treatise composed sometime in the late centuries BCE or early centuries AD – suggests that Mauryan rulers (ca. 324– 185 BCE) were concerned with establishing irrigation works. Moreover, numerous inscriptions and textual references indicate that the construction of reservoirs and irrigation facilities underwent remarkable development in Sri Lanka during the Early Historic Period, particularly on the island’s drier north and east sides. Northeast of Sigiriya, the Minneriya reservoir built during the reign of Mahasena (AD 276–303) is particularly noteworthy. This reservoir – fed by a canal as well as surface runoff – consisted of an embankment nearly 2 km in length and at places exceeded over 13 m in height. Inscriptions dated to the reign of Mahasena’s successor speak of “three harvests of [rice] paddy per year,” suggesting a marked impact on agricultural production (Gunawardana 1971: 8).

the purposes of agricultural intensification is most salient in South India after the Early Historic Period, and in Sri Lanka only slightly earlier. Numerous textual sources clearly indicate that reservoirs played an important role in the Early Middle Period (AD 500–1300). Although small dam-and-basin facilities for water impoundment continued to be built and used, Middle Period reservoirs typically consist of masonry-faced earthen dams thrown up across valleys, at the base of hills, and in other locations where seasonal runoff and small streams could be captured (Fig. 3) (Morrison 1993, in press). Like the Minneriya reservoir in Sri Lanka, some were supplied by canals, which took off via diversion weirs or anicuts, from perennial rivers or intermittent streams. Water was moved downstream from reservoirs to agricultural fields through masonrylined tunnels under the embankments, which were regulated by sluice gates (Figs. 4 and 5). Some water was also released over specially constructed waste weirs, facilities which range from boulder-filled cuts to elaborately built spillways (Fig. 6). Although the focus was clearly on the storage and downstream distribution of water, reservoir beds were also sometimes used for cultivation and reservoirs served as important sources of fish, silt and clay, and water for livestock. Reservoirs were also used to raise the water table around them, an important function even when the bed failed to fill. In fact, we have documented a consistent (but not universal) pattern of wells down-slope from sluice gates (Morrison in press). Reservoirs were particularly important in the far south, in present-day Tamil Nadu, where many of them were supplied by river-fed canals (Ludden 1999).

Middle Period Reservoirs: The “Traditional” System It should be clear from the above discussion that reservoirs and water storage features have a long history in South India, and South Asia more generally. However, the proliferation of large reservoir construction for

Water Management and Reservoirs in India and Sri Lanka. Fig. 3 A Middle Period reservoir embankment and sluice gate used to dam a valley below the Iron Age site of Hire Benkal (photo by Andrew M. Bauer).


2211

Water Management and Reservoirs in India and Sri Lanka. Fig. 4 Diagrammatic cross-section of a reservoir embankment and sluice gate.

Water Management and Reservoirs in India and Sri Lanka. Fig. 6 Masonry waste weir, or spillway, to regulate high water levels at the edge of a Middle Period reservoir embankment near the village of Kurugodu (photo by Kathleen D. Morrison).

Water Management and Reservoirs in India and Sri Lanka. Fig. 5 The northern sluice of the Daroji reservoir, south of Vijayanagara. This “double sluice” construction became prominent in the sixteenth century, when the Daroji embankment was constructed. This sluice and reservoir have been maintained and continue to operate (photo by Kathleen D. Morrison).

There it is possible to see perhaps the greatest elaboration of the so-called “system reservoirs” – long chains of reservoir facilities that flow one into the other, linking large areas into tightly knit watersheds (Sharma and Sharma 1990; see also Mosse 2003). Unfortunately, none of these systems has been specifically analyzed on the ground to determine precise construction sequences, so although we know of many specific single-reservoir projects dating as early as the seventh century AD, we cannot say exactly how the overall system functioned at this time or even how much of the landscape was under reservoir irrigation. It should be noted, however, that Early Middle Period reservoirs,

“traditional” by any reckoning, ranged widely in size from very small ponds to vast “seas,” the latter falling well within the contemporary definition of a “large dam.”2 Similar to the “systems reservoirs” of Tamil Nadu, large Sri Lankan Early Middle Period reservoirs were linked through vast networks of canals. The Minneriya reservoir, for example, was connected to other facilities via the construction of a “great canal” during the reign of Aggabodhi I (ca. AD 571–604), and composite maps of the irrigation facilities around Sigiriya show that it fed at least two large reservoirs – the Kaudulla and Kantalai reservoirs. Moreover, detailed survey has shown that smaller dam-and-basin features were also probably constructed and maintained throughout the period (Myrdal-Runebjer 1996). Again, it is difficult to know how the overall systems functioned simulta2 Crest length ≥ 500 m, maximum flood discharge ≥ 2,000 m3 s−1, “especially difficult foundation problems" or “unusual design" (ICOLD World Register of Dams 1998).

W

2212


neously. However, it is clear that Middle Period rulers and elite made demonstrative efforts to construct new irrigation facilities and repair older ones. In fact, the Minneriya inflow and outflow canals were repaired at least several times in the early centuries of the second millennium AD (Myrdal-Runebjer 1996; see also Brohier 1934; Gunawardana 1971). This pattern of extensive reservoir use in Sri Lanka and the far south of the Indian peninsula contrasts with that of drier regions in the northern interior of the subcontinent (Karnataka and parts of Andhra Pradesh). In these northern regions, reservoirs were (and are) almost exclusively runoff-fed and, given lower rainfall, they are generally not as closely spaced as those of the southern Tamil country. Still, many regions saw the use of both system and isolated reservoirs. In the area we have studied in northern Karnataka, reservoirs seem to have been only a minor component of Early Middle Period agricultural strategies. However, by the Late Middle Period (1300–1700 AD), and especially with the expansion of the large but loosely knit empire of Vijayanagara (ca. AD 1330–1600) across much of the peninsula, reservoir irrigation expanded considerably, especially in the drier zones where it had previously been limited. In and around the eponymous capital city of this empire, urban foundations in the early fourteenth century and the subsequent expansion of settlement and population explosion in the region propelled reservoirs into increasingly important components of larger agrarian and political strategies (Fig. 7). Important from the start of the Vijayanagara period, reservoirs also constituted a key form of agricultural intensification in the sixteenth century, or Late Vijayanagara period, especially in regions where canal irrigation was not feasible (Morrison 1995, 2001). Reservoirs played variable roles in the processes of Vijayanagara agricultural intensification and change; this variation was structured as much by political

Water Management and Reservoirs in India and Sri Lanka. Fig. 7 Small Middle Period reservoir and sluice gate in the Daroji Valley, south of Vijayanagara (photo by Kathleen D. Morrison).

factors and settlement dynamics as by environmental variables such as runoff and soil conditions. What is common to most parts of the urban hinterland, however, is the way in which the vast majority of reservoirs fell out of use after (in some cases, during) the Vijayanagara period. Very few of the reservoirs from the original system still effectively function, though there are a few notable “living” reservoirs with long histories of maintenance and reconstruction (Morrison 1993, 1995). For example, the Daroji reservoir, the terminus of one of the largest systems in our study area, continues to collect runoff from a catchment area of 955 km2 and provides water for downstream agricultural fields (Fig. 8). Research on Middle Period reservoirs includes analyses of pollen and charcoal from reservoir sediments (which allow the reconstruction of fire and vegetation histories), sedimentological studies of reservoir fill and changes to erosional regimes and local hydrology, and stylistic analyses of sluice and embankment construction (e.g., Morrison 1994, 1995; Myrdal-Runebjer 1994, 1996). In addition, we have also considered tens of thousands of textual and inscriptional records describing facility construction and

Water Management and Reservoirs in India and Sri Lanka. Fig. 8 ASTER satellite image of the Daroji reservoir first built in the sixteenth century, south of Vijayanagara. Notice the strong growth of agricultural crops indicated by the reflection of near-infrared wavelengths (shown here in red) downstream from the embankment. Vegetation growth is also occurring on the edges of the reservoir as receding water exposes moisture retaining sediments (data originally obtained from NASA).


maintenance as well as conflicts over water, land, labor, and rule (Kotraiah 1995; Morrison 1995; Morrison and Lycett 1994, 1997). All of these diverse lines of evidence suggest that Middle Period reservoirs, like their contemporary and colonial counterparts (Mosse 2003: 45–46), were highly unreliable sources of irrigation. Runoff-fed reservoirs, in particular, may fail to fill in dry years; in the drier districts, this meant not only that reservoirs could usually not support wet crops such as rice, but also that dry crops such as millets might not be assisted by the facility. The situation was somewhat better in areas of higher rainfall, but everywhere in southern India reservoirs are marked by high evaporation rates, high siltation rates, and ongoing maintenance challenges (Fig. 9).

Reservoirs, Politics, and Contemporary Development Across much of South Asia, contemporary “big dam” projects have been cast as the legacies of high modernist social and environmental engineering (cf. Gadgil and Guha 2000; Scott 1998). In India, well-organized and highly visible social protests have been made against specific projects such as the Sardar Sarovar project and others along the Narmada River. Often, antidam groups suggest that the answer to sustainable and equable development lies in a return to a “traditional” system of technology and management. It is not our purpose to argue the case for “modern” or “traditional” forms of water distribution. Rather, we wish to suggest that arguments that hinge on the dichotomies of large/small, equitable/inequitable, and political/apolitical that often accompany distinctions between “modern” and “traditional” forms of water management are inappropriate. The notion that large political dam projects are entirely a product of modernity is decidedly incorrect.

2213

This is no clearer than in South India, where some Middle Period reservoirs were created with embankments more than 3.5 km in length and over 15 m in height. As already noted, the Daroji reservoir in northern Karnataka pooled runoff over a total catchment area of nearly 1,000 km2; however, size alone is not the issue. Both small and large reservoirs were deeply political projects, tied to networks of patronage and power. Middle Period reservoir construction was sponsored by a wide range of political leaders from kings (rarely) to local chiefs (commonly) and was also connected with Hindu temples in a number of ways (Morrison 1995; Morrison and Lycett 1994, 1997). Moreover, even during the Iron Age and Early Historic Periods, shifts from a reliance on rain-fed agriculture to reservoir irrigation would have produced a new material order on the landscape that was necessarily related to sociopolitical fields. Indeed, the questions – How was labor mobilized? Who had access to irrigation water? And how was this decided? – are entirely appropriate in examining the entire history of water management. The answers to these questions undoubtedly had sociopolitical ramifications for ancient inhabitants. This may have especially been the case when irrigation accompanied the introduction of new cultigens and cultural values of cuisine shifted. It is not difficult to imagine the profound social consequences of being a dry-farming millet producer/consumer versus a wet-farming rice producer/consumer when rice was a high status cultigen and the preferred donation to Hindu temple gods during the Middle Period, for example. The above considerations of the (un)reliability and the political nature of historic reservoirs of South India and Sri Lanka are not to suggest that these impressive systems have no contemporary value with regard to (re) developing water management strategies. Quite the opposite is true: work on the 3,000-year history of irrigation in southern India shows both success and failure in equal measure, portents for a reasonably hopeful future. Thus, although there is no simple solution to the water problems of the dry tropics of South Asia, surely an informed perspective on the actual historical experiences of the region must provide a more secure basis for future planning than either a romantic and unrealistic view of “tradition” or a blind faith in “modern” science and technology.

References

Water Management and Reservoirs in India and Sri Lanka. Fig. 9 Photo showing the accumulation of an estimated 2–3 m of silt and clay, to near the top of a sluice gate. This reservoir is adjacent to the village of Avinamodugu, south of Vijayanagara (photo by Kathleen D. Morrison).

Allchin, F. R. The Development of Early Culture in the Raichur District of Hyderabad. Ph.D. Dissertation, University of London, 1954. Brohier, R. L. Ancient Irrigation Works in Ceylon. Colombo: Ceylon Government Press, 1934. Chakravarti, R. The Creation and Expansion of Settlements and Management of Hydraulic Resources in Ancient India. Nature and the Orient: The Environmental History of

W

2214

Water management in Peru

South and Southeast Asia. Ed. R. H. Grove, V. Damodaran, S. Sangwan. Delhi: Oxford University Press, 1998. 87–105. Devaraj, D. V., J. G. Shaffer, C. S. Patil, and Balasubramanya. The Watgal Excavations: An Interim Report. Man and Environment 20 (1995): 57–74. Fernando, A. D. N. Resource Maps of Sri Lanka. Colombo: Integrated Development Association of Sri Lanka, 1976. Gadgil, M. and R. Guha. The Use and Abuse of Nature. Delhi: Oxford University Press, 2000. Gordon, D. H. and F. R. Allchin. Rock Painting and Engravings in Raichur, Hyderabad. Man 55 (1955): 97–9. Gunawardana, R. A. L. H. Irrigation and Hydraulic Society in Early Medieval Ceylon. Past and Present 53 (1971): 3–27. India Meteorological Department. Climatological Atlas of India: Part A (Rain Fall). New Delhi: Government of India, 1981. Kotraiah, C. T. M. Irrigation Systems Under Vijayanagara Empire. Mysore: Directorate of Archaeology and Museums, 1995. Ludden, D. An Agrarian History of South Asia. Cambridge: Cambridge University Press, 1999. Marshall, J. The Monuments of Sanchi. London: Probsthain, 1940. Morrison, K. D. Supplying the City: The Role of Reservoirs in an Indian Urban Landscape. Asian Perspectives 32 (1993): 133–51. ---. Monitoring Regional Fire History Through Size-Specific Analysis of Microscopic Charcoal: The Last 600 Years in South India. Journal of Archaeological Science 21 (1994): 675–85. ---. Fields of Victory: Vijayanagara and the Course of Intensification. Berkeley: Contributions of the University of California Archaeological Research Facility, No. 53, 1995 (Rpt. New Delhi: Munshiram Manoharlal, 2000). ---. Coercion, Resistance, and Hierarchy: Local Processes and Imperial Strategies in the Vijayanagara Empire. Empires: Perspectives from Archaeology and History. Ed. S. Alcock, T. D’Altroy, K. Morrison, and C. Sinopoli. Cambridge: Cambridge University Press, 2001. 253–78. ---. The Daroji Valley: Landscape History, Place, and the Making of a Dryland Reservoir System. Vijayanagara Research Project Monographs. Delhi: Manohar Press, in press. Morrison, K. D. and M. T. Lycett. Centralized Power, Centralized Authority? Ideological Claims and Archaeological Patterns. Asian Perspectives 33 (1994): 312–53. ---. Inscriptions as Artifacts: Precolonial South India and the Analysis of Texts. Journal of Archaeological Method and Theory 3 (1997): 215–37. Mosse, D. The Rule of Water: Statecraft, Ecology, and Collective Action in South India. Delhi: Oxford University Press, 2003. Myrdal-Runebjer, E. Vävala väva – Sigiri Mahaväva Irrigation System: Preliminary Results from an Archaeological Case Study. South Asian Archaeology, Series B. 271. Ed. Asko Parpola, and Petteri Koskikallio. Helsinki: Annales Academiae Scientiarum Fennicae, 1994. 551–62. ---. Rice and Millet: An Archaeological Case Study of a Sri Lankan Transbasin Irrigation System. Goteborg: Goteborg University, 1996. Porembski, S. and W. Barthlott, ed. Inselbergs: Biotic Diversity of Isolated Rock Outcrops in Tropical and Temperate Regions. New York: Springer, 2000.

Scott, J. C. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven, CT: Yale University Press, 1998. Seneviratne, S. The Archaeology of the Megalithic – Black and Red Ware Complex in Sri Lanka. Ancient Ceylon 5 (1984): 237–307. Sharma, R. K. and T. K. Sharma. Textbook of Irrigation Engineering. Vol. I. Irrigation and Drainage. New Delhi: Oxford, 1990. Shaw, J. and J. Sutcliffe. Ancient Irrigation Works in the Sanchi Area: An Archaeological and Hydrological Investigation. South Asian Studies 17 (2001): 55–75. Thomas, M. F. Geomorphology in the Tropics: A Study of Weathering and Denudation in Low Latitudes. Chichester: John Willey & Sons, 1994.

Water Management in Peru C. R. O RTLOFF The Chimu Empire of ancient South America in the time period between 900–1480 CE dominated the north Peruvian coast from the Santa to the Lambeyeque Valleys west of the Andean Cordilera Negra mountain range. This region, in terms of present-day geographical locations, extended just north of the Peruvian capital city of Lima to the Ecuadorian border and eastward from the Pacific Ocean coast to the eastern slopes of the Andes. From the central administrative center at Chan Chan in the Moche Valley, successive generations of Chimu rulers exercised political and economic control of adjacent valleys through administrative centers charged with overseeing and maximizing agricultural production and development. Within the territorial domain of the empire, many of the westward-running rivers leading runoff water from highland Andean rainfall collection zones to fertile coastal fluvial valleys were intercepted and redistributed through extensive canal distribution systems to irrigate agricultural fields. Many valleys even today still contain well preserved canal networks from this era as land under cultivation by the Chimu far exceeded present-day cultivation areas by perhaps as much as 50 per cent. While only a few of the ancient valley irrigation systems have been extensively explored, mapped and analyzed to any extent (Eling 1986; Kosok 1965; Ortloff et al. 1982, 1985; Ortloff 1988, 1993), some novel canal hydraulic control features have been recently discovered that warrant analysis as they provide a window into the level of hydraulic science existing within the Chimu empire of ancient South America. Since most major valleys under Chimu control showed evidence of massive state-sponsored, hydraulic canal infrastructures to support irrigation agriculture, it


follows that a hydraulic science was co-developed to provide tailored flow rate canals to supply field systems distant from river inlets. As considerations of soil and crop types, crop water demands, alternate field system watering strategies, valley topography and defense from drought and large rainfall runoff events influence canal design and placement, an accompanying hydraulic science with flexibility to design and modify canals according to these considerations is expected in the archaeological record. Most probably, cumulative observations of water flow phenomena over time served to provide a database for empirical design principles used for canal layout and design. Canal volumetric flow rate relies on inlet geometry, bed slope, canal cross-section, wall roughness and subcritical, critical, and supercritical flow characterization. Allied technical disciplines related to route layout, surveying, water delivery sequencing and water routing through multiple canal branches are additional key technologies vital to understanding Chimu hydraulics practice. While investigation of all aspects of applied Chimu hydraulic science would give a complete picture of their technology base, the present investigation is focused on but one facet of the Chimu engineering repertoire: hydraulic control systems. This entry details an investigation of a recently discovered canal hydraulic feature of the Chimu Talambo-Farfán Canal located in the Jequetepeque Valley of northern Peru. While the remains of the canal system are of late Chimu origin due to association with the Chimu site of Farfán, some upstream versions of canal segments may be associated with earlier valley occupation by Gallinazo and late period Moche occupants. For the present analysis, however, only the last Chimu phase of canal construction is considered.

Analysis Results The Talambo-Farfán Canal originates far upstream in the valley neck of the Jequetepeque River and passes late hillside Chimu occupation zones in the Talambo region through a series of aqueducts and deep canal cuts through upvalley low hills before finally emerging onto the agricultural plains east of Farfán. The canal was further extended to provide water to the extensive agricultural field zones directly south of the Chimu mountain redoubt of Farfán Sur. In the canal extension region, deep quebradas (canyons) were formed over time from successive El Niño rainfall runoff events sculpting deep erosion channels into the soft soil deposits that formed natural obstacles to canal extension. In order to bridge the multiple quebradas, a series of three large earth-fill aqueduct structures and many small aqueducts were constructed to extend the canal to the vast Pampa de Faclo bordering the site of Pacatnamú. Two of these (largely destroyed)

2215

quebrada-crossing aqueducts are shown in Figs. 1 and 2; the Hoya Hondada aqueduct is the largest and final aqueduct in the downstream sequence and is shown in Figs. 3–5.

Water Management in Peru. Fig. 1 View of one of the three destroyed aqueducts breaching a deep, erosionally-incised quebrada upstream from the Hoya Hondada Aqueduct. Photo by C. R. Ortloff.

Water Management in Peru. Fig. 2 Another view of a further destroyed aqueduct in the upstream sequence from the Hoya Hondada Aqueduct. Photo by C. R. Ortloff.

W

Water Management in Peru. Fig. 3 View of the low Hoya Hondada aqueduct across the wide quebrada. Photo by C. R. Ortloff.

2216


Water Management in Peru. Fig. 4 Alternate view of the Hoya Hondada aqueduct. Photo by C. R. Ortloff.

Water Management in Peru. Fig. 5 View of the cross-section of the Hoya Hondada aqueduct at the destroyed south end; height of the channel approximately 5 m from ground level. Photo by C. R. Ortloff.

Due to the much larger depth and width of the Hoya Hondada quebrada as compared to upstream quebradas, aqueduct design changes from upstream high fill height designs breaching narrow and deep gorges to maintain slope to a long, low height aqueduct deep within the quebrada with a steep 40 degree approach chute from the southernmost upstream bluff. After the canal passes over this low aqueduct, it continues to the downstream northern sidewall of the Hoya Hondada quebrada to field systems located on the southern boundary of Pampa de Faclo approximately 8 km south of the ancient Moche-Chimu religious center at Pacatnamú. The intent of the canal extension through difficult terrain north and west of Farfán Sur was apparently to provide water to settlements and field systems in the south part of the pampa east of Pacatmanú. While further canal extension to the city limits of Pacatnamú may have been the ultimate intent of the canal builders, traces of a continuous connection path to the many canal fragments on the Pampa de Faclo are yet to be discovered.

Water Management in Peru. Fig. 6 Demonstration of a hydraulic jump in a hydraulic test channel induced by a supercritical flow interacting with a plate obstacle; a similar hydraulic phenomena is induced at the channel intersection point of the steep chute and low slope Hoya Hondada aqueduct channel. Photo by C. R. Ortloff.

The canal and aqueduct design within the Hoya Hondada quebrada area contains many novel hydraulic features indicative of the state of Chimu hydraulic knowledge. As opposed to using a massive earth-fill aqueduct that would span the 150 m width and 20 m height to connect opposite sides of the quebrada with a constant slope canal, a 5 m high aqueduct is constructed low in the quebrada with a 25 m long steep chute joining the upstream high elevation part of the canal to the downstream low aqueduct. This alternate design eliminates the possibility of a massive high dam structure trapping a large height of quebrada rainfall runoff behind it that would constitute a dam breakage hazard to downstream occupation areas. The low dam design, should breakthrough occur, would be easily repairable due to its low volume of fill material and contain relatively less impounded water than the alternate design. While some Chimu aqueducts spanning quebradas contained culverts or boulder bases to let water pass underneath to alleviate water damming during heavy rainfall events, use of a low aqueduct height design reduced backwater hydrostatic pressure effects that lowered destructive transverse forces that could lead to breakthrough failure. Continuing, with no upstream hydraulic controls, the effect of a high velocity, high volumetric flow rate down the steep 40-degree downward-sloped chute to the low slope aqueduct would be to create a massive hydraulic jump at the angle-change junction. Fig. 6 shows a typical hydraulic jump created by an obstacle placed in a high velocity, supercritical flow – in this case, the hydraulic jump is typical of that resulting from the effect of the rapid flow down the steep chute – low slope aqueduct junction. The severity and height of a large hydraulic jump is sufficient to destroy the aqueduct by turbulent erosion’s acting on the unlined


aqueduct structure unless a hydraulic control is in place upstream of the aqueduct to dissipate stream energy, lower stream velocity and/or remove excess water from the canal by a side weir – which is another mode of stream energy reduction. The height of the hydraulic jump also would cause sidewall overflow from the aqueduct leading to precious irrigation water loss for downstream field systems. By lowering the channel water velocity and/or volumetric flow rate, the height of the downstream hydraulic jump can be reduced to a level that does not imperil the integrity of the aqueduct structure or cause sidewall overflow water loss problems. This then constitutes the hydraulic design problem faced by Chimu engineers. The channel upstream of the Hoya Hondada aqueduct shows that an energy dissipation hydraulic structure had been installed to influence flow leading on to the aqueduct. Figs. 7 and 8 show a ground view of this hydraulic control structure in its unexcavated state. The hydraulic structure consists of two pairs of opposing boulders with a 70 cm separation distance between boulders and with a 13.2 m downstream separation distance between boulder pairs. The

2217

channel containing this structure also had variations in width and sidewall angle. A FLOW-3D computer model (FLOW-3D 2004) incorporating field measured channel geometry is shown in Fig. 9. In the distance between the boulder pairs, a side overflow weir that activates when water height exceeds a given height is in place that led to a channel that guided overflow water into the quebrada downstream of the Hoya Hondada aqueduct. The hydraulic function of the control structure is next described by use of numerical solution of the governing equations of fluid flow; results of the solutions indicate the free surface shape and internal velocity within the channel and provide the basis to understand the hydraulic functioning of the control section. The first upstream choke (defined as the open separation zone between boulder pairs) apparently forms a hydraulic jump at incoming stream velocities past about 7 m/s while the downstream choke apparently forms a hydraulic jump at inlet velocities about 3 to 4 m/s. The second downstream choke is therefore controlling as it actives first at a lower limiting velocity. For all cases, an increase in water velocity occurs in the zone downstream of the first upstream choke due to

W Water Management in Peru. Fig. 7 Unexcavated view of the hydraulic control structure upstream of the chute and the Hoya Hondada aqueduct; structure consists of opposing boulder pairs in the main channel separated by a 13.2 m distance. Photo by C. R. Ortloff.

Water Management in Peru. Fig. 8 Ground view of the hydraulic control structure upstream of the chute and Hoya Hondada aqueduct. Photo by C. R. Ortloff.

2218


Water Management in Peru. Fig. 9 FLOW-3D computer model of the excavated hydraulic control section upstream of the chute and Hoya Hondada aqueduct from measured field data. Note the presence of the side overflow weir. Photo by C. R. Ortloff. Water Management in Peru. Fig. 11 Solution results for a channel flow rate less than 3 m3/sec; note that flow is fully contained within the channel and the side overflow weir is not activated. Photo by C. R. Ortloff.

Water Management in Peru. Fig. 10 Solution results for water free surface geometry for a channel flow rate exceeding 3 m3/sec; note the spillage over the side weir from the action of the downstream choke limiting the transmitted flow rate on to the Hoya Hondada Aqueduct. Photo by C. R. Ortloff.

water gravitational acceleration on a hydraulically steep slope influenced by expansive channel cross-sectional geometry changes. These velocity changes account for the different hydraulic behavior of each choke. Since flow into the second choke is supercritical over the range of subcritical input flow rates, the second choke forms a hydraulic jump and the flow rate is limited by critical conditions at the throat. As the choked flow rate of the downstream choke is less than that of the upstream choke, the flow rate difference results in a water height

Water Management in Peru. Fig. 12 Plot of flow rate (m3/sec) into the control structure vs. flow rate exiting: the difference is the flow exiting over the side weir. Note that transmitted flow rate is limited past 3 m3/sec due to the action of the choke system and overflow weir activation. Photo by C. R. Ortloff.

change between chokes that activates water overflow from the side weir. The overflow flow rate is equal to the volumetric flow rate difference between up- and downstream chokes. Fig. 10 shows side weir overspillage typical of inlet flows exceeding about 7 m3/s; Fig. 11 shows conditions obtaining when the inlet volumetric flow rate is less than 3 m3/s and the side weir is inactive. Fig. 12 shows that the flow rate passed on to the Hoya Hondada aqueduct is automatically altered past 3 m3/sec inlet flow due to activation of the overflow weir. The upper curve represents a linear condition present when no control section is present and the input


flow rate is equal to the output flow rate in the channel. The lower curve indicates that when the control section is in place, the output flow exiting from the control section is reduced by the side weir overflow to a limiting value about equal to 3 m3/s. The limiting flow rate of 3 m3/s then ultimately limits the height of the downstream hydraulic jump before the aqueduct to acceptable design values to control turbulence generation and contain the flow in the aqueduct channel without sidewall overflow. The double choke system described is augmented with yet a further downstream energy dissipation system on the steep sloped chute. The Fig. 13 contour map, which shows the channel depression where the control section exists, contains a leadoff bifurcation channel that subtracts a fraction of the water downstream of the hydraulic control structure and leads it around a small hill into a low slope channel. This flow rejoins the main channel flow at a downstream junction point. The lower subcritical velocity stream from the lower slope channel acting on the main supercritical stream at the channel junction creates a further energy dissipating oblique hydraulic jump in the channel. The hydraulic jump converts velocity kinetic energy into random turbulent and potential, height change energy. This has the effect of subtracting further energy from the stream approaching the chute junction zone. Note that no energy gain exists as subtracted flow from the main channel is added back into the downstream flow to maintain the constant flow rate. The net effect of the two energy reduction controls – the first from subtracting energetic water by means of the overflow

Water Management in Peru. Fig. 13 View of a minor hydraulic control system found extensively in the Pampa Faclo field system. The presence of a dual, opposing stone choke downstream of a steep chute creates a hydraulic jump ahead of the choke. This reduces the velocity of water entering a field system to promote gradual water absorption into the soil over time. The inlet from the main canal is tailored only to let a certain amount of water into the chute to make the system work as planned, indicating careful design of this irrigation system control. Photo by C. R. Ortloff.

2219

weir, the second from hydraulic jump energy dissipation – is to limit the channel flow rate and also reduce entry velocity onto the Hoya Hondada aqueduct to limit hydraulic jump erosive and overflow damage. Although a large part of the steep slope channel is no longer extant, there is indication that stones placed in the chute bed provided yet a further energy dissipation mechanism to reduce stream velocity. The totality of these hydraulic controls then serve to limit erosion damage to the Hoya Hondada aqueduct – particularly under El Niño conditions. For example, as the coastal area is frequently subject to massive El Niño rainfall runoff from the nearby Cerro Faclo mountain range, large canal flow rates arising from water washing into the canal could easily produce excessive canal flow conditions beyond the design flow rate. An excessive flow rate could destroy the Hoya Hondada aqueduct by creating a massive hydraulic jump at the channelaqueduct junction. By diverting water over the weir into a channel leading to the downstream side of the aqueduct together with use of additional energy dissipation controls, excess water and energy is removed from the canal to provide a protective hydraulic feature for the aqueduct. As the canal can transport up to 3 m3/s before weir overflow activation, this flow rate represents a sustainable maximum flow value to a location 35 km from the river inlet. This flow rate may be available to the Pampa Faclo fields and settlements, despite evaporation and seepage losses, by blocking all other canal branches from the main canal to temporarily direct all flow to the Pampa Faclo. As the needed water volume delivered through each canal branch to a field sector was a known quantity to sustain crops in that sector according to their importance, yield, and water requirements, the assignment of water volumes therefore was metered for maximum agricultural productivity effect. As settlements on the outlying southern section of Pampa Faclo were somewhat remote from the Farfán center and field systems discovered to date rather limited in size, it may be surmised that water resources reserved for this sector were proportional to its potential expansion and importance. The fact that water could be directed to this secondary location perhaps indicates that further expansion of the area’s agricultural potential was anticipated but still in the construction phase. The major effort to construct multiple elaborate aqueduct structures to bring water to this zone through difficult topography perhaps indicated that a valuable agricultural resource in the Pampa Faclo land area overrode labor-to-build considerations – perhaps due to population increase pressures. The transition to Chimu from Moche eras in the seventh century CE sees a vast expansion of population, cities, administrative centers, agricultural and

W

2220

Water management in Petra

settlement zones as well as a development of hydraulic technology that permits intra- and intervalley canal network development to exploit available agricultural zones. The success of the agricultural program, supported by a knowledge base of technical achievements, undoubtedly underlies this expansion. The present study adds an example of an application of hydraulic technology that played a role to protect critical aqueduct structures from flood damage to ensure field system survival through time. Other examples of steep-slope channel constrictions formed by opposing stones set with a narrow opening are numerous within the Farfán field system area (Fig. 13) indicate an understanding of creative use of hydraulics knowledge to enhance the efficient use of field systems. Basically channel constrictions of this type create a hydraulic jump ahead of a constriction with a high height, low velocity flow leaving the throat that flows slowly into downstream distributive channels within the field system to regulate water seepage rates into the soil. The channel size, slope and inlet configuration supplying water to maintain a stable hydraulic jump without overflow must therefore be intelligently constructed to allow just a sufficient flow rate for that purpose. This design capability appears to have been well understood by Chimu engineers as demonstrated by many canal design examples. Examples of hydraulic controls in the form of flow rate limiting and velocity-reducing hydraulic structures point to a yet little explored creative aspect of Chimu irrigation agricultural practice and the Chimu hydraulic science knowledge base. While examples illustrating applications of Chimu hydraulics engineering are somewhat limited by scant exploration and analysis to date, undoubtedly more remains to be discovered as focus is given to exploration and analysis of hydraulic features. By computer analysis of such systems, aspects of Chimu hydraulic science will be revealed and point to a new evaluation of the contribution of indigenous South American cultures to the hydraulic sciences.

References Eling, H. PreHispanic Irrigation Sources and Systems in the Jequetepeque Valley, Northern Peru. Eds. M. Turpin, H. Eling and R. Matos. Andean Archaeology. UCLA Archaeology Publications, 1986: 130–49. FLOW-3D. V.8.4 Software-Flow Science, Inc., Santa Fe, New Mexico, 2004. Kosok, P. Life, Land and Water in Ancient Peru. New York: Long Island University Press, 1965. Ortloff, C. R. Chimu Hydraulics Technology and Statecraft on the North Coast of Peru. Economic Aspects of Water Management in the Prehispanic New World. Eds. V. Scarborough and B. Isaac. Greenwich, Connecticut: JAI Press, 1993. 327–87. ---. Canal Builders of Ancient Peru. Scientific American 256.12 (1988): 100–7.

Ortloff, C. R., Moseley, M. E. and R. Feldman. Hydraulic Engineering Aspects of the Chimu Chicama- Moche Intervalley Canal. American Antiquity 48 (1982): 572–99. ---. Hydraulic Engineering and Historical Aspects of the PreColumbian Canal Systems of the Moche Valley, Peru. Journal of Field Archaeology 12.1 (1985): 77–87.

Water Management in Petra C. R. O RT L O F F Many scholars have studied the political history of Petra (Taylor 2001; Guzzo and Schneider 2002; Glueck 1959, 1965; Hammond 1973; Levy 1999; Auge and Denzer 2000; Bowersock 1983; Bourbon 1999; Markoe 2003). Despite the fact that control of water is essential to an understanding of life in the desert, there has been little scholarship on hydraulic engineering at this site. Figure 1 shows details of the supply and distribution system leading water to the urban core of Petra. Numbered locations denote major buildings, temples, and site features listed in the Appendix. Shown are major dams (d-), minor dams (d), cisterns (c), water distribution tanks (T), and springs (S). The grid system (A,B,C;1,2,3) serves to define an area coordinate system composed of 1.0 km2 grid boxes to locate various features mentioned in the text. The urban core of Petra lies in a valley surrounded by high mountainous terrain. Seasonal rainfall runoff passes into the valley through many canyon streambeds (wadis) and drains out through Wadi Siyagh (A;2). While water storage is a partial key to the city’s survival, a number of springs internal and external to the city (Fig. 1: Ain Mousa, Ain Umm Sar’ab, Ain Braq, Ain Dibdiba, Ain Ammon, al Beidha, Ain Bebdbeh) provided water channeled and/or piped to the urban center. The main Petra water supply originated from the Ain Mousa spring about 7.0 km east of the town of Wadi Mousa (Fig. 1, D;1) combined with waters of the minor Ain Umm Sar’ab spring; this supply still serves the associated tourist complex (2), (3) located outside of the Siq (C;1) entrance (10). The Siq is a 2 km long, narrow passage through high mountains bordering the eastern part of the city core area. Early phases of water supply utilized Ain Mousa spring water channeled through the Siq (dashed line, 29, B;2) to the urban core of the city as far as Q’asr al Bint (29) with final drainage into the Wadi Siyagh (A;2). Due to dam and flood bypass tunnel construction at the Siq entrance and infilling and paving of the Siq floor both in ancient and modern times to reduce flooding, the channel now lies under the current pavement surface attributed to Nabataean construction under Aretas IV and later


2221

Water Management in Petra. Fig. 1 Petra site features and water distribution systems.

Roman paving efforts. Recent excavations in front of the Treasury (11, C;1) have revealed remnants of this early open channel. While this channel provided water to early, low population phases of the city, the later concentration of urban settlement areas north and south of Wadi Mousa (B;2) represented a transition towards full city status with a cosmopolitan society involved in trade and commerce. With demands to increase water supply to spreading urban settlement areas resulting from population increase and a desire to match the city’s prosperity from trade with an appropriate elevation in symbols of success, extensive use of pipelines followed to bring larger amounts of water to areas not reachable by the low elevation, open channel system. Pipeline systems, however, introduced new hydraulic design complexities that involved knowledge of ways to maintain stable piping flow whose maximum transport flow rate matched (or exceeded) spring flow rate input. Flows in poorly designed pipeline systems are capable of a number of transient, self-destructive hydraulic instabilities (e.g., water hammer, pressure surges, transient wave structure, flow intermittency, or internal oscillatory hydraulic jumps). Thus analysis of Petra’s piping systems provides insight into the available technical knowledge base applied to problem solution. Pipeline routing involved constant angle contour path surveying through rugged, mountainous terrain as well as choices of hydraulic technical parameters (slope, diameter, internal wall roughness, sinuosity, and supply head) that govern piping carrying capacity. Choices of these parameters, as extracted from the archaeological

Water Management in Petra. Fig. 2 The Zurraba (al Birka) reservoir.

record, as well as insights into the management strategy of these assets, indicate the level of technical achievement of Nabataean engineers. One example of later phase technological advances is a reservoir at Zurraba (1, D;2) (Figs. 1 and 2). It was constructed to store and transmit water along the Wadi Shab Qais (D;2) around the northern flank of the Jebel el Khubtha mountain (C;2) in an elevated channel (D;3, 40) containing piping (Fig. 3) that continued over royal tombs (22–24) to supply a typical large basin at its terminus (Fig. 4). Descending channels from this basin to cisterns at the base of the mountain added water to that collected from rainfall runoff for urban housing needs, celebratory rituals at nearby tomb complexes and for piping elements conducting water

W

2222


Water Management in Petra. Fig. 3 Elevated water channel/piping system on west face of el Khubtha Mountain.

Water Management in Petra. Fig. 5 (a) Piping channel elements on the north side of the Siq. (b) Channel trough on the north side of the Siq for piping placement. Water Management in Petra. Fig. 4 Elevated settling basin typical of the el Khubtha channel/piping system.

further on to city center locations. While runoff capture is one probable sourcing of the Zurraba reservoir, connection to local spring sources, including Ain Mousa, remains probable for additional charging. Although reservoir water could be used to supplement the Siq open channel flow, later city phases involved shifting Ain Mousa water supplies to a Siq piping system after the open channel was abandoned. In this case, rainfall runoff and spring charging still enabled the Zurraba reservoir to supplement Ain Mousa Siq pipeline flows when required. The Jebel el Khubtha pipeline appears to be the main outflow path for reservoir water along Wadi Mataha (B;2, C;2) and surplus water, after cistern topping, was most certainly directed to the main city fountain (Nymphaeum 42, B;2) through a pipeline (as some pipeline fragments in the area suggest). From a systems point of view, the reservoir served principally to maintain cistern levels by on-demand water release while the Ain Mousa spring provided the continuous spring supply source to the Nymphaeum through piping supported in a channel through the Siq (Fig. 5a, b; Ortloff 2005). The ability to provide an “on-demand” water supply from this backup

source would prove most useful to large caravans entering the city that would place a sudden demand on water supply capability.

Pipeline Carrying Capacity Considerations: The Zurraba–Jebel el Khubtha System While a spring produces a given volumetric flow rate, a limitation on how much can be transported by pipeline relates to pipeline technical characteristics (diameter, internal roughness, slope and supply head). Piping design considerations require the spring output flow rate to match (or be less than) the theoretical carrying capacity of the pipeline. Technical examination of Nabataean pipeline designs then yields insights into solutions to increase flow rate throughput. If the long Jebel el Khubtha piping system were to function in fullflow mode typical of a very low-flow rate, its flow rate would be somewhat less than that derived from an open channel, near critical flow mode due to internal wall frictional resistance; larger diameter piping would then be required to match the spring flow rate. For a steeper slope design, gravitational acceleration causes flows to become supercritical and tend to an internal free surface normal depth. Rapid supercritical flows, however,


may be subject to intermittent zones of subcritical full flow induced by internal piping wall roughness and curvature resistance effects as well as transient hydraulic jumps that create pulsations in delivery flow rate at the piping exit. Such effects can lead to destructive tensile forces that weaken mortared piping joints. The best piping design to produce a stable volumetric flow rate is therefore a partially full flow at near, but below, critical conditions that empties water gently into a terminal reservoir. Selection of this piping design would then be a measure of knowledge of hydraulic principles required to achieve a steady, high flow rate to the terminal elevated basin placed far left in C;2 and would explain the high elevation positioning of this pipeline (to maintain a low slope) around Jebel el Khubtha. The Nabataean design, given its slope and piping diameter, closely matches the near critical flow rate (Morris and Wiggert 1972) and provides for the largest possible flow rate from the Zurraba reservoir to meet on-demand large flow rate requirements. Additional benefit from the Nabataean design resides in the presence of partial, open channel flow in the piping to greatly reduce the leakage rate compared to a pressurized system. Since particles settle in the reservoir, no particle transport occurs to clog piping; this is particularly important as access to the high elevation piping (25 m above the ground) on the near vertical Jebel el Khubtha mountain face would limit the possibility of cleaning procedures. The combination of all these features designed into the Jebel el Khubtha pipeline indicates that much thought went into the best placement and design of this system to achieve the multiple goals that ensured not only system longevity but also rapid, on-demand water delivery capability with minimum leakage. From an exit pipeline leaving the basin to a ground level cistern, additional piping led to the Nymphaeum fountain to complete the Jebel el Khubtha circuit from the Zurraba reservoir. As the Nymphaeum was a major water supply to the urban core and market areas, much effort and innovation was employed to guarantee its year-round functioning from the Siq piping system (Ortloff 2005) supplemented, on-demand, by the long Jebel el Khubtha pipeline from the Zurraba reservoir.

2223

and replaced by a Siq north wall pipeline system (Fig. 5a, b) that extended to the area across from the theater district (B;1) and ended at the Nymphaeum. Thus at least two separate supply lines (Siq and Jebel el Khubtha) led to the Nymphaeum to ensure supply redundancy. The construction for the Siq pipeline system is generally attributed to Malichus II or his predecessors, Aretas IV or Obodas III, in the first century BCE or early in the first century CE (Guzzo and Schneider 2002). Since water demands south of Wadi Mousa (transecting urban Petra) are high due to the nearby marketplace, theater, temple, and housing districts and significant water resources are available from the north side piping systems, a pipeline transfer connection from the Nymphaeum to this area is logical for development of this area. While a bridge or piping from the north side of the Jebel el Khubtha system in the El Hubtar Necropolis area (20, B;2) across the Wadi Mousa may have existed in the vicinity of the theater (19, B;1) to carry water (at the same level) to the south side, traces are lost due to extensive erosion flood damage. In addition to water delivered by these means, the theater water source was supplemented from large, upper level reservoirs in the Wadi Farasa area, and pipelines originating from Ain Braq and Ain Ammon sources (Fig. 1) again indicating built-in supply redundancy from multiple sources. Some of the larger reservoirs appear to function in connection with a spring supply system and are situated to collect rainwater runoff; reservoir usage, therefore, is mainly to provide water for occasional peak requirements. Surface cisterns appear to be opportunistically placed to collect rainwater runoff; other than seasonal rain recharge, the numerous, widely scattered catchments appear to serve local community needs for supplemental supplies of lower quality water when piped water is not readily accessible. Traces of a south side piping system (Fig. 6) are found in front of the theater. Two parallel pipelines

Supplemental Water Supply Systems Supply Redundancy A number of cisterns and dams on Jebel el Khubtha (Akasheh 2003) (C;2) captured and stored rainfall runoff. Some of the upper level cisterns appear to have channels leading to ground level cisterns that led to urban housing or field areas to the west of Jebel el Khubtha to supplement the water supply from the Zurraba system. As previously mentioned, the Siq floor open channel was abandoned in late Nabataean phases

W

Water Management in Petra. Fig. 6 Dual pipelines continuing past the theater to supply tanks (T) above the Cardo.

2224


Water Management in Petra. Fig. 7 Great Temple on the south side of Wadi Mousa.

continue past the theater along the ridge (B;2) above the commercial district along the Roman Cardo (25), Hadrian’s Gate (43), upper and lower marketplaces and the Paradeisos water garden (Bedal 2004) to locations above the Great Temple (28, Figs. 1 and 7; Joukowsky 2001, 2003) where they form part of the water supply to structures located in (B;2). The two separate pipelines may indicate branch lines to separate destinations or a later elevation change that continues piping to Q’asr al Bint through the Great Temple to supply the Sacrificial Altar area, although no excavations exist to connect the multiplicity of subterranean canals below the altar to a specific water source. Hadrian’s Gate (43) separates the secular commercial district from the western sacred temple district containing the Great Temple, Temple of the Winged Lions (26), and Q’asr al Bint. The Paradeisos water garden complex west of the gate consisted of an open house structure situated on a platform island within a large water filled basin; bridge structures connected the island to outer precincts, and greenery added to the city’s elegance as indicated by reconstructions reported by Bedal (2004). The basin walls contain overflow channels as well as supply piping that may emanate from both the Nymphaeum piping extension into this area and water from a south side spring supply system. Distributed along this piping system, a number of elevated basins (T, Fig. 1) lined with hydraulic plaster (Fig. 8) served as receiving basins; earth-fill mound structures extended from these tanks to the lower Cardo area and served as pipeline support structures. As the basins are elevated 20 m on a bluff above the Cardo, sufficient head existed to provide pressurized water for fountains in the market area below as well as for the Great Temple (and possibly Q’asr al Bint). Because the south side urban

Water Management in Petra. Fig. 8 Fragment of one of the elevated water collection basins (T) above the Cardo area.

core region contains the marketplace area, water requirements were high; consequently, additional supplies are channeled to this area by means of an underground channel (B;1; B;2) from the combined flows from Ain Braq and Ain Ammon. Some as yet unexcavated branch of this system running through high elevation channels may be part of the system that provided water to piping located in front of the theater.


Water from these springs may be supplemented by elevated cistern storage water from the Jebel Attuf area (B;1) in one of the many high places (12, 13, B;1) of the city. Water to the Lion Fountain (14, B;1) and al Hamman pool area in the vicinity of elite tombs (16–18; B;1) came from this supply line that continued on to the Great Temple area and clearly indicated that a continuous spring supply is part of the system due to the presence of the Lion Fountain. A large elevated cistern located on a plateau above the Tomb of the Roman Soldier (16) (Browning 1982) also contributes rainfall runoff water supplies into this system. Details of the Wadi Farasa water system in this area (B;1) have been investigated (Schmid 2000) and indicate the existence of large reservoirs and piping systems that not only serve local usage, but also have sufficient capacity to transfer water further west to the Great Temple area. Numerous channels, pipelines, and multiple cisterns within, and leading from the Great Temple, indicate that water supplies within the temple were abundant (Joukowsky 1999, 2001, 2003). Water export lines to the marketplace area and the Q’asr al Bint region from the Great Temple served as part of the water system.

Water Supply System Management Operations The evolution of the water system to incorporate piping networks transformed the site to meet the demands of a large urban population estimated to reach 30,000 (Guzzo and Schneider 2002). The water system incorporated both intermittent, on-demand supplies piped from large reservoirs or drawn from cisterns and continuous supply piping systems from remote springs to provide daily requirements of city inhabitants. No water could be wasted. As a consequence, transfer piping from north side systems (Jebel el Khubtha and Siq pipelines to the Nymphaeum) provided water that could be transferred to south side downhill locations for further usage or storage before final discharge into the low elevation Wadi Siyagh. Water storage through use of major, on-site dams presented yet a further aspect of Petra’s water system. For example, on the north side of Wadi Mousa, numerous high status structures exist in the B;2 quadrant [Temple of the Winged Lions, Royal Palace (41), North Defense Wall and Fortress (35), Conway Tower (54)] and are logically associated with some water supply system. A dam (d) at Wadi Turkamaniya (B;2) may have trapped and stored sufficient runoff to provide water to the lower reaches of the Temple of the Winged Lions although no excavation data is available. Excavations reveal that lower portions of both the Temple of the Winged Lions and the Great Temple spanned the Wadi Mousa stream by means of bridging. It appears then that supply redundancy derived from pipelines from different spring/reservoir

2225

sources crossing from one to another part of the city is an aspect of the design approach. This design philosophy ensures that water supply to any area may be composed from different sources depending upon variations in individual spring flow rates and reservoir/cistern storage amounts and implies that management oversight was in place to monitor and control the system network. While cisterns are well dispersed through the urban settlement area, a main underground channel starting from Ain Bebdbeh north of (D;3) and running toward the convergence of Wadi Mataha and Wadi al Nassara (B;2), runs into the lower reaches of the north side below the temple areas. To illustrate the Nabataean mindset to utilize all water resources, a further element based upon on-site dams constituted yet further complexity to the water management picture. Local histories mention the existence of large dams – one on the Wadi Mataha (Taylor 2001), the other on the Wadi al Nassara (Fig. 1). Destroyed remains are found to verify that these dams provided water storage from rainfall runoff within urban Petra. Judging from Nabataean placement of the Wadi Mataha dam (d), piping to the nearby Nymphaeum must have been an additional third backup water source to the fountain. Since the Ain Mousa spring could also serve to place water behind the Wadi Mataha dam through the Wadi Shab Quais pipeline branch from Zurraba (in addition to seasonal rainfall runoff storage behind the dam from diversion of the Wadi Mousa stream through the bypass tunnel, 8 in Fig. 1), the water level behind the dam could be maintained to provide backup water to the Nymphaeum throughout the year. The Nymphaeum could then be supplied by contributions from stored runoff water behind the Wadi Mataha dam, a canal or pipeline from Ain Bebdbeh, the pipeline along the western face of Jebel el Khubtha feeding ground level cisterns and pipelines, and the north side Siq pipeline. This degree of redundancy indicates that planning for water supply variations was a consideration addressed by a complex design that could tap into various pipeline-water storage resources depending upon available supplies.

Flood Control, Groundwater Recharge, and the Great Temple Water Subsystem Floodwater drainage during the rainy season was a major concern. Since heavy rainfall and flooding characterize the Petra area, measures to divert Wadi Mousa floodwater from the Siq by means of a bypass tunnel (Fig. 1, 8 at C;1), a low dam at the Siq entrance and elevation of the Siq floor near the entrance provided flood control. While this strategy had proven effective in deflecting small flood events, continuous deliberate infilling and accumulating flood deposits in the Siq helped protect against floodwater incursion.

W

2226


Water Management in Petra. Fig. 9 FLOW-3D calculations of a rectilinear model of the Siq piping for entry full flow input flow rates of 1.0 (a), 2.0 (b), 5.0 (c), 10.0 (e) ft/s (0.305, 0.610, 1.52, 3.05 m/s) velocity indicating internal flow development for the observed wall roughness distribution. Fig. 9d shows results for smooth interior piping at 1.52 m/s initial, full flow inlet velocity.

While large flooding events had negative consequences, there were also ways to utilize the sudden water bounty: storage dams across the numerous wadis intersecting the urban core served to reduce floodwater entry into the city while seepage from the impoundments provided water table recharge suitable for well extraction during protracted drought. Thus a fraction of the seepage from dam storage, canals and pipelines ultimately can be recaptured and used as an ultimate groundwater defense against drought on a citywide basis. The same idea could also be used on a more localized basis for elite structures. Within the Great Temple, an elaborate south boundary wall drainage channel system collected infiltrated rainfall seepage and directed it to a nearby underground cistern with 50 m3 capacity located within the eastern, upper part of

the temple structure. A channel connected to the upper part of this cistern conducted overflow water to lower level structures before exiting to Q’asr al Bint and Wadi Siyagh in order to regulate its maximum capacity. Large channels located in an upper room north of the Theatron (Joukowsky 2003) of the temple most likely indicate the terminus of a subterranean channel from the Ain Braq, Ain Ammon system (B;1, B;2) with water transfer access to the cistern. Some additional water sources may have been available from springs in Wadis Kharareb and Ma’Aisert (Fig. 1) although piping connections await further excavations. Channel water, supplemented by cistern water to meet peak demands, was then distributed to subsidiary open cisterns located in east and west sides of the temple and then through subterranean channels under the lower temenos (a


Water Management in Petra. Fig. 10 Typical rippled wall pattern within the interior of some Nabataean piping elements.

sacred enclosure around a temple or holy site) platform to lower level rooms near the temple entrance stairway. Thus the cistern functioned as a reservoir adding stored runoff and seepage water to the channel-delivered base supply when required – much in the same way that previously described reservoir–pipeline systems worked to meet occasional peak demand requirements. Because such a system is contained within the temple itself, its position of importance as a major canal terminus and water distribution node is clear from the complexity of hidden channels, cisterns, and piping thus far discovered. Perhaps the internal water system of the temple, capable always of providing ample water supplies for rituals, had special significance to demonstrate the premier role of religion in the lives of the Nabataeans; only later under Roman rule are these supplies used for more utilitarian Cardo marketplace purposes indicating Roman predilection to practical concerns. While the north side piping of the Siq provided the main potable water supply, the south side channel system was probably meant for animal watering and may have been supplied by a channel from Ain Braq and supplemented from a cistern atop the bluffs with a drop hole to this channel (C0, Fig. 1).

Per Capita Water Availability For estimates of total water volumetric flow rate into the city, conservatively assuming about a third of the north side supply rate for the south side due to the less robust south side springs, and assuming a combined flow rate between the Siq and Wadi Shab Qais reservoir release lines to be about 40 m3 h−1, then the city could receive at minimum 50 m3 h−1 from these sources. While additional sources [Ain Braq (0.8 m3 h−1), Ain Dabdabah (2.5 m3 h−1), Ain Ammon, and Ain Siyagh (